Ret a penis Braue

ue

i Sri eons esate men ag

p - 2 pase sats bs seas Seo na SERPS Se
tart ep tribe ele ts separ
1 aan BUT oa ie de ' *

i N=

Perri

te Sinai
ae i cr fan “ eds

inet

oe a v4 Hieaaatote

Tae Rae saearn a
Cae fate va Pete 2
iit Aa

eae
i

shee
iReATeaeMMae eset ars

i ee pees
gy Lert aries oy ST ecietaeet isan eine wee
PAR oD By i a iene pies ES rae sae
ae reaticen ERAS pitas ie ris Ea Biers Sa
a : SEs aoteateietanarg page
i oe x emits
aie is peiciateviescoais

eur ey,

vise Seieie teee ty
patie areca

a sis ee Pee

ole oeciebekaucuaseasbsieeeneg pes Saar teste aepernthe eal SESS

iahwa oe Bo eels Zotar aieeolee Gece rae ipa ates eats ese

Panter aee : z E et pee
ates Pee pe Cele ee gerereemei Pees
piece eee ees a ie Siegert eens

weenie eedberansy iret atlantis Galea Sree reenter

Sich usc oe Biter eee Ee bos

ei Sue eeerts
: ty y natin toni atanonar et pitas as Sel eeertin
ay Thciebtein: Becton ier ete oe i ae Ee dethedcgee Sree oper se
Lew laaiel ih ea No tian dle Nat ag bmt aaa triebseses =Hindettoncs =
fees,
i

a aby babe iasbaeosioyes om soot ee Sine eerie
ni ay Sa usta cect eae ee ee tes
Shesateainy HERA Ge Pe = ; ee
es

Cornell University

The original of this book is in
the Cornell University Library.

There are no known copyright restrictions in
the United States on the use of the text.

http://www.archive.org/details/cu31924073870473

CORNELL UNIVERSITY LIBRARY

GAYLORO

Production Note

Comell University Library produced this volume to replace
the irreparably deteriorated original. It was scanned at 600
dots per inch resolution and compressed prior to storage
using CCITT/ITU Group 4 compression. The digital data
were used to create Cornell's replacement volume on paper
that meets the ANSI Standard Z39.48-1992. The production
of this volume was supported by the United States
Department of Education, Higher Education Act, Title 0-C.

Scanned as part of the A. R. Mann Library project to
preserve and enhance.access to the Core Historical Literature
of the Agricultural Sciences. Titles included in this
collection are listed in the volumes published by the Comell
University Press in the series The Literature of the
Agricultural Sciences, 1991-1996, Wallace C. Olsen, series
editor.

A\E\B 3%

iN

HANDBOOK OF
FLOWER POLLINATION

BASED UPON

HERMANN MULLER’S WORK

‘THE FERTILISATION OF
FLOWERS BY INSECTS’

BY

DR. PAUL KNUTH

FORMERLY PROFESSOR IN THE OBER-REALSCHULE IN KIEL, AND CORRESPONDING
MEMBER OF THE BOTANICAL SOCIETY DODONAEA IN GHENT

TRANSLATED BY
J. R. AINSWORTH DAVIS, M.A.

TRINITY COLLEGE, CAMBRIDGE

VOLUME II
(Il. BAND, I. TEIL OF THE GERMAN EDITION)

OBSERVATIONS ON FLOWER POLLINATION MADE IN
EUROPE AND THE ARCTIC REGIONS
ON SPECIES BELONGING TO THE NATURAL ORDERS

RANUNCULACEAE TO STYLIDIEAL

WITH 210 FIGURES IN THE TEXT AND A PORTRAIT OF
HERMANN MULLER

OXFORD
AT THE CLARENDON PRESS
1908

A vLe oS
HENRY FROWDE, M.A.
PUBLISHER TO THE UNIVERSITY OF OXFORD
LONDON, EDINBURGH
NEW YORK AND TORONTO

CONTENTS

ANGIOSPERMAE

Crass I. DicoTyLEDONES.

NaTuraL OrpER PAGE
I. Ranunculaceae /uss. i ; : ‘ , ; ; , i.

II. Calycanthaceae Lindl. . : : : ; ; ‘ : «= 63

III. Magnoliaceae DC. . ; : ‘ ‘ ; : : : aa
IV. Anonaceae Juss. . : ' ? ‘ ; : ; é me G5

V. Berberideae Veni. . ' : s ; : ; ‘ : . 5d

VI. Nymphaeaceae DC. : ‘ ; ‘ : ; ‘ ‘ . 59
VII. Sarraceniaceae Znd/. . , 5 : : i ‘ F . 6o
VIII. Papaveraceae DC. . ‘ : : ‘ ‘ ‘ : : . 61
IX. Cruciferae Juss. — , ; : : : : : ; : a 7

X. Capparideae Juss. . : : P : : : : . 128

XI. Resedaceae DC. . : : : . , : : 4 . 128
XII. Cistineae Dunai : : : 4 : : : : . 131
XIII. Violarieae DC... : : ; ; : , : ; . 135
XIV. Polygaleae Juss. . 2 5 : 3 F : : : . 146
XV. Caryophylleae Juss. ; : ; ; ‘ 5 : : . 149
XVI. Portulaceae Juss. . : ‘ : : : : : . 201
XVII. Tamariscineae Desv. . é ‘ : ; : , ; . 202
XVIII. Elatineae Camb. . ; : ; f ‘ : ; ‘ . 203
XIX. Hypericineae DC. . ; ; , : ‘ : : : 3203
XX. Malvaceae R. Br. . : ; ; : ; ; : ‘ . 206
XXI. Sterculiaceae Ven/. ‘ : ‘ ; ‘ ; 5 : . 213
XXII.\ Tiliaceae Juss. ; : : : : ‘ : : : res te)
XXIII. \Lineae DC. . 5 ‘ : ; : : ? , : . 214
XXIV. |Malpighiaceae Juss. ‘ : : : : : ; . 217
XXV. |Geraniaceae DC. . . : . : : i ; ; . 218
XXVI_ Rutaceae Juss. ' : 3 . ; é : é : . 239
XXVII. Hicineae DC. ; . : : : : : : : . 244
XXVIII. Celastrineae 2. Br. ‘ . ; ; : ; ; : . 244
XXIX. Rhamneae 2. Br. . ; ; : : : : ‘ . 246
XXX. Ampelideae #. B. ef A. . ‘ . : : , : ‘ . 250
XXXI. Sapindaceae Juss. . ‘ : ; : ‘ : : ‘ Beast
XXXII. Anacardiaceae Lind... : ; ‘ : ‘ : : . 258

XXXIII. Leguminosae /uss. : ; : ‘ : : . 259

iv

NaTuRAL ORDER

XXXIV.
XXXV.
XXXVI.
XXXVILI.

XXXVII.
XXXIX.
XL.

XLI.

XLII.
XLII.
XLIV.

XLV.
— XLVI.
NLVIL.

~ XLVIII.

XLIX.
L.
LI.

LII.

Rosaceae Juss.

Saxifrageae Ven/. .
Crassulaceae DC..
Droseraceae DC.
Bruniaceae &. Br.
Halorageae R. Br.

CONTENTS

Melastomaceae &. Br. .

Lythrarieae Juss. .
Onagrarieae /uss.
Loaseae Juss.

Passifloreae Juss. .

Cacteae DC.
Umbelliferae /uss.
Araliaceae Juss.
Cornaceae DC.

Caprifoliaceae /uss.

Rubiaceae DC.

. Valerianae DC.

Dipsaceae DC.

. Compositae Adaus.

. Stylidieae 2. Br.

Cucurbitaceae /uss.

Turneraceae H. B. ef XK.

NAW WN HOW

LIST OF ILLUSTRATIONS

. Nectaries of some Ranunculaceae (from nature) .

. Atragene alpina Z. (after Herm. Miiller)

. Pulsatilla vulgaris Z. (from nature). 3

. Pulsatilla vernalis Z. (after Herm. Miiller) .

. Ranunculus glacialis Z. (after Herm. Miiller)

- Ranunculus pyrenaeus Z. (after Herm. Miiller)

. Ranunculus parnassifolius 2. (after Herm. Miiller)

. Ranunculus Flammula Z. (after Herm. Miiler)

. Ranunculus auricomus Z. (after Herm. Miiller)

. Caltha palustris Z. (after Herm. Miiller)

. Trollius europaeus Z. (after Herm. Miiller) .

. Helleborus foetidus Z. (from nature)

. Helleborus viridis Z. (from nature)

. Nigella Z (from nature)

. Aquilegia vulgaris Z. (from nature) :

. Delphinium elatum Z. (after Herm. Miiller)

. Stages in Specialization of the Nectaries of Aconitum (after Kronfeld) :
; Map showing the distribution of the genera Aconitum and Bombus (after

Kronfeld)

. Aconitum Napellus LE (after Herm. Miiller)
. Aconitum Lycoctonum Z. (after Herm. Miller) .
. Berberis vulgaris Z. (after Herm. Miiller)

Chelidonium majus Z. (after Hildebrand)

. Diclytra spectabilis DC. (after Hildebrand) .

. Corydalis cava Schweigg. et Kort. (from nature) .

. Fumaria officinalis Z. (after Hildebrand)

. Nectaries of some Cruciferae (after Prantl) .

. Matthiola incana 72. By. (from nature) .

28. Nasturtium sylvestre 2. &r. (after Herm. Miiller)
. Arabis alpina Z. (after Herm. Miller) .

. Sisymbrium (from nature) .

. Lunaria annua ZL. (from nature)

. Draba aizoides Z. (after Herm. Miiller) :

. Kernera saxatilis Rezchd. (after Herm. Muller)

. Teesdalia nudicaulis A, B7. (after Herm. Miiller).

. Biscutella laevigata Z. (after Herm. Miiller)

. Reseda odorata Z. (after Herm. Miller)

. Viola calcarata Z. (after Herm. Miiller)

. Viola biflora Z. (after Herm. Miiller)

. Viola pinnata Z. (after Herm. Miiller) .

. Viola arenaria DC. (after Herm. Miiller)

. Polygala comosa Sché. (after Herm. Miiller)

. Polygala Chamaebuxus ZL. (after Herm. Miiller) .

. Polygala alpestris Rezchd. (after Herm. Miller)

. Gypsophila paniculata Z. (after Herm. Miiller)

. Gypsophila repens Z. (after Herm. Miiller) .

. Dianthus deltoides Z. (after Herm. Miiller).

. Dianthus superbus Z (after Herm. Miller)

. Dianthus Carthusianorum Z. (from nature) .

. Silene rupestris Z. (after Herm. Miiller)

. Silene acaulis Z. (after Herm. Miiller) .

. Cherleria sedoides Z. (after Herm. Miller) .

tac]
>
Q
&

HOOD MMNth

OS ol

: 180

LIST OF ILLUSTRATIONS

. Alsine verna Bardi. (after Herm. Miller)

. Honckenya peploides Er. (from nature) .

. Moehringia muscosa Z. (after Herm. Miller)

. Arenaria biflora Z. (after Herm. Miller)

. Stellaria graminea Z. (after Herm. Miiller) .

. Cerastium arvense Z. (after Herm. Miiller) .

. Cerastium trigynum V7//. (after Herm. Miller)

. Cerastium latifolium Z. (after Herm. Miiller)

. Myricaria germanica Desv. (after Herm. Miiller).
- Hypericum perforatum JZ. (after Herm. Miiller) .
. Malva sylvestris Z., and M. rotundifolia Z. (after Herm. Miller)
. Malva neglecta With. (from nature). ‘ .
. Anoda hastata Cav. (after Hildebrand)

. Linum catharticum Z. (after Herm. Miller).

. Geranium sylvaticum Z. (after Herm. Miiller)

. Geranium pyrenaicum Z. (after Herm. Miiller)

. Geranium molle Z. (after Herm. Miiller)

. Geranium pusillum Z. (after Herm. Miiller)

. Erodium cicutarium Z’Hérz¢. (from nature) .

. Tropaeolum majus Z. (from nature)

. Impatiens parviflora DC. (from nature)

. Impatiens glanduligera Zzzd/. (from nature)

. Dictamnus albus Z. (from nature) :

. Rhamnus pumila Z. (after Herm. Miiller)

. Rhamnus Frangula Z. (after Herm. Miiller)

. Vitis vinifera Z. (from nature) .

- Aesculus Hippocastanum Z. (after Herm. Miiller)
. Acer Z. (after F. Pax) . : ‘ :
. Rhus Cotinus Z. (after “Herm. Miiller).

. Apios tuberosa Joench. (after Taubert and Loew)
. Sarothamnus scoparius Koch (after Herm. Miller)
. Genista tinctoria Z. (after Herm. Miiller)

. Genista anglica Z. (after Herm. Miiller)

. Ulex europaeus Z. (from nature) .

. Cytisus Laburnum Z. (after Herm. Miiller) .

. Lupinus luteus Z. (after Herm. Miiller)

. Ononis spinosa Z. (after Herm. Miller)

. Medicago sativa Z. (after Herm. Miiller)

. Medicago falcata 2. (after Herm. Miiller) :
. Melilotus officinalis W/77/d. (after Herm. Miller)
. Trifolium repens Z. (after Herm. Miiller)

. Trifolium pratense Z. (after Herm. Miiller) .

. Trifolium alpinum Z. (after Herm. Miiller) .

. Trifolium pallescens Schred. (after Herm. Miller)
. Trifolium badium Schred. (after Herm. Miller) .
. Anthyllis Vulneraria Z. (after Herm. Miiller)

3. Lotus corniculatus Z. (after Herm. Miiller) .

. Robinia Pseud-acacia Z. (from nature)

. Astragalus depressus L. (after Herm. Miiller)

- Coronilla vaginalis Laz. (after Herm. Miller)

. Hippocrepis comosa Z. (after Herm. Miller)

. Hedysarum obscurum JZ. (after Herm. Miller)

. Onobrychis viciaefolia Sco. (after Herm. ae
. Vicia Cracca Z. (after Herm. Miller) .

. Vicia sepium Z. (after Herm. Miller) .

. Pisum sativum Z. (after Herm. Miiller)

. Lathyrus pratensis Z. (after Herm. Miiller).

. Phaseolus vulgaris Z. (from nature, and after Herm. Miiller)
. Rubus caesius Z. (from nature) ‘ : F :
. Rubus saxatilis Z. (after Herm. Miiller)

. Rubus arcticus Z. (after E. Warming) . :

. Dryas integrifolia VaA/ (after E. Warming).

No.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.

144.

145.
146.
147.
148.
149.

150.
ISI.

152.
153.
154.
155.
156.
157.
158.
159.

160.
161.
162.
163.

164.
165.
166.
167.
168.
169.
170.

171.
172.
173.

174.
175.

LIST OF ILLUSTRATIONS

Geum rivale Z. (from nature).

Potentilla minima Haller f. (after Herm. Miiller)
Sibbaldia procumbens Z. (after Herm. Miiller)
Alchemilla vulgaris Z. (after Herm. Miiller)
Alchemilla fissa Schum. (after Herm. Miiller)
Sanguisorba officinalis Z. (after Herm. Miiller)
Ulmaria pentapetala Gz/sd. (after Herm. Miiller) .
Spiraea sorbifolia Z. (after Herm. Miiller) :
Cotoneaster integerrima Medic. (after Herm. Miiller) :
Pyrus Malus Z. (after Hildebrand) .

Saxifraga Aizoon Jacg. (after Herm. Miller)
Saxifraga caesia ZL. (after Herm. Miiller)

Saxifraga oppositifolia Z. (after Herm. Miiller)
Saxifraga aizoides Z. (after Herm. Miiller) .
Saxifraga rotundifolia Z. (after Herm. Miiller)
Saxifraga stellaris Z. (after Herm. Miiller) ‘
Saxifraga stenopetala Gazd. (after Herm. Miiller)
Saxifraga granulata Z. (from nature)

Chrysosplenium alternifolium Z. (after E. Warming) .
Chrysosplenium tetrandrum /7zes (after E. Warming)
Parnassia palustris Z. (after Herm. Miiller) :
Ribes alpinum Z. (after Herm. Miiller)

Ribes nigrum JZ. (after Herm. Miiller) .

Ribes rubrum Z. (from nature) .

Ribes petraeum Wu/f. (after Herm. Miller)

Ribes Grossularia Z. (from nature)

Sedum acre Z. (after Herm. Miiller)

Sedum alpestre Vzd/. (after Herm. Miiller) .

Sedum atratum JZ. (after Herm. Miiller)

Sedum Telephium Z. (after Herm. Miiller) .

Sedum album JZ. (after Herm. Miller) .

Sempervivum Funckii Braun (after Herm. Miller)
Hippuris vulgaris Z. (from nature) 4

Lythrum Salicaria Z. (after Herm. Miiller) .

Scheme of the legitimate unions possible in Lythrum (after ‘Charles Darw in)

Epilobium Fleischeri Hochsz. (after Herm. Miiller)
Epilobium hirsutum Z. (after Herm. Miiller)
Epilobium parviflorum Schred. (after Herm. Miiller)
Epilobium alsinifolium V727. (after Herm. ee
Circaea lutetiana Z. (after Herm. Miiller)

Bryonia dioica Z. (after Herm. Miiller)

Geitonogamy of Chaerophyllum aromaticum Z. (after Kerner) ;

Hydrocotyle vulgaris Z. (after Drude) .
Astrantia major ZL. (after Herm. Miller)
Astrantia minor Z. (after Herm. Miiller)
Eryngium maritimum Z. (from nature) :
Eryngium campestre Z. (after Herm. Miller)
Conium maculatum Z. (after Herm. Miiller)

Pimpinella magna Z., var. B rosea, Koch (after Herm. Miiller) ;

Bupleurum stellatum ae (after Herm. Miiller)
Meum Mutellina Gaer/n. (after Herm. Miiller)
Gaya simplex Gazd. (after Herm. Miiller) . :
Anthriscus sylvestris Hoff. (after Herm. Miller)
Myrrhis odorata Scop. (after Herm. Miiller)
Cornus sanguinea Z. (after Herm. Miller) .
Adoxa Moschatellina Z. (after Herm. Miiller)
Sambucus nigra Z. (after Herm. Miiller)
Viburnum Opulus Z. (after Herm. Miiller) .
Symphoricarpos racemosa J/ichx. (after Herm. Miiller)
Linnaea borealis Z. (after Herm. Miller)
Lonicera Caprifolium Z. (after Herm. Miiller)
Lonicera Periclymenum Z. (from nature)

LIST OF ILLUSTRATIONS

. Lonicera nigra Z. (after Herm. Miiller)

. Lonicera caerulea Z. (after Herm. Miiller) .

. Lonicera alpigena Z. (after Herm. Miiller) .

. Asperula cynanchica Z. (after Herm. Miiller)

. Asperula taurina Z. (after Herm. Miiller)

. Galium Mollugo Z. (after Herm. Miiller)

. Galium sylvestre Poddich (after Herm. Miller)

. Galium verum Z. (after Herm. Miiller)

. Valeriana officinalis Z. (from nature) .

. Valeriana montana Z. (after Herm. Miiller) .

. Valeriana tripteris Z. (after Herm. Miiller) .

. Knautia arvensis Cozdt. (after Herm. Miller)

. Succisa pratensis AZoench. (after Herm. Miller).

. Styles and Stigmas of Compositae (from nature) .

. Eupatorium cannabinum Z. (after Herm. Miiller) g
. Geitonogamy of Eupatorium cannabinum Z. (after Kerner)
. Adenostyles alpina Bluff et Fing (after Herm. Miiller)

. Homogyne alpina Cass. (after Herm. Miiller)

. Petasites albus Gaertn. (after Herm. Miiller)

. Aster Tripolium Z. (from nature)

. Chrysocoma Linosyris Z. (after Herm. " Miiller)

. Bidens tripartita Z. (from nature)

. Gnaphalium Leontopodium LZ. (after Herm. * Miiller)

. Achillea Millefolium Z. (after Herm. Miller)

. Achillea moschata /Jucg. (after Herm. Miller)

. Tanacetum alpinum Sch. zp. (after Herm. Miiller)

. Chrysanthemum segetum Z., Anthemis arvensis /., Tanacetum "Parthenium we

and Matricaria Chamomilla Z. (from nature)

. Senecio carniolicus W2¢/d. (after Herm. Miller) .
. Echinops sphaerocephalus Z. (after Herm. Miiller)
. Cirsium arvense Scop. (after Herm. Miiller)

. Carduus defloratus Z. (after Herm. Miiller) .

. Centaurea Cyanus ZL. (after J. MacLeod)

. Arnoseris minima Dum. (from nature)

. Leontodon autumnalis Z. (after Herm. Miller)

. Hieracium umbellatum Z. (after Herm. Miiller)

OBSERVATIONS ON FLOWER POLLINATION

EUROPE AND THE ARCTIC REGIONS

(The abbreviations made use of in the bibliographical references are—where possible—those of
the ‘International Catalogue of Scientific Literature.’ The following are employed in the statements
regarding the flower-visits of insects:—nect-lkg., nectar-licking ; skg., sucking; po-cltg., pollen-
collecting; po-dvg., pollen-devouring ; freq., frequent. When there is no special reference to
locality, Knuth’s observations are to be regarded as having been made in Schleswig-Holstein
(more especially near Kiel), those of Hermann Miller in the neighbourhood of Lippstadt (West-
phalia), those of Buddeberg in Nassau, and those of Borgstette in Tecklenburg.]

ANGIOSPERMAE
CLASS I. DICOTYLEDONES

I. ORDER RANUNCULACEAE JUSS.

Lirerature.—Herm. Miller, ‘ Fertilisation.,’ pp. 69-89; Knuth, ‘Bl. u. Insek.
a. d. nordfr. Ins.,’ pp. 16-17, ‘Grundriss d. Bliitenbiol.,’ pp. 15-17.

The Ranunculaceae are characterized by a great variety of flower arrangements,
such as is met with in but few other orders. The conspicuousness of the flowers
is sometimes due to the corolla (Ranunculus, Batrachium, Adonis), sometimes to
the calyx (Clematis, Hepatica, Pulsatilla, Anemone, Caltha, Trollius, Helleborus,
Eranthis, Aconitum), sometimes to both (Aquilegia, Delphinium), and sometimes
even to the stamens (Thalictrum). The colour of the flowers is frequently white,
greenish, or yellow (Anemone, Batrachium, Ranunculus, Myosurus, Caltha, Trollius,
Helleborus, Eranthis, Actaea), more rarely red, blue, or violet (Pulsatilla, Atragene,
Hepatica, Adonis, Aquilegia, Delphinium, Aconitum, Paeonia). The modes of
secretion and concealment of nectar are quite as varied as the devices by which
conspicuousness is brought about. Nectar may be secreted by the sepals (some
peonies), by the stamens (Pulsatilla), or by the carpels (Caltha), but usually by
the petals, either at their bases (Batrachium, Ranunculus, Myosurus), or in specialized
nectaries resulting from their metamorphosis (Trollius, Helleborus, Eranthis, Aqui-
legia, Aconitum, Nigella). Comparison of some of the nectaries in the order
(Fig. 1) shows a gradual transition from the perfectly simple nectar-pit of Ranunculus
to the complex apparatus of Aconitum. In Trollius the petal is much reduced
and possesses an elongated nectar-groove above its base, in Helleborus the expanded

DAVIS, II B

2 ANGIOSPERMAE—DICOTYLEDONES

limb has already completely disappeared, so that only a nectar-cup remains. In
Aquilegia there is a similarly formed organ, but much larger, inverted, and curved
round at the tip, serving not only to secrete and conceal nectar, but also to attract
insects, being therefore brightly coloured. The nectary of Aconitum is shaped
much like that of Aquilegia, but it is smaller and provided with a long stalk—the
claw of the original petal—which serves to conceal nectar more deeply. The
remarkable nectary of Nigella is not connected by any intermediate gradations
with the preceding.

1 2 3
DB
n
n
Fic 1. Wectaries of some Ranunculaceae. (Enlarged. From nature.)
1, Ranunculus sceleratus, L. 4. Aguilegia vulgaris, L.
2. Trollius europaeus, L. 5. Aconttum Napellus, L.
3. Helleborus niger, L. 6. Nigella arvensis, L.
m, nectary. p, limb. s, stalk. d, cover. h, protuberance.

In not a few Ranunculaceae no nectar is secreted (Clematis, Thalictrum,
Anemone, Hepatica). These provide pollen for their insect-guests, as a reward for
their work in securing cross-pollination. There are therefore some Ranunculaceae
in almost all the flower-classes, as follows.—

Po or An: Clematis (most species), Thalictrum, Anemone, Hepatica, Adonis,
Actaea ;

E: Myosurus, some sp. of Ranunculus and Batrachium ;

EC: Ranunculus, Batrachium, Caltha, Eranthis, Isopyrum, Cimicifuga ;

C: Pulsatilla, Trollius, Helleborus ;

H: Aquilegia, Delphinium, Aconitum, Atragene, Nigella.

Species belonging to the flower-classes Po, An, EC, and C, are homogamous
or slightly protandrous, more rarely protogynous. In these—owing to the relative
position and time of maturation of stamens and carpels—automatic self-pollination
is possible in the later stages of flowering should insect-visits have failed. In species
belonging to class H—on the other hand—self-pollination is largely prevented by
marked protandrous dichogamy, and the visits of bees are often indispensable for
fertilization.

The visitors and pollinators belong to all insect-orders. The white, yellowish-
green, and yellow pollen flowers, and the similarly coloured flowers with readily
accessible nectar, are visited chiefly by short-tongued insects, especially flies and
beetles, more rarely by Hymenoptera, still more rarely by Lepidoptera. The blue
pollen flower Hepatica—on the other hand—is chiefly visited and pollinated by

RANUNCULACEAE 3

pollen-seeking bees. Trollius—a yellow flower with concealed nectar—receives
visits from Hymenoptera, Diptera, and Coleoptera, in about equal numbers, while
the violet Pulsatilla is pollinated almost exclusively by bees. Aquilegia, Delphinium,
and Aconitum, are characteristic humble-bee flowers; Nigella and Atragene are
equally characteristic bee flowers.

1. Clematis L.

Mostly pollen flowers with petaloid sepals, by which conspicuousness is secured.

1. C. Vitalba L. (Herm. Miller, ‘ Weit. Beob.,’ I, p. 312; Schulz, ‘ Beitrage,’
I, p. 1; Kirchner, ‘Flora v. Stuttgart,’ p. 258; Loew, ‘Blitenbiol. Floristik,’
p- 175; Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, IL; Knuth, ‘Bloemenbiol. Bijdragen,’
‘Bliitenbiol. Notizen.’)—Protogynous pollen flowers. The flowers are arranged in
dense cymes, carried high into the air by the climbing stem, so that the white
reflexed sepals, and the white stamens, make the plant conspicuous from a distance.
The hawthorn-like odour of the flowers (due to trimethylamine) serves as a further
means of attraction. The flowers are slightly protogynous (according to Schulz
occasional ones may be homogamous), and at the time of flowering the numerous
erect undehisced stamens are rather lower than the stigmas, which are already
mature. The filaments now elongate, and bend outwards as the anthers dehisce.
Since the outer stamens are the first to ripen, self-pollination is at first rendered
difficult. But as the stigmas continue receptive till the innermost stamens are ripe,
it becomes easy in the final stage of flowering.

The visitors and pollinators are pollen-collecting bees and pollen-devouring
flies. These necessarily effect cross-pollination when they pass from one blossom
to another, alighting in the middle of the flower upon the rather prominent stigmas.
It is very difficult to observe their actions because of the height of the plant.

Vistrors.—The following have been observed by Buddeberg (Budd.) in Nassau,
by Hermann Miiller (H. M.) in Westphalia, and by myself (Kn.) in Holstein.—
A. Diptera. (@) Muscidae: 1. Musca domestica Z. (Kn.); 2. Sarcophaga carnaria
L.(Kn.); 3. Scatophaga stercoraria Z. (Kn.); all po-dvg. (4) SyrpAzdae: 4. Eristalis
nemorum ZL. (Kn.); 5. E. tenax Z. (Kn.); 6. Rhingia rostrata Z. (Kn.); 7. Syrphus
balteatus Deg. (Kn.); 8. S. ribesii Z. 9 (Kn.); all po-dvg. B. Hymenoptera.
(a) Apidae: 9. Apis mellifica Z. (Kn., H. M., very freq. in Thuringia); ro. Andrena
albicans Jdl/. (Kn.); 11. Halictus calceatus Scop. (Kn.); 12. H. nitidiusculus
K. 9 (Budd.); all po-cltg. (4) Vespzédae: 13. Odynerus parietum L. 6 (Budd.).

Loew observed the following in Steiermark (‘Beitrage,’ p. 46).—A. Diptera.
Syrphidae: 1. Syrphus lunulatus J/g., po-dvg.(?). B. Hymenoptera. Apidae:
2. Halictus malachurus X. 9, po-dvg.

MacLeod observed in the Pyrenees one humble-bee, and 5 Diptera, as visitors
{‘Pyreneenbl.,’ p. 389).

2. C. recta Z. (Herm. Miiller, ‘ Fertilisation,’ p.69 ; Schulz, ‘ Beitrige,’ I, p. 1;
Beyer, ‘D. spont. Bewegungen. d. Staubgefasse u. Stempel’; Knuth, ‘ Bloemenbiol.
Bijdragen’; Kirchner, ‘Flora von Stuttgart,’ pp. 258-9.)

While the last species generally exhibits slight protogyny, C. recta is feebly
protandrous. The white flowers are arranged in terminal paniculate cymes. When
they open, the stigmas are not yet fully mature, and are covered by the stamens
which closely invest them. The outermost of these at once bend outwards and their

B2

4 ANGIOSPERMAE—DICOTYLEDONES

anthers dehisce, so that the flowers are in the first or male condition, in which
they yield pollen to insects, but are incapable of retaining on their stigmas any
that may be brought. Automatic self-pollination is equally impossible at this stage.
The outward curving and dehiscing of the stamens progresses centripetally, but
before the innermost ones have dehisced the stigmas have matured, and are liable
to be touched by such pollen-covered insects as may alight in the middle of the
flower. Bees collecting the abundant pollen almost invariably alight in the centre—
according to Hermann Miiller’s observations—while pollen-devouring flies are very
erratic in their mode of settling and, creeping about the flowers, may effect either
cross- or self-pollination. Failing insect-visits, self-pollination may easily be effected
by contact of the stigmas—which remain receptive—with the innermost stamens.

Vistrors.—Hermann Miller observed the following.—A. Coleoptera. Scara-
baezdae: 1. Trichius fasciatus Z., devouring the anthers. B. Diptera. (2) Aluscidae :
2. Prosena siberita F.; (5) Syrphidae: 3. Eristalis arbustorum Z.; 4. E. sepulcralis
L.; 5. Helophilus floreus Z.; 6. Syrphus pyrastri Z.; 7. Syritta pipiens Z.; 8. Xylota
ignava Pz.; 9. X. lenta Me.; all po-dvg. C. Hymenoptera. (a) Apzdae: 10.
Andrena albicans ful. 9; 11. A.gwynana A’ 9; 12. Apis mellifica Z. ¥; 13. Bombus
terrester Z.9; 14. Halictus sexnotatus A.9; 15. Osmia rufa Z.9; 16. Prosopis
signata Pz. 9; all po-cltg. (4) Sphegedae: 17. Gorytes mystaceus Z., perhaps only
hunting flies; 18. Oxybelus uniglumis Z., po-dvg. (c) Vespzdae: 19. Odynerus
parietum Z. 9, as 17. Handlirsch mentions as a visitor the fossorial wasp Gorytes
mystaceus Z.

On garden plants I saw only one pollen-devouring hover-fly, Eristalis tenax Z.

3. C. Viticella L. (Knuth, ‘ Bloemenbiol. Bijdragen’) is nectarless like the
preceding, despite its very large dark-violet, blue, or red pollen flowers. I have
only once observed the honey-bee collecting pollen on plants that were grown
in Kiel for the purpose of covering a bower. No observations are available from
the Mediterranean region, which is the home of this form.

4. C. balearica Rich. (=C. cirrhosa Z.) is a nectar flower indigenous in the
Mediterranean region. The outer stamens—according to Delpino—are modified
into spoon-shaped nectaries. The same observer mentions Bombus and Xylocopa
as visitors of this species.

5. C. integrifolia L. is also a nectar flower. According to Delpino (‘ Appli-
cazione d. teoria Darwiniana,’ p. 8) the inner stamens secrete nectar. The pendulous
flowers—according to Kerner (‘Nat. Hist. Pl.” Eng. Ed. 1, II, pp. 349-50)—are
protogynous for a short time, and therefore adapted for cross-pollination at the
beginning of flowering. The stamens lie close together so as to form a short tube,
in the base of which the numerous still immature stigmas are situated, while the
outer anthers have already dehisced, thus furthering cross-pollination. The anthers
of the inner stamens gradually dehisce, but, owing to the pendulous position of
the flower, would be unable to effect self-pollination were there not an elongation
of the carpels during the last two days of flowering, so that if pollination has not
been effected by insects, the stigmas—spreading out to some extent—receive some
of the pollen still adhering to the stamens.

6. C. angustifolia Jacq.—

Visitors. — Loew observed the following in the Berlin Botanic Garden.—
Hymenoptera. <Aprdae: Bombus terrester Z. 9, po-cltg.

RANUNCULACEAE 5

2. Atragene L.

Homogamous bee flowers. The large sepals serve for attraction; the small
petals are converted into nectaries.

7. A. alpina L. (=Clematis alpina Miller), (Herm. Miller, ‘ Alpenblumen,’
pp. 124-5; Ricca, ‘Oss. sulla fecondaz. incroc. d. veget. alp. e subalp.’; Kerner,
‘Nat. Hist. Pl.” Eng. Ed. 1, II, pp. 349-50; Schulz, ‘ Beitrage,’ p. r.)—A characteristic
bee flower. According to Hermann Miller, this plant is homogamous in the Alps,
but self-pollination is quite excluded, so that fertilization is entirely due to visitors,
which are chiefly Apidae (see Fig. 2).

Fic. 2. Alragene alpina, L. (after Hermann Miller). .4. Flower seen from the side(x 4). 2B. The
same after removal of asepal. C. One of the four large petals. 0. One of the innermost small petals,
with a small anther-lobe on one side at the end. £. A stamen seen from the side. #. The same seen
obliquely from within. |G. A carpel.

According to Kerner the flower mechanism of Atragene alpina completely
agrees with that of Clematis integrifolia, self-pollination of old flowers being brought
about automatically by elongation of the carpels.

Visttors.—Herm. Miiller observed one bee (Eucera) and one beetle; Ricca
observed humble-bees; Schulz noticed Apidae more particularly.

3. Thalictrum L.

Pollen flowers in which the stamens serve for attraction, or anemophilous
flowers occasionally visited by insects. Kerner observed opening and closing of
the anthers as a result of variations in the humidity of the air.

8. T. aquilegifolium L. (Herm. Miller, ‘ Fertilisation,’ p. 70, ‘Alpenblumen,’
p.115; Beyer, ‘D. spont. Bewegungen d. Staubgefasse u. Stempel’; Ricca, ‘ Oss.
sulla fecondaz. incroc. d. veget. alp. e subalp.’; Schulz, ‘ Beitrage,’ II, pp. 1-2;
Kerner, ‘Nat. Hist. Pl. Eng. Ed. 1, II; Knuth, ‘Bloemenbiol. Bijdragen.’)—This
tall plant is conspicuous from a distance, owing to the many stiff club-shaped
stamens of pale violet colour, which radiate from the centre of the numerous

6 ANGIOSPERMAE—DICOTYLEDONES

crowded flowers. In young flowers the stigmas—mature according to Miiller,
partly immature according to Schulz and Ricca—are covered by the unripe inner
stamens, being thus shielded from contact with insect visitors, which, in this first
condition of the flower, get dusted with pollen as they creep about collecting or
devouring this kind of flower-food. The inner stamens subsequently ripen and
spread out, so that insects alighting in the middle of the flower must necessarily
effect cross-pollination if they come from other flowers of the same species. Should
insect-visits fail, self-pollination takes place automatically, for the stigmas are situated
in the line of fall of pollen from the inner anthers.

It is also possible that the light and but slightly adhesive pollen may be carried
by the wind, but—as Schulz remarks—anemophily cannot often obtain, for the
closely crowded stamens keep pollen from the stigmas.

Vistrors—According to the observations of Hermann Miiller (H. M.), and
myself (Kn.), these are pollen-collecting bees or pollen-devouring flies and beetles.—
A. Coleoptera. (a) Nitidulidae: 1. Meligethes (Kn.). (6) Scarabaerdae: 2. Trichius
fasciatus Z., devouring the anthers (H.M.). B. Diptera. Syrphidae: 3. Eristalis
arbustorum Z. (H.M.); 4. E. nemorum Z. (H.M.); 5. E. pertinax Scop. (Kn.);
6. E. sepulcralis Z.(H. M.); 7. E. tenax Z.(Kn., H.M.); 8. Rhingia rostrata 2.
(H. M.); 9. Syrphus balteatus Deg. (Kn.); all po-dvg. C. Hymenoptera. Apidae:
10. Apis mellifica Z., po-cltg. (Kn., H.M.); 11. Halictus cylindricus /. 9 (Kn.);
12. H. sexnotatus X. 9, po-cltg. (H.M.); 13. Prosopis signata Pz. 8 9, po-cltg. (H. M.).

In the Alps Herm. Miiller also saw 4 flies, 3 beetles, 3 Hymenoptera.

Loew observed the following in the Berlin Botanic Garden.—A. Coleoptera.
Scarabaetdae: 1. Cetonia aurata Z., devouring the anthers. B. Diptera. Syrphrdae:
2. Syritta pipiens Z., po-dvg. C. Hymenoptera. Apidae: 3. Apis mellifica Z.,
po-cltg.

g. T. alpinum L.—A pollen flower. The stamens projecting far out of the
pendulous flowers indicate that the plant is anemophilous. According to Lindman
the stigmas mature before the stamens, but remain in a receptive condition during
the dehiscence of the anthers. According to Ekstam, the flowers are protogynous-
homogamous in Nova Zemlia. According to Kerner, the stigmas of this species
are at first concealed under the sepals: but after the sepals have fallen away they
may be geitonogamously pollinated by neighbouring flowers. The same is true of—

10. T. flavum L. (Herm. Miiller, ‘ Fertilisation,’ p. 70; MacLeod, ‘ Bot. Jaarb.
Dodonaea,’ Ghent, vi, 1894; Schulz, ‘ Beitrige,’ II, p. 2.)\—According to Warnstorf
(Verh. bot. Ver., Berlin, xxxvii, 1896) the flowers are slightly protogynous to homo-
gamous. The pollen is yellow, polyhedral, smooth, about 25-30 p» in diameter.

VisiTors.—I observed the following in the gardens at Kiel.—Apis mellifica
Z.%, and Bombus lapidarius Z.; po-cltg. Hermann Miiller also observed the former
in the Lippe meadows, near Lippstadt, as well as a number of pollen-devouring
Diptera :—(a) Syrphidae: 1. Eristalis arbustorum Z.; 2. E.nemorum Z.; 3. E. tenax
L.; 4. E. sepulcralis Z.; 5. Syritta pipiens Z. (6) Afuscidae: 6. Pollenia vespillo F.

1. T, minus L. (Herm. Miiller, ‘Weit. Beob.,’ I, pp. 312-13; Kerner, ‘ Nat.
Hist. Pl.’ Eng. Ed. 1, II; Schulz, ‘ Beitrige,’ II, p. 2; Knuth, ‘Bliitenbiol. Beob.
a. d. Ins. Riigen.’)—The filaments—which are narrowed at the base—project loosely
from the nectarless flowers, and are moved by every breath of wind. The plant is
therefore to be regarded as anemophilous. Owing to the sulphur-yellow colour of
the stamens, however, the flowers are very conspicuous, so that insects now

RANUNCULACEAE , 7

and then visit them. The plant may consequently be said to fluctuate between
anemophily and entomophily, for which reason I term its blossoms ‘wind flowers.’
Visitors effect cross- and self-pollination with equal ease. There being well-marked
protogyny, crossing is frequently secured, even when the pollen is carried by the
wind. While Herm. Miiller found the flowers to be strongly protogynous in
Thuringia, they are—according to Schulz—quite homogamous, or but feebly proto-
gynous in South Tyrol.

According to Kerner, the stigmas are at first hidden under the sepals. Geitono-
gamy becomes possible when these are shed.

Visitors. —On the island of Riigen I observed the following.—Diptera.
Syrphidae: Eristalis tenax Z. po-dvg. Buddeberg (in Nassau) noticed Syrphus sp.,
po-dvg. Hermann Miiller (in Thuringia) observed a beetle (Oedemera virescens L.),
po-dvg. Schulz noticed various flies, bees, and beetles. In Dumfriesshire (Scott-

Elliot, ‘Flora of Dumfriesshire,’ p. 1), one of the Muscidae was observed on the
flowers.

12. T. glaucophyllum Wend.—

Visitors. — Loew observed the following in the Berlin Botanic Garden.—
Coleoptera. Scarabaeidae: Cetonia aurata Z., devouring the anthers.

4. Hepatica Dillenius.

Pollen flowers. The calyx-like involucre—situated close to the flower—serves
for attraction. Sometimes gynomonoecious and gynodioecious.

13. H. triloba Gilibert (= Anemone Hepatica Z.). (Sprengel, ‘Entd. Geh.,’
p- 291; Herm. Miiller, ‘Weit. Beob.,’ I, p. 313; Schulz, ‘ Beitrage,’ II, p. 178;
Calloni, Justs bot. Jahresber., Leipzig, xiii (1885), 1887, p. 751; Kerner, ‘ Nat.
Hist. Pl.’ Eng. Ed. 1, II, pp. 194, 213; Schréter, Arch. Sci. Phys. Geneéve, xiv,
1885, p. 283; Knuth, ‘Bloemenbiol. Bijdragen.’)— The involucre—composed of
several bracts—is so near the open flower expanded in the sunshine, that it looks
like a calyx, and its dark-blue colour is very conspicuous among the fallen yellow
leaves of beeches and hazels. According to Kerner, the involucral bracts double
in length during the eight days of blooming, thus making the flower still more
conspicuous. Miiller states that the outer stamens develop simultaneously with the
carpels, from which they curve away, so that insects’ visits at this stage may result in
cross-pollination. The inner stamens mature later, and self-pollination must then
result automatically. According to Warnstorf (Schr. natw. Ver., Wernigerode, xi,
1896, p. 1) the flowers are protogynous. The connective of the two white anther-
lobes—which dehisce laterally—is white or violet. The stamens project beyond the
stigmas, so that autogamy is ultimately unavoidable. According to Schulz and
Schréter, gynomonoecious and gynodioecious plants occur sporadically.

Vistrors.—These have been observed chiefly by Hermann Miiller (H. M.) and
myself (Kn.). They are mostly pollen-devouring and pollen-collecting insects,
which creep about upon the open flowers and may lead to either cross- or self-
pollination. They are as follows——A. Coleoptera. Staphylinidae: 1. Staphylinus?
(Sprengel at Spandau). B. Diptera. Syrphidae: 2. Eristalis tenax L., po-dvg.
(Kn., H.M.). C. Hymenoptera. <Afzdae: 3. Apis mellifica Z. ¥, (Kn., freq. in
Kiel Garden, po-cltg.; H. M., very freq. in Westphalia); 4. Osmia rufa Z. 9 (H. M.),

8 4ANGIOSPERMAE—DICOTYLEDONES

possibly skg. nectar. D. Lepidoptera. Rhopalocera: 5. Rhodocera rhamni JZ.
(H. M.), settling for a considerable time and probing various parts of the base of
the flowers with the tip of its extended proboscis; 6. Vanessa urticae Z. (Kn.),
possibly skg. nectar.

14. H. angulosa DC.—

Loew observed on the flowers in the Botanic Garden at Berlin.—Hymeno-
ptera. Apidae: Apis mellifica LZ. ¥, po-clig.

5. Pulsatilla Tourn.

Protogynous flowers with concealed honey (rarely pollen flowers). Nectar
secreted by external vestigial stamens. The large brightly-coloured sepals serve
for attraction. In addition to the hermaphrodite flowers, some species (P. vulgaris,
vernalis, pratensis, and montana) possess others which are andromonoecious and
androdioecious, gynomonoecious, and gynodioecious.

15. P. vulgaris Miller (= Anemone Pulsatilla Z.). (Sprengel, ‘Entd. Geh.,’
p- 290; Herm. Miiller, ‘Weit. Beob.,’ I, pp. 313-14; Schulz, ‘ Beitrage,’ I, p. 2;
Knuth, ‘ Bl. u. Insek. a. d. nordfr. Ins.,’ pp. 17, 147, ‘Bloemenbiol. Bijdragen.’)—

Fic. 3. Pulsatilla vulgaris, L. (from nature). 1. Flower from which the two anterior sepals have
been removed. 7, nectaries. The numerous stigmas project beyond the anthers. 2. Stigma (s) of a central
carpel. 3. Stigma (s) of a marginal carpel. (2 and 3 highly magnified.)

The large blue-violet sepals form an effective means of attraction to the erect flowers.
In the first flowering-stage the stigmas are already receptive, and they remain in this
condition for the next two to four days, while the extremely numerous stamens are
dehiscing. Nectar—as in the case of the following species—is secreted by the outer-
most stamens, which are converted into stalked capitate structures. As the stigmas
project far above the longest stamens (see Fig. 3, 1), pollen-collecting and nectar-
sucking insects first come into contact with them, and effect cross-pollination if they
have already visited a flower of the same species.

Visitors. — Hermann Miiller (H. M.) and myself (Kn.) have chiefly observed
nectar-sucking and pollen-collecting bees as pollinators, and ants as nectar-thieves.—
A. Hymenoptera. (a) Apzdae: 1. Apis mellifica Z. 9, skg. and po-cltg. (Kn.,
freq. at Kiel; H.M., Thuringia); 2. Bombus lapidarius Z. 9, skg. (Kn., Kiel; H.M.,
Thuringia); 3. B. terrester Z. 9, skg., holding on by stamens and peduncles during
the process, and working round the flower to empty the nectaries in succession (Kn.,

RANUNCULACEAE 9

Kiel, freq.; H.M., Thuringia); 4. B. hortorum Z. 9 (Kn.), as the last. H.M. also
observed in Thuringia: 5. Andrena gwynana &.4 skg.; 6. Halictus cylindricus, F. 9,
po-cltg., freq.; 7. H. morio F. 9, po-cltg. (4) Formicidae (all nectar-thieves); 8. Lasius
alienus Mverst. §; 9. Leptothorax interruptus Schenck ¥; 10. Myrmica levinodis My. ¥;
11. M. ruginodis yl. §; 12. M. scabrinodis Wy/.¥; 13. Tapinoma erraticum Lar. ¥.
B. Coleoptera. (a) Mitidulidae: 14. Meligethes, nect-lkg. (4) Afeloidae: 15.
Larvae of Meloé. C. Hemiptera. 16. Aphanus vulgaris Sch. D. Thysano-
ptera. 17. Thrips, very freq.
Schenk observed in Nassau, Osmia rufa JZ. 9.

16. P. pratensis Miller. (Sprengel, ‘Entd. Geh.,’ p. 289; Francke, ‘ Einige
Beitr. zur Kennt. d. Bestaubungseinricht. d. Pfl.’; Loew, ‘ Bliitenbiol. Floristik,’
p- 390; Schulz, ‘ Beitrige,’ I, p. 3; Knuth, ‘ Bloemenbiol. Bijdragen.’)—The dark-
violet sepals of the large pendulous flowers are arranged in the form of a bell,
so that they constitute a roof by which the stamens and carpels are protected from
rain. The flower mechanism agrees with that of the preceding species. Protogyny
excludes autogamy. Warnstorf( Verh.
bot. Ver., Berlin, xxxviii, 1896)
describes the flowers found in the
neighbourhood of Ruppin as first
protogynous and subsequently ho-
mogamous. Pollen of a shining
white; the grains are about 37 » in
diameter.

Visitors. — Loew (L.) and my-
self (Kn.) have so far only observed
bees as pollinators. On alighting,
these first touch the stigmas—which
project beyond the stamens—and
then holding on to the crowded
stamens, collect pollen or suck
nectar.—Hymenoptera. Apidae:
1. Apis mellifica Z. ¥, and 2. Bom-
bus hortorum Z. 9 (Kn., Kiel), skg. Fic. 4. Pulsatilla vernalis, L. (after Herm. Miller), 4. A
and po-cltg. ; 3 Osmia bicolor few of the external stamens modified to form nectaries (#), and

eS behind them a few fully developed stamens (a). &. Apex of
Schr. 9 (L., Brandenburg meadow), tie ith stigma (); highly magnified.
po-cltg.

17. P. vernalis Miller. (Beyer, ‘D. spont. Bewegungen d. Staubgefasse u.
Stempel’; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, 11; Herm. Miiller, ‘ Alpenblumen,’
pp. 125-7; Ricca, ‘Oss. sulla fecondaz. incroc. d. veget. alp. e subalp.’; Schulz,
‘ Beitrage,’ II, pp. .2-4.)}—The sepals—which spread out in the sunshine—are white
on the inner side, bright violet to rose-red externally. They serve to attract insects.
The flower mechanism deviates in some points from that of the two preceding
species: protogyny is usually much less marked, and there may even be almost homo-
gamous flowers. Further, according to Beyer, dehiscence of the anthers progresses
outwards and inwards from a central zone. Schulz observed flowers which were
quite nectarless. Lastly, Kerner noted, in addition to cross-pollinated hermaphrodite
flowers with short stamens, others with long stamens, in which self-pollination auto-
matically took place on the closing of the flower. Plants that Lindman observed

ie) ANGIOSPERMAE—DICOTYLEDONES

on the Dovrefjeld were similarly capable of autocarpy as a result of automatic self-
pollination.
Visrrors.—Lindman observed a fly visiting the flowers on the Dovrefjeld. In

the Alps, Herman Miiller observed 6 Hymenoptera, 12 Diptera, 4 Lepidoptera,
and 2 Coleoptera.

18. P. patens Miller.—According to Kerner, this essentially agrees with
P. vulgaris in the arrangements of its large violet-blue protogynous flowers.

Vistror.—I observed in the Botanic Garden at Kiel, Bombus hortorum Z.,
skg. and po-cltg.

19. P. montana Hampe, and 20. P. transsilvanica Schur, agree with
P. vulgaris as regards their flower mechanism. Kerner mentions that old flowers of
P. transsilvanica may pollinate themselves by maturation of the inner stamens.

21. P. alpina Delarbre (= Anemone alpina Z.). (Herm. Miller, ‘ Alpen-
blumen,’ pp. 127-8; Ricca, ‘ Oss. sulla fecondaz. incroc. d. veget. alp. e subalp.’;
Kerner, ‘Nat. Hist. Pl... Eng. Ed. 1, II; Schulz, ‘ Beitrage,’ II, pp. 4-7.) —This
species, with its sulphur-yellow variety (Anemone sulphurea Z.), bears protogynous
pollen flowers. The variety predominates in the Riesengebirge and Central Alps
of the Tyrol, while the ordinary form characterizes the Eastern Limestone Alps.
According to Kerner, this plant—like P. vernalis—bears hermaphrodite flowers of
two kinds, respectively possessing few short stamens, and numerous long ones. As
before, the former form is cross-fertilized, while the latter is autogamous. Hermann
Miiller found some androdioecious flowers, in addition to hermaphrodite ones, and
Schulz found the still rarer andromonoecious flowers. In the Riesengebirge only
3-5 % are male flowers, in the Alps 80-95 %. They are smaller than the herma-
phrodite ones.

Visirors. — Hermann Miiller observed in the Alps 6 bees, 12 flies, and
2 beetles. Frey in Graubiinden observed one species of Lepidoptera, Lypusa

maurella S.V. Dalla Torre and Schletterer in the Tyrol observed Bombus alticola
Kriechb. 9 and 6, tolerably freq.

6. Anemone Tourn.

Monogamous, slightly protogynous, or slightly protandrous pollen flowers.
Petals wanting. Insects attracted by the sepals, which are usually white or yellow,
more rarely violet, red, or blue. Individual species (A. Richardsoni Mooker—in
Greenland) are also probably wind-pollinated (Warming).

22. A. sylvestris L. (Herm. Miller, ‘Weit. Beob. I, p. 314; Schulz,
‘ Beitrage,’ II, p. 7; Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, II.)—In the milk-white
homogamous, or slightly protogynous, or slightly protandrous flowers, the inner
stamens bend together above the stigmas, so that self-pollination must occur auto-
matically. The flowers when expanded in the sunshine attain a diameter of 70 mm.,

and consequently attract many insects, which, as they creep about, may effect either
self- or cross-pollination.

Visrtors.—Hermann Miiller observed the following in his garden at Lippstadt.—
A. Coleoptera. (a) Ceraméycidae: 1. Grammoptera ruficornis /., devouring the
anthers, (4) Dermestidae: 2. Byturus fumatus /., po-dvg. (¢) Malacodermata :

RANUNCULACEAE Ir

3- Dasytes flavipes ¥.; 4. Malachius bipustulatus ¥., devouring the anthers. (¢)
Mordellidae: 5. Anaspis rufilabris Gyi/., po-dvg. (e) Scarabaeidae: 6. Phyllopertha
horticola Z., gnawing the flower. B. Diptera. (a) Bibionidae: 47. Bibio hor-
tulanus Z., getting nothing. (6) Empidae: 8. Rhamphomyia sp.: 9. Tachydromia
connexa Mg. (c) Muscidae: 10. Anthomyia sp., po-dvg.; 11. Calliphora vomitoria
£.; 12. Chlorops hypostigma Mg. (d) Syrphidae: 13. Ascia podagrica F.; 14.
Eristalis arbustorum Z., freq.; 15. E. nemorum Z., freq.; 16. E. tenax Z., freq. :
17. Helophilus floreus Z.; 18. Pipiza funebris J/g.; 19. Rhingia rostrata Z.;
20. Syritta pipiens Z., freq.; all hover-flies, eagerly po-dvg. C. Hymenoptera.
Apidae: 21. Apis mellifica Z., po-cltg., freq., also skg.
Schulz also observed bees, flies, and—more rarely—beetles.

The flower mechanism agrees in the main with that of the next species.

23. A. nemorosa L. (Sprengel, ‘Entd. Geh.,’ p. 292; Hua, ‘ Anemone
nemorosa Z. var. anandra,’ Bul. soc. bot., Paris, xxxvi, 1889; Herm. Miller,
‘Fertilisation, p. 72, ‘Weit. Beob.,’ I, p. 314; Kirchner, ‘Flora v. Stuttgart,’
p- 260: Webster, ‘Change of colour in... Anemone nemorosa,’ J. Bot., London,
xxv, 1887; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894; Knuth, ‘ Bloe-
menbiol. Bijdragen.’)—- The flowers are considerably smaller than those of the
last species, and insect-visits are consequently distinctly fewer in number.

The flowers are generally white, marked externally with reddish lines, seldom
entirely red, and very rarely blue. The stamens at first project beyond the stigmas,
so that the latter are protected from contact. Later on the stamens bend outwards,
and as both they and the stigmas are now liable to be touched by visitors, either
cross- or self-pollination may take place. Should insect-visits fail, the inclined
position of the flower enables pollen to fall upon the stigma, thus ensuring autogamy.
Hua observed flowers with reduced stamens.

In the pheasantry at Treskow, Warnstorf (Verh. bot. Ver., Berlin, xxxviii.
1896, p. 16) observed white flowers of two different sizes, 35 mm. and 20 mm.
in diameter, respectively. The perianth leaves of the former are tinged with pale
violet on their under-side, while those of the latter are yellowish green underneath.
The peduncle of the large-flowered form attains a length of over 30 mm., but that of
the small-flowered form is only about 25 mm. long. The flowers are all slightly
protogynous : their inner and outer stamens are shorter than the middle ones, which
are inclined over the gynoecium, so that self-pollination is easy. The anthers dehisce
very irregularly. Pollen white, ellipsoidal to rounded tetrahedral, very finely tuber-
culated, about 37 » long and 25 pw broad.

Visttors.—Hermann Miiller (H. M.) and myself (IKKn.) have observed pollen-
collecting bees, and pollen-devouring flies more particularly.— A. Coleoptera.
(2) Mordellidae: 1. Anaspis frontalis 7., po-dvg. (H.M.); 2. Mordellistena pumila
Gyll. (H.M.). (6) Mitidulidae: 3. Meligethes, po-dvg. (Kn., Kiel and Wiesbaden ;
H. M., numerous in the bases of the flowers). B. Diptera. (a) Muscidae:
4. Scatophaga merdaria Z., po-dvg. (Kn., Kiel and Wiesbaden; H.M.); 5. S. ster-
coraria Z., po-dvg. (as preceding). (4) Syrphidae: 6. Eristalis tenax Z., nect-skg.,
po-dvg. (as preceding). C. Hymenoptera. Apzdae: 7. Andrena albicans Afull. 6,
po-dvg. (Kn., Wiesbaden; H.M.); 8. A. fulvicrus KX. 9, po-dvg. (H. M.); 9. A.
parvula K. 9, po-dvg. (H.M.); 10. Apis mellifica Z. ¥ (Kn., Kiel and Wiesbaden ;
H.M.). H. M. observed that the honey-bee not only collects pollen, but also sucks,
boring with its proboscis into the base of the flower, so as to obtain the sap which
it requires for moistening the pollen. 11. Bombus terrester Z. 9, po-cltg. (H. M.);

12 ANGIOSPERMAE—DICOTYLEDONES

12. Halictus cylindricus #. 9, po-clg. (H. M.); 13. Osmia bicolor Schr. 9, po-cltg.
(H.M.). D. Thysanoptera. 14. Thrips (H. M.).

Alfken and Hoppner (H.) observed in Bremen:—Afzdae: 1. Andrena albicans
Mull. 9; 2. A. parvula K.(H.); 3. Apis mellifica Z.¥; 4. Bombus hortorum Z. ¥;
5- B. pratorum Z.¥. All po-cltg. :

MacLeod noticed in Flanders :— Apis, a sp. of Halictus, 4 Muscids, an Empid,
a Lepidopterid, and 2 beetles. y

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 2), hover-flies, an
Empid, and a Muscid were observed.

Burkill (‘ Fertlsn. of spring flowers, &c.’) observed on the Yorkshire coast.—
A. Diptera. (a) Afuscrdae: 1. Anthomyia sp., po-dvg.; 2. Scatophaga stercoraria,
po-dvg. (4) Syrphidae: 3. Melanostoma quadrimaculata Verall, po-dvg. B. He-
miptera. 4. Anthocoris sp. C. Hymenoptera. Aprdac: 5. Bombus terrester LZ.
D, Thysanoptera. 6. Thrips sp.

24. A. ranunculoides L. (Herm. Miller, ‘Weit. Beob.,’ I, pp. 314-15;
Beyer, ‘D. spont. Bewegungen d. Staubgefiisse u. Stempel’; Warnstorf, Verh. bot-
Ver., Berlin, xxxviii, 1896; Knuth, ‘Bloemenbiol. Bijdragen.’)—The mechanism
of the golden-yellow flowers agrees with that of A. nemorosa. In the Treskow
pheasantry Warnstorf observed two hermaphrodite forms, one bearing large flowers
(30 mm. in diameter), the other small flowers (average diameter 18-20 mm.).
On the other hand, in the park at Wustrau, he noted a form with very small,
short-stalked flowers. Some of these possessed numerous green flower-leaves:
the gynoecium was often quite vestigial, and the stamens reduced to small green
sepaloid structures.

Visitors. — Hermann Miiller (H. M.) and myself (Kn.) have observed the
following.—A. Diptera. Bombylitdae: 1. Bombylius’ discolor Mzkan., vainly
nect-skg. (H. M., Thuringia). B. Hymenoptera. Apzdae: 2. Apis mellifica LZ. ¥
(Kn., po-cltg.; H. M., po-clig., and skg., Thuringia).

Loew observed in the Berlin Botanic Garden:—Coleoptera. Dermestidae :
Anthrenus scrophulariae Z., po-dvg.

25. A. narcissiflora L. (Herm. Miller, ‘Alpenblumen,’ p. 128; Schulz,
‘ Beitrage, I, p. 3.)— The flowers are protandrous: some of the stigmas of those
examined by Schulz in the Riesengebirge were dark-brown, and functionless. The
plants observed by Miiller in the Alps were capable of automatic self-pollination.

Visirors.—Hermann Miiller observed in the Alps 6 po-dvg. flies.

26. A. baldensis L. (Kerner, ‘ Nat. Hist. Pl.” Eng. Ed. 1, II.) —The white
periodically opening pollen flowers are partly protogynous and hermaphrodite, partly
pseudo-hermaphrodite and staminate. The former appear in two forms, one pos-
sessing short stamens, and adapted for cross-pollination, the other with longer
stamens and capable of automatic self-pollination.

27. A. trifolia L. (Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II; Schulz, ‘ Bei-
trage,’ II, p. 7.) —The white periodically opening blossoms are homogamous pollen
flowers, in which automatic self-pollination is easy, since anthers and stigmas are
closely approximated.

Vistrors.—In the South Tyrol Schulz chiefly observed flies, also some bees
and beetles. These may effect cross- as well as self-pollination.

RANUNCULACEAE 13

28. A. apennina L. (Knuth, ‘Bliitenbiol. a. d. Ins. Capri.) — Slightly proto-
gynous pollen flowers. When the violet blossoms have opened in the sunshine their
diameter is 5 cm. They exhale a faint odour of cumarin, The numerous blue-black
stamens are arranged in several whorls round the similarly coloured styles. The
stigmas are mature shortly before the anthers dehisce, so that at this stage cross-
pollination may result from insect-visits. The dehiscing anthers are so near the
stigmas that automatic self-pollination must necessarily take place, and this seems to
be the regular thing in Capri, where insect-visits are very rare, despite the very
conspicuous character of the flowers.

VisiTors.—I only once saw a small fly (Muscidae) po-dvg.

29. A. japonica Sieb. et Zucc. (Knuth, ‘Bloemenbiol. Bijdragen.’) — This
species—a native of Japan—is one of our ornamental garden plants. The diameter
of the homogamous flowers is about 7cm. The stamens at first lie close to the
sepals, rendering self-pollination difficult, and favouring cross-pollination. Subse-
quently the stamens become erect, so that automatic self-pollination may result from
contact of anthers and stigmas.

Visitors.—I observed the following po-cltg. and po-dvg. insects.—A. Diptera.
(a) Muscidae: 1. Musca domestica Z.; 2. Sarcophaga carnaria ZL. (4) Syrphrdae:
3. Eristalis tenax Z.; 4. Syrphus ribesii Z. B. Hymenoptera. Afzdac: 5. Bombus
terrester 1.9. Of these, Eristalis and Bombus regularly alighted upon the middle
of the flower, proceeding thence to the anthers, so that they almost always effected
cross-pollination. The other visitors sometimes alighted in the middle of the flower,
sometimes on the anthers, and consequently effected either cross- or self-pollination,
as the case might be.

Loew observed in the Berlin Botanic Garden.— A. Diptera. Syrphidae:
1. Eristalis tenax Z., po-dvg.; 2. Syritta pipiens Z., po-dvg.; 3. Syrphus balteatus
Deg., po-dvg. on the stamens; 4. S. ribesii Z., as the last. B. Lepidoptera.
Rhopalocera; 5. Pieris brassicae £., repeatedly inserting the proboscis between the
ovaries and apparently boring for sap with its tip. C. Orthoptera. 6. Forficula
auricularia Z.

Loew further observed—in the same place—the following upon the variety
purpurea.—A. Diptera. Syrphidae: 1. Syrphus balteatus Deg., po-dvg. on the
stamens; 2. S. corollae /., po-dvg. on the anthers. B, Hymenoptera. Vespzdae :
3. Vespa germanica /,, settling and vainly trying to suck.

7. Adonis Dill.

Protogynous pollen flowers, opening and closing periodically. Their vivid red
or yellow petals serve for attraction.

go. A. vernalis L. (Beyer, ‘D. spont. Bewegungen d. Staubgefasse u.
Stempel’; Kerner, ‘ Nat. Hist. Pl., Eng. Ed. 1, Il; Herm. Miller, ‘ Weit. Beob.,’ I,
p- 315; Knuth, ‘Bloemenbiol. Bijdragen.’)—- The yellow flowers—according to
Miiller’s account—expand in the sunshine to form a disk 40-70 mm. in diameter,
visible from a distance, and turning towards the sun. When the flower opens, the
numerous stigmas are already mature, while the still more numerous stamens are not
yet fully developed and are directed outwards, so that in this condition cross-pollina-
tion may result from insect-visits. The stamens gradually erect themselves centri-
petally, and at the same time the anther-lobes on the sides of the broad connective
dehisce laterally, When all the stamens are ripe they project a little above the

14 ANGIOSPERMAE—DICOTYLEDONES

stigmas, so that either cross- or self-pollination can be effected with equal ease by
insect-visits. At this later stage self-pollination is brought about automatically in
dull weather, for the flower then closes and the stigmas come into contact with the
pollen. Automatic self-pollination may also occur in sunshine, for as the flower
turns towards the sun, pollen readily falls upon the stigmas.

Visitors.—According to the observations of Hermann Miiller (H. M.), in
Thuringia, and myself (Kn.), in the garden at Kiel, the visitors are chiefly po-cltg.
bees, and po-dvg. flies and beetles A. Coleoptera. (a) Coccinelldae: 1. Micraspis
duodecimpunctata Z. (H.M.), 4 in a flower, one licking the stigmas. (4) Nitsdu-
lidae: 2. Meligethes (Kn., H. M.), in very large numbers, po-dvg. B. Diptera.
(a) Muscidae: 3. Scatophaga merdaria Z. (H.M.), po-dvg. (4) Syrphidae: 4. Enis-
talis sp. (Kn.); 5. E. tenax Z. (Kn.), both po-dvg. C. Hemiptera. 6. Lygaeus
equestris Z. (H. M.), very numerous, boring with their proboscis in the base of the
flower. D. Hymenoptera. (a) Apidae: Andrena nitida Fourcr.d, po-cltg. (H. M.);
8. A. parvula A. 9 (H.M.); 9. Apis mellifica Z. ¥, freq. (Kn., H.M.); 10. Bombus
terrester Z. 9, settling, but neither skg. nor po-cltg. (H. M.); 11. Halictus albipes
F.9(H.M.); 12. H. cylindricus F. 9, freq. (H. M.); 13. H. morio /. 9 (H. M.);
(3) Formicidae: 14. Formica congerens /Vy/. ¥ (H.M.), very freq., busying itself with
anthers (po-dvg.?) and stigmas (skg. their moisture?). E. Thysanoptera. 15.
Thrips (H. M.), not infrequent. In some flowers a spider was found lurking
for prey.

oe observed the following in the Berlin Botanic Garden.—A, Coleoptera.
Nitidulidae: 1. Meligethes sp., po-dvg. B. Diptera. (a) Afuscrdae: 2. Anthomyia
sp., po-dvg. (4) Syrphzdae: 3. Melithreptus scriptus Z., po-dvg. C. Hymenoptera.
Apidae: 4. Andrena nitida Fourc. 9, po-cltg.

gi. A. aestivalis L. (Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, IL; Knuth,
‘ Bloemenbiol. Bijdragen.’) — The red or (in the variety A. citrinus Hoffman, some-
times considered a distinct species) yellow petals spread out in the sunshine, attracting
pollen-collecting bees and pollen-devouring flies. The flower mechanism agrees with
that of the last species, but owing to the smaller size of the flower there are fewer
insect visitors.

Visitors. —I observed in the Garden at Kiel—A. Diptera. Syrphidae:
1. Eristalis tenax Z., po-dvg. B. Hymenoptera. Afzdae: 2. Apis mellifica ZL. ¥,
po-cltg.

32. A. autumnalis L. (Knuth, ‘Bloemenbiol. Bijdragen.’) — This agrees
with the last species as regards its flower mechanism. According to Warnstorf
{Abh. bot. Ver., Berlin, xxxviii, 1896), the flowers are homogamous and autogamous.
When the pollen is ripe, the stamens lie close to the purple stigmas. The pollen is
cinnamon-brown, and either irregularly elliptical with three longitudinal folds, or
tetrahedral with spheroidal faces. The grains in the former case are 43 » long
and 25 » broad, in the latter case 31 w in diameter.

Visitors.—I observed only Apis mellifica Z. ¥, po-cltg.

8. Myosurus Dill.

Flowers homogamous or protandrous with exposed nectar, secreted at the bases
of the small, inconspicuous, greenish-yellow petals.

33. M. minimus L. (Sprengel, ‘ Entd. Geh.,’ p. 443; -Delpino, ‘ Altri appa-
recchi dicog. recent. oss.,’ p. 57; Herm. Miiller, ‘ Weit. Beob.,’ I, pp. 316-18;

RANUNCULACEAE 15

MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1896, pp. 173-4; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ p. 17; Kirchner, ‘Flora v. Stuttgart,’ p. 262.) — When
the flower opens, the slender terminal lobes of the petals are directed outwards, and
each secretes in a shallow pit a drop of nectar that is at once visible. The stamens,
which are closely pressed against the carpellary cone, dehisce laterally by longitudinal
slits, so that their outer surfaces become completely covered with pollen. The small
visitors—minute flies and midges—get dusted with pollen below while licking the
nectar, and creeping about upon the central cone, transfer pollen from the same or
another flower to the stigmas. In young blossoms, where the carpels make up
a rounded projection, or at most a short cone, the visitors—according to Miiller—
usually alight upon its apex, thus effecting cross-pollination. As, however, owing to
the inconspicuousness of the flower, insect-visits are very few, automatic self-pollina-
tion takes place to a very great extent. For as the flower gets older, its axis—closely
beset with carpels—increases greatly in length, so that fresh stigmas are continually
being brought to the closely apposed anthers, and are pollinated in regular succession.

According to Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896), in some
just-opened flowers, the cone formed by the gynoecium is lower than the stamens
which cover it and the stigmas are already mature when the pollen is ripe, hence
these flowers are homogamous and autogamous. In other cases, the carpellary cone
is elongated and projects from the flower before it opens, and as the stigmas are not
mature at this stage, while some of the anthers of the deeply-placed stamens have
already dehisced, such flowers are decidedly protandrous. Self-pollination of the
central stigmas is in this case precluded, though it may take place in the lower
stigmas by further elongation of the axis bearing the carpels.

Visitors.— On the island of Féhr I noticed tiny Muscidae, but did not
determine their species. Hermann Miiller observed the following at Lippstadt—
Diptera. (a) Brbionidae: 1. Scatopse brevicornis Mg. (6) Cectdomyidae: 2. Ceci-
domyia sp. (c) Chironomidae: 3. Chironomus byssinus Schr. and other sp.
(d) Emprdae: 4. Microphorus sp. (¢) Ausczdae: 5. Anthomyia sp., several cases.
6. Hydrellia chrysostoma Mg. 4. H. griseola Fallen. 8. Oscinis sp. (/) Myceto-
philidae: 9. Sciara sp., 2 sp. in 7 cases. (g) Phoridae: 10. Phora sp. (A)
Syrphidae: 11. Melanostoma mellina Z., a single case.

g. Batrachium S. F. Gray

Flowers homogamous or slightly protogynous or protandrous, with half-con-
cealed nectar. The white petals which serve for attraction are usually ornamented
at the base with a yellow nectar-guide. Honey is secreted in a pit (open in several
northern species, according to Almqvist) at the base of each of them. The stem creeps
in mud or floats in water, so that the flowers are accessible only to flying insects and
not to creeping ones.

34. B. hederaceum S. F. Gray (=Ranunculus hederaceus Z.). (Knuth, ‘ Bl.
u. Insekt. a. d. nordfr. Ins.,? pp. 17-18, 147.)—The diameter of the homogamous
flowers is only 4-5 mm. Nectar is sparingly secreted. The 8-10 stamens are in a
single whorl, and since they are at the same level as the stigmas, and mature simul-
taneously, self-pollination must take place should insect-visits fail. This happens
very frequently, for, owing to the inconspicuousness of the flower, such visits are

16 ANGIOSPERMAE—DICOTYLEDONES

few. On the island of Féhr I observed small flies (Muscidae) visiting the flowers,
and of course these might effect self- or cross-pollination with equal facility.

Willis and Burkill (‘Fls. and Insects in Gt. Britain. I, p. 267) also found
flowers in central Wales of only 5 mm. in diameter. They did not observe any
secretion of nectar, and saw no insect visitors. The anthers dehisce when the fower
opens, and get covered all over with pollen, which they also shed on the stigmas.
The stamens move outwards after dehiscence of the anthers. Self-pollination is
completely effective. When flowering is over, the peduncle bends downwards during
the ripening of the fruit.

35- B. aquatile Wimm. (=Ranunculus aquatilis Z., in part). (Axell, ‘Om
anord. fér fanerog. Vaxt. befrukt.,’ p. 14; Hildebrand, ‘Die Geschlechtsvert. b. d.
PA,’ p.17; Herm. Miiller, ‘ Fertilisation,’ p. 74 ; ‘ Weit. Beob.,’ I, pp. 318-19; Beyer,
‘D. spont. Bewegungen d. Staubgefasse u. Stempel’; Crié, C. R. Acad. sci., Paris, ci,
1886, p. 1025; Kirchner, ‘ Flora v. Stuttgart,’ pp. 263-4; Knuth, ‘ Bl. u. Insekt. a. d.
nordfr. Ins.,’ pp. 18, 147.)—The faintly odorous flowers are homogamous or (on the
island of Féhr) slightly protogynous. They spread out their petals in the sunshine,
forming a yellow-centred white star 20-25 mm. in diameter, but the size varies (and
with it the number of stamens) to such an extent that, according to Kirchner, it may
be as low as 3-4 mm. (and the number of stamens may be reduced from over 20
to 8). Many insects may, however, be attracted, for the plants are usually associated in
large numbers, and not infrequently shallow ditches and ponds are completely filled
with them, so that the surface is covered with white blossoms. After the flower has
opened, the anthers at once dehisce, covering themselves with pollen. The stigmas
usually mature either simultaneously with the stamens or slightly earlier. In the
latter case, cross-pollination must result from the visits of insects, if these come
from a flower with dehisced anthers. When homogamy obtains, insects may effect
either cross- or self-pollination, for they sometimes alight on the centre, sometimes
on the margin of the flower, and then creep about. If insect-visits fail, automatic
self-pollination takes place, for pollen can readily fall from the anthers on the
adjacent stigmas.

When the flowers are submerged in consequence of flooding, they remain closed
and pollinate themselves (Axell, Hildebrand).

According to Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896), the flowers
are homogamous and autogamous in the neighbourhood of Ruppin. The pollen is
yellow, irregularly ellipsoidal, and tuberculated. The grains are about 25 pw
broad and 37 pu long.

Visirors.—The following have been observed by Hermann Miiller (H. M.)
and myself (Kn.).—A. Coleoptera. (a) Byrrhidae: 1. Pedilophorus aeneus F,,
with its head in the nectaries (H. M.). (4) Chrysomelidae: 2, Agelastica alni Z.,
resting idly on the flowers (H.M.); 3. Helodes phellandrii Z., devouring anthers
and flower-leaves (H.M.). (c) Elateridae: 4. Limonius cylindricus Payé., its head
and thorax yellow with pollen. B. Diptera. (@) Brbionrtdae: 5. Dilophus vulgaris
ig. and 9, freq. (H.M.). (3) Empidae: 6. Empis rustica Fall. (H. M.); 7. Hilara
maura FF. (H.M.). (c) Afusczdae: 8. Anthomyia sp., skg. and po-cltg. (Kn., Fohr ;
H.M.); 9. Cyrtoneura hortorum /a//. $ (H. M.); 10. Hydrellia griseola Fad/., skg.
and po-cltg., extremely abundant (H. M.); 11. Hylemyia sp. (H. M.); 12. Onesia
floralis R.-D. 8, freq. (H.M.); 13. O. sepulcralis A/ezg., freq. (H. M.); 14. Sarce-

RANUNCULACEAE 17

phaga carnaria Z., nect-lkg. (Kn., Kiel; H.M.); 15. Scatophaga merdaria F., po-dvg.
(H.M.); 16. S. stercoraria F., po-dvg. (Kn., Kiel); 17. Sepsis cynipsea Z. (Kn.,
Foéhr); 18. Thryptocera sp. (H.M.); 19. Small Muscidae (H.M.); (¢) Syrphidae:
20. Chrysogaster viduata Z., skg. and po-dvg. (H.M.); 21. Eristalis arbustorum Z.
(Kn.); 22. E. nemorum Z. (Kn.); 23. E. tenax Z. (Kn., all three po-dvg., freq.,
Kiel; H.M., po-dvg. or skg., feet abundantly smeared with pollen, so as to effect
cross-pollination when visiting a fresh flower); 24. Helophilus floreus Z. (H.M.),
and 25. H. pendulus Z., both po-dvg. (Kn., Kiel); 26. Melanostoma mellina Z.,
po-dvg. (H. M.); 27. Syrphus sp. po-dvg. (Kn., Kiel). C. Hymenoptera.
Apidae: 28. Apis mellifica Z. ¥, skg. and po-cltg., freq. (Kn., Kiel and Fohr;
H.M.); 29. Bombus terrester Z. 9, skg. (H.M.); 30. Halictus minutissimus KX. 9,
po-cltg., isolated cases (H. M.); 31. H. sexstrigatus Schenck 9, ditto (H. M.).
D. Neuroptera. 32. Psocus sp., nect-lkg. (Kn., Fohr).

MacLeod, in Flanders, noticed Apis, Megachile, and Eristalis (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 181).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire’) Muscidae have been
observed.

36. B. divaricatum Wimmer (=Ranunculus circinatus $7d¢h.), according to
Kirchner, has the same flower mechanism as the last species.

37. B. paucistamineum Sonder (=B. trichophyllum Cazx, in part). (Knuth,
‘Weit. Beob.,’ pp. 227-8.)— This plant is very common on the North Frisian
islands of Nordstrand and Pellworm. The numerous flowers remain open even in
rainy weather, and so abundant and closely massed are the plants in these islands
that some ditches look as if they were covered with a white table-cloth. The diameter
of the flower is 12-13 mm. Each petal is about 6 mm, long, and about 3 mm.
broad towards its tip; the base is marked with a yellow nectar-guide, and is so narrow
that there is a considerable space between the individual petals. The plant is slightly
protogynous : in just-opened flowers the stigmas are partly mature, while the anthers
of the few stamens (usually only 8-12) are still closed. Their ripening goes on
centripetally, and the filaments of the 4-6 outer ones elongate, bending towards the
petals, so that insects alighting on the middle of the flower effect cross-pollination.
The filaments of the 4-6 inner stamens elongate later on, but remain with their
dehisced anthers above the stigmas—now very markedly papillose—upon which
pollen necessarily falls, thus effecting self-pollination automatically. This must
be effective, for all the carpels constantly mature, though I have not observed any
insect-visits, despite long watching during favourable weather.

38. B. (Ranunculus) paucistamineum Tausch (not Sonder) was found
by Schulz—in central Germany—to be either homogamous or slightly protandrous,
with very great variation in the size of the flowers and the number of stamens. He
also observed gynomonoecism. According to Warnstorf (Schr. natw. Ver.,
Wernigerode, xi, 1896), the flowers at Neuruppin are protogynous, and _ their
diameter is 10-17 mm. ‘They possess aS many as 15 stamens, which are shorter
than the head of carpels. Pollen: golden yellow, roughly tuberculate, varied in
form, ellipsoidal or bluntly conical, with three longitudinal furrows, 30-43 » long
and 25-30 » broad. According to Freyn, the submerged flowers are sterile.

39. B. fluitans Wimmer (=Ranunculus fluitans Zmé.).—According to Freyn,
this is usually sterile because the flowers are submerged. The fruits were set,

DAVIS. II Cc

18 ANGIOSPERMAE—DICOTYLEDONES

however, in herbarium specimens, which were obtained from eastern Schleswig-
Holstein, and from the island of Rém.

40. B. carinatum Schur.—Freyn states that the very long peduncles—
which are at right angles to the surface—are not submerged by the rising of
the water.

41. B. Baudotii d. B—

VisiTors.—Verhoeff observed one of the Muscidae in Norderney (Anthomyia sp.,
skg., one instance).

1o. Ranunculus Z.

Flowers homogamous or—more rarely—slightly protogynous or protandrous,
with nectar half concealed. Insects are attracted by the petals, which are usually
yellow, but white or red in a few species. At the base of each petal there is
a nectar-pit, that is either upwardly prolonged into a membranous scale (in the
species with white or red flowers), or (in most yellow-flowered species) covered
by a fleshy upwardly directed scale, or else it is open (in R. sceleratus, and—
according to Almqvist—in a few northern species (R. pygmaeus Wg., R. hyperboreus
Roitb., R. nivalis Z.)). Many species have flowers that repeatedly open and close.
On dehiscence of the anthers the stamens incline towards the petals, so that the
pollen falls upon these, but not so readily upon the stigmas. Only the larger nectar-
sucking, pollen-collecting, or pollen-devouring insects come regularly into contact
with the stigmas, effecting self- or cross-pollination with equal ease. The larger
the flowers the less likely is automatic self-pollination, for with increased size the
distance between anthers and stigmas of course becomes greater, and there is less
chance of pollen falling upon the stigmas when the blossoms are bent by the
wind or any other agency. But as increase of size in the flowers furthers the visits
of insects—with resultant pollination—this disadvantage is counterbalanced.

The flowers are sometimes gyno-monoecious, while according to Schulz, gyno-
dioecism occurs in R. acris, auricomus, hybridus, and repens, and the same condition
has also been observed in England by Whitelegge in the case of R. bulbosus.

42, R. glacialis L. (Ricca, ‘Oss. sulla fecondaz. incroc. d. veget. alp.
e. subalp.’; Lindman, ‘ Bidrag till Kanned. om Skand. fijellvaxt. blom. o. befrukt.’ ;
Herm. Miiller, ‘Alpenblumen,’ pp. 128-9; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, I;
Andersson and Hesselman, ‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil,
Karlvaxtflora.’)—The flowers are homogamous or—according to Ricca—slightly
protandrous. In the Scandinavian highlands they are markedly protandrous. Some
may be female only. In old flowers automatic self-pollination can easily be effected
by the fall of pollen from the anthers of the inner stamens upon the stigmas. In
the Alps the size of the flower varies greatly (12-30 mm. in diameter). The structure
of the nectaries is equally varied (see Fig. 5). Besides hermaphrodite blossoms,
Kerner observed pseudo-hermaphrodite pollen flowers. He also found two forms
of hermaphrodite flower, corresponding to those of Anemone alpina. Andersson
and Hesselman state (op. cit., p. 42) that this species was frequently found in flower
in Spitzbergen during July.

Visitors.—Miiller observed—in the Alps—2 flies and 2 micro-Lepidoptera.

RANUNCULACEAE 19

43- R. lapponicus L.—This species flowers in Spitzbergen from the middle of
July to the middle of August, and though ripe fruits have not so far been seen,
they were observed to be set in 1897 and 1898. Of the pollen-grains 96 % were
normally developed (Andersson and Hesselman, op. cit., pp. 43-7). Ekstam (‘ Bliiten-
biol. Beob. a. Spitzbergen,’ p. 227) describes the flowers in the same locality as
being yellow, fragrant and protogynous-homogamous, with a diameter of 8-10 or
sometimes even 13 mm. Andersson and Hesselman state that the pouch-like
nectar-pits are relatively smaller than those of R. Pallasii.

For Nova Zemlia Ekstam gives the diameter of the protogynous-homogamous
flowers as 16 mm. (according to Kjellman, it is 12 mm. in Arctic Siberia). Self-
pollination is excluded, for the stigmas project beyond the anthers.

44. R. Pallasii Schlecht—This species flowers in Spitzbergen during the end
of July and the beginning of August, but perfectly ripe fruits have not so far been
observed. Andersson and Hesselman (op. cit., p. 42) state that 96-98% of the
pollen-grains are normally developed. According to Ekstam (‘Bliitenbiol. Beob.

Fic. 5. Ranunculus glactalis, L. (after Herm. Miller). 4. Petal of a flower with particularly small
perianth (x 7). B—Z. Bases of other petals with variously formed nectaries (x 7).

a. Spitzbergen,’ pp. 21-2), the white or greenish-white flowers are about 15 mm.
in diameter. Nathorst says that their odour is extremely pleasant, resembling that
of Platanthera. Andersson and Hesselman state that nectar is secreted in a
nectar-pit situated in the upper part of the small claw. Ekstam found the flowers
to be protogynous-homogamous.

Visrtors.—Ekstam (op. cit.) noticed one medium-sized and a number of small
‘industrious’ Diptera.

45. R. lapponicus x R. Pallasii—This hybrid was noticed beneath the parent
species in Spitzbergen by Nathorst in 1882, and subsequently in 1898 by Andersson
and Hesselman (op. cit, pp. 42-7). Its characters are an approximately equal
blend of those of the component species. The flowers are yellowish-white, fragrant,
and about 10-13 mm. in diameter. Only 8-12 % of the pollen-grains develop
normally. The flowers are protogynous-homogamous ; nectar is sparingly secreted
by the pouch-shaped nectar-pits. The plant was seen in flower from the end
of July on into August, but fruits have so far been observed only in the earlier
stages of development. The occurrence of this interesting hybrid shows that the
fruits of the mother-plant can actually ripen.

Cc 2

20 ANGIOSPERMAE—DICOTYLEDONES

46. R. sulphureus Sol.—According to Ekstam, the diameter of the flowers is
16 mm. ; in Arctic Siberia (Kjellman) and on Nova Zemlia it is distinctly greater.
In Spitzbergen this observer (op. cit., p. 23) found the flowers to be protogynous-
homogamous, and faintly fragrant. In expanded blossoms the gynoecium projects
so far beyond the stamens that autogamy is prevented. Andersson and Hesselman
(op. cit., p. 49) state that this species flowers in Spitzbergen from the end of June
until August, the fruits being ripe at the end of the latter month—or much earlier
near the deep fjords.

VisiTors.—Ekstam observed seven distinct cases in Spitzbergen where the flowers
were visited by numerous Diptera, and he also noticed these insects elsewhere.

47. R. pyrenaeus L. (Ricca, ‘Oss. sulla fecondaz. incroc. d. veget. alp.
e subalp.,’ p. 3; Herm. Miiller, ‘Alpenblumen,’ pp. 132-3 ; MacLeod, ‘ Pyreneenbl.,’
p. 114.)—In this species the nectaries are also very variable, as the accompanying figure
shows (Fig.6). According to Ricca, cross-pollination is at first favoured by feebly

Fic. 6. Ranunculus pyrenaeus, L. (after Herm. Miller). 1-12. Various forms of nectary.
A, nectar. 13. Transition from petal to stamen.

marked protogyny. Subsequently—according to Miiller—cross- and self-pollination

are equally possible as the result of insect-visits. During this stage, automatic self-

pollination by means of the inner stamens can easily take place.
Vistrors.—Hermann Miller observed—in the Alps—z2 beetles, 9 flies, one

Ichneumon, and one micro-Lepidopterid. MacLeod observed—in the Pyrenees—
2 flies.

48. R. alpestris L. (including R. Traunfellneri Hoppe). (Herm. Miiller,
‘Alpenblumen,’ pp. 130-1; Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, IL.) — According
to Miiller, cross-pollination is at first favoured in the homogamous or slightly proto-
gynous flowers, but afterwards self-pollination is possible. The hermaphrodite
flowers—according to Kerner—occur in two forms, as in Anemone alpina. Kerner
also observed pseudo-hermaphrodite pollen flowers.

Visirors. — Hermann Miiller observed 1g flies, one beetle, one humble-bee,
and 2 Lepidoptera.

RANUNCULACEAE 2i

49. R. aconitifolius L. (Herm. Miller, ‘Alpenblumen,’ p. 131; Schulz,
‘Beitrage.’)— According to Schulz, the flowers are markedly protandrous in the
Riesengebirge. Most of the plants in this district bear blossoms of very varied size,
and therefore have a characteristic appearance.

Vistrors.—Miiller observed the following in the Alps.—7 beetles, 18 flies, 6
Hymenoptera, and 4 Lepidoptera.

50. R. Seguieri Vill—At San Martino—according to Schulz—some stocks
bear purely male flowers.

51. R. parnassifolius L. (Herm. Miiller, ‘Alpenblumen,’ p. 132.) — The
flowers are protogynous with stigmas that persist for a long time, so that cross-
pollination is at first secured. Self-pollination by the inner stamens becomes possible

Fic. 7. Ranunculus parnasstfolius, L. (after Herm. Miller). 4. Flower in first (female) condition,
seen from above (x 7). All the stigmas are mature, while all the anthers are still closed. 2. Flower with
five sepals, and two petals, seen from the side (natural size). C. The same with one petal, oblique view
from below (natural size).

later on. Usually only one petal is developed, though sometimes there are two
or three (cf. Fig. 7).

Visrtors.—In the Alps these are chiefly flies (Muscidae and Syrphidae).

52. R. amplexicaulis L.—The flowers are white.

Visttors.—MacLeod observed—in the Pyrenees—one bee, one hover-fly, and
2 Muscidae.

53. R. Gouani Willd.—

Visttors.—MacLeod observed—in the Pyrenees—3 species of bee, one hover-fly,
and 5 species of Muscidae.

54. R. hyperboreus Rottb.—Lindman states that the flowers are slightly
protandrous in the Scandinavian highlands, and the numerous stigmas project so
far above the anthers that automatic self-pollination is impossible. In the Arctic
regions, throughout which the species is widely distributed, the flowers are remarkably
small and—according to Warming—they are autogamous.

22 ANGIOSPERMAE—DICOTYLEDONES

Vanhoffen (Abromeit, ‘Bot. Ergeb. v. Drygalski’s Grénlandsexped.,’ p. 30)
observed this species in partially flooded places in Greenland. The leaves projected
beyond the small flowers.

55. R. paucistamineus Tausch, var. eradicata Laestad. (=R. confervoides
Fries, in Lange’s ‘ Consp. flor. groenl.’).—Vanhéffen (Abromeit, op. cit., pp. 29-30)
found this species with mature fruits in ponds in Greenland.

56. R. pygmaeus Wg.—Ekstam states that the flowers are homogamous in
the Scandinavian highlands. Their diameter there is 7 or 4 mm., and on Nova
Zemlia 5-1o mm. The stigmas, especially in the small flowers, stand at the same
level as the anthers, so that automatic self-pollination must take place, and this
is effective in Scandinavia and Nova Zemlia; also in other parts of the Arctic
regions, as Warming has shown.

Andersson and Hesselman (op. cit., p. 48) say that the species flowers in
Spitzbergen from the end of July to the middle of August, and is in blossom on
Beeren Island by the middle of June. The fruits mature early, many of them
being ripe at the beginning of August, or later. Little pollen is produced, but this
develops normally. Self-pollination is brought about by contact of the dehisced
anthers with the stigmas of the lower carpels. According to Ekstam (‘Bliitenbiol.
Beob. a. Spitzbergen,’ pp. 22-3), the flowers are 5-8 mm. broad and non-fragrant.
They are feebly protandrous, all the anthers having as a rule dehisced before the
stigmas have completely matured.

57. R. Flammula L. (Herm. Miiller, ‘Fertilisation, pp. 74-6, ‘ Weit. Beob.,’
I, p. 319; Verhoeff, ‘ Bl. u. Insekt. a. d. Ins. Norderney,’ p. 127; Kirchner, ‘ Flora
v. Stuttgart,’ p. 265; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 1753
Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 18, 147.)—The bright yellow blossoms
are protandrous. Immediately after the flower has opened, the anthers of the
outermost stamens dehisce, so that their outer sides are covered with pollen, some
of which must adhere to insects seeking the nectar secreted by the bases of the
petals, At this stage the incompletely mature stigmas are completely, or almost com-
pletely, covered by the inner stamens, being thus guarded from contact with insect
visitors. Centripetal dehiscence of the anthers slowly takes place, each stamen
in turn bending outwards, so as to direct its pollen-covered side towards the petals.
The stigmas are mature before the anthers of the innermost stamens have dehisced.
Should insects covered with pollen alight on the middle of the flower, crossing must
result, while if they settle upon a petal, afterwards creeping over the stamens to the
stigmas, there is an equal chance of self-pollination, which at this stage may also
take place automatically. Herm. Miiller states that the two modes of alighting are
about equally common in this and the next three species.

Visttors.—In consequence of its relatively small flowers, R. Flammula attracts
but few insects. The following have been observed by Hermann Miiller (H. M.),
Verhoeff (V.) in Norderney, and by myself (Kn.) in Schleswig-Holstein.—A. Coleo-
ptera. Séaphylinidae: 1. Anthobium minutum /., freq. (H. M., Teutoburger
Wald). B. Diptera. (a) Muscidae: 2. Anthomyia sp. (Kn., Fohr; H. M., V.);
3. Scatophaga merdaria Z., po-dvg. (H. M.); 4. S. stercoraria #:, po-dvg. (Kn.,
Kiel). (6) Syrphdae: 5. Cheilosia sp., po-dvg. (H. M.); 6. Eristalis tenax Z.,
po-dvg. (Kn., Kiel); 7. Melithreptus taeniatus J/gn., po-dvg. and skg. (H. M.);

RANUNCULACEAE 23

8. Syritta pipiens Z., po-dvg. and skg. (Kn., Kiel, H.M.). C. Hymenoptera. Apzdae :
9. Apis mellifica Z. ¥, skg. and po-cltg. (Kn., Kiel); 10. Halictus cylindricus F. 9,
po-cltg. (H. M.); 11. H. flavipes /. 9, po-cltg. (H.M.). D. Lepidoptera: 12.
Coenonympha pamphilus Z., skg. (H. M.).

Alfken and Hoppner observed—at Bremen—the small bee Dufourea vulgaris
Schenck 9 and 4, skg., freq.

H. de Vries (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875), observed
in the Netherlands a bee, Trachusa serratulae Pz. MacLeod—in Flanders —saw
a short-tongued bee, 3 hover-flies, and a Muscid (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, pp. 175-6).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 4) Muscidae and
various hover-flies have been observed.

Fic. 8. Ranunculus Flammula, L. (after Herm. Miller). 1. Flower just opening: the anthers of
some of the marginal stamens have dehisced. 2. Flower in the first (male) condition : all the anthers
have dehisced; the stigmas are still immature. 3. Flower in the second (hermaphrodite) condition: the
stigmas are fully mature, some of the anthers are still covered with pollen. a@, immature stamens;
6, anthers about to dehisce; c, stamens with dehisced anthers; d, stamens with empty anthers;
¢, carpel.

58-60. R. acris L., R. repens L., and R. bulbosus L.—Herm. Miller
(‘ Fertilisation,’ pp. 76-7) states that these agree with R. Flammula in the structure
of their flowers, but receive more insect-visits owing to their greater conspicuousness.
The visitors are chiefly pollen-loving hover-flies (Syrphidae), which are attracted by
the bright blossoms and easily find the half-concealed nectar. Other frequent guests
are small bees, especially species of Halictus, which readily collect the abundant
pollen with their tarsal brushes, and with their moderately short proboscides easily
reach the nectar, for this—though hidden—is not difficult of access. These insects
and flowers—as Herm. Miiller puts it—are at correspondingly low stages of develop-

24 ANGIOSPERMAE—DICOTYLEDONES

ment, and exhibit complete mutual adaptation as regards both size and general
structure.

Lindman states that on the Dovrefjeld the flowers of R. acris sometimes possess
an agreeable, slightly sweet odour: the diameter of the flowers is 15-25 mm.: the
visitors are numerous flies, and also Lepidoptera. According to Ekstam, flowers of
the variety borealis Zraufv.—in Nova Zemlia—have a diameter of 30 mm., and are
slightly protogynous, protogynous-homogamous, protandrous-homogamous, or homo-
gamous. Flies were observed as visitors. In central Germany, Schulz also observed
gynomonoecism.

R. repens—according to Lindman—is also homogamous on the Trontfjeld.
Schulz states that this species also is gynodioecious in central Germany.

The same is true—according to Whitelegge—for R. bulbosus in England.

The species R. repens, R. acris, and R. Flammula? exhibit in the East Frisian
islands (Verhoeff, ‘Bl. u. Insekt. a. d. Ins. Norderney,’ pp. 108-14) three very
important and interesting degrees of adaptation to insects. The actual kind and
frequency of insect-visits harmonize most beautifully with these adaptational stages.
Verhoeff says that R. repens has an advantage over R. acris in the following
characters :—(1) the nectary and nectar-scale are better developed; (2) the stamens
diverge outwards more readily, usually projecting beyond the calyx—which seldom
happens in R. acris; (3) the petals are broader and more lustrous; and (4) the
flowers are more crowded.

Verhoeff makes the following comparison between R. acris and R. repens on
the one hand, and R. Flammula on the other, with regard to the times at which
the sexes develop.—

R, acris and R. repens. R. Flammula.

The stigmas are mature before any of The stigmas are undeveloped when
the anthers have dehisced, and are at the the first marginal anthers dehisce, but
same time exposed to contact with the mature while the rest of the outer anthers
under-sides of insects. are ripening.

In flowers where all or some of the
innermost anthers have dehisced, the
carpels are already swollen and their
stigmas shrivelled.

The stigmas may either be a little The stigmas project above the adjacent
higher or a little lower than the adjacent anthers from the first.
anthers.

Protogyny. Approach to protogyny.

In Westphalia, Hermann Miiller has not noticed any differences between
R. acris, R. repens, and R. bulbosus. On the contrary, he expressly remarks

1 Verhoeff says: ‘H. Miiller has made a mistake with regard to the species of Ranunculus,
wrongly asserting that R. acris, repens, and Flammula agree as regards structure and conspicuous-
ness.’ This, however, is not quite correct, for Hermann Miiller expressly states that R. Flammula
is much more rarely visited by insects than R. acris, R. repens, and R. bulbosus — ‘One reason

at least being that its flowers are much smaller, and less conspicuous.’ (‘ Fertilisation,’ p. 76.)

RANUNCULACEAE 25

that they agree ‘with each other in their habitat, in the conspicuousness of their
flowers, and therefore also in the insects which visit them ....I have even noticed
the hive-bee, which in general keeps strictly to one species of flower, passing from
Ranunculus acris to R. repens and R. bulbosus, or vice versa, without any distinction.’
{‘ Fertilisation,’ p. 76.)

Visrrors.—On the North Frisian islands I have noticed the following for these
three species.— Small bees, 7 hover-flies, 2 Muscidae, one butterfly, one beetle.
Also the following in Schleswig-Holstein (S-H.), Helgoland (H.), Thuringia (Th.),
and Rigen (R.).—A. Coleoptera. Mitidulidae: 1. Meligethes sp. (S-H.). B.
Diptera. (a) Muscidae: 2. Anthomyia sp. (S-H., Th.) ; 3. Aricia basalis Ze/t. (Th.) ;
4. A. incana Wied. (S-H.); 5. Coelopa frigida Fadi. (H.); 6. Fucellia fucorum Fall.
(H.); 7. Homalomyia scalaris F. § (H.); 8. Small Muscids, freq. (H., S-H.).
(3) Syrphidae: 9. Chrysogaster macquarti Loew (S-H.); 10. Eristalis arbustorum
L. 3, skg. (S-H., R.); 11. E. tenax Z. (S-H.); 12. Helophilus pendulus Z. (S-H.);
13. Melanostoma mellina Z. (S-H.); 14. Syritta pipiens Z. (S-H.); 15. Syrphus
cumulatus J/gn. (Th.); 16. S. sp. (S-H.). C. Hymenoptera. (a) Apzdae:
17. Short-tongued bees (S-H.). (6) Vespidae: 18. Vespa salonica F. 9 (Th.).
D. Lepidoptera. Rhopalocera: 19. Lycaena semiargus Roff (S-H.); 20. Leuco-
phasia sinapis Z. (S-H.).

MacLeod observed in Flanders :—Apis, 12 short-tongued Hymenoptera, 12

hover-flies, 11 other flies, 2 Lepidoptera, and 2 beetles (Bot. Jaarb. Dodonaea, Ghent,
vi. 1894, pp. 176-7).

By far the largest number of visitors has been observed by Hermann Miller
(H. M.) in Westphalia, and his friend Dr. Buddeberg (Budd.) in Nassau. The
following list gives the results of their observations, along with a few made by
Borgstette (Borg.) in Tecklenburg.—

A. Coleoptera. (a2) Supresizdae: 1. Anthaxia nitidula Z. (H.M., Budd.).
(6) Cerambycidae: 2. Strangalia nigra Z. (H.M.), gnawing the flowers. (c) Chryso-
melidae: 3. Cryptocephalus sericeus Z. (H. M.), as the last ; 4. Galleruca nymphaeae
ZL. (H.M.); 5. Prasocuris glabra /ést.(H. M.), gnawing the flowers. (d) Czstelidae:
6. Cistela murina Z. (H.M.), as the last. (¢) Coccenellidae: 7. Micraspis duodecim
punctata Z., vainly skg. (H.M.). (/) Curculionidae: 8. Bruchus sp., nect-lkg.
{H.M.). (g) Dermestidae: 9. Byturus fumatus /., po-dvg., freq. (H.M.). (2) Zlate-
yidae: 10. Limonius cylindricus Payk., nect-lkg. (H.M.). (2) Alalacodermata:
11. Malachius aeneus Z. (H. M.), and 12. M. bipustulatus ™, both devouring anthers
(H.M.); 13. Trichodes apiarius Z., po-dvg. (H.M.). (4) Mordellidae: 14. Mordella
aculeata Z. (H.M.); 15. M. pusilla Dey. (H.M.). (2) Metédulidae : 16, Meligethes
brassicae Scop., po-dvg. (H.M.); 17. M. sp., very freq., skg. and po-dvg. (H. M.).
{m) Oedemertdae: 18. Oedemera virescens L., freq. (mn) Staphylintdae: 19. Tachy-
porus solutus Zr.; 20. Anthobium minutum /”, very freq., Teutob. Wald.

B. Diptera. (a) Asz/zdae: 21. Dioctria atricapilla Mg. (Borg.). (4) Empidae:
22. Empis stercorea Z., skg. (H.M.); 23. E. tessellata #/, skg. (H.M.); 24. Rham-
phomyia umbripennis Jg., skg. (H. M.). (c) Musczdae: 25. Anthomyia sp. (H.M.);
26. Calobata cothurnata Pz. (H.M.); 27. Cyrtoneura caerulescens MMcq., skg. (H.M.).
{d@) Stratiomyidae: 28. Odontomyia tigrina F., skg. (H.M.). (e) Syrphidae: 29.
Cheilosia albitarsis AZg., freq., skg. and po-dvg. (H. M.); 30. C. pubera Zett,,
po-dvg., in large numbers (H.M.); 31. C. schmidtii Ze/#., skg. and po-dvg. (H. M.);
32. C. vidua Afy., skg. and po-dvg. (H.M., Budd.); 33. Chrysochlamys ruficornis F”,
po-dvg. (H. M.); 34. Chrysogaster macquarti Loew (H.M.); 35. C. viduata Z.,
very freq., skg. and po-dvg. (H. M.); 36. Chrysotoxum arcuatum Z., skg. and
po-dvg. (H.M.); 37. C. festivum Z., skg. (H.M.); 38. Eristalis arbustorum Z.
{H.M.); 39. E. nemorum Z. (H.M.); 40. E. sepulcralis Z. (H. M.), and 41. E.

26 ANGIOSPERMAE—DICOTYLEDONES

tenax Z., all four freq., skg. and po-dvg. (H. M.); 42. Melanostoma mellina Z., skg-
(H. M.); 43- Melithreptus pictus 1/g. (H. M.); 44. M. scriptus Z. (H. M.), and 45.
M. taeniatus J/g. (H. M.), all three freq., skg. and po-dvg.; 46. Pipiza chalybeata
Mg., po-dvg. (H.M.); 47. P. funebris AZg., skg.(H.M.); 48. Platycheirus albimanus
¥’., po-dvg. (Borg.); 49. Syritta pipiens Z., freq., skg. and po-dvg. (H. M.); 50-
Gee) pyrastri Z., po-dvg. (H. M.); 51. S. ribesii Z., freq., skg. and po-dvg-

C. Hymenoptera. (a) Apidae : 52. Andrena albicans AZull. 9 and 4, freq., skg-
and po-dvg. (H. M., Budd.); 53. A. albicrus X. 4, freq., skg. and po-dvg. (H. M.) ;
54. A. flavipes Pz. 9 and 4, freq., skg. and po-dvg. (H.M.); 55. A. gwynana XK. 6,
skg. and po-cltg. (Budd.); 56. A. trimmerana XK. 3, skg. (H.M.); 57. Apis mellifica
L. 9, skg. (H. M., Budd.); 58. Bombus agrorum F., skg. while on the wing (H. M.);
59. Eriades florisomnis Z. 6 and 9, skg. (H. M., Budd.) ; 60. E. nigricornis Vy/. 6, skg.
(Budd.); 61. Halictus albidulus Schenck 9, skg. and po-cltg.(H.M_); 62. H.albipes F. 9,
skg. and po-cltg.(Budd.); 63. H.cylindricus /. 9, po-cltg.and skg.(H. M., Budd.); 64. H.
flavipes /. 9, po-cltg.(H.M.); 65. H. leucopus &. 9, skg. (Budd.); 66. H. leucozonius
Schr. 9, skg. and po-clig. (H. M., Budd.); 67. H. longulus Sm. 9, skg. (H.M.); 68.
H. lugubris A. 9 and 4, skg. and po-cltg. (H.M.); 69. H. maculatus Sm. 9, skg. and
po-cltg. (H. M., Budd.) ; 70. H. morio F. 9, skg. (Budd.); 71. H. nitidiusculus X. 9,
skg. and po-cltg. (H. M., Budd.) ; 72. H. rubicundus Cér. 9, skg. and po-cltg. (H. M.,
Budd.); 73. H. sexnotatus K. 9, skg. and po-cltg. (H. M., Budd.); 74. H.sexsignatus
Schenck 9, skg.(H. M.); 75. H. smeathmanellus &. 9, skg. and po-cltg. (Budd.); 76.
H. tetrazonius A7g. 9, skg. and po-cltg. (H. M.); 77. H. villosulus X. 9, skg. and po-
cltg. (H. M., Budd.); 78. H. zonulus Sm. , skg. (H.M.); 79. Osmia aenea Z. @, skg.
(Budd.); 80. O. rufa Z. 9, po-cltg. (H. M., Budd.); 81. Panurgus calcaratus Scop.,
skg. (H.M.); 82. Prosopis brevicornis /Vy/. 5, skg. (Budd.); 83. P. clypearis Schenck
3, skg. (Budd.); 84. P. hyalinata Sm. 3, skg. and po-dvg. (H. M.). (4) Formicidae:
85. Lasius niger Z. ¥, nect-lkg. (H.M.). (c) Sphegidae: 86. Oxybelus uniglumis L.
(H.M.). (@) Zenthredinidae: 87. Amasis crassicornis Ross’ (H.M.); 88. Cephus
pallipes A7g., nect-lkg. (Budd.); 89. C. pygmaeus Z., by the hundred, skg. and
devouring the anthers (H.M.); go. C. small undetermined sp. (H.M.). (¢) Vespedae:
gi. Odynerus spinipes Z. 9 (H. M.).

D. Lepidoptera. (a) Rhopalocera: 92. Coenonympha pamphilus Z. (H. M.);
93- Lycaena icarus Rof/. (H. M., Budd.); 94. Pararge achine Scop. (Budd.); 95.
Polyommatus dorilis H/n. (H.M.); and 96. P. phlaeas Z.; all skg.(H.M.). (6)

Noctuidae: 97. Euclidia glyphica Z. (c) Zineidae: 98. Micropteryx calthella Z., very
frequent in R. repens, skg. (Dr. Speyer).

E. Thysanoptera. 99. Thrips, frequent (H. M.).

Hermann Miller (‘Alpenblumen,’ p. 135) observed in the Alps R. acris Z.
visited by 2 beetles, 2 hover-flies, 2 Tenthredinidae, and 11 Lepidoptera: R. repens
L. visited by 4 beetles, 5 flies, 5 Hymenoptera, and 6 Lepidoptera: R. bulbosus Z,
visited by Hymenoptera.

Verhoeff observed the following on Ranunculus acris in Norderney.—

A. Coleoptera. (a) Nitidulidae: 1. Meligethes brassicae Scop., skg.; 2. M-
coracinus Sf. B. Diptera. (a) Dolichopidae: 3. Dolichopus aeneus D.G., one 8.
(4) Empidae: 4. Hilara quadrivittata Mg., skg. (c) Muscrdae: 5. Anthomyia pratensis
iMg., po-dvg. and skg.; 6. A. sp., po-dvg. and skg.; 7. Aricia incana Wred., po-dvg. ;
8. Lucilia caesar Z., skg.; 9. Onesia floralis R.-D., 9 and 3, skg. (d) Syrphidae :
10. Eristalis sepulcralis Z., one 6; 11. Melanostoma mellina Z., one 9; 12. Pipizella
virens /.; 13. Platycheirus albimanus 7. 6, po-dvg.; 14. P. manicatus Mg. 9,
po-dvg. and skg.

RANUNCULACEAE 27

Von Dalla Torre observed in the Tyrol the bees Andrena rosae Pz.6; A.
tibialis K. $; A. bicolor /. (=A. Gwynana &.) 3; A. fulvicrus A. 9; Halictus
albipes /. 9; H. smeathmanellus K. 9 and 6; Osmia caerulescens Z. 6; Chelostoma
maxillosum Z. 4, very freq.

Loew observed the following on Ranunculus acris in the Berlin Botanic Garden.—
Apidae: Andrena nitida Fourc. 9, skg.and po-cltg.; in Brandenburg,—Pipiza quadri-
maculata Pz., skg.; in Silesia, —Syrphidae: 1. Syrphus tuniger M/g.; 2. S. cumulatus
Mg.; 3. Melithreptus scriptus Z., skg., and also Meligethes nect-lkg. He also
noticed the following in Switzerland (‘ Beitrige,’ p. 57).—

A. Coleoptera. Buprestidae: 1. Anthaxia quadripunctata Z. B. Diptera.
Muscidae: 2. Hydrotaea ciliata #.; 3. Tetanocera elata #7. C. Hymenoptera.
Apidae: 4. Panurgus banksianus X. 9, po-cltg.; Ricca (Atti Soc. ital. sc., Milano,
xiii. 1870) mentions this.

Schletterer enumerates the following bees for the Tyrol:—x. Andrena austriaca
Pz.; 2.A. flavipes Pz.; 3.A.gwynana X.; 4. A. tibialis. K.; 5. Eriades florisomnis Z.;
6. Halictus albipes #.; 7. H. smeathmanellus A.; Osmia caerulescens ZL.

MacLeod observed in the Pyrenees 4 short-tongued Hymenoptera, one of the
Lepidoptera, 2 beetles, 3 Syrphidae, and 6 Muscidae (‘ Pyreneenbl., p. 387).

Alfken—in Bremen—noticed the following on Ranunculus repens and R.
acris.—

A. Diptera. (a) Bibionidae: 1. Bibio marci Z.; 2. Dilophus vulgaris Z.

(6) Syrphidae: 3. Ascia podagrica Z.; 4. Eristalis arbustorum Z.; 5. E. sepulcralis
L.; 6. Melanostoma mellina Z. ; 7. Rhingia rostrata Z.; 8. Syritta pipiens L.

B. Hymenoptera. (a) Apidae: 9. Andrena albicans A/dll. 9; 10. A. nigro-
aenea K.9; 11. Bombus terrester Z.9; 12. Eriades florisomnis ZL. 9 and 6: 13.
Osmia rufa L. 9; () Zenthredinidae: 14. Cephus nigrinus Zs. (not on R. acris).

H. De Vries (‘Ned. Kruidk. Arch.,’ Nijmegen, 2. ser., 2. deel, 1875) observed—
in the Netherlands—z bees: Andrena trimmerana X. 9, and Eriades florisomnis Z. é.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 5), Muscidae, Empidae,
Syrphidae, Tenthredinidae, and Meligethes have been observed on R. acris and
R. repens.

Loew observed in Brandenberg (‘Beitrage,’ p. 38) on the flowers of R. repens,—
Eriades florisomnis Z. 9, po-cltg.; as did Schenck in Nassau, and Schletterer and
von Dalle Torre in the Tyrol.

Verhoeff observed the following on R. repens in Norderney.—

A. Coleoptera. (a) Witidulidae : 1. Brachypterus gravidus J//., skg.; 2. Meli-
gethes aeneus ZL. B. Diptera. (a) Zmpzdac: 3. Hilara quadrivittata Mg., skg.
(6) Muscidae: 4. Anthomyia sp.; 5. Aricia incana Wedem. 9 and 6, po-dvg. and skg. ;
6. Calliphora erythrocephala Mg. one 9; 7. Cyrtoneura hortorum Fall.; 8. Lucilia
caesar L.g and 6, skg.; 9. Onesia floralis R.-D.9 and 8. (¢) Syrphidae: to. Chryso-
gaster metallina /.; 11. Eristalis arbustorum L., three 5; 12, E. sepulcralis Z., one 4,
skg.; 13. Melanostoma mellina Z.; 14. Platycheirus manicatus Mg., po-dvg. and skg.
C. Hymenoptera. Formicidae: 15. Formica fusca Z. (= Rasse fusca Forel), one
9. D. Lepidoptera. Pvreridae: 16. Pieris brassicae Z., one 9, skg.

In Switzerland Loew observed (‘ Beitrage,’ p. 57).—

Diptera. Syrphidae: 1. Cheilosia antiqua JZg., po-dvg.; 2. Merodon cinereus
F., po-dvg.

28 ANGIOSPERMAE—DICOTYLEDONES

H. de Vries (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875)—in the
Netherlands—noted the following —Hymenoptera. 1. Apis mellifica L. 5 2.
Eriades florisomnis Z. 8; 3. Halictus leucozonius K. 9; 4. Panurgus banksianus
Lair. 8; a wood-wasp (Siricidae); Cephus pygmaeus ZL.

As visitors of R. bulbosus,—Cetonia hirtella Z., was observed by Loew in
Brandenburg (‘ Beitrage,’ p. 38); the bee Andrena humilis Zmt., by Schmiedeknecht
in Thuringia; the bee Andrena cingulata #., by Schenck, in Nassau; and the
bee Eriades florisomnis Z. 3, by H. de Vries, in the Netherlands (Ned. Kruidk.
Arch., Nijmegen, 2. ser., 2. deel, 1875).

MacLeod observed in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891,
p- 387-8) :—3 short-tongued Hymenoptera, 2 Lepidoptera, one of the Syrphidae,
and 3 Muscidae.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 5) there have been
observed—Apis, one humble-bee, one saw-fly, and 2 Muscidae.

61. R. Lingua L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins., pp. 18-20;
‘Bliitenbiol. Beob. a. d. Ins. Rtigen’) has been studied by myself on Fohr and al
Kiel. The stem is branched above and one metre high. It bears a number ol
large golden-yellow flowers, with a diameter of about 4 cm., so that the plant is
conspicuous at a distance. Each petal bears at its base a large richly secreting
nectary. The flowers are protogynous. After the numerous stigmas have maturec
centripetally, the anthers dehisce extrorsely in similar order. The stamens benc
outwards towards the expanded petals as they ripen, so that automatic self-pollinatior
is rendered very difficult. But owing to the oblique position of the flower it car
take place, though it appears to be ineffective, for frequently but few fruits are formed
and not rarely none at all. Cross-pollination is effected by the agency of flies
which nearly always alight on the middle of the flower, so that if already coverec
with pollen they dust the stigmas on their way to the anthers and nectaries.

Visitors. —I observed the following in Schleswig-Holstein.—Diptera. (a)
Muscidae: 1. Aricia incana Wred., nect-lkg. and po-dvg. (Féhr) ; 2. Small Muscidae,
as before; 3. Calliphora erythrocephala J/g. (Kiel); 4. Lucilia caesar Z., as before ;
5. Sarcophaga carnaria Z., as before (Féhr). (4) Syrphidae: 6. Eristalis arbustorum
L.; 7. E. tenax Z.; 8. Rhingia rostrata Z.; 9. Syrphus balteatus Deg.; 10. S. ribesii
L.; 11. Syritta pipiens Z., all skg. and po-dvg.

Besides these I observed in the island of Riigen.—A. Diptera. Odontomyidae:
1. Chrysomyia formosa Scop., skg. B. Lepidoptera. Hesperidae: 2. Hesperia

lineola O., skg.

62. R. hybridus Biria.—Schulz states that this species is homogamous, or
slightly protandrous, in the Tyrol. The flowers vary greatly in size and in the
number of the stamens. Schulz also observed gynomonoecism.

63. R. auricomus L. (Sprengel, ‘Entd. Geh.,’ p. 294; Herm. Miiller, ‘ Fer-
tilisation, pp. 78-9; Knuth, ‘Bloemenbiol. Bijdragen’; MacLeod, Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 179.)—The flowers agree in structure with those of
R. Flammula, but the corolla is rarely quite regular, almost always some or all
the petals are reduced, and they may even be entirely absent. The broad yellow-
margined sepals therefore take over the duty of attracting insects. The nectaries
are very variable in shape, as a rule they are pits devoid of covering-scales. Hermann
Miiller has represented various forms of nectary in the following diagram (Fig. 9).

RANUNCULACEAE 29

The flowers observed by Lindman on the Dovrefjeld were at first protogynous,
afterwards becoming homogamous. Their diameter was 5-22 mm. In that region
flies, and a Lepidopterid, were observed as visitors.

Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) observed in individual flowers
at Ruppin—though only very rarely—that certain stamens bore a papillated
stigma at the tip. Pollen-grains yellow, tuberculated, very irregular, usually rounded
tetrahedral, varying in size up to 43 » in diameter.

According to Focke (Abh. nat. Ver., Bremen, xii, 1891), Ranunculus auricomus
is self-fertile.

VisiTors.—In north and central Germany the following have been observed
by Herm. Miiller (H.M.) and myself (Kn.).— A. Coleoptera. 1. Meligethes, po-dvg.
(Kn.). B. Diptera. (2) Muscidae: 2. Anthomyia radicum Z. 9 and 4, particularly
freq. (H.M.); 3. Lucilia caesar Z., po-dvg. (Kn., Kiel.) ; 4. Scatophaga merdaria /.,
skg. and po-dvg. (H.M.). (6) Syrphzdae: 5. Cheilosia vernalis /a//., po-dvg. (H.M.);
6. Eristalis tenax Z., po-dvg. (Kn., Kiel); 7. Melanostoma mellina Z., po-dvg.(Kn.,H.M.);
8. Pipizella virens #., po-dvg. (H.M.). C. Hymenoptera. (ca) Apidae: 9. Andrena
fulvescens Sm. 3, skg. (H. M.); 10. A. parvula KX. 9, po-dvg. (Kn., H. M.); 11. Apis
mellifica Z. Y, po-clg. (Kn., Kiel); 12. Halictus albipes /. 9, po-clg. (H. M.); 13. H.
cylindricus F. 9, ditto (Kn.). (4) Formicidae: 14. An ant, nect-lkg. (Sprengel,
H.M.). D. Lepidoptera. TZineidae: 15. Micropteryx calthella Z., skg. (H. M.).
E. Thysanoptera. 16. Thrips (Sprengel, H. M.).

Warnstorf observed numerous pollen-devouring beetles on the flowers at Ruppin.
In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 4) 2 Muscidae, and a

beetle have been observed as visitors.
ey
! i
H
IWS,
3

W

n-

Fic.9. Ranunculus auricomus, L. (after Herm. Miller). 1-8. Petals with nectaries of various

forms. 9. Petals of Eranthis hiemalis, for comparison.

64. R. amoenus Ledeb.—This arctic species (Siberia, Dahuria, Spitzbergen)—
which belongs to the auricomus group—flowers in Spitzbergen from the beginning
of July until August or September, and regularly sets fruits. The bright yellow
corolla has a diameter of 15-20 mm. The few stamens (6-8) are closely applied to
the lower part of the gynoecium, the stigmas of which are therefore easily self-
pollinated, while the upper ones are cross-pollinated by insects (Andersson and
Hesselman, ‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’
PP. 53-4):

65. R. arcticus Richards, var. Wilanderi (Nath.) Freyn.— This species flowers
in Spitzbergen during the middle of July and beginning of August, and sets fruits.
The corolla has a diameter of 17-5-21-5 mm.

30 ANGIOSPERMAE—DICOTYLEDONES

66. R. nivalis L—This species is circumpolar. Lindman states that the
flowers are first protogynous, and afterwards homogamous. The mechanism is
similar to that of R. acris, but the corolla is deeper and narrower, and the floral
receptacle longer and more convex. On the upper surface of each petal there are
longitudinal hollow swellings. The diameter of the flower—which is 18 mm. in
Arctic Siberia—is considerably less on Nova Zemlia according to Ekstam.

Andersson and Hesselman state (‘ Bidrag till Kanned. om Spetsbergens o. Beeren
Eil. Karlvaxtflora,’ p. 49) that it flowers and sets fruit normally in Spitzbergen. In
specimens collected by Vanhdffen in Greenland (Abromeit, ‘Bot. Ergeb. v. Dry-
galski’s Grénlandsexped.,’ pp. 30-1) there were a large number of shrivelled and
seedless fruits in addition to the properly developed ones. Ekstam (‘Bliitenbiol.
Beob. a. Spitzbergen,’ p. 23) describes the feebly fragrant flowers as 10-12 mm.
in diameter; and the light or dark yellow petals as about half as long again as
the sepals, which are thickly covered with dark-brown woolly hairs. He found the
flowers to be protogynous-homogamous.

Visrrors.—Ekstam noted one small fly in Nova Zemlia, and a large number
on the flowers of one plant in Spitzbergen.

67. R. lanuginosus L. (Herm. Miller, ‘Fertilisation, p. 78, ‘Weit. Beob.,
I, p. 321; Knuth, ‘Bloemenbiol. Bijdragen.’)—The tloral mechanism agrees with
that of R. acris, but—in spite of the larger flowers—insect-visits are more numerous
only in the lighter parts of the woods.

Visttors.—The following have been observed by Hermann Miiller (H. M.) and
myself (Kn.).—A, Coleoptera. (a) Coccznell:dae: 1.Coccinella quattuordecimpunctata
L., nect. skg. (H.M.). (6) Dermestidae: 2. Byturus fumatus L., freq. po-dvg. (H.M.)
(c) Elateridae: 3. Athous haemorrhoidalis /. (H.M.), with its head in the base
of the flower. (d) Misédulidac: 4. Meligethes aeneus /, freq., gnawing petals anc
stamens (Kn., H.M.). B. Diptera. (a) Bibionidae: 5. Dilophus vulgaris Z. (H. M.)
(8) Empidae: 6. Empis livida Z., skg. (H.M.); 7. E. trigramma J/g., skg. (¢'
Muscidae: 8. Anthomyia sp. (Kn., H.M.); 9. Hylemyia conica Wied. (H.M.);
10. Scatophaga stercoraria ZL. (d) Syrphidae: 11. Ascia lanceolata Afg., occasional
individuals (H.M.); 12. A. podagrica #., freq. (H.M., Kn.); 13. Bacha elongata /.,
occasional individuals (H.M.); 14. Cheilosia albitarsis 47g. (H.M.); 15. C. pubera
Zeit. (H.M.) and other sp. (H.M.); 16. Eristalis arbustorum Z. (Kn.); 17. Melano-
stoma mellina Z., in large numbers (H.M.); 18. Pipiza notata Afg.(H.M.); 19.
Syrphus lunulatus J/g. (Kn., H.M.); 20. S. nitidicollis 4Zg. (H.M.); 21. S. ribesii
L.(Kn.); 22. S. venustus AZg., in large numbers (H. M.); 23. Volucella pellucens Z.
(H.M.); all po-dvg. C. Hymenoptera. Apidae: 24. Andrena cingulata F. 9,
po-cltg. (H. M.); 25. A. parvula X. 9, skg. (Kn., H. M.); 26. Apis mellifica Z. ¥,
po-cltg. (Kn.); 27. Bombus terrester Z.9, skg. (H. M.); 28. Eriades florisomnis Z. 4,
skg. (H.M.); 29. Halicius flavipes /. 9, skg. (Kn., H.M.); 30. Osmia bicolor
Schr. 9, po-cltg. and skg. (H. M.).

Loew observed in the Berlin Botanic Garden—Hymenoptera. Apidae:
Eriades florisomnis,’é skg., 9 po-cltg.

68, R. montanus Willd. (including R. Villarsii DC.). (Herm. Miiller, ‘ Alpen-
blumen, pp. 133-5.)—The flowers are about equal in size to those of the last
species, and are protogynous, but automatic self-pollination is possible.

Visitors.—Loew observed in Switzerland (‘Beitrage,’ p. 57)—Diptera. (ca)

Asilidae: 1. Lasiopogon cinctus /. (6) Syrphidae: 2. Cheilosia antiqua JZz.,
po-dvg.

RANUNCULACEAE 31

Herm. Miiller noticed in the Alps:—3 beetles, 20 flies, 7 Hymenoptera, and
19 Lepidoptera.
69. R. illyricus L.—

Vistrors.—Schletterer observed the following Apidae—1. Halictus calceatus
Scop.; and 2. H. fasciatellus Schenck; also 3. a saw-fly, Amsais laeta F.

70. R. sardous Crantz (=R. philonotis Z%rA.). (Knuth, ‘Bl. u. Insekt. a. d.
nordfr. Ins.,” pp. 18-20.)—The flower mechanism and circle of visitors are as in
R. repens and its associated species. Warnstorf (Schr. natw. Ver., Wernigerode,
xi, 1896) describes a large hairy form at Ruppin, with hollow broadly trapezoidal
nectar-scales, but no secretion. Flowers slightly protogynous, anthers extrorse, ripen-
ing centripetally, and somewhat higher than the carpellary heads. Self-fertilization
is therefore rendered difficult, but not wholly excluded. Pollen-grains yellow,
spherical or ovoid, with three longitudinal furrows, tuberculated, 30-37 » in diameter.

Visitors.—H. de Vries (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875)
observed — in the Netherlands—the honey-bee; and MacLeod—in Flanders—noted
2 Syrphidae, 2 Muscidae, and one of the Siricidae (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 178).

71. R. arvensis L. (Hoffmann, ‘Uber Sexualitat’; Kirchner, ‘Flora v.
Stuttgart,’ p. 266.)—Kirchner states that there is much variation in the size of the
flowers, in the order of development, and in the number of stamens and carpels.
The diameter of the sulphur-yellow flowers is 4-10 mm. When they open, the
stamens are at first directed inwards above the carpels, often completely covering
them, but as the anthers begin to dehisce upwardly and outwardly, the styles
elongate. These are beset with stigmatic papillae at the end, and along an inwardly
directed longitudinal line. Owing to the relative positions of pollen and stigmas
automatic self-pollination is hardly possible. In some flowers, however, the anthers
have dehisced while the stigmas are still situated below them, so as readily to
catch the falling pollen. But when such blossoms reach a later stage this cannot
happen, the flowers being erect, and the stigmas projecting above the anthers. The
number of stamens is usually 10-13, but not infrequently some of them degenerate,
so that the flower only possesses 5-2. They may even be entirely aborted, rendering
the plant gynomonoecious. These female flowers are much smaller than the herma-
phrodite ones, and the styles at the time of opening project from the corolla.
Hoffmann also observed markedly protandrous hermaphrodite flowers. Focke states
that R. arvensis is self-fertile (Abh. natw. Ver., Bremen, xii, 1893).

Vistrors.—MacLeod observed in Flanders one of the micro-Lepidoptera and
2 small Muscidae as visitors (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 180).

72. R. sceleratus L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 20, 147.)—
The numerous flowers—of which the diameter is usually less than 1 cm.—make
the plant conspicuous from a distance, in spite of the smallness of the individual
blossoms. Consequently many short-tongued insects seek the nectar, which is
secreted in a pit at the base of each petal. When the flower opens, the stamens—
with anthers still unripe—lie close to the carpels, but are not so high as the already
receptive stigmas. As the anthers dehisce centripetally the filaments bend away
from the carpels, approaching the horizontally expanded petals. When insect
visitors alight upon the carpels they effect cross-pollination, if otherwise—self-

32 ANGIOSPERMAE—DICOTYLEDONES

pollination. Should insect-visits fail, the pollen of the fading anthers may reach
the stigmas—which continue receptive in the oldest flowers—when the plant bends
before the wind.

Visitors.—Only the following Muscidae were observed by me in the North

Frisian islands, and at Kiel—1x. Lucilia Caesar Z.; 2. Musca corvina F.; 3. various
smaller species.

Verhoeff noticed the following in Norderney.—Diptera. (a) Dolichopidae :
1. Dolichopus aeneus Deg. (6) Muscidae: 2. Anthomyia sp., one 4, skg.; 3. Aricia
dispar. Fad. one 8; 4. Aricia incana Weedem. 9 and é, po-dvg.; 5. Myospila medita-
bunda /. one 9; 6. Scatophaga stercoraria Z. (c) Syrphidae: 7. Eristalis intricarius
L. one 9; 8. Platycheirus peltatus A/Zg. one 9; 9. Pyrophaena ocymi F. one 4.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 4), flies (AZuscedae,
E'mpidae, Dolichopidae) have been observed.

MacLeod noticed in Flanders,—one hover-fly, and one of the Muscidae (Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 179).

73. R. Ficaria L. (Herm. Miller, ‘Fertilisation,’ p. 78, ‘Weit. Beob.,’ I,
pp. 321-2; Chatin, C-R. Acad. sci., Paris, cxviii, 1894; Kerner, ‘Nat. Hist. PI.,’
Eng. Ed. 1, IL; Schulz, ‘ Beitrige,’ II, p. 179; Knuth, ‘ Bloemenbiol. Bijdragen.’)—
Hermann Miiller states that the mechanism of pollination in these golden-yellow
flowers—which expand in the sunshine to form a star about 20-25 mm. in diameter—
agrees with those of R. acris and R. auricomus. Besides homogamous or slightly
protandrous hermaphrodite flowers, there are also female ones. At the beginning
of the flowering season, flowers are frequently to be found in which the number
of developed petals is as low as three, or even two, but later the number increases
to 8-10. The size of the flowers, and the number of stamens, commonly exhibit
very considerable variation.

The flowers seldom set fruits. Irmisch and Hunger saw them in shady wet
localities, while Kerner—on the contrary—observed here and there ripe heads ol
fruits in sunny places, while he found the plant sterile in shady spots, with bulbils
in the axils of the leaves, Warnstorf states (Verh. bot. Ver., Berlin, xxxviii, 1896)
that at Ruppin, where the plant grows abundantly in shady ditches under walls,
fruits (usually 2-3 in a head) are common, the individuals bearing them also pos-
sessing bulbils in their leaf-axils.

Burkill (J. Bot., London, xxxv, 1897) rightly says that the extreme rarity
of fruits in Ranunculus Ficaria is a puzzle. The failure to form seeds cannot
be ascribed to a deficiency of insect guests, for a variety of these visit the flowers
in large numbers.

Chatin asserts that the bulbil-producing form produces no seeds, because it has
no pollen. Miiller—on the contrary—cultivated a plant with bulbils in the leaf-
axils, and it produced ripe seeds capable of germinating. But this species propagates
vegetatively—in the large majority of cases—by means of the axillary bulbils. These
fall off in early summer, when all the parts of the plant above the ground die down.

Visttors.—The following have been observed by Hermann Miiller (H.M.)
and myself (Kn.).— A. Coleoptera. 1. Meligethes, freq., skg., po-dvg., and
gnawing the flower-leaves (H. M., Kn.). B. Diptera. (a) Musczdae: 2. Anthomyia
radicum Z., very freq. (H.M.); 3. A. sp., po-dvg. (Kn.); 4. Scatophaga merdaria F.
(H.M.); 5. Sepsis, freq. (H. M.). (2) Syrphidae: 6. Brachypalpus valgus Pz., po-dvg

RANUNCULACEAE 33

(H.M.); 7. Rhingia rostrata Z., po-dvg. (Kn.). C. Hymenoptera. Apidae: 8.
Andrena albicans Afi. 9 and 6, po-cltg. and skg. (H.M.); 9. A. gwynana K. 9,
po-cltg. (H. M.); 10. A. parvula A. 9, skg. (H.M.); 11. Apis mellifica Z. ¥, freq.,
skg. and po-dvg. (H. M., Kn.); 12. Halictus albipes F. 9, skg. (H. M.); 13. H.
cylindricus /. 9, ditto (H.M.); 14. H. lucidus Schenck 9, ditto (H. M.); 15. H. niti-
diusculus A. 9, ditto (H.M.); 16. H. sp., skg. (Kn.); 17. Osmia rufa Z. 3, skg.
(Thuringia). D.Thysanoptera. 18. Thrips, very freq. (H. M.).

Sickmann mentions Salius sepicola Sm. (a fossorial wasp) as a visitor of the
plant at Osnabriick.

Alfken and Héppner (H.) observed the following at Bremen.—A. Hymeno-
ptera. Apidae: 1. Andrena albicans Mill. 9 and 3, skg.; 2. A. cineraria L. 3,
skg.; 3. A. clarkella K.8; 4. A. extricata Sm. 9; 5. A. flavipes Pz. 9 andé; 6.A.
gwynana K. 9 andé; 7. A. morawitzi Zhs. 9; 8. A. nitida Fourcr.d; 9. A. parvula
K. 9 and 6; 10. A. varians K. 9 and $; 11. Apis mellifica Z.¥; 12. Bombus
agrorum J”. 9; 13. B. terrester Z. 9; 14. Halictus minutus X. 9; 15. H. morio F. 9;
16. H. nitidiusculus KX. 9; 17. Nomada alternata K.d; 18. N. bifida Zhs.d; 19. N.
borealis Zest. 9 (H.); 20. N. fucata Pz. 8; 21. N. lineola Pz.d; 22. N. ruficornis
Z.8; 23. N. xanthosticta K. 9 and @, skg.; 24. Osmia rufa Z. 9; 25. Podalirius
acervorum Z. $6. B. Diptera. Syrphidae: 26. Brachypalpus valgus Pz.

MacLeod noted in Flanders, — 2 long-tongued bees, 2 short-tongued bees,
a saw-fly (Tenthredinidae), 3 Muscidae, and a beetle (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, pp. 181-2).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,” p. 41) there were
observed,—Apis, Syrphidae, Empidae, and Muscidae.

Burkill (‘ Fertilisation of Spring Flowers’) observed the following on the coast
of Yorkshire—A. Coleoptera. (2) Chrysomelidae: 1. Longitarsus fuscicollis Foudr.
(4) Colydizdae: 2. Coninomus nodifer Westw. (c) Nitidulidae: 3. Meligethes picipes
Sturm., freq., skg. B. Diptera. (a) Muscidae: 4. Onesia cognata Mg., one instance,
skg.; 5. Lucilia cornicina /., skg. and po-dvg.; 6. Pollenia rudis /., one instance;
7. Scatophaga stercoraria Z., freq., skg. and po-dvg.; 8. Sepsis nigripes JZg.,
occasional, skg. and po-dvg. (4) Empidae: 9. Empis sp., one instance, skg. (c)
Phoridae: 10. Phora sp. (d) Syrphidae: 11. Cheilosia nebulosa Verral; 12. Eristalis
arbustorum Z., one instance, skg.; 13. E. pertinax Scop., freq., skg. and po-dvg.;
14. Melanostoma quadrimaculatum Verral, 9 and 4, skg.; 15. Syrphus lasiophthal-
mus Ze/., skg. C. Hymenoptera. (a) Apidae: 16. Andrena clarkella X., skg.;
17. A. gwynana X., 9 and 4, occasional, skg.; 18. A. nigro-aenea K.9; 19. Apis
mellifica Z. ¥, one instance, skg. and po-cltg.; 20. Bombus agrorum F,, one instance.
(5) Formicidae: 21. Formica fusca LZ. (c) Jchneumonidae: 22. Ichneumon sp., skg.
D. Lepidoptera. Rhopalocera: 23. Vanessa urticae Z.,skg. E. Thysanoptera.
24. Thrips sp.

u. Coptis Salisb.

74. Coptis trifolia Salisb.—This species is native to Greenland, Norway,
Siberia, Kamchatka, and Japan, also ranging from Alaska to Labrador. There
are five white sepals veined with purple, and 6 mm. long. The five or more hood-
shaped orange-yellow petals are very much smaller, being only 2 mm. in length.
They secrete nectar, and the stamens project beyond them. Warming states that
this northern plant is homogamous in Greenland. He could not determine the
presence of nectar. The flower belongs either to C or Po.

12. Caltha L.

Flowers homogamous, with half-concealed nectar. The large yellow sepals
serve to attract insects: there are no petals. Nectar is secreted in two shallow
depressions, one on either side of each ovary (Fig. 10).

DAVIS. II D

34 ANGIOSPERMAE—DICOTYLEDONES

75. C. palustris L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ p. 147;
Sprengel, ‘Entd. Geh.,’ p. 298; Herm. Miiller, ‘ Fertilisation,’ pp. 79-80, ‘ Weit.
Beob.,’ p. 322, ‘ Alpenblumen,’ pp. 135-6; Kirchner, ‘Flora v. Stuttgart,’ p. 270;
Beyer, ‘D. spont. Bewegungen d. Staubgefiasse u. Stempel’; Schulz, ‘ Beitrage,’
II, p. 179; Haussknecht, Mitt. geogr. Ges., Jena, vi, 1887.)—In the large egg-yolk
yellow flowers—which spread out in the sunshine to a surface 4 cm. across—nectar
is so abundant that the drops from adjacent nectaries (Fig. 10) run together.
Although stamens and carpels mature simultaneously, cross-pollination is favoured, for
the anthers dehisce extrorsely and mature centripetally. In addition to the ordinary
plants with homogamous hermaphrodite blossoms, stocks with purely male flowers
have been observed in France and the Tyrol. In the Scandinavian highlands the
flowers are sometimes only 2 cm. in diameter, and Lindman states that in that region
they possess a slight odour, suggestive of gutta-percha. According to Ekstam, the
diameter of the flowers is 10-36 mm. in Nova Zemlia. Haussknecht observes that
large-flowered forms predominate in Thuringia, and small-flowered ones in South
Germany. Lecogq (‘Géographie botanique de |’Europe,’ IV, p. 488) speaks of this
species as being andromonoecious (Darwin, ‘ Different Forms of Flowers,’ p. 13).

VisiTors.—The following have been observed by
Hermann Miiller (H. M.) in Westphalia, and myself (Kn.)
in Schleswig-Holstein.—

A. Coleoptera. (a) Chrysomelidae: 1. Donacia
discolor Hoppe (H. M.); 2. Helodes marginella Z.(H. M.).
(6) Curculionidae: 3. Bruchus seminarius Z., nect-lkg. (?),
(H.M.). (c) Mitidulidae: 4. Epuraea aestiva Z. (H.M.);
5. Meligethes, very freq., skg. and po-dvg. (H.M.). (d)
Staphylinidae: 6. Tachyporus hypnorum F., nect-lkg. (?),
(H.M.). B. Diptera. (a) Bibionidae: 7. Dilophus vul-
Fic. 10. Caltha palustris, L. — garis Mg. 9, in large numbers (H.M.). (4) Empzdae: 8.
(after Herm. Miller). A single ro : 2 f
carpels of, Sitpena. vu, neceaty yrtoma spuria /a@//.(H.M.); 9. Empis opaca /”., skg. (H.
(with a drop of nectar). M.). (¢) Muscidae: 10. Anthomyia sp., extremely common,

po-dvg. (H.M., Kn.); 11. Aricia serva Mg. (H.M.);
12. Hydrotaea dentipes #. (H.M.); 13. Onesia floralis R.-D.(H.M.); 14. Scatophaga
merdaria /. (H.M.); 15. S. stercoraria Z., po-dvg. (H.M.). (d@) Stratiomyidae: 16.
Odontomyia argentata /. (H.M.). (¢) Syrphidae: 17. Ascia podagrica /., po-dvg.
(H.M.); 18. Cheilosia albitarsis A/g., skg. and po-dvg. (H. M.); 19. C. sp., po-dvg.
(H.M.); 20. C. pubera Zeét., po-dvg.(H.M.). 21. Eristalis arbustorum Z., freq., skg.
and po-dvg. (Kn., H.M.); 22. E. intricarius Z., skg. and po-dvg. (H. M.); 23. E.
nemorum Z., ditto(H.M., Kn.); 24. Melanostoma ambigua /a//. (H.M.); 25. Pipiza
tristis Zg. (H.M.); 26. Platycheirus manicatus J/g. (H.M.); 27. Rhingia rostrata Z.,
po-dvg. (H.M.). C. Hymenoptera. Apzdae: 28. Andrena albicans X. , skg.
(H.M.); 29. Apis mellifica Z. ¥, very freq., skg. and po-dvg. (H.M., Kn.); 30.
Bombus terrester Z. 9, skg. (H. M., Kn.); 31. Osmia rufa Z. 4, skg. (H.M.). D.
Neuroptera. Perlidae: 32. Perla sp., freq. (H. M.).

Von Fricken noticed—in Westphalia and East Prussia —the Chrysomelid
Prasocuris hannoverana /.; Rossler—at Wiesbaden—the moth Eriocephala cal-
thella Z.

In the Alps, Hermann Miiller observed four flies.

MacLeod noted—in Flanders—Apis, 3 hover-flies, and 4 Muscidae (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 182), and—in the Pyrenees—one of the Muscidae.

RANUNCULACEAE 35

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 6) there were
observed—Apis, one humble-bee, hover-flies, Muscidae, and one of the micro-
Lepidoptera.

Burkill (‘Fertilisation of Spring Flowers’) noticed the following on the coast
of Yorkshire.—

A. Coleoptera. Nitidulidae: 1. Meligethes picipes Sturm, one instance, skg.
B. Diptera. (ca) Muscidae: 2. Scatophaga stercoraria Z., skg. with difficulty.
(4) Syrphidae: 3. Syrphus sp., one instance, po-dvg. C. Hemiptera. 4. Deraeo-
coris sp. D. Hymenoptera. Afzdae: 5. Apis mellifica Z., one instance, po-cltg.

13. Trollius L.

Flowers mostly homogamous with concealed nectar. The large bright yellow
sepals are folded together into a sphere, and serve as the means of attraction. They
enclose the small linear petals, each of which possesses an uncovered nectar-pit
at its base.

76. T. europaeus L. (Herm. Miller, ‘Alpenblumen,’ pp. 136-7; Ricca,
‘Oss. sulla fecondaz. incroc. d. veget. alp. e subalp.’; Beyer, ‘D. spont. Bewegungen

Fic. 1. Zvollius europaeus, L. (after Herm. Miller). A. Flower seen from without, somewhat
reduced. B. Petal with nectary (7) seen from within. C. Stamen, before the dehiscence of the anther,
seen from within. D. The same, seen from the side. E. An anther with loculi almost empty. F. The
same, seen from the side. B-F x 44.

d. Staubgefasse u. Stempel’; Kerner, ‘Nat. Hist. Pl’, Eng. Ed. 1, II; Kirchner,
‘Flora v. Stuttgart,’ p. 270; Schulz, ‘Beitrage,’ II, p. 8; Knuth, ‘Bliitenbiol.
Herbstbeob.’}—The flowers have a slight odour (seeming to Kerner like that of
Auricula), and in dull weather are almost completely closed, though in sunshine
the sepals are not so close together. ‘The numerous stamens have curved inwards
before the pollen is shed, and as the anthers dehisce centripetally, they elongate
to some extent. Insects penetrating from above, to reach the pollen or nectar,
first alight on the stigmas in the middle of the flower, and effect cross-pollination
with tolerable constancy. Should there be no insect visitors, automatic self-pollina-
tion is inevitable, as the outer stamens project above the stigmas. It is, however,
questionable if this is effective.

Visrtors.—I have observed one of the Muscidae (Anthomyia sp.) covered with
pollen, and also Forficula auricularia Z., devouring the anthers. Hermann Miller

D2

36 ANGIOSPERMAE—DICOTYLEDONES

found in the Alps 3 beetles, 4 flies, and 3 small Hymenoptera. Ricca also sa
small flies covered with pollen in the flowers. Schulz observed numerous flic
Hymenoptera, and beetles.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 6) there have bei
observed,—one beetle, one of the Muscidae, 3 hover-flies, and a saw-fly.

Loew observed in the Berlin Botanic Garden a bee—Halictus minutissim
K. 9, po-cltg.

14. Eranthis Salisb.

Flowers homogamous, with half-concealed nectar. The elongated yellow sep:
serve to attract insects. The nectar is secreted by the modified petals, which <
converted into small, hollow nectaries, resembling inverted cones in shape, a:
almost bilabiate. (See Fig. 9, 9.)

77. E. hyemalis L. (Herm. Miller, ‘ Fertilisation,’ pp. 80-1; Kern
‘Nat. Hist. Pl, Eng. Ed. 1, I, p. 213; Knuth, ‘Bloemenbiol. Bijdragen.’)—Miil
states that the flower mechanism agrees with that of Ranunculus auricomus. T
flowers close in dull weather, and expand in sunshine. Their stamens and carp:
mature simultaneously, so that insects—which pay their visits only while the sun
shining, when they alight upon the middle of the flower—effect cross-pollination, t
in cloudy weather automatic self-pollination takes place in the closed flower
contact of the anthers with the stigmas. Kerner states that the anthesis of the
flowers—which remain open from eight o’clock in the morning till seven in t
evening—lasts for eight days. During this time the floral leaves increase to dout
their original size.

VisiTors.—In the Garden at Kiel, I observed the honey-bee collecting poll

and sucking nectar. Hermann Miiller noticed the same insect in Westphalia ‘
great numbers, sufficient to fertilize all the flowers.’ He also observed 3 Muscidae;
1. Pollenia rudis #., ‘stroking petals, anthers, and sometimes stigmas with the
end-flaps of its proboscis, but finally thrusting its proboscis into the nectaries’;
2. Musca domestica Z., ditto; 3. Sepsis, ‘busy about the anthers.’ I saw yet
another visitor—Vanessa urticae Z.—resting on the sepals, and sucking nectar,
but it touched neither anthers nor stigmas.

15. Helleborus Adans.

Flowers protogynous, with concealed nectar. The large sepals serve to attract
insects. The petals are modified into short-stalked nectaries of greenish colour,
which are in the form of short tubes and more or less distinctly bilabiate.

78. H. foetidus L. (Kirchner, ‘Flora v. Stuttgart,’ p. 271; Knuth, Bot.
Centralbl., Cassel, lvii, 1894.)—The ovoid flowers are green, usually spotted or
bordered with brown externally, and are moderately conspicuous owing to their
being associated in a crowded inflorescence. When they open, the stigmas are
mature, and are so placed in the narrow entrance of the flower—only 1 cm. in
diameter—that every moderately large insect that creeps in must inevitably brush
against them. Each carpel possesses stigmatic papillae not only on its slightly
clavate end, but also on its outer side, these being continued along a groove as

RANUNCULACEAE 37

far as the anthers, which in this first (female) condition are still unripe. The carpels
project 3-4 mm. beyond them. The extension of the stigmatic surfaces is an
adaptation to the movement executed by nectar-seeking insects.

The filaments now elongate—in centripetal order—until the extrorsely dehiscing
anthers completely fill the entrance to the flower, which is about 1-5-2 cm.
wide. The carpels also increase in length by a few millimetres. At the beginning
of this second (male) condition, the stigmas have not quite withered, but are still
receptive, so that an insect alighting on them may still bring about cross-pollination,
or the fall of pollen from the same flower may automatically effect self-pollination.
Looked at with the naked eye, or with a lens, the stigmatic papillae certainly look
dried up, but microscopic examination shows that at the beginning of this second
stage numerous pollen-grains adhere to them. Should, however, self-pollination
take place it is probably ineffective, for ripe fruits containing seeds are rarely found
if insect-visits fail owing to unfavourable weather.

Nectar is secreted and concealed in remarkable cup-shaped nectaries, formed by
specialization of the petals. The flowers being pendulous, and the sepals forming
a close investment, the secretion is well protected from rain. The nectaries are close
to the sepals, while the stamens and stigmas project so far beyond them that some of

Fic. 12. Aelleborus foetidus, L. (From nature, half schematic.) 1. Flower in first (female)
condition. 2. Flower in second (male) condition, 4, Outline of calyx; #, nectaries ; a, immature anthers ;
a’, mature anthers; s, stigma,
the insect visitors are quite unable to reach them. ‘These content themselves with
gathering pollen, plunging into the crowd of stamens after brushing past the stigmas.
The insects which are able to suck the nectar, climb over styles and stamens,
between the latter and the sepals, till they reach the nectaries, and so always come
into contact with the stigmas, and—should the flower be in the second stage—dust
themselves with pollen. It therefore follows that pollen-collectors and nectar-suckers
alike effect cross-pollination.
Vistrors. — These are chiefly Hymenoptera, of which I observed the
following.—

1. Apis mellifica Z. ¥, very freq., both po-cltg. and skg.; 2. Bombus terrester Z.
g, occasional, ditto; 3. B. lapidarius Z. 9, ditto, ditto; 4. Anthophora pilipes / 9,
ditto, ditto. Also Diptera. 5. Eristalis tenax Z., po-dvg.

79. H. viridis Z. (Sprengel, ‘Entd. Geh.,’ p. 298; Knuth, Bot. Centralbl.,
Cassel, lvii, 1894.)—The yellowish-green flowers are solitary, or in pairs, and owing

38 ANGIOSPERMAE— DICOTYLEDONES

to the early time of flowering, insect visitors are considerably fewer in number than
in the last species.

Here again the stigmas are mature when the flower opens, and curve outwards
pretty strongly. The calyx—which is at first 1-5 cm. in diameter—exceeds them in
length by several millimetres, they in their turn being longer by about 5 mm. than the
as yet quite immature anthers, which project but little beyond the nectaries.

An insect when sucking nectar is obliged to hold on by the downwardly-directed
styles, and—if it has come from a flower in the second condition—will dust the
stigmas with foreign pollen. The styles are therefore considerably stouter than
in the last species, and their curvature—in correspondence with the wider opening
of the flower—is greater. The tips of the styles are capitate.

The stigmatic papillae are larger than in H. foetidus. They closely cover the
dilated end of the style, along the inner side of which they are continued for some
distance. The pollen-grains of the two species are of about the same size (0-04 mm.
long, and 0-02 mm. broad) and are of a similar elongated ovoid form. In H. viridis the
stigmatic papillae are somewhat conical, so that a pollen-grain exactly fits into the
space between two of them, and to a certain extent is held fast.

Fic. 13. Helleborus virrdis, L. (From nature: the three front sepals have been removed.) 1. Flower
in the first (female) condition. 2. Flower in the second (male) condition. 4, sepal; 7, nectary; a, imma-
ture anther; a’, mature anther; s, stigma.

As the stigmas gradually wither, the stamens elongate centripetally, turning
outwards the sides which are covered with pollen to the place that the stigmas
occupied in the first stage of flowering. At the same time the sepals spread out
so as to increase the diameter of the flower to 3 cm.

The nectaries are considerably larger than in the last species, and—the flower
being pendulous—rain is kept away from their secretion. The blossom having
a much larger opening than in H. foetidus, insect visitors easily reach the nectar-
cups without much searching, and suck while holding on by the styles and stamens,
so that here again cross-pollination must ensue. None of the insect visitors I
observed were engaged solely in gathering pollen.

VistTors.—I observed the following in Kiel Garden.—

Hymenoptera.- 1. Apis mellifica Z.; 2. Bombus terrester Z.9; 3. B. lapi-
dorius L.

MacLeod noticed an Andrena in the Pyrenees, and Burkill (‘Fertilisation of
Spring Flowers’) saw on the coast of Yorkshire,—Bombus terrester Z., skg.

RANUNCULACEAE 39

80. H. niger L. (Knuth, Bot. Centralbl., Cassel, lvii, 1894.)—In spite of the
very large white flowers insect visitors are extremely few, no doubt because the un-
favourable flowering season usually deters them from flying about. The flowers
in all respects agree in structure with those of H. viridis. Warnstorf (Schr. natw.
Ver., Wernigerode, xi, 1896) says that the number of the nectaries is about 10-12.
The white, smooth, ellipsoidal pollen-grains average 53 y in length, and 28 p in
breadth.

Visitors.—I observed at Kiel only Apis mellifica Z.

81. H. siculus Schff.—native to Etna—agrees essentially in its anthesis (Nicotra,
Boll. Soc. bot. ital, Firenze, 1894) with the other species of Helleborus described
by me, more particularly with H. viridis. This Sicilian species is also protogynous.
The nectaries only begin to secrete when the anthers dehisce. Autogatay is com-
pletely prevented, for the stigmas have withered by the time the first anthers have
shed their pollen.

82. H. atrorubens Waldst. et Kit—

Visirors.—Loew observed—in the Berlin Botanic Garden—the honey-bee,
po-cltg,, and saw it also on

83. H. cyclophyllus Boiss., and
84. H. lividescens A. Br. et Sauer, as well as on

85. H. pallidus Host, on which species in the same place he also noticed
one of the Muscidae (Scatophaga stercoraria Z.).

16. Isopyrum L.

Flowers with half-concealed nectar. The sepals serve as the chief means
of attracting insects. The petals are modified into shovel-shaped nectaries, and are
considerably smaller than the sepals.

86. I. thalictroides L. (Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 120.)—
Soon after the white flower has opened the anthers of the outermost staminal whor]
dehisce, and at the same time the filaments bend over so as to bring them above
the nectaries, in which position they must necessarily be brushed against by nectar-
sucking insects. Next day these stamens move outwards towards the reflexed
sepals, while simultaneously the next whorl of stamens dehisce, and bend over the
nectaries. On the third day these in turn move outwards, their place being taken
by the members of the third whorl, and so on in succession till all the stamens have
brought their anthers above the nectaries. Insects alighting upon the middle of
the flower must of necessity effect cross-pollination if they have already visited
another blossom of the same species.

Visrtors.—Nothing is known about these.

17. Nigella Tourn.

Markedly protandrous bee-flowers. The large brightly coloured sepals serve
to attract insects. The eight petals are converted into nectaries of a characteristic
kind. They possess a hollow claw, bent like a knee, and a split limb provided with

40 ANGIOSPERMAE—DICOTYLEDONES

two processes. Above the angle of the claw, on its upper side, there is a slit
protected by a cover. The nectar is secreted internally on the under side of the
bend, and is concealed in the cavity of the tubular claw. The cover closes by
elasticity after it has been raised, and lies between two projections which prevent
it from being displaced. (See Fig. 14, B,C, D.)

87. N. arvensis L. (Sprengel, ‘Entd. Geh.,’ pp. 280-9 ; Terraciano, ‘ Intorno
alla strutt. fiorale ai process. d’impollinaz. in alc. Nigella,’ Boll. Soc. bot. ital.,
Firenze, 1892, pp. 46-51; Knuth, ‘ Bloemenbiol. Bijdragen.’)—Sprengel has given
a very thorough account of the structure of the flower. The description is one of
the most signal achievements of this great investigator.

The sepals are whitish below and bright blue at the tip. The small nectariform
petals are brownish or blue on the upper side, with two white or yellow-green
transverse bands. Their limbs present whitish or brown transverse striations: the
process of the nectar-cover is whitish and brown. It is further to be noted that
the white filaments are marked on the inner surface—not far from their insertion—
with a white spot that can be dimly seen by transparency on the outside. The

Fic. 14. Wigella, L. (From nature.) 4. Nigella damascena Z. Flower towards the end of the
second stage: the styles have curved spirally downwards, and are in contact with the anthers (a), which
are still covered with pollen, so that automatic self-pollination necessarily follows. (Natural size.) B, C,
D. Nectaries of N. Sativa, N.damascena, and N. arvensis (x 34). The form of the nectaries B and C
appears to be somewhat variable.

flower thus possesses a series of ten alternating light and dark rings, which serve
as annular nectar-guides, by which insect visitors (bees) are directed to one nectary
after another.

Above the nectaries there are eight groups of stamens, each consisting of six
members, one behind the other. When the flower opens all the forty-eight stamens
are erect. On the first day of flowering the outermost member of each group
curves downwards and outwards, which .causes the dehiscing side of its anther to
face downwards, so that the dorsal surface of an insect sucking from the nectaries
must necessarily be dusted with pollen. On the second day the eight outermost
stamens—which are now withered—assume a horizontal position, so as to lie upon
the sepals, and their place is occupied by the eight stamens next in order. On

RANUNCULACEAE 41

the third day these also have moved down, and are replaced by a third set of eight,
and so on in succession, until in six days’ time all the stamens are withered and
prostrate.

Meanwhile the styles are erect, but they gradually turn over in a somewhat
spiral fashion, curving outwards and downwards so as to be almost horizontal when
all the anthers have withered. The stigma, which is in the form of a longitudinal
seam extending from the base of the style to its tip, must now come into contact
with insects engaged in sucking nectar, and will be dusted with pollen should these
happen to come from a flower of the same species in the first condition. After
the styles have occupied this position for three or four days, they again become erect.
I observed in garden plants that automatic self-pollination sometimes took place
by a spiral downward movement of the styles, in cases where insects had not brought
about cross-fertilization. Terraciano states that the pollen of the lower stamens is
unable to effect self-fertilization, and that automatic self-pollination usually occurs
by the stigmas coming into contact with the anthers of the upper stamens. This
investigator observed no insect visitors on wild plants: on the contrary, he found
that this species, and also N. sativa Z., N. damascena L., N. Bourgaei Jord., N.
foeniculacea DC., and N. gallica /Jord., produced numerous seeds capable of
germinating without the aid of pollinating agents, i.e. as the results of automatic
self-pollination.

VisiTors.—Sprengel long ago observed bees which were precisely adapted to
the dimensions of the flowers. Bees alone are clever enough to lift the nectar-covers.
I have seen Apis, and also Bombus lapidarius Z. ¥, skilfully opening the nectaries,
and regularly effecting cross-pollination. I also saw Vanessa Io Z., as a useless
flower-guest.

Friese observed in Hungary,—the silk-bee Colletes punctatus Mocs. 6 not
infrequent, 9 very infrequent, and its parasite the cuckoo-bee, Epeolus fasciatus Friese
(= E. transitorius Frzese), not infrequent.

88. N. sativa L. (Knuth, ‘Bloemenbiol. Bijdragen’) agrees in the structure
of its flowers with the last species, but the nectaries are slightly different.

VistTors.—I saw in the Kiel Botanic Garden the honey-bee and 2 humble-
bees (Bombus terrester Z. ¥, and B. lapidarius Z. ¥ and 4) on this species and N.
damascena. These opened the nectaries one after the other and sucked nectar.
While doing so, the upper side of the thorax was in contact with the pollen-covered
anthers, or the receptive stigmas, thus effecting crossing.

89. N. damascena L. (Knuth, ‘ Bloemenbiol. Bijdragen,’ ‘ Bliitenbiol. Notizen ’;
Herm. Miiller, ‘ Weit. Beob.,’ I, p. 322.)—The flowers agree in structure with those
of the last two species. Here automatic self-pollination regularly takes place by
spiral twisting downwards of the styles, and, as the flowers always set fruits, it would
appear to be effective. (Cf. Fig. 14, 4.)

Visitors.—Vide supra, No. 88.

Buddeberg observed the two following bees in Nassau.—
1. Ceratina callosa #. 6, busy among the stamens; 2. Prosopis confusa
Nyl. 3, skg.

42 ANGIOSPERMAE—DICOTYLEDONES

18. Aquilegia Tourn.

Protandrous humble-bee flowers. The blossoms are rendered conspicuous by
the brightly coloured sepals and petals, and the stamens and carpels project from
among them like a yellow central column. Nectar is secreted and concealed at
the bottom of the spurs of the petals.

go. A. vulgaris L. (Sprengel, ‘Entd. Geh., pp. 279-80; Herm. Miiller,
‘Fertilisation,’ pp. 81-2; Beyer, ‘D. spont. Bewegungen d. Staubgefasse u. Stempel’;
Schulz, ‘ Beitrage’; Kirchner, ‘Flora v. Stuttgart.” p. 273; Knuth, ‘ Weit. Beob. wt.
Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 230, ‘ Bliitenbiol. Beob. a.d. Ins. Riigen.’)—The
violet-blue (rarely pink or white) flowers are pendulous, so that the nectar secreted in
the ends of the spurs of the petals is protected from rain. The spurs are 15-22 mm.
long, and their funnel-like openings are wide enough to readily accommodate the
head of a humble-bee. The much narrower terminal part of the spur curves inwards
and downwards, and conceals the nectar, which is secreted by a fleshy thickening at
the extreme tip. Humble-bees—which possess a proboscis long enough to reach
the nectar in the legitimate way—cling to the flowers from below, holding on to

Fic. 15. <Aguilegia vulgaris, L. (From nature: the front petal and two sepals have been removed.)
A. Flower in the first (male) condition : most of the anthers (a?) have already dehisced, a few of them (a!)
are still immature, with their short filaments still directed upwards. The stigmas (s) are still immature.
B. Flower in the second (bisexual) condition: all the anthers (a) have dehisced, and the stigmas (s) are
mature. 2, nectary,

the base of the spur with the fore-legs, and to the stamens and carpels with the
other two pairs of legs, and pushing the head into the spurs. In so doing the under-
side of the abdomen comes into contact—in the younger flowers—with the pollen-
covered anthers, which closely surround the pistil. In older flowers the same region
of the body touches the somewhat spreading stigmas which project from among
the stamens, so that cross-pollination is necessarily effected. Should insect-visits
fail, automatic self-pollination readily takes place, for the carpels grow down in
the middle of the stamens, and by elongation of the styles the stigmas ultimately
reach a lower level than the anthers.

VisiTors.—The usual pollinator is the garden humble-bee (Bombus hortorum Z.),
the long (19-21 mm.) proboscis of which easily reaches to the tip of the spur. It
is the most frequent visitor of the columbine: I observed it in Schleswig-Holstein,
Mecklenburg, Thuringia, North Pomerania, Riigen, the North Frisian islands, and

RANUNCULACEAE 43

elsewhere, while Hermann Miller noticed it in Westphalia. He also saw—but much
more rarely—the field humble-bee (Bombus agrorum /. 9) sucking nectar in the
legitimate way, and effecting crossing. This bee, however, in order to enable its
somewhat shorter (12-17 mm.) proboscis to reach the nectar, must have thrust
its head right into the opening of the spur, so as to diminish the distance to be
traversed by about 5mm. Bees with a still shorter proboscis are excluded from
legitimate enjoyment of the nectar, and they must bite through the spur to get at it.
Bombus terrester Z.—the proboscis of which is only 7-3 mm. long—is particularly
addicted to biting through the bend of spur and stealing nectar through the
hole. Hermann Miiller saw one of these bees settle upon the upper side of a
columbine flower, lick the bases of the sepals, and on finding nothing there, creep
to the lower surface and thrust its head into the spur. Still finding nothing, it once
more crept to the upper surface, again vainly licked the bases of the sepals, and
finally bit through the spur, thrust the tip of its proboscis into the hole, and thus
secured the nectar. It subsequently robbed the remaining spurs of the same flower,
and those of other flowers, without further deliberation, and it is probable that
every member of the species has to learn by experiment how to get at the nectar.
But, after having acquired the knowledge, it perforates the spurs even of unopened
flowers, thus stealing a march on legitimate visitors, as H. Miiller observed at
Lippstadt, and I myself at Kiel. Sprengel noticed that the honey-bee (with a
proboscis 6-7 mm. long) behaved in the same way as B. terrester. Hermann Miiller
confirms this, and adds that it also often uses the perforations made by B. terrester.
The holes bitten out by humble-bees when stealing nectar were also observed by
Schulz in the Tyrol and in Thuringia.

Both honey-bees and some of their smaller allies—among which Hermann Miiller
noted Halictus smeathmanellus A. 9 and H. leucozonius Schr. 9—collect pollen upon
columbine flowers, and in doing so may effect either cross- or self-pollination.

Visirors.—Schenck observed the following in Nassau.—

Hymenoptera. (a) Afidae: 1. Andrena convexiuscula K.; 2. A. curvungula
Thoms.; 3. Halictus xanthopus K. (4) Vesprdae: 4. Odynerus melanocephalus Z.

MacLeod noticed three humble-bees in the Pyrenees, of which only Bombus
hortorum Z. was skg. legitimately (‘ Pyreneénbl.,’ p. 386).

gi. A. atrata Koch. (Herm. Miiller, ‘Alpenblumen,’ p. 137.)—The flowers
essentially agree with those of the last species. It is, however, doubtful whether
automatic self-pollination occurs should insect-visits fail.

Vistrors.—Herm. Miiller observed three humble-bees, and two bees (species of
Andrena).

g2. A. pyrenaica DC. (MacLeod, ‘Pyreneénbl., pp. 385-6.)—The flowers
are darker than those of A. vulgaris, with which they agree in other respects. The
narrow spur of the petals is 20 mm. long, but 5-6 mm. broad at the entrance, so
that a proboscis about 15 mm. long is able to reach to the tip.

Visttors.—MacLeod observed none.

93. A. chrysantha A. Gray. (Knuth, ‘Bloemenbiol. Bijdragen.’)— This
species is indigenous to North America, and is a garden plant in West Europe. It
possesses flowers agreeing essentially with those of A. vulgaris, but the spur is

44 ANGIOSPERMAE—DICOTYLEDONES

45-50 mm. in length, so that the proboscis (20 mm. long) of Bombus hortorum Z.—
which I have repeatedly observed as a visitor—can reach only part of the nectar that
often accumulates in the spur to the depth of 30 mm., or more. The length of the
spur, and the pale colour of the blossoms, suggest that we have here to do with
hawk-moth flowers, but I have not observed any Sphingids as visitors at Kiel.

19. Actaea L.

Small protogynous pollen flowers, arranged in racemes. The whitish sepals,
petals, and stamens of the individual flowers serve to attract insects.

94. A. spicata L. (Ricca, ‘Oss. sulla fecondaz. incroc. d. veget. alp. ¢
subalp.’; Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II; Delpino, ‘ Ult. oss.,’ II: Herm
Miller, ‘ Weit. Beob.,’ I, p. 323; Kirchner, ‘ Beitrage,’ p. 18.)—According to Ricca
Kerner, and Kirchner, the flowers are protogynous. The ovaries are greenish, bul
otherwise the flowers are almost completely white, except that the sepals have violei
tips, and the filaments are also in some instances of a pale violet hue. The stamens
are spreading, and their ends clavate.

Visttors.—Buddeberg observed—in Nassau—a beetle (Byturus fumatus .’
and an earwig (Forficula auricularia Z.), the latter devouring pollen and apparently
the anthers as well.

20. Cimicifuga L.

Flowers with half-concealed nectar, secreted at the bases of the petals ir
cup-shaped pits.

95. C. foetida L. (Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, II, p. 202.)—The
small whitish flowers are in long racemes, and smell like new honey. They posses:
shovel-shaped nectaries. This is all that is known about the flower-mechanism, anc
information regarding visitors is entirely lacking.

21. Delphinium Tourn.

Protandrous humble-bee flowers. The nectar is secreted in the ends of one or
two spurs, belonging to the two upper petals, and is concealed so deeply that it can
only be reached by humble-bees with a long proboscis. The sepals serve as the

chief means of attracting insects. The upper one is spurred, and its spur ensheathes
those of the petals.

96. D.elatum L. (Herm. Miller, ‘Fertilisation,’ pp. 83-5, ‘ Weit. Beob.,’
p- 322; Beyer, ‘D. spont. Bewegungen d. Staubgefasse u. Stempel’; Schulz,
‘Beitrige,’ Il, p. 204; Knuth, ‘Bloemenbiol. Bijdragen.’)—As Hermann Miiller has
explained in a masterly fashion, the spur of the sepal not only serves to protect the
nectar, but also compels the humble-bees that alight in search of nectar to do so
in the only way that effects pollination. The hollow, sharply conical end of the
posterior process of each upper petal secretes nectar, getting so full of it that some
enters the semi-conical cavity of this process, which is open on its inner side. As
the two spurs are closely apposed, they together form a hollow cone, splitting at the
end into two horns filled with nectar. The proboscis of a humble-bee—if long

RANUNCULACEAE 45

enough—is thus unfailingly guided, while, at the same time, the length of the cone
denies access to shorter-tongued insects. The forwardly directed parts of the same
petals prolong the upper part of the hollow cone to the front, and as they expand
and turn up anteriorly, they afford a convenient approach to the proboscis of
a humble-bee, and also direct it with certainty to the nectar receptacle. These
anterior parts of the upper petals separate on slight pressure, so that the head of
a humble-bee can be entirely thrust between them, thus diminishing the distance
to the nectar by 6 to 7 mm. The length of the hollow cone from its entrance
to the beginning of the nectar-bearing horns is about 20 mm., and to the ends of
the latter 26-28 mm., so that when the head of a humble-bee is thrust into the
aperture, its proboscis must be 13-14 mm. long in order to reach the nectar, and
Ig-22 mm. long to empty the receptacle.

Fic. 16. Delphinium elatum, L. (after Herm. Miller). 1. Young flower after removal of the
petaloid calyx; front view. 2. The petals in their natural position, seen obliquely from in front and below.
3. Young flower, after removal of the right side of the calyx, seen from the right side. 4. Older flower, after
removal of the calyx, front view. 5. Young flower, after removal of the right side of the calyx and corolla,
seen from the right side. aa, The two upper petals, which are prolonged backwards into two spurs (a! a?)
that secrete and contain nectar (7), and which in front form an entrance for the humble-bee's proboscis ;
a* a*, bases of the same; 44, the two lower petals, of which the closely apposed surfaces form the lower
boundary of the entrance for the humble-bee’s proboscis; the upper surface of each bears a tuft of yellow
hairs to serve as a nectar-guide, while between their claws (2. 6’d’) is an interval (2. &*) in which the anthers
and stigmas successively appear (in the two stages of flowering), and which is in the path of the insect’s
proboscis ; ¢, dehisced anthers placed behind the entrance to the spur in the path of the insect's proboscis;
d, anthers which have not yet dehisced, and are bent downwards over the female organs; ¢, surface of
attachment of the stamens and carpels (which have been removed); /, withered stamens bent downwards ;
g, ovaries; 4, stigmas, which have placed themselves in the same spot that the ripe anthers occupied
in the first stage; 7, left half of the upper sepal, prolonged backwards into a corrugated sheath (2’)
for the spur; 4, left lateral sepal; 4 left lower sepal. (3 and § natural size; 1, 2, and 4 enlarged.)

There are bundles of erect yellow hairs upon the front surfaces of the two lower
petals—serving as nectar-guides—which are here closely apposed, thus bounding below
the way of entry to the nectar, and obliging humble-bees to probe in the only right
place. The claws of these petals, on the other hand, are sufficiently far apart to
enable the stamens and—after they have withered and bent back—the carpels to

46 ANGIOSPERMAE—DICOTYLEDONES

erect themselves in that part of the hollow cone lying immediately behind its entrance.
Here they inevitably come into contact with the under-surface of the proboscis or
head of a humble-bee engaged in sucking, so that cross-pollination is effected.

The stamens in their first immature condition are directed downwards, and
become erect as their anthers dehisce, thus putting themselves in the way of the
head of any humble-bee that may be sucking nectar. They bend down again after
withering, so as to make room for the styles, which are now turning up, and carrying
with them the newly-matured stigmas. Automatic self-pollination is thus excluded,
and cross-pollination is necessary for fertilization. Darwin states that artificial
self-pollination is ineffective.

VisiTors.—Only humble-bees with a proboscis 19-22 mm. long can suck out
all the nectar in the legitimate way as described. Of the bees native to Central and
North Germany, there are but two with such a proboscis, i.e. Anthophora pilipes /.
(= Podalirius acervorum Z.) and Bombus hortorum Z. The former of these,
however, is no longer on the wing when Delphinium elatum comes into flower,
so that the garden humble-bee alone remains as the regular pollinator of this
larkspur. As a matter of fact, Hermann Miiller and myself—making observations
in the Lippstadt and Kiel Gardens respectively—have seen hardly any other insect
visitor to Delphinium elatum sucking nectar in the legitimate way, and consequently
effecting cross-pollination. Some of our other native humble-bees, however, possess
a proboscis long enough to reach at least some of the nectar, e.g. B. agrorum /.
(proboscis ro—15 mm. long), and B. senilis Sm. (proboscis 14-15 mm. long).
Hermann Miiller also observed Anthophora personata J//. 9, visiting the flowers
at Strassburg. In the Riesengebirge, Schulz noticed holes in the spurs that had
no doubt been made by thieving short-tongued bees.

97. D. Staphisagria L.—Hildebrand states that this agrees in the main with
the last species as regards both structure and visitors.

98. D. Consolida L. (Herm. Miller, ‘Fertilisation,’ pp. 85-6, ‘ Weit. Beob.,’
I, p. 322; Kirchner, ‘Flora v. Stuttgart, p. 274; Schulz, ‘ Beitrage,’ II, p. 204;
Knuth, ‘ Bliitenbiol. Herbstbeob.,’ ‘ Weit. Beob.,’ p. 231.) —As first explained by Her-
mann Miiller, the flowers of this species chiefly differ in structure from those of the
preceding in the union of the four petals. The posterior prolongations of the upper
petals are fused into a single spur 15 mm. long, which secretes and conceals nectar
in itsend. The anterior lobes of the four petals form in addition a tube 7 mm. wide
at its commencement, which readily admits the head of a humble-bee. As it is only
open below, the pollen comes into contact with the under-side of a humble-bee’s
head at the beginning of anthesis, while the stigmas do so later, for stamens and
carpels develop in the same order as in the case of D. elatum. Cross-pollination
is consequently secured in D. consolida when its nectar is sucked legitimately.
Automatic self-pollination is excluded: artificial self-pollination is only moderately
effective.

A proboscis 15 mm. long would be able to reach the tip of the spur, which
is 15 mm. in length—exclusive of the entrance, that measures 7 mm.

Vistrors.—Among the humble-bees native to Central and North Germany, there
are—once more neglecting Anthophora pilipes #. with a proboscis 19-21 mm. long,

RANUNCULACEAE 47

but which is no longer on the wing when the field larkspur comes into flower—
a number of species with a proboscis 15 mm. in length (cf. vol. I, p. 160), but
Bombus hortorum Z.—with a proboscis 17-21 mm. long—is alone able to reach
the nectar without considerable loss of time, the other species being compelled to
force their heads between the four petal-lobes that form the entrance. Wherever
observations have been made on the guests of Delphinium consolida (Westphalia,
Thuringia, Schleswig-Holstein), this humble-bee has been shown to be the normal
visitor and pollinator. It flies with the greatest industry from flower to flower,
steadily effecting cross-pollination, and receiving all the nectar in return. Occasion-
ally, indeed, other guests appear, especially Lepidoptera (species of Vanessa, Pieris,
Satyrus, Hesperia), which usually reach the nectar with their long thin proboscis
without touching stamens or carpels. I sometimes saw the honey-bee—the proboscis
of which is only 5-7 mm. long—visiting the flowers and vainly seeking for nectar,
thrusting its proboscis into the spur, and therefore now and then effecting cross-
pollination. In Thuringia, Schulz observed spurs perforated by nectar-thieves, no
doubt short-tongued humble-bees, though he was not able to catch the plunderers
in the act. In the same part of Germany, Hermann Miiller also noticed B. lapi-
darius Z. sucking. Schletterer—at Pola—saw the bee Anthidium manicatum Z.
as a visitor.

99. D. Ajacis L. (Sprengel, ‘Entd. Geh., pp. 277-8; Knuth, ‘ Bliitenbiol.
Herbstbeob.’)—The flowers agree in structure with those of the last species. The
length of the spur in garden plants at Kiel was 15-18 mm.

Vistrors.—As a regular visitor and pollinator, I observed Bombus hortorum Z.,
and as an occasional visitor the honey-bee, which of course did not succeed in
reaching the nectar. As a nectar-thief, I now and then noticed Vanessa lo. L.

100. D. grandiflorum Jord.—Jordan states that thé anthers of the posterior
stamens are directed forwards, while those of lateral and anterior stamens are
directed laterally outwards.

22. Aconitum Tourn.

Lrrerature.—Kronfeld, M., ‘Uber d. biol. Verhiltn. d. Aconitumbliite,’ Bot
Jahrb., Leipzig, xi, 1889; Abstract in Bot. Centralbl., Cassel, xxxvi, 1888, p. 392.

Protandrous humble-bee flowers. The large sepals are blue, violet, vivid yellow,
or brightly mottled, and in conjunction with the smaller petals serve to attract
insects. Conspicuousness is enhanced by the aggregation of the flowers into
racemes, which usually bear numerous blossoms. The two upper petals are con-
verted into long-stalked hood-shaped nectaries, covered by the upper sepal, which
is galeate.

M. Kronfeld describes the stalk of the nectary of Aconitum as a hollow splint,
into which nectar-sucking humble-bees insert their proboscis, this being thrust
forwards and upwards to reach the actual source of the nectar.

As Kronfeld points out, the nectary is specialized to a varying extent in different
species (see Fig. 17). The simplest form occurs in A. heterophyllum Wall. an East
Indian form in which there is a tolerably thick stalk, expanding into a cap open

48 ANGIOSPERMAE—DICOTYLEDONES

below, with only a short lip on its free margin. In A. palmatum Wall. the spur
makes its first appearance as a slight projection; in A. Napellus Z. the lip is
elongated and expanded; in A. Anthora Z. and A. columbinum MWw//. the spur is
more prominent, and in A. paniculatum Zam. it is well defined. The spur of the
Japanese species A. Fischeri Rezchend. is prolonged and curved inwards like a
flamingo’s bill; in A. septentrionale Koelle it is drawn out in the shape of
a proboscis to a length of 6 mm.; and lastly, in A. Lycoctonum Z., it is coiled
into a spiral of 12 turns.

The blossoms of Aconitum are—as Kronfeld expresses it—‘ humble-bee flowers
par excellence’ The body of a humble-bee exactly fills the interior of the flower, and
if a plaster cast of this be made it wilt be found to correspond in a remarkable way to
the shape of a medium-sized female humble-bee. As a matter of fact, Aconitum is
dependent on Bombus, and must become extinct in places where humble-bee visits

Fic. 17. Stages 1n Specialization of the Nectarves of Aconttum (after M. Kronfeld). Nectary of :—
1. Aconitum heterophyllum Wa//. (with no spur). 2. A. palmatum Wa//. (witha slight projection). 3. A.
Napellus Z. (with a somewhat larger projection, and a longer lip). 4. A. Anthora Z. and A. columbinum
L. (with projection still more developed). 5. A. paniculatum Laz. (well-defined spur). 6. A. volubile
Pall. and E. villosum Rg/. (with buckle-shaped posterior projection. 7. A. Fischeri Retchend. (spur
curved inwards). 8. A. septentrionale Ae//e (spur rolled inwards). 9. A. Lycoctonum Z. (spur rolled
inwards to a greater extent).

fail, or where these insects merely rob the flowers by making lateral perforations.
The dependence of monkshoods on humble-bees is best made clear by comparing
their respective areas of distribution. A glance at the accompanying map (Fig. 18)
will show that the area of distribution of Aconitum is entirely included in that of
Bombus, while the two are coincident in the head quarters of the genera. In other
words, the range of monkshoods is closely related to that of humble-bees.

tor. A. Napellus Z. (Sprengel, ‘Entd. Geh.,’ pp. 278-9 ; Herm. Miiller,
‘Alpenblumen, pp. 137-9; Beyer, ‘D. spont. Bewegungen d. Staubgefasse u.
Stempel’; Knuth, ‘Bliitenbiol. Beob. a. d. Ins. Riigen,’ and other works.)\—The
large upper sepal of the erect flower not only helps to attract insects, but also roofs
over and protects the underlying nectaries, stamens, and carpels. The three smaller
lower sepals, together with the two lower petals, take part in the work of attraction;
and besides this they serve as alighting-places and platforms for the humble-bees
that creep into the flower, and complete the protective investment of the stamens and
carpels. The two upper petals are converted into characteristic nectaries, each with

RANUNCULACEAE 49

a stalk about 15 mm. long, which follows the curve of the upper sepal, and then
gradually expands into a receptacle that is open below and provided with tapering
mouth-lobes, while its upper end is closed and dilated. The dilatation is bluish-
black externally, greenish internally, and its inner surface secretes so abundantly that
a large drop of nectar hangs in the narrowed neck of the receptacle.

The numerous stamens lie at first in the entrance to the flower with their
immature anthers directed downwards. They then become erect as the anthers
dehisce, and are so placed that the pollen must dust the ventral surfaces of humble-
bees sucking nectar. During this first stage the 3-5 carpels are immature, and
are so closely surrounded by the stamens as to be completely covered. As the

CS}

Geographical Distribution

worneee- Aconitum

Fic. 18, Map showing the distribution of the genera Aconitum and Bombus (after Kronfeld).

stamens wither they bend downwards again, so that at the end of the first stage
the anthers of the anterior stamens are directed backwards, those of the lateral
stamens inwards, and those of the posterior stamens chiefly sideways. When all
the stamens have withered the stigmas become mature, and being now free from
their staminal investment occupy the entrance to the flower, so that humble-bees
dusted with pollen from a blossom in the first stage, necessarily touch the stigmas
and thus effect cross-pollination. Automatic self-pollination is thus normally excluded,
but it sometimes happens that one or two pollen-covered stamens have not bent
down again towards the base of the flower when the stigmas have matured, and
in such exceptional cases self-pollination may take place.

Visitors. — Those which act as pollinators are exclusively humble-bees.
Wherever I have observed this plant in gardens (Kiel, North Frisian Islands,
Mecklenburg, Riigen, Thuringia, and elsewhere), I have seen the garden humble-bee

DAVIS. IL E

50 ANGIOSPERMAE—DICOTYLEDONES

(Bombus hortorum Z.) sucking, and occasionally the ground humble-bee (Bombus
terrester Z.). Hermann Miiller—in the Alps—saw various humble-bees sucking
nectar or collecting pollen, and one of them (Bombus mastrucatus Gers/.) biting
through the upper sepal and stealing nectar, though some individuals sucked it in
the legitimate way. In the same region he also noticed one of the Lepidoptera
(Lycaena sp.) vainly trying to get at the nectar.

Frey-Gessner observed the following bees in Switzerland.—

1. Bombus agrorum F. 9, §, and 6; 2. B. alticola Krzechd. ¥ and 8; 3. B. brevi-
gena Z/s. (=B. mastrucatus Gers/.); 4. B. Gerstackeri Mor. 9, ¥, and é (especially
foundress-queens); 5. B. hortorum Z. 9 (which had deserted the nest), ¥, and 6;
6. B. mendax Gers?. one 9, many ¥, one 6; 7. B. pratorum L.

Dalla-Torre noticed B. alticola A7rzechd. in the Tyrol, and Schletterer B. hortorum
Z. in the same region.

Fic. 19. Aconzlum Napellus, L. (after Herm. Miller). A. Flower in first (male) stage. The dark-
+ coloured stamens have become erect; their anthers have dehisced, and are coated with white pollen.
B. The same flower seen from the side; a, a hole bitten by Bombus mastrucatus. C. The same in
longitudinal section. D. Stamens and carpels in the first (male) stage: some of the anthers are mature,
the stigmas are still immature; a’, undehisced anthers with filaments bent back; a’, stamens that
are becoming erect ; 2", stamens that are erect and covered with pollen; a’, stamens with empty anthers
bending backwards; a’, ditto, completely bent back. E. Stamens and carpels in the second (female)
stage: the anthers are all empty, and the stamens bent back: the stigmas are mature. (A—C natural
size; Dand E x 2.)

Gerstacker observed the following humble-bees at Kreuth.—

1. Bombus hortorum Z., ‘frequently biting off [the flowers] at their bases, like
other species of humble-bee’; 2. B. Gerstackeri AZor.; 3. B. mastrucatus Gers? 9
and ¥, ‘biting off [the flowers] at their bases’; 4. Psithyrus globosus Zz.

Alfken saw the following humble-bees at Bremen.—

1. Bombus agrorum /.; 2. B. hortorum Z.; 3. B. sylvarum Z. And on the
Schlern in the Tyrol—4. B. Gerstackeri Afor.

MacLeod observed in the Pyrenees three humble-bees, and the humming-bird
hawk-moth (Macroglossa stellatarum), (‘ Pyreneénbl.,’ pp. 381-2).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 7) a humble-bee was
noticed.

RANUNCULACEAE 51

Kronfeld states that Handlirsch observed eight species of humble-bee in lower
Austria, and Hoffer ten in upper Austria, those with a short proboscis (B. mastru-
catus, B. terrester, B. soroénsis, and B. mendax) obtaining the nectar by perforating
the flowers.

1o2, A. variegatum L.—

Visrtors.—Kronfeld—in Austria—saw Bombus agrorum (§ and 9) and B.
hortorum (9, ¥, and &) sucking legitimately, also Halictus morio and other short-
tongued insects vainly seeking for nectar. Schulz observed perforated flowers of
this species in Thuringia.

103. A. Lycoctonum L. (Sprengel, ‘Entd. Geh.,’ p. 279; Herm. Miiller,
‘Alpenblumen,’ pp. 139-40; MacLeod, ‘ Pyreneénbl.’; Aurivillius, C., ‘Uber d. BI.
u. Befrucht. v. Aconitum Lycoctonum L.,’ Bot. Centralbl., Cassel, xxix, 1887,
pp.125-8; Kronfeld, ‘ Uber d. biol. Verhaltn.
d. Aconitumbliite,’ Bot. Jahrb., Leipzig, xi,
1889; Loew, ‘Blumenbesuch,’ I, p. 28;
Knuth, ‘ Bliitenbesucher’).— This agrees
essentially with the species already de-
scribed in the structure of its flowers, but
the nectar is so deeply placed as to be
only accessible to humble-bees with a very
long proboscis. The upper sepal of the
yellow blossom is an almost vertical cylinder,
which serves as a protective envelope for
the nectaries. Each of these is produced
into a spiral tube (of one and a half turns),
which is full of nectar, that is here secreted
very abundantly. The stalk of the nectary
is about 20 mm. long, and a_ proboscis

5 FIG. 20. Aconttum Lycoctonum, L. (after Herm.
of about the same length Is necessary for Miller). A. Flower in the second (female) condi-

sucking. as humble-bees can get no foot- tion; seen from the side. Natural size. B. Longi-
ener 4 2 5 ; tudinal section of the same (nearly twice natural
hold in the cylindrical sepal, but are obliged _ size). The upper stamens have already fallen off.

to cling to the stamens and carpels.

Visirors.—In Central and North Germany the yellow monkshood is visited
exclusively by Bombus hortorum Z.; in the Alps almost solely by B. opulentus
Gerst. (=B. Gerstackeri Afor.)1. These two humble-bees have a longer proboscis

1 Frey-Gessner (‘ Exkursionen im Sommer 1880,’ Mitt. Schweiz. Ent. Ges., vi, 1881) first pointed
out that the old 9 of Bombus Gerstackeri Mor. consistently alights upon Aconitum Lycoctonum Z., the
% and 3$ upon A. Napellus Z.. and this phenomenon has been described by Dalla Torre as hetero-
trophy (cf. vol. 1, pp. 160-1). This investigator explains the adaptation as due to the extremely short
working period of this humble-bee, which only appears in July, and of course disappears at the end
of September or beginning of October, and to the fact that § and é appear simultanéously with
the foundress-queens (from Aug. 20 on), so that it is in the interest of the species that they should
visit different flowers. The term ‘heterotrophy’ (érepos, different ; tpopy, food) is reasonable in so
far that in districts where A.. Lycoctonum and A. Napellus occur’'side by sid: in great profusion, the
g of B. Gerstackeri appears actually only to visit A. Lycoctonum, the § and é on the other hand only

E 2

52 ANGIOSPERMAE—DICOTYLEDONES

than allied species native to the regions mentioned: that of B. hortorum is 21 mm.
long, and that of B. Gerstackeri 22 mm. Alfken tells me—in a private letter—that,
when ascending the Schlern in the Tyrol, he noticed large numbers of B. hortorum
L. (g and ¥) with B. Gerstackeri Aor. (9) on the flowers of A. Lycoctonum, diligently
sucking nectar.

In Jamtland (Sweden), Aurivillius observed’ B. hortorum Z., and frequently B.
consobrinus Dafld. as well. The latter resembles the former so closely that
Schmiedeknecht describes it as a variety of this (‘ Apidae Europ.,’ pp. 295, 297, 305):

MacLeod—in the Pyrenees—also met with B. hortorum Z. as a visitor of A.
Lycoctonum Z., var. pyrenaicum Ser. (=A. pyrenaicum). The same observer
further noticed many individuals of B. Gerstackeri Mor. 9, sucking nectar from this
flower, and effecting cross-pollination.

From these observations it appears that A. Lycoctonum Z. is everywhere visited
and pollinated by humble-bees with a proboscis of exceptional length. In the Alps
and Pyrenees, however, B. mastrucatus Gers/. has been observed on the flowers
either collecting pollen, or perforating the upper sepal at the level of the nectary,
in order to steal nectar. Both in Central Germany and in Sweden, short-tongued
humble-bees have been observed as nectar-thieves, e.g. B. terrester Z. and B. alticola
Kriechb. In Sweden, B. jonellus K.(=B. scrimshiranus KX.) has been seen collecting
pollen.

Aurivillius and MacLeod observed (in Sweden and the Pyrenees respectively)
two very well-marked kinds of flower, between which transitional forms were found
here and there. They may be described as :—

(a) orthocera Knuth: spur almost straight, relatively stout, with blunter tip ;

(2) campylocera Anu/h: spur more or less distinctly curved upwards—some-
times into almost a semicircle—relatively slender, tapering towards the tip. In the
bud the spur of this form also is straight.

104. A. anthora L. (MacLeod, ‘ Pyreneénbl.’)—The pale yellow flowers open
wide. The upper sepal is produced above the mouth of the flower into a forwardly
curved beak. The two lateral sepals are concave internally and clothed with woolly
hairs. The two or three other floral leaves serve as alighting-places and platforms
for insects. The black stamens are sharply defined against the otherwise uniform
colour of the flower, which is thus made very conspicuous. The stamens and carpels
develop in the same order as in A. Napellus. But few stamens mature simul-
taneously, so that the flower remains for a long time in the male condition.

Visirors.—None were observed by MacLeod. Hoffer noticed—at Graz—
Bombus Gerstackeri ¥, sucking.

105. A. septentrionale Koell.— Axell states (‘Om anord. f. fanerog. vaxt.
befrukt.,’ Stockholm, 1869, p. 34) that this species also is protandrous.

A. Napellus—at least so far as observations have been made—and there is thus, at least apparently,

a real difference as to diet (according to Hoffer, Natw. Miscell., 1889, pp. 21-2). Hoffer (op. cit., .
PP. 23-5) noticed in Steiermark, however, that in places where A. Napellus is very common, while

A. Lycoctonum is exceedingly rare, B. Gerstackeri 9 also visits A. Napellus, so that the heterotrophy

observed by Dalla Torre on the Schlern in the Tyrol does not there exist; all castes (9, Y, and 3)

feeding chiefly on A. Napellus.

CALYCANTHACEAE 53

106. A. Cammarum L. (=A. Stoerkianum Reichb.).—

Vistrors.—Schneider (Mus. Aarsh., Tromsg, 1894) observed Bombus hortorum

visiting this flower in the gardens of Arctic Norway, but saw it nowhere else in
that region.

23. Paeonia Tourn.
Protogynous pollen flowers (?). The large red petals serve to attract insects.

107. P. officinalis L.—Kerner states that the flowers, which are open only
in the day-time, possess the odour of nightshade.

Vistrors.—I observed Bombus terrester L. ¥ in the Kiel Garden, vainly seeking
for nectar.

108. P. Moutan Sims (=P. arborea Don.).—

Visrtors.—Delpino says that this species, indigenous to China, is regularly

pollinated by beetles (Cetoniae), which lick the fleshy disks on the bases of the
ovaries.

Il. ORDER CALVCANTHACEAE LINDL.
24. Chimonanthus Lindl.

tog. C. fragrans Lindl. (=Calycanthus praecox Z.).—The greenish-white
strongly odorous flowers appear before the leaves, and, according to Hildebrand
(Bot. Ztg., Leipzig, xxvii, 1869), are protogynous. In the first condition the still
immature anthers are remote from the stigmas, and these may be pollinated by
insects coming from other flowers in the second stage. In this second stage the
dehisced anthers project beyond the stigmas, so that insect visitors are obliged to
touch them.

The flowers are also described as protogynous by Entleutner (‘Die sommer-
griinen Ziergehdlze von Siid-Tirol, Meran, 1892), who says that in the first stage of
anthesis the unripe stamens of the newly opened flower form a funnel from the middle
of which the carpels project, their stigmas being already mature. In the second
stage the anthers—which so far have been curved outwards towards the perianth—
apply themselves to and cover the carpels, projecting well beyond them. The
anthers now dehisce, and insect visitors have to push between them and the perianth
in order to get at the nectar secreted in the base of the flower. If such an insect
next alights upon a flower in the first stage, it necessarily effects pollination, for
the only way to the nectar is through the conical space between anthers and stigmas.

Vistrors.—Delpino observed (‘ Altri app. dicog. recent. oss.,’ Nuovo Giorn. bot.
ital., Firenze, ii, 1870, p. 59) a bee (Osmia) at Florence.

25. Calycanthus L.
to. C. floridus L.—The brown nectarless flowers of this shrub—a native
of north-east America—exhale a slight odour of strawberries. Delpino states (‘Ult.
oss., Atti Soc. ital. sc. nat., Milano, xvii, 1874, and ‘Altri app. dicog. recent.
oss.,’ Nuovo Giorn. bot. ital., Firenze, ii, 1870, p. 58) that they are protogynous, with
stigmas that soon wither.
Visttors.—Beetles (Cetoniae) appear to effect pollination.

54 ANGIOSPERMAE—DICOTYLEDONES

III. ORDER AZAGNOLIACEAE DC.
26. Wlicium L.

ii. I. religiosum L.—According to Delpino (‘Applicaz. d. teor. Darwin ai
fori ed agli insetti visit. d. fiori,’ Boll. Soc. Entom., Firenze, ii, 1870, p. 10), there
are small juicy glands like stigmatic papillae in the middle of the flower that probably
serve to attract beetles (Cetoniae), which while licking nectar effect pollination.

27. Magnolia L.

12. M. Yulan Desf.—This species is a native of China. The erect white
lily-like blossoms are odorous, and—according to Delpino (‘ Ult. Oss.,’ Atti Soc. ital.
sc. nat., Milano), they are protogynous bee flowers. In the first (female) stage
of anthesis, the bees that visit them are not able to climb up the smooth petals,
nor to free themselves from the short erect carpels occupying the middle of the
flower, and therefore remain prisoners till the second (male) stage, in which the
anthers dehisce. They are then able to leave the flower, and being dusted with
pollen, may transfer this to the stigmas of another flower which is still in the
first stage.

113. M. grandiflora L.—This species is indigenous to Florida. According
to Delpino (op. cit., pp. 233-5), the white odorous protogynous flowers are visited
and pollinated by beetles (Cetoniae). During the first stage of anthesis, these insects
find under the three inner petals, which arch over the carpels, a warm nectar-
containing shelter, that they only leave when the petals are shed at the time
of dehiscence of the anthers. Dusted with pollen they then betake themselves
to another flower in the first stage, the mature stigmas of which they necessarily
pollinate. Self-pollination is prevented by the pronounced protogyny.

Vistrors.—Cetonia aurata Z., and Oxythyrea funesta Poda (= C. stictica Z.).

IV. ORDER ANONACEAE JUSS.

28. Asimina Adans.

114. A. triloba Dunal—Delpino states (op. cit, p. 231) that the stamens
project in the centre of the pendulous protogynous flower as a hemispherical mass,
from the middle of which a few styles with their stigmas project. In the first
(female) stage of anthesis the three inner petals lie so close to the stamens that
insect visitors (flies) cannot suck the nectar secreted at the bases of the former
without touching the already mature stigmas. In the second (male) stage the
stigmas have dried up and the inner petals have raised themselves, so that the
anthers— now covered with pollen—are touched by insects on their way to the
nectar. Cross-pollination of the younger flowers is therefore effected by trans-
ference from the older ones.

Visitors.—Delpino observed the following seven Muscidae, which were deter-
mined by Rondani.—

1. Calliphora erythrocephala J7/g.; 2. Lucilia sericata A/e.; 3. Cyrtoneura

pascuorum M/e.; 4. C. stabulans Fa//.; 5. C. assimilis Fal/.; 6. Homalomyia
prostrata Rosst: 7. Megaglossa umbrarum J/g.

BERBERIDEAE 55

V. ORDER BERBERIDEAE VENT.

29. Berberis L.

Homogamous flowers with half-concealed nectar, arranged in dense racemes,
so that in spite of the relative smallness of the individual blossoms, they are very
conspicuous in the aggregate. The inner sides of both sepals and petals are yellow
in colour. Nectar is secreted by two fleshy swellings on the inner side of each
petal near its base, and owing to the concave shape of the corolla these are
tolerably well concealed. Many species possess sensitive stamens, which suddenly
move towards the pistil when they are touched at the base of their inner surface.

15. B. vulgaris L. (Sprengel, ‘Entd. Geh.,’ pp. 203-6; Herm. Miller,
‘Fertilisation,’ pp. go-3, ‘Weit. Beob.,’ I, p. 323, ‘Alpenblumen,’ p. 142; Kirchner,
‘Flora v. Stuttgart, p. 255; Knuth, ‘Grundriss d. Bliitenbiol.,’ ‘Bloemenbio]. Bij-
dragen.’)—Sprengel, who first described the structure of the flowers, believed them
to be self-pollinated :—‘ When a stamen that has been touched by an insect applies
itself to the pistil, it presses the inner side of its apex—which is covered with pollen—

i

- SA
i es cial

¢ az 1

' Fic. 21. Berberts vulgaris, L. (after Herm. Miller). 1. Flower seen rom above: a, the three
large inner sepals, which by their size and colour serve to attract insects ; 5, outer, and J’, inner petals;
¢, nectaries: d, filaments; e, stigma. 2. Position of the stamens after moving inwards towards the
pistil. 3. Petal with its two thick, fleshy, orange-red nectaries (c). 4-7. Stamens in various stages of
dehiscence, erection, and rotation of the anther-valves, seen from the outer side. 4. Stamen with the anther-
lobes still closed. 5. The outer wall of the right anther-lobe has become free below, so as to form
a valve, of which the free end is beginning to turn upwards with the pollen attached to it. 6. Both valves
have nearly completed their upward movement. 7. Both valves have turned in such a way that the
pollen-masses are directed towards the middle of the flower. 7 4. Inner side of a stamen in this stage.

closely to the stigma, and this being moist, a part of the pollen must adhere to it.
The stigma is thus gradually pollinated all round, and the ovary is fertilized.’

Hermann Miiller proved this view to be erroneous, and explained the flower
mechanism as an adaptation to cross-pollination.

The flowers are either horizontal or hang obliquely downwards (not vertically
downwards as described and figured by Sprengel). They are, therefore—as Hermann
Miiller emphasizes—not fully sheltered by their position from the entry of rain,
though a tolerable amount of protection is afforded by the three inner concave

56 ANGIOSPERMAE—DICOTYLEDONES

sepals, which are curved inwards at the tip rather more sharply, and by the six
similarly shaped petals, which completely enclose the stamens so long as they are
not disturbed, and cover the anthers with their ends. ;

The nectaries are in the form of two thick orange-coloured bodies on the base
of each petal, placed so close together as to touch each other. The stamens—before
they have been stimulated—lie so near the nectaries that the nectar collects in the
angle between the filaments and the carpels. The adhesive margin of the disk on
the top of the pistil functions—as Sprengel recognized—as a stigma. It matures
simultaneously with the stamens.

When an insect in quest of nectar comes into contact with the broad
sensitive bases of the filaments, these suddenly move inwards towards the pistil,
so that the head or proboscis of the insect comes between the dehisced anthers
and the stigmatic margin, which is at the same level. As a rule the insect then
leaves the flower, and goes to another one, effecting cross-pollination in the latter
if it touches the stigma with that side of its body which is dusted with pollen.
Should insect-visits fail self-pollination automatically takes place when the flowers
fade, for then the anthers themselves come into contact with the stigma. Self-
pollination, however, does not always seem to be effective, for many flowers fail
to set fruits.

According to Pfeffer, the movement of the stamens is caused by the flow ot
water to the stimulated part. Chauveaud (‘Mécanisme des mouvements provoqués
de Berberis,’ C.-R. Acad. sci., Paris, cxix, 1894, pp. 103-5) describes a specia.
tissue concerned in the movement, and consisting of elongated narrow cells almos'
fused together, with small intercellular spaces, especially at their ends. The
transverse walls of these cells are thin, their longitudinal ones, on the contrary
are thick with numerous scattered pits. These last render possible both a very
rapid interchange between the cells, and also a quick bending of this elastic tissue
It is covered by thin-walled cells, the contents of which constitute the active irritable
element. When in the resting condition, the protoplasm of each cell of the motile
tissue forms a thick band applied to the posterior cell-wall. On stimulation, this
suddenly becomes lax, spreads out, curves like a bow, and while its edges pull upon
the transverse walls its convex central part presses against the outer wall, which
becomes more strongly arched. Hence the cell becomes shorter and thicker. This
change in the motile tissue causes the filaments to bend inwards.

Visitors.—In accordance with the position of the nectar, these are mostly
insects with a medium or short proboscis. They all suck nectar, only a few bees
also collecting pollen. H. Miiller (H.M.) and myself (Kn.) have observed the
following.—

A. Coleoptera. (a) Coccinellidae: Coccinella conglobata Z. (= C. quattuor-
punctata Z.), skg. (H.M.); 2. C. septempunctata Z., freq., skg. (Kn.); 3. C. varia-
bilis //és/, nect-lkg. (H.M.). (6) Dermestidae: 4. Attagenus pellio Z., skg. (H. M.).
B. Diptera. (a) Muscidae: 5. Musca corvina /., skg. (H.M.); 6. Musca domestica
L., skg. (H.M., Kn.); 7. Onesia cognata AZg., ditto (H.M.); 8. O. floralis R.-D.,
ditto (H. M.); 9. O. sepulcralis AZg., ditto (H. M.). (6) Syrphrdae: 10. Ascia
podagrica #, freq., skg. (H.M.); 11. Eristalis arbustorum Z., skg. (H.M., Kn.) ;
12. E. nemorum Z., ditto (H. M.); 13. E. pertinax Scop., ditto (Kn.); 14. E. tenax

BERBERIDEAE 57

L., ditto (H.M., Kn.) ;_ 15. Helophilus floreus L., ditto (H.M.); 16. H. pendulus Z.,
ditto (H. M., Kn.); 17. Rhingia rostrata Z., ditto (H. M., Kn.); 18. Syrphus
balteatus Deg., ditto (Kn.). C. Hymenoptera. (a) Apidae: 19. Andrena albicans
Mill. 9, ditto (H.M., Kn.); 20. A. fulva Schr. 9, tolerably freq., skg. and po-cltg.
(H.M.); 21. A. fulvicrus &. 9, in large numbers, skg. (H.M.); 22. A. helvola Z. ,
skg. (H.M.); 23. K. praecox Scop. 9, ditto (H.M.); 24. A. trimmerana KX. g, ditto
(H.M.); 25. Apis mellifica Z. ¥, freq., skg. (H. M., Kn.); 26. Bombus pratorum Z.
9, skg. (H.M.); 27. B. terrester Z. 3, ditto, freq. (H. M., Kn.); 28. Halictus rubi-
cundus Chr. 9, skg. (H.M.). (4) Formicidae: 29. Lasius niger LZ. ¥, nect-lkg.
(H.M.). (c) Vespidae: 30. Vespa holsatica F. 8% skg. (H.M.); 31. V. rufa Z. ¥,
ditto (H. M.).

Hermann Miiller also observed—in the Alps—rq flies, 3 beetles, and 9 Lepido-
ptera. In the Tyrol, von Dalla Torre noticed the bees Andrena trimmerana X. 9,
and A. atriceps A. 8; to which Kohl adds—for the same region—the ruby-wasp
Ellampus aeneus /. and the true wasp Leionotus nigripes Pz., while Schletterer
records Andrena tibialis X., and A. trimmerana XK. Schiner—in Austria—observed
the hover-fly Criorhina berberina F.

Ricca (‘ Oss. sulla fecondaz. incroc. d. veget. alp. e subalp.’) observed humble-
bees and wasps; H. de Vries (Ned. Kruidk. Arch., Nijmegen, see note, p. 23, 1887)
records Apis mellifica Z. ¥, as a very common visitor in the Netherlands.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 7) were observed—
2 humble-bees, one of the short-tongued Apidae, and 2 hover-flies.

116. B. aquifolium Pursh (=Mahonia aquifolium MVu/t.).—This is an orna-
mental shrub indigenous to North America. Its flowers agree in structure with those
of the last species.

Visitors.—I have observed Syrphids (Eristalis tenax Z., Syrphus ribesii Z.,
Rhingia rostrata Z.), the honey-bee, Andrena albicans Jill. 9, and 2 humble-bees
(Bombus terrester Z. 9, and B. lapidarius Z. 9); also a few Muscids. All skg.
Schletterer noted at Pola 2 of the Apidae (1. Bombus terrester Z., ‘ from Christmas-
time till the end of January’; and 2. Xylocopa violacea Z.).

30. Epimedium L.

Protogynous flowers with concealed nectar. The blood-red petals serve to
attract insects. The cup-shaped nectaries—which make up a corona—are yellow,
and possess a short protuberance secreting nectar.

117. E. alpinum L. (Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II, p. 234; Loew,
‘Bliitenbiol. Floristik,’ p. 182; Knuth, ‘Bloemenbiol. Bijdragen.’)—The anthers
dehisce by valves, which come together above the already mature stigma, though
they cannot pollinate it, as the flowers are pendulous in this stage. The blossoms
become erect later on, and pollen can now fall on the stigma. Automatic self-
pollination is ensured by elongation of the pistil till it comes into contact with
the anthers.

According to Warnstorf, the pollen is yellow, ellipsoidal, covered with delicate
papillae, and on the average 43 » long, and 31 » broad.

Visttors.—On May 2, 1896, I had the opportunity of watching the honey-bee
visiting and pollinating this interesting flower in the Botanic Garden of the Kiel

58 ANGIOSPERMAE—DICOTYLEDONES

Ober-Realschule. It went in a methodical way all round the flower and sucked
the nectaries in succession. On alighting it touched the stigma—which projected
to a distance of 1 mm.—and dusted it with pollen brought from another flower.
While the bee was sucking nectar a fresh supply of pollen adhered to its ventral
surface.

118. E. pinnatum Fisch. (Loew, ‘ Bliitenbiol. Beitrage,’ I, p. 5..—This species—
indigenous to Persia and the Caucasus—is protogynous in the Berlin Botanic
Garden. It agrees in structure—though not in size and colour—with E. alpinum.

VisiTors.—Loew observed Osmia rufa, Z., skg.

119, E. macranthum Lindl.—According to Loew’s investigations (‘ Bltitenbiol.
Beitrage,’ I, p. 6) in the Berlin Botanic Garden, this flower has long thin spurs,
which conceal the nectar in such a way that bees with a long proboscis are the
most likely pollinators.

120. E. violaceum Morr. et Decne.—This is similar in structure to the last
species, with which it is perhaps identical (Loew, ‘ Bliitenbiol. Beitrage,’ I, p. 6).

121. E. rubrum Morr. (Loew, ‘ Bliitenbiol. Beitrige,’ I, p. 6.)—Like the two
preceding species this is indigenous to Japan. Its flowers have a relatively thick
spur, and like those of the other species are protogynous.

Visitors.—Loew observed Bombus agrorum, /. 9, skg., in the Berlin Botanic
Garden.

31. Podophyllum L.
Pollen flowers devoid of nectar and nectar-guides.

122. P. Emodi Wall. (Loew, ‘ Bliitenbiol. Beitrage,’ I, p. 8..—This Himalayan
species is probably pollinated by visitors which settle on the stigma, and then pass
on to the stamens in order to collect pollen. Cross-pollination must result from
a visit to a second flower. As the stigma projects above the anthers, automatic
self-pollination is excluded.

123. P. peltatum L. (Loew, ‘Bliitenbiol. Beitrage,’ I, p. 9.}—The number of
floral members of this North American species frequently varies. The stamens
project even in the bud.

32. Achlys DC.

124. A. triphylla DC.—According to Calloni (Arch. Sci. Phys., Genéve,
xvili, 1886), this plant possesses three kinds of flowers in every inflorescence: the
lower are barren, the middle are in part fertile, the upper ones are all fertile.

33. Akebia Decne.

125. A. quinata Des.—<According to Francke (‘Einige Beitr. z. Kennt. d.
Bestaubungseinricht. d. Pfl.,’ Inaug.-Diss., Freiburg i. Br., Halle, 1883), cross-
pollination is effected either by the wind or by insects. The female flowers are
developed long before the male ones.

NYMPHAEACEAE 59
VI. ORDER NFMPHAEACEAE DC.

Lirrrature.— Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins., pp. 20-1; Caspary,
in Engler and Prantl., ‘D. nat. Pflanzen-Fam.,’ III, 2, pp. 1-10, Leipzig, 1888.

The large floating flowers are protected from creeping animals by their aquatic
habitat, and are only accessible to flying insects. The inner side of the sepals
is coloured like the petals, so that both whorls are conspicuous. A more or less
distinct odour of honey also serves as a further attraction to insects.

34. Nymphaea L.

Homogamous or slightly protogynous pollen flowers with an odour of honey.
The stigma secretes a moisture which is, perhaps, licked by insects: according to
Jordan, however, there are flat nectaries in front of the stamens, in which case
the flowers should be piaced in class EC or C. The white inner surface of the

sepals, and of the numerous petals—which gradually pass into the stamens—makes
the flower conspicuous.

126, N. alba L. (Delpino, ‘ Alc. appunti d. geog. bot.,’ Boll. Soc. geogr. ital.,
Roma, v, 1869; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp- 21, 148, ‘ Weit. Beob.,’
p. 231; Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892; Kerner, ‘Nat. Hist. Pl.,’
Eng. Ed. 1, I, p. 218; Schulz, ‘ Beitrage, II, p.g; Watson, Justs bot. Jahresber.,
Leipzig, xii (1884), 1886, p. 682; Caspary, op. cit.\—The faintly odorous, large,
white flowers, which open in the morning and close towards evening, are homo-
gamous according to my observations. Kerner states that the stigmatic papillae
are mature at the beginning of anthesis, remaining receptive for several days. The
anthers begin to dehisce when the flower opens, or a day—rarely a few days—
later. The filaments bend into the form of a sickle, so as to bring the anthers
above the stigmas, which spread out into a plate-like surface, so that self-pollination
must result from falling of the pollen. Insect visitors may effect either cross- or
self-pollination, but they are few in number. The flowers are usually 10 cm. or even
more in diameter. In dried-up marshes, however, on the Island of Féhr, I found
flowers of only 5 cm. diameter, and these I have described (‘Flora d. nordfr. Ins.,’
p. 32) as the variety ¢errestris.

Visitors.—On the Island of Féhr I observed a diminutive fly (Notiphila cinerea
Faill.), freq. Heinsius—in Holland—noted a species of the same genus (Notiphila
nigricornis Svenh.). Schulz—for central Germany—has recorded various flies and
beetles. MacI.eod—in Flanders—observed a beetle (Domicia) (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 183).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 7), Apis, a humble-
bee, and Muscids have been observed.

35. Victoria Lindl.
127. Victoria Regia Lind].—The flowers, which may be as large as dinner-plates,
are first white, and then rose-coloured.
Visirors.—Delpino believes that the flowers are pollinated by Cetoniae and
Glaphyridae.

60 ANGIOSPERMAE—DICOTYLEDONES

36. Nuphar Sm.

Flowers homogamous, or slightly protogynous, with nectar partly or completely
concealed, secreted by the backs of the petals, and collected in the angle between
sepals and petals. The sepals, which are yellow on their inner surfaces, and the
other floral whorls, which are of the same colour, serve to attract insects.

128. N. luteum Sm. (Sprengel, ‘Entd. Geh., p. 273; Herm. Miiller, ‘ Ferti-
lisation,’ p. 93; Caspary, op. cit.; Schulz, ‘Beitrage,’ II, pp. ro-11; Kirchner,
‘Flora v. Stuttgart,’ p. 276; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp. 183-4; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 21, ‘Weit. Beob.,’ p. 226,
note 1; Axell, ‘Om anord. for fanerog. vaxt. befrukt.,’ p. 104; Warnstorf, Verh. bot.
Ver., Berlin, xxxvii, 1895.)—The egg-yolk yellow flowers are strongly odorous. They
are either homogamous or—according to Caspary and Schulz—protogynous, the
stigmas being fully mature at the commencement of anthesis, while the anthers
dehisce somewhat later in centripetal order. The stamens move towards the petals as
they ripen, so that automatic self-pollination is excluded.

Warnstorf also describes the flowers as protogynous, and states that when the
sepals are closed the stamens are pressed together round the ovary beneath the stigma,
but later on, when the anthers dehisce, they bend back, so that the pollen on their
inner surfaces is encountered by small insect visitors. The spinose pollen-grains
are large, yellow, and ellipsoidal, with an average length of 63 m, and an average
breadth of 37-5 ». The spines may be as much as 8-75 » long. Insects creeping
about on the flowers effect either cross- or self-pollination.

Visitors. — Hermann Miiller (H. M.) and myself (Kn.) have observed the
following. —

A. Coleoptera. (a) Chrysomelidae: 1. Donacia dentata Hoppe (H.M.); 2. D.
sparganii Afr. (Kn.); (6) Witedulidae: 3. Meligethes (Kn.). B. Diptera. AMusczdae :
4. Calliphora vomitoria Z.(Kn.); 5. Scatophaga sp. (Kn.); 6. Onesia floralis &.-D.
(H.M.). C. Neuroptera, 7. Phryganidae (Kn.). Flies and beetles have also been
recorded by Schulz.

Heinsius observed in Holland numerous flies (Notiphila nigricornis Svenh., and
Cleigastra sp.) (Bot. Jaarb. Dodonaea, Ghent, iv, 1892, pp. 61-3).

37- Euryale Salisb.

129. E. ferox Salisb.—Goebel (‘ Pflanzenbiol. Schilder.,’ II, 2, p. 363) observed
only submerged cleistogamous flowers on cultivated plants in the Marburg Botanic
Garden, while in the Munich Garden the flowers rose above the surface and unfolded
their bluish-violet petals.

VII. ORDER SARRACENIACEAE ENDL.
38. Sarracenia L.

130. Sarracenia purpurea L.—Hildebrand states (Ber. D. bot. Ges., Berlin,
i, 1883) that this species is homogamous, though cross-pollination is promoted,
for there are recurved hooks on the stigma, which compel insect visitors to make
their exit to one side of the stigmatic surface.

PAPAVERACEAE 61

VII. ORDER PAPAVERACEAE DC.
1. Sus-oRDER PAPAVEREAE.

Homogamous, slightly protogynous, rarely protandrous pollen flowers. In the
bud the inner floral whorls are protected by the strong calyx, composed of two
or three sepals, or cap-shaped. After discharging this function the calyx falls off
when the petals unfold. The large, usually vividly-coloured, petals make the flowers
conspicuous from a distance. Sometimes this is enhanced by the colour of the
stamens.

39. Papaver Tourn.

Pollen flowers with large brilliantly coloured petals.

131. P. alpinum L. (Herm. Miller, ‘Alpenblumen,’ pp. 142-3; Kerner, ‘ Nat.
Hist. Pl.’ Eng. Ed. 1, I, p. 120; Hoffmann, referred to in Darwin, ‘Cross- and
Self-Fertilisation,’ p. 331.)—The flowers sometimes smell of hawthorn, sometimes of
musk. In the Alps the petals are citron-yellow, with a paler sulphur-yellow or
greenish base; in Carniola they are dark yellow; in Lower Austria and Steiermark
they are white, usually with a yellow base.

The flowers are homogamous. The pistil is situated in the middle of the
deeply concave corolla, which is 30 to 35 mm. wide. The 5-8 radiate stigmas are
mature when the flowers open, and at the same time the anthers of some of the
outermost of the numerous stamens dehisce. Insects visiting the flowers for the sake
of the very abundant pollen may therefore effect either cross- or self-pollination.

The flowers remain half-closed in dull weather, and Kerner states that they
are open only in the forenoon. The inner stamens converge together above the
stigmas, and dust them with pollen. This automatic self-pollination has, however,
very little result, for—according to H. Hoffmann—garden plants invariably proved
self-sterile (Darwin, ‘Cross- and Self-Fertilisation,’ p. 331).

Visttors.—Hermann Miiller observed several flies in the Alps.

132. P. nudicaule L. (=P. radicatum Ro/d.).—Warming states that self-
pollination is almost inevitable in the sulphur-yellow or white pollen flowers of this
Arctic species, and it must be effective, for ripe fruits have frequently been observed,
though owing to the dearth of insects in the Northern region, cross-pollination hardly
ever occurs. Focke states, however, that cultivated plants are self-sterile. According
to Ekstam, the diameter of the faintly odorous flower is 20-40 mm. in Nova Zemlia.
Self-pollination is possible here even in the bud.

This species flowers in Spitzbergen, according to Andersson and Hesselman
(‘ Bidrag till Kanned. om Spetsbergens 0. Beeren Fil. Karlvaxtflora,’ pp. 41-2), from
the end of June to the end of August, regularly setting fruits. The corolla is
25-82 mm.—on an average 58 mm.—in diameter. Its colour is greenish-yellow,
often shading into white at the bases of the petals; but frequently it may be pure
sulphur-yellow, or sometimes perfectly white. The peduncle becomes curved during
rain, so that the opening of the flower faces downwards, ahd wet is kept out.
The pollen-grains are not injured by moisture, and germinate in distilled water
after three or four hours. Ekstam (‘Bliitenbiol. Beob. a. Spitzbergen,’ p. 21) found
the flowers to be homogamous or feebly protogynous-homogamous in Spitzbergen,

62 ANGIOSPERMAE—DICOTYLEDONES

with a strong tendency to autogamy, for both kinds of sexual organ mature in
the bud, and the stamens bend toward the stigmas when the flower opens. The
latter are 4-7 in number in Greenland (Abromeit, ‘Bot. Ergeb. v. Drygalski’s
Groénlands exped.,’ p. 29).

Visitors.—In Nova Zemlia flies were observed by Ekstam, who also noticed
a small po-dvg. fly on the flowers in Spitzbergen, at a height of 1000’ above the
sea-level.

' Alfken saw at Bremen the following pollen-collecting Apidae on garden
plants.—

1. Andrena albicans Jf#ll.9; 2. A. nigro-aenea K.9; 3. A. nitida Fourcr. 9;
4. A. parvula A. 9; 5. Osmia rufa Z. 9. All po-cltg.

133. P. Rhoeas L. (Heim. Miiller, ‘Fertilisation,’ 93-4, ‘ Weit. Beob.,’ I, 323;
Hoffmann, ‘ Bot. Ztg.,’ Leipzig, xxxvi, 1878, p. 290; Beyer, ‘D. spont. Bewegungen
d. Staubgefasse u. Stempel’; Kirchner, ‘Flora v. Stuttgart,’ p. 277; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ pp. 22, 148.)—The petals are scarlet, with a black patch
at the base. Even in the bud the numerous stamens are mature, so that the pollen-
covered anthers come into contact with the lower parts of the flattened but already
mature stigmas, and dust them with pollen. Hoffmann states, however, that this
inevitable automatic self-pollination is ineffective. After the flowers have opened
either cross- or self-pollination may be brought about by insect-visits. Warnstorf
(Verh. bot. Ver., Berlin, xxxvii, 1895) describes the pollen-grains as grey-green, and
spherical or almost spherical when examined in water, very finely granular, with
an average of 37-5 pw.

Visitors. — Hermann Miller (H. H.) in Westphalia, and myself (Kn.) in
Schleswig-Holstein, have observed the following.—

A. Coleoptera. (a) Nitdulidae: 1. Meligethes, very freq., po-dvg. (H. M.);
(5) Oedemerrdae: 2. Oedemera virescens L., po-dvg. (H. M., Thuringia). (c) Scara-
baeidae: 3. Oxythyrea funesta Poda, very freq., dvg. the flowers(H. M.). B. Diptera.
(a) Empidae: 4. Empis livida Z.(H.M.). (6) Afuscidae: 5. Ulidia erypthrophthalma
Mg. (H.M., Thuringia). (c) Syrphidae: 6. Cheilosia, po-dvg. (H. M.); 7. Syrphus
ribesii Z., po-dvg. (Kn.); 8. S. umbellatarum /., po-dvg. (Kn.). C. Hymenoptera.
Apidae: g. Andrena dorsata #. 9, freq., po-cltg. (H.M.); 10. A. fulvicrus K. 9, freq.,
po-cltg. (H. M.); 11. Apis mellifica Z. ¥ (Kn.); 12. Bombus terrester Z. (Kn.); 13.
B. lapidarius Z. 9, and 14. Halictus cylindricus A. 9; all po-cltg.(H.M.); 15. H.
flavipes J. 9, freq., po-cltg. (H.M.); 16. H. leucopus &. 9, po-cltg. (H. M., Thuringia);
17. H. longulus Sm. 9, po-cltg. (H. M.): 18. H. maculatus Sm. 9 (H. M.); 19. H.
sexnotatus A’. 9, very freq., po-cltg. (H.M.); 20. H. smeathmanellus X. 9, po-cltg.
(H.M., Thuringia). D.Orthoptera. 21. Forficula auricularia 2. (H. M.). Friese
noted—in Mecklenburg—Osmia papaveris Z/r., occasional; Schletterer—at Pola—
Eucera longicornis, Z.; MacLeod—in the Pyrenees—Bombus terrester, L. 4, po-
cltg., also—in Flanders— 3 hover-flies (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp. 184-5).

134. P. Argemone L. (Herm. Miiller, ‘Fertilisation, p. 94; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 185-6; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’
pp. 22, 148; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The red petals are
blotched with black at their bases. The flowers agree in structure with those of
the last species, except that the anthers come into contact with a smaller part

of the stigmas. Warnstorf describes them as pseudo-cleistogamous, because the

PAPAVERACEAE 63

sky-blue anthers dehisce before they open, and pollinate the already mature stigmas.
The bluish, spherical pollen-grains are opaque owing to the presence of minute
wart-like projections. Their average diameter is 50 pe.

Vistrors.—I observed—at Kiel—a hover-fly, Platycheirus podagratus Le//.,
po-dvg.

135. P. somniferum L. (Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II, p. 190;
Kirchner, ‘Flora v. Stuttgart,’ p. 278; Knuth, ‘BI. u. Insekt. a.d. nordfr. Ins.,
pp. 22, 148.)—The petals are either carmine to violet in colour, often blackish at
the base, or else white with a lilac base. The flowers agree in structure with those
of P. Argemone and P. Rhoeas, but in the bud the anthers project over the stigma-
lobes—which are still folded down—so that before anthesis not only the lower but
all the stigmatic papillae are thickly covered with pollen. This automatic self-
pollination is partly effective. Owing to the size of the flower, however, insect-visits
are very numerous, so that in favourable weather cross-pollination is assured. The
Diptera (Syrphidae) which I observed visiting the flower almost always alighted upon
the large upper-surface of the pistil with its radiating stigmas—from which the
stamens move away when the flower opens—and then go to the anthers, so that
cross-pollination must be effected by a second visit. The humble-bees observed by
Me as visitors only now and then touched the stigmas, for they almost always
alighted upon the crowded stamens, and then wandered about among these collecting
pollen. Warnstorf (Ver. bot. Ver., Berlin, xxxviii, 1896) describes the whitish pollen-
grains as ellipsoidal, with a length of about 44 , and a breadth of 28 yp.

Visitors.—The following insects have been observed by Buddeberg (Budd.) in
Nassau, and by myself (Kn.) in Schleswig-Holstein.—

A. Coleoptera. (a) Scarabaeidae: 1. Cetonia stictica Z., devg. the flowers
(H.M.). (4) Mitdulidae: 2. Meligethes sp. (Kn.). B. Diptera. Syrphidae: 3.
Eristalis aeneus Scop. (H. M.); 4. E. arbustorum Z. (H.M., Kn.); 5. E. tenax Z.
(Kn.); 6. Platycheirus peltatus 47g. (Kn.); and 7. Syrphus sp. (Kn.), all po-dvg.
C. Hymenoptera. Apidae: 8. Apis mellifica Z. ¥, freq. (Kn.); 9. Bombus terrester
Z., freq. (Kn.); 10. Eriades campanularum A. 9 (H.M.); 11. E. truncorum Z. 9
(H.M.); 12. Halictus cylindricus /. 9 (H. M.); 13. H. leucopus A. 9 (H. M.); all
po-cltg.

136. P. dubium L. (Herm. Miller, ‘Fertilisation,’ p. 94.)— Automatic self-
pollination is rendered difficult in this species because the anthers stand a few
millimetres below the stigma; it can therefore only take place in flowers which
are bent downwards. Hermann Miiller is of opinion that the very great rarity of
this plant in some districts is perhaps due to the impossibility of autogamy. Warnstorf
(Verh. bot. Ver., Berlin, xxxvii, 1895) describes the pollen as yellow, spherical to
ellipsoidal when examined in water, marked with several longitudinal furrows, and
31-37 » in diameter.

Visirors.—MacLeod—in Flanders—observed small flies (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 186).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 8) there have been
noted 3 Muscidae, a hover-fly, and Meligethes.

137. P. argemonoides L.—Hildebrand states that this species is fertile when

self-pollinated.

64 ANGIOSPERMAE—DICOTYLEDONES

138. P. hybridum L.—According to Hoffmann, this species— at least in
gardens—possesses cleistogamous flowers.

Vistrors.—Schletterer—at Pola—observed the bee Halictus calceatus Scop.

139. P. bracteatum'Lindl.—

Vistrors.—Loew noticed Apis mellifica Z.¥ in the Berlin Botanic Garden,
po-cltg., and creeping over the stigma.

140. P. burseri Crantz.—

Visirors.—In the Berlin Botanic Garden, Loew noticed a long-tongued bee
(Osmia rufa Z. 9, po-cltg.).

40. Glaucium Tourn.

Homogamous or slightly protogynous pollen flowers, devoid of odour, and red
or yellow in colour.

141. G. flavum Crantz (=G. luteum Sm.). (Kerner, ‘Nat. Hist. Pl.’ Eng.
Ed. 1, II, p. 310; Kirchner, ‘ Beitrage,’ p. 19; Knuth, ‘ Bliitenbiol. Herbstbeob.’)—
The large citron-yellow petals fall off on the second day of flowering. The
stigma—which Kerner says develops somewhat earlier than the anthers—projects
a little beyond the stamens, so that automatic self-pollination is prevented.

Visirors.—On cultivated plants at Kiel I observed as pollinating agents numerous
individuals of a hover-fly species (Syrphus ribesii Z.), po-dvg., and also several butter-
flies (Vanessa Io Z., and Rhodocera rhamni Z.), vainly searching for nectar.
Kirchner—at Hohenheim—noticed the honey-bee and Thrips; while Loew—at
Bellagio—saw Xylocopa violacea L. 9, po-cltg. In Dumfriesshire (Scott-Elliot, ‘ Flora
of Dumfriesshire,’ p. 9) there have been recorded 2 Muscidae, a hover-fly, and
Meligethes.

142. G. corniculatum Curt. (= G. phoeniceum Crantz). (Kerner, op. cit.,
p. 213; Knuth, op. cit.)—The flower is bright red, with a black patch on the base
of each petal. Its structure is the same as in the last species.

VisiTors.—In the case of cultivated plants growing beside those of the last
species, the visitors to the two were identical.

41. Chelidonium L.
Homogamous pollen flowers with yellow petals.

143. C. majus L. (Sprengel, ‘Entd. Geh.,’ p. 271; Herm. Miiller, ‘ Fertilisation,’
p. 94, ‘ Weit. Beob.,’ I, p. 323; Hildebrand, ‘Die Geschlechtsvert. b. d. Pfl.,’ p. 60;
Kirchner, ‘ Flora v. Stuttgart, p. 279; Knuth, ‘ Bliiten-
biol. Herbstbeob.,’ ‘Bloemenbiol. Bijdragen’; Warns-
torf, Verh. bot. Ver., Berlin, xxxvii, 1896.)—The flowers
open in sunny weather. The anthers dehisce laterally,
and the stigma matures simultaneously. As the latter
projects somewhat beyond the stamens, cross-pollination
is effected by insects coming from other flowers of the
FIG. 22. Chelidontum majus, L. same species, and alighting in the centre of the flower,

fter Hildebrand). The stigma pro- en cs
ee beyand the eC ee P while insects settling on the edge of the blossom may

PAPAVERACEAE 65
effect either kind of pollination. During dull weather the flowers remain closed for
a long time, in which case the anthers dehisce in the bud and effect self-pollination.
Warnstorf describes these flowers as vatying from slightly protogynous to homo-
gamous, and from this to protandrous. The pollen-grains are of a beautiful yellow
colour, rounded, finely tuberculated, and measuring up to 37 p in diameter.

Vistrors.—These are chiefly insects vainly seeking for nectar, and next to
them come pollen-collecting bees and pollen-devouring flies. The larger bees
(humble-bees and the honey-bee) alight as a rule on the middle of the flower, thus
effecting cross-pollination, while smaller forms (species of Halictus) usually settle
on the edges of the blossoms, and only occasionally touch the stigma, in which case
either kind of pollination is equally probable. Hover-flies behave in the same way.

The following visitors have been determined by Hermann Miller (H. M.) in
Westphalia, and by myself (Kn.) in Schleswig-Holstein.—

A. Coleoptera. (a) Chrysomelidae: 1. Cryptocephalus sericeus Z., po-dvg.
(Kn.). (0) Nitidulidae: 2. Meligethes, po-dvg. (H. M.). B. Diptera. (a) Empidae :
3- Empis livida Z., vainly searching for nectar, or perhaps boring (H. M.). (4) Syr-
phidae: 4.sAscia podagrica F., po-dvg. (H. M.); 5. Eristalis arbustorum Z., po-dvg.
(H.M.); 6. E. nemorum Z., po-dvg. (Kn.); 7. E. pertinax Scop., po-dvg. (Kn.);
8. Helophilus pendulus Z., po-dvg. (Kn.); 9. Melanostoma mellina Z., po-dvg. (Kn:);
10. Melithreptus taeniatus Mg., po-dvg. (Kn.); 11. Rhingia rostrata Z., first vainly
searching for nectar, then po-dvg. (H.M.); 12. Syritta pipiens Z.. po-dvg. (H. M.);
13. Syrphus balteatus Deg., po-dvg. (H. M,, Kn.); 14. S. ribesii Z., po-dvg. (H. M.,
Kn.). C. Hymenoptera. (a) Apidae: 15. Anthophora pilipes F. 9, po-cltg. (Kn.) ;
16. Apis mellifica Z. ¥, po-cltg. (H.M., Kn.): 17. Bombus agrorum F. ¥, po-cltz.
(H. M., Kn.); 18. B. hortorum Z. ¥, po-cltg. (Kn.); 19. B. lapidarius Z. ¥, po-cltg.
(Kn.); 20. B. pratorum Z. ¥, po-clig. (H. M.); 21. B. rajellus X. ¥, po-cltg. (H. M.);
22. B. terrester Z., po-cltg. (Kn.). Herm. Miiller noticed that humble-bees settle on
the middle of the flower, sweep pollen very rapidly from the anthers with the tarsal-
brushes of the first two pairs of legs, and immediately transfer it to the ‘baskets’ of
the hind-legs, completing the operation in from two to three seconds, and then
at once flying away to another flower to treat it in the same fashion, thus effecting
cross-pollination. 23. Halictus cylindricus /. 9 (H.M.); 24. H. sexnotatus A. 9
(H.M.); 25. H. sexstrigatus Schenck 9 (H.M.); 26. H. zonulus Sm. 9 (H. M.); these
four species of Halictus fly to the anthers to collect pollen, occasionally touching
the stigma.

Loew observed—in the Berlin Botanic Garden—a hover-fly, Syrphus balteatus
Deg., po-dvg.; Alfken —at Bremen — 3 po-cltg. Apidae, Bombus lucorum Z. ¥,
Andrena nitida /ourc. 9, and A. nigro-aenea X. 9, the last two fiving laboriously
with disarranged wings, and carrying very heavy loads of pollen, destined to store
cells in the case of the last-named species; Hoffer records from Steiermark, Bombus
agrorum /. ¥. carrying huge loads of pollen, and Bombus terrester Z. § freq.;
MacLeod noted in Flanders, Apis, 3 humble-bees, 2 species of Halictus, 5 hover-
flies, and a Muscid (B. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 186-7).

42. Eschscholtzia Cham.

Homogamous pollen flowers, usually yellow in colour.

144. E. californica Cham. (F. Miller, Bot. Ztg., Leipzig, xxvi, 1868; Darwin,
op. cit., xxvii, 1869 ; Hildebrand, Jahrb. wiss. Bot., Leipzig, vii, 1869-70; Herm. Miller,
‘Fertilisation,’ pp. 94—5, ‘ Weit. Beob.,’ I, p. 323; Knuth, ‘ Bliitenbiol. Herbstbeob.’)—
The filiform stigmas are at first closely surrounded by the crowded stamens, but

DAVIS. I F

66 ANGIOSPERMAE—DICOTYLEDONES

in a later stage of anthesis, the latter bend out towards the somewhat remote petals,
after which the outer set of anthers dehisce, while those of the inner stamens are
still unopened. The centrally placed stigma—which is now mature—cannot therefore
receive pollen from the same flower, but may be cross-pollinated if insects that
have visited other blossoms alight upon it. At a still later stage the anthers of
the inner stamens dehisce so that self-pollination may take place if insect-visits have
failed. Fritz Miiller states, however, that this is ineffective in South Brazil, though
in England—according to Darwin—the contrary is true. It is remarkable that
Brazilian plants sent by Fritz Miller to Charles Darwin were to some extent self-
fertile. Hildebrand found the plant to be almost self-sterile in Germany.

VisiTors.—In sunny weather I observed the bright yellow flowers to be visited
by numerous individuals of a species of hover-fly (Syrphus ribesii Z.), which alighted
in an erratic fashion sometimes upon the stigma, sometimes upon the stamens or
petals, so that at times they effected cross-pollination, at other times self-pollination.
I frequently saw five or six of these flies in a single flower, and they remained there
so persistently that I could pluck the flower and examine it with a lens, while the
visitors remained undisturbed and continued to devour pollen. They were thickly
covered with this on the head, and especially on the upper- and under-surfaces
of the thorax. Herm. Miller observed another hover-fly (Helophilus floreus Z.)
visiting the flowers.

43. Sanguinaria L.

Pollen flowers devoid of nectar and nectar-guides.

145. S. canadensis L. (Loew, ‘Bliitenbiol. Beitrage,’ I, pp. 9, ro.)—In this
North American plant the inner stamens project a little beyond the stigma, while
the outer ones are somewhat shorter.

Visrtrors.—Loew observed in the Berlin Botanic Garden, Apis and Bombus
terrester Z. 9, both po-cltg.

2. Susporper FumarigazE DC.

Literature.—Hildebrand, Jahrb. wiss. Bot., Leipzig, 1869; Knuth, ‘Bl. u.
Insekt. a. d. nordfr. Ins.,’ p. 23.—The flower mechanism of the species of this family
has been thoroughly and carefully investigated by Hildebrand, and Hermann Miiller
has determined the visitors. The following remarks are chiefly based upon the
researches of these two investigators.

The Fumarieae bear homogamous bee flowers of peculiar structure. Being
as a rule large and brightly coloured, often aggregated into racemes, and not
infrequently exhaling a more or less strong odour of honey, these naturally attract
insects. The nectar is secreted and concealed in spurs or pouches of the petals,
and there are either two such nectaries (Diclytra and Adlumia), or only one (Cory-
dalis, Fumaria). The two inner petals are fused at the tip, and so form a hood-like
sheath, which encloses anthers and stigmas. This hood is pressed downwards or °
to the side by nectar-seeking bees, but when the pressure is removed it usually
springs back by elasticity to its original position. Bees dust themselves with pollen
in the younger flowers, and transfer some of it to the stigmas of older flowers that

PAPAVERACEAE 67

have already been robbed of their pollen. As they visit the flowers of an inflor-
escence in a systematic manner, working from below upwards, they regularly effect
cross-pollination, either geitonogamy or—when passing from one plant to another—
xenogamy.

44. Hypecoum L.

Protandrous pollen flowers, usually yellow in colour, with large inner petals
which conceal the pollen in a pouch. As regards flower-pollination they resemble
the Papavereae. Kerner (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, pp. 178, 182) regards them
as nectar flowers.

146. H. pendulum L.—F. Hildebrand (Jahrb. wiss. Bot., Leipzig, vii, 1869-70)
states that the two inner of the four petals are provided with a pair of lateral ligules,
while the middle part (the actua] petal) alters greatly in form during the course
of development. The anthers dehisce extrorsely before the flower opens, so that
the pollen is received by the inner petals, which at this stage are folded inwards
in the form of a spoon. Next the empty anthers shrivel and contract, and the
spoon-like petals form a case which completely encloses the pollen. When the flower
expands the pollen-pouches open if pressed from above, so that an insect alighting
upon one of them must dust its under-surface with pollen.

During this first stage the stigma is still not quite mature, but after a time
the style elongates so as to project beyond the pouches of the petals, and a pollen-
covered insect alighting upon the flower must necessarily dust the stigmatic papillae,
which are now well developed. Failing insect-visits, the tips and edges of the pollen-
pouches curve somewhat outwards, so that the pollen—if it has not already been
removed by insects—is brought into such a position that it can readily fall on the
stigma if the plant is shaken, or may be carried to it by the wind. Kerner says
that the flowers do not open in bad weather, and that in the closed condition
automatic self-pollination (pseudo-cleistogamy) takes place.

147. H. procumbens L.—Kerner states that the flowers of this species also
remain closed in unfavourable weather..

148. H. grandiflorum L.—Hildebrand asserts that this species is almost
always infertile when the stigma is dusted with pollen from the same flower, or
another flower of the same plant.

45- Diclytra DC.

Homogamous bee flowers, the nectar of which is usually secreted and concealed
in two pouches, one at the base of each semi-cordate outer petal.

149. D. spectabilis DC. (Hildebrand, Jahrb. wiss. Bot., Leipzig, vii, 1869-70 ;
Herm. Miller, ‘Fertilisation,’ pp. 95-6; Knuth, ‘Bloemenbiol. Bijdragen.’)—The
peduncles are so thin and flexible that the flowers always hang perpendicularly
downwards by their own weight. The two lancet-shaped sepals fall off very early.
Each of the semi-cordate petals encloses three stamens, the filaments of which
follow the bend of the outer edge of the petal, and together form a groove on their
outer side which leads to the nectar.

The end of this groove opens exactly where an aperture is left between the
outer petals and the winged bases of the inner ones, i.e. the two grooves begin

F 2

68 ANGIOSPERMAE—DICOTYLEDONES

at the only two places where there are openings into the interior of the flower.
The parts of the filaments projecting from the flower—together with the anthers—lie
close together, surrounding the stiff style with its stigma, and they themselves are
covered with a hood formed by the fusion of the tips of the two inner petals.

Some considerable time before the flower opens the anthers dehisce, discharging
their pollen upon the large lobed stigma—which is already mature—and there the
pollen lies surrounded by the hood. Automatic self-pollination would therefore be
inevitable, and the pollen could never escape from. its close envelope, were not
crossing effected by insects, which in this case are exclusively bees. When a bee
hangs on to the flower to suck nectar, it must push aside the hood and the flexible
stamens this encloses with the under-side of its body, and—with the hairs on its
ventral surface—sweep off the pollen from the stigma, which owing to the rigidity
of the style is not deflected. As soon as the bee leaves, the hood returns to its
former position, and again ensheaths the anthers and stigma. There being two

1

to
w
he

Fic. 23. Déclytra spectabili's, DC. (after Hildebrand). (1) Flower, natural size. (2) The same after
removal of half an outer petal: the hood is pressed aside: the dotted line beginning at ¢, indicates the
path of the insect's proboscis. (3) The sexual organs of a bud. (4) The pistil and the two middle stamens

of a bud before dehiscence of the anthers.

nectaries, these events happen twice during a visit to each blossom. In younger
flowers therefore the pollen on the stigma is removed by the bee, and carried to
the stigma of an older flower, which has already been robbed of its own pollen.

Visirors.—Since the curved channels of Diclytra spectabilis are 18-20 mm.
long, there are only two of our native bees which can reach the nectar in the
legitimate way, i.e. Bombus hortorum Z. 9 (proboscis 20-21 mm. long), and
Anthophora pilipes #. 9 (proboscis 19-20 mm.). These two bees are in fact the
normal visitors and pollinators of this species. Hermann Miiller observed both of
them sucking the flowers in Westphalia, and I myself saw B. hortorum doing so
in the Kiel Botanic Garden. Bees with a shorter proboscis steal the nectar by
biting through the flower. Bombus terrester Z. 9—the proboscis of which is
7-9 mm. long—climbs to the upper side of the flower and bites through the petals
in the neighbourhood of the nectaries, afterwards extracting the nectar through
the hole. Hermann Miiller also observed B. pratorum /. 9 (length of proboscis
11-12 mm.), and L. rajellus A. 9 (12-13 mm.), behaving in the same way, while

PAPAVERACEAE 69

1Osmia rufa Z. 9 (9 mm.), Megachile centuncularis Z. § (6-7 mm.), and Apis mellifica
J..% (6 mm.), steal nectar through the holes bitten by humble-bees. Owing to the
smoothness of the petals the honey-bee loses a great deal of time while stealing nectar.

150. D. eximia DC.—The flower mechanism essentially agrees with that of
the last species, but there is less room for lateral movement of the hood, and the
path to the honey is shorter. (Cf. F. Hildebrand, Jahrb. wiss. Bot., Leipzig, vii,
1869-79, pp. 434-6.)

151. D. cucullaria DC.—The flower possesses two long-spurred outer petals.
Nectar is secreted by two horn-shaped processes of the middle filaments, which
project into the spurs. (Cf. Hildebrand, op. cit., pp. 436, 637.)

46. Adlumia Rafin.

The flower mechanism resembles that of the last genus, but there is a larger
amount of fusion between the various parts. The pouches of the outer petals are
smaller, and the edges of these petals are united together below.

152. A. cirrhosa Rafin.—Only the upper part of the hood is well developed
and free, while the lower part is fused with the outer petals. The six filaments
cohere below into a ventricose tube. (Cf. Hildebrand, op. cit., pp. 437-9.)

47. Corydalis DC.

Homogamous bee flowers. The hood can only be moved downwards. The
upper of the two outer petals is prolonged posteriorly into a nectariferous spur.

153. C. cava Schweigg. et Kort. (Hildebrand, op. cit.; Herm. Miiller, ‘ Fer-
tilisation, pp. 97-8; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, IZ; Knuth, ‘Bloemenbiol-
Bijdragen.’)—The spur of the rose-red or white homogamous humble-bee flowers—
which exhale an odour of honey—projects at the back for about 12 mm. beyond
the peduncle. A common process of the upper filaments—thickened terminally—
extends into it as far as the place where it bends. This process secretes nectar,
which is concealed in the downwardly curved end of the spur. The two inner
lateral petals are fused together at their tips, and cohere at their bases with the
two outer petals. The hood thus formed encloses the anthers and stigma. The
latter is large and lobed, and its surface is granular. It is situated on a rigid style,
which cannot be bent down, and before the flower opens is covered with the pollen
of all the surrounding stamens, which readily adheres to its granular surface. The
empty anthers appear only as small appendages of the filaments beneath the stigma.
Insects—long-tongued bees—when probing for nectar, must insert their proboscis
between the hood and the upper spurred petal. In doing so they press down the
hood and—in younger flowers—dust their under-surface with the pollen which has
been shed on the stigma, and transfer it to the stigmas of older flowers that have
already been robbed of their pollen, thus effecting crossing. When the weight of
the insect visitor is removed the hood springs up by elasticity to its original position,
and again ensheaths the stigma. ;

Although the stigma—while within the hood—is surrounded by the pollen
of the same flower, automatic self-pollination does not take place. Hildebrand has

70 ANGIOSPERMAE—DICOTYLEDONES

proved by numerous experiments that the flowers are quite infertile when pollinated
with their own pollen, and very largely so when geitonogamously pollinated.
Xenogamy is necessary for complete fertility.

Visitors.—As, according to Hermann Miiller, the spur extends backwards for
12 mm. from its point of attachment to the peduncle, and the nectar fills only
its last 4-5 mm., there is—among the bees which are active when this plant is
in flower—only one species (Anthophora pilipes #. 9 and 6, with a proboscis
19-21 mm. long) that can reach the nectar in the legitimate way. This insect—
according to Hermann Miiller’s observations at Lippstadt and my own at Kiel—visits
the flowers so diligently, and in such numbers, that probably none of them remain
unfertilized.

Bombus terrester Z. here makes its appearance as a nectar-thief. It would
be just able to reach the beginning of the nectar with its proboscis (7-9 mm. long),

Fic. 24. Corydalis cava, Schweigg. et Kori. (from nature). A. Flower with closed hood, seen from
the side: the spur has been perforated by Bombus terrester. &. Flower seen from the front. C. Flower
with hood turned down so that the pollen-covered stigma is visible: seen from the side; the nectary
is seen shining through the spur (enlarged). 2. Lobed stigma covered with pollen: beneath it are
the withered empty anthers (more highly magnified).

and to get some of it, but instead of doing so it perforates the upper side of the
spur in the neighbourhood of the nectar, either at the point ef curvature or nearer
the base. It then inserts its proboscis into the hole and sucks the sweet fluid.
(Cf. Fig. 24 4.) By using the holes made by Bombus terrester, other short-tongued
bees are also able to get the nectar, e.g. the honey-bee (proboscis 6 mm. long),
as well as species of Andrena, Sphecodes, and Nomada. The honey-bee certainly
sometimes attempts to reach the nectar legitimately, but always fails, owing to
the shortness of its proboscis. In making these occasional vain attempts it—and
also Anthophora pilipes—effects cross-pollination. Crossing is also brought about
if—abandoning the search for nectar—such a bee remains in the flowers to collect
pollen. Hermann Miiller also observed a few hover-flies (Bombylius major Z., and
B. discolor A/ikan, with proboscis of 10 and 11-12 mm. respectively), hovering after

.

PAPAVERACEAE q1

the manner of their kind and sucking nectar legitimately, but the proboscis in these
forms is much too slender to be able to effect pollination.

Hoffer observed—in Steiermark—Bombus mastrucatus Gers?. 9, biting through
the spur, and stealing nectar.

154. C. intermedia P. M. E.(=C. fabacea Pers.). (Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896.)—The flowers are of a dirty purple colour, and arranged
in inconspicuous three- or four-flowered racemes. Their mechanism is the same
as that of the last species. The spur is about 9 mm. long. As insect-visits are
few when the plant is in bloom, and the flowers are frequently broken, there could
be only a small production of fruit and seeds were it entirely dependent on cross-
pollination. But—so far as Warnstorf was able to observe—each flower sets well-
developed fruit, so that the plant must be autogamous should insect-visits fail.
Kerner confirms the self-fertility of this species. The pollen is whitish, or yellowish
in quantity; the grains are spheroido-tetrahedral, and 37 pu in diameter.

155. C. solida Sm. (=C. digitata Pers.). (Hildebrand, op. cit.; Herm. Miiller,
‘Fertilisation,’ pp. 98-9; Kirchner, ‘Flora v. Stuttgart,’ p. 280; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 187-8; Warnstorf, Verh. bot. Ver., Berlin,
xxxvilil, 1896; Knuth, ‘Bloemenbiol. Bijdragen.’)—The flowers agree in structure
with those of C. cava, but the spur is sometimes rather shorter.

The bright violet flowers are arranged in many-flowered racemes, being therefore
very conspicuous. The two lateral petals which form the hood are—according to
Warnstorf—beset with large forked papillae at the place of fusion, and also along
the whole keel. These increase the friction, and are supposed to prevent the insect’s
feet from slipping so much as they would otherwise do. Hildebrand also states that
this species is self-sterile.

Visirors.—The normal visitor and pollinator is again Anthophora pilipes; here
too Bombus terrester and Apis mellifica get at the nectar by making perforations ;
while—as in C. cava—the species of Bombylius suck nectar legitimately without
benefiting the flower. Loew also observed—in the Berlin Botanic Garden—Antho-
phora and Apis.

156. C. nobilis Pers—Hildebrand says that this species resembles C. cava
as regards the mode of pollination.

157. C. capnoides Pers.—According to Hildebrand, this has a similar flower
mechanism, but the form of the outer petals is somewhat different, especially that
of the upper petal, the spur of which is curved inwards upon the peduncle. Kerner
asserts the species to be self-fertile.

Visirors.—Loew—in the Berlin Botanic Garden—noticed the honey-bee, skg.

158. C. ochroleuca Koch. (Hildebrand, op. cit.\—This species is distinguished
from the preceding ones in which the hood springs up after the weight of the
visiting insect has been removed, by the fact that when this structure has once
been displaced it does not return to its original position, but remains inclined down-
wards, while the stamens and pistil—just as in Medicago sativa—give a jerk like
a liberated spring, and conceal themselves in an excavation of the upper petal.
It follows that each flower can only once be visited so as to affect the stamens
and stigma. The visiting bee dusts its under-surface with pollen from the stigma,

72 ANGIOSPERMAE—DICOTYLEDONES

and if it has come from another flower of the same species effects cross-pollination.
This plant—according to Hildebrand’s experiments—is self-fertile. Kerner confirms
the fact of autogamy in the absence of insect-visits.

159. C. lutea DC. (Hildebrand, op. cit.; Herm. Miller, ‘Fertilisation, p. 99,
‘Weit. Beob.,’ I, p. 324.)—The flower mechanism essentially agrees with that of
the last species.

Vistvors.—Herm. Miiller—at Lippstadt—observed Bombus agrorum /’. 9, skg-
legitimately, and he also saw the following bees at Jena—

1. Anthophora aestivalis Pz. 9 and 4, skg.; 2. Bombus confusus Schenck 9, skg.;
3. B. lapidarius Z. 9, skg.; 4. B. pomorum Pz. 9, skg.; 5. B. rajellus A. 9, skg. ;
G. Eucera longicornis Z. 9, skg.; 7. Halictus xanthopus A’ 9, skg., or at least
attempting to do so; 8. Osmia aurulenta Pz. 9, skg.; 9. Psithyrus rupestris /. 9, skg.

Schenck—in Nassau—observed the following bees skg.:—Osmia cornuta L/r.,
and Podalirius acervorum Z. In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfries-
shire,’ p. 10) a humble-bee was noticed.

160. C. acaulis Pers.—Kerner states that the flower mechanism of this plant
resembles that of the last species.

161. C. bracteata Pers., and 162. C. kolpakowskiana Regel.—

Visirors.—Loew observed—in the Berlin Botanic Garden—Anthophora pilipes
F. 9, skg.

163. C. claviculata DC. (Knuth, Bot. Centralbl., Cassel, lii, 1892, pp. 1-2;
Hart, Nature, London, x, 1874, p. 5.)—The inconspicuous whitish flowers are only
6-8 mm. long and 2 mm. broad, and are arranged in racemes with few (at most six)
flowers. They are homogamous. The stamens are at first rather shorter than the
style, so that self-pollination can only occur when the pollen is discharged into
the dark lilac hood which ensheaths the stigma. The hood when pressed down
does not return to its original position: the stigma remains concealed under the
folded plate of the inner upper petal. The smallness of the flower makes an
exact study of its mechanism difficult.

VisiTors.—I have not observed any, but have seen traces of the activity of
nectar-sucking insects. In many flowers the connection between the spurred petal
and the three others had been forcibly severed, so that these latter must have formed
a convenient platform for bees. Willis (‘ Fls. and Ins. in Gt. Britain,’ Part 1) noticed
the following near the south coast of Scotland.—

Hymenoptera. Apidae: 1. Bombus agrorum /.,, freq., skg.; 2. B. terrester L.,
skg.; both effecting pollination. Every flower appears to set fruit.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 10) there have been
observed,—Apis, 4 humble-bees, a short-tongued bee, and a Muscid.

48. Fumaria L.

Homogamous bee flowers. The nectar is secreted by a short process from the
upper filament. It is concealed in a short rounded pouch of the upper petals. The
other arrangements are as in Corydalis.

164. F. officinalis L. (Hildebrand, op. cit., p. 450; Herm. Miller, ‘ Ferti-
lisation,’ pp. 99-100; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 188-90 ;

PAPAVERACEAE 73

Kirchner, ‘Flora v. Stuttgart,’ p. 281; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
p. 23.)—The dull purple flowers are coloured blackish-red at the tip. The flower
mechanism agrees with that of Corydalis cava. “Owing to the smallness of the
flowers, the lateness of the flowering time, and the concealed habitat, insect-visits
are few. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) believes that cross-
pollination by bees or humble-bees is barely possible, as forcible removal of the
upper petal from the two lateral ones—which are united at their tip, and enclose
the sexual organs—very easily breaks off the style at its base, for it is not elastic
and yielding as in Corydalis. According to the same authority (op. cit. xxxvi;
1895), the pollen-grains are whitish, spheroidal, and smooth, with large projecting
germinating processes, and 56-62 in diameter. As the flowers, in spite of the
scarcity of insect-visits, almost all set fruits even in continuously rainy weather, which

Fic. 25. Fumarita officinalts, L. (after Hildebrand). (1) Flower seen from the side (enlarged).
(2) The same after removal of half the upper petal, and depression of the inner ones.
entirely prevents visits from bees, there can be no doubt that the inevitable automatic
self-pollination is effective.

Visttors.—Herm. Miiller in Westphalia, and myself at Kiel and Foéhr, observed
the honey-bee visiting the flower and transferring pollen: Warnstorf noticed a humble-
bee, but does not mention the species. In Dumfriesshire (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 9) a butterfly was observed.

165. F. capreolata L., and 166. F. parviflora Lam.—Hildebrand (op. cit.)
states that the flower mechanism is quite similar to that of F. officinalis, except
that the hood is so feebly elastic that after pressure from above, it either does
not regain its original position above the sexual organs or does so but slowly.
Both species are capable of autogamy. Moggridge noticed a long-tongued bee
(Osmia) visiting the variety pallidiflora of F. capreolata Z., which is indigenous
in the west and south of Europe. In this case the flowers are white before fertilization,
but afterwards become rose-red or even carmine-red. This remarkable change of
colour in the fertilized flowers may probably be explained as an adaptation for
increasing the conspicuousness of the whole inflorescence, and for enabling the
highly specialized visitors to recognize and avoid such flowers as being devoid of
nectar. Similar phenomena are met with in Ribes aureum and R. sanguineum,
Weigelia rosea, Melampyrum pratense, and others (cf. vol. i, pp. 85-6).

167. F. spicata DC.—According to Hildebrand (op. cit.), the flower mechanism
of this species agrees with those of Corydalis lutea and C. ochroleuca. The column
formed by stamens and pistil when liberated from the hood, springs up by the
elasticity of the upper filament, and places itself in the sheltering excavation in
the upper petal. This species is also fertile with its own pollen.

74 ANGIOSPERMAE—DICOTYLEDONES

IX. ORDER CRUCIFERAE JUSS.

LiteraTurE.—Herm. Miiller, ‘ Fertilisation,’ pp. roo-14; Knuth, ‘BI. u. Insekt.
a. d. nordfr. Ins.,’ p. 24.

The Cruciferae are all entomophilous, and most of them are homogamous.
By elongation of the floral axis the inflorescence, which is usually an umbel to
begin with, becomes a raceme, that renders the plant more or less conspicuous
according to the size and number of the flowers. But few species of this order,
however, are so conspicuous that insect-visits are very numerous. It follows that
almost all Cruciferae are capable of automatic self-pollination.

The calyx not only protects the developing flower, but in many cases holds
the claws of the petal together in such a way that they form a short tube at the
bottom of which the nectar lies. Insects are attracted by the petals, which are
usually yellow or white, more rarely violet, blue, or red. In spite of marked
uniformity in the structure of their flowers, the Cruciferae present so much diversity
in the number and position of the nectaries, in the relation of the stamens to these
and to the stigma, as well as in the mode of storing and concealing nectar, that
hardly any two species completely agree in all these details.

The number and position of the nectaries have been investigated very thoroughly
by J. Velenovsky. His researches were made on 170 species, including all the
European Cruciferae and a few exotic forms. He gives drawings of the anterior
and lateral aspects of the nectaries in 123 species (belonging to fifty-five genera).
According to him, no species is devoid of them. When a stamen is reduced, the
nectary develops into a rounded swelling. The size of these organs is usually
proportionate to the size of the flower, but to this there are exceptions. Heliophila
amplexicaulis, for instance, has much smaller flowers than Malcolmia maritima,
yet its nectaries are much larger than those of the latter. The largest upper nectaries
are found in Crambe maritima: the smallest in Stenophragma thalianum, and
Lepidium ruderale. The lower nectaries—i.e. those at the bases of the short
stamens—are always present, though in many cases they may be small, or even
almost vestigial, as in Crambe maritima and C. cordifolia. The upper nectaries,
on the other hand, are very frequently absent, and some species of a genus may
possess them, while they are lacking in others. The position of these organs has
special relation to the structure and form of the fruits.

Velenovsky groups the Cruciferae with reference to the structure of the nectaries
as follows.—

1. Siliquosae.

Upper and lower glands always present, usually with distinct lateral ridges.

1. Cheirantheae. Only the lower nectaries are free anteriorly and posteriorly.
(Cheiranthus, Matthiola, Malcolmia, Hesperis, Chorispora.)

2. Erysimeae. Upper and lower nectaries either free, or united by a weak
lateral ridge: the lower open anteriorly, closed posteriorly, where they are usually
thickened. (Barbaraea, Nasturtium, Armoracia, Roripa, Erysimum, Conringia,
Alliaria.) :

CRUCIFERAE 75

3. Arabideae. Upper and lower nectaries either connected by a lateral ridge,
or quite distinct. The lower are always open posteriorly, and closed anteriorly,
where they are usually thickened and modified in various ways. The upper are
either simple or compound, and are of various forms. (Cardamine, Dentaria, Arabis,
Stenophragma, Turritis.)

4. Sisymbrieae. The lower nectaries embrace the bases of the short stamens
in the form of a uniform unbroken pentagonal ring; the upper are represented by a
straight transverse ridge, which is connected by a lateral ridge with the lower
nectaries. All six stamens are therefore enclosed by a continuous uniformly thickened
ridge. (Sisymbrium, Chamaeplium.)

II, Siliculosae.

(2) LaTISEPTAE.

Only the lower nectaries are developed, and these are always free, i.e.
unconnected internally and externally. They are distinctly or approximately
three-sided.

1. Alysseae. The lower nectaries open anteriorly and posteriorly, and are
without lateral processes. (Schievereckia, Alyssum, Vesicaria, Cochlearia, Draba.)

2. Lunarieae. The lower nectaries are without lateral processes, and either
united together into a ring or open posteriorly. (Aubrietia, Lunaria.)

(2) ANGUSTISEPTAE.

The upper nectaries are sometimes developed, and resemble the lower ones,
with which they are connected by a lateral transverse ridge.

(a) Only the lower nectaries are developed. They are three-sided, always open
posteriorly, open or closed anteriorly, and usually prolonged into ridges laterally.
(Thlaspi, Carpoceras, Capsella, Teesdalea, Aethionema, Eunomia.)

(8) Only the lower nectaries are developed. They are prismatic, truncated
above, free anteriorly and posteriorly, and not prolonged laterally. (Iberis.)

(y) Either both lower and upper nectaries developed, or only the former, which
then occupy the position of the latter, as e.g. in some species of Lepidium and
in Coronopus didymus. The lower nectaries are prolonged laterally into a strong
ridge, free posteriorly, closed or free anteriorly at their downwardly produced ends.
The upper nectaries are simple, and not connected with the lower ones. (Cardaria,
Physolepidium, Lepidium, Coronopus.)

Ill. Nucamentaceae.

The relations of the nectaries are not so constant as in the preceding groups,
and the various sections of Biscutella converge towards all of them. Either the
lower nectaries only are present, or both lower and upper. The lower ones are
either prismatic, in which case the upper are columnar with a terminal pit, or—when
they only are developed—are represented by a uniformly thickened annular ridge,
open anteriorly or posteriorly, or even at both ends. These lower nectaries are
produced laterally into long processes, which are connected with the upper nectaries

76 ANGIOSPERMAE—DICOTYLEDONES

when present. The. latter may be double, simple, or appear merely in the form

of transverse ridges. (Bunias, Ochthodium, Myagrum, Isatis, Peltaria, Neslea,
Camelina.)

IV. Brassiceae.

Both lower and upper nectaries are present, and are never united. The lower
are prismatic, flat at the top, and inserted at the posterior side of the short stamens.
The upper are always simple, either angular and columnar, or irregularly three-
sided, and never in the form of a transverse ridge. (Succowia, Erucastrum, Eruca,
Diplotaxis, Brassica, Melanosinapis, Sinapis, Moricandia, Rapistrum, Raphanus,
Crambe.) (Cf. the abstracts given by Poldk in the Bot. Centralbl., Cassel, xii, 1882,
pp. 264-6, and xix, 1884, pp. 9-11.)

The position of the nectaries relatively to the anthers is such that insects
probing for nectar must come into contact with all or at least some of the latter
with one side of their body, and touch the stigma with the other side. The more
unfavourable the position of stamens and pistil for this purpose, the more is automatic
self-pollination furthered. As regards concealment of nectar, most Cruciferae belong

C

Fic. 26. Nectaries of some Cruciferae (after Prantl). Seen rom the side, after removal of sepals,
petals, and stamens. 4. Hesperis Matronalis ZL. @&. Selenia aurea Nuét. C. Sisymbrium strictis-
simum Z. Brassica Napus L.

to the flower class EC, a few with vertical sepals (Sinapis, Erucastrum) to E, and
others, especially those with violet, red, or blue flowers (species of Matthiola, Cakile,
Cardamine) to C. The last are visited by decidedly more numerous and more
highly specialized insects, adapted to effect pollination, than is the case with the white
or yellow flowers of Cruciferae belonging to the class EC. While these are visited
chiefly by flies (especially Syrphidae), by the less specialized bees (Apidae), and
to a lesser extent by other Hymenoptera (Sphegidae), as well as by beetles and
Lepidoptera, the violet, red, or blue flowers with concealed nectar receive many
visits from long-tongued bees and from Lepidoptera. Individual species (Hesperis
tristis Z.) are characteristic Lepidopterid flowers (L).

49. Matthiola R. Br.

Conspicuous flowers with deeply concealed nectar, secreted by glands at the
bases of the two short stamens.

168. M. incana R. Br. (Knuth, ‘Bloemenbiol. Bijdragen, Bot. Centralbl.,
Cassel, lxx, 1896, pp. 337-8.)—Homogamous bright red flowers, smelling of cloves.
The sepals are erect, and cohere above. They closely surround the claws of the
petals, so that these form a tube 15 mm. long and 2 mm, wide, expanding above

CRUCIFERAE 77

to 4mm. The cordate shape of the base of the calyx indicates even externally
the position of the nectaries. The insertion of each short stamen is surrounded
by a moderately large swelling that secretes nectar, of which a large drop is produced
on each side, so that as much as half the corolla-tube may be filled with it. The
four long stamens are surrounded at their bases by a much smaller swelling that
does not secrete nectar. The two sepals enclosing them are therefore but slightly
gibbous.

The anthers of the four long stamens are placed immediately beneath the
opening of the flower, and dehisce by slits about 5 mm. long, which are turned
inwards. The anthers of the two short stamens are equally long, but their filaments
measure only 2-3 mm., so that they do not reach the stigma, which stands about
8 mm. up the corolla-tube. The four long stamens therefore serve for self-pollination,
which is automatically effected by the fall of pollen, or is due to insect-visits. The
short stamens—which also dehisce introrsely—serve for cross-pollination: the

Fic. 27. Matthiola incana, R. Br. (from nature). (1) Flower; natural size. A, pouch at the base of
the calyx. (2) Stamens and pistil after removal of sepals and petals, the stamens spreading ; natural size:
a’, anther of a long stamen; a”, anther of a short stamen; s, stigma; 2, the ridge-like nectary surrounding
the base of the shorter stamen. (3) Stamens and pistil; twice natural size: references as in (2).

proboscis of an insect endeavouring to get at one of the secreting nectaries, will
be dusted with some of the pollen from the adjacent short stamen, and will transfer
it to the stigma of another flower of the same species.

Visitors.—In the garden of the Ober-Realschnle at Kiel, I observed a butterfly
(Vanessa urticae Z.) visiting cultivated plants and sucking nectar, and as it visited
several flowers in succession cross-pollination must have been effected. The
proboscis (14-15 mm. long) of this insect can reach right down into the nectar-
secreting base of the flower. In the same place I also noted Pieris sp., skg.

169. M. annua Sweet. (Nobbe, Bot. Centralbl., xxxii, 1887, p. 253; Knuth,
‘Bloemenbiol. Bijdragen.’)—F. Nobbe observed in cultivated plants that when the
seeds germinated very rapidly (3-4 days), most or sometimes all of the flowers
were double, but that—on the other hand—plants raised from seeds which germinated
slowly usually bore fertile single flowers. The same investigator also found that
if a variety of annual stock naturally inclined to produce double flowers was crossed
with one chiefly bearing single flowers, the mongrel exhibited the peculiarities of
the parent from which pollen was taken, not so much as regards the colour of the

78 ANGIOSPERMAE—DICOTYLEDONES

flower—which was intermediate between those of the two parents—but rather in
the general form of the raceme, and in the proportion of ‘double flowers to single.
Similar experiments have also been made by Schmid, Richter, and Hiltner. In-
formation as to these is published in ‘ Landwirtschaftliche Versuchs-Stationen,’ Berlin,
xxxv, Heft 3, 1888.

Visirors.—I have observed on the species a butterfly (Pieris) ; and Schletterer—
at Pola—noticed Xylocopa violacea L.

170. M. valesiaca Boiss.—According to Briquet (‘Etudes d. biol. flor. dans
les Alpes occident.,’ Bul. Univ. Lab. Bot., Genéve, i, 1896, pp. 16-78) the throat
of the corolla is whitish, while the limbs of the petals are of a dirty violet colour,
and spread out to form a surface 30-35 mm. across. At the base of each short
stamen there are two secreting nectaries, i.e. four in all. The nectar is concealed
in a narrow tube—8-10 mm. deep—formed by the sepals and the claws of the
petals, and is sucked by butterflies and also by humble-bees. As the anthers of
the four long stamens project above the stigma, these insects chiefly effect self-
pollination, more rarely cross-pollination.

171. M. nudicaulis (L.) Trautv.—Ekstam says that the diameter of these
strongly odorous homogamous flowers is 10-20 mm. in Nova Zemlia, or sometimes
as much as 35 mm. The nectaries are at the bases of the short stamens.

This species has only once been found (1827) in Spitzbergen—according to
Andersson and Hesselman (‘Bidrag till Kanned. om Spetsbergens 0. Beeren Eil.
Karlvaxtflora,’ p. 32)—so that no new facts are as yet forthcoming as to its time
of flowering there.

Visitors.—Humble-bees have been observed.

50. Cheiranthus L.

Flowers of considerable size, odorous, homogamous, with nectar almost con-
cealed. The nectaries are two swollen ridges at the bases of the two short stamens.
The stigma possesses two small recurved plates.

172, C. Cheiri L. (Herm. Miiller, ‘Weit. Beob.,’ i, p. 324; Kirchner, ‘ Flora
v, Stuttgart,’ p. 285; Knuth, ‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.’)—The
wild plants investigated by Kirchner had bright yellow flowers. From the two
nectaries there project outwards to right and left two pointed processes, the nectar
secreted by which is stored up in dilatations of the sepals. The anthers dehisce
introrsely, and are so situated that they completely close the entrance to the flower.
The four upper ones touch the stigma with their lower ends, while the two lower
ones do so with their tips. It follows that automatic self-pollination is inevitable.
Insect-visitors, however, are more likely to effect cross-pollination, for they touch
the stigma and anthers with opposite sides of their proboscis. The flowers of
cultivated plants are usually orange- to brownish-yellow.

Visitors. — Besides the honey-bee, I have observed a hover-fly (Rhingia)
visiting the flowers; Hermann Miiller also noted Anthophora pilipes /. 9, skg.
nectar; and Schenck—in Nassau—saw Andrena flessae Pz. Burkill (‘Fertlsn. of
spring fls.’) observed on the Yorkshire coast a humble-bee, Bombus terrester Z., skg.

CRUCIFERAE 79

nectar. Schletterer records the following list of visitors noticed in the Tyrol (T.),
and at Pola.—

Hymenoptera. (a) Apzdae: 1. Andrena albicrus XK. 9 and 4 (T.); 2. A.
albopunctata Ross’= A. funebris Pz.; 3. A. carbonaria L.; 4. A. flavipes Pz.; 5.
A. morio #rull.; 6. A. schlettereri “7zese; 7. Bombus aryillaceus Scop., sky. ;
8. Eucera longicornis Z.; 9. Halictus calceatus Scop.; 10. H. levigatus K.9; 11. H.
morio /.; 12. H. scabiosae Ross?; 13. H. villosulus X.; 14. Podalirius acervorum
L.; 15. P.crinipes Sm.; 16. P. nigrocinctus Zep.; 17. P. retusus Z. var. meridionalis
Per.; 18. Xylocopa violacea LZ. (6) Lchneumonidae: 19. Bassus laetatorius /.; 20.
Homotropus tarsatorius Pz. The short-tongued bees were undoubtedly only po-cltg.

51. Nasturtium R. Br.

White or yellow homogamous flowers, with half-concealed nectar. Nectaries
four or six.

173. N. officinale R. Br. (Herm. Miiller, ‘ Weit. Beob.,’ p. 325, ‘ Alpenblumen,’
p- 153; Kirchner, ‘Flora v. Stuttgart,’ p. 286; Knuth, ‘Bl. u. Insekt. a. d. nordfr.
Ins.,’ pp. 24, 148.)—Flowers white. There are two green fleshy nectaries placed
close together on the inner side of the base of each short stamen. The short
stamens turn their pollen-covered sides towards the stigma, which projects far beyond
them. The four long stamens at first reach the level of the stigma, but later on
this is higher. They are so turned towards the short stamens that the head or
proboscis of an insect probing for nectar must simultaneously touch the stigma
and the pollen-covered sides of the three nearest anthers. In rainy weather the
flowers remain almost closed, so that automatic self-pollination is effected by the
pollen of the long stamens. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) says
that the flowers are slightly protogynous; that the longer stamens reach the level
of the stigma; and that after the pollen is shed the filaments and anther-lobes
become violet in colour.

Visirors.—The following were observed by myself (Kn.) on the island of
Foéhr, and by Hermann Miller (H. M.) in Thuringia.—

A. Coleoptera. Aridulidae: 1. Meligethes (H.M.). B. Diptera. (a) Cono-
pidae: 2. Physocephala rufipes /., occasional, skg. (H.M.). (4) Empidae: 3. Empis
livida Z., very freq., skg. (H. M.); 4. E. rustica Fallen. ditto (H.M.). (c) Muscidae:
5. Ocyptera cylindrica F., skg. (H. M.). (d) Syrphidae: 6. Eristalis arbustorum Z.,
freq., skg. (H. M.); 7. E. nemorum Z., ditto; 8. E. sepulcralis Z., ditto(H. M.);
g. E. sp., ditto (H.M.); 10. Helophilus floreus Z., in large numbers, skg. and
po-dvg. (H. M.); 11. Melithreptus sp., po-dvg. (H. M.); 12. Syritta pipiens Z. (Kn.),
and 13. Syrphus sp. (Kn.), both skg. C. Hymenoptera. Apzdae: 14. Apis mellifica
L. ¥, skg. (H. M., Kn.); 15. Halictus maculatus Sm. ¥, skg. and po-dvg. (H. M.).

Hermann Miiller also observed—in the Alps—2 Hymenoptera and 4 Diptera ;
while MacLeod noticed in Flanders—Apis, and an Eristalis (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 196).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 11) beetles and
numerous flies were observed.

174. N. amphibium R. Br. (Herm. Miiller, ‘Fertilisation,’ p. 102, ‘ Weit.
Beob.,’ i, p. 3243 Kirchner, ‘Flora v. Stuttgart, p. 287; Knuth, ‘Bl. u. Insekt. a. d.
nordfr. Ins.,’ p. 24.)—Flowers yellow. The six nectaries are placed in the intervals

80 ANGIOSPERMAE—DICOTYLEDONES

between the stamens, and are confluent so as to form a ring. The anthers of the
four long stamens reach the level of the stigma, those of the two short ones being
somewhat lower. In sunny weather the stamens spread out somewhat, and the
anthers dehisce introrsely, so that insects probing for nectar necessarily touch
the stigma and the pollen with different sides of their head. They therefore effect
either cross- or self-pollination. In rainy weather the flowers only half open, so
that the anthers of the long stamens come into contact with the stigma—which
is at the same level—and automatic self-pollination results.

Warnstorf (Verh. bot. ver., Berlin, xxxviii, 1896) says that the flowers are
slightly protogynous. At the time of flowering the stamens project beyond the
stigma, over which their anthers incline. During dehiscence these bend outwards
at right angles, so that the open loculi with the pollen are turned upwards away
from the stigma. Self-pollination is thus rendered difficult and cross-pollination
favoured. Between the filaments at the base of the ovary are six small dark-green
nectaries. The pollen-grains are yellow, ellipsoidal, densely tuberculated. They
are up to 44 » long and 25-32 mw broad.

Visirors.— The following were observed by Hermann Miiller (H. M.) and
myself (Kn.).—

A. Coleoptera. Mitidulidae: 1. Meligethes nect-skg. and po-dvg. (H. M.).
B. Diptera. (a) Lmpidae: 2. Empis livida Z., skg. (H.M.). (4) Muscrdae: 3.
Calobata cothurnata Pz. (H. M.); 4. Sp.of Lucilia po-dvg. (H.M.). (c) Syrphidae :
5. Eristalis arbustorum Z., skg. and po-dvg. (Kn.. H. M.); 6. Rhingia rostrata L.,
skg. (H. M.); 7. Syritta pipiens Z., skg. (H. M.). C. Hymenoptera. Apzdae:
8. Apis mellifica Z. §, skg. (Kn., H.M.). (0) Pleromalidae: 9. Pteromalids nect-skg.
(H.M.). (c) Zenthredinidae: (a) Allantus arcuatus Fors/., skg. (H. M.).

MacLeod—in Flanders—observed Apis, 9 short-tongued bees, a wood-wasp,
5 hover-flies, and 4 other flies (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 198).

175. N. sylvestre R. Br. (Herm. Miller,
‘Fertilisation, pp. 100-1, ‘ Weit. Beob..,’ i, p. 324.)—
The flower mechanism agrees with that of the last
species, but there are four, fleshy, non-confluent
nectaries.

Hermann Miiller (H. M.) and Buddeberg
(Budd.) observed the following.—

A. Diptera. (a) Bombyliidae: 1. Anthrax

hottentotta Z., skg. (Budd.). (4) Empidae: 2. Empis
livida Z., skg. (H. M.). (c) Syrphidae: 3. Chryso-

Fic. 28. Nasturtium sylvestre, R. Br.
(after Herm. Miller). Flower seen from

above ; in the centre is the stigma cover-
ing the ovary; around it are the four
larger (2) and two smaller (2’) nectar-
drops ; to right and left are the two short
stamens (a’), anteriorly and posteriorly
the two pairs of long stamens (a). In all
the stamens the pollen-covered surface of
the anthers is visible, turned towards the
stigma. The filaments are all consider-
ably foreshortened. s, sepal; 7, petal.

gaster macquarti /oew, skg. (H.M.); 4. Eristalis
arbustorum Z., skg. (IJ. M.); 5. Syritta pipiens Z.,
and 6. Syrphus sp., both skg. and po-dvg. (H. M.).
B. Hymenoptera. (a) Aprdae: 7. Andrena labiata
Schenck 9, po-cltg. (H. M.); 8. Apis mellifica L. ¥,
freq., po-cltg. (H.M.); 9. Halictus nitidiusculus &’. 9,
skg. (H.M.). (8) Sphegidae: 10. Crabro wesmaeli
v.d. L., ske. (H.M.); 11. ‘Viphia minuta vw. d. Z.,
ditto (H. M.).

Alfken noticed the following Apidae at Bremen.—Halictus nitidiusculus K’. 9,

and Andrena albicans J/i//. 9.

CRUCIFERAE 81

MacLeod—in Flanders—saw a humble-bee, 2 hover-flies, and one of the Lepi-
doptera (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 197.)

176. N. palustre DC. (Kirchner, ‘Flora v. Stuttgart,’ p. 287; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ pp. 25, 148.)—The bright yellow petals are only as long as
the sepals, so that the flowers are less conspicuous than those of allied species.
There are two nectaries at the sides of the base of each short stamen. The anthers
of the four long stamens are at the same level as the stigma, while those of the
two short ones curve away from it and are somewhat lower. All six dehisce
introrsely. The two short stamens are therefore exclusively concerned with cross-
pollination, while the four long ones effect automatic self-pollination if insect-
visits fail.

VisiTors.—I observed—at Féhr—one hover-fly (Eristalis sp.) ; while MacLeod—
in Flanders—noticed 3 insects of the kind (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p- 197).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 11), Meligethes and
2 Muscidae have been recorded.

177. N. lippiense DC.—
Visirors.—Schletterer observed the following Agzdae at Pola.—

1. Andrena albopunctata Ross?; 2. A. carbonaria Z.; 3. A. combinata Chr.;
4. A. convexiuscula A.; 5. A. flavipes Pz.; 6. A. nana A.; 7. A. parvula K.;
8. Halictus calceatus Scop.; 9. H. fasciatellus Schenck; 10. H. levigatus K. 9; 11.
H. morio #.; 12. Prosopis clypearis Schenck.

178. N. pyrenaicum R. Br. (= Roripa pyrenaica Rezchd.). (MacLeod,
‘Pyreneenbl.’)—The yellow flowers have a diameter of 5-5 mm, when expanded,
There are four nectaries, of which the two situated between the pairs of long stamens
are very small. At the end of flowering, automatic self-pollination is effected by
contact of anthers and stigma. Nectar-seeking insects which visit the flowers mainly
effect cross-pollination.

VisiTors.—MacLeod observed—in the Pyrenees—a short-tongued bee (Halictus),
and 2 Muscidae.

52. Barbarea R. Br.

Homogamous yellow flowers with half-concealed nectar. There are six nectaries,
of which the pair at the base of each short stamen frequently coalesce.

179. B. vulgaris L. (Herm. Miller, ‘Weit. Beob.,’ I, pp. 325-6; Kirchner,
‘Flora v. Stuttgart,’ p. 288; Knuth, ‘Bloemenbiol. Bijdragen.’)\—The golden yellow
petals spread out in the sun to a diameter of 7-g mm. At the base of each short
stamen is a small fleshy green nectary on either side, the two often being united into
a semicircular ridge. There is also a larger elongated tooth-like nectary outside and
between the bases of each pair of long stamens, and corresponding to two short
stamens which have disappeared. These last produce only a small drop of nectar,
while the four other nectaries—or the two ridges resulting from their union—secrete
abundantly in favourable weather. This nectar collects in dilatations at the bases
of the two outer sepals, and in such quantity that the stamens—to use Miiller’s

DAVIS, IL G

82 ANGIOSPERMAE—DICOTYLEDONES

expression—are placed just as they would be if the two drops of nectar between
the pairs of long filaments did not exist. The long stamens—which project above
the stigma—make a quarter-turn towards the adjacent short stamens, while these are
on the same level as the stigma and remain facing it. In sunny weather the flowers
open widely, and the short stamens curve far away from the stigma: when the
weather is continuously wet, these stamens effect self-pollination. Insects visiting
the flower and trying to get at the larger drops of nectar, chiefly effect cross-
pollination.

Vistrors——Hermann Miller states that—in addition to the honey-bee, also
observed by myself—there are the following beetles and flies.—

A. Coleoptera. (a) Mitidulidae: 1. Meligethes in large numbers, nect-skg. and
po-dvg. (6) Curculionidae: 2. Ceutorhynchus sp. (c) Scarabaeidae: 3. Phyllopertha
horticola Z., gnawing the flower. B. Diptera. (a) Muscidae: 4. sp. of Anthomyia,
skg.; 5. Aricia incana Wiedem., skg.; 6. Calobata cothurnata Pz., skg.; 7. Scatophaga
merdaria F., skg. (6) Syrphidae: 8. Ascia podagrica /., po-cltg.; 9. Rhingia rostrata
L., freq., skg. and po-cltg.

Leow observed the following Apidae in the Berlin Botanic Garden: 1. Andrena
extricata Sm. 9, po-cltg.; 2. Apis mellifica Z. ¥, skg.: and 3. Bombus lapidarius ZL. 9,
skg. MacLeod—in Flanders—noted 2 bees, a hover-fly, and Empidae (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 194).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 10) Apis, 2 short-
tongued bees, 2 Muscidae, 2 hover-flies, and one of the micro-Lepidoptera have been
recorded.

180. B. intermedia Bor. (Kirchner, ‘ Flora v. Stuttgart,’ p. 288 ; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 195.)—The flowers are smaller and of a brighter
yellow than in the last species. Their diameter is only6mm. The flower mechanism
resembles that of B. vulgaris, except that there are only four nectaries.

53. Turritis Dill.
Homogamous flowers with half-concealed nectar. Four nectaries.

181, T. glabra L. (Kirchner, ‘ Flora v. Stuttgart,’ p. 289; Knuth, ‘Bloemenbiol.
Bijdragen.’)—The flowers are not conspicuous, for the yellowish-white petals are
tolerably erect. Two of the four nectaries are situated outside the bases of the two
pairs of long stamens: each short stamen is upon a swelling, which is drawn out into
a conical process on either side. Not infrequently the four nectaries coalesce into
aring. The anthers dehisce introrsely ; those of the long stamens touch the stigma
with their lower ends, while those of the short stamens do so with their tips, so that
automatic self-pollination is inevitable. Insects probing for nectar may effect either
cross- or self-pollination. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) describes
the flowers as being protogynous, the stigmatic papillae maturing even before the
flower is fully open, while the anthers subsequently ripen at the same level as the
stigma, so that autogamy is rendered possible.

VisiTors.—These are but few; I saw at Kiel only 2 hover-flies (Rhingia rostrata
Z., and Syritta pipiens Z.), skg. nectar.

CRUCIFERAE 83

54. Arabis L.

Flowers usually small, seldom of considerable size; white or whitish, rarely
rose-red or lilac or blue; generally homogamous, infrequently protogynous. Nectar
half-concealed ; nectaries 2, 4, or 6.

182. A. alpina L. (Sprengel, ‘Entd. Geh.,’ p. 333; Axell, ‘Om Anord. for
Fanerog. Vaxt. Befrukt.’; Herm. Miiller, ‘ Alpenblumen,’ pp. 143-4; Schulz, ‘ Beitrige,’
II, pp. 11-2.)—The flowers are homogamous. Of the four nectaries those outside
the bases of the two short stamens are the largest; their nectar collects in the
dilatations of the sepals beneath them. The two smaller nectaries outside and
between the bases of each pair of long stamens, secrete hardly any nectar. The
form of the nectaries is very variable. The long stamens sometimes turn their
pollen-covered sides towards the adjacent short stamens, so that an insect probing
for nectar must touch them, and therefore cross-pollination is likely to take place.
In other cases these stamens turn towards the stigma, and then, especially in dull
weather (and always in Greenland), either pollen falls from them on the stigma, or

Fic. 29. Arabis alpina, L. (after Herm. Miller). .4, Flower after removal of two petals, seen from
the side; s#, nectar receptacles. Each of the long stamens is turned towards the adjacent short stamen.
BR. Flower seen from above. The pollen-covered sides of all the anthers are turned towards the stigma;
the filaments, however, are curved back in such a way that self-pollination does not take place at this
stage. C. Flower after removal of calyx and corolla: 7, functional nectaries; 2’, vestigial nectaries.

Stamens are asin B. (x 7.)

else this is touched by the anthers, so that automatic self-pollination is effected. In
Nova Zemlia, according to Ekstam, the feebly odorous flowers are 6-12 mm. in
diameter, and protogynous-homogamous. Nectar is abundantly secreted, and self-
pollination is easily possible. (See Fig. 29.)

This species was observed in flower, and bearing young fruits, in Spitzbergen on
July 29, 1896 (Anderson and Hesselman, ‘ Bidrag till Kanned. om Spetsbergens o.
Beeren Eil. Karlvaxtflora,’ p. 34).

Visitors.-—Herm. Miiller observed 2 flies in the Alps; while Schulz in the Tyrol
also saw various butterflies. MacLeod noticed 2 flies in the Pyrenees; and Loew—
in the Berlin Botanic Garden—saw the hive-bee sucking nectar.

183. A. Hookeri Lange.—This species is native only to West Greenland and
Arctic America. The white petals are as much as 5 mm. long, and project beyond
G 2

84 ANGIOSPERMAE—DICOTYLEDONES

the calyx, of which the upper part is purple in colour. The plant is apparently
biennial (see Abromeit, ‘Bot. Ergeb. von Drygalski’s Grénlandsexped.,’ pp. 27-8).

184. A. pauciflora Garcke (=A. brassicaeformis Wad/r., Brassica alpina Z.).
(Schulz, ‘ Beitrage,’ II, p. 11.)—The white flowers are homogamous. At the base of
each short stamen there is a ridge-like nectary, and at the base of each pair of long
stamens a small projection. Very little nectar is secreted. The nectaries vary greatly
in form. The stigma is usually at the same level as, and in contact with, the bases
of the anthers of the long stamens, and therefore automatic self-pollination is inevitable.
In favourable weather these anthers turn towards the short stamens, so that insect
visitors may effect cross-pollination.

VisiTors.—Schulz—in Thuringia—only observed Physopoda and flower-beetles
(Meligethes).

185. A. petraea (L.) Lam.—Ekstam says that the flowers of Nova Zemlian
plants exhale a moderately strong odour of almonds, and are homogamous. Self-
pollination is easy.

Visitors.—Ekstam observed a moderate-sized fly in Nova Zemlia.

186. A. hirsuta Scop. (Herm. Miller, ‘Fertilisation,’ p. 102.)\—The white
flowers are homogamous. Nectar is secreted only by the two nectaries inside
the bases of the short stamens. In most flowers the long stamens project beyond
the stigma, and therefore autogamy results from the fall of pollen if insect-visits
fail. More rarely the anthers of the long stamens are at the level of the stigma,
so that automatic self-pollination in effected by direct contact. Warnstorf (Verh. bot.
Ver., Berlin, xxxviii, 1896) says that the flowers are protogynous, and that the
stigma—which matures before the flowers open—projects somewhat beyond the
stamens.

Visitors.—Herm. Miiller observed the following.—

A. Diptera. Syrphidae: 1. Syritta pipiens Z.,skg. B. Hymenoptera. (ca)
Apidae: 2. Andrena albicrus K. 6, skg.; 3. Apis mellifica Z. ¥, skg.; 4. Halictus
sexnotatus . 9, po-cltg. (4) Sphegidae: 5. Ammophila sabulosa Z., skg. C. Lepi-
doptera. Bombyces: 6. Euchelia jacobaeae Z., skg.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 12) one of the
Empidae, 2 Muscidae, and 2 hover-flies have been recorded.

187. A. arenosa Scop.—Flowers lilac, or more rarely white.

Visitors.—Buddeberg in Nassau observed a butterfly (Thecla rubi Z., skg.),
and the following short-tongued Apzdae.—

1. Andrena albicans Af#Ul. 9, skg.; 2. A. cineraria Z. 9, po-cltg.; 3. A. cingulata
Fg and 6, skg.; 4. A. nigroaenea K. 9, skg.; 5. A. parvula X. 9, freq., skg. and
po-cltg. (12 individuals) ; 6. Halictus calceatus Scop. 9, skg. and po-cltg.; 7. H. flavipes
K. 9, skg.; 8. H. leucopus A’. 9, skg. and po-cltg.; 9. H. tetrazonius Aly. (quadri-
cinctus A’ olim) 9, skg.

Bail (Bot. Centralbl., Cassel, ix, 1882) records the following for West
Prussia.—

A. Coleoptera. Laseridac: 1. Athius subfuscus Afiél/. B. Diptera. (a)
Syrphidae: 2. Eristalis intricarius Z.; 3. Melanostoma mellina Z.; 4. Melithreptus
scriptus £.; and other sp. (4) Afuscidae: 5. Lucilia sp.; 6. Anthomyia sp. C.

CRUCIFERAE 85

Hymenoptera. (a) Apidae: 7. Apis; 8. Andrena nana K. (6) Zenthredinidae :
9. Dolerus vestigialis K/ug. D. Lepidoptera. (a) Noctuidae: 10. Euclidia glyphica
L. (6) Rhopalocera: 11. Pieris napi Z.; 12. Thecla rubi Z.; 13. Nemeobius lucina
L. E.Hemiptera. Pentatomidae: 14. Eurydema oleraceum Z.

188, A. Turrita. L.—Flowers white.

Vistrors.—MacLeod observed a short-tongued bee (Halictus cylindricus F’. 9)
in the Pyrenees.

189. A. sagittata DC.—

VisiTors.—MacLeod observed a moth (Adela sp.) in the Pyrenees.

190. A. pumila Jacq. (Schulz, ‘ Beitrage,’ II, pp. 12-13.)—The white flowers
are protogynous in the Tyrol. Outside the base of each short stamen there is
a crescentic nectary with two projections. The stigma usually matures in the bud,
and frequently protrudes between the petals before the flower opens. At the time of
anthesis the style almost always projects about 1 mm. beyond the anthers of the
long stamens, and 2-3 mm. beyond those of the short stamens. It is exceptional
for the long stamens to reach the level of the stigma. It follows that automatic
self-pollination is practically excluded. Insect visitors chiefly effect cross-pollination.
Schulz observed gyno-monoecism.

Visitors.— Schulz saw three small Diptera visiting the flowers in dull weather.

1g1. A. bellidifolia Jacq. (Herm. Miiller, ‘Alpenblumen,’ pp. 144-5.)—The
white flowers are protogynous with persistent stigmas. At the base of each short
stamen there is a green muriform fleshy nectary secreting abundant nectar; there
is also a little knob outside the base of each long stamen. The anthers of the long
stamens are at the same level as the stigma, and in dull weather they remain facing
it till they dehisce, and effect self-pollination: but in sunny weather these stamens
move outwards away from the stigma, so that an insect visitor may effect cross-
pollination.

Visrtors.—Herm. Miiller only observed a hover-fly (Eristalis tenax Z.).

192. A. alpestris Reichb.—
Visirors.—Hermann Miller observed the following in the Alps; the honey-bee,
2 butterflies, 2 hover-flies, and 2 Muscidae (‘ Alpenblumen,’ p. 145).

193. A. caerulea Haenke. (Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed.1, I], pp.192, 335;
Kirchner, ‘ Beitrage,’ p. 20; Schulz, ‘ Beitrage,’ II, p. 13.)—The flowers are at first
blue, but afterwards become pale. They are homogamous or slightly protogynous.
Outside the bases of the short stamens there are nectaries which are often very
insignificant. Others—resembling these but non-secreting—are situated at the bases
of the long stamens. The anthers of the long stamens are at the same level as the
stigma, to which they approximate in dull weather and at night, and the anthers
of the short stamens not infrequently reach the stigma. Automatic self-pollination
is therefore inevitable. Kerner asserts that in continuously rainy weather this takes
place within the still unopened flowers, which are therefore pseudo-cleistogamous.

Visrrors.—A few flies have been observed.

194. A. Holboellii Hornem.—The flowers are of considerable size, and
Warming describes them as homogamous. The anthers of the long stamens at

86 ANGIOSPERMAE—DICOTYLEDONES

first project beyond the stigma, but consequently come into contact with it, as a result
of growth of the style.

195. A. albida Stev.—
Visrrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Coccineliidae: 1. Coccinella septempunctata Z. B. Diptera.
Syrphidae: 2. Cheilosia sp., po-dvg.; 3. Eristalis aeneus Scop. C. Hymenoptera.
Apidae: 4. Andrena parvula X. 3, skg., 9, skg. and po-cltg.; 5. Apis mellifica L. ¥,
skg.; 6. Bombus hortorum Z. ¥, skg.; 7. B. lapidarius Z. 9, skg.; 8. Osmia rufa
L.%, skg. D. Lepidoptera. hopalocera: 9. Vanessa urticae L., skg.

196. A. deltoides DC.—

Vistrors.— Loew observed a long-tongued bee (Osmia rufa Z. 9), skg. and
po-cltg., in the Berlin Botanic Garden.

197. A. caucasica Willd.—

Visrrors.—Burkill (‘ Fertlsn. of spring fls.’) observed—on the coast of Yorkshire—
one of the Syrphidae, Eristalis pertinax Scop., freq., skg.

55. Cardamine L.

Flowers homogamous or protogynous, white or lilac in colour, with half-concealed
or completely concealed nectar. Nectaries 2 or 4.

198. C. pratensis L. (Sprengel, ‘ Entd. Geh.,’ p. 331; Herm. Miiller, ‘ Fertili-
sation,’ pp. 102-4, ‘ Weit. Beob.,’ I, p. 326; Kirchner, ‘ Flora v. Stuttgart,’ pp. 290-1 ;
Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896; Knuth, ‘Bl. u. Insekt. a. d. nordfr.
Ins., pp. 25, 148.)—The white or lilac flowers are large and conspicuous, and insect-
visits are therefore more numerous than in the case of most other plants of the order.
There are two larger nectaries at the bases of the two short stamens, and two other
smaller ones—one outside the bases of each pair of long stamens. The nectar
secreted by these four glands collects in the swollen bases of the sepals, and the
pouches of the two sepals beneath the larger and more vigorously secreting nectaries
are proportionately larger than those of the sepals beneath the two smaller nectaries.
As Hermann Miiller remarks, the position of the two short stamens can be recognized
by looking at the under-side of the calyx. The sepals are closely applied to the
petals, so that the claws of the latter are held together so as to form a tube several
milimetres long, in the bottom of which the nectar is concealed. C. pratensis therefore
belongs to the flower class C. Even in the bud the long stamens project beyond the
stigma. They also make a quarter-turn towards the adjacent short stamens, so that
insects probing for nectar touch the stigma and the pollen-covered anthers with
opposite sides of their heads. It follows that if they probe the flower first on one
side and then on the other self-pollination is effected, but if they work round it cross-
pollination may result. The two short stamens always turn the dehisced sides of the
anthers towards the stigma. These anthers in many flowers are at a lower level than
the stigma, or they may be at the same or a higher level. In the last two cases
automatic self-pollination is possible. During cold rainy weather the rotation of the
long stamens is either slight or does not take place at all, so that pollen falls upon
the stigma. Yet according to Hildebrand (Ber. D. bot. Ges., Berlin, xiv, 1896), the

CRUCIFERAE 87

plant is self-sterile. Warnstorf says that the flowers are protogynous, and that
the stigmatic papillae mature in the still unopened flower.

In the plants studied by Warming in Greenland, the anthers of the short stamens
are so close to the stigma that automatic self-pollination is possible; ripe fruits,
however, are seldom formed, and propagation is effected vegetatively by bulbils.

In Nova Zemlia—according to Ekstam—the diameter of the faintly odorous,
protogynous-homogamous flowers is 10-15 mm.; while Kjellman states that it is
usually 24 mm. in Arctic Siberia. The flowers agree in structure with those on
the North Frisian Islands which have been described by me.

This species flowers in Spitzbergen from the middle of July till the middle
of September (Andersson and Hesselman, ‘ Bidrag till Kanned. om Spetsbergens o.
Beeren Eil. Karlvaxtflora,’ pp. 32-3). According to the authorities quoted ripe fruits
are formed neither in Greenland (Kolderup Rosenvinge) nor arctic Norway (Norman).
Ekstam, however, has seen them in Spitzbergen (‘ Bliitenbiol. Beob. a. Spitzbergen,’
p. 19). Andersson and Hesselman regard this as a probable case of partheno-
genesis ; in two specimens they collected there were, respectively, 95 % and 100 % of
pollen-grains incapable of germination. According to Ekstam the flowers are white
or bright red—according to Andersson bright violet, with darker veins—attain a
diameter of 13-18 mm., and are slightly fragrant. The sexual organs are mature
at the time the flowers open.

Visitors.—Ekstam saw a small fly in Nova Zemlia. No visitors have been seen
in Spitzbergen.

The following have been observed by Hermann Miiller (H. M.), and myself
(Kn.).—

A. Coleoptera. (a) Witidulidae: 1. Meligethes sp., freq., nect-skg. (H. M., Kn.).
(4) Staphylinidae: 2. Omalium florale Pay’., extremely freq. (H.M.). B. Diptera.
(2) Bombyhidae: 3. Bombylius discolor AZg., skg. (H. M.); 4. B. major Z., skg.
(H. M.). (6) Empidae: 5. Empis opaca #., skg. (H. M.). (¢) Musctdae: 6.
Anthomyia sp., po-dvg. (H. M., Kn.). (d) Syrphidae: 7. Eristalis nemorum Z.,
po-dvg. (H. M.); 8. Helophilus pendulus Z., skg. (H. M., Kn.); 9. Melanostoma
mellina Z., po-dvg. (H. M.); 10. Rhingia rostrata Z., freq., skg. and po-dvg. (H. M.);
11. Syrphus nitidicollis A7g., skg. and po-dvg. (H. M.); 12. S. sp. (Kn.). C. Hy-
menoptera. Ajzdae: 13. Andrena cineraria Z. 9, one individual, po-cltg. and skg.
(H.M.); 14. A. dorsata A. 9, skg. and po-cltg. (H. M.); 15. A. gwynana X. 9, one
individual, po-cltg. (H. M.); 16. A. parvula A. 9 and 4, po-cltg. and skg. (H. M.);
17. Apis mellifica Z. ¥, very freq., po-cltg. or skg. (H.M., Kn.); 18. Bombus terrester
L. 9, skg. (H. M., Kn.); 19. Halictus cylindricus /. 9, po-cltg. and skg. (H. M.);
20. Nomada lateralis Pz. 9, skg. (H. M.); 21. N. lineola Pz. 5, skg. (H.M.); 22.
Osmia rufa Z. 4, skg.(H.M.). D. Lepidoptera. Rhopalocera, skg.: 23. Anthocharis
cardamines Z. (H. M.); 24. Pieris brassicae Z.(H.M.); 25. P. napi Z.(H. M., Kn.);
26. Rhodocera rhamni Z.(H.M.); 27. Vanessa urticae Z.(Kn.). E. Thysanoptera.
28. Thrips, skg. and po-dvg. (H. M.).

Alfken noticed the following in Bremen.—Agzdae: 1. Bombus derhamellus KX. ¥,
skg.; 2. B. pomorum Pz. 9, skg.; 3. Nomada succincta Pz. 9, skg. Also a butterfly,
Thecla rubi Z., skg. Rossler saw a hawk-moth, Macroglossa fuciformis Z., at
Wiesbaden.

In Dumfriesshire (Scott-Elliott, ‘Flora of Dumfriesshire,’ p. 13), one of the
Bibionidae, a Muscid, 4 hover-flies, and a Lepidopterid have been recorded.

88 ANGIOSPERMAE—DICOTYLEDONES

H. de Vries (Ned. Kruidk. Arch. Nijmegen, 1875)—in the Netherlands—
observed a bee, Halictus quadricinctus #. 9; and MacLeod—in Flanders—noticed
8 bees, 11 flies, 4 Lepidoptera, and a beetle (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp- 192-3).

199. C. amara L. (Ludwig, D. bot. Monatschr., Arnstadt, vi, 1888, p. 5;
MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 193-4; Kirchner, ‘Flora
v. Stuttgart,’ p. 291; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1296; Knuth,
‘Bloemenbiol. Bijdragen.’)—The nectaries are as in C. pratensis. The flowers are
narrowed like a funnel below: they therefore belong to the flower class C. The six
stamens are almost of the same length, and diverge, the pollen-covered sides of their
anthers facing inwards. The pistil is barely half as long as the stamens (according
to Warnstorf it is of the same length). Insects probing for nectar touch the anthers
and stigma with opposite sides of their heads, so that cross-pollination is usually
effected. Self-pollination results only when they suck nectar first on the right and
then on the left side of the pistil. Besides hermaphrodite flowers inflorescences
bearing small female ones have been observed.

VisiTors.—In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 12), flies,
Lepidoptera, and beetles have been recorded.

200. C.impatiens L. (Herm. Miiller, ‘ Weit. Beob.,’ I, p. 327 ; Kirchner, ‘ Flora
v. Stuttgart,’ p. 292.)—The white petals are very small or else absent, so that the
flowers are inconspicuous. There is a nectary at the base of each of the long pairs
of stamens, and also one at the base of each short stamen. They are united outside
the filaments by a prominent ridge. The stamens diverge widely outwards, and their
pollen-covered sides face inwards so that insects probing for nectar are likely to effect
cross-pollination.

VisiTors.—Buddeberg—in Nassau—observed a bee, Andrena albicans JZiU/. 9,
skg. nectar and po-cltg.

201. C. hirsuta L.—Jordan states that the anthers are applied to the stigma,
so that automatic self-pollination is inevitable.

Visttors.—MacLeod—in Flanders—observed a short-tongued bee, a Muscid,
and a beetle (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 193).

In Dumfriesshire (Scott-Elliott, ‘ Flora of Dumfriesshire,’ p. 14), a beetle, a hover-
fly, and 2 Muscidae were recorded.

202. C. latifolia Vahl—The lilac flowers have concealed nectar.

Visirors.—MacLeod observed 2 butterflies in the Pyrenees.

203. C. bellidifolia L.—In Greenland—according to Warming—the flowers
are autogamous, the anthers being for some time closely opposed to the stigma.

Ekstam states that in Nova Zemlia the odourless flowers are protogynous-
homogamous, and self-pollination is inevitable. According to Kjellman their diameter
in Arctic Siberia is 8 mm.

This species flowers in Spitzbergen from the end of June to the end of August:
at the end of this period ripe fruits have been observed; the pollen is abundant and
normal (Andersson and Hesselman, op. cit., pp. 33-4). The homogamous flowers
are faintly odorous, and have a diameter of 5-7 mm. (Ekstam, op. cit., p. 19).

Visirors.—None were observed in Spitzbergen.

CRUCIFERAE 89

204. C. resedifolia L.—Schulz (‘ Beitrage,’ II, pp. 13-14) says that automatic
‘self-pollination is inevitable in the homogamous flowers.

Vistrors.—Herm. Miiller saw 6 flies (Muscidae, Syrphidae, Empidae) and
a butterfly in the Alps.

205. C. alpina L.—According to Kerner this species is protogynous. The
stigma projects from the opening flower, the stamens being still immature, so that
at this stage pollination can only be effected by the agency of insects. Automatic
self-pollination is possible later on when the stamens elongate.

206. C. chenopodifolia L.—Grisebach states that in this species there are
subterranean cleistogamous flowers in addition to the open flowers above ground.

56. Dentaria Tourn.

Flowers of considerable size, whitish or reddish, usually with concealed nectar.
As a rule there are four nectaries.

207. D. enneaphylla L. (Schulz, ‘Beitrage,’ II, p. 14.)—The whitish-yellow
petals are 13-17 mm. long. Outside the base of each short stamen there is
a crescentic outwardly directed ridge, and at the middle of the bases of each pair
-of long stamens a broad process projecting upwards. These four nectaries secrete
an insignificant amount of nectar. The anthers of the long stamens usually project
a little above the petals, and are generally at the same level as the stigma—which is
frequently mature before anthesis—though in rare cases they are a little lower. As
even in warm weather the petals and stamens diverge but little, the anthers are so
near the stigma that automatic self-pollination necessarily takes place. The anthers
of the short stamens usually reach only to the middle of the long ones, and dehisce
at the same time as, or a little later than, the anthers of these. They serve exclusively
for cross-pollination.

Visitors.—Schulz—at Sanmartino and Paneveggio—saw flies and beetles and
especially Noctuidae creeping into the flowers.

208. D. bulbifera L. (Kirchner, ‘Flora v. Stuttgart, p. 291; Knuth, ‘ Bloe-
menbiol. Bijdragen.’)—In the large pale lilac, rose-red, or white flowers, there is
a nectary outside the base of each short stamen, and one that is usually cleft outside
the insertions of the two pairs of long stamens. Sometimes the four nectaries unite
to form a ring. Fruits are found only in sunny places where insects visit the flowers ;
in shady woods the plant is almost always sterile, propagating by dark-violet bulbils
produced in the axils of the leaves.

Visitors.—In spite of much watching I have never seen insects visiting the
flowers in the woods near Kiel and Flensburg, and have very rarely noticed fruits.

57. Hesperis L.

The odorous flowers are of considerable size, and their nectar is concealed.
‘They belong to the flower-classes C and L,

209. H. matronalis L. (Herm. Miller, ‘Fertilisation,’ pp. 108-9 ; Kirchner,
‘Flora v. Stuttgart,’ p. 293; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896 ; Kerner,
‘Nat. Hist. Pl,’ Eng. Ed. 1, II, p. 209; Knuth, ‘Weit. Beob. ii. Bl. u. Insekt. a. d.

90 ANGIOSPERMAE—DICOTYLEDONES

nordfr. Ins.,’ p. 23, ‘Bloemenbiol, Bijdragen.’)—The large violet flowers exhale
a strong odour of violets, especially in the evening. Two large green fleshy nectaries
surround the bases of the short stamens, and are especially well developed internally.
The nectar collects on either side of the flower between the insertions of the stamens
and the base of the pistil. The anthers of the long stamens stand in the throat of
the corolla; after dehiscence they grow somewhat and project beyond the flower.
When the anthers of the short stamens dehisce, their lower parts come into contact
with the stigma, which during anthesis grows so as to project out of the flower. The
anthers dehisce introrsely, so that all of them can pollinate the stigma, automatic
self-pollination being therefore inevitable. Insects sucking nectar regularly effect
cross-pollination, however, for they touch stigma and anthers with opposite sides
of their proboscis or head. In the case of those which collect pollen the chances
of cross- and self-pollination are equal.

Warnstorf says that the flowers vary from slightly protogynous to homogamous.
The pollen-grains are pale yellow, closely and finely tuberculated, and ellipsoidal,
measuring up to 37 mw in length and 25 » in breadth.

Visirors.—The following have been observed by Hermann Miller (H. M.),
Borgstette (B.), Buddeberg (Budd.), and myself (Kn.).—

A. Coleoptera. Zélephoridae: 1. Anthocomus fasciatus Z. (H.M.). B. Diptera.
(2) Stratiomytdae: 2. Nemotelus pantherinus Z., po-dvg. (H.M.). (4) Syrphidae:
3. Chrysogaster aenea A/g., po-dvg. (B.); 4. Eristalis arbustorum Z., po-dvg. (Kn.) ;
5. E. nemorum Z., po-dvg. (H. M., B.); 6. E. pertinax Scop., po-dvg. (Kn.); 7. E.
tenax L., po-dvg. (H. M., Kn.); 8. Rhingia rostrata Z., very freq., skg. and po-dvg.
(H. M., Budd., Kn.); 9. Volucella pellucens Z. (B.). C. Hymenoptera. Apzdae:
ro. Andrena albicans full. 9, po-cltg. (H.M.); 11. Apis mellifica Z. ¥, po-cltg.
(H.M., Kn.); 12. Bombus lapidarius Z. 9, ditto (Kn.); 13. Halictus leucopus A’ 9,
po-cltg. (H. M.). D. Lepidoptera. 14. Pieris brassicae Z., freq., ske. (H. M., Kn.);
15. P. napi Z., ditto (H. M., Kn.); 16. P. rapae Z., ditto (H. M., Kn.); 17. Vanessa
urticae Z., skg. (Kn.).

210. H. tristis L. (Herm. Miller, ‘Weit. Beob.,’ II, pp. 200-2; Kerner,
‘Nat. Hist. Pl.,” Eng. Ed. 1, I, p. 208.)—This species bears homogamous moth-
flowers. The petals are of a dirty greenish-yellow, and traversed by a network of
delicate dirty grey-green veins. They are therefore not very conspicuous among
the foliage, although their spreading limbs are 14-20 mm. long, or even more,
and 3-53 mm. broad. During the daytime there is no marked scent, and scarcely
any insect visitors. Between seven and eight o’clock in the evening, however, the
blossoms open and exhale a powerful and agreeable odour—of hyacinths, according
to Kerner—after the fashion of moth-flowers.

On the inner side of the bases of the short stamens—according to Hermann
Miiller’s account—there are two large green fleshy nectaries, the secretion of which
is so abundant that it completely fills on each side the angle between the base
of the short stamen, the two adjacent long stamens, and the pistil.

The narrow sepals are 11-15 mm. long. Their basal parts curve slightly
outwards, but their upper two-thirds are so closely apposed as to hold the claws
of the petals firmly together. At the beginning of anthesis there are thus but one
or two narrow passages to the nectar, which are only adapted to the proboscides.
of Lepidoptera. When the flowers open the pollen-covered anthers of the four

CRUCIFERAE gE

long stamens—which face inwards—are in the throat of the corolla, and the
simultaneously maturing stigma is 1-2 mm. beneath them. The stigma expands
to right and left, and is longitudinally cleft into two lobes, the narrow ends of
which are reflexed. The two short stamens also direct their pollen-covered surfaces
towards the middle of the flower, and are so close to the stigma that their tips
are at the same level as, and about 1 mm. distant from its reflexed lobes. The
one or two narrow nectar-passages run between the end of the pistil and the short
stamens, so that the proboscis of a moth probing for nectar must touch the stigma
with one side, and the pollen of a short stamen with the other. After the proboscis
has been dusted with pollen all round cross-pollination must be effected in every
new flower visited. Should insect-visits fail the stigma pushes up between the four
long stamens and dusts itself with pollen, effecting automatic self-pollination which—
according to Hermann Miiller’s experiments—is effective, though Hildebrand asserts
(Ber. D. bot. Ges., Berlin, xiv, 1896) that the species is self-sterile. In the early stage
of anthesis the only use of the long stamens is to keep unbidden guests from the
nectar by blocking the entrance of the flower with their anthers.

VisiTors.—Agnes Miiller—daughter of Hermann Miiller—observed the following
during a few mild evenings in May.—Three Woc/uidae :—1. Dianthoecia nana Hu/nx. ;
2. Hadena sp.; 3. Plusia gamma Z., freq.: also the Geometrid Iodis lactearia Z.,
and the Pyralid Pionea forficalis Z. In these moths the proboscis is from 11-
18 mm. long.

58. Malcolmia R. Br.

Flowers of considerable size, with concealed nectar.

2u1. M. maritima R. Br.—Kerner says that in this species there are two rows
of erect stiff-pointed bristles on the ovary, to prevent any insect from probing for
nectar, except in such a way that its proboscis and head touch the pollen-covered
anthers and the stigma. The plant is self-fertile according to Hildebrand (Ber. D.
bot. Ges., Berlin, 1896).

59. Sisymbrium L.

Small yellowish or whitish flowers, homogamous to slightly protogynous, with
half-concealed nectar. Nectaries two, four, or six.

212. S. officinale Scop. (Herm. Miller, ‘ Fertilisation,’ pp. 109-10, ‘ Weit.
Beob., II, p. 202; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 26.)—The small bright
yellow flowers are only 3 mm. in diameter. There is a nectary on either side the
base of each short stamen, and the nectar secreted coilects on each side in the angle
bounded by a short stamen, a long stamen and the pistil. The dehisced surfaces
of the anthers of the long stamens are at first on the same level as, and almost
touching, the stigma, both they and it projecting a little from the flower. The
two short stamens are meanwhile still concealed within the corolla, but their anthers
have also dehisced. All the stamens now grow a little, the long ones stretching
beyond the stigma, and inclining together above it, while the short ones—which
are now as long as the pistil—diverge outwards to some extent.

An insect visitor may therefore effect either cross- or self-pollination. As insect-
visits are few, automatic self-pollination frequently takes place by pollen falling on

92 ANGIOSPERMAE—DICOTYLEDONES

the stigma during the second stage of flowering. This self-pollination is effective,
according to Comes (‘Stud. s. impoll. i. alc. piante’).

Visitors.— Hermann Miiller (H. M.) and myself (Kn.) have observed the
following.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp., po-dvg. (H. M.). (4) Syrphidae :
2. Ascia podagrica #., in large numbers, po-dvg. (H.M.). B. Hymenoptera.
Apidae: 3. Apis mellifica Z., skg. (Kn.); 4. Andrena dorsata K., po-cltg. and skg.
(H.M.); 5. Halictus morio /. 4, skg. (H. M.). C. Lepidoptera. 6. Pieris brassicae
L., skg. (Kn.); 7. P. napi Z., skg. (H. M., Kn.); 8. P. rapae Z., skg. (H. M.).

Alfken noticed the following Apidae at Bremen.—1. Prosopis communis JVyi. 9 ;
2. Eriades nigricornis Ny/. 9, Schletterer saw at Pola the two bees Andrena florea
F., freq., skg., and Halictus calceatus Scop.; also the fossorial wasp Pemphredon
unicolor /., very freq.

Fic. 30. Szsymbrdum. (From nature. Semi-diagrammatic, and enlarged.) Two of the long
stamens, two petals, and the anterior sepal have been removed. (1) S. officinale 2. Flower in the first
stage; the anthers of the long stamensia) are at the same level as the stigma (s), those of the short ones(a’)
are lower. (2) S. officinale. Flower in the second stage: the anthers of the long stamens project beyond
the stigma, those of the short stamens are at the same level. (3) S. Sophia Z. Flower in the second
stage; 2, calyx; ¢, petal; #, nectary.

MacLeod observed 2 bees, 3 hover-flies, and a Muscid in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 199-200).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 14) a Muscid and
a hover-fly have been recorded.

213. S. Sophia L. (Kirchner, ‘Beitrage,’ pp. 20-1; MacLeod, Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 200; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, I1; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 26-7, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr.
Ins.,’ p. 231.)—Although this plant attains a height of 1 m., and possesses a many-
flowered inflorescence, it is not very conspicuous, for the diameter of the individual
flowers is only 3 mm., and their colour is yellowish-green. The petals are only
half as long as the sepals (see Fig. 30), and scarcely differ from them in colour,
so that they have almost entirely lost their original function. Stigma and stamens—
according to my own investigations—mature simultaneously, and have the same
relative position as in the last species. The nectaries are also in the same position,
judging from my observations on plants in the North Frisian Islands, though
Velenovsky figures an irregular swollen nectary covering the entire base of the
flower. Kerner states that there is slight protogyny, but that the difference between
the times of maturation of stamens and stigma only amount to a few hours.
Automatic self-pollination is therefore inevitable. Warnstorf (Verh. bot. Ver., Berlin,
xxxvii, 1895) describes the pollen-grains as pale yellow, ellipsoidal, very finely
papillated or almost smooth, 18-19 » broad and 25-31 p long.

CRUCIFERAE 93

VistTors.—I observed the following on the Island of Féhr.—

Diptera. (a) Muscidae: 1. Anthomyia sp. 9; 2. Sepsis sp.; 3. Themira minor
lal. (b) Syrphidae: 4. Syritta pipiens Z., all skg.

Von Fricken in Westphalia and East Prussia (also Redtenbacher at Vienna)
observed the Chrysomelid Colaphus sophiae Scha//.: and Schiner noted Thereva
anilis Z. in Austria.

214. S. austriacum Jacq.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z.; 2. E. nemorum Z.; 3.
Pipiza festiva Mg., skg.; 4. Syritta pipiens Z.; 5. Syrphus albostriatus Fal/., skg.
B. Hymenoptera. Apidae: 6. Andrena dorsata K. 9, skg. and po-cltg.; 7. A.
nitida Fourc. 9, po-cltg.; 8. A. propinqua Schenck 9, skg. and po-cltg.; 9. A. tibialis
XK. 9, skg. and po- cltg. ; ; 1o. Apis mellifica Z.¥, skg.; 11. Melecta armata Pz. 9, skg.;
12. Nomada lineola Pz. 9, skg.; 13. Osmia caerulescens L. 9, skg. and po-cltg. ;
14. O. fulviventris Pz. é and 9, “kg.

215. S. orientale L. (=S. Columnae /Jacg.).—

VisiTors.—Friese observed the following Apzdae at Fiume (F.), Trieste (T.),
and in Hungary (H.).—

1. Andrena carbonaria Z., not infrequent (F.); 2. A. decorata Sm. (H.); 3. A.
hypopolia Per. (H., freq. FE); 4. A. limbata Zz. (H.); 5. A. morio Brullé (F., fa
H.); 6. A. nobilis Aor. 9 and é (H.), not infrequent; 7. A. scita £v. (H.), not
infrequent; 8. A. sisymbrii /rzese, infrequent (F.); 9. A. suerinensis F’rzese (H.), not
infrequent; 10. A. tibialis A., 2nd generation (H.); 11. Nomada chrysopyga Mor.
(H.), freq. ; 12. Osmia bisulca Gers?. (F., H.); 13. O. fulviventris Pz.; 14. O. panzeri
Mor. (F., H.), freq.; 15. O.solskyi Aor. (F.).

216. S. acutangulum DC. (=S. austriacum, var. acutangulum Koch.).—

VisiTors.—MacLeod—in the Pyrenees—noticed 5 short-tongued Hymenoptera,
a Lepidopterid, 3 beetles, 7 hover-flies, a midge, 2 Empidae, and 7 Muscidae (B.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 392-3).

217. S. pinnatifidum DC.—

VistTors.— MacLeod observed Halictus in the Pyrenees (op. cit., p. 393).

218. S. strictissimum L.—

Visirors.—Loew observed the honey-bee skg. in the Berlin Botanic Garden.

60. Stenophragma Celak.

Flowers small, white, homogamous to slightly protogynous, with half-concealed
nectar. There are six nectaries at the insertions of the stamens, but only the two at
the bases of the short ones are functional, the other four being vestigial.

219. S. Thalianum Celak. (=Sisymbrium Thalianum Gawd.). (Herm. Miiller,
‘Weit. Beob.,’ II, pp. 202-3; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, Il; Kirchner,
‘Flora v. Stuttgart,’ p. 294; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896 ; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 27.)—The secretion of the nectaries at the bases
of the short stamens collects in the sepals underneath them, but sometimes nectar is
absent. The pollen-covered surfaces of the anthers of the long stamens surround
the stigma, so that automatic self-pollination is inevitable. Kirchner states that
the stamens are variable: as a rule all six are present, the length of the short ones
being four-fifths and one-third that of the long ones, but not infrequently both short

94 ANGIOSPERMAE—DICOTYLEDONES

ones are absent. Kerner noticed slight protogyny. Warnstorf describes this plant as

being homogamous at Ruppin. Hermaphrodite flowers and flowers with aborted

stamens are there found on the same plants, i.e. the plants are gyno-monoecious.

The pistil of the female flowers is bilateral and reddish-brown. The pollen-grains

are whitish, ovoid or ellipsoid, finely granulated, about 30 » long and 25 p» broad.
Visirors.—Hermann Miiller observed the following.—

A. Coleoptera. (a) Curculionidae: 1. Centorhynchus sp. (6) Mordellidae :
2. Anaspis rufilabris Gy//. (c) Nitédulidae: 3. Meligethes, B, Diptera. (a) Zmpidae:
4. Empis vernalis Jfg., skg. (6) Syrphidae: 5. Ascia podagrica /., po-dvg.; 6.
Rhingia rostrata Z., skg. C. Hymenoptera. Apidae: Apis mellifica L. ¥, skg.

MacLeod noticed a Muscid in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p- 200).
In Dumfriesshire a hover-fly was recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’
p- 12).
61. Hugueninia Reichb.

220. H. tanacetifolia Reichb. (= Sisymbrium tanacetifolium J.).—
Flowers homogamous, and smell like honey. Briquet (‘ Etudes d. biol. flor.
d. les Alpes occident.’) says that their diameter is 5mm. The calyx and
the yellow corolla are spreading. The stamens diverge and turn their anthers
horizontally with the dehisced side directed upwards. Kirchner adds that in the
Botanic Garden at Hohenheim, both petals and stamens are erect, so that the
stigma is closely enveloped by the four upper anthers, and automatically self-
pollinated. According to Hildebrand (Ber. D. bot. Ges., Berlin, xiv, 1896), the
plant is self-stenle.

Visirors.—Briquet states that these are flies, wasps, bees, and Lepidoptera,
which chiefly effect self-pollination.

62. Alliaria Adans.

Small white homogamous flowers, with half-concealed nectar. There are four
nectaries, but only the two at the bases of the short stamens are functional—
secreting on their inner sides. The two others—between the bases of the long
stamens of each pair—do not secrete.

221. A. officinalis Andrz. (= Sisymbrium Alliaria Scop.) (Herm. Miiller,
‘Fertilisation, p. 109, ‘ Weit. Beob., II, p. 202; MacLeod, Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 199; Knuth, ‘Bliitenbiol. Herbstbeob.’)—The nectar secreted
by the functional nectaries at first forms four drops at the base of the flower between
the short stamens and the adjacent long ones, and ultimately fills the space
between the bases of the stamens and pistil. As the nectar is secreted inwards,
and not towards the sepals, the latter are superfluous after the flower has opened,
and readily fall away. All the anthers dehisce introrsely, and those of the long
stamens so closely surround the stigma that automatic self-pollination must take
place, and according to Hildebrand this is effective (Ber. D. bot. Ges., Berlin, xiv,
1896). Insects, when probing for nectar or devouring or collecting pollen, must
sometimes effect cross-pollination owing to the relative position of the anthers
and stigma.

CRUCIFERAE 95

Visirors.—The following were observed by H. Miiller (H. M.), Borgstette (B.),
and myself (Kn.).— :

A. Coleoptera. (a) Curculionidae: 1. A minute sp. of Ceutorhynchus (H. M.).
(6) Dermestidae: 2. Byturus fumatus /., very freq., po-dvg. and skg.(?) (H. M.).
(c) Mitidulidae: 3. Epuraea (H.M.); 4. Meligethes, freq. (H.M.). B. Diptera.
{a) Brbionidae: 5. Dilophus vulgaris AZ. 8, skg.(?) (H.M.). (6) Empzdae: 6. Empis
nigricans A/g., freq., skg. (H.M.); 7. E. punctata #/, skg.(H.M.). (c) Muscédae:
8. Anthomyia, skg. (H.M.); 9. Sepsis sp. (H.M.). (d) Syrphidae: 10. Rhingia
rostrata Z., skg. (H.M.); 11. Syrphus decorus J/g. (B.). C. Hymenoptera.
Apidae: 12. Andrena nitida Yourc. 9, skg. (H.M.); 13. Apis mellifica Z. §, skg.
(H.M., Kn.).

Rossler noted a Tineid—Adela rufinitrella Scop.—at Wiesbaden.

Verhoeff saw the following in Norderney.—

A. Coleoptera. (a) Witidulidae: 1. Meligethes brassicae Scop. (6) S/aphy-
lintdae: 2. Tachyporus obtusus Z. B. Diptera. (a) Muscidae: 3. Anthomyia sp.
(6) Syrphidae: 4. Platycheirus peltatus Ag. one 4, po-dvg. C. Lepidoptera. (a)
Tinerdae: 5. Adela cuprella Zhédg. 9.

Ducke noted Andrena tscheki J/or. 9 at Trieste.

In Dumfriesshire a beetle, an Empid, 2 Muscidae, and 2 hover-flies have been
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 14).

63. Braya Sternb. et Hoppe.
Small yellow or white flowers with half-concealed nectar.

222, B. alpina Sternb. (Kerner, ‘ Nat. Hist. Pl.” Eng. Ed. 1, II, p. 248.)—As
in Malcolmia, the insects are kept on the right path to the nectar by two groups
of erect stiff-pointed bristles on the ovary, with the result that they necessarily
touch the pollen-covered anthers with their head or proboscis. The stigma matures
before the stamens, and is visible as soon as the petals of the opening bud separate
a little. In Nova Zemlia—according to Ekstam—the odourless flowers are homo-
gamous or slightly protogynous-homogamous. Self-pollination is possible.

223. B. purpurascens R. Br.—According to Andersson and Hesselman
(‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 34), this
species blooms in Spitzbergen from the beginning of July till the end of August,
and the fruits apparently ripen during the latter month.

64. Erysimum L.

Flowers yellow, homogamous or protogynous, with half-concealed nectar.
Nectaries two or four.

224. E. cheiranthoides L. (Herm. Miiller, ‘ Weit. Beob., II, pp. 203-4;
Kirchner, ‘Flora v. Stuttgart,’ p. 295.)—Two of the four nectaries are vestigial, and
are situated between the roots of the pairs of long stamens, while the two functional
ones are on the inner sides of the bases of the short stamens. ‘These produce so
much nectar that it fills on either side the space between the bases of the short
stamen, adjacent long stamen, and pistil, All the anthers turn their dehisced
pollen-covered surfaces towards the centre, but the short stamens curve outwards,
so as to free the approach to the nectar, and make cross-pollination by insect-

96 ANGIOSPERMAE—DICOTYLEDONES

visitors possible. The four long stamens surround the stigma and secure automatic
spontaneous self-pollination should insect-visits fail.

Visrrors.—Buddeberg—in Nassau—observed a short-tongued bee, Panurgus-
calcaratus Scop., ske. Loew (‘Beitrage,’ p. 30) saw Vanessa urticae Z., skg., in
Silesia, MacLeod—in Flanders—noted a short-tongued bee, a hover-fly, and a
Muscid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 198-9).

225. E, helveticum DC. (Herm. Miiller, ‘Alpenblumen,’ p. 150.)—Flowers
homogamous.

Visitors.—Miiller noticed Muscidae, 3 beetles, and 4 Lepidoptera.

226. E. orientale R. Br. (Knuth, ‘Bliitenbiol. Herbstbeob.’) — Warnstorf
(Verh. bot. Ver., Berlin, xxxviii, 1894) states that the pollen-grains are yellowish-
white, ellipsoidal, and tuberculated, 30-37 » long and 18-21 » broad.

Visrtors.—In the Kiel Botanic Garden I observed hover-flies (Eristalis sp.,
Platicheirus sp., Syritta pipiens Z., Syrphus balteatus Deg.) and Lepidoptera (Pieris.
napi Z.); all skg.

227. E, aureum Breb.—This species is self-fertile (Comes, ‘Stud. s. impoll.
i. alc. piante ’).

228. E. crepidifolium Reichb. (Schulz, ‘ Beitrage,’ Il, pp. 14-15.)—Flowers
bright yellow. The base of each short stamen is surrounded by a quadrangular
or polygonal nectar-secreting ridge, and there are three glandular processes directed
obliquely upwards in front of the bases of each pair of long stamens, the middle one
of the three being immediately opposite the cleft between the two filaments. The
stigma is mature immediately after the flower has opened. At first it projects about
3mm. beyond the long stamens. Subsequently the filaments elongate, the anthers
reaching the stigma, but they dehisce very late, so that at first only cross-pollination
is possible, and self-pollination can only take place towards the end of anthesis.

VistTors.—Schulz observed Lepidoptera, bees and flies, as well as numerous
small beetles (Meligethes): these probably effect both self- and cross-pollination.

65. Brassica L.

Flowers yellow, and homogamous or slightly protogynous. They are usually
aggregated into large inflorescences, being therefore tolerably conspicuous. The
nectar is half concealed, and there are four nectaries, of which two are on the inner
side of the short stamens, and the others between the insertions of the long stamens
of each pair.

229. B. oleracea L. (Herm. Miller, ‘ Fertilisation,’ pp. 111-12, ‘ Weit. Beob.,”
II, p. 204; Kirchner, ‘Flora v. Stuttgart,’ p. 297; Cobelli, Abh. ZoolBot. Ges.,
Wien, xl, 1890, pp. 161-4; Knuth, ‘Bl. u. Insekt. a. Helgoland,’ ‘Bl. u. Insekt. a.
d. nordfr. Ins.,’ ‘ Weit. Beob.,’ p. 231.)—Kerner states that the bright yellow flowers
are open from eight o’clock in the morning till nine o’clock at night. There are four
nectaries, of which two are on the inner side of the bases of the short stamens, and
the two others between the insertions of the long stamens of each pair. The drops
of nectar secreted by the former spread out on each side between the three adjacent
stamens and the ovary, while those secreted by the two others collect on the outside

CRUCIFERAE 97

between the pairs of long stamens, sometimes—says Miiller—in such quantity as to
touch the sepals outside these. Jordan states, however, that the anterior and posterior
nectaries are not functional. The two short stamens are usually not as long as the
pistil, though this is sometimes the case: they curve outwards away from it, and in
doing so turn the pollen-covered sides of their anthers inwards. The four long
stamens remain in the middle of the flower, but make a quarter- or half-turn, so that
the pollen-covered sides of their anthers are turned towards the adjacent short
stamens, or even completely outwards. I found the wild cabbage of Helgoland
to agree with this description of Hermann Miiller’s. Insects when probing for
the nectar secreted by the inner nectaries chiefly effect cross-pollination. The
nectar of the other glands can be got at without touching the stigma, so that these
are probably useless for pollination, as the observations of Jordan appear to indicate
(vide supra). Should insects fail to visit the flowers, the upper parts of the long
stamens usually curve so far towards the stigma as to touch it, thus effecting
automatic self-pollination. Lund and Kjaerskou (Justs Bot. Jahresber., Leipzig, xiii,
(1885) 1887, p. 753) say that this is effective, but the numerous fruits which result
do not usually contain so many seeds as those produced by crossing.

Visitors.—The honey-bee here takes a prominent place. Besides this, I have
observed—in Helgoland—another bee—Andrena carbonaria Z., skg., corresponding
to the flower in size, and—mostly in the same island—the white cabbage-butterfly,
Pieris brassicae Z. On the North Frisian Islands I further noticed various po-dvg.
and skg. Syrphidae—Helophilus, Eristalis, Syrphus, Rhingia—and also a humble-bee,
Bombus terrester Z., skg. A small beetle (Meligethes) is everywhere found po-dvg.
and gnawing the petals, usually without benefit to the flower.

Hermann Miiller (H. M.) and Buddeberg (Budd.) have observed the
following.—

A. Coleoptera. Mitidulidae: 1. Meligethes, very freq., po-dvg. or gnawing
the flower (H.M.). B. Hymenoptera. (a) Apzdae: 2. Andrena fulvescens Sm. 9,
po-cltg. (Budd.); 3. A. fulvicrus KX. 9, po-cltg. (H. M.); 4. A. nana X. 4, skg. (H. M.
5. A. gwynana X. 9, skg. and po-cltg. (M. M.); 6. A. nigroaenea X. 9, skg. (H. mM:
7. Apis mellifica Z. §, skg. and po-cltg. (H.M.); 8. Halictus cylindricus K. 9 (H.M.);
g. H. morio /. 9, po-cltg. and skg. (Budd.); 10. Osmia rufa Z. 6, skg. (Budd.).
C. Thysanoptera. 11. Thrips, freq. (H. M.).

Alfken and Héppner (H.) noticed the following at Bremen.—Afzdae: 1. Andrena
albicans full. 9, not infrequent; 2. A. humilis 7m’. $ (H.); 3. A. argentata KX. 9,
rare; 4. A. carbonaria Z. 9, infrequent; 5. A. convexiuscula X. 9, infrequent; 6. A.
nigroaenea X. 9, not infrequent; 7. A. propinqua Schenck 9(H.); 8. Bombus agrorum
f.9, infrequent; 9. B. derhamellus X. 9, infrequent; 10. B. sylvarum Z. 9, infrequent ;
11. Halictus calceatus Scop. 9, very freq.; 12. H. flavipes /. 9, freq.; 13. H. levis
K. 9, infrequent ; ‘14. H. minutus X. 9, infrequent; 15. H. nitidiusculus X. 9, freq. ;
16. H. punctulatus X. 9, freq.; 17. H. rubicundus Cfr. 9, freq.; 18. H. sexnotatulus
Nyl. 9, rare; 19. Nomada succincta Pz. 9 and 4, skg.; 20. Osmia rufa Z. 9 and
8, not infrequent; 21. Podalirius retusus Z. 9 and 4, infrequent.

Leege saw the following Apzdae on Juist.—1. Colletes cunicularis Z.; 2. Osmia
maritima Frese 4, freq., skg.

Loew observed the following in the Berlin Botanic Garden—Hymenoptera.
Apidae: 1. Andrena carbonaria L. 8, skg.; 2. A. extricata Sm. 9, po-cltg.; 3. Bombus
agrorum F. 9, skg. and po-cltg.; 4. B. lapidarius Z. 9, skg.; 5. B. terrester L. 9 and
¥, skg. and po-cltg.; Osmia rufa Z. 4, skg.

DAVIS. 1 H

2

98 ANGIOSPERMAE—DICOTYLEDONES

MacLeod saw in Flanders,—6 long-tongued and 8 short-tongued bees, 8 hover-
flies, 2 Muscidae, 3 Lepidoptera, and a beetle (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p: 204).

Schletterer noted two bees at Pola,—1. Halictus calceatus Scop. ; 2. H. fasciatellus
Schenck,

Cobelli (Verh. ZoolBot. Ges., Wien, xl, 1890) observed on the flowers of the
variety sabauda, 50 Apidae belonging to the genera Andrena, Anthophora, Apis,
Bombus, Chalcidoma, Chelostoma, Eucera, Halictus, Melecta, Nomada, Osmia, and
Xylocopa; while the later-flowering variety Jofry/is-asparagordes was visited only by
eleven species of Apidae, which were also less numerously represented.

230. B. Rapa L. (Kirchner, ‘Flora v. Stuttgart,’ p. 298; Schulz, ‘ Beitrage,’ I,
pp. 3-4; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 204-5; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 27-8.)}—The golden-yellow, slightly protogynous
flowers agree with the last species as regards the number and position of the nectaries,
but Kirchner says that the actively secreting ones on the inner side of the short
stamens are divided into two tubercles. When the flowers open the anthers are
still unripe, and those of the four long stamens lie close to the already mature
stigma. Before the corolla has fully expanded the anthers dehisce, and at the same
time the filaments make a half-turn so that their pollen-covered sides are directed
outwards. Schulz states that in some cases they make only a quarter-turn. The
anthers of the short stamens remain with their pollen-covered sides turned towards
the stigma, but they are 2-33 mm. below it. When the flower fades, the long
stamens—which project somewhat beyond the stigma—curve a little, so that auto-
matic self-pollination may take place, and according to Kirchner and Hildebrand
(‘Die Geschlechtsvert. b. d. Pf. p. 70), this is effective, though Lund and Kjerskou
(‘Morph. Beskriv. af B. oleracea &c.’), as well as Focke, describe the plant as
self-sterile. Insects visiting the flowers will, as in the last species, chiefly effect
cross-pollination, which results in the production of many seeds.

Vistrors.—Besides the honey-bee (skg. and po-cltg.), I have observed the
following at Kiel:—po-dvg. hover-flies (Helophilus pendulus Z., Syritta, Eristalis
tenax Z., E. nemorum Z., Syrphus), and also the useless guest Meligethes.

Krieger noticed Prosopis communis /Vy/. at Leipzig.

Schmiedeknecht observed the following Agzdae in Thuringia.—1. Andrena
flessae Pz.; 2. A. floricola Zv.; 3. A. dorsata K.; 4. Osmia bicolor Schr. 9;
5. O. rufa Z.; and—according to Piccioli—mentions 6. Andrena florentina Magr.,
for Florence.

Schenck noticed the following Agidae in Nassau.—

1. Andrena albicans J/i//.; 2. A. chrysosceles A.; 3. A. cineraria Z.; 4. A.
combinata CAr.; 5. A. convexiuscula K.; 6. A. extricata Sm.; 7. A. flavipes F.;
8. A. floricola Ev.; 9. A.gwynana K.; 10. A.parvula K.; 11. A. propinqua Schenck ;
12. A. punctulata Schenck; 13. A.nitida Fourcr.; 14. A.trimmerana K.; 15. Halictus
albipes /.; 16. H. interruptus Pz. 9; 17. Nomada alternata A.; 18. N. succincta
Pz.; 19. N. xanthosticta X.; 20. Osmia bicolor Schr.

231. B. Napus L. (Herm. Miiller, ‘ Weit. Beob.,’ II, p. 204 ; Kirchner, ‘ Flora v.
Stuttgart,’ p. 299; Knuth, ‘BI. u. Insekt. a.d. nordfr. Ins.,’ p. 28.)—The mechanism
of the golden-yellow slightly protogynous flowers entirely agrees with that of the last

CRUCIFERAE 99

species. Kirchner states, however, that the protogyny is somewhat more pronounced.
The anthers—which have a little red spot at the tip—dehisce very soon after the
flower opens. The flowers are somewhat larger than in the last species, but owing to
the elongation of the axis of the inflorescence, are less crowded.

Visitors.—I observed the same as in the last species.

Wiistnei saw Andrena carbonaria Z. in the island of Alsen. Alfken observed
the following on B. Napus and B. Rapa at Bremen.—

A. Diptera. Syrphidae: 1. Orthoneura nobilis Fad/.; 2. Platycheirus albimanus
F.; 3. Syrphus venustus JZg., very freq. B. Hymenoptera. Afsdae: 4. Andrena
albicrus K. 9, freq.; 5. A. argentata Sm.9; 6. A. carbonaria Z. 9 and 4, infrequent ;
7. A. cineraria Z. 9, infrequent; 8. A. cingulata F. 9, infrequent ; 9. A. flavipes Pz. 9,
freq.; 10. A. fucata Sm. 9, infrequent; 11. A. nigroaenea K. 9, not infrequent;
12. A. parvula K. 9, very freq.; 13. A. propinqua Schenck 9 and 4, very freq.;
14. A. tibialis A. 9, infrequent; 15. Eriades florisomnis Z. 9 and 4, infrequent;
16. Halictus calceatus Scop. 9, very freq.; 17. H. flavipes /. 9, freq.; 18. H. leucopus
X. 9, infrequent; 19. H. nitidiusculus X. 9, not infrequent; 20. H. rubicundus Cér. 9,
freq. ; 21. H. sexnotatulus Vy/. 9, infrequent; 22. Nomada bifida Ths. 9, infrequent ;
23. N. lineola Pz. 9, skg., infrequent; 24. N. ruficornis Z. var. flava Pz. 9, infrequent ;

25. Osmia rufa Z. 9 and 4, freq.; 26. Podalirius acervorum Z. 9, infrequent; 27. P.
retusus Z. 9 and 4, freq.

Schmiedeknecht noticed Osmia bicolor Schr. 9, in Thuringia.

MacLeod—in Flanders—observed the following on B. napus and B. Rapa.—
Apis, a humble-bee, 6 short-tongued bees, 4 hover-flies, 3 other Diptera, 2 Lepido-
ptera, and a beetle (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 205).

De Vries saw the bee Andrena dorsata A. in the Netherlands.

232. B. nigra Koch. (Kirchner, ‘Flora v. Stuttgart,’ p. 299; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 205-6; Knuth, ‘BI. u. Insekt. a. Helgoland,’
‘Bl. u. Insekt. a. d. nordfr. Ins.’)—The strong kumarin odour and the yellow colour of
the numerous flowers attract many insects wherever the plant is common, as for
instance on the high ground of Helgoland. The yellow sepals project obliquely ;
the petals are erect; the diameter of the flower is 11-12 mm. As the long stamens
are at the same level as and only about 1 mm. from the stigma, pollen can fall upon
it when the flowers are bent by the wind, thus effecting automatic self-pollination.
The iong stamens are turned towards the two short ones, which being 2-3 mm. lower
than the stigma, can never effect self-pollination, but serve for cross-pollination. This
is brought about by numerous insects, which probe for the nectar secreted in the base
of the flower, soon getting dusted with pollen, that they afterwards brush off on the
stigma projecting from the middle of some other flower. There are four green
nectaries of about the same size, one on the inner side of each short stamen and one
on the outer side of each pair of long stamens. They secrete copiously. Kirchner
says that on different stocks the styles are of varying length, so that the stigma
may be either at the level of the short stamens or of the long ones. I have not
observed such variations in Helgoland.

Visttrors.—I observed in Helgoland numerous po-dvg. flies and po-cltg. or
nectar skg. bees.—

A. Diptera. (a) Muscidae: 1. Calliphora erythrocephala, MZg., very freq. ;
2. C. vomitoria Z., freq.; 3. Coelopa frigida Fall, very freq.; 4. Cynomyia
mortuorum Z. 4, freq.; 5. Fucellia fucorum /all., very freq.; 6. Lucilia caesar

H 2

100 ANGIOSPERMAE—DICOTYLEDONES

L., very freq.; 7. Scatophaga stercoraria Z. 9 and 4, very freq.; 8. medium-sized
Muscids. (6) Syrphrdae: 9. Eristalis arbustorum Z. 9 and 4, very freq.; 10. E.
tenax Z. 9 and 4, freq,; 11. Helophilus trivittatus #. 9, occasional; 12. Syritta
pipiens Z., very freq. B. Hymenoptera. Afzdae: 13. Andrena carbonaria Z. 4,
2nd brood. C. Lepidoptera. Rhopalocera: 14. Pieris brassicae Z., occasional.
D. Orthoptera. 15. Forficula auricularia Z., very freq., dvg. the flowers. All
these insects were observed from July 8 to 11, 1895, on the high ground.

Verhoeff noticed the following in Baltrum.—A. Coleoptera. (a) Witdulidae:
1. Meligethes brassicae Scop. (5) Scarabaetdae: 2. Phyllopertha horticola Z.
B. Diptera. Muscidae: 3. Anthomyia sp.

Heinsius saw a Muscid—Scatophaga stercoraria Z. 3, and a hover-Ay—Eristalis
arbustorum Z. g9—in Holland (Bot. Jaarb. Dodonaea, Ghent, iv, 1892).

H. de Vries noticed a humble-bee—Bombus subterraneus Z. ¥—in the Nether-
lands (Ned. Kruidk. Arch., Nijmegen, v, 1877).

233. B. fruticulosa Cyril—tThis species is self-fertile (Comes, ‘ Stud. s. impoll. i.
alc. piante’).

66. Sinapis Tourn.

Flowers yellow, homogamous or slightly protogynous. In some species the
sepals project horizontally, so that the nectar is exposed, but in others it is
completely concealed. Four nectaries, situated as in Brassica.

234. S. arvensis L. (Herm. Miller, ‘Fertilisation, p. 112, ‘ Weit. Beob.’ II,
pp. 204~5; Knuth, ‘BI. u. Insekt. a. d. nordfr, Ins., pp. 28, 149, ‘ Bliitenbiol. Beob.
a. d. Ins. Riigen’; Kirchner, ‘ Flora v. Stuttgart,’ pp. 299, 300.)—As the sepals are
horizontal the nectaries are visible and accessible from the exterior, but the flowers
are so crowded that insect visitors find it more convenient to thrust the proboscis
between the stamens to the nectar, and this is their regular practice. The anthers of
the long stamens at first turn their dehisced sides towards the adjacent short stamens,
but on the third day of flowering direct them upwards, while the filaments curve
downwards, so that if the pollen has not been removed by visitors and pollination
effected, the stigma, by pushing up between the anthers, will be automatically pollinated.
Eggers observed—according to Hansgirg—pseudo-cleistogamy. Jordan says that, as
a rule, only the two nectaries opposite the short stamens are functional. According
to Kerner the flowers are protogynous. Warnstorf (Verh. bot. Ver., Berlin, xxxvii,
1895) describes the pollen-grains as yellow, ellipsoidal, with delicate regular ridges
which cross one another.

Visirors.—Herman Miiller (H. M.), Buddeberg (Budd.), and myself (Kn.) have
observed the following.—

A. Coleoptera. (a) Alleculidae: 1. Gonodera murina Z.(H.M.). (6) Ceram-
bycidae: 2, Leptura livida F., dvg. the anthers (H. M.); 3. Strangalia nigra Z., ditto
(H. M.). (c) Cocczneliidae: 4. Coccinella septempunctata Z., nect.-lkg. (H. M.).
(d) Nitidulidae: 5. Meligethes sp., freq. (H. M., Kn.). (e) Scaradaetdae: 6. Phyllo-
pertha horticola Z., gnawing the flower (H.M.). B. Diptera. (a) Conopidae:
7. Dalmannia punctata /., skg. (H.M.). 8. Myopa buccata Z., skg. (H. M.).
(4) Emprdae: 9. Empis sp., skg. (H. M.). (¢) Muscidae: 10. Lucilia sp., po-dvg.
(H.M.); 11. Scatophaga merdaria #. (H.M.). 12. S. stercoraria Z., po-dvg.
(H.M.). (d) Syrphidae: 13. Chrysogaster macquarti Loew, po-dvg. (H. M.);
14. Eristalis aeneus Scop., skg. and po-dvg. (H. M.); 15. E. arbustorum Z., freq.,

CRUCIFERAE Ior

ditto (H. M.); 16. E. pertinax Scog., not infrequent, ditto (H. M.); 17. E. sepulcralis
L., ditto (H. M.); 18. E. tenax Z., ditto (Kn.); 19. Rhingia rostrata Z., skg. and
po-dvg. (H. M.); 20. Syritta pipiens Z., po-dvg. (H.M.); 21. Syrphus umbellatarum
f., ditto (Kn.). C. Hemiptera. Pentatomidae: 22. Eurydema ornatum Z., boring
into the flower and skg. (Budd.). D. Hymenoptera. (a) Apidae: 23. Andrena
albicrus X. 4, very freq., skg. (H. M.); 24. A. cingulata F. 4, skg. (H. M.); 25. A.
dorsata A. 9, skg. and po-dvg. (H. M.); 26. A. nana XK. 4, skg. (H. M.); 27. Apis
mellifica L. %, freq., skg. (H. M., Kn.); 28. Bombus lapidarius Z. ¥, skg. (H. M.);
29. Eriades nigricornis JVy/. 5, skg. (Budd.); 30. Halictus leucozonius KX. 9, skg.
(H. M.); 31. H. malachurus X. 9, skg. and po-cltg. (H. M.); 32. H. sexnotatus X. 9,
occasional, skg. (H. M.); 33. H. sexsignatus Schenck 9, occasional, skg. (H. M.);
34. Nomada alboguttata H.-Sch. var. pallescens H.-Sch. 9, freq., skg. (H. M.); 35.
Prosopis hyalinata Sm. 4, skg. and po-cltg. (H. M.); 36. Prosopis confusa JVy/. 9,
ditto (H. M.). (4) Zenthredinidae: 37. Cephus pygmaeus Z., nect.-lkg. and po-dvg.
(H.M.). C. Lepidoptera. (a) Moctutdae: 38. Euclidia glyphica Z. skg. (H. M.).
(6) Rhopalocera: 39. Pieris napi LZ. (H. M.), and 40. P. rapae Z. (Kn.); both skg.

I noted the following in Helgoland (Bot. Jaarb. Dodonaea, Ghent, viii, 1896,
P- 38).—

Diptera. (2) Muscidae: 1. Calliphora vomitoria Z., po-dvg. (6) Syrphidae:
2. Eristalis tenax Z., skg.; and on Riigen: A, Diptera. (a) SyrphAzdae: 1. Eristalis
anthophorinus Ze/f.d; 2. E. arbustorum Z. 6; 3. E. pertinax Z.; 4. E. sepulcralis Z.;
5. E. tenax Z.; 6. Helophilus floreus Z.; 7. Syrphus pyrastri Z.; and 8. S.ribesii Z.; all
skg. and po-dvg. (5) Zabanidae: 9. Chrysops caecutiens Z.é. B. Hymenoptera.
Apidae: 10. Andrena carbonaria Z. 9; 11. Apis mellifica Z. ¥; 12. Bombus terrester
ZL. %; and 13. Halictus rubicundus CAr. 9; all skg. and po-cltg. C. Lepidoptera.
Rhopalocera: 14. Vanessa atalanta Z.; 15. V. urticae Z.; and 16. Pieris sp.; all skg.

Alfken observed the following in Bremen.—

Apidae. 1. Andrena albicans Mull. 9; 2. A. carbonaria Z. 9; 3. A. denticulata
K. 9; 4. A. flavipes Pz. 9; 5. Eriades florisomnis ZL. 9.

Heinsius saw the following in Holland (Bot. Jaarb. Dodonaea, Ghent, viii, 1896,
Pp. 63-5)-—

2 hover-flies—Eristalis arbustorum Z. 9, and E. horticola Deg. 9, a butterfly—
Pieris brassicae Z. 9, a short-tongued bee—Andrena carbonaria Z. 9—freq., and 4
long-tongued bees—Podalirius acervorum JL. 9, Apis mellifica Z. ¥, Bombus hortorum
L. %, and B. lapidarius Z. 9.

H. de Vries (Ned. Kruidk. Arch., Nijmegen, v, 1877) observed Apis mellifica
L.%: MacLeod records 5 hover-flies, a Muscid, and a Lepidopterid for Flanders (Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 207): and Schletterer noticed a ruby-wasp—
Arge cyanocrocea Fors/.—at Pola.

235. S. Cheiranthus Mert. et Koch, y. montana DC. (=Brassica montana
DC.). (MacLeod, ‘ Pyreneenbl.’)—The claws of the yellow petals are so closely held
together by the sepals as to form a tube 9-11 mm. long. The nectar is therefore
only legitimately accessible to the thin proboscis of Lepidoptera. Two of the four
nectaries—the smaller ones—are always at the bases of the short stamens. The two
larger ones are at the bases of the long stamens, but they secrete no nectar and (as in
the case of Diplotaxis muralis) may be reached from without by clefts between the
sepals. The smaller nectaries, on the other hand, are functional, and can only
be reached by two narrow passages at the sides of the stigma. When the thin
proboscis of an insect is introduced, it first touches the anthers of the four long

102 ANGIOSPERMAE—DICOTYLEDONES

stamens—which have dehisced introrsely, and are somewhat higher than the stigma—
and then the stigma with its opposite side, so that cross-pollination is promoted.
Crossing is subsequently rendered even more easy, for these anthers turn their
pollen-covered surfaces upwards.

Visitors.—MacLeod observed—in the Pyrenees—a butterfly—Anthocharis belia
Cr., var. simplonia Freyer, skg.—adapted to the structure of the flower.

236. S. alba L. (Hildebrand, Bot. Jahrb., Leipzig, xii, 1881, p. 26; Kirchner,
‘ Beitrige,’ pp. 22-3.)—The crowded golden-yellow flowers exhale an odour like that
of vanilla. The claws of the petals are 5 mm. long and at first erect; when the limbs
are expanded the diameter of the flower is 15mm, The stigma, and the anthers of
the long stamens, project 2-3 mm. beyond the corolla. Automatic self-pollination
does not take place, however, for the anthers turn their dehisced sides outwards, away
from the stigma. The two short stamens are 3-4 mm. below the stigma, with their
pollen-covered sides facing inwards. Two of the four nectaries are internal to the
bases of the long stamens.

VisiTors.—On cultivated plants at Kiel I noticed the honey-bee, skg., and also
a hover-fly—Eristalis tenax Z., po-dvg.

67. Erucastrum Presl.

The yellowish flowers are homogamous or slightly protogynous, with exposed
nectar. Four nectaries.

237. E. obtusangulum Reichb. (Kirchner, ‘ Beitrage,’ pp. 22-3.)—The sepals
of plants at Zermatt project horizontally, as in the case of Sinapis arvensis, so that
the nectar is accessible to insects from the outside. But the flowers are so
crowded that it is more easily reached from above. There are four functional
nectaries, of which two—according to Velenovsky’s figure—are broad and flat,
and situated on the inner sides of the bases of the two short stamens, projecting
between the claws of the petals. The latter are 5 mm. long, upwardly directed,
and closely apposed laterally. When the limbs are fully expanded the diameter
of the corolla is about 12 mm. Before dehiscence the anthers have a dark red
spot at the tip. They all turn the dehisced side inwards, but are remote from
the stigma, which is simultaneously mature, and is placed just above the entrance
of the flower, so that when the flowers are in the erect position, automatic self-
pollination cannot usually take place. The lower ends of the anthers of the long
stamens are at the same level as the stigma; those of the two short stamens are
somewhat lower.

Visttors.—MacLeod observed—in the Pyrenees—6 bees, 6 Lepidoptera, a
beetle, 5 Syrphidae, one of the Bombyliidae, and a Muscid. (‘Pyreneenbl.,’ p. 392.)

68. Diplotaxis DC.
The moderately large yellow flowers have a pleasant odour, and are homo-
gamous, with half-concealed nectar. Four nectaries.

238. D. tenuifolia DC. (MacLeod, ‘Untersuchungen ii. d. Befrucht. einiger
phanerog. Pfl. d. belg. Flora,’ II; Kirchner, ‘Flora v. Stuttgart,’ p. 301; Schulz,

CRUCIFERAE 103

‘ Beitrage,’ II, p. 15; Herm. Miller, ‘ Alpenblumen,’ p. 150.)—Only the two smaller
of the four nectaries are functional. These are situated on the inner side of the
bases of the two short stamens. The two others are much larger, are placed
external to the bases of the pairs of long stamens, and face obliquely outwards.
The two sepals opposite the functional nectaries are erect, while the two others
are horizontal. The short stamens turn the dehisced sides of their anthers inwards ;
the anthers of the long stamens are directed towards the short ones. Insects
probing for nectar must therefore usually effect cross-pollination. Failing insect-
visits self-pollination results from contact between the anthers and stigma.

Visttors.—Hermann Miller observed 2 Muscidae, 2 Apidae (Halictus), and a
Lepidopterid in the Alps. Schulz noticed numerous flies and Lepidoptera, more
rarely Hymenoptera and beetles. MacLeod—in Flanders—saw 2 hover-flies, and a
Lepidopterid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 202).

239. D. muralis DC. (Kirchner, ‘ Beitrage,’ pp. 23-4.)—The flower mechanism
essentially agrees with that of the last species, but all four nectaries are functional,
and the four sepals all oblique. The flowers have a diameter of 16-20 mm. The
limbs of the petals are so broad that they somewhat overlap. When the flower
opens the anthers are ripe, and the stigma fully mature. The latter is at first either
rather lower than the anthers of the four long stamens, or just at the same level.
These anthers turn their dehisced sides outwards, but automatic self-pollination
is inevitable, for they are coated almost all round with pollen, and quite near the
stigma. When the flowers have completely opened the stigma projects beyond
the anthers of the long stamens, so that cross-pollination is favoured when insects
visit the flowers. The anthers of the two short stamens are turned inwards, and
situated about 3 mm. below those of the long ones.

Visirors.—MacLeod observed the following in Flanders.—Apis, a species of
Halictus, 4 hover-flies, a Muscid, and a Lepidopterid (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 203).

69. Eruca DC.

Large yellowish homogamous flowers, with half-concealed nectar. Four nectaries.

240. E. sativa Lam. (Hildebrand, ‘ Vergleich. Untersuchungen ii. d. Saftdr. d.
Cruciferen’; Kirchner, ‘ Beitrage,’ p. 21.)—The delicate whitish-yellow petals with
their dark-brown veins spread out into a cross about 25 mm. in diameter. The
flowers are homogamous. The anthers dehisce introrsely, and are placed so close to
the stigma that automatic self-pollination is inevitable. Of the four nectaries only the
two large flat ones at the inner sides of the short stamens are functional, while the two
others—situated outside the bases of the two pairs of long stamens—do not secrete.

7o. Vesicaria Lam.
Yellow flowers, with half-concealed nectar.

241. V. arctica R. Br.—Warming observed in Greenland that fruits were set
even at a height of 700 m. Nothing is known about the flower mechanism.

242. V. utriculata L.—Briquet (‘Etudes d. biol. flor. d. les Alpes occident.’)
gives the diameter of the yellow corolla as 15 mm., and states that the claws of

104 ANGIOSPERMAE—DICOTYLEDONES

the petals along with the sepals form a tube 1-14 mm. wide and 15 mm. long.
At the bases of the two short stamens there are four nectaries, of which the secretion
collects in the bottom of the tube just mentioned. Cross-pollination predominates,
for the stigma projects somewhat beyond the anthers of the four long stamens—
which are in the entrance of the flower—so that automatic self-pollination is usually
impossible. The two outer stamens are sometimes as long as the four inner ones.
The flowers examined by Kirchner (Bot. Centralbl., Cassel, lxix, 1897, p. 20, note)
were odourless; they exhibited slight protogyny, and their diameter was 15-22 mm.

71. Alyssum Tourn.

Flowers rather small, yellow, homogamous to protogynous, with half-concealed
nectar. Usually four nectaries. There is sometimes no nectar.

243. A. calycinum L. (Kirchner, ‘Flora v. Stuttgart,’ p. 304; Kerner, ‘ Nat.
Hist. Pl.’ Eng. Ed. 1, II; Knuth, ‘ Bloemenbiol. Bijdragen.’)—The small flowers—
1-5-2 mm. in diameter—are devoid of nectar; they are at first bright yellow, but
subsequently colourless. The erect sepals closely ensheath the claws of the petals.
The anthers dehisce introrsely, and as those of the short stamens are at the same
level as the stigma, while those of the long ones project beyond it, automatic
self-pollination is inevitable. Kerner says that there is at first slight protogyny,
so that in the early stage insect-visits must effect cross-pollination, but failing these
the stamens incline towards the stigma towards the end of anthesis, and autogamy
results.

Visttors.—In the Kiel Botanic Garden I have observed Syritta pipiens Z. skg.;
Herm. Miiller (‘ Weit. Beob.,’ I, p. 327)—in Thuringia—saw one of the Conopidae—
Myopa testacea Z.—skg.

244. A. montanum L. (Kerner, ‘Nat. Hist. Pl.) Eng. Ed.1, II; Schulz,
‘Beitrage, II, p. 15.)—The homogamous yellow flowers are rather small, and smell
like honey. There are four functional nectaries, two in the angle at the bases of
the short stamens, and two between the long stamens of each pair. The anthers are
usually at the same level as the stigma, which matures at the same time. In
fine weather the petals and stamens spread out to some extent, when cross-pollination
is likely to result from insect-visits. In dull weather and at night the petals and
stamens are closely applied to the pistil, so that automatic self-pollination must
result. Kerner says that the conspicuousness of the inflorescence is greatly enhanced
by growth of the petals after the flowers have opened.

Visitors.—Schulz noticed flies. Hermann Miiller observed the following in
his garden.—

A. Coleoptera. Telephoridae: 1. Dasytes plumbeus A/i//., freq. B. Diptera.
(a) Muscidae: 2. Sp. of Anthomyia, freq., skg.; 3. Lucilia cornicina /, skg. per-
sistently. (4) Syrphrdae: 4. Eristalis sepulcralis Z., in large numbers, skg.; 5. Syritta
pipiens Z., freq., skg. and po-dvg. C. Hymenoptera. (a) Apidae: 6. Halictus
nitidiusculus A. 9, freq., skg. and po-cltg.; 7. Nomada ruficornis Z., skg.; 8. Prosopis
5, in large numbers, skg. (6) Sphegidae: g. Cerceris rybiensis Z., not infrequent, skg.

Friese saw in Hungary the rare species Andrena tscheki J/or. (=A. nigrifrons
Sm.). Ducke records from Trieste, Andrena tscheki Afor. 9, and A. (Biareolina)
neglecta Dours 6.

CRUCIFERAE 105

245. A. alpestre L. (Kirchner, ‘Beitrige,’ pp. 25-6.)—The homogamous
flowers smell of honey, and those growing on the ‘yellow wall’ at Zermatt are of
a golden-yellow colour. Their diameter is 3-4 mm. The four nectaries are on
either side the bases of the two short stamens. The anthers of the four long stamens
are at the same level as the stigma—which matures simultaneously—and both
project about 1 mm. above the entrance to the flower, in which the anthers of
the two short stamens are situated. The anthers dehisce introrsely and remain
facing inwards, but they are so far removed from the stigma that automatic
self-pollination is not completely assured.

246. A. saxatile L—

Visirors.—Loew observed the Syrphid Eristalis sepulcralis Z., skg., in the
Berlin Botanic Garden.

72. Aubrietia Adans.
247. A. Columnae Guss.—
VisrTors.—Loew saw the honey-bee skg. in the Berlin Botanic Garden.
248. A. spathulata DC.—
Visirors.—As A. Columnae.

73- Berteroa DC.

White homogamous flowers with half-concealed nectar. Four nectaries.

249. B. incana DC. (Schulz, ‘ Beitrage,’ I, p. 4; Kirchner, ‘ Flora v. Stuttgart,’
p- 304; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 209; Warnstorf,
Verh. bot. Ver., Berlin, xxxviii, 1896; Knuth, ‘ Bltitenbiol. Herbstbeob.,’ ‘Bloemen-
biol. Bijdragen.’)—There are four nectaries, one internal to either side the base
of each short stamen. The anthers of the long stamens project a little beyond
the stigma, and turn towards the anthers of the short stamens as soon as the flower
opens. The long stamens reach the level of the stigma, but their filaments are
bent in such a way as to remove them tolerably far from it. As, however, the
anthers of these stamens are usually slightly incurved at the tip, automatic self-
pollination is easily possible. The open passage to the secretion above each nectary
is narrowed by the tooth-like process of a short stamen, so that when the proboscis
of an insect is introduced it must be inserted between a short stamen and the
adjacent long one, which explains the rotation of the long stamens at the entrance
of the flower.

Warnstorf says that the flowers are protogynous, and that the stigma is already
mature in the half-open blossom, the long stamens being at about this stage much
shorter than the style, and with unripe anthers. When the petals expand these
stamens elongate and project a little beyond the stigma, so that autogamy can readily
take place. The pollen-grains are yellow, ellipsoidal, finely tuberculated, about
35 » long and 15 p» broad.

VisiTors.—On garden plants at Kiel I observed the following skg. hover-flies.—
Eristalis arbustorum Z., E. nemorum Z., Rhingia rostrata Z., Syritta pipiens Z., and
Syrphus ribesii Z.; salo a butterfly—Vanessa io. Z. Warnstorf noticed bees at
Ruppin ; and Alfken saw the bee Halictus brevicornis Schenck 9, skg., at Bremen.

106 ANGIOSPERMAE—DICOTYLEDONES

74. Lunaria L.

Large violet homogamous flowers, devoid of odour, and with concealed nectar.

250. L. annua L. (=L. biennis M€nch.). (Knuth, ‘ Bloemenbiol. Bijdragen,’
and Bot. Centralbl., Cassel, Ixx, 1897, pp. 339-40.)—The flower mechanism closely
resembles that of Matthiola incana, but the corolla-tube is only ro mm. long, so that
the nectar is accessible to comparatively short-tongued insects. The calyx is deeply
cordate at the base, and the sepals are closely apposed, so as to hold the claws
of the violet petals together to form a tube. The anthers of the four long stamens
project half-way out of the entrance of the flower, turning their closely-juxtaposed
pollen-covered sides inwards. Self-pollination may, therefore, take place automatically
by the fall of the pollen, or may be effected by insect-visits. The two short stamens
arch outwards at their bases, thus leaving room for the nectaries lying internal to them,
as well as for the secretion. The anthers of the two short stamens—like those
of the long ones—mature simultaneously with the stigma, and turn their dehisced

Fic. 31. Lunaria annua, L. (from nature). (1) Flower, natural size: A, pouch at the base of the
calyx; a', anthers, half projecting from the entrance of the flower. (2) Stamens and pistil, after removal
of sepals and petals: the short stamens diverging, natural size; a!, anthers of the long stamens; a”, anther
of one of the short stamens; _/ the lower part of the pistil visible between the filaments of the long stamens
which envelope it; 7, nectary with drop of nectar. (3) As (2), but x2.

sides towards it. Self-pollination by means of the short stamens is, however, scarcely
possible, for the filaments of the long stamens closely surround the pistil, thus
protecting the stigma from contact with the anthers of the short ones. When an
insect visits more than one flower the pollen of these short stamens is introduced
between the filaments of the long ones—which are pressed apart—dusts the stigma,
and effects cross-pollination.

The proboscis of an insect probing for nectar gets covered with pollen from
the two short stamens, for it has to be thrust between the inner sides of these
and the outer sides of the long stamens. Only a proboscis at least ro mm. long
can secure all the nectar, but one of half that length can reach that part of it which
ascends to the middle of the corolla-tube.

CRUCIFERAE 107

Small insects collecting or devouring pollen can only obtain it from the anthers
of the long stamens, which project somewhat from the flower, and in doing so
they may effect self-pollination by shaking down pollen upon the stigma. Failing
insect-visits, autogamy results automatically from the fall of pollen.

Visttors.—In the Garden of the Ober Realschule at Kiel I have observed
nectar-skg. butterflies —Vanessa urticae Z., and Pieris brassicae Z. 6. These
regularly fly from flower to flower and thus effect cross-pollination. I have further
seen, skg., Anthophora pilipes /. 4, which has a longer proboscis than any other
of our spring bees; and also Bombus lapidarius Z. 9 and ¥. Several honey-bees
made persistent efforts to suck nectar, and as they visited numerous flowers suc-
cessively I was able to detect the sucking movements, so that they evidently succeeded
in reaching the nectar with their proboscis (6 mm. long), and effected cross-pollination
in the same way as Lepidoptera. A small pollen-collecting bee—Andrena gwynana
. 9—occasionally effected self-pollination, as did a pollen-devouring hover-fly—
Syritta pipiens L.

251. L. rediviva L.—

Visirors.—Loew saw the honey-bee sucking the flowers of this species in the
Berlin Botanic Garden.

75. Schievereckia Andr.

Flowers with half-concealed nectar. Four nectaries.

252. S. podolica Andrz. (Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, I, pp.175, 339.)—
There is a nectary on either side the base of each short stamen. Self-pollination
is at first precluded by protogyny. Even after the dehiscence of the anthers it is
at first prevented, for the stamens remain at a distance from the stigma. Towards
the end of anthesis autogamy takes place, as the stamens bend towards the middle of
the flower.

Visttors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Mitidulidae: 1. Meligethes aeneus /., nect-lkg. B. Diptera.
Syrphidae: 2. Eristalisaeneus Scop., po-dvg. C. Hymenoptera. Apzdae: 3. Andrena
parvula X. 9, skg. and po-cltg.; 4. Apis mellifica Z. ¥, skg.; 5. Halictus nitidiusculus
K. 9, skg. and po-cltg.

76. Petrocallis R. Br.

Rose-coloured homogamous flowers, with half-concealed nectar. Four nectaries.

253. P. pyrenaica R. Br. (Schulz, ‘ Beitrage,’ LH, p. 16.)—There is a richly
secreting nectary on either side the base of each short stamen. At the beginning of
anthesis the anthers of these stamens are at the same level as the stigma—which
is already mature—but the filaments curve outwards at their bases, so that there
is no contact. The filaments of the long stamens are parallel for half their length,
and then bend outwards. Their anthers are curved downwards, and lie almost
immediately above those of the short stamens. They cannot easily effect automatic
self-pollination.

VisiTors,—Schulz—in the Tyrol—saw numerous flies and Lepidoptera, which
in many cases would effect self- as well as cross-pollination.

108 ANGIOSPERMAE—DICOTYLEDONES

77. Erophila DC.

Small white flowers, homogamous to slightly protogynous, with half-concealed
nectar. Four nectaries.

254. E. verna E. Meyer(=Draba verna Z.). (Herm. Miiller, ‘Fertilisation,’
p. 105, ‘ Weit. Beob.,’ I, p. 327; Hildebrand, ‘Die Geschlechtsvert. b. d. Pfl.. p. 70 ;
Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, Il; Kirchner, ‘Flora v. Stuttgart,’ p. 305;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 28.)—The four small green nectaries are
situated on either side the base of each short stamen. The pollen-covered sides of
the four long stamens lie close to the stigma—which matures simultaneously—and as
they discharge a cloud of pollen when even slightly shaken automatic self-pollination
is inevitable. Hildebrand’s experiments prove this to be effective. The anthers of
the short stamens are lower than the stigma, and serve for cross-pollination. While
Miiller describes the flowers as homogamous, Kerner says that they are protogynous
at the beginning of the first day of anthesis, though later on that day the anthers
dehisce and automatic self-pollination is effected by bending of the stamens towards
the middle of the flower.
Kerner also states that
the petals increase great-
ly in size during anthesis.
The flowers open about
nine in the morning, and
close about six o’clock in
the evening.

Jordan distinguishes
between __ short-fruited
and long-fruited forms
or varieties of Erophila.
The former have the
structure just described ;
in the latter the stigma

Fic. 32. Draba aizotdes, L. (after Herm. Miller). 4. Flower seen from projects beyond the an-
above. B Flower after removal of calyx and corolla. C. Nectaryand thers, and it therefore
the bases of the filaments. 4, drops of nectar; #, nectary; 4, short stamen;

Z, long stamens ( 7). frequently happens that
no fruits are set.

Visirors.—Owing to the smallness of the flowers insect-visits are rare. At Kiel
I only saw the honey-bee, skg. and po-cltg. Hermann Miiller—in Westphalia—
besides the honey-bee, observed 2 small short-tongued bees—Andrena parvula X. 9,
and Halictus sp., skg—and also a few po-dvg. Muscidae—Anthomyia sp., Hylemyia
cinerella #/g., and Sarcophaga carnaria LZ.

Alfken observed the following at Bremen.—

A. Hymenoptera. <Apzdae: 1. Andrena parvula KX. 9, po-cltg. and skg.,
$ skg.; 2. Apis mellifica Z. ¥, po-cltg. and skg.; 3. Bombus terrester Z. 9, skg.;
4. Halictus calceatus Scop. 9, skg. and po-cltg.; 5. H. morio F. 9, skg. and po-cltg. ;
6. Halictus nitidiusculus A. 9, skg. and po-citg. B. Diptera. Muscrdae: 7. Musca
domestica L. 4, skg.

CRUCIFERAE 109

MacLeod saw 2 Muscids in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p. 210), and Burkill (‘Fertlsn. of spring fis.’) noticed—on the Yorkshire coast—
a minute short-tongued Dipterid, skg. nectar. In Dumfriesshire (Scott-Elliot, ‘Flora
of Dumfriesshire, p. 17) a beetle and 2 flies have been recorded.

978. Draba L.

Small white or yellow flowers, homogamous or protogynous, with half-concealed
to completely-concealed nectar.

255. D. aizoides L. (Hildebrand, ‘ Vergleich. Untersuch. ii. d. Saftdr. d.
Cruciferen, p. 13; Herm. Miller, ‘Alpenblumen,’ pp. 145-6; Kerner, ‘Nat. Hist.
Pl.” Eng. Ed. 1, II.)—The flowers are at first golden yellow, subsequently whitish.
In the first (female) stage the stigma projects beyond the still unripe stamens. The
anthers only dehisce when these have grown so far that the long ones reach the level
of the stigma towards which they may then incline, effecting automatic self-pollination.
In sunny weather, however, the stamens spread out, rendering the nectar visible, and
cross-pollination is effected by insect visitors.

Vistrors—Hermann Miiller observed flies—7 Muscidae and 6 Syrphidae—
10 Lepidoptera, and a beetle in the Alps.

256. D. Zahlbruckneri Host. (Kirchner, ‘ Beitrage,’ p. 26.)—Self-pollination
is possible in a late stage of anthesis of the golden yellow protogynous flowers, as
a result of the fall of pollen. On either side of each short stamen there is a small
functional nectary.

257. D. Wahlenbergii Hartm. (Herm. Miiller, ‘Alpenblumen,’ p. 146;
Warming, ‘Biol. Optegn. om Grénl. Pl.’)—This species is homogamous. Failing
insect-visits, self-pollination regularly takes place.

In Greenland it produces ripe fruits (Abromeit, ‘Bot. Ergeb. d. Drygalski’s
Groénlands-Exped.,’ pp. 25-6).

258. D. Thomasii Koch.—

Visirors.—Hermann Miiller—in the Alps—observed chiefly flies—3 Muscidae
and a Syrphid.

259. D. frigida Sauter.—Flowers homogamous and capable of self-pollination.
Visirors.—Hermann Miiller observed a Muscid (‘ Alpenblumen,’ p. 147).

260. D. incana L.—Warming says that in Greenland this species is homogamous
and capable of automatic self-pollination. According to the same authority this is also
true in the same country for.—

261. D. nivalis Liljebl., 262. D. corymbosa R. Br., 263. D. arctica J. Vahl,
and 264. D. Hirta L., with its variety Har/m., while its variety leiocarpa Lindbl. is
not so easily automatically self-pollinated. Abromeit (op. cit., pp. 25-6) states that
D. arctica and D. hirta produce ripe fruits in Greenland, where D. nivalis with its
small white flowers ascends to a height of 3,000 feet above sea-level (op. cit.,

pp. 24-5).

I10 ANGIOSPERMAE—DICOTYLEDONES

265. D. aurea M. Vahl—Warming says that this species differs from the pre-
ceding in having the nectar more deeply concealed. The long claws of the petals
are closely apposed to form a tube, so that only insects with a long proboscis can
reach the nectar. In the homogamous flowers automatic self-pollination can only be
effected by the long stamens, while the short ones serve for cross-pollination.

266. D. alpina L.—Lindman says that this species is homogamous—on the
Dovrefjeld—and capable of automatic self-pollination.

According to Ekstam, the protogynous-homogamous flowers in Nova Zemlia
agree in structure with those of Scandinavia and Greenland. The same botanist has
observed plants in Spitzbergen bearing flowers and fruits during the height of
summer (‘ Bliitenbiol. Beob. a. Spitzbergen, p. 20). The flowers were homogamous,
5-7 mm. in diameter, faintly fragrant, and with yellow petals.

Visitors.—In Spitzbergen, Ekstam observed that the flowers were visited by
a number of small Diptera.

267. D. crassifolia L.—Warming says that this species is homogamous and
autogamous in the Arctic regions.

79. Kernera Med.

Small white homogamous flowers, with half-concealed nectar. Four nectaries.

268. K. saxatilis Reichb. (= Cochlearia saxatilis Zam.). (Herm. Miiller,
‘Alpenblumen,’ p. 147.)—The flowers are homogamous. There is a green fleshy
nodule secreting nectar on
either side the base of each
short stamen. The petals are
at first small and erect, but
subsequently expand. The
anthers of the four long
stamens are close to those of
the two short ones: all six
dehisce introrsely, and are so
placed that an insect sucking

Fic. 33. Hernera saxatilis, Reichb. (after Herm. Miller). 4. :
Flower seen from above. &. Flower after removal of two petals : nectar must touch them with

seen from the side; 2, short stamen. (xX 7.) one side, and the stigma with
the other side, thus favouring
cross-pollination. In dull weather the flowers remain half closed, when automatic
self-pollination takes place. The pistil turns to purple-brown in the older flowers.
Visirors.—Hermann Miiller—in the Alps—observed chiefly flies—s Muscidae,
an Empid, and 3 Syrphidae—also several bees—Andrena, and beetles—Meligethes.

80. Cochlearia L.
White, odorous, homogamous flowers, either with half-concealed nectar, or else
nectarless.

269. C. Armoracia. (Herm. Miiller,‘Weit. Beob.,’ IT, p. 198; Kirchner, ‘ Flora
v. Stuttgart,’ p. 305; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896; Kerner, ‘ Nat.

CRUCIFERAE III

Hist. Pl.” Eng. Ed. 1, I1; Knuth, ‘Bloemenbiol. Bijdragen.’)—There are muriform
nectaries at the bases of the stamens in the fragrant flowers, but the secretion is very
scanty. All the anthers dehisce introrsely ; those of the long stamens are at the same
level as the stigma, which matures simultaneously. This being in the entrance to the
flower, cross-pollination is likely to be effected by insect visitors. Self-pollination
may also easily occur, though Kerner says that it is almost or entirely ineffective.
According to Warnstorf, the flowers are protogynous, and the stigmatic papillae
mature before anthesis. The same authority states that all the stamens project
beyond the stigma. The pollen-grains are yellowish, ellipsoidal, tuberculated, on an
average 37-43 w long and 15-19 » broad.

Visitors.— Hermann Miiller (H. M.) and myself (Kn.) have observed the
following.—

A. Coleoptera. (a) Nitidulidae: 1. Meligethes sp. in immense numbers (Kn.,
H.M.). (4) Zelephoridae: 2. Malachius bipustulatus Z., dvg. the anthers (H. M.).
B. Diptera. (a) Bibionidae: 3. Bibio hortulanus L., searching for nectar (?) (H. M.).
(6) Empidae: 4. Empis punctata F., skg. (H. M.). (c) Musctdae: 5. Scatophaga
merdaria /., skg. (H.M.); 6. Sepsis sp. (Kn.). (d@) Syrphidae: 7. Eristalis sp.,
skg. and po-dvg. (Kn.); 8. Syritta pipiens Z., ditto (H. M., Kn.); 9. Syrphus
balteatus Deg. ditto (Kn.). C. Hymenoptera. (a) Apzdae: 10. Andrena albicans
Mill. 9, ditto (H. M.); 11. Halictus levis K. 9, skg. (H. M.); 12. H. zonulus Sm. 9,
skg. (H. M.). (0) Jchneumonidae: 13. Several sp., searching for nectar (H. M.).

270. C. officinalis L. (Knuth, ‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 29, 149.)—
The flowers are 8-10 mm. in diameter. Those I examined appeared to have
no nectaries. Burkill (‘Fertlsn. of spring fls.’), on the other hand, says that on the
Yorkshire coast there are four well-marked nectaries at the base of the flower. The
anthers of the four long stamens are at the same level as the stigma—which matures
simultaneously—and at first somewhat turned away from it. The anthers of the short
stamens dehisce rather later, and in early anthesis are lower than the stigma, though
subsequently they reach its level. Automatic self-pollination is therefore quite
possible. Either cross-pollination or self-pollination may be effected by insect visitors
while collecting pollen, or boring in the base of the flower for sap.

Andersson and Hesselman state (‘ Bidrag till Kanned. om Spetsbergens o. Beeren
Eil. Karlvaxtflora,’ pp. 34-40) that the diverse varieties included by Gelert in this
strongly polymorphous species—as f. groenlandica (L.) Gel., y. oblongifolia (DC.)
Gel., and 6. arctézca (Schlecht.) Gel.—bloom in Spitzbergen from early spring to
autumn, fruits being set from the end of July till autumn. The plant flowers
abundantly on Beeren Island in mid-July. According to Ekstam the flowers of
C. arctica Schlecht. have a diameter of 3-4 mm. and are apparently protogynous-
homogamous. Some of them are female.

Visitors.—I have observed various flies—Syrphidae and Muscidae—and beetles
—Meligethes. Loew saw the honey-bee in the Berlin Botani. Garden. Burkill
(‘ Fertlsn. of spring fis.’) noticed the following on the Yorkshire coast.—

A. Coleoptera. Mitidulidae: 1. Meligethes picipes Sturm, skg. B. Diptera.
Muscidae: 2. Coelopa sp., skg.; 3. Hylemyia sp., skg.; 4. Drosophila graminum Fal.,
skg.; 5. Scatophaga stercoraria Z., skg. and po-dvg.; 6. another small Muscid.
C. Hymenoptera. Jchneumonidae: 7. Ichneumon sp., skg.

112 ANGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 16) a Muscid and
Meligethes have been recorded.

In Spitzbergen Ekstam only saw a medium-sized Dipterid.

271. C. arctica Schlecht.—Ekstam says that in Nova Zemlia self-pollination
is brought about by the closing of the flowers.

272. C.danica L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ ‘Bl. u. Insekt. a.
Helgoland.’)—The diameter of the flower is only 4-5 mm. I could find no nectaries.
The anthers of the long stamens dehisce first, but from the beginning they are
directed towards the stigma, projecting a little beyond it. Those of the short
stamens dehisce soon afterwards, and then all the anthers incline towards the
middle of the flower, thus effecting automatic self-pollination, provided that autogamy
or allogamy has not previously resulted from insect-visits. Autogamy is effective in
any case.

Visitors.—In Schleswig-Holstein I observed (‘ BI. u. Insekt. a. d. nordfr. Ins.,”
p- 149) flies—Syrphidae and Muscidae—and also a nect.-lkg. ant. On Helgoland
(Bot. Jaarb. Dodonaea, Ghent, viii, 1896, p. 38) I also saw the following: Syrphzdae :
1. Eristalis sp.; 2. Syritta pipiens Z. Besides these I noticed—on June 5, 1897—
minute Muscidae persistently working in the bases of the flowers, and obviously
getting sap; but I did not succeed in repeating the observation.

273. C. groenlandica L.—Warming states that there are two non-functional
nectaries. The anthers do not touch the stigma—which matures simultaneously—
but autogamy takes place, apparently by closing of the flower at night or during
unfavourable weather. It must be effective, for numerous fruits are set. Kerner
says that autogamy is brought about as in Schievereckia. (See p. 107.)

81. Eutrema R. Br.

274. E, Edwardsii R. Br.—Ekstam states that the odourless flowers are homo-
gamous in Nova Zemlia. Their diameter in Arctic Siberia is usually 5 mm., according
to Kjellmann. Self-pollination is possible.

Andersson and Hesselman state (‘ Bidrag till Kanned. om Spetsbergens o. Beeren
Ejil. Karlvaxtflora,’ p. 34) that this arctic species ceases flowering in Spitzbergen
in the middle of July, but it is not known whether ripe fruits are produced.
Vanhéffen found a fruiting plant in Greenland on August 20, 1892 (see Abromeit,
‘Bot. Ergeb. d. Drygalski’s Grénlands-Exped.,’ p. 27).

82. Camelina Crantz

Yellow homogamous flowers with half-concealed nectar. Four nectaries.

275. C. sativa Crantz (=C. microcarpa A-rz.). (Kirchner, ‘ Flora v. Stuttgart,’
p. 306; Warnstorf, Verh. bot. Ver., Berlin, xxxvii, 1895, and xxxviii, 1896; Knuth,
‘B). u. Insekt. a. d. nordfr. Ins., pp. 29-30, ‘Bloemenbiol. Bijdragen.’)—There are
four nectaries, two outside the base of each short stamen. The diameter of the corolla
is only 4mm. The anthers of the long stamens are at the same level as and very
close to the stigma, so that they can effect self-pollination. Those of the short stamens
are lower than and curved outwards away from the stigma. They serve for cross-

CRUCIFERAE 113

pollination. Warnstorf describes the pollen-grains as pale yellow, ovoid or ellipsoidal,
very finely papillated, about 37-5 » in length and 27-5 » in breadth.

VisiTors.—On cultivated plants at Kiel I only observed Meligethes.

83. Subularia L.

Minute, homogamous, often cleistogamous flowers. I could not see any nectaries.

276. S. aquatica. (Axell, ‘Om Anord. for Fanerog. Vaxt. Befrukt.’; Knuth,
‘BI. u. Insekt. a. d. nordfr. Ins.,’ p. 30; Hiltner, ‘ Untersuch. ii. d. Gatt. Subularia’ ;
Hildebrand, ‘ Die Geschlechtsvert. b. d. Pfl.’)—This plant, which flowers under the
water, is cleistogamous. In the land variety I investigated the pollen-covered anthers
were very close to the stigma. Hiltner says that in the submerged—and consequently
cleistogamous—flowers there are large stigmatic papillae, which receive the pollen
directly. More seeds are set than in the chasmogamous variety living on river-banks.

Visitors.—In Dumfriesshire a fly was recorded (Scott-Elliot, ‘Flora of Dumfries-
shire,’ p. 17).

84. Thlaspi Dill.

White or lilac, homogamous and protogynous flowers, with half-concealed
nectar. Four nectaries.

277. T. arvense L. (Herm. Miiller, ‘ Weit. Beob.,’ II, pp. 198-9; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 211; Kerner, ‘ Nat. Hist. PL.’ Eng. Ed. 1,
II, p. 335; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896 ; Knuth, ‘BI. u. Insekt.
a. d. nordfr. Ins.,’ p. 30.)\—The small white flowers have a green fleshy nectary on
either side the base of each short stamen. The anthers of the four long stamens are
at the same level or a little higher than the stigma, which matures simultaneously.
They turn their pollen-covered surfaces towards the stigma, and are so near it that
automatic self-pollination is inevitable. The anthers of the two short stamens are
somewhat lower than the stigma, to which they also turn their dehisced sides. They
are further from the stigma, and serve for cross-pollination by insect agency.

Warnstorf says that all the stamens project beyond the stigma, and that their
introrsely dehiscing anthers are inclined over it, making autogamy inevitable. The
pollen-grains are yellowish-white, ellipsoidal, tuberculated, about 25-30 » long and
20-23 p broad.

According to Kerner there is slight protogyny, but automatic self-pollination
takes place later by contact of the anthers with the stigma.—Hieronymus observed
cleistogamous flowers.

Visitors.—Hermann Miller observed the following in Westphalia —A. Diptera.
Muscidae: 1. Anthomyia sp. 9; 2. Pollenia rudis *. B. Hymenoptera. Apzdae:
3. Andrena parvula X. 9, skg. and po-cltg.; 4. Apis mellifica Z. ¥, skg.

278. T. perfoliatum L. (Kirchner, ‘ Flora v. Stuttgart,’ p. 307.)—The flowers
are even smaller than those of the last species, and therefore more inconspicuous;
especially as the petals bend outwards but little. The conspicuousness of the
inflorescence is, however, enhanced by the fact that the petals do not fall off as soon
as fertilization is effected. The flower mechanism agrees with that of T. arvense.

DAVIS. I 1

114 ANGIOSPERMAE—DICOTYLEDONES

In dull weather the flowers are closed or only slightly open, while even in sunshine
they expand so little that their opening is only about 1 mm. in diameter.

279. T. montanum L. (Kirchner, ‘ Beitrage,’ pp. 26-7; the description refers
to plants from the Swabian Alps.)—The white flowers are of considerable size, with
confluent nectaries. The anthers of the four long stamens are at the same level as
the stigma—which matures simultaneously—and turn their pollen-covered surfaces
towards it. The anthers of the short stamens, which are at a somewhat lower level,
dothis also.

280, T. alpinum Crantz. (Kirchner, ‘ Beitrage,’ p. 27; the description refers to
specimens from the Riffelberg, near Zermatt.)—The nectaries are confluent, as in the
last species, forming a studlike projection at the base of the flower, into which the
stamens are inserted. The flowers are white, and when fully expanded their diameter
is 7mm. Although they are homogamous, automatic self-pollination is prevented by
projection of the stigma about 1 mm. beyond the anthers of the long stamens. All
six anthers turn their pollen-covered surfaces inwards: those of the long stamens
project a little from the entrance of the flower, while those of the short ones are about
1 mm. within it.

Visirors.—Pollination can only be effected by insects, but regarding these we
have as yet no knowledge.

281. T. alliaceum L.—Kerner (‘ Nat. Hist. Pl.’ Eng. Ed. 1, I, p. 333) says
that the flowers are protogynous, but that at a later stage automatic self-pollination
takes place by contact of the dehiscing anthers with the stigma.

282. T. alpestre L.—The homogamous flowers are white in colour, and their
anthers are first yellow, then purple-red, and finally black.

Visitors.—Hermann Miiller observed 9 flies and 2 Lepidoptera (‘ Alpenblumen,’
p. 147). Buddeberg saw 17 bees, 7 flies, 2 saw-flies, a solitary wasp, and a beetle
(Justs bot. Jahresber., Leipzig, xvi (1888), 1890, p. 564).

283. T. rotundifolium Gaud. (Schulz, ‘ Beitrage.’)—The bright violet flowers
are conspicuous among the white dolomitic gravel, on which the plant grows in the
south Tyrol, often in patches a square metre in size. Nectar is secreted abundantly
at the bases of the short stamens, and concealed at a depth of 3-4 mm. The anthers
of the long stamens are usually at the same level as the stigma, and ultimately turn
completely round towards those of the short stamens. They do not touch the
stigma—which matures simultaneously—nor can the anthers of the short stamens
do so, for they are at a lower level. Automatic self-pollination does not therefore
take place.

Visitors.—Schulz observed butterflies—Pieris, Vanessa cardui—and flies.

284. T. corymbosum Gay. (Kirchner, ‘Beitrage,’ pp. 27—8.)—On the Riffelberg,
near Zermatt, the flowers are bright lilac to violet in colour, fragrant, and aggregated
into relatively large inflorescences. The diameter of the individual flowers varies
from 6 to 1o mm. There is slight protogyny; at the beginning of anthesis the
anthers are unripe, while the stigma—which is situated in the entrance of the flower—
is already mature, By the time the flowers are fully expanded the anthers of the
four long stamens have opened, and those of the two short ones do so soon afterwards.

CRUCIFERAE II5

All of them dehisce introrsely, and remain in their original position; those of the
long stamens project a little from the entrance of the flower, while those of the short
stamens—together with the stigma—are in it. Automatic self-pollination is possible,
though the anthers are remote from the stigma.

285. T. praecox Wulf.—

Visitors.—Schletterer observed the following at Pola.—

Hymenoptera. (a) Apidae: 1. Andrena convexiuscula A.; 2. A. deceptoria
Schmiedekn.; 3. A. tscheki Mor. (5) Tenthredinidae: 4. Athalia rosae Z., var. liberta
Klug.

85. Teesdalia R. Br.

Small, white, bilaterally symmetrical flowers, with half-concealed nectar. Four
nectaries.

286. T. nudicaulis R. Br. (Herm. Miiller, ‘ Fertilisation,’ pp. 106-8, ‘ Weit.
Beob.,’ II, pp. 199-200; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 20, ‘ Weit. Beob.
ii. Bl. u. Insekt. a. d. nordfr. Ins.” p. 231.)—Hermann Miller says that during
anthesis the flowers are aggregated to form a flat surface, and the outer petals are
larger than the others, as in the Umbelliferae. In Teesdalia, however, as the flowers
successively fade the axis of the inflorescence lengthens, and the flat surface is drawn
out into a raceme, each flower becoming marginal in its turn. Hence all the flowers
have the outer side of the corolla enlarged, and not only—as in the Umbelliferae and
Compositae—those which are originally at the edge.

/ ve

Fic.34. Jeesdalia nudicaulis, R. Br.(after Herm. Miller). (1) Flower seen from above. (2) An-
terior half of a flower. (3) One of the long stamens with a nectary, seen from outside. (4) One of the

two short stamens, seen from outside. s, innersepal; s', outer sepal; s°, lateral sepal; , inner petal ;
pi, outer petal; a, short stamens; a', inner long stamens; a?, outer long stamens; x, 9, petaloid
appendages of the filaments; 2, nectary; ov, ovary.

The white-tipped sepals enhance conspicuousness, but this is chiefly attained
by the white petals, supplemented by petaloid appendages of the filaments. The
appendages of the four inner stamens closely surround the compressed ovary. Each
of these possesses a small pouch immediately above the middle of the base of the
adiacent petal, which is also pouched. Between the two, is a small green fleshy
nectary, which is functional, and appears to belong to the torus.

The anthers of the four long stamens project somewhat beyond the stigma, while
those of the two short ones are at the same level. All six make a quarter-turn when
the flower opens; each long one turning towards the adjacent short one, while each
of these is directed towards the outside of the inflorescence. The anthers now
dehisce, and the stigma simultaneously matures. Insects probing towards either

12

116 ANGIOSPERMAE—DICOTYLEDONES

of the two outer nectaries, touch two of the adjacent anthers with their head or
proboscis, while those trying to get at one of the two inner drops of nectar, only
come into contact with one anther. In either case they touch the stigma with the
other side of their head or proboscis. They may therefore effect either cross- or
self-pollination. Failing insect-visits, the long stamens effect automatic self-pollination.

Visttors.—-Hermann Miiller observed the following at Lippstadt.—

A. Coleoptera. (a) Chrysomelidae: 1. Cassida nebulosa Z.; 2. Aphthona
nemorum Z., skg.; 3. Chaetocnema concinna JZarsh., skg. (6) Curculionidae: 4.
Ceutorhynchidius pumilio Gy//., skg. (¢) Elater¢dae: 5. Limonius parvulus Pz. (¢)
Hydrophilidae: 6. Paracercyon analis Pk. B. Diptera. (a) Brbionidae: 7. Bibio
laniger Mg., skg. (6) Empidae: 8. Empis sp., skg. (c) Musczdae: g. Onesia floralis
R-D., po-dvg.; 10. Sarcophaga carnaria Z. 9; 11. Themyra putris Z., skg. (d)
Syrphidae: 12. Ascia podagrica F., po-dvg.; 13. Melithreptus sp., po-dvg. C.
Hymenoptera. Apzdae: 14. Halictus flavipes F. 9, skg. and po-cltg.; 15. H. luci-
dulus Schenck 9, ditto; 16. H. morio F. 9, ditto; 17. H. nitidiusculus X. 9, ditto ;
18. H. sexstrigatus Schenck 9, ditto; 19. H.smeathmanellus X. 9, ditto; 20. Sphecodes

ephippus Z., skg.
In the island of Fohr I saw Muscidae; while MacLeod in Flanders noticed
2 Dipterids (Bot. Jaarb. Dodonea, Ghent, vi, 1894, p. 211).

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 17) small flies have
been recorded.

86. Iberis L.

White to lilac homogamous flowers, with half-concealed nectar.

287. I. amara L. (=I. Forestieri Jord.) and 288. I. umbellata L.—The outer
petals of the marginal flowers are twice as large as the inner ones (Kerner).

Visitors of I. amara (= I. Forestieri_/ord.).—MacLeod noticed a Muscid in the
Pyrenees. Alfken observed the following Apzdae, skg., at Bremen.—

1. Andrena albicans A/il.$; 2. A. albicrus K.; 3. A. praecox Scop. $3 4.
Bombus lapidarius Z. 9; 5. B. lucorum Z.9; 6. B. terrester Z.9; 7. Osmia rufa Z.6.

289. I. pinnata L.—Hildebrand states that this species is almost self-sterile
(Ber. d. bot. Ges., Berlin, xiv, 1896).

290. I. saxatilis L—According to Briquet (‘Etudes d. biol. flor. d. les Alpes
occident.’) the sepals are spreading and the corolla white and zygomorphous. The
diameter of the corollas of the outer flowers of each inflorescence is about 5 mm., and
the inner flowers are only about half as large. The stigma is beneath the introrse
anthers of the four long stamens, but as these turn outwards, while those of the two
short ones dehisce laterally, insect agency is necessary for pollination, although the
flowers are homogamous. After fertilization the filaments and style assume a dark
violet colour (according to Kirchner).

Visirors.—Cross- and self-pollination are brought about by flies, wasps, bees,
and Lepidoptera.

87. Biscutella L.

Yellow homogamous flowers, with half-concealed nectar. Four nectaries, of
which, however, only two are functional.

CRUCIFERAE 117

291. B. laevigata L. (Herm. Miiller, ‘ Alpenblumen,’ pp. 148-9.) —The flowers
are aggregated into conspicuous inflorescences. Outside the base of each short
stamen there is a nectary, the secretion of which collects in the cavity of the sepal
immediately below it. There is a non-secreting nodule outside the base of each pair
of long stamens. Each petal possesses a basal lobe each side, the one on the side
next the short stamen being considerably larger than the other, and thus serving as
a nectar-cover for the functional nectaries, leaving free but a small passage to the
secretion. The reduction of the other lobes of the petals is correlated with the
vestigial condition of the remaining nectaries.

The anthers are so placed that every insect approaching the nectar must on
three sides touch a dehisced anther, and on the fourth come into contact,with the
simultaneously mature stigma. An insect passing from flower to flower will therefore
continually effect cross-pollination. Failing insect-visits, the anthers and stigma are

FIG. 35. Srscutella laevigata, L. (after Herm. Miiller). .4. Flower after removal of two sepals and
two petals: seen fromthe side. 2. Petal seen from the inner side. C. Ashort stamen with the two
adjacent petals(x 7). a, one of the lower sepals with a cavity (4) in its base, which serves as a nectar
reservoir; ¢, functional nectary; d, vestigial nectary ; ¢, the two short stamens, which dehisce introrsely ;
J, the four long stamens, which dehisce towards the short ones; g, one of the upper sepals; 4, petals; 2,
ovary ; &, style; /, stigma; 2, larger lobe of a petal ; #, smaller lobe of a petal ; 0, entrance to the nectar.

brought into contact by the closing of the flower, so that automatic self-pollination
ensues.

Visitors.—Herm. Miller observed 23 flies, 5 Hymenoptera, 6 Lepidoptera, and
Meligethes in the Alps. Loew noticed one of the Pyralidae, skg., in Switzerland
(‘ Beitrage,’ p. 56).

88. Lepidium L.

Small white or yellow flowers, homogamous or protogynous, with half-concealed
nectar. Four or six nectaries. The corolla is sometimes absent.

292. L. Draba L. (Kirchner, ‘Flora v. Stuttgart,’ pp. 308-9; Kerner, ‘ Nat.
Hist. Pl’ Eng. Ed. 1, 1, p. 432.)}—The small white flowers are not individually
conspicuous, but many of them are aggregated together. In favourable weather
the flower opens so widely at the beginning of anthesis that its diameter is 6-7 mm.,

118 ANGIOSPERMAE—DICOTYLEDONES

and the six small green nectaries—situated outside and between the bases of the six
stamens—are readily accessible even to short-tongued insects. The anthers project
beyond the stigma and are turned towards it, but automatic self-pollination is at first
prevented by the filaments curving outwards. Insects visiting the flower at this stage
will touch the stigma and the pollen with different sides of their body, and may easily
effect cross-pollination. At a later stage the parts of the flower are somewhat closer
together, reducing the diameter to 4-5 mm. At the same time the anthers approach
the stigma so closely that automatic self-pollination must result. According to Kerner,
the flower is slightly protogynous. The same authority states that the anthers of the
long stamens conceal themselves behind the petals during the first stage of anthesis,
so that they cannot be touched by insect visitors.

VisiTors.—Redtenbacher noticed—in Austria—a Nitidulid, Meligethes lepidii
Mill., and an Oedemerid, Nacerdes viridipes, Schmzd/. Schletterer observed the
following bees at Pola.—1. Halictus interruptus Pz.; 2. H. malachurus &.; and
3. H. minutus A.

293. L. sativum L. (Herm. Miller, ‘ Fertilisation,’ pp. 110-11, ‘ Weit. Beob.,’
II, p. 204; Kirchner, ‘ Flora v. Stuttgart, p. 310; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1,
II, p. 508.)—The white flowers are inconspicuous, but being strongly odorous are
readily discovered and much visited by insects. The four nectaries are situated
in the intervals between the long and short stamens. The anthers dehisce introrsely ,
but in sunny weather they curve so far outwards that automatic self-pollination
cannot take place, though cross-pollination may be effected by insect visitors. In
dull weather, or failing insect-visits, autogamy obtains as a last result by closure
of the flowers. Kerner states that the species is slightly protogynous.

Visitors. — Herm. Miiller (H. M.) and Buddeberg (Budd.) observed the
following.—

A. Coleoptera. (a) Dermestidae: 1. Andrenus pimpinellae #. (H. M.). (4)
Telephoridae: 2. Anthocomus fasciatus Z. (H. M.); 3. Dasytes plumbeus J4/7//. F.
(H.M.); 4. Malachius bipustulatus /., gnawing anthers and petals (H.M.). B. Di-
ptera. (a) Bombyliidae: 5. Argyromoeba sinuata Fall. (H. M.). (4) Muscidae: 6.
Siphona cristata #.(H.M.). (c) Syrphidae: 7. Ascia podagrica F., very freq., skg.
and po-dvg. (H. M.); 8. Eristalis arbustorum Z., skg. and po-dvg. (H. M.); 9. E.
nemorum Z., ditto (H. M.); 10. E. sepulcralis Z., ditto (H. M.); 11. Helophilus
floreus Z., ditto (H. M.); 12. Melithreptus taeniatus JZ., ditto (H. M.); 13. Pipiza
chalybeata JZg., ditto (H. M.); 14. Syritta pipiens Z., freq., ditto (H. M.). C.
Hymenoptera. (a) Aprdae: 15. Andrena carbonaria Z. #. 6 (H.M.); 16. A.
parvula A. 9, skg. (H. M.); 17. Halictus lucidulus Schenck 9, skg. (H.M.); 18. H.
nitidiusculus A. 9, skg. (H. M.); 19. Prosopis bipunctata #. 6, skg. (Budd.); 20.
P. communis JVy/. 6 and 9, ditto (H.M.); 21. P. hyalinata Sm., & and 9, very freq.,
skg. and po-cltg.(H. M.). (2) Chryszdidae: 22. Hedychrum nobile Scop. #.8 (H. M.).
(c) Ichneumonidae: 23. An undetermined sp., occasional (H.M.). (d) Sphegidae:
24. Cerceris rybiensis Z., very freq. (H.M.); 25. Pemphredon unicolor /. (H. M.);
26. Oxybelus bellus Dah/d., freq. (H. M.); 27. O. uniglumis Z., very freq. (H. M.).
D. Lepidoptera: 28. Sesia tipuliformis CZ, skg. repeatedly (H. M.).

294. L. ruderale L. (Kirchner, ‘Flora v. Stuttgart,’ p. 310; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins., p. 30; Warnstorf, Verh. bot. Ges., Berlin, xxxviii, 1896.)—
The small greenish-white flowers only occasionally possess petals. Of the original
six stamens only the two short ones are retained, a small nectary taking the place

CRUCIFERAE 119

of each of the four long ones. The stamens are at the same level as the stigma, and
mature simultaneously. They regularly effect automatic self-pollination, which is
effective, according to Comes (‘Stud. s. impoll. i. alc. piante’). Warnstorf says that
the anthers of the two stamens are pressed against the mature stigma by two sepals
when the flower opens.

295. L. campestre L.—According to Kirchner (‘Beitrige,’ pp. 28-9), the
very small white flowers are only 2 mm. in diameter when expanded. On either
side of the base of each short stamen there is a small green nectary (Velenovsky
figures 6 nectaries). The six anthers—which mature simultaneously—turn their
dehisced sides towards the stigma. The sepals when they wither close together
in such a way as to press all the stamens against the stigma, thus effecting automatic
self-pollination, which is probably indispensable. Kerner says that this species is
slightly protogynous.

296. L. graminifolium L—

VisiTors.—Schletterer observed the following at Pola.—

Hymenoptera. (a) Apzdac: 1. Prosopis genalis Thoms (=P. confusa Fors?.).

(0) Ichneumonidae: 2. Amblyteles litigiosus Wesm. (c) Sphegidae: 3. Pemphredon
unicolor /.

89. Hutchinsia R. Br.

Small, white, homogamous or protogynous flowers, with half-concealed nectar.
Four nectaries.

297. H. alpina R. Br. (Herm. Miiler, ‘Alpenblumen,’ p. 150; Schulz,
‘ Beitrage,’ II, p. 17.)—Hermann Miiller describes the flowers as protogynous, with
persistent stigmas. Only some of the plants he examined—at Albula Hospice—were
capable of automatic self-pollination: in these the four long stamens reached the
stigma. A. Schulz describes the flowers—for the South Tyrol—as homogamous
or nearly so, and automatically self-pollinated by contact of the anthers of the long
stamens with the stigma. Kerner states that autogamy takes place as in Schievereckia

(see p. 107).
go. Capsella Vent.

Small, white, homogamous flowers with half-concealed nectar. Four nectaries.

298. C. Bursa-pastoris Moench. (Herm. Miller, ‘Fertilisation,’ p. 110, ‘ Weit.
Beob.,’ II, p. 204; Kirchner, ‘Flora v. Stuttgart,’ p. 311; Knuth, ‘BI. u. Insekt. a. d.
nordfr. Ins.,’ pp. 31, 149; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The
four nectaries are at the sides of the short stamens. All the anthers remain turned
towards the stigma, those of the four long stamens being at the same level, and so
near it that automatic self-pollination—which is effective—regularly takes place.
Insect visitors may effect either cross- or self-pollination.

Breitenbach observed relatively large female flowers in addition to the herma-
phrodite ones (Justs bot. Jahresber., Leipzig, xii (1884), 1886, p. 676).

Willis (Proc. Phil. Soc., Cambridge, 1893) also observed gynomonoecism and
gynodioecism in England. Burkill’s investigations (‘Fertlsn. of spring fis.’) prove
that cold produces gynodioecism and gynomonoecism in this species. Plants that

120 ANGIOSPERMAE—DICOTYLEDONES

flowered on the Yorkshire coast soon after the rigorous weather of January and
February, 1895, possessed only reduced stamens, and it was not till the beginning
of April that hermaphrodite flowers appeared. After the mild winter of 1896 the
first plants were female, but hermaphrodite ones were seen towards the end of March.
The female flowers had an average diameter of 3 mm., so that in Yorkshire they do
not exceed the hermaphrodite ones in size, though Breitenbach (Kosmos, Lwéw,
iii, 1878, p. 206) observed larger female flowers in Germany.

It was also noticed by Warnstorf that the stamens are frequently reduced in the
first flowers that appear at Ruppin: later in the season only hermaphrodite flowers
are found with anthers at the same level as the stigma, making self-pollination
inevitable. Anna Bateson (‘Effect of cross-fertlsn. on inconspicuous fis.) determined
by culture experiments that the plants produced by crossing are not noticeably larger
than those resulting from self-pollination, but they are somewhat heavier, the relative
weights being as 100: 88.

Visirors——Hermann Miiller in Westphalia (H. M.), Buddeberg (Budd.) in
Nassau, and myself (Kn.), have observed the following.—

A. Coleoptera. Mordellidae: 1. Anaspis rufilabris Gy//.(H.M.). B. Diptera.
(a) Muscidae: 2. Anthomyia, skg. (H. M.). (6) Syrphidae: 3. Ascia podagrica F.,
skg. (H. M.); 4. Chrysotoxum bicinctum Z., po-dvg. (Budd.); 5. Eristalis nemorum
L., skg. and po-dvg. (H. M.); 6. E. sp., ditto (Kn.); 7. Melithreptus pictus J/g., skg.
and po-dvg. (H. M.); 8. M. scriptus Z., ditto (H. M.); 9. M. taeniatus JZg., ditto
(H. M.); 10. Syritta pipiens Z., ditto (H. M., Kn. in Helgoland); 11. Syrphus
balteatus Deg., ditto (H. M.). C. Hymenoptera. (a) Apzdae: 12. Prosopis pictipes
Wyl. 3, skg. (Budd.); 13. P. bipunctata /. 8, skg. (Budd.). (6) Sphegzdae: 14. Sapyga
clavicornis Z., skg. (Budd.). D. Lepidoptera. TZinerdae: 15. Adela violella 7Zr.,
skg. (H.M.). E. Thysanoptera: 16. Thrips, freq. (H. M.).

Schmiedeknecht noticed the bee Andrena distinguenda Schenck in Thuringia, and
Alfken saw A. flavipes Pz. 9, skg., at Bremen. Verhoeff observed the following in
Baltrum.—

Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Cynomyia mortuorum Z. (8)
Syrphidae: 3. Syritta pipiens, Z., po-dvg. and skg.

Von Dalla Torre noticed the bee Andrena rosae Pz. $ in the Tyrol, where
it was also seen by Schletterer. The latter further observed the following at Pola—

Hymenoptera. (2) Apidae: 1. Andrena parvula X.; 2. Eucera longicornis
L.; 3. Halictus malachurus K. (0) Zenthredinidae: 4. Athalia spinarum /.; 5. A.
rosae Z., var. liberta Alug.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 18) Apis, a short-
tongued bee, 3 hover-flies, and 4 Muscidae have been recorded.

MacLeod—in Flanders—observed Apis, 9 short-tongued Hymenoptera, hover-
flies, a Muscid, a beetle, and a Lepidopterid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p. 212), and—in the Pyrenees—a Muscid and a Lepidopterid (‘ Pyreneenbl.,’ p. 396).

299. C. pauciflora K.— Kirchner (Jahresber. Ver. Nat., Stuttgart, v, 1893,
p. 100) states that this species does not secrete nectar in its natural habitat (under
overhanging rocks in the South Tyrol). Individual plants cultivated under very
favourable circumstances possessed very small dark-green functional nectaries on
either side the base of each short stamen.

CRUCIFERAE 121

gi. Aethionema R. Br.

goo. A, saxatile R. Br.—According to Briquet (‘Etudes d. biol. flor. d. les
Alpes occident.’) the upright sepals are edged with white, while the red-veined white
or bright rose-red petals spread out above. The stigma is at first beneath the
anthers, but subsequently the style elongates. Kirchner adds that the flowers are
slightly protogynous, that the diameter of the upper part of the corolla is 3-4 mm.,
and that automatic self-pollination is regularly effected by the anthers of the four
long stamens.

Visitors.—These are flies and small beetles, which chiefly effect self- but
occasionally cross-pollination.

go1. A. grandiflorum Boiss. et Hohen.—Hildebrand says that this species is
self-sterile (Ber. D. bot. Ges., Berlin, xiv, 1896, p. 324).

92. Coronopus Haller.

Small, white, homogamous to protogynous flowers, with half-concealed nectar.
Four nectaries.

goz, C. Ruellii All. (Kirchner, ‘Flora v. Stuttgart,’ p. 312; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 213; Knuth, ‘ Weit. Beob. i. Bl. u. Insekt. a. d.
nordfr. Ins.,’ ‘ Bl. u. Insekt, a. Helgoland’; Warnstorf, Verh. bot. Ver., Berlin, xxxviil,
1896.)—The small white flowers are arranged in dense helicoid inflorescences in
the forks of the branches, especially in the middle of the plant, which is closely
applied to the ground. The diameter of the flower is only 4mm, On either side
each short stamen—and therefore apposed to a petal—there is a relatively large
green nectary, which secretes so abundantly that the base of the ovary glistens
all round. When the flower opens all the anthers are still unripe, and at the same
level as the stigma, which would appear to be already mature. When the petals
expand the stamens curve away from the stigma, on the side facing which they
dehisce almost simultaneously. Cross-pollination may therefore result from insect-
visits. Failing these, however, automatic self-pollination is ensured later on, by
the petals inclining together and bringing the anthers into direct contact with the
stigma. Warnstorf says that two of the sepals bend inwards during anthesis, pressing
the long stamens against the stigma so that autogamy takes place. The pollen-
grains are whitish, ellipsoidal, closely tuberculated, 25-30 » long and 15-18 p
broad.

Visttrors.—In Helgoland I observed 2 small Afuscidae, i.e. Coelopa frigida Pad.
and Fucellia fucorum Fa//., both skg.

93. Isatis L.

Small, yellow, homogamous flowers, with half-concealed nectar. Six nectaries.
303. I. tinctoria L. (Kirchner, ‘Flora v. Stuttgart,’ p. 313 ; Knuth, ‘Bloemen-
biol. Bijdragen.’)—In spite of the smallness of the individual flowers the inflorescences
are very conspicuous owing to their size. Kirchner says that the six nectaries are
situated between the six stamens, which curve outwards so as to be a long way

122 ANGIOSPERMAE—DICOTYLEDONES

from the stigma, and turn their dehisced sides upwards. Insect visitors therefore
chiefly effect cross-pollination.

VisiTors.—On garden plants at Kiel I observed the following. —

A. Diptera. Syrphidae: 1. Syritta pipiens Z. skg. B. Hymenoptera.
Apidae: 2. Andrena parvula X. 9, skg.; 3. Apis mellifica Z., skg. C. Coleoptera.
4. Meligethes.

Loew noticed the following in the Berlin Botanic Garden.—

A. Coleoptera. Zélephoridae: 1. Cantharis rusticus Fal. B, Diptera. (a)
Bibronidae: 2. Bibio hortulanus Z.,skg. (4) Syrphidae: 3. Eristalis nemorum L., skg.

‘

94. Myagrum Tourn.

Small, yellow, homogamous flowers, with half-concealed nectar. Two functional
and two vestigial nectaries.

304. M. perfoliatum L.—Kirchner (‘Flora v. Stuttgart,’ p. 313) says that
the flower possesses a well-developed nectary internal to the base of each short
stamen, while the nectaries belonging to the longer stamens are only represented
by narrow green stripes. Automatic self-pollination is possible and effective.

Visirors.—Schletterer observed the following Apzdae at Pola.—

1. Andrena carbonaria Z.; 2. A. deceptoria Schmiedekn.; 3. A. flavipes Pz. ;
4. A. lucens Jmh.; 5. A. morio Bruli.; 6. A. parvula K.; 7. Halictus levigatus X. ;
8. H. quadricinctus &.; 9. H. scabiosae Rossz.

95. Neslia Desv.

Small, yellow, homogamous flowers, with half-concealed nectar. Two nectaries.

305. N. paniculata Desv.—Kirchner (‘Flora v. Stuttgart,’ p. 314) states that
the two nectaries are but feebly represented by small swellings, upon which the short
stamens are seated. All the anthers turn their dehisced sides towards the stigma.
Automatic self-pollination is easily possible, for, according to Warnstorf (Verh. bot.
Ver., Berlin, xxxviii, 1896), the anthers project a little beyond the stigma. The
same authority (op. cit, xxxvii, 1895) describes the pollen-grains as pale yellow,
ellipsoidal, finely papillated, about 31 yw long and 25 mw broad.

96. Bunias L.
Yellow, homogamous flowers, with half-concealed nectar. Two nectaries.

306. B. orientalis L.—The golden-yellow odorous flowers, which are aggre-
gated into large inflorescences, possess but two nectaries, according to Kirchner
(‘Flora v. Stuttgart,’ pp. 314-15). These are placed on the inner sides of the short
stamens, and their secretion is scanty. The diameter of the flowers is 11 mm.
The anthers of the long stamens project beyond the stigma and turn their dehisced
sides upwards. Those of the two short stamens are at about the same level as
the stigma, but curve outwards away from it, and remain perpendicular. They
dehisce somewhat later than the anthers of the long stamens, and their dehisced
sides are directed inwards. It follows that either cross- or self-pollination may result
from insect-visits. Automatic self-pollination takes place by the fall of pollen from

CRUCIFERAE 123

the anthers of the long stamens upon the stigma, and according to Comes (‘ Stud.
s. impoll. i. alc. piante’) it is effective. Warnstorf (Verh. bot. Ver., Berlin, xxxviii,
1896) describes the flower as protogynous. The pollen-grains are pale yellow,
ellipsoidal, with reticulated ridges, about 44 » long and 25 p broad.

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Bibionidae: 1. Bibio hortulanus Z. 9 andé, skg. (4) Syrphidae:
2. Ceria conopsoides Z., skg.; 3. Eristalis arbustorum Z. B. Hymenoptera.
(a2) Ag:dae: 4. Andrena propinqua Schenck 9, skg. and po-cltg.; 5. Prosopis com-
munis Vy/. 8, skg. (6) Tenthredinidae: 6. Cephus sp. 9.

307. B. Erucago L.—Comes says that this species is self-fertile.

VistTors.—Schletterer observed the following bees at Pola.—

1. Andrena flavipes Pz.; 2. A. nana K.; 3. Halictus fasciatellus Schenck; 4. H.
morbillosus K7rchd.; 5. H. morio F.

97- Cakile ‘Tourn.

Moderately large homogamous flowers, bright violet to almost white, with
concealed nectar. Four nectaries.

308. C. maritima Scop. (MacLeod, Bot. Jaarb. Dodonaea, Ghent, i, 1889;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 31-2, 149-50, ‘ Weit. Beob. ii. BI. u.
Insekt. a. d. nordfr. Ins.,’ p. 231, ‘Bl. u. Insekt. a. Helgoland.’)—The fragrant
flowers possess four nectaries: a larger triangular one between and outside the two
long stamens of each pair, and a small bilobed one internal to each short stamen.
The closely apposed sepals hold the claws of the petals together so as to form
a vertical tube 4-5 mm. long, in which nectar collects, often so abundantly as to
half fill it. The anthers of the long stamens project beyond the corolla, so that
automatic self-pollination may take place by the fall of pollen upon the stigma,
which is placed in the entrance of the flower and matures simultaneously. The
anthers of the short stamens remain enclosed in the flower, and reach the level
of the stigma.

Insects are as likely to effect cross- as self-pollination. When probing for
nectar—as in the case of all Cruciferae—they thrust their head or proboscis between
the stigma and anthers so as to dust only one side of it, provided they work round
the flower, and do not repeat the process. If the other side of the head or proboscis
had previously been dusted in another flower, the stigma will be cross-pollinated.
If an insect thrusts its head into the flower first on one side and then on the
other, self-pollination will result. After several flowers have been visited by an
insect, both sides of its head will be dusted, and every fresh visit will effect cross-
pollination.

Visirors.—I have observed the following at Kiel and in the North Frisian
Islands.—

A. Coleoptera. 1. Meligethes. B. Diptera. (a) Muscidae: 2. Aricia albo-
lineata /all.; 3. Musca domestica Z.; 4. Onesia sepulcralis AZg.; 5. Scatophaga
merdaria /.; 6. S. stercoraria Z.; all po-dvg. (4) Syrphrdae: 7. Eristalis arbustorum
L.; 8. E. pertinax Scop.; 9. E. sp.; 10. E. tenax Z.; 11. Platycheirus podagratus Zeéz. ;
12. Rhingia campestris J7/g.; 13. Syrphus arcuatus fad//.; 14. S. umbellatarum /.;
15. Tropidia milesiformis /a//., all skg. and po-dvg. C. Hymenoptera. (a) Apzdae:

124 ANGIOSPERMAE—DICOTYLEDONES

16. Apis mellifica Z.; 17. Bombus lapidarius Z.; 18. Halictus calceatus Scop. D.
Lepidoptera. (a) Noctuidae: 19. Plusia gamma LZ. (6) Rhopalocera: 20. Epinephele
janira Z.; 21. Hipparchia hyperanthus Z.; 22. Pieris napi Z.; 23. P. rapae Z.;
24. Wanessa urticae Z. (c) Zygaentdae: 25. Zygaena filipendulae Z.; all skg.

On the dunes of Helgoland—where there are no bees—I noticed the following
on July 9, 1895.—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella septempunctata Z. (4) Oede-
mertdae: 2. Nacerdes melanura LZ. (c) Telephoridae: 3. Psilothrix cyaneus Ol.
(=Dolichosoma nobilis Ross?). B. Diptera. (a) Syrphidae: 4. Syrphus arcuatus
Fall.gandé; 5. S. pyrastri Z.9 andé; 6. Eristalistenax Z.; 7. E.sp. (6) Muscidae:
8. Calliphora vomitoria Z.é C. Lepidoptera. Nociuzdae: 9. Plusia gamma /.
All the insects very freq., the beetles po-dvg., the flies and moth skg. nectar.

Alfken and Leege recorded the following from Juist.—

A. Diptera. Syrphidae: 1. Syritta pipiens Z. B. Hymenoptera. (a) Aprdae:
1. Bombus lapidarius ¥ and @, very freq., skg.; 2. B. lucorum Z. ¥ and 4, very freq.,
skg.; 3. B. ruderatus /. 4, infrequent, skg.; 4. B. terrester Z. ¥, skg.; 5. Psithyrus
rupestris /. 8, freq., skg.; 6. P. vestalis Mourcr. 8, freq., skg. (5) Chrysididae:
7. Chrysis ignita Z. (c) Scoltidae: 8. Tiphia femorata /. (d) Pompilidae: 9. Pompilus
chalybeatus Schjédfe; 10. P. plumbeus &. C. Lepidoptera. (a) Prerzdae: 11. Pieris
brassicae Z.; 12. P. napi LZ. (4) Satyridae: 13. Hipparchia semele Z., freq.

Verhoeff observed the following in Norderney.—

A. Coleoptera. (2) Nitidulidae: 1. Meligethes aeneus Z., freq. (6) Scara-
baetdae: 2. Phyllopertha horticola Z. B. Diptera. (a) Bombyliidae: 3. Phthiria
canescens Low. (6) Syrphidae: 4. Eristalis arbustorum L., skg.; 5. E. intricarius Z.,
skg. and po-dvg.; 6. E. tenax Z., one 4, skg. and po-dvg.

MacLeod saw a small nocturnal moth at Blankenberge.

In Dumfriesshire (Scott-Elliot, ‘Flora of Dumfriesshire, p. 19) a Muscid and
Meligethes have been recorded.

98. Rapistrum Boerh.

Yellow, homogamous flowers, with half-concealed nectar. Four nectaries.

309. R. rugosum Bergt. (Kirchner, ‘ Beitrage,’ pp. 24-5; Hildebrand, ‘ Ver-
gleich. Untersuch. ti. d. Saftdr. d. Cruciferen,’ p. 25.)—Of the four nectaries
(figured by Velenovsky), two are actively secreting swellings on the inner side of
the bases of the short stamens: their nectar collects in pouches of the sepals. The
two other nectaries are smaller and papilliform. They lie outside and between
the long stamens of each pair, and secrete but little. The diameter of the flower
is 10 mm.: the erect sepals are 5 mm. long, and hold the claws of the petals—
which are of equal length—in a vertical position. The anthers of the four long
stamens are 1-14 mm. above the entrance of the flower, and the stigma—which
matures simultaneously—is at the same level. Although these anthers turn their
dehisced sides outwards they get covered with pollen all round, and are so near
the stigma that automatic self-pollination no doubt sometimes occurs. The anthers
of the two short stamens only reach the entrance of the flower, where they spread
out to some distance from the pistil, thus serving for cross-pollination (Kirchner).
The species is almost self-sterile according to Hildebrand (Ber. D. bot. Ges., Berlin,
1896).

CRUCIFERAE 125

99. Crambe Tourn.

Moderately large, white, slightly protogynous flowers, with half-concealed
nectar. Four nectaries.

gio. C. maritima L. (Knuth, Bot. Centralbl., Cassel, xliv, 1890, pp. 305-8.)—
The flowers—of which the diameter is 12 mm.—possess an odour of honey, and
are aggregated into large crowded inflorescences. The reddish-white sepals project
outwards and upwards, supporting the expanded petals, the white limbs of which
are almost horizontal. The claws of the petals are at first yellowish-green, sub-
sequently becoming bright violet-red. The filaments and style pass through the
same changes of colour, but the anthers and stigma are yellow throughout anthesis.
The interior of a young mature flower is therefore yellowish-green, while that of
an older one is of a discoloured violet. At the base of each pair of long stamens
there is a large rounded green nectary, to which the drop of nectar remains clinging.
On the inner side of each short curved stamen there is also a much smaller nectary
of the same colour. The filaments of the long stamens are forked, and their anthers
are borne on the branches adjacent to the short stamens. Owing to the forking
of the filaments the heads of nectar-seeking insects have to be thrust into the flower
at defined spots, contact with anthers and stigma being thus secured.

The stigma matures in the bud—while the anthers are still unripe—and at
the beginning of anthesis is situated in the entrance of the flower. The filaments
elongate very soon after, thus bringing up the anthers from beneath the stigma.
Dehiscence then takes place. The anthers of the long stamens are now somewhat
higher than the stigma, while those of the short ones are at the same level.

VisiTors.—Insects while probing for nectar—provided they are large enough to
touch anthers and stigma simultaneously—will regularly effect crossing if they thrust
their heads into the flower only once. The honey-bee behaved in this way. Other
insects of about the same size, e.g. some of the hover-flies (Eristalis tenax Z.,
Syrphus ribesii Z.) did not always do so, and therefore sometimes effected self-
pollination. Another hover-fly (Syritta pipiens Z.) when sucking nectar was too
slender to touch both stamens and carpels, being therefore of no use to the plant.
The same is true for two Muscidae (Borborus sp., and Phora pulicaria /a/.). I also
found in the flowers numerous small po-dvg. beetles (Meligethes brassicae Scop.,
more rarely M. viridescens /.), as well as their larvae. The adults in most cases
effect self-pollination, but may also occasionally bring about cross-pollination. The
larvae of Meligethes are found not only in the fully developed flowers, but also
in the buds, where they are very abundant. They destroy the stamens and pistil,
so that many flowers are infertile. The question arises whether the beetles and
their larvae are to be regarded only as enemies of the plant. I am inclined to
maintain the negative, for these small insects are apparently among the most
important, if indeed they are not the chief pollinators of the species; and if they
and their larvae are present in but small numbers, many flowers would remain
unfertilized, though it is true that many others would escape destruction. On
the other hand, should the beetles be too numerous the damage outweighs the
benefit. The beetles, however, will then suffer proportionately, for only a small
number will be able to develop, and this again will react unfavourably upon pollination.

126 ANGIOSPERMAE—DICOTYLEDONES

A certain intermediate number of beetles will therefore be of use to the plant,
and this mean will always be maintained for the mutual benefit of plant and insect
(cf. vol. I, p. 102).

In Dumfriesshire 2 Muscidae and also Meligethes have been observed (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 19).

gu. C. tatarica Wulf.— According to Kerner’s researches artificial self-
pollination is ineffective.

giz. C. pinnatifida R. Br.—

Visitors.—Loew observed the following in the Berlin Botanic Garden, all skg.—

A. Diptera, Syrphidce: 1. Eristalis arbustorum Z.; 2. Syritta pipiens Z.;
3. Syrphus ribesii Z. B. Hymenoptera. Aprdae: 4. Apis mellifica Z.

313. C. grandiflora DC.—

Vistrors.—Loew observed a hover-fly (Melithreptus scriptus Z., skg.) in the
Berlin Botanic Garden.

100. Raphanus Tourn.

Whitish, homogamous flowers, with half-concealed nectar. Four nectaries.

314. R. Raphanistrum L. (Herm. Miller, ‘Fertilisation,’ p. 113, ‘Weit. Beob.,’
II, p. 205; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 208; Kirchner,
‘Flora v. Stuttgart,’ p. 302; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 32, 140,
‘Bl. u. Insekt. a. d. Ins. Rigen’; Warnstorf, Verh. bot. Ver., Berlin, xxxvii, 1895.)—
The nectaries are situated as in Sinapis arvensis, but owing to the erect position
of the sepals their secretion is not visible and easily accessible from without. The
petals are either white with violet veins, or bright yellow with dark yellow veins.
All the anthers turn their dehisced sides towards the stigma, beyond which the
long stamens project, while those of the short ones are at the same level. Automatic
self-pollination therefore appears to be even more favoured than in Sinapis, but
it is ineffective. Warnstorf describes the pollen-grains as being pale yellow in
colour, ellipsoidal, very finely reticulated and tuberculated, about 37-5 » long and
31 p broad.

VistTrors.—Hermann Miiller (H. M.) in Westphalia, myself (Kn.) in Schleswig-
Holstein, have observed the following.—

A. Coleoptera. Mitidulidae: 1. Meligethes brassicae Scop. (Kn.). B. Diptera.
Syrphidae: 2. Melanostoma gracilis A/g., skg. (Kn.); 3. Rhingia rostrata Z., freq.,
skg. and po-dvg. (H. M.); 4. Syritta pipiens Z., po-dvg. (H. M., Kn.);_ 5. Syrphus
ribesii Z., po-dvg. (H.M.); 6. S. sp. (Kn.). C. Hymenoptera. (a) Apidae: 7. Apis
mellifica Z. §, skg. and po-cltg. (H.M., Kn.); 8. Bombus lapidarius Z., skg. (Kn.); 9.
B. muscorum /. 9, skg. (H.M.); 10. B. pratorum Z., skg. (Kn.); 11. B. variabilis
Schmied. 9, skg. (H.M.); 12. Halictus flavipes #. 9, skg. (H.M.); 13. H. smeath-
manellus A. 9, skg. (H. M.). (6) Zenthredinrdae: 14. Cephus pygmaeus Pz. (H. M.).
D. Lepidoptera. Ahopalocera: 15. Coenonympha pamphilus Z.(H.M.); 16. Rho-
docera rhamni Z. (Kn.); 17. Lycaena sp. (Kn.); 18. Pieris napi Z. (Kn.), and 19.
P. rapae LZ. (Kn.); all skg.

Alfken noticed the bee Halictus nitidiusculus A. 9 at Bremen.

On the island of Riigen I also observed the following, all skg.—

A. Diptera. Syrphidae: 1. Volucella bombylans Z. B. Hymenoptera.
Apidac: 2. Apis mellifica Z. ¥. C. Lepidoptera. Rhopalocera: 3. Pieris sp.;
4. Vanessa urticae L.

CRUCIFERAE 127

Schletterer saw the small green bee Halictus morio /. at Pola.
In Dumfriesshire, Apis, a humble-bee, Muscids, and Meligethes have been
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 19).

315. R. sativus L.'—Kirchner (‘Flora v. Stuttgart,’ pp. 302-3) has described
the structure of the flowers from garden specimens. The petals are white or lilac
with darker veins. The diameter of the expanded flower is about 20 mm. There is
a large cushion-shaped nectary internal to the base of each short stamen, and a
slender peg-like one outside the base of each pair of long stamens. The two outer
sepals possess dilatations at their bases for reception of the nectar. The stamens do
not undergo rotation, but bend horizontally outwards away from the stigma. The
anthers of the four long ones are at the same level as the stigma; those of the
two short ones—which curve more markedly outwards—are 2-3 mm. below it.
When the flower fades the anthers of the long stamens come into contact with
the stigma so that, failing insect-visits automatic self-pollination is effected, which
results in normal fruits being set, though only about half of the ovules become seeds.
Insect visitors chiefly bring about cross-pollination.

Visrtors.—Kirchner observed bees (Apis, species of Bombus), hover-flies, butter-
flies (Pieris), and beetles (Meligethes). Schletterer gives the following Apzdae for
Pola in the Tyrol.—

1. Andrena carbonaria Z.; 2. A. deceptoria Schmiedekn.; 3. A. flavipes Pz.;
4. A. gwynana A.; 5. A. nana &.; 6. A. thoracica /.; 7. Eucera clypeata £7. ;
8. E. longicornis Z.; 9. Halictus calceatus Scop.; 10. H. malachurus K.; 11. Poda-
lirius acervorum Z.; 12. P. nigrocinctus Lep.; 13. P. retusus Z., var. meridionalis

Pér.; 14. Xylocopa violacea Z.

Andrena gwynana KX. was also observed in the Tyrol by von Dalla Torre.
MacLeod noticed 3 flies and 2 Lepidoptera in Flanders (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 209).

1o1. Lobularia Desd.

316. L. maritima Desd.—Hildebrand states that this species is self-sterile
(Ber. D. bot. Ges., Berlin, xiv, 1896).

317. L. nummularia Stend.— According to Kerner (‘Nat. Hist. Pl.,’ Eng.
Ed. 1, II, p. 339), this species is protogynous, but towards the end of anthesis
autogamy is brought about by the stamens moving towards the middle of the flower,
so that the pollen of the long ones reaches the stigma.

102. Clypeola L.
318. Clypeola Messanensis.—As No. 317 (cf. loc. cit.).

103. Sobolewskia Bieb.

31g. S. clavata Frenzl—Hildebrand (Ber. D. bot. Ges., Berlin, xiv, 1896) says
that this species is self-fertile.

1 R. sativus and R. Raphanistrum are one species, according to Carriére (André, Belg. Hortic.,
Liége, xix, 1869, p. 151), and Hoffmann (Bot. Ztg., Leipzig, xxx, 1872; xxxi, 1873; and xii,
1884).

128 ANGIOSPERMAE—DICOTYLEDONES

104. Succowia Medic.
320. S. balearica.—As No. 319.

105. Pugionium Gaertn.

321. P. dolabratum Maxim.—Batalin (Acta horti Petr., St. Peterburg, x, 1889)
describes this species as being protandrous.

X. ORDER CAPPARIDEAE JUSS.

Some of the species belonging to this order are pollinated by humming-birds.
Details will be given elsewhere, when dealing with the extra-European flora,

106, Capparis L.; 107. Cleome Cl.; and 108. Polanisia Rafin.

According to Delpino (‘ Sugli app. d. fecondaz. n. piante autocarp.’), species of
the above three genera are cleistogamous.

XI. ORDER RESEDACEAE DC.
109. Reseda L.

Flowers whitish or yellow, homogamous or feebly protandrous, with half-
concealed to completely concealed nectar. The petals are split into radiating clavate
threads. The torus broadens out at the back into an erect, four-sided disk, the
velvety anterior surface of which serves as a nectar-guide, while its posterior smooth
surface secretes and conceals the nectar. The expanded claws of the posterior
and middle petals protect the nectar from rain and useless visitors (flies). They
closely adjoin the posterior side of the disk, and embrace its upper and lateral
margins with their forwardly directed lobes. Wilson compares the nectary to a pot
of which the lid must be opened by nectar-seeking insects, and short-tongued bees
(Prosopis) are better adapted for this than long-tongued ones. The flower is open
even in the bud, and the beginning of anthesis is marked by the first secretion
of nectar. The ovary projects from the middle of the flower, and serves as the
most convenient alighting-place for insects. It follows that these will regularly effect
cross-pollination when they have previously visited another flower of the same species
(cf. Fig. 36).

322. R. luteola L. (Herm. Miiller, ‘Fertilisation, p. 116, ‘ Weit. Beob.,’ II,
p. 205; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 214-15; Kirchner,
‘Flora v. Stuttgart,’ p. 316.)—The inconspicuous bright yellow flowers—open in
the bud—are aggregated into moderately conspicuous inflorescences. They quickly
wither. The stamens are symmetrically arranged around the pistil, and the three
stigmas project a little beyond them. As there is no movement of the stamens
during anthesis, automatic self-pollination easily takes place. Beyer states that
the stamens dehisce in centrifugal order, unlike those of the other species.

Visitors. — Herm. Miiller (H. M.) and Buddeberg (Budd.) observed the
following.—

RESEDACEAE 129

A. Coleoptera. Anthribidae: 1. Urodon conformis Suffr. (Budd.); 2. U.
rufipes Ov. (Budd.). B. Hymenoptera. 3. Agzdae: Andrena nigroaenea X. 9,
in large numbers, skg. (H. M., Thuringia); 4. Apis mellifica Z. ¥, skg. and po-cltg.
(H.M.); 5. Prosopis bipunctata /. 9 (Budd.); 6. P. communis JVy/. 9 and 4, very
freq. (H.M.); 7. P. hyalinata Sm. 8 and 9, freq., skg. and po-cltg. (H. M.).

323. R. lutea L. (Herm. Miller, ‘ Fertilisation,’ p. 116, ‘Weit. Beob.,’ II,
p. 205; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 213; Warnstorf, Verh.
bot. Ver., Berlin, xxxviii, 1896; Kirchner, ‘Flora v. Stuttgart,’ p. 315; Schulz,
‘Beitrage,’ I, p. 4.}—The odourless flowers are bright greenish-yellow, and either
homogamous (Kirchner) or slightly protandrous (Schulz). The stamens are at first
curved over the pistil. When the secretion of nectar begins some of the anthers
dehisce, and their filaments bend up towards the disk. Kirchner says that the
stigmatic papillae mature at this stage, but Schulz asserts that they do not do so
till all but the innermost anthers have shed their pollen. Failing insect-visits, auto-
matic self-pollination takes place, for the anthers dehisce above the stigma. Autogamy,
however, has little or no result (Darwin, Focke). Besides hermaphrodite flowers,
Schulz observed here and there andromonoecious ones with undeveloped stigmas.
Warnstorf describes the pollen-grains as pale yellowish in colour, ellipsoidal, finely
tuberculated, about 44 » long and 19 p» broad.

VisiTors.—Hermann Miiller observed the following in Thuringia.—

A. Coleoptera. (a) Anthribidae: 1. Urodon rufipes O/v., vainly searching for
nectar. (6) Curculionidae: 2. Baris abrotani Germ.,do. (c) Mordellidae: 3. Anaspis
rufilabris Gy/2. B. Diptera. Musczdae: 4. Ulidia erythrophthalma MMg., do.
C. Hymenoptera. (a) Apidae: 5. Apis mellifica Z.¥, skg. and po-cltg.; 6. Halictus
sp. 9, skg.; 7. Prosopis pictipes /Vy/. 9, skg.; 8. Pr. signata Pz. 9 and 4, very freq.,
skg. (6) Formicidae: g. Lasius niger LZ. ¥,do. (c) Ichneumonidae: 10. Undetermined
sp.,do. (d) Sphegidae: 11. Cerceris arenaria Z., skg.; 12. C. labiata /., freq., skg.;
13. C. rybiensis Z., very freq., skg.; 14. Crabro (Entomognathus) brevis Z. 9 and 4,
res 15. Diodontus tristis Z.9, occasional. (e) Vespzdae: 16. Odynerus parietum
L. 3, skg.

Loew noticed Prosopis sp. in Steiermark (‘ Beitrage,’ p. 51). The following bees
were seen by von Dalla Torre in the Tyrol:—1. Halictus quadricinctus Fr. 9;
2. H. sexnotatus K. 9. MacLeod observed 7 short-tongued Hymenoptera, a Lepi-
dopterid, 2 Syrphids, and a Muscid, in the Pyrenees (‘ Pyreneenbl.,’ p. 396). Smith
records Prosopis bipunctata /. (=P. signata Pz.) for England.

The following were observed by Schletterer at Pola in the Tyrol.—

Hymenoptera. (a) Apidae: 1. Anthidium diadema Z/r.; 2. A. oblongatum
Ltr.; 3. Andrena albopunctata Ross7; 4. A. convexiuscula A.; 5. A. convexiuscula
K. v. fuscata K.; 6. A. flessae Pz.; 7. A. labialis K.; 8. A. morio Brull.; 9. A.
parvula X.; 10. A. thoracica /.; 11. Ceratina cucurbitina Ross7; 12. Colletes lacu-
natus Dours; 13. C. niveofasciatus Dours; 14. Eucera longicornis Z.; 15. Halictus
calceatus Scop.; 16. H. interruptus Pz.; 17. H. quadricinctus /.; 18. H. sexnotatus
K.; 19. Nomada nobilis A. Sch. ; 20. Nomia diversipes La/r, ; 21. Prosopis clypearis
Schenck. (6) Ichneumonidae: 22. Pristomerus vulnerator Pz. (c) Pompilidae: 23.
Pseudagenia albifrons Dalm.; 24. Salius notatus Lep. (d) Sphegidae: 25. Cerceris
arenaria Z.; 26.C.emarginata Pz.; 27. C. quadrifasciata Pz.; 28. C. specularis Costa;
29. Crabro clypeatus LZ. (e) Zenthredinidac: 30. Allantus fasciatus Scop. (f°) Vespidae :
31. Eumenes pomiformis Pz. ; 32. Odynerus parietum Z.; 33. Polistes gallica Z.

DAVIS. If K

130 ANGIOSPERMAE—DICOTYLEDONES

324. R. odorata L. (Herm. Miller, ‘Fertilisation,’ pp. 114-16, ‘ Weit. Beob.,’
II, p. 205; Knuth, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.” p. 231.)—The
flower mechanism agrees with that of the last species. The yellowish-white, odorous,
homogamous flowers attract numerous small bees, which collect pollen or suck
nectar, and effect pollination. Failing insect-visits, automatic self-pollination takes
place, and this is effective.

Fic. 36. Reseda odorata, L. (after Hermann Miller’. (1) Flower before the anthers have dehisced,
seen from the front. (2) Flower after some of the anthers have dehisced, seen from the front. (3) Young
fruit, seen from the side. (4) Left upper petal. (5) Left lateral petal. (6) Left lower petal. a, sepals;
4, petals; ¢, expanded claws of the upper and lateral petals, which embrace the shield-shaped swelling (4)
of the receptacle ; d, undehisced stamens bent downwards; e, dehiscing stamen raising itself; dehisced
and erect stamen; yg, pistil; 4, shield-shaped projection of the receptacle; z, nectary with nectar.

Visirors.—Herman Miiller (H. M.) in Westphalia, and myself (Kn.)in Schleswig-
Holstein, observed the following —

A. Diptera. Syrphidae: 1. Syritta pipiens Z., po-dvg. (H. M.). B. Hymeno-
ptera. (a) Apidae: 2. Andrena nigroaenea KX’ 9, po-cltg.(H. M.); 3. Apis mellifica
L. %, freq., skg. and po-cltg. (H. M., Kn.); 4. Halictus smeathmanellus X. 9, po-cltg.
(H. M.); 5. H. zonulus Sm. 9, po-cltg.(H.M.); 6. Prosopis annularis Sm. 9 (=P.
panzeri Férsf, according to Dalla Torre) (H.M.), 7. P. bipunctata /. 9 and 6, freq.
(H. M.), 8. P. communis WVy/. 9 and 4, very freq. (H.M.), 9. P. hyalinata /. 9 and &
(H. M.), and ro. P. pictipes yl (H.M.); all po-cltg. and skg. (4) Sphegidae:
11. Cerceris rybiensis Z. 9 and 4, skg. and po-dvg. (H.M.). C. Lepidoptera.
12. Pieris sp. (Kn.). D. Thysanoptera. 13. Thrips, very freq. (H. M.).

CISTINEAE 131

Loew noticed Halictus rubicundus Chr. 3, skg., in Mecklenburg (‘ Beitrage,’
p. 41), and the following (op. cit., p. 33) in Silesia:—A. Diptera. Syrphidae:
1. Syrphus balteatus Deg., skg. B. Hymenoptera. Apidae: 2. Apis moellifica
L. 4%, skg.

Schenck observed 3 bees in Nassau:—1. Anthidium oblongatum Zér.; 2. A.
punctatum Zir.; 3. A. strigatum Lr.

Alfken saw the following at Bozen—

A. Hymenoptera. Apzdae: 1. Coelioxys rufocaudata Sm. 9 and 6, not in-
frequent, skg.; 2. Halictus flavipes F. 9, freq., skg. and po-cltg.; 3. Megachile pacifica
Pz. 9, freq. B. Coleoptera. (a) Buprestidae: 4. Acmaeodera flavo-fasciata P7il.
(4) Cerambycidae: 5. Clytus massiliensis Z.; 6. C. ornatus Hébs¢.

Friese records the following Agzdae for Baden (B.), Alsace (A.), Mecklenburg
(M.), Nassau (N.), and Hungary (H.).

1. Prosopis bipunctata 7’. (B., A., M.); 2. P. confusa Vy. (M.); 3. P. dilatata X.

(H.), not infrequent; 4. P. nigrita # (M.,H.); 5. Stelis signata Zr. (N., according
to Schenck).

325. R. glauca L.—The white flowers—which were investigated by MacLeod
in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 397-8)—are aggregated
into conspicuous inflorescences. The nectary is a semicircular white disk on the
posterior side of the ovary. Nectar is secreted by its central part, which is partly
covered by the claw-lobes of the two upper petals, so that the secretion is only
visible from in front. The flowers therefore belong to the class EC.

When the flower opens the four stigmas are mature, and may therefore be
cross-pollinated by insects. The anthers of the upper stamens dehisce subsequently,
and those of the lower ones still later. Automatic self-pollination may result from
the fall of pollen out of the upper anthers on to the stigma.

VisiTors.—MacLeod observed Hymenoptera—6 species of Andrena, 2 of
Halictus, and one of Polistes—and Diptera—Syrphidae and Muscidae.

XII. ORDER CIJSTINEAE DUNAL.

White or brightly coloured pollen flowers, generally large, homogamous or
slightly protogynous. They open only in the sunshine, and usually remain open
but for a short time (a few hours). The absence of nectar is compensated for
by the production of large quantities of pollen. When cross-pollination is not
effected, autogamy is brought about by closure of the flowers. There are often
cleistogamous flowers, as, for example—according to M. Kuhn—in species of
Lechea (Bot. Ztg., Leip., xxv, 1867, p. 67).

uo. Helianthemum Tourn.

Homogamous or protogynous pollen flowers, of which all the stamens are
fertile.
326. H. vulgare Gaertn. (=H. Chamaecistus 4/2/., and Cistus Helianthemum
L.). (Herm. Miiller, ‘Fertilisation,’ p. 117, ‘ Weit. Beob.,’ II, p. 210, ‘ Alpenblumen,’
pp. 161-2; MacLeod, ‘ Pyreneenbl., pp. 124-5 ; Knuth, ‘ Bloemenbiol. Bijdragen’ ;
K 2

132 A NGI OSPERMAE—DICOTYLEDONES

Warnstorf, Verh. bot. Ver., Berlin, xxxvii, 1895, xxxviii, 1896.)—The citron-yellow,
or more rarely white pollen flowers, expand in the sunshine into a disk of from
25 to over 30 mm. in diameter. The numerous stamens radiate from the simul-
taneously mature stigma, so as to be tolerably far removed from it. Cross-pollination
is consequently effected by insects alighting in the middle of the flower and bringing
with them pollen from other blossoms. Even when the flower is half closed the
pollen-covered anthers touch the stigma, so that automatic self-pollination necessarily
takes place if insect-visits fail. The flowers close completely at night and during
rainy weather.

Warnstorf describes the flowers as homogamous or protogynous. The thick
greenish stigma is often mature before the flower has fully expanded. The pollen-
grains are of a beautiful dark yellow colour, biscuit-shaped, with a longitudinal
groove, and covered by lines of very delicate tubercles; about 75 » long and
31 p broad.

Visttors.—MacLeod observed 7 of the smaller bees, 10 Diptera, 2 Lepidoptera
(vainly attempting to suck), and 5 beetles, in the Pyrenees. Hermann Miiller saw 5
beetles, 19 Diptera, 13 Hymenoptera, and 16 Lepidoptera, in the Alps. Loew noticed
a hover-fly—Merodon cinereus /’. po-dvg.

Hermann Miiller (H.M.) and myself (Kn.) observed the following in Central
and North Germany.—

A. Coleoptera, (a) Buprestidae: 1. Anthaxia nitidula Z. (H.M.); 2. A.
quadripunctata Z. (H.M.). (6) Bruchidae: 3. Spermophagus cardui Svev., po-dvg.(?)
(H.M.). (¢) Cerambycidae: 4. Strangalia nigra Z., dvg. the anthers (H.M.). (¢) Mor-
dellidae: 5. Mordella aculeata Z., vainly skg. (H.M.). (e) Oedemeridae: 6. Oedemera
virescens Z., po-dvg. (H.M.). (/) Zelephoridae: 7. Dasytes plumbeus AZi//., po-dvg.
(H.M.). B. Diptera, Syrphidae: 8. Ascia podagrica /., po-dvg. (H.M.); 9. Chryso-
toxum fasciolatum Deg., po-dvg. (H.M.); ro. Eristalis nemorum Z., po-dvg.
(Kn.); 11. Helophilus pendulus Z., po-dvg. (H. M., Kn.); 12. Melithreptus scriptus
£., po-dvg. (H. M.); 13. M. taeniatus M/g., po-dvg. (H.M., Kn.); 14. Merodon
aeneus Mg., po-dvg. (H.M.); 15. Syrphus pyrastri Z., po-dvg. (H.M.); 16. S. ribesii
L., po-dvg. (H. M., Kn.). C, Hymenoptera. Agsdae: 17. Andrena fulvicrus KX. 9,
po-cltg. (H.M.); 18. Apis mellifica Z. ¥, freq., po-cltg. (H. M., Kn.); 19. Bombus
lapidarius Z. ¥ and 9, po-cltg. (Kn.); 20. B. agrorum /. ¥, po-cltg. (H.M.); ar.
Halictus sp., po-cltg. (Kn.); 22. H. villosulus A. 9, po-cltg. (H.M.); 23. Prosopis
annularis Sm. 9 (=P. panzeri Fors/, according to Dalla Torre), po-cltg. (H. M.).
D. Lepidoptera, 24. Melithaea athalia Ros/., trying to suck on the wing (H. M.).

Willis (‘ Fls. and Insects in Gt. Britain,’ Part II) observed the Muscid Anthomyia
radicum Z., very abundant, po-dvg., on the seaboard of South Scotland. In
Dumfriesshire numerous Diptera were recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’
p- 20).

Schletterer observed the following Agrdae at Pola in the Tyrol:—1. Andrena
parvula K.; 2. Bombus derhamellus #.; 3. Halictus calceatus Scop.; 4. H. morio
F.; 5. Melitta melanura WVy/.

327. H. alpestre DC. (Herm. Miller, ‘Alpenblumen,’ pp. 160-2; Kerner,
‘Nat. Hist. Pl,’ Eng. Ed. 1, II, p. 310.)—The expanded flower is 12-20 mm. in
diameter, and its mechanism essentially agrees with that of the last species. Miiller
describes the plant as homogamous, Kerner as feebly protogynous. The stamens

CISTINEAE 133

are sensitive, and this favours cross-pollination. Automatic self-pollination is effected
by closure of the flowers at night and during dull weather.

Visttors.—These are similar to those of H. vulgare in the Alps (Herm. Miiller),
but owing to the smaller size of the flowers are less numerous and belong to fewer
species.

328. H. Fumana Mill. (Schulz, ‘ Beitrage,’ II, pp. 17-18.)—Schulz states that
the yellow flowers are homogamous and of variable size. They open only in the
forenoon during sunshiny weather. At first self-pollination is prevented by the position
of the stigma, but as the stamens move inwards either spontaneously or by insect
agency, so as to touch it, automatic self-pollination can be effected.

Vistrors.—Schulz observed bees, flies, and more rarely beetles, in the South
Tyrol.

329. H. oelandicum Wahlenb. (=H. vineale Pers.).—Schulz (‘Beitrage,’ II,
p- 18) describes the flowers as homogamous. They are smaller than those of the
last species, but not so ephemeral. As the style is less bent than in H. Fumana,
there is either contact between anthers and stigma at the beginning of anthesis
or the stigma is placed so that pollen falls upon it.

VisiToRS.—Schulz observed flies, bees, and beetles.

MacLeod saw a Syrphid visiting the yellow pollen flowers in the Pyrenees.

330. H. guttatum Mill. (Verhoeff, ‘Bl. u. Insekt. a. d. Ins. Norderney.’)—
The citron-yellow flowers open only for a forenoon. There are usually dark-brown
pollen-guides on the bases of the petals. The anthers being at a higher level than
the large whitish stigma, automatic self-pollination by the fall of pollen can easily
take place. It is inevitable after the petals have been shed, for the sepals then close
together so as to press the anthers against the stigmas, to which they remain
clinging as the fruit matures. Linnaeus (‘Amoenitates, III, p. 396) observed
cleistogamy in plants brought from Spain and cultivated in Upsala.

331. H. salicifolium Pers.—Linnaeus (op. cit.) observed at Upsala that ripe
fruits were produced when the flowers remained closed.

Visitors.—Schletterer noticed the small bee Halictus morio F. at Pola.

332. H. polifolium DC.—Briquet (‘Etudes d. biol. flor. d. les Alpes occident.’)
states that the flowers open and close repeatedly, owing to movements of the sepals.
The bases of the white petals are of a citron-yellow colour. The numerous yellow
stamens are crowded into a bundle in the middle of the flower; their anthers are
introrse, but while shedding their pollen turn their dehisced sides more or less
outwards. There is no nectar. The style is curved like the letter S, and the stigma
is large. The irritability of the filaments—which has long been known—persists
throughout anthesis, and affects all the stamens equally. It is most marked in dry
weather at a temperature of 18-25°C. Within 1-5 seconds after a filament is
touched it has passed from an almost erect position into a nearly horizontal one,
owing to the bending of a zone about 4 mm. long immediately above its insertion.
After about 15 seconds it gradually resumes its original position, and can then be
stimulated afresh. The motile mechanism resembles that of the pulvini of Mimosa
leaves. Its oecological significance is found in the fact that when insects (humble-

134 ANGIOSPERMAE—DICOTYLEDONES

bees and bees) alight upon the corolla they irritate the stamens, which move outwards
and dust them with pollen, that they frequently transfer to the stigmas of other
flowers, especially those in which the style protrudes laterally from the bundle of
stamens. Automatic self-pollination is usually excluded, for the stigma projects
o-5-0-7 beyond the anthers of the surrounding stamens. Besides the ordinary plants
with hermaphrodite flowers only, andromonoecious ones have been observed, bearing
male flowers in the proportion of one to fifty or eighty of the usual kind. Kirchner
says that the male flowers are smaller than the others, with fewer stamens, and no
pistil.

333. H. canum Dun.—According to Briquet (op. cit.), the protogynous pollen
flowers of this species, which are visited by humble-bees and bees, open and close
by movement of the sepals. The filaments are not irritable. The diameter of the
bright yellow corolla is 12-13 mm. when fully expanded. Kirchner says that
automatic self-pollination very rarely takes place, as the flowers are protogynous
and the anthers extrorse.

334. H. kahiricum Delile, and 335. H. Lippii Pers—These two Egyptian
species frequently possess cleistogamous flowers, according to Ascherson (Bul. soc.
linn., Paris, i, 1880, pp. 250-1; Sitzber. Ges. natf. Freunde, Berlin, 1880, pp. 97-108).

336. H. villosum Thib., and 337. H. ledifolium L.—Ascherson (Sitzber.
Ges. natf. Freunde, Berlin, i, 1880) says that the flowers of these species open and
are cross-pollinated only in the morning. If crossing is not then effected, self

pollination is brought about by closure of the flower. The same applies to species
of the genus

mz. Cistus Tourn.

338. C. hirsutus L. and 33g. C. villosus L.

Visttors of C. villosus—Schletterer observed the following bees, po-cltg., at
Pola.—

1. Andrena convexiuscula A.; 2. A.cyanescens /Vy/.; 3.A.nana X.; 4. Halictus
calceatus Scop., var. obovatus K.; 5. H. fasciatellus Schenck; 6. H. interruptus 72. ;
4. H. levigatus K.é; 8. H. minutus A.; 9. H. quadrinotatus K.; 10. H. scabiosae
Ross?; 11. H. tetrazonius Avug ; 12. H. varipes Mor.

340. C. monspeliensis L._—

VisiTors.—Schletterer observed the following bees at Pola—

1. Andrena cyanescens /Vy/.; 2. A. morio Brud/.; 3. A. nana K.; 4. Ceratina
cucurbitina Rossz; 5. Colletes lacunatus Dours; 6. Halictus calceatus Scop.; 7. H.
minutus A.; 8. H. morio #.; 9. H. quadrinotatus .; 10. H. scabiosae Ross?; 11.
Prosopis clypearis Schenck ; 12. P. genalis Zhs.; 13. P. variegata F.

341. C. salvifolius L. (Knuth, ‘BI. u. Insekt. a. d. Ins. Capri.’)—The flower—
which smells slightly of jessamine—expands in the sunshine into a disk 5 cm. in
diameter. There are yellow pollen-guides on the bases of the white petals. The
anthers of the numerous stamens have already dehisced when the flower opens, and
the large capitate, strongly papillated stigma is simultaneously mature. The stamens
at first lie upon the recurved petals, but subsequently become erect, so that the
anthers are brought above the stigma, when automatic self-pollination may take

VIOLARIEAE 135

place by the fall of pollen. This kind of pollination is inevitable when the flowers
close at night, or in dull weather.

Visitors.—In Capri I noticed that a beetle—Oxythyrea squalida Scop—was
the almost exclusive visitor and pollinator, the pollen readily adhering to its hairy
body. Less frequently a bee of medium size—Halictus sp.—made its appearance
and collected pollen on the tibiae of its hind-legs. Both insects usually first alighted
on the stigma, and therefore effected cross-pollination.

Schletterer observed the following at Pola.—

Hymenoptera. (a) Apidae: 1. Andrena cyanescens /Vy/.; 2. A. dubitata
Schenck; 3. A.nana K.; 4.A.parvula K.; 5. Halictus interruptus Pz. (4) Pompzlidae :
6. Pompilus rufipes Z. (c) Zenthredinidae: 7. Amasis laeta F.

XIII. ORDER VIOLARIEAE DC.

The most important genus of this order is—

12. Viola Tourn.

The species of this genus mostly possess large brightly-coloured flowers, in
which yellow, violet, and blue predominate. The anterior (lower) petal is spurred,
giving the flowers their characteristic form, from which alone we might infer that
they were adapted to particular groups of insects. Most species of violets and the
like are bee flowers, Diptera and Lepidoptera playing a comparatively unimportant
part as pollinators. In certain species, however—e. g. V. calcarata—the spur is so
long that only the proboscis of Lepidoptera can reach the nectar. ‘There are, on the
other hand, violets with so short a spur—e. g., V. biflora—that they must be described
as fly flowers. The species of Viola therefore chiefly belong to the flower class Hb,
but some of them are included in the classes L and F. All are homogamous.

The anther of each of the two lower stamens possesses—as Sprengel long ago
admirably described—a nectar-secreting process, which projects into the spur of the
corolla where the secretion is stored. The connective of each of the five stamens
is produced into a membranous appendage. As these appendages overlap one
another laterally and also clasp the style underneath the stigma, they form
a conical chamber, into which the dry pollen falls when the anthers dehisce.
The stigma projects beyond this cone, and closes the entrance to the flower, so that
an insect probing for nectar must first touch it, and then raise it up so as to open
the anther cone, from which pollen falls on to the upper surface of its proboscis.
And since visitors are in the habit of thrusting the proboscis only once into each
flower, they must regularly effect cross-pollination. In many species cleistogamous
flowers with vestigial corollas have been observed, as well as the ordinary open ones.
(Cf. Vol. I, pp. 51, 58.)

342. V. odorata L. (Sprengel, ‘Entd. Geh.’, p. 394; Herm. Miller, ‘ Fertilisa-
tion,’ p. 119, ‘ Weit. Beob.,’ II, p. 209; Hildebrand, ‘ Die Geschlechtsvert. b. d. Pf.’ ;
Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, II, p. 200; Schulz, ‘Beitrage, II, p. 205;
MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 221-2, Arch. biol., Paris-
Bruxelles, vii, 1886 ; Knuth, ‘ Bloemenbiol. Bijdragen’ ; Kirchner, ‘ Flora v. Stuttgart,’
p. 318; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The inconspicuousness

136 ANGIOSPERMAE—DICOTYLEDONES

of the dark-blue flowers, which are almost hidden by the leaves, is partly compensated
for by their powerful fragrance. The corolla is whitish in the middle, the white patch
on the lower (spurred) petal being traversed by dark-blue veins, which converge to
the opening of the spur, and serve as nectar-guides. The end of the style—which
bears the stigma—is dilated distally, bent down like a hook, and at some little
distance from the lower petal. MacLeod states that a fluid is secreted in the stigmatic
cavity, a drop of it being pressed out when an insect pushing into the spur touches
and raises the stigma. This drop moistens the head of the insect and prepares it
for the reception of the dry, white, smooth pollen, of which the separate grains are
about 44 » long and 25 » broad.

According to Hildebrand and Kerner, the lobe on the under-side of the
stigmatic cavity is covered with pollen when an insect visitor pushes its proboscis
into the flower, and when this is withdrawn the lobe is pressed up against the head
of the stigma so as to introduce the pollen into its cavity.

Vis1tors.—Bees are the most important visitors. Sprengel depicts the honey-bee
as the pollinator on the title-page of his ‘Entdeckte Geheimnisse’; and as a matter
of fact it is the most frequent visitor of this violet. Long-tongued bees have also
been observed—especially by Hermann Miiller—sucking nectar and effecting cross-
pollination. This may also be regularly brought about, though more rarely, by
Bombyliidae and butterflies (Vanessa, Rhodocera), which are attracted by the pleasant
odour and suck the nectar. Cross-pollination is essential to the chasmogamous
flowers, for Sprengel’s experiments long ago proved that no fruits are set if insects
are excluded. Short-tongued humble-bees occasionally bite through the spur and
steal the nectar (Schulz).

Kirchner states that—should insect-visits fail—cleistogamous flowers are developed
on the runners during August. These are axillary, with peduncles 3-5 cm. long,
and are bent downwards, sometimes even penetrating the loose soil. Within the
closed calyx there are five small pale petals folded as in a bud, five stamens with
small non-dehisced anthers, the pollen-grains of which send out tubes that penetrate
the stigma. These cleistogamous flowers are fertile: their capsules bury themselves
in the earth—when this is loose enough—and there ripen.

The following insects have been observed by Hermann Miller (H. M.) in
Westphalia, by Buddeberg (Budd.) in Nassau, and by myself (Kn.) in Schleswig-
Holstein.— :

A.Coleoptera. Nrtidulidae: 1. Meligethes(H.M.). B. Diptera. Bombylitdae :
2. Bombylius discolor Akan, skg. (H.M.). C.Hymenoptera. <Apsdae: 3. Andrena
fulva Schr. 9, vainly skg. (H.M.); 4. Anthophora pilipes / 4, skg. (H. M.); 5. Apis
mellifica Z.¥, skg. and po-cltg. (?) (H. M., Kn.); 6. Bombus derhamellus X. 9 (H. M.);
7. B. hortorum Z. 9 (H. M.); 8. B. lapidarius Z. 9, skg. (H. M., Kn.); 9. Halictus
calceatus Scop. 9, vainly skg. (H. M.); 10. Osmia cornuta La/r. 9, skg. (H. M.); 11.
O. rufa Z. 9 and 4, very freq., skg. (H. M., Budd.). D. Lepidoptera. Rzopalocera:
12. Rhodocera rhamni Z., skg. (H. M.); 13. Vanessa cardui Z., very freq., skg.
(H.M.); 14. V. urticae Z., skg. (H. M.).

Schmiedeknecht noticed the following bees in Thuringia :—1. Bombus jonellus
K. 9; 2. B. pratorum Z. 9; 3. Osmia bicolor Schr. 9; 4. O. uncinata Gersi.;
Schenck saw Osmia rufa Z.%, in Nassau; and Alfken observed 8 bees at Bremen,

VIOLARIEAE 137

i,e.:—1, Andrena albicans A/wl/.$; 2. A. albicrus K.5; 3. A. praecox Scop. 6; 4.
Bombus lapidarius Z.9; 5. B. lucorum Z. 9; 6. B. terrester Z. 9; 7. Osmia rufa L.
g and é; 8. Podalirius acervorum Z. 9.

MacLeod observed Apis, 4 long-tongued bees, 2 short-tongued bees, and 3
Lepidoptera in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 222).

Friese records the following bees for Fiume (F.), Innsbruck (I.), Mecklenberg
(M.), Trieste (T.), and Hungary (H.):—1. Osmia acuticornis Duf. ef Per. (=O.
dentiventris Mor., and O. hispanica Schmiedekn.) (F., T., H.); 2. O. bicolor Schr. 9,
skg. (M., sometimes H.); 3. O. cornuta Zyr. (I.); 4. O. pilicornis Sm. (T., H., occa-
sionally M.); 5. O. rufa Z. (M.).

343. V. hirta L.—Kirchner states that the mechanism of the odourless brightly
coloured flowers essentially agrees with that of V. odorata (‘Flora v. Stuttgart,’ p. 318).
They are mostly infertile. Schulz observed that the spur was sometimes bitten
through by humble-bees. The cleistogamous flowers—according to Kirchner—
resemble those of the last species.

Warnstorf describes the pollen-grains as white, irregularly ellipsoidal, smooth,
about 37 » long and 25-30 » broad.

Calloni says that the variety Salvatoriana possesses both chasmogamous and
cleistogamous flowers. The former are visited by bees and butterflies (Argynnis).

344. V. collina Bess.—Kerner says there are cleistogamous flowers in this
species. Schulz observed that the spurs of the chasmogamous flowers were sometimes
perforated by humble-bees.

345. V. sylvatica Fr. (=V. sylvestris Zam., in part)—_The mechanism of the
odourless flowers is similar to that of the last species, according to Miiller (‘ Fertilisa-
tion,’ p. 119). The corolla is violet, with a spur 7 mm. long of somewhat darker
colour. Kirchner states that the cleistogamous flowers of this species—which were
discovered by Corry and Bennett—agree in structure with those of V. odorata, but
the tips of the sepals are bent outwards.

VistTors.—Herm. Miiller observed the following, all skg—A. Hymenoptera.
Apidae: 1. Bombus agrorum F. 9. B. Diptera. Bombyliidae: 2. Bombylius
discolor Mikan. C.Lepidoptera. Rhopalocera: 3. Pieris brassicae L.; 4. P. rapae
L.; 5. P. napi Z.; 6. Rhodocera rhamni Z.; 7. Anthocharis cardamines L.

346. V. Riviniana Reichb.—The flower mechanism agrees with that of V.
sylvatica, but the coroila is larger and of a brighter blue, and the spur is yellowish-
white. Kirchner states that there are both chasmogamous and cleistogamous flowers.

VisrTors.—I observed at Kiel a nect-skg. humble-bee—Bombus agrorum F. 9.

347. V.canina L. (Herm. Miller, ‘Fertilisation, p. 121, ‘ Weit. Beob.,’ II,
p. 209; MacLeod, ‘Bot. Jaarb. Dodonaea,’ Ghent, vi, 1894, pp. 222-3, Arch. biol.,
Paris-Bruxelles, vii, 1886; Kirchner, ‘Flora v. Stuttgart, p. 320; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ p. 33.)—The flower mechanism agrees with that of V.
odorata. MacLeod describes the stigma as swollen in the bud, with a wide opening
and a small valve ; it subsequently grows in a straight line, and finally bends round
into a hook. Failing insect-visits, the flowers are infertile (Darwin). According
to Kirchner, the cleistogamous flowers are situated as in V. odorata, but the petals
have almost completely disappeared, while the stamens are very small, and only the

138 ANGIOSPERMAE—DICOTYLEDONES

two lower ones possess anthers, which are small and contain but little pollen. The
pollen-grains send out pollen-tubes through an opening at the upper end of each
anther-lobe. The capsules of the cleistogamous flowers ripen much more rapidly
than those of the chasmogamous ones.

Visirors.—I observed a humble-bee, a butterfly (Pieris), and Podalirius acervorum
L. at Kiel; all skg. I saw a humble-bee—Bombus lapidarius Z.—in Sylt, skg. the
var. flavicornis Sm., which has conspicuous dark flowers with bright orange spurs.

Hermann Miiller observed the following in Westphalia and Thuringia.—
A. Diptera. Bombylidae: 1. Bombylius discolor Afikan, 9, skg.(?); 2. B. major
L., skg. B.Hymenoptera, Apzdae: 3. Bombus lapidarius LZ. 9, skg.; 4. B. terrester

Fic. 37. Viola calcarata, L. (after Herm. Miller). A. Front view of flower from Piz Umbrail.
B. Side view of the same. C. Front view of flower from Albula. 2. Lower lip of the same with the
nectar-spur. £. Front view of the entrance of 4. F. Section of 4. a, anthers; c, appendages of
the connectives ; 4’, pollen-collecting hairs ; 4°, hairs which in bee-pollinated species of Viola help the bees
to cling to the flower: here they are useless vestiges; 2, stigma; /, lip-like appendage below the entrance
(s¢) to the cavity of the stigma; sd, nectar receptacle. (4—J natural size; E F x 3.)

L. 9, skg. (Thuringia); 5. Osmia bicolor Schr. 9, skg.; 6. O. rufa Z.8, skg. C.
Lepidoptera. Rhopalocera: 7. Anthocharis cardamines Z., skg.; 8. Pieris brassicae
L., skg.; 9. P. napi Z., skg.; 10. P. rapae Z., skg.; 11. Rhodocera rhamni Z., skg.

Alfken saw the following 4 bees at Bremen :—1. Bombus arenicola Zhs. 9; 2.
B. derhamellus A.; 3. B. muscorum /. 9; 4. Podalirius acervorum /. 9.

Verhoeff observed the following in Norderney.—

A. Coleoptera. (a) Mitidulidae: 1. Meligethes coracinus S/, in perforated
spurs. B. Hymenoptera. (a) Apzdae: 2. B. lapidarius Z. 9, skg.; 3. B. terrester
L. 9, skg.; 4. Osmia maritima Friese, one 9; 5. Psithyrus vestalis Mourcr. 9, skg.
C. Lepidoptera. Preridae: 6. Pieris brassicae L., 2 9.

VIOLARIEAE 139

MacLeod noticed 3 humble-bees, Anthophora, an ant, and a beetle in Flanders
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 223):

Burkill (‘Fertlsn. of Spring Fls.’) observed the following on the Yorkshire
coast.—

A. Diptera. Jfuscidae: 1. Cephalia nigripes /g., skg. B. Hymenoptera.
Apidae: 2. Bombus terrester Z., skg.

In Dumfriesshire 2 humble-bees, an Empid, and a hover-fly have been recorded
(Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 21).

348. V. canina x stagnina Ritschl. (Warnstorf, Verh. bot. Ver., Berlin, xxxviii,
1896.)—Both chasmogamous and cleistogamous flowers are present. The former
are borne on distinctly four-winged peduncles, up to 7o mm. in length. They are
bright blue; the lower and two lateral petals are marked with dark violet veins ;
the spur is blunt, grooved at the tip, greenish-yellow, quite or nearly as long as the
appendages of the sepals; the
stigma projects 1 mm. beyond
the cone of anthers. The cleis-
togamous flowers—which are
situated in the leaf-axils of the
upper branches—have very
short peduncies, are destitute
of corolla, and the anthers of
their broad leaf-like stamens
are extremely small. The spurs
of the chasmogamous flowers
are almost always broken.

349. V. calcarata L.—
This species bears Lepidopterid
flowers (Herm. Miiller, ‘ Alpen-
blumen,’ pp. 154-6), with a spur
13-25 mm. long (Fig.37). The
nectar is concealed in the end

G a Fic. 38. Véola btfora, L. (after Herm. Miller). 4. Front view
of the Spur, and is only readily of flower (x 34). B. Section of flower, omitting the nectar-guides
accessible to Lepidoptera. (Xx 7). C. One of the stamens with a nectary (), seen from the

: oo inner side. JD. Outer side of the same. E. Pistil seen from below,
Autogamy isexcluded. Kerner F. Lateral view of the same. G. Stamens and pistil, seen from

says that the flowers are blue 2>ove: % filament: a, anther; ¢, appendage of connective: 7,
: nectary.

in the West-Central Alps and

yellowish in Carniola.

VisiTors.—These are partly moths, partly butterflies, according to the length of
the spur. The varieties with the longest spurs can only be properly sucked by
a diurnal moth—Macroglossa stellatarum Z., with a proboscis 25-8 mm. long.
This insect appears to be the most successful pollinator. Hermann Miiller, for
instance, saw it visit and pollinate in 63 minutes, no less than 194 flowers belonging
to different plants.

350. V. biflora L.—This species bears fly-flowers (Herm. Miiller, ‘Alpenblumen,.
pp. 152-4). The spur is so short that a proboscis not more than 2 or 3 mm. long

140 ANGIOSPERMAE—DICOTYLEDONES

can reach the nectar (Fig. 38). The flowers are, in fact, chiefly visited by flies. It
has yet to be determined whether automatic self-pollination can take place should
insect-visits fail.

Lindman observed on the Dovrefjeld flowers exhibiting a transition to cleistogamy,
in addition to the chasmogamous ones, which resemble those of the Alps. In these
the lateral petals, and sometimes even the anterior ones, are greatly reduced. In
individual flowers the style may be very short, so that the pollen-surrounded stigma
occupies the position which is usual in cases of cleistogamy.

Visitors.—Herm. Miiller observed the following in the Alps.—(a) Flies, par-
ticularly Syrphidae—yz species—which sucked from above, and regularly effected
cross-pollination ; the larger J/uscédae behaved similarly. (6) Short-tongued bees—
Halictus cylindricus #.—which first attempted to suck from below, but quickly
learned the proper way of securing nectar from above. (c) A few Lepidoptera, skg.

MacLeod noticed 2 Muscids in the Pyrenees (‘ Pyreneenbl.,’ pp. 398-9).

351. V. lutea Sm.—

VisiTors.— Willis and Burkill (‘Fls. and Insects in Gt. Britain,’ Part 1) observed
the following Muscids in Central Wales:—1. Anthomyia sp.; 2. Hylemyia lasciva
Zett., skg.; 3. Siphona geniculata Deg., skg.

Wittrock noticed near Stockholm—on the variety grandiflora V777,—humble-bees
and Lepidoptera.

352. V. sepincola Kern.—Kerner states that the flowers of this species are
chasmogamous in sunny places, but cleistogamous in shady woods. Both hemi-
cleistogamous and eucleistogamous flowers occur, according to Calloni (Bul. Soc. bot.,
Geneve, v, 1889).

353. V. sciaphila Koch.—Calloni states that in this species there are hemi-
cleistogamous or eucleistogamous flowers, in addition to the chasmogamous ones
which appear in spring.

354. V. stagnina Kit—Corry says that this species possesses cleistogamous
flowers.

355- V. montana L.—Linnaeus long ago observed cleistogamous flowers in
this species.

356. V. elatior L., and 357. V. lancifolia L.—Both these species bear
cleistogamous flowers, according to Daniel Miiller (Bot. Ztg., Leipzig, xxv, 1867).

358. V. bicolor L.—Hermann Miller (‘ Fertilisation,’ p. 121) states that there
are cleistogamous flowers in this species.

359. V. mirabilis L—There are cleistogamous flowers in this species, according
to Dillenius,

360. V. palustris L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ p. 33.) —The
small flowers are of a bright lilac colour, and the lower petal is streaked with dark
violet.

Visitors.—In Dumfriesshire a Muscid has been observed (Scott-Elliot, ‘ Flora of
Dumfriesshire,’ p. 20).

361. V. cornuta L.—The flower exhales a strong odour at night, and, according
to Hart (‘Fertlsn. of Viola tricolor and V. cornuta’), is adapted for pollination by
moths. The spur is longer than the corolla.

VIOLARIEAE 141

Visttors.—Hart observed an owlet-moth—Cucullia umbratica Z.—a butterfly—
Hipparchia janira Z.—and humble-bees.
Wittrock noticed Bombus subterraneus Z. and several butterflies near Stockholm.

362. V. pinnata L. (Herm. Miiller, ‘Alpenblumen,’ p. 151.)—The lower lip
possesses no hairs for the reception of the pollen which falls from the cone of anthers.
The proboscis of an insect when thrust into the spur is, therefore, not covered with
pollen from below, but from above, so that the lower margin of the stigma is pollinated
and crossing is effected. Self-pollination is prevented by a pronounced broadening
of the rim of the stigma (see Fig. 39). Linnaeus long ago observed cleistogamous
flowers.

Visitors.—These are probably bees.

Fic. 39. Viola pinnata, L. (after Herm. Miiller). 4. Front view of flower. B. Section through
flower. C. Reproductive organs. J. Pistil seen from the side. 2. Front view of style and stigma.
(A and B x3}; C—£ x7.)

363. V. arenaria DC. (Herm. Miiller, ‘Alpenblumen,’ p. 152.)—The flower
mechanism agrees with that of the last species. The proboscis of an insect is dusted
with pollen from above, and self-pollination is prevented. The margin of the stigma
is not expanded, but beset with stiff, projecting hairs (see Fig. 40).

At Kongsvold on the Dovrefjeld Lindman found that the flowers were cleisto-
gamous during the first three weeks of July, and set fruits. Kerner observed in the
Tyrol cleistogamous flowers on the prostrate runners, besides the chasmogamous ones.

Visitors.—Herm. Miiller observed butterflies (Vanessa).

364. V. tricolor L. (Sprengel, ‘Entd. Geh.,’ pp. 386-400; Hildebrand, ‘ Die
Geschlechtsvert. b. d. Pfl.’; Herm. Miiller, ‘ Fertilisation,’ pp. 117-18, ‘ Weit. Beob.,’
II, pp. 206-9; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 215-20;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 33, 150; Kirchner, ‘Flora v. Stuttgart,’
p. 320.)—Hermann Miiller says that to get at the nectar insects must introduce their
proboscis immediately below the spheroidal stigmatic head, and so on into the spur.

142 ANGIOSPERMAE—DICOTYLEDONES

This head, however, rests on a hair-fringed groove in the lower petal, into which the
pollen falls either of itself or owing to the pressure of the insect. The proboscis—as
it passes along the groove—gets dusted with pollen on its under-side. The three
following varieties may be distinguished.—

(a) V. vulgaris Koch. This variety possesses large flowers, 20-30 mm. long and
34-16 mm. broad. The petals are longer than the calyx, and are either all violet,
or the four upper ones are violet, while the lower one is yellow with violet veins; or
else the lateral petals may also be yellowish. This variety can only be fertilized by
cross-pollination. The opening of the spheroidal stigmatic head is turned outwards,
so that the pollen, when it escapes from the anther-cone, cannot of itself fall into the
stigmatic cavity. On the lower edge of the stigma there is a flexible lip-like valve,
which prevents the proboscis of an insect as it is being withdrawn from effecting
self-pollination. When—on the other hand—the proboscis is introduced, it may
dust the upper-side of the valve with pollen from another flower, thus bringing about

Fic. 40. V7ola arenarta, DC (after Herm. Miller). 4 and B, as in Fig. 39. C. Reproductive
organs, seen from below. D and E. Pistil seen from below and from the side. (A and B x 24;
C—E x 7.)

cross-pollination. After the flower has been open a few days the pollen falls of itself
out of the anther-cone into the hairy groove of the lower petal. If insect visitors are
kept away, the flowers do not wither for two or three weeks, and either no fruits are
set, or only a few with seeds incapable of germinating.

Vistrors.—I have seen Anthophora pilipes #., and Bombus hortorum Z., both
skg., in this large-flowered variety. Hermann Miiller observed only long-tongued
bees :—Apis, Bombus lapidarius, B. terrester, B. hortorum, and Anthophora pilipes F’.
9; all skg. The last-named bee was also noticed by Delpino.

As further visitors Herm. Miiller observed a small bee—Andrena albicans JZ#l/.
é—with a proboscis 2-24 mm. long. Self-pollination was effected during its fruitless
attempts to reach the nectar, which was concealed at a depth of 3mm. He also
observed a hover-fly—Syritta pipiens Z.—po-dvg., which frequently touched the
stigma, and must therefore have effected self-pollination.

On one occasion Alfken noticed a humble-bee—Bombus hortorum £.—skg.,
in Juist.

VIOLARIEAE 143

Verhoeff observed the following in Norderney and Juist (J).—

A. Hymenoptera. (a) Apidae: 1. Bombus cognatus S%eph. (=B. muscorum
F,), one 9, skg. (J.);_ 2. B. lapidarius Z. 9, skg.; 3. B. latreillelus A. (=B. subterraneus
L.), 29; 4. B. terrester Z. 2 9, skg., ¥§ unusual; 5. Psithyrus vestalis Mourcr. one 9,
skg. (5) Vespidae: 6. Odynerus parietum Z. one 9, skg. perforated spurs. B. Lepi-
doptera. Pureridae: 7. Pieris brassicae L. 4 9, one 4, skg.

Friese records Podalirius acervorum Z. for Central Europe.

Dalla Torre and Schletterer noticed Bombus hortorum Z. é in the Tyrol.

The second morphological and oecological variety is:

(8) V. arvensis Murr. The flowers are 8-13 mm. long, and 6-8 mm. broad
The petals are small, hardly as long as the calyx, and yellowish-white in colour,
though more rarely the upper petals are bluish or violet, and the lower one dark
yellow. The nectar guides are more or less reduced. This variety is autogamous
and autocarpous. The opening of the spheroidal stigmatic head is turned inwards,
enabling pollen-grains to fall into it. The lip-like valve is wanting, so that the
proboscis of an insect when being withdrawn may effect self-pollination, which is
effective.

VisiTors.—I saw the following insects, all skg., on the small-flowered variety ot
the pansy at Kiel:—the honey-bee, Anthophora pilipes /. 6, Bombus agrorum F.,
and a white cabbage-butterfly (Pieris napi Z.). Hermann Miiller observed the
following.—

A. Coleoptera. Nitidulidae: 1. Meligethes. B. Diptera. Syrphidae: 2.
Rhingia rostrata Z., skg. C. Hymenoptera. Apidae: 3. Apis mellifica LZ. ¥, skg.;
4. Bombus hortorum Z. 9, skg.; 5. B. agrorum F’. 9, skg.; 6. B. rajellus X. 9, skg. ;
4. Osmia rufa Z. 6, skg. on the wing. D. Lepidoptera. Rhopalocera: 8. Pieris
napi L., skg.; 9. P. rapae Z., skg.; 10. Polyommatus dorilis W/n., skg.

Loew noticed the following in Silesia (‘ Beitrage,’ pp. 34-5).—

A. Hymenoptera. Afzdae: 1. Diphysis serratulae Pz. 4, skg. B. Lepi-
doptera. hopalocera: 2. Pieris brassicae L., skg.

MacLeod observed a wasp, and a Lepidopterid, in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1896, p. 220), and—in the Pyrenees—a hover-fly vainly
endeavouring to penetrate into the flower (op. cit., iii 1893, p. 398).

Schneider (Mus. Aarsh., Troms¢g, 1894) saw B. pratorum Z. ¥, and B. terrester
Z.%, on garden plants in arctic Norway. Wittrock noticed only Apis mellifica Z.
as a cross-pollinator near Stockholm.

The third variety is—

(y) V. alpestris, (Herm. Miiller, ‘Alpenblumen,’ p. 156.) The variety is inter-
mediate between V. tricolor var. vulgaris and V. calcarata (see p. 139). The
flowers are 25-30 mm. long, and 18-22 mm. broad when full grown. The spur
is intermediate in length between that of V. tricolor (3-4 mm.) and V. calcarata
(13-25 mm.). Should insects visit the flowers, crossing is certain, while automatic
self-pollination is usually excluded.

Visirors.—These are intermediate in character between those of the two other
varieties, and include Syrphidae, Apidae, and Lepidoptera.

Konig (SitzBer. Isis, Dresden, (1891) 1892) calls attention to the fact that the
large-flowered variety of the pansy is much more variegated than the small-flowered

144 ANGIOSPERMAE—DICOTYLEDONES

one, which is usually yellowish-white with ill-developed nectar-guides. The upper
petals of the latter form are rarely bluish or violet (cf. p. 143).

Miiller says that the blossoms of the small-flowered variety—which are protected
against the ravages of insects—wither after two or three days, by which time the
fruits have been set. The flowers of the large-flowered variety, on the other hand,
do not wither for two or three weeks, and then usually without having set fruits
(cf. p. 142).

MacLeod found on sand-dunes in Belgium—at Blankenberge —a variety
possessing remarkably large flowers, with black spots on the under-side of the
stigma. The stigmatic papillae of this variety are more numerous than in flowers
from other habitats.

Wittrock (‘ Viola-Studier’) says that the two constituent parts of the nectar-
guide on the odd petal, i.e. the ‘nectar-spots’ at its base, and the ‘nectar-streaks’
which radiate forwards from it, are constant in colour in the same variety of
the pansy, even though the colour of the petals varies. The nectar-spots are
yellow or orange, and the honey-streaks dark violet. The spur is always violet.

The hair-fringed groove on the lower petal serves—according to Wittrock—
as a ‘pollen-magazine,’ in which the falling pollen is collected and stored till an
insect visits the flower. The hairs possess knot-like thickenings, which appears
to be an adaptation for holding the pollen-grains fast. The front part of the groove
forms a ‘pollen cavity’ open above, while its posterior narrower part constitutes
a ‘pollen-canal.’ The pollen falls into the pollen cavity through an opening between
the membranous appendages of the two lowest stamens, which is situated exactly
above it.

Wittrock has also demonstrated by direct experiment that the hairs on the
bases of the limbs of the lateral petals serve to protect the reproductive organs
and the pollen magazine against rain. They are also—as earlier investigators sup-
posed—of use as footholds for pollinating insects.

The petals are much smaller and relatively much broader in the younger flowers
than in the older ones.

The flowers of the plant differ according to the season: in spring and early summer
the petals are much larger, broader, and more brightly coloured than in late summer.
The nectar-streaks are often absent in late summer and autumn, but the nectar-spots
and nectaries are constant.

It is exceptional to find flowers of very different colour on the same plant.
In one such case that was accurately investigated this peculiarity was limited to
the hottest part of the summer, when the colour varied from violet to white. In
spring and autumn, on the contrary, the plant produced only violet flowers. Reasoning
from this and other cases, Wittrock thinks it probable that a high temperature has
an injurious effect upon flower production in the varieties of the pansy, causing
smaller and paler blossoms to be developed.

For the first two or three days of anthesis—which lasts about a week—the
petals of V. tricolor are nyctitropous. The two upper petals bend forwards in
the evening, assuming an almost horizontal position, the lateral petals curve a little
inwards, and the lowest petal becomes slightly concave by the upward bending
of its edges. Besides these movements—which were overlooked by earlier investi-

VIOLARIEAE 145

gators—mention must be made of the nyctitropous curvature of the peduncle
described by Kerner. During the last three or four days of anthesis the nutations
of the petals and peduncle are scarcely perceptible.

The anthers do not all dehisce simultaneously. When the flower opens, or
even a day earlier, the upper anther opens, the two lateral ones do so a few days
later, and the two lower stamens are the last to dehisce.

The pollen-grains are dimorphous or trimorphous; as seen from the front
they are quadrangular or triangular, more rarely pentagonal; in side view they
are ellipsoidal.

With regard to the functions of the various parts of the female apparatus during
pollination, Wittrock arrives at the following conclusions, which differ considerably
from those hitherto accepted.—

The form and structure of the stigmatic valve appear not to have been
correctly known before, its function being therefore misunderstood. Wittrock
describes it as an epidermal outgrowth, short, fan-shaped, and very small, and
composed of hyaline club-shaped cells, moderately stiff, and beset with papillae.
In the middle part of the valve these cells are arranged in four or five layers, those
of the middle layer being the longest, while those of the other layers get successively
shorter as the surface is approached. The sides of the lobe are made up of three
layers, the cells of the middle one being the longest. Wittrock finds that the valve
is only slightly flexible, so that when the proboscis of an insect is withdrawn from
the flower it neither closes the stigmatic cavity nor presses pollen into it. The
valve therefore plays but a subordinate part in preventing self-pollination during
the visits of insects. The knee-like joint in the lower part of the style is much
more effective for this purpose. Its elasticity enables the well-known upward
curvature of the stigmatic head from the pressure of an insect’s proboscis.

Wittrock made observations at Stockholm on the visits of insects to wild plants
of V. tricolor Z., var. versicolor Wittr., and from these he draws the following main
conclusions :—The blossoms of V. tricolor are both Lepidopterid and Hymenopterid
flowers in Central Scandinavia. Most insect visitors effect cross-pollination. Among
nectar-thieves are a few of the smaller Hymenoptera, e.g. Odynerus oviventris Z.,
and also the fly Ocyptera brassicaria Fadr. The beetle Cetonia aurata Z. devours
stamens and petals. The po-dvg. Thysanoptera may in certain cases effect self-
pollination. Legitimate insect visitors are few during the height of summer.

In V. arvensis AZurr. the pollen-magazine is quite open in front, so that there
is no sharply-defined pollen-cavity, and the pollen-grains can fall unhindered into
the stigmatic cavity. Self-pollination, therefore, regularly takes place. The August
flowers are usually entirely devoid of nectar-streaks. In autumn flowers the upper
petals are obviously smaller, in proportion to the sepals, than in spring and summer
ones. In V. patens, blossoms in which the petals are sometimes greatly reduced
may also occur in the earlier parts of the year. These are usually borne on axes
of a high order, and possess normally developed and fully functional reproductive
organs, even when the petals are reduced to small scales; in this respect, and also
in having an open corolla, they differ from cleistogamous flowers.

The busiest visitor observed by Wittrock, in the neighbourhood of Stockholm,
was Bombus subterraneus Z., which, on account of its activity, may be compared

DAVIS. II L

146 ANGIOSPERMAE—DICOTYLEDONES

with the hawk-moth, Macroglossa stellatarum Z., rendered famous by Hermann
Miiller (cf. Vol. 1, p. 172; Vol. II, p. 139). (Cf. the review by Grevillius in Bot.
Centralbl., Cassel, xxi, 1897.)

113. Ionidium Vent.

Some species of this genus are cleistogamous, according to Bernoulli (Bot.
Zig., Leipzig, xxvii, 1869).

XIV. ORDER POLYGALEAE JUSS.

The genus here to be considered is—

14. Polygala L.

LiteraTuRE.—Hildebrand, Bot. Ztg., Leipzig, xxv, 1867, p. 281; Knuth, ‘Bl.
u. Insekt. a.d. nordfr. Ins.,’ p. 35.

Homogamous bee flowers. The chief means of attracting insects are two large
lateral petaloid sepals. The petals are usually of minor importance for this purpose,
mainly serving to protect the stamens and pistil. Cleistogamy has also been
observed (Kuhn).

Fic. 41. Polygala comosa, Schk. (after Hermann Miller). (1) Flower seen from the side (the flower
is represented erect instead of in the natural horizontal position): s, sepal; s!, one of the two lateral sepals
that serve to attract insects; ~, petal; #', lower petal provided with digitate appendages, to which the

insect clings when it alights. (2) Flower seen from below. (3) Lower petal enclosing the essential
organs, seen from above; a, anthers; s, stigma (with adhesive matter); /, spoon-shaped end of the style,
which receives the pollen from the adjacent anthers. (4) Pistil, seen from above; st, stigma; /, spoon-

shaped process; ov, ovary. (5) The same, seen obliquely from above. (6) The same, seen from the side. (7)

The lower petal of a flower just about to expand, split to show the anthers enclosed by it. (8) The coherent

anthers. (9) A dehisced anther ; fo, pollen-grains.

Chodat (‘Révision et critique des Polygala suisses,’ Bul. Soc. Bot., Genéve,

v, (1888) 1889, pp. 123-85) is of opinion that all the Swiss species of Polygala
are capable of self-pollination, i.e. P. vulgaris Z., P. comosa Sch&., P. amara Jacq.,
P. calcarea Schk., P. nicaeénsis Resso, P. depressa Wender., P. alpina Long. et Perr.,
and P. Chamaebuxus Z.

365. P. comosa Schk. (Hildebrand, Bot. Ztg., Leipzig, xxv, 1867; Herm.
Miiller, ‘Fertilisation,’ pp. 122-3, ‘ Weit, Beob.,’ II, p. 213, ‘Alpenblumen,’ p. 169;

POLYGALEAE 147

Kirchner, ‘Flora v. Stuttgart, pp. 353-4; Schulz, ‘Beitrage,’ II, pp. 18-19.)—
Serrated processes of the lower petal are made use of by insects as an alighting-
place. There are two folds on the upper side of the same petal which enclose
the anthers and the spoon-shaped end of the style. Behind this ‘spoon’ is the
stigma, a sticky hook-shaped eminence. The anthers are so placed above the end
of the style that when they dehisce the pollen necessarily falls into the spoon,
where it is stored while the stamens wither. The proboscis of an insect probing
for the nectar secreted in the base of the flower must therefore first encounter the
pollen in the spoon, and then the stigma. This does not, however, result in self-
pollination, for it is not till the proboscis of the insect has been smeared with
glutinous matter from the stigma that pollen adheres to it, and this may be carried
to the stigma of the next flower visited. Hildebrand states that when insect-visits
fail the stigma bends down to the pollen in the spoon far enough to bring about
self-pollination. Schulz considers this curvature of the stigma to be an exceptional
occurrence, though there is no doubt that the pollen may come into contact with
the adhesive stigmatic surfaces. He asserts that self-pollination frequently takes
place at the beginning of anthesis, so large a quantity of pollen being discharged
from the anthers into the spoon that this is filled up to the level of the stigma,
against which some is necessarily pushed by the proboscis of an insect visitor, seeing
that the stigmatic surface immediately adjoins the posterior end of the spoon.

Visitors.—Herm. Miiller saw 3 Lepidoptera in the Alps. Buddeberg observed
the following in Nassau; all skg.—

A. Hymenoptera. Afzdac: 1. Andrena albicans A/#/l. 9; 2. A. fulvago Chr.
9; 3. Eucera longicornis Z. 8. B. Lepidoptera. Rhopalocera: 4. Lycaena sp.

366. P. vulgaris L. (Hildebrand, op. cit.; Herm. Miiller, ‘Fertilisation,’
p. 123, ‘Weit. Beob.,’ II, p. 213; Kirchner, ‘Flora v. Stuttgart,’ p. 354; Schulz,
op. cit.; MacLeod, Bot. Jaarb, Dodonaea, Ghent, vi, 1894, pp. 241-6; Knuth, ‘BI.
u. Insekt. a. d. nordfr. Ins.,’ p. 35.)—The flower mechanism agrees with that of the
last species.

Visttors.—The following have been observed by Herm. Miiller (H.M.) and
myself (Kn.); all skg.—

A. Hymenoptera. Apidae: 1. Apis mellifica Z. (Kn., H.M.); 2. Bombus
lapidarius Z. (Kn., H.M.); 3. B. terrester Z. (Kn., H.M.). B. Lepidoptera. (a)
Geometridae: 4. Odezia chaerophyllata Z. (H.M.). (4) Rhopalocera: 5. Polyom-
matus hippothoé Z.(H.M.). C. Diptera. £mpzdae: 6. Empis livida Z. (H.M.).

In the Pyrenees MacLeod observed a humble-bee skg. nectar, and a wasp
trying to suck it.

367. P. amara L.—The flower mechanism of this species-agrees with that of
P. comosa. Kirchner states that the flowers of the variety austriaca Koch are in
all respects smaller.

Schulz (‘ Beitrage,’ I, p. 19), in a very large number of cases, was unable to
perceive the outward curving of the stigmatic process towards the spoon, which
Hildebrand described as taking place towards the end of anthesis.

368. P. calcarea Schulz.—

Visitors.— MacLeod observed a Lepidopterid skg. in the Pyrenees.
2

148 ANGIOSPERMAE—DICOTYLEDONES

369. P. Chamaebuxus L. (Hildebrand, op. cit.; Herm. Miiller, ‘ Alpen-
blumen,’ pp. 165-8.)—Kerner says that the flowers possess a plum-like odour.
Their structure—which was first described by Hildebrand—bears a certain resem-
blance to that of some Papilionaceae (Lotus). In both cases the stamens and style
lie in the lower part of the horizontal flower, and bend upwards at the tip. Anthers
and stigma occupy a laterally compressed chamber (carina), which is only open
above, and this is pressed down by an insect visitor in such a way that not only
are the anthers themselves pressed against its body, but also a part of the pollen
discharged into the surrounding chamber before anthesis. At the same time the
stigma is pressed against the under-surface of the visitor—which is probably a bee—

Fic. 42. Polygala Chamaebuxus, L. (after Herm. Miller). 4. Flower seen from the side (natural
size). B. Flower seen from the front (x 2}). C. Flower after removal of the two petaloid sepals,
seen from the side. D. The same, seen from above. £. The same, seen from below (x 3). F. The
two petals of the left side (x 24). G. Bud, after removal of the sepals (= 54). A. Mature flower in longi-
tudinal section (x 54). /- Base of flower after removal of the upper sepal, seen from above (x7). s', upper
sepal; s°, lateral sepal; s*, lower sepal ; #! lower petal; #° lateral petal; 4" upper petal; 4, spoon; e, en-
trance to the flower ; 4/, adhesive matter on the stigma; fo, pollen; ov, ovary; gr, style ; sch, carina with
hinge (x).

and perhaps first becomes receptive as a result of the friction. It would seem that
either the pollen of the same flower is ineffective or the foreign pollen prepotent,

cross-pollination being thus ensured.

Visitors.—Herm. Miiller—in the Alps—observed 5 humble-bees, in part skg.
nectar, in part getting at it by perforating the flower; also 3 Lepidoptera, skg. nectar,
but useless as visitors. Dalla Torre, and also Schletterer, saw Bombus sylvarum

LZ. 9 and ¥, in the Tyrol. Hoffer (Kosmos, Stuttgart, xvi, 1885) in the Lower Alps
(Steiermark) noticed the following Apzdae.—

CARYOPHYLLEAE 149

1. Anthophora pilipes #.; 2. Andrena fulva Schr.; 3. Apis mellifica Z.;
4. Bombus agrorum /.; 5. B. hortorum Z.; 6. B. lapidarius Z.; 7. B. mastrucatus
Gers/.; 8. B, pomorum Pz.; g. B. pratorum Z.; 10. B. rajellus A.; 11. B. sylvarum
L.; 12. B. soroénsis F.; 13. B. terrester £.; 14. Osmia bicolor Schr.; 15. O.
cornuta Lair.

Of these, Bombus mastrucatus and B. terrester steal the nectar by perforation.
Apis, Bombus pratorum, and B. soroensis steal nectar through the holes made by
these humble-bees.

Ricca found that 95 % of the flowers were broken, and Schulz noticed the same
thing in many instances at Bozen.

370. P. serpyllacea Weihe (=P. depressa Wender.)—

Visirors.—MacLeod observed Bombus agrorum and B. hortorum in Flanders
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 246).

371. P. alpestris Reichb.
(Herm. Miller, ‘Alpenblumen,’
p. 168-9.)—The flower me-
chanism of this species agrees
pretty well with that of P.
comosa, but owing to the fusion
of the three lower petals, there
is also a certain resemblance
to that of P. Chamaebuxus.

Visirors.— Herm. Miller
saw only 4 sp. of butterflies in

the Alps. Fic. 43. Folygala alpestrts, Reichb. (after Herm. Miller). 4.
" , Flower seen from below (x 7). &. The same, seen from the front,
372. P. myrtifolia L.— more highly magnified. e, entrance to the flower.

Delpino (‘ Ult. oss.,’ pp. 185-7)
states that the flower mechanism agrees with those of certain Papilionaceae (Lathyrus,
Phaseolus).

VisiTors.—The same bee—Xylocopa violacea £.—which pollinates Lathyrus
and Phaseolus does the same for this species, and in the same manner.

XV. ORDER CARFOPAYVLLEAE.

1. Tripe SirengaE DC.

LitrraturE.—Knuth, Grundriss d. Bliitenbiologie, pp. 29-30.

The flowers are frequently large and brightly coloured, and are rendered more
conspicuous by the fact that the inflorescence is often much branched. In the
individual flower, the petals—which are usually long-clawed—are so held together
by the gamosepalous calyx as to form a more or less elongated tube, the length
of which is often increased by the presence of a corona. The nectar secreted in
the base of the flower, or the juices there obtainable by boring, are therefore usually
only accessible to long-tongued insects. It follows that many Sileneae are Lepi-
dopterid flowers, the red-flowered species being butterfly flowers, and the white-
flowered ones moth flowers or hawk-moth flowers.

150 ANGIOSPERMAE—DICOTYLEDONES

Some species, however—Tunica prolifera, Gypsophila—possess so short and
wide a tube that the nectar is accessible even to beetles and short-tongued flies.
One species (Silene Otites Sm.) is mainly anemophilous. Of the ten stamens usually
present the five outer ones almost always mature first, and the five inner ones
last. Many species are conspicuously protandrous, the stigmatic papillae only
developing after the stamens have withered. Homogamy is rare (Tunica prolifera).
A few vary from protandry to homogamy. Certain species show different relations
in different regions. In addition to the hermaphrodite flowers, many species possess
some that are purely female, or may exhibit dioecism—the female flowers being
usually a little smaller than the male ones, and these again inferior in size to the
hermaphrodite flowers. In hermaphrodite flowers individual stamens are occasionally
absent. Many species are gynodioecious (gd.), gynomonoecious (gm.), androdioe-
cious (ad.), or andromonoecious (am.), e. g.:—

Gypsophila repens Z.: gd., more rarely gm. (Ludwig) ;

G. fastigiata Z.: gd. and gm. (Schulz);

Tunica saxifraga Scop.: gd. (Breitenbach), rarely gm. ;

T. prolifera Scop.: gd. and gm. (Schulz) ;

Dianthus Seguierii V7//.: gd., rarely gm. (Schulz) ;

D. caesius Sm.: gd. (Kirchner) ;

D. deltoides Z.: gd. (Schulz) ;

D. Armeria Z.: gd. and gm. (Kirchner) ;

D. Carthusianorum Z.: gd., rarely gm. (Schulz);

D. atrorubens 4//., D. superbus Z., D. monspessulanus Z., D. sylvestris Wulf:
ditto ;

Saponaria officinalis Z.: gd., rarely gm. (Schulz) ;

S. ocymoides Z.: gd. and gm.,rarely ad. and am., also trimonoecious (Hildebrand);

Vaccaria parviflora Moench: gd., rarely gm. (Schulz) ;

Cucubalus baccifer Z.: gd. and gm. (Schulz);

Silene Armeria Z.: gd. (Breitenbach) ;

S. nutans Z.: gm., gd., am., ad, (Schulz) ;

. Otites Sm.: dioecious, rarely ad. (Knuth) ;

. inflata Sm.: gm., gd., am., ad. (Schulz, Magnus, Knuth, and others) :
. Saxifraga Z.: am, and gm. (Lalanne) ;

- noctiflora Z.: gm. (MacLeod), gd. and am. (Schulz) ;

. dichotoma Ffrh.: gd. (Warming, Kirchner);

Viscaria vulgaris Roeh/.: gm., gd., rarelygad. and am. (Schulz) ;
Coronaria Flos-cuculi 4. Br.: gd., gm., rarely ad. and am. (Schulz) ;

C. tomentosa A. Br.: gd., rarely gm. (Schulz);

Melandrium rubrum Garcée : trioecious, rarely gm. or am. (Schulz) ;
M. album Garcke: dioecious.

NNnNNNNM

115. Gypsophila L.

Flowers protandrous, usually with concealed nectar secreted by a fleshy ring
formed by the thickened bases of the filaments.

373. G. paniculataL. (Herm. Miiller, ‘ Fertilisation,’ pp. 127-8, ‘Weit. Beob.,’
I, p. 230.)—Numerous blossoms are borne on the same stock, so that, in spite

CARYOPHYLLEAE 151

of their diminutive size, the plant is very conspicuous. The cup-shaped flower is
only 4-5 mm. in diameter, and 24 mm. deep. Nectar is secreted at the bottom
of the cup by a green fleshy ring, and is readily accessible even to very short-tongued
insects. The anthers of the five stamens which alternate with the petals mature
before those of the stamens opposite the petals. They project from the flower—
at the same time inclining towards the petals—and dust insect visitors with pollen.
When they wither the stamens
curve outwards and down-
wards, while the styles—at
first bent inwards—now elon-
gate and diverge from one
another, projecting from the
flower, so as to be pollinated
by the numerous visitors that
hasten from blossom to
blossom.

Visitors.— Hermann Miil-

ler observed the following.—

: Fic. 44. Gypsophila paniculata, L. (after Herm. Miller). (1)
A. Diptera. (2) Mus- Flower in the first (male) condition. (2) Flower in the second (female)

ctdae: 1. Anthomyia Sp.; 2. condition. a, anthers; s¢, stigma.

Lucilia sylvarum Jfe., skg.;

3. Miltogramma sp., skg.; 4. Mosillus arcuatus Za/r., skg.; 5. Onesia floralis #.-D.,
skg.; 6. Pyrellia cadaverina Z., skg.; 7. Sarcophaga carnaria Z., skg. (4) Syrphidae:
8. Ascia podagrica F., skg. and po-dvg.; 9. Eristalis aeneus Scop., do.; 10. E.
arbustorum Z., do.; 11. E. nemorum Z., do.; 12. Melithreptus pictus JZg., do. ;
13. M. taeniatus A/g., do.; 14. Syritta pipiens Z., freq., do.; 15. Syrphus balteatus
Deg., do. (¢) Tabanidae: 16. Chrysops caecutiens Z., skg. B. Hymenoptera.
(a) Aprdae: 17. Prosopis armillata Vy/. 9 and 4, skg.; 18. P. brevicornis /Vy/. 4, skg.;
19. P. communis /Vy/. 9, skg.; 20. Sphecodes ephippium Z. 6. (6) Evanitdae: 21.
Gasteruption jaculator #., skg. (c) Formicidae: several sp. (d) Sphegidae: 22. Oxy-
belus quattuordecimnotatus Jur. 9 andé, skg.; 23. O. uniglumis Z., skg. (e) Vespidac:
24. Odynerus parietum Z., skg.; 25. O. quadrifasciatus /., skg.

374. G. muralis L.—The flowers are flesh-coloured, and traversed by red
veins; their diameter is about 5 mm. Kirchner (‘Flora v. Stuttgart,’ p. 242) states
that the anthers of the five outer stamens mature before those of the five inner
ones; the filaments elongating so that the anthers project beyond the opening of
the flower.

After the pollen is shed, the filaments curve so far outwards that the styles—-
which now diverge and project out of the flower—cannot come into contact with
the anthers, automatic self-pollination being thus prevented. In this second (female)
condition the petals—which so far have been flat—roll themselves up longitudinally
to some extent, so that the flowers are less conspicuous than in the first (male)
condition, when they are consequently more noticed and visited by insects.

375. G. fastigiata L.—Schulz (‘Beitrage, I], p. 180) observed at Kyffhauser
gynomonoecious and gynodioecious plants, in addition to those bearing protandrous
flowers. Warnstorf describes the pollen-grains as white, roundish-polyhedral,
delicately papillated, 30-37 » in diameter (Verh. bot. Ver., Berlin, xxxvii, 1895).

152 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Loew observed a bee, Prosopis communis JVy/., skg., in the Berlin
Botanic Garden.

376. G. repens L. (Herm. Miiller, ‘Fertilisation, p. 128, ‘Alpenblumen,’
pp. 191-2; Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, I, p. 511; Ludwig, Bot. Centralbl.,
Cassel, xxxvi, 1888; Schulz, ‘ Beitrage,’ II, pp. 19-20.)—The rose-red flowers vary
from slight’ to marked protandry. Their diameter is scarcely 10 mm., but as the
plant forms great plots on the stony declivities of the Alps it is very conspicuous.
Nectar is secreted in great abundance. In good weather the visits of insects are,
therefore, numerous, so that crossing takes place. Should insect-visits fail, automatic
self-pollination obtains in the more unfavourable places. Ludwig observed cases of
gynodioecism, and more rarely of gynomonoecism.

Visrrors——Herm. Miiller—in the Alps—observed chiefly flies (14 species),

humble-bees (2), and Lepidoptera (5). A. Schulz noticed a similar set of visitors
in the Tyrol, i.e. flies, bees, and Lepidoptera, as well as a few beetles.

Fic. 45. Gypsophtla repens, L. (after Herm. Miller). 4. Flower at the beginning of the first
(male) condition. 2. Flower at the end of this condition, C. Flower in the second (female) condition.
a, anthers ; s¢, stigma; , nectary.

MacLeod saw a bee and 13 flies in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent,
iti, 1891, pp. 375-6).

377. G. perfoliata L—

Vistrors.— Loew observed the following Syrphids in the Berlin Botanic
Garden :—1. Eristalis nemorum Z.; 2. Syritta pipiens Z.

378. G. elegans Bilb.—This species is protandrous and self-fertile (Comes,
‘Stud. s. impoll. in alc. piante’).

116. Tunica Scop.

Flowers protandrous or homogamous, with concealed nectar: sometimes gyno-
dioecious, rarely gynomonoecious.

379. T. Saxifraga Scop.—Schulz (‘Beitrage, II, pp. 20-1) states that the
hermaphrodite flowers vary greatly in size: their diameter from 6-ro mm., and
their depth from 4-53 mm. The five outer stamens first mature, then the five
inner ones, and lastly the stigma, at so late a stage that self-pollination is almost
completely prevented. As already mentioned by Breitenbach (Kosmos, Stuttgart, xiv,

CARYOPHYLLEAE 153

1884), there are purely female flowers in addition to the hermaphrodite ones. In
the Botanic Gardens of both Marburg and Géttingen the size of these is very
variable. Nectar is very abundantly secreted.

VisiTors.—Schulz—at Bozen—observed numerous flies (30 species), some of
the smaller bees (about the same number of species) and Lepidoptera, and also
some beetles.

Loew noticed the bee Halictus minutissimus 4’. 4, skg., in the Berlin Botanic
Garden, and Schletterer saw the following bees at Pola:—1. Andrena nana K.; 2. A.
parvula K.; 3. Halictus morio /%

380. T. prolifera (=Dianthus prolifer Z.). (Schulz, ‘Beitrage, I, p. 21;
Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II, pp. 213, 298.)—This plant also—according
to Schulz—is gynodioecious and gynomonoecious. The flowers are homogamous—
at Halle and Bozen—so that automatic self-pollination regularly takes place. This
only is of importance, for the small inconspicuous flowers produce but little nectar
and have few insect visitors—some nectar-sucking butterflies and pollen-devouring
flies. Further, according to Kerner, the anthesis lasts but two days, and the
flowers are only open from 8 a.m. till 1 p.m.

Visirors.— Vide supra.

117. Dianthus L.

The protandrous flowers are usually large, and often beautifully coloured.
The claws of the petals are long and winged. ‘They are held together—so as to
form a long tube—by the calyx, which is usually surrounded by tough granular

1
Ab
)

J/

Fic. 46. Déanthus deltoides, L. (after Herm. Miller). (1) Flower in the first (male) condition, seen
from above. Five anthers covered with pollen protrude from the flowers; two that have not yet dehisced
are visible in the mouth of the tube. (2) Pistil at the end of the first stage. All the anthers have dehisced,
and the two styles are still twisted together. (3) The same, with the bases of stamens and petals (more
highly magnified): a, nectary; c, petals; 4, filaments. (4) Flower in the second (female) condition after
removal of the petals, seen from the side. Most of the anthers have fallen off, and the styles have
separated. (5) Pistil of the same flower. The separated styles retain their spiral form, so that stigmatic
papillae are turned towards all sides.

bracts, serving as a protection against the bites of humble-bees trying to break
in laterally. Nectar is secreted and concealed at the bottom of the corolla-tube,
which is usually so long and narrow that it is accessible only to Lepidoptera, some-

154 ANGIOSPERMAE—DICOTYLEDONES

times even to only such of these as possess a very long proboscis (hawk-moths).

The flowers therefore belong to class L. Many species are gynomonoecious or
gynodioecious.

381. D. deltoides L. (Herm. Miller, ‘Fertilisation, pp. 125-6, ‘ Weit. Beob.,’
II, p. 230; Kirchner, ‘Flora v. Stuttgart,’ p. 244; Knuth, ‘Bloemenbiol. Bij-
dragen.’)— This species bears protandrous butterfly flowers. Hermann Miiller
(op. cit., p. 125) says that—‘ The stamens and petals spring from an annular ridge
of the receptacle encircling the stalk of the ovary. This ridge bears on its inner
border a yellow fleshy cushion which secretes honey.’ As the calyx tube is 12-
14 mm. long, the passage to the nectar is of the same length, while it is only
2mm. in diameter. At the beginning of anthesis this passage is made so narrow
by the five inner stamens it encloses as to leave only enough room for the intro-
duction of a butterfly’s proboscis. The path to the nectar is indicated by a nectar-
guide on the rose-red corolla, of which the whitish centre is surrounded by a
purple-red ring with whitish spots. Of the ten stamens, the five outer ones first
elongate so that their anthers project from the corolla-tube when they have dehisced.
After the outer stamens have withered the other five behave similarly. When all
the pollen is shed the two styles—which have so far been twisted together within
the corolla-tube—elongate, and their stigma-bearing ends project from the flower,
dominating its entrance. When the styles separate they remain spirally twisted,
so that the proboscis of a butterfly approaching the flower from any side must
necessarily touch some of the stigmatic papillae, effecting cross-pollination if the
insect has previously visited a younger flower.

Visirors.—The following were observed by Herm. Miiller (H. M.) in Westphalia,
and myself (Kn.) in Schleswig- Holstein. —

Lepidoptera. (a) Bombycidae: 1. Gnophria quadra L., skg.(?) (H. M.).
(6) Tinetdae: 2. Nemotois metallicus Poda (H. M.). (c) Rhopalocera: 3. Hesperia
lineola O., very freq., skg. (H. M.); 4. H. thaumas 47/n., skg. (H.M.); 5. Lycaena
icarus Roét., skg. (Kn., H.M.); 6. Pieris napi Z., skg. (H.M.); 7. P. rapae Z., skg.
(H.M.); 8. P. sp., skg. (H.M.); 9. Polyommatus phlaeas Z. skg. (Kn.); 10. Rho-
docera rhamni Z., skg. (Kn.); 11. Epinephele janira 2. (H. M.).

I saw in Thuringia, as a useless visitor, the small bee Halictus morio /. 9, vainly
trying to suck nectar, and afterwards po-cltg.
Loew noticed the following in Silesia (‘ Beitrage,’ p. 35).—

A. Diptera. Syrphidae: 1. Volucella bombylans Z., trying to suck. B. Lepi-
doptera. XRhopalocera: 2. Argynnis pandora S. 1’, skg.; 3. Pieris brassicae J. ;
4. Rhodocera rhamni Z., skg.

382. D. superbus L. (Sprengel, ‘Entd. Geh.,’ p. 248; Herm. Miller, ‘ Fer-
tilisation,’ p. 127, ‘Alpenblumen,’ pp. 202-4.)—The fragrant protandrous diurnal
hawk-moth flowers are red in colour, and elegant in appearance, owing to the
delicately dissected petals. Their mechanism agrees with that of the last species,
though the nectar is so deeply concealed (20-25 mm.), and the access to it so
narrow, that even butterflies cannot reach it, but only the diurnal hawk-moths
(species of Macroglossa). Self-pollination is excluded. Besides hermaphrodite
flowers, infrequent female ones have been observed, which are smaller than the

CARYOPHYLLEAE 155

others, Schulz states that the large-flowered variety grandiflora Tausch, which occurs
in the Riesengebirge, also possesses much smaller female flowers.

Warnstorf (Ver. bot. Ver., Berlin, xxxviii, 1896) was able to distinguish the
following varieties at Ruppin.—

1. Large-flowered variety. Diameter of corolla about 6 cm.; all the filaments
developed, but a larger or smaller number of the anthers brown and vestigial.

2. Medium-flowered variety. Diameter of corolla about 4 cm.; all the stamens
fully developed, so that the flowers are all hermaphrodite. A few medium-sized
hermaphrodite flowers are often found on the first variety.

Both these varieties are markedly protandrous. The stamens develop gradually,
ultimately projecting far out of the calyx-tube, which is about 23-25 mm. long.
The whitish anthers are introrse; after dehiscence they bend back through an angle
of go°, so that the anther-lobes—which spread out so as to be almost flat—lie at
right angles to the filaments. The pollen-grains are white, dodecahedral, tuber-
culated, and with an average diameter of 50 yu.

Fic. 47. Dianthus superbus, L. (after Herm. Miller). 4. Hermaphrodite flower in the first (nale)
condition, seen directly from above (natural size). B. Stamens and pistil in the same condition, seen
from the side. C. A similar flower in the second (female) condition (x 2). D. Sexual organs of a
purely female flower. Eight of the stamens possess minute vestigial anthers, and are only as long as the
ovary. The other two stamens are devoid of anthers, and twice as long.

3. Small-flowered variety. Diameter of corolla only about 3 cm.; all the
stamens reduced to small vestiges at the base of the calyx-tube, so that the flowers
are female. These female plants are rare at Ruppin.

The dissected blade of the petals of all the varieties is either from light to dark
violet in colour, or else pure white. In the former case, the patch at the base of the
blade is dirty-green, and beset with long purple hairs, while the rest of the blade
is covered with very short violet hairs: in the latter case the basal patch is of
a beautiful bright green, and covered with colourless hyalineh airs, as also is the
rest of the blade. It is also remarkable that these white-flowered plants are dis-
tinguishable, even at a distance, from dark-flowered specimens growing near them,
by the pale-green colour of stem, branches, leaves, and calyx (Warnstorf).

383. D. Armeria L. (Kirchner, ‘Flora v. Stuttgart, p. 245; Schulz, ‘ Beitrige,’
III, p. 21.)—Kirchner says that though the flowers essentially agree with those of
D. deltoides, yet—in accordance with their inconspicuousness—automatic self-pollina-

156 ANGIOSPERMAE—DICOTYLEDONES

tion is possible, the styles being developed before the outer anthers have lost all
their pollen. The corolla is bright red with clearer spots; its diameter is 13 mm.,
its tube 15 mm. long, and scarcely 2 mm. wide. Besides the hermaphrodite flowers
there are some in which one staminal whorl is vestigial, and others again that are
purely female, the yellow anthers remaining enclosed in the corolla-tube, and not
dehiscing. The plant is therefore gynodioecious and gynomonoecious.

Visttors.—According to the observations of A. Schulz, these are very infre-
quent: he only saw one butterfly (Vanessa urticae).

384. D. Carthusianorum L. (Sprengel,‘ Entd.Geh.,’ pp.250-1 ; Herm. Miiller,
‘Fertilisation,’ pp. 126-7, ‘Weit. Beob.,’ Il, p. 230; Schulz, ‘Beitrage,’ I, p. 5;
Knuth, ‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 36-7, ‘ Bloemenbiol. Bijdragen.’)—The
mechanism of the protandrous diurnal hawk-moth flowers agrees with that of D.
deltoides. There are female flowers in addition to the hermaphrodite ones. The
plant is gynodioecious, more rarely gynomonoecious.

Fic. 48. Dzanthus Carthustanorum, L. (Natural size, after removal of calyx and three of the

petals. From nature, semi-diagrammatic.) (1) a, two stamens of the outer whorl, one with pollen,
the other dehisced and empty ; a', two stamens of different lengths belonging to the inner whorl; s, stigma
(undeveloped); c, two petals; 7, nectaries. (2) a, two withered stamens of the outer whorl: a}, two

mature stamens of the inner whorl; s, c, 22, asin (1). (3) a, a', withered stamens of the outer and
inner whorls respectively; s, mature stigmas ; ¢, #, as in (1). (4) Variety with vestigial stamens.
(5) Flower in the first (male) condition, seen from above. (6) Flower in the second (female) condition,
seen from above.

Warnstorf (Verh. bot. Ver.. Berlin, xxxvii, 1895) found both large and small
flowers at Ruppin. The former are androdynamous-protandrous. At the time when
the pollen is mature the lilac anthers project far beyond the style. Pollen-grains
large, rounded, covered with a delicate net-work of tubercules, 44-50 » in diameter.
The small flowers are imperfectly hermaphrodite. At the time when the stigma
is mature the stamens are much shorter than the style. The yellowish anthers
are smaller; their pollen-grains are polyhedral and papillose, with a maximum
diameter of 31 p.

Visirors.—On the North Frisian Islands I saw Lepidoptera, and—as unbidden
guests—small bees, as well as various flies, beetles, and grasshoppers.

CARYOPHYLLEAE 157

Hermann Miiller (H. M.) and myself (Kn.) observed the following in Central
Germany.—

Lepidoptera. (a) Noctucdae: 1. Plusia gamma JZ. (Kn., H.M.), freq., skg.
(6) Rhopalocera: 2. Coenonympha arcania L. (H. M.),skg.; 3. Colias hyale Z. (Kn.) ;
4. Hesperia sp. (H.M.), repeatedly ; 5. H. lineola O. (H. M.), very freq., skg.; 6. H.
silvanus £sp. (H. M.), do.; 7. Melanargia galathea Z. (H.M.), skg.; 8. Polyommatus
phlaeas Z. (Kn., M.H.); 9. Rhodocera rhamni Z.(Kn., H.M.), freq.; 10. Syrichthus
malvae Z. (H.M.), freq., skg. (c) Sphingidae: 11. Macroglossa stellatarum Z.
(H.M.); 12. Zygaena carniolica Scop. (H.M.); 13. Z. lonicerae Esp. (H. M.), skg.
freq.; 14. Z. pilosellae Zsp. (Kn., H.M.), do.; 15. Z. trifolii Zsg. (Kn.), do.

All skg.

As useless visitors Herm. Miiller also observed the following beetles :—Oedemera
podagrarieae Z.; Danacea pallipes Pz.; and Spermophagus cardui Svev.: and
Rossler—at Wiesbaden—saw 2 Lepidoptera, nect-skg., i.e. Ino geryon fdd., and
Dianthoecia compta /.

385. D. chinensis L.—This species bears protandrous Lepidopterid flowers.

Visrrors.—Herm. Miiller saw moths—Plusia gamma Z., Agrotis pronuba /..,
Brotolomia meticulosa Z.—on the flowers in his garden.

3&6. D. barbatus L.—This species bears protandrous diurnal Lepidopterid
flowers (Sprengel, ‘Entd. Geh.,’ p. 251).

Visirors.—I observed Macroglossa stellatarum Z. in the gardens of Féhr and
Helgoland, and in the latter some butterflies—Pieris brassicae Z., Vanessa urticae
L.—as well. All skg.

387. D. sylvestris Wulf. (Herm. Miiller, ‘ Fertilisation,’ p.127, ‘ Alpenblumen,’
pp. 204-5; Schulz, ‘ Beitrage, II, pp. 22-3.)—This species possesses protandrous
diurnal Lepidopterid flowers. They are fragrant and rose-coloured, and expand
into a disk 25-35 mm. in diameter. The nectar is so deep—according to Schulz
as much as 18-25 mm.—that it is as much as diurnal Lepidoptera can do to reach
it, and Miiller says that a proboscis 18-20 mm. long is required. Otherwise the
mechanism of the flower agrees with that of D. deltoides, the other species already
described. According to Schulz, the plant is sometimes gynodioecious, more rarely
gynomonoecious.

Visirors.—Macroglossa stellatarum Z. has been observed by Miiller in the
Suldenthal, and by Schulz at Bozen.

388. D. atrorubens All.—This species bears protandrous diurnal Lepidopterid
flowers, of which the dark-red petals are covered with darker hairs and marked with
darker spots. The nectar is deeply seated—13~-15 mm. according to Miiller, ro-
17 mm. according to Schulz—and is accessible to many diurnal Lepidoptera.
Besides fully-developed protandrous hermaphrodite flowers, Schulz observed some
female ones, distributed gynodioeciously, more rarely gynomonoeciously.

Visttors.—Herm. Miiller observed butterflies—4 species, skg.—in the Alps,
and also Zygaena minos W. V. (=Z. pilosellae £sf.), vainly trying to reach the
nectar. A. Schulz (‘ Beitrage,’ II, p. 22) saw 2 species of Lepidoptera at Bozen
in the Tyrol.

389. D. arenarius L.—This species bears protandrous nocturnal hawk-moth
flowers (?), the mechanism of which has been described by Kirchner (‘Beitrige,’

158 ANGIOSPERMAE—DICOTYLEDONES

p. 18) from garden specimens. The calyx is 16 mm. in length, and only 24-3 mm.
wide. It closely ensheaths the claws of the white petals, which project about g mm.
beyond it. The stamens and carpels develop in the usual order.

Visttors.—The deeply seated nectar, and the white colour of the corolla,
suggest that the flowers are pollinated by nocturnal hawk-moths,

390. D. monspessulanus L.—According to Schulz (‘ Beitrage,’ II, p. 23) the
nectar is placed at a depth of 14-25 mm. in the flesh-coloured or white Lepidopterid
flowers, of which the diameter varies between 25 and 35mm. The order of develop-
ment of stamens and carpels is the same as in other species. Near Bozen Schulz
also observed female flowers, with a minimum diameter of 8 mm.

VisiTors.—Macroglossa stellatarum Z., which possesses a proboscis 25-28 mm.
long, that can easily reach all the nectar, was observed by A. Schulz at Bozen,
and also by G. E. Mattei (‘I lepidotteri e la dicogamia,’ 1888, p. 16). MacLeod
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 377) did not see any normal visitors
in the Pyrenees, but only a flower-beetle.

391. D. Caryophyllus L.—According to Darwin, this species is self-sterile.

392. D. neglectus Loisel.—Kerner says that the flowers of this species are
protandrous, though later on automatic self-pollination is possible. The flowers
are open in the morning, and between 6 and 7 o'clock in the evening.

393. D. glacialis L—According to Kerner, the flowers are at first protandrous,
but automatic self-pollination may afterwards take place. The plant is also
gynodioecious.

394. D. caesius Sm.—Kirchner (‘Beitrage,’ pp. 17-18) states that the
mechanism of the rose-coloured flowers—which smell strongly of cloves—agrees
with that of D. sylvestris, Besides the protandrous hermaphrodite flowers, some
stocks—at Uberlingen—also bear female flowers of the same size.

395. D. Seguierii Vill—Besides the protandrous hermaphrodite flowers, Schulz
observed female ones, which were either borne on the same plants as the former
or on different ones.

396. D. plumarius L. (Knuth, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.’
p. 231.)—

Visitors.—I observed Bombus hortorum Z., skg., in the gardens on the island
of Fohr.

118. Saponaria L.

Markedly protandrous Lepidopterid flowers. The petals narrow abruptly into
long-winged claws. The somewhat ventricose calyx—which is not surrounded by
bracts—holds these together to form a long tube in which the nectar is concealed,
and which is prolonged upwards by a corona consisting of bifid ligules. There
may be gynomonoecism and gynodioecism.

397. S. officinalis L. (Sprengel, ‘Entd. Geh.,’ p. 248; Herm. Miller,
‘Fertilisation,’ pp. 128-9, ‘ Weit. Beob.,’ II, p. 232; MacLeod, ‘Pyreneenbl.,’ p. 101,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 151-3; Kirchner, ‘Flora v. Stuttgart,’
p- 246; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.’ pp. 37-8, 151; Schulz, ‘ Beitrage,’

CARYOPHYLLEAE 159

I, p. 6.)—The white or flesh-coloured petals of the protandrous hawk-moth flowers
are devoid of nectar-guides. A fragrant odour is exhaled, becoming much stronger
in the evening. Nectar is secreted and concealed at the bottom of the calyx-tube—
which is 18-21 mm. in length, and is prolonged a few mm. by the corona—and can
only be reached by Lepidoptera possessing a very long proboscis. The five outer
stamens first project from the flower, above the entrance to which their anthers
dehisce. After shedding their pollen they diverge, leaving the entrance of the flower
free for the inner whorl of stamens. When these have shed their pollen the two
styles grow up, spreading out their stigmas at the level previously occupied by
the anthers. Self-pollination is therefore excluded. The species is sometimes gyno-
dioecious, more rarely gynomonoecious.

Vistrors.—These are chiefly hawk-moths (Sphinx and Macroglossa), Hermann
Miiller saw Sphinx ligustri Z., skg.; I myself noticed Macroglossa stellatarum Z. in
the Kiel Botanic Garden. MacLeod very frequently saw this species in the Pyrenees;
also Sphinx convolvuli Z., of which a single individual visited twenty-nine flowers in
two minutes. Kerner observed Noctuidae belonging to the genera Dianthoecia
and Mamestra. Owing to the shortness of their proboscis, butterflies cannot secure
the nectar: I observed, for example, Vanessa io Z. vainly trying to suck the flower
in the North Frisian islands. Nor was the honey-bee able to reach the nectar.
I also noticed there po-dvg. hover-flies and Lucilia caesar Z. Buddeberg found
a small bee—Halictus morio /. g—po-cltg. on the flowers as an unbidden guest.

In the Netherlands, H. de Vries (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel,
1875) observed a humble-bee—Bombus terrester Z. 9—probably only po-cltg.

398. S. ocymoides L.—This species bears protandrous Lepidopterid flowers.
Hermann Miller (‘ Alpenblumen,’ pp. 200-1) states that the stamens and carpels
develop in the same order as among the Sileneae already described, and that
automatic self-pollination is exceptionally possible as a last resort. A divergent
feature is that each of the five outer stamens expands at its base into a fleshy
red appendage, which probably secretes the nectar. Besides protandrous herma-
phrodite flowers, there are also purely female ones, and in rare instances purely
male ones as well. The plant, according to Hildebrand (‘Die Geschlechtsvert. b.
d. Pfi.” p. r1), is gynodioecious, gynomonoecious, androdioecious, andromonoecious,
or even trimonoecious (¥=6 >9).

Visitors. — Herm. Miiller observed numerous Lepidoptera (more than 30
species) in the Alps; also some humble-bees (3 species) and Bombyliidae (2 species),
which reached the honey only with difficulty. A. Schulz (‘ Beitrage,’ II, pp. 24-6)
also noted a preponderance of Lepidoptera (35 species) in the South Tyrol, together
with a few humble-bees nect. skg., and po-dvg. flies. This observer very frequently
found the flowers of which the calyx had been perforated and the nectar stolen by
Bombus mastrucatus Gers/., or more rarely by B. terrester Z.

119. Vaccaria Medicus.

Protandrous, homogamous, or feebly protogynous Lepidopterid flowers. Calyx
ventricose—as a protection against humble-bees—acutely pentagonal, almost winged,
and devoid of basal bracts; petals with winged claws; no corona. Sometimes
gynomonoecious and gynodioecious,

160 ANGIOSPERMAE—DICOTYLEDONES

399. V. parviflora Moench (=S. Vaccaria Z.). (Herm. Miiller, ‘ Weit. Beob.,’
II, pp. 231-2; Schulz, ‘ Beitrage,’ II, pp. 23-4; Kirchner, ‘Flora v. Stuttgart,’
p. 247.)—This species bears butterfly flowers. Herm. Miiller says that the expansion
of the calyx is so great that its diameter below the middle may be as much as 7 mm.
Its ventricose surface is deeply enfolded between the sharp projecting longitudinal
ribs. The protection against robbery by humble-bees—e.g. Bombus terrester—
is thus rendered more effective, as the bees cannot perforate the folds, and if they
bite through the edges can scarcely reach the nectar. The folds also hold together
the claws of the petals, which are flesh-coloured to rose-red. The calyx-tube narrows
above, so that it is almost closed by the petals, stamens, and pistil, and can only
be traversed by the proboscis of a butterfly. Nectar is secreted in small quantities
at the bottom of the calyx, at a depth of 15-18 mm. The flowers are sometimes
slightly protogynous, not infrequently protandrous, and often homogamous. Auto-
matic self-pollination is always possible. Kerner says that it is brought about
by continued growth of the stamens. Although autogamy is indicated by the
inconspicuousness of the flowers, cross-pollination by insects is ensured at the
beginning of anthesis. Female flowers have been observed, in addition to the herma-
phrodite ones. The plant is gynomonoecious and gynodioecious.

Vistrors.—Schulz observed butterflies in eastern Westphalia, i.e. species of
Pieris, especially P. brassicae Z.

120. Cucubalus Tourn.

The ventricose companulate calyx protects the flower from humble-bees; the
petals gradually narrow into the long claws.

400. C, baccifer L.—Schulz (‘ Beitrage,’ II, p. 81) states that the hermaphrodite
flowers of this species are protandrous. There are also a small number of female
{lowers on the same, or on special plants.

121. Silene L.

Protandrous, homogamous, or protogynous flowers, with nectar concealed in
various ways. The calyx is tubular to ventricose, serving in the latter case as
a protection against the bites of nectar-stealing humble-bees. A corona is sometimes
present. The long claws of the petals are often held together so firmly by the
calyx that the approach to the nectar secreted in the base of the flower is only
available to Lepidoptera; many species of this genus, therefore, belong to the flower
class L. In other species the nectar is easily accessible to long-tongued bees,
and such species are placed in class H. There are also many flowers in which
the nectar is still less deeply concealed, and these belong to class C. One species—
S. Otites—is peculiar in being chiefly anemophilous. Gynomonoecism and gyno-
dioecism are common.

According to Rohrbach (‘ Monogr. d. Gatt. Silene,’ Leipzig, 1868, pp. 41-3),
the following species are exclusively autogamous :—S. antirrhina Z., S. apetala Willd.,
S. cerastoides Z., S. clandestina Jacg., S. gallica Z., S. hirsuta Lag., S. inaperta Z.,
S. longicaulis Pourr., S. tridentata Des/-

CARYOPHYLLEAE 161

Batalin observed cleistogamous flowers on the following species of Silene :—
S. vilipensa Kunze, S. hirsuta Lag., S, gallica Z., S. cerastoides Z., S. tridentata Des/,,
S. clandestina Jacg., S. longicaulis Pourr., S. apetala Willd., S. inaperta Z., S. antir-
rhina Z,

401. S. inflata Sm. (=S. vulgaris Garcke). (Axell, ‘Om Anord. for Fanerog.
Vaxt. Befrukt.,’ p. 46 ; Herm. Miiller, ‘ Fertilisation,’ p. 129, ‘Alpenblumen,’ pp. 198-9 ;
Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, II; Kirchner, ‘Flora v. Stuttgart,’ p. 248;
A. Schulz, ‘ Beitrige, I, pp. 9-10; MacLeod, Bot. Jaarb. Dodonaea, Ghent, iii, 1891,
PP. 374-5, vi, 1894, p. 154; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 38-9, 151,
‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 231; Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896.)—The species bears protandrous-trioecious Lepidopterid and
humble-bee (?) flowers. The white bilobed petals possess no nectar-guides. The
nectar is concealed at a depth of ro-rz mm., but the enuance to the flower is
Not very narrow, admitting even the proboscis of a humble-bee. The female flowers
are smaller than the male and hermaphrodite ones, and the first two kinds each
possess vestigial reproductive organs of the opposite sex. The hermaphrodite flowers
are protandrous, and capable of self-pollination. The inflated calyx is not always
an effective protection against the bites of nectar-stealing humble-bees, for Bombus
terrester and B. mastrucatus sometimes—though not always, as MacLeod observed
in the Pyrenees—succeed in stealing the nectar after perforating the calyx.

According to Schulz, the sexes in this species are distributed in a five-fold
manner; there being hermaphrodite, male, female, gynomonoecious and andro-
monoecious stocks. The occurrence of these forms is very variable; in some places
male plants seem to be wanting altogether, or at least to be rarer than female ones.

Visttors.—These are partly Lepidoptera (mostly moths), partly humble-bees,
skg. in both cases. In the North Frisian Islands I saw 2 Lepidoptera—Plusia gamma
Z£. and Epinephele janira Z.—and a humble-bee—Bombus lapidarius Z. Kerner
observed Noctuidae—Dianthoecia and Mamestra—in the Tyrol. Loew (‘ Beitrige,’
p- 28) noticed Bombus agrorum F. ¥, skg., in Lower Silesia. Réssler—at Wies-
baden—saw the moth Dianthoecia nana Roff., skg. Herm. Miiller observed Lepi-
doptera—2 moths and 3 butterflies—and 7 species of humble-bee in the Alps.
MacLeod—in the Pyrenees—noticed 3 species of humble-bee, a wasp, a Bombylius
(skg. while resting on the flower!), and a po-dvg. Muscid, but no Lepidoptera.
Lindman saw a humble-bee, a Lepidopterid, and a fly by the Hardanger Fjord,
and Warnstorf—in Brandenburg—observed numerous ants as unbidden guests.

In Dumfriesshire an Empid, and numerous Muscids and hover-flies have been
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 23).

402. S. nutans L. (Sprengel, ‘Entd. Geh., p. 252; Ricca, Atti Soc. ital. sc.
nat., Milano, xiv, 1871; Herm. Miller, ‘Fertilisation, p. 129, ‘Alpenblumen,’
pp. 197-8; Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, II; Schulz, ‘ Beitrage,’ I, pp. 6-7;
Kirchner, ‘Flora v. Stuttgart,’ p. 248; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’
pp. 40-1.}—The protandrous moth flowers—which smell like hyacinths at night—
are of a dirty white colour, devoid of nectar-guides, and conceal their nectar at
a depth of 13-14 mm. Kerner says that in the Tyrol they mature their stamens
and pistil on three consecutive nights. The hermaphrodite flowers—which are the

DAVIS. I M

1€2 ANGIOSPERMAE—DICOTYLEDONES

commonest—are markedly protandrous. The anthers of the outer whorl of stamens
mature during the first night, those of the inner whorl during the second night,
and the stigmas during the third. Self-pollination is therefore excluded. The
flowers are closed and devoid of odour during the day. Besides the hermaphrodite
blossoms, male and female ones have been observed. The plant is gynomonoecious
and gynodioecious, andromonoecious and androdioecious. The female blooms are
smaller than the others, but a small-flowered hermaphrodite variety also occurs.
The anthers project far out of the flower, and the stigmatic papillae are long, which
characters Schulz regards as indications of anemophily. This investigator—working
at Halle, and in Thuringia, the Tyrol and North Italy—was unable to confirm
Kerner’s observations except in a few points. He found, on the contrary, that the
stamens and pistil may mature at any time, and that the interval between the three
stages of development is not so regular as Kerner asserts. Schulz also questions
the general correctness of Kerner’s statement that the petals always roll up in the
daytime, thus making the flowers quite inconspicuous; for on high regions (2,000-
2,200 m, above sea-level) this rolling-up only occurs in very sunny spots during the
middle of the day. The plants I investigated in Sylt and subsequently in the garden
of the Realschule at Kiel, agreed with Kerner’s accov...: in the daytime they looked
faded, and were quite odourless; at dusk their petals became turgid, and a powerful
scent of hyacinths was exhaled. Unfortunately, I did not see any guests.

Visirors.— Owlet-moths (Noctuidae) have been observed — Dianthoecia and
Mamestra—which, Kerner says, lay their eggs in the flowers, and according to
Buchanan White (Justs bot. Jahresber., Leipzig, 1873, p. 377), they stand in much
the same relation to Silene as the Yucca-moth to Yucca (cf. Vol. I, pp. 102-3).

Rossler observed the following Lepidoptera at Wiesbaden.—1. Cidaria hydrata
Tr.; 2. Dianthoecia albimacula Bkh.; 3. Cucullia chamomillae Schiff.

Herm. Miiller—in the Alps—saw 2 butterflies and a moth, also 3 species of
humble-bee, of which two obtained nectar by perforating the flowers. Loew (‘ Bei-
trage,’ p. 63) observed Bombus hortorum Z. ¥, skg., by Lake Como, and Frey saw
the moth Pterogon proserpina Pal/. Schulz noted butterflies and moths, besides
humble-bees, of which some obtained nectar by perforation.

403. S. dichotoma Ehrh.—The white flowers are protandrous. Gynodioecism
was observed by Warming in Denmark, and by Kirchner in Wurtemberg.

404. S. Armeria L.—MacLeod says that the rose-red flowers are distinctly
protandrous, with a corolla-tube 16-18 mm. long, and nectar accessible only to
Lepidoptera.

Besides the hermaphrodite flowers, Breitenbach has observed (Kosmos, Stuttgart,
xiv, 1884) female ones on distinct plants in the Botanic Gardens of Marburg and
Gottingen. The flowers are protected against creeping animals (ants) by the sticky
nature of the upper internodes.

Visrrors.—MacLeod observed diurnal hawk-moths—Macroglossa—and moths—
Plusia—in Belgium.

405. S. longifiora Ehrh.—The white flowers smell like hyacinths at night,
and open between eight and nine in the evening.

VisiTors.— Kerner says that moths are the pollinating agents,

CARYOPHYLLEAE 163

406. S. viridiflora L—According to Kerner, the flowers exhale an odour of
hyacinths at night. ,

VisiTors.— The species appears to be. moth-pollinated. Schletterer observed
the bee Halictus patellatus or. at Pola.

407. S. Otites Sm. (Herm. Miller, ‘ Weit. Beob.,’ IT, p. 234; Schulz, ‘ Bei-
trage, I, p. 78; Verhoeff, ‘Bl. u. Insekt. a. d. Ins. Norderney’; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ pp. 39-40, 151, ‘ Weit. Beob. ti. Bl. u. Insekt. a. d. nordfr.
Ins.,’ p. 232.)—The plant is almost dioecious (entirely so in the island of Rém).
The male blossoms are much more numerous than the female ones, while hermaphro-
dite flowers occur occasionally. The petals only project 2-3 mm. beyond the calyx—
which is 4 mm. long—within which they are almost entirely concealed in female
flowers from some localities. Nectar is secreted and concealed in the base of the
flower. The nectaries of the male flowers are non-functional in Central Germany ;
those of the female flowers in the same region are not accessible to insects in the
normal way, owing to the closeness of the calyx to the ovary (Schulz). In the North
Frisian island of Rém, and in the Tyrol, both kinds of flower secrete nectar, which
is accessible to and secured by insects. It would appear, therefore, as if the flowers
were in part anemophilous, and in part entomophilous. Of the ten stamens of the
male flower, only the five of the outer whorl at first mature, at the same time pro-
jecting 3-4 mm. beyond the calyx-tube. They are afterwards replaced by the five
stamens of the inner whorl. In the female flower the stigma also projects a few mm.
We must suppose that, even in these nectar-producing flowers, wind is the chief
pollinating agent, for none of the female flowers I observed in the island of Rém
remained unfertilized, in spite of the very limited number of insect visitors. This is
confirmed by the great preponderance of male flowers. The hermaphrodite flowers

are distinctly protandrous. The plant is usually protected against creeping insects
by a sticky stem.

Visitors. —I observed 4 nectar-sucking Lepidoptera in the North Frisian
Islands, i.e. Epinephele janira Z., Coenonympha pamphilus Z., Plusia gamma Z.,
and Zygaena filipendulae Z.; and also a few Hemiptera, vainly seeking for nectar.
Verhoeff saw Plusia gamma Z., skg.,on Norderney. Herm. Miiller noticed fossorial
wasps—Cerceris variabilis Schr. 9 and 4, and Philanthus triangulum F’. é—nect. skg.,
at Kitzingen.

408. S. gallica L.—

VisiTors.— Buddeberg observed a small po-cltg. bee, Halictus smeathma-
nellus K. 9 (Herm. Miller, ‘ Weit. Beob.,’ II, p. 235).

409. S. Saxifraga L. (= S. petraea Waldst. ef Kit.).—Lalanne (Justs bot.
Jahresber., Leipzig, xvi, 1888, p. 563) states that sometimes the ovaries, sometimes
the anthers are vestigial. Kerner describes the plant as trioecious, with strongly pro-
tandrous hermaphrodite flowers, that only open in the evening between eight and
nine o'clock.

Visirors.—These are apparently moths.
410. S. rupestris L. (Herm. Miiller, ‘Alpenblumen,’ pp. 193-4; Schulz,

‘ Beitrage,’ II, pp. 29-30.)—The hermaphrodite flowers are protandrous, but the
, M 2

164 ANGIOSPERMAE—DICOTYLEDONES

possibility of automatic self-pollination is apparently not precluded. Plants observed
by Warming in the Scandinavian highlands were also strongly protandrous, as were
those examined by Schulz in the Tyrol. The latter describes the plant as gynodioe-
cious, rarely gynomonoecious, very rarely androdioecious or andromonoecious.

Visitors.—Loew observed the following in Switzerland (‘ Beitrage,’ p. 60).—
Diptera, Bombylidae: 1. Argyromoeba sinuata Fall.; 2. Bombylius minor Z.

Herm. Miiller—in the same country—saw numerous Lepidoptera, especially
‘Noctuidae, and flies; also some bees.

A. Schulz observed a similar circle of guests in the Tyrol.
MacLeod noticed a Muscid in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii,
1591, Pp. 375)-

_ Fic. 49. Silene rupesiris, L. (after Herm. Miiller). A, Flower in the first (male) stage. B.
Pistil of the same flower with the branches of the style still closely apposed, and stigmas not yet mature.
C, Flower in the second (hermaphrodite) stage. D., Flower in the third (female) stage.

4u. S. acaulis L. (Herm. Miller, ‘ Fertilisation, p. 129, ‘Alpenblumen,’
pp. 194-7; MacLeod, Bot. Jaarb, Dodonaea, Ghent, iii, 1891, pp. 375-6; Ricca,
Atti Soc. ital. sc. nat., Milano, xiii, 1870.)—This species is trioecious. The markedly
protandrous hermaphrodite red flowers are crowded together in great numbers, and
are visited by so many insects that there is scarcely any or no necessity for automatic
self-pollination. In some places, however, none of the flowers are bisexual (Schulz,
Warming). Ekstam says that in dioecious plants in Nova Zemlia the diameter of
red or white corolla is 6-12 mm. The only unisexual flowers there observed were
male; the bisexual ones were—as elsewhere—protandrous.

According to Andersson and Hesselman (‘ Bidrag till Kanned. om Spetsbergens
o. Beeren Eil. Karlvaxtflora ’), this species flowers in Spitzbergen from the beginning
of July to the end of August; during the latter month fruits are regularly ripened.

CARYOPHYLLEAE 165

The colour of the corolla varies from pure white—as in males infested by Ustilago
violacea—to bright violet and red violet. In Spitzbergen—as in Greenland—there
are male and female flowers as well as hermaphrodite ones. On a clump of the
plant containing the remains of several hundred female flowers of the preceding year,
the above-named investigators only found two ripe seed-containing fruits, In their
opinion the fact indicated the ineffectiveness of cross-pollination in Spitzbergen. The
clumps of Silene acaulis often show very beautifully the effect of direct isolation on
anthesis, for the southern half—exnosed to the sun—presents a dense mass of expanded
flowers while the northern half remains more or less in bud.

Fic. 50. Szlene acaulr's, L. (after Herm. Miller). 4. Flower half-way through the first (male) condi-
tion. B. Smaller male flower at the end of anthesis. C. Female flower, seen from above. D. The
same in section. 2. Hermaphrodite flower at the end of the male stage. F. Pistil of the same flower.

Vistrors.— Herm. Miiller observed—in the Alps—numerous Lepidoptera—
28 Macro-lepidoptera, and 4 Micro-lepidoptera—as well as some bees, Muscidae,
Syrphidae, and anther-eating beetles.

Frey—in the same region—saw the moth Anarta melanopa Zhund., and in
Switzerland the species Anarta nigrita Bsd. MacLeod observed 6 Lepidoptera, and
3 pollen-eating beetles in the Pyrenees. Lindman noticed a humble-bee on the.
Dovrefjeld, as did Ekstam in Nova Zemlia. Schneider (Tromse Mus. Aarsh., 1894)
saw Bombus agrorum Z. and B. lapponicus Z. in Arctic Norway. For Spitzbergen,
Ekstam (‘ Bliitenbiol. Beob. a. Spitzbergen,’ p. 24) records two cases of visits by
small Dipterids, and one by a spider (!), which was doubtless hunting flies.

412. S. noctiflora L. (= Melandrium noctiflorum /7ries)—The hermaphrodite
flowers are fragrant when they open in the evening, between seven and eight o’clock,
according to Kerner, though Warnstorf has seen them open as early as between five

166 ANGIOSPERMAE—DICOTYLEDONES

and six p.m. MacLeod states that the protandry is so marked that self-pollination
is excluded. The nectar is concealed at a depth of 18 mm.

In Belgium the plant is gynomonoecious (MacLeod); Warnstorf observed
gynodioecism at Ruppin; and Schulz says that this is the most frequent condition,
but that in some places androdioecism and andromonoecism may occur. Schulz
now and then observed flowers which had been perforated by bees. The female
flowers—which were long since observed by Girtner—are only 12 mm. deep.
According to Hansgirg, pseudo-cleistogamous flowers occasionally make their
appearance.

Visitors.— These are probably moths.

413. S. conica L.— According to Kerner (‘Nat. Hist. Pl.’ Eng. Ed. 1, II,
pp- 336), the flowers are protandrous. The anthers of the outer whorl open first,
and after they have fallen off, the mature stigmas spread out. The anthers of the
inner whorl now dehisce, and by lengthening of their filaments are brought into
contact with the stigmas. These stages in anthesis are passed through in the
course of a single day.

414. S. vespertina Retz.—Kerner states that the flowers of this species
open in the evening between seven and eight o'clock.

415. S. Elisabethae Jan.—According to Loew—who studied garden speci-
mens—the flowers of the species belong to class H. They are protandrous. Though
the calyx is wide open, and the claws of the petals diverge considerably, the flowers
are often perforated by humble-bees. Kerner states that fruits with seeds capable of
germinating are rare.

416. S. Pumilio Wulf—Flowers of this species perforated by humble-bees were
found by Kerner in the Taurn.

417. S. valesiaca L.—According to Kerner, the flowers of this species open
between eight and nine o’clock in the evening.

418. S. maritima With—Plants observed by Warming in the Altenfjord, bore
hermaphrodite flowers which exhibited fairly well-marked protandry, though self-
pollination was ultimately possible. According to Gibson (‘ Flora of St. Kilda’), the
plant is probably pollinated by flies in St. Kilda, the outermost island of the Scottish
west coast—except Rockall, which is barren—for fruits are occasionally formed,
though there are no Lepidoptera, bees, nor wasps.

419. S. inaperta L. (= S. vilipensa A’unze?).—Batalin states that the flowers
of this species are completely cleistogamous, with their openings entirely closed by the
calyx-teeth.

420. S. linicola Gmel.—Kirchner was unable to recognize any secretion of
nectar in cultivated specimens of this species. ‘The diameter of the somewhat incon-
spicuous flowers is at first only 4-5 mm. At the beginning of anthesis five of the
ten stamens are so far developed that they dehisce, their pollen-covered anthers
coming into contact with the three stigmas, so that automatic self-pollination results.
The limbs of the petals subsequently spread out horizontally, increasing the diameter
of the flower to 8-9 mm. At this stage the other five stamens elongate until their
anthers occupy the mouth of the flower, while the five older anthers wither and
fall off.

CARYOPHYLLEAE 167

Visirors.—Only Thrips (larvae and adults) has so far been observed.
421. S. Bastardi Bor.—

VisiTors.—Loew observed the bee Halictus sexnotatus K. 9 creeping into the
flower in the Berlin Botanic Garden.

422. S, petraea.—Lalanne and Caille (Actes soc. linn., Bordeaux, Ixi, 1887),
states that this species is dimorphous.

122. Viscaria Riv.

Protandrous butterfly flowers, more rarely homogamous or protogynous. Petals
red, with corona and linear claws. Gynomonoecious and gynodioecious, rarely
androdioecious or andromonoecious.

423. V. vulgaris Roehl. (= Lychnis Viscaria Z.). (Herm. Miiller, ‘Weit. Beob.,’
II, pp. 233-4 ; A. Schulz, ‘ Beitrage,’ II, p. 32 ; Kirchner, ‘ Flora v. Stuttgart,’ p. 250 ;
Knuth, ‘Bloemenbiol. Bijdragen’; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—
This species bears protandrous butterfly flowers. The red calyx helps to make the
flower conspicuous. It is 13mm. long, but a proboscis of 7-8 mm. is able to reach
the nectar without forcing open the mouth of the flower, for the floral axis is
prolonged 5 mm. beyond the insertion of the calyx before the petals, stamens, and
pistil take origin from it. The rose-red stellate corolla is 18-20 mm. in diameter.
The claw of each petal is prolonged into a deeply bifid ligule 3 mm. long, and which
is curved outwards in such a way that the entrance of the flower is widened to the
extent of 3-5 mm. At the commencement of the anthesis of the hermaphrodite
blossom, the five outer stamens, their anthers covered with pollen, are situated
between the ligules. The anthers of the five inner stamens—which dehisce either
somewhat later or else simultaneously—are placed rather further down in the mouth
of the corolla-tube. When the anthers have shed their pollen, the filaments bend
outwards and downwards beyond the corolla, while the styles elongate so that the
stigmas come to lie above the tips of the ligules. According to Schulz, the herma-
phrodite flowers are sometimes homogamous. Besides the bisexual blossoms, uni-
sexual ones have been observed (female, rarely male) on the same or on different
stocks (gynomonoecism, gynodioecism, rarely andromonoecism and androdioecism).
According to Schulz, the stigmas of purely female flowers do not mature till after the
beginning of anthesis.

Warnstorf states that larger and smaller flowers occur at Ruppin. The former
are completely hermaphrodite and protandrous. At the time when the pollen is
ripe, the styles are still very short, the long stamens with their lilac anthers projecting
far beyond them. They elongate subsequently, and protrude far out of the
flower. The smaller flowers are at first imperfectly hermaphrodite, afterwards
becoming purely female by the absorption of the small yellow anthers of the short
stamens, beyond which the styles always project. The pollen-grains of the normal
anthers are spherical, white, transparent, almost smooth, and measure about
31-37°5 #, rarely as much as 50 » in diameter; while those of the vestigial anthers
of the smaller flowers are rounded-polyhedral, delicately papillose, and only about
25 » in diameter.

168 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Schulz observed numerous butterflies in the Tyrol, and also flies as
unbidden guests. Herm. Miiller—in Westphalia—saw a few Lepidoptera effecting
cross-pollination (Ino statices Z. and I. pruni ScAif., skg.), and—as unbidden guests
—certain Sphegidae (Gorytes quinquecinctus F. 9) and beetles (Meligethes).

In Thuringia I observed two humble-bees (Bombus soréensis /., var. proteus,
Gerst. §, and B. terrester Z. ¥, both skg.). Loew noticed Pieris brassicae Z., skg., in
the Berlin Botanic Garden.

424. V. alpina Don. (Axell ‘Om Anord. for Fanerog. Vaxt. Befrukt.,’ p. 33;
Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, II; Kirchner, ‘ Beitrage,’ p. 17.)—This species
bears protandrous butterfly flowers which possess an odour of vanilla. Kirchner
states that they are for the most part bisexual and protandrous at Zermatt, though
female stocks are not infrequent. The hermaphrodite blossoms are 10-12 mm. in
diameter ; the female ones, in which the stamens are reduced so much that they
scarcely attain the length of the ovary, are 6-8 mm. in diameter. Warming, in
Greenland, besides plants similar to those found by Kirchner at Zermatt, also observed
protogynous flowers, and noted the occurrence of smaller gynodioecious female
flowers, as well as of intermediate forms between female and bisexual ones. It
is doubtful whether purely male flowers exist in Greenland. The same investigator
found that this plant is gynodioecious, gynomonoecious, and andromonoecious in
Scandinavia. The hermaphrodite flowers are capable of self-pollination in the later
stages of anthesis.

Visitrors.—The only one hitherto observed (in Scandinavia) is a butterfly,
Argynnis pales Schiff.

123. Coronaria L.

Protandrous Lepidopterid flowers. Petals with a ligule and divided or undivided
limb. Nectar secretion as usual.

425. C. Flos-cuculi A. Br. (=Lychnis Flos-cuculi Z.). (Sprengel, ‘ Entd. Geh.,’
p. 261; Herm. Miller, ‘Fertilisation,’ pp. 129-30, ‘Weit. Beob.,’ II, p. 232; Kerner,
‘Nat. Hist. Pl,’ Eng. Ed. 1,11; Schulz, ‘ Beitrage,’ I, pp. 11-12; Loew, ‘ Bliitenbiol-
Floristik,’ pp. 392, 395; Kirchner, ‘Flora v. Stuttgart,’ p. 251; Knuth, ‘BI. u. Insekt.
a. d. nordfr. Ins.,’ pp. 42, 151.)—The flesh-coloured protandrous flowers, devoid of
nectar-guides, secrete nectar at the base of the stamens. The calyx is 6-7 mm. long,
with teeth 3 mm. in length, and holds together the claws of the petals. The order of
maturation of the stamens and carpels is as usual. The five outer stamens develop
first ; their anthers occupy the entrance of the flower, and turn their pollen-covered
sides towards the centre; when the pollen is shed their filaments elongate and curve
outwards, so as to make room for the inner five stamens. The anthers of these now
dehisce, and fill the entrance of the flower. The five styles develop last of all, taking
the place of the inner stamens. Their ends are spirally twisted, so that the proboscis
of an insect, when inserted into the flower, must touch them. Automatic self-pollina-
tion may be effected by contact with the stigmas of pollen-grains which have
remained clinging to the edge of the corolla-tube.

Besides protandrous hermaphrodite flowers, female or male ones have also been
observed, these being gynodioecious and gynomonoecious, or more rarely androdioe-
cious and andromonoecious. In the female flowers, according to Schulz, the stigmas

CARYOPHYLLEAE 169

do not mature till a considerable time after the commencement of anthesis, and the
male flowers also remain perfectly fresh after their pollen is shed—a fact which has
no significance for the plant. Schulz is of opinion that this peculiarity is obviously
inherited from ancestors in which the stigmas were developed subsequently to the
shedding of the pollen.

Vistrors.—In the island of Féhr, I saw 2 butterflies, Apis, 2 species of Bombus,
and a po-dvg. hover-fly (Syrphus).

Loew—in Brandenburg (Br.) and Hesse (H.)—observed the following (Beitrage,
P- 45)—

A. Diptera. Syrphidae: 1. Volucella bombylans Z. (Br.).

B. Lepidoptera. Sphingidae: 2. Macroglossa fuciformis Z. (Br., H.). Réssler
noticed the moth Dianthoecia nana Rof/. in Wiesbaden; Kerner Noctuids—Dian-
thoecia and Mamestra—in the Tyrol ; and Schletterer the following Hymenoptera at

Pola :—(a) Afzdae: 1. Eucera interrupta Baer; 2. E.longicornis Z. (6) Jchneumo-
nidae: 3. Tryphon rutilator Gr.

Herm, Miiller observed the following in Westphalia.

A. Diptera. Syrphidac: 1. Rhingia rostrata Z., skg.; 2. Syrphus pyrastri Z.,
po-dvg.; 3. Volucella plumata Z., po-dvg. B. Hymenoptera. Apzdae: 4. Andrena
nitida Fourcr. 9, vainly seeking nectar; 5. Apis mellifica L. ¥, freq., skg. and po-cltg. ;
6. Bombus agrorum /. 9; 7. B. lapidarius Z. 9 and ¥; 8. B. rajellus AK. 9; 9. B. ter-
rester Z.Y; 10. Osmia rufa Z.9; 11. Psithyrus vestalis Fourcr. 9, skg. C. Lepi-
doptera. (a) Noctuidae: 12. Euclidia glyphica Z., very freq. (6) Rhopalocera :
13. Lycaena icarus Roft.; 14. Pieris brassicae Z.; and 15. P. rapae Z., both freq.
{c) Sphingidae: 16. Ino statices Z.; 17. Macroglossa fuciformis LZ.

MacLeod saw Apis, 3 humble-bees, 2 hover-flies, and 4 Lepidoptera in Flanders ;
and H. de Vries (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875) noticed Apis
mellifica Z. ¥ and 2 humble-bees—Bombus agrorum /. and B. subterraneus L. ¥
in the Netherlands.

In Dumfriesshire 2 humble-bees, 2 hover-flies, and 2 Muscidae have been
recorded. (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 24.)

426. C. Flos-Jovis Lam. (=Lychnis Flos-Jovis Z.). (Herm. Miiller, ‘Alpen-
blumen,’ pp. 199-200.)—This species bears markedly protandrous butterfly flowers, red
in colour. The nectar is concealed at a depth of about ro mm., and owing to the
narrowness of the mouth of the flower—which is only 1-2 mm. wide, and largely
blocked up by the anthers or styles—it is only conveniently accessible to the proboscis
of Lepidoptera. The order of maturation of the five outer and five inner stamens,
and the five stigmas is as usual. It may be, however, that automatic self-pollination is
possible, for the styles—with half-developed stigmatic papillae—project from the
mouth of the flower while pollen still remains clinging to the last anthers, owing
to the failure of insect-visits. According to Briquet (‘Etudes de biol. flor. dans les
Alpes occident.’), a proboscis of about 15 mm. is necessary to secure the nectar,
which is secreted on the inner sides of the bases of the stamens. The visitors are
Lepidoptera, which effect cross-pollination, as self-pollination is excluded by the very
marked character of the protandry (Kirchner).

Visitors.—Herm. Miiller observed butterflies as regular pollinators (Argynnis,
Colias), nect.-skg., and, as an occasional guest, a po-dvg. hover-fly (Eristalis tenax Z.).

170 ANGIOSPERMAE—DICOTYLEDONES

427. C. tomentosa A. Br. (=Agrostemma Coronaria Z.).—At Bozen, accord-
ing to Schulz (‘ Beitrage,’ II, p. 33), the large purple-red Lepidopterid flowers of this
species conceal their nectar, which is sparingly secreted, at a depth of 12-15 mm.
Automatic self-pollination must be a rare occurrence in the protandrous hermaphro-
dite flowers, for though the stigmas come into contact with the anthers towards the
end of anthesis, no pollen remains clinging to these as a rule. Besides bi-sexual
flowers, smaller female ones have been observed, which are distributed either gyno-
dioeciously or gynomonoeciously.

Visrrors.—Schulz observed many of the larger butterflies (species of Pieris and
Vanessa, Papilio machaon Z. and P. podalirius Z.).

124. Melandrium Roehl.

Mostly dioecious or trioecious moth or butterfly flowers ; more rarely protan-
drous to protogynous hermaphrodite flowers. Petals with ligule and bifid limb.
Secretion of nectar as usual.

428. M. album Garcke (=Lychnis vespertina S7d¢h., and in part Lychnis
dioica Z.). (Sprengel, ‘Entd. Geh.,’ pp. 255-60; Herm. Miller, ‘ Fertilisation,’ p. 131;
Delpino, ‘Ult. oss.” pp. 161-4; Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1,11; Schulz,
‘Beitrage,’ I, p. 13, II, pp. 33-5 ; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp. 156-7 ; Kirchner, ‘ Flora v. Stuttgart,’ p. 251 ; Knuth, ‘BI. u. Insekt. a. d. nordfr.
Ins.,’ pp. 41-151.)—The species bears moth flowers, and is almost dioecious. The
white petals are devoid of nectar-guides, and appear limp and faded in the daytime.
They are odourless and almost completely closed. In the evening they open, the
petals spread out, and a strong odour is exhaled. In shady places they are frequently
open in the daytime as well, but if exposed to bright sunshine are usually closed from
9 am.to6p.m, The nectar—as usual—is secreted by the fleshy base of the ovary,
and is thoroughly concealed at a depth of 20-25 mm. in female flowers and
15-18 mm. in male flowers. The stamens and styles vary in length according to
Schulz. Besides unisexual flowers, hermaphrodite ones have also been observed, which
are distinctly protandrous, and are associated with male flowers on the same stocks.

Magnin states (‘ Recherches sur le polymorphisme floral . . . du Lychnis Vesper-
tina,’ Lyon, 1889) that the male flowers are smaller than the female and hermaphro-
dite ones. The latter are produced by the action of a fungus (Ustilago antherarum
Fries) upon female blossoms. This fungus causes only a slight change of form of
the anthers in male flowers, but in female flowers the styles and upper part of the
ovary become vestigial, while the anthers develop because they afford the only nidus
in which the fungus can flourish. At the same time, there is an elongation of the
internode between calyx and corolla, such as is characteristic for purely male flowers.
This phenomenon of ‘castration parasitaire androgéne’ was discovered by Tulasne,
and described by Cornu and by Giard (‘Sur la castration parasitaire du Lychnis
dioica Z. par l'Ustilago antherarum F7.,’ C.-R. Acad. Sci., Paris, cvii, 1888).

VisiTors.—In the island of Amrum I very frequently noticed a moth, Plusia
gamma Z.; Herm. Miiller observed a nocturnal hawk-moth—Deilephila porcellus Z.—
in Westphalia; Réssler saw the moth Dianthoecia nana Ro#/. at Wiesbaden: all skg.

In Dumfriesshire several flies and moths, probably useless guests, have been
recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 23).

CARYOPHYLLEAE 17I

429. M. rubrum Garcke (=Lychnis diurna Své/h., and, in part, L. dioica Z.).
(Sprengel, op. cit.; Herm. Miiller, op. cit.; MacLeod, Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, pp. 155-6; Schulz, ‘ Beitrage,’ I, p. 12; Kerner, op. cit.; Loew, ‘ Bliiten-
biol. Floristik,’ p. 400; Knuth, op. cit.—The species bears butterfly flowers, and is
trioecious. The flower mechanism agrees with that of the last species, but the
nectar is concealed at a depth of only 12-15 mm. The calyx-tube is 14 cm. long in
male flowers and 13 cm. in female ones. The petals have claws of equal length, and
their limbs are nearly 1 cm. longer. They are held so closely together by the calyx
that an opening of only 4 mm. in diameter is left. This opening is surrounded by
a corona 3-4 mm. high, which contains the anthers in male flowers and the stigmas
in female ones. In male flowers the five stamens opposite the sepals mature before
the other five, but there are never more than two or three anthers in the entrance of the
flower, filling it completely, so that the proboscis of an insect, however slender, must
touch them when probing for nectar. The stigmatic papillae are all directed inwards,
leaving a passage to the nectar in the middle, so that an insect coming from a male
flower must touch them with the part of its body which is dusted with pollen if it
wishes to get at this. In the female flowers nectar is secreted on the base of
the ovary, in male flowers on the inner sides of the bases of the filaments, where the
vestigial pistil is situated. In male flowers the nectar is protected by hairs on
the lower thirds of the filaments, and perpendicular to their surfaces, while in female
flowers the somewhat overhanging ovary serves the same purpose (Knuth, ‘Bl. u.
Insekt. a. d. nordfr. Ins.,’ p. 41).

Besides unisexual flowers, hermaphrodite ones have been locally, though rarely,
observed, e. g., by Schulz at Halle a.S. They are markedly protandrous.

Visirors.—As a regular pollinator of the faintly odorous flowers, I observed—
at Kiel—only Bombus hortorum Z. Apis only paid brief visits, soon going off to
some other species. The male flowers were also visited by po-dvg. Syrphidae
(Eristalis, Melanostoma). On the Dovrefjeld Lindman also noticed humble-bees
and flies, while Herm. Miiller in the Alps saw 12 Lepidoptera and a hover-fly
(‘ Alpenblumen,’ p. 200).

Réssler observed the moths Dianthoecia filigrana Zsp. and B. nana Roét. at
Wiesbaden; Loew—at Varenna (‘Beitrage,’ p. 63)—saw a hover-fly, Leucozona
lucorum Z., endeavouring to suck nectar.

H. de Vries (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875) noticed a
humble-bee, Bombus terrester Z., in the Netherlands.

Willis (‘Flowers and Insects in Gt. Britain,’ Part I) observed the following in the
neighbourhood of the south coast of Scotland:—A. Diptera, Syrphidae: 1. Platy-
cheirus albimanus /., on & flowers only, po-dvg. B. Hymenoptera, Apzdae:
2. Bombus terrester Z., freq., skg.

In Dumfriesshire 3 humble-bees and 2 hover-flies have been recorded (Scott-
Elliot, ‘ Flora of Dumfriesshire,’ p. 24).

The flowers are sometimes perforated by humble-bees (Schulz).

430. M. apetalum Fenzl (= Wahlbergella apetala /.).—This northern species
certainly does not bear Lepidopterid flowers. Lindman could not find any nectar in
the flowers, although there are nectaries in the form of small swellings on the inner

172 ANGIOSPERMAE—DICOTYLEDONES

sides of the bases of the stamens. The small petals project from the calyx but little
or not at all. In Greenland plants self-pollination is inevitable. They develop fruits
as far as 70° or 71° N. Jat. Lindman was able to distinguish two varieties in Norway,
the larger of which possessed a greater proportion of female flowers, with enclosed
dirty red petals and shorter stamens; while the smaller variety had more male flowers,
with projecting expanded petals of yellow colour, and longer stamens. Warming
observed in Nova Zemlia a variety intermediate between these two forms, with petals
of medium length, and capable of autogamy. The same observer saw in Greenland
and Norway flowers that looked normal, but had no pollen and were therefore female.
According to Ekstam, Nova Zemlian plants agree with those native to Greenland
(Loew, ‘ Bliitenbiol. Floristik,’ p. 100).

Andersson and Hesselman (‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil.
Karlvaxtflora, p. 55) state that the arctic variety of this species (Wahlbergella
apetala Fy., var. arctica Zh. Fr.) flowers in Spitzbergen from the beginning of July
to the middle of August, regularly setting fruits in the latter month. As a rule the
plant only bears one flower, or more rarely two. The calyx is dark red—in plants
from Greenland (Abromeit, ‘Bot. Ergeb. von Drygalski’s Grénlandsexped.,’ p. 15)
with reddish-violet veins on a green ground—and the scarcely longer corolla dirty
violet. The entrance of rain is prevented by the pendulous nature of the flower, and
by its narrow opening. The pollen-grains are normal, and quickly germinate in
distilled water. Self-pollination is the rule, for the anthers are in immediate contact
with the twisted stigmas.

Ekstam (‘ Bliitenbiol. Beob. a. Spitzbergen,’ p. 24) describes the plant as dio-
ecious (?), perhaps on the strength of two somewhat different sexual forms observed
by Lindman on the Dovrefjeld, He only makes special mention, however, of her-
maphrodite and female flowers.

Visrtors.—Ekstam saw none in Spitzbergen.

431. M. involucratum Cham. et Schlecht, var. Baffine Rofrd. (= Wahlbergella
affinis /rzes.).—This plant has been studied by Warming in Greenland. The faintly
odorous flowers possess more or less projecting petals. It is doubtful whether they
belong to Class L. The flowers are at first protogynous, but automatic self-pollina-
tion is possible in the later stages of anthesis. Fruits are ripened as far as 70° or 71°
N. lat., and—in Grinell Land in Arctic America—even as far north as 84°; they
have also been found in Spitzbergen and Nova Zemlia. In Norway purely female
flowers have been observed (Loew, op. cit.).

According to Andersson and Hesselman (op. cit., p. 56), this species flowers in
Spitzbergen from the middle of June on. Ripe fruits were collected by Ekstam on
August 7, 1897. The plants usually bore only one or two flowers, rarely three.
The petals project beyond the cylindrical calyx for 3-3 its length. Ekstam (op. cit.,
p- 25) determined the diameter of the flowers to be 12-13mm. The hermaphrodite
blossoms are protogynous to homogamous, and secrete nectar on the inner sides of
the bases of the filaments.

Visitors.—Ekstam observed none in Spitzbergen,

432. M. triflorum J. Vahl—According to Warming, who studied this northern
species in Greenland, it is doubtful whether it belongs to Class L. The flowers are

CARYOPHYLLEAE 173

faintly odorous, and the petals more or less spreading. There is slight protogyny,
but self-pollination is inevitable later on, and is effective, for fruits are regularly set,
even in 76° N. lat. In Greenland, at 73° N. lat., purely female flowers have also
been observed (Loew, op. cit.).

433. M. divaricatum Nym. (=M. macrocarpum W7//k.).—Focke considers
that the greatly dilated calyx of this South European species is a means of protection
against the ovipositor of insects. The opposite is true—according to the same
investigator—for

434. M. album x Silene noctiflora, the calyx of which is narrower than that
of M. album. This hybrid is therefore not so well protected against such attacks.

125. Agrostemma L.

Protandrous to homogamous butterfly flowers. Petals red, undivided, without
ligules; the lower parts of the claws are winged and held together by the calyx,
which is contracted above. Secretion of nectar as usual.

435. A. Githago Z.(=Lychnis Githago Z., Githago segetum Desf). (Sprengel,
‘Entd. Geh., pp. 254-5; Herm. Miller, ‘ Fertilisation,’ p. 131, ‘ Weit. Beob.,’ II,
p- 234; Tullberg, Bot. Not., Lund, v, 1868, p. 10; Kemer, ‘ Nat. Hist. Pl.,’ Eng. Ed.,
1, IL; Kirchner, ‘Flora v. Stuttgart, pp. 252-3; Schulz, ‘Beitrage,’ I, p. 11;
MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 157; Knuth, ‘BI. u. Insekt. a. d.
nordfr. Ins.,’ pp. 42, 151.)—The purple petals of this species close neither at night
nor during bad weather. At the bases of their limbs there are whitish spots with
dark purple lines and blotches. The secretion of nectar is as usual. The stamens
and stigmas mature serially after the fashion of most other species of the order.
Besides plants bearing the ordinary hermaphrodite blossoms there are also varieties
with smaller flowers, distinguished by better developed carpels and less pronounced
nectar-guides. Tullberg observed transitions between protandry and homogamy in
Sweden. Schulz also describes the hermaphrodite flowers as varying from protandry
to homogamy, autogamy taking place in the lattercase. Kerner states that continued
growth of the stamens always brings about automatic self-pollination towards the end
of anthesis. According to Schulz, autogamy and xenogamy may both take place in
the same field, though they are often restricted to different localities. Besides
hermaphrodite flowers female ones also occur, these being gynodioeciously, or rarely
gynomonoeciously distributed.

Visrrors.—In the island of Amrum I saw only a butterfly—Pieris brassicae Z—
skg. normally; also a fly, which was a useless guest. Herm. Miiller observed the
following in Central Germany.—

A. Diptera. Syrphidee: 1. Rhingia rostrata Z., vainly seeking nectar.
B. Lepidoptera. (a) Rhopalocera: 2. Hesperia lineola O., skg.; 3. H. sylvanus
Esp., skg.; 4. H. thaumas H/x., skg.; 5. Pieris brassicae Z., very freq., skg.
(4) Sphingidae: Ino statices L., skg.

2. Tripe ArsingarE DC.

LiteraTurE.—Herm, Miiller, ‘ Fertilisation, pp. 131-8 ; Knuth, ‘Grundriss d.
Bliitenbiol.’ ; Schulz, ‘ Beitrage,’ I, pp. 25-6, II, pp. 52-5.

174 ANGIOSPERMAE—DICOTYLEDONES

The small flowers are usually but slightly conspicuous, even when aggregated
into inflorescences. The polysepalous calyx permits the petals to spread out. This
takes place in the sunshine, the nectar secreted in the base of the flower at the same
time becoming visible, so that the Alsineae all belong to Class EC. The nectar is
therefore accessible to insects with a very short proboscis; flies and the less special-
ized bees have chiefly been observed as visitors. Many Alsineae are dichogamous, in
which case they are almost always protandrous, but sometimes protogynous. Homo-
gamy is less frequent. The degree of dichogamy is proportionate to the conspicuous-
ness of the flowers and the number of insect visitors. Automatic self-pollination is
probably possible in all species, but is more certain in its action when the flowers are
inconspicuous, and insect visits few, either on this account or owing to unfavourable
weather.

Schulz (‘ Beitrage, I, pp. 25-6), adds the following particulars.—In many cases
the normal number (ten) of stamens is not developed. In certain species (Spergularia
salina Pres?, Holosteum umbellatum Z., Cerastium semidecandrum Z., [also C.
tetrandrum Cur/.]) the full number is never or rarely present; in others (Sagina
Linnaei Pres/, Stellaria media Cyrii/.) this is more frequently the case. In most
instances several, or all, of the inner stamens disappear, while even some of the
outer ones do so; e.g. Spergularia salina, Holosteum umbellatum, and Stellaria media
commonly possess but three outer stamens. Vestiges of the filaments are usually
retained, and commonly small anthers devoid of pollen; more rarely the stamens
have completely disappeared. In most species purely female stocks occur, often in
great numbers, but sometimes sporadically. In certain species hermaphrodite and
female flowers are found on the same plant. Male flowers have not been recorded.
The female flowers are usually at once distinguishable by their smaller size. Among
small-flowered species, female stocks are much less frequent than among large-flowered
ones. In several species where some of the stamens are almost always aborted—
it may be even some of the outer ones—female flowers are rare. There are both
individual species possessing small flowers—and therefore attracting insects but little
—which are protandrous, and large-flowered species that are homogamous. On the
whole, however, the small-flowered species are either homogamous or but slightly
protandrous,

With regard to the development of the reproductive organs, and so forth, Schulz
(Beitrige, II, pp. 52-5) speaks somewhat as follows.—

The anthers of the outer stamens usually dehisce very soon after the flower
opens, their filaments, in many species, bending inwards, and frequently meeting
above the ovary. Sooner or later, after the dehiscence of the outer anthers, those of
the inner stamens, of which the filaments are either erect or else more or less inclined
towards the petals, also begin to shed their pollen. At this time the outer anthers
in some species still contain abundant pollen; in other species they are almost or
entirely empty; and in yet other species, e.g. in Alsine verna, some or all of them
have dropped off. There are but a few species in which both sets of anthers dehisce
simultaneously. The anthers of the same whorl dehisce either altogether or within
a short time—usually only a few minutes—of one another. In the latter case no
species exhibits any definite order. The anthers at first have an introrse position,
but in most species, either before or during dehiscence, or rarely towards its end,

CARYOPHYLLEAE 175

they assume a horizontal or completely extrorse position. The styles and stigmas in
most species are not fully developed when the flower opens, or at the time the
shedding of pollen begins.

In some species, however, the stigmas are regularly mature and receptive during
the dehiscence of the outer or at any rate of the inner anthers. In other species the
stigmas are mature only when the inner stamens have discharged almost all their
pollen, and in yet other species—e.g. Alsine verna and Stellaria graminea—they
become receptive only after all the empty inner anthers, or at least those of them
which first dehisce, have fallen to the ground. There are but few species in which
the stigmas are receptive when the outer anthers are beginning to open. Automatic
self-pollination is almost inevitable in those species where the stigmas are receptive
while the anthers of the outer stamens—which often bend inwards—are dehiscing.
Self-pollination is much less likely when the stigmas first mature while pollen is
being shed from the anthers of the erect or more or less outwardly inclined inner
stamens. It is, of course, quite impossible when the stigmas become receptive only
after the anthers have lost their pollen, or when all or some of them have fallen to
the ground. The hermaphrodite flowers of but few species regularly possess the
typical number of stamens: in most species some or all of the inner stamens are
wanting in a larger or smaller proportion of the flowers, and less frequently one, two,
or even three of the outer stamens are absent. In some species the normal number
of stamens is rare, in a few it is very rare, and in a small percentage it seems never
to have been observed. In almost all cases—Moenchia erecta and Moehringia
trinervia apparently alone excepted—female flowers make their appearance, being
far more numerous in species which possess the typical number of stamens than in
those which rarely or never do so. The female flowers are usually to be found
on separate plants. They are much less frequently—in some species but very rarely
associated with hermaphrodite flowers on the same stocks. In some species,
however, such an association is almost always the rule. The female flowers are
smaller than the hermaphrodite ones, but in nearly all species both kinds often vary
greatly in size. In female flowers the stamens have either entirely disappeared
or are reduced to more or less considerable vestiges. In the latter case remains
of the anthers commonly occur, the largest of these being often but little smaller
than the normal and typical ones. They are almost always white or discoloured
yellow, but now and then contain a few normal pollen-grains capable of germination,
as well as small abnormal polyhedral or rounded ones. The styles of female flowers
are often somewhat longer, and stigmas stouter, closer, and beset with longer papillae,
than those of hermaphrodite ones. In almost all Alsineae there is a tendency for the
flowers to close more or less completely—or at least to contract—during the night,
and in cool, moist weather. In many species when the flowers close completely at
night, and in bad weather, they remain open all day when the weather is bright and
warm. In other species the flowers remain open in such weather only during the
middle of the day, while in some species, e.g. Sagina Linnaei, var. macrocarpa, it
appears that they only open when there has been warm weather for at least five
or six hours previously. Other species, again, such as Sagina Linnaei, var. micro-
carpa, and Stellaria media, var. pallida (S. Boraeana /ord.), have taken a further step
towards cleistogamy, for they frequently do not open even during long periods of

176 ANGIOSPERMAE—DICOTYLEDONES

warm weather. Stellaria media, var. pallida, is indeed in some places completely
cleistogamous. Most species secrete nectar very abundantly. This usually trickles
down from the nectaries between the bases of the petals on to the spoon-shaped
sepals, which often project horizontally. The larger flowers of some species relatively
poor in nectar are visited much less than the smaller flowers of other species which
are more richly supplied. But the small flowers of some species which secrete
a comparatively large quantity of nectar—e.g. those of Arenaria serpyllifolia and
Sagina Linnaei, var. macrocarpa—remain almost unvisited. Probably this difference
as regards visits depends upon variations in the composition of the nectar: presum-
ably that of Arenaria serpyllifolia and similar species are deficient in certain odorous
substances, so that insects have difficulty in detecting it.

The following species are known to be gynodioecious (gd.) or gynomonoe-
cious (gm.).—

Sagina nodosa Fenz/, gd. in Denmark (Warming) and Belgium (MacLeod);

S. Linnaei Pres/, gd. and gm. (Schulz) ;

Spergula arvensis Z., gm., more rarely gd. (Schulz);

S. vernalis Will/d., and S. pentandra Z., gm. and gd. (Schulz);

Spergularia media Pres/, gd., rarely gm. (Schulz);

S. salina Pres/, ditto;

S. rubra Pres/, gm. and gd., as also Alsine verna Bart/., Cherleria sedoides Z.,
Mochringia muscosa Z., Arenaria serpyllifolia Z., A. biflora Z., A. ciliata Z., Holo-
steum umbellatum Z., Stellaria nemorum Z., S. media Cyr., S. Holostea Z., S. uligi-
nosa Murr. (mostly according to Schulz);

S. graminea Z., gm. (MacLeod) ;

S. palustris ZArh., gd. (Warming, Ludwig, Miiller);

S. graminea Re/z., gd. (Tullberg, Warming, Miiller, Ludwig, Schulz);

Malachium aquaticum /7zes, gd. (Ludwig) and gm. (Schulz);

Cerastium arvense Z., gd. and gm. (Schulz);

C. triviale Lenk, gd. (Ludwig), and gm. (Schulz) ;

C.-glomeratum Zuill., gd. (Ludwig);

C. brachypetalum Desp., gm., moderately gd. (Schultz), so also for C. semi-
decandrum Z., C. pallens #. Schultz, C. obscurum Chaud., C. trigynum V7il., and
C. latifolium Z. ;

C. alpinum Z., gd. (Ludwig).

126. Sagina L.

Small whitish, protandrous, homogamous or protogynous flowers, with half-
concealed nectar secreted at the bases of the stamens.

436. S. procumbens L. (Schulz, ‘Beitrage,’ II, pp. 38-9.)—There are four
small nectaries at the bases of the filaments. The white petals, usually four in
number, are smaller than the sepals, of which an equal number are present. The
four or five stamens develop simultaneously with the four or five stigmas. Automatic
self-pollination is inevitable, for the flowers remain closed in dull weather. In
Greenland, according to Warming, the anthers, even in the open flower, come into
contact with the short, greatly spreading styles. The same investigator observed
female flowers as well as hermaphrodite ones in Denmark.

CARYOPHYLLEAE 177

VisiTors.—Schulz noticed a few small flies and bees. MacLeod observed ants,
Poduridae, and Acaridae in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 159).

In Dumfriesshire several ants have been recorded (Scott-Elliot, ‘Flora of Dum-
friesshire,’ p. 25).

437. S. apetala L. (MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 159 3
Kirchner, ‘Flora v. Stuttgart,’ p. 234.)—The flower mechanism essentially agrees
with that of the last species. The flowers are very feebly protandrous, and their
petals—which are very small, wanting altogether, or at least soon disappear—
open in sunshine, and the nectaries are situated as in the last species. During
anthesis the stamens incline inwards to such an extent that the anthers touch the
stigma, and automatic self-pollination necessarily results. In dull weather it takes
place at once, for the flowers remain closed.

Vistrors.—Acaridae have been observed in Belgium, and MacLeod has also
seen Poduridae, ants, and mites which are able to effect cross-pollination.

438. S. maritima D. Don.—The flower mechanism of this species, which
I studied in the Halligen, agrees essentially with that of the last species.

Visitors.—I did not observe any.

439. S. Linnaei Pres] (=S. saxatilis Wimm.).—In the Alps, according to
Schulz (‘Beitrige,’ II, pp. 14-15), the hermaphrodite flowers of this species are homo-
gamous or slightly protogynous. In dull weather automatic self-pollination takes
place, for the flowers remain closed. Kerner says that when the flowers have
opened the five outer stamens serve for cross- and the five inner stamens for self-
pollination. Besides hermaphrodite blossoms there are also female ones, distributed
either gynodioeciously, or gynomonoeciously. In the Riesengebirge Schulz observed
a large-flowered variety rich in nectar, and with partially degenerate stamens.
Warming saw ripe fruits in Greenland.

Visitors.—Schulz noticed flies and small beetles.

440. S. nivalis Fries—According to Lindman, this species is autogamous on
the Dovrefjeld, and Warming states that it occurs and sets fruits in Spitzbergen and
on the north coast of Siberia. For Nova Zemlia Ekstam describes the flowers as
being odourless, protogynous-homogamous (sometimes homogamous) and 5 mm. in
diameter. Self-pollination is inevitable.

Andersson and Hesselman state (‘ Bidrag till Kanned. om Spetsbergens o. Beeren
Eil. Karlvaxtflora,’ p. 65) that this species flowers in Spitzbergen from the second
half of July to the middle of August, its fruits maturing regularly and abundantly
towards the end of the latter month and the beginning of September. The petals of
Greenland specimens are either shorter than the calyx or of the same length (Abro-
meit, ‘ Bot. Ergeb. v. Drygalski’s Grénlandsexped.,’ p. 17).

441. S. caespitosa J. Vahl.—Warming states that this species is homogamous
and autogamous in Greenland, where it has been observed to set fruits. Gynodioe-
cism obtains in Norway, the female flowers possessing more or less degenerate
stamens.

442. S. nodosa Fenzl.—The hermaphrodite flowers of this species are pro-
tandrous in Norway and Denmark—according to Warming—and also in Russia

DAVIS. 11 N

178 ANGIOSPERMAE—DICOTYLEDONES

(Batalin, Bot. Ztg., Leipzig, xxviii, 1870). During unfavourable weather the flowers
remain closed, and self-pollination takes place. MacLeod observed (Bot. Centralbl.,
Cassel, xxix, 1887) female as well as hermaphrodite flowers on the dunes of the
Flemish coast. Female stocks were also noticed by Warming in Denmark.
Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) observed gynodioecism and
imperfect gynomonoecism at Ruppin. He distinguished—(1) larger hermaphrodite
flowers of ro mm. diameter, frequently with more or less reduced anthers or stamens:
and (2) smaller flowers of only 5-6 mm. diameter, and female by abortion of all the
stamens. The hermaphrodite flowers are protandrous. The first stamens to mature
are the outer ones, which have nectaries at their bases. These incline over the

still apposed stigmatic branches. The inner stamens ripen subsequently. Nectar is
abundantly secreted.

Visttors.—Hermann Miller (‘ Alpenblumen,’ p. 183) observed one of the Bom-
byliidae—Anthrax sp.—in the Alps.

443. S. subulata Torr. et Gray.—Garden plants were found by Warming to
be sometimes protogynous, sometimes slightly protandrous.

127. Spergula L.

Flowers white, usually homogamous, rarely protogynous, with half-concealed
nectar, secreted by nectaries situated as usual.

444. S. arvensis L. (Herm. Miiller, ‘Weit. Beob., II, p.225 ; Kerner, ‘ Nat.
Hist. Pl.,’ Eng. Ed. 1, II; Schulz, ‘Beitrige,’ I, pp. 15-16; Kirchner, ‘ Flora v. Stuttgart,’
p. 232; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 43, ‘Bloemenbiol. Bijdragen ’.)—
The white homogamous flowers of this species open widely in the sun, and secrete
nectar near the bases of the stamens. The filaments are meanwhile curved outwards to
such an extent that insects in quest of nectar touch the anthers with one side of their
body, and the stigma with the other, so that cross-pollination is promoted. During
unfavourable weather the flowers remain closed, and self-pollination takes place.
Further—according to Kerner—automatic self-pollination occurs towards the end of
anthesis, when the flowers are beginning to close. The same writer states that they
remain closed from 1o a.m. till 4 p.m. According to Schulz, female flowers fre-
quently occur, and over 50% of them may be gynomonoecious in a given
locality, while gynodioecism is rare. Schulz also says that the number of stamens
varies, for some of them are often more or less vestigial. Normal and abnormal
flowers are sometimes associated on the same stock, or may occur on different ones.

Visirors.—Hermann Miiller (H. M.) in Westphalia, and myself (Kn.) in
Schleswig-Holstein, have observed the following.—

A. Diptera. (a) Muscidae: 1. Lucilia sp., skg. (H. M.). (4) Syrphzdae:
2. Eristalis arbustorum Z., skg. and po-dvg. (H. M.); 3. E. tenax Z., ditto (Kn.);
4. Helophilus pendulus Z., ditto (H. M.); 5. Melanostoma ambigua /ail., ditto
(H. M.); 6. Melithreptus menthastri Z., ditto (H. M.); 7. M. strigatus Séaeg.,
ditto (H. M.); 8. Syritta pipiens L., ditto (H. M., Kn.); 9. Syrphus balteatus Deg.,
ditto (H. M., Kn.); 10. S. corollae #., ditto (H. M.); 11. S. ribesii Z., ditto (H. M.,
Kn.). B. Hymenoptera. (a) Apidae: 12. Andrena albicrus AZill. 9, po-cltg.
(H.M.); 13. A. convexiuscula A. 3, skg. (H. M.); 14. Apis mellifica Z., skg. (Kn.);

15. Halictus malachurus &. 9, skg. and po-dvg. (H.M.). (6) Sphegedae: 16. Crabro
wesmaéli v.d. L. 9, skg. (H. M.).

CARYOPHYLLEAE 179

MacLeod saw two hover-flies and a Muscid in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 158).

In Dumfriesshire an Empid, three Muscidae, and five hover-flies have been
recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 31).

445. S. pentandra Z.—According to Schulz (‘ Beitrage,’ II, p. 41), the flowers
of this species are open from noon till 5 p.m. They are homogamous, and, as
a rule, only five stamens are present. These are usually erect in the open flower, or
spread somewhat outwards, so that the anthers do not generally come into contact
with the stigma. As the flowers, however, soon close and in dull weather do not
open at all, automatic self-pollination takes place, while in the open flowers cross-
pollination is favoured by insect-visits. In addition to hermaphrodite flowers female
ones sometimes occur, these being distributed gynomonoeciously or gynodioeciously.

446. S. Morisonii Bor. (=S. vernalis Wélld., in part).—Schulz states
(‘ Beitrage, II, pp. 39-41) that the hermaphrodite flowers of this species are homoga-
mous in North Thuringia, but when the flowers are open—between noon and 5 p.m.—
the anthers do not usually touch the stigmas. Of the ten stamens a few are com-
monly vestigial, and sometimes none are developed, so that the flowers become
female. Such female flowers are smaller than the hermaphrodite ones, and are
distributed either gynomonoeciously or gynodioeciously. Cross-pollination is favoured
by insect-visits, but automatic self-pollination takes place when the flowers close.
Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) observed only homogamous and
autogamous hermaphrodite flowers at Ruppin.

128. Spergularia Presl.

Flowers white or red, varying from marked protandry to homogamy, with nectar
half concealed.

447. S. media Poir. (= S. marginata A7v//., and Arenaria marginata DC.).—
Schulz states that the red or white hermaphrodite flowers of this species are distinctly
protandrous at Halle. All ten stamens are usually developed. Besides hermaphrodite
flowers, smaller female ones have been occasionally observed—e. g. by MacLeod on
the Belgian coast—and the plants in such cases are gynodioecious, more rarely
gynomonoecious.

Visttors.—I observed small Diptera—sp. of Hilara and Hydrellia—in the
Halligen.

448. S. salina Pres] (= Arenaria rubra 8 marina Z., S. marina Grdsed., Arenaria
marina oth, and Lepigonum medium Waehlend.).—The flowers of this species
investigated by MacLeod on the Belgian coast were very similar to the female flowers
of the last species. The petals are rose-red; only 1-3 stamens are developed ;
automatic self-pollination is ensured.

Schulz examined the flowers by the Salt Lake at Eisleben, where they appear
to be considerably smaller than in Belgium. The petals are shorter than the sepals ;
frequently but three stamens are developed, which usually mature rather sooner than
the stigmas, these being at the same level. The nectar is secreted by a fleshy ring
internal to the bases of the stamens. In unfavourable weather the flowers remain
closed, so that automatic self-pollination necessarily takes place.

N 2

180 ANGIOSPERMAE—DICOTYLEDONES

All the flowers observed by Magnus at Kissingen showed a transition to
cleistogamy, the pale corolla remaining closed.

Besides hermaphrodite flowers, Schulz noticed female ones, distributed
gynomonoeciously, more rarely gynodioeciously.

VisiTors.—I saw the honey-bee skg. in the island of Sylt (‘ Weit. Beob.,’ p. 232).

Verhoeff observed the following in Norderney.—

A. Diptera. (a) Empzdae: 1. Hilara quadrivittata MZg.; (6) Muscidae: 2.
Anthomyia sp., skg.; 3. Aricia incana Wredem., skg.; 4. Lucilia caesar Z. 6, skg.;
(c) Syrphidae: 5. Syritta pipiens L., skg.

449. S. rubra Pres] (=Arenaria rubra and campestris Z., Alsine rubra
Wahlend., and Lepigonum rubrum Wahlend.).—According to Schulz (‘Beitrage,’I, p.17),
the flower mechanism of this species resembles that of S. satina, and there is also
an agreement as regards the small number of stamens. The flowers vary from
homogamy to slight protandry, and, as in the last species, self-pollination often takes
place when they remain closed. There are female flowers as well as hermaphrodite
ones. These are usually smaller, and may be distributed either gynomonoeciously
or gynodioeciously. Schulz regards this species as oecologically intermediate
between S. media and S. salina.

Visitors.—MacLeod observed an Empid in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 157).

129. Cherleria L.

Flowers protandrous to homogamous, and very inconspicuous, with half-
concealed nectar secreted between the roots of the stamens.

Fic. 51. Cherleria sedosdes, L. (after Herm. Miller). 4. Flower seen from above (x 16). 2B. Styles
and stigmas in the first (male) condition. C. The same in the second (female) condition. 5, sepals;
p, petals; 7, filaments; #, nectaries; ov, ovary.
450. C. sedoides L. (Herm. Miiller, ‘Alpenblumen,’ pp. 184-5; Schulz,
‘ Beitrage,’ II, pp. 44-5.)—In this species the petals scarcely attain one-third the
length of the sepals. The expanded calyx is star-shaped and 4-5 mm. in diameter.
Hermann Miiller says that in the Alps the hermaphrodite flowers are distinctly
protandrous, so that automatic self-pollination is usually prevented. Schulz examined

CARYOPHYLLEAE ; 181

plants not far from the same district, and found them homogamous or slightly
protandrous. Besides hermaphrodite flowers, he also observed female ones,
distributed gynodioeciously or gynomonoeciously.

Visirors.—Herm. Miiller observed numerous small nect-skg. flies.

130. Alsine L.

Flowers white, usually small, protandrous, homogamous or protogynous, with
half-concealed nectar.

451. A. verna Bartl. (=A. Gerardi Wahlend.). (Herm. Miiller, ‘Alpenblumen,’
pp. 183-4; Schulz, ‘Beitrage,’ I, p. 18.)—In the Alps, according to Hermann
Miller, the flowers of this species attain a diameter of 6 mm. Schulz states that at
a height of 2,000-3,000 m. it is as much as 7-9 mm., while in the Riesengebirge
it averages 1omm. ‘There is marked protandry (see Fig. 52), and Schulz asserts
that self-pollination does not take place, though Kerner says that it occurs towards
the end of anthesis, while according to MacLeod it is brought about by the closing
of the flowers in the evening. Smaller female flowers have been observed, in addition
to hermaphrodite ones, especially in the Hochgebirge. They are distributed
gynomonoeciously and gynodioeciously.

Fic. 52. Alsine verna, Bartl. (atter Herm. Miller). 4. Flower before the beginning of the first
stage. B. Flower in the first half of the first (male) stage. C. Flower in the second half of the same
stage, seen fromabove. . The same in section seen from the side. &. Flower in the second (female)
stage. F. Upper part of the stigma. G. Lower part of the same.

The hermaphrodite flowers seen by Warming in Greenland—belonging to the
variety (2) Airfa Lange—were almost homogamous, and capable of self-pollination.
In Spitzbergen the same botanist observed cases of protogyny, representing perhaps
a transition towards the female condition.

182 ANGIOSPERMAE—DICOTYLEDONES

Vistrors.—Those observed by Miiller and A. Schulz in the Alps were chiefly
Diptera (Muscidae, Syrphidae, Empidae, Bombyliidae), some beetles, ants, and
Lepidoptera (Pyralidae). MacLeod also saw Diptera in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, pp. 379-81).

452. A.recurva Wahlenb. (Herm. Miiller, ‘Alpenblumen,’ p. 182.)—Hermann
Miiller says that some of the flowers remain open, while others half close.

VisiTors.—Diptera (Syrphidae, Muscidae) and some Lepidoptera have been
observed in the Alps.

453. A. stricta Wahlenb.—According to Warming; the flowers of this species
are homogamous in Greenland and Norway. Automatic self-pollination takes place
and is effective, for ripe fruits are regularly produced even as far north as
70°=71° N. lat.

454. A. rubella Wg.—Ekstam states that the odourless, protogynous to
homogamous flowers of this species are 5-8 mm. in diameter in Nova Zemlia. The
stigmas being usually at a higher level than the anthers, self-pollination is rendered
difficult.

This species flowers on Spitzbergen throughout July, and fruits are ripened very
early (Andersson and Hesselman, ‘ Bidrag till Kanned. om Spetsbergens o. Beeren
Eil. Karlvaxtflora,’ p. 65).

455. A. groenlandica Fenzl.—In Greenland—according to Warming—the
flowers of this species are either slightly protandrous or else quite homogamous,
so that self-pollination is almost inevitable.

456. A. biflora Wg.—The hermaphrodite flowers of this species are protandrous
on the Dovrefjeld, but automatic self-pollination takes place when the flowers close.
The plants examined in Greenland by Warming bore slightly protandrous, homo-
gamous, or even slightly protogynous hermaphrodite flowers, in which automatic
self-pollination necessarily occurred. This resulted in the abundant production of
fruits. In Spitzbergen the flowers are smaller, and gynodioecism has been observed
in Norway.

According to Andersson and Hesselman (op. cit., p. 64), this species flowers
in Spitzbergen from the beginning of July till September. By the beginning of
August fruits are abundantly and regularly ripened. Ekstam (‘Bliitenbiol. Beob.
a. Spitzbergen,’ pp. 28-9) gives 5-7 mm. as the diameter of the flowers. There
appear to be variations in regard to the maturation of the sexual organs. In
Greenland specimens the petals are about half as long again as the calyx (Abromeit,
‘Bot. Ergeb. v. Drygalski’s Grénlandsexped.,’ p. 17).

457. A. Rossii Fenzl_—This species has only once been found in flower in
Spitzbergen (Andersson and Hesselman, op. cit., p. 64).

131. Honckenya Ehrh.
Flowers white, protandrous, with half-concealed nectar secreted as usual.

458. H. peploides Ehrh. (=Ammodenia peploides Rup., Arenaria peploides Z.,
and Halianthus peploides Frzes)—The plants examined by me in the North Frisian
Islands (‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 44) spread out their flowers in the

CARYOPHYLLEAE 183

sunshine, forming an almost plate-like disk, measuring about 8 mm. in diameter.
The five white spathulate petals are about the same length as the bright green sepals.
Of the ten stamens, the five opposite the sepals mature first. They are then nearly
erect, and project from the flower about 1mm. The five other stamens—which have
meanwhile rested on the petals—next mature, elongating till they are as long as the
outer ones. It is only at this stage that the stigmas unfold. At the base of
the ovary there is a large yellow nectary between each pair of stamens, which
secretes so abundantly that the intervals between the pairs of stamens are completely
filled with nectar.

In spite of its abundant nectar the flower is seldom visited by insects. I saw
no visitors in the island of Rém, though I watched for a long time during fine
weather. From dehisced anthers, however, pollen
frequently falls into the flower, and this may
be carried by the wind to the stigmas of the
same plant, or of neighbouring ones. Small
grains of sand are constantly to be found in
the flowers, into which they have been drifted
by the wind, and as these may be blown along
from blossom to blossom they possibly serve as
occasional agents of pollination. During dull
weather the flowers close, so that automatic Fic. 53. Honckenya peploides, Ehrh.
self-pollination is then possible. tar ied eis ae

In Greenland, Iceland, north Norway, Spitz- fromabove. 4, sepal; , petal; a, stamen of

r 5 p the outer whorl with dehisced anther: a’,
bergen, and Nova Zemlia—according to Warming stamen of the inner whorl, with anther still
—hermaphrodite flowers are very rare, and closed; s, immature stigma; 9, sand-grain.
dioecism, polyoecism, or monoecism almost
always obtains. Warming observed fruits to be set in Greenland.

This species flowers in Spitzbergen from the beginning of July to August
(Andersson and Hesselman, ‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil.
Karlvaxtflora,’ p. 64). Ekstam observed ripe fruit there on August 24, 1897.
Both female and hermaphrodite flowers were found. The latter are protandrous
and odourless, with a diameter of 9-11 mm., and with functional nectaries at the
bases of the filaments.

Visrrors.—On the dunes of Helgoland, I observed (June 8, 1895) 2 Muscidae—
Lucilia caesar Z., and Fucellia fucorum Fad/—both skg. Ekstam (‘ Bliitenbiol.
Beob. a. Spitzbergen,’ p. 28) observed no visitors in Spitzbergen.

Verhoeff observed the following 2 Muscidae in Norderney :—1. Lucilia caesar
L., 9 and é, skg. 2. Scatophaga stercoraria L.

In Dumfriesshire 2 Muscidae have been recorded, freq. (Scott-Elliot, ‘Flora
of Dumfriesshire,’ p. 26).
132. Moehringia L.
Flowers white, homogamous, protandrous or protogynous, with half-concealed
nectar.
459. M. trinervia Clairv. (=Arenaria trinervia Z.). (Herm. Miiller,
‘Fertilisation’ 136-7, ‘ Weit. Beob.,’ II, p. 225; Warnstorf, Verh. bot. Ver., Berlin,

184 ANGIOSPERMAE—DICOTYLEDONES

xxxviii, 1896; Schulz, ‘Beitrage, II, pp. 46—7; Kirchner, ‘Flora v. Stuttgart,’
p- 235-)—Hermann Miiller describes the flowers as protogynous, with stigmas
persisting for a long time, while A. Schulz found them to be almost always
homogamous, though sometimes slightly protandrous or slightly protogynous. At
the base of each of the five outer stamens there is a fleshy swelling, which secretes
a relatively large drop of nectar. The anthers of the five outer stamens dehisce
first. Insect visitors regularly effect cross-pollination, for they touch the stigmas
before the anthers. Failing insect-visits, automatic self-pollination takes place,
the stamens gradually curving inwards till they touch the stigmas. Sometimes
the anthers of the five outer stamens are vestigial.

Warnstorf states that the flowers are homogamous and autogamous at Ruppin.
The long stigmatic branches—usually three, more rarely two—curve outwards, often
becoming hook-shaped, and arch over the stamens in such a way that self-pollination
necessarily results. Self-pollination also takes place by apposition of the stamens
to the stigmas. The styles are rarely vestigial. The parts of the flower are
frequently in fours.

Visitors.—The inconspicuous odourless flowers, with petals shorter than the
sepals, are but rarely visited by insects.
Hermann Miller observed the following.—

A. Coleoptera. (a) Nit:-
dulidae: 1. Meligethes, nect-
Ikg. (6) Phalacridae : 2.Olibrus
affinis Sturm., nect-lkg.

B. Diptera, (a) Brbi0-
midae: 3. Dilophus vulgaris
Afg., nect-lkg. (4) Muscidae :
4. Sapromyza rosida Feal.,
nect-lkg.

MacLeod observed an
Empid in Flanders (Bot.
Jaarb. Dodonaea, Ghent, vi,
1894, p. 162).

In Dumfriesshire an Em-
pid, several other Dipters, and

Fic. 54. Moehringia muscosa, L. (after Herm. Miller). 4. a wasp, have been recorded
Flower in the first (male) stage. 2. Stamens of the same flower, Ss Elli ‘FI
seen from without. C. Flower in the second (female) stage, after ( cott- lot, ora of Dum-

removal of calyx and corolla. friesshire,’ p- 26).

460. M. muscosa L. (Herm. Miiller, ‘Alpenblumen,’ pp. 187-8; Schulz,
‘ Beitrage,’ II, pp. 45-6.)—-This species bears protandrous hover-fly flowers. There
are eight stamens, of which the four outer ones first raise themselves and open their
anthers, then the four inner stamens do the same, and when all eight have withered
the styles and stigmas develop. Self-pollination is, therefore, only occasionally
possible (see Fig. 34). Schulz says that besides hermaphrodite flowers there are
also female ones on gynodioecious, more rarely on gynomonoecious plants.

Visitors.—Herm. Miiller frequently observed small hover-flies—especially Sphe-
gina clunipes /a//—which hover in front of a flower, alight to lick nectar or devour
pollen, and then go off to another. A. Schulz noticed small bees as well as flies.

CARYOPHYLLEAE 185

133. Krascheninikovia Turcz.

Kuhn states that cleistogamous flowers occur in this genus.

134. Arenaria L.

Flowers small and white, homogamous or protandrous, with half-concealed
nectar secreted in the usual place.

461. A. serpyllifolia L. (Herm. Miiller, ‘ Weit. Beob.,’ II, p. 226; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 161 ; Kirchner, ‘Flora v. Stuttgart,’ p. 754;
Schulz, ‘Beitrige,’ I, p. 19, II, p. 47.)—In this species the stamens and stigmas
mature simultaneously, and in sunshiny weather drops of nectar can be seen in the
base of the flower. Schulz also observed female flowers on gynomonoccious, rarely
on gynodioecious plants. In hermaphrodite flowers the stamens are frequently
reduced in number; and automatic self-pollination by contact of stigmas and
anthers is unavoidable.

Fic. 55. Arenaria biffora, L. (after Herm. Miller). 4. Flower in the first (male) stage. 2B. Flower
a styles) in the second (female) stage, seen from above. C. Stamens and nectaries of the same

Vistrors.— Herm. Miiller observed 2 small short-tongued bees—Sphecodes
ephippius Z. 9, and Halictus lucidulus Schenck 9, skg.—both capable of effecting
cross-pollination.

MacLeod saw a hover-fly in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent,
iii, 1891, p. 377).

In Dumfriesshire, a hover-fly and Thrips have been recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 26).

462, A. biflora L. (Herm. Miiller, ‘Alpenblumen,’ pp. 185-7; Schulz,
‘Beitrage,’ I], pp. 47-8.)}—The white flowers are protandrous, but the periods of
maturation, not only of the outer and inner stamens, but also of the inner stamens

186 ANGIOSPERMAE—DICOTYLEDONES

and the stigmas, are not sharply demarcated. Failing insect-visits, automatic self-
pollination therefore takes place (see Fig. 55). Besides hermaphrodite flowers, Schulz
also observed female ones on gynodioecious, rarely on gynomonoecious plants.

Visirors——In the Alps, Herm. Miiller only saw Diptera, i.e. 11 Muscidae,
3 Syrphidae, and an Empid.

463. A. ciliata L—Kirchner examined the hermaphrodite flowers at Zermatt
(‘ Beitrage,’ p. 14).

They are protandrous, and when fully expanded their diameter is 12 mm.
After the anthers have dehisced, the styles develop and the stigmatic papillae
mature. Besides hermaphrodite flowers, Kirchner also observed smaller female ones
with a diameter of 7-10 mm. The stamens of these are in various stages of
reduction: though all ten are sometimes present, all or most of them are quite short,
while some of them may be entirely absent. Ludwig (Bot. Centralbl., Cassel, iii,
1880, p. 1021) first noticed the occurrence of female flowers on gynodioecious plants
in Switzerland; Warming found the species to be gynomonoecious in Norway
(‘Om Caryophyll. blomst.,’ 1890, pp. 32-3).

Warming saw ripe fruits at Disko on the variety (0) Aumzfusa Rink, the flowers
being slightly protandrous at first and afterwards homogamous.

According to Andersson and Hesselman (‘ Bidrag till Kanned. om Spetsbergens
o. Beeren Eil. Karlvaxtflora, p. 64), the variety /rig¢da Koch flowers in Spitzbergen
from the beginning of July till August. Ekstam observed ripe fruits on August 7,
1897 (‘ Bliitenbiol. Beob. a. Spitzbergen,’ p. 28). The pollen-grains are of two sizes ;
in one case there was 23% of small grains. The flowers are very fragrant;
Ekstam gives their diameter as 11-14 mm. In flowers not fully open the stigmas
already possess glistening papillae, while the anthers are still undehisced. When
the flowers are quite expanded autogamy takes place, for the dehiscing anthers
come into contact with the stigmas by bending of the filaments. Nectar is secreted
on the outer sides of the stamens superposed to the sepals.

VisiTors.—On one occasion Ekstam observed a small fly in Spitzbergen.

464. A. graminifolia Schrad.—

Visirors.—Loew observed a Syrphid—Eristalis nemorum Z.—and two bees—
Apis mellifica Z. ¥, skg., and Prosopis communis /Vy/. 6, po-dvg—in the Berlin
Botanic Garden.

465. A. arctica Stev—This species is native to Siberia and Alaska.
According to Alice Eastwood (Bot. Gaz., Chicago (IIl.), xxxiii, 1902, pp. 137-8),
the petals are over 7 mm. long and 5 mm. wide, with short yellow claws. The
filaments thicken abruptly at their bases, so that there is perhaps a secretion of nectar.

135. Holosteum L.

Flowers homogamous, slightly protandrous or protogynous flowers, with half-
concealed nectar.

466. H. umbellatum. (Herm. Miiller, ‘ Weit. Beob., II, pp. 226-7; Schulz,
‘Beitrage,’ II, pp. 48-9 ; Warnstorf, Verh. natw. Ver., Berlin, xxxviii, 1896; Schr.
natw. Ver., Wernigerode, xi, 1896.)—According to Herm. Miller, the small white
flowers are protandrous, and cross-pollinated by insect-visits, though automatic self-

CARYOPHYLLEAE 187

pollination can take place early. There are usually only three stamens, more rarely
five, four, or two. A green fleshy nectary is situated at the base of each stamen. At
the beginning of anthesis, when the anthers are dehiscing, the styles with their incom-
pletely developed stigmas are erect. The stamens, however, are so curved inwards
as to bring the anthers immediately above the stigmas, so that when these are mature
automatic self-pollination must result from the fall of pollen. It also often takes
place in the closed flower. Dehisced stamens gradually incline outwards, and the
stigmas spread out more and more. Warnstorf says that the outer stamens ripen earlier
than the inner ones; their filaments are longer, and their bases are provided with
yellow nectaries. The anthers are yellow, and after dehiscence undergo a rotation
through go°, so as to become horizontal. The pollen-grains are golden yellow,
regularly dodecahedral, and closely beset with short spines; average diameter 37 p.

Besides hermaphrodite flowers, female ones have also been observed, distributed
gynodioeciously, or more rarely gynomonoeciously. The ordinary flowers are
locally—e. g. in Denmark—
homogamous, or even pro-
togynous.

Visitors. — Herm.
Miiller observed a Muscid—
Anthomyia sp. 9 — and 3
bees, i.e. Andrena gwynana
A. 9, skg.; A. parvula KX. 9,
skg.; and Halictus sp. 9,
skg.

136. Stellaria L.

Flowers white, protan-
drous, homogamous, or pro-
togynous, with half-con-
cealed nectar secreted at
the bases of the stamens.

467. S. graminea L.
(Herm. Miiller, ‘ Fertilisa-
tion, pp. 133-4, ‘ Weit.
Beob.,’ II, p. 227; Knuth,
‘Bl. u. Insekt. a. d. nordfr.
Ins.,’ p. 45, ‘ Bloemenbiol.
Bijdragen’; Kirchner, ‘Flora
v. Stuttgart,’ p.238; Ludwig,
Bot. Centralbl., Cassel, iii, 5, he ‘Gor halt of the fret male) stage, the five outer stamens have
1880 : Schulz, Beitrage,’ 1, curved inwards, and their anthers are covered with pollen. 2. Flower

in the second (female) stage: all the anthers are empty and shrivelled ;

p- 20, IL, pp. 50-1.) —In the styles have spread out and curved backwards, thus turning upwards

this species the five nec- their papillose sides. @! and a?, outer and inner stamens; 7, nectaries
P st, stigmas.

taries are in the form of
green, fleshy ridges at the bases of the five outer stamens. The flowers are

188 ANGIOSPERMAE—DICOTYLEDONES

protandrous, and when they open the five outer stamens first bend inwards and
dehisce, the five inner ones being as yet unripe and curved outwards, and the
stigmas immature. Before the anthers of the five outer stamens have withered,
those of the five inner ones dehisce, but still remain directed outwards. As they
wither, the styles elongate, and the stigmas unfold above the contracting and
shrivelling stamens. Every insect that is not too small must, therefore, when trying
to get at the nectar, dust itself with pollen in the younger flowers, whether it alights
in the middle or at the edge; while in older flowers it is obliged to touch the
stigmatic papillae, thus effecting cross-pollination. Failing insect-visits, the stigmatic
branches bend back still more, coming into contact with the anthers to which pollen
still adheres, and in this way automatic self-pollination takes piace as a last resource
(see Fig. 56).

Besides these protandrous hermaphrodite flowers, the mechanism of which
has been thus described by Hermann Miiller, there are small female ones, with
quite vestigial stamens of white colour, and also medium-sized transitional forms,
with 2-3 stamens developed (e.g. in Belgium, according to MacLeod). The plant
is gynodioecious in Sweden (Tullberg), and by the Altenfjord (Warming). Schulz
observed gynomonoecism, as well as gynodioecism, in Central Germany, where in
some localities the stocks may be mainly or exclusively female. He noticed herma-
phrodite flowers of three different sizes, i.e. 8-ro mm., 10-14 mm., and 16-18 mm.
in diameter. These seem to be local. The larger flowers are not visited by more
numerous insects than the smaller ones.

VisiTors.—Schulz observed flies, small bees, and beetles. In the Alps, where
Lepidoptera abound, Herm. Miiller saw a butterfly. He (H.M.) and myself (Kn.)
have observed the following in North and Central Germany :—A. Coleoptera.
Nitidulidae: 1. Meligethes, skg. and po-dvg. (H.M.). B. Diptera. (a) Empidae:
2. Empis livida Z., skg. (H.M.). (6) Syrphidae: 3. Eristalis tenax Z., skg. (Kn.);

4. Helophilus pendulus L., skg. (Kn.); 5. Syritta pipiens Z., skg. and po-dvg.
a. M.); 6. Volucella bombylans L., skg. (H.M.).

Verhoeff saw the following in Norderney.—A. Coleoptera. (a) NMitédulidae:
1. Brachypterus gravidus //.., skg. B. Diptera. (a) Empzdae: 2. Hilara quadri-
vittata Afg., skg. (0) Musctdae: 3. Anthomyia sp. (c) Syrphidae: 4. Melanostoma
mellina Z., skg.; 5. Syritta pipiens Z., skg. C. Hymenoptera. (a) Formicidae:
6. Lasius niger Z., skg.

MacLeod noticed Apis, 6 Syrphids, an Empid, 2 ichneumon-flies, a Siricid,
and a Lepidopterid in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 164).

In Dumfriesshire, an Empid, 2 hover-flies, and 4 Dolichopidae have been
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 29).

468. S. cerastioides L. (Herm. Miiller, ‘Alpenblumen,’ pp. 188-9.)—The
number of carpels is variable in the homogamous flowers of this species.

Visirors.—Herm. Miiller saw Diptera (an Empid, 2 Muscidae, and 4 Syrphidae)
in the Alps.

469. S.Holostea L. (Herm. Miiller, ‘Fertilisation, p. 135; ‘ Weit. Beob.,’ II,
p- 228; Kirchner, ‘Flora v. Stuttgart, p. 238; MacLeod, Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 378; vi, 1894, pp. 162-3; Schulz, ‘ Beitrage,’ I, p. 22; Knuth,
‘ Bloemenbiol. Bijdragen.’)

CARYOPHYLLEAE 189

Hermann Miller states that the flower mechanism of this species agrees in the
main with that of S. graminea, but the flowers are larger, and insect visitors are
consequently more numerous. The nectaries are yellow. During anthesis, the five
outer and the five inner stamens successively occupy the middle of the flower, those
which are not dehiscing being curved outwards. Besides hermaphrodite flowers,
female ones have been observed, and also transitional forms; e.g. MacLeod—in
Belgium—noticed flowers with partly reduced stamens. Schulz states that the
female blossoms are distributed gynodioeciously, or more rarely gynomonoeciously.
In addition to the protandrous hermaphrodite flowers, Schulz describes homogamous
ones in which automatic self-pollination is inevitable.

Visitors.—Herm. Miiller (H.M.), Borgstette (B.), Buddeberg (Budd.), and
myself (Kn.) have observed the following in Central and North Germany.—

A. Coleoptera. (a) Nitidulidae: 1. Meligethes, freq., skg. (H. M., Kn,).
(4) Oedemeridae: 2. Oedemera virescens ZL. (B.). B. Diptera. (a) Bombylitdae:
3. Bombylius canescens 4M2., skg. (Budd.); 4. B. major Z., skg.(Kn.). (6) Empidae :
5. Empis ciliata /. 9, skg. and po-dvg. (Budd.); 6. E. opaca &, skg. (H.M.); 7.
E. tessellata 7, skg. (H.M.). (c) Muscidae: 8. Anthomyia sp., skg. (H.M.); 9.
Hydrotaea dentipes #., skg. (H.M.); 10. Scatophaga merdaria Z., skg. (Kn.);
11. Siphona geniculata Deg., skg. (H.M.). (d) Syrpfidae: 12. Eristalis arbustorum
L., skg. and po-dvg. (H.M.); 13. E. nemorum Z., ditto (Kn.); 14. Platycheirus
peltatus J/g., ditto (H.M.); 15. Rhingia rostrata Z., ditto (Kn.); 16. Syrphus
balteatus Deg., ditto (Kn.); 17. S. ribesii Z., freq., ditto (H. M.). C. Hymenoptera.
(a) Apidae: 18. Andrena cineraria Z. 9, skg. (H.M., Budd.); 19. A. parvula &. 9,
skg. (H.M.); 20. A. gwynana &. 9. skg. (H.M.); 21. A. labiata Schenck et Ny. 9,
skg. (Budd.); 22. Apis mellifica Z. ¥, skg. (H. M., Kn.); 23. Halictus cylindricus /.
9, skg. (H. M., Budd., Kn.); 24. H. albipes X. 9, skg. and po-cltg. (Budd.); 25.
H. flavipes F. 9, skg. (Budd.); 26. H. nitidiusculus A’. 9, skg. (Budd.); 27. H. rubi-
cundus Chr. 9, skg. (Budd.); 28. Nomada flavoguttata A. 9, skg. (H.M., Budd.) ;
29. N. ruficornis Z. 9, skg. (H. M., Budd.). (0) Zenthredinidae: 30. Cephus pallipes
Kl, skg. (H.M.). D. Lepidoptera. Ropalocera: 31. Pieris napi L., skg. (H. M.) ;
32. P. rapae Z., skg. (H.M.). E. Thysanoptera. 33. Thrips, freq. (H. M.).

Alfken noticed the following at Bremen.—A. Hymenoptera. Apidae: 1.
Andrena chrysopyga Schenck 9, skg.; 2. Nomada bifida Zhs., skg.; 3. N. flavoguttata
K.qand 4, skg. B. Diptera. Syrphidae: 4. Platycheirus albimanus /.

Schenck saw the bee Andrena cingulata /. in Nassau, and Rossler the moth
Asychna modestella Dup. at Wiesbaden.

MacLeod observed 6 hover-flies, 12 other Diptera, 2 beetles, and 3 Lepidoptera
in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 162-3); a bee, 3 Syrphidae,
and 4 Muscidae in the Pyrenees (op. cit., iii, 1891, p. 378).

In Dumfriesshire, several flies, Meligethes, and another beetle have been recorded
(Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 30).

Burkill (‘ Fertlsn. of Spring Flowers’) noticed a Muscid—Sepsis nigripes M/g.—
on the coast of Yorkshire.

470. S. scapigera Willd.—Breitenbach found this species to be gynodimor-
phous in the Marburg Botanic Garden and Géttingen (Kosmos, Stuttgart, xiv, 1884).

471. S. media Cyrill. (Herm. Miiller, ‘Fertilisation,’ pp. 135-6 ; ‘ Weit. Beob.,’
II, p. 228; Schulz, ‘Beitrage,’ I, p. 20; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’
Pp. 45, 151; ‘Bloemenbiol. Bijdragen’; Kirchner, ‘Flora v. Stuttgart,’ p. 237;

190 ANGIOSPERMAE—DICOTYLEDONES

Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.}—Hermann Miiller states that some
of the ten stamens are almost always vestigial, these being usually the inner five, and
often one or two of the outer whorl in addition. ‘The nectaries are situated on the
bases of the five outer stamens, and secrete during sunny weather. The anthers
dehisce successively, either simultaneously with the maturation of the stigmas, or
somewhat earlier, or rather later. According to Kerner, autogamy takes place when
the flowers begin to close. Warnstorf says that at Ruppin the plant bears either
hermaphrodite flowers or pseudo-hermaphrodite fruiting flowers. The former possess
two to five, or more rarely six to eight stamens, which have violet anthers projecting
beyond the stigmas or at the same level, and effect autogamy by moving inwards.
In the pseudo-hermaphrodite fruiting flowers all or some of the stamens have
degenerated. The variety decandra is markedly protandrous. Automatic self-pollina-
tion takes place when the flowers close, and results in autocarpy. According to
Anna Bateson, plants produced by crossing are somewhat larger and heavier than
those resulting from autocarpy in the proportion of 100: 91; Celakovsky states that
the variety doraeana is cleistogamous.

In Greenland, according to Warming, the hermaphrodite flowers agree in their
mechanism with those of Europe, and there are also female ones, distributed
gynodioeciously or gynomonoeciously. Cleistogamy obtains in the same country.

Plants observed by Vanhéffen in Greenland (Abromeit, ‘Bot. Ergeb. v. Dry-
galski’s Grénlandsexped.,’ p. 20), and by Andersson and Hesselman in Spitzbergen
(‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 86), were
sterile, like several other introduced weeds.

This species flowers so early that it has few rivals, and occurs in such numbers
that visitors are moderately numerous in spite of the smallness of its flowers. In
the hermaphrodite blossoms these may effect either cross- or self-pollination.

Visitors.—Hermann Miller (H. M.) in Westphalia and myself (Kn.) in Schles-
wig-Holstein have observed the following.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp., skg. (H.M.); 2. Chlorops cir-
cumdata J/g., industriously skg. (H.M.); 3. Lucilia cornicina &., po-dvg. (Kn.);
4. Musca corvina /., ditto (H.M.); 5. M. domestica Z., ditto (H.M., Kn.); 6.
Pollenia rudis /., po-dvg. (Kn.); 7. Scatophaga sp., po-dvg. (Kn.); 8. Sepsis sp.,
skg. (H.M.). (6) Syrphidae: 9. Ascia podagrica F., skg. (H.M.); 10. Cheilosia
sp., skg. (H. M.); 11. Eristalis arbustorum, Z., po-dvg. (Kn.); 12. Syritta pipiens
L., skg. and po-dvg. (H. M., Kn.); 13. Syrphus corollae /., po-dvg. (Kn.); 14. S.
ribesii Z., po-dvg. (Kn.). B. Hymenoptera. (a) Apidae: 15. Andrena albicans
Mill. 9, skg. (H. M); 16. A. albicrus X. 4, skg. (H. M.); 17. A. chrysosceles X. 4,
skg. (H.M.); 18. A. dorsata K. 9, skg. (H.M.); 19. A. fasciata Wesm. 3, skg.
(H.M.); 20. A. florea 9 and 4, skg. (H.M.); 21. A. fulvicrus &. 4, skg. (H. M.);
22. A. gwynana &. 9, skg. and po-cltg. (H.M.); 23. A. smithella X. 4, skg. (H. M.);
24. Apis mellifica Z. ¥, skg. (Kn.); 25. Halictus cylindricus #., skg. (Kn.); 26.
H. flavipes #. 9, skg. (H.M., Kn.); 27. H. leucopus X. 9, skg. (H.M.); 28. H.
sexstrigatus Schenck 9, skg. (H.M.); 29. Osmia rufa Z. 8, skg. (H. M.); 30.
Sphecodes gibbus Z. 9, skg. (H.M.). (0) Cynipidae: 31. Eucoela sp. (H.M.).
C. Thysanoptera. 32. Thrips, po-dvg. (H.M.).

MacLeod noticed Apis, 14 other short-tongued Hymenoptera, 7 Syrphidae,
10 other Diptera, and 2 beetles in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp. 166-7).

CARYOPHYLLEAE IQI

Verhoeff observed the following in Norway.—

A. Diptera. (a) Bibzonidae: 1. Scatopse notata ZL. (6) Muscidae: 2. Antho-
myia sp., freq., skg.; 3. Lucilia caesar Z. 9 and 4, skg.; 4. Nemopoda stercoraria
Rob.-Desv., skg. (c) Syrphidae: 5. Platycheirus clypeatus Mg. 4, skg.; 6. Syritta
pipiens Z., skg. B. Hymenoptera. Jormicidae: 7. Lasius niger L., skg.

Alfken saw the following at Bremen.—

A. Hymenoptera. Apidae: 1. Andrena parvula A. 9, skg.; 2. Halictus
nitidiusculus A. 9, skg.; 3. Podalirius acervorum Z. &. B. Diptera. Syrphidae:
4. Chrysogaster macquarti Loew.

Bees have also been recorded as follows.—By Schmiedeknecht in Thuringia—
1. Andrena congruens Schmuedekn.; 2. A. dorsata K. 3; 3. A. eximia Sm.; 4. A.
floricola Ev.

By Friese in Baden (B.) and Mecklenburg (M.)—1. Andrena gwynana &., 2nd
generation (M.); 2. A. parvula X., freq. (M.), very freq. (B.).

By von Dalla Torre and Schletterer in the Tyrol—r. Andrena eximia Sm. $;
2. Halictus albipes /%r. 9; 3. Nomada alternata X. 6.

Burkill observed the following—all skg.—on the Yorkshire coast (‘Fertlsn. of
Spring Flowers’).—

A. Diptera. (a) Bibionidae: 1. Scatopse notata Z. (6) Muscidae: 2. Lucilia
comicina /.; 3. Phorbia muscaria M/g.; 4. Scatophaga stercoraria Z.; 5. Sepsis
nigripes Mg. (c) Phoridae: 6. Phora, sp. C. Hymenoptera. Jchneumonidae:
4. Pezomachus, sp. C. Thysanoptera. 8. Thrips, sp.

In Dumfriesshire—Meligethes and numerous Diptera have been recorded (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 28).

472. S. nemorum L.—The hermaphrodite flowers of this species are every-
where more or less distinctly protandrous. Besides these there are smaller female
ones, distributed gynodioeciously in Thuringia (Ludwig) and the Riesengebirge
(Schulz) or more rarely gynomonoeciously.

VistTors.—Lindman observed small and medium-sized flies on the Dovrefjeld.

In Dumfriesshire—Meligethes and numerous flies have been recorded (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 28).

473. S. Frieseana Lange.—Lindman states that the variety alpestris is pro-
togynous on the Dovrefjeld, becoming homogamous later on by persistence of the
stigmas. In Atnedal, however, there are also many protandrous plants. Automatic
self-pollination easily takes place towards the end of anthesis, by the contact of
stigmas and anthers.

474. S. palustris Ehrh. (=S. glauca With.).—In this species gynodioeciously
distributed female flowers have been observed in Denmark (Warming) and Germany
(Ludwig, Hermann Miiller), as well as strongly protandrous hermaphrodite ones.

Visrrors.—MacLeod observed a Muscid in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 164).

475. S. bulbosa Wulf.—Kerner made observations on this species in Carniola.
The flowers are moderately large, and yet receive very few visits from insects (some
flies), and are quite infertile. Propagation is effected by means of numerous little
tubers on the thread-like underground stem.

192 ANGIOSPERMAE—DICOTYLEDONES

476. S. crassifolia Ehrh—In Denmark, Warming observed gynodioeciously
distributed female flowers in addition to the markedly protandrous hermaphrodite
ones. Warnstorf describes the flowers at Ruppin as being also protandrous.

477. S. longipes Goldie —Warming states that the hermaphrodite flowers of
this species are protandrous or homogamous in Greenland, while self-pollination
appears to be prevented by the distance between anthers and stigmas. The female
flowers are gynodioeciously distributed, as also in Spitzbergen, where, however, they
are remarkably small.

Ekstam describes the flowers in Nova Zemlia as odourless, protogynous-
homogamous, and 8-12 mm. in diameter. In Spitzbergen and Greenland, according
to Warming, the hermaphrodite flowers are protandrous-homogamous, or homogamous,
and there are also purely female ones.

Andersson and Hesselman state (‘ Bidrag till Kanned. om Spetsbergens o. Beeren
Eil. Karlvaxtflora,’ p. 56) that this species flowers in Spitzbergen from the middle
of July till the end of August, and ripens its fruits. Ekstam (‘Bliitenbiol. Beob. a.
Spitzbergen,’ pp. 25-6) gives the diameter of the flowers of the variety Aum@lis Fenzl
as 8-12 mm., and he noticed some instances of apetaly. The sexual organs mature
simultaneously after the flowers have opened, and autogamy is rendered possible
by bending of the stamens towards the stigmas. The floral receptacle obviously
secretes nectar between the bases of the stamens.

In Greenland this extremely variable plant is so abundant that it serves as food
for reindeer (Abromeit, ‘ Bot. Ergeb. v. Drygalski’s Gronlandsexped.,’ pp. 21-2).

Visitors.—On four different days Ekstam noticed several small Diptera in
Spitzbergen, and he observed a medium-sized fly in Nova Zemlia.

478. S. humifusa Rottb.—In Greenland, according to Warming, the
hermaphrodite flowers of this species are usually protandrous, rarely protogynous,
but in later stages they are always homogamous, The female flowers are distributed
in Spitzbergen like those of the last species. Warming saw no fruits, and there
is probably a large amount of vegetative propagation by means of buds.

For Nova Zemlia Ekstam gives the diameter of the protogynous-homogamous
flowers as 10-15 mm. Self-pollination is possible. There is sometimes a tolerably
strong odour of honey, associated with the secretion of a considerable quantity
of nectar.

According to Andersson and Hesselman (op. cit., p. 56), this species flowers
in Spitzbergen from the middle of July to the end of August, and ripens fruits.
Ekstam (op. cit., p. 26) gives the diameter of the flowers as 6-8 mm.; in Greenland
it is 8-10 mm., according to Abromeit (op. cit., p. 20). Ekstam observed slight
protandry.

Visttors.—Ekstam observed a small fly in Nova Zemlia.

479. S. borealis Bigel.—Lindman states that the flowers of this species are
homogamous on the Dovrefjeld. Towards the end of anthesis the anthers of the
longer stamens come into contact with the stigmas, and self-pollinate them. In
Greenland, according to Warming, the flowers are apetalous, homogamous, and
automatically self-pollinated. Gynodioecism has also been observed.

CARYOPHYLLEAE 193

480. S. uliginosa Murr. (MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp. 164-5; Knuth, ‘ Bliitenbiol. Beob. a. d. nordfr. Ins.,’ p. 145; Schulz, ‘ Beitrige,’
I, pp. 22-3.)—MacLeod describes plants of this species in Belgium as possessing
inconspicuous protandrous flowers, with petals shorter than the sepals. The outer
stamens first mature, and remain in the middle of the flower during the whole
period of anthesis, while the inner ones curve outwards. After dehiscence, the styles
develop, and the spreading stigmas come into contact with the outer anthers.
Towards the end of anthesis the outer stamens, in contact with the stigmas, also
incline inwards, so that automatic self-pollination is always assured.

The flowers are more or less protandrous in Denmark (Warming) and at
Halle, but they are homogamous during autumn in the latter locality (Schulz).
Schulz found them to be commonly homogamous and automatically self-pollinated
in the Riesengebirge. Besides hermaphrodite flowers, this observer also noticed
female ones, distributed gynodioeciously, or more rarely gynomonoeciously.

VisiTors.— MacLeod observed an Empid in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 165).

In Dumfriesshire, several flies have been recorded (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 29).

137. Moenchia Ehrh.

Small white protogynous flowers, with half-concealed nectar.

481. M. erecta Gaertn. Mey. et Scherb.—According to Schulz (‘ Beitrage,’ II,
p- 51), some of the stamens in this species are usually not developed. Even
in the bud the four (more rarely three or five) stigmas are receptive. When the
flowers are open the stigmas but rarely come into contact with the anthers: automatic
self-pollination is therefore not very likely to occur. In dull weather, however,
effective autogamy takes place in pseudo-cleistogamous closed flowers. During
bright sunshine nectar is secreted in tolerable abundance.

VisiTors.—Schulz only saw a few flies.

138. Malachium Fries.
Flowers white, protandrous, with half-concealed nectar.

482. M. aquaticum L. (=Cerastium aquaticum Z.). (Herm. Miller, ‘ Fertili-
sation,’ p. 133, ‘ Weit. Beob.,’ II, p. 230; Knuth, ‘ Bloemenbiol. Bijdragen’; Kirchner,
‘Flora v. Stuttgart,’ p. 239; Schulz, ‘ Beitrage, I, p. 23; Ludwig, D. bot. Monatschr.,
Arnstadt, vi, 1888, p. 5, Bot. Centralbl., Cassel, viii, 1881, p. 79; Kerner, ‘ Nat. Hist.
Pl.” Eng. Ed. 1, Il; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)— Hermann
Miiller states that the mechanism of the protandrous flowers resembles that of Stellaria
Holostea. The petals are about 14 times as long as the sepals. Should insect-
visits fail, the ends of the spreading (3-5) stigmatic branches regularly come into
contact with the pale anthers, to which pollen still remains clinging. According
to Schulz, self-pollination is very rare; while Kerner has observed that it occurs
towards the end of anthesis, at the time when the flowers begin to close.
Gynodioecism has been observed (Ludwig), rarely gynomonoecism (Schulz); but
not in Denmark, according to Warming. The petals of the female flowers are

DAVIS. It (a)

194 ANGIOSPERMAE—DICOTYLEDONES

only as long as the sepals, and their stamens possess yellow degenerate anthers.
Warnstorf usually found only hermaphrodite flowers at Ruppin, 1-4 stamens being
vestigial in rare instances. The five outer stamens bend over the stigmatic branches,
and discharge their pollen when these are still closely apposed. Meanwhile the
inner stamens—with anthers as yet immature—lie remote from the centre between
the petals, and only dehisce when the stigmatic branches begin to spread out.
After dehiscence the inner stamens curve back again, so that all the anthers are in
a peripheral circle when the stigmatic branches have fully diverged. Self-pollination
is thus rendered very difficult. The pollen clings to the anthers for a time. The
pollen-grains are white, dodecahedral and smooth, with an average diameter of 37-43 p.

Visirors.— Hermann Miiller (H. M.) in Westphalia, Buddeberg (Budd.) in
Nassau, and myself (Kn.) in Schleswig-Holstein, have observed the following.—

A. Coleoptera. Nitidulidae: 1. Meligethes, freq., nect-lkg. (H. M.). B.
Diptera. (a) Muscidae: 2. Anthomyia, skg. (H.M.). (4) Syrphidae: 3. Ascia
podagrica /., freq., skg. (H.M.); 4. Eristalis arbustorum Z., skg. and po-dvg. (H.M.,
Kn.); 5. Helophilus lineatus /, freq., skg. (H. M.); 6. Rhingia rostrata Z., skg. and
po-dvg. (Kn.); 7. Syritta pipiens Z., ditto (H.M., Kn.); 8. Syrphus sp., ditto (Kn.).
C. Hymenoptera. Apzdae: 9. Colletes daviesanus K. 8, skg. (Budd.); 10. Halictus
quadricinctus /. 9, skg.(H.M.); 11. H. sexnotatus A. 3, skg. (H.M.); 12. Prosopis
communis /Vy/. 9, skg. (H.M.); 13. P. hyalinata Sm. 9, skg. (H.M.). D. Thysano-
ptera. 14. Thrips, very freq. (H. M.).

MacLeod saw a short-tongued bee in Flanders (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 378), as well as 3 hover-flies and 3 other Diptera (op. cit., vi, 1894, p. 170).

139. Cerastium L.

Flowers mostly white and protandrous, with half-concealed nectar secreted
as in the preceding genera.

483. C. arvense L. (Herm. Miiller, ‘ Fertilisation, pp. 131-2, ‘ Weit. Beob.,’
II, p. 229; Schulz, ‘Beitrage,’ I, p. 24; Kirchner, ‘Flora v. Stuttgart,’ p. 240;
Knuth, ‘Bloemenbiol. Bijdragen’;
Loew, ‘ Bliitenbiol. Floristik,’ pp.
389, 397; Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896.)—The flowers
of this species are protandrous, and,
according to Hermann Miiller, agree
with S¥ellaria Holostea as regards
the position of the nectaries, and the

Fic. 57. Cerastium arvense, L. (after Herm. Miller. Order of maturation of stamens and

(t) Flower in the first part of the first (male) stage: the : oe
anthers of the outer stamens are covered with pollen, those of stigmas. There 18) therefore, sas

the inner ones are still closed, and the stigmatic branches agreement as to the probability of
are still curved inwards. (2) Flower in the almost exclusively ‘ i . oe
female stage; the anthers of the outer stamens have partly cross-pollination by insect-visits, and
fallen off, partly shrivelled ; those of the inner whorl are still ibilj sf ‘ a
sparingly covered with pollen ; the stigmas are mature. the possibility of 5 self pollination
when such visits fail. The plants

examined by Warming in Greenland, as far north as 67° N. lat., had somewhat
smaller protandrous flowers.

Besides the hermaphrodite flowers, smaller female ones with degenerate whitish

stamens have been observed. Schulz states that their distribution is usually

CARYOPHYLLEAE 195

gynodioecious, more rarely gynomonoecious. At Ruppin, according to Warnstorf,
there are hermaphrodite flowers of two sizes, respectively 15 mm. and ro mm. in
diameter. The small-flowered variety is further characterized by the possession
of glandular hairs on the stem, especially on the upper part of this.

VisiTors.—Herm. Miiller (H. M.) in Westphalia, and myself (Kn.) in Schleswig-
Holstein, have observed the following.—

A. Coleoptera. (a) Carabidae: 1. Amara sp. (H.M.). (4) Cerambycidae:
2. Leptura livida /., vainly searching for nectar (H. M.); 3. Malachius bipustulatus
fF. (H.M.). (¢) Witidulidae: 4. Meligethes, nect-lkg. (H.M., Kn.). (d) Staphy-
“inidae: 5. Omalium florale Payk. (H.M.). (¢) Zelephoridae: 6. Dasytes sp., po-dvg.
(H.M.). B. Diptera. (a) Conopidae: 7. Dalmannia punctata F., skg. (H.M.). (4)
Empidae: 8. Empis livida Z., skg.(H.M.); 9. E. opaca &., very freq., skg. (H. M.);
to. E, rustica Fad/., ditto(H.M.). (c) Lepidae: 11. Leptis strigosa Mg., skg.(H. M.).
(d) Musczdae: 12. Anthomyia aestiva 7g.,skg.(H.M.) ; 13. Onesia sepulcralis M/g.,skg.
(H.M.); 14. Pyrellia aenea Ze//., po-dvg. (H. M.); 15. Scatophaga merdaria F., skg.
(H.M.); 16. S. stercoraria Z., skg.(Kn.). (e) SyrpAzdae: 14. Eristalis arbustorum
Z., skg. (H.M., Kn.). 18. E. nemorum Z., skg. (H. M., Kn.); 19. E. sepulcralis Z.,
skg. (H.M.); 20. Helophilus sp., skg. (Kn.); 21. Melanostoma mellina Z., freq., skg.
(H.M., Kn.); 22. Melithreptus scriptus Z., skg. (H.M.); 23. M. strigatus Svaeg.,
po-dvg. (H.M.); 24. Platycheirus manicatus J/Zg., freq., skg. (H.M.); 25. Syritta
pipiens Z., skg. (H. M., Kn.): 26. Syrphus pyrastri Z., skg. (Kn.); 27. S. sp., skg.
(H.M.). C. Hymenoptera. (a) Aprdae: 28. Andrena albicans Mill. 9, skg.
(H.M.); 29. A. argentata Sm. 9, skg. (H.M.); 30. A. cineraria Z. 9, skg. (H. M.);
31. Halictus leucozonius Schr. 9, skg. (H.M.); 32. H. sexnotatus K. 9, in large
numbers, skg.(Kn.); 33. H. sp. 9, skg.(H.M.). (0) Ichneumonidae: 34. Ichneumon
sp., skg. (H.M.). D. Lepidoptera. (a) Noctuzdae: 35. Euclidia glyphica Z., skg.
(H.M.). (8) Rhopalocera: 36. Pieris napi Z., skg. (Kn.); 37. Polyommatus dorilis
ffn., skg. (H.M.); 38. P. phlaeas Z., skg.(H. M.). E. Thysanoptera. 39. Thrips,
freq. (H. M.).

Herm. Miller saw 20 Diptera, 2 bees, and 3 Lepidoptera in the Alps (‘Alpen-
blumen,’ p. 171).

Loew noticed the following on the variety s/rictum Haencke in Switzerland
(‘Beitrage,’ p. 57).—

A. Diptera. Syrphidae: 1. Melithreptus dispar Zw. B. Hymenoptera.

Apidae: 2. Halictus cylindricus F. 9, skg. C. Lepidoptera. Pyralidae: 3. Un-
determined sp.

MacLeod saw a bee, an ant, 3 Syrphidae, an Empid, and 7 Muscidae in the
Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 378), and 4 Diptera, a Lepi-
dopterid, and a beetle in Flanders (op. cit., vi, 1894, p. 167).

In Dumfriesshire, several Diptera have been recorded (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 27).

484. C. triviale Link (=C. vulgatum Z. and C. caespitosum Gz/zd.). (Axel,
‘Om Anordning. for Fanerog. Vaxt. Befrukt.,’ pp. 16-17; Herm. Miiller, ‘ Fertilisa-
tion, p. 132; Kirchner, ‘Flora v. Stuttgart,’ p. 240; Schulz, ‘ Beitrage,’ I, p. 24.)
Hermann Miller says that the flowers have the same mechanism for pollination
as those of C. arvense, but they are smaller, in accordance with which protandry
is less pronounced, and the visits of insects less numerous. Failing insect-visits,
automatic self-pollination takes place and is effective (Axell). The stamens of the
hermaphrodite flowers are often more or less vestigial. Protogyny has been observed

02

196 ANGIOSPERMAE—DICOTYLEDONES

by Warming in Denmark. Schulz states that—in the Riesengebirge—protandrous
and protogynous flowers sometimes occur on the same plant. Besides hermaphrodite
flowers, female ones have now and then been observed, distributed gynodioeciously
(Ludwig), or more frequently gynomonoeciously (Schulz). Kerner describes the
variety /ongzrostre Wichura as protandrous, and in this also autogamy takes place
when the flowers close.

Visirors.—Herm. Miiller observed several flies—Syritta pipiens Z., Empis
livida Z., and Melithreptus scriptus LZ. 6.

Verhoeff noticed the following in Norderney.—A. Coleoptera. Carabidae:
1. Amara familiaris Duff. B. Diptera. (a) Empzidae: 2. Hilara quadrivittata Mg.
é and 9, freq., skg. (6) Afuscrdae: 3. Anthomyia sp.; 4. Aricia incana Wredem., skg.
and po-dvg.; 5. Lucilia caesar Z., skg. (c) Syrphidae: 6. Eristalis arbustorum Z. ;
7. Platycheirus manicatus J/g. one 6.

MacLeod saw 2 bees and a fly in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi,
1894, p. 168).

Burkill (‘Fertlsn. of Spring Flowers’) observed the following on the coast of
Yorkshire.— .

A. Diptera. (a) Muscidae: 1. Helomyza sp. (6) Phortdae: 2. Phora sp.,
skg. B. Thysanoptera. 3. Thrips sp., skg.

485. C. semidecandrum Z. (Herm. Miller, ‘ Fertilisation,’ pp. 132-3, ‘ Weit.
Beob.,’ II, pp. 229-30; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.” pp. 46, 151,
‘Bloemenbiol. Bijdragen’; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 168 ;
Kirchner, ‘Flora v. Stuttgart,’ pp. 241-2.)—Hermann Miiller says that the flower
mechanism of this species agrees with that of the last, but the flowers are even less
conspicuous, so that insect-visits are still fewer, and the protandry less distinct.
The flowers may even be quite homogamous (Schulz). In the absence of visitors
automatic self-pollination regularly takes place. The inner stamens are devoid
of nectaries, and almost always vestigial, there being, with rare exceptions, remains
of filaments only. In dull weather the flowers remain closed. Besides hermaphrodite
flowers, female ones of equal size have been observed, distributed gynomonoeciously,
more rarely gynodioeciously (Schulz, ‘ Beitrage,’ I, pp. 23-4).

Visitors:— Herm. Miller (H. M.) and myself (Kn.) have observed the follow-
ing.—A. Diptera. (a) Musczdae: 1. Pollenia rudis #., skg. (H.M.); 2. P. vespillo
F., skg. (H. M.). (4) SyrpAddae: 3. Rhingia rostrata Z., skg. (H.M.). B. Hymeno-
ptera. Apzdae: 4. Apis mellifica LZ. ¥, skg. (H.M., Kn.); 5. Sphecodes ephippius
L. 9, busily skg. (H. M.)

MacLeod saw 2 short-tongued bees, a Lepidopterid, and a beetle in Flanders
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 168).

486. C. obscurum Chaub. and 487. C. pallens F. Schultz are closely related
to the last species, with which they agree, according to Schulz, as regards the flower
mechanism and distribution of the sexes.

488. C. tetrandrum Curt.—As observed by me on the dunes of Helgoland,
this species is remarkable for the variations in number of its flower-leaves. The
sepals and petals are usually in fours and the stamens in fives, while there are generally
four carpels, though these may be reduced to three.

The flowers, when expanded in the sun, have a diameter of 3 to 4 mm., and
are as high as broad. The petals are greenish and thickened at their bases, where

CARYOPHYLLEAE 197

a small amount of nectar seems to be secreted, for insect visitors busy themselves
persistently in this region. During my frequent visits to Helgoland in 1895 and
1897, however, I was unable to find any nectar, though I examined the bases of the
petals under a tolerably high magnification. The sepals, which are beset with
glandular hairs, are almost as long as the petals, and help a little in the attraction
of insects.

The flowers are homogamous. At first the anthers project about 1 mm. beyond
the stigmas, but near the end of anthesis incline towards and dust them with pollen.
This self-pollination is clearly effective, for though insect-visits are exceedingly rare
all the flowers set fruits. I always found both stamens and carpels developed, and
examined numerous specimens without finding a single case where either kind of
sexual organ was entirely aborted. Pentamery predominated in the more vigorous
plants, tetramery in the weaker ones.

An insect visitor thrusts its head into the base of the flower, and in doing
so touches the anthers—which in bright weather rest on the reflexed petals—and
also the simultaneously developed stigmas, which are at the same level. Hence
a visit to a second blossom must effect cross-pollination. In dull weather the flowers
close, so that the anthers come into direct contact with the stigmas, and automatically
self-pollinate them.

In Helgoland, on June 5, 1895, I saw a hover-fly (Syritta pipiens Z.) visiting
the flowers. This observation interested me very greatly, for on the morning of
the same day I noticed on the high ground of that island the same species (also
Eristalis sp.) sucking nectar in the flowers of Cochlearia danica. This confirms
to a certain degree the ideas to which W. J. Behrens gave expression in 1878
(Flora, Marburg, New Series, xxxvi, 1878, pp. 225-32), with regard to the oecological
connexion between Cerastium tetrandrum and Cochlearia danica. The flower
mechanisms of the two species agree to such an extent that Behrens thought it
possible that the former might have arisen under the influence of an insular environ-
ment from Cerastium semidecandrum, mimicking Cochlearia danica, a form better
adapted to attract insects. ‘The agreement as to insect visitors appears to give to
this supposition a higher degree of probability, although Syritta pipiens also visits
other flowers of similar structure in Helgoland (e.g. Brassica nigra and Capsella
Bursa-pastoris). This, however, appears natural, as the size of this insect is adapted
to these other flowers quite as well as to Cerastium tetrandrum and Cochlearia danica.

VisiTors.— Vide supra.

489. C. glomeratum Thuill_—Henslow states that the flowers of this species
are autogamous ; they sometimes remain closed (Warming). According to Ludwig
(Bot. Centralbl., Cassel, iii, 1880, p. 1021), there are female flowers, distributed gyno-
dioeciously, as well as hermaphrodite ones. Now and then the petals are vestigial
(Kirchner). According to Warnstorf (Schr. natw. Ver., Wernigerode, xi, 1896), the
homogamous and autogamous flowers open only a little, or remain closed, and the
introrse anthers lie upon the branches of the style. The pollen-grains are white,
and roundish-dodecahedral, with six well-marked germinating processes around the
equatorial zone, and a diameter of about 37 p.

VistTors.—Schletterer observed Halictus calceatus Scop. at Pola.

198 ANGIOSPERMAE—DICOTYLEDONES

490. C. brachypetalum Desp.—According to Schulz (‘ Beitrage,’ I, pp. 51-2),
the number of stamens varies in this species, as in its allies. The stigmas are
receptive, even before anthesis. When the flowers open, contact between stigmas
and anthers very rarely, if ever, takes place. Automatic self-pollination regularly
occurs, however, when they close. Besides hermaphrodite flowers, female ones have
also been observed, distributed gynomonoeciously, more rarely gynodioeciously.

VistTors.—Schulz noticed two flies. Schletterer observed the two small bees
Andrena parvula A. and Halictus morio /. at Pola.

491. C. tomentosum L.—Warming observed protandry, with transitions to
homogamy in cultivated plants. Automatic self-pollination takes place towards the
end of anthesis.

492. C. viscosum L.—Batalin observed that in plants of this species developed
from seeds the flowers of the first summer often remained closed, while open flowers
were produced the following year.

493. C. trigynum Vill. (=Stellaria cerastoides Z.). (Ricca, ‘ Oss. sulla fecondaz.
incroc. d. veget. alp. e subalp.’; Herm. Miiller, ‘ Alpenblumen,’ pp. 188-9 ; Schulz,

\ \\ Y SS
My
B. x
Fic. 58. Cerasttum trigynum, Vill. (after Herm. Miiller). A. Section of a flower in the middle
of anthesis (x 7). B. Half-closed flower in the act of automatic self-pollination.

‘Beitrage,’ II, pp. 49-50.)—Hermann Miiller and Ricca describe the odorous flowers
as being homogamous, but Schulz says that they are sometimes slightly protogynous
or protandrous. In sunny weather the anthers are so far from the stigmas that
insects alighting in the middle of the flowers effect cross-pollination. In cold,
dull weather the flowers hardly open at all, but in somewhat warmer cloudy weather
they do so rather more. Under such circumstances automatic self-pollination always
takes place. The styles vary in number from three to five. The stamens are
now and then vestigial. Schulz states that the female flowers are distributed
gynomonoeciously, or more rarely gynodioeciously.

On the Dovrefjeld also, according to Lindman, the flowers are homogamous
and autogamous, and here self-pollination takes place at the beginning of anthesis.

CARYOPHYLLEAE 199

Vistrors.— Herm. Miiller observed only Diptera—- chiefly Syrphidae and
Muscidae, together with some Empidae.

494. C. latifolium L. (Herm. Miller, ‘Alpenblumen,’ pp. 189-90.)—Hermann
Miiller describes the flowers of this species as protandrous, though A. Schulz says
that they are also homogamous. Even in the former case automatic self-pollination
is possible (see Fig. 59). Dovrefjeld plants are autogamous and slightly protandrous ;
Warming once observed slight protogyny. Kerner considers that the sticky calyx
serves to protect the flowers against creeping animals. Besides hermaphrodite
flowers, Schulz has observed female ones, distributed gynodioeciously, or more
rarely gynomonoeciously.

Visitors.—Herm. Miiller chiefly observed Diptera (8 species) in the Alps,
where he also saw various bees (Halictoides), beetles (1), and Lepidoptera (4).

Fic. 59. Cerastium latifolium, L. (after Herm. Miller). 4. Flower in the first (male) stage.
B. Flower in the second (bi-sexual) stage (x 7). C. Stamens and carpels of <4 (x 7).

495. C. alpinum L. (=C. lanatum Zem.)—Besides the protandrous, ulti-
mately homogamous hermaphrodite flowers described by Kerner (‘ Nat. Hist. Pl.’
Eng. Ed. 1, I, p. 355), F. Ludwig observed female ones in the Alps, distributed
gynodioeciously. On the Dovrefjeld the flowers are at first protandrous, and self-
pollination is not effected until the stigmas bend back so as to come into contact
with the anthers (Lindman). According to Warming, the flowers are also pro-
tandrous in Greenland and Spitzbergen, but to such a slight extent that homogamy
and automatic self-pollination very soon obtain, sometimes even in the half-
opened bud. As Warming in Greenland found the stigmas of the gynodioeciously
or gynomonoeciously distributed female flowers covered with pollen, they must
have been visited by insects. Ekstam gives the diameter of the protandrous-
homogamous, or homogamous flowers as ro-20 mm. for Nova Zemlia. In the

200 ANGIOSPERMAE—DICOTYLEDONES

larger flowers automatic self-pollination readily takes place; in the smaller it is
sometimes possible. Andersson and Hesselman state that this species flowers in
Spitzbergen from the middle of July till the end of August (‘Bidrag till Kanned.
om Spetsbergens o. Beeren Eil. Karlvaxtflora’). Fruits were observed to be set,
though in an irregular fashion. According to Ekstam (‘ Bliitenbiol. Beob. a. Spitz-
bergen,’ p. 26), the flowers are homogamous and 12-15 mm. in diameter. The

anthers bend towards the stigmas during dehiscence, so that self-pollination easily
takes place.

Visirors.—H. Miiller (‘ Alpenblumen,’ p. 190) observed Diptera (3 Muscidae
and a Syrphid) and a Lepidopterid in the Alps. On the Dovrefjeld large and
small Diptera and a Lepidopterid were also noticed by Lindman. Ekstam saw
Diptera in Nova Zemlia, and on five occasions in Spitzbergen (large and small flies).
In the latter island Holmgren observed Hymenoptera—Hemiteles septentrionalis
Holmgr., and Orthocentrus pedestris Holmgr.—as well as Diptera—Aricia (Spilo-
gaster) dorsata Ze//., A. (S.) denudata Holmgr., A. (S.) megastoma Bohem., and
Sciara atrata Holmgr., very freq.

496. C. Edmonstonii (Wats.) Murb. et Ostenf—Andersson and Hesselman
state (‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ pp. 59-61)
that this species begins to flower in Spitzbergen somewhat earlier (the second half
of June) than the last species, but ripens its fruits at the same time. The variety
caespitosum Malmgr. also flowers early. The diameter of the flower varies from
9-5 to 11mm. There are both hermaphrodite and female blossoms, the former
being protogynous, homogamous, or protandrous. In all fully-developed flowers
autogamy is possible by contact of the coiled reflexed stigmas with the dehisced
anthers. The pollen-grains are very sensitive to damp, and at once burst when
placed in distilled water. Transitional forms between hermaphrodite and female
flowers are variable in number, and the latter set fruits irregularly. In one case

only twelve per cent. of the last year’s blossoms—of which remains were present—
had done so.

497. C. uniflorum Murith. (=C. subacaule Hegetschw., and C. glaciale Gaud.).—
Kirchner (‘ Beitrage,’ pp. 15-16) describes as follows the mechanism of flowers of
this species—from Gorner Grat near Zermatt—which agrees in the main with that
of C. latifolium.— The flowers are protandrous, but the possibility of automatic
self-pollination remains. When expanded their diameter is about 15 mm.; the
white petals are streaked with dark veins converging to the base of the flower;
and nectar is secreted at the bases of the stamens. After anthesis the stamens
diverge but little, and their anthers dehisce successively, first those of the outer,
and then those of the inner ones. After the pollen is shed all the stamens curve
outwards. The five styles are closely apposed in the just-opened flower, and
are scarcely 2 mm. long; they elongate, however, till at the time when the five
outer anthers have dehisced their length is about 5 mm. At this stage they diverge
somewhat, their stigmas being receptive, and remaining so till all the stamens have
shed their pollen. During anthesis, therefore, each flower passes through a male
stage, then becomes hermaphrodite, and finally enters on a female condition.

PORTULACEAE 201
XVI. ORDER PORTULACEAE JUSS.

140. Portulaca Tourn.

Small, yellow, homogamous flowers, not infrequently cleistogamous or pseudo-
cleistogamous.

498. P. oleracea L. (Kirchner, ‘ Flora v. Stuttgart? p. 254; Kerner, ‘ Nat. Hist.
P].’ Eng. Ed. 1,11; Battandier, Justs bot. Jahresber., Leipzig, xi (1883), 1885, p. 472 3
Halstead, op. cit., xvi (1888), 1890, p. 562.)—Kerner says that the golden-yellow
flowers are devoid of honey and odour, and only open for about five hours on
sunny mornings. Between the bases of the stamens and petals there is a fleshy
ridge beset with glassy papillae. These do not secrete, but, according to Kerner,
insects readily feed upon them. Stamens and stigmas mature simultaneously.
Kirchner states that the latter lie between the anthers in such a way that automatic
self-pollination is inevitable. Crossing, however, would appear to take place
occasionally, for flies and ants have been seen as visitors. According to Kerner,
automatic self-pollination only occurs when the flower closes, and also in the pseudo-
cleistogamous flowers that remain shut during bad weather. Halstead describes
the stamens as sensitive, and Hansgirg states that they curve in the direction of
the stimulus. Cleistogamous flowers have been observed by Battandier.

499. P. grandiflora Lindl.— De Bonis says that this species produces
cleistogamous flowers.

141. Montia Mich.
Small white flowers, often pseudocleistogamous.

500. M. minor C. C. Gmel. (Axell, ‘Om Anordning. for Fanerog. Vaxt. Befrukt.,’
p- 13; Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, I, p. 391.)—Axell describes the open
flowers as homogamous. Both he and Kerner state that numerous flowers are
pseudo-cleistogamous during bad weather, remaining closed and fertilizing themselves.

Visttors.—In Dumfriesshire, 2 Muscidae have been observed (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 31).

501. M. rivularis C. C. Gmel.—According to Vanhéffen this species flowers
and fruits regularly in Greenland (see Abromeit, ‘Bot. Ergeb. v. Drygalski’s Grén-
landsexped.,’ pp. 22-3).

142. Claytonia Gronov.

502. C. alsinoides Sims.—The flowers of this species in the Cambridge
Botanic Garden are protandrous, according to Willis (J. Linn. Soc. Bot., London,
xxx, 1895). They secrete nectar at the bases of the filaments. The stamens are
at first erect, but afterwards bend back towards the petals, giving free access to
the stigma, so that small insects covered with pollen are able to effect crossing.
Self-pollination is not excluded, but its effect is doubtful.

503. C. sibirica Pall_—Agrees with the last species (Willis, op. cit.).

202 ANGIOSPERMAE—DICOTYLEDONES

504. C. perfoliata Donn.—According to Kerner (‘ Nat. Hist. Pl., Eng. Ed. 1,
II, p. 365), autogamy takes place towards the end of anthesis, the pollen-covered
anthers being pressed against the stigma by the closing of the perianth.

143. Calandrinia H. B. et K.

When the flowers fade, the petals become pulpy, the surface being covered with
a thin layer of fluid which oozes out of the tissues. This is sought out and licked
by flies, which effect cross-pollination (Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, W,
p- 171).

505. C. compressa Schrad.—In this species the anthers of the ephemeral
blossoms are at first remote from the stigma. At a later stage they are applied
to the stigma by the closing of the flower (Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1,
Il, pp. 346-7).

XVII. ORDER ZJAMARISCINEAE DESV.
144. Myricaria Desv.

Small red, slightly protogynous flowers, with concealed nectar, secreted by
the inner sides of the filaments. All are entomophilous, according to Niedenzu
(Engler and Prantl, ‘D. nat. Pflanzenfam.,’ III, 6, p. 290).

Fic. 60. Myricaria germanica, Desv. (after Herm. Miller). A. Section through an open flower;
seen from the side. B. Section through a flower in the act of automatic self-pollination ; seen from the
side. C. Sexual organs of a bad, with stigma already receptive. D. Sexual organs of a flower
completely closed during rain. £. Some of the stamens seen from within, with the nectary (7).

506. M. germanica Desv. (Herm. Miiller, ‘ Alpenblumen,’ pp. 164-3.)—In
this species the stigmas are receptive, even before the flowers open, and the anthers
dehisce successively soon after this takes place. The stamens and stigma both
continue functional, so that in bad weather automatic self-pollination must take
place in the half-closed or completely closed flower. During favourable weather
insect visitors may effect crossing. (See Fig. 60.)

Visitors.— Herm. Miiller observed a fly and a Lepidopterid.

ELATINEAE—HYPERICINEAE 203

XVIII. ORDER ELATINEAE CAMB.

There is only one plant included in this order, of which the flowers have been
studied as regards pollination.

145. Elatine L.

507. Elatine hexandra DC.—Vaucher says that automatic self-pollination
takes place in the small reddish-white flowers of this species, the anthers dehiscing
introrsely, and shedding pollen directly upon the three stigmas.

XIX. ORDER APPERICINEAE DC.

This order is represented by the genus

146. Hypericum L.

The flowers are rendered conspicuous not only by the usually large, bright
yellow corolla, but also by the similarly coloured branched stamens (described as
‘bundles’), and the three styles. They are homogamous pollen flowers. The
glands on the calyx in many species keep away creeping insects. The flower
mechanisms of the various species agree as regards size, number of stamens and
possibility of automatic self-pollination, with those of—

508. H. perforatum L. (Herm. Miller, ‘ Fertilisation,’ pp. 139-40; Kirchner,
‘Weit. Beob.,’ II, pp. 211-12; Kirchner, ‘ Flora v. Stuttgart,’ p. 325 ; Knuth, ‘Bl. u.
Insekt. a. d. nordfr. Ins.,’ pp. 49, 152, ‘ Weit. Beob.
ti. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 232, ‘ Bliiten-
biol. Beob. a. d. Ins. Riigen,’ ‘ Bloemenbiol.
Bijdragen.’)—In the flowers of this species the
three spreading styles are situated between the
three bundles of filaments. The anthers dehisce
upwards—the innermost first—and do not usually
touch the stigmas, which are on the same level,
so that cross- and self-pollination alike depend
upon insect-visits. When anthesis is over, the
petals and stamens are drawn inwards, the stig-
mas being thus generally brought into contact
with the anthers, which are still covered with Fic. 61. Hypericum perforatum, L. (after

= : : eB Herm. Miller). Flower seen obliquely from
pollen, so that automatic self-pollination takes ayove. a, a, a, the three stigmas.

place if insect-visits have failed.

Visitors. — Hermann Miiller (H. M.) in Westphalia, Buddeberg (Budd.) in
Nassau, and myself (Kn.) in Schleswig-Holstein and Pomerania, have observed
the following.—

A. Coleoptera. Chrysomelidae : 1. Cryptocephalus sericeus Z., po-dvg. (H. M.).
B. Diptera. (a) Bombyhidae: 2. Anthrax flava A/g., po-dvg. (?) (H. M., Thuringia);
3. A. maura Z. (H. M., Thuringia); 4. Argyromoeba sinuata /a/l., vainly searching
for nectar (H. M.); 5. Bombylius canescens Afskan., skg. (H. M.). (6) Lmpidae:
6. Empis livida Z., do. (H.M.). (¢) Muscidae: 7. Musca sp. (H. M.). (d) Syrphidae:
8. Ascia podagrica /., po-dvg. (H. M.); 9. Eristalis aeneus Scop. 9, po-dvg. (Kn.,

204 ANGIOSPERMAE—DICOTYLEDONES

Riigen); 10. E. arbustorum Z., do. (H. M.); 11. E. nemorum Z. (H.M., Kn.); 12. E.
sepulcralis Z., do. (H. M.); 13. E. sp. (Kn.); 14. E. tenax Z., do. (H. M.); 15. Helo-
philus pendulinus Z., do.(H. M.); 16. H. trivittatus /., do. (H. M.); 17. Melanostoma
mellina Z., do. (H. M.); 18. Melithreptus pictus JZg., do. (H. M.); 19. M. scriptus
L., do. (H. M.); 20. Syrphus balteatus Deg., do.(H. M.); 21. S. ribesii Z.,do.(H. M.,
Kn.); 22. S. sp., do.(Kn.). C. Hymenoptera. (a2) Apidae: 23. Andrena shawella
K. 9, po-cltg. (H. M.); 24. A. dorsata K. 9, do.(H.M.); 25. A. fulvicrus X. 9, do.
(Budd.); 26. A. nigriceps K. 9, do. (Kn.); 27. Apis mellifica Z. ¥, do. (Kn.); 28.
Bombus agrorum F. 9, do. (H. M., Kn.); 29. B. lapidarius Z., do. (H.M.); 30. B.
rajellus X. 9, do. (H.M.); 31. B. terrester Z. ¥, do. (H. M., Kn.); 32. Cilissa
melanura Wy/. 9, do. (H.M.); 33. Halictus cylindricus /. 9, do. (Budd.); 34. H.
malachurus &, 9, do. (Budd.); 35. H. morio F. 9, do. (Budd.); 36. Nomada lateralis
Pz. 9, skg.(H.M.); 347. N. lineola Pz. 9, do.(H.M.); 38. Prosopis armillata JVy/.,
po-cltg. (H.M.); 39. Saropoda bimaculata Pz., skg. (H.M.). (4) Yenthredinidae :
40. Tenthredo sp., vainly searching for nectar (H.M.). D. Lepidoptera. Rho-
palocera: 41. Hesperia sylvanus £sp., trying to bore into the tissues (H.M.); 42.
Melitaea athalia Rots, do. (H.M.); 43. Pieris rapae Z., do. (H.M.); 44. Epinephele
janira Z., do. (H. M.).

The following were noticed by Loew in Silesia (‘ Beitrage,’ pp. 28, 46).—

A. Coleoptera. Chrysomeitdae: 1. Cryptocephalus sericeus Z., po-dvg;
2. Chrysomela varians Schall. B. Diptera. Syrphidae: 3. Didea intermedia Zw.,
po-dvg.; 4. Eristalis horticola Deg., do. C. Hymenoptera. Afzdae: 5. Bombus
terrester Z., po-cltg.; 6. Diphysis serratulae Pz. 9, do. D. Lepidoptera. Rhopa-
locera: 7. Argynnis paphea Z., vainly seeking for nectar.

In Bremen Alfken saw 3 po-cltg. humble-bees,—1. Bombus lapidarius Z. ¥ ;
2. B. terrester Z. ¥; 3. B. hortorum Z. ¥. MacLeod noticed a humble-bee, 3 hover-
flies and a Muscid in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 225-6),
and Bombus terrester Z., po-cltg., in the Pyrenees (op. cit., iii, 1891, p. 400).

Willis observed the following in the neighbourhood of the south coast of Scotland
(‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera, MWitdulidae: 1. Meligethes aeneus /, freqg., po-dvg. B.
Diptera. (a) Empidae: 2. Tachydromia sp., po-dvg.; 3. Anthomyia radicum Z.,
very freq., po-dvg.; 4. A.sp., po-dvg. (6) Musczdae: 5. Calliphora erythrocephala
Mg., do.; 6. C. vomitoria L., do.; 7. Morellia sp.,do.; 8. Mydaea sp., freq., po-dvg.;
g. Stomoxys calcitrans Z., do. (c) Syrphidae: 10. Eristalis pertinax Scop., freq.,
po-dvg.; 11. Platycheirus albimanus /, po-dvg.; 12. P. peltatus J/g., freq., po-dvg. ;
13. Syritta pipiens Z., do.; 14. Syrphus balteatus Deg., do.; 15. S. topiarius AZg.,
po-dvg. C. Hemiptera. 16. An undetermined sp. D. Hymenoptera. (a)
Apidae: 17. Bombus agrorum F., vainly searching for nectar. (4) /ehneumontdae :
18. An undetermined sp.

In Dumfriesshire, Apis, 4 humble-bees, an Empid, 8 hover-flies, and 4 Muscids
have been recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 32).

509. H. hirsutum L. (Herm. Miiller, ‘Fertilisation,’ pp. 140-1; Kirchner,
‘Flora v. Stuttgart,’ p. 327.)—This species has somewhat smaller flowers and fewer
anthers than the last. As Hermann Miller explains, the three groups of anthers
are therefore separated by wider spaces, so that automatic self-pollination is prevented
in the open flower. It regularly takes place, however, by the closing of the flower
even before the end of anthesis. The result, according to Hermann Miiller, seems
to be complete fertility.

Visttors.—In Dumfriesshire, 2 humble-bees, an Empid, 3 Muscids, and a
hover-fly have been recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 34).

HYPERICINEAE 205

510. H. quadrangulum L. (Herm. Miiller, ‘ Fertilisation,’ p. 141, ‘Weit. Beob.,’
II, p. 212; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 266; Kirchner,
‘Flora v. Stuttgart,’ p. 325; Knuth, ‘ Bliitenbiol. Beob. a. d. Ins. Rigen,’ ‘ Bloemen-
biol. Bijdragen.’)—Hermann Miiller says that this species is intermediate between
the two foregoing ones in respect of the size of the flowers and the number of
stamens. He observed that in open flowers the stigmas were not in immediate
contact with the anthers. Probably, however, when fading takes place, there is
a contraction of parts followed by automatic self-pollination.

Visitors. — Herm, Miiller (H. M.) and myself (Kn.) have observed the
following.—

A. Diptera. (a) Muscidae: 1. Aricia vagans Fall., po-dvg. (H. M.). (4)
Syrphidae: 2. Syritta pipiens Z., po-dvg. (Kn.); 3. Syrphus balteatus Deg., do.
(H. M.); 4. S. ribesii Z., do. (Kn.). B. Hymenoptera. Apidae: 5. Apis mellifica
L.%, po-cltg. (Kn.); 6. Bombus agrorum F” ¥, do. (Kn., Riigen); 7. B. terrester

£. ¥%, do. (Kn.). The latter humble-bee was also observed by Loew in the Berlin
Botanic Garden, po-cltg.

In Dumfriesshire, Apis, 3 humble-bees, a Dolichopid, and 3 Muscids have been
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 32).

511. H. commutatum Nolte (=H. perforatum x H. quadrangulum).

Visitors.—Loew observed Apis mellifica Z. § and Bombus terrester Z. %, both
po-cltg., in the Berlin Botanic Garden.

512. H. tetrapterum Fries. (Herm. Miiller, ‘ Weit. Beob.,’ II, p. 212 ; Kirchner,
‘Flora v. Stuttgart,’ p. 325.)—In this species again the flower mechanism is similar to
those of allied forms. Kirchner states that automatic self-pollination is usually
impossible in the open flower.

Visitors.—Herm. Miiller gives the following.—

A. Coleoptera. Nitidulidae: 1. Meligethes aeneus F., po-dvg. B. Diptera.
(2) Muscidae: 2. Aricia incana Wiedem., freq., po-dvg.; 3. A. vagans Fall, do.
(4) Syrphidae: 4. Syrphus balteatus Dey., po-dvg. C. Hymenoptera. Apidae:
5. Apis mellifica Z. ¥, po-cltg.; 6. Bombus terrester Z. 9 and ¥, do.

MacLeod observed a humble-bee and a Muscid in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 226).

513. H. pulchrum L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ p. 49.)—The
flower mechanism once more agrees with those of related species. The diameter of
the flowers of plants studied by me in the North Frisian Islands is about 1-5 cm. ;
the number of stamens is about fifty.

Visitors.—MacLeod observed small Diptera in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 378)-

In Dumfriesshire, Apis and numerous Diptera were recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 53).

514. H. humifusum L. (Herm. Miiller, ‘Fertilisation,’ p. 141; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 226; Kirchner, ‘Flora v. Stuttgart,’ p. 326;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 52.)—The flowers of this species observed
by me in the North Frisian Islands had only 10-15 stamens. Automatic self-
pollination regularly takes place in closing flowers, and often also even in open
ones (Herm. Miller). Kerner says that the flowers do not open in unfavourable

206 ANGIOSPERMAE—DICOTYLEDONES

weather, and then automatic self-pollination takes place as a result of the pseudo-
cleistogamy.

According to Warnstorf (Verh. bot. Ver., Berlin, xxxviii. 1896), the diameter
of the flower may be as much as 8 mm.; the petals have black marginal glands.
The pollen is yellow, ovoid, delicately tuberculated, about 31 » long and
15 p broad.

Visitors.—In Dumfriesshire, several Muscidae have been recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 33).

515. H. Elodes Huds. (=Elodes palustris Spach). (MacLeod, Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 226-7.)—The base of each yellow petal bears a
split scale, which perhaps secretes nectar. Above these scales, between the staminal
bundles, there are small liguliform bifid glands (? modified stamens) which are
applied to the ovary, and perhaps also secrete nectar.

Visrtors.—In Dumfriesshire, 2 Muscid has been observed (Scott-Elliot, ‘ Flora
of Dumfriesshire,’ p. 34).

XX. ORDER MALVACEAE R. BR.

Literature.—Knuth, ‘Grundriss d. Bliitenbiol.,’ p. 32.

The bright colour of the large corolla, and often also of the pyramid of stamens,
renders the flower conspicuous. Nectar is secreted between the bases of the petals,
or at the bottom of the calyx. Many species, therefore, belong to the flower class C.
Individual genera (Hibiscus) include nectarless species, which consequently belong
to class Po. Almost all the Malvaceae are markedly protandrous. The South
Brazilian species of Abutilon are pollinated by humming-birds, which discharge
their function so actively that the possibility of autogamy has been lost (Herm. Miiller,
‘Fertilisation,’ p. 146).

147. Malva L.

Flowers protandrous, with concealed nectar, secreted as above described.

516. M. sylvestris. (Sprengel, ‘Entd. Geh.,’ pp. 347-50; Herm. Miller,
‘Fertilisation,’ pp. 142-4, ‘Weit. Beob.,’ IJ, p. 221; Knuth, ‘Bl. u. Insekt. a. d.
nordfr. Ins.,’ pp. 47-8, 152; Kirchner, ‘Flora v. Stuttgart,’ p. 331.)—The red petals
are marked with dark circles, which serve as nectar-guides, and above the secre-
tion are ciliary hairs serving as a protection. In the first stage of anthesis the
anthers meet in the form of a pyramid above the filaments—which are united below—
completely covering the still immature stigmatic branches enclosed in the staminal
tube, so that the mature anthers alone occupy the middle of the flower. After they
have dehisced the stamens curve downwards, while the stigmatic branches develop,
and spread out so that the papillae on their inner surfaces now occupy the place
where the anthers were at first situated. Insects visiting the flower must, therefore,
regularly effect crossing. Automatic self-pollination is excluded, and is unnecessary,
owing to the large number of insect visitors.

Visitors.—Herm. Miiller (H. M.) and myself (Kn.) in Central and North
Germany have observed the following).—

MALVACEAE 207

A. Coleoptera. (2) Chrysomelidae: 1. Mantura fuscicornis Z., po-dvg. (H.M.).
(6) Nitedulidae: 2. Meligethes, po-dvg. (H. M.). (c) Zelephoridae: 3. Danacea pallipes
Pz., resting in the flowers (H. M.). B. Diptera. (a) Muscidae: 4. Ulidia erythroph-
thalma A7fg., resting in the flowers (H.M.). (4) Siratiomyidae: 5. Sargus cuprarius
L., a useless guest (H. M.). (c) Syrphidae: 6. Rhingia rostrata L., freq., skg. (H. M.).
C. Hemiptera. 7. Pyrrhocoris apterus Z., skg. D. Hymenoptera. (a) Apzdae:
8. Andrena fulvicrus K. 8, skg. (H. M.); 9. A. gwynana X. 9, do. (H.M.); 10. A.
parvula A. 8, do. (H.M.); 11. Apis mellifica Z. ¥, freq., skg. (H.M., Kn.); 12.
Bombus agrorum /. ¥, skg. (H.M.); 13. B. hortorum Z. 9, do. (H. M.); 14. B.
lapidarius Z. 9, freq. skg. (H. M., Kn.); 15. B. agrorum F. 9 and ¥, skg. (H.M.);
16. B. pratorum Z. 6,9 and ¥, in large numbers, skg. (H.M.); 17. B. sylvarum ZL. 9,
skg. (H. M.); 18. Chelostoma campanularum Z. é, do. (H.M.); 1g. C. nigricorne
Nyi. 3 and 9, very freq., skg. and po-dvg. (H.M.); 20. Cilissa haemorrhoidalis /. 6
and 9, skg. ti. M.); 21. Coelioxys conoidea ///, 8, do. (H. M.); 22. C. elongata Lep.
é and 9, do. (H. M.); 23. Halictus albipes F. 9, freq., skg.(H. M.); 24. H. cylindricus
fF. 9, skg. (H.M.); 25. H. flavipes #. 9, do. (H. M.); 26. H. maculatus Sm. 9, do.
(H. M.); 27. H. morio F.8 and 9, do. (H.M.); 28. H. nigerrimus Schenck 9,
do. (H.M.); 29. H. smeathmanellus X. 9, do. (H. M.); 30. H. zonulus Sm. 8, do.
(H. M.); 31. Megachile ligniseca X. 8, do. (H.M.); 32. M. willughbiella X. 4,
do. (H. M.); 33. Nomada lateralis Pz. 9, do. (H.M.); 34. Osmia aenea ZL. 3, do.
(H.M.); 35. O. aurulenta Pz. 9, do. (H.M.); 36. P. communis My/. é and 9, seen
repeatedly, skg. (H. M.); 37. P. dilatata A. 3, skg. (H.M.); 38. P. hyalinata Sm. 4,
do. (H. M.); 39. P. pictipes Wy/. 8, do. (H. M.); 40. P. signata Pz. , do. (H. M.);
41. Stelis aterrima Pz. $, do. (H.M.); 42. S. minuta Zep. 6, do. (H.M.). (6) Lch-
neumonidae: 43. Ichneumon sp., vainly searching for nectar (?) (H. M.). (c) Vesprdae:
44. Odynerus melanocephalus Z. 9, do. (H. M.). E. Lepidoptera. Rhopalocera:
45. Pieris rapae Z., skg. (H. M.).

Schenck noticed Osmia caerulescens Z. in Nassau, and Alfken saw the
following Apidae in Bremen.—

1. Bombus agrorum Ff. ¥, freq., skg. and po-cltg.; 2. B. arenicola Zs. ¥, do. ;
3. B. derhamellus X. ¥, skg. and po-cltg., 6, skg.; 4. B. proteus Gerst.9; 5. B.
sylvarum ZL. ¥; 6. Coelioxys elongata Zep. 9, skg.; 7. C. rufescens Zep. § and 9, do.;
8. Eriades florisomnis Z. 9, very freq., skg. and po-cltg.; 9. E. nigricornis Vy/. 9, do. ;
10. E. truncorum Z.9; 11. Megachile centuncularis Z. §; 12. M. willughbiella
K.9; 13. Osmia caerulescens Z.9; 14. Podalirius furcatus Pz. §; 15. Prosopis
communis JVy/. and 9.

Schletterer observed the following bees at Pola.

1, Andrena albopunctata Rossi’; 2. Colletes fodiens Lir.; 3. Halictus scabiosae
Ross; 4. Megachile muraria Z.; 5. Osmia andrenoides Spzn.; 6. O. rufohirta Zér.

Loew saw two bees—Apis mellifica Z. ¥, po-cltg., and B. terrester Z. 9, do.—
and a butterfly—Pieris brassicae Z., skg.—in the Berlin Botanic Garden. MacLeod
noticed Apis and a small fly in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,

. 224-8).

In Deainianee 2 humble-bees and 2 hover-flies have been recorded (Scott-
Elliot, ‘ Flora of Dumfriesshire,’ p. 36).

Smith observed the following Afzdae in England. —1. Andrena gwynana X.,
and generation; 2. Stelis aterrima Pz.; 3. S. phaeoptera K.

The very numerous bees observed by Herm. Miiller—with the exception of
one species (Chelostoma nigricorne Wy/.)—never collected pollen, but were always
nectar-suckers, though they constantly covered themselves with the spinulose pollen-
grains. The above-named species of bee, however, collected unusually large
pollen-balls.

208 ANGIOSPERMAE—DICOTYLEDONES

Herm. Miiller calls attention (‘ Fertilisation,’ p. 14) to a peculiarity of the flowers
of this species, i.e. that they are not adequately protected against nectar-thieves.
In the afternoon, when the flowers begin to close, the honey-bee frequently thrusts
its proboscis between the five sepals of blossoms which are still fresh though shut,
and thus empties the nectar receptacle from outside. On some few occasions Miiller
even observed bees which had rifled several such flowers in this way, continuing
their depredations on neighbouring ones that were still open.

517. M. rotundifolia L. (=M. borealis Wallm.). (Herm. Miiller, ‘ Fertilisation,’
pp. 142-4, ‘ Weit. Beob., II, p. 221; Warnstorf, Verh. bot. Ver., Berlin, xxxvii,
1895.)—The flower mechanism of this species at the beginning of anthesis is the
same as that of the last one. But, in correlation with its much smaller and less
brightly coloured flowers, M. rotundifolia is capable of automatic self-pollination,
which is necessary for the maintenance of the species since the number of visitors
is naturally much smaller. The stamens remain so far erect that the pollen-covered
anthers are touched by the recurved stigmatic branches (see Fig. 60).

Fic. 62. Malva sylvesiris, L., and M. rotundifolia, L. (after Herm. Miiller). MM. sylvestrzs:
(1) Column of stamens in the bud, enclosing the styles. (2) Sexual organs in the first (male) stage.
(3) The same in the transition from the first to the second stage. (4) The same in the second (female)
stage. (5) M. rotundzfolia in the last stage, showing self-pollination. a, anthers; st, stigmas.

Warnstorf gives a somewhat different account, as follows.—Most of the German
species of Malva, such as M. Alcea, M. sylvestris, and M. neglecta, possess strongly
protandrous flowers, so that self-pollination appears to be excluded, at least in the
first stage, though later on it is possible, since some of the large thickly spinulose
pollen-grains remain adhering to the dehisced anthers. But M. rotundifolia, on
the other hand, has very small inconspicuous flowers, usually concealed under a
dense covering of the leaves. They can scarcely depend upon insect-visits, and
are almost homogamous. (See, however, the appended list of visitors.) Even at
the beginning of anthesis, the stigmas have more or less curved outward to receive
pollen, and are at once seen on looking down into the open flower, Owing to
the smallness of the hidden flowers, insect-visits are rendered impossible, or at least
very unlikely, but homogamy completely compensates for this. In Buslar (Pome-
rania), where this species—as well as M. neglecta—is quite common, Warnstorf
only on a few occasions observed in the flowers a few winged ants, upon the wings
of which were numerous pollen-grains, so that they were clearly able to effect
cross- pollination.

MALVACEAE 209

The diameter of the pollen-grains in M. rotundifolia is about roo p, in M.
neglecta about 112 », and in M. sylvestris as much as 144 pw. In all these species
they are closely beset with long spines (Warnstorf).

Vistrors.—Herm. Miiller observed the following.—

A. Hymenoptera. Apidae: 1. Anthophora quadrimaculata 7. 6; 2. Apis
mellifica Z. ¥; 3. Bombus agrorum F. ¥; 4. Halictus morio F. 3; 5. H. tetrazonius
Kil. 9; allskg. B. Hemiptera. 5. Pyrrhocoris aptera Z., skg.

MacLeod saw Apis, 2 sp. of Halictus, Syritta, and a Muscid in Flanders (Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 229), and an Apid in the Pyrenees.

518. M. neglecta With. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 48,
152.)—The flower mechanism of this species is intermediate between those of M.
sylvestris and M. rotundifolia. Here again, at the beginning of anthesis, the pyramid of
anthers borne on the united filaments completely encloses the still completely immature
stigmatic branches. After dehiscence the upper free ends of the filaments bend
downwards, so as to uncover the previously enclosed stigmas. These now spread
out in all directions, curving backwards so far that the papillae on their inner sides

FiG. 63. Malva neglecta, With. (Magnified about 5 times, after removal of calyx and corolla. From
nature.) (1) Bud condition: column of stamens with unripe anthers. (2) Early male condition: column
of stamens with dehiscing anthers surrounding the immature stigmas. (3) Later male condition: the
styles project somewhat from the column of stamens, otherwise as before. (4) Early hermaphrodite
condition (adapted to cross-pollination): the receptive stigmas project above the stamens (which are
directed downwards, but still covered with pollen) and spread out like a star in the entrance to the flower.
(5) Later hermaphrodite condition (adapted to automatic self-pollination): the stigmas have rolled them-
selves spirally round the anthers, which are still covered with pollen. a, stamens; s, stigmas;
m, annular nectary ring.

project freely, and occupy the place where the anthers were at first situated. Insects
coming from a flower in the first stage to one in the second stage must consequently
effect cross-pollination. Towards the end of anthesis, the stigmatic branches curve
downwards so far as to come into contact with the stamens which are still covered
with pollen to some extent; and thus automatic self-pollination takes place, if insect-
visits have failed.

Vistrors.—In the island of Féhr I only observed the honey-bee, skg. and
po-cltg.

sig. M. mauritiana L.—Kirchner (‘Flora v. Stuttgart,’ p. 332) states that
the flower mechanism of this species agrees with that of M. sylvestris.

520. M. Alcea L. (Herm. Miiller, ‘Fertilisation,’ p. 144, ‘Weit. Beob.,’ II,
p. 221; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—As No. 519. .

DAVIS. IT Pp

210 ANGIOSPERMAE—DICOTYLEDONES

Visirors.—Herm. Miiller (H. M.) in Westphalia, and Buddeberg (Budd.) in
Nassau, observed the following Apzdae.—

1. Andrena schrankella Wy. 5, skg. (H. see 2. Apis mellifica Z. ¥, very freq.,
skg. (H.M.); 3. Chelostoma nigricorne My/. 3, skg. (H.M.); 4. Cilissa haemor-
thoidalis #. 6, do. (H. M., Budd.); 5. Halictus cylindricus /. 9, do. (H. M.);
6. Rhophites canus Zz. 6, do. (Budd.).

Friese saw Eucera malvae Rossz at Bozen.
Loew noticed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syrphus balteatus Deg., po-dvg. B. Hymeno-
ptera. Apzdae: 2. Apis mellifica Z. ¥, po-cltg. C. Lepidoptera. Rhopalocera:
3. Pieris brassicae Z., skg.; 4. Spilothyrus alceae Lsp., do.

521. M. moschata L. (Herm. Miller, ‘Fertilisation,’ p. 144.)—As No. 519.
Visitors.—Herm. Miiller observed the following.—

A. Diptera. Bombylidae: 1. Systoechus sulphureus Afzkan., skkg. B. Hymeno-
ptera. Apzdae: 2. Andrena coitana X. 4, skg.; 3. Apis mellifica Z. ¥, do.;
4. Chelostoma nigricorne Z. 9, do. C. Lepidoptera. Rhopalocera: 5. Hesperia
sylvanus L'sp., skg.

MacLeod saw Bombus lapidarius Z. § in the Pyrenees (Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 401).

In Dumfriesshire, Apis and 2 humble-bees were recorded (Scott-Elliot, ‘ Flora of
Dumfriesshire,’ p. 35).

148. Lavatera L.

Like the last genus.

522, L. thuringiaca L—The large pale rose-red flowers of this species are
protandrous, according to Schulz (‘Beitrage,’ I, p. 26). The anthers of the 7o-
go stamens often remain closed for a time after the flower opens. Dehiscence
begins above, and after it is over the filaments do not curve downwards. Before
the lower anthers have shed their pollen, the styles—which have so far been enclosed
in the tube formed by the filaments—elongate, and curve outwards until the
stigmatic branches touch the anthers. But the latter are usually destitute of pollen
at this stage, so that automatic self-pollination is unusual.

Visitors. — Loew observed the honey-bee, po-cltg., in the Berlin Botanic
Garden.

523. L. trimestris L.—

VisrTors.—Schenck observed the fossorial wasp Crabro serripes Pz. in Nassau.

149. Kitaibelia Willd.
524. K. vitifolia Willd.—
VisiTors.—Loew observed the hover-fly Syrphus balteatus Deg., po-dvg., in the
Berlin Botanic Garden.
150, Althaea L.
Like Malva.

525. A. ficifolia Cav.— This species is self-sterile (Comes, ‘Ult. stud. e
consideraz. sulla impollinaz. delle piante’).

MALVACEAE 211

VisiTors.—Loew observed two bees in the Berlin Botanic Garden—Bombus

terrester Z. 8, skg. and dusting itself thickly with pollen, and Psithyrus vestalis
Fourcr. 8, do.

526. A. rosea Cav. (Kirchner, ‘Flora v. Stuttgart,’ p. 333; Knuth, ‘ Bloemen-
biol. Bijdragen.’)—The very large white, yellow, red or blackish flowers of this well-
known ornamental plant are markedly protandrous. Their diameter is 6-7 cm.;
each petal is about 4 cm. long, and 5-6 cm. broad above. Nectar is secreted by
five yellow areas at the bottom of the calyx, between the bases of the petals. It
is protected from rain and small insects by hairs on the petals. Automatic self-
pollination takes place if insect-visits fail, the stigmas curving back among the
anthers that have not yet lost all their pollen.

VisiTors.—I observed Apis mellifica Z. and Bombus terrester Z. Both sucked
persistently, although I was unable to taste the nectar. They flew from flower to
flower, skg. in each, and constantly effecting cross-pollination. Loew noticed the
same visitors in the Berlin Botanic Garden.

Alfken observed the humble-bees Bombus hortorum Z. 4, and B. agrorum F. é,
skg., at Bremen. Réssler at Wiesbaden saw the Geometrid Ortholitha cervinata S.V.

For the Tyrol, Schletterer describes Bombus pascuorum Scop., which is widely
distributed in South Europe.

527. A. officinalis L. (Knuth, ‘Bloemenbiol. Bijdragen.’)—The flowers are
the same in structure as those of the last species, but considerably smaller, their
diameter being 2-3 cm., while the petals are 2 cm. long, and about as broad.

VisiTors.—In the Kiel Botanic Garden I observed the honey-bee and Bombus
terrester, skg. Schletterer describes the garden humble-bee for the Tyrol.

528. A. cannabina L.—

VisiTrors.—In the Botanic Garden at Berlin Loew observed the following.—

A. Diptera. Syrphzdae: 1. Eristalis nemorum Z., po-dvg.; 2. E. tenax Z., do.
B. Hymenoptera. Apidae: 3. Apis mellifica Z. ¥, po-cltg.

151. Hibiscus L.

Protandrous pollen flowers.

529. H. Trionum L.—The flowers of this species are yellow with a purple
base. Kerner (‘ Nat. Hist. Pl.’ Eng. Ed. 1, I, p. 212) says that they open between
8am. and noon. The pollen-covered anthers rise up in the middle of the newly
opened flower, and the free parts of the filaments soon curve downwards so that
the stigmatic branches, as they become receptive, may occupy the place of the
anthers. Insect visitors must therefore effect cross-pollination. After a few hours
the styles become S-shaped, and curve downwards to such an extent that the
stigmatic papillae come into contact with the anthers, which are still covered with
pollen.

152. Abutilon Tourn.

530. A. Avicennae DC. (= Sida Abutilon Z.).— Kerner states that the
flowers of this species open from ro to 6 o’clock, and present the same arrangements

as those of Hibiscus Trionum.
P 2

212 ANGIOSPERMAE—DICOTYLEDONES

153- Anoda Cav.

Protandrous pollen flowers.

531. A. hastata Cav. (Hildebrand, ‘Die Geschlechtsvert. b. d. Pfi.,’
pp. 48-9.)—As in the case of Malva, the anthers at first form a pyramid, which
encloses the still undeveloped styles. The upper stamens are erect, the lower ones
are curved back. The anthers of the upper stamens dehisce first; those of the
lower ones follow them, and at the same time their filaments become upright. We
can now see the styles, which are curved downwards, arranged in five bundles,
and closely apposed to the column of filaments. This is beset with hairs, which
project between the bundles of styles, protecting the reddish stigmas from contact
with insect visitors. When the dehisced stamens curve backwards, the styles erect
themselves, so that the stigmas take up the position occupied by the pollen-covered
anthers during the first stage of anthesis.

Fig. 64. Anoda hastata, Cav. (after Hildebrand). a. Condition inthe bud. 4. Beginning of the first
(male) stage: some of the stamens are erect and covered with pollen. c and d. Male stage: all the
stamens are erect and their anthers are covered with pollen, while the styles (¢) are still curved downwards.
e. Second (female) stage ; the styles have become erect.

154. Goethea Nees.

532. G. coccinea. (Delpino, ‘ Altri appar. dicog. recent. osserv.,’ p. 59; Hilde-
brand, ‘ Die Geschlechtsvert. b. d. Pfl.,’ p. 19.}—In this species the nectar is secreted
by five glands at the bottom of the calyx. The corolla forms a nectar-cover, and
the four petals make the protogynous flower conspicuous.

Vistrors.—Delpino supposes that these are bees or humming-birds.

155. Pavonia Cav.

533. P. hastata Cav.—Heckel (C.-R. Acad. sci., Paris, lxxxix, 1879) states
that this species possesses cleistogamous flowers.

“STERCULIACEAE, TILIACEAE 213

156. Malope L.

534. M. grandiflora F. G. Dietr. (Knuth, ‘ Bliitenbiol. Notizen.’)—This species
exhibits the same marked protandry as Malva.

VisiTrors.—On October 10, 1897, I observed Apis mellifica Z., skg., in the
Garden of the Kiel Ober-Realschule. This insect dusted the under-side of its body
with pollen in flowers that were in the first stage, and transferred some of it to
the stigmas of flowers in the second stage.

Schenck observed the fossorial wasp Crabro serripes Pz. in Nassau.

XXI. ORDER STERCULIACEAE VENT.

157. Pterospermum Schreb.

535. P. acerifolium Willd.—Lanza (‘ Note di biol. fior.,’ 1894) states that this
species is protandrous in the Palermo Botanic Garden. The adynamandrous
blossoms are moth flowers.

158. Cheirostemon Humb. et Bonpl.

536. C. platanoides Humb. et Bonpl.—Lanza (op. cit.) states that autogamy
cannot take place. The plants in the Palermo Botanic Garden never set fruits,
apparently because the appropriate pollinating agent does not occur there.

159. Rulingia R. Br.
537- R. pannosa R. Br.—This species is protandrous (Urban, Ber. D. bot.
Ges., Berlin, i, 1883, pp. 53-6).
538. R. corylifolia R. Grah.—This species is homogamous (op. cit.).

539. R. parviflora Endl.—The petals are yellowish-white at first, but become
rose-coloured after the pollen has been shed, as in Weigelia (op. cit.).

XXII ORDER 7IZIACEAE JUSS.

This family is represented in Europe by the genus.

160. Tilia L.
Flowers usually whitish, and belonging to class E.

540. T. platyphyllos Scop. (=T. grandifolia Zhrh.) and 541. T. ulmifolia
Scop. (=T. parvifolia Lzrh.). (Sprengel, ‘Entd. Geh.,’ pp. 275-6; Herm. Miller,
‘Fertilisation,’ pp. 146-7, ‘ Weit. Beob.,’ II, p. 219; MacLeod, Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 227; Hildebrand, Bot. Ztg., Leipzig, xxvii, 1869; Kirchner,
‘Flora v. Stuttgart,’ p. 329; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 48, 152
‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ ‘Bloemenbiol. Bijdragen.’)— The
flower mechanism and visitors are the same in these two species, but the latter
blossoms about a fortnight later than the former. As the flowers are pendent, the

214 ANGIOSPERMAE—DICOTYLEDONES

nectar, secreted and sheltered by the hollowed sepals, is protected from rain. The
yellowish flowers smell strongly of honey, and are protandrous, as Hildebrand first
pointed out (op. cit.). The numerous stamens curve outwards, projecting beyond
sepals and petals. Insects settling upon the hanging blossoms can therefore only
find footing on the stamens and stigmas, or in the space between them. In the
younger flowers, therefore, they dust themselves with pollen, and transfer this to
the stigmas of the older flowers, Automatic self-pollination is scarcely possible,
for the stamens remain curved outwards to the end of anthesis. The flowers, how-
ever, are visited by so many insects that crossing is certain. Warnstorf (Verh. bot.
Ver., Berlin, xxxvii, 1895) says that the pollen-grains of T. platyphyllos are white,
tetrahedral, closely tuberculated, opaque, with three germinating processes in the
middle of the edges, and averaging 31 « in diameter. Jordan states that the exposed
nectar collects in two little pits on the bases of the sepals, and is accessible even
to very short-tongued insects. Besides the honey-bee which visits the flowers in
thousands, and collects nectar only, not pollen, other Apidae, as well as Syrphidae
and Muscidae, are very frequent visitors.

Visirors.— Herm. Miiller (H. M.) in Westphalia, and myself (Kn.) in Schleswig-
Holstein, have observed the following.—

A. Diptera. (a) Muscidae: 1. Lucilia cornicina &., ske. (H.M.); 2. Musca
domestica Z., do. (H. M., Kn.); 3. Sarcophaga carnaria Z., do. (H.M., Kn.). (4)
Syrphidae: 4. Eristalis arbustorum Z., skg. (H.M.); 5. E. nemorum Z., do. (H. M.,
Kn.); 6. E. sepulcralis Z., do. (H. M.); 7. E. tenax Z., do. (H. M., Kn.); 8. Helo-
philus floreus Z., very freq., skg. and po-dvg. (H.M.); 9. Volucella bombylans Z.,
skg. (Kn.); 1o. V. pellucens Z., do. (H.M.). (c) Zadantdae: 11. Tabanus bovinus
Z., do. (H.M.). B. Hymenoptera. (a) Apzdae: 12. Apis mellifica Z. ¥, skg.
(H.M., Kn.); 13. Bombus agrorum /” 9, freq., skg. (H. M.); 14. B. lapidarius Z.,
skg. {Kn.); 15. B. soroénsis /., var. proteus Gers?., do. (Kn.); 16. B. terrester Z., do.
(Kn.); 17. Prosopis, freq. (H.M.). (6) Sphegzdae: 18. Oxybelus uniglumis Z., freq.,
nect-lkg. (H. M.).

Alfken saw the following at Bremen.—

A. Diptera. Zmpidae: 1. Empis tessellata #. B. Hymenoptera. (ca)
Apidae: 2. Bombus agrorum /. ¥ and 9; 3. B. muscorum /. ¥ and 9. (6) Vespedae:
4. Vespa crabro Z. ¥ and 9.

542. T. tomentosa Moench. (Kirchner, ‘ Flora v. Stuttgart,’ p. 330.) —This
Hungarian species possesses bright yellow homogamous flowers, in which the stigma
projects beyond the anthers, so that cross-pollination is assured in the event of
insect-visits.

543. T. sylvestris Desf. --

Visrrors.—MacLeod saw a humble-bee and 4 Diptera in the Pyrenees (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 400).

XXIII. ORDER ZINEAE DC.

Flowers homogamous, with concealed nectar. Dimorphism frequent.

Alefeld says that many European, Asiatic, and North African species are
dimorphous, while the Cape species, and those native to North and South America,
are monomorphous.

LINEAE 215

161. Linum L.

544. L. catharticum L. (Herm. Miller, ‘Fertilisation, p. 147; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 238-9; Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896.)}—The filaments of the small white homogamous flowers are
fused at the base into a fleshy ring, which, as Herm. Miiller explains, secretes
on its outer side five drops of nectar from five small, flat pits situated in the middle of
the filaments. The five petals are inserted into the same ring, a little higher than
the nectar-pits, and alternating with them. The lower halves of their edges are
closely applied, but their bases suddenly narrow, so as to leave a small round
aperture—giving access to the nectar—immediately above each nectar-pit. The
anthers are at the same level as the stigmas, but are at first remote from them,

Fic. 65. Linum catharticum, L. (after Herm. Miller). (1) Young flower, seen from above: the
anthers are still remote from the stigmas. (2) The same seen obliquely from above. (3) Somewhat
older flower seen from above: the anthers are applied to the stigmas (4) Flower after removal of the
calyx to show the insertion of the petals and the position of the nectaries. (5) Stamens and stigmas in
the act of automatic self-pollination. (6) Sepa} seen from the inner side, with a drop of nectar.
a, anthers; a', the united filaments; s¢, stigma; 7, nectaries.

so that insect visitors may effect either self- or cross-pollination. Failing such
visitors, automatic self-pollination takes place, the stamens bending more and more
inwards, and the flowers closing in the evening.

Warnstorf describes the flowers as slightly protogynous, the stigmas maturing
before the flower opens. The pollen-grains are large, golden-yellow, spherical or
ellipsoidal, tuberculated, with a maximum length of 50 » and breadth of 30-37 p.

Visrrors.—Herm. Miiller saw the following.—2 nect-skg. flies, a Bombylid—
Systoechus sulphureus J/:kan in Westphalia—and an Empid—Empis livida Z. in
Thuringia.

MacLeod observed a Syrphid and a Bombylid in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, p. 406).

In Dumfriesshire, an Empid, a Muscid, and a hover-fly were recorded (Scott-
Elliot, ‘ Flora of Dumfriesshire,’ p. 35).

216 ANGIOSPERMAE—DICOTYLEDONES

545. L. usitatissimum L. (Sprengel, ‘Entd. Geh.,’ p. 175; Hildebrand, ‘ Die
Geschlechtsvert. b. d. Pf.” p. 75; Herm. Miiller, ‘Fertilisation, p. 148, ‘ Weit.
Beob.,’ II, p. 219.)—The bright blue flower, according to Hermann Miiller, possesses
a mechanism which completely agrees with that of the last species. But being
More conspicuous, insect-visits are more numerous, so that crossing takes place
more frequently. Hildebrand has proved that when automatic self-pollination occurs
it is effective.

Vistrors.—Sprengel saw a humble-bee; Herm. Miiller observed bees—Apis,
Halictus cylindricus #.— and Lepidoptera — Plusia gamma JZ., Pieris rapae L.;
MacLeod noticed one humble-bee and a Muscid in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 239). Cf. L. grandiflorum.

546. L. tenuifolium L.—Kerner says that the slightly odorous rose-red
flowers of this species fall off as early as the second day of anthesis. According
to Kirchner (‘Beitrage,’ pp. 29-30), who studied the mechanism of the flowers
in Wallis, they agree with those of L. usitatissimum as regards homogamy, and
the secretion and concealment of nectar, but from the relative positions of the
stigmas and anthers—which mature simultaneously—the regular occurrence of cross-
pollination may be inferred. Owing to the conspicuousness of the flowers, which
have a diameter of 22 mm., this is sure to take place. The five styles which spring
from the ovary diverge, while the five filaments—united together below—stand erect
between the styles, and at the same level as the stigmas, which are outside them, and
at a distance of 3 mm. Even if towards the end of anthesis the petals and styles
come into contact, automatic self-pollination does not as a rule take place, for the
stigmas are at a higher level than the anthers in the closed flower.

547. L. Lewisii L.— Planchon states that each plant of this species bears three
different kinds of flower, possessing medium, long, and short styles, respectively.

548. L. austriacum L.—

Visirors.—Friese observed Andrena braunsiana /7zese in Hungary; v. Dalla
Torre saw the bee Osmia leucomelaena X. é and 9 in the Innsbruck Botanic Garden,
and Schletterer noticed it in the Tyrol.

549. L. grandiflorum Desf—This species was the starting-point of Darwin’s
research ‘On the Existence of Two Forms and their Reciprocal Sexual Relations in
several species of the genus Linum’ (1863). He showed that fertility was greatest
when the long-styled variety was crossed with pollen from the short-styled one,
and vice versa. Darwin’s investigations also demonstrated that the short-styled
variety was more fertile when self-fertilized than the long-styled one, which was
almost self-sterile. When both kinds of pollen were put upon one kind of stigma, as
a rule only that germinated which was derived from the opposite sort of flower,

while that from the same sort either failed to germinate, or did so only to a
slight extent.

Visitors.-— Frey-Gessner observed two bees—Nomia diversipes Z/r. and
Systropha curricornis Scop.—in Switzerland. Plateau noticed that small Syrphidae

went straight from the red flowers of L. grandiflorum to the blue ones of L.
usitatissimum.

MALPIGHIACEAE 217

550. L. perenne L.—Darwin found (op. cit.) that three-quarters of the legiti-
mately pollinated flowers of both varieties were completely fertile, while illegitimately
pollinated flowers of the long- and short-styled varieties were, respectively, completely
infertile and almost completely fertile. According to Hildebrand’s investigations
(1864), the short-styled variety is completely infertile both with its own pollen
and with that from other flowers of the same plant, and even with pollen from
other short-styled plants, while, on the other hand, it is completely fertile with
pollen from long-styled flowers.

162. Radiola Dill.
Minute white flowers, probably with concealed nectar.
551. R. linoides Gmel.—MacLeod (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
P- 379) states that in the flowers of this species the four anthers come into contact
with the four stigmas, so that automatic self-pollination is inevitable. Owing to the
smallness of the flowers, nectaries could not be recognized.
Vistrors.—Herm. Miiller observed several minute flies.

XXIV. ORDER MWALPIGHIACEAE JUSS.
163. Camarea St. Hil. and 164. Janusia A. Juss.

In species of these genera, there are cleistogamous as well as chasmogamous
flowers.
165. Aspicarpa Rich. 7
552. A. urens Rich—H. von Moh! states (Bot. Ztg., Leipzig, xxi, 1863) that
this species bears cleistogamous flowers. The same is true for the species of
the genus

166, Gaudichaudia H. B. et K. (Kuhn, Bot. Ztg., Leipzig, xxv, 1867.)

167. Bunchosia Rich.
553. B. Gaudichaudiana A. Juss. (Delpino, Bot. Ztg., Leipzig, xxviii, 1870 ;
Hildebrand, op. cit., xxviii, 1870.)—
Visitors.—These are bees—Tetrapodia, Epicharis—which cover their ventral
surfaces with pollen, and transfer it to the stigmas of other flowers.

168. Coriaria Niss.
554. C. myrtifolia L.—According to Hildebrand (Bot. Ztg., Leipzig, xxvii,
1869, pp. 494-5), the flowers of this species are markedly protandrous, being
purely male in their first stage.

169. Hiptage Gaertn.

555. H. Madablota Gaertn.—Lanza (‘ Note di biol. fior.,’ 1894) states that the
flowers of this species are protogynous in the Palermo Botanic Garden. There is
a single nectary between the two upper petals turned towards the floral axis. The
flower mechanism resembles that of Aesculus.

Visrrors.—The pollen is transferred by bees, as in the horse-chestnut.

218 ANGIOSPERMAE—DICOTYLEDONES

170. Cratoxylon Blume.

556. C. formosum Bentham et Hook. f.— Darwin (‘Different Forms of
Flowers’) states that the species possess dimorphous flowers.

XXV. ORDER GERANIACEAE DC.

1. Trrpe GERANIEAE.

Lirerature.—Herm. Miiller, ‘ Fertilisation,’ pp. 149-60; Knuth, ‘Grundriss d.
Bliitenbiol.,’ p. 37.

The flowers are usually bright in colour, and very frequently red. They differ
greatly in size in the various species, and with increasing inconspicuousness, there is
proportionate diminution in the number of insect visitors, but increased probability
of automatic self-pollination. In most species nectar is secreted by the outer sides of
the bases of the five external stamens. Almost all the flowers belong to class C
as regards concealment of nectar, but certain species may be placed in class H, not
because the nectar is more deeply situated, but because the flowers hang in such
a way that only very expert visitors can reach it. The flowers are mostly pro-
tandrous, rarely homogamous or protogynous. The greater the probability of
insect-visits the more marked is the dichogamy. Besides hermaphrodite flowers,
there are smaller female ones in some species, distributed gynodioeciously, as a rule.

171. Geranium L.

Flowers protandrous, rarely protogynous (G. dissectum and G. pusillum), with
concealed nectar secreted by the outside of the bases of the five inner stamens.
Some of them are bee flowers. Jordan states that in the erect flowers belonging
to class C, insect visitors alight on the petals, while in the case of the pendulous
flowers—belonging to class H—they settle on the stamens and pistil.

557. G. palustre L. (Sprengel, ‘Entd. Geh., pp. 335-7; Herm. Miller,
‘Fertilisation,’ p. 149; Schulz, ‘ Beitrige,’ I, p. 28; Kirchner, ‘Flora v. Stuttgart,’
p- 335; Knuth, ‘Bloemenbiol. Bijdragen.’)—The markedly protandrous flowers of
this species spread out their purple-red petals with paler claws to form a surface
30-40 mm. in diameter, which is turned to the sun. The nectar-guides are dark
lines on the corolla, converging to the middle of the flower. Nectar is secreted
in abundance by five glands on the outer side of the bases of the five inner stamens.
Hairs on the bases of the petals serve to protect the nectar, and prevent the access of
rain-drops. The five inner stamens ripen first, and then the five outer ones, and
only when these have shed their pollen do the stigmas—that have so far been closely
apposed—spread out and project from the middle of the flower. After dehiscing
all the stamens bend outwards so far that automatic self-pollination is impossible.

Besides hermaphrodite flowers, Schulz has observed female ones distributed
gynodioeciously, or more frequently gynomonoeciously. He also distinguished
between large- and small-flowered varieties of the hermaphrodite flowers.

VistTors.— Herm. Miiller (H. M.) in Westphalia, and myself (Kn.) in Schleswig-
Holstein, have observed the following.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp., skg. (H.M.). (2) Syrphidae:
2. Eristalis tenax Z., skg. (H.M.); 3. Helophilus pendulus Z., do. (Kn.); 4. Meli-

GERANIACEAE 219

threptus scriptus Z., do. (H.M.); 5. Platycheirus peltatus J/g, do. (H.M.); 6.
Rhingia rostrata Z., do. (Kn.). B. Hymenoptera. Ajidae: 7. Andrena dorsata
K. 3, skg. (H.M.); 8. A. fulvicrus XK. 8, do. (H.M.); 9. Apis mellifica Z. ¥, do.
(Kn.); 10. Halictus albipes /., do. (H.M.); 11. H. cylindricus #. 8, do. (H. M,);
12. H. flavipes F. 6, do. (H.M.); 13. H. longulus Sm. 9, do. (H.M.); 14. H. niti-
diusculus A. 9 and 4, do. (H. M.); 15. H. zonulus Sm. 6, do. (H. M.); 16. Prosopis
communis /Vy/. 9, do. (H. M.). C. Lepidoptera. Rfopalocera: 17. Pieris rapae
L., skg. (Kn.).
Loew saw the following in the Berlin Botanic Garden.—

A. Diptera. (2) Muscidae: 1. Anthomyia sp., skg.; 2. Lucilia caesar L.
(6) Syrphidae: 3. Eristalis nemorum Z., skg.; 4. E. tenax Z.; 5. Syritta pipiens Z.
B. Hymenoptera. Apidae: 6. Apis mellifica Z. ¥, skg.; 7. Chelostoma nigricorne
Wyi.%, do.; 8. Coelioxys rufescens Lep. 9, do.; 9. Halictus cylindricus F. 4, do.; 10.
H. nitidiusculus XK. 9, do.; 11. H. rubicundus Chr. 9, do.; 12. H. sexnotatus X. 9,
do. ; 13. H. villosulus A. , do.; 14. Prosopis communis Wy. 4, do.

Fic. 66. Gerantum sylvaticum, L. (after Herm. Miller). 4. Petals from different plants of the
large-flowered variety, natural size. At their bases are the hairs serving as a nectar-cover, which led
Sprengel to make his investigations. 2B. Petals from different plants of the small-flowered variety,
natural size. C. Stamens and pistil of a small female flower. DD. The same parts of a large flower at
the end of the first (male) stage: the anthers have all shed their pollen, the stigmas are still close
together. £. The same in the second (female) stage. F. Sexual organs of a homogamous flower ;
m, nectaries; s¢, stigmas. (C—FX7.)

558. G. sylvaticum L. (Sprengel, ‘Entd. Geh.,’ p. 1; Axell, ‘Om Anordning.
for Fanerog. Vaxt. Befrukt.,’ p. 36 ; Herm. Miiller, ‘Alpenblumen,’ pp. 174-8; Schulz,
‘Beitrage, I, pp. 26-7; Loew, ‘ Bliitenbiol. Floristik, p. 398; Kirchner, ‘Flora v.
Stuttgart,’ pp. 335-6.)—This species served as the starting-point for Christian Konrad
Sprengel’s classical investigations. The flower mechanism agrees with that of the
last species; here again self-pollination is excluded by marked protandry. Besides
large flowers with a diameter of about 27 mm., Lindman observed small ones with
a diameter of only 15 mm. Schulz states that there are smaller female flowers,
in which the stamens are quite short and the anthers vestigial. Among the plants

220 ANGIOSPERMAE—DICOTYLEDONES

with hermaphrodite flowers there are rarely some with homogamous flowers, in
which automatic self-pollination is possible. In the South Tyrol Schulz observed
large-flowered male varieties, in which the branches of the style never spread out.
Unisexual flowers are gynodioecious or more rarely gynomonoecious, as well as
androdioecious and andromonoecious. In the Scandinavian highlands, Ekstam
observed female flowers with vestigial stamens, and small male flowers, in addition
to the protandrous hermaphrodite ones.

Vis1rors.—Schulz states that many insects visit the flowers in Central Germany.
Herm. Miiller saw 8 beetles, 21 Diptera, 24 Neuroptera, and 20 Lepidoptera,
in the Alps.

In the Berlin Botanic Garden Loew observed two bees, skg., i.e. Apis mellifica
Z£.% and Bombus hortorum LZ. 9; also Prosopis communis JVy/. 9, skg., in the variety
robustum. We further noticed two Syrphids—Platycheirus manicatus Mg. 6 and
Syrphus annulipes Ze/f—and a bee (Andrena sp.) in Switzerland.

Schneider (Troms¢ Mus. Aarsh., 1894) saw Bombus hypnorum ZL. and B. kirby-
ellus Curt. $ and 9, in Arctic Norway, and Lindman observed Diptera and humble-
bees on the Dovrefjeld.

In Dumfriesshire, Apis, freq., 2 humble-bees, 2 short-tongued bees, 3 Empidae,
5 Muscidae, and 2 hover-flies have been recorded (Scott-Elliot, ‘ Flora of Dumfries-
shire,’ p. 34).

Glandular hairs on the stem keep off creeping insects from below.

559. G. pratense L. (Hildebrand, ‘D. Geschlechts-Vert. b. d. Pf.” p. 27;
Herm. Miiller, ‘ Fertilisation,’ p. 150, ‘ Weit. Beob., II, p. 167; Schulz, ‘ Beitrage,’ I,
pp. 27-8; Kirchner, ‘Flora v. Stuttgart,’ p. 336.)—The mechanism of the markedly
protandrous flowers of this species agrees essentially with that of G. palustre. The
stamens at first lie on the petals, but become erect when the anthers dehisce, pass to
the middle of the flower, and after the pollen has been shed return to their former
position. Hildebrand has established by experiments that the stigmas are usually
still immature at the time when the anthers are covered with pollen, but become
mature when the dehisced stamens bend back. They cease to be receptive when
the petals fall away.

Schulz says that the flowers vary considerably in size. In addition to herma-
phrodite flowers, he also observed female ones, distributed gynodioeciously or
gynomonoeciously.

Warnstorf describes the pollen-grains as white, spherical, coarsely tuberculated,
and 100 w in diameter.

This species also is protected against creeping insects by the sticky nature of
the stem.

Vistrors.—The only one I have seen is the honey-bee, skg. Hermann Miiller
observed the following in Westphalia and Thuringia.—

A. Coleoptera. Curculionidae: 1. Coeliodes geranii Pays., skg.(?); 2. Miarus
campanulae Z., do.(?). B. Diptera. (a) Stratomyzdae: 3. Nemotelus pantherinus L.
(4) Syrphidae: 4. Melithreptus pictus M/g., po-dvg. C. Hymenoptera. Apidae:
5. Andrena coitana K. 9 and @, skg.; 6. A. gwynana K. 9, do.; 7. Apis mellifica Z.
§, very freq., skg.; 8. Chelostoma campanularum KX”. 9 and 4, freq., skg.; 9. C. nigri-
corne L, 9 and 6, very freq., skg.; 10. Coelioxys conoidea Ji/. 8, skg.; 11. C. elongata
Lep., do.; 12. C. quadridentata Z., do.; 13. C. rufescens Lep. 9 and 4, do.; 14.

GERANIACEAE 221

Halictus albipes F. 5, do.; 15. H. cylindricus /. 4, do.; 16. H. leucozonius X. 9, do. ;
17. H. lucidulus Schenck 9, po-cltg.; 18. H. maculatus Sm. 4, skg. (Thuringia) ;
19. Heriades truncorum Z., do.; 20. Osmia fulviventris /. 9, do.; 21. O. rufa LZ. 9,
do.; 22. Prosopis hyalinata Sm. 9, do.; 23. Stelis aterrima Pz. 9 and 6, do.; 24. S.
breviuscula Nyl, @ and 4, do.; 25. S. minuta Zep. $, do.; 26. S. phaeoptera A. 9 and
$,do. D. Lepidoptera. Rhopalocera: 27. Pieris napi Z., skg. |

In Dumfriesshire, Apis (stealing nectar), 2 humble-bees, a short-tongued bee,
a Muscid, and a hover-fly have been recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’
P- 38).

Loew observed the following bees in the Berlin Botanic Garden.—

1. Apis mellifica Z. ¥, skg. (also on the variety forum album; 2. Chelostoma
nigricorne WVyl. $, do.; 3. Coelioxys elongata Zep. 9, do.; 4. Megachile argentata
F. 3, do.; 5. M. ericetorum Zep. 6, skg.

560. G. argenteum L.—According to Kerner (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, I,
Pp. 311), the flowers of this species are as markedly protandrous as those of G.
pratense and G. sylvaticum, self-pollination being excluded.

561. G. sanguineum L. (Herm. Miller, ‘Fertilisation,’ p. 132, ‘ Weit. Beob.,’
II, p. 217, ‘Alpenblumen,’ p. 174; Schulz, ‘ Beitrage,’ II, p. 56; Knuth, ‘ Bloemen-
biol. Bijdragen.’)—Hermann Miiller describes the purple-red flowers of this species
as protandrous, but capable of automatic self-pollination when insect-visits fail. This
arrangement is necessary owing to the shady habitat. When the flower opens,
the five inner stamens become erect, and the anthers—which dehisce above and
externally—project beyond the still immature stigmas. Meanwhile the five outer
stamens curve downwards. On the following day, they too become erect and their
anthers dehisce. The stigmas begin to spread out one or two days later, and
elongate so as to reach the same level as the anthers, so that if the pollen has
not already been removed by insects, either cross- or self-pollination may result
from their subsequent visits. The latter must take place automatically should insect-
visits fail.

Schulz occasionally observed female flowers, distributed gynomonoeciously or
more commonly gynodioeciously. On the Dovrefjeld, the ordinary flowers are quite
as large as in Central Germany, but smaller ones—hermaphrodite, male, and female—-
also occur.

Vistrors.—Lindman observed flies and humble-bees on the Dovrefjeld. In
Schleswig-Holstein I only noticed the honey-bee, skg.

In Westphalia and Thuringia Herm. Miiller did not observe a large number
of visitors. ‘They were chiefly flies and bees, which are the actual pollinators.
These sometimes alight on the middle of the flower, or may use a petal as a platform
from which to suck. He gives the following list—

A. Coleoptera. Curculionidae: 1. Coeliodes geranii Payk., skg.(?); 2. Miarus
graminis Schénh. B. Diptera. Syrphidae: 3. Merodon aeneus AZg., freq., skg.;
4. Pelecocera tricincta Mg., po-dvg.; 5. Pipiza sp., do.; 6. Rhingia rostrata Z., skg.
C. Hymenoptera. (a) Apidae: 7. Bombus pratorum ZL. ¥, po-cltg.; 8. Halictus
maculatus Sm. 9, nect-lkg.; 9. H. sexnotatus A. 9, do.; 10. Prosopis sp., skg.
(6) Sphegidae: 11. Oxybelus sp., skg. (c) Zenthredinidae: 12. Megalodontes cepha-
lotes #., very freq., skg. D. Lepidoptera. Sphingidae: 13. Ino globulariae
Hin., skg.

222 ANGIOSPERMAE—DICOTYLEDONES

Von Fricken saw the small beetle Trachys nana Aés/., very infrequent, in
Westphalia. Von Dalla Torre and Schletterer record the cuckoo-bee Nomada
guttulata Schenck $ for the Tyrol.

Herm. Miiller saw 2 Hymenoptera in the Alps, and MacLeod noticed 2 Hymeno-
ptera, a Bombylius, and a Muscid in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii,
1891, p. 402).

In Dumfriesshire, Apis, a humble-bee, 2 short-tongued bees, and several Diptera
have been recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 37).

Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Helophilus pendulus Z. B. Hymenoptera. (ca)
Apidae: 2. Halictus cylindricus /. $6, skg. (0) Sphegidae: 3. Cerceris variabilis Schr.
9; 4. Oxybelus sericatus Gerst. 9.

562. G. pyrenaicum. (Herm. Miller, ‘Fertilisation,’ p. 151, ‘ Alpenblumen,’
pp. 173-4; Schulz, ‘ Beitrage,’ II, p. 185; Knuth, ‘Bloemenbiol. Bijdragen.’)— For
Central Germany, Hermann Miiller describes the flowers of this species as agreeing
with those of G. sanguineum. A. Schulz states that the female flowers are distributed
gynodioeciously, or much more rarely gynomonoeciously.

vi
sy

vat

FIG. 67. Gerantum pyrenatcum, L. (after Herm. Miller). (1) Stamens and pistil before the beginning
of the first stage: all the anthers are still immature, and the undeveloped stigmas are in the middle
of them. (2) The same in the first half of the first (male) stage: the outer stamens are erect, and their
anthers are covered with pollen. (3) The same in the second part of the same stage. (4) The same in
the second (hermaphrodite) stage: all the anthers are covered with pollen, and the stigmas are spread out.
a) and a’, anthers of the outer and inner stamens; sé, stigmas; #, nectaries.

In the Alps, Herm. Miiller (‘ Alpenblumen,’ pp. 173-4) observed a variety with
lilac petals, which spread out into a flat surface, and with stamens curving strongly
outwards before dehiscence. The styles do not diverge till later on, so that automatic
self-pollination is impossible.

Visitors.—Herm. Miiller observed 5 bees, nect-skg., 2 Syrphids, skg., and a
Lepidopterid upon this Alpine form. I only saw the honey-bee in Schleswig-
Holstein. Borgstette noticed the following—mostly bees and flies—in Central
Germany.—

A. Coleoptera. (a) Cuistelidae: 1. Cistela murina Z. (6) Dermestidae: 2.

Byturus fumatus Z. (c) Zelephoridae: 3. Malachius aeneus Z. B. Diptera. (qc)
Muscidae: 4. Echinomyia fera Z.; 5. Scatophaga stercoraria Z. (6) Syrphidae:

GERANIACEAE 223

6. Ascia podagrica #.; 7. Chrysotoxum bicinctum Z.; 8. Helophilus floreus Z.;
9. Melithreptus pictus A/g.; 10. M. taeniatus Mg.; 11. Pelecocera tricincta Mg.;
12. Rhingia rostrata Z.; 13. Syrphus balteatus Deg.; 14.S.pyrastriZ.; 15. S. ribesii
L.: allskg. C. Hymenoptera. (a) Aprdae: 16. Andrena dorsata K.9; 17. A.
fulvago Chr. 9; 18. A. gwynana K.9 and$; 19. A. parvula K. 9; 20. Chelostoma
nigricorne L.9; 21. Halictus cylindricus F. 9; 22. H. maculatus Sm.9; 23. H.
smeathmanellus K. 9; 24. Osmia fusca Chr. 9; 25. Sphecodes gibbus Z. 9: all skg.
(2) Sphegidae: 26. Ammophila sabulosa Z. (c) Vespidae: 27. Odynerus spinipes Z.

MacLeod observed 5 Hymenoptera, 2 species of Bombylius, 2 species of Empis,
and 2 Muscidae, in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 401).
Loew saw the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp., skg. (6) Syrphidae: 2. Syrphus
pyrastri Z., hovering for a long time in front of a flower, and then skg. B. Hymeno-
ptera. Apzdac: 3. Apis mellifica Z. Y, skg.; 4. Bombus lapidarius Z. ¥, do. ;
5- Stelis phaeoptera . 9, do.

563. G. cinereum Cav.—MacLeod states that this Pyrenean species possesses
thoroughly protandrous flowers (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 403-5).
Their diameter—when fully expanded—is 3-34 cm. The petals are dark violet
with numerous violet veins. The stamens are at first curved outwards, with their
anthers still closed. They then become erect, and the pollen is shed. After dehi-
scence, the stamens again curve outwards, and the stigmas unfold. Self-pollination
is consequently prevented. Besides hermaphrodite flowers, MacLeod observed smaller
female ones, distributed gynodioeciously. Their filaments, though devoid of anthers,
executed the same movements as those of hermaphrodite blossoms. They even
retain protandry, for the stigmas remain apposed until the flower has been open
for some time. This arrangement, as MacLeod points out, is not merely useless
but even injurious. For it exposes the stigmas and nectar for a while to wind
and rain.

Vistrors.—MacLeod observed bees (Bombus), Lepidoptera (Pyralidae), Syrphidae
(Eristalis), Empidae, and, more particularly, Muscidae (species of Anthomyia).

564. G. phaeum L. (Ricca, ‘Oss. sulla fecondaz. incroc. d. veget. alp. e
subalp.’; MacLeod, ‘ Pyreneénbl.,’ p. 130; Schulz, ‘ Beitrage,’ II, p. 184; Kirchner,
‘Flora v. Stuttgart,’ pp. 336-7; Errera, ‘Geranium phaeum’; Knuth, ‘ Bloemenbiol.
Bijdragen.’)—According to MacLeod, the dark reddish-brown or violet blossoms
of this species are bee flowers. It is true that the nectar is not more deeply placed
than in the other large-flowered species of geranium, but the blossoms are vertical, or
even slightly pendulous. In consequence of this position, they can only be sucked
by very skilful visitors, such as bees. Kirchner says that they are markedly pro-
tandrous. At the beginning of anthesis, the petals spread out to form a surface
22 mm. across, but soon recurve so much that the diameter is reduced to 18 mm.,
and the stamens and stigmas successively project freely. The nectaries are situated,
as usual, outside the bases of the stamens which alternate with the petals. They
are from the first protected from rain by the corolla, for though the claws of the
petals are vertical the lower part of the limbs arches over the nectary to form a
roof. The stamens mature successively, the inner ones first. At first they all curve
towards the base of the flower, but when the anthers dehisce, the filaments straighten
out, and project horizontally from the flower for about ro mm, The anthers fall

224 ANGIOSPERMAE—DICOTYLEDONES

off after they have shed their pollen, and the filaments bend back again to their
original position. After all the anthers have dropped off, the styles—which at
first project only about 7 mm. from the flower, and are close together—gradually
grow to a length of ro-11 mm. and spread out, the five stigmatic branches being
placed at the level which the anthers first occupied. After being pollinated, they
come together again. Besides these distinctly protandrous flowers, Schulz—on
garden plants—observed female ones, distributed gynodioeciously.

Visitors.— Kirchner noticed numerous honey-bees at Hohenheim (Wurtem-
berg). I myself observed the same visitors at Kiel; they hang upon the stamens
or styles in the manner described by Jordan.

MacLeod saw 4 humble-bees in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii,
1891, pp. 405-6). Darwin mentions humble-bees; Ricca humble-bees and Apis;
Plateau Eucera longicornis Z.; and Errera 29 Hymenoptera.

Loew observed the following bees in the Berlin Botanic Garden.—

1. Apis mellifica Z. ¥ (also the variety Zvzdum), very freq., skg.; 2. Bombus
hortorum L. 9, skg.; 3. B. lapidarius Z. ¥, do. ; 4. B. rajellus X. 9, do.; 5. Coelioxys
elongata Zep. 9, do.; 6. Halictus albipes /. 9, do.

565. G. macrorrhizum L.— This species also, according to Jordan, has
pendulous flowers that are visited in the same way as those of the last species.
Hildebrand (Bot. Ztg., Leipzig, xxvii, 1869, pp. 479-81) states that it is equally
protandrous ; at first purely female flowers appear.

566. G. dissectum Z. (Herm. Miiller, ‘ Fertilisation,’ p. 156, ‘ Weit. Beob.,’
II, pp. 217-18; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 233; Kirchner,
‘Flora v. Stuttgart,’ p. 338; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—This
species, unlike those hitherto dealt with, bears protogynous flowers, with very
persistent stigmas. In the sunshine they open only far enough to be funnel-shaped,
with an entrance 6-8 mm. The stigmas are mature, and their branches expanded
when the flower opens, while the anthers which closely surround them are still unripe.
Later on the anthers dehisce successively, covering the stigmas with pollen. Hermann
Miiller’s researches show that this automatic self-pollination is effective. Insects
visiting the flowers may effect self- as well as cross-pollination, but the visitors
are very few.

Warnstorf describes the flowers as homogamous and autogamous; the stigmatic
papillae are mature at the time when the anthers dehisce. The latter are blue in
colour, and closely applied to the stigmatic branches, so that self-pollination is
inevitable. The pollen-grains are bluish-white, spherical, densely tuberculated,
cohering, with an average diameter of 63 p.

Vistrors.—Herm. Miiller observed in Thuringia only a bee—Andrena gwynana
K. 9 and 4, skg.—and 2 Diptera—Occemyia atra #. and Merodon aeneus JZz.,
both skg.

Schletterer saw the beautiful bee Osmia versicolor Z/r., and the saw-fly Amasis
laeta /., at Pola.

MacLeod noticed a Lepidopterid and a fly in the Pyrenees (Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 402).

In Dumfriesshire a hover-fly and 2 Muscids have been recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 39).

GERANIACEAE 225

In this species again the access of creeping insects to the flowers is prevented
by the glandular character of the calyx.

567. G. lucidum L.—Kerner states that the small flowers of this species are
open from 7 a.m. to 8 p.m. Besides the protogynous hermaphrodite flowers—capable
of self-pollination—there are female ones distributed gynomonoeciously.

Visitors. —In Dumfriesshire 6 hover-flies have been recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 38).

568. G. columbinum L.—Each of the bright rose-coloured petals of this
species is marked with three dark veins that serve as nectar-guides. Kerner says
that the flowers are open from 8 a.m. till 5 p.m. Their mechanism seems to vary,
for while Kerner describes them as protogynous and autogamous, Schulz (‘ Beitrage,’
II, p. 185) asserts that they are slightly protandrous. According to the latter there
are female flowers as well as hermaphrodite ones, distributed gynodioeciously and
gynomonoeciously.

569. G. rotundifolium L.—The flowers of this species studied by A. Schulz
(‘ Beitrage,’ II, p. 56) at Bozen, were red, and measured about 5-7 mm. across when
expanded. Soon after anthesis the outer anthers mature, and the inner anthers
do not dehisce as a rule until these have shed their pollen. The stigmas are at
the same level as the anthers, and mature simultaneously, so that, failing insect-visits,
automatic self-pollination is inevitable. Owing to the inconspicuousness of the

Fic. 68. Geranium molle, L. (after Herm. Muller), 1-5. Stamens and pistil in successive stages of
development. a!, outer stamens, with nectaries at their bases; a’, inner stamens; s/, stigmas.

flowers, and the small amount of nectar secreted, there are but few visitors. These
may effect either self- or cross-pollination.

Protection against creeping animals is afforded by glandular hairs.

Visttors.—Schulz saw occasional Diptera—mostly hover-flies, e.g. Rhingia—
and 2 Lepidoptera (Lycaena).

F. F. Kohl observed the wasp Odynerus tarsatus Sawss. in the Tyrol.

570. G. molle L. (Sprengel, ‘Entd. Geh.,’ p. 338; Herm. Miiller, ‘ lertilisa-
tion, pp. 153-4, ‘Weit. Beob.,’ II, p. 217; Kirchner, ‘ Flora v. Stuttgart,’ p. 340;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins., p. 51, ‘Bliitenbiol. Notizen’; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 230-33 Loew, ‘ Bliitenbiol. Floristik,”
p. 398; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The rose-coloured
flowers of this species are slightly protandrous. When the flower opens, the stigmatic
branches are apposed, so that their receptive parts are covered. The still immature
anthers are directed outwards. The outer stamens now bend inwards in succession,
their anthers resting on the pistil, and there dehiscing. As the top of the stigmatic

DAVIS. Ul Q

226 ANGIOSPERMAE—DICOTYLEDONES

branches are still unreceptive, the flower is purely male at this stage. But, even
before all the five outer anthers have dehisced, these branches begin to diverge.
The outer stamens next bend towards the middle of the flower, and their anthers
dehisce. In the final stage, the anthers stand between and somewhat above the
stigmas, so that insect visitors may effect either cross- or self-pollination. In the
absence of such visitors, automatic self-pollination necessarily occurs.

Warnstorf describes the flowers as homogamous and autogamous, and states
that the pollen-grains are yellowish, spherical, covered with reticulately arranged
tubercles, and about 63 » in diameter.

At Blankenberge, besides the protandrous hermaphrodite flowers, MacLeod
observed female ones with anthers devoid of pollen, and also intermediate forms,
in which only some of the stamens were sterile.

Visirors.—Hermann Miiller observed the following.—
A. Diptera. (a) Conopidae: 1. Dalmannia punctata /, skg.; 2. Myopa

testacea Z., do. (0) Afluscidae: 3. Scatophaga merdaria /., skg. “(c) Syrphidae :
4. Ascia podagrica /”, very freq., skg.; 5. Helophilus pendulus Z., skg.; 6. Rhingia
rostrata LZ., do.; 7. Syritta pipiens Z., do. B. Hymenoptera. (a) Apidae: 8.
Andrena gwynana K. 9, skg.; 9. Apis mellifica Z. Y, do.; 10. Chelostoma campanu-
larum 4. 9, do. (Buddeberg); 11. Halictus nitidus Schenck 9, do.; 12. H. sp. do.

On June 5, 1897, I saw the following in Helgoland.—r. Eucera difficilis (Duf.)
Per. %, skg.; 2. Lucilia caesar Z., do.; 3. Syritta pipiens Z., hovering for some time
before the flower, then skg. and po-dvg.; 4. Andrena labialis A’ 4, skg.

H. de Vries in the Netherlands (Ned. Kruidk. Arch., 2. ser., 2. deel, Nijmegen,
1875) noticed a humble-bee, Bombus terrester 7. 9; and MacLeod observed 6 bees,
4 hover-flies, 2 Muscidae, and a Lepidopterid (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 232-3).

Loew (‘Beitriige,’ p. 60) saw the bee Melithreptus
menthastri Z. in Switzerland; while at Pola Schletterer
observed a saw-fly, Cladius pectinicornis Fourcr., and the
following bees:—r1. Andrena dubitata Schenck; 2. A. fla-
vipes Pz.; 3. A. parvula K.; 4. Halictus calceatus Scop. ;
re oe a 5 Osmia versicolor Ltr.

Stamens and pistil of a just In Dumfriesshire, a short-tongued bee, a hover-fly,
ee Hower. a anthers: sand several Muscidae have been recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 39).

571. G. pusillum L. (Herm. Miiller, ‘Fertilisation, pp. 154-5, ‘Weit. Beob.,
II, p. 217; Kirchner, ‘Flora v. Stuttgart,’ p. 339; Knuth, ‘ LI. u. Insekt. a. d. nordfr.
Ins.,’ pp. 50-1.)—As Hermann Miller explains, the small lilac flowers of this plant
are even less conspicuous than those of the last species, so that insect visitors
are still less numerous, and automatic self-pollination constantly takes place before
the end of anthesis. Although the flowers of the two species are externally very
similar, their mechanisms are quite different. G. pusillum is protogynous, with
persistent stigmas. Only the five inner stamens—those with nectaries at their
bases—bear anthers. When the flower opens, the stigmatic branches have already
half spread out, while the anthers lying between them are still immature. On
dehiscence of the latter, the stigmatic branches diverge still more, and the stamens

GERANIACEAE 227

bend towards the middle of the flower, so that the pollen-covered anthers lie above
them. It follows that automatic self-pollination is effected by the fall of pollen,
should insect-visits fail. Even after the anthers have dropped off, the stigmas remain
receptive.

Visirors.—Hermann Miiller observed only a few Syrphidae.

572. G. Robertianum L. (Sprengel, ‘Entd. Geh., p. 337; Herm. Miiller,
‘Fertilisation,’ pp. 156-7, ‘ Weit. Beob.,’ II, p.218; MacLeod, Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, pp. 229-30; Kirchner, ‘Flora v. Stuttgart,’ pp. 340-1; Schulz,
‘Beitrage,’ II, pp. 57-8; Knuth, ‘Bloemenbiol. Bijdragen.’)— Hermann Miiller
describes the flowers of this species as slightly protandrous. The petals are rose-
coloured, with three paler streaks, and their claws remain erect, so that the flower
does not open widely. The nectar collects in the slightly. excavated bases of the
sepals, and a proboscis 7 mm. long is required to reach it. When the flower opens
the five inner stamens are in the middle of the flower, their anthers dehiscing at
a somewhat higher level than the still apposed stigmatic branches, which consequently
get covered with pollen. The five outer stamens are curved widely outwards.
While the anthers of the five inner stamens still retain pollen, the stigmatic branches
elongate and spread out above them. When, however, these stamens have shed
their pollen, the five outer ones move towards the middle of the flower, and surround
the styles. Insect-visits effect cross-pollination, at first owing to protandry, and later
on because the fully mature stigmas project above the pollen-covered anthers. Self-
pollination, however, is not excluded.

According to Schulz, the hermaphrodite flowers are occasionally homogamous.
Besides these, unisexual ones occur, distributed gynodioeciously and gynomonoe-
ciously, or androdioeciously and andromonoeciously.

Warnstorf describes the pollen-grains as large, spherical, closely tuberculated,
and opaque, measuring 70 p in diameter.

Visitors.—Hermann Miiller (H. M.), Buddeberg (Budd.), and myself (Kn.)
have observed the following in Central and North Germany.—-

A. Coleoptera. (a) Staphylinidae: 1. Anthobium sp. (H.M.). (6) Zele-
phoridac: 2. Dasytes flavipes #., skg., and gnawing the petals. B. Diptera. ()
Empidae: 3. Empis sp., vainly searching for nectar (Budd.). (2) Syrphidae: 4. Rhingia
rostrata Z., freq., skg. and po-dvg. (H. M., Budd.). C. Hymenoptera. Apidae:
5. Andrena gwynana A. é (Budd.); 6. Bombus agrorum /”. ¥, skg. (H.M.); 7. B.
hortorum ZL. ¥, skg. persistently (H. M.); 8. B. lapidarius /., skg. (Kn.); 9. B. ter-
rester /.., do. (Kn.); 9. Chelostoma campanuiarum A’ 6, do. (Budd.); ro. C. nigricorne
Nyl. 8, do.(Budd.) ; 11. Halictus cylindricus /. 9, do. (FH. M., Thuringia); 12. Osmia
adunca Pz. 6, do. (Budd.); 13. O. rufa Z. 9, do. (Budd.). D. Lepidoptera. /o-
palocera: 14. Pieris napi Z., freq., skg. (H. M.).

Krieger noticed the following Agrdae at Leipzig.—1. Anthidium manicatum J. ;
2. Andrena gwynana X., rst generation; 3. Coelioxys rufescens Lep.; 4. Eriades
nigricornis My/.; 5. Osmia caerulescens Z. (=O. aenea L.); 6. O. solskyi A/or. ;
7. Stelis phaeoptera A’.

Hermann Miller further saw 4 humble-bees, 2 hover-flies, and 3 Lepidoptera
in the Alps.

Willis observed a Syrphid (Syrphus sp., skg.), a bee (Bombus agrorum /, freq.,
skg.), and a butterfly (Pieris napi Z., skg.), in the neighbourhood of the south coast
of Scotland (‘Fls. and Insects in Gt. Britain,’ Part I).

Q 2

228 ANGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire, 2 humble-bees, 2 Empidae, several Muscidae, a Lepidopterid,
and Meligethes have been recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 38).

MacLeod observed 4 bees, 4 Lepidoptera, and a Muscid in the Pyrenees (Bot.
Jaarb. Dodonaea, Ghent, ili, 1891, p. 402); and a humble-bee, an Empis, a wasp,
and a Lepidopterid in Flanders (op. cit., vi, 1894, p. 230).

573. G. rivulare Vill.—According to Briquet (‘ Etudes de biol. flor. dans les
Alpes occident.’), the flowers of this species are protandrous to such a degree, that
automatic self-pollination can only occur exceptionally. The petals are white, each
with red veins. The yellow anthers, after dehiscing, become extrorse, and assume
a violet colour. The bases of the filaments are beset with hairs, which serve to
protect the nectar.

Visitors.—These are Diptera, Hymenoptera, and butterflies (Kirchner).

According to the observations of Loew, in the Berlin Botanic Garden, the
following species of Geranium are visited as stated.—

574. G. albanum Bieb.; a Muscid, skg.

575. G. Arnottianum Steud.; the honey-bee, skg.

576. G. ibericum Cav.—

A. Diptera. Syrphidae: 1. Helophilus pendulus Z. B. Hymenoptera.
Apidae: 2. Apis mellifica LZ. ¥, skg.; 3. Bombus lapidarius Z. ¥, skg. (also, on the
var. platypefalum, Apis mellifica and Prosopis communis Vy/. 9).

577. G. Pseudosibiricum J. Meyer; the honey-bee, skg.

578. G. reflexum L.; do.

579. G. rubellum Moench.—Two bees—Coelioxys elongata Zep. 9, skg., and
Osmia aenea L. 9, do.

580. G. ruthenicum Uechtritz; a bee—Halictus cylindricus F. 4, skg.

581. G. sibiricum L.; 2 Muscid—Anthomyia sp.—skg.

582. G. striatum L.; bees—Apis and Halictus leucozonius Schr. é—skg.

172. Erodium L’Hérit.

Protandrous, homogamous, or protogynous flowers, with concealed nectar,
secreted as in the last genus. The petals are frequently dissimilar, the lower ones
being longer. Stamens ten, the five opposite the petals broader than the others,
and devoid of anthers, while those alternating with the petals bear anthers and
possess nectaries at their bases.

583. E. cicutarium L’Hérit. (Sprengel, ‘Entd. Geh., pp. 338-40; Herm.
Miiller, ‘Fertilisation, pp. 158-9; Ludwig, Kosmos, Leipzig, viii, 1881, pp. 357-
62, Irmischia, Sondershausen, ii, 1881, pp. 5-7, Bot. Centralbl., Cassel, viii, 1881,
pp. 87-8, D. bot. Monatschr., Arnstadt, i, 1883, pp. 5-7, Bot. Centralbl., Cassel,
xix, 1884, p.118; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 234-7;
Schulz, ‘ Beitrage,’ II, pp. 58-9, 185; Knuth, ‘BL. u. Insekt. a. d. nordfr. Ins.,’
Pp. 51-3, 152; Nirchner, ‘Flora v. Stuttgart,’ pp. 341-2; Loew, ‘Bliitenbiol.
Floristik,’ p. 212; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.) — Sprengel
has given an admirable account of the flower mechanism. F. Ludwig was the first to

GERANIACEAE 229

call attention to the differences between the two varieties of this species as regards
pollination. He distinguishes between them as follows.—

(2) Genuinum.—The petals of this, the common variety, are generally of a
uniform red, and equal in size, though the upper ones are sometimes rather shorter
and of a darker tint than the others. The nectaries are all equally developed,
as in Geranium. The flowers are homogamous or slightly protogynous. The three
upper anthers, while dehiscing, lie close to the stigmatic branches, as do the two
lower ones subsequently, so that automatic self-pollination is inevitable. This is
effected an hour after the flower opens, which is at 7 o’clock in the morning.
By midday the flowers have shed their petals. This variety is perfectly fertile
with its own pollen.

(2) Pimpinellifolium Willd.—This variety is markedly entomophilous. The
flowers are as a rule relatively large, and the two upper petals smaller, broader,
and of a brighter red than the three lower ones. The latter are elongated, and
serve as an alighting-place for visitors. Each upper petal generally has a dark
nectar-guide, but this may be absent, and some or all of the other petals may be
similarly provided. The deeper parts of the flower are at first so blocked up by
the stamens that no insects can penetrate into them. The upper sepal and petals
are so far remote from the upper stamens that the dark posterior nectaries become
visible, while the lower ones are almost hidden by the hairs of the petals. The
latter are distinctly smaller than the former, and secrete much less nectar. When
the flower opens the style is still short and undeveloped, and the anthers are at
some distance from it. First the upper anthers, and afterwards the lower ones,
dehisce introrsely. The stamens soon curve completely outwards, and usually shed
their anthers before the stigmatic branches open and expand, which is generally
on the second day of anthesis. Only occasionally, especially when the nectar-guides
are not very conspicuous, do the stamens bend back again to the stigma in such
a way as to make automatic self-pollination possible as a last resort; but this is
effective only to a slight extent. The petals usually fall off on the second day.
The nectar-guides are frequently wanting, or only slightly developed, in the smaller
female flowers that occur in addition to the hermaphrodite ones, either on the same
or on different stocks.

The variety prmpinelli/olium, however, does not always exhibit the characteristics
described by Ludwig'. Schulz, in various parts of Germany and the Tyrol, has
studied the varieties of Erodium cicutarium for several years. The chief variety
(genuinum), according to him, occurs—e.g. at Halle—in two different oecological
forms.

1. Flowers of a uniform red, in many cases quite radially symmetrical (actino-
morphous), and with a diameter of about 8 to 13 mm. Sometimes the two upper
petals are short and broad, sometimes they are also more deeply coloured than
the upper ones, and may exhibit one or several greyish-white patches, occasionally
streaked with red. Such flowers are almost always homogamous, less often slightly

* Ludwig described gevddzum from the neighbourhood of Greiz, and pimpinellifolium from
Schinalkalden, Schleusingen, &c.

230 ANGIOSPERMAE—DICOTYLEDONES

protandrous, very rarely protogynous. As a rule, self-fertilization alone takes place
in this form. Insect-visits are extremely rare, even when nectar-guides are present.

2. Flowers usually bilaterally symmetrical (zygomorphous) to a marked degree,
and very large, their diameter being 12-15 mm. Nectar-guides may be present,

and if so are clearly marked, moderately large, often nearly bisected by the mid-ribs,
and covered with numerous dark streaks and dots.

protandrous, and self-pollination is usually excluded.

Schulz says that, at Halle, the flowers of prmpinellifolium are almost always
larger than the average of those borne by genuznum, bilaterally symmetrical as a rule,
and possessed of nectar-guides, though more rarely they are radially symmetrical,
in which case the petals may be blotched or unblotched. They are almost always
protandrous, but homogamy may occur sporadically. In many cases the plant
is entirely dependent on cross-pollination.

In the south Tyrol, Schulz found only genucnum, mostly with unblotched homo-
gamous flowers, presenting perfect or almost perfect radial symmetry. Elsewhere
he also saw the large-flowered form of this variety growing together with pempinell-
folium; the flowers of both were visited by insects to the same extent. With a
pipette, Schulz carefully removed every trace of nectar from the nectaries of a number
of newly-opened flowers of pzmpznellifolium, and smeared the glands with shellac.

The flowers are markedly

Fic. 70. Variations in the nectar-guides on the petals of Evodtum ciculartum, L' Hert.
(From nature.)

In spite of their nectar-guides, such flowers were visited by only a few insects,
while neighbouring blossoms— both those of genurnum, and untouched ones of
pimpinellifolium—treceived, just as before, a considerable number of visits. Here
again, the odour of the nectar serves as the chief means of attracting the insects,
the nectar-guides being of secondary importance in this respect, except when very
brightly coloured, and therefore conspicuous by contrast.

At a time when unacquainted with Schulz’s work, I arrived at similar con-
clusions while studying the variation of the nectar-guides of Erodium cicutarium
in the North Frisian Islands. The flower is here markedly protandrous, and always
bilaterally symmetrical, the upper petals being shorter, broader, and of deeper hue
than the lower ones. The nectar-guides are developed to a very varying extent;
in some flowers they have all but disappeared, in others they are very conspicuous.
But in all cases the insect visitors are the same, and blossoms which possess well-
marked nectar-guides are by no means more sought out than those almost lacking
them. Insects fly from one kind of flower to another without their selection being
influenced by the degree of prominence of the nectar-guides. This is noteworthy, it
being generally assumed that the guides indicate the position of the nectar to insects,
which in flowers without such markings, either do not find, or do not suspect the

GERANIACEAE 231

presence of, the desired secretion. When, however, insects have once found
the nectar in a flower possessing guides, it is easy for them to discover it in flowers
of the same species devoid of guides. But the various flower-forms of Erodium
cicutarium somewhat limit Sprengel’s theory of nectar-guides (‘Entd. Geh.,
Pp. 15, 16), since they show that these—though helping insects to find nectar—
are not absolutely necessary, it being sufficient if a certain number of blossoms
possess them.

Loew summarizes as follows the results he obtained by studying the flower-
pollination of Erodium cicutarium (‘Bliitenbiol. Floristik, p. 212)—The absence
or presence of nectar-guides is not an essential distinction between the autogamous
and allogamous varieties, nor are these markings restricted—as Ludwig maintained—
to pimpinellifohum. Both great varieties of the species present various stages between
allogamy and autogamy, differing chiefly in their feebler or stronger protandry,
smaller or larger flowers, and possession of actinomorphy or zygomorphy.

Besides the hermaphrodite flowers, Schulz has observed unisexual ones, distributed
gynodioeciously and gynomonoeciously, or androdioeciously and andromonoeciously.

Warnstorf describes the flowers as protogynous, and adds the following.—The
stamens are shorter than the styles, so that self-pollination is prevented. In
the small-flowered variety (at Ruppin) the petals are usually not blotched at the
base, but the two upper ones, which are smaller and of a deeper red than the others,
frequently have the yellowish basal blotch that is characteristic of the large-flowered
variety. In the latter there may be, though rarely, three to four blotched petals.
In the smaller flowers the stamens are often vestigial, so that female stocks occur.
The stigma is purple, rose-coloured, or pale yellow.

Visirors.—Sprengel saw humble-bees and the honey-bee. In the island of
Rom, I noticed a hover-fly (Helophilus pendulus Z.), remarkably freq., and I have
seen other hover-flies in Féhr and at Kiel, as well as short-tongued bees.

Herm. Miller gives the following list.—

A. Coleoptera. Coccznellidae: 1. Coccinella septempunctata Z., nect-lkg.
B. Diptera. (a) Conopidae: 2. Myopa buccata Z., skg. (6) AMfuscidae: 3. Calliphora
vomitoria Z., ske.; 4. Lucilia cornicina #, do.; 5. L. sp. do. (c) Syrphidae:
6. Rhingia rostrata Z., skg.; 7. Syritta pipiens Z.,do. C. Hymenoptera. (ca)
Apidae: 8. Andrena gwynana &. 9, skg. (Thuringia); 9. A. parvula K. 9, do.;
ro. Apis mellifica Z. ¥, skg. and po-cltg.; 11. Halictus cylindricus /. 9, skg.;
12. H. leucozonius Schr. 9, do.; 13. H. nitidiusculus A. 9, do.; 14. Sphecodes
ephippius Z.,do. (4) Sphegedac: 15. Ammophila sabulosa Z., skg. D. Lepidoptera.
Rhopalocera: 16. Pieris napi L., skg. persistently; 17. P. rapae L., do.

Verhoeff observed the following in Norderney.—

A. Coleoptera. (a) Nrtidulidac : 1. Meligethes aeneus 7, po-dvg. B. Diptera.
(a) Muscidae: 2, Anthomyia sp.; 3. Miltogramma sp., po-dvg. (4) Syrphidae: 4.
Melithreptus menthastri Z. C. Hymenoptera. (a) Chrysididae: 5. Holopyga
amoenula Dahlb. (5) Pteromalidac: 6. Pteromalus sp. (c) Sphegidae: 7. Oxybelus
uniglumis Z., skg.

Loew saw Apis mellifica Z. ¥, skg., in Silesia (‘ Beitrage,’ p. 25); as did also
H. de Vries in the Netherlands (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).
MacLeod observed Apis, a humble-bee, 2 short-tongued bees, 3 hover-flies, 3 Muscidae,
and a Lepidopterid in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 236-7).

232 -ANGIOSPERMAE—DICOTYLEDONES

Schletterer noticed the following bees at Pola:—1. Andrena ventricosa Dours. 9 ;
2. Ceratina cucurbitina Ross’; 3. Halictus calceatus Scop.

In Dumfriesshire 3 Muscidae have been recorded (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 37).

Ludwig saw some Afidae, 2 ichneumons, 5 bees, and the following flies,
visiting the variety pempznellifolium (D. bot. Monatschr., Arnstadt, i, 1883).—

(a) Syrphidae: 1. Syrphus pyrastri Z.; 2. S. cinctellus Ze//.; 3. S. lineola Zevz. ;
4. S. corollae #.; 5. S. balteatus Deg.; 6. S. arcuatus Fe//.; 7. Eristalis sepulcralis
L.; 8. Syritta pipiens Z.; 9. Melithreptus scriptus Z.; 10. M. pictus A/g.; 11. M.
taeniatus AZg.; 12. Melanostoma mellina Z.; 13. M. gracilis A/g.; 14. Ascia poda-
grica /.; 15. Xylota segnis Z.; 16. Platycheirus albimanus /.; 17. P. scutatus A/g.;
18. P. clypeatus Afe.; 19. P. fasciculatus Loew; 20. Lucilia caesar £.; 21. L.
sylvarum Ag. (0) Afuscidae: 22. Anthomyia radicum Z.; 23. Spilogaster duplicata
iMfg.; 24. Chortophila cilicrura Rond.; 25. C. dissecta A/g.; 26. C. floccosa J/g.

584. E. malachoides Willd.—

VisiTors.—Schletterer observed the bee Halictus calceatus Scop. at Pola.

585. E. gruinum Willd. (Ludwig, Bot. Centralbl., Cassel, vili, 1881, pp. 357-
62.)—Ludwig states that this South European and North African species bears large
blue protogynous flowers, exhibiting radial symmetry, and having a diameter of
28mm. The stamens at first curve outwards, subsequently bending inwards, so
that self-pollination may ultimately be effected.

586. E. macradenum L’Hérit. (Ludwig, op. cit., viii, 1881, pp. 87-88.)—
Ludwig says that the flowers of this Pyrenean species are so markedly protandrous
that self-pollination is prevented. Each of the two upper petals possesses a large
and conspicuous nectar-cuide. Ludwig also states that the species is adynamandrous.

587. E. Gussoni Tenore.— This small-flowered South European species is very
nearly homogamous.

588. E. Manescavi Coss.—Ludwig states that the petals of this Pyrenean
species are of a purple-violet colour, with darker veins, while the upper one possesses
a basal nectar-guide. The anthesis of the protogynous flowers lasts from 13 to 3
days, and the species is in bloom for four months.

Both autocarpy and allocarpy obtain to a certain extent. Out of forty-four
flowers pollinated from the same stock, twenty-six set fruits, but only 4% of these
ripened.

589. E. moschatum L’Hérit.—Ludwig says that the short-lived inconspicuous
purple flowers of this species are either homogamous, or slightly protogynous.

590. E. maritimum L’Hérit.—According to Ludwig this species is sometimes
pseudo-cleistogamcus.

2. Tripe PELARGONIEAE.

173. Pelargonium L’Hérit.

According to Hildebrand (Bot. Ztg., Leipzig, xxvii, 1869, pp. 479-81), some
species of this genus are protandrous, while the first flowers are purely female.

591. P. triste L—The greenish flowers of this species exhale a strong odour
at night.

GERANIACEAE 233

VisiTors.—During the day Plateau only saw a small Muscid. The time when
the flowers are odorous suggests that they are probably pollinated by nocturnal
insects.

592. P. zonale L’Hérit.—Darwin states that this species is self-sterile.

174. Tropaeolum L.

Protandrous hymenopterid flowers, the spurred calyx of which secretes and
conceals nectar.

593- T. majus L. (Sprengel, ‘Entd. Geh., pp. 213-27; Delpino, ‘ Sugli appar.
d. fecondaz. nella piante autocarp.,’ p. 30; Knuth, ‘ Bloemenbiol. Bijdragen.’)\—This
species bears markedly protandrous humble-bee flowers. There is a red blotch,
serving as a nectar-guide, on the base of the lamina of each yellow petal. When the
flower opens, the stamens are all curved downwards, with anthers still unripe,
the style is still very short, and the stigmas are closely apposed. Now one stamen

lis

Fic. 71. Tvropaeolum majus, L. (From nature.) A..Flower in the first (male) condition: an
anther (a) stands in the entrance of the flower. &. Flower in the second (female) condition: the stigmas
(s) occupy the entrance of the flower. Natural size.

becomes erect, and its anther dehisces exactly opposite the flower-entrance, the
filament again curving downwards when, on the second day, a second stamen raises
itself so that its pollen-covered anther occupies the mouth of the flower. After
all the eight stamens have successively executed these movements, and are all bent
downwards with empty anthers, the style has become so long that the gradually
maturing stigmas take up the position previously assumed by the dehiscing anthers.
An insect visiting one of the younger flowers in search of nectar, will therefore dust
the under-surface of its body with pollen, and when subsequently visiting older
flowers will necessarily transfer some of this to the stigmas.

VisiTors.—Sprengel observed an ant (in the spur), also small spiders (‘which
presumably chase the little insects that creep into the flowers’), and a fly (‘ which,

234 ANGIOSPERMAE—DICOTYLEDONES

however, was not adapted to effect fertilization, for the lazy and stupid insect
mistook the nectar-cover for the nectar-reservoir, and thrust its proboscis into it
to find, as it had been raining, only drops of water’).

In Schleswig-Holstein, Mecklenburg, Pomerania, and Thuringia, I observed
Bombus hortorum Z., also the variety xzgrzcans Schmiedekn., skg., and spending some
seconds in each flower. The spur is 25 mm. long, so that the proboscis of this
humble-bee—being only 21 mm. in length—would not reach its tip, but the entrance
to the spur is so wide, that the insect can push its head 5 mm. down the tube,
and thus easily exhaust its contents. The garden humble-bee is, therefore, a regular
fertilizer of the species. Besides this insect, I once saw an earwig (Forficula auri-
cularia) half-buried in the spur, and remaining there so persistently that I was able
to pluck the flower, and observe its behaviour. It was evidently able to get some
of the nectar that filled the spur to an unusual height. A third visitor that I observed
was Apis mellifica Z. §. This first penetrated as far as possible into the spur,
and then made vain efforts to suck. When it did not succeed in getting nectar
in sufficient quantity, it proceeded to collect pollen, and, taught by experience, made
no more fruitless attempts at sucking. Alfken saw Bombus hortorum Z. 4, skg.,
at Bremen.

594. T. minus L. (Knuth, ‘Bloemenbiol. Bijdragen.’)—Like the last species,
this plant is a native of Peru, and its flower mechanism is very similar. The spur,
however, is usually considerably longer, i.e. 25-35 mm. None of our German
humble-bees or bees are able therefore to: reach to the ends of the longest spurs,
even by thrusting in their head, which is about 5 mm. in length. One stamen after
another becomes erect, placing its pollen-covered anther in the entrance of the
flower, and after shedding its pollen again bending back against the corolla. When
all the anthers have dehisced, the three-lobed stigma occupies the entrance of the
flower, so that it is touched by nectar-sucking visitors. The three lower petals
are fringed on their inner surface, so as to prevent visitors from forcing their way
into the lower part of the flower. They are therefore compelled to seek the entrance
to the spur, from above the upwardly dehiscing anther, or the stigma.

3. Tripe Oxatipeaz DC.

This tribe is represented by the genus

175. Oxalis L.

(H. von Mohl, Bot. Ztg., Leipzig, xxi, 1863; Hildebrand, Abh. Ak. Wiss.,
Berlin, 1866, Bot. Ztg., Leipzig, xxix, 1871, ‘Die Lebensverhaltnisse der Oxalis-
Arten,’ Jena, 1884.)

Flowers homogamous, with half-concealed nectar secreted in their bases.

While the three German species appear in only one form, a large number
of foreign species are trimorphous or dimorphous. Experiments in artificial fer-
tilization made by Hildebrand with trimorphous species, confirm the law established
by Darwin for dimorphous plants, and for Lythrum Salicaria, i.e. that there is
greatest fertility with legitimate fertilization. In some species (including O. Aceto-

GERANIACEAE 235

sella), cleistogamous flowers have been observed. Our three native species agree
almost completely in their flower mechanisms.

595. O. Acetosella L. (Herm. Miiller, ‘Fertilisation,’ p. 159; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 237-8; Kirchner, ‘Flora v. Stuttgart,’
pp. 342-3; Knuth, ‘Grundriss d. Bliitenbiol.,”’ p. 39.)— Kerner states that the
chasmogamous flowers are open from 9 a.m. till 6 p.m. Their white petals are
streaked with longitudinal violet veins that serve as nectar-guides, and are also
marked with a yellow basal spot immediately above the nectaries. The nectar
collects in five depressions in the base of the corolla. These are formed by fleshy
appendages of the claws of the five petals, reaching as far as the filaments. The
length of the style being variable, the stigma may either project beyond the anthers,
or be situated in the middle of them.

Hugo von Mohl (Bot. Ztg., Leipzig, xxi, 1863) was the first to describe the
cleistogamous flowers. (Cf. Vol. 1, pp. 51-2.)

VisiTors.—These are rare. Alfken saw Apis and Bombus terrester ZL. 9, at
Bremen.

Herm. Miiller observed the following.—

A. Coleoptera. (a) Nitidulidac: 1.Meligethes, freq. (0) Staphylinidac: 2.Oma-
lium florale Pays., freq. B. Thysanoptera. 3. Thrips, freq. He also noticed
7 Diptera, an ant, and Thrips in the Alps (‘ Alpenblumen,’ pp. 178-9).

In Dumfriesshire a Muscid has been recorded (Scott-Elliot, ‘Flora of Dumfries-
shire,’ p. 40).

Burkill observed the Nitidulid Meligethes picipes S/erm, and Thrips sp. on the
coast of Yorkshire (‘Fertilsn. of Spring Fls.’).

596. O. stricta’ L. (Kirchner, ‘ Flora v. Stuttgart,’ p. 343; Schulz, ‘ Beitrage,’
I, p. 31; Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, II, p. 391.)—Kerner says that the
bright yellow flowers of this species are open from 8 a.m. till 4 pm. The flower
mechanism is very similar to that of the last species, but, according to Schulz,
the anthers of the longer stamens are at the same level as the stigma—which matures
simultaneously—and applied to it in such a way that automatic self-pollination is
inevitable. The shorter stamens serve for cross-pollination. Kerner states that
the flowers remain closed in bad weather. Cleistogamous flowers have not hitherto
been observed.

Visitors.—MacLeod observed Apis, 2 hover-flies, and 2 Lepidoptera in Flanders
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 238).

597. O. corniculata L.—The flower mechanism of this species completely
agrees with that of O. stricta. And here too, according to Kerner, the flowers remain
closed in bad weather, fertilizing themselves pseudo-cleistogamously.

598. O. cernua Thunb.—This species has been introduced into Sicily and
Sardinia, where, according to Nicotra (‘Oss. antobiol. sull’ Oxalis cernua’), it is
always microstylous, which explains its sterility in those islands, Stigmatic papillae

1 (In the Index Kewensis, O. stricta L. is given as a synonym of O. corniculata L.—TR.]

236 ANGIOSPE RMAE—DICOTYLEDONES

can scarcely be said to exist. The pollen-grains are not homogamous. As the
shorter stamens grow more rapidly than the style, the stigmas are often seen to
be covered with pollen, but this apparently does not lead to aulocarpy. Nicotra
also observed cleistogamous flowers, or at least transition-forms towards such, and
these perhaps now and then set fruits.

4. Trisz BacsaMINEAE.

176. Impatiens L.

Markedly protandrous bee flowers, or more rarely hover-fly flowers. The nectar
is secreted in the end of the calyx-spur. Some species (including I. Noli-tangere),
especially North American ones, occasionally possess cleistogamous flowers, according
to Hugo von Mohl (Bot. Ztg., Leipzig, xxi, 1863). Von Mohl’s description of these
is given in Vol. I, pp. 52-3. Some North American species of the genus are visited
by humming-birds.

599. I. Noli-tangere L. (Kirchner, ‘Flora v. Stuttgart,’ pp. 346-7; Knuth,
‘Grundriss d. Bliitenbiol.,” pp. 38-9, ‘Bloemenbiol. Bijdragen.’)—In the large
pendulous flowers of this species, which are golden-yellow, dotted with red in the
throat, the united anthers of the five stamens are situated on the upper lip, and
surround the stigma. When the flower opens, the anthers have dehisced, so that
a humble-bee after getting at the nectar must carry away pollen on its back.
At a later stage the stamens separate and the stigma matures, assuming the position
previously occupied by the anthers. Automatic self-pollination, by pollen that
remains clinging to the anthers, is possible.

VisiTors.—I saw at Eutin two species of humble-bee skg., Bombus lapidarius Z.
g and é, and B. hortorum Z. 9 and 4 (also the variety mzgricans Schmiedekn.).
I further noticed a wasp—Vespa media e/g. Y—at work on the flowers, and it
seemed to reach the deeply concealed nectar. At Flensburg I also saw a humble-
bee—Bombus terrester Z.— which, however, was unable to completely empty the
spur. In Central Germany, Herm. Miiller also saw humble-bees, without being able,
however, to determine their species. Loew noticed a small bee, Halictus cylindricus
fF. 9, po-cltg., and, at the same time, incidentally effecting cross-pollination. As
unbidden guests, Herm. Miiller observed a species of Halictus (H. zonulus Sz. 9),
2 beetles (Meligethes and Dasytes flavipes 7), and a fly (Sargus cuprarius Z. 4).

Hermann Miiller (‘Alpenblumen,’ p. 179) found the flowers in Prattigau
frequently perforated by Bombus mastrucatus Gers/., and remarks that the flower
there probably often has recourse to automatic self-pollination, its capacity for which
was demonstrated by Darwin (‘ Cross- and Self-Fertlsn., p. 367).

600. I. parviflora DC. (Bennett, ‘On the Floral Structure of Impatiens fulva
Nutt. ; Henslow, ‘On the Self-fertilisation of Plants’; Knuth, ‘Bloemenbiol. Bij-
dragen. )—The small, bright yellow flowers of this plant possess the same flower
mechanism as the last species, but, according to Henslow, automatic self-pollination
is favoured. Bennett states that cleistogamous flowers do not occur.

GERANIACEAE 237

Vistrors.—In the Kiel Botanic Garden I observed no bees, but repeatedly saw
a small hover-fly, Syrphus balteatus Deg., which sucked persistently, and constantly
visited several flowers in succession.

On September 10, 1897, I noticed (‘Bliitenbiol. Notizen’) hover-flies, almost
exclusively Syrphus corollae /., in the Kiel Ober-Realschule Garden. This insect
first hovers in the sunshine in front of the flower, approaches to within a few
millimetres, again hovers a little further away, and repeats this sport several times
till finally it settles down to suck, and to devour pollen. I also observed Syrphus
tibesii Z., the same morning, visiting the flowers several times, while Apis avoided
them, flying between the inflorescences and steadily visiting the flowers of Sedum
maximum. I therefore regard J. parviflora as a hover-fly flower, not as a bee flower
or humble-bee flower, like the other cultivated species.

The insect visitors of the species are generally
very few, but as in spite of this all the flowers set
fruits, it must be assumed that the plant is self-
fertile.

601. I. Balsamina L. (Sprengel, ‘Entd. Geh.,’ =
p- 400; Hildebrand, Bot. Ztg., Leipzig, xxv, 1867; =
Delpino, ‘Sugli appar. d. fecondaz. nelle piante
autocarp.,’ pp. 30-1.)—The mechanism of the flower
of this species agrees with that of I. Noli-tangere.
In younger flowers the visitors are dusted with
pollen, which they transfer to the stigmas of older Fic. 72. Impatiens parviflora,

oe s DC. (From nature.) Flower seen from
ones, in which the stamens have already dropped off. ic ront. a, anthers.

j~—a

Ome)

Va

VisiToRs.— These are bees (humble-bees). Prunet
(Rev. gén. Bot., Paris, iv, 1892) observed numerous insects, especially Apis, Bombus
hortorum Z., B. terrester Z., and Polistes gallica Z.

602. I. glanduligera Lindl. (=I. Roylei Wa/p.). (Delpino, ‘Ult. Oss.,’ II;
Hildebrand, Bot. Ztg., Leipzig, xxv, 1867; Stadler, ‘ Beitrage’; Loew, Bot. Jahrb.,
Leipzig, xiv, 1891, pp. 166-82; Knuth, ‘ Bloemenbiol. Bijdragen.’)—This commonly
cultivated East Indian species bears well-marked humble-bee flowers. The large
markedly protandrous purple-red blossoms have a short spur, and are of such a size
that the body of a humble-bee just finds sufficient room in them. These insects,
when visiting flowers in the first stage, get their backs dusted with pollen by contact
with the anthers, and when visiting flowers in the second stage, they brush the stigma
with the same part of the body. They disappear altogether in the blossoms when
sucking nectar from the short blunt spurs, and leave flowers in the male condition
with a streak of pollen 1-3 mm. in length on their backs.

This view of the flower mechanism, represented in my figure, is also taken
by Delpino and Hildebrand. Though Loew does not dispute the possibility of such
a mechanism, it does not appear to him to be the normal one indicated by the
structure of the flowers. He points out the remarkable stigma-like structure of the
ligules springing from the bases of the filaments, the bringing together of these
in a narrow space readily accessible from the front, the position of this space at
the most anterior point of the androecium immediately above the surface where

238 ANGIOSPERMAE—DICOTYLEDONES

the pollen is shed, and lastly the occurrence of a mass of germinating pollen-grains
on the upper surface of the closed stigma. All these facts, in Loew’s opinion, lead
to the conclusion that the cleft between the anterior stamens is a ‘pollinating
chamber,’ while the ligular tips are ‘pseudo-stigmas, or pollen-catchers.’ When,
for example, a humble-bee of suitable size—Bombus agrorum /., and B. terrester
Z£., on the Peacock Island at Potsdam—flies to the wide opening of the flower,
it first alights on the under-lip, the lateral teeth serving as points by which to hold on.
It then tries to thrust its head under the sexual apparatus that hangs down from
the roof of the flower-entrance, so as to penetrate into the wide sac-like sepal that
secretes nectar at its end. When pressing its head against the androecium, it
probably pushes forward a little the end of the contained stigma, which is directed
obliquely to the front, and with the stigma it moves the pollen-catchers (i. e. the ligules).
But even without this assumption, a humble-bee must in many cases, when pressing
against the anterior margin of the androecium, introduce its prominent cephalic
hairs into the cleft, so as to come into contact with the top of the pisti]l. When
in this way a bee brings pollen from a flower previously visited, this pollen will
be retained by the funnel-shaped ligular cap, and will germinate on the stigmatic
surface at the top of the ovary. The fact that the stigma of I. Roylei remains
closed, the ligular cap obviously occupying its place, lends special probability to
this interpretation.

Loew also observed a dwarf-flower, intermediate in character between chasmo-
gamous and cleistogamous blossoms. Actual cleistoramous flowers, such as have

Fic. 73. Jmpattens glanduligera, Lind/. (Longitudinal section. From nature.) .4. Flower in the
first (male) condition: the pollen-covered anthers (2) are above the entrance to the tlower. 2. Flower
in the second (female) condition: the stigma (s) is above the entrance to the flower. 2, nectary.
(Natural size.)

been described for numerous species of this genus, have not hitherto been noticed
in I. Roylei.

Vistrors.—In the Kiel Botanic Garden I observed the following 3 humble-bees,
all very freq., skg.—1. Bombus agrorum /. 9 and 6; 2. B. lapidarius Z.8; 3. B.
terrester Z. 9 and 6. I also noticed Apis mellifica Z. ¥, which readily touches the
anthers or stigma with its dorsal surface when entering or leaving the flower.
It can therefore effect cross-pollination.

603. I. latifolia L.—According to Loew (op. cit.), this species bears lepido-
pterid flowers, possessing stigmas that are not covered, but project in the form
of thin, feebly lobed membranous lamellae.

RUTACEAE 239

XXVI. ORDER RUTACEAE JUSS.

1. Trise Rutear.

Litrrature.—Knuth, Grundriss d. Blitenbiologie, p. 39; Urban, Jahrb. bot.
Gart., Berlin, ii, 1883, pp. 366-404.

The species of the genus Ruta bear nauseous flowers, which, by their penetrating
odour and dull yellow colour, attract Hymenoptera, and particularly Diptera that are
fond of putrid matter. On the other hand, the species of Dictamnus are visited
by Apidae. Owing to their marked protandry, cross-pollination results in the species
of both genera from insect-visits.

177. Ruta Tourn.

Flowers with exposed nectar, secreted by a fleshy disk beneath the ovary.

604. R. graveolens L. (Sprengel, ‘Fntd. Geh.,’ p. 236; Herm. Miiller,
‘Fertilisation,’ pp. 160-2, ‘ Weit. Beob.,’ I, p.213; Schulz, ‘ Beitrage,’ II, pp. 59, 60:
Kirchner, ‘Flora v. Stuttgart,’ p. 548 Knuth, ‘ Bliitenbiol. Herbstbeob.’) — As
Hermann Miiller justly remarks, the flower-mechanism of this species is very like
that of Parnassia palustris, for the stamens first of all ripen successively, and then
the stigma matures. The transfer of pollen is secured in both by the fact that
the middle of the flower, which offers the most convenient alighting-place, is occupied
first by a dehisced anther, and afterwards by the stigma. Finally, in both, the
exposed nectar attracts similar guests (flies and short-tongued Hymenoptera), but
while the white flowers of Parnassia are also visited by beetles, the dull yellow
flowers of Ruta are not.

There is, however, this difference. In Ruta, according to Urban, all the
stamens once more curve upwards before the stigma withers, so that should the anthers
retain any pollen this falls upon the stigma, and thus renders automatic self-pollina-
tion possible towards the end of anthesis. Schulz, however, states that autogamy
is impossible.

Visirors.—Hermann Miiller (H. M.) in Westphalia, Buddeberg (Budd.) in
Nassau, and myself (IXn.) in Schleswig-Holstein, have observed the following.—

A. Diptera. (@) Muscidac: 1. Anthomyia obelisca A/y., skg. and po-dvg.
(Winnertz); 2. A. pratensis J/g., do. (Winnertz); 3. a radicum L., do. (Winnertz) ;
4. Calliphora erythrocephala AZg., do. (Kn., H. M.); 5. Lucilia caesar Z., do. (Kn.);
6. L. cornicina /., do. (H. M.); 7. L. sylvarum J/e., “do. (H.M.); 8. Pollenia rudis
f., do. (H. M.); 9. Sarcophaga albiceps sl/y., do. (H. M.); ro. S. carnaria /., do.
(H. M., Kn.); 11. S. haemorrhoa A/g¢., do. (H. M.); 12. Scatophaga stercoraria /..,
do. (Kn.); 13. Sepsis sp., do. (H. M.). (2) Sératvomyrdae: 14. Chrysomyia formosa
Scop., skg. (Budd.); 15. Sargus cuprarius Z., skg.and po-dvg. (H. M.). (c) Syrphzdae :
16. Ascia podagrica /., skg. and po-dvg. (H.M.); 17. Eristalis sepulcralis Z., do.
(H.M.); 18. E. tenax Z., do. (IXn.); 19. Helophilus floreus 7, do. (H.M.); 20.
Melithreptus pictus J/g., do. (H.M.); 21. Syritta pipiens Z., do. (H.M., Kn.);
22. Syrphus nitidicollis /g., do. (H.M.); 23.S. ribesii 7., do. (H. M.). B. Hymeno.
ptera. (a) Apidae: 24. Apis mellifica Z. ¥, skg. (H.M., Kn.); 25. Halictus
sexnotatus A. 9, do. (H.M.); 26. H. tetrazonius KE 9, do. (H. M., Kn.); 27. Pro-
sopis sinuata Schenck 9, do. (H. M., Budd.); 28. Sphecodes gibbus 4. 9, do. (Budd.).

240 ANGIOSPERMAE—DICOTYLEDONES

(6) Chrysididae: 29. Chrysis ignita Z., skg.(H.M.). (c) Evaniidae: 30. Gasterup-
tion affectator /., skg. (H.M.); 31. G. jaculator /., do. (H. M.). (¢) Ichneumonidae:
32. Ichneumon sp., skg. (H.M.). (e) Scoliidae: 33. Tiphia minuta v. d. LZ. 3, skg.
(H.M.). (f) Sphegidae: 34. Crabro chrysostoma Leg. , skg. (Budd.); 35. C.
clavipes Z., do. (H.M.); 36. C. dives #.-Sch. 8, do. (Budd.); 37. C. elongatulus
v. d. L. 9, do. (H.M.); 38. C. guttatus v. d. Z. , do. (Budd.); 39. Oxybelus bellus
Dahlb., do. (H.M.); 40. Pseudagenia carbonaria Scop. 8, do. (H.M.) 5 41. Trypoxylon
figulus Z., do. (H.M.). (g) Vespidae: 42. Odynerus parietum ZL. 8, skg. (H.M.,
Budd.); 43. Polistes gallica £., do. (Budd.).

Loew observed a Syrphid, Syritta pipiens Z., skg., and the honey-bee, Apis
mellifica Z. ¥, skg., in the Berlin Botanic Garden; F. F. Kohl saw the wasp Eumenes
pomiformis /., in the Tyrol.

The variety dvaricafa Tenore was seen by Schletterer to be visited at Pola
by the bee Prosopis clypearis Schenck.

605. R. bracteosa DC. (Knuth, ‘Bliitenbiol. Beob. a. d. Ins. Capri.’)—The
flower mechanism and means of attraction in this species agree with those of
R. graveolens.

Vistrors.—I observed at Capri only a few Diptera and a species of ant.

Fic. 74. Déctamnus a/bus, L. (From nature.) A. Flower in the first (male) condition: the
anthers (a2) occupy the entrance to the flower. B. Flower in the second condition: the stigma (s)
protrudes from among the stamens. (Natural size.)

178. Dictamnus Tourn.

Protandrous hymenopterid flowers.

606. D. albus L. (Delpino, ‘ Ult. oss., p. 145; Hildebrand, Bot. Ztg., Leipzig,
xxviii, 1870; Loew, ‘ Bliitenbiol. Floristik,’ p. 214; Urban, Jahrb. bot. Gart., Berlin,
ii, 1883, pp. 36-404; K. F. Jordan, Ber. D. bot. Ges., Berlin, v, 1887, pp. 327-44 ;
Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II, pp. 203, 225; Knuth, ‘Bloemenbiol. Bij-
dragen.’)—The flower mechanism closely resembles that of Aesculus Hippocastanum.
As Delpino first demonstrated, the pollen-covered stamens project in the first stage
from the flowers (which exhale an odour of citron), while in the second stage the
mature stigma protrudes. These parts serve as alighting- and resting-places for
insect visitors. During the male stage the stamens lie upon the under-lip of the
flower, curving upwards to its entrance, while the style is hidden in the middle

RUTACEAE 241

of them. After the pollen has been shed, the filaments straighten, and the style
with the mature stigma bends upwards at right angles, so that a female condition
ensues. An insect with proboscis of suitable length probing for the nectar secreted
in the base of the flower, must therefore introduce its tongue between the petals
and the stamens, or stigma, so that when flowers of different ages are visited, cross-
pollination necessarily takes place.

Visirors.—In the Kiel Botanic Garden, I observed the following three bees, all
freq., skg.:—1. Megachile willughbiella K. 9 and 8; 2. Bombus lapidarius Z. ¥;
3. Apis mellifica Z. ¥. Of these visitors, however, the honey-bee alone regularly
alighted in such a way as to touch the anthers or the stigma, and so constantly
effected cross-pollination, the two others often approached laterally, so that they
only touched the filaments.

Loew saw two bees in the Berlin Botanic Garden :—r. Apis mellifica Z. ¥, skg.
and po-cltg.; 2. Bombus agrorum /. 9 and ¥, do. On the variety roseus he saw two
others :—1. Megachile centuncularis Z. 9, po-cltg., hovering above the anthers and
removing pollen with its ventral brush; 2. M. circumcincta X. 4, skg.

2. Tripe Boronigae.

179. Correa Andr.

Flowers protandrous (Delpino, ‘ Ult. oss.,’ p. 170). Numerous species have been
studied in the Berlin Botanic Garden by Urban with regard to pollination (cf. abstract
of his paper in Bot. Centralbl., Cassel, xiv, 1883, pp. 200-4).

3. Trise ToppALigAE.

180, Ptelea L.

Flowers greenish, pseudo-hermaphrodite, dioecious, in pseudo-umbellate racemes;
with concealed nectar secreted in their bases.

607. P. trifoliata L. (Knuth, ‘Bloemenbiol. Bijdragen.’)— This North
American shrub is occasionally cultivated in our gardens. The whitish-green flowers
exhale a strong odour of hyacinths; they secrete a very small quantity of nectar
below the ovary, by which it is concealed in the female flower, and in the male
flower by the staminal hairs as well. The female flower possesses vestigial stamens,
the anthers of which are sterile; the terminal stigma projects 1-2 mm. beyond them.
The male flower possesses a moderately large ovary, but this develops no further.
The five stamens are closely beset with white hairs of moderate length on the
inner side of the lower halves of their filaments. These hairs serve to protect
the nectar from rain, and keep off unbidden guests. The anthers dehisce simul-
taneously, and turn their pollen-covered sides upwards, so that a nectar-seeking
insect dusts its head or thorax with a ring of pollen, and when it visits a female
flower, this must be applied to the stigma. The male flowers are considerably

DAVIS. 1 R

242 ANGIOSPERMAE—DICOTYLEDONES

larger (diameter 14 mm.) than the female flowers, and have also, it seems to me,
a somewhat stronger odour, so that they are first visited by insects.

Visirors.—In the Kiel Botanic Garden I observed on June 20, 1896, the honey-
bee, freq., skg., well dusted with pollen, and a hover-fly, Syritta pipiens Z., solitary,
po-dvg. and skg.

At the close of his memoir (Jahrb. bot. Gart., Berlin, ii, 1883), Urban gives the
following summary of the flower mechanisms observed by him in the Rutaceae.—

I. Prants Mownoctinovs.
A. Flowers protandrous.

1, The filaments bring the anthers one after the other to the place subsequently
occupied by the mature stigma, and afterwards return to their original position.
(a) Style (and stigma) undeveloped during the male stage.

(2) Ruta. The filaments are at first horizontal, then elongate considerably,
apply themselves to the ovary, again move back and become erect. Petals
horizontal. Self-pollination usually impossible.

(8) Coleonema. The filaments are at first short and erect; they elongate,
bend over, and again become straight. Petals meeting together below
to form a tube. Automatic self-pollination possible by fall of pollen.

(b) Style developed in the male stage (though sometimes imperfectly), but so
placed that self-pollination is impossible.

* Flowers zygomorphous.

(a) Dictamnus. The filaments lie upon the under-lip; they become curved,
the lower ones first, bend upwards over the middle of the flower, and
straighten again after shedding their pollen. The style is at first curved
somewhat downwards; after the pollen has been shed, it bends upwards
at right angles.

(8) Calodendron. The filaments are curved upwards; they elongate, the
anterior ones first, and become almost straight before shedding their pollen,
finally curving outwards. The style, which is at first curved downwards,
straightens after the pollen is shed.

** Flowers actinomorphous. The filaments (successively) elongate considerably
after dehiscence.

(a) Diosma. The style is at first curved horizontally over the ovary. The
petals finally become erect, and the filaments curve outwards between them.

(8) Adenandra, As before, but the staminodes, not the petals, incline together
at the end. The filaments of the fertile stamens curve outwards but little.

(y) Barosma. The style curves outwards and downwards between the
staminodes after the flower opens. The petals remain in a horizontal
position, the staminodes are applied to the ovary, and the filaments of

fertile stamens resume their original horizontal position after the pollen
has been shed.

RUTACEAE 243

2. The filaments make only one movement, which they execute simultaneously.
In the male stage they are erect, or else somewhat inclined towards one another,
so that the anthers touch; in the female stage the stamens are curved outwards.
(a) The anthers fall off when the filaments separate. As the stigmatic branches
do not diverge till this stage, self-pollination is impossible. Ravenia.
(b) The anthers persist after the filaments have separated.

* In the male stage pollen may fall from the anthers upon the still sessile or
immature stigma, and thus self-pollination may subsequently result. And
later on, owing to the wind or the position of the flowers, pollen from
the separate anthers may reach the mature stigma.

(a) Zieria and Eriostemon. The style continues to grow in the second
stage of anthesis.

(8) Boronia (ex parte). The stigma matures only at a late stage.

(y) Erytrochiton. The style elongates only at a late stage, but the stigma
comes into contact with the anthers before they have separated.

** Neither in the male nor in the female stage can pollen automatically reach
the stigma from the anthers, owing to their position, and the adhesive
character of the pollen. Metrodorea.

3. The filaments execute no movements either before or after the shedding

of the pollen.
(a) Correa. Self-pollination of the pendulous flowers is ultimately possible after
the stigmatic lobes have spread out.
(b) Agathosma (ex parte). The style is enclosed by the staminodes in the
male stage. In the female stage, the stigmas may still be pollinated,
should insect-visits fail, by pollen from the anthers of neighbouring flowers.

B. Flowers homogamous.

1. Automatic self-pollination is impossible—
(a) in Boronia (ex parte), owing to the adhesive nature of the pollen;
(b) in Triphasia, because the stigma projects considerably beyond the anthers.
2. Automatic self-pollination is impossible, owing to the position of the filaments,
but mutual automatic pollination of neighbouring flowers is favoured by the position
and torsion of the anthers. Agathosma (ex parte).

3. Automatic self- and cross-pollination are not likely to occur; self-pollination
by insect-visits is inevitable. Crowea.

4. Automatic self-pollination is possible, but cross-pollination is favoured.
Cusparia, Choisya, Skimmia (ex parte), Murraya, Citrus.

II. Prants Dicrinous.

Self-pollination is impossible; cross-pollination furthered. Ptelea, Skimmia
(ex parte).
R2

244 ANGIOSPERMAE—DICOTYLEDONES

XXVIII ORDER JZICINEAE DC.

181. Ilex L.

Flowers white, often dioecious, with exposed nectar secreted in their bases.

608. I. Aquifolium L.—This species is described by Vaucher and Darwin
as dioecious, but A. Schulz found that cultivated plants bore normal hermaphrodite
flowers which set fruits.

MacLeod describes Belgian plants as dioecious (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, pp. 246-7). The male flowers possess a vestigial pistil. The anthers
dehisce when the flower opens. The female flowers have a much larger ovary,
green in colour. In both forms there is but very scanty secretion of nectar.

Visitors.—H. de Vries observed Apis mellifica Z. §, very freq.

XXVIII. ORDER CELASTRINEAE R. BR.

Flowers hermaphrodite or unisexual; mostly inconspicuous, with exposed or
half-concealed nectar.

182. Euonymus Tourn.

Inconspicuous protandrous flowers, with exposed nectar secreted by a fleshy
disk surrounding the style.

609. E. europaeus L. (Delpino, ‘Altri appar. dicog. recent. oss.,’ p. 52;
Herm. Miller, ‘ Fertilisation, pp. 162-3; Kirchner, ‘Flora v. Stuttgart,’ p. 357;
Schulz, ‘Beitrage,’ II, p. 61; Knuth, ‘Bloemenbiol. Bijdragen.’)— The nectar is
spread out in such a thin layer in the green trioecious flowers of this species, and
is so readily accessible that it is chiefly sought for by short-tongued insects. The
hermaphrodite flowers are protandrous. The four stamens are remote from the
stigma, and their filaments are stiff. The anthers dehisce extrorsely, while the stigma
is still immature. This unfolds its lobes only several days later, closing them when
fertilization has been effected. Automatic self-pollination is therefore entirely excluded
(but w2de znfra). When insects visit the flowers cross-pollination almost always takes
place. Self-pollination may occur if there have been no insect-visits in the first
days of anthesis. Warnstorf describes the pollen-grains as white, ellipsoidal, markedly
tuberculated, up to 50 long and 25 w broad. In addition to the hermaphrodite
flowers, unisexual ones also occur, in which non-functional vestiges of the opposite
sex-organs are present. According to Schulz, they are distributed gynomonoeciously
and andromonoeciously, rarely gynodioeciously and androdioeciously.

VistTors.—Schulz noticed Diptera, ichneumon-flies, ants, and beetles, in the South
Tyrol. Herm. Miiller failed to observe beetles as visitors of the dull yellow flowers
of Euonymus, just as he did in the case of the somewhat similarly coloured flowers of
Ruta. I, too, have never seen beetles. In North and Central Germany Herm.
Miiller (H. M.) and myself (Kn.) have observed the following.—

A. Diptera. (a) Brbionidae: 1. Bibio hortulanus Z., skg. (H. M.); 2. Many
minute midges (H.M.). (4) Musczdae: 3. Calliphora erythrocephala A/g., skg. and
po-dvg. (H. M.); 4. C. vomitoria Z., skg. (H.M., Kn.); 5. Echinomyia fera Z., skg.
and po-dvg. (Kn.); 6. Lucilia cornicina #., do. (H. M., Kn.) ; 7. Musca domestica Z.,

CELASTRINEAE 245

do. (H.M.,-Kn.); 8. Sarcophaga carnaria Z., do. (H. M., Kn.); 9. Scatophaga mer-
daria F., do. (Kn.); 10. S. stercoraria Z., skg. (H.M.). (c) Syrphidae: 11. Eristalis
nemorum Z., skg. and po-dvg. (Kn.); 12. E. tenax Z., skg. (H. M.); 13. Helophilus
floreus Z., skg. (H. M.); 14. Syritta pipiens Z., do. (H. M.); 15. Syrphus ribesii Z.,
do. (H. M.); 16. S.sp., skg. and po-dvg.(Kn.); 17. Xanthogramma citrofasciata Deg.,
skg. (H.M.). B.Hymenoptera. Formicidae: 18. Formica sp., skg. (H. M.).

Schiner saw the hover-fly Criorhina asilica Fall, freq., in Austria.
610. E. latifolius Scop.—

VisiTors.—Loew saw the Muscid Calliphora erythrocephala Mg., skg., in the
Berlin Botanic Garden.

611. E. americanus L.—
Visitors.—Loew noticed the honey-bee, skg., in the Berlin Botanic Garden.

612, E. japonicus Thunb.—This species is a native of Japan.

Visirors.—In the Tyrol, F. F. Kohl observed the following ruby wasps.—
1. Chrysis leachii Shuck.; 2. C. viridula Z.; 3. C. splendidula Ross?; 4. C. rutilans
Oliv.; 5. C. scutellaris Fabr.; 6. C. analis Spen.; 7. C. distinguenda Spzm.; 8. C.
comparata Lefel.; 9. C. inaequalis Dahlé.; 10. Stilbum nobile Su/z.; 11. Hedychrum
nobile Scop.; 12. H. rutilans Dahld.; 13. Holopyga rosea Ross?; 14. H. chrysonota
Forst.; 15. Ellampus caeruleus Pad/.: and also the wasps—16. Vespa crabro ZL. ;
17. V. germanica #.; 18. V. saxonica Yadr.; 19. Polistes gallica Z.; 20. Eumenes
pomiformis /.; 21. Odynerus floricola Sauss.; 22. O. modestus Sauss.

Handlirsch, according to Kohl, records the fossorial wasp Gorytes pleuri-
punctatus Cos/a.

613. E. variegatus.—

Visitors.—F. F. Kohl] records for the Tyrol the wasp—tr. Polistes gallica Z.;
and the ruby-wasps—z. Chrysis leachii Schuck; 3. C. bidentata Z.; 4. C. scutellaris
Fabr.; 4. C. distinguenda Spzn.; 5. C. inaequalis Dahlé.; 6. Holopyga rosea Rossz.

183. Celastrus L.

614. C. Orixa Thunb. (=C. japonicus C. Koch).—The flowers of this species
are green, except the anthers, which are yellow.

VistTors.—In the Ghent Botanic Garden, Plateau saw po-dvg. Muscidae—Musca
domestica Z., Calliphora vomitoria Z.

184. Staphylea L.

Inconspicuous, homogamous flowers associated in racemose inflorescences ;
nectar half concealed, secreted at the base of the ovary.

615. S. pinnata L. (Kirchner, ‘Flora v. Stuttgart, p. 356; Knuth, ‘ Bloemen-
biol. Bijdragen.’)—According to Kirchner, the sepals—which are white, usually
tinged with red externally—spread out till in the end they are almost horizontal.
The five small white petals are placed vertically in the pendulous flower, and sur-
round the five stamens pretty closely. The cup-shaped green receptacle forms a
groove round the base of the ovary, bounded externally by a pentagonal
ridge, outside which the stamens are inserted. The approximated stigmas of the
two styles are fused together. They are matured simultaneously with the anthers,

246 ANGIOSPERMAE—DICOTYLEDONES

and placed almost at the same level, but sometimes they project a little beyond them.
In such cases cross-pollination is likely to result from insect-visits. As the stamens
closely surround the style, and the anthers dehisce introrsely, the sticky pollen—
owing to the pendulous nature of the flower—readily falls upon the stigma.

VisiTors.—In the Kiel Botanic Garden, I observed only the following Diptera,
skg. or po-dvg.—(a) Syrphidae: 1. Eristalis tenax Z.; 2. Syrphus ribesii Z.; 3.
Melanostoma mellina Z. (0) Musczdae: 4. Scatophaga stercoraria Z.; 5. Lucilia
caesar Z.; 6. Sarcophaga carnaria ZL.

XXIX. ORDER RHAMNEAE R. BR.
Inconspicuous protandrous flowers, with exposed nectar. Dioecism frequent,
dimorphism occasional.
185. Rhamnus L.

Inconspicuous, frequently dioecious flowers with exposed nectar secreted by the

calyx. In some species there are dimorphous flowers (e.g. R. lanceolata, according
to Darwin).

i hy)

i
i

SiMe

Fic. 75. Rhamnus pumila, L. (After Herm. Miller.) 4. A tetramerous flower with two petals;
two anthers have dehisced, and two have not. 2. A tetramerous flower without any petals; all the
anthers have dehisced. C. A pentamerous flower with five petals; all the anthers have dehisced. JD. The
same in longitudinal section.

616. R. cathartica L. (MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp- 248-9; Schulz, ‘Beitrage,’ II, p. 185; Kirchner, ‘Flora v. Stuttgart, p. 363;
Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896 ; Knuth, ‘ Bloemenbiol. Bijdragen.’)—
Kirchner states that in this species each kind of the greenish, dioecious, fragrant
flowers possesses vestiges of the opposite sex-organs. The male flowers are larger
than the female: their pistil is either quite rudimentary, and devoid of stigma, or
it may be somewhat better developed. The female flowers possess vestigial stamens,
and are dimorphous.

According to Warnstorf, shrubs with pseudo-hermaphrodite pollen-flowers are
rare at Ruppin and flower more freely than the female stocks. The petals only
cover the filaments of the four stiff, erect stamens; the anthers dehisce introrsely.

The pollen-grains are white, spheroidal, alipeoidal, or ovoid, on an average 31 u
long and 25 » broad.

RHAMNEAE 247

Vistrors.—At Kiel I only saw a hover-fly (Eristalis nemorum Z.), skg. Hoffer
observed Bombus hypnorum Z. 9 in Steiermark.

617. R. pumila L. (Herm. Miller, ‘Alpenblumen,’ pp. 169-71.)—The small
flowers of this species—which Kerner describes as possessing an odour of honey—
are said by Miiller to be usually hermaphrodite, while Koch (‘ Synopsis’) describes
them as dioecious-polygamous. Cross-pollination is favoured in the hermaphrodite

flowers owing to the fact that the stamens and stigmas are on opposite sides of
the nectar.

Visirors.—Herm. Miiller observed Hymenoptera (Chrysididae, Formicidae),
beetles, and Diptera (Muscidae, Empidae, Syrphidae), in the Alps.

618. R. saxatilis L—Kerner describes this species as dioecious, with pseudo-
hermaphrodite pollen-flowers and fruiting-flowers.

619. R. Frangula L. (=Frangula Alnus JZ//.). (Herm. Miiller, ‘ Fertilisation,’
pp. 163-4, ‘ Weit. Beob.,’ II, p. 212;
Kirchner, ‘Flora v. Stuttgart,’ pp.
363-4; Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896; Schulz, ‘Bei-
trage,’ I, p. 31, II, p. 61; Knuth,
‘Bloemenbiol. Bijdragen.’)— The in-
conspicuous greenish-white herma-
phrodite flowers of this species,
according to the investigations of
Hermann Miiller and A. Schultz, are
protandrous to a less (Thuringia) or
greater (Westphalia, South Tyrol)
degree. The cup-shaped calyx is
also a hemispherical nectary. Between
the five triangular whitish tips of the
sepals, are five small white bifid petals,
which almost cover the five inwardly
curved introrse stamens. In the base
of the calyx lies the short-styled pistil,
the bilobed stigma of which is at Fic. 76. Rhamnus Frangula, L. (After Herm.

Miller.) 1. Young flower seen from above. 2. The same
a lower level than the anthers. At (in section) seen from the side, after removal of the anterior

the time when the latter are ripe the Malfofthecalys, 3 Older lowersten rom above, s, sepals
stigma is still ill-developed. Insect
visitors usually effect cross-pollination while sucking nectar, since they touch
the anthers with one side of their body and the stigma with the other. As the
flowers, however, are rather inconspicuous, the number of visitors is small, and
automatic self-pollination is often effected as a last resort, the withering stamens
dropping pollen on the mature stigmas.

The account given by A. Schulz differs from that of Miiller in some points.
He found that in plants growing at Halle and in North Thuringia, the petals are
never so deeply cleft as represented by Miiller (see Fig. 76). He describes the
anthers as being enveloped for a considerable time by the white longitudinally

248 ANGIOSPERMAE—DICOTYLEDONES

folded petals, which only become erect and free the anthers at a later stage.
Warnstorf gives a similar account of the flower mechanism as observed at Ruppin.
The blossoms are homogamous, and when they open the stigmatic papillae are
fully developed, and probably ready for pollination. The stamens are at first covered
by the small cap-like folded white petals, and subsequently incline inwards towards
the stigmas. As the anthers dehisce introrsely, autogamy can take place in the
absence of insect-visits. The small smooth pollen-grains are white and irregular,
varying from rounded-tetrahedral to nearly ellipsoidal, about 30 » long and 19 p
broad.

Schulz states that there are two forms of flower, short-styled and long-styled
respectively. In the one, as figured by Miiller, the style does not attain the level
of the anthers, in the other it reaches at least to their bases, and usually to their
middle, or even higher. These two forms are local.

VisiTors.—Schulz saw at Bozen numerous bees (including Apis), wasps, ich-
neumon-flies, flies, and beetles ; in all about 300 visitors in fourteen days. He noticed
similar visitors in Central Germany.

F. F. Kohl saw the wasp Polistes gallica Z. in the Tyrol.

Hermann Miiller observed the following in Westphalia.—

A. Diptera. 1. Culex pipiens Z. 4, skg. B. Hymenoptera. (a) Apidae:
2. Apis mellifica Z. ¥, skg. and po-cltg.; 3. Bombus agrorum /. 9 and 9, skg.; 4.
Macropis labiata F. 6, do. (4) Vespidae: 5. Eumenes pomiformis /., skg.; 6. Vespa
sylvestris Scop. 9. do.

Alfken saw the following at Bremen.—

A. Coleoptera. Liaterzdae: 1. Corymbites sjaelandicus Afiiller; 2. Elater
balteatus Z.; 3. E. pomonae S¥eph.; 4. Sericus brunneus Z. B. Hymenoptera.
Apidae: 5. Apis mellifica Z. ¥, skg. and po-cltg.; 6. Bombus jonellus A. 9, skg.;
7. B. proteus Gerst. 9; 8. B. terrester Z. 9 and ¥.

Schiner noticed a Muscid—Lophosia fasciata J7¢.—in Austria. MacLeod saw
Apis, a humble-bee, an Empis, and a beetle in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 248). And H. de Vries observed the honey-bee in the Nether-
lands (Ned, Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).

620. R. Alaternus L.—

Visirors.—Schmiedeknecht, according to Piccioli, records Andrena schmiede-
knechti A/agr., for Florence.

186. Paliurus Tourn.

621. P. aculeatus Lam. (=P. australis Gaertn.)— According to Delpino
(‘Altri appar. dicog. recent. oss., pp. 51-2), this species is markedly protandrous.
The stamens are at first erect or slightly bent inwards, shed their pollen, and then
curve outwards, while the stigmas mature.

VisiTors.—Schiner saw a hover-fly—Spilomyia speciosa Ross?—in Austria.

Schletterer observed the following Hymenoptera at Pola.—

(a) Apedae: 1. Anthidium diadema Zir.; 2. A. variegatum F.; 3. Andrena
austriaca Pz.; 4. A. colletiformis MMor.; 5. A. flavipes Pz.; 6. A. nana X.; 4. Cera-
tina cucurbitina Fossz; 8. Colletes lacunatus Dours.; 9. Epeolus scalaris ///.; 10.
Eriades campanularum #’.; 11. Halictus calceatus Scop.; 12. H. interruptus Pz.;

RHAMNEAE 249

13. H. leucozonius Schr. 9; 14. H. tetrazonius Kig.; 15. Nomia diversipes Zyr.;
16. Osmia cephalotes AZor.; 17. Prosopis clypearis Schenck; 18. P. hyalinata Sm.,
var. subquadrata Forsi.; 19. P. pictipes Vy/.; 20. P. variegatus /.; 21. Sphecodes
gibbus Z.; 22. S. subquadratus Sm. (4) Braconidae: 23. Bracon castrator F.;
24. B. nominator #.; 25. B. terrefactor Vill. 9; 26. B. urinator /.; 27. B. xantho-
gaster K7chd. one 9, and one 8; 28. Isomecus schlettereri Krchd.; 29. Microgaster
subcompletus Mees; 30. M. tibialis Wees. (c) Chalcididae: 31. Brachymeria minata
L.; 32. Leucaspis dorsigera F., rare; 33. L. intermedia Ji, rare. (d) Chrysta:dae:
34. Chrysis igniventris 44.; 35. C. chevrieri Mocs.; 36. C. inaequalis Dahlb.; 34. C.
indigotea Duf. et Per.; 38. C. jucunda Mocs.; 39. C. pustulosa Ad.; 40. C. refulgens
Spin.; 41. C. splendidula Ross7; 42. C. succincta Z.; 43. Ellampus spina Zep.;
44. Holopyga amoenula Dahid.; 45. H. chrysonota Forst.; 46. H. curvata Forst.;
47. H. gloriosa #. (e) Evaniddae: 48. Gasteruption affectator Z.; 49. G. granuli-
thorax Zourn.; 50. G. kriechbaumeri ScAlef/.; 51. G. opacum TZourn.; 52. G.
pedemontanum Zourn.; 53. G. rubricans Guér.; 54. G. terrestre Zourn. ; 55. G. tibiale
Lourn.; 56. G. tournieri Schleit. (jf) Ichneumonidae: 57. Amblyteles armatorius
Forst.; 58. Casinaria tenuiventris Gr.; 59. Crypturus argiolus Ross?; 60. Cryptus
bucculentus Zschek.; 61. C. viduatorius /.; 62. Exephanes hilaris Wesm.; 63. Exe-
tastes guttatorius Gr., var. procera Krchd. ; 64. Glypta ceratites Gr.; 65. Hoplismenus
armatorius Pz.; 66. Ichneumon balteatus Wesm.; 67. 1. consimilis Wesm.; 68. I.
monostagon Gr.; 69. IJ. pisorius (L.) Gr.; 70. I. sarcitorius Z.; 71. Limneria
chrysosticta Gr.; 72. Linoceras macrobatus Gr., var. geniculata Archd.; 73. Lissonota
folii Zhs.; 74. L. verberans Gr., var. procera Krchd.; 75. Mesostenus grammicus Gr. ;
76. M. grammicus Gr. var. nigroscutellata Krchd.; 77. Metopius dentatus /.; 78. M.
micratorius /.; 79. Onorga mutabilis Hg7.; 80. Ophion (Eremotylus) undulatus Gr. ;
81. Phygadeuon (Campoplex) nitens Gr.; 82. Pimpla illecebrator Pz.; 83. P.
instigator G7v.; 84. P. turionellae Z.; 85. P. vesicaria Ra/zedb.; 86. Pristomerus
vulnerator Pz.; 87. Sagaritis annulata Gr.; 88. S. annulata Gr., var. fuscicarpus
Krchb.; 89. Spilocryptus claviventris Krchd.; 90. Trachynotus foliator /., extremely
freq.; 91. Trichomma enecator Ross7; 92. Trychosis plebeius Zschek., var. nigritarsis
Krchb. (g) Pompilidae: 93. Agenia variegata Z.; 94. Ceropales variegatus F.;
95. Pompilus aterrimus Rossz; 96. P. cellularis Dahlb.; 97. P. cingulatus Ross7;
98. P. latebricola KoAl.; gg. P. nigerrimus Scop.; 100. P. quadripunctatus #.; 101.
P. ursus /.; 102. P. vagans K/ug.; 103. P. viaticus Z.; 104. Pseudagenia albifrons
Daim.; tog. P. carbonaria Scop.; 106. Salius affinis v.d.Z.; 107. S. elegans Spzm.;
108.8. fuscus #. (4) Scolivdae: tog. Myzine tripunctata Ross7; 110. Tiphia femorata
F., rare; 111. T. morio &. (2) Sphegidae: 112. Astata bodps Schr.; 113. A. minor
Kohl.; 114. Cerceris arenaria Z.; 115. C. bupresticida Duf; 116. C. conigera Dahid.;
117. C. emarginata Pz.; 118. C. quadricincta Vil/.; 119. C. quadrimaculata Duf ;
120. C. quinquefasciata Moss?; 121. C. specularis Cos/a; 122. Crabro clypeatus
Schreb.; 123. C. meridionalis Cos/a; 124. C. vagus Z., freq.; 125. Gorytes consan-
guineus Hand/.; 126. G. pleuripunctatus Costa; 127. G. procrustes Hand/.; 128. G.
quinquecinctus #.; 129. Larra anathema Jossz, one 5; 130. Nysson scalaris JU. ;
131. Pemphredon shuckardi A. Mor., one 9; 132. P. unicolor /.; 133. Psen pallidipes
Pz., one $; 134. Sceliphron destillatorium ///., very freq.; 135. S. omissum Kodi,
rather rare; 136. 5S. spirifex Z., several $; 1347. Tachysphex nitidus Spzv.; 138. T.
rufipes Arch. (4) Zenthredinidae: 139. Allantus viduus Ross?; 140. Arge cyaneo-
crocea /érst.; 141. Athalia rosae Z., var. cordata Lep.; 142. Cephus (Philoecus)
pareyssei Spex.; 143. Emphytus balteatus AZg.; 144. Macrophya diversipes Scér. ;
145. M. neglecta Alg.; 146. M. rustica Z.; 147. Tenthredopsis austriaca Anw. ;
148. T. dorsalis Zep.; 149. T. raddatzi Anw., var. vittata Anw.; 150. T. thomsoni
Knw., var. femoralis S/eph.; 151. T.thomsoni Avw., var. nigripes Knw. (/) Vespidae:
152. Eumenes mediterranea Archd.; 153. E. pomiformis /.; 154. Odynerus alpestris
Sauss.; 155. O. bidentatus Lep.; 156. O. dantici Ross?; 157. O. floricola Sauss. ;
158. O. levipes Shuck.; 159. O. modestus Sauss.; 160. O. parietum Z.; 161. Polistes
gallica Z.

250 ANGIOSPERMAE—DICOTYLEDONES

XXXK. ORDER AMPELIDEAE H.B.K.

Small, green, but fragrant flowers, homogamous or protandrous.

187. Ampelopsis Michx.
Protandrous flowers with concealed nectar, secreted at the base of the ovary.

622. A. quinquefolia Michx.—Kirchner (‘Flora v. Stuttgart,’ p. 362) states
that the nectar is secreted in minute drops under the base of the ovary. After the
flower has opened the green petals become completely reflexed, while the five
stamens become erect, and their anthers dehisce introrsely. These then turn their
pollen-covered surfaces upwards, and at this stage project about 1 mm. beyond the
still immature stigma. The stigma does not become receptive until the petals and
stamens have dropped off.

VisiTors.—Kirchner observed the honey-bee. I myself observed Lucilia caesar
Z£. at work on the anthers. According to Kerner (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II,
p. 206), the flowers are much visited by bees, attracted by an odour which is
imperceptible to us. Plateau also noticed Apis in Belgium.

188. Vitis L.

Flowers homogamous, with exposed nectaries. Perhaps there is also wind-
pollination. Focke (Abh. natw. Ver., Bremen, xiv, 1897, p. 302) states that some
of the species are androdioecious. According to Beach (Bot. Gaz., Chicago (IIl.),
xvii, 1892), self-pollination frequently takes place in the still unopened flowers.

623. V. vinifera L. (Kirchner, ‘Flora v. Stuttgart,’ p. 361; Rathay, see
Bibliography ; Kronfeld, Ber. D. bot. Ges., Berlin, vii, 1889, p. 42 ; Knuth, ‘ Bloemen-
biol. Bijdragen.’}—The inconspicuous, small, yellowish-
green flowers of this species attract insects by their
delightful odour. At the base of the ovary, between
the filaments, are five (more rarely six) yellow, fleshy
nectaries. The flowers, as is well known, open by the
five (rarely six) green petals becoming detached at
the base, and falling off in the form of a cap. The
five (or six) stamens spread out, and the upper sides
of the anthers become covered with pollen. The
Sr ee ees stigma matures simultaneously, but, according to

Waakarey ae neckag. Kirchner, remains receptive afler the anthers have
withered. As it is at a lower level than the anthers,
automatic self-pollination is possible, and Kirchner says that it is effective.

Rathay asserts that the five nectaries are non-functional, but according to
Delpino, they secrete freely. Portele says that the stigmatic secretion is strongly
saccharine, but Rathay was only able to find traces of grape-sugar. The latter
authority considers the vine to be anemophilous—because he proved that the wind
can carry away individual pollen-grains from the dehisced anthers (‘ Geschlechtsver-
haltnisse der Reben,’ I, pp. 31 et seq.)—and entomophilous as well—for he observed

AMPELIDEAE 251

(op. cit., II, pp. 16 et seq.) twenty-seven different insect visitors (vzde znfra). These
appeared on specially warm days, and a 9, belonging to one of the four species
of Halictus, was loaded with large balls of pollen derived entirely from grape-flowers.
Kirchner (‘ Uber einige irrtiimlich fiir windbliitig gehaltene Pflanzen,’ Jahreshefte Ver.
Natk., Stuttgart, xlix, 1893, pp. 98 et seq.) points out that the possibility of transport
of pollen-grains by the wind does not prove that anemophily takes place to any
considerable extent, for with the wind blowing steadily in one direction, a pollen-
grain took 200 hours to reach a stigma tolerably near. And he asserts that this is
in direct opposition to all our experimental knowledge of undoubtedly wind-pollinated
plants, while the whole structure of the vine-flower as regards stigma and pollen
is equally contradictory of anemophily. At the time when the stigma is receptive,
it is covered with short papillae, and secretes an abundant supply of glistening
stigmatic fluid, which is well adapted to hold any pollen-grains that reach it, but
not at all fitted for catching grains, its surface being much too small for such a
purpose. Besides, not a single undoubtedly wind-pollinated flower is known with
a stigma secreting a sticky fluid. Kirchner goes on to say that the pollen is but
slightly coherent, the individual grains possessing a smooth extine, not beset with
oil-droplets. It is produced in remarkably small quantity for a supposed anemo-
philous plant, and is by no means readily blown away from the dehisced anthers.
But were the vine-flower wind-pollinated, the opposite might be expected to a marked
extent, since the filaments are rigid, and the anthers firmly and immovably attached
to them. The inconspicuousness of the small flowers is counterbalanced by their
extreme fragrance, which would undoubtedly suffice to attract numerous insects,
if the booty offered were of proportionate amount. But the quantity of pollen is
small, and, according to all available accounts, nectar never appears to be secreted—
at least in Central Europe—a fact that may well restrain the clever bees from
visiting the flowers at a time when so many other copious sources of nectar are
available. This deficiency, however, the last piece of evidence supporting the theory
of anemophily, is not absolute, for when an observer of the rank of Delpino asserts
that nectar is abundantly secreted by the five glands at the base of the ovary, we
are obliged to conclude that in warmer regions than ours the vine behaves dif-
ferently, and that its flowers do actually produce nectar. According to Rathay,
the glands in question are the odour-producing organs. Kirchner is confirmed by
Delpino’s account in a view long entertained by him, i.e. that in our climate the
vine has lost the power of producing nectar which it formerly possessed. It is
really a native of warmer regions, but has been cultivated for the sake of its fruit
to the utmost possible northern limit.

The fertilization of the hermaphrodite flowers obviously results from automatic
self-pollination, for it is specially favoured by warm still weather. Besides this
autogamy, there is no doubt that geitonogamy frequently takes place (Kerner, ‘ Nat.
Hist. Pl.,’ Eng. Ed. 1, II, p. 326), the stigmas receiving pollen from neighbouring
flowers. The position and direction of the stigmas do not alter, but the filaments
elongate, and curve sufficiently to enable the transfer of pollen to adjacent flowers.
Cross-pollination is chiefly effected by insects, as already stated, but it is quite

possible that the wind sometimes carries the loose pollen from one flower to others
which are not far distant.

252 ANGIOSPERMAE—DICOTYLEDONES

The flowers of the vines I examined at Kiel were all thickly covered with yellow
pollen, from which circumstance it appeared to me to be likely that autogamy and
geitonogamy may occasionally be effected by the wind, while xenogamy may be
brought about by insects.

Rathay states that cultivated vines may be gynodioecious or andromonoecious,
and wild ones dioecious with pseudo-hermaphrodite pollen and fruiting flowers.
According to Focke (Abh. natw. Ver., Bremen, xiv, 1897, p. 302) the hermaphrodite
form of Vitis vinifera Z. is perfectly fertile in itself; in the case of V. cordifolia
Mchx. the pollen of the male form is necessary for the production of good fruit.
This latter species is therefore almost dioecious.

Visrrors.—At Kiel I observed the honey-bee and B. terrester Z., po-cltg., as they
flew from flower to flower. The humble-bees in particular were so active that
they constantly came back again, though repeatedly driven away by the owner of
the vines, who feared that the flowers might be injured. Kronfeld (Ber. D. bot. Ges.,
vii, 1889) also saw numerous honey-bees in a garden at Ober-St.-Veit.

Rathay has observed the largest number of visitors, as follows (‘Die Ge-
schlechtsverhaltnisse der Reben,’ Part 2, 1889, pp. 17-23).—

A. Coleoptera. 1. Adrastus humilis Hr.; 2. Agriotes ustulatus Schaller ;
3. Anaglyptus mysticus Z.; 4. Anaspis pulicaria Cos/a.; 5. Clytra musciformis Goze;
6. Clytus figuratus Scop.; 7. C. ornatus Herédst.; 8. Danacea nigritarsis Adés?. ;
g. Dasytes plumbeus JZi//.; 10. Adoxus obscurus Z., var. vitis ad7.; 11. Limonius
lythrodes Germ.; 12. Malachius elegans Oliv.; 13. M. geniculatus Germ.; 14. Meli-
gethes brassicae Scop.; 15. Nacerdes austriacus Ggd.; 16. Notoxys cornutus Faér. ;
17. N. monoceros Z.; 18. Oxythyrea funesta Poda.; 19. Phyllopertha horticola Z. ;
20. Spermophagus cardui Svev.; 21. Epilachna globosa Schneid. B. Diptera. 22.
Sciara sp. C. Hymenoptera. 23. Andrena sp. 9; 24. Apis mellifica Z.;
25. Halictus albipes /., var. affinis Schenck; 26. H. morio /.; 27. H. villosulus &.
D. Hemiptera. 28. Two undetermined specimens.

Rathay also gives (op. cit.) the following visitors for different varieties of the
vine.—

A. Coleoptera. 1. Adrastus humilis Z7., 1 on ‘Zimmet-traube’; 2. Anaspis
pulicaria Cos/a., 2 on ‘Zimmet-traube’; 3. Clytra musciformis Goeze, 1 on ‘ Zimmet-
traube’; 4. Clytus figuratus Scop., 6 on ‘Zimmet-traube’; 5. Danacea nigritarsis
Kist., 2 on ‘Zimmet-traube’; 6. Dasytes plumbeus J/zd//., 1 on ‘ Zimmet-traube,’
1 on V. riparia; 7. Adoxus obscurus Z., var. vitis /adr., 1 on ‘Zimmet-traube’ ;
8. Limonius lythrodes Germ., 1 on ‘Zimmet-traube’; 9. Malachius geniculatus Germ.,
2 on ‘Zimmet-traube,’ 3 on V. riparia; 10. Meligethes brassicae Scop., 3 on blue
Kardaka (V. vinifera), 10 on ‘Zimmet-traube’ (V. vinif.); 11. Nacerdes austriaca
Ggé., 5 9 and 24 on ‘Zimmet-traube’; 12. Oedemera lurida A/arsh., 1 on V. riparia;
13. Oxythyrea funesta Poda., 1 on ‘Zimmet-traube’; 14. Phyllopertha horticola Z.,
2 on ‘ Zimmet-traube’; 15. Spermophagus cardui S/ev., 3 on V. riparia, 2 on ‘ Zim-
met-traube’; 16. Subcoccinella 24-punctata Z., 1 on ‘Zimmet-traube. B. Diptera.
17. Sciara sp., 2 on ‘Zimmet-traube’; 18. Syritta pipiens Z. 1 on V. riparia.
C. Hymenoptera. 19. Halictus albipes /., var. affinis Schenck, 2 on ‘ Zimmet-
traube’; 20. H. morio /., 1 on ‘ Zimmet-traube.’

Supplementary list.—
A. Coleoptera. 1. Coccinella bipunctata Z., 1 on V. riparia; 2. Agriotes
ustulatus Scha//., 1 on V. vinifera; 3. Anaglyptus mysticus Z., 1 on V. viniferas

4. Anaspis melanostoma Cos‘, 1 on V. rupestris candicans; 5. Ceutorrhynchus
suturalis Fadr., 1 on V. riparia; 6. Cis hispidus Pays. 1 on V. riparia; 7. Clytus

SAPINDACEAE 253

figuratus Scop. 1 on V. vinifera, 1 on Taylor-seedling; 8. C. ornatus Hés¢., 2 on V.
vinifera ; 9. Coccinella septem-punctata Z., 1 on riparia; 10. Dasytes plumbeus JZi/.,
1 on V. riparia, 1 on V. cordifolia rupestris; 11. Clytra affinis Hed/w., 1 on Clinton;
rz. Limonius bructeri Panz., 1 on V. riparia; 13. Malachius aeneus Z., 2 on V.
riparia; 14. M. elegans Odv., 3 on V. riparia, 2 on V. rupestris, 1 on V. vinifera,
1 on Othello (V. riparia vinifera, American variety), 1 on Taylor-seedling (V. riparia
labrusca), 2 on Clinton (V. riparia labrusca); 15. Meligethes brassicae Scop., 3 on V.
riparia, 1 on V. arizonica; 16. M. pedicularis Gy//., 1 on V. riparia; 17. Nacerdes
austriacus (y//., 4 on Clinton, 3 on Solonis (V. riparia, V. rupestris, V. candicans) ;
18. Notoxys cornutus /'aé7., 6 on V. vinifera; 19. N. monoceros L., 4 on V. vinifera;
20. Oedemera lurida Marsh., 1 on V. riparia; 21. Omophlus longicornis Bert, 1 on
V. riparia; 22. Oxythyrea funesta Poda., 1 on V. riparia; 23. Spermophagus cardui
Stev., 1 on V. riparia, 1 on V. vinifera, 1 on V. Solonis. B. Diptera. 24. An
Anthomyid, 1 on V. riparia; 25. Pipizella virens Fadr., 1 on V. riparia, r on V.
Solonis; 26. Syritta pipiens Z., 4 on V. riparia, 4 on V. rupestris. C. Hymeno-
ptera. 27. Andrena sp.? 9, 1 on V. vinifera; 28. Apis mellifica Z., 5 on V. riparia,
8 on V. vinifera; 29. Halictus morio F. 9, 1 on Clinton; 30. H. sp. 9, 1 on Clinton;
31. H. villosulus A77d. 9. D. Hemiptera. 32. Two undetermined specimens on
V. vinifera.

XXXI. ORDER SAPINDACEAE.

1. SuB-oRDER SaPINDEAE. ’

The large flowers of the species cultivated in Europe are rendered very con-
spicuous by their aggregation into large candelabra-like inflorescences of considerable
size. The flowers belong to class B, for nectar is secreted and concealed in their
bases.

189. Aesculus L.

624. A. Hippocastanum L. (Sprengel, ‘Entd. Geh., pp. 209-14; Herm.
Miller, ‘Fertilisation,’ pp. 164-6; Kirchner, ‘Flora v. Stuttgart,’ p. 349; Knuth,
‘“Grundriss d. Bliitenbiol.,’ pp. 35-6, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 50, ‘ Bloe-
menbiol. Bijdragen’; Hildebrand, ‘D. Geschlechts-Vert. b. d. Pfl.,’ pp. 11, 26;
Beyer, ‘D. spont. Bewegung. d. Staubgefiasse u. Stempel’; Martelli, Nuovo Giorn.
bot. ital., Firenze, xx, 1888, pp. 401-4; Ogle, Pop. Sci. Rev., London, ix, 1870,
p- 54; K. Fr. Jordan, Ber. D. bot. Ges., Berlin, v, 1887; Focke, Verh. bot. Ver.,
Berlin, xxxi, 1889, pp. 108-12.)—The species is coenomonoecious. It is remarkable
that Sprengel describes the hermaphrodite white flowers as protandrous, though
they are really protogynous, as first correctly pointed out by Hildebrand. The two
upper petals are the largest, while the lowest one is the smallest. They possess
nectar-guides which are at first yellow, afterwards assuming a carmine hue. Focke
says that this coloration renders the inflorescences more conspicuous as a whole.
Nectar is secreted in the bottom of the calyx, between the claws of the upper petals
and stamens. It is protected by the horizontal position of the flowers, the folding of
the petals, and woolly hairs on the petals and stamens. Hermaphrodite, male, and
female flowers are associated in the same inflorescence. In hermaphrodite flowers
the stamens are curved downwards so long as the anthers are immature, while the
style projects horizontally from the flower. In the second stage, the stamens with
their ripe anthers curve upwards, but after shedding their pollen return to their
first position. In harmony with the protogyny of the hermaphrodite flowers is the

254 ANGIOSPERMAE—DICOTYLEDONES

fact that the first flowers of an inflorescence to open are all purely male (with
vestigial pistils), while it usually happens that in the lower part of the inflorescence,
some flowers are actually female, since their anthers fall off without dehiscing,
although the pollen-sacs are full of pollen-grains (Miiller). Martelli states (op. cit.)
that the lowest flowers are the only fertile ones, and that not more than two to
four of them produce seeds, these being the fourth (rarely the third) to the seventh
in regular succession, counting from the base of the cyme.

Warnstorf (Schr. natw. Ver., Wernigerode, xi, 1896, pp. 1-12) gives a similar
account of the distribution of the sexes, The lower flowers of the panicle are male,
and are the first to open; towards the middle isolated pseudo-hermaphrodite pollen-
flowers, with no style and a sessile stigma are often to be found. The upper flowers
are hermaphrodite and protogynous, with projecting styles; their ovaries are beset
with stalked red glands of large size. The smooth pollen-grains are vermilion in
colour and ellipsoidal, with several longitudinal furrows; size about 20 » broad
and 37-40 » long.

The dimensions of the flowers correspond with those of the chief visitors,
humble-bees, which when alighting at once settle in the most convenient position

Fic. 78. Aesculus Hippocastanum, L. (After Herm. Miiller.) 1. Male flower in section. 2. Herm-
aphrodite flower in the first (male) condition, seen obliquely from the front. 3. The same in the second
(female) condition, in section. a, anthers; 7,nectary; ov, ovary; ov', vestigial ovary; /, petals; s, sepals;
st. stigma.

for sucking, and in so doing touch either the stigma or the anthers with the under-
side of the abdomen, so that cross-pollination is always effected. The other bees

observed by Herm. Miiller (Apis, Eucera, Osmia rufa Z., Halictus rubicundus Cfr.,
Andrena) do not correspond with the flowers in dimension.

VisiTors.—The height of the trees makes it very difficult to observe visitors,
but I was able to carefully note the following actively engaged on the flowers.—

A. Diptera. (a) Muscidae: 1. Musca domestica Z., skg. from the side without
touching stigma or stamens; 2. Scatophaga merdaria L., do.; 3. S. stercoraria Z., do.
(4) Syrphidae: 4. Syritta pipiens Z., skg. from the side as before, also po-dvg., but
not touching the stigma; 5. Syrphus balteatus Deg.,as Musca. B. Hymenoptera.
Apidae: 6. Apis mellifica Z. ¥, skg. while hanging on the filaments from below,
and therefore touching neither stigma nor anthers ; sometimes also po-cltg.; 7. Bombus
lapidarius Z. 9, skg. legitimately and effecting pollination; 8. B. terrester Z. 9
and ¥, do.

Loew saw the honey-bee, skg. and po-cltg., in the Berlin Botanic Garden.

Alfken and Hopper (H.) noticed the following Apzdae at Bremen:—1. Apis
mellifica Z. ¥; 2. Bombus terrester Z. 93; 3. Podalirius retusus Z., var. obscurus
Friese 9 (H.). All freq., skg.; 1 and 2 also po-cltg.

SAPINDACEAE 255

625. A. carnea Willd.—lIn this species again, according to Martelli (op. cit.),
the flowers are not all fertile. Such are to be found, singly or in pairs, among
sterile ones on the same inflorescence. And as before, only the lower part of the
whole inflorescence is fertile. Sterile flowers occur in A. flava, but the great
majority are fertile. There is no difference between the parts of the entire in-
florescence, or its component branches. Focke (Abh. natw. Ver., Bremen, xiv, 1897,
p- 302) states that the species of Aesculus (Pavia), cultivated in Germany, are andro-
monoecious, like the horse-chestnut. For the production of fruit, cross-pollination
(by humble-bees) is generally necessary. .

626. A. Pavia L. (=Pavia rubra Zznk). (Warnstorf, Schr. natw. Ver., Werni-
gerode, xi, 1896.)—In this species the lower flowers of the branches of the panicle
are hermaphrodite and fertile, and the next above them pseudo-hermaphrodite; or
all the flowers may be pseudo-hermaphrodite. The two posterior and larger petals
possess a yellow nectar-guide, which later on assumes an intensely red colour. The
stamens are about as long as these petals. The pollen-grains are vermilion in
colour, ellipsoidal, with longitudinal furrows; on an average they are 25-30 mw broad
and 42 p» long.

Visirors.—Alfken observed the following Apzdae at Bremen.—1. Apis mellifica
ZL. %; 2. Bombus hortorum Z.¢; 3. B. lucorum Z.9; 4. B. muscorum F. 9; 5. B.

ruderatus /. 9; 6. Psithyrus barbutellus A. 9, skg.; 7. P. vestalis Pourcr. 1-5 skg.
and po-cltg.

627. A. rubicunda Lodd. (‘D. Geschlechts-Vert. b. d. Pfl.,’ pp. 26-7.)—The
species is andromonoccious with protogynous hermaphrodite flowers. While Hilde-
brand found all the earliest flowers of the panicle to be purely male, Kirchner (‘ Neue
Beob. iti. d. Bestéubungseinricht. einheim. Pfl.,’ p. 31) says they are hermaphrodite,
occurring chiefly in the lower part of the inflorescence, where male flowers greatly
predominate as a tule.

628. A. flava Ait—DMartelli found most of the flowers to be fertile in this
species (cf. A. carnea). Focke saw many blossoms that had been perforated by
Bombus terrester.

629. A. macrostachya Michx. (Kirchner, ‘Beitrage,’ p. 30; Knuth, ‘ Bloe-
menbiol. Bijdragen.’)—The species is andromonoecious, with protandrous hermaphro-
dite blossoms, which are perhaps moth flowers. They are horizontal, white in colour
except for the red anthers, and exhale an odour of lilies. Nectar is secreted outside
the bases of the upper stamens. The tubular gamosepalous calyx is 7-8 mm. long.
The petals are slender, with long claws, and at first 12 mm. in length. To begin
with, the stamens are of the same length, but afterwards project 20-25 mm. out
of the corolla, when their anthers dehisce successively. After these have shed their
pollen, withered, and curved downwards, the stigma becomes fully mature, and at
this stage the style is over 30 mm. long. In male flowers the pistil is vestigial.

Visitors.—Kirchner observed the honey-bee, but supposes from the structure,
colour, and odour of the flowers, that nocturnal hawk-moths are the real pollinators.
In the Kiel Botanic Garden I too saw the honey-bee, skg., and also Bombus hortorum
L.%. These two insects did not fly, but crept from flower to flower.

256 ANGIOSPERMAE—DICOTYLEDONES

190. Melianthus L.

630. M. major L.—Francke describes the flowers of this species as protandrous
(‘ Beitrage z. Kennt. d. Bestaubungseinricht. d. Pfi.’).

2. SuB-oRDER ACERINEAE.

This suborder is represented in Europe only by the genus.

191. Acer L.

Several or many of the small greenish-yellow flowers are crowded together into
inflorescences, and thus rendered conspicuous. Some species blossom before the
leaves unfold, by which the same end is attained to a marked degree. Nectar is
secreted by a thick fleshy central disk, and is fully exposed. The flowers therefore
belong to class E. They are usually monoecious, rarely dioecious.

631. A. platanoides L. (Herm. Miiller, ‘ Weit. Beob.,’ I], pp. 212-13 ; Kirchner,
‘Flora v. Stuttgart, p. 351; Wittrock, Bot. Centralbl., Cassel, xxv, 1886, p. 553
K. Fr. Jordan, Ber. D. bot. Ges., Berlin, v, 1887 ; Knuth, ‘Grundriss d. Bliitenbiol.’ ;

Fic. 79. Acer, Z. (After F. Pax.) Flowers of various species in longitudinal section. 4 8, BQ,
of A.Negundo. Cand D, do. of A. Pseudo-Platanus. £. A. Hookeri. F. A. campestre.

Warnstorf, Schr. natw. Ver., Wernigerode, xi, 1896.)—The flowers appear betore the
leaves in this species. The stamens, usually eight in number, arise from pits in
the fleshy disk, which is covered with minute drops of nectar lying quite exposed.
Wittrock states that the inflorescences are of five kinds, according to the distribution
of the male and female flowers on the inflorescence. These are composed, respec-
tively, of:—1. female flowers exclusively; 2. the flowers first developed are female,
and the later ones male; 3. the earliest flowers are male (at the apex), then follow
male and female, and the last to open are mostly male; 4. male flowers are
first developed, and then female; 5. all the flowers are male. Only one of these
varieties of inflorescence is found on most trees, but two, or even three of them, may
be associated on the same tree in exceptional cases. The commonest kind is 2
(about 40 % in the trees examined by Wittrock), then follow in order 4 (22 %),
5 (12%), 3 (4%), 1 (not quite 1 %).

The female flowers possess stamens that appear normal, as do the numerous
pollen-grains contained in the anthers, which, however, never dehisce. The filaments

SAPINDACEAE 257

are considerably shorter than those of male flowers, in which they are so long that
the anthers about reach the tips of the petals: they surround a vestigial pistil. After
fertilization the female flowers close, the sepals and petals becoming erect. Warnstorf
describes the pollen-grains as pale yellow in colour, ellipsoidal, with three longitudinal
grooves, very delicately striated, about 50 » long and 20 « broad.

Visitors.—Herm. Miiller observed the honey-bee.

632. A. campestre L.—The greenish flowers of this species open at the
same time that the leaves unfold, they are much less conspicuous than those of
A. platanoides, with which, according to Wittrock, they entirely correspond as regards
mechanism and distribution of sexes.

Vistrors.—Alfken observed 3 Apzdae at Bremen :—1. Andrena nigro-aenea K. 9;
2. A. trimmerana K. 9; 3. Apis mellifica Z. ¥.

H. de Vries saw Apis mellifica Z.¥, freg., in the Netherlands (Ned. Kruidk.
Ach., Nijmegen, 2. ser., 2. deel, 1875).

633. A. Pseudo-Platanus L.—This species does not blossom till the leaves
have unfolded. Wittrock states that the flowers agree essentially with those of
A. platanoides, except that purely male or purely female inflorescences have not so
far been observed. According to Jordan, there is a nectar-cover, in the form of
white hairs at the bases of the stamens.

Warnstorf (Schr. natw. Ver., Wernigerode, xi, 1896) gives the following
account.—Inflorescence a racemose panicle; flowers markedly protandrous. Lower
floral axes branched, with male and pseudo-hermaphrodite fruiting-flowers; the
middle ones being either almost purely female, and the upper ones mixed male and
female, or the former mixed female and male, and the latter purely female. The
filaments of the male flowers project far beyond the petals, while those of the female
flowers are very short, and do not exceed the petals in length.

VistTors.—Herm. Miiller observed the following.—

A. Diptera. Syrphzdae: 1. Eristalis arbustorum Z., skg.; 2. E. tenax Z., do.;
3. Syrphus ribesii Z., po-dvg. B. Hymenoptera. <Afzdae: 4. Andrena albicans
Mill. 9, skg.; 5. Anthophora aestivalis Pz. 9, do.; 6. Bombus hortorum JZ. 9, do. ;
4. B. lapidarius Z. 9, do.; 8. B. rajellus K. 9, do.; 9. B. terrester Z. 9, do.; ro.
Melecta luctuosa Scop. 9, do.; 11. Osmia emarginata Zep. 9, do.; 12. Psithyrus bar-
butellus XK. 9, do.

Loew noticed the following in the Berlin Botanic Garden.—

A. Diptera. &2dzonzdae: 1. Bibio hortulanus Z. 9, skg. B. Hymenoptera.
Apidae: 2. Apis mellifica Z., skg.

Friese saw the following Apzdae in Hungary.—1. Andrena bucephala S/p4.,
freq.; 2. A. gwynana K., 2nd generation; 3. A. mitis Per.; 4. A. rufula Pér.; 5. A.
trimmerana K.; 6. Nomada alternata K.; 7. N. bifida Zhs.; 8. N. ruficornis L.;
g. N. succincta Pz.

634. A. dasycarpum Ehrh.—In this North American species, occasionally
cultivated in Europe, the flowers are arranged in dense clustered inflorescences, and
open long before the leaves unfold. Kirchner (‘Flora v. Stuttgart,’ p. 352) says that
the diameter of the male flowers is only about 2mm. The stamens project 6 mm.
from the calyx, which is yellowish with a reddish margin, and 4mm. long. No
pistil is visible. The female flowers are compressed like the ovary; the calyx is

DAVIS. IL Ss

258 ANGIOSPERMAE—DICOTYLEDONES

5 mm. by 2 mm. broad, and 3-4 mm. long. Vestigial stamens surround the hair-
covered ovary, but their anthers do not dehisce.
Visrrors.—Kirchner observed the honey-bee.

635. A. rubrum L.—The flower mechanism of this species, which is also
native to North America, essentially resembles that of A. dasycarpum (Kirchner,
op. cit.).

636. A. tataricum L.—Francke states that this species, indigenous to Carniola
and Russia, bears male flowers with vestigial ovaries, and female flowers with vestigial
stamens. The hermaphrodite flowers are crossed with pollen from male flowers,
since their own matures late.

XXXII. ORDER ANACARDIACEAE Lindl.

192. Rhus Tourn.

Hermaphrodite or unisexual greenish flowers, with exposed nectar secreted
by the receptacle. The hermaphrodite flowers are homogamous or often pro-
tandrous (Darwin).

FIG. 80. Rhus Cotinus, L. (after Herm. Miller). (1) A purely male flower. (2) A hermaphrodite
flower. (3) A purely female flower. s, sepals; #, petals; a, anthers; st, stigmas; #, nectary.

637. R. Cotinus L. (Herm. Miiller, ‘Fertilisation,’ pp. 166-7; Schulz,
‘Beitrage, II, pp. 62-4.)—In this species, according to the observations which
Hermann Miller made on garden plants, numerous intermediate stages between
purely male, hermaphrodite, and purely female flowers are present on the same
stock. The purely male flowers, being the largest, and opening the widest, are
the most conspicuous. The female ones are the smallest, and open least, so that
they are the most inconspicuous. The flowers are consequently visited in the
order best adapted to secure cross-pollination. Schulz, who studied wild plants
in the South Tyrol, also observed three distinct forms differing in the development
of stamens and carpels. The plants, however, were dioecious, like the cultivated
varieties at Halle. Schulz states that there are two series of female flowers, one
(diameter 31-4 mm.) possessing anthers shaped like those of the male flowers,
but with abnormal pollen-grains. The stamens of the second form (diameter
3-34 mm.), on the other hand, are quite vestigial. The diameter of the male
flowers is 5-6 mm. According to Schulz, the distribution of the flowers is dioecious,
or more rarely monoecious. In all cases there is a secreting yellow or orange-

LEGUMINOSAE 259

coloured disk in the base of the flower, and the nectar lies exposed, so that it is
accessible even to very short-tongued insects. Hermann Miiller states that visitors
chiefly effect cross-pollination in hermaphrodite flowers, owing to the fact that
there is a tolerably large interval between anthers and stigmas.

Vistrors.—In Westphalia Herm. Miiller chiefly observed Diptera and short-
tongued Hymenoptera, but only very few beetles, which do not seem to care much
for the dull yellow colour of the flowers. Schulz, on the other hand, saw numerous
beetles in the Tyrol, as well as Diptera, wasps, ichneumon-flies, and other short-
tongued Hymenoptera. These were so numerous that, on one small bush, he
captured 350 individuals belonging to about fifty species in half an hour.

Hermann Miiller’s list is as follows.—

A. Coleoptera. Dermestidae: 1. Andrenus pimpinellae /., nect-lkg. B. Di-
ptera. (¢) Muscidae: 2. Calliphora erythrocephala A/g.; 3. Lucilia cornicina F., skg.;
4. Sarcophaga carnaria Z. (6) Syrphidae: 5. Helophilus floreus Z., very freq., skg.
and po-dvg.; 6. H. pendulus Z., do.; 7. Syritta pipiens Z., do. C. Hymenoptera,
(2) Apidae: 8. Andrena albicans AZi/. 9, po-cltg.; 9. Apis mellifica Z. ¥, skg. ;
10. Halictus sexnotatus K. 9, do.; 11. H. sexstrigatus Schenck 9, do. (0) Sphegidae:
12. Gorytes campestris Z., nect-lkg.; 13. Oxybelus uniglumis Z., do. (c) Zenthre-
dinidae: 14. Allantus marginellus /., nect-lkg. (d) Vesprdae: 15. Eumenes pomi-
formis /., nect-lkg.; 16. Odynerus sinuatus /., do.; 17. O. spinipes Z., do.

638. R. typhina L. (Herm. Miiller, ‘ Fertilisation, p. 167.)—The dioecious
flowers of this species are moderately conspicuous, and secrete nectar that is
accessible to all.

Visitors. — Herm. Miiller observed certain nect-skg. bees— Apis, Prosopis
communis JVy/. 9 and é, skg.—and a Neuropterid, Panorpa communis Z., skg.

XXXIIL ORDER ZEGUMINOSAE.
1. Sus-orDER PapiLionacEak,

LitrraTuRE.—Sprengel, ‘Entd. Geh.,’ pp. 358-9; Herm. Miller, ‘ Fertilisation,’
pp. 167-220; Delpino, ‘Sugli appar. d. fecondaz. nella piante autocarp.,’ pp. 24-8,
“Ult. Oss.,” pp. 39-66; Kirchner, ‘Flora v. Stuttgart,’ pp. 467-8; Loew, in Engler
and Prantl, ‘ D. nat. Pflanzenfam.,’ III, 3, pp. 88 et seq.; Knuth, ‘ Flora v. Schleswig-
Holstein,’ p. 231, ‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 53-5, ‘Grundriss d. Bliiten-
biol.,’ pp. 40-2.

C. K. Sprengel long ago described the uses of the individual parts of the
papilionaceous flower. But he did not understand the various special adaptations
presented, and for an explanation of these we are indebted to F. Delpino and
Hermann Miiller. In the following descriptions the admirable accounts of the
latter will be followed so far as possible. All the papilionaceous species indigenous
to Germany bear homogamous, rarely slightly protandrous bee flowers’ (in the
wider sense).

The characteristically shaped, usually brightly coloured flowers are often aggre-
gated into very conspicuous racemose, or capitulate inflorescences, admirably adapted

1 For this reason the Hymenoptera will be placed first in some of the lists of visitors.
$2

260 ANGIOSPERMAE—DICOTYLEDONES

to attract insects. Their power of attraction is often enhanced by more or less
marked fragrance. The gamosepalous calyx holds the petals erect, in a position
adapted to the insect visitors. The vexillum serves in the bud to cover the inner
parts of the flower. In the open blossom it is erect and acts as a signboard;
it often possesses nectar-guides in the form of colour-streaks. The vexillum is also
used as a support by bees, which press their heads against it as they suck. The
alae have a threefold function.—1. They serve as a resting-place for bees; 2. They
act as levers, depressing the carina, so that the stigma and pollen protrude during
insect-visits, and are brought into contact with the under-sides of the visitors
(bees); 3. They hold the carina in place as regards stamens and pistils, and
after the departure of a visitor help these parts to regain their original position.
The carina is a protective structure, sheltering the stamens and pistil from rain,
and keeping away unbidden guests (Lepidoptera and flies). When all ten filaments
cohere, the flowers afford only pollen, but when the upper one is free, there is a
slit on either side of it, leading to the nectar secreted inside the bases of the
stamens. The closed or split cylinder formed by the filaments envelops the pistil,
of which the style is usually upwardly curved at the tip, and projects somewhat
beyond the anthers, so that the stigma first protrudes from the carina when an
insect visits the flower, and first touches its under-side. Cross-pollination therefore
results if the insect has previously visited another flower of the same species. In
some species the stigma is completely covered by the pollen of the same flower,
but self-fertilization does not usually follow, for it only becomes receptive after
visitors have rubbed against its papillae.

In our native Papilionaceae Delpino distinguishes four types of flower
mechanism connected by transitional forms.—

1. Simple valvular arrangement. Stamens and pistil project from the carina
as long as the pressure of the bee continues, and then return to their former position.
Such flowers permit of a number of effective visits.

(a) Nectar: Melilotus, Trifolium, Galega, Onobrychis, Astragalus, Oxytropis,
Phaca, Ornithopus, Hedysarum.
(2) Enclosed sap, reached by boring: Cytisus (some of the species are
transitional to 3 (a2)).
2. Explosive arrangement. Stamens and pistil suddenly spring out of the
carina. Such flowers permit of only one effective visit.
(2) Nectar present: Medicago.
(2) No nectar.
(a) The ventral surface of the bee comes into contact with the pollen and
stigma: Genista, Ulex.
(8) The bee is struck on the back by the pollen and stigma: Sarothamnus.
3. Pump arrangement. The thickened ends of the filaments press out the
pollen in successive portions from the tip of the carina. Several insect-visits are
necessary for pollination.
(2) Nectar present: Lotus, Anthyllis, Tetragonolobus, Hippocrepis.
(5) No nectar: Ononis, Lupinus, Coronilla.

LEGUMINOSAE 261

4. Brush arrangement. A brush of hairs on the style sweeps the pollen out

from the tip of the carina. Here again repeated insect-visits are usually necessary
for pollination.

(2) The tip of the style is straight: Lathyrus, Pisum, Vicia, Lens, Robinia.
(2) The tip of the style is helicoid: Phaseolus.

Papilionaceae possessing the brush arrangement are divided by Taubert (in
Engler and Prantl, ‘D. nat. Pflanzenfam.,’ III, 3, p. 92) into two subdivisions,
according as the sweeping apparatus acts exactly in the median plane of the flower
or not. In the former case the pollen is deposited on the ventral surface of the
visitor (‘ pollinazione sternotriba,’ Delpino), e.g. in Vicia Cracca, V. sepium, V. Faba,
and also in Pisum sativum, which shows a combination of the pump and brush
arrangements. In the second case the brush emerges laterally and obliquely, not
in the median plane of the flower, so that the pollen can only be deposited on
the right or left side of the body of the visitor (‘pollinazione pleurotriba,’ Delpino).
An indication of a unilateral mechanism of the kind occurs in some species
of Lathyrus (L. sylvestris, L.
grandiflorus), while other species
of the same genus (e.g. L.
pratensis) have a median polli-
nating apparatus. The asym-
metry is more marked in species
of Phaseolus (P. vulgaris, P.
multiflorus), in which it is con-
ditioned by the helicoid spiral

of the tip of the style. In P. Fic. 81. Aptos tuberosa, Moench (after Taubert and Loew).

Caracalla this coiling of the -4. Flower seen from the side after removal of half the calyx, half
¥ the vexillum, and the right ala (x3). #. Sexual apparatus after
style 1s most pronounced, aS removal of the corolla; the stamens project on the right side and

there are 4-5 turns in the the style on the left (x 4).
spiral.

Apios tuberosa presents a transition to flower mechanisms of other kind, par-
ticularly characteristic of extra-European species. Here, according to Loew (Flora,
Marburg, Ixxiv, 1891), the sickle-shaped tip of the carina is held fast in a cap-
shaped protuberance of the vexillum in such a way as to render impossible the
movements which usually take place in papilionaceous flowers, so that cross-
pollination has to be secured in another way (see Fig. 81).

The species of Erythrina exhibit still another modification in the construction
of the flower. In E. Crista-galli, according to Hildebrand (Bot. Ztg., Leipzig, xxviii,
1870), the flower is so twisted that the pollinating apparatus is upside down, while
at the same time the alae and carina are much reduced. The latter forms a stiff,
immobile sheath, which surrounds the markedly projecting sexual organs above,
and expands below into a nectar-receptacle. Delpino supposed that species of
Trochilus and Nectarinia acted as pollinators. This supposition has been confirmed
by the direct observations of Scott-Elliot, who saw species of Nectarinia on E. caffra
Thunb. He further states that E. Indica Zam. and Sutherlandia frutescens &. Br.
are also ornithophilous. The alae and carina are entirely suppressed in Amorpha

262 ANGIOSPERMAE—DICOTYLEDONES

fruticosa, which, according to Hermann Miiller (‘ Weit. Beob.,’ pp. 244-5), is also
distinguished by its protogyny from the other Papilionaceae, these being typically
homogamous or protandrous (Taubert, op. cit.).

Kuhn states (Bot. Ztg., Leipzig, xxv, 1867) that some genera include species
with cleistogamous flowers, e.g. Arachis Z., Chapmannia Zorr. ef Gray, Hetero-
carpaea PAil., Lesperdeza Rich., and Stylosanthes Swar/s.

The following Papilionaceae have so far been recognized as self-sterile:—
Trifolium pratense, T. repens, T. incarnatum; Phaseolus multiflorus, Lathyrus
grandiflorus, Vicia Faba, Erythrina sp., Sarothamnus scoparius, Melilotus officinalis,
Lotus corniculatus, Cytisus Laburnum (Darwin), Astragalus alpinus (Axell), Wistaria
sinensis (Gentry).

193. Sarothamnus Wimm.

Yellow, homogamous, hymenopterid flowers devoid of nectar; with explosive
mechanism and spirally coiled style. Only one effective visit possible.

639. S. scoparius Koch (=Spartium scoparium Z.). (Darwin, Proc., Linn.
Soc., London, 1867, p. 358; Herm. Miiller, ‘ Fertilisation,’ pp. 195-8, ‘ Weit. Beob.,’
II, p. 2573; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 329-32; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 55-6, 152, ‘ Bloemenbiol. Bijdragen.’) —The
flower mechanism of this species can only be liberated by humble-bees and the
honey-bee ; while smaller and less skilful Apidae, as well as a few Syrphidae and
beetles, are merely found collecting or devouring the pollen of already exploded
blossoms. The mechanism is described by Hermann Miiller somewhat as follows—

Insects are attracted by the large yellow flowers, and although these are
nectarless the vexillum is marked with streaks which converge below so that the
presence of nectar is suggested to insects. When a honey-bee settles upon a flower
that has not yet been visited, it grasps the alae with its middle and hind-legs,
and pushes its fore-legs and head under the middle of the vexillum. The alae
are therefore pressed forcibly down, and also the carina, of which the proximal
third is united with them by interlocking folds. The free upper edges of the
carina consequently begin to separate from base to tip, and as soon as they have
half done so, the five shorter stamens, which even in the bud dehisce upwards,
spring up and scatter some of their pollen on the ventral surface of the bee, without
disturbing it in its work. The cleft now rapidly extends towards the tip of the
carina till it reaches the point where the tip of the style presses against the united
part of the carinal petals, when a second and much more violent explosion takes
place. Until then the long style lies like a stretched spring in the carinal cavity,
the extremity of which it fills, while its end is pressed against the tip of the carina.
Scarcely, therefore, have the edges of the carina separated as far as this point,
when the style shoots out, and strikes the back of the bee with its papillose tip.
Immediately afterwards, most of the pollen swept out by the lamelliform part of
the style is thrown upon the back of the insect, while at the same time the five
long stamens, which have long since dehisced, bend inwards and spring out of
the carina. The visitor now frees itself from the style, which has usually coiled
round it, and collects the pollen still adhering to the anthers. This is so abundantly

LEGUMINOSAE 263

present that despite the absence of nectar, and the whip-like action of the style,
the bee continues to visit other flowers.

While the honey-bee is obliged to make considerable efforts before it can effect
explosion, the stronger and heavier humble-bees (Bombus terrester and B. lapidarius)
do this with the greatest ease.

Cross-pollination takes place because the style springs out of the carina a
moment sooner than the stamens, so that the stigma of the second flower visited
is dusted with foreign pollen. Even the first flower visited will very probably be
crossed, though its own pollen lies all around, for the style rolls up to such an
extent that the stigma once more faces upwards, so that subsequent visitors may
still effect cross-pollination. The honey-bee and humble-bees hardly ever settle on
exploded flowers, which are almost exclusively visited by smaller bees, hover-flies,
or flower-beetles. Failing visits from humble-bees or the honey-bee, explosion does
not take place. Darwin states that in this case the flowers remain unfertilized.

Fic. 82. Sarothamnus scopartus, Koch (after Herm. Miller). (1) Flower seen from the side.
(2) The same with vexillum somewhat more erect, to show the nectar-guides; seen from the right front.
(3) The same after removal of the vexillum, seen from above. (4) The same, after removal of the alae.
(4 6) The left ala, seen from the inner side, showing the fold 4, which lies upon a projection (4 ¢ and 67”) of
the carina, (4c) The projection of the carina, seen from the front. (5) Position of the stamens and
pistil in the unexploded flower. (6) Flower after explosion of the short stamens and removal of the
vexillum and alae; seen from the side. (7) Position of the parts after explosion. (8) Staminal tube, split
longitudinally to the right of the posterior stamen (1) and spread out. (9) End of the style with the
stigma (7) seen from the inner side; £/, the plate which scatters the pollen.

VisiTors.—Only strong eutropous bees with a long proboscis (Apis, Bombus,
Eucera) are able to set the flower mechanism in motion. Other bees (mostly
hemitropous), po-dvg. hover-flies, and po-dvg. beetles, can only, as already stated,
plunder flowers that have already exploded.

Hermann Miller (H. M.) in Westphalia, Loew (L.) in Brandenburg, Alfken (A.)
in Bremen, Verhoeff (V.) in Norderney, and myself (Kn.) in Schleswig-Holstein
have observed the following bees, all po-dvg., as legitimate visitors.—

264. ANGIOSPERMAE—DICOTYLEDONES

1. Apis mellifica Z. ¥ (H.M., A., Kn.); 2. Bombus agrorum /”. 9, infrequent
(A., Kn.); 3. B. distinguendus Jor. 9, freq. (A.); 4. B. lapidarius Z. 9 (H.M., V.,
Kn.); 5. B. hortorum Z. (A., Kn.); 6. B. muscorum /” 9, freq. (A.); 7. B. terrester
L. 9 (H.M., V., Kn.); 8. Eucera longicornis Z. 6 (L.).

H. de Vries saw the honey-bee in the Netherlands, and MacLeod noticed Apis,
3 humble-bees, 3 Andrenae, and 3 hover-flies in Flanders. In Dumfriesshire, Apis,
a humble-bee, and several Diptera have been recorded (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 42). In England Saunders observed Eucera longicornis Z. with
its parasite Nomada sexfasciata Pz.

As unbidden guests, Hermann Miiller observed bees (Andrena fulvicrus A. 9,
Halictus zonulus Sm. 9, Osmia fusca Chr. 9), hover-flies (Rhingia rostrata Z.), and
beetles (Anthobium abdominale Gr., A. florale Gr., and Meligethes). In Wiesbaden
Réssler noticed moths — Trifurcula immundella Z., Fidonia famula £sf., and
Threnodes pollinalis Schiff. In Westphalia and East Prussia von Fricken saw
the Curculionids Bruchus villosus /. and Tychius venustus /’. destroying the flowers,
and the Chrysomelids Cryptocephalus vittatus #. and Gonioctena olivacea Forst,,
po-dvg.

194. Spartium L.

640. S. junceum L.—There is an explosive mechanism in this species.

VisiTors.—Delpino (‘ Ult. oss.,’ I) more particularly noticed the carpenter bee,
Xylocopa violacea Z. At Pola Schletterer observed 3 bees—r1. Andrena flavipes Pz.,
2. A. morio Brui/., and 3. the mason bee Megachile muraria Z., the last as ‘one
of the few nectar-loving guests.’

195. Genista L.

Yellow, homogamous, nectarless bee flowers; with an explosive mechanism.
The stamens and pistil come into contact with the ventral surface of the visiting
bees. Only one effective visit is possible. More rarely there is a simple valvular
arrangement,

641. G. tinctoria L. (G. Henslow, J. Linn. Soc. Bot. London, x, 1869;
Herm. Miiller, ‘Fertilisation,’ pp. 188-92, ‘ Weit. Beob., II, p. 257; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 332-3; Knuth, ‘Bl. u. Insekt. a. d. nordfr.
Ins.” pp. 56-7, 152, ‘ Weit. Beob. i. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 232.)—
The yellow flowers of this species are associated in racemes, and devoid of nectar
and nectar-guides. The two pentamerous whorls of stamens, and the projecting
style, are closely surrounded by the carina. The anthers of the four upper stamens
of the outer whorl dehisce in the bud, discharging their pollen into the carina,
while their filaments shrivel. This pollen remains lying above the style, and is
pushed into the tip of the still growing carina by the elongation of the five inner
stamens. Shortly before the vexillum unfolds the pollen of the six undehisced stamens
is discharged, so that now the upper part of the carina closely surrounds all the pollen,
and the lower part the style. The latter and the staminal tube together make up
a stretched spring exerting an upward pressure, while the claws of the carinal
petals and the interlocked alae constitute a second spring which presses downwards.
These opposing forces are in equilibrium, and the parts concerned remain horizontal,

LEGUMINOSAE 265

until the upper edges of the carinal petals are caused to separate. Each ala
possesses a fold which interlocks with an acutely angular projection on the upper
margin of the corresponding half of the carina. It therefore follows that when
a bee settles on the flower, resting with its limbs on the alae, and thrusting its
head under the vexillum, the alar folds slip down the column formed by the
filaments and the pistil, and at the same time the carina splits open above from
base to tip. When the splitting has extended as far as the end of the style, the
stretched parts suddenly separate, the carina and alae moving down, the style
with its superimposed pollen springing up. The stigma thus first touches the ventral
surface of the bee, and if this has previously visited another flower of the same
species it gets pollinated. Immediately afterwards the pollen is pressed against the
under-side of the visitor. If cross-pollination is not effected, self-pollination is

Fic. 83. Gentsta tinctoria, L. (after Herm. Miller) (1) The stamens with style and stigma;
taken froma bud. (2) Position of the parts enclosed in the carina in a flower not yet visited by insects.
(3 Right ala, seen from within. (4) Exploded flower. (5) Unexploded flower, after removal of calyx
and vexillum; seen from above. (6) The same, after the carina has been split open almost to the tip by
pressure from above. (7) Unexploded flower, after removal of vexillum and alae; seen from above
(8) Anterior part of a flower split open and about to explode; seen from above (enlarged to twice the
scale of the other figures). a, stamens with style and stigma; 4, the four outer stamens that have
temained short (2, 4, 8, 10 in Fig. 1); c, the five inner stamens (1, 3, 5, 7, 9); @, the outer stamen (6) that lies
under the style; e, tip of the style; 7, stigma; g, pollen; 4, lateral projections of the carina, with which
corresponding alar folds (4') interlock ; 4/, that part of the upper margin of the carina which is split before
explosion ; 2, alae; #, carina.

brought about by the insect as it backs out of the flower. Explosion has not been
observed to occur without the application of external pressure.

Visitors. — According to Hermann Miller, the visitors are chiefly pollen-
collecting bees, which sometimes also vainly seek for nectar. All of them, even
the nectar-seeking males, bring about explosion, and effect pollination in the way
already described. Among useless visitors are wasps, Conopidae, Syrphidae, Lepi-
doptera, and destructive beetles which gnaw the flowers (Cryptocephalus).

266 ANGIOSPERMAE—DICOTYLEDONES

I have so far observed Apis and some humble-bees (Bombus cognatus S#ph.,
B. lapidarius Z., B. terrester Z.) in Schleswig-Holstein.
Hermann Miller’s list of visitors is as follows.—

A. Coleoptera. (a) Chrysomelidae: 1. Cryptocephalus moraei Z., gnawing the
flowers; 2. C. sericeus Z.; 3. C. vittatus #. (6) L£latertdae: 4. Agriotes gallicus
Lac., vainly skg.; 5. A.ustulatus Scholl, do. B. Diptera. (2) Conopidae: 6. Myopa
testacea L., vainly skg.; 7. Sicus ferrugineus Z., do. (4) Syrphtdae: 8. Chrysotoxum
bicinctum Z., vainly skg. C. Hymenoptera. (a) Apzdae: g. Andrena albicrus
K.%, po-cltg.; 10. A. fulvescens Sm. 4, do.; 11. A. fulvicrus K.9, do.; 12. A.
xanthura A. 9, do.; 13. Anthidium punctatum Za/r. , vainly seeking for nectar,
po-cltg.; 14. Apis mellifica Z. ¥, freq., po-cltg.; 15. Bombus terrester Z. 9, po-cltg. ;
16, Colletes daviesanus K. 9, do.; 147. Diphysis serratulae Pz. $, do.; 18. Halictus
albipes F. 9, do.; 19. H. rubicundus Chr., do.; 20. Megachile centuncularis Z. 9,
very freq., po-cltg.; 21. M. circumcincta A. 9, do.; 22. M. versicolor Sm. 9, po-cltg.;
23. M. willughbiella X. 9, do.; 24. Osmia platycera Gerst., do. (4) Vespidae:
25. Odynerus trifasciatus /. 9, po-dvg. D. Lepidoptera. 26. Lycaena damon

S.V.; 27. Melitaea athalia Ro/.; 28. Pararge megaera Z., vainly seeking for
nectar.

Réssler observed a moth, Grapholitha scopariana #.-S. at Wiesbaden.

In Dumfriesshire two humble-bees were noticed (Scott-Elliot, ‘Flora of Dum-
friesshire,’ p. 42).

642. G. germanica L.—Kirchner (‘Flora v. Stuttgart,’ pp. 473-4) states that
the flower mechanism resembles that of the last species, but in this case there is
no explosion by release of parts in a state of tension. The stamens and style
protrude from the carina on the application of pressure, so that there is a simple
valvular arrangement. There is a slit in the upper side of the carina, extending
right to its tip, and just in front of the claw of either carinal petal there is a rounded
prominence which fits into a pouch in the corresponding ala. The anthers are
disposed in two sets in the bud, one close behind the other, and the hooked inwardly
bent style projects beyond them. They dehisce before anthesis. The anterior
surface of the style is pressed against the inner wall of the carina, so that there
is a certain amount of strain—though this is inconsiderable—between the style and
stamens on the one hand and the carina on the other. In this stage the vexillum
is still folded down on the alae and carina, and as the stigma is already mature
automatic self-pollination must take place, according to the above account. As
the vexillum erects itself the backwardly-curved style elongates, and projects from
the tip of the carina of the horizontal flower. Insect visitors must, therefore, first
touch the stigma as they alight, and will effect cross-pollination if they have
previously visited another flower. During the first visit almost all the pollen is
at once discharged from the depressed carina. If the carina is only slightly pressed
down it returns slowly to its original position when the pressure is removed, owing
to the slight elasticity of its interlocking projections. But if it is so far depressed
by heavier and more powerful insects that these projections are brought right
under the style, a return to the first position is impossible. Such flowers look
like the exploded blossoms of G. tinctoria.

Visitors.—I observed Bombus lapidarius Z. 9 in Schleswig-Holstein (‘Bloemen-
biol. Bijdragen’).

LEGUMINOSAE 267

643. G. sagittalis L—Kirchner (‘Flora v. Stuttgart,’ p. 474) states that,
like G. germanica, this species possesses a simple non-explosive valvular mechanism.
The column, made up of style and stamens, protrudes from the carina during insect-
visits, moving back again when the pressure is removed. The anthers dehisce
in the bud, and as the slightly up-curved style extends but little beyond them
at this stage the stigma is self-pollinated. After the vexillum has become erect,
the style, which is still bent slightly upwards, projects about 1 mm. beyond the
anthers, so that when a bee settles on the flower the stigma first projects from
the carina and touches its ventral surface. The second flower visited by an insect
is therefore cross-pollinated. Should considerable pressure be exerted, the carina
remains depressed as in G. germanica.

Visitors.—Kirchner observed undetermined species of Apidae. Schenck saw
two sternotribous bees, i.e. Megachile circumcincta A. and Trachusa serratulae Pz.

644. G. anglica L—In this species, which was first thoroughly described by
Hermann Miiller (‘ Fertilisation, pp. 192-3), the opposite tensions of style and
stamens on the one hand, and of the carina and alae on the other, are much less
pronounced. The carina and
alae sink down but little when
explosion takes place, and
only the style curves upwards,
its tip at the same time curling
inwards.

VisiTors.— Herm. Miiller
observed the honey-bee as
a visitor, This almost ex-

clusively visited unexploded Fic. 84. Gentsta anglica, L. (after Herm. Miller). (1) Un-

: sys visited flower, seen from the side. (2) The same, seen from the front.
flowers, assuming a position (3) Right ala, seen from the inner side. (4) An exploded flower, with
as if it would suck concealed _ the style curved back less than usual. (5) A flower that has exploded

e normally, seen from the left. e¢, tip of style; 4 stigma; 4, alar fold.
nectar from their bases, and
filling its pollen-baskets with its second legs. Herm. Miiller also repeatedly
saw two short-tongued bees po-cltg., i.e. Andrena fulvicrus A. 9 and Halictus
cylindricus /. 9.

Alfken and Héppner (H.) observed the following bees at Bremen.—

1. Andrena nigro-aenea KX. 9, infrequent, po-cltg.; 2. A. convexiuscula XK. 9,
po-cltg.; 3. Apis mellifica Z. ¥, do.; 4. Bombus muscorum, F. 9; 5. B. terrester
L.9; 6. Halictus flavipes /. 9, freq., po-cltg.; 7. H. leucopus K.9; 8. H. rubi-
cundus Chr. 9, freq., po-cltg.; 9. Osmia uncinata Gerst. 9, one individual, po-cltg. ;
10. Nomada alternata Pz. 9 (H.); 11. N. succincta Pz. 9 (H.).

In Amrum and Fohr, I only saw the honey-bee (‘BI. u. Insekt. a. d. nordfr.
Ins.,’ p. 152).

In Fohr and Sylt, I observed a moth (Zygaena filipendulae Z.), as an unbidden
guest, vainly trying to suck.

645. G. pilosa L.—The flower mechanism of this species agrees completely
with that of G. anglica. It was first described by Delpino (‘Ult. oss.,’ pp. 48-52).
He found that the flowers are self-sterile.

268 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Herm. Miiller (‘ Fertilisation,’ p. 193) in Westphalia, and myself in
Fohr and Amrum, observed the honey-bee.

At Wiesbaden Réssler noticed the moth Threnodes pollinalis S.V. as an
unbidden guest.

196. Ulex L.
As the last genus.

646. U. europaeus L. (Ogle, Pop. Sci. Rev., London, ix, 1870, pp. 164-5 ;
Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892, pp. 101 et seq.; Knuth, ‘ Bloemen-
biol. Bijdragen.’)—Ogle states that the flower mechanism of this species agrees with
that of G. tinctoria, and Kerner also describes it as an explosive arrangement.
But, according to my own observations, the contrary tensions of stamens and style
on the one hand, and carina and alae on the other, are not so strong, so that the
mechanism much more closely resembles those of G. anglica and G. pilosa.

Fic. 85. Ulex europaeus, L. (from nature). .4. Exploded flower; s, stigma. 2, C. An ala, seen
from within and without. J. Stamens and pistil enclosed in the carina, and seen by transparency. The
style is bent like a spring, so that the stigma is pressed against the cohering upper edges of the carinal
petals. £. The same, removed from the carina.

The carina and alae are only united together at one point on the upper side
of their claws, where some of the epidermal cells interlock, and an alar projection
fits into a carinal pit on either side. The union is so slight that the petals can
easily be separated without tearing.

Although explosion is but feeble, the pollen is so completely discharged on
to the ventral surface of the visiting bee, that when this has flown away scarcely
a single grain can be found upon the anthers.

VisiTors.—In the island of Foéhr (‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 85) I saw
numerous well-developed fruits, indicating insect-visits, although I never actually
observed these. The size of the flowers suggested that humble-bees were the
pollinators. And I actually noticed Bombus terrester Z. 9 as a visitor at Kiel
(May 9 and 23, 1896). I also saw Meligethes as an unbidden guest. In Flanders,
MacLeod noticed Apis, Bombus terrester Z. 9, 2 Halictus, and 2 flies (the last four
only on exploded blossoms) (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 329).

Burkill observed the following on the Yorkshire coast (‘Fertlsn. of Spring
Fis.’).—

LEGUMINOSAE 269

A. Araneidae. 1. Philodromus aureolus Clerck, lurking on unexploded flowers
and the central parts of exploded ones. B. Coleoptera. 2. Apion ulicis ors;
3. Meligethes picipes S/urm, po-dvg.; 4. Cryptophagus vini Panz., po-dvg. and
searching for nectar. C. Diptera. (a) Musczdae: 5. Hylemia sp., searching for
nectar; 6. Lucilia cornicina #., do.; 7. Sepsis nigripes AZg., do. (4) Syrphidae:
8. Eristalis arbustorum Z., po-dvg.; 9. E. pertinax Scop., searching for nectar;
1o. Melanostoma quadrimaculata Verral, po-dvg. D. Hymenoptera. Aprdae:
11, Andrena clarkella X., po-cltg.; 12. Apis mellifica Z. ¥, po-cltg. and at times
searching for nectar; 13. Bombus lapidarius Z., searching for nectar; 14. B. ter-
rester Z., do. and po-cltg. E.Thysanoptera. 15. Thrips sp., very freq.

197. Cytisus L.

Yellow, homogamous to protandrous, monadelphous bee flowers; with sap
enclosed in the tissues of the base of the flower, requiring to be bored for. The
pressure exerted by a visitor causes stamens and pistil to protrude from the carina,
to which they return when the pressure ceases. Several effective visits are therefore
possible. In some species there is a transition to a pumping arrangement. (Cf.
C. nigricans.)

647. C. Laburnum L. (Herm. Miiller, ‘ Fertilisation, pp. 193-5; Kirchner,
‘Flora v. Stuttgart,’ pp. 475-6; Knuth, ‘ Bloemenbiol. Bijdragen.’)—The flowers of this
species are of considerable size,
and aggregated into conspicuous
many-blossomed inflorescences.
Kerner says that when they open
the pedicels twist round so as
to turn the vexillum upwards
and the carina downwards.
The insertion of the vexillum
is bounded in front by a thick
fleshy swelling, full of sweet sap.

As nectar-guides, the vexillum Fic. 86. Cytisus Laburnum, L. (alter Herm. Miller). (1) Basal
possesses dark streaks, con- part of an older flower, after removal of calyx and vexillum ; seen

: b ll hi velli from above: aa, cut surface of calyx ; 4, insertion of vexillum; c, the
verging basally to this swelling, nectar-yielding swelling, into which insects appear to bore; d, claw

from which nectar is obtained of ala; e, flat projection of ala, which fits into a corresponding pit
on the upper side of the carina ; * carina; g, open slit of the same.

by boring. The union of alae (2, 3, 4) Stigmas of younger flowers. (5) Stigma of an older flower.

and carina is but slight, and is

effected by a flattish alar projection, which fits into a corresponding carinal pit

on either side.

Towards the end of the bud-period the stigma lies in the tip of the carina,
surrounded by stiff, glassy, erect hairs, that project beyond it, and at the beginning
of anthesis bend somewhat over its papillae, which are thus guarded from contact
with the under-sides of insect visitors. These hairs gradually shrivel up, so that
in older flowers the papillae are quite uncovered. At the same time the style curls
more and more inwards, and stretches out to an increasing distance through the
open cleft of the carina, so that insect visitors first touch the stigma, and cross-
pollination is ensured. Automatic self-pollination is excluded.

270 ANGIOSPERMAE—DICOTYLEDONES

Visirors.—Hermann Miiller saw po-cltg. bees, most of which also bored for
nectar (Bombus lapidarius Z. 9 and ¥, skg. and po-cltg.; B. terrester Z. 9, skg.;
Andrena albicans Afill. 9, po-cltg.; A. tibialis K.; A. xanthura XK. 9, po-cltg.; Apis
mellifica Z. ¥, freq., po-cltg.), also moths (Plusia), skg.; and Meligethes, creeping
about the flowers. He noticed both bees and butterflies that, not merely in one
flower but in several successively, inserted their proboscis under the vexillum, and
remained in this attitude for some time. Under such circumstances the collecting
apparatus of the bees remained quite empty after repeated visits. From this it
may be concluded that bees and butterflies actually perforate and suck the
nectariferous swelling.

In Kiel Botanic Garden, I saw (May 21, 1896) the just-opened flowers of
laburnum visited by our three commonest humble-bees, po-cltg. (Bombus hortorum
L. 9, B. terrester Z. 9, B. lapidarius Z. 9).

Alfken observed the following bees at Bremen, all skg.—

1. Bombus agrorum F. 9; 2. B. hortorum Z. 9; 3. B. ruderatus /. 9; 4.
Psithyrus vestalis Pourcr.

648. C. decumbens Spach.—According to Briquet (‘Etudes de biol. flor.
dans les Alpes occident.’), the flowers of this species are nectarless, and possess
an explosive mechanism which acts but once. It is set free by humble-bees, and
cross-pollination often results. Automatic self-pollination takes place in wet weather
(Kirchner).

649. C. hirsutus L.—

VisiTors.—Schletterer records the bees Podalirius acervorum Z., and P. tarsatus
Spin., for the Tyrol.

650. C. nigricans L.—Hermann Miller describes (‘ Weit. Beob.,’ II, pp. 254-6)
the flower mechanism of the golden yellow flowers of this species as being an
intermediate stage between the pumping arrangement of Lotus (vde supra), and
the simple valvular arrangement of C. Laburnum. The alae enclose the uppermost
part of the carina, which narrows into a sharp edge, and are slightly convex
externally. Their lower margins rest on the expanded sides of the carina. In
the young bud the five very large outer stamens (i.e. those alternating with the
petals) project completely beyond the five very small inner stamens (superposed
upon the petals). The anthers of the large stamens dehisce before anthesis, and
quickly shrivel, so that their pollen lies loosely between them, surrounded only
by the carina. The tips of the filaments of the small stamens have so far been
bent inwards, but they now straighten themselves, forcing their anthers between
those which have already dehisced, and pushing the pollen shed by these into the
empty upwardly-bent tip of the carina. The thickened outer filaments are stiff,
and when the carina is depressed push out the pollen through the opening at its
apex. These filaments therefore act as piston-rods, while the small anthers, filling
as they do the lower part of the pollen receptacle, serve as pistons. In a young
flower, depression of the carina—the upper edges of which are closely apposed—
causes some pollen to be squeezed out of its apex, so that it will adhere to the
under-side of an insect visitor. But, in older flowers, the upper carinal edges are
so loosely in contact that when the carina is depressed it opens widely, allowing

LEGUMINOSAE 271

stamens and stigma to protrude. It follows that pollen-collecting insects will carry
the pollen of younger flowers to the stigmas of older ones, and thus effect crossing.

Visrrors.—In the Oberpfalz, Herm. Miiller only saw a po-cltg. bee—Andrena
xanthura K. 9. - In Steiermark, E. Loew observed a long-tongued bee—Megachile
sp., po-cltg. ; and Hoffer the humble-bee Bombus mastrucatus Gers?. 9.

651. C. sagittalis Koch. (Herm. Miiller, ‘ Weit. Beob.,’ II, p. 254.)—

Vistrors.—These are po-cltg. bees. Herm. Miiller saw the following in the
Vosges,—

1. Andrena convexiuscula X. 9; 2. Bombus lapidarius Z.4; 3. B. terrester Z. 9;
4. Halictus rubicundus Chr. 9; 5. Osmia fulviventris Pz. 9.

Buddeberg observed the following in Nassau.—

1. Bombus variabilis Schmiedekn., var. tristis Secd/.$; 2. Diphysis serratulae
Pz.8; 3. Megachile circumcincta K. 9.

Réssler noticed the following Lepidoptera as unbidden guests at Wiesbaden.—

1. Grapholitha asseclana H.; 2. G. fuchsiana Rssir.; 3. G. succedana Frii.;
4. Threnodes pollinalis S. V.

652. C. canariensis Steud., and 653. C. albus Link.—The flowers of these
species possess an explosive mechanism, in which the anthers are at first slightly
depressed with the carina, and then spring out above. Hildebrand considers it
an adaptation to self-pollination, because some of the scattered pollen reaches the
stigma of the same flower (Bot. Ztg., Leipzig, xxiv, 1866, p. 75). It appears very
probable that cross-pollination is favoured by insect-visits.

654. C. austriacus L.—

Visirors.—Loew observed a humble-bee, Bombus agrorum F., po-cltg., in the
Berlin Botanic Garden.

198. Sophora L.
655. S. flavescens Ait—
Visirors.—Loew observed Bombus terrester Z. 5, skg., in the Berlin Botanic
Garden.
199. Thermopsis R. Br.
656. T. fabacea DC.—

VisiTors.—Loew observed Bombus hortorum Z., skg., in the Berlin Botanic
Garden.

200. Lupinus Tourn.

Yellow, blue, or white, nectarless bee flowers, with a pumping arrangement
from which strings of pollen are extruded.

657. L. luteus L. (Herm. Miiller, ‘Fertilisation,’ pp. 187-8 ; Knuth, ‘ Bliiten-
biol. Beob. a. d. Ins. Riigen,’ ‘ Bloemenbiol. Bijdragen.’)—The dark yellow flowers
of this species are devoid of nectar but very fragrant. Hermann Miiller describes
the alae as being united together by fusion of their anterior margins, while each
of them is interlocked with the carina by a latero-basal fold fitting into a cor-
responding depression. The anthers of the five outer stamens are very much larger

272 ANGIOSPERMAE—DICOTYLEDONES

than those of the five inner, and dehisce before anthesis, after which they com-
pletely shrivel up. Their pollen is stored up in the hollow cone constituted by
the tip of the carina. The five inner stamens, which have so far remained short,
now begin to grow actively, and compress the pollen in the end of the carina.
During insect-visits they act as pistons, by which a string of pollen is squeezed
out from the carinal apex. When the weight of the visitor is removed, the alae
and carina return to their old position, so that further visits cause more pollen to
be extruded. At a later stage the stigma also protrudes, and visitors carrying
pollen from younger flowers will therefore effect crossing. As in Cytisus Laburnum,
automatic self-pollination is prevented, or at any rate hindered, by a circlet of stiff
erect hairs.

Visitors.—Herm. Miller saw three po-cltg. bees.—1. Apis mellifica Z. ¥, freq. ;
2. Bombus lapidarius Z. ¥, occasional; 3. Megachile circumcincta K. 9. The two
first of these I noticed in Mecklen-
burg, and also Bombus terrester Z. ¥,
po-cltg. In Riigen I observed the
following.—

A. Hymenoptera. Apzdae:
1. Apis mellifica Z. ¥, skg. and po-
cltg., with large masses of orange-
coloured pollen in its baskets; 2. Bom-
bus agrorum /. ¥, skg.and po-cltg.
B. Lepidoptera. Mhopalocera: 3.
Argynnis paphia Z., skg., an unbidden
guest.

Alfken saw the humble-bee
Bombus lapidarius Z. 9 in Bremen,
and Loew the following bees in Silesia
Fic. 87. Lupinus luteus, L. (after Herm. Miller). (‘ Beitrage,’ p- 34).—

(1) Stamens and pistil in the bud. (2) The same in the
mature flower. 1, 3, 5, 7, 9, the five inner, and 2, 4, 6, 8, 10, 1, Bombus cognatus Steph. ¥,

the five outer stamens. -, stigma. po-cltg. ; 2. B. rajellus K. 8, do. ; 3.
Megachile maritima &”. 9, do.

658. L. angustifolius L.—The blue flowers of this species are odourless and
devoid of nectar. Kirchner states that their mechanism agrees with that of the
last species (‘Flora v. Stuttgart, p. 478).

Visitors.—I noticed the same bees in Mecklenburg as for the last species.

659. L. polyphyllus Lindl.—

Visirors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Andrena dorsata A. 9, transferring pollen pumped out from the carina to
the tibial and tarsal brushes of its hind-legs; 2. Anthidium manicatum JZ. 9, po-cltg.
and trying to suck in spite of the absence of nectar, 6 swarming about the flowers ;
3. Apis mellifica Z. ¥, cltg. the extruded pollen, and vainly skg.; 4. Megachile
centuncularis Z. 9, do.; 5. M. circumcincta 4. 9, do.; 6. M. ericetorum Lep. 9, do.,
é, vainly skg.; 7. Osmia aenea ZL. 9, do.

I saw the humble-bee Bombus lapidarius Z. 4, vainly skg., at Kiel.

LEGUMINOSAE 273

660. L. albus L.—Delpino states that this species essentially agrees with
L. luteus (‘ Ult. oss.,’ pp. 46-17).

661. L. hirsutus L.—
Visirors.—Schletterer saw the humble-bee Bombus terrester Z. at Pola.

201. Ononis L.

Generally red, rarely white or yellow, nectarless bee flowers, with a pumping
arrangement from which threads of pollen are extruded. Bentham says that apetalous
cleistogamous flowers are often present in South European species.

662. O. spinosa L. (Herm. Miller, ‘Fertilisation,’ pp. 174-5; Kirchner,
‘Flora v. Stuttgart, pp. 478-9; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 57-8;
Loew, ‘Bliitenbiol. Floristik,’ p. 392; Warnstorf, Verh. bot. Ver., Berlin, xxxviii,

1896.)—The flowers of this species are rose-red in colour, more rarely white, without
nectar or nectar-guides.

Hermann Miller describes
the alae as flat plates, that
are separate below but
converge above to en-
sheath the upper part of
the carina, with which they
interlock by means of a
pointed process on either
side, that projects forwards
and downwards internal to
the upper alar margin,
and fits into a deep carinal

fold. There is also a
E 3 Fic. 88. Oxonts spinosa, L. (after Herm. Miller). (1) Reproductive
basal lappet projecting organs of a bud. (2) The same of a mature flower (x 7). (3) Flower

after removal of calyx and vexillum, seen from the side. (4) Some
backwards from the upper stamens, more highly magnified, to show the difference between the outer
margin of each ala, and and inner filaments as regards thickening. (5) Left ala seen from the

: inner side, with the upper margin directed downwards. (6) Outer surface
resting upon the sexual of the same. a and 4, outer and inner stamens; ¢, pollen in the tip of the

column, without touching carina, seen by transparency; d, pointed alar process ; e, alar lappet.
its fellow.

The ten cohering filaments are somewhat thickened just below the anthers,
the five outer ones much more strongly than the others. The anthers of the five
inner stamens, on the other hand, produce a larger quantity of pollen. Before
anthesis, the anthers extend to the base of the hollow cone formed by the tip of
the carina, completely fill this with pollen, and then shrivel up. The stigma is
situated a little below the carinal apex.

The upper margins of the carinal petals are at first united for the whole of
their extent, except that a small opening is left at the tip. If now the carina is
slightly depressed, the thickened ends of the filaments are pushed further into the
pollen-chamber, and a corresponding amount of pollen is squeezed out. The carina
returns to its original position when the pressure ceases. With repeated depression
the united upper margins of the carinal petals are split apart, when stamens and

DAVIS. It T

274 ANGIOSPERMAE—DICOTYLEDONES

style protrude, returning again to the cavity of the carina if the pressure is not
too strong. But if this is considerable they remain entirely or partially projecting
out of the carina.

Visttors.—These are bees, especially Dasygastres, of which Herm. Miiller
observed the following in Westphalia.—

1. Anthidium manicatum Z. 9 and 4, freq.; 2. A. punctatum Zir. 9 and 4;
3. Megachile circumcincta K. 9, freq.; 4. M. lagopoda Z. 9 and 4, do.; 5. M.
versicolor Sm. 9; 6. Osmia aenea Z. 9, freq.; and in Thuringia, 7. O. aurulenta
Pz. 9, do.

Herm. Miiller also saw the following Scopulipedes in Westphalia.

1. Apis mellifica Z. %; 2. Bombus lapidarius Z. ¥; 3. B.terrester Z.9; 4. Cilissa
leporina Pz. 9; and in Thuringia, 5. Podalirius vulpinus Pz. 9 and 6, freq.

In Schleswig-Holstein I noticed only the Scopulipedes Apis, Bombus terrester,
and B. lapidarius, while Loew saw the dasygastrid Megachile maritima &. 9, po-cltg.,
in North Germany. In the Berlin Botanic Garden he also observed Anthidium
manicatum LZ. 9, po-cltg., and also skg. persistently in spite of the absence of nectar ;
when the 9 was caught a é immediately visited the same flower, obstinately returning
after being driven away.

The following have been recorded by the naturalists and at the places
mentioned.—

Alfken (Bremen), the bees, 1. Andrena flavipes Pz. 9, 2nd generation; 2. Bombus
arenicola 7hs.9; 3. B. distinguendus J/or.9; 4. Megachile maritima X. 9; all po-cltg.
Sickmann (Osnabriick), the fossorial wasp Astata minor KoA/. Alfken (Juist), the
bees, 1. Bombus lapidarius Z. 9; 2. B. muscorum /. ¥; 3. B. terrester L. 9;
4. Megachile maritima X. 9 and é. Réssler (Wiesbaden), the following Lepidoptera
as unbidden guests—1. Grapholitha microgammana Gz.; 2. Acidalia humiliata Hu/n.;
3. Hesperia actaeon Rof/.; 4. Lycaenaargus Z. MacLeod (Flanders), Apis, 4 humble-
bees, and a Lepidopterid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 335-6).

663. O. repens L. (=O. procurrens Wallr.). (Herm. Miiller, ‘ Weit. Beob.,’
IJ, p. 254; Kirchner, ‘Flora v. Stuttgart,’ p. 479; Knuth, ‘ Bloemenbiol. Bijdragen’ ;
Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The flower mechanism agrees
with that of the last species, except that the blossoms are rather larger (Kirchner).
Warnstorf states that the downwardly directed pointed alar process is longer and
sharper than in QO. spinosa. The pollen-grains are golden yellow in colour, of
broad or narrow ellipsoidal shape, about 37 » long and 25 p broad.

Visrrors.—These are the following bees, po-cltg., or vainly skg.—

A, Dasygastres. 1. Anthidium manicatum Z. 9 and é (Buddeberg, Nassau) ;
2. A. oblongatum Z/r. (do.); 3. Megachile argentata J. (do.); 4. M. circumcincta
XK. 9 (do.); 5. M. fasciata Sm.d, skg. (do.); 6. Osmia spinulosa A. 9 (Herm. Miller,
Thuringia). B.Scopulipedes. 7. Bombus agrorum /. (Knuth, Holstein); 8. B.
variabilis Schmzedekn., var. tristis Secd/. (Herm. Miiller, Thuringia); 9. Cilissa leporina
Pz. 9, skg. (Buddeberg, Nassau).

I saw the honey-bee, skg., in Riigen.

H. de Vries observed the humble-bee Bombus terrester Z. ¥ in the Netherlands
{Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).

664. O. arvensis L. (Syst. Nat.)—Kirchner describes the mechanism as
agreeing with those of the last two species, though the flowers are often smaller.

LEGUMINOSAE 275

Visitors.—In Dumfriesshire, the honey-bee and a humble-bee have been
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 43).

665. O. Natrix L. (MacLeod, ‘Pyreneenbl.’; Kirchner, ‘ Beitrage,’ p. 40.)—
The flowers of this species are yellow, with dark red streaks on the vexillum.
Kirchner states that the mechanism agrees with those of the last two species.

Vistrors.—In the Pyrenees, MacLeod observed three Dasygastres (a Megachile
and 2 Osmia), and 6 Scopulipedes (an Andrena, 4 Bombus, and an Eucera).

666. O. rotundifolia L.—According to Briquet (‘Etudes de biol. flor. dans
les Alpes occident.’), the rose-red flowers of this species possess a pumping arrange-
ment extruding threads of pollen, as in other species of the genus. Insect visitors,
mostly bees and Lepidoptera, are very numerous. They usually effect cross-
pollination, for the stigma projects beyond the anthers, and does not become
sticky until its papillae have rubbed against an insect’s body. Automatic self-
pollination can take place at the end of anthesis. The uppermost stamen is not
fused with the others. Kirchner, however, found it to be united with its neighbours
for about 3 mm. at the base, but otherwise free. He also states that the flowers
exhale an odour of roses.

Visitors.— Vide supra.

202. Medicago L.

Yellow or bluish nectar-yielding bee flowers, the stamens and pistil of which
Spring out of the carina.

667. M. sativa L. (Henslow, J. Linn. Soc., Bot., London, ix, 1867; Hilde-
brand, Bot. Ztg., Leipzig, xxiv, 1866, pp. 74-5, and xxv, 1867, p. 283; Delpino,
‘Sugii appar. d. fecondaz. nelle piante autocarp.,’ pp. 26-8, ‘ Ult. oss.,’ pp. 47-8;
Herm. Miller, ‘Fertilisation,’ pp. 175-8, ‘ Weit. Beob.,’ II, p. 252; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 336-8; Knuth, ‘Bloemenbiol. Bijdragen’;
Loew, ‘ Bliitenbiol. Floristik,’ p. 391.)—The bluish or violet blossoms are aggregated
into tolerably conspicuous many-flowered racemes. Each individual flower is 7 to
11 mm. in length. Nectar is secreted in the usual place, and there is a passage
to it on either side of the free stamen, as in similar cases. The pressure exerted
by an insect visitor causes stamens and style to spring out of the carina, to which
they are unable to return. The elasticity necessary to bring about explosion lies
entirely in the upper stamens, and there are two arrangements in the unvisited
flower by which this is kept in check.—(1) At the upper basal angle of either carinal
petal there is an internal hollow process, forwardly directed, and closely apposed
to its fellow, the two together gripping the front part of the sexual column from
above. A still more pronounced alar process fits into each of them. (2) From
the upper edge of each ala a long fringe-like process runs back, passing upwards
and inwards so as to grasp, with its fellow, the upper side of the column at a point
about one-third of its Jength from the base.

The various processes described forcibly maintain the sexual column in a
horizontal position. But should an insect visitor press down the alae and carina,
the stamens and pistil spring up out of the carina against the under-side of its
body or proboscis. The stigma projects beyond the anthers, and therefore first

T2

276 ANGIOSPERMAE—DICOTYLEDONES

comes into contact with the insect, getting cross-pollinated if this has previously
visited another flower of the species. The first flower visited by an insect will be
self-pollinated as the visitor backs out of it. Should insect-visits fail, automatic
self-pollination of unexploded flowers is possible, and may be effective under certain
conditions (wide infra).

Burkill (Proc. Phil. Soc., Cambridge, viii, 1894) aptly describes the basal pro-
cesses of alae and carina as two triggers, by which the flower is, so to speak, fired
off. He states that the upper surfaces of the alae are beset with papillae, serving
as foot-holds to insect visitors. There is also a marginal row of papillae on either
side of the vexillum’s inner surface, to which long-legged insects would appear to
cling. The stigma remains unreceptive until its papillae have been subjected
to friction. By covering a number of inflorescences with nets, in order to keep
away insects, Burkill was able to confirm the conclusion at which Urban had
previously arrived (Verh. bot. Ver.,
Berlin, xv, 1873), i.e. that unex-
ploded flowers do not set fruits,
although their stigmas are sur-
rounded by pollen. He succeeded,
however, in inducing such flowers
to fruit by (1) squeezing the stigma
through the tip of the carina, (2)
piercing the carina with a needle
and scratching the stigma, (3)
cutting off the tip of the carina and
rubbing the stigma with a paint-
brush.

Fic. 89. Medicago sativa, L. (after Herm. Miiller). (1)

Unvisited flower, seen from below. (2) The same, after re-
moval of the vexillum and upper half of the calyx; seen from
above. (3) Right ala seen from inside. (4) Carina, seen
from above and from the right. (5) Exploded flower, seen
from above and from the left, after removal! of the vexillum
and upper half of the calyx (x 3}). a, calyx; 6, vexillum; c, ala;
d, alar claw ; ¢, anterior alar process; 7, cavity of this process;
&, posterior finger-shaped alar process; 4, carina; 2, carinal
process into which the anterior alar process fits; 2, the cohering
filaments; /, the free uppermost filament ; 7, nectar-passages ;
nm, anther ; 0, stigma.

VisiTors.— These are bees
and Lepidoptera. The slender
proboscis of the latter is undoubted-
ly able to bring about explosion,
providing it is thrust down to the
nectar from the middle of the flower,
while, on the other hand, explosion
is not effected by insects that suck

laterally, as e.g. the honey-bee, which inserts its proboscis into the base of the

flower from the side of the vexillum.

Herm. Miiller (H. M.), Buddeberg (Budd.), and myself (Kn.) have observed the

following.—
Kn.);

(Kn.
chile argentata 7.

A. Hymenoptera. (a) Apzdae: 1. Apis mellifica L. ¥, very freq., skg. (H. M.,
2. Bombus agrorum ZL. 9 and ¥, skg. (H.M., Kn.); 3. B. terrester Z., do.
io ; 4. Cilissa leporina Pz. 6, do. (H. M.); 5. Coelioxys rufescens Zep. $, do.
; 6. Colletes sp. 6, do. (H.M.); 7. Halictus morio F. 9, do. (Budd.); 8. Mega-

g and 4, do. (H. M., Budd.); 9. M. pyrina Lep., do. (H. M.);

10. M. willughbiella A. , do. (H.M.); 11. Osmia aenea Z. 9, freq., skg. and po-cltg.
(H. M.); 12. O. rufa LZ. 9, skg. (H. M.); 13. Rhophites canus £v., do. (H. M.);

14. Xylocopa violacea LZ. $, do. (H. M.).
skg. (H. M.).

(6) Sphegrdae: 15. Bembex rostrata Z.,
B. Lepidoptera. (2) Noctuidae: 16. Plusia gamma L.

(6) Rho-

LEGUMINOSAE 277

palocera: 17. Colias edusa Z., skg. (Budd.); 18. C. hyale Z. (H. M., Budd.); 19.
Hesperia lineola O., skg. (Budd.); 20. H. thaumas Hu/n. (H. M.); 21. Lycaena
argiolus Z. (H. M.); 22. Pieris brassicae Z. (H. M.); 23. P. napi Z. (H. M.);
24. P. rapae Z. (H. M.); 25. Rhodocera rhamni Z., skg. (Budd.); 26. Satyrus
hyperanthus Z. (H. M.); 27. Vanessa urticae Z. (H. M.).

Réssler observed the butterfly Colias edusa #., at Bremen, while bees were
recorded by and for the authors and places named in the following list.—

Loew (Brandenburg), Cilissa leporina Pz. 3, skg. (‘ Beitrage ’); Alfken (Bremen),
1. Anthidium manicatum Z. 9 and é, 2. Bombus variabilis Schmdedekn. ¥, 3. Cilissa
leporina Pz. 9 and 8; Frey-Gessner (canton Valais), Eucera hungarica Fzese 9 and
8; Friese (Baden), Cilissa leporina Pz., occasional ; Krieger (Leipzig), Eucera longi-

cornis Z.9; Schenck (Nassau), Cilissa leporina Pz., occasional; Dalla Torre and
Schletterer (Tyrol), Bombus pomorum Pz. é.

Burkill observed the following at Cambridge (Proc. Phil. Soc., Cambridge, viii,
1894).—

A. Coleoptera. MWitidulidae: 1. Meligethes viridescens /. B. Diptera.
(a) Muscidae: 2. Caricea tigrina F.; 3. Lucilia sericata Mg. (6) Syrphidae:
4. Eristalis pertinax Scop.; 5. Helophilus floreus Z.; 6. Melithreptus scriptus Z. ;
7. Platycheirus albimanus /.; 8. P. manicatus AZg.; 9. P. scutatus M/g.; 10. Syritta
pipiens Z.; 11. Syrphus balteatus Deg.; 12. S. corollae #.; 13. S. ribesii Z.
C. Hymenoptera. (a) Apidae: 14. Andrena convexiuscula K.9; 15. A. extri-
cata Sm.d; 16. Apis mellifica Z. ¥, very freq.; 17. Bombus agrorum /.; 18. B.
hortorum Z., common; 1g. B. lapidarius Z.; 20. B. pratorum Z.; 21. Megachile
centuncularis Z. 9. (4) Vespidae: 22. Vespa vulgaris Z. 6. D. Lepidoptera.
(a) Noctuidae: 23. Agrotis pronuba Z.; 24. Phasiane clathrata Z.; 25. Plusia
gamma LZ. (6) Rhopalocera: 26. Lycaena icarus Z.; 27. Pieris brassicae Z., freq.;
28. P. napi Z.; 29. P. rapae Z.; 30. Polyommatus phlaeas Z.; 31. Vanessa
urticae Z.

All the visitors suck nectar, though the flies would seem but seldom to reach
it, and can only get pollen from exploded flowers. As Hermann Miiller previously
noticed, the honey-bee does not explode the flowers, but steals the nectar by thrust-
ing in its proboscis from the side. Ona hot summer afternoon Burkill saw Bombus
hortorum Z. in large numbers on the flowers, skg. legitimately, and effecting
explosion.

Burkill states that the flowers are not all explosive to the same degree, the
hotter the weather the more vigorously they explode. In cold weather they remain
unexploded for eight or nine days, and then wither; in hot sunny weather anthesis
lasts for three days at most. Explosion is not brought about by the shaking of the
wind.

Schulz (Thuringia) and Urban (Berlin) have observed flowers perforated by
Apis and humble-bees.

668. M. falcata L. (Herm. Miller, ‘ Fertilisation,’ pp. 179-80, ‘ Weit. Beob.,’
II, p. 252; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 338.)}—Hermann
Miiller states that the mechanism of the yellow flowers of this species entirely agrees
with that of M. sativa, but the springing up of the sexual column as the result
of pressure from above is facilitated by the loose way in which the alae and carina
grasp the upper side of this. It is, however, more difficult for bees to extract nectar

278 ANGIOSPERMAE—DICOTYLEDONES

without causing explosion, for the basal parts of the relatively short and broad alae
rest on the carina for some little distance.

Visttors.—These are once more bees and Lepidoptera. As a result of the
peculiarity last named, the former always effect explosion, while Lepidoptera, owing
to the slender character of their proboscis, can extract nectar from unvisited flowers
without exploding them.

Herm. Miiller observed the following at R6hmberge, near Miihlberg.—

A. Hymenoptera. Aprdae: 1. Andrena denticulata K. 9, skg.; 2. A. fulvi-
crus K. 9, do.; 3. Apis mellifica Z. ¥, freq., skg.; 4. Bombus agrorum ff. 9, skg. 5.
5. Cilissa leporina Pz. 9 and 4, skg. and po-cltg.; 6. Halictus quadricinctus /. 9,
po-cltg.; 7. Nomada ferruginata K. 9, skg.; 8. N. solidaginis Pz. 9, do.; 9. N.
fucata Pz. 9, do.; 10. Osmia aurulenta Pz. 9, freq., skg. and po-cltg. ; 11. Rhophites
canus Ev. 9 and 4, skg. B. Diptera. (a) Bombyliidae : 12. Systoechus sulphureus
Mikan, skg. (6) Syrphidae: 13. Helophidus trivittatus 7. C. Lepidoptera.

Fic. 90. Medtcago falcata, L. (after Herm. Miller). (1) Flower seen obliquely from below. (2) The
same, after removal of the calyx and vexillum; seen from above. (3) Left ala, seen from above and
from the right. (4) Carina, seen from above and from the right. (5) Exploded flower, after removal of
the vexillum; seen from above and from the right. The sexual column is much foreshortened (x 7).
a, calyx; 6, vexillum; c¢, nectar-guides; d, alar claw; ¢,ala; / anterior alar process; g, posterior alar
process ; 4, fused claws of carinal petals; z, limbs ofdo.; &, carinal pit into which an anterior alar process
fits; 4, sexual column; 72, upper filament; 7, nectar-passage ; 0, stigma.

(a) Nocturdae: 14. Euclidia glyphica Z., skg. (6) Rhopalocera: 15. Epinephele
janira Z., skg.; 16. Hesperia sylvanus Z£sf., do.; 17. Lycaena coridon Poda, skg. ;
18. Melitaea athalia Ro//., do.; 19. Pieris rapae Z., do.; 20. Vanessa urticae Z., do.;
(c) Sphingidae: 21. Sesia asiliformis Roét., skg.; 22. Zygaena carniolica Scop., freq.
Herm. Miiller also saw a humble-bee and two Lepidoptera in the Alps (‘ Alpen-
blumen,’ p. 248).

Rossler noticed the butterfly Colias hyale Z. at Wiesbaden, as an unbidden
guest.

Burkill observed the following at Cambridge (Proc. Phil. Soc., Cambridge,
viii, 1894).—

A. Diptera. Syrphzdae: 1. Syritta pipiens Z.; 2. Syrphus balteatus Deg.; 3. S.
luniger 17g. B. Hymenoptera. (a) Apidae: 4. Apis mellifica Z. ¥; 5. Bombus.

LEGUMINOSAE 279

hortorum ZL. (0) Formicidae: 6. Formica rufa ZL. (c) Ichneumonidae: 4. Cryptus
analis Gr.

Schletterer saw the small leaf-cutting bee Megachile argentata /. at Pola.

Schulz (Thuringia) and Urban (Berlin) have observed flowers perforated by
Apis and humble-bees.

The flowers are more explosive than those of M. sativa, and Burkill says
that they are made more so by warmth. To such an extent is this the case that
the pressure exerted by flies (Syrphids, and even Muscids) may be sufficient to
liberate the mechanism. When the blossoms are at maximum tension, explosion
may be effected by a sudden heavy shower.

669. M. media Pers. (=M. falcata x M. sativa). (Knuth, ‘ Bloemenbiol.
Bijdragen.’)—The colour of the corolla is variable in this hybrid. At frst it is
generally yellow, then becomes greenish, and lastly bluish or violet. It grows
abundantly on the walls of the fortress at Coburg, where I noticed that some
individuals approximated more nearly to M. falcata, and others to M. sativa.

Vistrors.—I noticed the honey-bee, common, at Coburg. Loew observed a
beetle, and 2 bees in the Berlin Botanic Garden, as follows.—

1. Coccinella octodecimpunctata Scop., settling on the outside of the flower ;
2. Andrena fasciata Wesm. 8, skg.; 3. Cilissa tricincta A. 9, po-cltg. and skg.
legitimately.

670. M. prostrata Jacq. (Burkill, Proc. Phil. Soc., Cambridge, viii, 1894.)—
The flower mechanism of this species agrees with that of M. falcata, but the flowers
are smaller, and exploded by less pressure on the alae.

671. M. sylvestris Fries (=M. falcata, according to Urban, Verh. bot. Ver.,
Berlin, xv, 1873, p. 56).—According to Burkill (Proc. Phil. Soc., Cambridge, viii,
1894), this species has a closer oecological resemblance to M. sativa than to M.
‘ fdlcata, for its mechanism never attains the highly explosive condition characteristic
of the latter.

VistTors.—The honey-bee is by far the commonest visitor in the Cambridge
Botanic Garden, sucking the nectar from the side as in M. sativa. As only humble-
bees are able to cause explosion, 99 per cent. of the flowers remain unfertilized
(at Cambridge).

Visttors.—Burkill observed the following.—

A. Diptera. (a) Muscidae: 1. Caricea tigrina ¥.; 2. Lucilia sericata Mg.;
3. Sarcophaga carnaria ZL. (6) Syrphidae: 4. Eristalis pertinax Scop.; 5. Melithreptus
scriptus Z.; 6. Platycheirus manicatus Mg.; 7. Syritta pipiens Z.; 8. Syrphus
balteatus Deg.; 9. S.corollae #.; 10. S.luniger Mg.; 11.58. ribesii Z. B. Hymeno-
ptera. Apidae: 12. Apis mellifica Z. ¥, very freq.; 13. Bombus hortorum JZ. ;
14. B. lucorum Z.; 15. Odynerus parietum Z.9. C. Lepidoptera. Rhopalocera:
16. Pieris brassicae L.

672. M. lupulina L. (Darwin, Gard. Chron., London, 1857; Herm. Miller,
Fertilisation, p. 180, ‘Weit. Beob., II, p. 252; Kirchner, ‘Flora v, Stuttgart,’
p- 483; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.” pp. 59, 152.)—The mechanism
of the small (scarcely 2 mm. in length) golden-yellow flowers of this species agrees
with that of M. sativa, but the elasticity of the upper stamens is slight. An insect-
visit causes the column of reproductive organs to spring out of the carina, to which

280 ANGIOSPERMAE—DICOTYLEDONES

it does not return when the pressure is removed. According to Darwin, automatic
self-pollination readily takes place, but is far less productive than cross-pollination.

Visitors. — Darwin in England, MacLeod in Flanders, Hermann Miiller in
Westphalia, and myself in Schleswig-Holstein have observed the honey-bee. Hermann
Miiller remarks in this connexion :—‘It is characteristic of the honey-bee that it
does not disdain even this flower’s tiny store of honey. The whole inflorescence
is pulled down by the weight of the bee, which then sucks the flower’s head down-
wards. It visits a few flowers, at most four, on one inflorescence, and then flies away
to another plant of the same species. Xenogamy is thus effected to a great extent.’

While as a rule the honey-bee has been observed as the chief pollinator,
Burkill (Proc. Phil. Soc., Cambridge, viii, 1894, names a hover-fly (Platycherrus
manicatus) as the most active visitor at Scarborough, a short-tongued bee (Halzctus
morto) and a Muscid (Scafophaga) at Cambridge. He gives the following list for
Scarborough (June, 1893) and Cambridge (July and August, 1893).—

A. Coleoptera. 1. Anthobium torquatum JZarsh.; 2. Ceutorhynchidius floralis
Payk.; 3. Meligethes aeneus /. B. Diptera. (2) Amthomyzdae: 4. Anthomyia sp.;
5. Caricea tigrina #.; 6. Chortophila cinerella Fal/.; 7. C. sepitorum Meade ;
8. C. sp.; 9. Homalomyia armata Mg.; 10. Hydrotea irritans Fa//.; 11. Hylemyia
pullula Ze/.; 12. Pogonomyia alpicola Rud.? (4) Brbionrdae: 13. Scatopse brevi-
cornis J/g. (c) Chironomidae: 14. Chironomus sp. (d) Chloropidae: 15, 16, 17.
Three sp. of Chlorops; 18. Oscinis sp.?, very freq. (e) Lmpidae: 19. Empis
punctata #. (/) Muscidae: 20, 21. Two undetermined sp. (g) Sarcophagidae :
22. Sarcophaga sp., very freq. (h) Scatophagidae: 23. Scatophaga stercoraria L.
(7) Sepstdae: 24. Hydrellia griseola Fal/.; 25. Sepsis cynipsea Z. (h) Syrphidae:
26. Paragus tibialis Fad/.; 27. Pipizella virens /.; 28. Platycheirus albimanus /.;
29. P. manicatus A/g.; 30. P. scutatus M/g.; 31. Syrphus balteatus Deg.; 32. S.
corollae F.; 33. Syritta pipiens Z. (2) Zadanzdae: 34. Ptiolina crassicornis Pz.
(m) Pachinidae: 35. Myobia inanis Fal/.; 36. Siphona cristata #.; 37. S. genicu-
lata Deg. C. Hemiptera. 38. Aphis sp.; 39. Siphonophora artemisiae Koch. D.
Hymenoptera. Agrdae: 40. Andrena parvula A.9; 41. Apis mellifica Z.%, rare;
42. Bombus hortorum Z.; 43. Halictus minutissimus A.6; 44. H. morio #.é and 9,
not infrequent; 45-50. Six undetermined species. E. Lepidoptera. (a) Noctur-
dae: 51. Hadena fasciuncula Haw. (6) Pyralidae: 52. Crambus pratellus Z.
(c) Tinerdae: 53. Porrectaria sp. (d) Tortricidae: 54, 55. Sp. of Tortrix? Fy
Neuroptera. 56. Thrips sp.

Sickmann mentions the fossorial wasp Gorytes lunatus Dahid., infrequent.

Besides the honey-bee, Herm. Miiller observed the following. —

A. Hymenoptera, Apidae: 1. Andrena convexiuscula A. 9, skg.; 2. A.
xanthura A’. 9, po-cltg.; 3. Bombus agrorum /. 9 and 4%, skg. (Strassburg); 4.
Halictus flavipes /.9, po-clig. B. Diptera. Conopidae: 5. Myopa buccata Z.,

skg.; 6. M. testacea Z., do. CC. Lepidoptera. Rhopalocera: 7. Thecla rubi
L. 9, skg.

Miiller further records three Lepidoptera, skg., for the Alps (‘ Alpenblumen,’
p- 248).

MacLeod, in addition to Apis, observed Halictus sp. in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 338). H. de Vries observed Bombus terrester Z. ¥
in the Netherlands (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).

673. M. arabica All.— Kirchner (‘Flora v. Stuttgart,’ p. 484) states that
the flower mechanism of this South European species agrees with that of M. sativa.

LEGUMINOSAE 281

The flowers are yellow and about 6 mm. long; the vexillum is streaked, the carina
is about 4 mm. long, and the alae somewhat shorter.

674. M. hispida Gaertn—The flowers of this species, which is indigenous
to South Europe, are about as long again as those of M. lupulina, according to
Kirchner (‘Flora v. Stuttgart,’ p. 483). Their mechanism is similar.

675. M. carstiensis Jacq.—

VisiTors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Bombus rajellus A. ¥, skg.; 2. Cilissa tricincta K. 9, skg. legitimately and
po-cltg., skg.; 3. Megachile centuncularis Z. 6, skg.; 4. M. circumcincta &. 9,
po-cltg.; 5. M. lagopoda L. 9, do.

203. Dorycnium L.

676. D. hirsutum Ser. (=Bonjeania hirsuta Rezchd.).—In the flowers of this
species there is a pumping arrangement, the ends of the filaments being thickened
(Delpino, ‘ Ult. oss.,’ p. 45).

VisiTors.—Schletterer observed the following at Pola.—

Hymenoptera. (a) Apzdae: 1. Andrena convexiuscula &., var. fuscata X.;
2. A. morio Brull.; 3. Bombus argillaceus Scop.; 4. B. terrester Z.; 5. Eucera hispana

Lep.; 6. E. interrupta Baer; 7. Megachile muraria Z.; 8. Podalirius retusus Z.,
var. meridionalis Per. (4) Sphegzdae: g. Cerceris specularis Coséa.

677. D. herbaceum Vill.—

VistTors.—Schletterer observed the following at Pola—

Hymenoptera. (a) Apzdae: 1. Anthidium strigatum Z/r.; 2. Andrena dubitata
Schenck; 3. A. limbata £v.; 4. A. morio Brull.; 5. Coelioxys aurolimbata Fors¢.;
6. Colletes lacunatus Dours.; 7.Eucera alternans Brull.; 8. E. clypeata Z7.; 9. E. in-
terrupta Baer.; 10. E. ruficollis Brull.; 11. Halictus calceatus Scop., var. obovatus A. ;
12. H. maculatus Sm., one $; 13. H. morbillosus Avzecht.; 14. H. quadricinctus F”. ;
15. H. scabiosae Ross?; 16. H. villosulus A.; 17. Nomada nobilis A. Sch.;
18. N. ochrostoma X.; 19. Nomia diversipes Z/r.; 20. Osmia andrenoides Spzm. ;
21. O. crenulata Afor.; 22. Prosopis clypearis Schenck; 23. P. variegata F.;
24. Sphecodes gibbus Z.; 25. S. subquadratus Sm.; 26. Xylocopa cyanescens
Brull. one 9. (6) Braconidae: 27. Bracon terrefactor Vill. (c) Chalcididae:
28. Leucaspis dorsigera F.; 29. L. gigas F.; 30. L. intermedia J/2. (d) Evantidae :
31. Gasteruption pedemontanus Zourn.; 32. G. rubicans Guev.; 33. G. tibiale
Schlett. (e) Pompiliidae: 34. Agenia erythropus KoA/.; 35. Pompilus quadripunc-
tatus F.; 36. Pseudagenia carbonaria Scop. (/) Scoktdae: 34. Myzine tripunctata
Rossi; 38. Scolia hirta Schr.; 39. S. insubrica Scop.; 40. S. quadripunctata F- ;
41. Tiphia minuta v.d. Z. (g) Sphegidae: 42. Cerceris arenaria L.; 43. C. bupre-
sticida Duf; 44. C. emarginata Pz.; 45. C. ferreri v.d. Z.; 46. C. labiata F.; 47. C.
leucozonica Schlef.; 48. C. quadrimaculata Duf; 49. C. rybiensis Z.; 50. C. specu-
laris Costa; 1. Gorytes quinquecinctus #.; 52. Oxybelus melancholicus Chevr. ;
53. Tachytes europaeus KoA/.; 54. T. obsoletus Rosse. (h) Tenthredinidae: 55.
Cyphona furcata V2d/., var. melanocephala Pz. (7) Vespedae : 56. Eumenes pomiformis
F.; 57. Polistes gallica L.

204. Indigofera L.

(Hildebrand, Bot. Ztg., Leipzig, xxiv, 1866, pp. 74-5.)

Flowers with explosive mechanism. Carina and alae fold downwards when
the tension is released, so that the column of stamens and pistil remains horizontal.
Automatic self-pollination is possible when the flowers fade.

282 ANGIOSPERMAE—DICOTYLEDONES

678. I. speciosa.—This species, according to Henslow (J. Linn. Soc., Bot.,
London, ix, 1867), possesses the flower mechanism described by Hildebrand (vde
supra), and represents it as expressly adapted to secure cross-pollination.

679. I. macrostachya Vent.—

Visitors.—Delpino saw Bombus italicus (‘ Ult. Oss.,’ p. 54).

205. Parochetus Buch.-Ham.

Kuhn states that cleistogamous flowers occur in this genus.

206. Melilotus Tourn.

Yellow or white bee flowers, arranged in racemes. They secrete nectar and
smell like cumarin. There is a simple valvular arrangement by which the stamens
and pistil are caused to protrude from the carina so long as pressure is exerted
by visitors (bees).

680. M. altissima Thuill. (=M. officinalis Wel/d.). (Herm. Miller, ‘ Ferti-
lisation, pp. 180-1; Kerner, ‘Nat. Hist. Pl., Eng. ed. 1, IL; Schulz, ‘ Beitrage,’ II,
p. 208; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 338-9; Loew, ‘ Bli-
tenbiol. Floristik,’ pp. 392, 395; Knuth, ‘Bloemenbiol. Bijdragen.’)—The bright
golden-yellow flowers of this species exhale an odour of cumarin. Hermann Miller
states that the calyx is only 2 mm. long, and moderately wide, so that the nectar
can be reached by short-tongued insects. The alae and the carina are partially
fused together on either side, so that both are depressed when the flower is visited
by an insect, and simultaneously return to their original position when the pressure
is removed. The latter movement is due to the fact that there are two finger-
like processes directed backwards and inwards on the upper basal angle of the
alae, and embracing the column formed by the stamens and pistil. These processes
slide along the upper side of the column during depression, returning to their
original position when the pressure is removed, and guiding the carina and alae
back to theirs.

As the stigma projects beyond the anthers, cross-pollination is secured by
insect-visits ; failing these, automatic self-pollination is difficult, but Kerner states
that it is effective.

Visitors.—Herm. Miller observed the following.—

Hymenoptera. (a) Apidae: 1. Andrena dorsata K. 9, skg. and po-dvg.;
2. Apis mellifica Z. %, very freq., do.; 3. Coelioxys quadridentata Z. é, skg.; 4.
Heriades truncorum JL. 9, po-cltg.; 5. Osmia sp. (4) Sphegzdae: 6. Ammophila
sabulosa Z. 4, skg. (c) Zenthredinidae: 4. Tenthredo sp., vainly searching for
nectar.

In Schleswig-Holstein I saw only the honey-bee, in great numbers, po-cltg.
and skg.

Loew observed a short-tongued bee (Halictus zonulus Sm. 9) po-cltg. in Central
Germany, and the following at Warnemiinde.—

A. Hymenoptera. (a) Apidae: 1. Andrena pilipes F. 9, po-cltg.; 2. Anthidium
strigatum Zér. 9, do.; 3. Coelioxys quadridentata Z. $, skg.; 4. C. elongata Zep. 9,
do.; 5. C. rufocaudata Sm. 6, do.; 6. Halictus rubicundus Cér. 9, po-cltg.; 7. Osmia

LEGUMINOSAE 283

claviventris Zhoms.d, skg. (b) Sphegzdae: 8. Oxybelus furcatus Lep., skg. B. Diptera.
Limpidae: g. Empis sp.

Schletterer records the following for Pola.—

Hymenoptera. (a) Apidae: 1. Anthidium strigatum Z/r.; 2. Andrena con-
vexiuscula K.; 3. A. flessae Pz.; 4. A. limbata Zv.; 5. A. lucens /mA.; 6. A. morio
Brull,; 7. A. nana K.; 8. A. thoracica #.; 9. Halictus calceatus Scop.; 10. H.
morbillosus A7rchd. ; 11. H. patellatus Mor.; 12. H. sexcinctus /.; 13. H. tetrazonius
Kig.; 14.Nomia diversipes L/r. (6) Tenthredinidae: 15. Amasis laeta F'.; 16. Cephus
haemorrhoidalis F.; 17. C. pygmaeus Z. (c) Vesprdae: 18. Polistes gallica Z.

Schulz saw bees perforating the flowers.

681. M. officinalis Desr. (=M. arvensis Wadir.). (Knuth, ‘Bloemenbiol.
Bijdragen.’)\—The golden-yellow flowers of this species also smell like cumarin, and
possess the same mechanism as those of M. altissima.

VistTors.—I have seen
the honey-bee.

Loew observed the
following in Mecklenburg
(‘ Beitrage,’ p. 45).—

A. Diptera. Lmpi-
dae: 1. Empis sp. B.
Hymenoptera. (a) Aj:-
dae: 2. Andrena pilipes
F. 9, po-clig.; 3. Anthi-
dium strigatum Zr. 9,
do.; 4. Coelioxys conica
L. 3, skg.; 5. C. elongata
Lep. 9, do.; 6. C. octo-
dentata Zep. 4, do.; 7.
Halictus rubicundus Chr.
3, po-cltg.; 8. Osmia cla-

viventris Zhoms. 5, skg. Fic. 91. Melrlotus officinalis, Willd. (after Herm. Miller). (ay

(4) Sphegidae: 9. Oxybelus Flower seen from the side. (2) The same, after removal of the vexillum

furcatus Lep skg and calyx; seen from above. (3) The same, after the alae and carina
- Hi

have been pressed down; seen from the side. a@, anthers; d, pivot on which
Schletterer and Dalla the carina turns; e, depressed parts of the alae, the inner surfaces of
é which are united with the outer surfaces of the two carinal petals, by the
Torre record the following fusion of superficial cells; _/ finger-like processes of the superior basal angles
of the alae; g, column of stamens and pistil; 4, nectar-passage ; gv, style;
bees for the Tyrol.— oe, oie
1. Bombus hortorum

L.; 2. B. mastrucatus Gerst.; 3. Ceratina cyanea XK.

682. M. alba Desr. (=M. vulgaris Wil/d.). (Herm. Miiller, ‘ Fertilisation,’
p. 181, ‘ Weit. Beob.,’ II, p. 252; Loew, ‘Bliitenbiol. Floristik,’ p. 392; Kirchner,
‘ Beitrige,’ p. 40; Knuth, ‘Bloemenbiol. Bijdragen..)—The white flowers of this
species, which also smell of cumarin, possess the same mechanism as those of
M. officinalis. The calyx is 2 mm. long; the blade of the oblique vexillum is
4 mm. in length; the carina and alae project for 24 mm. from the calyx. The
anthers mature shortly before anthesis, and do not reach the level of the stigma.

Visrrors. — Besides the honey-bee (po-cltg. and nect-skg.) numerous other
insects have been observed. Loew saw the following at Warnemiinde.—

284 ANGIOSPERMAE—DICOTYLEDONES

A. Hymenoptera. (a) Apzdae: 1. Andrena cineraria Z. 9, po-cltg.; 2. A.
fulvicrus K. 9, do.; 3. A. gwynana X,, var. aestiva Sm. 9, do.; 4. A. pilipes F’. 9, do.;
5. Coelioxys conica Z. 9, skg.; 6. C. elongata Zep. 4, do.; 7. C.sp., do.; 8. Colletes
fodiens K. 8, do.; 9. Macropis labiata Pz. 9, do. (4) Sphegidae: 10. Cerceris
arenaria Z. 9 and 4, skg. (c) Vespidae: 11. Odynerus parietum Z., var. renimacula
Lep., skg.; 12. Eumenes coarctata Z., do. B. Diptera. (a) Chironomidae: 13.
Ceratopogon fasciatus Mg. 9. (4) Conopidae: 14. Physocephala rufipes /., skg.
(c) Muscidae: 15. Olivieria lateralis F. (d¢) Syrphidae: 16. Eristalis intricarius Z.,

skg.; 17. Helophilus pendulus Z., do.; 18. Melithreptus sp.; 19. Volucella bombylans
L., do.

Friese, quoting Konow and Sajo respectively, gives Systropha curvicornis Scop.,
not infrequent, for Mecklenburg, and Osmia grandis AZor. (a 5) for Hungary.

Alfken observed the following bees at Bremen.—

1. Andrena flavipes Pz.9; 2. A. propinqua Schenck 9; 3. Bombus lapidarius Z. ¥.

Ducke saw the following bees at Aquileja—

1. Eucera (Macrocera) ruficornis #. $; 2. E. (M.) salicariae Zep. 9.

Schletterer and von Dalla Torre noticed the bee Halictus rubicundus Chr. 6,
in the Tyrol.

Schmiedeknecht mentions Andrena nasuta G7r. as a rare visitor.

Schulz observed flowers perforated by bees.

683. M. dentata Pers.—

VisiTors.—In this species also Schulz observed flowers perforated by bees.

684. M. caerulea Desr. (=Trigonella caerulea Ser.). (Kirchner, ‘ Beitrage,’
p. 41.)—The bright blue flowers of this plant, which are arranged in short crowded
racemes, possess a mechanism agreeing with those of the other species of Melilotus ;
but the stigma is either surrounded by the anthers or projects but little beyond
them. The calyx is 3-4 mm., the limb of the vexillum 5 mm., the alae 3 mm.,
and the carina 2 mm. in length. Each ala possesses a basal process overlapping
the staminal tube, so that the alae and carina can only be depressed by the
application of considerable force. The stamens and pistil then spring up towards
the vexillum, but return again into the carina when the pressure is removed.

VisiTors.—Kirchner observed the honey-bee.

207. Trifolium Tourn.

Yellow, white, or red bee flowers (very rarely Lepidopterid flowers as well)
arranged in heads; nectar-secreting and odorous. There is a simple valvular
arrangement as in the last genus. According to Kuhn (Bot. Ztg., Leipzig, xxv, 1867),
some species bear cleistogamous flowers; Darwin states that this is the case in
T. polymorphum (‘Forms of flowers’).

685. T. repens L. (Herm. Miller, ‘Fertilisation,’ pp. 181-3, ‘ Weit. Beob.,’
II, p. 246; Darwin, Ann. Mag. Nat. Hist, London, Ser. 3, ii, 1858, p. 4603
Lindmann, ‘ Bidrag till Kanned. om Skand. Fjallvaxt. Blomn. o. Befrukt.’; Verhoeff,
“BI. u. Insekt. a. d. Ins. Norderney’; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi,
1894, pp. 342-9; Loew, ‘ Bliitenbiol. Floristik,’ p. 395; Knuth, ‘Bl. u. Insekt. a. d.
nordfr. Ins.,’ pp. 59-60, 153, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 232,
‘Bl. u. Insekt. a. d. Halligen,’ ‘Bl. u. Insekt. a. Helgoland,’ ‘ Bloemenbiol. Bijdragen,’

LEGUMINOSAE 285

‘Bliitenbiol. Beob. in Thiiringen,’ ‘Bliitenbiol. Beob. a. d. Ins. Riigen,’ &c.)—The
white or reddish flowers of this species smell like honey. The nectar is secreted
as usual, i.e. on the inner side of the base of the staminal tube. The calyx-tube
is only 3 mm. long, so that even short-tongued bees are able to reach the nectar.
Hermann Miiller states that the alae are partly fused with the carina on either
side, so that both move simultaneously upwards and downwards. Their depression
by the weight of a nectar-seeking insect is facilitated by the fact that the claws
are very slender. These claws are fused for the most part with the staminal tube,
which is cleft above. The return of the parts to their original position is chiefly
due to the action of the vexillum and alae. The broad claw of the former grasps
the other petals, as well as the stamens and pistil, and, owing to its elasticity,
guides their bases back to the original position when the pressure is removed.
The front parts of the petals, and the reproductive column, return to their places
because the dorso-basal processes of the alae are converted into two elastic swellings
lying close together above the staminal tube.

In order to reach the nectar,
an insect visitor must thrust its
head beneath the vexillum, and
in doing so has only the alae
for a support. These and the
carina are consequently pressed
downwards, and the vexillum
upwards, while stamens and
stigma protrude from the carina.
As the stigma projects some-
what beyond the anthers, cross-
pollination is greatly favoured.
Bees alone are able to work
the flower mechanism properly,
so as regularly to effect cross-
pollination; other visitors, e. g.

Diptera and Lepidoptera, only Fic. 92. Trtfoléum repens, L. (after Herm. Miller). (1)
ss 5 Flower seen from below. (2) The same after removal of calyx
do this occasionally. and vexillum; seen from above. (3) Right ala, seen from the

MacLeod gives a detailed Marist 6 secanyongrs (4,0 2 Oe cto re
account of the flower mechan- processes.

ism, which deviates in some

respects from Miiller’s description. He arrives at the following conclusions (Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, pp. 342-9).—

The flower of T. repens consists of two co-operating (synergic) parts, i.e. the
alae and carina, which are united together, and move simultaneously in the same
direction. Each of these parts, however, may execute independent movements.
In this connexion two other organs have to be considered, the filaments (which
are passive) and the claw of the vexillum (which seems to play an active part). When
the alae and the carina of an unvisited flower are pressed down from 30 to 40 times
in succession, the mechanism does not appear to be exhausted: the reproductive
organs always return into the carina after the pressure has been removed. If,

286 ANGIOSPERMAE—DICOTYLEDONES

however, pressure is exerted on the carina alone or the alae alone several times
in succession, these parts resume their original position again, but slowly, and the
mechanism gives distinct indications of exhaustion. By the union of various co-
operating (synergic) organs, Nature has not only afforded greater certainty of
movement, but has even been able to minimize exhaustion.

The movements of the higher animals are similarly rendered effective by the
grouping of co-ordinated muscles: in this way the same advantages are obtained as
by the co-operation of alae, carina, and vexillum in the flower of Trifolium repens.

According to Darwin’s experiments (‘Cross- and Self-fertlsn., p. 361) cross-
pollination is necessary for the complete fertility of white clover; when insects are
excluded the flowers are to a great extent self-sterile.

Visirors.—Hermann Miller in Westphalia (H. M.), Buddeberg (Budd.) in
Nassau, and myself (Kn.) in Schleswig-Holstein (S.-H.), Rigen (R.), Helgoland (H.),
and Thuringia (T.) have observed the following.—

A. Diptera. (a) Conopidae: 1. Myopa buccata Z., skg. (H. M.); 2. M.
testacea L., do. (H. M.). (4) Syrphzdae: 3. Eristalis sp. (H. M.); 4. Volucella bom-
bylans Z., skg. (H. M.). B. Hymenoptera. Apzdae: 5. Andrena fulvicrus X. 9,
skg. (H. M.); 6. A. nigriceps K.9, do. (H. M.); 7. A. sp. (Kn., H.); 8. Antho-
phora quadrimaculata Pz. 9 (Kn., S.-H. and H.); 9. Apis mellifica Z. ¥, very freq.,
skg., and po-cltg. (H. M.; Kn., S.-H. and R.); 10. Bombus cognatus Steph. ¥ and 9
(Kn., S.-H.); 11. B. derhamellus A. (Kn., S.-H.); 12. B. lapidarius Z. § and 9 (Kn.,
S.-H. and T.); 13. B. pratorum Z. ¥, skg. (H.M.); 14. B. rajellus «. 9, do.
(Kn., R.); 15. B. terrester Z. (Kn., T.); 16. Cilissa leporina Ps. 6, skg. (Budd.) ;
17. Colletes balteatus My/. (Kn., S.-H.); 18. Eucera difficilis (Duf.) Per. (Kn., R.);
19. Halictus maculatus Sm. 9, po-cltg. (H. M.); 20. H. sexnotatus A’ 9, skg. (H.M.);
21. H. smeathmanellus A. 9, do. (Budd.); 22. H. tarsatus Schenck 9, do. (H. M.);
23. H. zonulus Sm. 9, do. (H. M.); 24. Megachile willughbiella A. é (H. M.);
25. Psithyrus quadricolor Zep. 6, skg. (H.M.). C. Lepidoptera. Rhopalocera: 26.
Coenonympha pamphilus Z., skg. (H. M., T.); 27. Epinephele janira Z., skg. (Kn.,
S.-H.); 28. Hesperia, do. (H. M.); 29. Melitaea athalia Z's. do. (H. M., T.);
30. Lycaena semiargus fo#t., do. (Kn., S.-H.); 31. Pieris brassicae Z., do. (H. M.;
Kn., S.-H.) ; 32. P. napi Z., do. (H. M.).

Herm. Miiller also noticed 11 bees, a hover-fly, and 10 Lepidoptera in the Alps
(‘ Alpenblumen,’ p. 244).

Von Dalla Torre saw the bee Stelis aterrima Pz. 9 in the Tyrol.

Hoffer records the rare Alpine humble-bee Bombus alpinus Z., an old nest-
female, for Steiermark. Loew observed Halictus zonulus Sm. 9, po-cltg. in the
same province (‘ Beitrage,’ p. 53), and the honey-bee in Silesia (op. cit., p. 34).

Schiner saw the Muscid Ocyptera pusilla J/g. (probably vainly endeavouring
to suck) in Austria. Réssler noticed the butterfly Colias hyale Z. as a useless visitor
at Wiesbaden. Schenck records the bee Melitta leporina Pz. for Nassau.

Schletterer mentions the following bees for the Tyrol (T.) and Pola—

1. Andrena dubita Schenck; 2. Megachile muraria Z.; 3. Osmia tridentata
Duf. et Perr.; 4. Stelis aterrima Pz. (T.).

Friese, on the authority of Konow, gives the two following rare bees for
Mecklenburg.—

1. Colletes nasutus Sw.; 2. Meliturgus clavicornis L/r.
Alfken and Hoppner (H.) observed the following bees at Bremen.—

LEGUMINOSAE ; 287

1. Andrena flavipes Pz. 9 (2nd generation); 2. Bombus agrorum /. ¥; 3. B.
arenicola Zhs. §; 4. B. derhamellus &. 9, ¥, and 4; 5. B. distinguendus Zor. ¥ skg.,
9 (H.); 6. B. hortorum Z. ¥, skg. and po-cltg.9; 7. B. lapidarius Z. 4; 8. B.
lucorum Z.¥; 9. B. muscorum /. 9; 10. B. sylvarum Z. ¥; 11. B. terrester L. 8;
12. Coelioxys rufescens Zep. 9 and $; 13. Melitta leporina Pz. 9 and 6; 14. Psi-
thyrus barbutellus X. 6.

Verhoeff saw the following in Norderney.—

A. Diptera. (a) Syrphidae: 1. Syrphus corollae /. 9, not infrequent ; 2. S.
pyrastri Z., occasional. B. Hymenoptera. (a) Apzdae: 3. Bombus hortorum Z.
¥; 4. B. lapidarius Z. 9 ¥ and 4, freq., skg.; 5. B. terrester Z. ¥, freq. C. Lepi-
doptera. (a) Noctuidae: 6. Plusia gamma Z.

Alfken noticed the following bees in Juist.—

1. Bombus distinguendus Mor. ¥ po-cltg. and skg.; 2. B. hortorum JZ. ¥, do. ;
3. B. lapidarius Z. ¥, do.; 4. B. muscorum /. ¥, do.

MacLeod observed Apis, a humble-bee, a fossorial wasp, and 7 Lepidoptera
in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 349); and 8 humble-bees,
2 short-tongued bees, a lepidopterid, and a beetle in the Pyrenees (op. cit., iii, 1891,
p- 436). H. de Vries saw the honey-bee in the Netherlands (Ned. Kruidk. Arch.,
Nijmegen, 2. ser., 2. deel, 1875).

E. D. Marquard noticed the bee Melitta leporina Pz. in Cornwall.

In England Smith saw the bee Melitta leporina Pz. and Saunders the bee
Colletes marginatus LZ.

In Dumfriesshire Apis and a hover-fly have been recorded (Scott-Elliot, ‘Flora
of Dumfriesshire,’ p. 46).

Schneider gives the following bees for Arctic Norway (Tromsg. Mus. Aarsh.,
xvii, 1895).

1. Bombus lapponicus Z. ¥ and 8; 2. B. pratorum Z. ¥ and $; 3. B. scrimshi-
ranus K.¥ and 6; 4. B. terrester Z. ¥ and 6; 5. Psithyrus quadricolor Zep. 6;
6. P. vestalis Fourcr. 6.

Lindman noticed numerous humble-bees in the Dovrefjeld.

Loew saw the honey-bee on the variety a/ropurpureum in the Berlin Botanic
Garden.

686. T. hybridum L.—The flowers of this species being at first white and
erect, afterwards rose-red and curved downwards, the heads present a white centre
and a rose-red margin, and are thus rendered more conspicuous. The mechanism
agrees with that of the last species.

Visirors.—Buddeberg saw a bee, Melitta leporina Pz. 8, skg., in Nassau.

1 observed the honey-bee and the two following humble-bees in Thuringia
(‘Bloemenbiol. Bijdragen’).— 1. Bombus agrorum /. $ (Coburg, July 4, 1894);
2. B. terrester L. ¥ (Inselsberg, July 16, 1894).

687. T. fragiferum L. (Herm. Miller, ‘ Fertilisation, p. 183, ‘Weit. Beob.,’
p. 246; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 349; Knuth, ‘Bl. u.
Insekt. a. d. nordfr. Ins., pp. 60, 153, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’
‘Bl. u. Insekt. a. d. Halligen.’)—As Hermann Miiller states, the mechanism in this
species is the same as in T. repens, but all the parts of the flowers are smaller.

Vistrors.—In Kiel I observed the following nect-skg. bees.—

288 ANGIOSPERMAE—DICOTYLEDONES

1. Apis mellifica Z. ¥ (and in Sylt); 2. Bombus lapidarius Z. ¥; 3. B. sylvarum
£.%. In the island Langeness (the Halligen), so far as I was able to observe, the
only pollinator is Anthophora quadrimaculata Pz. 9.

In Westphalia Hermann Miiller only noticed the honey-bee, passing from
T. fragiferum to T. repens, and vice versa.

Alfken saw the humble-bee Bombus terrester Z. ¥ in Juist; also at Bremen,
freq., skg. and po-cltg. Even when covered by his collecting-net it continued
quietly to collect pollen.

Heinsius observed two bees in Holland.—

1. Apis mellifica Z. §¥; 2. Bombus lapidarius Z. and ¥.

MacLeod records Bombus lapidarius Z. ¥ for Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 346), and Schiner the Muscid Ocyptera pusilla AZg., vainly trying
to suck, for Austria.

688. T. montanum L. (Herm. Miiller, ‘Fertilisation,’ p. 181, ‘ Weit. Beob.,’
II, pp. 250-2, ‘Alpenblumen,’ p. 143.)—The white flowers of this species, according
to Hermann Miiller, are about 5 mm. in length from the base to the tip of the carina,
while the calyx is 2-3 mm. long. The nectar is therefore accessible to all insects
with a proboscis 5 mm. in length. Bees visiting the flowers cause the reproductive
organs to protrude as in T. repens. Lepidoptera also effect cross-pollination, for
they touch the stigma and stamens as they push their proboscis along the groove of
the folded vexillum into the dorsal cleft of the carina. The plant is therefore adapted
to cross-pollination by both bees and Lepidoptera. The mechanism agrees in other
respects with that of T. repens, except that the rounded alar processes are more
feebly developed.

VisiTors.—Hermann Miiller observed the honey-bee in Westphalia, and the
following insects in Thuringia.—

A. Hymenoptera. (a) Apzdac: 1. Apis mellifica Z. ¥, very freq., skg.;
2. Bombus pratorum Z. 4, skg.; 3. Nomada roberjeotiana Pz. 9, do.; 4. N. rufi-
cornis Z. 9, do. (4) Sphegidae: 3. Ammophila campestris Z/r. 9 and 4, skg.
B. Lepidoptera. Rhopalocera: 6. Hesperia sylvanus £sp., skg.; 7. Lycaena
aegon W.V. 9, do.; 8. L. corydon Poda, do.; 9. Melitaea athalia Zsp., freq., skg.
persistently.

Miiller saw 8 bees and 8 Lepidoptera in the Alps. Dalla Torre and
Schletterer noticed Bombus mastrucatus Gers/. 5 in the Tyrol; and MacLeod
records a humble-bee and an Andrena for the Pyrenees (Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, pp. 436-7.

689. T. pratense L. (Darwin, ‘Origin of Species,’ Chap. III; Herm.
Miiller, ‘Fertilisation, pp. 184-6, ‘Weit. Beob.,’ II, pp. 246-7; Lindman, ‘ Bidrag
till Kanned. om Skand. Fjallvaxt. Blomn. o. Befrukt.’; Loew, ‘ Bliitenbiol. Floristik,’
p- 396; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 339-42; Schulz,
‘ Beitrage,’ II, p. 208; Kerner, ‘ Nat. Hist. Pl.,’ Eng. ed. 1, IL; Knuth, ‘Bl. u. Insekt.
a. d. nordfr. Ins.,’ pp. 59, 152-3, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’
p. 232, ‘Bl. u. Insekt. a. d. Halligen,’ ‘ Bliitenbiol. Beob. a. d. Ins. Riigen,’ ‘ Bl. u.
Insekt. a. Helgoland,’ ‘ Bliitenbiol. Beob. in Thiiringen,’ ‘ Bloemenbiol. Bijdragen.’)—
The red, rarely white, flowers of this species, which smell like honey, are aggregated
into conspicuous spherical heads, abundantly visited by insects. Herm. Miiller

LEGUMINOSAE 289

states that the nectar is concealed in the base of a tube 9-10 mm. long, formed by
fusion of the nine lower filaments with the claws of the petals. The upper free stamen
lies on one side of the flower, so that the whole of the cleft in the staminal tube
serves as a nectar passage. When a bee thrusts its proboscis under the vexillum
towards the nectar, it holds on with its fore-legs to the alae, which are united with
the carina, while its other two pairs of legs are supported by deeper parts of the
capitulum. The pressure exerted depresses the carina and alae, so that the stigma
protrudes, to be immediately followed by the upwardly dehisced anthers, which press
against the under-side of the bee’s head. The stigma consequently receives pollen
brought from another flower, and this is replaced by a fresh supply. Cross-pollina-

tion is thus assured. Self-pollination may also take place when the head of the bee
is withdrawn, but Darwin says

that this is ineffective (though
Kerner is of the opposite opinion),
and is rendered inoperative by
the previously effected
pollination.

The return of the parts to
their original position, after re-
moval of the pressure exerted by
the insect, is brought about by
the intrinsic elasticity of the base
of the carina. The thin easily
rotated claws of the alae separate
from the common tube, and the
two large vesicular processes of
these petals grasp the reproductive
column from above, their elasticity
keeping the stamens, pistil, and

cross-

FIG. 93.
(1) Flower seen from below.
the vexillum; seen from above.

Trifolium pratense, L. (after Herm. Miiller).
(2) The same, after removal of
(3) Anterior part of the same

the parts which enclose them,
in their proper positions. The
staminal tube is continued into
free, stiff, upwardly curved fila-
ments, somewhat thickened at the
end; the style curves upwards
between them in such a way that
the stigma projects above the
anthers to some extent.

with the edges of the carina pressed apart (enlarged to twice the

scale of 1 and 2). (4) Right ala, with the claw torn off; seen
from the inner side. (5) Right half of the carina, with the claw
torn away ; seen from the outside. (6) The depressed carina,
with the protruding anthers and stigma (at a higher level); seen
from the side. a, calyx; 5, tube formed by fusion of 9 filaments
with the claws of the petals; c, vexillum; @, hollow on the inner
side of the ala; ¢, lower border of ala; / outside of ala; yg, base
of ala dilated into a vesicular process; 4, carina; z, style; &, free
upper stamen; /, stigma ; 7, anthers; 2, point of fusion between
alae and carina ; 0, point at which the carina is moved up and
down; #, outwardly curved part of the upper margin of the ala;
9, expansion of the vexillum, folded over below.

In order to reach the nectar legitimately, an insect must have a proboscis of

at least 9-10 mm. in length, corresponding to that of the corolla-tube. Many
species of humble-bee and other bees are regular pollinators, and some Lepidoptera
are casual agents of cross-pollination. The pollen, on the other hand, is accessible
to all short-tongued insects which are skilful enough to depress the carina. Such
insects, e.g. the honey-bee, also effect regular cross-pollination. Nectar is also
stolen from the red clover, especially by Bombus terrester (with a proboscis only
7-9 mm. long), and the honey-bee (with a proboscis of 6 mm.); these perforate

DAVIS. IT U

290 ANGIOSPERMAE—DICOTYLEDONES

the flowers from without, and thrust their proboscis through the hole to the desired
booty. The opening is made use of by other thieving insects.

Vistrors.— Lindman saw humble-bees and Lepidoptera on the Dovrefjeld.
Hermann Miller (H. M.) in Westphalia, Buddeberg (Budd.) in Nassau, and myself
(Kn.) in Schleswig-Holstein (S.-H.), in Rigen (R.) on the high ground of Helgoland
(H.), and in Thuringia (T.), observed the following.—

A. Diptera. (a) Bombyliidae: 1. Systoechus sulphureus A/zkan, vainly seeking
for nectar (H.M.). (4) Conopidae: 2. Sicus ferrugineus Z., vainly seeking for nectar
(H.M.). (c) Syrphidae: 3. Volucella bombylans, vainly seeking for nectar (H. M.).
B. Hymenoptera. Apzdae: 4. Andrena convexiuscula &. 4, vainly trying to suck
(Budd.); 5. A. fasciata Wesm. 9 and 4, do. (H.M.); 6. A. fulvicrus A. 9, do. (H. M.);
7. A. labialis K. 8, do. (H.M.); 8. A. schrankella JVy/. 9 (H.M.); 9. A. xanthura
K. 9, po-cltg. (H.M.); 10. Anthidium manicatum Z. 9 and 4, skg. (H.M.); 11.
Anthophora aestivalis Pz. do. (Budd.; Kn., R.); 12. A. pilipes #., do. (H.M.);
13. Apis mellifica Z. ¥, stealing nectar from flowers perforated by B. terrester, and
po-cltg. (H.M.; Kn., S.-H.); 14. Bombus agrorum /. ¥ and 9, skg. (H.M.; Kn.,
S.-H., R., and T.); 15. B. confusus Schenck ¥ and 9, do. (H.M.); 16. B. derhamellus
.,, do. (Kn., S.-H.); 17. B. distinguendus Aor. 9, do. (H. M.); 18. B. hortorum Z.,
var. hortorum L. 4, do. (Kn., T. and R.); 19. B. lapidarius Z. ¥ and 9, do. (H.M.,
Kn., S.-H.); 20. B. muscorum /., do. (H.M.); 21. B. pratorum Z. ¥ (proboscis
8 mm.), 9 (proboscis ro mm.), do. (Budd.); 22. B. rajellus K.¥% and 9, do. (H.M.;
Kn., R.); 23. B. sylvarum Z. 9, do. (H.M.; Kn., R.); 24. B. terrester Z., perforating
the corolla and stealing the nectar, even of buds (H.M.; Kn,, T.); 25. Cilissa
leporina £2. 4, vainly trying to suck (Budd.); 26. Colletes fodiens K. 9, po-cltg.
(H.M.); 27. Diphysis serratulae Pz. 9, do. (H.M.); 28. Eucera difficilis (Duf.) P&r.,
skg. (Kn., Greifswalder I. near R., and H.); 29. E. longicornis 9 and 6, do. (H. M.;
Kn., S.-H. and H.); 30. Halictus cylindricus /. 9, vainly trying to suck (Budd.);
31. H. flavipes /. 9, po-cltg. (H.M.); 32. H. interruptus Pz. 9, do. (H.M., T.);
33. H. malachurus A. 9, do. (Budd.); 34. H. sexnotatus X. 9, vainly trying to suck
(Budd.); 35. H. tetrazonius A7/. 9, do. (Budd.); 36. Megachile circumcincta LZ. 9,
skg. and po-cltg. (H.M.); 37. Osmia aurulenta Pz. 9, po-cltg. (H.M., T.); 38. O.
aenea Z. 9, skg. and po-cltg. (H.M.); 39. Psithyrus barbutellus X. 9, skg. (H.M.);
40. P. campestris Pz. 9 (H.M.); 41. P. rupestris /. 9, skg. (H.M.; Kn,, S.-H.);
42. P. vestalis Fourcr. 9, do. (H.M.). C. Lepidoptera. (a) Bombycidae: 43.
Gnophria quadra Z., settling on the flowers (H.M.). (6) Mocturdae: 44. Plusia
gamma Z., skg. (H.M.; Kn., S.-H.). (c) Rhopalocera: 45. Argynnis adippe Z. skg.
(Kn., T.); 46. Coenonympha pamphilus Z. 9, do. (H.M.; Kn., S.-H.); 47. Epine-
phele janira Z., do. (H.M.; Kn., S.-H.); 48. Hesperia sylvanus £sp., do. (H. M.);
49. H. thaumas A/n., skg. (Budd.); 50. Melanargia galatea Z., freq., skg. (H. M..
T.); 51. Papilio podalirius Z., skg. (H. M.); 52. Pieris brassicae Z., skg. (H.M.;
Kn., S.-H., and H.); 53. Pararge megaera Z., skg. (H.M.); 54. Vanessa urticae Z.,
skg. (H.M., Kn., H.). (d) Zygaentdae: 55. Zygaena filipendulae Z., skg. (Kn.,
S.-H.); 56. Z. sp., effecting cross-pollination only occasionally (Kn., R.).

Krieger saw two bees (1. Andrena labialis A.; 2. Eucera longicornis Z. 9}
at Leipzig.
Schmiedeknecht gives the following bees for Thuringia —

1. Bombus agrorum /. 6; 2. B. distinguendus Mor. 9 and 8; 3. B. hortorum
£,9 ¥Y%and 6; 4. B. lapidarius Z. 9 ¥ and 4; 5. B. latreillellus A. (=B. subterraneus
L.)9 Yandé; 6. B. mastrucatus Gers/.d; 7. B. mesomelas Gerst. 9 ¥ and 6; 8. B.
muscorum /. 9 ¥andé; 9. B. pomorum Z.9 and 4, freq., 6 occasional; ro. B.
ruderatus /. 9 ¥ and é; 11. B.sylvarum Z. 9 and é; 12. B. variabilis Schmuedekn.
gYand d; 13. Psithyrus vestalis Fourcr. 9.

LEGUMINOSAE 291

Schenck records the following bees for Nassau.—

1. Andrena labialis AK. 9 and 6; 2. A. convexiuscula X.; 3. Bombus confusus
Schenck z: 4. B. derhamellus KX. 9 ¥ and $; 5. B. lapidarius Z.; 6. B. muscorum F.
9 ¥ ands; 7. B. pomorum Pz. 9; 8. Halictus tetrazonius A7p. 9.

The following lists of bees are given as stated.—

Loew (Hesse).—Eucera longicornis Z. 9, po-cltg.

Schletterer and von Dalla Torre (Tyrol).—1. Andrena fulva Schr. 9; 2. A. nana
&. 9 and 8; 3. Bombus sylvarum Z. 9 and ¥; 4. Chalicodoma pyrenaica Zep. 9.

Schletterer (Pola).—1. Bombus sylvarum Z.; 2. B. terrester Z.; 3. Eucera
alternans Brull.; 4. Megachile pyrenaica Zep.

Ducke (Trieste).—1. Andrena korleviciana Friese 9; 2. Eucera difficilis (Duf.)
Per. 9; 3. Osmia aurulenta Pz. 9 and 4, freq.; 4. Rophites canus Zv. 9 and é.

Hoffer (Steiermark).—1. Bombus agrorum /. 9 and ¥; 2. B. hortorum LZ. 3;
3. B. lapidarius Z. 9 and ¥; 4. B. mesomelas Gersz. 9 (v. Dalla Torre); 5. Psithyrus
vestalis Fourcr. 6.

Friese (Alsace (A.), Fiume (F.), Mecklenburg (M.), and Hungary (H.)).—
1. Andrena convexiuscula X., very freq. (M.); 2. A. labialis A., occasional (H.);
3. Eucera seminuda Brui/. 9, do. (H.); 4. Melitta dimidiata or., not infreq. (F. and
H.); 5. M. leporina Pz., do. (M.); 6. Podalirius fulvitarsis Brul/. 9, freq. (A.); 7. P.
parietinus /’., not infreq. (M.).

Alfken observed the following at Bremen.—

A. Diptera. Muscidae: 1. Prosena siberita /. B. Hymenoptera. Afzdae:
2. Andrena convexiuscula K. 9 and 6; 3. A. labialis A. 9 and 6; 4. Bombus agrorum
F.% and; 5. B. arenicola Zhs. 9 and ¥; 6. B. derhamellus X.9 4 and ¥; 7. B.
distinguendus Zor. 9, skg. and po-cltg., ¥ and ; 8. B. hortorum ZL. 9 and ¥, skg. and
po-cltg., é skg., var. nigricans Schmzedekn. ¥ and 6, skg.; 9. B. lapidarius ZL. 9 and ¥,
skg. and po-cltg.; 10. B. lucorum Z. 9 and ¥ (perforating the tube of the corolla) ;
11. B. muscorum F. 9; 12. B. pomorum Pz. 9 and ¥; 13. B. ruderatus /. 9 and ¥,
skg. and po-cltg.; 14. B. sylvarum Z. 9 and ¥, do.; 15. B. subterraneus Z. 9 and §,
do.; 16. Coelioxys quadridentata Z. 6, skg.; 17. Eucera difficilis (Duf.) Pé&. 9 and 6;
18. Megachile circumcincta A. 9; 19. M. willughbiella AK. 6; 20. Melitta leporina
Pz. 9; 21. Osmia caerulescens Z. 6; 22. O. claviventris Zhs. 8; 23. Podalirius
borealis Mor. 9 and 4, skg.; 24. P. parietinus #9, do.; 25. P. retusus Z. 6 and 9;
26. Psithyrus barbutellus A. 9, skg.; 27. P. campestris /z. 9, do.; 28. P. rupestris ;
2g. P. vestalis Fourcr. 9, skg.

Verhoeff noticed the following bees in Norderney.—

1. Bombus lapidarius Z. 9, skg.; 2. B. latreillellus A. (=B. subterraneus Z.),
one 9; 3. B. terrester Z. 9 and ¥ (nectar-thieves), skg.

Alfken saw the following bees in Juist.—

1. Bombus hortorum Z. ¥ and 6; 2. B. muscorum Ff. 9; 3. B. ruderatus F”. 6.

Morawitz records Podalirius borealis Mor. for St. Petersburg; H. de Vries
observed 3 bees (1. Andrena labialis XK. 9; 2. A. xanthura K.9; 3. Apis mellifica
L. ¥), and 6 humble-bees (1. Bombus agrorum F. ¥ and 8; 2. B. hortorum Z.
Yand 6; 3. B. pratorum Z. $; 4. B. sylvarum Z.¥; 5. B.subterraneus Z.; 6. B.
terrester Z. ¥) in the Netherlands (Ned. Kruidk. Arch., Nijmejen, 2. ser., 2. deel,
1875).

MacLeod saw the following in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,

pp. 341-2).—Apis, 11 humble-bees, an Eucera, an Andrena, a hover-fly, and 12
Lepidoptera ; also 8 humble-bees, an Anthophora, 11 Lepidoptera, a Bombylius,
and a hover-fly in the Pyrenees (op. cit., ili, 1891, pp. 435-6).
In Dumfriesshire 3 humble-bees were recorded (Scott-Elliot, ‘ Flora of Dumfries-
shire,’ p. 45).
U2

292 ANGIOSPERMAE—DICOTYLEDONES

The variety (4) mzvale (=the species T. nivale Sved.), which lives for the most
part above the limit of trees, and bears dirty white instead of red flowers, was seen
by Hermann Miller to be visited in the Alps by 7 humble-bees and 17 Lepidoptera,
i.e. 71 % of the visitors were of the latter kind. The same investigator noticed that
the ordinary variety was visited in the same region by 15 humble-bees, 21 Lepido-
ptera, and a beetle, i.e. 55% of the visitors were Lepidoptera (‘ Alpenblumen,.
Pp. 241-3).

I may here introduce an observation of Hermann Miller’s on a well-known
series of inferences.—‘ The more cats there are, the fewer are the mice; the fewer
mice there are, the more abundant are the humble-bees; the more numerous the
humble-bees, the more prolific is the red clover; therefore, the more numerous
the cats, the more prolific is the red clover.’ It is true that humble-bees are the
chief pollinators (though not, as Darwin thought, the only ones) of the red clover,
but when they are excluded there are numerous nectar-sucking and pollen-collecting
insects that can effect the cross-pollination necessary for complete fertility. Hence
one link in the above chain of inferences will not hold, for it is wrong to say ‘ the
more numerous the humble-bees, the more prolific is the red clover’. The red
clover of New Zealand became wonderfully prolific after the introduction of about
100 humble-bees, there being no indigenous species (Dunning, Proc. Ent. Soc.,
London, 1886).

690. T. incarnatum L.— According to Kirchner (‘Flora v. Stuttgart,’
pp. 491-2), the flower mechanism of this species essentially agrees with that of
T. pratense. The length of the corolla-tube of the bright blood-red flowers is
8-9 mm., and that of the calyx-tube 5mm. The vexillum is folded up, and there-
fore serves to guide the proboscis of the longer-tongued insects to the nectar. The
base of its limb also completely embraces the claws of the alae and carina, while its
own claw is free. The alae possess well-developed vesicular processes, gripping the
upper side of the sexual column; also a longitudinal fold, united internally with
the epidermis of the carina.

Automatic self-pollination is much less productive than cross-pollination.

VisiTors.—I observed the long-tongued bees Bombus lapidarius Z. ¥, an Eucera
longicornis Z. 9 and 6, in Mecklenburg; both skg. (‘ Bloemenbiol. Bijdragen ’).

H6ppner saw three humble-bees in Bremen.—1. Bombus agrorum /.; 2. B.
muscorum /.; 3. B. variabilis Schmiedekn.

691. T. alpestre L. (Herm. Miller, ‘Weit. Beob.,’ Il, pp. 247-8; Schulz,
«Beitrage,’ p. 209.)— The flower-heads of this species, according to Hermann
Miiller, are larger and more brightly coloured than those of T. pratense, with which
the flower mechanism agrees in most respects. But while in the latter species the
corolla-tube is 7 mm. in length up to the point where carina and vexillum separate,
and 11 mm. long measured to the tip of the carina, in the former species these
dimensions are 11 mm. and 14 mm. respectively. A great many humble-bees are
therefore prevented from sucking nectar legitimately. The carina and alae being
shorter than the vexillum, introduction of the proboscis into the flower is rendered
more difficult for bees and easier for Lepidoptera. Lastly, the carina is considerably
higher than the corolla-tube, and curves sharply upwards, so that the proboscis of
a Lepidopterid cannot reach the base of the flower except through the open cleft,

LEGUMINOSAE 293

first touching the stigma and then getting dusted with pollen. Hence cross-pollina-
tion is regularly effected if another flower has previously been visited. In T. pratense,
on the other hand, the proboscis of a Lepidopterid, when similarly introduced, does
not touch the stigma and pollen. The blossoms of this species are therefore purely
humble-bee flowers, while those of T. alpestre are adaptcd for cross-pollination by
both humble-bees and Lepidoptera.

VisiTors.—Those observed by Hermann Miiller in Thuringia correspond with
the above interpretation. He found the following.—

A. Hymenoptera, <Afzdae: 1. Eucera longicornis Z. 4, skg.; 2. Psithyrus
rupestris /. 9, do. B. Lepidoptera. Rhopalocera: 3. Coenonympha arcania L.,
trying to suck; 4. C. pamphilus Z., do.; 5. Epinephele janira Z., do.; 6. Hesperia
thaumas Hfn., very freq., skg.; 7. Lycaena semiargus Ro//., trying to suck; 8. Melan-
argia galatea Z., in large numbers, skg. or trying to do so; 9. Melitaea athalia Rovt.,
trying to suck; 10. Pieris rapae Z., in large numbers, skg.; 11. Syrichthus malvae
L., trying to suck.

Loew observed the bees—Eucera longicornis L. 9, po-cltg., in Silesia (‘ Beitrage,’
p- 53); and Bombus pomorum Pz., var. elegans Sed/. ¥, in Switzerland (op. cit.,
p- 62).

Schletterer and von Dalla Torre noticed the leaf-cutting bee Megachile nigri-
ventris Schenck, in the Tyrol.

692. T. medium L. (=T. flexuosum /acg.).—Kirchner states (‘Flora v.
Stuttgart,’ p. 492) that the ower mechanism of this species is the same as that
of T. pratense, but the flowers are of a brighter red.

Visrrors.—The following have been observed by Herm. Miiller (H. M.) in
Westphalia, and Buddeberg (Budd.) in Nassau—

A. Hymenoptera. Apidae: 1. Andrena dorsata KX. 9, po-cltg. (H. M.);
2. Bombus agrorum /. 9, skg. normally (H. M.); 3. B. muscorum /”. 9, skg.
(Budd.); 4. B. terrester Z. 9, perforating the flower-tube and stealing nectar (H. M.,
Thuringia); 5. Halictus smeathmanellus X’. 9, trying to suck (Budd.); 6. Psithyrus
barbutellus X. 9, skg. (Budd.); 7. P. campestris Pz. 9, do. (Budd.). B. Diptera.
Syrphidae: 8. Volucella plumata Z., trying to suck (Budd.). C. Lepidoptera.
Rhopalocera: 9. Coenonympha pamphilus Z., skg.(H. M., Thuringia) ; 10. Hesperia
lineola O., do. (H. M.);_ 11. Lycaena semiargus Rofs., do. (H. M.); 12. Melanargia
galatea Z., do. (H. M., Thuringia).

Alfken and Héppner noticed the following bees at Bremen.—

1. Andrena convexiuscula K. 9, po-cltg., 6; 2. Colletes daviesanus K. 9; 3.
Megachile circumcincta K. 9, po-cltg.; 4. Podalirius borealis AZor. 9, do., é.

In Dumfriesshire Apis and 2 humble-bees were recorded (Scott-Elliot, ‘ Flora of

Dumfriesshire,’ p. 45).

693. T. rubens L. (Herm. Miller, ‘Fertilisation,’ p. 187. ‘ Weit. Beob.,’ II,
pp. 248-9.)—According to Hermann Miiller, who studied this species in Thuringia,
the flower mechanism is intermediate between those of T. pratense and T. alpestre.
The purple-red flowers are borne on an elongated axis, all of them being similarly
placed and directed obliquely upwards. The upper part of the flower-tube is curved
strongly outwards, giving a flexure which corresponds to the most convenient
position for the proboscis of long-tongued bees. The corolla-tube is 8-9 mm. long
up to the separation of vexillum and carina, and 13-14 mm. to the tip of the carina.

204 ANGIOSPERMAE—DICOTYLEDONES

The vexillum projects 1-13 mm. beyond the carina. The alae are curved outwards
almost horizontally, providing visiting bees with a surface to grasp while they press
down the carina, as convenient as that of T. pratense, which is formed by the elonga-
tion of the vexillum. On the other hand, the place suited for the introduction of the
proboscis of Lepidoptera is almost as readily visible as in T. alpestre.

Visirors.— Those observed by Herm. Miller in Thuringia are adapted to the
imtermediate character of the flower mechanism. They were the following.—

A. Hymenoptera. Agrdae: 1. Anthophora retusa Z. 9, skg.; 2. Bombus
muscorum F’. 9, skg., ¥, po-cltg.; 3. B. proteus Gerst. ¥, do.; 4. B. sylvarum 9, do. ;
5. B. variabilis Schmzedekn., var. tristis Sed/. ¥, skg.; 6. Psithyrus rupestris 7. 9, do.
B. Lepidoptera. (a) Rhopalocera: 7. Epinephele hyperanthus Z., skg.; 8. Hesperia
sylvanus £'sp., do.; 9. Lycaena corydon Poda., do.; 10. Melanargia galatea Z., skg.
persistently ; 11. Pieris napi Z., skg. (4) Sphingidae: 12. Zygaena filipendulae L.,
skg.; 13. Z. lonicerae Esp., do. C. Coleoptera. Liateridae: 14. Corymbites
holosericeus O/iv., vainly trying to suck.

Schletterer and von Dalla Torre record the humble-bee Bombus alticola A7rzechb.,
common on Alpine roses, and the leaf-cutting bee Megachile nigriventris Schenck
(=M. ursula Gers?.).

694. T. arvense L. (Herm. Miller, ‘Fertilisation, p. 186, ‘Weit. Beob.,’
II, p. 248; Knuth, ‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 222.)—The
small inconspicuous whitish or pink flowers of this species possess a corolla-tube
scarcely 2 mm. long. They are fertile when automatically self-pollinated.

Visirors.—Herm. Miiller chiefly observed bees, less frequently Lepidoptera,
as follows.—

A. Hymenoptera. (a) Apzdae: 1. Andrena carbonaria Z. 6, skg.; 2. A.
denticulata A., do.; 3. A. fuscipes A., do.; 4. A. xanthura A. 9, do.; 5. Apis
mellifica Z. §, do.; 6. Bombus lapidarius Z. ¥, do.; 7. B. rajellus A. 9 and 4, very
freq., skg.; 8. Cilissa leporina Pz. 9, skg.; 9. Colletes marginatus Z. 6, do.; 10.
Diphysis serratulae Pz. 6, do.; 11. Epeolus variegatus Z., do.; 12. Halictus flavipes
F. 9, do.; 13. H. quadricinctus F. 9, do.; 14. H. zonulus Sm. 9, do.; 15. Megachile
argentata /. 6, do.; 16. M. maritima A. 4, do.; 17. Osmia spinolae Schenck &
(Thuringia); 18. Saropoda bimaculata Pz. 4, freq., skg. (4) Sphegidae: 19. Ammo-
phila affnis XK. 9, skg. B. Diptera. Muscidae: 20. Gonia capitata Deg., skg.
(Buddeberg). C. Lepidoptera. Rhopalocera: 21. Coenonympha pamphilus Z.,
skg. (Thuringia) ; 22. Hesperia thaumas H/n., do.; 23. Lycaena aegon S. V., do. ;
24. Polyommatus phlaeas Z., do.

On the island of Riigen, I observed the humble-bee Bombus lapidarius Z. ¥, skg.

Friese mentions for Mecklenburg, on the authority of Brauns and Konow, the
very rare bee Andrena nigriceps K.

Alfken observed the following bees at Bremen.—

1. Bombus agrorum /. §; 2. B. derhamellus A. %; 3. B. lapidarius Z.¥; 4. B.
pomorum Pz. ¥; 5. B. soroénsis /, var. proteus Gers/. ¥; 6. Coelioxys quadridentata
L. 9, skg.; 7. Colletes marginatus Z., freq., 9 skg. and po-cltg., skg.; 8. Megachile
argentata /’. 9 and 4, do.; 9g. Cilissa leporina Pz. $; 10. Podalirius bimaculatus Pz.
g and 4, do. And on the island of Juist—A. Diptera. (a) Aszizdae: 1. Asilus
albiceps Jfg., very freq., skg. (4) Syrphedae: 2. Eristalis tenax Z.; 3. Melithreptus
sp.; 4. Syrphus pyrastri Z., very freq. B. Hymenoptera. Apidae: 5. Bombus
lucorum L. ¥ and 6; 6. B. muscorum /. 4, freq., skg.; 7. B. terrester LZ. 9 ¥ and 6;
8. Colletes marginatus Z., freq., skg.

ens

LEGUMINOSAE 295

MacLeod saw the bee Halictus flavipes /. 9, in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 350).

In Dumfriesshire a humble-bee and 2 hover-flies have been recorded (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 45).

695. T. nigrescens Viv.—

Visirors.—Schletterer noticed the following bees at Pola.—

1, Andrena flavipes Pz.; 2. A. lucens Jmh.; 3. A. parvula K.; 4. Eucera
parvula Friese, very frequent; 5. Halictus interruptus Pz.; 6. H. levigatus K. 9;
7. H. variipes Mor.; 8. Megachile argentata F.; 9. M. muraria Z.; 10. Osmia
gallarum Szv.; 11. O. tridentata Duf ef Pér.; 12. O. versicolor Zr.

696. T. parviflorum Ehrh.—

VisiTors.—Schletterer observed the bee Halictus variipes Mor. at Pola.

697. T. alpinum Z. (Herm. Miiller, ‘Alpenblumen,’ pp. 240-1.)—In this
species the expanded base of the vexillum surrounds the inner parts of the flower for
a length of about ro mm.,
so that among all the
Alpine Hymenoptera only
humble-bees can reach the
nectar.

Visitors. — Herm.
Miiller observed in the
Alps 8 species of humble-
bee, nect-skg. or po-cltg.
{besides Bombus terrester
Z£. as a nectar-thief), and
4 species of Lepidoptera
skg. or attempting to suck.

Loew noticed the

. : : Fic. 94. 7vrifolium alpinum, L. (after Herm. Miller). 4. Flower
following in Switzerland seen from below (x 34). B. The same, after removal of the calyx and
(‘ Beitrage,’ p. 62).— vexillum ; seen from above: d, alar claw. C. Anterior part of the same
flower, after removal of the right ala, and depression of the carina and lett
A. Hymenoptera. ala; seen from the side. d, claw of the carina; ca, calyx; / vexillum;
. : Ji, alae; sch, carina; 4, nectar-passage ; a, anthers ; 4, upper free filament ;
Apidae : 1. Bombus alti- c, fused filaments; s/, stigma; x, point where the right carinal petal is

cola Kriechd. ¥, skg. 5; 2. connected with the right ala.

B. mucidus Gers? ¥, do.;
3. B. rajellus X. 9, do.; 4. Halictus xanthopus A. 9, po-cltg. B. Lepidoptera.
Rhopalocera: 5. Lycaena sp.

698. T. pallescens Schreb. (Herm. Miiller, ‘Alpenblumen,’ pp. 244-6.)—
As the calyx-tube in this species is only 1 mm. long, and the length up to the tip
of the carina only 4-5 mm., the nectar concealed in the base of the flower is
accessible even to short-tongued bees. The flower mechanism agrees with that of
T. repens. Automatic self-pollination is easily effected (see Fig. 95).

Visrrors.— Loew observed a short-tongued bee (Andrena) in the Alps (‘ Beitrage,’
p. 63). Herm. Miiller, besides the honey-bee and 6 species of humble-bee, also
saw 8 species of nect-skg. Lepidoptera.

296 ANGIOSPERMAE—DICOTYLEDONES

699. T. badium Schreb. (Herm. Miiller, ‘Alpenblumen,’ pp. 246-7.)—The
tiny golden-yellow flowers of this species are scarcely 8 mm. long, and the distance
from the tip of the carina to the nectar is hardly 4 mm., so that it is accessible
to quite short-tongued bees. Lepidoptera are also easily able to effect cross-
pollination, for the stigma is at about the same level as the anthers, surrounded
by them, and situated quite at the top of the broad open cleft of the carina (see
Fig. 96). In the absence of insect-visits automatic self-pollination readily occurs.

Visrtors.—Herm. Miiller observed 4 humble-bees and 11 Lepidoptera.

yoo. T, agrarium L. (Herm. Miiller, ‘ Weit. Beob.,’ II, p. 250; MacLeod,

Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 380.)}—Here again automatic self-pollination
is effective.

Fic. 95. Trifolium pallescens, Schreb. (after Herm. Miller). 4, Flower seen from below. &. The
same, after removal of the calyx and vexillum ; seen from above. C. The same, seen from the side.
D. Flower after removal of the vexillum, with depressed alae and carina. &. Flower after removal of
the calyx, vexillum, and right ala; seen from the right side. References as in Fig. 94.

Visitors.—Hermann Miiller gives the following.—

A. Hymenoptera. Apzdae: 1. Apis mellifica Z. ¥, skg. B. Lepidoptera.
Rhopalocera: 2. Epinephele hyperanthus Z., skg. (Bavarian Oberpfalz); 3. Hesperia
lineola O., do.; 4. Lycaena aegon S.V. 6, do.

MacLeod observed the bee Halictus flavipes /. 9 in Flanders.

qo. T. campestre Schreb. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 60-1,
153-)—This is a variety of T. procumbens Z. with larger dark-yellow flowers, which
afterwards become brown. In the bud, the large vexillum closely, and almost
completely, surrounds the other parts of the flower. On anthesis, the vexillum,
which is stiffened by a number of longitudinal veins, serves as a roof by which
the alae and the tiny carina, as well as the stamens and pistil, are sheltered. The
limbs of the alae are fused with the carina, so that during insect-visits both are
depressed together, or laterally displaced, while stamens and pistil project. As
the stigma protrudes somewhat further than the anthers, a visitor must first touch
the former and then the latter with its ventral surface. When a second flower

LEGUMINOSAE 297

is visited, cross-pollination is consequently effected. As anthesis progresses, the
brownish tint which replaces the original golden-yellow colour becomes more pro-
nounced. At the same time, the vexillum with its 12-16 wavy ribs, becomes applied
to the other parts of the flower, and closes the entrance to them.

VistTors.—I observed the honey-bee, and Bombus pratorum ZL. ¥ at Kiel; in
Sylt only the former.

702. T. procumbens L. (main variety). (Herm. Miiller, ‘ Fertilisation,’ p. 187,
‘Weit. Beob.,’ II, p. 250.)—

VisiTors.— Herm. Miiller (H.M.) and Buddeberg (Budd.) record the following
insects, all skg.—

A. Hymenoptera. Apidae: 1. Andrena schrankella Ny/. 9 (Budd.); 2. Apis
mellifica Z. ¥ (H.M.); 3. Halictus flavipes F. 9 (H.M.); 4. H. nitidiusculus A. 9
(Budd.). B. Diptera. Muscidae: 5.
Ocyptera brassicaria #. (Budd.). C.
Lepidoptera. Xhopalocera: 6. Epi-
nephele janira Z. (H.M., Thuringia);
7. Lycaena icarus Koff. (H. M.).

MacLeod observed Apis and a Muscid
in Flanders (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 350).

In Dumfriesshire 3 hover-flies were
recorded (Scott-Elliot, ‘Flora of Dum-
friesshire,’ p. 46).

703. T. minus Relhan. (Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 153.)—

Visitors.—In Sylt and Kiel I ob-
served Apis mellifica Z. 9, skg. Herm.
Miiller saw, in addition, Halictus albipes
F’, 9, skg., and H. cylindricus /. 9, po-cltg.

Alfken noticed the following bees at
Bremen.—

1. Andrena parvula 4.9; 2. Nomada
succincta Pz. ‘ c FiG. 96. Zri/olium badium, Schreb. (after Herm.

MacLeod observed 3 Lepidoptera in Mailer). 4. Flower seen from below (x 7). 8. The
the Pyrenees (: Pyreneénbl., p. 437). same, after removal of calyx and vexillum; seen from

above. C. Stigma (s¢) and stamens (a) in their natural
Position (x 35). References as in Fig. 94, also: a’,

704. T. subterraneum L. (Warm- upper anther ; 7, upper filament; ov, ovary; za, alar
ing, Bot. Jahrb., Leipzig, iv, 1883, p.502;
Glaab, D. bot. Monatsschr., Arnstadt, ix, 1890, pp. 20-2 ; Ross, Malpighia, Genova,
v, 1891.)—According to E. Warming, the inflorescence bears but few (usually 3-4)
normal flowers which set fruits, and are capable of self-fertilization, though perhaps
they do not always effect this. The inflorescence turns downwards, and penetrates
into the ground. In order to protect it from injury, the upper vestigial flowers
become converted during the period of fruit-forming into peculiar hook-like organs
which fix it in the ground, and protect the maturing fruits. The normal flower is
almost sessile, while the modified ones possess particularly strong stalks, 2-4 mm. in

298 ANGIOSPERMAE—DICOTYLEDONES

length. The lowest of these altered flowers still retain the five calyx-teeth, but all the
other parts have become aborted. The higher the flowers on the inflorescence, the
feebler are the calyx-teeth; the uppermost flowers are only represented by thick
skittle-shaped stalks, somewhat curved, and without a trace of leaves.

The observations of L. Glaab confirm those of Warming, and were made inde-
pendently. In its first stage the capitulum possesses 3-5 fully formed corollas.
While these are fading, the axis of the inflorescence turns downwards, and elongates
till at last it reaches the ground, into which it presses the fruits till they sometimes
appear half-buried in the soil. Those heads which are more or less surrounded by
earth produce the greatest number of fruits and vestigial flowers, as well as the most
numerous (3-4) and largest seeds; while those which encounter obstacles, e.g.
stones, are relatively retarded in their development.

705. T. pannonicum L.—
VisiTors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Andrena dorsata £. 9, po-cltg.; 2. Anthophora parietina F. é skg., 9 skg.
and po-cltg.; 3. Bombus hortorum JZ. ¥, skg.; 4. Megachile centuncularis /. §,
skg. and po-cltg.

208. Anthyllis L.

Yellow nectar-producing bee flowers with a pumping arrangement, extruding
threads of pollen. The tips of all ten filaments are clavate.

706. A. Vulneraria L. (Delpino, ‘ Ult. oss.’ p. 45; Herm. Miiller, ‘ Fertili-
sation,’ pp. 172-3, ‘Alpenblumen,’ pp. 248-9 ; Knuth, ‘BI. u. Insekt. a. d. nordfr.
Ins., pp. 58, 152; Kerner, ‘ Nat. Hist. Pl.’ Eng. Ed. 1, II; Frey, ‘ Lepidopteren der
Schweiz,’ pp. 16, 20; Schulz, ‘Beitrage,’ II, p. 208.)—The flower mechanism of
this species was first described by Delpino, and afterwards more fully by Hermann
Miiller, who gives the following account of it—The greatly elongated claws of the
petals are enclosed by a calyx 9-10 mm. long, and somewhat dilated in the middle.
From this the vexillum, which is expanded and flat at the end, projects as much as
6-7 mm. It extends somewhat beyond the alae, embracing these above with the
grooved base of its limb, while at the same time it grasps and almost completely
enfolds them from below by means of two rounded lobes, one on either side of its
base. The alae ensheath the carina, with which they are so closely united that it
is depressed along with them during insect-visits. The union of these parts is
effected in three ways: (1) A narrow, deep groove on the upper side of each ala,
near its base, fits into a fold of the carina below it. (2) A pointed triangular tooth,
projecting from the outside of this carinal fold fits into the space behind the alar
groove. (3) Two internal folds, one from the upper margin of each ala, fairly far
forwards, hold these petals firmly together above the carina. The tip of the carina
projects immediately in front of these folds, and opens by a slit when the alae are
depressed. Behind this slit the upper edges of the carinal petals are fused. The
pressure of a bee visiting the flower squeezes out a string of pollen from the slit,
pushed from behind by the thickened ends of the ten stamens. The pollen is
stored up in the tip of the carina while the flower is still unopened. When the
pressure on the alae and carina is removed, these return to their former position,
and when the pressure is renewed fresh pollen masses are extruded. At a later

LEGUMINOSAE 299

‘stage the stigma projects through the cleft. To begin with, it is surrounded by the
pollen-grains of its own flower, but these do not adhere to its papillae, which are not
yet viscid. Not till the pollen has been carried away on the ventral surfaces of insect
‘visitors, are the delicate epidermal cells of the stigma partly rubbed away, making
it receptive for pollen from other flowers.

VistTors.—These are chiefly humble-bees. Herm. Miiller observed the follow-
ing insects.—

A. Hymenoptera. Apidae: 1. Bombus agrorum F. 9, skg.; 2. B. hortorum

L. 9, do.; 3. B. sylvarum Z. ¥, do. B. Lepidoptera. Rhopalocera: 4. Lycaena
minima /vessi. 9, do. C. Hemiptera. 5. Capsus sp., attempting to suck.

FIG. 97. Anthyllis Vulneraria, L. (after Herm. Miller). (1) Flower seen from below. (2) Flower
-after removal of the calyx, seen from the side. (3) Flower after removal of the calyx and the vexillum,
‘seen from above. (4) Anterior half of the carina, seen obliquely from above and from the left. (5)
Anterior half of the carina and an ala, seen from the left side. (6) Left ala (except the base), seen from
within. (7) Vexillum, seen from below ( 3}). (8) Tip of the carina after removal of its left half, with
the stamens and style, seen from the left side (x 7). a, calyx ; 6, lower side of the vexillum; 5’, groove in
the same; ¢, outside of the vexillum ; d, lobes of the vexillum grasping alae and carina ; e, inner side of the
alae; /, outer side of the alae; g, deep and narrow fold in the upper external surface of the vexillam,
projecting inward asa sharp ridge, This latter fits into a deep fold (2) in the upper side of the carina (4),
and is fixed still more firmly by an acute process (/) of the carina, which fits into the space (#/) behind the
sharp ridge (4); 2, opening in the carina through which the pollen passes out; 22’, anterior fold in
the upper margin of the ala; #, column of reproductive organs; 0, nectar-passages ; #, thickened ends of
the filaments; z, empty anthers; 7, pollen; s, stigma; ¢, claws of the alae.

Kerner saw a butterfly, Lycaena hylas Zsg.: the female lays her eggs in the
ovary. Frey states that the caterpillars of this butterfly live only on Thymus
Serpyllum and Coronilla varia, while the caterpillars of Lycaena minima Fess/.
and L. semiargus Ao//. live on Anthyllis. This also agrees with Herm. Miiller’s
-observations.

Herm. Miiller observed 10 Apidae, 10 Lepidoptera, and 2 beetles in the Alps.

Alfken noticed 3 humble-bees in Juist—

1. Bombus hortorum Z. ¥, very freq.; 2. B. lapidarius Z. ¥, do.; 3. B. mus-
-corum £” ¥, do.

MacLeod saw 4 humble-bees, an Anthophora, 5 Lepidoptera, and a Dipterid in
the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 435). Loew observed two
bees in Switzerland (‘ Beitrage,’ p. 61), i.e. Bombus pomorum Pz., var. elegans Sezdl. ¥,
po-cltg., and Eucera longicornis Z. 9, do.; also Bombus agrorum /”. ¥, skg., in the

Berlin Botanic Garden.

300 «INGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire a humble-bee was recorded (Scott-Elliot, ‘Flora of Dumfries-
shire,’ p. 47).

Schulz noticed the perforation of flowers by humble-bees.

For the variety maritima Schwezy., the flower mechanism of which agrees with
the type-form, I observed humble-bees (Bombus agrorum F.) in Schleswig-Holstein,
and, in the island of Féhr, 2 Lepidoptera as useless guests, nect-skg., i.e. Epinephele
janira Z. and Zygaena filipendulae Z.

Schletterer saw the Scoliid Scolia flavifrons /., var. haemorrhoidalis /., on the
variety A. Dillenii Schu/¢., at Pola.

707. A. montana L.—According to Briquet (‘Etudes d. biol. flor. dans les
Alpes occident’) the bright rose-red flowers of this species possess a pumping
arrangement essentially agreeing with that of A. vulneraria.

VisiTors.—These are honey-bees, humble-bees, and also Lepidoptera, which

effect cross-pollination after removing the pollen. Automatic self-pollination is
improbable (Kirchner).

209. Lotus Tourn.

Like the last genus. Only the five outer filaments are club-shaped at the tip.

708. L. corniculatus L. (Delpino, ‘Sugli appar. d. fecondaz. nelle piante
autocarp.,’ p. 25; Herm. Miller, ‘Fertilisation,’ pp. 167-71, ‘Weit. Beob.,’ II,
pp. 245-6, ‘Alpenblumen,’ pp. 238-40; MacLeod, Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, pp. 350-3; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 61-2, 153, ‘ Weit.
Beob. ti. Bl. u. Insekt. a. d. nordfr. Ins., p. 233, ‘Bl. u. Insekt. a. d. Halligen,’
‘ Bliitenbiol. Beob. a. d. Ins. Riigen’; Loew, ‘ Bliitenbiol. Floristik,’ pp. 391, 395, 399 ;
Schulz, ‘Beitrage,’ II, p. 209; Verhoeff, ‘Bl. u. Insekt. a. d. Ins. Norderney’;
Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.) — The flower mechanism of this
species was first indicated by Delpino, and subsequently fully described in a masterly
way by Hermann Miiller. The following is an abstract of his account.—The
vexillum of the golden-yellow flowers, of which there are five in a head, is per-
pendicular, and frequently possesses red streaks as nectar-guides. The nectar,
secreted in the usual place, is sought out by numerous insects. Only Hymenoptera
are effective pollinators; various Lepidoptera and Diptera are unbidden guests.
Each ala possesses a deep depression near the base of its limb, which fits into a
corresponding pit in the upper side of the carina. Immediately behind this point,
the upper edges of the alae are fused together, so that when a suitable insect visits
the flower both alae and carina must simultaneously be depressed. Even in the
bud, before the petals are fully developed, the ten anthers dehisce, discharging their
pollen into the tip of the carina, after which they shrivel. As the flower grows,
only the filaments of the five outer stamens elongate, while at the same time their
ends thicken, and completely cut off the conical tip of the carina from its lower part.
A pollen-chamber is thus constituted; it conceals the stigma, and there is a slit
along its upper margin. As a result of pressure exerted by a nectar-seeking insect,
the five thickened ends of the filaments penetrate further into the tip of the carina,
and a corresponding quantity of pollen passes out piecemeal through the carinal
opening. As the downward pressure increases, the stigma also protrudes, so that

be rubbed before becoming receptive.

j LEGUMINOSAE 301

either cross- or self-pollination may take place. The latter, however, is ineffective.
When the pressure is removed, the parts of the flower return to their original
positions, Automatic self-pollination, by the pollen surrounding the stigma of an
unvisited flower, does not take place, for it appears that the stigmatic papillae must
Kerner states, however, that the flowers are
fertile when insects are excluded. Warnstorf describes the pollen-grains as very
small, shining, white, and smooth, resembling blunt-ended prisms in shape, with

a slight constriction in the middle. On an average they are 25 » long and 12 broad.

Fic. 98. Lotus corniculatus, L. (after Herm. Miller). (1) Flower seen from the front. (2) Flower
seen obliquely from the side. (3) Flower after removal of the vexillum, seen from the side. (4) The
same, seen from above. (5) Flower after removal of the vexillum and alae, seen from the side, and
more highly magnified. (6) Flower after removal of the right half of the carina, seen from the right
side. (7) Flower after removal of the vexilluin and alae, seen from above. (8) The stamens, style, and
stigma, enclosed in the anterior part of the carina, more highly magnified than in 6. (9) Stamens, style,
and stigma of a bud, immediately after the pollen has been shed; removed from the flower and seen from
the side. The outer filaments have become thicker than they were in 8. (10) The same, seen
from above; the outer filaments, thickened at the end, spread out, being freed from the pressure of the
carina. (11) The nine united filaments of a fully developed flower spread out. a, nectar-passage; 4, up-
ward bend of the free filament; ¢, depressions in the two carinal petals, into which projections (c’) of the
alae fit; d, the five inner filaments that remain short; ¢, the five outer filaments, which elongate and
become club-shaped ; 7 stigma ; eg, hollow cone of the carina filled with pollen; ¢ carinal opening from
which pollen is pressed out ; xy indicates corresponding parts of 6 and 8.

Visirors.— Hermann Miiller (H.M.) in Westphalia, Buddeburg (Budd.) in

Nassau, and myself (Kn.) in Schleswig-Holstein (S.-H.), Riigen (R.) and Thuringia

(T.), observed the following.—
A. Hymenoptera. Apidae: (2) Dasygastres: 1. Anthidium manicatum ZL. 9

(H.M.); 2. A. oblongatum Zir., freq., 9, skg. and po-cltg., (H. M., Budd.); 3. A.

punctatum Zr. 9 and 6, do. (H. M., Budd.); 4. A. strigatum Zé. 9 and 4, do.

302 ANGIOSPERMAE—DICOTYLEDONES

(Budd., Kn., T.); 5. Chelostoma nigricorne Wy/. 6, skg. (Budd.); 6. Diphysis ser-
ratulae Pz. 9 and 6, skg. and po-cltg. (H. M., Budd.); 7. Megachile argentata /’. 9
and é, skg. (H. M., Budd.); 8. M. circumcincta X. 9 and 4, freq. (H. M., Budd., Kn.,
S.-H. and R.); 9. M. fasciata Sm. 9 and 4, skg. and po-cltg. (Budd., H.M., T.);
ro. M. analis Wy/., var. obscura AJf&. (Kn., Langeness); 11. M. pyrina Le. 9 and 4,
freq. (H.M.); 12. M. willughbiella . 9 and 4, skg. and po-clig. (H. M., Budd.) ;
13. Osmia adunca Pz. 9 and 4, do. (Budd.); 14. O. aenea Z. 9 and 3, freq., skg. and
po-cltg.; 15. O. aurulenta Pz. 9, very freq. (H. M., Budd.) ; 16. O. claviventris Ths.
(H. M.); 17. O. fuciformis L/r. 9, skg. (H. M., T.); 18. O. pilicornis Sm. 9 (Budd.) ;
19. O. rufa LZ. 9 and 4, skg. (H.M., Budd.). (4) Scopulipedes: 20. Andrena con-
vexiuscula X. 9, skg. and po-cltg. (H. M.); 21. A. labialis K. 9, skg. (H.M.); 22. A.
xanthura A. 9, po-cltg. (H. M.); 23. Anthophora quadrimaculata Pz. 9 (Kn., S.-H.);
24. Apis mellifica Z. ¥, very freq., skg., more rarely po-cltg. (H. M.; Kn., S.-H., and
R.); 25. Bombus agrorum F. é 9 and ¥, skg., more rarely po-cltg. (H. M., Budd.,
Kn.); 26. B. derhamellus XK. ¥ (Kn.); 27. B. lapidarius Z. ¥, skg. (H. M., T.; Kn,
S.-H. and T.); 28. B. pratorum Z. ¥ (Budd.); 29. B. muscorum /. ¥, skg. (H. M..
T.; Budd.), 9 (Kn.); 30. B. sylvarum Z. 9, skg. (H.M., T.); 31. B. terrester L. ¥,
skg., more rarely po-cltg. (H.M.; Kn., S.-H. and T.); 32. Cilissa haemorrhoidaus
F.%, skg. (Budd.); 33. C. leporina Pz. 9, do. (Budd.); 34. Eucera longicornis L. ¢
and 4, do. (H. M., Budd.); 35. Halictus flavipes /”. 9, do. (H.M.); 36. H. leucopus
K. 9 (Budd.); 37. H. leucozonius Schr. 9 (Budd.); 38. H. levigatus A. 9 (Budd.) :
39. H. rubicundus Cfr., skg. and po-cltg. (H.M.); 40. H. sexnotatus K. 9 (Budd.);
41. H. smeathmanellus K. 9 (Budd.); 42. Rhophites canus Eversm. 9 and é (H. M.,
T.). (c) Cuckoo-bees : 43. Coelioxys elongata Le. 9, skg. (H.M., T.); 44.C. sp. 8, do.
(H.M.); 45. Nomada ruficornis Z. 9, do.(H.M.). B. Coleoptera. (a) Elaterzdae:
46. Agriotes sputator Z., vainly seeking for nectar (H.M., T.). (6) Mordellidae :
47. Mordella fasciata #., vainly seeking for nectar (H.M., T.). C. Diptera. (a)
Conopidae: 48. Conops flavipes Z., skg., inserting its proboscis under the vexillum
(H.M.); 49. Myopa testacea Z., skg. (Budd.). (4) Syrphzdae: 50. Melanostoma
mellina Z., po-dvg. (H.M.). D. Lepidoptera. (a) Bombycidae: 51. Porthesia
similis /uessl., vainly skg. (H.M.). (6) Moctuzdae: 52. Euclidia glyphica Z., skg.
(H. M.); 53. Plusia gamma Z. (H.M.). (c) Rhopalocera: 54. Coenonympha arcania
L., skg. (H.M.); 55. C. pamphilus Z., do. (Kn.); 56. Epinephele janira Z., do.
(KXn.); 547. Syrichthus malvae Z., do. (H. M.); 58. Nisoniades tages Z., do. (H.M.);
59. Lycaena aegon S. V., do. (H.M., T.); 60. L. damon S. V. (H.M., T.); 61. L.
icarus Ro/ft., skg. (H. M.); 62. L. semiargus off. (Kn.); 63. L. sp.(Kn.); 64. Thecla
spini S. V., skg. (H.M.). (d) Sphengidae: 65. Sesia empiformis Zsp., skg. (H. M.,
T.); 66. Zygaena filipendulae Z., do. (H. M., T.; Kn.); 67. Z. sp., do. (Kn., R.).

Herm. Miiller saw 17 Apidae, 25 Lepidoptera, and a hover-fly in the Alps.

For the Tyrol von Dalla Torre gives the humble-bees Bombus mastrucatus
Gerst. 6, and B. pratorum Z. ¥ and & Schletterer saw both these, and also the
following bees, at Pola. —

1. Andrena albopunctata Ross¢; 2. A. convexiuscula K.; 3. A. cyanescens /Vy/.;
4. A. deceptoria Schenck; 5. A. flavipes Pz.; 6. A. parvula A.; 7. Halictus levigatus
K.9; 8. Osmia andrenoides Sp:v.; 9. O. aurulenta Pz.; ro. O. latreillei Spz.; 11.0.
ligurica AZor.

Ducke noticed the following bees at Trieste.—

1. Meliturga clavicornis Z/r.; 2. Osmia aurulenta Pz.; 3. O. tiflensis Aor. 9.
occasional; 4. O. versicolor Zér. 9 and é, very freq.; 5. The southern var. of the
parasitic species Psithyrus barbutellus A. (=P. maxillosus K/ug) 9.

Alfken saw the following bees at Bad Ratzes in the Tyrol.

1. Anthidium strigatum Z/r. 9 and 4, freq.; 2. Megachile willughbiella K. 9.

freq., skg.; 3. Trachusa serratulae Pz., very freq.; to which Kohl adds a mason-bee,
4. Osmia claviventris Ths.

LEGUMINOSAE 303

Alfken observed the following bees at Bremen.—

1. Anthidium strigatum Pz. 9, skg. and po-clig., 4 skg.; 2. Andrena con-
vexiuscula K. 9, skg. and po-cltg.; 3. A. labialis A. 9; 4. Bombus agrorum /. 9
and ¥; 5. B. arenicola Zhs. ¥; 6. B. derhamellus X.9 and ¥; 7. B. distinguendus
Mor. 4; 8. B.hortorum Z. ¥; 9. B. lapidarius Z. ¥; 10. B. muscorum /. 9 and §;
11. B, sylvarum Z. ¥; 12. B. variabilis Schmiedekn. $; 13. Coelioxys quadridentata 9,
skg.; 14. C, mandibularis Wy/. 9, do.; 15. C. rufescens Lep. 9, do.; 16. Eucera
longicornis Z.9; 17. Halictus calceatus Scop.; 18. H. rubicundus Chr. 9; 19. H.
tumulorum L. 9; 20. Megachile analis Vy/. 9 and ; 21. M. centuncularis Z. 9 and 3;
22. M. circumcincta X. 9 and $; 23. M. maritima &.9 and é; 24. M. willughbiella
K.9 and $; 25. Melitta leporina Pz. 6; 26. Nomada jacobaeae Pz. 9, skg.; 27. N.
ochrostoma X. 9; 28. Osmia claviventris Zhs. 9; 29. Podalirius vulpinus Pz. 9;
30. Trachusa serratulae Pz. 9 and 6.

Alfken (A.) and Leege (L.) record the following for Juist.—

A. Hymenoptera. <Afzdae: 1. Bombus hortorum JZ. ¥, po-cltg. and skg.;
2. B. lapidarius Z. (L.); 3. B. muscorum /. (L.); 4. Megachile circumcincta X. 9,
rare, po-cltg. and skg. (A., L.); 5. Osmia maritima /7zese 9, very freq., po-cltg. and
skg. (A., L.); 6. Psithyrus rupestris Z. (L.). B. Lepidoptera. Sphingidae:
7. Deilephila galii Ro#z., very freq. (L.); 8. D. porcellus, Z., do. (L.).

Verhoeff gives the following for Norderney and Juist (J.).—

A. Coleoptera. Staphylinidae: 1. Anthobium torquatum J/arsh., abnormal (J.).
B. Hymenoptera. <Afzdae: 2. Bombus cognatus Steph. (=B. muscorum Ff.) 9,
skg. (J.); 3. B. hortorum Z. 9, do. (J.); 4. B. lapidarius Z. 9, freq., skg. and po-cltg.;
5. B. terrester Z. Y, skg., sometimes abnormally (J.); 6. Halictus minutus K. 9,
abnormal; 7. Megachile circumcincta K. 9, skg.; 8. Osmia maritima Frese 9 and
8, do. 9 skg. (J.). C. Lepidoptera. (a) Nymphalidae: 9. Vanessa cardui L.,
abnormal. (6) Lycaenidae: 10. Lycaena icarus o//., abnormal (also J.). (¢) Prerz-
dae: 11. Pieris brassicae Z., abnormal (also J.).

Friese observed the following bees in Alsace (A.), Baden (B.), at Innsbruck (I.),
in Mecklenburg (M.), Switzerland (S.), Thuringia (T.), and the Tyrol (Ty.).—

1. Anthidium montanum J/or., infrequent (S.); 2. A. oblongatum Z/r., not un-
common (T.); 3. A. punctatum Zyr. (A.,M.; rarein S.,T., and Ty.); 4. A. strigatum
Pz.; 5. Coelioxys elongata Zep. (T.); 6. C. quadridentata Z.; 7. Eucera difficilis
Duf., occasional (B.); 8. E. interrupta Baer, do. (B.); 9. E. longicornis Z., freq.
(B.); 10. Megachile apicalis Sgz7., occasional (M.); 11. M. argentata /., freq. (M.);
12. M. centuncularis Z., do. (M.); 13. M.circumcincta &., do. (A., B., M., T.); 14. M.
ericetorum Zép., occasional (A.); 15. M. muraria Reéz., freq. (A.); 16. M. pyrenaica
Lep.; 17. Osmia aurulenta Pz., freq. (M.; rare in A. and B.); 18. O. bicolor Schr.
9, skg. (A.); 19. O. claviventris 7ds. (B.; rare in M.; occasional, A., T., S.); 20.
O. nigriventris Ze//., infrequent (S.); 21. lepeletieri Per. 9 and 6 (I., S.); 22. O. leu-
comelaena X., occasional (M.) ; 23. O. maritima F’zese, Freq. (M.); 24. O. mora-
witzi Gerst.; 25. O. vulpecula Gers. (S.); 26. Podalirius bimaculatus Pz. not
infrequent (M.); 27. Trachusa serratulae Pz., occasional (A., B., M., S., T.).

Rossler saw the moth Butalis aeneospersella Ros/r. at Wiesbaden, and Krieger
the following bees at Leipzig.—

1. Anthidium strigatum Pz.; 2. Bombus derhamellus A. ¥; 3. Megachile
centuncularis Z.; 4. Osmia rufa Z.; 5. Podalirius vulpinus /z.; 6. Trachusa
serratulae Pz.

Schenck records the following Hymenopterids.—

(2) Apidae: 1. Anthidium oblongatum Z/r.; 2. A. punctatum Z/r.,; 3. Andrena
labiata Schenck; 4. Megachile argentata /.; 5. M. maritima X.; 6. Podalirius
bimaculatus Pz. (0) Vespzdae: 7. Odynerus xanthomelas H.-Sch.

304 ANGIOSPERMAE—DICOTYLEDONES

Schmiedekneckt gives the bees Osmia aurulenta Pz. for Thuringia, and O.
difformis Pér. for the Pyrenees; while Gerstacker noticed two bees at Berlin—
Coelioxys quadridenta Z. and Osmia tridentata Duf. ef Perr (one 9).

Loew observed the hover-fly Eristalis tenax Z., po-dvg., in Silesia, and the
following bees in Brunswick (B.) (‘Beitrage,’ p. 53), Mecklenburg (M.), (op. cit.,
p- 44), Switzerland (Sw.) and the Tyrol (T.) (op. cit., p. 61).—

1. Diphysis serratulae Pz. 9, po-cltg. (B.); 2. Cilissa tricincta K. 9, do. (M.) ;
3. Colletes fodiens K. 9, do. (M.); 4. Megachile argentata /. 9, do. (M.); 5. M.
willughbiella K. 8, skg. (M.); 6. Chalicodoma muraria Re/z. 9, po-cltg. (T.); 7.
Eucera longicornis LZ. 9, do. (Sw.); 8. Megachile analis My/. 9, do. (Sw.); 9. Osmia
angustula Zeft. (T.).

MacLeod saw Apis, 5 humble-bees, Diphysis, and 5 Lepidoptera in Flanders
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 352-3); and 11 long-tongued bees,
7 Lepidoptera, and a fly in the Pyrenees (op. cit., iii, 1891, pp. 437-8).

In Dumfriesshire, Apis, 2 humble-bees, a short-tongued bee, a hover-fly, and
a beetle were recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 47).

In England, Saunders observed the leaf-cutting bee Megachile versicolor Sw.
and Smith the wall-bee Osmia aurulenta Pz.

Schulz in Central Germany noticed flowers perforated by humble-bees.

Only bees are able to liberate the flower mechanism ; other visitors are unbidden
guests.

70g. L. uliginosus Schkuhr (= L. major Sm.). (MacLeod, Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 353; Kirchner, ‘Flora v. Stuttgart,’ p. 494; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 62, 153, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d.
nordfr. Ins.,’ p. 233, ‘Bltitenbiol. Beob. a. d. Ins. Riigen’; Schulz, ‘Beitrage,’ II,
p- 209; Warnstorf, Verh. bot. Ver., Berlin, xxxv, (1893) 1894.) — The flower
mechanism of this species is exactly like that of L. corniculatus, but the carina is
longer and narrower, and is not exactly vertical—but directed obliquely upwards.
Perhaps, therefore, rather less pressure suffices to work the pumping apparatus.
Warnstorf states that the pollen-grains are only 18-19 » long and 12 » broad,
but otherwise like those of L. corniculatus.

VisiTorS.—In the island of Féhr, I only observed the honey-bee; in Riigen
I saw Bombus rajellus K. ¥, skg., and also a moth (Zygaena filipendulae Z.) as
an unbidden guest. Schulz noticed flowers perforated by humble-bees in Central
Germany.

In Thuringia I only observed a moth, Zygaena trifolii Hsp. (an unbidden guest)
(‘ Bliitenbiol. Beob. in Thiiringen,’ p. 42).

H. de Vries saw a humble-bee (Bombus subterraneus Z. ¥) in the Netherlands
(Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875). MacLeod observed Apis,
2 humble-bees, a hover-fly, and 2 Lepidoptera in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 353)-

Willis noticed a humble-bee, Bombus agrorum ¥#., skg., in the neighbourhood
of the south coast of Scotland (‘Fls. and Insects in Gt. Britain,’ Part I).

LEGUMINOSAE 305

210. Tetragonolobus Scop.
Agrees with Lotus.

710. T. siliquosus Roth.—Hermann Miller (‘ Alpenblumen,’ p. 238) describes
the mechanism of the large yellow humble-bee flowers of this species. A proboscis
12-14 mm. in length is required to secure all the nectar. Kirchner (‘Beitrage,’
p. 42) adds that the somewhat S-shaped style thickens towards its end, becoming
thinner again for the last millimetre. The stigma is situated here, on the side
directed outwards and upwards, and indented.

VisiTors.—These are undoubtedly humble-bees, but the nectar is concealed
so deeply that it can only be reached by long-tongued species. A. Schulz says
that, in Central Germany, shorter-tongued species steal the nectar by perforating
the flowers. Loew observed Bombus lapidarius Z. ¥, skg. in the Berlin Botanic
Garden. MacLeod noticed 2 humble-bees and an Osmia in the Pyrenees (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 437).

2u. Amorpha L.

Protogynous nectar-yielding flowers, devoid of carina and alae.

qu. A. fruticosa L.—As emphasized by Delpino (‘Ult. oss.,’ pp. 64-8), and
subsequently by Hermann Miller (‘ Weit. Beob.,’ II, pp. 244-5), the flowers of this
species, which has been introduced into Europe from North America, have neither
alae nor carina, so that the vexillum alone enfolds the stamens and pistil in the
bud. When first the flower opens, only the style with its stigma already receptive
projects beneath the vexillum, while the still immature anthers are concealed beneath
it. The stamens soon elongate, however, so that they often project beyond the
stigma. If the latter has not already been pollinated it remains receptive till the
anthers dehisce, so that self-pollination takes place should insect-visits fail. If,
however, these take place, cross-pollination is secured as a result of protogyny.

Visirors.—Herm. Miiller observed the honey-bee, very freq., skg. and po-cltg.
As the individual flowers do not possess alighting- or supporting-surfaces (alae and
carina), bees use the whole inflorescence as such,

712. A. canescens Nutt.—The flower mechanism is like that of A. fruticosa.

212. Galega Toum.

Nectarless bee flowers, lilac or white in colour; with a simple valvular
arrangement.

713. G. officinalis L. (Kirchner, ‘ Beitrage,’ p. 42.)—The flowers of this species
are arranged in erect racemes of considerable size. The calyx-tube is 24 mm.,
and the lamina of the vexillum 9 mm. in length. At the base of the latter there
is a bright median streak. The carina projects from the calyx as much as the
vexillum, but the alae are a little shorter. Each alar lamina possesses a posterior
process which is directed obliquely upwards and overlaps the staminal tube. In
front of this process there is a deep projection which fits into a corresponding fold
of the carina, and brings about a firm union. The stigma and anthers project

freely from the carina when the alae are depressed, resuming their position when
DAVIS. I 35

306 ANGIOSPERMAE—DICOTYLEDONES

the pressure is removed. The upper filament is free distally, but its proximal half
is fused with the other nine, so as to leave no access to the base of the inner
surface of the staminal tube. This is in correspondence with the absence of nectar.
The anthers shed their reddish-yellow pollen before the flower opens.

213. Colutea L.

Nectar-yielding bee flowers, mostly yellow in colour ; with a brush arrangement.

714. C. arborescens L. (Kirchner, ‘ Beitrage,’ pp. 42-3; Loew, ‘ Bliitenbiol.
Floristik,’ p. 395; Knuth, ‘Weit. Beob. ui. BI. u. Insekt. a. d. nordfr. Ins.,’ p. 233,
‘Bloemenbiol. Bijdragen.’)—In this species, according to Kirchner, the thick-walled
calyx and the strong elastic claws keep the petals in place, and cause them to return
to their original position when the pressure exerted by an insect visitor is removed.
The vexillum is erect, and possesses a faintly marked nectar-guide. At the base
of its lamina there are two swellings, closely apposed to the alae. The latter are
small, and not united with the carina. Each possesses a finger-shaped process,
carried obliquely downwards and grasping the sexual column behind. The large
strong carina possesses two posteriorly directed triangular lappets which serve the
same purpose: its front part is thickened, so that the upper edges of the carinal petals
are kept close together.

The style projects about 3 mm. beyond the anthers, and is rolled at the end
in such a way that its apex points downwards. On its inner side it bears a brush
about 5 mm. long directed obliquely upwards. The end of the style is obliquely
truncated, presenting a surface from the middle of which the small papilliform stigma
projects. This is surrounded by hairs which prevent self-pollination from
taking place.

Considerable pressure is required to depress the carina. When this is effected,
the style with its adherent pollen first protrudes, and then the anthers. These
dehisce shortly before the flower opens, and part of the pollen which covers them
gets transferred to the hairs on the style.

Kirchner observed a great many honey-bees visiting this species. Some of
them sucked legitimately, having settled on the middle of the flowers and succeeded
in depressing the alae and strong carina, so that the stigma and anthers protruded.
In doing so, they frequently effected self-pollination, though many of them dusted
the stigma with pollen brought on their legs from other flowers. Most of the
visitors, however, preferred to thrust in their proboscis laterally between vexillum and
alae, without causing protrusion of the sexual organs from the carina. Sometimes
all the bees visiting one particular shrub sucked legitimately, while all those which
settled on another stole the nectar. Humble-bees were also seen inserting the
proboscis obliquely into the bases of the flowers without depressing the carina.
My own observations essentially agree with those of Kirchner.

VisiTors.—Apis mellifica Z. ¥ appears frequently as a visitor. At Kiel (17. 6.’96)
and Sonderburg, in the island of Alsen (1. 8. 96), I almost always observed it to
settle on the side of the blossoms, skg. laterally without operating the flower
mechanism. Sometimes a bee would try to suck legitimately, but was too weak to
force its way into the almost closed flowers. A humble-bee (Bombus lapidarius

LEGUMINOSAE 307

Z. 9) succeeded in doing this without any special effort, at the same time effecting
cross-pollination. I observed the same species of humble-bee in the island of
Pellworm (4. 6. ’93).

Schletterer records two bees for the Tyrol :—1. Melecta luctuosa Scop.; 2. Poda-
lirius tarsatus Spzz. He observed the Ichneumonid Perithous mediator /. at Pola.

Loew noticed the bee Megachile lagopoda Z. 9, po-cltg. in the Harz district
(‘ Beitriige,’ p. 52).

Kerner states that a butterfly (Lycaena baetica Z.) visits the flowers, the female
laying her eggs in the ovary (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 156).

214. Glycyrrhiza Tourn.
715. G. grandiflora Tausch.—
Visitors.— Loew saw the honey-bee, skg., in the Berlin Botanic Garden.

215. Tephrosia Pers.

716. T. heterantha Griseb.—Hieronymus states that this species produces
cleistogamous flowers (Jahresber. Ges. vaterl. Cultur, Breslau, 1897).

216. Robinia L.

White or reddish nectar-yielding bee flowers; with brush arrangement.

717. R. Pseud-acacia L. (Kirchner, ‘Flora v. Stuttgart,’ pp. 495-6; Knuth,
‘Bloemenbiol. Bijdragen.’)—The odorous white flowers of this species are arranged
in large pendulous racemes. The
vexillum has a green nectar-guide.
Kirchner states that the upper edges

of the carinal petals are closely ap- z

posed: there is the usual posterior

hollow for union with the alae.

Thecarina; dlae,andsemnalicolamn cio from aha aide” a. Bllgenn (o) sora from above

are mainly held together by the (Enlarged.)

vexillum, the lower part of which

grasps all of them by means of two well-developed elastic lobes. The posterior pro-
cesses of the alar laminae also surround the sexual column, so long as the alae are
gripped by the vexillum, for the alar claws are twisted outwards at the back, thus
pressing the laminae inwards and downwards. The anthers dehisce before the flower
opens; pollen collects among the hairs of the stylar brush, but protective bristles
prevent it from reaching the stigma. The perpendicular style is 6 mm. long, and the
terminal capitate stigma is surrounded by a circlet of protective bristles directed
obliquely upwards. Below these comes a hairless region about + mm. long, while
the part underneath this carries a brush of collecting-hairs crowded together ex-
ternally into a tract about 4 mm. long, more loosely arranged internally, and stretching
over a distance of from 13 to 2 mm. (see Fig. 99). I was able to satisfy myself
that the stigma remains sticky and receptive long after the anthers dehisce and

‘the pollen has been removed.

308 ANGIOSPERMAE—DICOTYLEDONES

During insect-visits the stigma first projects from the tip of the carina, and then
the pollen, both returning to the carina on removal of the pressure.

VisiTors.—These are bees, of which I observed:—r. Apis mellifica Z., skg.;
2. Bombus agrorum F,, do.

718. R. viscosa Vent. (=R. glutinosa Sims). (Knuth, ‘Bloemenbiol. Bij-
dragen.’)—This is a North American species, not infrequently cultivated in Europe
as an ornamental tree. The flowers are of a bright flesh-colour, and arranged in
dense racemes. There is a bright yellow nectar-guide on the vexillum. The brush
on the style resembles that of R. Pseud-acacia. The upper free filament is fused
for about half its length with the staminal tube.

Visitors.—At Kiel and Rendsburg I observed two bees, freq. skg. (28. 6 to
1. 7.’96):—1. Apis mellifica Z. ¥; 2. Bombus lapidarius Z. ¥.

217. Caragana Lam.
jig. C. arborescens Lam. (=Robinia Caragana Z.).—

Visitors. — Kirchner observed a humble-bee (Bombus lapidarius Z. ¥), skg.
legitimately, in Wurtemberg (‘Beitrage,’ p. 43).

218. Phaca L.

Nectar-yielding bee flowers, usually yellowish or violet in colour; with simple
valvular arrangement.

720. P, alpina Jacq. (Herm. Miiller, ‘Alpenblumen, pp. 236-7.)—In the
flowers of this species the nectar is concealed at a depth of 9-10 mm. The petals
are so close together that it is doubtful whether the numerous Lepidoptera that are
attracted by the marked conspicuousness of this plant succeed in reaching the nectar,
even when they have a proboscis of sufficient length. Probably only long-tongued
humble-bees are able to do this. It is doubtful whether self-pollination takes place.

Visitors. — Herm. Miiller observed 4 humble-bees and 9 Lepidoptera in
the Alps.

Loew noticed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syritta pipiens Z., flying in numbers around the
flowers and settling both on the alae and the carina, po-dvg.(?). B. Lepidoptera.
Rhopalocera: 2. Pieris napi L., skg.

721. P. frigida L. (Herm. Miiller, ‘Alpenblumen,’ pp. 237-8.)—In_ this
species self-pollination is possible in the Alps in individual cases. According to
Axell (‘Om Anord. fér Fanerog. Vaxt. Befrukt.,’ p. 17), the flowers are homogamous
in the Scandinavian highlands; but Lindman says that the anthers ripen in the
‘bud, before the stigma becomes receptive. In this region automatic self-pollination
takes place in fully matured flowers when the weather is favourable, but when it
is fine crossing is effected by the agency of humble-bees. Hermann Miiller observed
that, from the first, the stigma usually projects a little beyond the stamens; and
it is only in individual flowers that the pollen surrounds the stigma so that automatic
‘self-pollination takes place.

VistTors.—These are undoubtedly humble-bees, but the species which actually
effect cross-pollination have not, so far, been determined.

LEGUMINOSAE 309

219. Oxytropis DC.

Nectar-yielding bee flowers, mostly yellow or violet in colour; with a simple
valvular arrangement.

722. O. uralensis DC. (=O. Halleri Bunge). (Herm. Miiller, ‘Alpenblumen,’
pp. 232-4.)—A humble-bee which forced the vexillum and alae as far as possible
apart with its head would require a proboscis at least 10 mm. long to reach the
nectar in this species. As the stigma projects but little beyond the anthers, it gets
covered with their pollen, but does not appear to become receptive till a later stage.

Visrtors.—Herm. Miiller saw Bombus mendax Gers¢. 9 and ¥, skg.

723. O. Gaudini Reut—Kirchner (‘Beitrage,’ p. 44) says that plants of this
species growing at Zermatt have the same flower mechanism as O. uralensis, but
the calyx-tube is only 4 mm. long, so that even short-tongued bees can reach
the nectar.

724. O. montana DC. (Herm. Miiller, ‘ Alpenblumen,’ p. 234.)—In order to
reach the nectar of this species a proboscis 8-9 mm. long is necessary: otherwise
the flower mechanism agrees with that of O. uralensis.

Visitors.—Herm. Miiller observed a humble-bee and 2 Lepidoptera.

725. O. lapponica Gaud. (Herm. Miiller, ‘ Alpenblumen,’ pp. 234-5.)—As in
this species the calyx only surrounds the petals for a distance of 3 mm., the nectar
is more readily accessible than in O. montana, with which the flower mechanism
otherwise agrees.

Visttors.—Herm. Miiller only saw Lepidoptera, 2 butterflies and a Zygaena,
the latter probably effecting cross-pollination. Lindman observed humble-bees
paying short visits on the Dovrefjeld.

726. O. campestris DC. (Herm. Miiller, ‘Alpenblumen,’ pp. 235-6.)— In
this species the calyx-tube is 7-8 mm. long and 3-4 mm. broad. It ensheaths the
claws of the petals so closely that a proboscis 11-13 mm. in length is required
to reach the nectar. The vexillum possesses a nectar-guide, and the carina a pollen-
guide. In other respects the flower mechanism agrees with those of O. uralensis
and other species.

In the Alps the calyx is often perforated by Bombus mastrucatus 5 mm. above
its base. Forficula also gnaws through the flowers.

Visitors. — Herm. Miiller observed humble-bees (5) skg. and po-cltg., and
Lepidoptera (10), skg., as invited guests. Loew noticed in the Alps (‘Beitrage,’
p. 62) a humble-bee, Bombus pomorum Pz., var. elegans Sezd/. ¥, skg., and a
butterfly, Argynnis pales S.V.

Ekstam says that in Nova Zemlia the tolerably odorous flowers are visited by
Bombus hyperboreus Schénh. and B. nivalis Dahld.; and also by moderately large
flies.

727. O. pilosa DC. (Herm. Miiller, ‘Weit. Beob.,’ II, pp. 253-4; Loew,
Flora, Marburg, Ixxiv, 1891, pp. 84-91, ‘ Bliitenbiol. Floristik,’ pp. 220, 339; Schulz,
‘Beitrage,’ II, p. 269.)—In the flowers of this species observed by Hermann Miller
in Thuringia, the calyx ensheaths the petals for a distance of 6mm. The vexillum

310 ANGIOSPERMAE—DICOTYLEDONES

is folded together in the median plane. This fold, with the projecting tip of the
carina, guides the proboscis of bee visitors, which must be 6-7 mm. long to get
at the nectar. In this species too the stigma is surrounded by the pollen of its own
flower, but probably this does not adhere to it without pressure.

Loew has studied this species in the Uckermark, and compared it with those
described by Hermann Miller (‘ Alpenblumen,’ pp. 232-6), i.e. O. uralensis DC.,
O. montana DC., O. lapponica Gaud., and O. campestris DC. He describes it
as intermediate between O. uralensis and O. campestris on the one hand, and
O. lapponica on the other hand; for a proboscis 10-13 mm. long is requisite for
securing the nectar of the first two species, while one 4-5 mm. long is necessary
in the case of O. lapponica. O. montana requires a proboscis 8-9 mm. in length.

The very prominent tip of the carina, and the double union between alae and
carina are specially characteristic of O. pilosa. The carina, in accordance with the
mechanical arrangements of papilionaceous flowers, is exposed to the greatest
pressure and pulling-strain during insect-visits. Loew says that those parts of it
most liable to these stresses are best provided with epidermal cells of specifically
mechanical nature, the walls of these being either wavy or strongly ribbed.

It must be added that the outer walls of the epidermal cells covering the folds
of the double basal union between carina and alae are strongly papillose, while
increased firmness is given by cuticular ridges radiating from the tips of the
individual cells.

Visttors.—Loew observed long-tongued dasygastrid bees in the Uckermark.—
Eucera longicornis Z. 9 and 4, and Osmia aurulenta Pz. 9; also some Scopulipedes
{humble-bees). In Thuringia Herm. Miiller saw the honey-bee, skg., and Pieris
rapae Z., do. Schulz noticed flowers perforated by humble-bees.

220. Astragalus Tourn.

Nectar-yielding bee flowers, mostly yellowish or violet in colour; with simple
valvular arrangement.

728. A. Glycyphyllos L. (Herm. Miiller, ‘Weit. Beob.” IJ, pp. 232-3;
Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892, pp. 87-91; Schulz, ‘Beitrage,’
II, p. 209; Knuth, ‘Bloemenbiol. Bijdragen.’)—The flowers of this species are
greenish-yellow in colour. Hermann Miiller states that the edges of the carina
are so close together in the front part containing the anthers, that they scrape off
some of the pollen and leave it outside when the depressed carina moves up again.
The alae only interlock with the anterior part of the carina; the lower edges of
their finger-like processes, which are broad and flat, abut closely on the sexual
column. Only the upper half of the flower is ensheathed by the broad base of
the vexillum, which passes gradually into its erect portion. Along the middle of this
is a deep groove serving as a guide for the proboscis of bees. An open cleft remains
between the claws of the carina and alae through which the honey-bee is in the
habit of stealing nectar from the side.

Visitors.—Humble-bees and other long-tongued bees suck the nectar legiti-

mately and are effective visitors. Schulz also observed flowers perforated by
humble-bees,

LEGUMINOSAE gir

Hermann Miiller (H. M.), Buddeberg (Budd.), and myself (Kn.) observed the
following in North and Central Germany.—

A. Hymenoptera. Apidae: 1. Apis mellifica Z. ¥, skg. (H.M.); 2. Bombus
agrorum F. 9, in large numbers, skg. (H.M., Budd., Kn.); 3. B. hortorum Z. 9 ¥
and 6, in large numbers, skg. legitimately (H. M.); 4. B. lapidarius Z. 4, skg. (H. M.);
5. B. rajellus K.9, po-cltg. and skg. legitimately (H.M.); 6. B. variabilis Schmzedekn.,
var. tristis Sed/. ¥, skg. (H.M.). B. Lepidoptera. (a) Geometridae: 7. Odezia
chaerophyllata Z. (H.M.). (4) Rhopalocera: Melanargia galatea Z., skg. (H. M.).

Loew saw 2 bees in the Berlin Botanic Garden.—1. Megachile willughbiella
XK. 6, skg.; 2. Osmia rufa LZ. 9, po-cltg.

729. A. aristatus L’Hérit.—According to Briquet (‘Etudes de biol. flor. dans
les Alpes occident.’), the flowers of this species secrete abundant nectar, and are
therefore eagerly visited by bees and humble-bees. They possess an explosive
mechanism that acts only once, though the alae and carina subsequently return
to their original position, and during subsequent visits the sexual organs protrude,
as a result of elasticity. Automatic self-pollination may take place (Kirchner).

730. A. Cicer L—The flowers of this species are yellowish-white in colour
and possess an agreeable odour. Kirchner states that their mechanism essentially
tesembles that of A. Glycyphyllos.

Vistrors.—Loew saw a humble-bee, Bombus agrorum F. ¥, skg., in the Berlin
Botanic Garden.

Schulz (‘ Beitrage,’ II, p. 209) observed flowers perforated by humble-bees.

731. A. danicus Retz. (=A. hypoglottis Z.).—

VisrTors.—Schulz (loc. cit.) noticed flowers perforated by humble-bees.

732. A. exscapus L.—Automatic self-pollination is inevitable in this species,
should insect-visits fail, for the stigma lies between the anthers (Schulz).
Visirors.—Two humble-bees, Bombus hortorum Z. and B. agrorum /., skg.

legitimately and effecting cross-pollination. B. terrester Z. perforates the flowers
(Schulz, loc. cit.).

733. A. depressus L. (Herm. Miller, ‘Alpenblumen,’ pp. 230-1.)—In the
flowers of this species, the depressed carina very often fails to completely regain
its old position, so that the stigma and stamens remain protruding from it more
or less. Automatic self-pollination takes place if insect-visits fail (cf. Fig. 100).

Visttors.—Herm. Miiller observed 2 humble-bees and a moth (Plusia).

734. A. monspessulanus L.— Hermann Miiller’s idea (‘ Alpenblumen,’
p. 231), that the flowers of this species, conspicuous on account of their size and
purple colour, receive a greater number of visits from humble-bees than those of
A. depressus, is confirmed by MacLeod. He observed four species of humble-bee,
sucking legitimately in the Pyrenees. Miiller himself only saw Vanessa cardui L.,
sucking diligently and persistently.

Visrrors.—Loew noticed 2 bees in the Berlin Botanic Garden.—1z. Andrena
dorsata K. 9, po-cltg.; 2. Bombus agrorum /. 9, skg. And vide supra.

735. A. alpinus L. (=Phaca astragalina DC.). (Axell, ‘Om Anord. for
Fanerog. Vaxt. Befrukt.,’ p. 17; Herm. Miiller, ‘ Alpenblumen,’ pp. 231-2; Lindman,

312 ANGIOSPERMAE—DICOTYLEDONES

‘ Bidrag till Kanned. om Skandin. Fjellvaxt. Blomn. o. Befrukt.’; Loew, ‘ Bliitenbiol.
Floristik,’ p. 400.)—A proboscis 6 mm. in length is sufficient to reach the nectar
of this species, but bees of the kind appear to be wanting in the Alps. The agents
of pollination are humble-bees and other long-tongued bees, as well as numerous:
Lepidoptera.

Visitors.—Herm. Miiller observed Bombus alticola A7zechd. ¥, skg. and po-cltg.,
and also 6 Lepidoptera. Loew saw Bombus mastrucatus Gers. ¥, skg., and Osmia
morawitzi Gers¢. 6, do., in the Alps (Albula). Lindman noticed humble-bees and
Lepidoptera on the Dovrefjeld.

FIG. 100. Astragalus depressus, L. (after Herm. Miiller). 4. Older flower that has already been
visited (x43). 8. Flower after removal of the calyx, vexillum, and right ala. C. The tight ala, seen from
the inner side. JD. Flower after removal of calyx and vexillum, seen from above. Z. Carina after removal
ofits right side. _¥. Anterior portion of the carina, depressed. (B-Fx7.) a, anthers; 4 and v, staminal
tube; ¢, alar fold (seen from inside) ; e’, do. (seen from outside); # alar lappet ; 72’, alae; 72, nectaries ;
o, upper filament; ov, ovary; ss’ sch, parts ofcarina; sé, stigma.

Schneider observed the following humble-bees in Arctic Norway (Mus. Aarsh.
Tromse, xvii, 1895) :—1. Bombus alpinus Z. 9 and §; 2. B. hyperboraeus Schonh. 9;

3- B. hypnorum Z. 9 and ¥; 4. B. lapponicus Z. 9 and ¥; 5. B. scrimshiranus &. Q
and ¥; 6. B. terrester Z. 9 and ¥.

Ekstam saw small humble-bees visiting the fragrant flowers in Nova Zemlia.
736. A. oroboides Hornem.—Axell describes (op. cit., p. 17) the flowers of
this species as homogamous, and says that their mechanism agrees with that

of A. alpinus. They are pale blue, passing into violet at the bases of the vexillum
and carina, and markedly asymmetrical.

LEGUMINOSAE 313

VisrTors.—Lindman says that the flowers are sparingly visited on the Dovrefjeld
by humble-bees and Lepidoptera.

737. A. alopecuroides L.—

Vistrors.—Loew saw Bombus hortorum Z. 9 and ¥, steadily skg., in the Berlin
Botanic Garden.

738. A. arenarius L.—

VisiTors.—Bombus pratorum Z. ¥, skg. (Loew, Berlin).
739. A. glycyphylloides DC.—

VisiTors.—Bombus agrorum /. 9, skg. (Loew, Berlin).
740. A. narbonensis Gouan.—

VisitTors.—Bombus hortorum Z. ¥, skg., and Megachile fasciata Sm. 6, do.
(Loew, Berlin).

741. A. onobrychis L.—

Visitors.—Megachile fasciata Sm. 6, skg. (Loew, Berlin). Von Dalla Torre
and Schletterer record the following bees for the Tyrol.—

1. Andrena curvungula Zhoms.; 2. Bombus confusus Schenck; 3. B. hortorum
L.; 4. B. variabilis Schmzedekn.; 5. Eucera longicornis Z.; 6. Megachile muraria
L.; 7. Melecta luctuosa Scop.; 8. Osmia aurulenta Pz.; 9. O. cornuta L/r.; 10. O.
spinolae Schenck; 11. Podalirius fulvitarsis Zep.; 12. P. parietinus /.; 13. P. retusus
L.; 14. Sphecodes similis Wesm.

Schulz observed flowers perforated by humble-bees at Bozen.

221. Coronilla L.

Yellow nectarless bee flowers, with pumping arrangement, from which threads
of pollen are extruded.

742. C. vaginalis Lam. (=C. montana Schr.). (Herm. Miiller, ‘ Alpenblumen,’
pp. 249—-52.)—The flower mechanism of this species agrees on the whole with that of
Lotus. There are differences, however, in the order of development of the inner
ard outer filaments, and as to the part they play in pressing out pollen; also as
regards the relative size of the alae and carina and the way they are united together.
The carina too is less easily depressed. It is doubtful if self-pollination takes place
(cf. Fig. ror).

Visrrors.—These are very rare. In good weather, and after watching for days,
Herm. Miiller only once observed a po-cltg. bee (Andrena ?).

743. C. varia L. (Delpino, ‘Ult. oss.” p. 45; Herm. Miiller, ‘ Fertilisation,’
pp. 198-9; Kirchner, ‘Flora v. Stuttgart,’ p. 498; Loew, ‘Bliitenbiol. Floristik,’
Pp- 399-)—In this species, again, the flower mechanism is similar to that of Lotus,
except that the thickened ends of all ten filaments act as pistons. The two
openings at the base of the free filament are wanting, for the flowers do not secrete
nectar in the usual place, but on the outside of the fleshy calyx, where it is sought
out by bees. These settle upon the alae in the normal fashion, and probe beneath
the vexillum. The proboscis passes through the wide space between the unusually
slender bases of the petals to the outside of the flower, encountering the nectar on
the calyx (Kirchner).

314 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Herm. Miiller noticed the honey-bee in Thuringia. Loew saw 2
bee (Andrena propinqua Schenck 9) po-cltg. in the Alps; and, in Silesia, a butterfly
(Hesperia comma J.) vainly trying to suck. He also observed the following in the
Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis nemorum Z., settling on the corolla.
B. Hymenoptera. <Apzdae: 2. Anthidium manicatum Z. 9, po-cltg., and trying to
suck; 3. Bombus agrorum /. ¥, vainly skg.; 4. B. hortorum Z. ¥, po-cltg.; 5. B.
lapidarius Z. ¥, po-cltg. and vainly skg.; 6. B. rajellus A. ¥, skg.; 7. Megachile
centuncularis Z. 9, po-cltg. and vainly skg.; 8. M. fasciata Sm.9, do.; 9. M. lagopoda
L. 9, po-cltg.; 10. Osmia aenea Z. 9, do.

Fic. 101. Coronilla vaginalis, Lam. (after Herm. Miller). 4, Flower seen from the front (x 3}).
B. The same, after removal of the right half of the vexillum, and of the right ala. C. Left ala seen
from the inside (x 3$). J. Flower after removal of calyx, vexillum, and alae; seen from above (x 7).
£&. Young bud in longitudinal sections. #. Calyx and carina with enclosed stamens and pistil (in
section). ”’. Style and stigma. G. The nine united stamens, removed from a bud and spread out.
a, anthers; ca, calyx; ¢, alar process, fitting into carinal pit (e’); ex, tip of carina; /, vexillum;
#, filaments; 77, alae; gv, style; 0, posterior filament ; ov, ovules; #0, pollen; s, carina; s’, carinal claw;
sé, stigma; v, staminal tube.

Réssler saw the butterfly Lycaena argus Z. at Wiesbaden. Schletterer noticed

the mason-bee Megachile (Chalicodoma) muraria Ae/z. in the Tyrol; and, at Pola,
the rare Xylocopa cyanescens Brui/., which is the smallest European carpenter-bee.

744. C. montana Scop., 745. C. glauca L., and 746. C. minima L.—These
species, according to Farrer (Nature, London, x, 1874), agree with C. varia as
regards the flower mechanism and secretion of nectar.

Vistrors.—For C. montana Loew observed a humble-bee (Bombus rajellus
&. ¥), po-cltg., in the Berlin Botanic Garden.

747. C. Emerus L.—It was in this species that Delpino (‘Ult. oss.,’ pp. 39-44)
first recognized and described in detail the pump arrangement by which threads
of pollen are extruded (‘apparecchio che offre una curiosa analogia col meccanismo
con cui si fabbrica la pasta da vermicellajo’).

LEGUMINOSAE 315

Visitors.—Delpino noticed long-tongued bees :—Bombus, Eucera longicornis
L., Anthophora pilipes /., and Xylocopa violacea Z. Friese saw the beautiful bee
Podalirius tarsatus Sprn., freq., at Bozen. Ducke, at Trieste, observed the bees
Eucera caspica Mor. 9 and é, and Megachile (Chalicodoma) manicata Grr. 9 and @.
Schletterer records the following bees for Pola.—

1. Andrena carbonaria Z.; 2. A. flavipes Pz.; 3. A. parvula K.; 4. Eucera

interrupta Baer.; 5. Halictus patellatus Wor.; 6. H. sexcinctus F.; 7. Podalirius
tarsatus Spzz. (also in the Tyrol).

222. Ornithopus L.
Bee flowers with simple valvular arrangement.

748. O. perpusillus L. (Herm. Miller, ‘ Weit. Beob.,’ IJ, pp. 262-3; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 62.)—In the diminutive yellowish flowers of this
species, the vexillum is streaked with purple. The bases of the petals and stamens
are fused with the calyx. Hermann Miiller supposed this fusion to indicate that
in favourable weather the base of the flower would be quite full of nectar, but on
examination found none at all. Nor was I able to detect any in numerous flowers
examined in the island of Féhr. The stamens and pistil mature simultaneously,
and are of equal length. Despite careful watching I observed no visitors, yet fruits
are regularly set, so that automatic self-pollination is undoubtedly effective.

Visttors.—Herm. Miiller, in Westphalia, observed only one diminutive bee
(Halictus flavipes F. 9), skg. and po-cltg., and a minute digging-wasp (Passaloecus
turionum Dads. 3), skg.?

MacLeod saw Bombus agrorum /. 9, skg.?, in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 354).

A hover-fly has been recorded for Dumfriesshire (Scott-Elliot, ‘Flora of Dum-
friesshire,’ p. 48).

749. O. sativus Brot. (Kirchner, ‘ Beitrage,’ pp. 44-5 ; Knuth, ‘ Bloemenbiol.
Bijdragen.’)—In cultivated plants of this species, Kirchner described the flower
mechanism as follows.—The calyx is about 5 mm. in length, and its tube 23 mm.
The erect lamina of the vexillum, which projects from it, is 7-8 mm. long, and rose-
coloured with dark veins. The alar laminae are of a brighter hue, about 6 mm.
long, and with a deep longitudinal fold running parallel to their upper margin.
These folds overlie the carina and stamina] tube in such a way as to be in con-
tinuous contact with them. The alae also possess rounded elastic processes, directed
backwards. At the posterior end of each alar lamina its process firmly interlocks
with a dorso-lateral carinal pit. Carina and alae are therefore intimately united in
this region. The spheroidal stigma is closely surrounded by the mature anthers,
both projecting together from the greenish carina (only 1 mm. long) when the alae
are depressed. They return into the carina when the pressure is removed.

Although a moderately large aperture on either side the base of the upper
free filament leads into the staminal tube, Kirchner found no nectar even in sunny
weather; nor could I detect any. Kirchner thinks secretion possibly takes place
only under particularly favourable circumstances, or in the native habitat of the
species in the South.

316 ANGIOSPERMAE—DICOTYLEDONES

The flower is slightly asymmetrical, the vexillum being twisted a little to the
right at its base, and the left ala to the left. Of the two longitudinal alar folds,
that on the left is the deeper. The right ala is nearly perpendicular, or curved
somewhat to the left. The front ends of the filaments are also somewhat twisted
to the left.

Stigma and anthers being close together, automatic self-pollination is inevitable.
Cross-pollination can be effected by insects.

Visirors. — Kirchner (Wurtemburg) and myself (Schleswig-Holstein) have
observed the honey-bee. It introduces its proboscis normally as if about to suck,
and therefore possibly bores for sap in the base of the flower. Kirchner also saw
Meligethes. MacLeod noticed Apis and Eristalis tenax Z. in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 380).

Fic. 102. Aippocrepis comosa, L. (after Herm. Miller). 4. Flower seen from the side (x 4).
B. The same, seen from the front. C. Flower after removal of the vexillum and the upper part of the
calyx; seen from above (x 7). C! and C?. Lower part of the vexillar claw, with the plate that closes
the nectar-passages. LJ. The same flower, after removal of the alae. Z. Right ala, seen from the
inside. . Carina, seen from the side. G. The same in section, more strongly magnified. a, anthers:
ca, calyx; e¢, alar process, fitting into carinal pit (e); ex, tip of carina; fand /, vexillar lamina and claw;
Ja, alar fold, fitting into carinal groove ( fa’); 7, filaments; 7 and’, alar lamina and claw; 4, nectar-
Passage; 0, upper free filament; #0, pollen; s and s’, carinal lamina and claw; sé, stigma; vz,
staminal tube.

223. Hippocrepis L.

Yellow nectar-yielding bee flowers, with pumping arrangement extruding threads
of pollen.

750. H. comosa L. (Herm. Miiller, ‘Alpenblumen,’ pp. 252-4.)—Hermann
Miiller says that the flower mechanism agrees essentially with that of Lotus. The
union of alae and carina, however, is much firmer, for each of the former possesses
a fold and a deep saccular process, which fit into corresponding carinal depressions.

LEGUMINOSAE 317

The nectar is also concealed in a remarkable way, for the claw of the vexillum
is so slender and bends so far upwards out of the short calyx, that it is possible
to look into the flower laterally between it and the stamens. It would, therefore,
appear as if insect visitors could easily steal nectar from the side, without setting
the flower mechanism in motion. This, however, is not the case, for the claw of
the vexillum bears a projecting triangular plate on the under-side of its base, and
by this the two nectar-passages are completely closed. Insects can only remove
this obstruction by thrusting in their heads under the vexillum (cf. Fig. 102).

Visitors.—Herm. Miiller chiefly observed bees (12) and Lepidoptera (9) in
the Alps. Schulz noticed flowers perforated by humble-bees in Central Germany.

Schmiedeknecht saw 3 bees in Thuringia:—1. Osmia aurulenta Pz.; 2. O.
uncinata Gers/.; 3. O. xanthomelaena X. (=O. fuciformis Gerst.).

Friese observed the following bees in Baden (B.), Hungary (H.), Switzerland (S.),
Thuringia (T.), and Trieste (Tr.).—

1. Megachile muraria Re/z. (B.); 2. Osmia acuticornis Duf. ef Perr. 6 (H.);
3. O. andrenoides Sgzx., infrequent ; 4. O. aurulenta Pz., freq. (B.); 5. O. gallarum
Spin., not infrequent (H., Tr.); 6. O. lepeletieri Pev.; 7. O. leucomelaena X., freq.

(H., Tr.); 8. O. rufohirta Zep. 9 and 4, skg. (H., T.); 9. O. uncinata Gers?. (S., T.);
10. O. xanthomelaena XX. (T., 5.).

Loew saw Apis mellifica Z. ¥, po-cltg., in Hesse, and, skg., in the Berlin Botanic
Garden, where he also noticed a humble-bee (Bombus lapidarius Z. ¥), do. (‘ Beitrage,’
P- 53):

Ducke observed the following Afzdae at Trieste —

1. Eucera cinerea Zep. 9 and é; 2. Megachile (Chalicodoma) pyrenaica Lep. ;
3. Osmia andrenoides Spzn. 9 and 4, freq.; 4. O. campanularis Jor. 6; 5. O. giraudi
Schmiedekn., not very infrequent; 6. O. fulviventris Pz. 4, not infrequent; 7. O.
jheringi Ducke 9 and 4, very freq.; 8. O. longiceps Zor. 9 and 6, not infrequent ;
g. O. pallicornis Friese 9 and 4, freq.; 10. O. rubicola Friese 9, freq., 6, very
occasional ; 11. O. rufohirta Li. 9, freq., 5, rather rare; 12. O. solskyi Mor., rather
rare; 13. O. tergestensis Ducke 9 and 4, infrequent; 14. O. tiflensis AZor. 9 and 4,
occasional; 15. O. tridentata Duf. ef Perr., rare.

MacLeod noticed 2 long-tongued bees and a Lepidopterid in the Pyrenees
{Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 440).

224. Desmodium Desv.

751. D. canadense DC.—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Melithreptus scriptus Z., settling. B. Hymeno-
ptera. Apzdae: 2. Megachile centuncularis L. 9, po-cltg.; 3. M. fasciata Sm., 9, do.
C. Lepidoptera. Rhopalocera: 4. Pieris brassicae Z., skg.

225. Hedysarum L.

Red nectar-yielding bee flowers, with simple valvular arrangement.

752. H. obscurum L. (Herm. Miiller, ‘Alpenblumen,’ pp. 254-5; Schulz,
‘Beitrige,’ I, p. 32, II, p. 210.}—Hermann Miiller says that for this species a
proboscis 9-10 mm. long is required to suck legitimately. The flower mechanism
is the simplest in the sub-order. When a humble-bee visits a flower, the stigma

318 ANGIOSPERMAE—DICOTYLEDONES

and anthers protrude from the carina, pressing against its ventral surface. As the
stigma projects about 2 mm. beyond the anthers, it is the first to emerge, so that
cross-pollination is assured, and self-pollination rendered difficult.

Vistrors.—Herm. Miiller chiefly observed skg. or po-cltg. humble-bees (5), and
skg. Lepidoptera (13), which usually effected cross-pollination. Bombus mastrucatus
Gers?. also obtained nectar by perforating the flowers.

In the Riesengebirge, A. Schulz noticed humble-bees, and also flowers per-
forated by them.

Loew saw the following bees in the Berlin Botanic Garden.—1. Apis mellifica

L.¥, skg.; 2. Bombus
sr hortorum ZL. 9, do.;
3. B. lapidarius Z. ¥,
do.; 4. Osmia rufa Z.

» @

P
Ly i 9, skg. and po-cltg.
8 >,

Z 753. H. sibiricum
Poir.—

Visitors. — Loew
observed 2 humble-
bees (Bombus agrorum
F. ¥, and B. rajellus
&.9) skg., in the Berlin
Botanic Garden.

754. H. corona-
rium L.—This is an
Italian species.

Visitors. — Von
Dalla Torre saw 3
bees in the Innsbruck

Fic. 103. Hedysarum obscurum, L. (after Herm. Miiller). 4. Flower
seen from the side (x 14). &. Flower after removal of calyx, vexillum, and Botanic Garden.—t.

alae, and depression of the carina, seen from the side. J. The same seen Megachile ericetorum
from above. C. Right ala from the inside (B-D x 34). a, anthers; ca, calyx;

e’, alar fold; / vexillum; 7 and_//7’, alar lamina and claw; 2, nectar-passage ; Lep. 3, effecting pol-
o, free upper filament; ov, ovary; s and s’, carinal lamina and claw; v, : : 7 .
staiminal tabe: lination; 2. Halictus

leucozonius ., var.
nigrotibialis D.-7., do.; 3. Megachile maritima X. 6, freq.

Schletterer records the same for the Tyrol.

226. Onobrychis Tourn.
Red nectar-yielding bee flowers, with simple valvular arrangement.

755. O. viciaefolia Scop. (=O. sativa Lam.). (Herm. Miller, ‘Fertilisation,’
pp. 200-1, ‘Weit. Beob.,’ I, p. 263; Schulz, ‘Beitrage’; Knuth, ‘Bloemenbiol.
Bijdragen.’)—Hermann Miiller says that the flower mechanism of this species agrees
essentially with those of Melilotus and Trifolium. The stigma and anthers pro-
trude from the carina when this is weighed down by an insect visitor, resuming
their original position as soon as the pressure is removed. The vexillum is rose-red
with darker streaks, and the carina of a brighter red. The alae are greatly reduced,

LEGUMINOSAE 319

covering only the claws of the carina, and serving as nectar-covers, by which
lateral access is prevented or at least rendered difficult. The platform for insect
visitors is therefore constituted by the carina alone. As the result of its own
elasticity, this returns to its original position on the conclusion of a visit. The
stigma projects beyond the anthers, so that cross-pollination is ensured should such
visits take place. Automatic self-pollination is excluded, especially as the style
becomes more and more erect as anthesis progresses, and ultimately projects
1-1} mm. from the carinal cleft. Since the calyx-tube is only 2-3 mm. long, the
nectar and pollen are accessible, even to the shortest-tongued bees. Schulz noticed
flowers perforated by humble-bees in Central Germany.

Visitors.—Herm. Miiller says that the honey-bee (skg. and po-cltg.) is by far

the most frequent guest, paying at least nine-tenths of all the visits. I myself
noticed it in great num-

bers in Mecklenburg.
Herm. Miiller also re-
cords the following.—

A. Hymenoptera.
Apidae: 1. Anthidium
manicatum L. 9, skg. and
po-cltg., 4, skg.; 2. An-
drena labialis ., do.;
3. A. nigroaenea XK. 6;
4. Apis mellifica Z. ¥, skg.;
5. Bombus agrorum /.
g and ¥, skg. and po-cltg. ;
6. B. confusus Schenck 9,
do.; 7. B. muscorum /,
g, do.; 8. B. pratorum
£. 9 and ¥, do.; 9. B.
scrimshiranus K. 9, do. ;
10. B. sylvarum Z. 9, do. ;

11. B. Terkester L. 3 do. ¥ Fic. 104. Onobrychis victaefolia, Scop. (after Herm. Miiller). (1)
12. Chalicodoma muraria Flower from below (x 3). (2) The same, after removal of the vexillam
F. 9, do. (Thuringia) ; 13. and upper half of the calyx; from above. (3) Stamens and pistil, from

: . the side (x 7). a, calyx; 6, carina; ¢, ala; d, vexillum; e, alar claw;
Coelioxys conoidea iil. z J, united filaments; g, free filament; 4, nectar-passage; 7, “cleft in the
skg. > 44. C. umbrina carina, through which stamens and stigma protrude.

Sm. 6, in large numbers,

skg.; 15. Eucera longicornis Z. 9 and %, skg. and po-cltg.; 16. Halictus
albipes /. 9, do.; 17. H. flavipes #. 9, do.; 18. H. lugubris X. 9; 19. Mega-
chile argentata /. 4, skg.; 20. M. centuncularis Z. 4, do.; 21. M. circum-
cincta K. 9, skg. and po-cltg.; 22. M. fasciata Sm. 3, skg.; 23. M. willughbiella
&. 9, skg. and po-cltg., 5, skg.; 24. Osmia aenea Z. 9, freq., skg. and po-cltg. ;
25. O. aurulenta Pz. 9, skg. and po-cltg. (Thuringia); 26. O. fulviventris Pz. 9, in
large numbers, skg. and po-cltg.; 27. O. rufa Z. 9, skg.; 28. O. spinulosa K. 9, do.
(Thuringia); 29. Psithyrus campestris Pz. 9, do.; 30. P. rupestris /. 9, do.; 31.
Xylocopa violacea Z. $6, do. B. Diptera. Syrphzdae: 32. Volucella bombylans Z.,
var. plumata 7g. C. Lepidoptera. (a) Noc/uidae: 33. Euclidia glyphica Z., freq.,
skg., but apparently not effecting pollination; 34. Plusia gamma Z. do. (4) Rhopalo-
cera: 35. Lycaena aegon S.V. 3, skg.; 36. L. corydon Poda., do.; 37. L. icarus
Rott, do.; 38. L. sp., freq., skg., but apparently not effecting pollination; 39. Pieris
napi Z., skg.; 40. Thecla ilicis Bsp., do. (c) Sphingidae: 41. Zygaena carniolica
Scop., freq., skg., but apparently not effecting pollination (Thuringia).

320 ANGIOSPERMAE— DICOTYLEDONES

Herm. Miiller also saw 4 bees in the Alps (‘ Alpenblumen,’ p. 254).

Loew noticed 5 bees in the Berlin Botanic Garden :—1. Anthidium manicatum
L. 9, skg. and po-cltg.; 2. Bombus agrorum Ff”. 9, skg.; 3. B. lapidarius Z. ¥, do. ;
4. Megachile fasciata Sm. 6, do.; 5. Osmia aenea LZ. 9, po-cltg.

Schletterer records a bee (Meliturga clavicornis Z/r.) for the Tyrol.

Réssler observed a moth (Grapholitha caecana Schl.) at Wiesbaden. Ducke
records the following bees.—

1, Anthidium cingulatum Z/r. 9 and $; 2. Melitta dimidiata M/or.; 3. Osmia
rubicola Friese 9, freq., 8, occasional; 4. O. rufohirta Zér. 9, freq.; 5. O. tergestensis
Ducke 9 and 3; 6. O. tiflensis A/for. 9, occasional; 7. O. versicolor Zir. 9 and 4,
very freq.

756. O. aurea Stev.—

Vistrors.— Loew saw a bee (Osmia aenea Z. 9), skg. and po-cltg., in the Berlin
Botanic Garden.

757. O. montana DC.—

VisiTors.—T wo bees :—Anthidium manicatum Z. 9, skg. and po-cltg., Megachile
fasciata Sm. 3, skg. (Loew, Berlin).

758. O. arenaria DC.—

Vistrors.—A humble-bee, Bombus rajellus A’. ¥, skg. (Loew, Berlin). Friese (on
the authority of Mocsary) gives the bee Nomia femoralis Pail., freq., for Hungary.

227. Vicia Tourn.
Nectar-yielding bee flowers, with a stylar brush.

In many species of this genus there are extra-floral nectaries. These are deeply
coloured spots situated on the under-side of the stipules. They secrete nectar in
sunny but not in dull weather. The secretion is eagerly sought out by ants, which
for their part serve to protect the plants against caterpillars and the like.

9759. V. Cracca L. (Delpino, ‘ Ult. oss.’ p. 58; Herm. Miiller, ‘ Fertilisation,’
pp. 202-4, ‘ Weit. Beob.,’ II, p. 262; Lindman, ‘ Bidrag till Kanned. om Skandin.
Fjellvaxt. Blomn. o. Befrukt.’; Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892,
p. 100; MacLeod, op. cit., vi, 1894, pp. 354-6; Knuth, ‘Bl. u. Insekt. a. d. nordfr.
Ins.,’ ‘ Bliitenbiol. Beob. a. d. Ins. Riigen,’ ‘ Bloemenbiol. Bijdragen’ ; Loew, ‘ Bliiten-
biol. Statistik,’ p. 400.)—The purple flowers of this species are arranged in crowded
racemes. The flower mechanism was first described by Delpino, and subsequently
in greater detail by Hermann Miiller. Each ala is united with the carina at two
places, About the middle of its upper margin there is a hollow process, small
but deep, which fits closely into a pit in the upper surface of the carina. Immediately
behind this is a second alar process, much broader but equally deep. It is connected
with a broad, but rather shallow depression in the upper surface of the carina,
the epidermal cells of the two petals being so firmly interlocked that it is difficult
to separate without at the same time tearing them. On the back of the vexillum,
at the bend between claw and lamina, there are two grooves diverging anteriorly ;
these project below as ridges which are applied to the alae so as to block up the
lateral passages to the nectar.

LEGUMINOSAE 321

When a bee visits the flower it settles on the alae, and these, being firmly
united with the carina as above described, act like the long arm of a lever and
cause its depression. The carina and alae return to their original position after
the insect visitor has departed. This is partly due to their elasticity, and partly
because the upper basal corners of the alae are produced into backwardly and
inwardly directed processes, which lie upon the upper surface of the sexual column.
This is also grasped by the two upper basal lobes of the carina, between which
only a narrow space is left. And lastly, the broad base of the vexillum curves
so far forwards on each side that it completely surrounds the claws of the alae
and carina. More than half the length of the extremely short style (only about
1% mm. in length), is beset with hairs from immediately below the terminal stigma
downwards. These are longer and closer on the outer than on the inner side.

Fic. 105. Victa Cracca, L. (after Herm. Miiller). (1) Flower seen from the side (x 3). (2) The
same, after removal of calyx and vexillum, seen from above (further enlarged). (3) The same, after
the alae also have been removed. (4) Left ala from the inside. (5) Style (still more enlarged).
a, anterior alar process; a’, corresponding carinal pit; 6, posterior alar process; 6’, corresponding
carinal depression; c, backwardly and inwardly directed alar process; d@, alar claws; e¢, carinal claws;
J, dilatation of the carina containing pollen; g, upper basal lobes of the carina; 4, nectar; 0, opening
through which the style passes out; #, stylar brush; s/, stigma.

The flowers have barely reached half their full size, when the anthers, which
closely surround the stylar brush, dehisce and shed their pollen among the hairs
of the brush, the stigma being also covered with it. When a bee visits the flower,
pollen adheres to its under-side and the stigma is at the same time rendered sticky
and receptive by rupture of its papillae.

Visitors.—These are bees and Lepidoptera, though the latter suck nectar
without effecting pollination. I saw Apis, 2 Bombus, and a Zygaena on the North
Frisian Islands; Bombus agrorum F., skg., at Flensburg; and, in Riigen, 2 humble-
bees, skg.—Bombus hortorum Z., and B. sylvarum, var. albicauda Schmzedekn.

Alfken observed the following bees at Bremen.—

1. Anthidium manicatum Z.9; 2. Bombus arenicola Zhs. 9 and ¥; 3. B. der-
hamellus A. ¥ and 6; 4. B. distinguendus Aor. ¥; 5. B. muscorum F. 9; 6. B.
sylvarum Z. 9 and ¥; 7. Coelioxys rufescens Zep. 9, skg.; 8. Eucera difficilis (Duf.)
Per. 9; 9. Megachile centuncularis Z. 9 and 4; 10. Podalirius borealis Aor. 9.

Krieger saw Eucera longicornis Z. (once) at Leipzig. De Vries noticed
numerous honey-bees, skg., in the Netherlands. Heinsius saw Zygaena filipendulae
Z., and Lycaena icarus Ro//. 6 in Holland. MacLeod observed 2 long-tongued
bees, and 2 Lepidoptera in Flanders; a humble-bee and a Lepidopterid in the

DAVIS. IL Y

322 ANGIOSPERMAE—DICOTYLEDONES

Pyrenees. Loew noticed Psithyrus globosus Zv. 4, skg., in the Alps, and Lindman
saw several humble-bees and Lepidoptera on the Dovrefjeld.

Hermann Miller gives the following list for Westphalia.—

A. Hymenoptera. (a) Apidae: 1. Apis mellifica Z.¥, freq., skg. (Thuringia) ;
2. Bombus agrorum F”. 9 and ¥, skg.; 3. B. hortorum Z. ¥, do.; 4. B. rajellus X. ¥,
do.; 5. B. scrimshiranus X. 6 9 and ¥, do.; 6. Eucera longicornis Z. 9 and 6, do. ;
7. Diphysis serratulae Pz. 9, skg. and po-cltg.; 8. Megachile circumcincta X. 9, do. ;
9. M. maritima 4. 9, do.; 10. M. versicolor Sm. 9, do.; 11. M. willughbiella &. 9,
do. ; 12. Osmia adunca Z/r. 9, do. ; 13. Psithyrus vestalis Fourc.d,do. (4) Vespidae:
14. Odynerus quadrifasciatus /’. 9, vainly skg. B. Diptera. mpzdae: 15. Empis
livida Z., freq., skg. C. Lepidoptera. (a) Rhopalocera: 16. Hesperia lineola O.,
skg.; 17. Lycaena arion Z., do.; 18. Melanargia galatea Z., do.; 19. Pieris rapae Z.,
do., but without effecting pollination. (2) Sphzngidae: 20. Zygaena meliloti Esp., skg.

Herm. Miiller also saw 4 Apidae and 5 Lepidoptera in the Alps (‘ Alpenblumen,’
p- 249).

In Dumfriesshire, 2 humble-bees, an Empid, and a hover-fly were recorded
(Scott-Elliot, Flora, p. 49).

760. V. hybrida L.—

VisiTors.—Schletterer observed three bees at Pola :—1. Eucera interrupta Baer. ;
2. E. longicornis Z.; 3. Halictus interruptus Pz.

761. V. dumetorum L.—Kirchner (‘Flora v. Stuttgart,’ p. 503) describes the
flower mechanism as similar to that of V. Cracca. The style is 3 mm. in length,
and surrounded by hairs for a length of 1 mm. from its tip; the hairs on the outer
being noticeably longer than those on the inner side. The anthers dehisce in the
young bud, but the stigma is moderately well protected from the pollen of its own
flower by the stylar brush. Each ala is united with the carina by a smaller anterior,
and a much larger and deeper posterior process: the epidermal cells of the latter
interlock.

Visitors.—These are bees. Loew observed Bombus agrorum F. ¥, skg., in
the Berlin Botanic Garden. The honey-bee steals nectar from the side by pushing
the petals apart. Schulz noticed flowers perforated by humble-bees.

762. V. villosa Roth.—Kirchner (‘Flora v. Stuttgart,’ p. 502) states that the
flower mechanism of this species also agrees essentially with that of V. Cracca,
but the anthers dehisce and shed their pollen on the stylar brush when the flowers
are almost full grown.

Visrtors.—Hd6ppner saw a bee (Podalirius retusus Z.), ske., at Bremen.

Schletterer observed the variety varia Host. to be visited by the following bees
at Pola.—

1. Anthidium manicatum Z.; 2. Colletes lacunatus Dours.; 3. Eucera alternans
Brull.; 4.E.longicornis Z.; 5. E. parvula Przese; 6. E. ruficollis Brull.; 7. Poda-
lirius retusus Z., var. meridionalis Per.; 8. P. tarsatus Sp. Also g. the Mutilla
viduata Pall.

763. V. sepium L. (Sprengel, ‘Entd. Geh.” pp. 356-7; Herm. Miiller,
‘Fertilisation, pp. 204-6, ‘ Weit. Beob.,’ II, p. 262; Schulz, ‘ Beitrage’; de Vries,
Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875; Knuth, ‘Bloemenbiol. Bij-
dragen’; Loew, ‘ Bliitenbiol. Floristik,’ pp. 392, 395.)—The blossoms of this species
are dirty lilac in colour, with a yellowish base. Hermann Miiller describes the

LEGUMINOSAE 323

flower mechanism as similar to that of V. Cracca, except as regards the nature
of the brush arrangement. For here the style (21 mm. long) bears two brushes,
immediately beneath the stigma. They are completely separate, about 1 mm. long,
and respectively situated internally and externally. The inner brush consists of
a single row of short hairs slanting upwards, while the outer one is composed
of hairs which radiate obliquely upwards towards the stigma, and enclose a plate-
like depression. The anthers dehisce when the flowers have attained a considerable
size, shed their pollen into a swelling at the tip of the carina, and are then retracted.
Access to the nectar is more difficult than in V. Cracca, for the petals are
thicker and firmer, the calyx-tube surrounds their claws for a greater distance,
and the vexillum is tumid at the entrance to the nectar-passage, which is bounded
by it and the alae. In addition to this, the arms of the alar levers, which depress

Fic. 106. Vicia sepium, L. (after Herm. Miller). (1) Flower seen from the front. (2) The same,
after removal of calyx and vexillum; seen from above. (3) The same, after the alae also have been
removed; seen from above. (4) The same, seen from the side. (5) Style with brushes and stigma, seen
from the side. (6) Brushes and stigma, seen from above. (7) Stamens and pistil of a bud. References
as in Fig. 105.

the carina, are relatively shorter than in V. Cracca. Hence only powerful bees
(Bombus, Anthophora) are able to suck legitimately and effect crossing. The flies
and Lepidoptera which so often steal nectar from V. Cracca, sucking legitimately
but without benefit to its flowers, are here excluded, and this is a distinct advantage.
On the other hand, Bombus terrester regularly perforates the flowers from the side
and steals the nectar, although it possesses the requisite strength, skill, and length
of proboscis. Nectar is stolen by feebler bees with a shorter proboscis (Apis, Osmia
rufa) through the holes thus made.

Visttors——The following have been observed by Herm. Miiller (H.M.) in
Westphalia, Buddeberg (Budd.) in Nassau, and myself (Kn.) in Schleswig-Holstein.—
A. Hymenoptera. Apidae: 1. Anthophora aestiva Pz. 6, skg. (Budd.);
2. A. pilipes #. 9 and 4, skg. legitimately (H.M.); 3. Apis mellifica Z. ¥, using the
holes made by B. terrester Z. (H.M., Kn.); 4. Bombus agrorum /. 9 and ¥, skg.
(H. M., Kn.);_ 5. B. lapidarius Z. 9 and ¥, do. (H.M., Kn.); 6. B. muscorum F. ¥,

Yo 2

324 ANGIOSPERMAE—DICOTYLEDONES

do. (Budd.); 7. B. rajellus A. 9 and 9, do. (H.M., Kn.); 8. B. sylvarum Z. 9, do-
(H.M.); 9. B. terrester Z. Y, perforating the flowers, and stealing nectar (H.M., Kn.);
10. Eucera longicornis Z. ¥ and 4, skg. (Budd., Kn.); 11. Megachile circumcincta
K. 9, do. (Budd.); 12. Osmia aurulenta Pz. 9, in large numbers, skg. (Budd.); 13. O.
rufa L. 9, freq., skg. (Budd.), stealing nectar through holes made by B. terrester Z_
(H.M.). B. Diptera. Bombyhidae: 14. Bombylius canescens Afrkan., skg. (Budd.),

Wiistnei saw Eucera longicornis Z. in the island of Alsen.

Alfken noticed the following bees at Bremen.—

1. Andrena convexiuscula XK. 8; 2. A. xanthura A. 9; 3. Bombus arenicola
Ths.9; 4. B. derhamellus X.9; 5. B. lapidarius Z.9; 6. B. muscorum F. 9; 7. B.
sylvarum LZ. 9 and ¥; 8. B. terrester Z. 9 (perforating the corollas); 9. Eucera difficilis
(Duf.) Per. 9.

Loew also saw Eucera in Brandenburg; Bombus agrorum F. ¥, skg., in the
Berlin Botanic Garden; and Megachile sp. in Silesia.

The following bees, &c., were noticed by the observers and at the places stated.—

Schenck (Nassau), 1. Bombus confusus Schenck; 2. B. lapidarius Z.; 3. B.
pomorum Pz.; 4. Eucera longicornis Z.; 5. Podalirius retusus Z.; and 6. the
Sphegid Gorytes mystaceus Z.: Réssler (Wiesbaden), the moth Toxocampa craccae
.: Friese (Baden), Andrena xanthura A., n. sp.: Hoffer (Steiermark), 1. Bombus
lapidarius Z.9 and ¥; 2. B. derhamellus A. 9 and ¥: Dalla Torre and Schletterer
(Tyrol), 1. Bombus derhamellus A. ¥; 2. B. variabilis Schmzedekn., var. tristis Sezd/. § ;
3. Andrena xanthura KX. 9; 4. Halictus major Wy. 6.

Herm. Miiller, in the Alps, saw Bombus mastrucatus Gers/. § getting the nectar
by perforation (‘Alpenblumen,’ p. 249). Schulz also noticed humble-bees stealing
in the same way.

MacLeod saw Bombus variabilis Schmzedekn. 9 in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, ili, 1891, p. 438); and, in Flanders, 2 humble-bees, an Andrena,
a Lepidopterid, Bombus terrester Z. stealing nectar by perforation, Apis and Osmia
sucking nectar through the holes thus made (op. cit., vi, 1894, p. 358). H. de Vries
noticed a humble-bee, Bombus sylvarum Z. 9, skg., in the Netherlands (Ned. Kruidk.
Arch., Nijmegen, 2. ser., 2. deel, 1875).

In Dumfriesshire 3 humble-bees were recorded (Scott-Elliot, ‘ Flora of Dumfries-
shire,’ p. 50).

Von Fricken, at Arnsberg, saw the Curculionid Bruchus pisi Z. as an injurious
visitor.

764. V. sativa L. (Sprengel, ‘Entd. Geh.” p. 357; Heinsius, Bot. Jaarb.
Dodonaea, Ghent, iv, 1892, pp. 96-100; Kirchner, ‘Neue Beob.,’ p. 44, ‘Flora v.
Stuttgart,’ p. 506; Schulz, ‘ Beitrage,’ II, p. 211; Knuth, ‘ Bloemenbiol. Bijdragen.’)—
In this species the alae are usually violet in colour, the vexillum lilac, and the carina
whitish with a blue tip. Kirchner says that the union of the alae with the carina,
effected in the usual way, is rendered so firm by the mutual interlocking of the
epidermal cells of these petals, that the alae tear when pulled apart. The posterior
angles of the carina are drawn out into processes which lie upon the sexual column.
There are also finger-shaped alar processes, running backwards parallel to each
other. The upper filament is united with the other nine, but two nectar-passages
are left at its base. The style is about 2 mm. long, and bears a brush on its upper
half, the hairs of which are disposed all round it, and directed obliquely upwards.
Externally there is a tuft of longer protective hairs, projecting beyond the stigma.

LEGUMINOSAE 325

Even in the bud the anthers dehisce, so that automatic self-pollination is
inevitable; it is thoroughly effective.

Visitors. —I saw 3 long-tongued bees in Schleswig-Holstein.—1. Bombus
agrorum F.9; 2. B. lapidarius Z. 9 and ¥; 3. Eucera longicornis Z. 9 and , all skg.
legitimately. Sprengel records Sphinx (Deilephila) euphorbiae Z.

Loew noticed Bombus sylvarum Z. 9, skg., in Silesia (‘Beitrige,’ p. 34). Heinsius
observed B. hortorum JZ. as an invited guest, and the brimstone butterfly (Rhodocera
rhamni Z.) as an unbidden one; also Bombus terrester Z., stealing nectar by per-
foration. Schulz, in Central Germany, also noticed flowers perforated by humble-bees.

MacLeod saw Bombus and Eucera in Flanders (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 361).

In Dumfriesshire humble-bees were recorded (Scott-Elliot, ‘Flora of Dumfries-
shire,’ p. 50).

Heinsius noticed the extra-floral nectaries to be visited by Vespa sylvestris Scop.,
V. rufa Z., Apis, and a fly (Cleigastra sp.).

765. V. angustifolia All. (H. von Mohl, Bot. Ztg., Leipzig, xxi, 1863,
p. 312; Treviranus, op. cit., xxi, 1863, p. 143; Kuhn, op. cit., xxv, 1867, p. 67;
Herm. Miiller, ‘Weit. Beob.,’ II, p. 262; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’
pp. 64, 153-)—This is regarded as the ancestral form of V. sativa. As early as
1863 Treviranus and Hugo von Mohl called attention to the underground flowers
and fruits of V. angustifolia (var. amphicarpos); about 10% of the plants (at Berlin)
possess subterranean cleistogamous flowers produced on runners bearing scale-
leaves. The mechanism of the ordinary chasmogamous flowers agrees with that
of V. sativa.

Visirors.—I saw Bombus cognatus Sveph., skg., and B. agrorum #., do., in
the North Frisian Islands. Schulz, in Central Germany, noticed flowers perforated
by humble-bees.

Herm. Miiller observed the following.—

A. Hymenoptera. (a) Apidac: 1. Bombus agrorum &. ¥, skg. persistently ;
2. B. muscorum F. ¥, skg.; 3. B. sylvarum Z. 9, do.; 4. Saropoda rotundata Pz., do.
B. Lepidoptera. (a) Rhopalocera: 5. Lycaena aegon W.V., skg. (4) Sphingidae :
6. Ino pruni ScAzf., skg.

Alfken saw a bee, Osmia solskyi Jor. 9, at Bremen.

766. V. narbonensis L.—Treviranus (loc. cit.) says that this species pro-
duces subterranean fruits.

767. V. pyrenaica Pourr.—There are subterranean fruits in this species
(Treviranus, loc, cit.).

Visttors.—MacLeod observed 4 species of humble-bee, an Anthophora, an
Eucera, and a Bombylius in the Pyrenees.

768. V. pannonica Jacq.—Kirchner (‘ Beitrage,’ p. 46) says that the mechanism
of the yellowish-white flowers of this species is much the same as that of V. sativa.
The vexillum is bright rose-red in colour, with brownish streaks as nectar-guides.
The carina is greenish-yellow with a brownish tip. The claw of this petal is so
broad that it surrounds all the other internal parts of the flower, and its lower
margins curve under the staminal tube, touching each other somewhere about the

326 ANGIOSPERMAE—DICOTYLEDONES

middle. The alar laminae possess longitudinal folds, and arch inwards so as to
touch each other at the middle of the curvature, in front of the tip of the carina.
Near the posterior end of the upper margin of each ala there is a long conical
process, which fits into a corresponding depression in the upper margin of the
carina. Still further back there are two deep alar folds, and also a shallower one,
the inner surfaces of all which are so intimately united with the carina that the
alae cannot be separated without tearing them. The rounded posterior lobes of
the alar and carinal laminae are directed backwards, and lie above the staminal
tube. Their elasticity causes this tube to return into the carina. The upper filament,
as in V. sativa, is fused with the others, permitting access to the nectar only by
means of passages at its base. There is abundant secretion at the usual place.
The claws of the petals are in contact with one another, and 13 mm. long. They
are enclosed by the calyx-tube for a distance of 6-7 mm. The stipules serve as
extra-floral nectaries, but only a few of them are functional.

Visitors.—Kirchner, in Wurtemberg, saw a humble-bee (Bombus lapidarius
Z£.%) on cultivated plants.

769. V. Faba L. (Sprengel, ‘Entd. Geh.,’ pp. 357-60; Herm. Miller,
‘Fertilisation,’ pp. 206-7; Darwin, Ann. Mag. Nat. Hist., London, Ser. 3, ii, 1858,
p- 460; Mattei, Justs bot. Jahresber., xvii, (1889) 1891, p. 480; Knuth, ‘ Blitenbiol.
Beob. a. d. Ins. Riigen.’)—The fragrant flowers of this species are white with a black
spot on each ala. Their mechanism agrees, according to Hermann Miiller, with that
of V. sepium, but they are much larger. The nectar, however, is more easily
accessible, for the vexillum and alae are less firmly held together, and the carina
is more readily depressed. The swollen processes on the under-side of the vexillum
are also wanting, so that the claw of this petal (13-16 mm. long) is but loosely
held by the calyx-tube. On the other hand, the two involutions connecting the
alae and carina are present, but the union is less firm than in V. sepium, and
the backwardly directed alar processes are much more feebly developed. The alae
and carina are but slightly elastic, so that if forcibly depressed they do not return
to their original position.

Darwin found that when insects were excluded the plant was only one-third
as fertile as when they were allowed to visit the flowers. If, however, the flowers
protected from such visits were shaken, healthy seeds were abundantly produced.
Mattei describes an Indian variety as adynamandrous.

Only long-tongued bees are able to suck the nectar legitimately, and in
doing this to effect cross-pollination. Short-tongued bees either collect pollen
on visited flowers with exposed anthers (simultaneously bringing about crossing)
or steal nectar by perforation. Bombus terrester Z. 9 (proboscis 7-9 mm. long)
is notorious in this respect, and very seldom tries to suck legitimately. The
honey-bee either steals nectar through the holes made by B. terrester, or collects
pollen.

VisiTors.—I observed the following bees in the Island of Riigen.—

1. Apis mellifica Z. ¥, stealing nectar as above mentioned; 2. Bombus terrester
£., perforating the flowers; 3. B. hortorum Z. 9, very freq., skg.; 4. B. rajellus
A. 9, skg.

LEGUMINOSAE 327

Hermann Miiller records the following.—

A. Hymenoptera. Apidae: 1. Andrena convexiuscula X. 9, po-cltg.; 2. A.
labialis A’. 4, vainly trying to suck; 3. Apis mellifica Z. ¥, stealing nectar as above,
and po-cltg.; 4. Bombus confusus Schenck 9, freq., skg.; 5. B. hortorum Z. 9, do.;
6. B. lapidarius Z. 9, do.; 7. B. muscorum F. 9, do.; 8. B. sylvarum ZL. 9, do.; 9. B.
terrester LZ. 9, perforating ; 10. Osmia rufa Z.9, skg. B. Coleoptera. Malacoder-
mata: 11. Malachius bipustulatus Z., po-dvg.

The following were observed by the authorities and at the places stated.—

Alfken (Bremen), the bees, 1. Bombus ruderatus /. 9; 2. B. terrester L. 9, skg.
illegitimately: Verhoeff (Norderney and Baltrum), the bees 1. Bombus lapidarius
L. 9, skg.; 2. B. cognatus Sveph. (=B. muscorum #.), a 9, do.: Alfken (A.) and
Leege (L.) (Juist), 1. the Syrphid Syrphus pyrastri Z., very freq., and the bees
2. Bombus hortorum JZ. (A., L.); 3. B. muscorum /. (L.); 4. B. terrester LZ. (L.);
H. de Vries (Netherlands), a bee, 1. Apis mellifica Z. ¥, very freq., and a humble-bee,
2. Bombus agrorum F”. ¥ (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).

770. V. hirsuta S. F. Gray (=Ervum hirsutum Z.). (Herm. Miiller, ‘ Weit.
Beob.,’ II, pp. 260-2; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 361;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 62-3.)—The small bluish-white flowers
of this species, though only 4 mm. long, secrete nectar very abundantly. Hermann
Miiller describes their mechanism as particularly interesting on account of its great
simplicity. There are only 6-12 hairs on the style, instead of a brush. The
stamens closely surround and partly project beyond the stigma, that therefore gets
covered with pollen when the anthers dehisce, which they do in the bud. The
carina is open above throughout its whole length, so that the stigma and anthers pro-
trude when it is depressed. When the pressure is removed, the depressed parts return
to their original position. This is brought about by the elasticity of the alae and
carina, aided by the broad vexillum that ensheaths them; also by the action of
the calyx, which holds together the bases of all the petals. Each ala is united
with the carina only in one place, where there is an ill-marked fold and the epidermal
cells slightly interlock.

Automatic self-pollination regularly takes place, and Hermann Miiller says that
it is thoroughly effective. Insect visitors effect cross- and self-pollination with equal
ease, and, despite the small size of the flowers, their number is considerable. This
is obviously due to the relatively great abundance of nectar. While in other species
the secretion remains concealed between the base of the pistil and the filaments,
it here flows out through the nectar-passage on either side the base of the free
filament, collecting to form a large drop. This extends beneath the vexillum
beyond the level of the calyx, and hence can be seen from the outside.

VisiTors.—These are small bees and Lepidoptera. I saw the honey-bee in
the island of Fohr, and Herm. Miiller noticed it in Westphalia. Herm. Miiller
(H. M.) and Buddeberg (Budd.) also observed the following.—

A. Hymenoptera. (a) Apidae: 1. Andrena convexiuscula X. 6, skg. (H. M.);
2. Halictus flavipes K. 9, do. (Budd.). (2) S¢hegzdae: 3. Ammophila sabulosa Z. 6,
making casual attempts at skg. (H.M.). B. Lepidoptera. Rhopalocera: 4. Coe-
nonympha pamphilus Z., skg. (H.M.); 5. Lycaena aegon W.V., do. (H. M.).

H. de Vries saw a bee, Apis mellifica Z. ¥, in the Netherlands (Ned. Kruidk.
Arch., Nijmegen, 2. ser., 2. deel, 1875).

328 ANGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire a Muscid and several Dolichopodids were recorded (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 48).

771. V. tetrasperma Moench (= Ervum tetraspermum Z.).—The mechanism
of the bright yellow flowers of this species is not, according to Kirchner (‘Flora
v. Stuttgart,’ p. 504), so much simplified as in the last species, but essentially agrees
with that of V. Cracca. The alae possess the finger-shaped processes overlapping
the sexual column, and in front of these there are two outgrowths on either side
which fit into and are loosely united with corresponding carinal depressions. The
upper margins of the carina lie close together; the base of the vexillum ensheaths
the claws of the other petals. The anthers dehisce just before the bud opens, and
some of the pollen falls upon the collecting-brush.

Visirors.—I observed the honey-bee and Halictus sp., skg., at Kiel (‘ Bloemen-
biol. Bijdragen ’).

772. V. pisiformis L. (=Ervum pisiforme Peferm.).—The greenish to yellowish-
white flowers of this species were investigated by Herm. Miiller in Thuringia (‘ Weit.
Beob.,’ II, pp. 258-60). Their mechanism is intermediate between those of V.
sepium and V. Cracca. The style is provided with a very regular brush for almost
half its length from the stigma, and the anthers (which dehisce in the bud) deposit
the greater part of their pollen upon it. The upper margins of the carina are
readily separated, so that when it is depressed the stigma, stylar brush, and all
the stamens protrude. Union between carina and alae is effected very much as
in V. Cracca and V. sepium. But the finger-shaped processes on the bases of the
alar laminae are broader and thicker, triangular in form, and only gradually becoming
slender and flattened towards their tips. They consequently cause the return of
all the parts of the flower to their original position more efficiently than in the
other two species. The vexillum is pressed against the alae by means of two slight
folds situated at the junction of its claw and erect lamina, and diverging to the
front superiorly. Unbidden guests are thus excluded. Since the claw of the vexillum
is 8-10 mm. long, the proboscis of insect visitors must be of the same length.
Many shorter-tongued bees, however, are able to thrust their heads under the claw,
thus reaching the nectar.

Visitors.—Herm. Miiller observed the following.—

A. Hymenoptera. Afidac: 1. Bombus lapidarius Z. ¥, skg.; 2. B. rajellus
K. 9 and §, do.; 3. B. sylvarum L. 9, skg. and po-clig.; 4. Halictus tetragonius
Kilg. 9, po-cltg.; 5. Megachile circumcincta K. 9, skg. and po-cltg.; 6. M. versicolor
Sm. 9, do. B. Diptera. Syrphidae: 7. Syrphus balteatus Deg., hovering and vainly
skg. C. Lepidoptera. hopalocera: 8. Coenonympha arcania Z., skg.

A. Schulz saw flowers perforated by humble-bees.
773. V. sylvatica L. (=Ervum sylvaticum Peferm.).—

Visitors.—In Dumfriesshire 2 humble-bees were recorded (Scott-Elliot, ‘Flora
of Dumfriesshire,’ p. 49). Schulz saw flowers perforated by humble-bees.

774. V. cassubica L. (=Ervum cassubicum Peferm.).—

VisiTors.—Schulz saw flowers perforated by humble-bees.

LEGUMINOSAE 329

775- V. Orobus DC. (=Ervum Orobus K7.).—

Vistrors —MacLeod, in the Pyrenees, saw Bombus mastrucatus Gers/. §, stealing
the nectar by perforation (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 439):

In Dumfriesshire a humble-bee, probably Bombus agrorum /., was observed,
but did not appear to suck (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 50).

776. V. Ervilia Willd. (=Ervum Ervilia Z., and Ervilia sativa Zzni)—The
odourless white flowers of this species are marked with dark violet veins on the
vexillum, and a dark blotch on either side of the carina. Kirchner (‘Flora v.
Stuttgart,’ p. 507) describes the vexillum, which grasps the claws of the other petals
from above, as possessing two processes at the base of its lamina, These are closely
apposed to the parts of the flower beneath them. Each ala bears a finger-shaped
process, in front of which there is a deep fold interlocking with the carina. The
brush on the style is about half its length, and composed of uniform fine hairs.

777. V. onobrychioides L.—

VisiTors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Andrena dorsata K. 9, po-clig.; 2. Bombus agrorum F’. ¥, steadily skg. ;
3. Megachile willughbiella A. 3, skg.; 4. Osmia aenea LZ. 9, po-cltg.

778. V. unijuga A. Br.—

Visrtors.— Loew observed the following in the Berlin Botanic Garden.—

1. Bombus hortorum Z. ¥, skg.; 2. B. lapidarius Z. ¥, do.; 3. B. pratorum A’. ¥,
do.; 4. Megachile circumcincta A’. 9, skg. and po-cltg.; 5. M. willughbiella Z. 3,
skg.; 6. Osmia rufa Z. 9, po-cltg.

228. Lens L.

Nectar-yielding bee flowers with a stylar brush.

779. L. esculenta Moench (=Ervum Lens Z.).—The bluish-white flowers
of this species are marked with blue lines on the vexillum (nectar-guides), and
a small blue patch on the tip of the carina (pollen-guide). Kirchner (‘Flora v.
Stuttgart,’ p. 508) states that the vexillum, which but slightly ensheaths the other
parts of the flower, comes into close contact with the alae by means of two forwardly-
directed folds of a projecting ridge. The style bears collecting hairs only on its
inner side. Otherwise the mechanism agrees with that of Vicia Ervilia. Kerner
says that the flowers are fertile when insect-visits are prevented.

VisiTors.— Herm. Miiller saw the honey-bee, skg., and a butterfly (Coe-
nonympha pamphilus Z.), do. (‘Weit. Beob., II, p. 258).

229. Pisum L.

Nectar-containing bee flowers with a brush arrangement transitional to a
pumping one.

780. P. sativum L.—The mechanism of the white flowers of this species has
been very fully described by Herm. Miiller (‘ Fertilisation,’ pp. 211-14). The strong
sickle-shaped carina is produced into a leaf-like outgrowth along the junction of
its two petals, and is thus strengthened. The alae and carina are very firmly united
with one another, and with the sexual column. At the base of each alar lamina

330 ANGIOSPERMAE—DICOTYLEDONES

there is a deep involution directed forwards and downwards, and fitting into a
corresponding pit in the upper surface of the adjacent carinal petal. The epidermal
cells are here so closely interlocked that it is scarcely possible to separate the carina
and alae without tearing them. There is also an anterior alar fold, which lies in
a carinal depression. The upper side of the base of the vexillum possesses two
deep but narrow indentations which project ventrally as firm ridges. These diverge
in front and fit into the anterior alar folds. The base of each carinal petal expands
into an upwardly and inwardly directed lobe, which rests on the sexual column.

FiG. 107. Prsum sattvum, L. (after Herm. Miller). (1) Flower after removal of the left ala, seen
from the left. (2) Left ala seen from within. -(3) Vexillum seen from within, (4) Carina seen from above
(further enlarged). (5) The same surrounded by the alae; anterior part omitted (still further enlarged).

(6) Basal part of the left ala, seen from the outside. (7) The sexual organs removed from the bud. (8: Upper
part of the style, seen from within (x 7). (9g) A stamen. a, apical carinal swelling; 6, median carinal
outgrowth; c’ alar involution, fitting into carinal pit (c); @’, anterior alar fold, resting in carinal
depression (d@), and overlaid by vexillary ridge (@”); ¢, basal carinal lobe; e’ posterior alar process ;
JF, alar plate, held in place by rounded vexillar swelling (/’); 4 tip of carina; sé, stigma; 0, stylar
opening.
It is kept in place by a posterior alar process, directed backwards and inwards.
These processes, in their turn, are grasped by narrow alar plates which project
horizontally from them, and are backwardly directed. A rounded swelling on the
very broad and firm vexillar base presses against each of these plates.

The style runs up vertically from the end of the horizontal ovary; its end
is curved inwards to such an extent that the terminal stigma faces almost directly
downwards towards the base of the flower. The inner side of the style is beset
with long bristly hairs for almost half its length. The tip of the carina is also
directed towards the base of the flower, and dilates into a pair of swellings which
enclose the anthers in the bud. The conical space thus bounded possesses an

opening through which the style can protrude.

LEGUMINOSAE 331

Towards the end of the bud period, the anthers dehisce, filling the conical
space at the end of the carina with pollen; at the same time the filaments are
retracted. The stigma and stylar brush get covered with pollen, some of which
is squeezed out at the tip of the carina when this is depressed. When the parts
of the flower resume their original position the edges of the stylar aperture scrape
off the pollen, which then of course remains outside the carina. The ends of the
filaments lying in the lower part of the conical anther-chamber are somewhat
thickened and club-shaped after the anthers have dehisced. When the carina is
depressed they push the pollen before them, so that the supply on the stylar brush
is constantly renewed.

The interlocking and union of the parts of the flower are advantageous, because
nectar-seeking insects are compelled to apply the necessary force in order to
operate the mechanism by which the stigma is rubbed against the pollen-covered ventral
surface of a visitor, and this is dusted afresh. The firm union of parts also secures
return to their original position when the pressure is removed. Besides which, access
to the nectar is only permitted to a select number of powerful insects. Bees pos-
sessing the requisite strength, however, are scarce, and visitors of this species are
consequently few in number. Although the majority of flowers remain unvisited,
they are just as fertile when self-pollinated as when crossed (Ogle, Miiller, Kerner).

Visrtors.—In the course of four summers, Herm. Miiller only observed 3 bees :—
1. Eucera longicornis Z.; 2. Halictus sexnotatus A. 9, skg. and po-cltg., 6, skg.;
3. Megachile pyrina Le., do.

A. Schulz noticed flowers perforated by humble-bees.

Alfken saw 2 bees, skg., at Bremen.—1. Anthidium manicatum F. 9; 2. Mega-
chile maritima A’ 3.

230. Lathyrus L.

Nectar-containing bee flowers with a stylar brush.

781. L. pratensis L. (Delpino, ‘Ult. oss.,’ pp. 55-9; Herm. Miller, ‘ Ferti-
lisation, pp. 207-10, ‘ Weit. Beob.,’ II, p. 257, ‘Alpenblumen,’ p. 249; Schulz,
‘Beitrage,’ II, p. 211; Lindman, ‘ Bidrag till Kanned. om Skandin. Fjellvaxt. Blomn.
o. Befrukt.’; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 362-4; Loew,
‘Bliitenbiol. Floristik,’ p. 395; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ p. 201, ‘Weit.
Beob. ti. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 233.)—The mechanism of the yellow
flowers of this species, which was first described by Delpino, and afterwards more
fully by Hermann Miller, is similar to that of Pisum. The union of alae and carina
with the sexual column is effected by two long vesicular alar processes. These
tun backwards, and rest upon the column, their tips meeting in the middle line.
Their elasticity brings about the return of the carina to its original position, on
removal of the pressure exerted by an insect visitor. On either side the end of the
carina there is a sacculation, bounded behind by a deep fold, and opening only at
the carinal tip. These two sacculations jointly enclose all the anthers in the bud;
dehiscence takes place at the beginning of anthesis. The style is almost vertical,
and expands beneath the oval terminal stigma into an elongated oval plate. The
inner side of this is beset with short hairs directed obliquely upwards, which sweep

332 ANGIOSPERMA E— DICOTYLEDONES

the pollen clinging to them through the tip of the carina on to the ventral surfaces
of bees. This pollen dusts the projecting stigma of the next flower visited, so
that crossing is effected. Although the stigma is surrounded by the pollen of the
same flower, automatic self-pollination apparently does not take place, the stigmatic
papillae needing to be rubbed (by visiting bees) before they become receptive.

Visitors.—I saw the following bees in the North Frisian Islands.—1. Apis
mellifica Z. §. skg.; 2. Bombus derhamellus X., do.; 3. B. terrester Z., do.; and also
B. agrorum F’. 9, do. (Riigen).

Fic. 108. Lathyrus pratensis, L. (after Herm. Miller). (1) Flower seen from the side (slightly
enlarged). (2) Flower after removal of calyx and vexillum, seen from above (further enlarged). (3) Left
ala, seen from within. (4) Flower after removal of vexillum and alae, seen from above. (5) Bud shortly
before opening, after removal of calyx, vexillum, and alac; seen from the side. (6) The anterior part
of the same, seen from above. (7) Style with brush and stigma, seen from the side. (8) The same,
seen from within. a, alar sac, lying in a deeper pouch (a’) of the carina; 6, process of alar margin,
directed forward and downward: it interlocks with the narrowest part of the pocket-like carinal sac ;
¢, transverse depression in the ala, immediately behind the anterior dark yellow lobe, an angular
ridge-like vexillar fold (0) is closely connected with it; d, posterior process of the upper alar margin;
¢, alar claw; /, margin of the ala folded over; gg, margins of the carina; 4, uppermost filament;
7, fused filaments; &, nectar passages, with basal nectaries: /, carinal claws; 2, place at which the
tip of the style and the stigma project, when the carina is depressed; 7, leaf-like expansion along the
line of fusion of the carinal petals; 9, ridge-like processes of vexillum; #, terminal sacculation of
the carina.

Alfken observed the following bees at Bremen.—

1. Bombus arenicola 7hs. ¥; 2. B. derhamellus A. 9 and ¥; 3. B. distinguendus
Mor.g and ¥; 4. B. lapidarius Z.¥; 5. B. lucorum Z. ¥; 6. B. muscorum /. 9;
7. B. sylvarum LZ. 9; 8. Eucera difficilis (Duf.) Per. 9 and 6; 9. Megachile circum-

cincta A. 8; 10. M. willughbiella A”. 9.

Loew noticed Diphysis serratulae Pz. 9, po-cltg., in Steiermark (‘Beitrage,’
P- 53):

Herm. Miiller (H. M.) and Buddeberg (Budd.) noticed the following.—

1 Bombus agrorum 9, in large numbers, skg. (H. M., Thuringia); 2. Diphysis
serratulae Pz. 9 and 4, skg. (H.M., Budd.); 3. Eucera longicornis Z. 9, do. (H.M.,

Budd.); 4. Megachile maritima A. 4, do. (H.M.); 5. M. versicolor Sm., skg. and
po-cltg. (H. M.).

LEGUMINOSAE 333

Herm. Miiller saw a humble-bee and a Lepidopterid in the Alps, and von
Dalla Torre in the Tyrol noticed the bee Xylocopa violacea Z. 9. Schletterer
records the latter for the Alps, and Polistes gallica Z. for Pola.

Lindman saw a Lepidopterid in Scandinavia ; and MacLeod observed Bombus
sylvarum Z. 9 and ¥, in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 364).

Apis, 3 humble-bees and a saw-fly have been recorded for Dumfriesshire (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 51).

Only bees are able to effect pollination. Lepidoptera are no doubt able to
suck the nectar by means of their slender proboscis, but cannot liberate the flower
mechanism.

Schulz observed flowers perforated by humble-bees.

782. L. maritimus Bigel. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 64-5,
153, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 233.)—I was originally of
the opinion that in this species self-fertilization is possible even in the bud. It
appears, however, that though the stigma is from the first surrounded with the pollen
of the same flower, cross-pollination is ensured by insect-visits. For the said
pollen does not adhere to the stigma, which only becomes receptive after an insect
has rubbed against it.

The mechanism essentially agrees with that of L. pratensis. The large brightly
coloured blossoms are associated in racemes of 5-8 flowers. The erect vexillum
is violet in colour with darker veins; it is 2 cm. long, and its upper part about
14cm. broad. At the junction of claw and lamina there is a process 3 mm. long,
which fits into corresponding alar folds, so that complete union is effected. The
violet alae (x cm. long, with lamina 5 mm. broad) interlock in their turn with
carinal depressions. They escape from the vexillum when depressed by insect
visitors, diverging so that the stigma first protrudes, and afterwards the pollen-
covered stylar brush. The two alar processes interlock so firmly with corresponding
depressions in the carina that they are not separated. during insect-visits. Hence,
when the pressure is removed, the margins of the alae resume their original position,
and this movement is furthered by the stiff claws of the carina. The latter is bent
almost at right angles; externally it is bright violet in colour, but white elsewhere ;
its boat-shaped region is 8 mm. long, and its claws of about the same length. The
two carinal petals are fused along the whole of their ventral margins; above they
gape a little, but are covered by the overlapping alae. The staminal tube is about
1 cm. long, and the free part of the filaments about the same length.

Pollination is effected by long-tongued bees (humble-bees). These hold on
to the flower in the way already described, and suck the nectar secreted at the
usual place. Lepidoptera play the part of nectar-thieves, and do not operate
the flower mechanism. I observed perforated flowers in the Island of Féhr; there
was a hole in the alar claw probably made by short-tongued humble-bees.

Vistrors.—In Féhr and Sylt, I saw more particularly 5 species of humble-bee
and 3 Lepidoptera, the latter being unbidden guests.
Loew observed the following bees in the Berlin Botanic Garden.— 1. Bombus

agrorum F.9, skg.; 2. B. hortorum LZ. ¥, do.; 3. B. lapidarius Z. ¥, do.; 4. B.
pratorum Z. ¥, do.

334 ANGIOSPERMAE—DICOTYLEDONES

Schneider saw 2 bees (Bombus nivalis Dako. % and 4, and B. alpinus L.9%
and 6) in Arctic Norway (Mus. Aarsh. Tromse, xvi, 1894).

783. L. sativus L.—Kirchner gives the following account of the bright blue
or white flowers of this species (‘Flora v. Stuttgart,’ pp. 511-12). The claw of the
large vexillum grasps the bases of the alae from above only. It is, however, attached
to them very firmly, for its base possesses two pairs of folds placed almost at right
angles to one another, and projecting inwards, so as to fit closely into corresponding
depressions of the alae. The front edge of the carina is strengthened by a wing-like
appendage, and is curved like an S, in such a way that its end lies somewhat to
the left. The tip of the right carinal petal is arched outwards, while the left one
has a deep terminal fold, in front of which the style lies in the carina. As in Pisum,
the alae are firmly united with the carina; the right ala, however, where it lies
above the carinal tip, possesses a dilated fold contracting from above downwards,
through which the end of the style, with its small stigma, projects when the carina
is depressed. The style broadens above, and is compressed from before backwards,
but it is turned through an angle of go° in such a way that its morphological inner
side, which bears collecting-hairs directed obliquely upwards, is directed to the left,
and its hairless outer surface to the right. The anthers dehisce in the bud, and dis-
charge their pollen into the stylar brush, by which it is transferred to insect visitors.

VisiTors.—Kirchner observed the honey-bee on cultivated plants in Wurtemberg.
When it settled in the middle of the flower, it was dusted with pollen on its right
side behind the head; it regularly effected cross-pollination. It frequently stole
nectar by thrusting its proboscis into the right side of the flower, occasionally
touching the style with its feet.

784. L, sylvestris L. (=L. pyrenaicus Jord.). (Delpino, ‘ Ult. oss.,’ pp. 57-8 ;
Herm. Miiller, ‘Fertilisation’ p. 210; Kirchner, ‘Flora v. Stuttgart, p. 512;
MacLeod, Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 439.)—Delpino describes the
flower of this species as asymmetrical, with an oblique stylar brush; but the
asymmetry is less pronounced than in L. sativus. Kirchner says that the vexillum
(which is of a rose-purple colour, greenish externally) has a longer claw than
V. sativus, but is devoid of inwardly directed folds. Here too the greenish carina
is twisted, leaving a narrow nectar-passage on the right side, that is regularly
used by the honey-bee when stealing nectar, in doing which it only now and then
touches the stigma and pollen with its legs.

VisiTors.— Besides the honey-bee, Kirchner observed (Wurtemburg) Lepi-
doptera, as unbidden guests.

MacLeod saw 3 species of humble-bees (skg. legitimately), and a Lepidopterid
in the Pyrenees. Delpino (‘ Ult. oss.,’ I) noticed more particularly Xylocopa, Apis,
Bombus, Eucera, and Anthophora; and Loew saw Apis, skg., in the Berlin Botanic
Garden. Herm. Miiller also observed the honey-bee (skg. and po-cltg.), as well
as the following Lepidoptera, all skg., but conferring no benefit :—Pieris rapae Z.,
Plusia gamma Z., Rhodocera rhamni Z., Vanessa io Z., V. urticae Z.

Alfken observed the following bees at Bremen.—

1. Bombus agrorum /. 9; 2. B. derhamellus A.; 3. B. hortorum Z. 9 and ¥;
4. B. sylvarum LZ. ¥; 5. Megachile centuncularis Z. 9, po-cltg.; 6. M. circumcincta
&. 9, do.; 7. M. maritima A’. 9, do.; 8. Trachusa serratulae Pz. 9, do.

LEGUMINOSAE 335

785. L. tuberosus L.—The purple-red, fragrant flowers of this species are
markedly asymmetrical, and according to Kirchner (‘ Flora v. Stuttgart,’ p. 511) their
mechanism agrees with that of L. sylvestris, even as regards the torsion of the
carina and style.

VisiTors.—Herm. Miiller observed the following (‘ Fertilisation,’ p. 210, ‘ Weit.
Beob.,’ II, p. 257).—

A. Hymenoptera. Apidae: 1. Apis mellifica Z. 9, skg. and po-cltg.

B. Lepidoptera. Rhopalocera: 2. Hesperia sp., skg.; 3. Lycaena damon S.V., do.;
4. Pieris rapae Z.,do. C.Thysanoptera. 5. Thrips, freq.

Schletterer and von Dalla Torre (Tyrol) saw the bee Halictus sexcinctus For. 9 ;
and Loew (Berlin Botanic Garden) 2 long-tongued bees (Megachile circumcincta
K. 9, skg. and po-cltg.; M. fasciata Sm. 3), skg. Schulz, in Central Germany, noticed
flowers perforated by humble-bees.

786. L. heterophyllus L.— :

VisiTorSs.—Schulz (Central Germany) saw flowers perforated by humble-bees.

787. L. palustris L.—The flowers of this species are asymmetrical. Their
mechanism has been fully described and illustrated by Heinsius (Bot. Jaarb.
Dodonaea, Ghent, iv, 1892, pp. 91-4).

Visitors.—Heinsius, in Holland, observed 2 humble-bees, Bombus agrorum
f., skg. legitimately, and B. scrimshiranus X. (proboscis only 9-10 mm. long), po-cltg.;
also a butterfly Hesperia sylvanus Zsp. 6, perhaps an unbidden guest.

Schulz (Central Germany) saw flowers perforated by humble-bees (loc. cit.).

788. L. latifolius L.—

Visitors.—Schenck (Nassau) saw the leaf-cutting bee Megachile maritima X.;
von Dalla Torre and Schletterer (Tyrol) the bee Nomada lineola Pz.

Loew observed the following bees and butterflies (‘ Beitrage,’ p. 34).—

(2) In Silesia.—1. Apis mellifica Z. ¥, trying to suck; 2. Megachile maritima
K. 9, po-cltg.; 3. Xylocopa violacea Z. 9, skg.; 4. Rhodocera rhamni Z., trying to
suck. (4) In the Berlin Botanic Garden.—1. Apis mellifica Z. ¥, skg. through holes
made by humble-bees; 2. Bombus terrester Z. 9, vainly trying to suck from the
outside; another 9 was biting holes with its mandibles immediately above the calyx ;
3. Megachile fasciata Sm. and 9, skg.; 4. Vanessa cardui Z.,skg. (c) Ditto, on the var.
ensifolius.—1. Megachile fasciata Sm. 9, skg. and po-cltg.; 2. Colias rhamni Z., skg. ;
3. Pieris brassicae Z., do. (d) Ditto, on the var. intermedius.—1. Bombus sylvarum
L. 9, skg. legitimately and po-cltg.; 2. Megachile fasciata Sm. 9, do.; 3. Lycaena
bellargus Jeof/., skg.; 4. Pieris brassicae Z., do.

Plateau noticed the following Hymenoptera.—

1. Bombus muscorum F.; 2. B. terrester Z.; 3. Eucera longicornis Z.; 4. Mega-
chile ericetorum Zep.; 5. Odynerus quadratus Pz.; 6. Stelis sp.

789. L. luteus Gren. (=Orobus luteus Z.).—The flowers of this species are
at first yellow, but towards the end of anthesis take on a fiery red colour. According
to MacLeod’s observations in the Cottian Alps, their mechanism is the same as
that of L. pratensis.

Vistrors.—MacLeod saw a species of humble-bee.

790. L. montanus Bernh. (=L. macrorrhizus Z., and Orobus tuberosus
Wimm.).—The flowers of this species are at first rose-red, then lilac, and finally of a
faded brown colour.

336 ANGIOSPERMAE—DICOTYLEDONES

Kirchner (‘Flora v. Stuttgart,’ p. 513) describes their mechanism as agreeing
almost completely with that of L. pratensis, but the end of the style is somewhat
broader.

Visttors.—Alfken observed the following bees at Bremen.—

1. Andrena convexiuscula K. 9 and 4; 2. A. xanthura X. 9, skg. and po-cltg. ;
3. Coelioxys quadridentata Z. 9, sk.; 4. Halictus nitidiusculus A. 9; 5. H. punctatis-
simus Schenck 9; Megachile circumcincta &. 6.

Schmiedeknecht saw the humble-bees Bombus hortorum JZ. 9, and B. mastrucatus
Gerst. 9, in Thuringia. MacLeod noticed 2 humble-bees in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, pp. 434, 440). Loew observed 3 bees in the Berlin
Botanic Garden.—1. Anthophora pilipes F. 9, skg.; 2. Bombus agrorum /”. 9, do. ;
3. B. lapidarius Z. ¥, do. In Dumfriesshire 2 humble-bees were recorded (Scott-
Elliot, ‘ Flora of Dumfriesshire,’ p. 52).

Schulz noticed flowers perforated by humble-bees.

791. L. odoratus L.—The flowers of this species smell like honey.
Visirors.— Herm. Miiller saw a bee (Anthidium manicatum Z.), skg., in
Strasburg.

792. L. niger Bernh. (=Orobus niger Z.). (Knuth, ‘Bloemenbiol. Bij-
dragen.’)—The flowers of this species studied by me in Schleswig-Holstein possess
a mechanism agreeing with that of L. pratensis. The vexillum is purple-red in
colour, with a darker nectar-guide ; it is about 10 mm. broad, and rises to a height
of 8mm. The tips of the alae are of a bluish-violet; the alar and carinal claws
are colourless. The processes and depressions of the alar laminae and carina are
even better developed than in A. pratensis; but there is hardly any interlocking
of the epidermal cells, so that these petals can easily be separated. The distance
from the entrance of the flower to the nectar is 7mm. The petals become dis-
coloured after anthesis.

Visitors.—I saw Bombus agrorum /., skg. Loew observed 3 bees in the
Berlin Botanic Garden.—1. Bombus agrorum F. 9, skg.; 2. B. lapidarius Z. 9, do. ;
3. Osmia rufa LZ. 9, skg. and po-cltg.

MacLeod noticed Bombus agroruin /. 9 in the Pyrenees (Bet. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 439). Bombus terrester Z. also visits the flowers. It bites
through the base of the vexillum from above, or at the tip, and steals nectar through
the opening, which is sometimes 4 mm. long and 2 mm. broad.

Schulz (Central Germany) also saw flowers perforated by humble-bees.

793. L. variegatus Ten.—

Visitors.—Loew noticed 2 long-tongued bees in the Berlin Botanic Garden.—
1. Bombus hortorum Z. ¥, skg.; 2. Osmia rufa Z. 9, skg. and po-cltg.

Schenck saw 2 bees in Nassau.—1. Megachile ericetorum Lep.; 2. Xylocopa
violacea L.

Schulz (Central Germany) noticed flowers perforated by humble-bees.

794. L. setifolius L.—Kiefer states that this species bears cleistogamous
flowers.

LEGUMINOSAE 337

795. L. vernus Bernh. (=Orobus vernus L.). (Knuth, ‘Bloemenbiol. Bij-
dragen.’)—The flower mechanism of this species is almost exactly like that of
L. pratensis. The vexillar lamina is 12 mm. broad and 6 mm. long, and dark
purple-red in colour, with delicate darker markings. The vexillar claw is 10 mm.
long, white at the base, and almost enclosed by the calyx, like the remaining parts
of the flower. The exposed portions of the vexillum and carina are violet in colour,
while the parts enclosed by the vexillar claw and calyx are white. The union
between carina and alae is tolerably firm, but with care they can be separated
from one another without tearing. The style gradually tapers upwards; its brush
is 3 mm. long.

The anterior parts of the petals become blue towards the end of anthesis.

Visrrors.— Schmiedeknecht mentions Bombus mastrucatus Gers‘. 9 for
Thuringia, and Hoffer records it for Steiermark.

I observed 3 humble-bees at Kiel (2. 5.’96).—1. Bombus hortorum ZL. 9, freq.,
skg. legitimately, and diligently flying from flower to flower; 2. B. lapidarius Z. 9,
occasionally do.; 3. B. terrester Z. 9, perforating the vexillar claw just between the
two upper calyx-teeth, and thus stealing nectar. Schulz also observed flowers
perforated by humble-bees.

Loew, in the Berlin Botanic Garden, observed Bombus hortorum J. 9, skg.,
and B. terrester Z. 9, perforating the flowers; also, on the var. flaccidus A7z/., Osmia
rufa Z. 9, skg. and po-cltg.

796. L. Aphaca L.—The bright or dark yellow odourless flowers of this
species possess nectar-guides in the form of dark streaks on the vexillum. Kirchner
(‘Flora v. Stuttgart,’ p. 514) states that their mechanism agrees with that of L.
pratensis, except that the style enlarges but slightly and gradually towards its end.

797. L. Nissolia L.—Kirchner (‘Flora v. Stuttgart,’ p. 515) says that the
rather small carmine-red flowers of this species frequently do not open at all,
but nevertheless set healthy fruits, being therefore cleistogamously fertilized.

798. L. grandiflorus Sibth. et Sm.—This species is very rarely visited by
insects in England. The flowers are more fertile when shaken (Darwin, Ann. Mag.
Nat. Hist., London, Ser. 3, ii, 1858, p. 459).

Visrtors.—Loew observed a bee (Megachile fasciata Sm. 9), po-cltg. and skg.,
in the Berlin Botanic Garden. : .

799. L. brachypterus Alef.—
Visrtors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Bombus agrorum F. ¥, skg.; 2. B. lapidarius Z. ¥, skg. and po-cltg.; 3. B.
terrester Z. 4, trying to perforate just above the calyx; 4. Megachile centuncularis
L. 9, skg. and po-cltg.; 5. M. fasciata Sm. 6 and 9, em copuld on the flowers,
the 9 previously skg. and po-cltg.

800. L. cirrhosus Ser.—
Vistrors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Bombus agrorum /. ¥, skg.; 2. B. hortorum Z. ¥, skg. and po-cltg.; 3. B.
lapidarius Z. ¥, do.; 4. Eucera longicornis LZ. 9, do.

DAVIS ik Zz

338 ANGIOSPERMAE—DICOTYLEDONES

8o1. L. incurvus Roth.—

Visitors.—Loew (Berlin) saw a humble-bee (Bombus agrorum /. ¥) pressing
the alae together normally, skg. and po-cltg.

802. L. rotundifolius Willd.—

VisiTors.— Loew (Berlin) observed 2 bees :—1. Bombus agrorum J”. 9, inserting
its proboscis laterally under the vexillum; 2. Megachile fasciata Sm. 3. skg.

231. Orobus L.
803. O. aureus Stev.—

Visttors.—Loew (Berlin) saw 2 humble-bees. — 1. Bombus agrorum #”. 9, skg. ;
2. B. lapidarius LZ. 9, do.

804. O. hirsutus L.—

Visirors. — Loew (Berlin) saw 2 bees.—1. Bombus agrorum Ff. 9, skg.;
2. Megachile fasciata Sm. 8, do.

805. O. Jordani Tenore.—

Visitors.—-Loew (Berlin) saw a humble-bee (Bombus hortorum Z. ¥), skg. and
po-cltg.

232. Erythrina L.

806, E. Crista-galli L. (Delpino, ‘ Ult. oss., pp. 64-8; Herm. Miiller, ‘ Weit.
Beob.,’ II, p. 264.)—The flower of this Brazilian species is twisted through an angle
of 180°, so that the large vexillum is directed downwards, and serves as an
alighting-place and platform for visitors. The stamens and style are enclosed by
the upwardly directed carina, the lower part of which is expanded into a nectar-
reservoir. The alae are only represented by two vestigial laminae. Cross-pollination
is favoured by the fact that the stigma projects a little beyond the anthers.

Visitors.—Delpino supposes that humming-birds are the pollinating agents.

807. E. velutina. (Delpino, op. cit.)—In this species the alae and carina
are reduced to minute vestiges, so that the column of sexual organs lies free under
the vexillum.

Visttors.—Delpino believes these to be bees that must thrust in their heads
between the sexual column and the vexillum to reach the nectar, which is secreted as
in papilionaceous flowers generally.

233. Glycine L.

808. G. chinensis Sims. (= Wistaria chinensis DC.).—The large blue flowers
of this species are associated in many-blossomed racemes.

Vistrors.—Herm. Miiller observed the following bees at Strasburg (‘ Weit.
Beob.’ II, p. 263).—

1. Anthidium manicatum Z. é, skg.; 2. Anthophora personata ///. 9 and 4, do.;
3. Megachile willughbiella A. 6, do.; 4. Osmia aenea Z. 9, do.; 5. O. rufa Z. 9, do.

Loew saw the honey-bee, skg., in the Berlin Botanic Garden.

Schletterer records the widely distributed carpenter-bee Xvylocopa violacea Z.
for the Tyrol.

LEGUMINOSAE 339

234. Phaseolus L.

Nectar-yielding bee-flowers with a stylar brush. The spirally twisted end
of the style, with the stigma and the pollen adhering to the brush, projects from
the tip of the carina (which is also spirally twisted) when this is depressed, return-
ing again when the pressure is removed. Delpino (‘ Ult. oss.,’ p. 55) says that’
in some species the style is twisted to the right, in others to the left, and that all
intermediate stages exist between a simple falciform curve (P. angulosus, &c.) and
a helicoid spiral of 4 to 5 turns (P. Caracalla).

809. P. vulgaris L. (Herm. Miller, ‘Fertilization,’ pp. 216-17; Kirchner,
‘Flora v. Stuttgart,’ pp. 515-16; Knuth, ‘Bloemenbiol. Bijdragen.’)—The flower
mechanism of this species was first described by Darwin (Gard. Chron., London,
1857, p. 725, 1858, pp. 824-44, Ann. Mag. Nat. Hist., London, Ser. 3, ii, 1858,
pp. 459-64). He also demonstrated by experiment that insect-visits are necessary #

'

ee es

——

Fic. 109. Phaseolus vulgaris, L. (1) Flower scen obliqucly from the front (from nature). (2) Pistil
enlarged (after Herm. Miller). ¢, ovary; 7, style; g, stylar brush; 4, stigma.

for fertilization. The bean, however, is able to fertilize itself with complete effect.
Darwin’s account is here substantially repeated. The left ala is larger than the right.
The base of the alar lamina is contracted, and bears an oblique tooth-like process,
firm and succulent, which fits into a carinal pit. On the lower third of the ala,
internally, there is a crescentic fold, adapted to a corresponding groove in the carina.
The latter is small, and the opening at its tip faces downward, lying above the tooth-
like process of the right ala. The end of the style is somewhat expanded, and the
oblique stigmatic surface is beset with a dense fringe of short hairs. This not only
prevents the proboscis of an insect while being withdrawn from coming into contact
with the stigma of the same flower, but also prevents the escape of the stigmatic
fluid, which is abundantly secreted by the stigmatic papillae when torn by contact
with the rough bodies of insects.

The anthers surround the style and shed their pollen upon it, but the stigma is
never dusted. The upper free filament broadens out so much immediately in front
of the two nectar-passages that it grasps the edges of the staminal tube, completely
closing it. There is here an oblique scale-like appendage directed upwards and

ae

340 ANGIOSPERMAE—DICOTYLEDONES

forwards. Insects can only obtain nectar legitimately by alighting upon the left ala
and pushing their proboscis under the opening on the right side of the carinal tip.
To prevent access in any other way, there is here an upwardly and forwardly directed
scale-like appendage.

Only large humble-bees are able to work the flower mechanism. When the
carina is depressed the end of the style with its pollen-covered brush springs out of
the opening in the carina, and a narrow canal appears, beginning just below the
carinal opening, and running past the end of the style, along the right margin of
the groove in the staminal column as far as the base of the nectary. The upper
free stamen retains its position, while the nine united ones are bent downwards.
As the stigma is touched by the proboscis of an insect visitor before the pollen, it
follows that cross-pollination is regularly effected by visits. Automatic self-pollination
is excluded: unvisited flowers remain infertile. as already stated.

Visitors.—I saw, at Kiel, Bombus hortorum Z. 9, skg. legitimately. In spite
of constant watching, I have but seldom observed insect-visits, and automatic self-
pollination takes place in the large majority of cases, though Darwin was of the
contrary opinion (vzde supra). Some humble-bees obtain the nectar by perforation :
I observed Bombus terrester Z. as a nectar-thief at Kiel.

Alfken saw a bee (Megachile maritima A. 9) at Bremen, and another (Osmia
maritima Friese 9) in Juist. Leege observed (Juist) a bee (Osmia maritima Friese 9),
freq., skg. and po-cltg., a hawk-moth (Deilephila galii Ro//.), and a Noctuid (Chari-
clea umbra H//n.) very freq.

810. P. multiflorus Willd. (Herm. Miller, ‘ Fertilisation,’ p. 217; Kirchner,
‘Flora v. Stuttgart,’ pp. 575-6; Knuth, ‘Bloemenbiol Bijdragen.’)—The flower
mechanism of this species agrees completely with that of P. vulgaris, and was first
described by Farrer (Ann. Mag. Nat. Hist., London, Ser. 4, ii, 1868, pp. 256-60).

According to Ogle (Pop. Sci. Rev., London, ix, 1870, p. 166), the flowers are
infertile when bees are excluded, but Kirchner says that they can produce perfect
fruits by self-fertilization. The honey-bee and other small bees which are too feeble
to depress the carina, use the holes made by Bombus terrester Z. in the calyx for
stealing nectar. Hermann Miiller states that more powerful bees, with a sufficiently
long proboscis, alight upon the left ala and first touch the stigma with the base of
their proboscis, when they thrust it into the base of the flower, and thus effect cross-
pollination as in P. vulgaris. As the interlocked alae and carina are pressed down-
wards, the spirally twisted end of the style protrudes from the tip of the carina, which
is also coiled into a spiral of two turns. The stigma faces downwards and to the left,
while the pollen-covered stylar brush rubs against the base of the bee’s proboscis,
and dusts it afresh. When insects visit the flowers cross-pollination is therefore
assured, and self-pollination prevented. Hermann Miiller states that automatic
self-pollination is impossible, as the stigma projects from the tip of the carina, while
the pollen is enclosed within it.

VisiTors.—I several times observed (at Kiel) Bombus hortorum Z. 9 as an
invited guest. Schletterer records Eucera longicornis Z. for the Tyrol.

LEGUMINOSAE 341

235. Apios Moench.

8u. A. tuberosa Moench.—<According to Loew’s investigations (Flora, Mar-
burg, Ixxiv, 1891) this species bears Lepidopterid flowers in which the carina,
pp. 160-71, is fixed, and there is no mechanical union between it and the alae. The
protrusion of anthers and stigma from the carina, which is the usual way of dusting
the ventral surface of visitors and effecting pollination, is thus rendered impossible.
In place of this mechanism the flower has acquired another method of securing cross-
pollination, by the remoteness of stigma and anthers from one another, and besides,
the nectar passages are shorter and more accessible, so that insects can suck more
easily. (Cf. p. 261.)

236. Alhagi Tourn.

812. A. camelorum Fisch.—

Visrtors.—Morawitz saw the Buprestid Sphenoptera karelini Fa/derm. in the
Caucasus.

2. SuB-oRDER CaESALPINIACEAE R. Br.

Flowers bilaterally symmetrical (zygomorphous). Scarcely or not at all papi-
lionaceous. Petals 5, sometimes entirely or partly absent; stamens ro, frequently
fewer or more (2-15), free or united in various ways; some of them may be sterile.

237. Gleditschia 1.

Inconspicuous green flowers which secrete nectar abundantly in the hollow
of the calyx. The hermaphrodite flowers are protogynous.

813. G. triacanthos L.—Kirchner (‘Neue Beob. i. d. Bestaubungseinricht.
einheimisch. Pfl.,’ Stuttgart, 1886) says that the fragrant richly nectariferous flowers
of this species are monoecious-polygamous, and possibly dioecious. The four green
sepals and petals are united below into a cup, which secretes nectar abundantly
on its inner surface. It is protected by hairs on the bases of the stamens. The
hermaphrodite flowers are protogynous. The hairy elongated ovary bears a large
terminal stigmatic cushion, which projects several millimetres from the flower, while
the stamens are still enclosed in the perianth. The male flowers usually possess
5-7 projecting stamens, and there is no trace of a pistil. The female flowers bear
stamens with vestigial anthers.

Visttors.—Kirchner noticed numerous insects, especially bees.

238. Cercis L.
814. C. Siliquastrum L.—
Visrrors.—Loew noticed honey-bees, skg., in the Berlin Botanic Garden.

239. Parkinsonia.
815. P. aculeata L.—Lanza (‘Note di biologia fiorale,’ 1894) states that this
species is dichogamous in the Palermo Botanic Garden. The vexillum of the
fertilized flower changes colour.

Visrrors.—Lanza observed Xylocopa cyanescens Bru/l.

342 ANGIOSPERMAE—DICOTYLEDONES

240. Cassia L.

816. C. marylandica L.—In this species the anthers do not dehisce, but
remain covered by a thin membrane, which is ruptured by humble-bees (Nature,
London, xxxv, 1886).

XXXIV. ORDER ROSACEAE JUSS.
(Including Drupacear DC. (=AmyGpacaceak Juss.) and Pomacgae Lindl.)

Literature.—Hermann Miller, ‘ Fertilisation,’ pp. 221-43; Knuth, ‘Bl. u.
Insekt. a. d. nordfr. Ins.,’ pp. 65-6, ‘ Grundriss d. Bliitenbiol.,’ pp. 51-2.

In this order the flowers are made conspicuous in very different ways in the
various genera: there are numerous transitions from the insignificant little blossoms
of Alchemilla to the large ones of roses, which are conspicuous from a distance.
The inflorescences are equally varied, even in the species of the same genus:
sometimes the flowers are solitary or in pairs (species of Mespilus, Cydonia, Dryas,
Geum, Rosa, and so forth); sometimes they are aggregated into more or less
copiously branched inflorescences—-cymes, capitula, racemes, umbels, or panicles
(Spiraea, Crataegus, Pyrus, Sorbus, Alchemilla, Sanguisorba, Amygdalus, Prunus,
Potentilla, Agrimonia, and so on). Many secrete nectar from an annular ridge
on the inner surface of the receptacle. The quantity varies greatly: from large drops
(e. g. in species of Rubus and in Geum rivale), to a scarcely perceptible layer, which,
however, is eagerly licked by insects (Alchemilla and Potentilla); and there are
also intermediate stages. Some species are quite nectarless, others are even anemo-
philous. Our native forms, therefore, belong to the following flower classes.—

An: Sanguisorba minor.

Po: Rosa, Ulmaria, Aruncus, Kerria.

E: Alchemilla, Sibbaldia, Amelanchier vulgaris.

EC: Amygdalus, Prunus, Geum, Potentilla, Spiraea, Crataegus.
C: Rubus, Comarus, Sorbus, Fragaria, Persica.

It follows from what has been said that the visitors of the various species are
very diverse. Flies, especially hover-flies, and short-tongued bees (Andrena,
Halictus), are commonest, while the more conspicuous flowers, and those rich in
nectar, are also visited by long-tongued bees, as well as beetles and even Lepido-
ptera. Cross-pollination by insects is frequently favoured or secured by protogyny
(Prunus, Amygdalus, species of Spiraea, Geum, Fragaria, Crataegus, Sorbus, and
Pyrus), by divergence of the stamens from the stigma in homogamous flowers
(species of Persica, Prunus, Rosa, and Potentilla), rarely by protandry (Rubus
caesius), or partial dicliny (Sanguisorba minor). Automatic self-pollination appears
regularly to take place in hermaphrodite flowers should insect-visits fail.

241. Amygdalus L.
Flowers protogynous, bright rose-red or white in colour, with half-concealed
nectar secreted in the lower part of the cup-shaped receptacle.

817. A. communis L. (=Prunus Amygdalus Séokes).—Kirchner (‘Flora v.
Stuttgart,’ pp. 460-1) says that in this species nectar is secreted by the yellow inner

ROSACEAE 343

surface of the lower part of the cup-shaped receptacle. It is protected from rain and
unbidden guests by woolly hairs which cover the ovary and the lower part of the
style. The numerous stamens are inserted into the receptacle at very different heights,
so that some of the anthers are on a level with the stigma, while others project
beyond it. When the flower opens, the stigma is already mature; the anthers
then gradually dehisce and get covered with pollen on all sides, so that in the
event of an insect-visit either cross- or self-pollination may take place. The latter
may also be effected automatically.

Visttors—I have observed the following on cultivated plants (‘ Bloemenbiol.
Bijdragen ’).—

A. Diptera. Syrphidae: 1. Eristalis tenax Z., skg. and po-dvg. B. Hymeno-
ptera. (a) Apzdae: 2. Bombus terrester Z. 9, skg.; 3. Halictus cylindricus /. 9,
do. The humble-bee usually alighted on the petals, and then crept under the
stamens to get at the nectar. In doing so it brushed the anthers with its back,
but did not touch the stigma. More rarely, it settled on the middle of the flower,
i.e. on the stigma, and then crept to the stamens, so that its under-surface was
covered with pollen: in this case cross-pollination was effected. (4) Vesprdae:
4. Vespa sp., skg.

Ducke saw the rufous bee Osmia cornuta Z/r. 9 and 4, freq., at Trieste.
Schletterer noticed 2 bees at Pola:—1. Bombus terrester Z.; 2. Xylocopa violacea Z.

On the foliage-leaves of this species there are extrafloral nectaries (Kirchner,
‘Flora v. Stuttgart,’ p. 461), which are visited by ants and wasps, that afford
protection against caterpillars and other harmful animals.

818. A. nana L. (=Prunus nana Séokes)—Kirchner has also described the
pollination of cultivated plants of this species (‘ Neue Beob. ii. d. Bestaubungseinricht.
einheimisch. Pfi.,’ Stuttgart, 1886, p. 36). The length of the receptacle is ro mm.
and the diameter of its throat is 4 mm.; it narrows somewhat below. The
lower part of its inner surface is yellow in colour, and secretes nectar, which is
protected from rain and unbidden guests as in the last species, the ovary and
that part of the style included in the receptacle being beset with many woolly
hairs. The upper glabrous region of the style projects 2-3 mm. from the
receptacle. Here also the stamens are inserted at very different heights on
the cup-shaped floral receptacle, and the filaments are of various lengths,
so that the anthers of the shortest stamens are on a level with the stigma
or even lower, while those of the longer stamens project beyond it. When the
flower opens the anthers are still closed; the already mature stigma is at first
covered by the stamens, which project vertically upwards. Subsequently the anthers
successively dehisce in no particular order, and get dusted with pollen all round,
so that automatic self-pollination can now readily take place. Whether this is
effective or not seems doubtful, as the numerous bushes observed by Kirchner but
rarely set fruits.

242. Persica Tourn.

Homogamous, bright rose-red flowers with concealed nectar, secreted in the

base of the receptacle.

819. P. vulgaris Mill. (=Prunus Persica Svokes and Amygdalus Persica L.).
(Herm. Miiller, ‘ Weit. Beob.,’ II, p. 244; Nirchner, ‘ Neue Beob. ii. d. Bestaubungs-

344 ANGIOSPERMAE—DICOTYLEDONES

einricht. einheimisch. Pfl.,’ Stuttgart, 1886, pp. 36-7; Knuth, ‘Bloemenbioi. Bij-
dragen.’)—In this species the cup-shaped receptacle is 8 mm. long; for a height
of 5 mm., according to Herm. Miiller, it is lined with an orange-coloured layer
that secretes nectar. The flowers are therefore better adapted for visits from
long-tongued insects than the other flowers of this order. Kirchner states that
the roots of the filaments are so curved towards the style as to lie closely
together, thus blocking the entrance to the receptacle and its contained nectar.
The size of the flowers varies greatly in the different varieties.

Vistrors.—In the Kiel Botanic Garden I noticed the honey-bee, skg., and also
2 humble-bees.—1. Bombus lapidarius Z. ¥, skg.; and 2. B. terrester Z. 9, do.
Kirchner saw Apis, Bombus sp., and Vanessa urticae Z. in Wurtemburg. Herm.
Miiller observed, besides Meligethes, several bees.—1. Andrena albicans Mfill. 9
and 6, po-cltg. and skg.; 2. Bombus terrester Z. 9, skg.; 3. Osmia cornuta Z/r. 9
and 4, do.; 4. O. rufa Z.4, do. Schletterer saw 2 bees at Pola.—1. Bombus terrester
L.; 2. Xylocopa violacea Z. Plateau noticed 3 bees in Belgium.—1. Apis, freq. ;
2. Bombus lapidarius Z.; 3. Osmia bicornis LZ.

243. Prunus L.

White, homogamous or protogynous flowers, with half or completely concealed
nectar, secreted in the receptacle.

820. P. Armeniaca L.—The flowers of this species are white, with a reddish
tinge; they exhale an odour of honey, and Kirchner says that they are homogamous
(‘Neue Beob. ii. d. Bestaubungseinricht. einheimisch. Pf.” Stuttgart, 1886, p. 37).
The red receptacle forms a cup 7-8 mm. deep, and nectar is secreted by the orange-
yellow lower part of its wall, so as to be quite concealed. The stamens are either
erect or directed somewhat outwards, so that the access to the nectar is not closed.
The ovary and lower part of the style are covered with hairs, that serve to protect
the nectar.

Visirors—Herm. Miiller observed only Hymenoptera (‘ Weit. Beob.,’ I
Pp. 244).—

(2) Apidae: 1. Andrena fasciata Wesm. 9, po-cltg.; 2. A. parvula X. 9, do. ;
3. Halictus leucozonius Schr. 9, skg.; 4. H. sexstrigatus Schenck 9, po-cltg. and
skg.; 5. Osmia rufa ZL. 6, freq, skg. (6) Péeromalidac; 6. Chalcis sp., skg.

Schletterer saw the carpenter-bee Xylocopa violacea L. at Pola.

?

821. P. domestica L. (Kirchner, ‘ Beitrage,’ p. 35; Herm. Miiller, ‘ Fertili-
sation,’ p. 222; Knuth, ‘ Bioemenbiol. Bijdragen.’)—The flowers of this species are
white with a greenish tinge, and Kirchner says that they are protogynous, though
Herm. Miiller describes them as homogamous. The first female condition, accord-
ing to Kirchner, lasts almost two days, after which the anthers dehisce; the flowers
remain for three days in the hermaphrodite condition, so that the whole period of
anthesis is five days. Cross-pollination is therefore only possible in the first stage.
Since the stigma projects beyond the inner stamens, while the outer ones are at the
same level, insects are likely to bring about cross-pollination, as they suck the nectar
secreted by the fleshy lining of the receptacle, for they usually touch the stigma and
anthers of the same flower with different sides of their bodies. In flowers that are

ROSACEAE 345

not quite erect, automatic self-pollination can easily take place during the second
stage, should insect-visits fail, the pollen falling from the outer longer stamens upon
the stigma.

Visitors.—I saw the following at Kiel.—

A. Diptera. Syrphidae: 1. ¥ristalis tenax 1. B. Hymenoptera. Apidac:
2. Apis mellifica Z. ¥, skg.; 3. Bombus terrester Z. 9, do. C. Lepidoptera.
Rhopalocera: 4. Pieris sp., skg.

Herm. Miiller gives the following list for Prunus domestica, P. avium, and
P. cerasus (‘ Fertilisation,’ p. 222).—

A. Diptera. Syrphidae: 1. Eristalis arbustorum /., skg.; 2. E. tenax Z.;
3. Rhingia rostrata Z., freq., skg. B. Hymenoptera. Afzdae: 4. Andrena
albicans Ail). 9 and 3, very freq., po-cltg. and skg.; 5. A. fulva Schr. 9, skg. and
po-cltg.; 6. Apis mellifica Z. ¥, very common, skg.; 7. Bombus hortorum /. 9, do. ;
8. B. lapidarius Z. 9, do.; g. B. terrester Z. 9, do.; 10. Osmia cornuta Zér. 9 and
6, do.; 11. O. rufa Z. 9 and 6, freq., do. C. Lepidoptera. hopalocera: 12.
Pieris brassicae Z., skg.; 13. P. napi Z., do.; 14. P. rapae Z., do.

MacLeod observed Apis in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
P. 325).

822. P. cerasifera Ehrh. (Knuth, ‘ Bloemenbiol. Bijdragen.’)—Focke says that
this species is somewhat infertile when isolated.

Visitors.—I observed Andrena albicans J/i//. 9, skg.

823. P. insititia L—Kirchner (‘ Beitriige,’ p. 35) says that the flower mechan-
ism of this species agrees with that of P. domestica, but cross-pollination is more
likely to occur, for the style sometimes exceeds the longest stamen in length.

Vistrors.—In the Kiel Botanic Garden I only observed Andrena albicans A/i//.
9, skg.

824. P. avium L.—For this species Kirchner (‘Beitrage,’ pp. 32-4) gives the
following account, supplementing the observations of C. K. Sprengel and Herm.
Miller (‘ Fertilisation, p. 222). The pure white flowers are faintly fragrant. The
corolla does not usually flatten itself out, but forms an almost hemispherical bell,
about 10-12 mm. deep, and 17-25 (on an average 22) mm. in diameter. The flower
is generally pendulous, and therefore faces downwards. Owing to the shape and
arrangement described the stamens and pistil of the sweet cherry are much better
protected against rain than is the case in other fruit trees. The length of anthesis
is 7-8 days. The flowers are homogamous, but automatic self-pollination cannot
normally take place. The stamens are of different lengths: those nearest the centre
being only 2-3 mm. long, while the outermost ones are 9-11 mm.

The stigma is mature when the flower opens; it stands at about the same levei
as the longest stamens. The stamens are divergent and, except in the case of a few
of the short (inner) ones, their anthers are still unripe. The pollen of those which
have dehisced can scarcely reach the stigma. Dehiscence progresses centrifugally in
a not very regular fashion, so that on the second day of anthesis a number of the
outer anthers are still closed, but these dehisce extrorsely in the course of the day.
The style projects in the middle of the flower beyond the obliquely diverging
stamens. These relative positions are maintained till the flower fades, so that
automatic self-pollination is infrequent, and can only occur accidentally. Insects

346 ANGIOSPERMAE—DICOTYLEDONES

probing the base of a flower to get the nectar secreted on the inner surface of the
receptacle generally touch the stigma and pollen with opposite sides of their heads,
and therefore effect cross-pollination. Pollen-devouring insects, on the other hand,
bring about cross- and self-pollination indifferently.

Vistrors.—I observed the following at Kiel (‘ Bloemenbiol. Bijdragen ’).—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z., skg.; 2. Rhingia rostrata

£.,do. B. Hymenoptera. Apzdac: 3. Apis mellifica Z. ¥, skg.; 4. Bombus
lapidarius Z. 9 and 4, do.

Herm. Miiller records the following for Jena (‘ Weit. Beob.,’ II, p. 244).—

A. Coleoptera. (a) Cerambycidae: 1. Tetrops praeusta LZ. (6) Chrysomelidae :
2. Haltica sp. B. Hymenoptera. <Afzdae: 3. Anthophora aestivalis Pz. § and 9,
skg. and po-clig.; 4. Apis mellifica Z. ¥, do.; 5. Halictus maculatus Sm. 9, po-cltg.;
6. Osmia aurulenta Pz. $ and 9, skg.; 7. O. fusca Chr. 9, po-clig.

Loew noticed the following bees in Brandenburg (‘ Beitrage,’ p. 37).—

1. Andrena combinata Chr. 9, skg.; 2. A. nigroaenea K. 9, do.; 3. A. pilipes
9, do.; 4. A. tibialis K. 8, do.; 5. A. varians K., var. helvola Z. 9, do.; 6. Nomada
alternata A. 6; 7. Osmia rufa Z. 4, skg.

MacLeod saw Apis, and Bombus terrester Z., in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 323).

825. P. Cerasus L.—The results of Kirchner’s investigations on this species
(‘ Beitrage,’ pp. 34-5) differ in not unimportant particulars from the account given
by Sprengel and Miiller. The flowers smell like bitter almonds, and are mostly
borne upon horizontal stalks, though not infrequently these may be directed obliquely
upwards or downwards. The petals spread out to form a flat surface, of which the
diameter is 28-31 mm. (on an average 30 mm.). The anthesis of individual flowers
fasts 7-8 days. The flowers are protogynous (Herm. Miiller described them as homo-
gamous). The stigma is at about the same level as the anthers of the longest stamens.

When the flower opens, the stigma is already mature, but the anthers are all
closed, so that at this stage insect visitors must effect cross-pollination. While the
flower is fully expanding, the anthers of the inner stamens begin to dehisce in
the course of the first day. Dehiscence is centrifugal and extrorse, so that in the
great majority of flowers automatic self-pollination cannot take place.

Visirors.—I observed a small bee, Andrena albicans, AZi//. 9, skg., at Kiel
(‘ Bloemenbiol. Bijdragen ’).

Schmiedeknecht saw Bombus pratorum Z. 9, in Thuringia, and Schenck 2

cuckoo-bees in Nassau.—r1. Nomada fabriciana Z., var. nigrita Schenck; 2. N.
thenana Jor.

Alfken and Hoppner (H.) observed the following bees at Bremen,—
1. Andrena albicans Mull. 9; 2. A. albicrus K. 9; 3. A. argentata Sm. 9:
4. Bombus agrorum F. 9 (H.); 5. B. derhamellus &. 9 (H.); 6. B. Tasnderia ag

B
(H.); 7. B. terrester Z. 9 (H.); 8. Nomada alboguttata A.-Sch. 6; 9. Osmia rufa
L.9

Bees were also noticed as follows.—

Friese (Mecklenburg), Osmia rufa Z.9. Loew (Brandenburg), Andrena pro-
pinqua Schenck (‘ Beitrage,’ p. 37). Plateau (Belgium), 1. Andrena fulva Schr. (=A.
vestita #”.); 2. Apis; 3. Osmia bicornis Z.

In Dumfriesshire Apis and a parasitic humble-bee have been recorded (Scott-
Elliot, ‘ Flora of Dumfriesshire,’ p. 53)-

ROSACEAE 347

826. P. spinosa L. (Herm. Miller, ‘Fertilisation, p. 221, ‘Weit. Beob.,’ IJ,
p- 244; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 323-4; Knuth,
‘Bloemenbiol. Bijdragen.’)—In the principal variety of this species the fragrant white
flowers appear before the leaves. They so abundantly bedeck the numerous dark
branches of the thorny shrub that it can be seen from a distance, so that at the
time of anthesis the blackthorn is the most conspicuous member of our flora.
The blossoms are consequently visited by numerous insects, which either seek
out the nectar secreted abundantly in the base of the receptacle or else collect
pollen. They are protogynous: when the flower-buds expand the anthers are
still closed, while the style projects a few millimetres beyond the crowded central
mass of stamens, and as the stigma is already receptive, cross-pollination must be
effected by insects that have previously visited an older flower in the male stage.
The stamens subsequently elongate, spread themselves out, and open their anthers.
The style also becomes longer, so that it projects somewhat beyond the shorter
stamens. The stigma remains receptive, so that, should insect-visits fail, automatic
self-pollination can take place.

Visitors.— Hermann Miiller (H.M.) and myself (Kn.) have observed the
following.—

A. Coleoptera. Nitidulidac: 1. Meligethes, nect-skg. (H. M., Kn.). B. Di-
ptera. (a) Brbiontdac: 2. Bibio marci L., nect-lkg. (H. M.). (4) Emprdae: 3. Empis
rustica /a/l., skg. (H.M.). (c) Afusczdae: 4. Species of Anthomyia, skg. (H. M.); 5.
Chlorops sp., do. (H. M.); 6. Musca domestica Z., do. (Kn.); 7. Scatophaga merdaria
£, do. (H.M.); 8. S. stercoraria Z., do. (H. M., Kn.); 9. Sepsis, freq., skg. (H. M.).
(d) Syrphidae: 10. Eristalis arbustorum Z., skg. and po-dvg. (H.M.); 11. E. intri-
carius Z., do. (H.M.); 12. E. nemorum Z., do. (H. M.); 13. E. tenax Z., do (H. M.,
Kn.); 14. Rhingia rostrata Z., skg.(Kn.). C. Hymenoptera. (a) Apidac: 15.
Andrena albicans Mull. 9 and 4, skg. and po-cltg. (H.M., Kn.); 16. A. atriceps A. 9.
and 4, skg. (H.M.); 17. A. dorsata K. 9, po-clig. (H.M.); 18. A. fasciata Wesm. 6,
skg. (H.M.); 19. A. fulva Schrank 9, skg. and po-clig. (H. M.); 20. A. fulvicrus XK.
g and 4, skg. (H.M.); 21. A. gwynana A. 9, skg. and po-cltg. (H.M.); 22. A.
parvula A. 9, do. (H.M., Kn.); 23. A. eximia Sm. 9, do. (H.M.); 24. A. schran-
kella Wy/. 9, po-cltg. (H.M.); 25. Apis mellifica Z. ¥, skg. and po-cltg. (H. M., Kn.);
26. Bombus lapidarius Z. é, skg.(H.M.); 27. Halictus albipes #. 9, freq., skg. and
po-cltg. (H.M.); 28. H. cylindricus F. 9, do. (H.M.); 29. Nomada succincta Pz. 6,
skg. (H.M.); 30. Osmia rufa Z. 4, do. (H.M.). (4) Zenthredinidae: 31. Dolerus
gonager A/, skg. (H.M.). D. Lepidoptera. Rhopalocera: 32. Vanessa io L.,
persistently skg. (H. M.).

Alfken noticed the following at Bremen.—

A. Diptera. (2) Bombylizdac: 1. Bombylius major L., freq., skg. (0) Afuscidae :
2. Sarcophaga carnaria Z., skg. (c) Syrphrdae: 3. Eristalis arbustorum Z., very
common, skg. and po-dvg.; 4. E. intricarius Z., do.; 5. Helophilus pendulus Z., do.;
6. Platycheirus albimanus #.; 7. Syritta pipiens 2. B. Hymenoptera. (a) Apidae:
8. Andrena albicans J/#//.9 and $; 9. A. albicrus A. 9 and é; 10. A. extricata
Sm.9; 11. A. flavipes Pz.9; 12. A. helvola Z.d; 13. A. nitida Fourc.9; 14. A.
varians A.9 and é; 15. Apis mellifica Z.¥, very common, skg.; 16. Halictus flavipes
F.9; 17. H. morio #.9; 18. H. nitidiusculus AK. 9; 19. Nomada alternata X. 9;
20. N. lineola Pz. 8, skg.; 21. N. succincta Pz. 4; 22. Osmia rufa LZ. 8. (6) Zen-
thredintdae: 23. Hoplocampa ferruginea /*.; 24. H. rutilicornis A7g.

Gerstacker records a bee, Osmia aurulenta Pz., freq., for Berlin.

348 ANGIOSPERMAE—DICOTYLEDONES

Schiner (Austria) noticed a hover-fly, Mallota fuciformis ¥.; and von Dalla
Torre (Tyrol) a bee, Halictus smeathmanellus X. 9 and é. Schletterer mentions the
latter insect for the Tyrol, and observed the following bees at Pola——x. Andrena
carbonaria Z., freq.; 2. A. deceptoria Schmizedekn.; 3. A. theracica #.; 4. Bombus
terrester Z.; and a wasp, 5. Polistes gallica Z.

Bees were observed by the following authorities in the localities stated.—

Schmiedeknecht (Thuringia).—1. Andrena congruens Schmiedekn.; 2. A.eximia
‘Sm. Saunders (England), the rare Andrena bucephala S/epf., with its parasite, the
beautiful Nomada xanthosticta A. Smith (England), Andrena bimaculata A.

827. P. Padus L. (Herm. Miller, ‘Fertilisation,’ pp. 221-2, ‘ Weit. Beob.,’ II,
p- 244; Knuth, ‘ Lloemenbiol. Bijdragen.’)—The white, strongly fragrant blossoms
of this species are arranged in many-flowered racemes, usually pendulous. Hermann
Miiller states that their mechanism agrees with that of P. spinosa in being proto-
gynous; but the stamens remain curved somewhat inwards during the whole period
of anthesis; so here self-pollination is more likely to be effected by insects during
the second (hermaphrodite) stage. The inner stamens dehisce while they are
curved down under the stigma, so that in becoming erect their anthers must brush
against its edge. Automatic self-pollination must, therefore, regularly take place,
should insect-visits fail.

Visirors. — On garden plants at Kiel I saw only Muscids as visitors of the
flowers, which smell of trimetbylamide.—1. Calliphora vomitoria Z.; 2. Lucilia caesar
L.; 3. Musca domestica Z.; 4. Sarcophaga carnaria Z.; all skg.

Herm. Miiller observed the following.—

A. Coleoptera. (a) Cerambycidae: 1. Grammoptera ruficornis #., nect-lkg.
(4) Afalacodermata: 2. Dasytes sp., nect-lkg. (¢) Mordellidae: 3. Anaspis rufilabris
Gyill., nect-lkg. (d) Netidulidae: 4. Meligethes, nect-lkg. B. Diptera. Lmpzdae:
5. Empis livida Z., skg.; 6. E. rustica “all, do. C. Hymenoptera. Aprdae:
7. Andrena parvula A. 9, skg.

F. F. Kohl (Tyrol) saw the ruby wasp Ellampus aeneus /.

828. P. Mahaleb L.—This species also is slightly protogynous, according to
Kirchner (‘ Neue Beob. ii. d. Bestaubungseinricht. einheimisch. Pf.,’ p. 37). At first
the stamens are erect or inclined somewhat inwards, but subsequently the outer ones
spread outwards. At the beginning of anthesis the style is of the same length as
the shortest stamens, but later on attains that of the longest.

Visirors.—Schletterer observed three bees and a wasp at Pola.—

x. Andrena morio Srull.; 2. A. Thoracica /.; 3. Bombus argillaceus Scop. ;
4. Polistes gallica Z. (wasp).

244. Rosa Toum.

Flowers homogamous, sometimes delightfully fragrant, usually large; rose-
coloured, white, or more rarely yellow ; devoid of nectar (some species, perhaps, have
a thin layer on the edge of the receptacle). The lack of nectar is made up for by
the abundant production of pollen.

829. R. canina L. (Herm. Miiller, ‘Fertilisation,’ pp. 236-8, ‘ Weit. Beob.,’
II, p. 239; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 307-8; Heinsius,

ROSACEAE 349

op. cit., iv, 1892, pp. 55-7; Knuth, ‘ Bl. u. Insekt. a. d. nordfr. Ins., pp. 70, 154.)—
The bright pink fragrant flowers of this species are homogamous, and probably
devoid of nectar. Hermann Miiller says that the upper margin of the receptacle,
internal to the insertion of the stamens, exhibits a thick fleshy ring, but this certainly
does not appear to secrete any nectar. Heinsius describes this ring as having the
structure of a nectary, the secretion, however, being so scanty that it is impossible to
include the blossoms in the class of nectar-flowers. As the stamens curve outwards
when the flower opens, and the petals remain tolerably erect, the ring just mentioned
and the stigmas which project from the middle of it afford the most convenient
alighting-place to insect visitors, and cross-pollination isin this way favoured. Should
insect-visits fail, automatic self-pollination by the fall of pollen upon the stigmas takes
place in all flowers, except such as happen to be quite erect.

Visitors.—I saw the honey-bee, po-cltg., in the island of Amrum. MacLeod
(Flanders) noticed a humble-bee, a Muscid, and 2 beetles (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, pp. 308, 380). Heinsius (Holland) records 3 hover-flies (Didea
intermedia Loew 9, Eristalis arbustorum Z. 9, E. horticola Deg. 8), 2 Muscids
(Anthomyia sp. 4, Aricia vagans Fa//. 5), and a beetle (Cetonia metallica #. = C.
floricola A/és?.) (op. cit., iv, 1892, p. 57).

Herm. Miiller (H. M.) and Buddeberg (Budd.) observed the following.—

A. Coleoptera. (a) Buprestidae: 1. Anthaxia nitidula Z., in the flowers
(Budd.). (4) Cerambycidae: 2. Stenocorus inquisitor #. (H.M.); 3. Strangalia
maculata Poda (H.M.). 4. S. nigra Z., gnawing the anthers and the delicate parts
of the flowers in general (H.M.). (c) Chrysomelidae: 5. Luperus flavipes Z. (H. M.).
(d) Clertdae: 6. Trichodes alvearius /. 9 (Budd.). (¢) Dermestidae: 7. Anthrenus
pimpinellae /., freq., po-dvg. (H. M.); 8. A. scrophulariae Z., do.(H.M.). (/) Afor-
dellidae: 9. Anaspis frontalis Z. (H.M.); 10. Mordella aculeata Z. (H.M.).
g) Nitidulidae: 11. Meligethes, freq. (H. M.). (A) Scarabaerdae: 12. Cetonia aurata
L., feeding on the stamens and stigmas, and biting large holes in the petals (H. M.,
Budd.); 13. Oxythvrea funesta Poda, do. (H.M.); 14. Phyllopertha horticola Z.,
do. (H. M.). (2) Zelephoridae: 18. Anthocomus fasciatus £.(H.M.). B. Diptera.
Syrphidae: 16. Helophilus floreus Z., po-dvg. (H. M.); 17. Syritta pipiens Z., freq.,
po-dvg.(H.M.). C. Hymenoptera. Aprdae: 18. Andrena albicans A7#/.9 and 4,
po-cltg. and po-dvg. (H. M.); 19. A. fucata Sm. 9, po-cltg. (H.M.); 20. Apis mellifica
£. %, do. (H.M.); 21. Halictus nitidus Schenck 9, do. (H.M.); 22. Megachile
circumcincta K. 9, do. (H.M.); 23. Osmia rufa Z. 9, do. (H.M.); 24. Prosopis
communis /Vy/. 9 and 4, freq., po-dvg. (H. M.).

Schenck (Nassau) saw the bee Andrena labialis A. 6; Redtenbacher (Vienna)
the Chrysomelid Cryptocephalus duodecimpunctatus #.; von Dalla Torre and
Schletterer (Tyrol) the bee Andrena propinqua Schenck 9.

In Dumfriesshire Apis, freq., 3 humble-bees, a short-tongued bee, a ruby-wasp,
a Tenthredinid, 2 Muscids, and 5 hover-flics were recorded (Scott-Elliot, ‘ Flora of
Dumfriesshire,’ p. 62).

830. R. repens Scop. (= R. arvensis Huds.).—The flowers of this species
are white, fragrant, and devoid of nectar. Kirchner describes their mechanism as
agreeing with that of R. canina. Kerner states the flowers are open from 4 a.m.
till 9 p.m., and that anthesis lasts two days.

831. R. pimpinellifolia 7. (= R. spinosissima Z.). (Knuth, ‘Bl. u. Insekt.
a. d. nordfr. Ins.,’ pp. 69-70, 154.)—I have been able to study the flower mechanism

350 ANGIOSPERMAE—DICOTYLEDONES

of this species on the dunes, especially those of the island of Rém, but have had
little opportunity of observing visitors there. The diameter of the white corolla is
about 3cm. Anthers and stigmas mature simultaneously when the flowers open.
Although the stamens curve away from the stigmas, so that insects-visits render
cross-pollination possible, yet automatic self-pollination takes place in the absence of
these insects, for when, owing to the frequent violent winds on the islands, the little
plant strikes against the ground pollen can easily be transferred to the adjacent
stigmas.

Visrrors.—I saw several po-dvg. Muscids in the island of Rém, as well as
Forficula and a few po-dvg. beetles.

Verhoeff (Norderney) observed a beetle (Phyllopertha horticola Z., freq., po-dvg.)
and a humble-bee (Bombus terrester Z., a ¥, po-cltg.).

832. R. rubiginosa L. (Herm. Miller, ‘ Weit. Beob.,’ IJ, pp. 239-40;
Knuth, ‘ BI. u. Insekt. a.d. nordfr. Ins.” pp. 70, 154.)—In this species not only the
flowers but also the foliage-leaves are fragrant, so that the latter help to attract insects.
The bright pink flowers afford them not only pollen, as in the other species of the
genus, but also nectar, as shown by Hermann Miiller. It is secreted in quite a thin
layer on the broad fleshy margin of the receptacle. The flowers of this species are
slightly protogynous, so that cross-pollination necessarily takes place when insects
visit them at the beginning of anthesis. At this stage numerous fully mature stigmas
are crowded together in the middle of the flower, as projecting pillow-like swellings,
affording to insects a convenient alighting-place and platform. The stamens at this
time are still curved outwards, and their anthers as yet unripe. Later on, when the
anthers dehisce, the filaments bend together above the middle of the flower, so that
automatic self-pollination takes place. Kerner states that the flowers are open from
5 a.m. till 9 p.m.

Visirors.—In the island of Amrum I only observed the honey-bee, po-cltg.

Herm. Miiller gives the following list. —-

A. Coleoptera. (a) Chrysomelidae: Crytocephalus sericeus Z., devouring the
flowers; 2. Lupcrus flavipes Z., freq. (0) Zéelephoridae: 3. Danacea pallipes Pz., in
very large numbers within the flowers. B. Diptera. Svraszomyidae: 4. Oxycera
pulchella J/g., occasional. C. Hymenoptera. Apidae: 5. Bombus pratorum ZL.
¥, po-cltg.; 6. B. terrester Z. ¥, do.

833. R. alpina L.
VisiTors.—Herm. Miiller (‘ Alpenblumen,’ p. 215) saw a small bee (Halictus).

834. R. centifolia L. (Herm. Miiller, ‘ Fertilisation, pp. 237-8, ‘ Weit.
Beob.,’ I], p. 239; Knuth, ‘Bloemenbiol. Bijdragen.’)

VisiTors.—I (Kn.) noticed that this and other cultivated double species received
a moderately large number of visits; Herm. Miiller (H. M.) records more numerous
visitors for R. centifolia alone.—

A. Coleoptera. (a) Cerambyctdae: 1. Clytus arietis Z., destroying the delicate
parts of the flowers, especially the anthers (H.M.); 2. Grammoptera ruficornis /.,
very numerous, do. (H.M.); 3. Strangalia atra Larch. as 1 (H.M.); 4. S. attenuata
L., do. (H.M.). (8) Custelzdae: 5. Cistela murina Z., as 1(H.M.). (c) Dermestidae :
6. Anthrenus fuscus Z/r., rare (H.M.); 7. A. pimpinellae /. (H.M.); 8. A. scrophu-

ROSACEAE 351

lariae Z., freq. (H.M.). (@) Mordellidae: 9. Anaspis ruficollis #., as 1 (H.M.);
10. Mordella aculeata Z., do. (H. M.) (¢) Metzduledae: 11. Meligethes, in very large
numbers (H.M., Kn.). (/) Scarabaerdae: 12. Cetonia aurata Z., devouring the
flowers (H. M., Kn.) ; 13. Melolontha vulgaris /., do. (H. M., Kn.); 14. Phyllopertha
horticola Z., do. (H.M., Kn.). (g) Zéelephoridae: 15. Anthocomus fasciatus Z.,
freq. (H. M.); 16. Dasytes sp., rare (H. M.).

Kohl records a fossorial wasp, Crabro peltarius Schred. 9 and 6.

835. R. alba L.—On bushes with double flowers in gardens at Vészt6
(Hungary), Borbés observed 2-3 well-formed fruits containing seeds capable of
germinating.

836. R. pomifera Herm.—

Visitors.—The following bees were observed by von Dalla Torre and
Schletterer in the Tyrol.—z. Halictus albipes /.; 2. H. interruptus Pz.; 3. H.
tumulorum Z.; 4. Osmia leucomelaena A.; 5. Prosopis sinuata, Schenck 8.

245. Rubus L.

Flowers white or reddish in colour ; homogamous, feebly protandrous, or feebly
protogynous, sometimes dioecious (R. Chamaemorus); with concealed nectar abun-
dantly secreted by a fleshy ring on the margin of the receptacle internal to the
stamens.

The recognition of the numerous recently demarcated species of Rubus is in
itself a special study, and perhaps later investigators will consider it necessary to
distinguish between the flower mechanisms and visitors of the various species;
though it is hardly likely that any important differences will make themselves
apparent. For the purposes of Flower Pollination, in the meantime, one collective
name will serve, i.e.—

837. R. fruticosus L. (Herm. Miiller, ‘Fertilisation, pp. 227-8, ‘ Weit.
Beob.,’ II, pp. 240-1; Kirchner, ‘Flora v. Stuttgart,’ p. 451; Loew, ‘ Bliitenbiol.
Floristik,’ p. 391 ; Knuth, ‘ Bl. u. Insekt. a. d. Ins. Riigen,’ ‘ Bloemenbiol. Bijdragen,’
&c.)—In this species, according to Hermann Miiller’s account, the usually white
petals spread out flat, so that conspicuousness is fairly marked. The stamens
diverge so widely that even very short-tongued insects can easily thrust their heads
between the filaments and carpels so as to reach the nectar-secreting ring in the
base of the flower. The anthers of the outermost stamens dehisce first, and turn
their pollen-covered sides upwards, while the stigmas mature simultaneously. It
follows that most visitors effect crossing, so that, as a rule, the flowers are fertilized
before all the anthers are ripe. Automatic self-pollination is rendered somewhat
difficult, for only the dehisced anthers of the innermost stamens sometimes touch
the outermost stigmas. It is only in the rarest cases, during continued bad weather,
that self-pollination is resorted to, for otherwise insect-visits are extremely numerous.

Vistrors.—I have not observed in North Germany the species of Halictus
which were seen by Buddeberg in large numbers in Nassau.

Herm. Miiller (H. M.) and Buddeberg (Budd.) record the following.—

A. Coleoptera. (a) Cerambycidae: 1. Clytus arietis Z., sometimes nect-Ike.,

352 ANGIOSPERMAE—DICOTYLEDONES

sometimes dvg. the flowers (H. M.); 2. Leptura livida /., do. (H.M.); 3. L. macu-
licornis Deg., very numerous, in the flowers (H.M.); 4. Judolia cerambyciformis
Schr., do. (H.M.); 5. Strangalia armata Hést., do. (H.M.); 6. S. atra Lazch., do.
(H.M.); 7. S. melanura Z., do. (H.M.); 8. S. nigra Z., do. (H.M.). (8) Curcu-
ftonidae: g. Spermophagus cardui Svev., busy on the anthers (H.M.). (c) Dermestidae:
1o. Byturus fumatus #., skg., and dvg. the flowers (H.M.). (d) Llateridae: 11.
Corymbites aeneus Z., dvg. the delicate parts of the flowers (H. M.); 12. Lacon
murinus Z.(H.M.) ; Limonius cylindricus Payk.,as 11. (e) Mordellidae: 13. Mor-
della aculeata Z., in the flowers (H. M., Thuringia). (/) Wetedulidae: 14. Meligethes,
freq. (g) Oedemeridae: 15. Oedemera virescens Z., nect-lkg., and dvg. the delicate
parts of the flowers(H.M.). (4) Scarabae¢dae: 16. Phyllopertha horticola £., grazing
in the flowers (H.M.); 17. Trichius fasciatus Z., dvg. the delicate parts of the
flowers (H.M.). (z) Zelephoridae: 18. Cantharis rustica Fa//., as 17 (H.M.);
19. Malachius bipustulatus Z., do. (H.M.). B. Diptera. (a2) Conopidae: 20.
Physocephala rufipes #., skg. (H.M.); 21. Sicus ferrugineus Z., do. (H. M.).
(5) Empidae: 22. Empis livida Z., freq., skg. (H.M.); 23. G. tessellata #, skg.
(H.M.). (c) Muscidae: 24. Echinomyia grossa Z., skg. (H.M.); 25. Lucilia sp.,
do. (d) Syrphidae: 26. Ascia podagrica F., skg. and po-dvg. (H. M.); 27. Chryso-
toxum arcuatum Z., do. (H.M.; 28. Eristalis tenax Z., do. (H.M.); 29. Helophilus
pendulus Z., do. (H. M.); 30. Rhingia rostrata Z., do. (H. M.); 31. Syritta pipiens
L., do. (H.M.); 32. Volucella inanis Z., skg. (Budd.); 33. V. pellucens Z., do.
(H.M., Budd.) (e) Stratiomyidae: 34. Chrysomyia formosa Scop., skg. (H. M.);
35. Sargus cuprarius Z., do. (H.M.). (/) Zipulidae: 36. Tipula oleracea L., skg.
(H.M.). C. Hymenoptera. (a) Apidae: 37. Andrena albicrus X. 4, skg. (H.M.) ;
38. A. gwynana K. 9, do. (H.M.); 39. A. thoracica #. 9, do. (H. M.); 40. Apis
mellifica Z. ¥, very common, skg. and po-cltg. (H.M.); 41. Bombus agrorum F.,
skg. and po-cltg. (H. M., Kn.) ; 42. B. hortorum Z. ¥, do. (H. M.); 43. B.hypnorum
L. 8, freq., skg. (Kn.); 44. B. lapidarius Z. 6, skg. (Kn.); 45. B. pratorum Z. 9 and
6, numerous, skg. (H.M.); 46. B. scrimshiranus A. ¥, skg. and po-cltg. (H.M.) ;
47. B. sylvarum Z. 9, do. (H.M.); 48. B. soroénsis #, var. proteus Gers/., freq.,
skg. (Kn.); 49. B. terrester Z.9 and 4, skg. and po-cltg. (H.M.); 50. Coelioxys
elongata Lep. 9 and 4, skg. (Budd.); 51. C. rufescens Zep. 9, dvg. (H.M.); 4, do.
(H.M., Budd.); 52. Diphysis serratulae Pz. 9, do. (H.M.); 53. Halictus albipes F.,
var. affinis Schenck, do. (H.M.); 54. H.cylindricus /. 9 and 6, do. (H.M.); 55. H.
flavipes F. 9, do. (Budd.); 56. H. leucopus K. 9, do. (Budd.); 57. H. leucozonius
Schr. 9, po-cltg. (H.M.); 58. H. lucidulus Schenck 9, skg. (H.M.); 59. H. mala-
churus £’. 9, do. (Budd.); 60. H. quadricinctus X. 9, do. (Budd.) ; 61. H. sexnotatus
£. 9, do.(H. M., Budd.) ; 62. H.smeathmanellus X. 9, do. (Budd.); 63. H. villosulus
K. 9, skg. and po-cltg. (H.M., Budd.); 64. H. zonulus Sm. 9, skg. (H.M.);
65. Macropis labiata Pz. 6, do. (H.M.) ; 66. Nomada fabriciana Z. 9, do. (H.M.) ;
67. N. lateralis Pz. 9, do. (H. M.); 68. N. lineola Pz. $, do.(H.M.); 69. N. ruficornis
L., do. (H.M.); 70. Osmia fusca Chr. 9, do. (H.M.); 71. Prosopis communis
yi. 3, do. (H.M.); 72. P. pictipes Wy/. 8, do. (H.M.); 73. P. variegata F. 6, do.
(H.M.); 74. Psithyrus campestris Pz. 9, do. (H.M.); 75. P. quadricolor Les. 4,
do. (H. M., Kn.); 76. P. vestalis Fourcr.9, do. (H.M.); 77. Stelis breviuscula Wy/. é,
do. (H.M.). (6) Hormicidae: 78. Formica pratensis Deg. ¥, nect-lkg. (H.M.);
79. F.sp., skg.(Kn.). (c) Sphegidac: 80. Ammophila campestris Zir. 6, skg. (H. M.);
81. A. hirsuta Scop., nect-lkg. (H.M.); 82. A. sabulosa Z. 9 and 6, skg. (H.M.);
83. Cerceris quinquefasciata Ross? 6, do. (H.M.); 84. C. rybiensis Z. 9, nect-lkg.
(Budd.); 85. Crabro peltarius Schred. 9 and 4, skg. (H.M.); 86. Oxybelus uniglumis
ZL. ¢ and 4, do. (H.M.). D. Lepidoptera. Rhopalocera: 87. Argynnis paphia
L., skg. (H.M.); 88. Epinephele janira Z., do. (H.M.); 89. Erebia ligea Z., freq.,
ske. (H.M.); 90. Carterocephalus palaemon Pail, skg. (H.M.); 91. Melithaea
athalia Esp., freq., skg. (H.M.); 92. Pieris crataegi Z., skg. (H.M.); 93. P. napi
“., do. (H.M.); 94. Thecla ilicis £sp., do. (Budd.).

ROSACEAE 353

The following bees were recorded as stated.—

Schenck (Nassau), 1. Andrena florea F.; 2. A. trimmerana K.; 3. Coelioxys
conoidea ///.9; 4. Halictus albipes #.9; 5. H. calceatus Scop. 9; 6. H. morio F. ;
7. H. pauxillus Schenck 9 and $; 8. H. sexnotatus K. Schletterer (Tyrol), Bombus
variabilis Schmzedekn, von Dalla Torre (Tyrol), B. muscorum F. ¥.

Kohl (Tyrol) records a fossorial wasp (Crabro peltarius Schred.9 and $), and
a ruby-wasp (Hedychrum nobile Scop.).

Loew observed the following in Silesia (‘ Beitrage, pp. 33, 51).—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida F., nect-lkg.; 2. L.
maculicornis Deg., do.; 3. Strangalia bifasciata Afu//., do. (6) Malacodermata: 4.
Dasytes flavipes 7, nect-lkg. (c) Witidulidae: 5. Meligethes sp. B. Diptera.
(a) Muscidae: 6. Dexia rustica F., skg. (6) Syrphidae: 7. Eristalis intricarius L.,
skg.; 8. E. tenax Z., do.; 9. Helophilus floreus Z., do.; 10. H. pendulus Z., do. ;
11. Syrphus grossulariae AZ/g., do.; 12. Volucella bombylans Z., do. ; 13. V. pellucens
L., do. C. Hymenoptera. Apidae: 14. Bombus pratorum ZL. 6, skg.; 15. B.
scrimshiranus A. 6, do.; 16. B. soroénsis #., var. B. proteus Gers/. 6, do.; 17.
Diphysis serratulae Pz. 6, do.; 18. Macropis labiata Pz. 5, do. D. Lepidoptera.
Rhopalocera: 19. Argynnis paphia Z.; 20. Coenonympha arcania Z.; 21. Hesperia
comma L.; 22. Lycaena argiolus Z.; 23. Melitaea parthenie Bkh.; 24. Epinephele
janira Z.; 25. Pieris brassicae Z.; 26. Polyommatus alciphron Roff.; 27. Vanessa
prorsa Z. All skg.

Loew further noticed 2 bees in Mecklenburg—1. Prosopis confusa Wy. 5, skg. ;
2. P. sp., do. (op. cit., p. 41): and 2 bees in Switzerland—1. Andrena propinqua
Schenck 9, skg.; 2. A. thoracica F. 9, do. (op. cit., p. 60).

I saw the following in the island of Rigen.—

A. Diptera. (a) Muscidae: 1. Aricia sp. (6) Syrphidae: 2. Eristalis pertinax
Scop.; 3. E. tenax Z.; 4. Syrphus ribesii Z.; 5. Volucella bombylans Z. 9, and
the var. plumata JZ. B. Hymenoptera. (a) Apzdae: 6. Apis moellifica Z.;
7. Bombus agrorum /. 9; 8. B. lapidarius Z. ¥; 9. B. terrester Z. 9; ro. Psithyrus
quadricolor Zep. 6. (6) Sphegidae : 11. Ammophila sabulosa Z. 6. C. Lepidoptera.
Rhopalocera: 12. Argynnis paphia Z., and the var. valesina Zsf.; 13. Epinephele
janira Z.; 14. Limenitis sibylla Z.; 15. Pieris sp. All freq., skg.

Gerstacker noticed the following bees in Berlin.—

1. Osmia acuticornis Duf. ef Perr.; 2. O. leucomelaena X.; 3. O. uncinata
Gerst.

Alfken and Héppner (H.) observed the following at Bremen.—

A. Diptera. (a) Asildae: 1. Dioctria oelandica Z.9 and 6 (4) Syrphidae :
2. Sericomyia borealis Fa//.; 3. Volucella pellucens Z. B. Hymenoptera. (ca)
Apidae: 4. Andrena albicans Mill. 9; 5. A. albicrus K. 9; 6. A. tibialis K. 9;
7. Bombus agrorum /. 9 and ¥; 8. B. lucorum LZ. ¥; 9. B. pratorum Z.6; 10. B.
proteus Gerst.9; 11. B. sylvarum Z. 9; 12. Eriades truncorum Z. 8; 13. Halictus
calceatus Scop.9; 14. H. levis A.9; 15. Macropis labiata. K.9 and 6; 16. Megachile
circumcincta K. $; 17. Nomada mutabilis A/or. 9 (H.); 18. N. ochrostoma K. Q;
1g. N. roberjeotiana Pz. §(H.); 20. N. similis Mor. 9 (H.); 21. Prosopis bipunctata
F. 9; 22. P.communis Vy/. 9 and é; 23. P. confusa Myl. 9; 24. P. dilatata X. 9;
25. P. hyalinata Sm.9; 26. P. nigrita / 9; 27. P. pictipes Wy/.9; 28. P. rinki
Gorski, 8. (6) Sphegidae: 29. Crabro subterraneus /. 9.

Visitors were recorded as follows by the authorities and at the places named.—

Hoffer (Steiermark), the bee Bombus hypnorum Z.¥. Schiner (Dalmatia), the
Therevid Xestomyza Kollari gg. Friese (Hungary), 2 bees, 1. Andrena albo-
punctata Rossz, 2. A. fucata Sm.; (Baden) the bee Bombus jonellus K.

DAVIS, II Aa

354 ANGIOSPERMAE—DICOTYLEDONES

MacLeod observed Apis, 7 humble-bees, 7 short-tongued bees, a Tenthredinid,
5 hover-flies, 6 other flies, 4 beetles, and 10 Lepidoptera in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 318-19); and 3 bees, a Lepidopterid, a beetle, and
a hover-fly in the Pyrenees (op. cit., iii, 1891, p. 432).

H. de Vries saw a bee, Halictus cylindricus /. 9, in the Netherlands (Ned.
Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).

Willis observed the following in the neighbourhood of the south coast of
Scotland (‘ Fis. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera. Nitidulidae: 1. Meligethes viridescens /., freq., skg. and
po-dvg. B. Diptera. (a) Muscidae: 2. Anthomyia radicum Z., freq. (6) Syr-
phidae: 3. Eristalis pertinax Scop., freq., skg.; 4. Platycheirus albimanus /., skg.;
5. Syrphus balteatus Deg., po-dvg.; 6. S. topiarius M/g., skg. C. Hymenoptera.
Apidae: 7. Bombus agrorum Z., freq., skg.; 8. B. hortorum Z., do. D. Lepi-
doptera. (a) Microlepidoptera: 9. Simaéthis oxyacanthella Z., freq., skg. (6)
Rhopalocera: to. Pieris napi L., freq., skg.

Saunders (Sd.) and Smith (Sm.) record the following bees for England.—

1. Andrena austriaca Pz.(= A. rosae Saund.) (Sd., Sm.); 2. A. bimaculata K,
(= A. decorata Sm., and A. vitrea Sm.), 2nd generation (Sd., Sm.); 3. A. carbonaria

1

Fic. 110. Rubus caestus, L. (Diagrammatic longitudinal sections of flowers, magnified about
2diameters. Fdéhr, July, 1892.) 1. Flower before the beginning of the first stage: all the anthers are
closed and incline together above the equally immature styles; the receptacle is flat. 2. Flower in the

beginning of the male stage: the anthers of the outer stamens (which have now curved away from
the centre are ripe ; the stigmas are not yet fully mature; the receptacle has begun to bulge. 3. Flower
in the hermaphrodite stage: most of the stamens have curved away from the centre, and their anthers are
ripe; only a few remain immature with filaments curved under the fully developed stigmas, which are at
the ends of the elongated styles; the receptacle is distinctly dome-shaped. , receptacle; &, sepal;
¢, petal; a2, immature anther; @’, mature anther; s, stigma.

L. (=A. pilipes 7), 2nd generation (Sd.); 4. A. dorsata A. (Sd., Sm.); 5. Halictus
sexnotatus K. 6 (Sm.); 6. Prosopis bipunctata F. (= P. signata Pz.) (Sd.); 7. P.
communis /Vy/. (Sd.); 8. P. confusa Wy/. (Sd.); 9. P. hyalinata Sm. (Sd., Sm.)

Marquard (Cornwall) saw 2 bees, 1. Andrena austriaca Pz.; 2. A. minutula KX,

838. R. caesius. (Knuth, ‘Bl. u. Insekt. a.d. nordfr. Ins.,’ pp. 66-7, 154,
‘ Weit. Beob. ii. Bl. u. Insekt. a.d. nordfr. Ins.,’ p. 233.)—When the flowers of this
species open the anthers are all immature, and lie above the middle of the flower
with the filaments inclining inwards. Subsequently the outer stamens bend back
towards the petals, and their anthers dehisce. As dehiscence progresses centripetally
the originally flat receptacle becomes more and more convex, the styles begin to
grow, and the stigmas appear at their tips. Several series of stamens with dehisced
anthers have meanwhile bent back towards the petals, while those with undehisced
anthers remain with their filaments curved under the stigmas. When insects visit
the flower and alight in its centre cross-pollination is effected, but if they alight
elsewhere self-pollination results. The latter takes place automatically when the inner
stamens erect themselves, for they dehisce directly upon the stigmas.

ROSACEAE 355

Visitors.—I observed the following in the island of Fohr.—

A. Coleoptera. Witidulidae: 1. Meligethes sp. B. Diptera. (a) Muscidae:
2. Anthomyia sp.9; 3. Drymeia hamata Fal/.; 4. Lucilia caesar Z.; 5. L. sp.;
6. Musca sp.; 7. Onesia sepulcralis Mg.; 8. Sarcophaga carnaria LZ. (6) Syrphidae:
9. Eristalis arbustorum Z.; 10. Helophilus floreus Z. 9; 11. H. pendulus Z.;
12. Syrphus ribesii Z. C. Hymenoptera. <Apzdae: 13. Apis mellifica Z.; 14.
Bombus lapidarius Z.; 15. B. terrester Z.; 16. Coelioxys acuminata Myl.; 17. C.
rufescens Lep.; 18. Colletes picistigma Zhoms.; 19. Megachile centuncularis Z. 9.
D. Lepidoptera. Rhopalocera: 20. Epinephele janira Z.; 21. Lycaena semiargus
Rott. To these must be added a fossorial wasp, 22. Ammophila sabulosa Z.,
noticed in 1897 in the island of Amrum, though not previously seen by me on the
North Frisian Islands: all skg.

Alfken observed the following in Juist—

A. Diptera. (a) Stratiomyidae: 1. Sargus cuprarius LZ. (6) Syrphidae: 2.
Syrphus trilineatus Z. B. Hymenoptera. Afzdae: 3. Bombus lucorum Z. ¥, freq.,
skg.; 4. B. muscorum /. §, do.; 5. B. terrester Z. ¥, do.; 6. Colletes marginatus
L. 9, freq., po-cltg. and skg.; 7. Megachile maritima X. 9, skg. and po-cltg., 6 skg.

Verhoeff saw a small Muscid in Norderney.

Schenck noticed the following bees in Nassau.—

1. Ceratina cyanea K.; 2. Macropis labiata 7, and the var. fulvipes /.;
3. Stelis breviuscula Vy/.

Schletterer observed the following Hymenoptera at Pola.—

(2) Apidae: 1. Bombus variabilis Schmzedekn.; 2. Eucera interrupta Baer. ;
3. Halictus minutus K.; 4. H. morbillosus Avrchd.; 5. H. quadricinctus /.; 6. H.
scabiosae Rossz; 7. H. variipes Mor.; 8. Osmia aurulenta Pz.; 9. Prosopis genalis
Ths. (b) Sphegidae: 10. Tachysphex nitidus Spzn.

H. de Vries observed, in the Netherlands, 2 humble-bees (Ned. Kruidk. Arch.,
Nijmegen, 2. ser., 2. deel, 1875),—Bombus subterraneus Z. ¥, and B. terrester Z. ¥.

839. R. odoratus L. (Knuth, ‘Bloemenbiol. Bijdragen.’)— This is a
Canadian species, often cultivated in German gardens as an ornamental shrub.
The flower mechanism is as follows.—At first the numerous still unripe anthers
completely cover the immature stigmas; those of the outer stamens then dehisce,
and pari passu the stigmas become freed by an increase in the diameter of the
flower. An insect dusted with pollen will therefore necessarily effect crossing if it
alights in the middle of the flower, and when such an insect creeps to the mature
peripheral anthers it covers its under-surface with pollen afresh. Dehiscence of
the anthers goes on centripetally, so that the pollen of the innermost may effect
automatic self-pollination by falling upon the stigmas should insect-visits fail. In
spite of the large size of the flower (40-50 mm. in diameter), and of the deep red
colour of the petals, the number of visitors is small. Automatic self-pollination is
not always effective in Germany, for I have very rarely observed the setting of fruits.

Visttors.—I have seen Bombus lapidarius Z. 6, po-cltg., in the island of Riigen.

Loew has noticed a humble-bee in Silesia (‘ Beitrage,’ p. 51) ; also B. hypnorum Z.
, skg., in the Berlin Botanic Garden.

840. R. Idaeus L. (Herm. Miiller, ‘Fertilisation,’ pp. 226-7; Kirchner,
‘Flora v. Stuttgart, pp. 450-1; Knuth, ‘Bl. u. Insekt. a. d. Ins. Riigen.’)— In
the flowers of this species the small narrow petals, which Kerner says drop off on
the second day of anthesis, remain erect, according to Hermann Miiller’s account, so

Aa2

356 ANGIOSPERMAE—DICOTYLEDONES

that the stamens are crowded together between them and the carpels, and hence
a nectar-seeking insect can only insert its proboscis into the receptacle. This
mostly results in cross-pollination, for the styles are generally made use of as an
alighting-place. Automatic self-pollination regularly takes place, for some of the
stigmas always come into contact with the anthers.

Visirors.—I saw the following in the island of Riigen: all freq., skg.—

A. Diptera. (a) Muscidae: 1. Lucilia caesar Z.; 2. Scatophaga merdaria LZ.
(8) Syrphidae: 3. Eristalis arbustorum Z.; 4. E. pertinax Scop.; 5. E. sepulcralis, Z.;
6. E. tenax Z.; 7. Helophilus floreus Z. 8; 8. Syritta pipiens Z. B. Hymeno-
ptera. Apidae: 9. Bombus lapidarius Z. ¥; 10. B. terrester Z.9 andy. C. Lepi-
doptera. Rhopalocera: 11. Pieris sp.

Alfken observed the following at Bremen.—

A. Coleoptera. (a) Zrixagidae: 1. Trixagus fumatus /., freq. (4) Ceram-
bycidae: 2. Strangalia nigra Z., very common. B. Diptera. (a) Lmpidae: 3.
Empis tessellata #., very common. (6) Syrphrdae: 4. Ascia podagrica FP. ;
5. Syrphus ribesii Z. C. Hymenoptera. (a) Afzdae: 6. Andrena albicans
Mill.9; 7. A.albicrus K.9; 8. A. fucata Sm. 9, freq., skg., po-cltg.; 9. A. fulvida
Schenck 9, rare, skg. and po-cltg.; ro. A. nigroaenea A. 9; 11. A. parvula X. 9;
12. Apis mellifica Z.; 13. Bombus agrorum /. 9 and ¥; 14. B. derhamellus K. 9,
skg. and po-cltg.; 15. B. hortorum Z. ¥, skg.; 16. B. jonellus K. 9, do.; 17. B.
lapidarius Z. 9, po-cltg. and skg.; 18. B. lucorum Z. 9 and ¥; 19. B. muscorum F”,
skg. and po-cltg.; 20. B. pratorum Z. ¥ and 4, do.; 21. B. proteus Gers/. 9 and ¥,
po-citg., skg.; 22. B. sylvarum Z. 9; 23. B. terrester LZ. 9, skg. and po-cltg.;
24. Coelioxys quadridentata Z. 6, skg.; 25. C. elongata Zep. d; 26. C. rufescens
Lep.d, skg.; 27. Eriades florisomnis Z.9 and é; 28. E. truncorum Z.4; 29. Halictes
calceatus Scop. 9; 30. H. leucopus K. 9; 31. H. levis K. 9; 32. H. minutus KA. 9;
33. H. punctulatus K. 9; 34. H. quadrinotatulus Schenck 9; 35. H. tumulorum Z. 9;
36. Megachile centuncularis Z. 9; 37. Prosopis communis /Vy/. 6; 38. P. confusa
Vyl. 9; 39. Psithyrus rupestris /. 9, skg. (6) Ichneumonidae: 40. Mesostenus
ligator Gr., skg. (c) Vespidae: 41. Odynerus antilope Pz. d; 42. O. parietum Z.
gand 6; 43. Vespa sylvestris Scop. 9.

Schenck (Nassau) saw 2 bees (Andrena fucata Sm., and A. varians X.) and
a wasp (Vespa norwegica F., one 9).

Handlirsch records the fossorial wasp Gorytes mystaceus Z.

von Fricken noticed the following Coleoptera in Westphalia and East Prussia.—

(2) Nitedulidae: 1. Cychramus luteus Ov. (6) Byturidae: 2. Byturus fumatus
f., very freq. (c) Malacodermata: 3. Dasytes niger Z., not rare. (@) Curculionidae :
4. Anthonomus rubi /Zést., freq.

Herm. Miiller observed the following in Westphalia.—

A. Coleoptera. (a) Cerambycidae: 1. Pachyta octomaculata /., freq., nect-
Ikg. and dvg. the flowers. (6) Dermestidae: 2. Byturus fumatus Z., dvg. the anthers
and nect-lkg. B. Diptera. Syrphidae: 3. Rhingia rostrata Z., skg. and po-dvg. ;
4. Volucella pellucens Z.,do.(H.M.). C. Hymenoptera. (a) Afzdae: 5. Andrena
albicrus X., skg. (H. M.); 6. A. nigroaenea X. 6, do. ; 7. Apis mellifica Z. ¥, extremely
common, skg. and po-cltg.; 8. Bombus agrorum /. 9, freq., skg.; 9. B. hortorum
L. 9, po-clig.; ro. B. pratorum ZL. ¥ and 4, freq., skg. and po-cltg.; 11. B. muscorum
F. 9, sky.; 12. B. sylvarum Z. 9, do.; 13. Halictus lucidulus Schenck 9, do.
() Tenthredinidae: 14. Macrophya rustica L.

Herm. Miiller saw a hover-fly, 7 bees, a wasp, and a Lepidopterid in the Alps
(‘ Alpenblumen,’ p. 215).

ROSACEAE 357

Loew (Silesia) saw a wasp, Vespa media Re/z. 9, skg. (‘Beitrage,’ p. 33);
and Warnstorf (Brandenburg) numerous bees and humble-bees. MacLeod (Flanders)
noticed 5 long-tongued bees, 4 short-tongued bees, 2 wasps, an ant, 3 hover-flies,
5 other flies, 5 beetles, and a few Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p. 317). Plateau (Belgium) saw Apis, Bombus hypnorum Z., B. lapidarius Z.; and
also numerous moths, e.g. Scoliopterys libatrix Z.

In Dumfriesshire, Apis (common), 3 humble-bees, and 2 hover-flies were
recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 55).

Morawitz gives a bee, Nomada ochrostoma X., for St. Petersburg.

Friese observed 2 bees in Mecklenburg: 1. Andrena fucata Sm., freq.; 2. A.
fulvida Schenck, rare.

Hoffer gives the following bees for Steiermark,—

1. Bombus agrorum F. 9 and ¥; 2. B. hypnorum Z. ¥, occasional; 3. B.
pratorum, Z. 9 and ¥, very common, 6; 4. B. terrester Z. 6.

Schmiedeknecht, on the authority of S. Brauns, records Bombus jonellus X. 6.

Fic. 111. Rubus saxatilis, L. (after Herm. Miller). 4. Lateral view of flower (x 7).
&. The same, in longitudinal section.

841. R. spectabilis Pursh._—

Visirors.—Alfken observed the following bees at Bremen : allskg. and po-cltg.—

1. Apis mellifica Z. ¥, freq.; 2. Bombus jonellus A. 9; 3. B. lucorum Z. 9;
4. B. muscorum F. 9; 5. Podalirius acervorum ZL. 9.

842. R. saxatilis L—Hermann Miiller (‘ Alpenblumen,’ pp. 215-16) describes
the white flowers of this species as protogynous, with persistent stigmas. The
stamens are about 40 in number. When the flowers open the stigmas are already
mature: the outer stamens now become erect, and their anthers dehisce, while the
inner ones remain curved inwards, so that the stigmas are for a time protected
against automatic self-pollination. As the petals incline together above the nectar-
secreting receptacle only a small passage of access is left. Visitors at first effect
cross-pollination, subsequently self-pollination as well. The latter always takes place
automatically should insect-visits fail (Miiller, Warming).

Visirors.—Herm. Miller observed 3 bees and an Empid in Switzerland.

358 ANGIOSPERMAE—DICOTYLEDONES

843. R. Chamaemorus L.—This species is dioecious. The white flowers
appear to be visited but seldom, as fruits are rarely seen. According to Warming,
propagation is largely effected by underground shoots. Vahl says that in some parts
of Greenland female plants alone occur, and in other parts only males. This state-
ment is confirmed by Andersson and Hesselman (‘Bidrag till Kanned. om Spets-
bergens o. Beeren Eil. Karlvaxtflora,’ p. 18). Such an arrangement is intelligible as
a result of the marked vegetative reproduction.

In Spitzbergen the species is usually sterile. Female flowers only were seen
there in bloom (7.8.83 and 23. 7.98), but no fruits were observed (Andersson and
Hesselman, loc. cit.).

Schulz states that female flowers with almost normal stamens are found in the
Riesengebirge.

Visitors.—Schneider observed the humble-bees Bombus alpinus Z. and B.
scrimshiranus £. in Arctic Norway (Mus. Aars. Tromsg, xvii, 1895).

In Dumfriesshire an Empid and 3 Muscids were recorded (Scott-Elliot, ‘Flora
of Dumfriesshire,’ p. 57).

Fic. 112. Rubus arcticus, L. (after E. Warming). 4. A flower seen from above. 4. Longitudinal
section through a flower; the outermost anthers are ripe. C, D, &. Stamens of different series. £. An
outermost one. ¥F. A carpel.

844. R. arcticus L.—According to Warming’s investigations at Bosekop
(Norway) the dark rose-red stellate flowers of this species are slightly protandrous
(‘ Arkt. Vaxt. Biol.,’ pp. 37-40). The outer anthers dehisce early, and the stamens
are more closely crowded together above the stigmas than in other species of the
genus, so that automatic self-pollination necessarily takes place. Propagation is
chiefly effected by subterranean shoots.

845. R. serpens Weihe.—
VisiTors.—Loew saw Apis (skg. and po-cltg.) in the Berlin Botanic Garden.

246. Dryas L.

Flowers white; protogynous, homogamous, or protandrous; with concealed
nectar, secreted by a fleshy ring internal to the insertion of the stamens. Not rarely
androdioecious or andromonoecious.

846. D. octopetala L. (Ricca, Atti Soc. ital. sc. nat., Milano, xiv, 1871; Herm.
Miller, ‘Alpenblumen, pp. 227-8; Schulz, ‘Beitrage’; Lindman, ‘Bidrag till
Kanned. om Skandin. Fjallvaxt Blomn. o. Befrukt.’ ; Warming, ‘ Arkt. Vaxt. Biol.’}—

ROSACEAE 359

In this species, besides the hermaphrodite flowers described by Ricca, A. Schulz,
and Lindman as protogynous, and by Hermann Miller as functionally protandrous,
there are unisexual ones distributed androdioeciously (Miiller), or andromonoeciously
(Schulz), and these male flowers are on an average smaller than those which are
hermaphrodite. Failing insect-visits, pollen must fall upon the stigmas from the
stamens (which almost reach their level) owing to the oblique position of the flowers,
so that automatic self-pollination results. On the other hand, owing to the pro-
jection of the stigmas, crossing is usually effected by such visits.

Kerner (‘ Nat. Hist. Pl.” Eng. Ed. 1, II, p. 381) describes the flowers as proto-
gynous, and gives the following account of their mechanism.—The stamens, which
curve inwards in the bud, straighten themselves only a short time before the anthers
dehisce. The anthers of the outermost whorls dehisce first, so that to begin with
autogamy is excluded owing to the distance between stigmas and anthers.
At this stage insect visitors which alight in the middle of the flower and creep out-
wards to collect pollen or lick nectar effect cross-pollination. The inner stamens
now straighten themselves, and their anthers dehisce. These are now at the same
level as the stigmas, which are still receptive, and consequently get self-pollinated.
This is favoured by the outward inclination of the external carpels. The automatic
self-pollination of the stigmas of the inner carpels, should insect-visits fail, is brought
about at the end of anthesis by the bending of the peduncle to an extent that brings
them into the line of fall of the pollen.

On the Dovrefjeld, Lindman noticed protogyny followed by homogamy, but
not dioecism. Warming, on the other hand, observed that in Greenland the flower
mechanism is the same as in Europe. Ekstam gives 1o-25 mm. for the diameter
of the odourless, homogamous flowers examined in Nova Zemlia. Automatic self-
pollination is possible.

The species flowers in Spitzbergen from the end of June to the end of August,
or even longer, and sometimes sets numerous fruits (Andersson and Hesselman,
‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ pp. 20-1).
Ekstam (‘Bliitenbiol. Beob. a. Spitzbergen,’ p. 10) says that the flowers there are
faintly fragrant, and 20-7 mm. (exceptionally 28-38 mm.) in diameter. Self-
pollination can easily take place, for the anthers dehisce almost immediately over
the stigmas. The peduncle elongates considerably after anthesis.

The variety unguzculata, discovered by Nathorst, possesses a form of flower
which deviates markedly from the ordinary type, for the petals narrow at their bases
into a claw 4 mm. long, and radiate squarrosely, while the petals are bent inwards
(Andersson and Hesselman, op. cit.).

According to Abromeit, the flowers of the variety zufermedia Nathorst, which
unites the group-forms D. octopetala Z. and D. integrifolia Vahl, exhibit slight
protogyny and slight protandry, as well as homogamy (‘ Bot. Ergeb. von Drygalsky’s
Grénlandsexped.’).

Visrrors.—Herm. Miiller saw in the Alps numerous bees (especially Halictus

sp.), and Diptera (particularly Muscids), as well as some beetles and Lepidoptera.

rey (Switzerland) noticed the Tineids Ergatis heliacella A-S., and Tinagma
dryadis Ségr.

360 ANGIOSPERMAE—DICOTYLEDONES

von Dalla Torre (Tyrol) saw the bee Halictoides dentiventris My/. 9, and
Schletterer records the same visitor.

Lindman (Dovrefjeld) observed 2 species of flies; and Holmgren (Spitzbergen)
Hymenoptera (Hemiteles septentrionalis Holmgr., and Orthocentrus pedestris
Hfolmgr.) and Diptera (Aricia (Spilogaster) dorsata Zef//., A. (Chortophila) megastoma
Bohem., and Scaeva dryadis Holmgr.).

Ekstam noticed several small and medium-sized flies in Nova Zemlia. MacLeod
saw a short-tongued bee, a hover-fly, and a Muscid in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, p. 427).

847. D. integrifolia Vahl.— Warming observed this species in Greenland, and
states that both hermaphrodite and male flowers occur there (‘ Bestévningsmaade af

Fic. 113. Dryas integrifolta, Vah/ (from Greenland: after E. Warming). A. Entire plant (x 2).
B. Hermaphrodite flower. C. Male flower. D. Protogynous hermaphrodite flower. £. A carpel.
F. Tip of a style with germinating pollen-grains. G. Anther. #7. A foliage-leaf.
nogle grénlandske Blomst.,’ pp. 127-8). The hermaphrodite flowers are homoga-
mous or slightly protogynous, or even slightly protandrous: automatic self-pollination
can readily take place. (Cf. Fig. 113.)

247. Geum L.

Flowers yellow; protogynous or rarely homogamous; with concealed nectar
secreted in the receptacle. Sometimes androdioecious or andromonoecious.

848. G. rivale Z. (Herm. Miller, ‘ Fertilisation,’ p. 229; Kerner, ‘ Nat. Hist.
Pl, Eng. Ed. 1, I1; Loew, ‘Bliitenbiol. Floristik,’ p. 390; Schulz, ‘ Beitrage,’ I,
pp. 33-4; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896; Knuth, ‘ Bloemenbiol.
Bijdragen.’)—In the flowers of this species the calyx is brownish-red, and the
petals bright yellow streaked with red; nectar is secreted in numerous drops in
the receptacle. According to Hermann Miller, the hermaphrodite flowers are
slightly protogynous: at first the mature stigmas project beyond the still unripe

ROSACEAE 361

anthers, so that at this stage cross-pollination must result from insect-visits. The
stamens now elongate, bringing the pollen-covered anthers to the level of the outer
stigmas, so that when the flower closes automatic self-pollination is readily effected.

Insect visitors usually hang on to the flowers from below, by means of the
second and third legs, and thrust their heads and first legs into them. Some
humble bees, especially Bombus terrester Z., also steal nectar from the outside,
without rendering any service in return. The proboscis is inserted between the
calyx and petals.

A. Schulz states that, besides hermaphrodite flowers, there are unisexual ones,
distributed androdioeciously or andromonoeciously. These male flowers are of the
same size as the ordinary kind. There is a group of vestigial carpels in the middle
of the stamens. Warnstorf also observed andromonoecism and androdioecism, the
male flowers, however, being much smaller than the others. He describes the pollen-
grains as being of a beautiful yellow colour, very irregular, roundish, tetrahedral
or ovoid, slightly tuberculated, up to 43 » long and 25 » broad. Warnstorf also
states that the flowers are very frequently visited and robbed by humble-bees.

Fic. 114. Geum rivale, L. (from nature; enlarged). A. Flower (after removal of the anterior
sepals and petals) in the first stage, with stigmas (s) mature, and anthers (a) still unripe. B. The same,
in the second stage, with stigmas still receptive, and anthers partly dehisced (a”), partly still unripe (@').

Visitors.—Herm. Miiller (H. M.) in Westphalia, Loew (L.) in Brandenburg,
and myself (Kn.) in Schleswig-Holstein observed the following.—

A. Coleoptera. Midulidae: 1. Meligethes, freq., completely covered with
pollen (H.M., Kn.). B. Diptera. Syrpfidae: 2. Eristalis nemorum L., po-dvg.
(Kn.); 3. Rhingia rostrata Z., skg. and po-dvg. (H. M., Kn.). C. Hymenoptera.
Apidae: 4. Andrena helvola Z. 9, vainly seeking for nectar (H. M.); 5. Apis melli-
fica L. ¥, freq., skg. from the outside (H.M., L., Kn.); 6. Bombus agrorum fF. 9,
skg. (H. M., L., Kn.);_ 7. B. confusus Schenck 9, do. (H. M.); 8. B. distinguendus
Mor. 9, very occasionally, skg. (H. M.); 9. B. hortorum ZL. 9 and ¥, very common,
skg. (H. M.); 10. B. hypnorum JZ. 9, skg. (H. M.); 11. B. lapidarius Z. 9, do.
(H. M., L., Kn.); 12. B. pratorum ZL. 9, do., ¥ also po-cltg. (H. M.); 13. B.scrim-
shiranus K. ¥ and 9, skg. (H. M.); 14. B. muscorum F”. 9, do. (H. M.); 15. B.
sylvarum L. 9, freq., skg. and po-cltg. (H.M.); 16. B. terrester L. 9; skg. (H. M.).

Herm. Miiller saw 2 humble-bees in the Alps (‘ Alpenblumen,’ p. 227).
Gerstiicker observed Osmia bicolor Schr. 9, po-cltg., at Berlin.

362 ANGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire 2 humble-bees and a hover-fly (very common) were recorded
(Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 54).

Schneider saw Bombus hypnorum Z. ¥ and ¢ in Arctic Norway (Mus. Aarsh.
Tromss, xvii, 1895).

849. G. urbanum L. (Herm. Miller, ‘ Fertilisation,’ p. 230; Schulz,
‘ Beitrage,’ I, p. 34, I, p.186; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.’ p. 66.)—
In the much smaller golden-yellow flowers of this species, according to Hermann
Miiller, nectar is secreted by a green fleshy ring situated internal to the filaments.
When the flower opens, the stamens are curved inwards, so that the anthers lie close
to the outer carpels, while the inner stigmas are mature and project from the middle
of the flower. The outermost stamens at once begin to curve outwards, their anthers
dehisce, and turn their pollen-covered sides upwards; the more central stamens then
do the same. The innermost ones almost always shed some of their pollen upon the
outermost stigmas. Cross-pollination is therefore assured by early insect-visits, and
even later is favoured by the position of the stigmas. Usually, however, automatic
self-pollination comes into play, the number of visitors being very small owing to the
time of flowering, and because the flowers are associated with many others that are
more conspicuous.

Schulz states that, besides hermaphrodite flowers, there are also male ones
distributed andromonoeciously or more rarely androdioeciously. The same investi-
gator observed large-flowered and small-flowered forms; the former as a rule
markedly protogynous, the latter homogamous. The two are connected by inter-
mediate stages.

VisiTors.—Herm. Miiller observed the following.—

A. Coleoptera. Dermestidae: 1. Byturus fumatus /, po-dvg. B. Diptera.
Syrphidae: 2, Melithreptus scriptus Z., skg. and po-cltg. ;

I saw the humble-bee Bombus terrester Z. ¥, skg., in Kiel (20. 6.’97).

MacLeod observed a Muscid in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi,
1894, p. 310).

Verhoeff noticed the following in Norderney.—

A. Coleoptera. Jalacodermata: 1. Dasytes plumbeus J/il/., po-dvg.
B. Diptera. (a) Aluscidae: 2. Aricia incana Wredem., a 9. (6) Syrphidae: 3.
Melanostoma mellina Z., a 9, po-dvg.

In Dumfriesshire 4 muscids were recorded (Scott-Elliott, ‘Flora of Dumfries-
shire,’ p. 54).

850. G. reptans L. (Herm. Miiller, ‘Alpenblumen,’ pp. 225-6.)—The
hermaphrodite flowers of this species are markedly protogynous. They grow to
such an extent during anthesis that the original diameter of 12-15 mm. is increased
to 30-35 mm. Kerner states that there are two forms of the hermaphrodite flowers,
one with short and the other with long stamens. Besides these flowers there are
others, which have become purely male by degeneration of the carpels. They occur
either on stocks of their own or are associated with hermaphrodite flowers.

851. G. montanum L. (Ricca, Atti Soc. ital. sc. nat., Milano, xiv, 1871;
Schulz, ‘Beitrage,’ I, p. 33; Herm. Miiller, ‘Alpenblumen, pp. 226-7.) — The
mechanism of this species agrees with that of G. reptans, but the flowers are usually

ROSACEAE 363

somewhat smaller. Here again Kerner distinguished two forms, Besides the
hermaphrodite flowers, there are also unisexual ones distributed androdioeciously
and andromonoeciously (Schulz).

Kerner (‘ Nat. Hist. Pl.” Eng. Ed. 1, II, p. 381) states that the flowers behave
like those of Dryas octopetala (q. v.), so that at first cross-pollination is favoured ;
subsequently they become autogamous.

Visirors.—Herm. Miller observed in the Alps numerous Diptera (especially
Syrphidae and Muscidae), as well as some bees, Lepidoptera, and beetles. MacLeod
saw 3 Hymenoptera and 4 Diptera in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent,
ili, 1891, p. 427). ,

852. G. coccineum Sibth. et Sm.—The flower mechanism of this species
agrees with that of G. montanum.

Visirors.—Loew observed the bee Halictus nitidiusculus X. 9, po-cltg., in the
Berlin Botanic Garden.

853. G. japonicum Thunbg.—

VisiTors.—Loew saw the bee Halictus sexnotatus X. 9, po-cltg., in the Berlin
Botanic Garden.

854. G. inclinatum Schleich

Visirors.—Loew saw the honey-bee Apis mellifica Z. ¥, among the stamens,
skg., in the Berlin Botanic Garden.

248. Waldsteinia Willd.

Flowers protogynous, with concealed nectar, secreted just below the insertions
of the stamens at the upper margin of the top-shaped receptacle.

855. W. geoides Willd. (Loew, ‘ Bliitenbiol. Beitrage,’ I, pp. 14-16.)—This
species ranges from Galicia and the Siebengebirge to the Crimea. The nectar is
covered by small heart-shaped plates that project at an acute angle from the bases of
the stamens. The flower mechanism is similar to that of Geum rivale.

Visrrors.—Loew saw a small bee (Halictus nitidiusculus A.) and a flower-fly
(Anthomyia), both skg., in the Berlin Botanic Garden.

856. W. trifolia Koch. (Loew, op. cit.)—This species belongs to the Sieben-
gebirge and Siberia. Its flower mechanism is similar to that of W. geoides, but
nectar-covers are wanting.

857. W. fragarioides Tratt— The flower mechanism of this North American
species agrees with that of W. trifolia.

VisiTors.—Loew (op. cit.) observed the same insects as in W. geoides.

249. Fragaria L.

Flowers white, protogynous, with concealed nectar. This is secreted by a
narrow fleshy ring on the receptacle, which is covered by the outer carpels from
within, and the stamens from without. Unisexual flowers are also present.

858. F. vesca L. (Herm. Miller, ‘ Fertilisation, pp. 230-1, ‘ Weit. Beob., ITI,
p- 241, ‘Alpenblumen,’ p. 216 ; Schulz, ‘ Beitrage,’ II, p. 187; Millardet, Mém. Soc.

364. ANGIOSPERMAE—DICOTYLEDONES

sci. phys. nat., Bordeaux, Ser. 4, iv, 1894; Knuth, ‘Bloemenbiol. Bijdragen.’)—
Hermann Miiller was the first to describe the mechanism of the protogynous
hermaphrodite flowers of this species. The petals spread out flat, and therefore
offer to insects a convenient alighting-place and platform. As the anthers dehisce
long after the stigmas mature, insects probing down to the nectar-ring usually effect
cross-pollination. In the absence of visits, automatic self-pollination generally takes
place, owing to the oblique position of the flower. Schulz observed gynomonoecism
and gynodioecism, as well as andromonoecism and androdioecism. Darwin says
that in the United States, among the numerous cultivated varieties of the strawberry,
plants of three kinds are recognized by growers, i.e. 1. female, which produce
fruits in great abundance; 2. hermaphrodite, which yield a scanty crop; 3. male,
which of course set no fruits. Such forms may also be recognized among plants
cultivated in Germany, but purely male individuals are rare.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) observed the
following.—

A. Coleoptera. (a) Cerambycidae: 1. Grammoptera ruficornis /., not infre-
quent, nect-lkg., and dvg. the anthers; even when pairing the 9 continues to gnaw
an anther (H.M.). (6) Dermestidae: 2. Anthrenus pimpinellae F., nect-lkg. (H. M.);
3. A. scrophulariae Z., do. (H. M.). (c) Zelephoridae: 4. Dasytes flavipes /.,
nect-lkg. and dvg. the anthers (H. M.); 5. Malachius bipustulatus Z., do. (H. M.).
(@) Mordellidae: 6. Mordella aculeata Z., nect-lkg. (H. M.). (e) Mitedulidae :
7. Meligethes, freq. (H. M.). B. Diptera. (a) Empzdae: 8. Empis chioptera
Fall, skg. (H. M.); 9. E. livida Z., do. (H. M.); (6) A@uscidae: 10. Anthomyia sp.
(H. M.); 11. Musca corvina #. (H. M.); 12. Scatophaga merdaria Fall., skg.
(H.M.). (c) Syrphidae: 13. Eristalis sepulcralis Z., skg. (H.M.); 14. Melithreptus
menthastri £., do. (H.M.); 15. Paragus bicolor F., skg. and po-dvg. (Budd.);
16. Rhingia rostrata Z., skg. (H. M.); 17. Syritta pipiens Z., freq., skg. (H. M.);
18. Syrphus sp., skg.(H. M.). C. Hymenoptera. (a) Apidae: 19. Andrena dorsata
&. 9, po-cltg. (H. M.); 20. Apis mellifica Z. ¥, do. (H. M.); 21. Halictus leucopus
XK. 9, skg. and po-cltg. (Budd.); 22. H. lucidulus Schenck 9, skg. (H. M.); 23. H.
sexstrigatus Schenck 9 (H. M.); 24. Nomada ruficornis Z. 9, skg. (H. M.); 25. N.
tuficornis L., var. signata Jur. 4, skg. (H. M.); 26. N. sexfasciata Pz. 6 (H. M.);
27. Prosopis communis yl. 9 (H.M.). (4) Formicidae: 28. Myrmica levinodis Wy.
¥%, nect-Ikg. (H. M.). (c) Sphegzdae: 29. Oxybelus uniglumis Z., nect-lkg. (H. M.).
D. Thysanoptera. 30. Thrips, freq., skg. (H. M.).

Herm. Miiller saw 6 Hymenoptera, 2 beetles, 8 flies, and a bug in the Alps.

Alfken noticed a Syrphid (Pipiza sp.), freq., at Bremen.

Friese observed the bee Osmia caerulescens Z., not rare, in Mecklenburg ; and
Schenck the following bees in Nassau.—

1. Andrena flessae Pz.; 2. Halictus albipes #9; 3. H. calceatus Scop.; 4. H.
morio /.; 5. Osmia bicolor Schr.

MacLeod (Pyrenees) saw an ant and a beetle (Bot. Jaarb. Dodonaea, Ghent, iii,
1891, p. 432); and (Flanders) a short-tongued bee, 2 Muscids, an Empid, and
4 beetles (op. cit., vi, 1894, pp. 312, 380).

In Dumfriesshire, a hover-fly and 2 Muscids were recorded (Scott-Elliott, ‘Flora
of Dumfriesshire,’ p. 57).

In the Berlin Botanic Garden, on the var. semperflorens Hayne, Loew observed
2 Syrphids (Eristalis aeneus Scop., skg., and Syritta pipiens Z., po-cltg.).

ROSACEAE 365

859. F. elatior Ehrh. (=F. moschata Duchesne)—The mechanism of the
flowers of this species, according to Kirchner (‘Flora v. Stuttgart,’ p. 442), agrees
with that of F. vesca. Schulz (‘ Beitrage,’ II, p. 187), observed andromonoecism
and androdioecism, rarely gynodioecism and gynomonoecism. The species is
sporadically purely dioecious, in some places there are only male-pleogamous or
female-pleogamous stocks, while elsewhere 10 % or more of the flowers are
hermaphrodite.

Visttors.—Herm. Miiller observed beetles, a wasp, a Lepidopterid, a Bombylius,
and 4 hover-flies in the Alps (‘Alpenblumen,’ p. 216). Loew noticed a Syrphid
(Chrysogaster coemeteriorum Z.), skg., in the Berlin Botanic Garden.

860. F. collina Ehrh. (=F. viridis Duchesne)—The yellowish-white flowers
of this species are imperfectly dioecious; pseudo-hermaphrodite plants usually pre-
dominate. Schulz (‘Beitrage,’ II, p. 187) observed androdioecism and gynodioe-
cism, and also andromonoecism and gynomonoecism. Kirchner (‘Flora v. Stuttgart,’
p. 441) states that the male and female flowers differ in size; in the former the
stamens are twice as long as the group of carpels, in the latter the anthers (which do
not dehisce) are at the same level.

VistTors.—Loew observed 2 Syrphids (Eristalis nemorum Z., po-dvg., and E.
sepulcralis Z.) in the Berlin Botanic Garden.

250. Comarum L.

Dark purple-red protandrous flowers, with nectar half- or completely concealed
and secreted in the usual place.

861. C. palustre L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.” pp. 67-8, 154,
‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 234.)—The flowers of this species
are about 24 cm. in diameter, and secrete abundant nectar from a green ridge-like
disk situated between the stamens and the carpels. The large sepals are dark
purple-red to almost brown internally, while the much smaller petals are rather
brighter in colour. The anthers dehisce laterally some time after the perianth has
unfolded. The stamens, which number about twenty and are in two whorls, are at
first erect, the inner anthers being above the carpels, so that pollen must fall upon
the stigmas. As these are still immature, however, such self-pollination is not
effective. After the anthers have fallen off, the filaments bend back towards the
perianth, leaving the space they occupied in the middle of the flower free for
the styles. These meanwhile increase in length, and their tips are beset with small
yellow papillae. It follows that insect-visits must effect crossing, while automatic
self-pollination is excluded, even when some of the pollen remains adhering to
the stigmas.

After fertilization has been effected the broad sepals shut together above, and
with them the small (5 mm. long, 13 mm. broad) pointed petals, so that the opening
of the flower is closed, and the less conspicuous reddish-green outer surfaces of the
sepals once more become visible (as in the bud), The epicalyx remains perpendi-
cular to the peduncle. Warnstorf describes the pollen-grains as yellowish in colour,
smooth, spherical, and 25-31 p in diameter.

366 ANGIOSPERMAE—DICOTYLEDONES

Visirors.—In the island of Rém I saw the honey-bee, skg., and Muscids
(especially Aricia lardaria #.) at Kiel. In the island of Féhr I observed another
Muscid (Nemoraea consobrina J//g.), as well as a Lepidopterid (Epinephele janira Z.).

Heinsius noticed the following in Holland.—

A. Diptera. (a) Muscrdae: 1. Aricia incana Wired. $; 2. Lucilia caesar Z. $;
(2) Stratiomyidae: 3. Odontomyia viridula #. 8 and 9; (c) Syrphidae: 4. Eristalis
pertinax Scop. 9; 5. Helophilus lineatus #. 8; 6. Tropidia milesiformis Fall. 9.
B. Hymenoptera. Afzdae: 7. Bombus scrimshiranus X. ¥. C. Lepidoptera.
Rhopalocera: 8. Epinephele janira Z. (Bot. Jaarb. Dodonaea, Ghent, iv, 1892, p. 65).

Schneider saw Bombus nivalis Daf/d. ¥ in Arctic Norway (Mus. Aars. Tromsg,
Xvii, 1895).
251. Potentilla L.

Flowers usually homogamous, yellow or white, with half-concealed nectar,
secreted as a rule only as a thin, annular, shining layer on the inner wall of the
receptacle, and not forming actual drops.

862. P. Anserina L. (Herm. Miller, ‘Fertilisation,’ p. 233, ‘ Weit. Beob.,’
II, p. 242; Schulz, ‘Beitrage,’ II, pp. 187-8; Knuth, ‘Bl. u. Insekt. a. d. nordfr.
Ins.,’ pp. 68, 154.)—Hermann Miiller describes the yellow flowers of this species as
homogamous, but Schulz says that they may also be slightly protandrous or slightly
protogynous. The former states that there is a thin layer of nectar on the annular
part of the inner wall of the receptacle that surrounds the roots of the filaments,
and is of a dark or sometimes reddish-yellow colour. Insect visitors sometimes
alight in the centre, sometimes upon the petals. In the former case, they effect
cross-pollination ; in the latter, they often do not touch the stigmas (which are then too
near the middle), but only the anthers, that are covered all over with pollen. In
dull weather the flowers are half shut, and during the night close completely, so that
automatic self-pollination necessarily takes place if insect-visits fail.

Schulz also observed gynomonoecism and gynodioecism.

Visitors.—I saw 2 flies (Eristalis and Anthomyia) in the island of Fohr, and
one (Eristalis arbustorum Z. 4, po-dvg.) in the island of Riigen.

Wiistnei noticed Andrena pilipes /%ér. in the island of Alsen.

Herm. Miiller observed the following, of which the short-tongued bees were
commonest.—

A. Coleoptera. (a) Mitedulidae: 1. Meligethes, freq. (6) Staphylinidae :
2. Tachyporus sp., nect-Ikg. (c) Yelephoridae: 3. Dasytes sp., nect-lkg. B. Diptera.
Muscidae: 4. Anthomyia sp. 9, skg.; 5. Scatophaga merdaria &., do. C. Hemi-
ptera. 6. Aphanus vulgaris Schzi/., skg. D. Hymenoptera. (a2) Apidae: 7. Apis
mellifica Z. 3, skg. ; 8. Halictus flavipes /. 9, po-cltg.; 9. H. sexstrigatus Schenck 9,
do.; 10. H. zonulus Sm. 9, skg.; 11. Sphecodes gibbus Z., do. (4) Formicidae:

12. Lasius niger Z. ¥, necl-Ikg. (c) Sphegrdae: 13. Oxybelus bellus Dahid.;
14. O. uniglumis Z.

Herm. Miiller also observed Bombus terrestet, po-cltg., in the Alps (‘ Alpen-
blumen,’ p. 221); and von Dalla Torre saw two bees in the Tyrol—1z. Andrena
proxima A.6; 2. Melecta luctuosa Scop. 9. The latter is recorded by Schletterer
for the same region.

Loew noticed the Syrphid Pyrophaena rosarum F., skg., in Silesia (‘ Beitrage,
p- 30).

ROSACEAE 367

Verhoeff observed the following in Norderney.—

A. Coleoptera, (a) Nitédulidae: 1. Meligethes aeneus F. (5) Staphylinidae :
2. Tachyporus hypnorum Z. B. Diptera. (a) Empidae: 3. Hilara quadrivittata
Mg.,skg. (b) Muscidae: 4. Anthomyia sp., freq., skg. and po-dvg.; 5. Aricia incana
Wiedem. 6, skg. and po-dvg.; 6. Calliphora erythrocephala AZg.; 7. Cyrtoneura
hortorum Fa//. $, po-dvg.; 8. Lucilia caesar LZ. 9 and 4, skg.; 9. Myospila medita-
bunda /.; 10. Scatophaga stercoraria Z.d; 11. Sepsis cynipsea Z. (c) Syrphidae :
12. Cheilosia sp. 9; 13. Chrysogaster macquarti Zéw 9; 14. Eristalis arbustorum
L. 9, skg.; 15. E. intricarius Z.; 16. Helophilus pendulus Z. 9; 17. Melithreptus
menthastri Z. 9, po-dvg. ; 18. Pipizella virens /. 6; 19. Platycheirus clypeatus Ag.
3, skg.; 20. P. peltatus M/g. 6; 21. Syrphus ribesiiZ. (d@) Therevidae: 22. Thereva
anilis Z.6. C. Hymenoptera. (2) Apidae: 23. Andrena albicans Mull. 9, skg.
and po-cltg.; 24. Colletes cunicularius Z. 9, do.; 25. Osmia maritima /7rzese.
(6) Chrystdidae: 26. Holopyga ovata Dahib. (c) Formicidae: 24. Formica fusca L.
¥, skg.; 28. Lasius niger Z. ¥, do. D. Lepidoptera. Lycaenidae: 29. Polyom-
matus phlaeas Z.

Leege noticed the following in Juist.—

A. Diptera. Syrphidae: 1. Volucella bombylans Z., very common. B. Hy-
menoptera. (a) Aprdae: 2. Andrena albicans Aull. 8, freq., skg.; 3. A. albicrus
XK. 9, freq., skg. and po-cltg.; 4. Colletes cunicularius Z., rare; 5. Halictus rubi-
cundus Chr. 9, once; 6. Nomada ruficornis Z. 4, do.; 7. Prosopis brevicornis JVyl.
é, rare. (6) Spheg¢dae: 8. Ammophila sabulosa Z., rare, skg.

MacLeod saw 3 Muscids in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
P- 313):

In Dumfriesshire several Diptera were recorded (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 59).

863. P. Wiemanniana Guenth. Grab. et Wimm. (= P. Guentheri Pof/).—

Vistrors.—F. F. Kohl observed the ruby-wasp Chrysis dichrora Ag. in the
Tyrol.

864. P. reptans L. (Herm. Miiller, ‘ Fertilisation, pp. 232-3, ‘ Weit. Beob.,’
Il, p. 2413; Schulz, ‘ Beitrage,’ Il, pp. 187-8 ; ‘ Knuth, ‘ Bloemenbiol. Bijdragen.’)—
The flower mechanism of this species agrees with that of P. Anserina. Schulz
noticed gynomonoecism and gynodioecism.

Visrrors.—I only observed Volucella bombylans Z. and Meligethes.

Herm. Miller gives the following list, partly after Buddeberg (Budd.).—

A. Coleoptera. 1. Notoxys monoceros Z., in the flowers in large numbers.
B. Diptera. (a) Emprdae: 2. Empis livida Z., skg. (Thuringia). (6) Auscrdae:
3. Aricia sp., skg. (Thuringia). (c) Syrphzdae: 4. Eristalis arbustorum Z., skg.
(Thuringia); 5. Syritta pipiens Z., skg. and po-dvg. (Thuringia); 6. Syrphus
arcuatus Fall, po-dvg. C. Hymenoptera. (a) Apidae: 7. Andrena albicrus
K.%; 8. A. nana X. 4, skg.; 9. Halictus cylindricus /. 9, do.; 10. H. flavipes /.
9, skg. and po-cltg. (Budd.) ; 11. H. leucozonius Schr. 9, po-cltg.; 12. H. maculatus
Sm. 9 and , skg. and po-cltg.; 13. H. sexstrigatus Schenck 9, do.; 14. H. tetrazonius
Kig.9 and , do. (Thuringia, Budd.); 15. Nomada flavoguttata K. 9, skg. (Budd.) ;
16. N. succincta Pz. , do.; 17. N. xanthosticta K. 4, do.; 18. Prosopis hyalinata
Sm., do.; 19. Sphecodes gibbus Z. 8, do. (4) Sphegidae: 20. Ammophila sabulosa
L.8; 21. Oxybelus bellus Dahlé., nect-lkg.

MacLeod (Flanders) observed 2 hover-flies and a Muscid (Bot. Jaarb. Dodo-
naea, Ghent, vi, 1894, p. 313), and (Pyrenees) a Muscid (op. cit. iii, 1891, p. 431).

368 ANGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire, Dolichopodids, a Muscid, and 2 hover-flies were recorded
(Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 59).

865. P. sylvestris Neck. (= P. Tormentilla Weck. and Tormentilla erecta Z,).
(Herm. Miller, ‘ Fertilisation,’ pp. 233-4, ‘ Weit. Beob.,’ II, p.242; Loew, ‘ Bliitenbiol.
Floristik,’ p. 393; Schulz, ‘Beitrage,’ I, p. 35 ; Knuth, ‘ Weit. Beob. ii. BI. u. Insekt.
a. d. nordfr. Ins.,’ p. 234.)—The flower mechanism of this species agrees with that
of P. Anserina, but Hermann Miiller says that nectar is secreted more abundantly,
and the anthers are only covered with pollen on their narrow external edges, and
not all over. Schulz states that the flowers, even of the same plant, are partly
homogamous, partly protogynous, and partly slightly protandrous. They vary in size,
as do the stamens in number, and there is also considerable variation in the pistil.

VisiTors.—Herm. Miiller gives the following list —

A. Diptera. (2) Bombylidae: 1, Systoechus sulphureus Afzkan., skg. (6)
Syrphidae: 2. Cheilosia sp., po-dvg.; 3. Chrysotoxum bicinctum Z., in large
numbers; 4. Melithreptus scriptus Z., po-dvg. B. Hymenoptera. Apzdae:

5. Andrena argentata Sm. 9, po-dvg.; 6. A. denticulata K. 9 and 4, skg. and po-cltg. ;
7. A. parvula X. 9, po-cltg. C. Lepidoptera. Rhopalocera: 8. Pieris rapae L.,skg.

Herm. Miiller also saw 2 Lepidoptera, a Muscid, and a beetle in the Alps
(‘ Alpenblumen,’ p. 222).

In the island of Féhr I noticed a bee, Andrena tibialis A. 9, skg.

Loew saw a Leptid (Leptis sp.) and 2 Syrphids (Syrphus cinctellus Ze//., and
Didea intermedia Zoew, skg.) in Silesia (‘ Beitrige,’ pp. 30, 49).

Alfken and Hoppner (H.) give the following list of bees for Bremen.—

1. Andrena shawella A. 9, po-cltg. and skg.; 2. A.tarsata Vy/.(H.); 3. Dufourea

vulgaris Schenck (H.); 4. Nomada jacobaeae Pz. 9 and 6; 5. N. obtusifrons Wy/. 9
and é; 6. N. solidaginis Pz. 9 and 4.

Verhoeff saw the following in Norderney.—

A. Diptera. (a) Dolichopodidae: 1. Dolichopus sp. (6) Syrphidae: 2.
Melithreptus taematus (7/g., a 6, po-dvg. B. Hymenoptera. Afrdae: 3. Colletes
cunicularis Z., a 9, po-cltg.

MacLeod observed (Flanders) 7 short-tongued bees, 2 other short-tongued
Hymenopterids, 3 Syrphids, 4 Muscids, and a beetle (Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, p. 314); also (Pyrenees) 2 Syrphids and 2 Muscids (op. cit., iii, 1891,
P- 432).

In Dumfriesshire a humble-bee, a short-tongued bee, and several flies were
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 58).

Willis and Burkill (‘Fls. and Insects in Gt. Britain,’ Part I) observed the following
in Central Wales.—

A. Diptera. (@) Muscrdae: 1. Anthomyia radicum Z., skg.; 2. Lucilia
cornicina ¥., do.; 3. Siphona geniculata Deg., do. (4) Syrphidae: 4. Eristalis
horticola Z.,skg.; 5 Sphaerophoria scriptaZ.,do. B. Lepidoptera. Rhopalocera:
6. Polyommatus phlaeas L.

And for the neighbourhood of the south coast of Scotland they record.—

A. Coleoptera. Scarabaerdae: 1. Aphodius contaminatus Herés/., resting on
the flowers. B. Diptera. (a) Muscidae: 2. Anthomyia radicum Z., very freq.,
skg. and po-dvg.; 3. Cyrtoneura curvipes Meg., skg.; 4. Hydrellia griseola Fad/.,
po-dvg.; 5. Hylemyia lasciva Zed, do.; 6. Oscinis frit Z., do. (6) Syrphrdac:
7. Sphaerophoria scripta Z., po-dvg.; 8. Syritta pipiens Z., do.

ROSACEAE 369

866. P. argentea L. (Herm. Miiller, ‘ Weit. Beob., II, p. 242; Knuth, ‘BI.
u. Insekt. a. d. nordfr. Ins.,’ p. 154.)—

Visirors.—I only observed Meligethes in the island of Sylt, but bees and flies
are the actual pollinators. Herm. Miiller (H. M.) in Thuringia, and Buddeberg
(Budd.) in Nassau, observed the following.—

A. Coleoptera. (a) Buprestidae: 1. Anthaxia quadripunctata Z. (H. M.);
2. Coraebus elatus /. (H.M.). (4) Witidulidae: 3. Meligethes, nect-lkg. (H. M.).
B. Diptera. (a) Muscidae: 4. Anthomyia sp. 9, freq., skg. (H. M.); 5. Aricia sp.,
skg. (H. M.); 6. Ulidia erythrophthalma J/g., in large numbers, skg. (H. M.).
(5) Syrphidae: 7. Paragus bicolor /, skg.(Budd.). C. Hymenoptera. (a) Apidae:
8. Andrena dorsata A. 9, skg. and po-cltg. (Budd.) ; 9. Halictus leucopus &. 9, skg.
(Budd.); 10. H. maculatus Sm. 9, do. (H. M.); 11. H. morio #. 9, do. (Budd.);
12. H. villosulus A’. 9, skg. and po-cltg. (Budd.); 13. Nomada fabriciana Z. 9, skg.
(Budd.); 14. Prosopis communis /Vy/. 9, do. (H. M.); 15. Stelis breviuscula /Vy/. 9,
do. (H.M.). (4) Lvaniidae: 16. Foenus affectator /., nect-lkg. (H. M.).

Fic. 115. Potentzlla minima, Laller f. (after Herm. Miller). 4. Flower, seen trom above (x 7)
B. Longitudinal section through the same. C. Upper part of a stamen, with laterally dehiscing
anther (x 35). a, epicalyx; 6, calyx; c, corolla: d, stamen; ¢, yellow nectar-secreting fleshy ring, into
which the stamens are inserted ;_/ inner orange-coloured part of this ring, on which a layer of nectar is
found; g, ring of hairs by which nectar is secreted, 4, orange-coloured blotch at base of petal (nectar-
guide); z, carpel.

867. P. procumbens Sibth.—

Visitors.—MacLeod noticed 3 hover-flies and a Siricid in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 313-14); and Verhoeff saw a Sphegid (Oxybelus
uniglumis Z., skg.), in Norderney.

868. P. minima Hall. f. (Herm. Miiller, ‘ Alpenblumen,’ p. 217.)—The small
yellow blossoms of this species are homogamous, but at the beginning of anthesis
the petals are not quite fully expanded, so that the middle of the flower is the most
convenient alighting-place, and visitors which come dusted with pollen from an older
flower may effect either cross- or self-pollination. Failing insect-visits, automatic

DAVIS. II Bb

370 ANGIOSPERMAE—DICOTYLEDONES

self-pollination takes place, and this would frequently appear to be resorted to, for
Hermann Miiller only observed 2 Muscids and a micro-Lepidopterid as visitors.
(Cf. Fig. 115.)

Visitors.— Vide supra.

869. P. salisburgensis Haenke (=P. alpestris Hal/. and P. maculata Powrr.).
(Herm. Miiller, ‘Alpenblumen,’ p. 218.)—The flowers of this species agree as regards
their mechanism with those of P. minima, but are larger, so that insect visitors are
more numerous.

Visirors.—Herm. Miiller observed 8 Muscids, 7 Syrphids, 2 beetles, 3 bees,
and 3 Lepidopterids. (Cf. P. verna.)

MacLeod noticed a short-tongued bee, a Syrphid, and a Muscid in the Pyrenees
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 431).

In Dumfriesshire an Empid and 2 Muscids were recorded (Scott-Elliot, ‘ Flora of
Dumfriesshire,’ p. 58).

870. P. aurea L. (Herm. Miiller, ‘Alpenblumen,’ pp. 218-19 ; Schulz, ‘ Bei-
trage,’ II, p. 68; Loew, ‘Bliitenbiol. Floristik,’ p. 397.)—The flowers agree in
mechanism with those of P. salisburgensis, but are still larger, so that the visitors are
still more numerous. Schulz observed gynomonoecism and gynodioecism, and
according to him the hermaphrodite flowers in the Riesengebirge are slightly
protogynous.

Visitors.—Herm. Miiller observed 18 Muscids, 8 hover-flies, 3 beetles, 7 bees,
and 15 Lepidopterids; and Loew noticed 2 hover-flies, a Muscid, a Lepidopterid,
and a beetle.

Loew records the following for Switzerland (‘ Beitrage,’ p. 57).—

A. Coleoptera. JJalacodermata: 1. Dasytes alpigradus Kiesew. B. Diptera.
(a) Muscidae: 2. Anthomyia sp. (4) Syrphidae: 3. Cheilosia brachysoma Lgg. (?) ;

4. Pelecocera scaevoides all. C. Lepidoptera. Zygaenidae: 5. Zygaena
exulans Achw.

871. P. frigida Vill_—The flower mechanism of this Alpine species, according
to Kirchner (‘ Beitrage,’ p. 39), agrees essentially with that of P. minima, though the
corolla does not spread out flat, but is saucer-shaped. Hence the diameter of
the flower is only 7-10 mm., although each of the yellow petals is 5 mm. long.
The base of each petal is marked with an orange-yellow blotch. The flowers are
homogamous: dehiscence of the anthers begins externally, and continues centri-
petally, so that at the beginning of anthesis cross-pollination is more readily effected
by insect-visits than in P. minima. Later on, when the anthers of the inner stamens
have dehisced, automatic self-pollination is inevitable.

872. P. multifida L—The flowers of this Alpine species agree as regards
their mechanism (Kirchner, loc. cit.) with those of P. frigida, but all the anthers
are ripe at the beginning of anthesis. As they are close to the stigmas automatic
self-pollination is inevitable.

Plants with buds, flowers, and ripe fruits were found (6. 8.’82) at Cape Thordsen
in Spitzbergen (Andersson and Hesselman, ‘ Bidrag till Kanned. om Spetsbergens o.
Beeren Eil. Karlvaxtflora,’ p. 20).

ROSACEAE 371

Visitors.— Loew (Jahrb. Kgl. bot. Gart. zu Berlin, iv, 1886, p. 159) observed
a po-dvg. hover-fly (Eristalis sepulcralis Z.) on cultivated plants in the Berlin Botanic
Garden.

873. P. supina L.—This species agrees with P. Anserina as regards its flower
mechanism, and the occurrence of gynomonoecism and gynodioecism (Schulz,
‘ Beitrage,’ IT, p. 187).

874. P. recta L.—Kerner states that the flowers of this species open between
11 and 12 o’clock in the morning, and that the petals fall off on the second day of
anthesis.

875. P. grandiflora L. (Herm. Miiller, ‘Alpenblumen,’ 219-20.)—The flowers
of this species are even larger than those of P. aurea; they are also borne on longer
stalks, which make them still more conspicuous. The number of insect visitors is
therefore greater. The flowers are protandrous, so that at first crossing is favoured,
and automatic self-pollination is rarer.

Visitors,—Herm. Miiller observed 10 Muscids, 4 Syrphids, 4 beetles, 12 bees,
and 13 Lepidopterids.

876. P. verna L.—According to Hermann Miller (‘ Fertilisation,’ pp. 231-2),
the flowers agree as regards mechanism with those of P. Anserina ; but their size is
less. They are of a brilliant yellow colour, and rich in nectar. Schulz describes
them as slightly protogynous. The inner anthers are situated above the stigmas, so
that automatic self-pollination regularly takes place.

VisiTors.—Schulz observed numerous flies, beetles, and small bees in Central
Germany; and Lindman noticed flies, a beetle, and a butterfly on the Dovrefjeld.

In the Alps Herm. Miiller (‘Alpenblumen,’ p. 221) saw 17 Hymenopterids,
2 Dipterids, a beetle, and 2 Lepidopterids visiting this species, and with P. alpestris
Hall. f,, the large-flowered form of P. salisburgensis (cf. p. 370).

I noticed a bee (Andrena albicans A///. 9), and a hover-fly (Eristalis tenax Z.),
both skg., in Schleswig-Holstein (‘Bloemenbiol. Bijdragen’). Herm. Miiller gives
the following list for Westphalia and Thuringia.—

A. Coleoptera. (a) Curculionidae: 1. Spermophagus cardui Sfev. (6) WVite-
dulidae: 2. Meligethes freq., nect-lkg. B. Diptera. (@) Muscidae: 3. Onesia
cognata Mg., skg.; 4. O. floralis R.D., do.; 5. Pollenia vespillo ¥., do. (4) S/ratio-
myidae: 6. Odontomyia argentata F., skg. (c) SyrpAidae : 7. Cheilosia modesta Ege.,
skg.; 8. C. praecox Ze#/., freq., skg.; y. Rhingia rostrata Z., skg.; 10. Syritta
pipiens Z.,do.; 11. Syrphus sp., do. C. Hymenoptera. (a) Apidae: 12. Andrena
albicans JZ#l/., 9 and 4, freq., po-cltg. and skg.; 13. A. albicrus A. 4, skg.; 14. A.
argentata Sm. 6, do.; 15. A. dorsata X. 9, po-cltg.; 16. A. fulvicrus £. 8, skg.;
17. A. parvula KX. 9, do.; 18. A. nana A. é,do.; 19. A. xanthura A. 9, do.; 20.
Apis mellifica Z. ¥, po-cltg. and skg.; 21. Bombus terrester Z. 9, po-clig.; 22.
Halictus albipes 7. 9, skg.; 23. H. cylindricus J. 9, po-cltg.; 24. H. flavipes /.
9, skg.; 25. H. leucopus K. 9, skg. and po-cltg.; 26. H. maculatus Sm. 9, po-cltg. ;
27. H. morio £. 9, skg.; 28. H. nitidiusculus #. 9, do.; 29. H. semipunctulatus
Schenck 9, do.; 30. H. sexstrigatus Schenck 9, po-cltg.; 31. Nomada ruficornis L. 4;
32. Osmia fusca Chr. 9, skg. and po-cltg. (4) Formicidae: 33. Formica praten-
sis Deg. ¥, nect-lkg.

Bb2

372 ANGIOSPERMAE—DICOTYLEDONES

Bees were also observed as follows by the authorities and in the places stated. —

Alfken (Bremen)—1. Andrena albicans J/ull. 9, occasional, skg. and po-cltg. ;
2. A. albicrus K. 9, freq., skg. and po-cltg.; 3. Nomada bifida Zhs. 9, do. Schmiede-
knecht (Thuringia)—1. Andrena cyanescens MVyl.; 2. A. parvula A. Schenck
(Nassau)—Andrena cingulata /. Friese (Hungary), Andrena genevensis, Schmzedekn. ;
(Innsbruck)—Osmia bicolor Schr. 9, only po-cltg. von Dalla Torre (Tyrol)—
1. Andrena parvula K.9; 2. Halictus nanulus Schenck $; 3. Osmia aurulenta Pz.
and 9; 4. Prosopis borealis Vy/. 9. Schletterer (Tyrol)—1. Halictus albipes /- ;
2. Prosopis communis /Vy/.

877. P. cinerea Chaix (=P. arenaria Borck.), and

878. P. opaca L.—Schulz (‘Beitrage,’ II, pp. 67-8) states that these two
species agree with P. verna as regards flower mechanism, and are visited by numerous
insects, especially flies, beetles, and the smaller or more rarely the larger bees. Most
of the visitors collect or devour pollen.

Vistrors.—Loew (Brandenburg) gives the following list for P. cinerea (‘ Beitrage,’
p- 38).—

A. Diptera. Syrphidae: 1. Cheilosia praecox Ze/t., po-dvg. B. Hymeno-

ptera. Apidae: 2. Halictus morio /. 9, po-cltg.; 3. H. tumulorum Z. 9, do.; 4.
Osmia bicolor Schr. 4, skg.

Schletterer observed the following bees at Pola

1. Andrena parvula &.; 2. Halictus calceatus Scof.; 3. H. interruptus Pz. ;
4. H. levigatus X.9; 5. H. malachurus X.; 6. H. morio /.; 7. H. quadrinotatus
&.; 8. Osmia versicolor Lr.

879. P. caulescens L.—The flowers of this species are homogamous, or
(according to Kerner) slightly protogynous. Visitors may effect either cross- or self-
pollination, and the latter can also take place automatically (Schulz).

Visttors.—Herm. Miiller observed Apis, a Bombus, and a Melithreptus (‘ Alpen-
blumen,’ p. 222).

880. P. atrosanguinea Lodd.—Delpino (‘Ult. Oss.,’ p. 233) describes the
flowers of this species as protogynous, with stigmas that are only receptive for a
short time. At first the stamens are curved away from the middle of the flower
and their anthers are unripe, while the mature stigmas occupy the centre. The
stamens subsequently raise themselves to the level of the stigmas.

Visitors.—Delpino observed small bees (sp. of Andrena and Halictus).

881. P. fruticosa L. (Herm. Miiller, ‘ Fertilisation,’ p. 233; Knuth, ‘ Bloemen-
biol. Bijdragen.’)—In this species, according to Hermann Miiller, nectar is so scantily
secreted that it does not accumulate into little drops; but the smooth glistening
ring of the receptacle surrounding the roots of the filaments is so frequently licked
by insects, even by the honey-bee, that it is doubtless covered by a thin layer of it.

The flowers are homogamous. When insects alight they sometimes first touch
the stigmas, sometimes the laterally dehiscing anthers. The chances of cross- and
self-pollination are therefore about equal. Failing insect-visits, automatic self-
pollination sometimes takes place, the withering stamens partly bending inwards,
and thus bringing their anthers (to which pollen still clings) into contact with
the stigmas. The number of insect visitors, however, is so large that this kind
of pollination is scarcely likely to occur.

ROSACEAE 373

Vistrors.—Herm. Miiller (H. M.), and myself (Kn.) observed the following. —

A. Coleoptera. (a) Malacodermata: 1. Dasytes flavipes /., nect-lkg. and dvg.
the anthers (H.M.). (6) Witidulidae: 2. Meligethes, very freq., po-dvg. (H. M.).
B. Diptera. (a) Conopidae: 3. Sicus ferrugineus Z., nect-Ikg. (H.M.). (6) Cuh-
cidae: 4. Culex pipiens Z., skg. (H. M.). (c) Musctdae: 5. Anthomyia, very freq.,
nect-lkg. (H. M.); 6. Lucilia cornicina /,, freq., nect-Ikg. (H.M.); 7. L. sylvarum
Mg., do. (H. M.); 8. Sarcophaga carnaria Z., freq., nect-lkg. (H. M.); 9. Scatophaga
merdaria /, do. (H.M.); 10. Sepsis, very freq., nect-Ikg. (H.M.); 11. Smaller
Muscids, nect-lkg. (Kn.). (@) Stratiomy¢dae: 12. Sargus cuprarius Z., freq., nect-lkg.
and po-dvg. (H.M.). (e) Syrphzdae: 13. Eristalis arbustorum Z., freq., nect-lkg. and
po-dvg. (H.M.); 14. E. nemorum Z., skg. or po-dvg. (Kn.); 15. E. sepulcralis Z.,
do. (H.M.); 16. Helophilus floreus Z., as 13.(H.M.); 17. H. pendulus Z., do.
(H.M.); 18. Melithreptus taeniatus A/g., do. (H.M.); 19. Syritta pipiens Z., do.
(H. M.); 20. Syrphus pyrastri Z., skg. or po-dvg. (Kn.); 21. Volucella bombylans Z.,
do.(Kn.); 22. V. pellucens Z., do.(H.M.). (/) Zabanzdae: 23. Chrysops caecutiens
L. 3, nect-lkg. or po-dvg. (H.M.). C. Hymenoptera. (a) Apidae: 24. Apis
mellifica Z. ¥, freq., nect-lkg. (H.M., Kn.); 25. Halictus zonulus Sm. 9, nect-lkg.
(H.M.). (8) Sphegidae: 26. Oxybelus bellus Dahld., very freq., nect-lkg. (H. M.) ;
27. O. uniglumis Z., occasional, nect-lkg. (H. M.).

882. P. alchemilloides Lapeyr.—MacLeod (‘ Pyreneenbl.,’ pp. 25-9) has fully
described the mechanism of this Pyrenean species. The diameter of the white
flowers is 20 mm. Between the carpels and the bases of the stamens there is a
nectar-groove bounded by two hairy ridges, and most conveniently accessible at five
spots. The flowers are almost homogamous: probably the anthers dehisce a little
before the stigmas are mature. Insect visitors may effect either cross- or self-
pollination. The latter can only take place automatically with difficulty, for the
anthers are turned away from the stigmas; but now and then it is brought about
by a stamen remaining erect.

Visitors.—MacLeod observed 2 Empids and 5 Muscids in the Pyrenees.

883. P. sterilis Garcke (=Fragaria sterilis Z., P. Fragariastrum £4rh., and
P. Fragaria Porr.). (MacLeod, Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 429-39,
vi, 1894, pp. 314-15.)—MacLeod examined the mechanism of the white flowers of
this species in the Pyrenees and in Belgium. In the former region the diameter
of the flower is 20 mm.; in the latter only 11-12 mm. Between the stamens
and the carpels there is a pentagonal hairy orange-coloured (Belgium) or reddish-
brown (Pyrenees) nectar-ring. In late anthesis the stamens bend inwards so much
that automatic self-pollination results from the contact between anthers and stigmas.
In Belgium the flowers are slightly protogynous.

Visitors.—MacLeod observed small beetles and flies, and an Acarid in Flanders ;
and a bee, a Muscid, and a Lepidopterid in the Pyrenees.
Burkill (‘ Fertlsn. of Spring Fls.’) noticed the following on the Yorkshire coast.—

A. Coleoptera. Curculionidae: 1. Apion nigritarse X., skg. B. Diptera.
(2) Muscidae: 2. Coelopa sp., po-dvg.; 3. Lucilia cornicina /., skg.; 4. Onesia
cognata J7fg., do.; 5. Sepsis nigripes Mg., skg.; 6. Siphona geniculata Deg., skg.;
4. one other Muscid. (4) Phoridae: 8. Phora sp. C. Hymenoptera. (a) Apidae:
g. Andrena clarkella K. 6 and 9; 10. A. gwynana X. 4, skg., 9 po-cltg. (4) Formi-
ctdae: 11. Formica fusca Z., skg. (c) Jchneumonidae: 12. three small sp.

884. P. Sommerfeltii Lehm., 885. P. Ranunculus Lange, 886. P. Vah-
liana Lehm., 887. P. Frieseana Lange, and 888. P. tridentata Ait—These and

374 ANGIOSPERMAE—DICOTYLEDONES

other Arctic species (Nos. 860, 872, and 889-891) rapidly set fruits, and are otherwise
well adapted to the conditions of life obtaining in high latitudes. Warming describes
them as probably homogamous, and relying upon automatic self-pollination (cf. Loew,
‘ Bliitenbiol. Floristik,’ p. 99).

889. P. fragiformis Willd., var. parviflora Trautw. (the species=P. emargi-
nata Pursh, according to Abromeit, ‘Bot. Ergeb. von Drygalski’s Grénlandsexped.,’
p. 8).—In Spitzbergen this species is one of the earliest spring plants, and was
observed in flower 21.6.’96, 30. 6.82, 1.7.'98, and 7.7.’61. The last two dates apply
to North-east Land in 80° N. lat. Andersson and Hesselman state that abundant
fruits are set (‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’
pp. 19-20).

The specimens which Vanhdéffen collected in Greenland and described as the
variety (a) fypica (=P. emarginata Pursh) bear flowers 15-17 mm. in diameter,
with dark yellow petals traversed by darker veins. In the variety (c) elator (=P.
fragiformis W7//d.) the flowers are larger, their diameter being not infrequently
20 mm.: petals pale yellow, but darker and distinctly veined at their bases. The
development of the petals is clearly dependent upon oecological conditions.

Ekstam says that this species and its variety parvzflora are also native to
Nova Zemlia (Bot. Jahrb., Leipzig, xxii, 1897, p. 195).

890. P. pulchella R. Br.—This species was found flowering in Spitzbergen
from June to the beginning of September, chiefly, however, in the second half of
July. Fruits are set and ripened in a normal way (Andersson and Hesselman,
‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 19). Several
varieties of this species occur in Greenland (Abromeit, ‘ Bot. Ergeb. von Drygalski’s
Gronlandsexped.,’ pp. 7-8).

891. P. nivea L.—This species was found flowering in Spitzbergen on 12. 7.’90,
and ripe fruits were collected on 6.8.’82 (Andersson and Hesselman, op. cit., p. 20).
It is very polymorphous in Greenland. Here the yellow petals project but little
beyond the calyx : numerous fruits were noticed (Abromeit, op. cit., pp. g—11).

Visitors.—Lindman noticed a medium-sized fly on the Dovrefjeld.

892. P. rupestris L.—This species, which was studied at Bozen by Schulz
(‘ Beitrage, p. 68), is capable of automatic self-pollination, for the inner anthers
incline somewhat towards the middle of the flower, so as to be above the simul-
taneously mature stigmas upon which pollen can therefore fall. As but little nectar
is secreted, and the white flowers are not conspicuous, the number of insect-visits
is small.

VisiTors.—Schulz observed flies, beetles, and bees. MacLeod saw 5 Syrphids
and 5 Muscids (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 431). Loew noticed
Apis, skg., in the Berlin Botanic Garden.

893. P. alba L.—The flower mechanism of this species agrees, according to
Kirchner (‘Flora v. Stuttgart,’ p. 447), with those of P. verna, P. opaca, and P.
Anserina.

894. P. micrantha Ram.—lIn the flowers of this species, according to Kerner,
the stamens form a hollow cone which covers the nectar-secreting receptacle.

ROSACEAE 375

895. P. hirta Vill.—

Visitors.—Schletterer observed the following Hymenoptera at Pola.—

(a) Apidae: 1. Andrena lucens Jmh.; 2. A. thoracica /.; 3. Halictus fasciatellus
Schenck; 4. H. villosulus #.; 5. Prosopis clypearis Schenck. (6) Tenthredintdae :
6. Amasis laeta /.

896. P. delphinensis Gren. et Godr., and 897. P. Kurdica Boiss. et
Hohen.—

Visitors.—Loew saw Apis, skg. and po-cltg., in the Berlin Botanic Garden.
898. P. chrysantha Trevir.—

VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis nemorum Z.; 2. Syritta pipiens Z.,
po-dvg. B. Hymenoptera. Apidae: 3. Apis mellifica Z. ¥, skg. and po-cltg.

899. P. Meyeri Boiss., var. Fenzlii Lehm.—

VisiTors.—Loew saw the bee Prosopis communis WVy/. 9, po-dvg., in the Berlin
Botanic Garden.

252. Sibbaldia L.

Homogamous greenish-yellow flowers, with exposed nectar secreted in the
usual place.

goo. S. procumbens L. (Herm. Miiller, ‘Alpenblumen,’ p. 222.)—In this
species the exposed nectar is secreted by the
broad fleshy disk which surrounds the ten carpels.
It is eagerly visited by short-tongued insects
(Muscids, ants, Ichneumonids), and these effect
cross- and _ self-pollination. The possibility of
automatic self-pollination seems to be excluded,
for though the anthers mature simultaneously
with the stigmas, they are so far from them
that transfer of pollen cannot take place auto-
gamously. Lindman, however, says that self-
pollination is a much easier matter in plants of Ris a16i Scena Pebconehene B.
the species growing in the Scandinavian High- (after Herm. Miller). Flower seen

3 directly from above (x7). a, anther; a&
lands. Warming makes the same statement for saci a &, sepal; #, nectary; p, petal”
Greenland.

253. Alchemilla L.

Small, greenish apetalous flowers; with exposed nectar secreted by a fleshy
ring on the inner wall of the receptacle.

gor. A. vulgaris L. (Herm. Miller, ‘Fertilisation, pp. 234-5; Lindman,
‘ Bidrag till Kanned. om Skandin. Fjellvaxt. Blomn. o. Befrukt.’; Schulz, ‘ Beitrage,’ II,
p. 188; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, IT; Loew, ‘ Bliitenbiol. Floristik,’ p. 396.)—
Schulz states that this species is very commonly gynomonoecious and gynodioecious,
as well as andromonoecious and androdioecious; and that in some districts herma-
phrodite flowers are entirely absent. According to Hermann Miller, the fleshy yellow
ring on the inner side of the receptacle, which surrounds the style at the time of

376 ANGIOSPERMAE—DICOTYLEDONES

flowering, secretes a thin layer of nectar that gives a yellowish appearance to the
whole inflorescence. The stamens and pistil are rarely developed normally: either
the stamens are developed, while the style remains so short that the stigma hardly
projects beyond the nectary, or the style is long, and 1-3 or all four of the
stamens are reduced. Automatic self-pollination is therefore rendered very difficult.
Kerner, however, says that it may take place in the hermaphrodite flowers, which
are protogynous and therefore primarily adapted for cross-pollination, for the stigma
remains receptive till the anthers are ripe, and is brought into contact with them
by elongation of the style.

Visirors.—Herm. Miiller saw a hover-fly (Xanthogramma citrofasciata Deg.),
and Lindman noticed flies on the Dovrefjeld.

Herm. Miller observed 3 Lepidoptera and 6 flies in the Alps (‘ Alpenblumen,’
pp. 223-4); Loew saw the hover-fly Melithreptus scriptus Z. in Switzerland
(‘Beitrage,’ p. 55); and MacLeod noticed a beetle and 5 flies in the Pyrenees
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 433).

Plateau observed flies (Calliphora, Musca, Scatophaga, and Syritta pipiens Z.)
and small Hymenoptera in Belgium.

Fic. 117. Alchemilla vulgaris, L. (after Herm. Miller). (1) Flower with normal stamens and
a short style, seen directly from above. (2) The same, seen obliquely from above. (3) Flower with one
normal and three vestigial stamens, and normal style ; seen obliquely from above. (4) Flower with all
the stamens vestigial, and well-developed style. a, epicalyx ; 4, calyx; c, stamens; c', vestigial stamens;
d, stigma ; ¢, nectary.
In Dumfriesshire, a long-tongued bee, a Tenthredinid, 2 Empids, several other
flies, and 2 Lepidoptera were recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 59).

In spite of repeated watching, I have never observed any insects on the flowers.

go2. A. alpina L., 903. A. fissa Schum., and g04. A. pentaphyllea L.—
Hermann Miller (‘ Alpenblumen,’ pp. 222-3) states that these species agree with
A. vulgaris and Sibbaldia procumbens, as regards their flower mechanism, the
distribution of sexes, and the degeneration of various organs. Trimerous and
pentamerous flowers are not infrequent. (Cf. Fig. 118.)

Visitors.—Herm. Miiller saw small Muscids, beetles, ants, and Ichneumonids.

MacLeod observed (on A. alpina) 2 beetles and 4 flies in the Pyrenees (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 433). Loew noticed a small bee (Sphecodes
gibbus Z. 9), skg., in the Berlin Botanic Garden.

g05. A. arvensis Scop.—The very insignificant green flowers of this species
are associated in small dense sessile glomerate cymes, but in spite of this are extremely
inconspicuous. Kirchner (‘Flora v. Stuttgart,’ p. 449) says there is a nectary, but it
is green and non-functional. The single stamen slants obliquely inwards, so that its
anther is situated above the stigma, and automatic self-pollination is inevitable.

ROSACEAE 377

906. A. acutiloba Stev.—

VistTors.—Loew observed a Bombyliid (Anthrax morio Z., skg.) and a Syrphid
(Eristalis tenax Z., do.) in the Berlin Botanic Garden.

254. Sanguisorba L.

Flowers apetalous, associated in capitula; either with half-concealed nectar, or
anemophilous.

907. S. officinalis L. (Herm. Miller, ‘Fertilisation,’ p. 236, ‘Alpenblumen,’
pp. 224-5.)—In the homogamous flowers of this species the receptacle encloses the
ovary, and possesses a nectar-secreting ring that surrounds the base of the style.
There are four ovate sepals, concave at their bases, and coloured red above. They
serve as nectar receptacles, and also as a means of making the flower more con-
spicuous. The 50-100 florets of a capitulum open in succession from below
upwards in such a way that only a zone one floret deep is in bloom at the same time.
In favourable weather insects appear in considerable numbers, and usually effect

Fic. 118. Alchemeilla fissa, Schum. (after Herm. Miller). A. Hermaphrodite tetramerous
flower. A}. Pistilofsame. 2B. Female tetramerous flower. #'. Pistil of same. C. Hermaphrodite
trimerous flower, with one vestigial stamen. C'. Pistilofsame. DD. Flower in section. ah, epicalyx;

k, calyx; fi, filament; s, sepal; sf, stigma; ov, ovary; #, nectary.
crossing, for as a rule they touch the stigma and anthers with different sides of their
heads. Sometimes, however, they bring about self-pollination, which may easily take
place automatically.

Visitors.—Herm. Miiller observed 4 Muscids, 2 Syrphid, and 11 Lepidoptera in
the Alps. Loew noticed one species of hover-fly (Didea alneti Fad/.) in the same
region, and another (Syritta pipiens Z.) in the Berlin Botanic Garden. Rdéssler saw
a butterfly (Lycaena euphemus /7é.) at Wiesbaden.

In Dumfriesshire 5 Muscids were recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’
p- 60).

Kerner gives the butterfly Lycaena arcas Fo/t., for the caterpillars of which this
species is the food-plant.

908. S. minor Scop. (=Poterium Sanguisorba Z.). (Herm, Miiller, ‘ Fertili-
sation,’ p. 236; Axell, ‘Om Anord. fér Fanerog. Vaxt. Befrukt., p. 54.)—This
species is nectarless, anemophilous, and, according to Kirchner (‘Flora v. Stuttgart,’
Pp. 456-7), coenomonoecious. The male flowers are at the bases of the capitula,
the hermaphrodite ones in the middle, and the female ones at the top. Schulz
‘ Beitrage,’ II, pp. 69-70, 188) observed gynomonoecism, andromonoecism, and even
pure monoecism. The hermaphrodite flowers are usually homogamous, and the
distribution of the sexes varies very much in individual plants. At Ruppin, according

378 ANGIOSPERMAE—DICOTYLEDONES

to Warnstorf (Schr. natw. Ver., Wernigerode, xi, 1896), only the uppermost flowers
of the capitula are female, as a rule, and are in a minority; the other flowers are
male, often with an isolated hermaphrodite flower here and there between them.
The anthers are yellow, and pendulous on long reddish filaments. The hermaphro-
dite flowers possess but few stamens. The pollen-grains are of a dirty yellowish-
white colour, rounded-polyhedral, smooth, up to 37 » in diameter. Ludwig says that
not infrequently there are variations in the colour of the stamens of different male
flowers, from which the anthers hang down limply on long thin filaments. As a rule
the anthers are yellow, and the filaments white; but stocks also occur in which
the filaments are red, and the anthers yellowish-red to red. In the female flowers the
style and the large spreading stigma are red to wax-yellow or white.

Visitors.—Herm. Miiller (‘ Fertilisation,’ p. 236) saw a wasp (Odynerus parie-
tum Z. 9) settle on the flowers, flying away again after a short and profitless search.
I noticed a hover-fly (Melanostoma mellina Z.), po-dvg., on the inflorescence.

In Dumfriesshire a Tenthredinid and 2 hover-flies were recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 61).

Fic. 119. Sangutsorba officinalis, L. (after Herm. Miller). A. Flower seen directly from above.
B. The same, seen from the side. C. The same, in longitudinal section. D. Single sepal, seen from
within. (x 7) a, anther; dr, bract; ca, receptacle; 7, filament; gv, style; 4, drop of nectar:
u, Nectary; ov, ovary; s, sepal; s¢, stigma.
gog. S. alpina Bunge.—In this species, as in Thalictrum aquilegiifolium, the
filaments are claviform, and therefore shaken by the slightest breath of air, so that
the pollen is scattered (Kerner, ‘Nat. Hist. PL.’ II, p. 145).

255. Poterium L.

gio. P. spinosum L.— Pirotta states that this species is anemophilous
(Annuario R. Inst. Bot. Roma, iii, 1887). Hermaphrodite flowers are only to be
found on cultivated plants: wild ones (in Sardinia) bear only unisexual flowers, and
purely female inflorescences are commoner than such as bear both kinds of flowers,
while the number of male flowers in the latter is seldom larger than that of the

female ones. Cultivated plants more frequently bear polygamous inflorescences
than purely female ones.

gu. P. polygama Waldst. et Kit.—Kerner describes this species as trimonoe-
cious. The number of stamens in the hermaphrodite flowers is sometimes reduced
from 8 to 1.

ROSACEAE 379

256. Agrimonia Tourn.
Yellow, homogamous, pollen flowers with pseudo-nectaries.

g12. A. Eupatoria L. (Herm. Miller, ‘ Fertilisation,’ pp. 235-6; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 319-20; Kirchner, ‘ Flora v. Stuttgart,’
p- 457; Knuth, ‘Bloemenbiol. Bijdragen.’)—At the base of the two styles in this
species there is a fleshy ring, that looks like a nectary, though no secretion has been
observed. The 5-7 stamens on the margin of this disk attain the same level as the
stigmas, and their anthers dehisce laterally. The anthers incline inwards, and there-
fore come into contact with the stigmas. The individual flowers bloom for a single
day only, and open very early in the morning. The stamens, which are at first
divergent, bend inwards in the course of the day, until they touch one another and
the stigmas. Comparatively few insects visit the flowers, but these may bring about
either cross- or self-pollination. From the above description it is clear that the
latter occurs automatically, and it is obviously effective.

Visttors.—Herm. Miiller (H. M.) and myself (Kn.) observed the following.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp., po-dvg. (H. M.). (6) Syrphrdae:
2. Ascia podagrica #., po-dvg. (H. M.); 3. Eristalis nemorum Z., do. (Kn.); 4. E.
tenax Z., do. (H.M.); 5. Melanostoma mellina Z., do. (H.M.); 6. Melithreptus
dispar Loew, do. (H.M.); 7. M. pictus M/g., do. (H.M.); 8. M. scriptus Z., do.
(H.M.); 9. M. taeniatus Mg., do. (H. M.); 1o. Rhingia rostrata Z., do. (H. M.);
11. Syritta pipiens Z., do.(H. M.); 12. Syrphus ribesii Z., do. (Kn.). B. Hymeno-
ptera. Aprdae: 13. Apis mellifica Z. Y, po-clig. (Kn.); 14. Bombus terrester Z.
g and ¥, do. (Kn.); 15. Halictus, small sp. 9, do. (H. M.).

Schletterer records Bombus pascuorum Sco. for the Tyrol.
g13. A. odorata Mill—
Visirors.—Alfken noticed bees (Apis and Prosopis sp.) at Bremen.

257. Ulmaria Hill.

White, hermaphrodite, homogamous pollen flowers, devoid of nectar.

914. U. pentapetala Gilib. (= Filipendula Ulmaria Maxim., and Spiraea
Ulmaria Z.). (Herm. Miiller, ‘Fertilisation, pp. 222-3, ‘Weit. Beob., II, p. 243;
Lindman, ‘ Bidrag till Kanned. om Skandin. Fjellvaxt. Blomn. o. Befrukt.’; Schulz,
‘Beitrage, II, p. 186; Knuth, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’
p- 234, ‘Bloemenbiol. Bijdragen.)—

The flowers of this species are yellowish-white in colour, and associated in dense
crowded inflorescences. Their conspicuous appearance, and the strong odour of
almonds they exhale, attract numerous insect visitors, to which a large quantity
of pollen is afforded. According to Hermann Miiller’s account the stamens at first
incline together in the middle of the flower, so as completely to cover the stigmas.
They then gradually erect themselves in centripetal order, and bend somewhat out-
wards. Meanwhile the anthers dehisce, and get covered with pollen all round.
When the innermost stamens have become erect, the centre of the flower occupied
by the stigmas is the most convenient alighting-place for insects. These can
therefore easily effect crossing, though they are just as likely to bring about self-
pollination. Failing insect visitors automatic self-pollination takes place. Owing to

380 ANGIOSPERMAE—DICOTYLEDONES

the crowded nature of the inflorescences automatic cross-pollination may also happen

by the fall of pollen upon the stigmas of neighbouring flowers. Schulz observed
andromonoecious stocks.—

Visrtors.—I observed the following, all po-dvg.—

A. Coleoptera. (a) Cerambycidae: 1. Gaurotes virginea L.; 2. Judolia ceram-
byciformis Schr.; 3. Leptura livida ¥.; 4. L. maculicornis Deg.; 5. Stenocorus
mordax Deg. (6) Chrysomelidae: 6. Cryptocephalus sericeus Z. (c) Scarabaezdae :
7. Trichius fasciatus Z. B. Diptera. Syrphzdae: 8. Syritta pipiens Z., freq.

Herm. Miiller gives the following list.—

A. Coleoptera. (a) Cerambycidae : 1. Judolia cerambyciformis Schr. ; 2. Leptura
maculicornis Deg., dvg. the flowers; 3. Pachyta quadrimaculata Z., dvg. the anthers ;
4. Strangalia attenuata Z.,do.; 5. S. quadrifasciata Z., dvg. the flowers. (4) Clertdae :
6. Trichodes apiarius Z., dvg. the flowers. (c) Dermestidae: 7. Anthrenus pimpi-
nellae F., dvg. the flowers. (d) Afordellidae: 8. Mordella aculeata Z. (e) Mit-
dulidae: 9. Cychramus luteus Oliv. (/) Scarabaeidae: 10. Cetonia aurata L., dvg.
the flowers; 11. Trichius fasciatus Z., do. (g) Telephoridae: 12. Malachius bi-

Ww h | <n
BN); FG a

/F ae pat EG AS
WEA a
Bi

ff 2 Sec

iy

G
5

Fic. 120. Ulmarsra pentapetala, Giltb. (after Herm. Miiller). (1) Young flower.
(2) Older flower. (a) anthers; s/, stigmas.

pustulatus Z., dvg. the anthers. B. Diptera. (@) Afuscidae: 13. Anthomyia sp.
(6) Syrphidae (all po-dvg.): 14. Eristalis arbustorum Z., very freq.; 15. E. horticola
Deg.; 16. E. nemorum Z.; 17. E. sepulcralis Z.; 18. E. tenax Z.; 19. Helo-
philus floreus ; 20. Syritta pipiens Z.; 21. Volucella bombylans Z.; 22. V. pellucens
£. C. Hymenoptera. (2) Apidac: 23. Andrena coitina K. 9, po-cltg.; 24. Apis
mellifica Z. ¥, freq., po-cltg.; 25. Prosopis armillata WVy/. 3, freq., po-dvg.; 26. P.
clypearis Schenck 6, po-dvg.; 27. P. communis JVy/.é, do.; 28. P. confusa Wyi. ,
do. ; 29. Xylocopa violacea ZL. 9, po-cltg. (4) Chrystdidae: 30. Chrysis ignita L. ;
31. Ellampus auratus Z.; 32. Hedrychum nobile Scop. (c) Sphegidae: 33. Pem-
phredon unicolor #.; 34. Crabro larvatus Wesm. 9; 35. C. wesmaeli v.d. LZ. 3.
D. Lepidoptera. 36. Zygaena pilosellae Zsf., trying to suck.

Von Fricken records 2 Cerambycids (Clytus figuratus Scop. and Grammoptera
ruficornis /.) and a Curculionid (Apoderus erythropterus Zschoch.=A. intermedius
li.) for Westphalia and East Prussia. Redtenbacher gives 2 Cerambycids (Molor-
chus minimus Scop. and Obrium brunneum /.) for Austria.

Alfken saw the humble-bee Bombus terrester Z. ¥ at Bremen, and Loew
observed the following in Silesia (‘ Beitrage,’ p. 28).—

ROSACEAE 381

A. Coleoptera. (a) Mordellidae: 1. Anaspis frontalis Z. (5) Nitidulidae :
2. Meligethes sp. (c) Scarabaeidae: 3. Cetonia aurata Z., dvg. the anthers. B.
Diptera. (a) Muscidae: 4. Anthomyia sp. (4) Syrphidae: 5. Chrysogaster
coemeteriorum Z., po-dvg. C. Lepidoptera. Rhopalocera: 6. Argynnis pandora
S. V., vainly searching for nectar.

Willis noticed the following in the neighbourhood of the south coast of Scotland
_(‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera. Nitidulidae: 1. Epuraea melina Z7., po-dvg.; 2. Meligethes
aeneus /”, freq., po-dvg.; 3. M. viridescens F., do. B. Diptera. (a) Muscidae:
4. Anthomyia radicum Z., freq., po-dvg.; s. Mydaea sp., po-dvg.; 6. Tricho-
phthicus hirsutulus Ze/z., do. (4) Syrphidae (all po-dvg.): 7. Eristalis aeneus Scop.;
8. E. horticola Deg.; 9. E. tenax Z.; 10. Melanostoma mellina Z. (c) Chirono-
midae: 11. Corynoneura sp., po-dvg.

Herm. Miiller saw the beetle Cetonia aurata Z., freq., in the Alps (‘Alpen-
blumen,’ p. 228). MacLeod observed (Pyrenees) a short-tongued bee and 2 beetles
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 426-7): also (Flanders) Apis, a humble-
bee, 12 hover-flies, a saw-fly, 3 beetles, and a Lepidopterid (op. cit. vi, 1894,
pp. 321-2, 380). Heinsius noticed 2 po-dvg. flies (Helophilus floreus Z. 9 and
Cyrtoneura curvipes Jacg. 9) in Holland (op. cit., iv, 1892, p. 57). Lindman saw
numerous flies on the Dovrefjeld.

g15. U. Filipendula Hill (=Spiraea Filipendula Z., and Filipendula hexa-
petala Grld.). (Herm. Miller, ‘Fertilisation, pp. 223-4.)—The faintly fragrant
inflorescences of this species attract far fewer insects than the much larger ones
of U. pentapetala. Hermann Miiller (‘ Beitrage,’ II, p. 12) describes the white
petals as possessing such slender claws that they are very easily bent down, and
cannot serve as alighting-places for insects. They are somewhat reflexed when the
flower has fully opened. The stamens also bend far outwards before the anthers
dehisce, so that the 9-12 broad, bilobed styles that radiate from the centre of the
flower serve as the most convenient platform for insect visitors, which therefore
regularly effect crossing. Should insect-visits fail, automatic self-pollination takes
place ; the innermost stamens often remaining bent inwards till their anthers dehisce,
so that the pollen comes into contact with the stigmas. Schulz says that late-flower-
ing stocks are sometimes andromonoecious.

Visttors.—Hermann Miller observed the following.—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia bifasciala AZul/. 9, po-dvg.
(0) Oedemeridae: 2. Oedemera podagrariae Z., po-dvg. (c) Scarabaetdae: 3.Cetonia
aurata Z., gnawing the anthers; 4. Trichius fasciatus Z., rapidly gnawing the
anthers from below upwards. B. Diptera. Syrphidae (all po-dvg.): 5. Eristalis
arbustorum Z.; 6. E. nemorum /.; 7. Helophilus floreus Z.; 8. Syritta pipiens /.
C. Hymenoptera. Aprdae: 9. Halictus sexnotatus A. 9, po-cltg.; 10. H. zonulus
Sm. 9, do.

Loew saw a Syrphid (Eristalis tenax Z., po-dvg.) and a bee (Apis mellifica LZ. ¥,
po-cltg.) in the Berlin Botanic Garden.

258. Spiraea L.

White or red hermaphrodite flowers, often protogynous; mostly smelling like
hawthorn ; with half-concealed nectar secreted in abundance by an annular orange-
yellow thickening on the inner wall of the receptacle, internal to the insertions of the

382 ANGIOSPERMAE—DICOTYLEDONES

stamens. In some species, according to Kerner (‘ Nat. Hist. Pl., Eng. Ed. 1, II,
Pp. 326), geitonogamy takes place. For though the direction of the style and the
position of the stigma remain unchanged the filaments elongate, and curve in such
a way that the pollen can reach the stigmas of neighbouring flowers.

916. S. sorbifolia L. (Herm. Miller, ‘Fertilisation,’ pp. 224-6.)—This
Siberian species, cultivated in our parks and gardens as an ornamental shrub, attracts

3 EQN \ 4

Fic. 121. Sprraea sorbtfolia, L. (after Herm. Miiller). (1) Flower immediately after
opening. (2) Older flower, with some of the stamens dehisced. a, dehisced anthers;
sé, stigma; 2, nectary.
numerous insects by its large fragrant inflorescences, and because of its richness in
nectar and pollen. Hermann Miller says that the flowers are distinctly protogynous,
and that even in the bud the broad capitate stigmas are provided with papillae and
project beyond the stamens, which are bent inwards towards one another. When
the flower opens the stamens gradually become erect, and successively dehisce from
without inwards. Insects therefore effect crossing at the beginning of anthesis; later

ROSACEAE 383

on they may also bring about self-pollination, for the stigmas remain receptive till the
innermost anthers dehisce. The latter may also occur automatically.
Visitors.— Vide infra.

917. S. salicifolia L., and 918. S. ulmifolia Scop. (Herm. Miller, ‘Fertili-
sation,’ pp. 224-6, ‘Weit. Beob.,’ II, p. 243.)—These species are cultivated along
with S. sorbifolia as ornamental shrubs.

Visirors.—Herm. Miiller gives the following list for these two species and
S. sorbifolia, all three of which flower at the same time.—

A. Coleoptera. (a) Cerambycidae: 1. Clytus arietis Z., nect-lkg.; 2. Gram-
moptera ruficornis F., do.; 3. Leptura livida #., very freq., nect-lkg.; 4. Strangalia
armata //ést., nect-lkg.; 5. S. attenuata Z., freq., nect-Ikg.; 6. S. nigra Z., nect-lkg.
(4) Cistelidae: 7. Cistela murina Z., freq., dvg. the anthers and petals. (c) Der-
mestidae: 8. Anthrenus muscorum Z., very freq., nect-lkg.; 9. A. pimpinellae /., do. ;
10, A. scrophulariae Z., do.; 11. Attagenus pellio Z., do.; 12. Byturus fumatus &.,
do. (d) Elateridae: 13. Cardiophorus cinereus Hés/., nect-lkg.; 14. Lacon murinus,
L.,do. (¢) Lagritdae: 15. Lagria hirta Z., nect-lkg. (/) Mordellidae: 16. Anaspis
frontalis Z., freq., nect-lkg.; 17. A. maculata Fourc., nect-lkg. (g) Mitidulrdae :
18. Meligethes, freq. (2) Scarabaerdae: 19. Cetonia aurata Z.; 20. Phyllopertha
horticola Z., dvg. the flowers; 21. Trichius fasciatus Z., do. (7) Zelephoridae:
22. Cantharis fulva Scop.; 23. Dasytes flavipes /.; 24. Malachius bipustulatus Z..,
dvg. the anthers. B. Diptera. (a) Brbionzdae: 25. Bibio hortulanus Z., nect-lkg.
(4) Chironomidae: 26. Ceratopogon sp., in large numbers, skg. (c) Conopidae:
27. Myopa polystigma Rond., skg.; 28. Physocephala rufipes /., do. (d) Empzdae:
29. Empis opaca /., freq., skg.; 30. E. punctata #., skg.; 31. E. tesselata #., very
freq., skg. (e) M/uscidae: 32. Sp. of Anthomyia; 33. Cyrtoneura simplex Loew;
34. Echinomyia fera £.; 35. E. magnicornis Ze//. (Borgstette); 36. Gymnosoma
rotundata Z.; 37. Lucilia cornicina /., skg.; 38. L. sylvarum A/g., do.; 39. Mesem-
brina meridiana Z.; 40. Musca corvina #.; 41. Onesia cognata J/g.; 42. O. floralis
R.-D.; 43. Sarcophaga carnaria Z., skg. (/) Stratiomyidae: 44. Odontomyia
viridula /., skg.; 45. Stratiomys riparia AZg., do. (g) Syrphrdae: 46. Ascia lanceo-
lata Afg., skg.; 47. A. podagrica #., do.; 48. Cheilosia gilvipes Ze//, skg. and
po-dvg.; 49. Chrysogaster viduata Z.; 50. Chrysotoxum festivum /.; 51. Eristalis
arbustorum Z., freq., skg. and po-dvg.; 52. E. intricarius Z., do.; 53. E. nemorum
L., do.; 84. E. pertinax Scop., do.; 55. E. sepulcralis Z., do.; 56. E. tenax L., do.;
57. Helophilus floreus Z., freq., skg.; 58. Melithreptus strigatus S/aeg.; 59. Pipiza
funebris 47g.; 60. Rhingia rostrata Z., in great numbers, skg.; 61. Syritta pipiens
L., do.; 62. Syrphus excisus Ze/f.; 63. S. ribesii Z., po-dvg.; 64. Volucella bombylans
f., var. plumata JZg.; 65. Xylota ignava Pz.; 66. X. lenta A/g.; 67. X. segnis L.
(2) Tabanidae: 68. Chrysops caecutiens Z. $, skg. (2) Tipulidae: 69. Pachyrhina
pratensis Z., nect-lkg. C. Hymenoptera. (a) Apidae: 70. Andrena albicans Ail.
9, freq., skg. and po-cltg.; 71. A. albicrus A. 9 and 4, do.; 72. A. dorsata A’ 9, do.;
73. A. fucata Sm. 9, do.; 74. A. fulvicrus A’ 4, skg.; 75. A. nigroaenea A’. 4, do.;
76. A. parvula A”. 9, freq., skg. and po-cltg.; 77. A. schrankella Ay/. 8, skg.; 78. A.
trimmerana A. 9, do.; 79. Apis mellifica Z. ¥, skg. and po-cltg.; 80. Bombus
muscorum /. 9, po-cltg.; 81. B. scrimshiranus A. ¥, running quickly over the
inflorescences and po-cltg.; 82. B. terrester Z.9, skg. and po-cltg.; 83. Halictus
flavipes 7.9; 84. H. sexnotatus A. 9, po-cltg.; 85. H. sexstrigatus Schenck 9, skg.;
86. H. villosulus A. 9, do.; 87. Nomada ruficornis Z. 9, do.; 88. Osmia rufa L. 9,
po-cltg.; 89. Sphecodes gibbus Z. 9, skg. (Budd.). (4) Chrysrdidae: go. Hedrychum
lucidulum #7. 3. (c) Evanitdae: 91. Foenus sp., nect-lkg. (Budd.). (d) Formicidae :
92. Lasius niger Z. ¥, nect-lkg.; 93. Myrmica levinodis Vy/. ¥; 94. Numerous small
ants, nect-lkg., and preying on the minute black midges often found nect-lkg.
(e) Ichneumontdae: 95. Various sp. (/) Sphegidae: 96. Ammophila sabulosa Z. ;

384 ANGIOSPERMAE—DICOTYLEDONES

97. Cerceris arenaria Z., not infreq.; 98. Crabro lapidarius Pz. 3, skg.; 99. Oxybelus
bellus Dahill., very freq., skg.; 100. O. uniglumis Z., do.; 101. Passaloecus insignis
Shuck. 9, skg.; 102. Psen atratus Pz., do. (g) Pompilidae: 103. Pompilus minutus
Dahlb., skg. (h) Tenthredinidae: 104. Allantus temulus Scop., nect-lkg. (z) Vespidae:
105. Odynerus spinipes Z. D. Lepidoptera. 106. Adela croessella Scop., freq.,
skg.; 107. Dichrorampha plumbagana Zr. E. Neuroptera. 108. Agrion settled
on the flowers not infrequently, apparently only to sun itself; 109, Panorpa com-
munis Z., nect-lkg. F.Orthoptera. 110. Ectobia lapponica Z., nect-lkg.(?).

Loew observed the following on S. salicifolia in Silesia (‘ Beitrage,’ p. 30).—

A. Coleoptera. (a) Malacodermaia: 1. Dasytes flavipes F., nect-lkg. (6) Wite-
dulidae: 2. Meligethes sp. B. Hymenoptera. Vespidae: 3. Odynerus sinuatus
F. 9, skg.

Schenck noticed the bee Andrena gwynana &. in Nassau.

H. de Vries records a bee (Apis mellifica Z. $) and a humble-bee (Bombus
terrester Z. 3) for the Netherlands (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel,

1875).

gig. S. opulifolia LF. Ludwig expresses the opinion (Kosmos, Stuttgart,
xvii, 1885, p. 203) that the red colour assumed by the ovary in this species after the
flower has faded serves to keep away unbidden guests from blossoms that are
still fresh and retain their original hue.

Visitors.—Alfken observed 2 bees (Prosopis communis /Vy/. $, and Andrena
albicans A4@#//. 9) at Bremen; and F. F. Kohl saw a wasp (Odynerus oviventris
Wesm.) in the Tyrol.

920. S. digitata Willd —

Visttors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Scarabaeidae: 1. Cetonia aurata Z., freq., dvg. the flowers ;
2. Phyllopertha horticola Z., do. B. Diptera. Syrphidae: 3. Eristalis nemorum J..,
po-dvg. C. Hymenoptera. Apzdae: 4. Apis mellifica ZL. ¥, po-cltg.

259. Aruncus L,

Yellowish-white dioecious pollen flowers, devoid of nectar, and aggregated into
large inflorescences.

g21. A. sylvestris Kostel. (=Spiraea Aruncus Z.), (Herm. Miiller, ‘ Fertilisa-
tion,’ p. 224.)—The flowers of this species are polygamous-dioecious. Kerner
says :—‘It produces true hermaphrodite flowers and pseudo-hermaphrodite male and
female flowers. The three kinds of flowers are arranged thus: (1) some plants
bear only pseudo-hermaphrodite female flowers, (2) others only pseudo-hermaphrodite
male flowers, (3) some bear both hermaphrodite flowers and pseudo-hermaphrodite
male flowers, and (4) in addition there are yet other plants whose flowers are all
hermaphrodite’ (‘ Nat. Hist. Pl.’ Eng. Ed. 1, I, p. 300).

Visitors.—Herm. Miiller (loc. cit. and ‘ Weit. Beob.,’ IJ, p. 243) gives the
following list.—

A. Coleoptera. (2) Dermestidae: 1. Anthrenus claviger Zr., occasional ; 2. A.
muscorum Z.; 3. A. pimpinellae /., very freq.; 4. A. scrophulariae Z., not infreq. ;
5. Attagenus schaefferi Ast. (4) Nitrdulzdae: 6. Meligethes freq. B. Diptera.
(a) Muscidae: 7. Sp. of Anthomyia, po-dvg. (4) Syrphidae: 8. Syritta pipiens /.,
very freq., po-dvg. C. Hymenoptera. (a) Apidae: 9. Prosopis armillata My. ;

ROSACEAE 385

to. P. clypearis Schenck 3, freq., po-dvg.; 11. P. communis Wy. 8, do.; 12. P.
signata Pz.9 and 4, po-dvg. (4) Sphegzdae: 13. Oxybelus bellus Dahlb., po-dvg. ;
14. QO. uniglumis Z. (c) Vespidae: 15. Odynerus sinuatus 7.
Herm. Miiller saw a Cerambycid in the Alps (‘ Alpenblumen,’ p. 228).
Sickmann noticed the following Hymenoptera at Osnabriick.—

Hymenoptera, Sphegzdae: 1. Crabro cetratus Shuck.; 2. C. chrysostoma
Sep., freq.; 3. C. dives H.-Sch., rare; 4. C. leucostoma Z., infreq.; 5. Psen atratus
Pz., freq.

Von Dalla Torre observed the following bees in the Tyrol.—

1. Andrena albicrus X. 9 and ; 2. Osmia leucomelaena X. 8 and 9; 3.
Prosopis borealis Wy/. 9 and $; 4. P. nigrita ¥.; 5. P. bipunctata F%ér. (also
recorded by Schletterer).

260. Kerria DC.

Homogamous pollen flowers.

922. K, japonica DC.—The flowers of this species are odourless and nectarless.
Their mechanism is described as follows by Kirchner (‘Beitrage, p. 40). The
anthers of the most external stamens ripen, and the stigmas mature in the bud.
The inner stamens are short in proportion to their nearness to the centre of
the flower. At first they are curved inwards, but later on become erect. The
markedly diverging styles attain a length almost equal to that of the longest
stamens. Self-pollination is therefore inevitable, and apparently takes place even
before the flower opens. Focke states that the plant is self-sterile in Europe, but
produces succulent fruits in Central China, its native region (Abh. natw. Ver.,
Bremen, xiv, 1897). The petals are at first of an orange-yellow colour, but become
inconspicuous before the innermost anthers are ripe.

VistTors.—Kirchner did not see any.

261. Mespilus L.

White, conspicuous, homogamous hermaphrodite flowers; with half-concealed
nectar secreted by a yellow fleshy ring in the receptacle, internal to the stamens.

923. M. germanica L.—Kirchner (‘Flora v. Stuttgart,’ p. 427) states that
when the white flowers of this species open the five styles lie close together, but their
stigmas are already mature, and directed outwards. The stamens are inclined inwards,
and the innermost anthers are situated beneath the stigmas, while the others are
at the same or a higher level, and as they dehisce introrsely automatic self-pollination
must regularly take place. Crossing is only possible at later stages, when the
stamens incline more outwards and the styles curve away from one another above.

262. Crataegus L.

White, protogynous flowers smelling like herring-brine (of trimethylamide). The
nectar is half-concealed and secreted by a ring in the receptacle. On account of
their odour the blossoms are referred to the class of nauseous flowers, visited by
flies that are fond of putrefying substances.

924. C. Oxyacantha L. (Herm. Miller, ‘Fertilisation, pp. 240-1, ‘ Weit.
Beob.,’ II, p. 239; Kirchner, ‘Flora v. Stuttgart,’ p. 426; Loew, ‘Bliitenbiol.

DAVIS. 11 cc

386 ANGIOSPERMAE—DICOTYLEDONES

Floristik, pp. 388-9; Knuth, ‘Weit. Beob. i. Bl. u. Insekt. a. d. nordfr. Ins.,’
p- 234.)—According to Hermann Miiller’s account of this species the stigmas are
already mature when the flower opens, and project from its centre; but the anthers
are still unripe. The outer stamens are erect, but the inner ones are so bent
inwards as to lie beneath the stigmas. After from one to two days the outermost
anthers begin to dehisce, and get covered with pollen all round.

In cold dull weather the inner stamens remain incurved, while the outer ones
project beyond the stigmas, and are bent inwards to an extent that renders automatic
self-pollination easy. In warm sunshine, on the other hand, the stamens spread
out away from the stigmas, so that the nectar previously concealed by the woolly
hairs on the bases of the styles comes into view. Owing to the protogyny of the
flower, nectar-seeking insects always effect crossing at the beginning of anthesis,
and generally do so during later stages.

Visttors.—I observed the following in the island of Pellworm (4. 6.’93).—

A. Diptera. (a) Muscidae: 1. Scatophaga sp.; 2. Larger and smaller Muscids.
(2) Syrphidae: 3. Helophilus pendulus Z.; 4. Rhingia sp.; 5. Syritta pipiens Z.
B. Hymenoptera. Apidae: 6. Andrena albicans AZiéll.9; 7. Apis mellifica LZ. ;
8. Bombus terrester Z. ; all skg.

Wiistnei noticed Andrena trimmerana X. in the island of Alsen.

Alfken records the following for Bremen.—

A. Diptera. (a) Empidae: 1. Empis ciliata F.; 2. E. opaca /.; 3. E. tesse-
lata F. (2) Muscidae: 4. Cynomyia mortuorum Z.; 5. Cyrtoneura hortorum falZ. ;
6. Lucilia caesar Z.; 7. Scatophaga stercoraria Z. (c) Syrphzdae: 8. Ascia lanceolata
Mg.; 9. A. podagrica #.; 10. Eristalis arbustorum Z.; 11. Helophilus pendulus Z. ;
12. Melanostoma mellina Z.; 13. Syritta pipiens Z.; 14. Syrphus pyrastri Z.; 15. S.
ribesii Z. B. Hymenoptera. (a) Agzdae: 16. Andrena albicans Afu//l. 9; 17. A.
albicrus AK. 9; 18. A. carbonaria Z.$; 19. A. cineraria Z.9; 20. A. fucata Sm. $;
21. A. humilis Jmz.9; 22. A. nigroaenea K.9; 23. A. parvula K.9; 24. A. pro-
pinqua Schenck 9; 25. A. trimmerana K. 9; 26. A. varians A. 9; 27. Bombus
hortorum Z. 9, po-clig.; 28. Eriades florisomnis Z. 9 and 4; 29. Halictus calceatus
Scop., var. elegans Lep. 9; 30. H. levis K.9; 31. H. morio /. 9; 32. H. rubicundus
Chr. 8, po-cltg.; 33. H. sexnotatulus Vy/.9; 34. Osmia rufa Z.9; 35. Psithyrus
vestalis Fourcr.9, skg. (3) Zenthredinidae: 36. Pamphilius sylvaticus Z. (c) Vespidae:
37. Vespa germanica F.; 38. V. sylvestris Scop. 9.

Von Fricken saw a Cantharid (Cantharis haemorrhoidalis /.), a Cerambycid
(Grammoptera ruficornis ¥.), and a Chrysomelid (Cryptocephalus violaceus Zazch.)
in Westphalia and East Prussia.

Loew observed the following in Brandenburg (‘ Beitrage,’ p. 36).—

A. Coleoptera. (a) Cerambycidae: 1. Molorchus minor L. (8) Dermestidae:
2. Anthrenus scrophulariae Z. (c) Mordellidae: 3. Anaspis frontalis Z. (d) Ano-
bidae: 4. Anobium paniceum LZ. (e) Scarabaeidae: 5. Cetonia aurata L. (f
Telephoridae: 6. Cantharis rustica Fa//.; 7. Malachius bipustulatus Z. B. Diptera.
(a) Empidae: 8. Empis sp. (4) Afusc¢dae: 9. Anthomyia pluvialis Z.; 10. Hydrotaea
ciliata #. (c) Syrphidae: 11. Criorhina oxyacanthae AZg., skg.; 12. Syritta pipiens
L., do. ©. Hymenoptera. Afzdae: 13. Andrena albicans Aull. 9; 14. A.
propinqua Schenck 9; 15. A. tibialis A. 9; 16. Halictus sexnotatus A.9; 17. Nomada
ruficornis Z.; 18. Osmia bicornis Z. 9; all skg.

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list —

A. Coleoptera. (a) Buprestidae: 1. Anthaxia nitidula Z. (H.M.). (3) Ceram-
byctdae: 2. Clytus mysticus Z., nect-lkg. (H.M.); 3. Grammoptera ruficornis &,

ROSACEAE 387

freq., nect-lkg. (H.M.). (c) Chrysomelidae: 4. Clytra cyanea #., damaging the
petals (H.M.). (d) Dermestidae: 5. Anthrenus claviger Zr., occasional, nect-lkg.
(H.M.); 6. A. pimpinellae /., very freq., nect-lkg. (H.M.); 7. A. scrophulariae Z..,
freq., nect-Ikg. (H. M.); 8. Attagenus pellio Z., nect-Ikg. (H.M.). (e) Mordellidae:
g. Anaspis frontalis Z., nect-lkg. (H.M.); 10. Mordellistena abdominalis F., do.
(H.M.). (/) Mitidulidae: 11. Epuraea sp., nect-lkg. (H. M.); 12. Meligethes, very
freq., nect-lkg. (H.M.). (g) Ocdemertdae: 13. Asclera coerulea Z. (H.M.). (A)
Scarabaetdae: 14. Oxythyrea stictica Z., dvg. the anthers (H. M.). (2) Tedephoridae :
15. Malachius elegans O2.(?), dvg. the anthers (H. M.); 16. Cantharis testacea L.
(H.M.). B. Diptera. (a) Brbiontdae: 17. Bibio marci L., skg. (H.M.); 18. Dilo-
phus vulgaris J/g., very freq. (H.M.). (6) Emprdae: 19. Empis livida Z., in very
large numbers, skg. (Kn.); 20. E. opaca /., freq., skg. (H.M.); 21. E. punctata F.,
do. (H.M.); 22. Microphorus velutinus Jacg.(H.M.); 23. Tachydromia connexa
Mg., freq. (H.M.). (c) Muscidae: 24. Aricia serva Mg. (H.M.); 25. Cyrtoneura
sp. (H.M.); 26. Echinomyia fera Z. (H.M.); 27. Graphomyia maculata Scop.
(H.M.); 28. Mesembrina meridiana Z. (H.M.); 29. Onesia floralis A.-D., skg.
(H.M.); 30. O. sepulcralis AZg., do. (H.M.); 31. Sarcophaga carnaria Z., do.
(H.M.). (d@) Syrphidae: 32. Eristalis arbustorum Z., very freq., skg. and po-dvg.
(H.M.); 33. E. intricarius Z., skg. and po-dvg. (H.M.); 34. E. nemorum Z.,
do. (H.M.); 35. E. pertinax Scop., do. (H.M.); 36. E. sepulcralis Z., do. (H. M.);
37. E. tenax Z., do. (H.M.); 38. Helophilus floreus Z., freq. (H.M.); 39. H.
pendulus Z., do. (H.M.); 40. Pipiza notata Mg. (H.M.); 41. Rhingia rostrata L.,
freq., skg. (H. M.); 42. Xylota segnis Z. (H.M.). C. Hymenoptera. Apzdae:
43. Andrena albicans J/z//. 9 and 4, exceedingly numerous, skg. and po-cltg. (H.M.);
44. A. atriceps K. 9 and 4, do. (H.M.); 45. A. chrysosceles X. 9, do. (H. M.); 46.
A. connectens A. 9, do. (H.M.); 47. A. dorsata A. 9, do. (H.M.); 48. A. fulva
Schr. 9, do. (H. M., Kn.); 49. A. fulvicrus K. 9 and 4, skg. (H. M.); 50. A. gwynana
XK. 9, skg. and po-cltg. (H.M.); 51. A. helvola Z. 9, skg. (H.M.); 52. A. nitida
Fourcr.9 and é, exceedingly numerous, skg. and po-cltg. (H. M.): 53. A. parvula K. 9,
do. (H.M.); 54. A. schrankella Vy/. 6, skg. (H.M., Budd.); 55. A. smithella X. 9,
do. (Budd.); 56. A. trimmerana X. 9 and 4, skg. and po-cltg. (H. M.); 57. A. varians
K. 9, do. (H.M.); 88. Apis mellifica Z. ¥, freq., do. (H.M.); 59. Eucera longicornis
L.%, skg. (H. M.); 60. Halictus cylindricus F. 9, do. (H. M.); 61. Nomada ruficornis
L. 9 and 4, do. (H. M.); 62. N. ruficornis Z., var. signata Jur. 9 (H.M.).

Sickmann records the fossorial wasp Gorytes mystaceus Z., rare, for Osnabriick.
Schmiedeknecht noticed a bee (Andrena ferox Sm.) in Thuringia, and Krieger

observed the following bees at Leipzig. —

1. Andrena albicans A/#//.; 2. A. carbonaria Z.; 3. A. fucata Sm.; 4. A. labialis
K.; 5. A. nigroaenea K.; 6. A. tibialis A.; 7. A. trimmerana X.; 8. A. varians K. ;
9. Nomada lineola Pz.; 10. N. succincta Pz.

Friese saw the bee Andrena combinata Chr., one 9, in Baden.

Schenck observed a wasp (Odynerus melanocephalus Gme/.) in Nassau, and
Schiner the following flies in Austria. —

(a) Stratiomytdae: 1. Stratiomys furcata F. (4) Syrphidae: 2. Criorhina asilica
Fail., freq.; 3. C. berberina /., rare; 4. C. floccosa Mg.; 5. C. oxycanthae Mg.,
rare; 6. Mallota fuciformis /.; 7. Plocota apiformis Schr., very rare. (c) Zherevidae :
8. Thereva praecox Lgg.

Redtenbacher noticed the following beetles ir. Austria.—

(a) Cantharidae: 1. Cantharis sp. (4) Chrysomelidae: 2. Cryptocephalus lobatus
F. (c) Dermestidae: 3. Hadrotoma nigripes /, rare.

MacLeod (Flanders) observed Apis, a short-tongued bee, 2 hover-flies, an Empid,
a Muscid, and 8 beetles (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 305): also

cc2

388 ANGIOSPERMAE—DICOTYLEDONES

(Pyrenees) a short-tongued bee, a beetle, and 2 flies (op. cit., ili, 1891, pp. 433-43):
H. de Vries saw the honey-bee in the Netherlands (Ned. Kruidk. Arch., Nijmegen,
2. ser., 2. deel, 1875).

In Dumfriesshire, Apis (freq.), a humble-bee, a Dolichopodid, and another fly
were recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 64).

Hermann Miller also observed extra-floral nectaries on the hawthorn: a sweet
sap sometimes oozes from the tips of the young twigs, which is licked by bees
(Anthophora pilipes /. 6, Bombus terrester Z. 9, Andrena sp. 6) and wasps
(Odynerus parietum Z. 9).

H. Schiitte (Elsfleth) saw large numbers of wasps (Vespa germanica 4")
attracted by the juice exuding from the Psylla larvae which live on the hawthorn.
He also noticed a humble-bee (Bombus terrester Z.) licking this juice.

925. C. monogyna Jacq.—The flowers of this species agree with those of
C. Oxyacantha as regards their mechanism.

Visrtors.—I observed the same insects in the island of Pellworm as for C.
Oxyacantha (q. v.).

Fic. 122, Cotoneaster integerrima, Medic. (after Herm. Miiller.. A. Flower seen from
the side and somewhat obliquely from above. B. The same, seen trom above. C. The same,
in longitudinal section. (x 7.) nm, nectary.

263. Cotoneaster Rupp.

Flowers white or red; homogamous or protogynous; with concealed nectar
secreted by the fleshy inner wal] cf the receptacle.

926. C. integerrima Medic. (=C. vulgaris Zzmd/., and Mespilus Cotoneaster
£.). (Herm. Miiller, ‘ Alpenblumen,’ pp. 214-15; Schulz, ‘ Beitrage,’ II, pp. 7o-1.)—
The blossoms of this species are wasp flowers, with persistent stigmas. They are
protogynous in the Alps, but homogamous to protogynous at Halle and in North
Thuringia (Schulz). The petals and stamens bend so closely together over the
nectar as to leave but a small passage of access. In the protogynous flowers crossing
is effected by insect visitors before the anthers are ripe; in the homogamous ones
self-pollination is inevitable, for the stamens are situated just under the anthers
of the permanently incurved stamens.

Visitors.—Herm. Miller, in the Alps, only noticed a wasp (Polistes biglumis Z.).
Schulz (Central Germany) observed not only wasps, but also several other Hymeno-
ptera, together with some flies and beetles.

Morawitz (St. Petersburg) saw the bee Andrena fucata S7.

ROSACEAE 389

927. C. nigra Wahlenb. (=Crataegus nigra Waldst. et Kit., and Mespilus nigra
Willd.).—Focke says that when anthesis is over in this species the white colour
of the petals changes into pink.

264. Amelanchier Medic.

Flowers white ; homogamous, protogynous, or protandrous; either with exposed
nectar, or yielding only pollen (?).

928. A. vulgaris Moench (=A. rotundifolia C. Koch, Mespilus Amelanchier
£., and Aronia rotundifolia Pers.). (Herm. Miiller, ‘Alpenblumen,’ pp. 213-14;
Schulz, ‘ Beitrage,’ II, pp. 70, 72; Ricca, Atti Soc. ital. sc. nat., Milano, xiv, 1871.)—
Both in the Alps, and according to Schulz in Central Germany, the flowers of this
species are protandrous, sometimes to so marked an extent that the stigmas do
not become receptive till all the anthers have dropped off. The nectar is exposed,
and therefore accessible even to short-tongued insects. Visitors effect crossing in
flowers which exhibit well-marked protandry, but where this is less marked, self-
pollination may also take place, and this may be automatically effected by the fall
of pollen on the stigmas, should insect-visits fail. Ricca describes the flowers as
protogynous and devoid of nectar, and says that the four stamens develop in
succession.

Visitors.—Herm. Miiller saw 7 beetles, a Hymenopterid, 2 Muscids, and 4
Syrphids in the Alps. Schulz observed flies, Hymenoptera, and beetles in Central
Germany.

929. A. canadensis Torrey et Gray (=A. Botryapium DC.). (Kirchner,
‘Beitrage,’ pp. 38-9.)—The flowers of this species are white, and aggregated into
conspicuous racemose inflorescences: their odour resembles that of Prunus Padus.
Kirchner, for cultivated shrubs, describes the mechanism of the flowers as follows.—
They are slightly protogynous, as when they open the five stigmas are mature,
and occupy the middle of the flower, projecting 1-2 mm. beyond the still unripe
anthers. Before complete expansion, however, the outermost anthers dehisce, and
their filaments become erect, attaining the level of the stigmas, but at the same
time diverging so as to be several millimetres from them. The inner stamens
subsequently behave in the same way. Automatic self-pollination can easily take
place in the obliquely placed flowers by fall of pollen on the stigmas. Secretion
of nectar was not observed, probably however only because the weather was dull
during the days when the observations were made. The fact that the inner side
of the receptacle and the bases of the styles are coated with hairs allows us to
infer that nectar is actually secreted by the former.

265. Cydonia Tourn.

Flowers reddish-white, and of considerable size; protogynous or homogamous ;
with half-concealed nectar, secreted by a fleshy ring at the base of the style.

930. C. japonica Pers. (=Chaenomeles japonica Zznd/.). (Herm. Miiller,
‘Weit. Beob.,’ II, p. 288; Stadler, ‘ Beitrage’; Focke, Abh. natw. Ver., Bremen, xiv,
1897, p. 303.)—Hermann Miiller says that the flowers of this species are homo-
gamous; Stadler describes them as protogynous, with a style that varies in length,

390 ANGIOSPERMAE—DICOTYLEDONES

When the homogamous flowers open the anthers of the outer stamens dehisce,
while those of the inner ones remain for some time beneath the receptive stigmas.
As most of the visitors settle in the middle of the flower, and therefore touch
the stigmas first, they regularly effect crossing. Only the honey-bee usually pushes
in between the petals and stamens to reach the nectar, and so may effect self-
pollination as well. Failing insect-visits, Stadler says that automatic self-pollination
is possible, while Focke and Waite state that the species is undoubtedly self-fertile.
Large fruits often contain only empty pips.

Focke describes the species as andromonoecious. When hermaphrodite flowers
are dusted with pollen taken from the male flowers of the same plant they are
almost always sterile, though they are fertile with pollen taken from another plant.

Visttors.—Herm. Miiller observed the following.—

A. Coleoptera. Coccinellidae: 1. Rhizobius litura ¥., creeping about in the
flowers. B. Diptera. Muscidae: 2. Lucilia cornicina /. C. Hymenoptera.
Apidae: 3. Andrena albicans J/#/l. 9, vainly searching for nectar, and then po-dvg. ;
4. A. fulva Schr. 9, po-cltg.; 5. A. gwynana K. 9, do.; 6. Anthophora pilipes #.
and 9, skg.; 7. Apis mellifica Z. ¥, generally skg., now and then also po-cltg. ;
8. Bombus muscorum Ff’. 9, skg.; 9. B. pratorum 9 and 4, persistently skg.; ro. B.
rajellus K.9, skg.; 11. B. terrester Z. 9, persistently skg.; 12. Halictus rubicundus
Chr. 9, po-cltg.

Alfken noticed the following bees at Bremen :—1. Bombus agrorum Ff. 9; 2. B.
derhamellus X.9; 3. B. lucorum Z. 9 and ¥; 4. Halictus calceatus Scop. 9.

Schletterer, at Pola, saw the southern humble-bee Bombus argillaceus Scop.,
flying about the flowers on quiet sunny days in fanuary.

931. C. vulgaris Pers. (=Pyrus Cydonia Z.). (Dodel-Port, ‘Biol. Atlas d.
Botanik’; Kirchner, ‘Flora v. Stuttgart,’ p. 428.)—The large reddish-white flowers
of this species are protogynous. The nectar is protected against small unbidden
guests by hairs on the style, and by the incurved bases of the filaments. Small
creeping insects are kept out by the recurved sepals, which are covered with
glandular hairs below, and also by the thick hairs on the bases of the petals.
In other respects the mechanism agrees with that of Crataegus Oxyacantha.
Automatic self-pollination is not excluded.

VisiTors.—Loew observed a bee (Halictus nitidiusculus A. 9), po-cltg., in the
Berlin Botanic Garden.

266. Pyrus L.

White or red protogynous flowers of considerable size; with half concealed
nectar secreted by the receptacle.

932. P. Malus L. (Hildebrand, ‘D. Geschlechts-Vert. b. d. Pfl.,’ p.60; Herm.
Miiller, ‘ Fertilisation,’ p. 238; Kirchner, ‘ Beitrage,’ pp. 36-8; Waite, ‘ Pollination
of Pomaceous Fruits.’)—Hildebrand was the first to figure the projecting position
of the stigmas owing to which cross-pollination is favoured, and Hermann Miiller
first noticed protogyny in this species. ‘To Kirchner we are indebted for the most
exhaustive study of the flower mechanism. The size of the widely expanded reddish-
white or rose-coloured blossoms varies with the variety; in small-flowered ones
the average diameter is 38 mm., in large-flowered ones 49 mm. During the day

ROSACEAE 391

the blossoms possess only a slight odour of honey; but at night (according to
Dr. Steudel of Stuttgart) they exhale an agreeable fragrance which attracts numerous
Noctuids. The erect stamens are at first crowded together in the middle of the
flower. Their unripe yellow anthers are either at the same level as the five already
mature stigmas, or (as figured by Hildebrand) as much as 5 mm. below them.
About two days after the bud has opened the anthers of the outer stamens begin
to dehisce, and afterwards those of the inner stamens. Meanwhile the stamens
diverge but little, so that in varieties where they are long automatic self-pollination
can readily take place. This may also occur when the flower fades, for at this
stage the styles curve outwards to such an extent that the stigmas are brought
into contact with the but slightly diverging stamens. Anthesis lasts from five to
six days. The interior of the flower is usually exposed without any protection
to the action of rain, to the influence of which
it appears to be very sensitive. Crossing is neces-
sary for the production of a good crop. Waite
says that self-pollinated flowers but rarely set fruits
(cf. p. 392).

Visitors. —I noticed the following in my
garden (‘ Bloemenbiol. Bijdragen ’).—

A. Diptera. Syrphidae: 1. Syritta pipiens L.
(14.5.’96); 2. Syrphus balteatus Deg. (27. 5.'96);
both skg. and po-dvg. B. Hymenoptera. Apzdac:
3. Andrena parvula X. 9, skg. and po-cltg.

Herm. Miiller noticed the following.—

A. Diptera. (a) Bibionidae: 1. Dilophus vul-
garis Mg., in large numbers, skg. (4) Bombylidac : Fic.123. Pyrus Malus,L. (after Hilde-
z. Bombyling major Z., skg. (c) 2mpidac; ,. band). Flowst seen from fneside, ihe
Empis livida Z., skg. (@) Muscidae: 4. Onesia floralis RAs eee j
R.-D., skg. (e) Syrphidae: 5. Rhingia rostrata L.,
exceedingly numerous, mostly skg., but also po-dvg.; 6. Syrphus pyrastri Z., skg. and
po-dvg. B. Hymenoptera. (a) Apzdae: 7. Andrena albicans A/ull. and 9, skg.
and po-cltg.; 8. Anthophora pilipes /’. 9, do.; 9. Apis mellifica Z. ¥, do.; 10. Bombus
agrorum /.9, very common, do.; 11. B. hortorum ZL. 9, do.; 12. B. lapidarius Z., do.;
13. B. terrester Z.9, do.; 14. Halictus sexnotatus A. 9, skg.; 15. Osmia rufa Z. 8,
do. (6) Formicidae: 16. Several species, freq., skg.

Alfken, at Bremen, noticed a dragon-fly (Agrion minimum Zarr.), freq., busy
about the flowers, though the nature of its activity was not determined; also the
following bees.—

1. Andrena albicans M/zli.9; 2. A. albicrus A. 9; 3. A. convexiuscula X. 9;
4. A. varians K.9; 5. Bombus agrorum #. 9; 6. B. hortorum Z. 9; 7. B. terrester

£.9; 8. Halictus caiceatus Scop. 9; 9. H. levis A.9; 10. Osmia rufa Z. 9 and $;
11. Podalirius acervorum JZ. 9.

The following were recorded by the observers and at the places stated —

Krieger (Leipzig), a bee (Bombus hortorum Z. 9). Smith (England), a bee
(Andrena fulva Schr.). Plateau (Belgium), bees (Apis, Andrena fulva Schr., and
Bombus terrester Z.), a wasp (Vespa germanica /.), and flies (Calliphora vomitoria
L., Musca domestica Z., Lucilia caesar Z., and Eristalis tenax Z.).

392 4NGIOSPERMAE—DICOTYLEDONES

933- Pyrus communis L. (Herm. Miller, ‘Fertilisation, p. 239, ‘ Weit.
Beob.,’ p. 234; Kirchner, ‘ Beitrage,’ pp. 35-6; Swayne, Trans. Hort. Soc., London,
v, 1881, p. 208; Waite, ‘The Pollination of Pear Fls.,’ ‘Pollination of Pomaceous
Fruits,’ ‘The Fertilisation of Pear Fls.’; Knuth, ‘Bloemenbiol. Bijdragen,’ ‘ Weit.
Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 234.)—Hermann Miiller was the first
to call attention to the pronounced protogyny of the flowers of this species, and
also briefly described their mechanism. Kirchner has given the following exhaustive
account of the same. The anthesis of individual flowers lasts from seven to eight
days. They possess an odour like that of hawthorn (due to trimethylamide), or,
as Kirchner puts it, smell like cockchafers. There are many differences in size
and form, according to the variety. Sometimes the flowers are bell-shaped, the
petals inclining obliquely upwards; the average diameter in such cases is 18 mm.
In other varieties the corolla spreads out flat, so that its diameter attains 42-48 mm.
When the flowers open the central styles, with their already receptive stigmas,
stand tolerably erect: all the stamens are so bent inwards that their red unripe
anthers are crowded together in a heap in the middle, a little below the level
of the stigmas. They guard the passage to the nectar. It follows that insect
visitors usually alight upon the stigmas, and effect crossing if they have previously
visited an older flower. This female stage lasts from two to four days, according
to the weather. Meanwhile the outermost stamens gradually erect themselves,
diverge obliquely outwards, and open their anthers. The inner stamens in turn
behave in the same way, till in from five to seven days all the anthers have dehisced.
The petals have sometimes been shed before the anthers of the innermost five
stamens have opened. When the flower fades the stamens retain their divergent
position, but the styles curve away from one another so as to bring the stigmas
into contact with the shrivelled anthers. Since some pollen still clings to these,
automatic self-pollination is effected. The flowers are not protected against rain,
but are not very sensitive to its influence.

Cross-pollination is absolutely essential. This was first proved by George
Swayne, and confirmed by Merton B. Waite. The latter states that, as a rule,
perfect fruits only result from crossing, while pollen from another tree of the same
variety has no more effect than that of the same flower.

Waite drew the following conclusions from his pollination experiments.—

1. Many of the common varieties of pears require to be crossed, and produce
imperfect fruits or none at all when fertilized with their own pollen.

2. Some varieties are fertile with their own pollen.

3. In effecting crossing it is not sufficient to use pollen from another tree
of the same variety, for this is not more effective than the pollen of the same
tree. It is necessary to make use of pollen from another variety.

4. This ineffectiveness of pollen is not absolute, but is due to want of affinity
between the pollen and ovules of the same variety.

5. It follows that pollen taken from two varieties may be perfectly ineffective
when transferred, respectively, to the stigmas of flowers of the same variety, but
each may produce excellent results when applied to stigmas of the other variety.

11. Pears resulting from self-fertilization contain imperfectly formed seeds,
that are usually quite vestigial. The fruits that result from crossing contain well-
developed healthy seeds.

ROSACEAE 393

12. Even in varieties which are fertile with their own pollen that of other
varieties is more efficient, and when crossing is not prevented by the exclusion of
insects most of the fruits seem to result from it.

13. Typical fruits, and, as a rule, the largest and best examples from all the
varieties, are the result of crossing, whether the varieties are self-sterile or self-fertile.

Waite extended his experiments to apples and quinces. The varieties of the
former exhibit even a greater tendency towards infertility when dusted with their
own pollen than is the case with pears. The quince, on the other hand, is almost
equally fertile with self- and cross-pollination.

Vistrors.—Herm. Miller (H. M.) in Westphalia, and myself (Kn.) in Schleswig-
Holstein, observed the following.—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella conglobata Z., nect-lkg.
(H.M.). (8) Curculionidae: 2. Rhynchites aequatus L., nect-lkg. (H.M.). (c) Wet-
dulidae: 3. Meligethes, freq.(H.M.). (d) Phalcridae: 4. Olibrus aeneus /., nect-lkg.
(H.M.). B. Diptera. (2) Muscidae: 5. Anthomyia radicum Z. é and 9, very freq.,
skg. (H.M.); 6. A. sp. (Kn.); 7. Calliphora erythrocephala A/g., skg. (H.M.);
8. Lucilia cornicina &., do. (H. M., Kn.); 9. Musca corvina /., do. (H. M.); 10. M.
domestica Z., do. (H.M., Kn.); 11. Pollenia rudis /., do. (H.M.); 12. P. vespillo
F., do. (H. M.); 13. Sarcophaga carnaria Z.(Kn.); 14. Scatophaga merdaria /, skg.
and po-dvg. (H.M.); 15. Sepsis sp., skg. (H.M.). (2) Syrphzdae: 16. Ascia poda-
grica #., freq., skg. and po-dvg. (H.M.); 17. Eristalis arbustorum Z., skg. and
po-dvg. (H.M.); 18. E. intricarius Z., do. (H.M.); 19. E. nemorum Z., freq., do.
(H.M.); 20. E. tenax Z., do. (H.M., Kn.); 21. Melanostoma mellina £., skg. and
po-dvg. (H.M.); 22. Rhingia rostrata Z., do. (Kn.); 23. Syritta pipiens Z., do.
(H.M., Kn.). C. Hymenoptera, (a) Apzdae: 24. Andrena albicans A/il/. 9 and
3, freq., skg. and po-cltg. (H.M., Kn.); 25. A. collinsonana X. 9, do. (H.M.) ;
26. A. gwynana K. 9, do. (H.M., Kn.); 27. A. parvula A. 9, do. (H.M., Kn.) ;
28. Apis mellifica Z. ¥, do. (H. M., Kn.); 29. Bombus terrester Z. 9, skg. (H. M.);
30. Halictus rubicundus Chr. 9, skg. and po-cltg. (H.M.). (4) Formzcidae: 31. Lasius
niger Z. ¥, nect-lkg. (H. M.). (c) Zenthredinidae: 32. Dolerus gonager F’, occasional,
skg. (H.M.); 33. Nematus capraeae £. (=Nematus gallicola Svepf.-L£.), in large
numbers, skg. (H.M.). D. Thysanoptera. 34. Thrips, freq. (H. M.).

Alfken noticed the following bees at Bremen:—1. Andrena nigroaenea K. 6;
2. A. varians &. 9; 3. Bombus agrorum /. 9; 4. B. lucorum Z. 9.

934. P. salicifolia L. (Kirchner, ‘ Beitrage,’ p. 38.)—This oriental species
was studied by Kirchner in the Hohenheim Botanic Garden. He found its flowers
to be protogynous, and to agree in other respects with those of P. communis.

267. Sorbus L.

Flowers white or rose-red in colour, and aggregated into dense panicles;
bomogamous, protogynous, or protandrous; with half-concealed nectar secreted
by a ring at the base of the style.

935. S. Aucuparia L. (Herm. Miiller, ‘Fertilisation,’ pp. 239-40; Warnstorf,
Schr. natw. Ver., Wernigerode, xi, 1896.)—Hermann Miiler describes the flowers
of this species as protogynous, agreeing as regards their mechanism with those of
Crataegus Oxyacantha, which they also resemble in odour (that of trimethylamide).
.As many flowers are aggregated into large conspicuous inflorescences, insect visitors

394 ANGIOSPERMAE—DICOTYLEDONES

are numerous. Warnstorf describes the pollen-grains as white in colour, irregular,
rounded to ellipsoidal, almost smooth, about 37 » long and 25 « broad.

Visitors. — Herm. Miiller (H. M.), Buddeberg (Budd.), and myself (Kn.)
(‘Bloemenbiol. Bijdragen’), have observed the following.—

A. Coleoptera. (a) Cerambycidae: 1. Clytus arietis Z., skg. (H.M.). (6)
Chrysomelidae: 2. Lochmaea sanguinea /., skg. (H.M.). (c) Curculionidae: 3. Apion,
skg. (H.M.); 4. Phyllobius maculicornis Germ., skg. (H.M.). (d) Dermestidae :
5. Attagenus pellio Z., occasional (H. M.); 6. Byturus sp., by the hundred (H. M.).
(e) Elateridae: 7. Agriotes aterrimus Z.(H.M.); 8. Corymbites holosericeus Oliv.
(H.M.); 9. Dolopius marginatus Z. (H.M.); 10. Limonius cylindricus Pays.
(H.M.); 11. L. parvulus Pz. (H.M.). (/) AMordellidae: 12. Anaspis rufilabris GyZh.
(H.M.). (¢) Metidulidae: 13. Epuraea, by the hundred (H. M.); 14. Meligethes, do.
(H.M.). (4) Scarabaeidae: 15. Cetonia aurata Z., dvg. all the parts of the flowers
(H.M.); 16. Melolontha vulgaris F., do. (H.M.). (¢) Zelephoridae: 17. Malachius
aeneus /”, nect-lkg. and dvg. the anthers (H. M.). (4) Zenebrionidae: 18. Microzoum
tibiale #. (H.M.). B. Diptera. (a) Bibionidae: 19. Dilophus vulgaris AZ@e., freq.,
skg. (H.M.). (4) Conopidae: 20. Myopa testacea Z. (H.M.). (c) Empzdae: 21.
Empis livida Z., freq., skg. (H. M.); 22. E. rustica Fadl., do. (H.M.). (@) Muscidae:
23. Echinomyia fera Z. (H.M.); 24. Lucilia caesar Z., skg. and po-dvg. (Kn.);
25. Musca domestica Z., do. (H.M.); 26. Onesia floralis R.-D., freq., skg. (H.M.);
27. Sarcophaga carnaria Z., skg. and po-dvg. (Kn.); 28. Scatophaga merdaria F.,
freq., skg. (H. M.); 29. S. stercoraria Z., do. (H. M., Kn.); 30. Sepsis, freq. (H. M.).
(e) Syrphidae: 31. Eristalis arbustorum Z., freq., skg. and po-dvg.(H.M.); 32. E.
horticola Deg., do. (H. M.); 33. E. nemorum Z., do. (H.M.); 34. E. pertinax Scop.,
do. (Kn.); 35. E. tenax Z., do. (Kn.); 36. Helophilus floreus Z., do. (Kn.); 37.
Melanostoma mellina Z., do. (Kn.); 38. Rhingia rostrata Z., do. (H.M.); 39. Syritta
pipiens Z., do. (H.M.). C. Hymenoptera. (a) Apzdae: 40. Andrena albicans
Mill. 8, skg. (H. M., Budd.); 41. A. albicrus X, 9 and 4, skg. and po-cltg. (H. M.) ;
42. A. atriceps K. 9, do. (H. M.); 43. A. convexiuscula A. 9, do. (H.M.); 44. A.
dorsata A. 9 and 6, do. (H.M.); 45. A. smithella X. 9, po-cltg. (H.M.); 46. Apis
mellifica Z. ¥, very freq., skg. and po-cltg. (H.M., Kn.); 47. Halictus rubicundus
Chr. 9, skg. and po-cltg. (H.M.); 48. H. zonulus Sm. 9, do. (H.M.); 49. Nomada
tuficornis Z. 9 and é, skg. (H.M.); 50. N. ruficornis Z., var. signata Jur. 9, skg.
(H.M.). (4) Hormicidae: 51. Formica pratensis Deg. 9 and 4, freq., skg. (H. M.);
52. F. rufa Z. ¥, nect-lkg. (H.M.); 53. Lasius niger Z.9, freq., skg. (H. M.);
54. Myrmica sp. §, do. (H.M.). D. Lepidoptera. Rhopalocera: 55. Thecla rubi
L., skg. (Budd.).

Loew noticed the following in Brandenburg (‘ Beitrage,’ p. 37).—

A. Coleoptera. Nitidulidae: 1. Meligethes aeneus /. 3B. Diptera. (a)
Lmpidae: 2. Empis punctata Mg., skg.; 3. E. tessellata F., do. (6) Strakomyrdae :
4. Odontomyia tigrina #. (c) Syrphidae: all skg.; 5. Eristalis arbustorum Z.; 6. E.
nemorum Z.; 7. E. tenax L.; 8. Helophilus floreus Z.; 9. H. pendulus Z.; 10. H.
trivittatus #.; 11. Syrphus corollae # C. Hymenoptera. Apidae: all skg.;
12. Andrena fulva Schr. 9; 13. A. nigroaenea A.9; 14. A. varians X., var. helvola
L. 9.

Alfken records the following for Bremen.—

A. Coleoptera. (a) Scarabaetdae: 1. Cetonia aurata Z.; 2. C. floricola Hést.
(6) Cerambycidae: 3. Cerambyx scopolii ess’. B. Hymenoptera. (a) Apidae:
4, Andrena albicans Jill. 9 and 6; 5. A. albicrus A. 9; 6. A. apicata Sm.9; 7. A.
cingulata 7.3; 8. A. flavipes Pz. 9; 9. A. nigroaenea K.9; 10. A. nitida Fourcr. Q3
11. A. praecox Scop. 9; 12. A. tibialis A. 9; 13. A. varians A.9; 14. Bombus
agrorum #9; 15. B. terrester Z.9; 16. Halictus nitidiusculus A. 9. (4) Vespidae :
17. Odynerus parietum Z. 6.

ROSACEAE 395

MacLeod saw a humble-bee in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii,
1891, p. 434); and Apis in Flanders (op. cit., vi, 1894, p. 307). Redtenbacher
noticed the Cerambycid Rhopalopus insubricus Germ.

H. de Vries noticed 2 bees (Apis mellifica Z. ¥, and Andrena pilipes /. 9) in the
Netherlands (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).

In Dumfriesshire, a humble-bee, 2 Empids, 3 Muscids, a hover-fly, a Dolicho-

podid, Meligethes, and undetermined beetles were recorded (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 64).

936. S. Chamaemespilus Crantz (=Pyrus Chamaemespilus Z%rh., Mespilus
Chamaemespilus Z., and Crataegus Chamaemespilus /acg.). (Schulz, ‘Beitrage,’ II,
p. 72-)}—Schulz describes the rose-red flowers of this species as cither homogamous
or more or less protogynous. As the stigmas are situated in the line of fall of
the pollen, automatic self-pollination frequently occurs, but this can be dispensed
with during sunny weather, for nectar is then secreted in great abundance, and
the flowers are visited by numerous nectar-sucking or pollen-devouring insects
(long-tongued flies, small beetles, and particularly bees and wasps). Flowers that
have already lost their petals, but which are still fresh enough to secrete plenty
of nectar, are also sucked by short-tongued flies and the larger beetles. These

are kept from visiting the younger flowers by the erect petals, which are tolerably
close together.

937. S. scandica Fries.—

Visirors.—Loew observed the following visitors in the Berlin Botanic Garden.—

A. Coleoptera. Malacodermata: 1. Dasytes flavipes F., nect-lkg. B. Diptera.
Empidae: 2. Empis trigramma A/g., skg. C. Hymenoptera. Afidae: 3. Apis
mellifica Z. ¥, skg. and po-cltg.

XXXV. ORDER SAXIFRAGEAE Vent.

(including PuirapELpHacraE Don and GrossuzartacEakE DC.
(=RupestaceaE Endl.)).

268. Saxifraga Tourn.

LirzraTure.— Hermann Miiller, ‘ Fertilisation,’ pp. 243-5, ‘Alpenblumen,’
pp. 109-11; Engler u. Prantl, ‘D. nat. Pflanzenfam.,’ III, 209, Bot. Ztg., Leipzig,
xxvi, 1868, pp. 833-42.

Flowers pure white or yellow to purple-spotted or dirty yellow in colour,
rarely rose-red or blue. The nectar is exposed, rarely half-concealed, and is
secreted by the outer wall of the ovary. This position attracts numerous short-
tongued insects, among which flies predominate to such an extent that most of
the species must be placed in flower class F. Owing to the large number of visitors,
many species are able to dispense with automatic self-pollination, which indeed
is rendered nearly or quite impossible by the occurrence of more or less pronounced
dichogamy. Most species are protandrous, but a few are protogynous (S. androsacea,
S. muscoides, S. Seguieri). In the latter the flowers are considerably smaller in
the first (female) than in the second (male) stage. After the stigmas shrivel the

396 ANGIOSPERMAE—DICOTYLEDONES

flower increases to twice its original diameter or even more, so that the visits
paid by any particular insect will usually be in the order most favourable for
crossing.

We are indebted to A. Engler for an excellent monograph on the genus.
He regarded all the species as protandrous. ‘Treviranus (Bot. Ztg., Leipzig, xxi,
1863) had already observed the movement of the stamens towards the middle
of the flower, and inferred that automatic self-pollination takes place. As some
species of this genus possess nectar-guides, while others do not, Engler called in
question the correctness of Sprengel’s interpretation of these markings. Hermann
Miiller remarks in this connexion (‘ Fertilisation,’ p. 243), that in plants where the
nectar is sometimes fully exposed, and at other times concealed, there is very
naturally a corresponding variation in the nectar-guides, so that this peculiarity
of the genus cannot be regarded as opposed to Sprengel’s view, while there is no
other explanation to take its place.

938. S. Aizoon Jacq. (Herm. Miiller, ‘Alpenblumen,’ pp. 100-2.) —The
flowers of this species are markedly protandrous, and in the Alps self-pollination
is almost or entirely prevented. The abundant pollen and readily accessible nectar

Fic. 124. Saxtfraga Atzoon, Jacq. (after Herm. Miller). 4. Flower at the beginning of the first
(male) stage. &. The same, at the end of the first stage. C. Flower in the second (female) stage (x 3).
a, nectary.

attract numerous visitors, especially flies. In Greenland, where insects are few,
the flowers are also strongly protandrous, but here effective automatic self-pollination
is ultimately possible by contact of the anthers (still covered with pollen) with the
widely divergent stigmas (Warming, Bot. Tids., Kjébenhavn, xvi, 1888, pp. 27-9).

Visirors.—Herm. Miiller observed (in the Alps) no fewer than 61 species of
flies (including 37 Muscids); also 5 beetles, 11 Hymenoptera, and 10 Lepidoptera.

Loew saw the Syrphid Cheilosia modesta £eg.(?) in Switzerland (‘ Beitrage,’
p. 56). MacLeod noticed a short-tongued Hymenopterid and 5 Muscids in the
Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 420).

Loew observed 2 Syrphids in the Berlin Botanic Garden—1. Ascia podagrica
F’., passing from flower to flower, hovering a considerable time over each, and then
alighting to suck; 2. Melithreptus scriptus Z., do.

SAXIFRAGEAE 307

939. S. mutata L.—The flowers of this species secrete exposed nectar, and
Stadler describes them as protandrous, though self-pollination is not excluded. The
stamens at first move towards the centre, and later on away from it.

940. S. Burseriana L.—Kerner states that the flowers of this species are
protogynous, but during the twelve days of anthesis the stamens move towards
the centre, so that automatic self-pollination ultimately takes place.

941. S. caesia L. (Herm. Miiller, ‘Alpenblumen,’ pp. 102-4; Engler, Bot.
Ztg., Leipzig, xxvi, 1868.)—The blossoms of this species are fly flowers, secreting
exposed nectar. Engler was the first to notice their protandry. Self-pollination
is entirely prevented in the Alps.

Vistrors.—These are chiefly flies, of which Herm. Miiller observed 15 species
in three days, besides 3 beetles, 3 Hymenoptera, and 3 Lepidoptera.

Fic. 125. Saxtfraga caesta, L. (after Herm. Miller), 4. Flower in the first (male) stage (x4%).
B. Flower in the middle of the same stage. C. Flower in the second (female) stage. 1-6, anthers;
Ji filaments; 2, nectary; ov, ovules; #, petals; s, sepal; s/, stigma.

942. S. exarata Vill. (=S. nervosa Lapeyr.). (Herm. Miiller, ‘ Alpenblumen,’
p. 104.)—The blossoms of this species are fly flowers secreting exposed nectar.
Their mechanism essentially agrees with that of S. caesia. Here again self-pollination
is prevented (in the Alps) by well-marked protandry.

Visrtrors—Herm. Miller observed 4 species of flies and an ant. MacLeod
saw a fossorial wasp and a hover-fly in the Pyrenees.

943. S. oppositifolia L. (Ricca, Atti Soc. ital. sc. nat., Milano, xiv, 1871;
Warming, Bot. Tids., Kjébenhavn, xvi, 1888, pp. 29-33, Vet. Ak. Overs., Kjében-
havn, 1886-7, p. 13; Herm. Miiller, ‘ Alpenblumen,’ pp. 98-1o0.)—The flowers of
this species are variously described by Engler as protandrous, Axell as slightly
protandrous, Ricca as homogamous, and Hermann Miller as protogynous (on the
Piz Umbrail and the Albula). Schulz also observed protogyny, as did Warming
in Greenland, and Lindman on the Dovrefjeld. Ekstam, on the other hand, noticed
protandry in Nova Zemlia. The blossoms are Lepidopterid flowers with concealed
nectar. Lindman observed large-flowered and small-flowered varieties on the Dovre-
fjeld. Failing insect-visits, self-pollination frequently takes place, and appears to
be effective, for the flowers in Greenland set many fruits, despite the early anthesis
and the lack of insects. Lindman also observed ripe fruits on the Dovrefjeld,
but no visitors.

398 ANGIOSPERMAE—DICOTYLEDONES

In Spitzbergen this species is sometimes seen as a creeping form, sometimes in
dense clumps. According to Andersson and Hesselman (‘ Bidrag till Kanned. om
Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 1), it begins to bloom in the middle of
June, and some plants continue to do so till September. Fruits were observed
to be set from the middle of July (19. 7. ’98) or the beginning of August (8. 8. 97)
onwards. The flowers vary from 9 to 11 mm. in diameter, or exceptionally 18 to
zo mm. Ekstam describes them as faintly fragrant, dark-red to bright violet
in colour, or sometimes white (‘Bliitenbiol. Beob. a. Spitzbergen,’ p. 12). As to
the development of the sexual organs the protogynous to homogamous flowers agree
with those examined by Warming. Andersson and Hesselman (op. cit., p. 24)

Fic. 126. Saxifraga opposttifolia, L. (after Herm. Miller). 4. Newly opened flower, seen from
above. 2. Older flower, seen from above. C. Another older flower, in longitudinal section (x 5).
D. Flower with mature stigmas and unripe anthers, partly in section (Xx 3). a, anthers; 2, nectary ;
ov, ovules; p, petals; s, sepals; sé, stigmas.
noticed distinct protogyny. Ekstam says that characteristic more or less closed
flowers appear in August, with reduced stamens and greatly enlarged pistils.

Schulz observed gynodioecism in the Tyrol. The nectar is so deeply situated
that short-tongued insects can reach it only with great difficulty or not at all, while
it is conveniently placed for Lepidoptera.

Vistrors.—Herm. Miiller noticed that Lepidoptera formed the majority of
individuals (3 species), and also observed a beetle, a Syrphid, and 3 Muscids. Ricca
saw a humble-bee and several Lepidoptera. Ekstam (Nova Zemlia) observed humble-

SAXIFRAGEAE 399

bees (the sole resort of which during the summer is this species) and flies. In
Spitzbergen he noticed several small Muscids on three days during July. Schneider
(Mus. Aarsh. Tromsg, xvii, 1895) saw Andrena sp. in the Christiania Botanic
Garden. He further states (op. cit., p. 142) that humble-bees only play a sub-
ordinate part as pollinators in Arctic Norway.

944. S. aizoides L. (Herm. Miiller, ‘Fertilisation, p. 245, ‘ Alpenblumen,’
pp. 94-8; Warming, Bot. Tids., Kjébenhavn, xvi, 1888, pp. 26-7; Axell, ‘Om
Anord. for Fanerog. Vaxt. Befrukt.,’ p. 35; Engler, Bot. Ztg., Leipzig, xxvi, 1868 ;
Schulz, ‘ Beitrage.’)—Axell and Engler were the first to describe the protandry of
this species, though Schulz states that the
terminal flower is frequently female. The
nectar is exposed, and the flower mechanism
adapted to the visits of flies. According to
Hermann Miller, cross-pollination, as the
result of insect-visits, is ultimately secured
by the slow successive development of the
individual stamens and the stigmas. Self-
pollination is not completely excluded.
Warming states that in Greenland, Spitz-
bergen, and Finmark, the flowers are at first
markedly protandrous, afterwards becoming
homogamous. Ripe fruits were observed at

Fic. 127. Saxifraga atzordes, L. (after Herm. Miller). 4. Flower in the first (male) stage,

B. Flower in the second (female) stage (x 34). C. The same in longitudinal section (x 7). a@, anther;

fi, filament; 7, nectary; ov. ovary; #, petal; s, sepal; s/, stigma.
Jacobshavn and Franz-Josef’s Fjord. Abromeit describes Greenland specimens as
possessing the orange-red spots on the petals, marked more or less clearly (‘Bot. Ergeb.
von Drygalski’s Grénlandsexped.’). Andersson and Hesselman saw sterile flowers
in Beeren Island. They state that the species flowers in Spitzbergen from the first
half of July till August and September, but whether fruits are set was not determined
(‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 28). Ekstam,
for Spitzbergen, describes the flowers as being usually protandrous, though cases
of homogamy were also noticed (‘ Bliitenbiol. Beob. a. Spitzbergen,’ p. 15). Ekstam
says that the odourless flowers are 10-12 mm. in diameter in Nova Zemlia.

400 ANGIOSPERMAE—DICOTYLEDONES

VisiTors.—Small flies and ants were seen in Nova Zemlia, but no visitors in
Spitzbergen.

Herm. Miiller observed in the Alps no fewer than 85 Diptera (mostly Muscidae).
besides 8 beetles, zo Hymenoptera, and 13 Lepidoptera; Loew saw a hover-fly
in the same region (‘ Bliitenbiol. Floristik,’ p. 397). Lindman noticed flies, Hymeno-
ptera, and a beetle on the Dovrefjeld. MacLeod saw 8 short-tongued Hymenoptera,
a Phryganid, a beetle, 4 Syrphids, and 19 other Diptera in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, pp. 420-2).

945. S. Hirculus L. (Warming, Bot. Tids., Kjébenhavn, xvi, 1888, p. 25.)—
Warming says that the flowers of this species are markedly protandrous in Spitz-
bergen. Andersson and Hesselman (‘ Bidrag till Kianned. om Spetsbergens o. Beeren
Eil. Karlvaxtflora,’ pp. 27-8) say that this species flowers in Spitzbergen during
the second half of July and in August; feebly developed flower-buds were noticed in
Beeren Island on June 18, 1898. Fruits appear to be ripened with difficulty at
the end of August or the beginning of September. The petals are bright yellow,

Fic. 128. Saxt/raga rotundifolia, L. (after Herm. Miller). 4. Flower at the beginning of the
first (male) stage. &. Flower towards the end of the same stage. C. Flower in the second (female)
stage (44). a-a°, anthers; /, filament: 4, drop of nectar; ov, ovary; #, petal; s, sepal; sé, stigma.

with a basal swelling on either side, and citron-yellow spots. There is marked
protandry, according to Andersson and Hesselman, confirmed by Ekstam. Stocks
with purely female flowers were seen, and also gynomonoecious transition forms.
Ekstam thinks that the basal folds of the petals, the roots of the filaments, and
the base of the ovary possibly secrete nectar. In Nova Zemlia, according to Ekstam,
the diameter of the odourless, nectarless(?), slightly protandrous flowers is 12-
25 mm.

Visitors. —On seven days of July and August (1897) Ekstam noticed an
extremely large number of flies. These insects have also been seen in Nova Zemlia.

946. S. rotundifolia L. (Herm. Miller, ‘ Fertilisation,’ p. 245; ‘Alpenblumen,’
pp. 89, 90.)—The flowers of this species are white, spotted with purple-red, and
secreting half-concealed nectar. They are so markedly protandrous that automatic
self-pollination cannot take place. They belong to the class of fly flowers. Insect-
visits necessarily bring about crossing, for even the smallest nectar-seeking guest
touches the anthers of younger flowers, and one cr both stigmas in those which
are older.

SAXIFRAGEAE 401

VisiTors.—Herm. Miiller observed almost exclusively flies (2 Empids, 7 Muscids,
5 Syrphids); also an Ichneumonid. Schiner mentions the Syrphid Sphegina clunipes
Fail. as a common visitor.

947. S. stellaris L. (Herm. Miiller, ‘Fertilisation,’ p. 244, ‘Alpenblumen,’
pp: 90-2.)—The flowers of this species are stellate, with half-concealed nectar.
Stamens and carpels mature in the same order as in S. rotundifolia, but the times
of ripening of the individual stamens overlap to a greater extent. Automatic self-
pollination does not usually take place, but perhaps occurs in dull weather or if
insect-visits fail. Schulz says that the terminal flower is often female. According
to Ekstam, the flowers are protandrous in the Swedish Highlands at Dovre, and
also in Nova Zemlia, while on the Ronde and Tronfjall they are almost homogamous.

Fic. 129. Saxifraga stellaris, L. (after Herm. Miller). 4. A bilaterally symmetrical flower in
the middle of the first (male) stage. 2. A radially symmetrical flower at the end of the same stage.
C. Flower in the second (female) stage. a-a", anthers; 7, filament; 7, nectary; ov, ovary; s, sepal;
Pp, petal; sé, stigma.

On the Dovrefjeld, on the other hand, according to Lindman, the flowers
are markedly protandrous, but on the Tronfjeld, and in Langluplad, they are
almost homogamous, and self-pollination is ultimately possible. In Greenland
Warming observed homogamy and protogyny as well as protandry (Bot. Tids.,
Kjébenhavn, xvi, 1888, pp. 10-14). He found ripe fruits on the Sukkestop. In
63° N. lat. there occurs a variety, comosa Poir., that propagates by rosettes of leaves,
which fall off. These originate from flowers.

Abromeit also made observations on the variety in Greenland, and noticed
properly developed terminal flowers, or else flowers with minute petals (var. crypéo-
petala) at the ends of rosette-bearing branches (‘Bot. Ergeb. von Drygalski’s
Groniandsexped.,’ p. 33). The normal flower possesses white, clawed petals, with
two yellow spots at the base of the lamina. The anthers vary in colour from white
to pale violet and purple; the pollen is vermilion,

Ekstam says that in Spitzbergen the variety propagates entirely by rosette-shaped
leaf-buds, which easily fall off and take root (‘Bliitenbiol. Beob. a. Spitzbergen,’
p. 12). Andersson and Hesselman state that the process commences at the beginning
of August and goes on for a considerable time (‘ Bidrag till Kanned. om Spetsbergens
o. Beeren Eil. Karlvaxtflora,’ p. 23).

DAVIS. I pd

402 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Herm. Miiller observed flies, almost exclusively (a Dolichopodid,
an Empid, 8 Muscids, 2 Syrphids); also occasional beetles, Lepidoptera, and
Hymenoptera. In Dumfriesshire, an Empid, 3 Muscids, and 3 Syrphids were
recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 72).

948. S. aspera DC. (Herm. Miiller, ‘Fertilisation, p. 244, ‘Alpenblumen,’
Pp- 92-3-)—The blossoms of this species secrete exposed nectar, and belong to the
class of fly flowers. As Engler observed, they are markedly protandrous. Auto-
matic self-pollination is as a rule excluded.

Visttors.—Herm. Miiller noticed 2 Muscids; and Loew (Pontresina) saw a
Muscid.

949. S. bryoides L. (=S. aspera Z.). (Herm. Miller, ‘Fertilisation,’
p- 244, ‘Alpenblumen,’ pp. 93-4.)—Engler was the first to observe protandry in
this species. The flower mechanism completely agrees with that of S. aspera.
Kerner states that anthesis lasts for eight days.

Visirors.—Herm. Miiller chiefly observed flies (an Empid, 6 Muscids, 2 Syr-
phids); also occasional beetles and Ichneumonids.

950. S. cuneifolia L. (Kirchner, ‘ Beitrige,’ pp. 31-2.)—The flowers of this
species secrete exposed nectar. Delpino was the first to call attention to their
marked protandry. Kirchner says that the flower mechanism, as observed at Zermatt,
comes nearest to that of S. stellaris. Self-pollination is excluded by protandry:
it is only when all the anthers have shrivelled and dropped off that the styles
separate and expose the stigmas.

VisiTors.—Kirchner observed 2 species of flies.

gsi. S. hieracifolia Waldst. et Kit—The nectar is half-concealed in this
species. Kerner, whose observations were made in the Tyrol, noticed that the
peduncle bends downwards towards the end of anthesis, so that the stigmas are
brought into the line of fall of the pollen, automatic self-pollination resulting.
Warming found that self-pollination can also easily take place in Greenland, where
the flowers remain more or less closed (Bot. Tids., Kjébenhavn, xvi, 1888, pp. 16-22).
Ekstam says that in Nova Zemlia the odourless, inconspicuous, greenish-yellow
flowers are markedly protandrous, and 5-10 mm. in diameter.

Andersson and Hesselman state that the species flowers in Spitzbergen from
the beginning of July till the beginning of August (‘Bidrag till Kanned. om Spets-
bergens o. Beeren Eil. Karlvaxtflora,’ pp. 21-2). Fruits are regularly set, and are
in some cases fully ripe by the end of August. Ekstam, for Spitzbergen, describes
the flowers as 5-10 mm. in diameter, odourless, and protogynous to homogamous:
the stigmatic surfaces separate and assume a glistening appearance before the anthers
are completely ripe (‘ Bliitenbiol. Beob. a. Spitzbergen,’ pp. 10-11).

Visirors.—Ekstam noticed a medium-sized fly in Spitzbergen.

952. S. Seguieri Spreng. (Herm. Miiller, ‘Fertilisation,’ p. 244, ‘Alpen-
blumen,’ pp. 106—7.)—This species secretes exposed nectar, and its flower mechanism
is adapted to the visits of flies. Unlike most others of the genus it exhibits marked
protogyny, and the stigmas persist only for a short time. Self-pollination is excluded,
for the outer anthers only begin to dehisce after the stigmas have shrivelled up.

Visirors.— These are flies.

SAXIFRAGEAE 403

953. S. muscoides Wulf. (Herm. Miiller, ‘Fertilisation,’ p. 244, ‘ Alpen-
blumen,’ pp. 106-7.)—The flower mechanism of this species agrees with that of
S. Seguieri.

Visirors.—Herm. Miiller observed 6 flies, a beetle, an Ichneumonid, and a
Lepidopterid.

‘MacLeod noticed 2 short-tongued Hymenoptera, a beetle, and 5 flies, in the
Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 422-3).

954. S. androsacea L. (Herm. Miller, ‘ Fertilisation,’ p. 244, ‘Alpenblumen,’
pp. 107-8.)—This is a third protogynous Alpine species, and its flower mechanism
agrees with those of S. Seguieri and S. muscoides. In this case, however, self-
pollination is ultimately possible, for the stigmas remain receptive till the dehiscence
of the first anther.

Visitors.—Herm. Miiller noticed a hover-fly (Eristalis tenax JZ.) at a height
of over 3,000 m.

955- 9. decipiens Ehrh. (=S. hypnoides Z. and S. caespitosa Sm.). (Warming,
Bot. Tids., Kjébenhavn, xvi, 1888, pp. 18-22.)—Warming describes the flowers of
this species as slightly protandrous, homogamous, or even protogynous. Self-
pollination is possible and effective, for ripe fruits have been found in Spitzbergen,
Beeren Island, &c. Besides hermaphrodite flowers, female ones have been observed
in Spitzbergen, on the Dovrefjeld, and in Greenland.

Abromeit states that the species is widely distributed in Greenland up to a height
of 5,000 ft. above sea-level, and is represented by several varieties (‘ Bot. Ergeb. von
Drygalski’s Grénlandsexped.,’ p. 35). The calyx is covered with glandular hairs,
sometimes black in colour, and sometimes yellow.

Visrtors.—Loew observed the following in the Berlin Botanic Garden.—
A. Diptera. (a) Muscidae: 1. Lucilia caesar Z.; 2. Scatophaga scybalaria Z.

(4) Syrphidae: 3. Eristalis nemorum Z., skg.; 4. Syritta pipiens Z., do. B. Hymeno-
ptera. Afidae: 5. Halictus minutissimus X. 9, skg.; 6. H. nitidiusculus &. 9, do.

956. S. caespitosa L.—The nectar is exposed in this species. Lindman
observed, on the Dovrefjeld, homogamous flowers in which effective self-pollination
was possible, as evidenced by fruits. Ekstam states that the faintly fragrant flowers
are 5-12 mm. in diameter in Nova Zemlia. Self-pollination is prevented in Aowers
that are either almost homogamous or very markedly protandrous, but it is possible
in those which are protogynous-homogamous.

According to Andersson and Hesselman (‘ Bidrag till Kanned. om Spetsbergens
o. Beeren Eil. Karlvaxtflora,’ pp. 30-1), this species is among those indigenous to
Spitzbergen in which anthesis is prolonged. It begins when the snow melts in
the middle of June, and continues all through the summer till the end of August,
or even into September. In Beeren Island the earliest blossoms were noticed on
June 13, 1898; fruits are ripened in abundance from about the second half of
July. The observers mentioned regard the flowers as protogynous, but Ekstam
describes them as strongly protandrous-homogamous (‘ Bliitenbiol. Beob. a. Spitz-
bergen,’ p. 18). They are 10-15 mm. in diameter, odourless, and white, yellowish-
green or reddish in colour. After the anthers have completely ripened the stamens

pd2

404 ANGIOSPERMAE—DICOTYLEDONES

bend over the matured stigmas in such a way that autogamy would seem to be
inevitable. Andersson and Hesselman observed an apetalous variety, in which the
petals were modified into stamens, the pollen of which was partially abnormal.

Visirors.—Numerous flies were observed by Lindman on the Dovrefjeld. As
frequent visitors in Spitzbergen Holmgren noticed the Ichneumonids Hemiteles
septentrionalis Ho/mgr., and Orthocentrus pedestris Holmgr.; also a Muscid, Aricia
(Chortophila) megastoma Bohkem. On seven different days of July and August
Ekstam saw many small Diptera in Spitzbergen.

957. S. rivularis L. (Warming, Bot. Tids., Kjébenhavn, xvi, 1888, pp. 7-
10.)—The inconspicuous flowers of this boreal and arctic species are, according
to Lindman and Warming, at first slightly protogynous, and afterwards homogamous,
self-pollination being easily possible. Fruits are set early and rapidly. Warming
observed purely female plants with vestigial stamens in Spitzbergen.

Andersson and Hesselman describe the species as flowering in Spitzbergen from
the beginning of July to the end of August; in Beeren Island open blossoms were
noticed in favourable spots as early as June 18 (‘ Bidrag till Kanned. om Spetsbergens
o. Beeren Eil. Karlvaxtflora,’ pp. 29-30). The pollen is normally developed, and
fruits appear to ripen regularly. Ekstam says that in Spitzbergen the flowers are
odourless, white or pale red in colour (bright red in some Beeren Island specimens),

Fic. 130. Saxtfraga stenopetala, Gaud. (after Herm. Miller). 4. Flower at the beginning of the
first (male) stage. &. Flower towards the end of the same stage. C. Flower in the second (female)
stage, in section. a, outer stamen; 7, nectary ; ov, ovary.

and 6-8 mm. in diameter, or sometimes 5-10 mm. (‘Bliitenbiol. Beob. a. Spitz-
bergen,’ pp.17-18). In still incompletely open flowers, the anthers are beginning to
dehisce, and the stigmas are papillose ; homogamy may therefore be inferred.

According to Abromeit, several varieties have been recorded for Greenland,
including purpurascens Lange with reddish petals (‘Bot. Ergeb. von Drygalski’s
Groénlandsexped.,’ pp. 34-5).

Visirors——Ekstam noticed large and small flies on two days of July in
Spitzbergen.

958. S. stenopetala Gaud. (=S. aphylla Svernd.). (Herm. Miiller, ‘ Fertilisa-
tion,’ p. 244, ‘Alpenblumen,’ pp. 108-9.)—In this markedly protandrous species
self-pollination is excluded (cf. Fig. 130). The nectar is exposed, and the flower
mechanism is adapted to flies.

VisiTors.—Flies have been observed.

SAXIFRAGEAE 405

959. S. adscendens L.—The nectar is exposed in this species. Kerner
describes the flowers as protogynous. At first only the stigmas are mature, so that
cross-pollination may take place; later on the outer anthers bend over the stigmas,
and shed their pollen upon them, so that failing insect-visits automatic self-pollination
may be effected. In the third stage of anthesis the stigmas shrivel up, while the
inner anthers dehisce and offer their pollen to insect visitors.

Lindman says that, on the Dovrefjeld, the flowers are homogamous, and
effectively autogamous.

960. S. controversa Sternb.—Kerner describes this species as trimonoecious.

961. S. longifolia Lapeyr.—The white blossoms of this species are aggregated
into many-flowered inflorescences: the nectar is exposed. MacLeod describes them
as protandrous in the Pyrenees, becoming homogamous towards the end of anthesis,

Fic. 131. Saxtfraga granulaia, L. (from nature). 4. Flower in the first part of the first (male)
stage: several of the outer anthers have dehisced, or are already empty, while the inner ones are
still unripe, and the stigmas are immature. J. Flower in the second half of the same stage: all the
outer anthers are empty, those of the inner whorl are partly covered with pollen, partly unripe, and
the stigmas are still immature. C. Flower in the second (female) stage: all the anthers are empty,
and the stigmas are mature. a! and a’, outer and inner anthers; s, stigma.

so that automatic self-pollination is ultimately possible. Anthesis apparently lasts
for several weeks, the various stages following one another very slowly.

Visirors.—MacLeod observed a few Muscids (Bot. Jaarb. Dodonaea, Ghent, iii,
1891, Pp. 425).

962. S. ajugifolia L—

Visrtors.— MacLeod observed 4 species of flies in the Pyrenees.

963. S. granulata L. (Sprengel, ‘Entd. Geh.,’ pp. 242-4; Herm. Miiller,
‘Weit. Beob.,’ I, pp. 296-7; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 154,
‘Bloemenbiol. Bijdragen.’)—The white flowers of this species are markedly pro-
tandrous. Sprengel describes the green nectary as being situated on the upper
side of the ovary. The calyx holds the petals so closely together that they form
a tube, in the base of which the nectar is sheltered from rain. When the flower
opens the anthers are still unripe, and the filaments short. Two of the latter
rapidly elongate, and assume an oblique position, in such a way that their anthers,
which have meanwhile dehisced, lie immediately above the pistil. When these
stamens have shed their pollen, they bend back towards the petals, and two or

406 ANGIOSPERMAE—DICOTYLEDONES

three others take their place. The pollen continues to be shed for about three
days, and during this time the styles with their immature stigmas lie close together.
It is only after the anthers have emptied themselves that the styles elongate and
diverge, so that the stigmas occupy the position taken up by the anthers in the
first (male) stage.

Kirchner says that the flowers vary in size without any other correlated
differences.

Visirors.—I have seen a hover-fly (Eristalis arbustorum Z.), skg., and also
a beetle (Meligethes) at Kiel. Sprengel describes pollination by the blow-fly
Calliphora vomitoria Z.

Herm. Miiller (H. M.) and Buddeberg (Budd.) observed the following.—

A. Coleoptera. (a) Curculionidae: 1. Miarus graminis Schinh, (6) Der-
mestidae: 2, Anthrenus scrophulariae Z. (H.M.). B. Diptera. (a) Lmpzdae: 3.
Empis tessellata /., skg. (H.M.). (4) Syrphidae: 4. Eristalis arbustorum Z., skg.
(H.M.). C. Hymenoptera. (a) Apidae: 5. Andrena schrankella Wy/. 3, skg.
(H.M.); 6. Halictus malachurus &. 9, skg. and po-cltg. (H.M.); 7. H. minutissimus
K. 9, do. (H.M.); 8. H. morio Z. 9, do. (H.M.); 9. H. nitidiusculus K. 9, do.
(H.M.). (8) Zenthredinidae: 10. Cephus sp. (H. M.).

MacLeod noticed 2 short-tongued bees and 4 Muscids in the Pyrenees (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, pp. 423-4).

964. S. tridactylites L. (Sprengel, ‘Entd. Geh.,’ pp. 244-6; Herm. Miller,
‘Weit. Beob.,’ I, p. 297.)—Sprengel says that the flower mechanism of this species
agrees entirely with that of S. granulata. He also quotes the observation of Linnaeus,
‘sub florescentia germen stylo stigmatibusque destitutum,’ which cannot be in-
terpreted otherwise than as meaning that the styles and stigmas do not mature
until the anthers have withered.

Hermann Miller, on the other hand, describes the small white flowers as feebly
protogynous. When they open the stigmas are already mature. Soon afterwards
the anthers dehisce successively, first those of the outer whorl, then those of the
inner. As this takes place they regularly come into contact with the stigmas,
effecting automatic self-pollination at an early stage, and this is completely effective.
In dull weather the flowers remain shut, or close if they had previously opened.
During such weather the nectary, which surrounds the style as a yellow fleshy
Ting, does not secrete, but in the sunshine it produces glistening drops in the
midday hours.

Kerner states that, besides hermaphrodite blossoms, there are pseudo-herma-
phrodite pollen flowers and fruiting flowers on the same plant. Moreover, Warnstorf
also says that the protogynous hermaphrodite flowers are associated on the same
stock with pseudo-hermaphrodite ones of both kinds; sometimes the carpels aborting,
sometimes the stamens.

965. S. tricuspidata Rottb. (Warming, Bot. Tids., Kjébenhavn, xvi, 1888,
pp. 22-5.)—Warming states that the stellate flowers of this species are at first
slightly protandrous in Greenland, soon becoming homogamous. Self-pollination
is therefore always possible, though the probability of its occurrence varies in different
cases. As ripe fruits have been observed, it must be effective. The nectar is
exposed. Besides hermaphrodite flowers, Warming noticed purely female ones.

SAXIFRAGEAE 407

Abromeit describes two varieties of flower for Greenland; one with narrow
white petals spotted with yellow and purple-red, the other with small petals of
distinctly yellow colour (‘ Bot. Ergeb. von Drygalski’s Grénlandsexped.,’ pp. 35-7).

966. S. flagellaris Willd. (Warming, Bot. Tids. Kjébenhavn, xvi, 1888,
pp- 25-6.)—Of three plants of this species brought from Spitzbergen, Warming
describes two as protogynous; in the third the outer anthers were applied to the
stigmas in such a way that automatic self-pollination must have taken place. For
Nova Zemlia, Ekstam describes the odourless flowers as slightly protandrous or
homogamous, self-pollination being quite possible.

Andersson and Hesselman state that the species is in flower in Spitzbergen
from the beginning of July till September; ripe fruits were observed August 20, 1897
(‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ pp. 26-¥).
The bell-shaped, bright yellow flowers are 8-o-11-5 mm. long, and 9-5-13-0 mm.
broad; there is a swelling on either side the base of each petal. In Van Meyen
Bay the observers mentioned found the flowers to be protogynous at the beginning
of anthesis, the anthers subsequently ripening and coming into contact with the
stigmas. Ekstam, on the other hand, describes specimens from Advent Bay as
homogamous.

967. S. Cotyledon L.—Briquet says that in this protandrous species the outer
and inner stamens move in succession towards the middle of the flower (‘Etudes
de biol. flor. dans les Alpes occident.’). The diameter of the corolla may be as
much as 15 mm. The exposed nectar is secreted by a green disk. Flies alighting
upon the petals regularly effect cross-pollination. Kirchner points out that the
flower mechanism was long ago described by Sprengel (‘Entd. Geh.,’ p. 246), and
subsequently by Lindman (‘Bidrag till Kanned. om Skandin. Fjellvaxt. Blomn. o.
Befrukt.,’ p. 60).

On the Dovrefjeld, according to Lindman, the flowers smell rather strongly
of apples, and are markedly protandrous.

Visitors. — Lindman observed numerous flies and a humble-bee on the
Dovrefjeld.

968. S. hypnoides L.—

Visitors.—In Dumfriesshire, an Empid and 3 Muscids have been recorded
(Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 71).

969. S. cernua L. (Lindman, op. cit.; Warming, Bot. Tids., Kjébenhavn,
xvi, 1888, pp. 3-6.)—The conspicuous flowers of this species, according to Lindman
and Warming, are markedly protandrous as a rule on the Dovrefjeld, and also
in Greenland, Nordland, Finmark, and Spitzbergen. Occasionally, however, they
are protogynous, and this perhaps indicates a transition towards the production
of female flowers. The nectar is exposed. The inadequate production of fruits
is compensated for by the development of bulbils taking the place of flowers, both
in northern regions, and also, according to Kerner, in the Tyrol.

Andersson and Hesselman state that the species flowers in Spitzbergen from the
beginning of July to the beginning of August, here and there to the end of the latter
month: in Beeren Island a stock with fully developed flower-buds ready to open
was seen on June 18, 1898 (‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil.

408 ANGIOSPERMAE—DICOTYLEDONES

Karlvaxtflora,’ pp. 28-9). A well-marked small-flowered female form, and a specimen
with normal flowers, but only 56 % of functional pollen-grains, were also observed.
Ekstam (‘Bliitenbiol. Beob. a. Spitzbergen’) describes the flowers as possessing
a strong and agreeable odour of almonds: diameter 14-16 mm. (in some cases
18-22 mm.). The colour is usually pure white, but the centre of the corolla is
sometimes marked with red lines, and tinged with pink. Ekstam, confirming the
observations of Warming, states that marked protandry obtains for Spitzbergen.
Only the uppermost flower develops as a rule, the others being transformed into
bulbils. The flowers are extremely sensitive to light, and assume a fixed light-
position when the illumination is favourable. Ekstam did not find any ripe fruits
in Spitzbergen.

Abromeit says that terminal flowers with small petals completely hidden in the
calyx (var. crypfopetala K. Rosenvinge) were observed in Greenland (‘ Bot. Ergeb.
von Drygalski’s Grénlandsexped.,’ p. 34).

For Nova Zemlia, Ekstam describes the flowers as conspicuous (diameter up to
20 mm.) and smelling slightly of almonds. They are usually protogynous-homo-
gamous, sometimes protandrous-homogamous.

VisiTors.—On six days of July and August Ekstam saw small and medium-sized
flies in Spitzbergen. He also noticed a medium-sized fly in Nova Zemlia.

970. S. nivalis L. (Lindman, ‘Bidrag till Kanned. om Skandin. Fjellvaxt.
Blomn. o. Befrukt.’; Warming, Bot. Tids., Kjébenhavn, xvi, 1888, pp. 14-17.)—
The flowers of this species are small and inconspicuous, the erect petals being little
longer than the calyx. The nectar is half concealed. Warming says that protogyny
is not well marked, while homogamy is not infrequent, and cultivated plants may
even exhibit protandry. Lindman observed homogamy with a tendency to slight
protandry on the Dovrefjeld. According to Warming, self-pollination is inevitable
in protandrous plants; but this is less easily possible in the case of the Norwegian
form, which possesses reflexed stamens. Both Lindman and Warming observed
ripe fruits.

Ekstam says that the flowers are commonly protandrous in Nova Zemlia,
though some are homogamous or slightly protogynous-homogamous. The diameter
of the flowers in Arctic Siberia is 10 mm., according to Kjellman.

Andersson and Hesselman state that the species flowers in Spitzbergen from
the beginning of July to the end of August or the beginning of September, and sets
fruit every year (‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’
p- 22). The flowers are 5-5-8 mm. in diameter, and white or bright greenish-
yellow in colour, sometimes with a reddish spot near the tip of each petal. The
stamens are orange-coloured, and the ovary greenish-yellow. The pollen-grains
quickly swell up in distilled water. Ekstam found the flowers in Spitzbergen proto-
gynous-homogamous, and noticeably fragrant. He states that the stigmas are
situated at the same or a lower level than the anthers, which bend inwards when
they dehisce, and autogamy is thus made possible (‘ Bliitenbiol. Beob. a. Spitzbergen,’
p. 11).

For the variety 8 éenuzs Wahl., in Greenland, Abromeit gives the colour of the
flowers as white or light purple, and that of the filaments as more or less purple
(‘ Bot. Ergeb. von Drygalski’s Grénlandsexped.,’ p. 32).

SAXIFRAGEAE 409

Vistrors.—Ekstam saw a large and a small fly in Spitzbergen. He also noticed
medium-sized flies in Nova Zemlia.

971. S. juniperifolia Adams. (Kerner, ‘Nat. Hist. Pl.” Eng. Ed. 1, p. 326.)—
Kerner says that geitonogamy takes place in this species, for though the direction of
the styles and the position of the stigmas remain unchanged, the filaments elongate
and curve in such a way that the pollen can reach the stigmas of. neighbouring
flowers.

972. S. umbrosa L. (Herm. Miller, ‘ Fertilisation,’ p. 245.)—Hermann Miller
says that, ‘the prettily dotted petals of S. rotundifolia and S. umbrosa L., seem to be
due to the influence of certain elegant and prettily coloured Syrphidae, by which they
are especially visited and cross-fertilized.’

Visitors.—Herm. Miller noticed the Syrphid Ascia podagrica Z. Plateau
observed a Muscid (Lucilia caesar Z.), Syrphids (Melanostoma mellina Z., Helophilus
pendulus Z., Syrphus corollae /.), and bees (Apis, Andrena nana K., Megachile
ericetorum Lep. (=M. fasciata Sm.), Odynerus quadratus P2z.(?) ).

973- S. crassifolia L.(=Bergenia biflora Moench). (Herm. Miller, ‘Fertilisation,’
Pp. 243, 247, ‘Weit. Beob.,’ I, p. 298.)—The flowers of this species are protogynous.

Visitors.—Herm. Miiller observed bees, skg. (Apis, Bombus hortorum Z. 9,
B. pratorum Z. 9). Loew saw a humble-bee (B. terrester Z. 9), skg., in the Berlin
Botanic Garden.

269. Bergenia Moench.
974. B. subciliata A. Br.—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Hymenoptera. (a) Apidae: 1. Anthophora pilipes #. 4, skg.; 2. Apis
mellifica Z. ¥, skg.; 3. Osmia rufa Z., skg. (4) Vespzdae: 4. Odynerus parietum L. 9
andé. B. Lepidoptera. Rhopalocera: 5. Colias rhamni Z., skg.

270. Chrysosplenium Tourn.

Flowers inconspicuous, golden-yellow, or greenish in colour; homogamous,
protogynous, or slightly protandrous. The nectar is exposed, and secreted by a disk
surrounding the styles. The foliage-leaves around the flowers are mostly streaked
with golden-yellow, and increase conspicuousness. Besides hermaphrodite flowers,
there are often some purely male ones as well.

975- C. alternifolium L. (Sprengel, ‘Entd. Geh.,’ p. 241; Ricca, Atti Soc.
ital. sc. nat., Milano, xiii, 1870, p. 257; Herm. Miiller, ‘Fertilisation,’ pp. 245-6;
MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 290-1; Kirchner, ‘Flora v.
Stuttgart,’ p. 406; Warming, ‘ Arkt. Vaxt. Biol.’ p. 7; Lindman, ‘ Bidrag till Kanned.
om Skandin. Fjellvaxt. Blomn. o. Befrukt.’; Kerner, ‘ Nat. Hist. Pl.” Eng. Ed. 1, I,
p- 379; Knuth, ‘Bloemenbiol. Bijdragen.’)—Hermann Miler describes the in-
conspicuous yellow flowers of this species as homogamous, but Ricca says they
are slightly protogynous, with persistent stigmas. Ekstam, for Nova Zemlia, speaks

410 ANGIOSPERMAE—DICOTYLEDONES

of them as protogynous-homogamous. I found only homogamous flowers in the
neighbourhood of Kiel. Each blossom is in the form of a shallow cup, 5-7 mm.
in diameter; from the middle of this the two projecting styles (about 1 mm. long)
project and curve outwards. They carry the somewhat thickened smooth stigmas
at their tips, and are surrounded at the base by a broad, fleshy, yellowish disk,
on which numerous drops of nectar spread out to form a thin layer. The eight
stamens are erect; the anthers rise about 1 mm. above the disk, and are at the
same level as the stigmas, which remain receptive throughout anthesis. The anthers
dehisce in succession and get covered with pollen all round. As there are no
petals, and the four sepals spread out flat, the 6-12 or more flowers that are
usually crowded together into a cyme make up a practically level surface, which is
extended by the uppermost golden-yellow foliage leaves, so that there is a considerable
area by which numerous small short-tongued insects are attracted. As these visitors
usually touch one or several stamens with one side of their bodies, and the stigmas
with the other side, they commonly effect cross-pollination, but owing to the erratic
way in which they creep about the flowers
and inflorescences, self-pollination is also
oes a frequent occurrence. It can only take

place automatically when in exceptional

cases the flowers are nearly or quite

~ vertical, so that pollen can fall upon the
: stigmas. Kerner says that at a later
stage the peduncles curve downwards, so

that the flowers come to be inclined or
cater © Warnsag teapeamhatereition.Z. pendulous, and the stigmas being brought
a flower (x 8). into the line of fall of the pollen autogamy
necessarily follows.

Lindman found the flowers to be homogamous on the Dovrefjeld, but owing
to the distance between anthers and stigmas automatic self-pollination is here
scarcely possible. The flowers observed were larger (7 mm. diameter) than those
from the neighbourhood of Stockholm.

Andersson and Hesselman state that the var. ée/randrum Lund flowers in
Spitzbergen from the end of June till at least the end of August, when ripe fruits
were also observed (‘ Bidrag till Kanned. om Spetsbergens o. Beeren Ei]. Karlvaxt-
flora,’ p. 31).

Hermann Miiller calls attention to the fact that the pollination of this plant
is also occasionally due to snails. He found small snails (young Succinea) on many
flowers, sometimes creeping about, sometimes devouring the styles or one or several
stamens. He could generally see a number of pollen-grains in the slimy trail left
on the flowers by these guests, and in several cases noticed the direct transfer to
the stigmas. I can partly confirm these observations, for though I have not actually
seen the snails at work, I have often noticed little slimy trails among the flowers,
obviously made by them, and have also frequently remarked the results of their
activity, in the form of gnawed edges and surfaces of the foliage-leaves and flowers.
Besides this, I have observed many ants and minute Muscids licking nectar, but did
not collect them and determine their species.

SAXIFRAGEAE 411

VisiTors.—Herm. Miiller observed the following, all nect-lkg.—

A. Coleoptera. (a) Colyditdae: 1. Corticaria gibbosa Host. (b) Curculionidae:
2. Apion onopordi K.; 3. A. varipes Germ. (c) Phalacridae: 4. Olibrus aeneus F.
B. Diptera. (a) Cectdomyzdae: 5. 6 individuals. (4) Chironomidae: 6. 3 indi-
viduals, belonging to minute species. (c) Muscidae: 7. Sciomyza cinerella Fall.
(d) Mycetophilidae: 8. 5 individuals. (e) Simudidae: 9. Simulia sp. C. Hymeno-
ptera. (a) Cynipidae: 10. Eucoila Westw. sp. (4) Formicidae: 11. Lasius niger
L.%; 12. Myrmica levinodis Wy/. ¥; 13. M. ruginodis My/. ¥.

Herm. Miiller saw 12 Diptera, an ant, 2 Ichneumonids, and a beetle in the
Alps (‘ Alpenblumen,’ p. 89).

Alfken noticed 2 bees (Andrena gwynana X. 9 and 4, skg., and A. parvula K. 9
and , do.) at Bremen. MacLeod observed 2 moths, 3 short-tongued Hymenoptera,
3 beetles, and a Neuropterid in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
Pp. 291).

Burkill records the following for the coast of Yorkshire, all skg. (‘Fertlsn. of
Spring Fls.’).—

A. Coleoptera. 1. Lathrimaeum atrocephalum Gy//.; 2. Tachyporus chryso-

melinus Z. B. Diptera. Afuscidae: 3. Cecidomyia sp. and 3 other small flies.
C. Hemiptera. 4. One sp.

976. C. oppositifolium L. (Herm. Miiller, ‘ Fertilisation,’ p. 247,‘ Weit. Beob.,’
I, p. 298.)—Hermann Miller states that the hermaphrodite flowers of this species
are protogynous with persistent stigmas. In other respects their mechanism agrees
with that of C. alternifolium, though they and the bracts surrounding them are
smaller and less conspicuously coloured, besides which automatic self-pollination
is easy. Kobus describes the species as andromonoecious, developing numerous
purely male flowers when it grows in thick clumps (D. bot. Monatsschr., Arnstadt, i,
1883, P- 74).

Burkill (‘Fertlsn. of Spring Fls.’) remarks that the species is gynodioecious
on the Yorkshire coast, and that female plants are common throughout the whole
district. They are readily recognized by their green flowers, for the golden-yellow
colour of the hermaphrodite blossoms is almost entirely absent. Female flowers
are also smaller than hermaphrodite ones. No trace of pollen is to be found in
them: even anthers are almost completely absent, and any which may remain are
functionless.

Visirors.—Herm. Miiller observed 2 beetles (Coccinella bipunctata Z. and C.
impustulata Z.) and 2 flies (Chlorops scalaris M/g. and Musca domestica Z.) seeking
the drops of nectar which are distinctly visible.

Burkill observed the following on the Yorkshire coast.—

A. Araneida. 1.1 sp., lying in wait. B. Collembola. 2. Lepidocyrtus sp.
C. Diptera. (a) Muscrdae: 3. Lonchoptera sp.; 4. Sepsis nigripes AZ. (6) ALyceto-
philidae: 5. Exechia sp.; 6. Sciara sp. (c) Syrphidae: 4. Melanostoma quadri-
maculata Verrall. (d) Tipulidae: 8. Chironomus sp. D. Hymenoptera. Jchneu-
monidae: g. One small sp. E.Thysanoptera. 1o. Thrips sp.

In Dumfriesshire, an Ichneumonid, 3 Muscids, and a beetle have been recorded
(Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 72).

412 ANGIOSPERMAE—DICOTYLEDONES

977. C. tetrandrum Fries (= C. alternifolium, according to the Jndex
Kewensis). (Warming, ‘ Arkt. Vaxt. Biol.,’ pp. 4-7.)—Warming investigated plants
from Spitzbergen. The flowers are greenish, less open than in C. oppositifolium,
and with a scarcely developed nectary. Besides homogamy, slight protandry was
observed. Automatic self-pollination regularly obtains by contact of the stigmas with
the anthers of the two outer stamens. This must be effective, for almost every
flower sets fruit.

Fic. 133. Chrysosplentum tctrandrum, Fries (after E. Warming). 4. Longitudinal section through
an almost closed flower. Anthers unripe, styles short. 2. Do. through a fully open young flower. The
stigmas touch some of the ripe anthers, which deposit a quantity of pollen upon them (cf. D). C. A
fertilized flower. The formation of seeds and setting of fruit have begun; the free parts are more erect,
and have closed in somewhat. J. Style of B more highly magnified, showing the stigma covered with
numerous pollen-grains. £, F. Sterile stamens, the latter in connexion with its perianth leaf. G. Longi-
tudinal section through a sterile stamen. A. Normalanther. _/. Scarcely ripe fruit. A. Do. through
a similar fruit; the seeds have been removed, but some of the funicles are still present in part. (4, B, C,
J, & xX 8.)

271. Heuchera L.
978. H. cylindracea Lind!l.—

Visirors.—Loew observed 2 bees in the Berlin Botanic Garden (Apis mellifica
L.%, skg., and Halictus cylindricus /. 9, do.).

272. Tellima R. Br.
979. T. grandiflora R. Br.—

Visirors.—Loew observed a bee (Apis mellifica Z. ¥) and a butterfly (Pieris
brassicae Z.), skg., in the Berlin Botanic Garden,

273. Tiarella L.
980. T. cordifolia L.—Francke describes this species as protogynous (‘ Inaug.-
Dissert.,’ Halle, 1883). The anthers ripen in succession at long intervals.

274. Parnassia L.

Flowers white in colour, markedly protandrous, with half-concealed nectar.
There are five fringed glandular staminodes opposite the petals. Each of them
swells into a disk, with two shallow depressions on the upper side, into which
the tolerably exposed nectar is sparingly secreted.

SAXIFRAGEAE 413

981. P. palustris L. (Sprengel, ‘Entd.Geh., pp. 166-73; C. W. Ritter,in Hoppe’s
“Bot. Taschenbuch,’ Regensburg, 1803, Nachschrift, p. 181; Delpino, ‘ Ult. Oss.,’
p. 168 ; Herm. Miiller, ‘ Fertilisation,’ pp. 247-8, ‘ Alpenblumen,’ pp. 111-13 ; Kerner,
‘Nat. Hist. Pl.,’ Eng. Ed. 1, II, pp. 209, 213, 249, 251, 307; Verhoef, ‘Bl. u. Insekt.
d. Ins. Norderney’; Knuth, ‘Bl. u. Insekt. a.d. nordfr. Ins.,’ pp. 34-5, 150, ‘ Bliiten-
biol. Notizen.’)—This species bears protandrous deceptive flowers. Their mechanism
was very fully described by Sprengel, but he remained in doubt as to whether
they were day or night flowers. Since Sprengel’s time this most interesting flower
has been studied by Ritter, Hermann Miller, and others, and we are particularly
indebted to Miiller for an explanation of the significance of its individual parts.
The five white petals are furrowed by colourless veins, and in front of them there
are five characteristic yellow-green organs, the staminodes, each of which possesses
a short broad stalk that expands into a disk bearing 7-13 or even as many as
25 Stalked glands, serving to attract insects. Some nectar is secreted on either
side the insertion of the stalk. When the flower opens the anthers are still

Fic. 134. Parnassia palustris, L. (after Herm. Miller). 4. Flower after removal of three sepals
and four petals, seen from above, immediately after opening. One stamen has lengthened, and its anther
lies upon the middle of the pistil, the stigma being still immature. This stamen is beginning to dehisce,
and to cover its upwardly directed outer surface with pollen. &. Flower after removal of calyx, corolla.
and staminodes. Four stamens have already shed their pollen and curved back; the fifth, which is covered
above with pollen, lies upon the still immature pistil. C. The same flower in the second (female) stage.
The stamens have all dehisced; the stigma, which is here represented as trilobed, but is usually four-lobed,
ismature. DD. Staminode, more highly magnified. #, nectar; #, petal; s, sepal.

unripe, their filaments are short, and they lie close to the conical pistil, the stigma
of which is also still immature. The stamens now ripen successively, their filaments
elongating to such an extent that the anthers lie exactly upon the top of the ovary,
their dehisced pollen-covered surfaces facing upwards. After about a day an anther
has shed its pollen and bends outwards, when another takes its place at the top
of the pistil, and so on. At the end of four days, when all the anthers are empty,
the apical stigma unfolds, occupying exactly the place taken up in the first (male)
stage by a ripe anther.

The stalked glands of the staminodes attract insects by their glistening appearance,
which apparently suggests the presence of abundant nectar. The more intelligent

414 ANGIOSPERMAE—DICOTYLEDONES

insects, however, do not allow themselves to be deceived by this, but the stupider
ones (flies and beetles) are repeatedly attracted, and effect cross-pollination as they seek
out the scanty nectar. The smaller flies, however, usually go round and round the
flower, licking the nectar without touching either pollen or stigma, and therefore render
no service to the flower. But larger insects mostly settle upon the middle of the
flower to suck, turning round from one nectary to another, so that in younger
blossoms they cover their under-side with pollen, which they transfer to the stigmas
of older ones}.

The flowers are very small (25-13 mm. in diameter) in the Alps, and possess
only three-lobed stigmas (cf. Fig. 134), though elsewhere there are four-lobed ones.
In the Scandinavian highlands, according to Lindman, the flowers are also often
very small, sometimes not more than 11 mm. in diameter. They are fragrant,
smelling like honey.

During September, 1896, I made a noteworthy observation on flowers placed
in water in a shady part of my room, and which kept fresh for a week. Under
these circumstances the buds developed into purely homogamous flowers; all five
stamens diverged and turned their dehisced anthers outwards. The stigmas matured
simultaneously, so that, the flowers being obliquely placed, pollen could fall upon
them, as actually happened in some cases. Flowers kept in a room, therefore,
behaved quite differently from those in the open air, confirming in this case the
advice of Sprengel (‘Entd. Geh.,’ p. 22) not to bring flowers from the garden or
the fields, but rather to study them among their natural surroundings.

Visitors. — Herm. Miiller (H.M.) and myself (Kn.) have observed the
following.—

A. Coleoptera. Coccznellidae: 1. Coccinella conglobata Z., very common,
nect-lkg. (H.M.); 2. C. septempunctata Z., do. (H. M., Kn.). B. Diptera. (a)
Muscidae: 3. Aricia sp. (Kn.); 4. Lucilia caesar Z.(Kn.); 5. Pollenia vespillo 7,
skg. (H.M.); 6. Sarcophaga carnaria Z. (H. M., Kn.); 7. Smaller Muscidae (H. M.).
(6) Syrphidae: all skg.: 8. Eristalis arbustorum Z. (H.M., Kn.); 9. E. nemorum Z.
(H. M., Kn.); 10. E. pertinax Z. (Kn.); 11. E. tenax Z. (Kn.); 12. Helophilus
floreus Z., freq. (H.M.); 13. H. pendulus Z., especially freq. (Kn.); 14. Melanostoma
mellina Z. (H. M., Kn.); 15. Melithreptus menthastri Z. (H. M.); 16. M. scriptus Z.
(H.M.); 17. M. taeniatus 7g. (H.M., Kn.); 18. Syritta pipiens Z., freq. (H. M.,
Kn.) ; 19. Syrphus balteatus Deg., very common, now and then po-dvg. (H. M., Kn.);
20. S. excisus Ze//. (H.M.); 21. S. pyrastri Z., freq. (H.M., Kn.); 22. S. ribesii
L., do. (H.M., Kn.). (c) Zipulidae: 23. Tipula oleracea Z. (H.M.). C. Hymeno-
ptera. (a) Formicidae: 24. Formica sp.(Kn.). (0) Jchneumonidae: 25. Numerous
small species, skg. (H.M.). (c) Sphegzdae: 26. Gorytes campestris Afi//. (H.M.);
27. Pompilius viaticus Z.(H. M.). (d) Zenthredinzdae: 28. Tenthredo sp., skg. (H. M.).

1 In my work, ‘Blumen und Insekten auf den nordfriesischen Inseln’ (1892, pp. 34-5), I have
pointed out that the term ‘deceptive flower’ is not strictly applicable to Parnassia, which secretes at
least as much nectar as most Umbelliferae. Though fairly abundant it does not taste sweet to
us. That it is, however, palatable to insects attracted by the pronounced honey-odour of the flower,
is shown by the zeal with which these visitors seek it out; and, besides, the numerous lacerations of
the ‘nectar apparatus’ indicate that boring and sucking take place. Professor Ludwig of Greiz has
told me in a letter that he shares my opinion, adding that he could not understand why the blossoms
have been described as ‘deceptive flowers.’

SAXIFRAGEAE 415,

Alfken (A.) and Leege (L.) record the following for Juist.—

A.Coleoptera. Zelephoridae: 1. Cantharis fulva Scop., freq. (A.,L.). B. Diptera.
(a) Dolichopodidae: 2. Dolichopus plumipes Scop. (L.). (4) Muscidae: 3. Cynomyia
mortuorum Z. (A.); 4. Lucilia caesar Z. (A.); 5. Spilogaster quadrum /., freq. (A.);
6.S. sp. (A.). (¢) Stratiomyidae: 7. Nemotelus notatus Ze/t., freq., po-dvg. and skg.
(A.); 8. Odontomyia viridula /., one (A.). (d) Syrphidae: 9. Eristalis arbustorum
L.(A.); 10. Melithreptus strigatus S/aeg. 9 and 4, freq. (A.); 11. Platycheirus sp.
(A.); 12. Syrphus balteatus Deg., freq. (A.); 13. S. trilineatus Z., rare (A.).
C. Hymenoptera. (a) Formicidae: 14. Lasius niger Z. (A.). (4) Ichneumonidae :
15. Glypta fronticornis Gr. (L.); 16. Lissonota commixta Her. (L.). (c) Scolitdae :
17. Tiphia femorata /. 9, one (A.). D. Lepidoptera. (a) Satyridae: 18. Hip-
parchia semele Z., very common, skg. (A.). (4) Moc/uidae: 19. Plusia gamma Z..,
freq., skg.

Verhoeff saw the following in Norderney.—

A. Coleoptera. Mitidulidac: 1. Meligethes aeneus Z., one. B. Diptera.
Bibiontdae: 2. Dilophus vulgaris JZg., very common.

Lindman observed numerous flies, a beetle, and a Lepidopterid on the Dovrefjeld.
Herm. Miiller saw 43 species of flies, 2 beetles, 8 Hymenoptera, and 6 Lepidoptera
in the Alps. MacLeod noticed an Ichneumonid, a Lepidopterid, a hover-fly, and 7
Muscids in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 424-5).
Delpino saw a hover-fly, Helophilus floreus Z., at Florence.

Burkill observed the following on the East coast of Scotland (‘ Fls. and Insects in
Gt. Britain,’ Part I).—

A. Coleoptera. Nitzdulidae: 1. Meligethes picipes S/urm., skg. B. Diptera.
(a) Bibionidac: 2. Scatopse brevicornis AZg. (0) Muscidae: 3. Anthomyia brevicornis
Zell., po-dvg.; 4. A. radicum Z., freq., skg.; 5. Calliphora erythrocephala AZg., skg. ;
6. Coelopa sp., po-dvg.; 7. Hydrellia griseola Fad/.; 8. Phytomyza sp.; 9. Sarco-
phaga sp.; 10. Sepsis cynipsea Z. (c) Phoridae: 11. Phora sp. (d) Syrphidae:
12. Eristalis tenax Z.; 13. Helophilus pendulus Z., skg.; 14. Melanostoma mellina
L.; 1g. Platycheirus albimanus #.; 16. Sphaerophoria scripta Z., skg. (e) Myce/o-
philidae: 17. Sciara sp. C. Hemiptera. 18. Onesp. D. Hymenoptera. (a)
Formicidae: 19. Formica fusca Z., skg.; 20. Myrmica rubra Z., skg. (0) Lchneu-
monidae: 21. 3 species.

275. Philadelphus L.

Large, white, very fragrant flowers; with half-concealed nectar, secreted by
a disk on the inferior ovary.

982. P. coronarius L. (Sprengel, ‘Entd. Geh.,’ p. 267; Herm. Miiller,
‘Fertilisation,’ pp. 248-9, ‘ Weit. Beob.,’ II, pp. 237-8; Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896; Knuth, ‘Blumenbiol. Beob. a. d. Ins. Riigen.’)—In this
species, according to Hermann Miiller, the stigmas are mature when the flower
opens, so that crossing is necessarily effected by insect visitors which have previously
been dusted with pollen. After a time the anthers dehisce, those of the outer
stamens doing so first. As the numerous anthers are very close to the stigmas,
which are situated in the line of fall of the pollen, automatic self-pollination readily
takes place should insect-visits fail. If a flower is not visited till the anthers have
dehisced, self- and cross-pollination are both possible. Warnstorf describes the
normal pollen-grains as yellow, ellipsoidal, closely tuberculated, about 25 » long
and 12-13 » broad. They are mixed with much smaller, apparently abortive grains.

416 ANGIOSPERMAE—DICOTYLEDONES

The large white flowers attract numerous insects owing to their conspicuousness
and pronounced fragrance.

Vistrors.—MacLeod saw 2 Noctuids in Belgium. Herm. Miiller (H.M.) chiefly
records bees, while in Schleswig-Holstein and Riigen I myself (Kn.) observed Diptera
for the most part. Our joint list is as follows. —

A. Coleoptera. (a) Dermestidae: 1. Anthrenus pimpinellae # (H.M.); 2. A.
scrophulariae Z. (H.M.). (0) Alordellidae: 3. Mordella aculeata Z. (H.M.). (c¢)
Nitidulidae: 4. Meligethes, po-dvg. (Kn., H.M.). (d) Scaradbaerdae: 5. Phyllopertha
horticola Z., gnawing the flowers (H.M.). (e) Zélephoridae: 6. Dasytes, freq.
(H. M.); 7. Malachius bipustulatus Z., dvg. the anthers (H.M.). B. Diptera.
(a) Muscidae: 8. Sepsis sp. (H. M.). (6) Syrphidae: g. Ascia podagrica /., freq.,
skg. and po-dvg.(H. M.); ro. Eristalis arbustorum Z., po-dvg. (Kn.);_ 11. E. pertinax
Scop., do. (Kn.); 12. E. tenax Z., do. (Kn.); 13. Helophilus floreus Z., do. (H. M.);
14. Rhingia rostrata Z., skg. (Kn., H.M.); 15. Syritta pipiens Z., po-dvg. (Kn.,
H.M.); 16. Syrphus ribesii Z.,skg. and po-dvg. (Kn., H.M.); 17. Volucella bomby-
lans Z., po-dvg. (Kn., H.M.); 18. V. pellucens Z. 9, do. (Kn.). C. Hymenoptera.
(a) Apidae: 19. Andrena albicans A/i//. 6 and g, very numerous, skg. and po-cltg.
(H. M.); 20. A. dorsata A. 9, po-cltg. (H. M.); 21. A. fasciata Wesm. 9, do. (H. M.);
22. A. fulvicrus K. 9, do. (H. M.); 23. A. nitida Fourcr. 9, do. (H. M.); 24. A.
tibialis A. 9, skg. (H.M.); 25. A. trimmerana A’ 9, skg. and po-cltg. (H.M.); 26.
Apis mellifica Z. 9, freq., skg. and po-cltg. (Kn., H.M.); 27. Bombus agrorum F.
(Kn.), skg. for a short time (H. M.); 28. B. lapidarius Z. §, skg. (Kn.); 29. B.
pratorum Z. ¥, skg. and po-clig. (H. M.); 30. Halictus leucozonius Schr. 9, po-cltg.
(H.M.); 31. H. sexnotatus A. 9, do. (H.M.); 32. Osmia rufa Z. 9, freq., po-cltg.
(H.M.); 33. Prosopis armillata Wy/. 8, po-dvg. (H.M.); 34. Psithyrus barbutellus
&. 9, skg. (6) Formicidae: 35. Lasius niger Z. ¥, skg. (H.M.). D. Lepidoptera.
Rhopalocera: 36. Pieris brassicae Z., skg. (H.M.); 37. P. napi Z., do. (H.M.);
38. P. rapae Z., do. (Kn.).

Cobelli gives 9 Hymenoptera not observed by Herm. Miiller (Nuovo Giorn. bot.

ital., Firenze, xxv, 1893).
276. Deutzia Thunb.

983. D. crenata Sieb. et Zucc.—

Vistrors.—I observed 3 bees, skg., in the island of Riigen—r. Apis mellifica
L.%; 2. Bombus terrester Z. 9; 3. B. lapidarius Z. ¥.
Alfken noticed the bee Andrena nigroaenea X. 9 and 4, skg., at Bremen.

277. Ribes L.

Flowers mostly greenish-yellow in colour, more rarely red or yellow; frequently
associated in many-blossomed racemes. The nectar is free or concealed, and
secreted by an epigynous disk. It may be so deeply placed (R. aureum) as to be
accessible only to long-tongued bees. Sometimes the flowers are gynodioecious.

Hermann Miller (‘ Weit. Beob.,’ I, pp. 298-300) gives the following oecological
series of our wild and cultivated species of Ribes.—R. alpinum is the least specialized,
its nectar being secreted in a shallow receptacular depression, and accessible to
the shortest-tongued insects. In R. rubrum the depression is much deeper, and its
bottom is covered with nectar. In shape it is approximately hemispherical, but
its outer part is expanded. The downwardly directed bells of the gooseberry
(R. Grossularia), though hardly deeper than those of R. rubrum, are somewhat

i”

SAXIFRAGEAE 417

narrowed near the mouth by stiff hairs projecting from the edge of the receptacle
and from the style, constituting a grating-like floor to the bell. The flowers are
also rendered more difficult of access to flies, and better adapted to bees by being
pendulous. The bells of R. nigrum are yet deeper, being almost spherical. They
too are pendulous, and are still better adapted to bees than those of the last species.
The flowers of R. sanguineum are tubular, though scarcely deeper than those of
R. nigrum (3 mm.). But owing to the erect position of the petals, the tube is
considerably prolonged (to more than 5 mm.). In spite of the tolerably erect
position of the flowers, they are usually visited only by bees. The last term of
the series is constituted by the flowers of R. aureum, which are in the form of
long tubes (10-11 mm.). These are further prolonged some 3 mm. by the erect
petals, so that the nectar is only accessible to very long-tongued bees. Cross-
pollination by insects is secured by dioecism in the flowers of R. alpinum, but in
all the other species by the relative pggitions of stamens and stigmas, which are
successively touched by different sides of insect visitors. Since the hermaphrodite
species are homogamous, the possibility of automatic self-ppllination, is not excluded
in them.

984. R. alpinum L. (Herm. Miiller, ‘Fertilisation,’ p. 249.)—The flowers of
this species belong to class E. The greenish-yellow flowers are rendered con-

Fic. 135. Rrbes alpinum, L. (after Herm. Miller). (1) Femaic flower seen from above. (2) The
same in longitudinal section, seen from the side. (3) Do. of male flower. a, anthers: a’, vestigial anthers ;
nm, nectary ; 7, petal; s,sepal; s/, stigma; s/’, vestigial stigma.

spicuous by the sepals, under which the minute petals are almost completely hidden.
Hermann Miller describes the free upper part of the receptacle as forming a slightly
concave depression, in which nectar is secreted. ‘The species is dioecious, the male
flowers being a little larger than the female ones, and of a yellowish-green colour.
They are therefore more conspicuous than the greener female flowers, and conse-
quently receive earlier visits. The female flowers possess vestigial stamens, and
the male ones a vestigial pistil.

Visitors.—Herm. Miiller observed the following. —

A. Diptera. (a) Afuscidae: 1. Scatophaga merdaria /.; 2. S. stercoraria L.
(6) Syrphidae: 3. Syritta pipiens Z., all three freq., skg. B. Hymenoptera.
Apidae: 4. Andrena albicans AZ#//. 9 and 4, very freq., skg. and po-dvg.; 5. A.
ewynana XK. 3, skg.; 6. A. parvula X.%, do.; 7. Halictus nitidus Schenck 6, do. ;
8. H. nitidiusculus K. 9, po-cltg.; 9. Sphecodes gibbus Z. 9, skg.

985. R. nigrum L. (Herm. Miller, ‘Fertilisation, p. 250; MacLeod, ‘ Nouv.
recherches sur la fertlsn. de quelques pl. phanérog.’)—The flowers of this species
belong to class C. They possess a characteristic odour, and Hermann Miiller

DAVIS. 1 Ee

418 ANGIOSPERMAE—DICOTYLEDONES

describes them as homogamous. The tips of the sepals are of a reddish colour,
and the small petals are whitish. The latter incline together above, bringing the
introrse anthers so close to the stigma that an insect probing for the nectar secreted
in the base of the flower must touch one or two of the ripe anthers with one side
of its head and the stigma, which projects somewhat beyond the anthers, with the
other side. Cross-pollination necessarily follows. There are, however, but very
few insect visitors, and it is usual for automatic self-pollination to take place, the
pollen falling from the anthers upon the reflexed margin of the stigma.

Visirors.—Herm. Miiller saw Apis mellifica Z. ¥ visiting the campanulate
flowers (3 mm. deep). MacLeod made the same observation in Belgium. The
latter also noticed that the honey-bee extracts nectar not
only from open blossoms, but also opens the older buds
with its jaws, and during this process dusts the already
mature stigma with pollen brought from other flowers.
MacLeod further observed ants, which use a lower flower
as a ladder in their attempts to reach the nectar, from
which the pendulous character of the flowers and the
reflexed sepals keep them away. From below they

Fic. 136. Ribes nigrum,L. licked the stigmatic secretion, but were unable to
(after Herm. Miller). Flower gm , bd e

seen from the side. @, anther; penetrate into the base of the flower. Plateau (Belgium)
. etal; s, sepal ;
pee ely Ai) eae eta saw Bombus terrester Z. Schenck (Nassau) observed

Bombus hypnorum Z. 9 and B. pratorum Z. 9.

986. R. rubrum L. (Herm. Miller, ‘ Fertilisation,’ p. 250.)—The flowers of
this species belong to class C. They are greenish-yellow in colour, and Hermann
Miiller describes them as homogamous. The bells are tolerably flat, and widely
open, so that the nectar is easily accessible. As in R. nigrum, insect visitors usually
effect crossing. Automatic self-pollination is only possible for flowers that hang

obliquely, in which case pollen falls from
x the stamens upon the underlying stigma.

Visitors. — Herm. Miiller observed
the following. —

A. Hymenoptera. (a) Aprdae:
1. Andrena fulva Schr. 9, freq., skg. and
po-cltg.; 2. A. parvula A. 4, skg.; 3. A.

é - i : :

Fic. 137. Rebes rubrum, L. rom nature). smithella X. 6, do. ae Apis mellifica L. %,
A. Flower in section. 2. Flower seen from above. freq., skg. and po-cltg. (8) Tenthredinidae :
a, anthers; ca, calyx; co, corolla; s, stigma. 5. Pteronus hortensis H7z., skg.

Alfken saw 2 bees at Bremen, both
skg.—1. Apis mellifica Z. %; 2. Nomada borealis Ze//. 6.
The following were observed by the authorities, and at the places stated.—

Loew (Brandenburg) the hover-fly Syrphus lunulatus J/g. (‘ Beitrage,’ p. 37).
F. F. Kohl (Tyrol) the ruby-wasps Chrysis austriaca /. and C. fulgida Z., and the
true wasp Odynerus trifasciatus #. Plateau (Belgium) Apis.

987. R. aureum Pursh. (Herm. Miiller, ‘Fertilisation,’ p. 251, ‘ Weit. Beob.,’
I, p. 301; Knuth, ‘Bloemenbiol. Bijdragen’; Warnstorf, Verh. bot. Ver., Berlin,

SAXIFRAGEAE 419

xxxviii, 1896.)—In this species the free part of the receptacle is 10-11 mm. long,
forming a tube which is prolonged another 3 mm. by the erect petals. The nectar
is therefore only accessible to long-tongued bees, and the flowers belong to class Hh.
In other respects their mechanism agrees with that of R. rubrum. Warnstorf
describes them as protogynous: the stigma matures in the bud, and is very sticky,
The style projects beyond the stamens, so that autogamy is prevented. The pollen-
grains are white, irregularly roundish-tetrahedral, up to 41 » in diameter.

The flowers are at first bright yellow, but become carmine-red in colour when
the style and anthers have withered. Delpino was the first to attempt an explanation
of this colour-change: he thinks that it enables insect visitors to recognize flowers
which are over, thus being spared fruitless efforts to get nectar. Hermann Miller
justly objects to this explanation on the ground that, if it were correct, flowers
exhibiting colour-change would have no advantage over those which wither or
fall off immediately anthesis is over. The persistence and brighter colouring of
flowers that are over are better explained as serving primarily to enhance the con-
spicuousness of the inflorescences as a whole, so that insects are attracted in greater
numbers, though the fact that flowers which have completed anthesis are readily
recognized as such gives maximum value to the adaptation.

Visitors.—Herm. Miiller saw Anthophora pilipes /., skg., easily inserting its
proboscis (20 mm. long) into the base of the flower. Delpino records the same bee
for Italy, and J have noticed it in the Kiel Botanic Garden. Warnstorf also observed
bees.

Alfken gives the following for Bremen.—

Hymenoptera. Apzdae: 1. Andrena apicata Sm.9; 2. A. nigroaenea &. 6;
3. A. varians A’.8; 4. Nomada borealis Ze//.6; 5. Osmia rufa Z.9. B. Diptera.
Muscidae: 6. Cynomyia mortuorum ZL.

Gerstacker noticed the bee Osmia aurulenta Pz. 6, freq., at Berlin.

988. R. sanguineum Pursh. (Herm. Miiller, ‘Fertilisation, p. 251, ‘ Weit.
Beob.,’ I, p. 300.)-—The flowers of this species belong to class Hb, Hermann Miller
describes the free part of the receptacle as a tube 3 mm. long, prolonged to more
than 5 mm. by the erect petals. Otherwise, the flower mechanism agrees with that
of R. rubrum. Warnstorf (Schr. natw. Ver., Wernigerode, xi, 1896) says that there
is slight protogyny. The style with its yellow very sticky stigma projects about
1mm. beyond the stamens. The pollen-grains are white, rounded three- to five-
sided, or shaped like an obelisk, smooth, with distinctly projecting germinating
papillae, 37 » in diameter on an average. The petals are at first quite white, but
after fertilization they become rose-red in colour, so that, as in R. aureum, the
inflorescence is rendered more conspicuous.

Visitors.—Herm. Miiller observed 3 bees, skg.—1. Apis mellifica 1. $; 2.
3ombus pratorum Z. ¥; 3. Osmia rufa Z. 9.

Plateau noticed Apis and Osmia bicornis Z. in Belgium. Burkill saw Bombus
terrester Z. on the coast of Yorkshire (‘Fertlsn. of Spring Fls.’).

989. R. petraeum Wulf. (Ricca, Atti Soc. ital. sc. nat., Milano, xiv, 1871;
Herm. Miiller, ‘ Fertilisation,’ p. 251, ‘Alpenblumen,’ pp. 111-12.)—The flowers of
this species belong to class C. Ricca describes them as slightly protogynous,

Ee 2

420 ANGIOSPERMAE—DICOTYLEDONES

Hermann Miiller as homogamous. As in the other species, cross-fertilization is
favoured when insects visit the flowers, but automatic self-pollination readily takes
place should they fail to do so.

Visirors.—Herm. Miiller noticed 2 hover-flies.

ggo. R. niveum DC. (Loew, ‘Bliitenbiol. Beitrage, I, pp. 11-14.)—
Loew observed that the
bees which visit the pro-
tandrous flowers of this
species insert their: pro-
boscis just above the petals
into the slit between the
bases of the filaments,
thus pressing the anthers
against the under-side of
their body. When visiting
a flower in the second

FIG. 138. Rihes pefraenm, Wulf. (after Herm. Miller). 4. Flower (female) stage, pollen is
from above (= 7. &. The same in longitudinal section. @, anther; ‘ly d ited
a, nectary; ov, ovary; #, petal; s, sepal; sv, stigma. necessarily eposite on

the receptive stigmas.
Visitors.—Loew observed a bee (Anthophora pilipes 7.) and a humble-bee
(Bombus agrorum /.) in the Berlin Botanic Garden.

gg. R. Grossularia L. (Herm. Miiller, ‘ Fertilisation,’ 230-1, ‘ Weit. Beob.,’
I, p. 300; Kirchner, ‘ Flora v. Stuttgart,’ p. 409; Knuth, ‘ Bloemenbiol. Bijdragen.’)—
The flowers of this species belong to class C. The anthers dehisce as they open,

Fic. 139. Rébes Grossularia, L. (from nature). A, Flower in the first (male) stage: the anthers have
dehisced, the stigma is stillimmature. 4. Flower in the second (hermaphrodite) stage; the stigma is now
mature. C. Flower after fertilization: the sepals have curved inwards. a, anther; ca, calyx; co, corolla;
ov, Ovary; s, stigma.

but the styles are not yet of full length, nor are the stigmas receptive (Hermann
Miller). The reflexed sepals of these protandrous flowers are green, usually with
reddish tips, while the petals, which are directed vertically downwards, are white.
Nectar is secreted in the base of the bell-shaped extension of the receptacle, and
the access to the secretion is narrowed by contraction of the receptacular margin,

SAXIFRAGEAE 421

and blocked by stiff hairs projecting perpendicularly from the style. The anthers
at first project somewhat beyond the stigmas, but ultimately stand at the same level,
so that automatic self-pollination must take place in horizontal or oblique flowers,
though cross-pollination is favoured during insect-visits. After fertilization, the sepals
curve upwards.

Besides hermaphrodite flowers, Kirchner noticed shrubs bearing female ones.
The stamens of these are so short that their anthers (which do not dehisce) are
at the level of the petals, or may even be situated more deeply in the receptacle.

Visitors. — Herm. Miiller (H.M.), Buddeberg (Budd.), and myself (Kn.)
observed the following.—

A. Diptera. (a) Afuscidae: 1. Calliphora erythrocephala A/g., skg. (H. M.):
2. Sarcophaga carnaria Z., do. (Kn.); 3. Scatophaga stercoraria Z., do. (H.M.).
(2) Syrphidae: 4. Eristalis aeneus Z., skg. and po-dvg. (H.M.); 5. E. tenax Z., do.
(H. M., Kn.); 6. Syrphus ribesii Z., po-dvg. (Kn.). B. Hymenoptera. Afzdae:
4. Andrena albicans JZ#//. § and 9, skg. and po-dvg. (H.M., Budd.); 8. A. fasciata
Wesm. 6, skg. (Budd.); 9. A. fulva Schr. 9, skg. and po-dvg. (H. M., Budd.); ro. A.
gwynana XK. é and 9 (H.M.); 11. A. nigroaenea A. 3, skg. (Budd.); 12. A. nitida
Fourc. $, do. (H.M.); 13. A. parvula A’ 9, po-cltg. (Budd., Kn.); 14. A. smithella
K. 9, skg. (Budd.); 15. Apis mellifica Z. §, do. (H. M., Kn.); 16. Bombus pratorum
£.9(H.M.); 17. B. scrimshiranus X. 9 (H.M.); 18. B. terrester 7. 9 and ¥, skg.
(H.M., Kn.); 19. Halictus cylindricus A. 9, po-cltg. (Budd.); 20. H. rubicundus
Chr. 9, skg. (H. M.).

Wiistnei saw the bee Halictus flavipes 7”. 9 in the island of Alsen.
* Alfken and Héppner (H.) record the following for Bremen.—

A. Diptera. Syrphidac: 1. Eristalis pertinax Scop.; 2. Helophilus pendulus Z.;
3. Syrphus ribesii Z. B. Hymenoptera. (a) Apidae: 4. Andrena albicans AZiU/. 8,
skg.; 5. A. propinqua Schenck 9, do.; 6. A. trimmerana A’. 6, do.; 7. A. varians A.
gandé, do.; 8. Apis mellificaZ.,do.; 9. Bombus agrorum £/. 9, do.; 10. B. derhamellus
K.9(H.); 11. B. jonellus A’. 9, skg.; 12. B. lapidarius Z. 9, po-cltg. (H.); 13. B.
lucorum L. 9, skg.; 14. B. pratorum ZL. 9, do.; 15. B. sylvarum Z. 9 (H.); 16. B.
terrester Z. 9, skg.; 17. Halictus calceatus Scop. 9 (H.); 18. Nomada alternata K. 9
and $; 19. N. bifida Zhs.qg and; 20. N. ruficornis Z. 9 and; 21. N. succincta
Pz.9 and é, skg.; 22. N. xanthosticta A. 9 and $; 23. Osmia rufa Z.é and 9.
(2) Tenthredinidae: 24. Pteronus ribesii Scop. (c) Vespidac: 25. Odynerus callosus
Ths. 9.

Friese observed the following bees in Mecklenburg and Baden.—

1. Andrena carbonaria Z., not infrequent; 2. A. fulva Schr. (Baden), one
individual; 3. A. nigroaenea K., freq.; 4. A. tibialis A’, do.; 5. A. varians A’, with
the vars. helvola /. and mixta Schenck, do.

Krieger saw the following bees at Leipzig.—

1. Andrena fulva Schr.; 2. A. nitida Mourcr.; 3. A. tibialis .A.; 4. A. trimmerana
A.; 5. A. varians A.; 6. Bombus hypnorum /. 9; 7. B. lapidarius Z.9; 8. B. pra-
torum 1.9; 9. B. terrester Z.9; 10. Nomada alternata A. = N. marshamella £.;
11. N. lineola Pz.; 12. Osmia rufa Z.

Schmiedeknecht gives the following bees for Thuringia.—

1. Andrena albicans Adi//.; 2. A. fulva Schr.; 3. A. propinqua Schenck; 4. A.
trimmerana KX. 9 and $; 5. A. varians A. 9 and 6; 6. Bombus hypnorum Z. 9;
7. B. pratorum LZ. 9; 8. B. terrester 7.9; 9. Nomada fabriciana /.; 10, N. ochro-
stoma A”, var. hillana A.; 11. N. ruficornis Z., var. flava Ps.

422 ANGIOSPERMAE—DICOTYLEDONES

Schenck records the following Hymenoptera for Nassau.—

(2) Apidae: 1. Andrena albicans Afil/.; 2. A. cineraria /.; 3. A. combinata
Chr.; 4. A. convexiuscula A.; 5. A. flavipes Pz.; 6. A. fulva Schr.; 7. A. gwynana
K.; 8. A. nitida Fourcr.; 9. A. parvula K.; 10. A. propinqua Schenck; 11. A.
tibialis K.; 12. A. trimmerana K.; 13. A. varians A, with the var. helvola Z.; 14.
Bombus hypnorum £. 9; 15. B. pratorum Z. 9; 16. B. terrester Z.9; 17. Halictus
albipes F.; 18. H. calceatus Scop.; 19. Nomada alternata X.; 20. N. ruficornis Z. ;
N, flava Pz. 6; 21.N. succincta Pz. (6) Vespidae: 22. Vespa germanica /. 9, very
common; 23. V. vulgaris Z. 9, freq.

Plateau noticed bees (Andrena sp., Apis, Bombus terrester /., and Osmia
bicornis Z.) and a Muscid (Calliphora vomiteria Z.) in Belgium.

Hoffer mentions Bombus terrester Z. 9 for Steiermark.

Von Dalla Torre and Schletterer give 2 bees for the Tyrol.—1. Andrena cineraria
Z., rare; 2. A. tibialis A.

Burkill saw a wasp (Vespa sylvestris Scop., skg.) and a Muscid (Scatophaga
stercoraria Z., skg.) on the coast of Yorkshire.

E. D. Marquard observed the bee Andrena fulva Scr. in Cornwall.

XXXVI. ORDER CRASSULACEAE DC.

Literature.— Knuth, ‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 73.

In this order the petals serve to attract insects, but some flowers are so small
(Tillaea, Bulliarda) that they are only very occasionally visited. In the large-flowered
species of Sempervivum and Sedum, on the other hand, self-pollination is often
entirely excluded by marked protandry, and insects are absolutely necessary for
fertilization. Protogyny is rarer. Nectar is usually secreted by glands at the base
of the ovary. In our native species it is only slightly concealed, and most of the
flowers must be referred to class EC. In some foreign species, however, it is very
deeply situated.

278. Tillaea Mich.

992. T. muscosa L.— This species bears insignificant reddish or white flowers,

solitary in the leaf-axils. They are apparently almost entirely dependent upon
automatic self-pollination.

279. Bulliarda DC.

993. B. aquatica (= Tillaea aquatica /., and T. prostrata Schtvhr).—According
to Ascherson, the minute almost sessile white flowers of this species possess four
nectaries, situated between the stamens and the ovary. In my herbarium specimens
the pollen-covered anthers almost lie upon the stigmas.

280, Rhodiola L.

Flowers dioecious, or sometimes trioecious; with half-concealed nectar. The
hermaphrodite flowers are protandrous.

994. R. rosea L. (=Sedum Rhodiola Z.). (Ricca, Atti Soc. ital. sc. nat.,
Milano, xiv, 1871; Schulz, ‘ Beitrage,’ II, p. 188; Warming, Vet. Ak. Overs.,
Kjébenhavn, 1886-7.)—The yellowish-red flowers of this species are dioecious in

"

CRASSULACEAE 423

the Riesengebirge, according to Schulz, and in the male and female flowers there
are vestiges of the other sex-organs. Axell also found only dioecious plants, as
did Lindman on the Dovrefjeld. Ricca, in the Alps, observed protandrous herma-
phrodite flowers, and so did Warming in Greenland, but the latter noticed also
cases of trioecism.

Ekstam says that in Nova Zemlia the flowers smell like honey, and secrete
abundant nectar.

Visitors.—Small flies were noticed by Ekstam in Nova Zemlia. Ricca observed
flies and ants in the Alps. In Dumfriesshire, an Empid and a Muscid were recorded
(Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 68).

281. Sedum L.

Flowers protogynous, or homogamous to markedly protandrous; with half-
concealed nectar, secreted basally between petals and stamens.

995. S. acre L. (Herm. Miller, ‘Fertilisation, pp. 251-3; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 289; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
PP. 74, 154, ‘ Weit. Beob.
ii. Bl. u. Insekt. a. d. nordfr.
Ins.,’ p. 234.)—The flowers
of this species are bright
yellow in colour, and the
five outer stamens super-
posed to the sepals are
the first to mature. They
direct their filaments (about
5 mm. long) obliquely up-
wards. When they have
shed their pollen they bend
back towards the petals,
while the five inner anthers

dehisce, and assume the

position of the first five. Fic. 140. Sedum acre, L. after Herm. Miller). (1) Flower in the
The small terminal stigmas first stage, seen obliquely from above. a! and a’, outer and inner

stamens; z, nectary; ov, ovary; f, petal; s, sepal. (2) Tip of a style

of the five carpels do not in the first stage. (3) The same in the second stage, after all the anthers
: % have dehisced.

mature till the inner anthers

have shed all their pollen. I observed this marked protandry, which completely

prevents self-pollination, in the island of Féhr.

Elsewhere the flowers are not so distinctly protandrous. Hermann Miller states
that in Westphalia the stigmas mature before the five inner stamens have finished
shedding their pollen, so that automatic self-pollination is possible if insect-
visits fail.

Visrtors.—I observed the following in Schleswig-Holstein, all skg.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Calliphora erythrocephala
Ae.; 3. Lucilia sp.; 4. Nemotelus uliginosus L.9; 5. Spilogaster carbonella Ze?s.
(6) Syrphidae: 6. Eristalis tenax Z.; 7. Melithreptus taeniatus JZ¢.; 8. Syritta pipiens

424 ANGIOSPERMAE—DICOTYLEDONES

/..; 9. Syrphus balteatus Deg. 6. B. Hymenoptera. ro. Andrena nigriceps &. 9;
11. Bombus rajellus A.C, Lepidoptera. 12. Epinephele janiraZ. Also 2 Muscids
(1. Lucilia caesar Z., and 2. Scatophaga stercoraria Z.) and a Syrphid (Syrphus sp.),
all skg., in Helgoland.

Herm. Miiller noticed the following in Westphalia.—

A. Diptera. (a) JAJuscrdac: 1. Pyrellia aenea Zeél., skg. (0) Syrphidae: 2.
Fristalis tenax Z., po-dvg. B. Hymenoptera. Afrdac: 3. Andrena cingulata
£9, skg.; 4. A. parvula A. 9 and 4, freq., skg.; 5. Bombus rajellus X. ¥, skg.;
6. Cilissa tricincta A. 9, do.; 7. Megachile centuncularis Z. 9, po-cltg.; 8. M. cir-
cumcincta K. 9, skg.; 9. Nomada ferruginata K. 9, do.; 10. Prosopis armillata Vy7. 4,
do.; 11. P. brevicornis WVy/. 8, do.; 12. P. variegata F. $, do.; 13. Sphecodes gibbus
L. 9, freq., skg.

The following were recorded by the authorities, and at the places stated.—

Alfken (Bremen), 3 bees (1. Andrena parvula A. 9; 2. Halictus punctulatus
K.9; 3. Prosopis hyalinata Sy. 9), and a Syrphid (4. Melithreptus menthastri Z.).
Verhoeff (Norderney),—A. Diptera. (a) Aluscidac: 1. Cynomyia mortuorum JZ. 4,
skg.; 2. Lucilia caesar Z. 6, do.; 3. Miltogramma sp., do.; 4. Sarcophaga striata F.,
do. (4) Syrphidae: 5. Eristalis arbustorum ZL. 9, skg.; 6. E. sepulcralis Z. one 9, do.
B. Hymenoptera. Sphegidae: 7. Oxybelus uniglumis Z. Schenck (Nassau) the
bee Anthidium oblongatum L/r. Réssler (Wiesbaden) the Tineid Glyphipteryx
equitella Scop. Frey (Simplon) the butterfly Lycaena orion Pa//. Herm. Miiller
(Alps) a humble-bee, 2 flies, and 3 Lepidoptera. MacLeod (Pyrenees) a Muscid.
Scott-Elliot (Scotland) a humble-bee.

Schletterer saw the following Hymenoptera at Pola—

(2) Apidae: 1. Andrena limbata Zv.; 2. Crocisa major Afor.; 3. Halictus
variipes Mor.; 4. H. virescens Zep.; 5. Osmia fulviventris Pz.; 6. O. versicolor Z/r.;

4. Prosopis clypearis Schenck. (0) Ichneumonidae: 8. Anilasta rapax (Gr.) Ths.
(c) Sphegrdae: 9. Trypoxylon figulus Z.

996. S. reflexum L. (=S. rupestre Z.). (Herm. Miiller, ‘ Fertilisation,’ p. 253,
‘Weit. Beob.,’ p. 295.)— Hermann Miiller states that the citron-yellow flowers
of this species are imperfectly protandrous, like those of S. acre.

Visrrors.—Herm. Miiller
(H. M.) and Buddeberg
(Budd.) observed the fol-
lowing.—

A. Diptera. (a) Jhisei-
dae: 1, Anthomyia sp., po-
dvg.(Budd.). (4) Syrphidae:
2. Eristalis tenax Z.(H.M.);
3. Syrphus arcuatus Fa//.,
skg. (Budd.). B. Hymeno-
B ptera. <Aprdac: 4. Anthi-

: dium oblongatum //r. 6,

skg. (Budd.); 5. A. punc-
Fic. 141. Sedus alpestre, Vill. alter Herm. Maller). 4. Flower tatum J//r. 9 and 3, in great

in the first (female) stage (x 7). &. Three carpels of the same flower, : e 3
with the attached nectaries; seen from without. a, anther; #, nectary ; number Si skg. (Budd.) 3 6.
ov, ovary; ~, petal; s, sepal; s/, stigma. Halictus morio £. 9 skg.

(Budd.); 7. H. sexnotatus
K.9, do. (Budd.); 8. Megachile maritima X. $(H.M.). C. Lepidoptera. Rhopa-
focera: g. Vpinephele janira 7. 4, skg. (Budd.); 1o. Vanessa urticae /., do. (Budd.).

CRASSULACEAE 425

Friese (Thuringia) noticed parasitic bees (1. Coelioxys elongata /ep., and 2.
Stelis signata //.) and po-cltg. bees (3. Anthidium lituratum Pz., and 4. A. punc-
tatum //,.).

997- S. boloniense Loisel. (=S. sexangulare Z.). (Schulz, ‘ Beitrage,’ 1,
p- 39-)— Schulz states that in the yellow flowers of this species the stigmas are
perfectly mature when the outer stamens are dehiscing. Automatic self-pollination
is quite possible, for the stamens bend towards the stigmas; but would appear to
be seldom resorted to, since numerous insects are attracted by the yellow colour of
the blossoms, as well as by the abundant nectar, which is secreted as in S. acre.

998. S. alpestre Vill. (=S. repens ScAlerch.). (Herm. Miiller, ‘ Alpenblumen,’
pp. 82-3.)—This species is native to the high Alps. Its flowers are protogynous,
but the stigmas continue functional till the anthers dehisce, so that automatic self-
pollination is possible should insect-visits fail. (Cf. Fig. 141.)

Visitors.—Herm. Miiller observed 2 Hymenoptera, a Dipterid, and 2 Lepidoptera.

g99. S. albescens Haw. (=S. rupestre 7..)—

Visitors.— MacLeod observed 2 bees (Bombus and Andrena) and a butterfly
(Lycaena) in the Pyrenees.

tooo. S. annuum I. (Schulz, ‘ Beitrage,’ ». 77.)—The flowers of this species
are mostly pale-yellow in colour. Schulz states that when they open the stigmas are

Fic. 142. Sedui atratum, L. (after Herm. Miller). .4. Flower in the first (female) stage. #. Three
carpels of the same, seen from without. C. Flower towards the end of the second (male) stage. J. Flower
after fading. (+ 7.) @, anther; 7, lament; #2, nectary; ov, ovary; , petal; s, sepal; s/, stigma.

receptive, remaining so throughout anthesis. ‘Vhe anthers of the outer stamens
dehisce as soon as the flower opens, and then those of the inner ones. The stigmas
and anthers being close together and at the same level, automatic self-pollination
is at first possible, or even inevitable ; but the inner stamens curve outwards towards
the end of anthesis, thus favouring cross-pollination. Lindman too found the
flowers to be protogynous at first, afterwards becoming homogamous and capable
of automatic self-pollination. Kerner says that the outer stamens serve for self-
and the inner ones for cross-pollination. He also states that this annual species
hybernates by means of rosette-shaped offshoots, should winter set in early and
prevent the ripening of fruits.
Visitors.—Schulz observed occasional flies and Ichneumonids at Bozen.

1oo1. S. atratum L. (Ricca, Atti Soc. ital. sc. nat., Milano, xili, 1870, p. 250;
Herm. Miiller, ‘ Alpenblumen,’ pp. 79-80; Kerner, ‘ Nat. Hist. Pl’ Eng. Ed. 1, JI.

426 ANGIOSPERMAE—DICOTYLEDONES

Pp. 175, 344-)—This species is native to the high Alps. Hermann Miiller describes
the flowers as protogynous, but with persistent stigmas, so that automatic self-
pollination takes place normaliy and tolerably soon. Ricca says the flowers are
protogynous with short-lived stigmas. Kerner adds that anthesis lasts four days,
that the outer stamens serve for cross- and the inner ones for self-pollination, and
that the scale-like nectaries are cleft at the tip. (Cf. Fig. 142.)

Visirors.—Herm. Miiller only observed a Chrysidid and a Pyralid.

1002. S. Telephium L. (Herm. Miiller, ‘ Fertilisation,’ pp. 253-4.)—The two
species (S. maximum Sw/er and S. purpureum Zzzé) into which S. Telephium is now
divided possess the same flower mechanism, except that the inner stamens of S,
purpureum are inserted into the petals one-sixth of the way up. Hermann Miiller
states that the anthers of the five outer stamens dehisce first, and then those of
the five inner ones; and it is only when the latter have withered that the stigmatic
papillae develop. The stamens lie close to the widely radiating petals, so that
self-pollination is excluded even if some pollen remains clinging to the anthers
until the stigmas are mature.

Fic. 133. Sedusz Telephinzn, 1. (after Herm. Maller’. (1) Flower seen from above, (2) The same
after removal of the carpels, to show the five nectarics.

The nectaries are situated as in S. acre, but their form is somewhat different :
in S. Telephium, they are on the tips of longish scales at the bases of the petals
and below the ovaries. Nectar-sucking or pollen-collecting insects that creep about
on the crowded inflorescences touch the anthers and stigmas of numerous flowers
in succession, and since these are protandrous effect crossing: they may also,
however, occasionally effect sclf-pollination in old flowers with mature stigmas,
should some pollen still cling to the anthers.

Visitors.—Herm. Miiller observed the following.—

A. Diptera. J/uscrdac: 1. Echinomyia magnicornis Ze//., skg. B. Hymeno-
ptera. (a) Apidae: 2. Bombus agrorum /”. 4, skg.; 3. B. lapidarius /.. ¥, po-cltg. ;
4. B. sylvarum Z. 9 and ¥, in great numbers, skg.; 5. Halictus zonulus Sm. 9, skg. :
6, Psithyrus campestris Ps. 8, do. (6) Zenthredintdae: 7. Allantus arcuatus Fors/.
(Borgstette).

MacLeod saw Bombus terrester Z. ¥, po-clig. and skg., in the Pyrenees (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 419).

Alfken noticed Bombus agrorum /. é at Bremen.

CRASSULACEAE 427

Loew observed the following visitors in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syritta pipiens Z. B. Hymenoptera. A/rda:
2. Apis mellifica Z. ¥, skg.; 3. Bombus sylvarum Z. 9, do.; 4. B. terrester Z. 8, do.

1003. S. dasyphyllum L. (Schulz, ‘Beitrage,’ II, pp. 77-8.)—The flowers
of this species are white with a reddish tinge. Schulz describes them as protandrous,
to a degree inversely proportional to the height of the habitat above sea-level:
the plants of low-lying districts are most markedly so. In such places (e. g. in the
Etschthal) the styles with their immature stigmas lie close together at the time
when the flower opens: the stigmas only become receptive when the anthers have
shed all their pollen, often indeed, only when they have dropped off; it follows
that self-pollination is almost entirely excluded. In higher regions (e.g. the Ortler
district), the stigmas ripen somewhat sooner. so that automatic self-poliination is
usually tolerably easy when they spread out.

Kerner states that the stigmas are receptive when the flower opens; and that
the outer stamens serve for cross-, the inner ones for self-pollination. It appears,
therefore, that homogamy also obtains.

Fic. 144. Sedum alonm, L. (after Herm. Miller). 4. Just opened flower. 4. Flower in the second
half of the first (male) stage. C. Flower in the second ifemale) stage. 2. Centre of the flower, after
removal of the carpels. (* 7.) a, anther; 77, filament; 2, nectary; , petal; s, sepal; s/, stigma.

The nectaries are small, heart-shaped, stalked scales, yellow or orange-red
in colour, and placed opposite the ovaries.

Visitors. —Schulz observed numerous short-tongued insects (flies and the
smaller Hymenoptera), but does not record their species. MacLeod saw a bee
in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 418).

1004. S. altissimum Poir. (=S. nicaeénse A//.)\—The flowers of this species
are yellow.
Visirors.—MacLeod saw a bee (Halictus morio) in the Pyrenees.

1005. S. album L. (Herm. Miiller, ‘ Weit. Beob., I, p. 296, ‘ Alpenblumen,’
pp. 80-1; Schulz, ‘ Beitrage,’ I, p. 77; Loew, ‘ Bliitenbiol. Floristik,’ p. 397.)—The
flowers of this species are distinctly protandrous, and Hermann Miller says that
self-pollination is scarcely possible. Schulz confirms this statement for the Tyrol.
(Cf. Fig. 144.)

Visrrors.— Herm. Miiller (Alps) saw 3 bectles. 7 flies, 2 bees, and 3 Lepidoptera.

428 -ANGIOSPERMAE—DICOTYLEDONES

Herm. Miiller (H. M.) in the Fichtelgebirge and Buddeberg (Budd.) in Nassau
also observed the following.—

A. Coleoptera. (2) Byrrhidae: 1. Byrrhus pilula Z., skg. (H.M.). (8)
Cerambycidac: 2. Leptura maculicornis Deg., freq., skg. (H.M.). B. Diptera.
(a) Bombyliidae: 3. Bombylius canescens Afzkan, skg. (Budd.). (6) Muscidae: 4.
Echinomyia fera Z., skg. (H.M.); 5. E. grossa Z., do. (H.M.). C. Hymenoptera.
Apidae: 6. Chelostoma campanularum A’. 9, skg. (Budd.); 7. Halictus albipes F.%,
do. (H.M.); 8. H. flavipes /. 9, do. (H.M.): 9. H. interruptus Pz. 9, do. (Budd.) ;
10. Prosopis armillata Vy/. 9, do. (Budd.); 11. P. signata Pz. $, do. (H.M.); 12.
Psithyrus quadricolor /ep. 6, do. (H. M.).

Schenck saw the bee Anthidium lituratum /2z., and quotes Friese as recording
the parasitic bee Stelis signata Z/r. for Nassau.

Loew observed the following in Switzerland.—

A. Coleoptera. (a) Cerambycidac: 1. Stenopterus rufus Z.; 2. Strangalia
armata Hés/.; 3. S. melanura Z. (2) Cleridae: 4. Trichodes apiarius Z. (c) Ocde-
meridac: 5. Oedemera coerulea Z.; 6. O. flavescens /.; 7. O. flavipes / B.
Hymenoptera. Apzdac: 8. Prosopis alpina Afor. C. Lepidoptera. (a) Sesidae:
g. Sesia formicaeformis L'sp. (0) Zygaenidav : 10. Syntomis phegea Z.; 11. Zygaena
filipendulae Z.

F. F. Koh] noticed 3 wasps (1. Vespa crabro Z., 2. Eumenes pomaformis /”.,
and 3. Odynerus bidentatus Ze.) in the Tyrol.

MacLeod saw 2 Lepidoptera, 7 beetles, and 13 flies in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent. iii, 1891, p. 419).

1006. S. Aizoon L.—

Visirors.—Loew noticed the humble-bee Bombus lapidarius Z. ¥., skg., in the
Berlin Botanic Garden.

1007. S. spectabile Ber.—

Visirors.—Loew observed a Syrphid (Eristalis tenax 7.) and a bee (Halictus
minutissimus A’. 9) in the Berlin Botanic Garden.

1008. S. anglicum Huds.—The flowers of this species are of a whitish-rose
colour.

Visirors.—MacLeod saw a bee, a beetle, a Syrphid, and a Muscid in the
Pyrenees.

In Dumfriesshire 2 Strauomyids and a Muscid were recorded (Scott-Elliot,
‘Flora of Dumtfriesshire,’ p. 69).

282. Sempervivum Rupp.

Flowers protandrous, red or yellowish-white; with concealed nectar secreted at
the bases of the carpels.

1009. S. Wulfeni Hoppe. (Herm. Miller, ‘Alpenblumen, pp. 83-4; Schulz,
‘Beitrage,’ 11, pp. 79-80.)—Hermann Miiller describes the flowers of this species
as being so markedly protandrous that automatic self-pollination appears to be
excluded. According to A. Schulz, who also studied the plants in the Ortler district,
the protandry is not so marked, autogamy sometimes taking place, though perhaps
not often.

CRASSULACEAE 429

The abundant nectar is secreted by a disk beneath the pistil, and is protected
from rain by hairs.

Visitors.—Herm. Miiller states that the nectar is specially sought out by bees
(8 species), more rarely by Ichneumonids (1), beetles (r), and hover-flies (2). Schulz
mentions flies, bees, and Lepidoptera as frequent; beetles as less so.

toro. S. Funckii F. Braun. (Herm. Miller, ‘Alpenblumen,’ pp. 84-6.)—The
flowers of this species are also protandrous, but individual stigmas often become
Teceptive when only the first stamens have withered, so that self-pollination is
commoner than in S. Wulfeni. (Cf. Fig. 145.)

Visrrors.—Herm. Miiller observed 2 beetles, 6 bees, 9 Lepidoptera, and 3 flies.

FIG. 145. Senepervivuie Funcki?, Braun (after Herm. Miller). <f. Flower in the second (female)
stage (x 24). &. Part of the reproductive organs in the first (predominatingly male) stage (~ 7). C. Part
of the same in the second (purely female) stage. 0. A few carpels with mature stigmas. a@-a, anthers
/, Glament; 7, nectary; ov, ovary; p, petal; s, sepal; sé, stigma.

ou. S. montanum L. (Herm. Miiller, ‘ Alpenblumen,’ p. 86; Kerner, ‘ Nat.
Hist. Pl.,’ Eng. Ed. 1, Il, pp. 344-5.)—The flower mechanism of this species agrees
with that of S. Funckii. Kerner states that the inner anthers do not dehisce till
the stigmas have withered, so that they serve for cross-pollination, while the outer
ones are concerned with autogamy.

430 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Herm. Miiller saw a bee and 4 Lepidoptera in the Heuthal. Loew
noticed a Noctuid (Agrotis ocellina S.V.) in the same locality. Alfken observed the
bee Anthidium manicatum Z. 9 and 4, very freq., skg., at Bremen.

1o1z, S. tectorum L. (Herm. Miiller, ‘Alpenblumen, pp. 86-7; Schulz,
‘Beitrage,’ 11, p. 79.)—Hermann Miillér says that the flower mechanism of this
species agrees with that of S. Funckii. According to Schulz, the stigmas do not
usually become receptive till the anthers have shed all their pollen, so that self-
pollination is excluded.

Visitors.—Herm. Miiller saw 3 beetles, 6 bees, and 7 Lepidoptera. Schulz
observed humble-bees and other Hymenoptera, more rarely Lepidoptera and flies.

1013. S. arachnoideum J.. (Herm. Miiller, ‘Alpenblumen,’ p. 87.)—In the
flowers of this species, according to the observations of Hermann Miiller (made in
the canton of Graubiinden), the styles are often apposed till the anthers have all
discharged their pollen. Schulz (‘Beitrage’) states that the styles (in the Tyrol)
are often inclined towards one another when the flower opens, but subsequently
spread out gradually till they occupy an almost horizontal position. The stigmas
are usually receptive at the time when the last of the inner stamens shed their
pollen, so that automatic self-pollination may be effected.

VisiTors.—Herm. Miller observed 8 flies, 7 bees, and 11 Lepidoptera. Schulz
also observed Lepidoptera (3), bees, and flies.

1014. S. ruthenicum Koch.—Kerner states that in this species the inner
stamens serve for cross-pollination, while the outer ones bend towards the stigmas
and bring about autogamy.

283. Bryophyllum Salisb.

1015. B. calycinum Salisb. (Delpino, ‘ Altri appar. dicog. recent. oss.,’ p. 56.)—
Delpino states that the elongated tubular and pendulous flowers of this species
secrete abundant nectar from four glands. They are protandrous.

Visitors.—Delpino supposes these to be humming-birds, despite the incon-
spicuous greenish or brownish colour of the flowers.

284. Cotyledon Tourn.

1016. C. umbilicus L. (Willis, ‘Contributions to the Nat. Hist. of the Flower,’
(2).}—The corolla-tube of this species is about 10 mm. deep and 3 mm. wide,
and the nectar is secreted in its base by the five carpels. The ten anthers dehisce
when the flower opens. They are at the same level as the stigmas, though these
are at first not fully mature. It follows that insects probing for nectar in the first
stage of anthesis cover themselves with pollen, which they transfer to the stigmas
of flowers in the second stage. Towards the end of anthesis automatic self-pollination
is inevitable, owing to the proximity of stigmas and anthers.

Visrrors.—-Willis only observed Thrips.

DROSERACEAE 431

XXXVI. ORDER DROSLRACLAE DC.
285. Drosera L.

Most of the flowers, according to my observations in Schleswig-Holstein, are
cleistogamous (cf. Vol. I, p. 54). It is only in very favourable continuously sunny
weather that the small white blossoms open, and anthesis is confined to a single
morning. Hansgirg describes the species of this genus as pseudo-cleistogamous.

Kerner states that nectar is secreted by the yellow claws of the petals in the
chasmogamous flowers. After the brief anthesis the flowers close again, and the six
stigmatic lobes of the style curve in such a way that their receptive papillae touch
the pollen-covered anthers. This self-pollination is effective.

1017. D. rotundifolia L. (Kirchner, ‘Flora v. Stuttgart,’ p. 322; Knuth, ‘BI.
u. Insekt. a. d. nordfr. Ins.,’ p. 34.)—The white flowers of this species open but
rarely, and are only 3 mm. in diameter, so that there can scarcely be any question of
insect-visits. The five anthers dehisce when the flower opens, and are at the
same level as the simultaneously matured stigmas, from which they are only
4mm. distant, so that automatic self-pollination can easily take place, especially
when the flower closes towards the end of anthesis. Either cross- or self-pollination
may be effected by insect-visits to the open flowers. In those which remain closed
the stamens and carpels touch one another when mature. Such cleistogamous
flowers produce abundant seed.

Visitors. —In Dumfriesshire, several Muscids were observed (Scott-Elliot, ‘ Flora
of Dumfriesshire,’ p. 73).

1018. D. intermedia Hayne. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 34.) —
The flower mechanism of this species agrees with that of D. rotundifolia. Cleisto-
gamous flowers are even more frequent.

101g. D. anglica Huds. (Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—
The flowers of this species are mostly pseudo-cleistogamous, many of them opening
about mid-day, but soon closing again. The stigmatic branches and the anthers
mature simultaneously before anthesis; the latter are very small, and borne-upon
thick filaments. The pollen-grains are golden-yellow in colour, densely covered
with spinous tubercles, and aggregated into 3-4 celled pollinia.

286. Aldrovanda Mont.
The flowers are mostly cleistogamous, according to Bentham and Hooker.
1o20, A. vesiculosa L. (Korzchinsky, ‘Zur Kenntnis der Aldrovanda vesicu-
losa.’)—Korzchinsky states that in the flowers o! this species each anther contains at
most thirty-five pollen-grains. The anthers get bound to the stigma by pollen-tubes,
but most of the ovules remain unfertilized, even though they swell up with the ovary.

NXXVITJ. ORDER BRUNIACEAE R. Br.

The flowers are aggregated into small capitula, and their mechanism resembles
that of the Compositae. According to Delpino (‘ Ult. oss.,’ p. 98) and Hildebrand
(Bot. Ztg., Leipzig, xxviii, 1870, p. 636), the end of the style bears a glabrous bilobed
collecting-cup, which grows through the anther-tube and thus takes up the pollen.
The stigmatic surface appears upon it later.

432 ANGIOSPERMAE—DICOTYLEDONES

XXXIAX. ORDER AHALURAGLEAL R. Br.

(including Hrepuriacear Link, Gunneracear Endl., and CarritricHacer Link.)

287. Myriophyllum L.

Monoecious wind-pollinated water-plants. The anthers are versatile, and contain
abundant easily dispersed pollen. The stigmas are large, and very rough. Some
of the species are possibly water-pollinated.

1021. M. verticillatum L. (Ludwig, Kosmos, Stuttgart, x, 1881-2, pp. 7-12.)—
The small, greenish-yellow flowers of this species are arranged in spiked whorls
in the leaf-axils. They project above water, and are wind-pollinated. Ludwig states
that there are also submerged water-pollinated flowers. These, however, do not
appear to be universally distributed; at any rate I did not observe them in the
island of Fohr.

1022. M. spicatum L. (Ludwig, op. cit.\—Ludwig says that the reddish
flowers of this species possess the same mechanism as those of M. verticillatum,
but submerged flowers have not been observed. According to Kerner, the female
flowers develop before the male ones.

Only sterile plants of the species had been observed in Greenland, until
Vanhoffen discovered richly flowering specimens in a pond at Ikerusak (Abromeit,
‘Bot. Ergeb. von Drygalski’s Grénlandsexped..” pp. 11-12). As in Europe, the
stigmas of the female flowers were mature before the opening of the male ones.

1023. M. alterniflorum DC.—Here again only wind-pollinated flowers pro-
jecting above the water-level have so far been observed.

288. Hippuris L.

Water-plants with inconspicuous protogynous wind-pollinated flowers in the
leaf-axils.
1024. H. vulgaris L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 171-2.)—
In this species the part of the stem projecting
from the water bears on each of its nodes (fifty
to sixty, or even more in number) a whorl of
ten leaves. There is a small flower in the axil
of each of these. At first the white, very papillose
stigma projects 3 mm. beyond the ovary, while
the sessile anther is still closed. When the
; Ee ; stigma has withered a thin filament (13 mm.
bic. 146. Aippurtsiulgaris, L.(from : e :
nature, several times enlarged). (1) Flower long) develops, and at its tip the dehisced

in the first (female) stage: a, sessilt, anther exposes its pollen to the wind. I am
immature anther; s, papillose stigma. a

(2) Flower in the second (male) stage: unable to confirm Vaucher’s assertion (‘ Hist.
a, stalked dehisced anther; s, witheret

physiolog. des pl. d'Europe,’ II, p. 362), that the
oily yellow pollen is directly applied to the stigma.

In Fohr, besides hermaphrodite stocks, I observed others that were purely
female, or on which only some flowers possessed stamens. Jirchner also observed
gynodioecism at Stuttgart (‘Flora v. Stuttgart,” p. 419), as did Willis in England
(Proc. Phil. Soc., Cambridge, 1893).

stigma.

HALORAGEAE 433

Vanhéffen states that the species flowers and fruits normally in Greenland
between 60° and 70° N. lat. (Abromeit, ‘ Bot. Ergeb. von Drygalski’s Gronlandsexped.,’
p. 12).

289. Gunnera L.

1025. G. manicata Linden. (Jonas, ‘Uber d. Inflores. u. Bl. v. Gunnera
manicata,’ Inaug.-Dissert., Breslau, 1892.}—Jonas describes this species as gyno-
monoecious. Nectaries are absent. The hermaphrodite flowers are anemophilous,
and they alone set ripe fruits.

290. Callitriche L.

This genus includes inconspicuous monoecious species; anemophilous and
protogynous so far as my own observations go, but generally described as entomo-
philous or hydrophilous in some instances.

1026. C. vernalis Kuetz. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 72.)—
This species is anemophilous and obviously protogynous in the North Frisian
Islands, for pollen remains clinging to the anthers of the male flowers when the
female ones in the same plant have already set fruits. I have not observed the early
stages of anthesis.

1027. C, stagnalis Scop. (Axell,‘Om Anord. for Fanerog. Vaxt. Befrukt.,’
p. 36; Knuth, op. cit., p. 72; Hegelmaier, ‘Monographie d. Gatt. Callitriche,’
Stuttgart, 1864; Ludwig, Kosmos, Stuttgart, x, 1881-2, p. 32; Warnstorf, Verh. bot.
Ver., Berlin, xxxviii, 1896.)—In the North Frisian Islands this species also is
anemophilous and protogynous. Hegelmaier describes the flowers as anemophilous,
but also says they are protandrous, the male flowers maturing before the female
ones of the same inflorescence. The former are mostly situated in the axils of
the upper leaves, and the latter in those of the lower ones. The pollen-grains are
covered with a thick, somewhat tuberculated membrane.

Ludwig says that in this species and the other Eucallitricheae (C. vernalis and
C. hamulata) the fibrous anther-layer characteristic of aerial flowers is present, a
layer which plays an important part in dehiscence. He describes the stamens as
1 mm. long, with stiff filaments, and producing pollen-grains about 25 m» long
and 21 » broad. The two styles are almost entirely covered with papillae.

Besides aerial flowers Ludwig also observed submerged ones, fertilized under
the surface of the water. Hegelmaier found such flowers to be infertile, and Kerner
says that their anthers do not usually dehisce, but decay with the pollen they contain.
The flowers of terrestrial stocks growing in mud or damp woods behave like those
of aquatic ones.

Warnstorf describes the flower mechanism of the variety (a) vera Aschs. some-
what as follows.—Owing to the extreme shortening of the terminal nodes of the
shoots, the opposite leaves are crowded into a floating rosette. The diclinous
(monoecious) flowers are situated in their axils. As a rule the female flowers
first make their appearance. The two long styles are beset with stigmatic papillae
for two-thirds of their length, and project for about 3 mm. beyond the leaves of
the rosette. After pollination and elongation of the nodes the flowers become sub-
merged and ripen their fruits. At a later stage the male flowers appear in their

DAVIS. II F f

434 ANGIOSPERMAE—DICOTYLEDONES

axils, each with a single stamen 4-5 mm. long, the yellow anther of which is
borne upon a stiff filament and projects from the rosette to this distance, afterwards
dehiscing. The pale-yellow tuberculated pollen-grains are very irregular in shape
(prismatic, tetrahedral, or bluntly pyramidal), very easily scattered, and transferred
by the wind to adjoining younger leaf-rosettes bearing female flowers. In rarer
cases I noticed a male flower in one leaf-axil and a female flower in that immediately
opposite. Self-pollination was then easily possible, for pollen could fall directly
on the two branches of the style. Besides the anemophilous flowers now described
I also found, on lateral shoots of varying length, female flowers devoid of perianth
leaves, relatively small ovaries and very long styles. These were apparently
pollinated under water, though I did not observe submerged male flowers. The
species may therefore be described as anemo-hydrophilous, and autogamous under
certain conditions.

1028. C. hamulata Kuetz. (Ludwig, op. cit.; Hegelmaier, op. cit.)}—This
species also bears submerged flowers, which are sterile according to Hegelmaier.

102g. C. autumnalis L. (Ludwig, op. cit; Hegelmaier, op. cit., p. 61;
Jénsson, Univ. Arsskr., Lund, xx, 1883-4.)—This species belongs to the sub-genus
Pseudocallitriche, in which the pollen-grains possess no external coat (extine), an
indication of hydrophily. Jénsson describes them as oil-containing and lighter than

water, so that they can drift to the stigmas of the female flowers and effect
pollination.

XL. ORDER AZELASTOMACEAE R.Br.

291. Heeria Schlecht.

Hermann Miiller states (Nature, London, xxiv, 1881, pp. 307-8) that in the
species of this genus there are two kinds of stamen differing in function: the one
kind serving to attract insects, while the other produces pollen. (Cf Vol. I, p. 108.)

292. Centradenia G. Don (1030. C. floribunda Planch.), 293. Rhexia

L. (1031. R. glandulosa Bertol.), and 294. Monochaetum Naud.
(1032. M. ensiferum Naud.)—

Darwin describes the above three species as self-sterile.

295. Pleroma D. Don (=Tibouchina Awé/.).

1033. P. Sellowiana Cogn. (Ludwig, Biol. Centralbl., Erlangen, vi, 1886-7.) —
Ludwig says that the flowers of this species are at first pure white, afterwards assum-
ing a purple-red colour. (Cf Vol. I, p. 86.)

Darwin describes species of Pleroma as self-sterile.

XLI. ORDER LYTHRARIEAE JUSS.
(including GranaTgEaz Don.)

This order includes a large number of trimorphous and dimorphous species.
Among the former are Lythrum Salicaria Z., L. Graefferi Zenore, and species of
the genera Nesaea and Lagerstroemia. Among dimorphous forms are Lythrum

LYTHRARIEAE 435

Thymifolia Z. (homomorphous according to Koehne), and twenty other species of
the same genus, as well as some belonging to the genera Pemphis, Rotala, and
Nesaea. (Cf Vol. I, p. 491.) Numerous species are homomorphous, among which
Koehne enumerates no fewer than 340 belonging to Lythrum, including our native
species L. Hyssopifolia Z.

Koehne gives the following particulars in his newest monograph of the order
(‘Lythraceae,’ in Engler u. Prantl, d. nat. Pflanzenfam., III, 7, pp. 4-5).—Eight
species exhibit trimorphous heterostyly (Lythrum maculatum Ajersk., L. flexuosum
Lag., L. virgatum Z., L. Salicaria Z., Nesaea Kilimandscharica Koehne(?), N.
Schinzii Koehne, N. sagittifolia (Sond.) Koehne, Decodon verticillatum (L.) Ell.) and
possess two whorls of stamens. Dimorphous heterostyly obtains in five species
possessing two staminal whorls (Lythrum rotundifolium Hochs¢., Pemphis acidula
Forst., Nesaea rigidula (Sond.) Koehne, N. mucronata Koehne, N. Volkensii Koehne).

Fic. 147. Lythrum Salicaria, L. (after Herm. Miiller). (1) Long-styled flower, after removal of one-
third of the calyx, corolla, and stamens; seen from above. (Pollen green.) (2) Medium-styled flower,
ditto. (Pollen yellow.) (3) Short-styled flower, ditto. (Pollen yellow.) (4) Medium-styled flower, seen
obliquely from the right front. a, long style or stamens; 4, medium style or stamens; c, short style or
stamens; d, nectar.

The following fourteen species possessing but one whorl of stamens are dimor-
phously heterostylous (Rotala myriophylloides Welw., R. floribunda (night) Koehne,
R. nummularia Welw., Lythrum lineare Z., L. album #. 2. and K., L. ovalifolium
(Engelm.) Koehne, L. acinifolium (DC.) Koehne, L. gracile Benth., L. lanceolatum L.,
L. californicum S. Wais., L. alatum Pursh, L. Vulneraria Azz, Nesaea lythroides
Welw., N. Woodii Koehne). About 6% of the 450 species included in the order
are heterostylous. The most specialized forms as regards pollination are the strongly
zygomorphous flowers of species of Cuphea and Pleurophora.

Some species bear cleistogamous or pseudo-cleistogamous flowers. Ammannia
latifolia Z., for instance, is often cleistogamous, according to Koehne. Treviranus
states that the following are self-pollinated before their flowers open :—Cuphea
silenoides WVees, C. floribunda Zelm., and C. Melvilla Lznmd/. (Bot. Ztg., Leipzig, xxi,
1863).

296. Lythrum L.

Flowers red, trimorphous, dimorphous, or homomorphous; with concealed
nectar, secreted at the bottom of the tubular receptacle.
Ff2

436 ANGIOSPERMAE—DICOTYLEDONES

1034. L. Salicaria L. (Charles Darwin, ‘On the Sexual Relations of the three
Forms of Lythrum Salicaria,’ J. Linn. Soc. Bot., London, viii, 1865, pp. 169-96,
‘Different Forms of Flowers’; Herm. Miiller, ‘Fertilisation,’ pp. 255-60.)—Charles
Darwin investigated the flower mechanism of this species in an extremely thorough
manner, and proved by numerous experiments that the ‘reproductive organs, when of
different length, behave to one another like different species of the same genus
in regard both to direct productiveness and the character of the offspring; and that
consequently mutual barrenness, which was once thought conclusive proof of
difference of species is worthless as such, and the last barrier that was raised
between species and varieties is broken down’ (Herm. Miiller, op. cit., p. 258).

Before considering Darwin’s experiments in detail, a description of the flower
mechanism will be given.

Fic. 148. Scheme of the legitimate unions possible in Ly¢srum (atter Charles Darwin). 4. Long-
styled form. B. Medium-styled form. C. Short-styled form. The dotted lines indicate the anthers

from which pollen must pass to the stigma of each of the three forms, so as to give a legitimate union
with complete fertility.

Hermann Miiller says that the nectar is secreted by the fleshy base of the
receptacle, surrounding the short stalk of the ovary, and filling the space between
this and the receptacular wall. The red inner surface of the calyx, and the dark
veins of the petals (which converge to the middle of the flower), serve as nectar-
guides. The flowers are horizontal, usually hexamerous (more rarely pentamerous),
and not strictly actinomorphous. The petals are inserted into the edge of the
cylindrical receptacle (5-7 mm. long), and the three lower petals are usually some-
what longer than the upper ones, which are 6-10 mm. in length. When the flower

LYTHRARIEAE 437

has completely opened, the lower petals are directed somewhat obliquely forward,
while the upper petals spread out vertically. The stamens and style are on the
lower side of the flower, so that a nectar-seeking insect cannot penetrate between
them, but has to insert its proboscis above them into the base of the receptacle.
The ends of these organs bend upwards in such a way that the insect is obliged
to touch the stigma and anthers.

Owing to the varying relative lengths of the stamens and style, nectar-seeking
insects normally effect the crossing of different stocks. In each flower the two
whorls of stamens and the style are at three different levels: the shortest of these
are concealed in the receptacle, those of medium length project 3-4 mm., and the
longest 7-8 mm. There are therefore the three following forms of flower.—

1. Long-styled flowers: the style is longer than the stamens; half of the latter
are of medium length, the others short.

2. Medium-styled flowers: the style is of medium length; half of the stamens
are longer than the style, the others shorter.

3. Short-styled flowers: the style is short; half of the stamens are long, the
others of medium length.

The anthers of the long stamens are green in colour, possibly as a protection
against pollen-devouring insects; those of the medium and short stamens are yellow.
The long stamens produce the largest pollen-grains, the medium stamens grains
of medium size, and the short stamens the smallest grains. In correspondence with
this the stigmatic papillae of the long styles are distinctly longer than those of
the medium and short styles.

The experiments of Darwin, to which reference has been made, show that of
the eighteen possible modes of fertilization (f Vol. I, p. 47) only those six lead
to complete fertility in which each kind of stigma receives pollen from anthers
situated at a corresponding level (‘legitimate union,’ % Fig. 148).

Insects of a size adapted to the flower mechanism (medium-sized bees, and
certain hover-flies) regularly bring about legitimate fertilization when they probe
for the nectar secreted in the fleshy base of the receptacle. Holding firmly to the
long and medium organs, they thrust their proboscis into the receptacle, so that
after visiting the various kinds of flower, three different parts of their proboscis
and body are dusted with pollen, which is rubbed off on stigmas occupying the
three corresponding levels.

Visirors.—The most noteworthy is a bee, Melitta (Cilissa) melanura Wy. 9
and 6, which Herm. Miller (op. cit., p. 259) observed ‘wherever Lythrum Salicaria
grows, both sucking honey and gathering pollen, and almost confining itself to
this one plant.’ Curiously enough, however, I myself have never succeeded in seeing
this bee on the flowers, though I have often watched them in Schleswig-Holstein,
Mecklenburg, and the island of Riigen, under very favourable conditions (in calm
weather, and during sunshine). Herm. Miiller (loc. cit.) adds, ‘Since its proboscis
is only 3-4 mm. long, it must thrust a great part of its head, which is 2-3 mm,
broad, into the tube: it then touches the shortest reproductive organs with the under-
surface of its head, the next with the ventral surface of its thorax, and the longest
with the ventral surface of its abdomen; so that its dimensions suit the flower,

438 ANGIOSPERMAE—DICOTYLEDONES

just as well as the flower must suit it, for the insect to confine its visits so exclusively
to it. I have only found one exception to this exclusiveness, for I once found
a male of Melitta (Cilissa) melanura sucking honey on Thrincia hirta DC,’

Friese saw the rare bee Melitta (Cilissa) haemorrhoidalis F., var. nigra Priese,
in Switzerland (canton Valais).

Herm. Miiller (H. M.), Buddeberg (Budd.), and myself (Kn.) have observed,
in addition, the following.—

(Insects regularly effecting all three kinds of legitimate union are indicated by
an asterisk (*); those which regularly effect only one or two kinds of legitimate
union are not specially designated ; the names of those which only effect pollination
accidentally, bringing about either legitimate or illegitimate union, are enclosed in
brackets [ ].)—

A. Coleoptera. (a) Curculionidae: 1. [Nanophyes lythri #. (H.M.)]. (2)
Nitidulidae: 2, [Meligethes (H.M., Kn.)]. B. Diptera. Syrphidae: 3. Eristalis
intricarius Z., po-dvg. (H. M.); 4. E. sp. (Kn.);_ 5. *Helophilus pendulus Z., skg.
(H.M., Kn.); 6. *H. trivittatus #, do. (H.M.); 7. [Melithreptus taeniatus JZ.
po-dvg. (H. M.)]; 8. Rhingia rostrata Z., skg. and po-dvg. (H. M., Kn.); 9. Syritta
pipiens Z., do. (H. M., Kn.); 10. Syrphus balteatus Deg., do. (H. M., Budd., Kn.) ;
11. S. ribesii Z. (Kn.); 12. Volucella bombylans Z., skg. and po-dvg. (Kn.); 13. *V.
bombylans Z., var. plumata J@g., skg. (H.M.). C. Hemiptera. 14. [Capsus
(H.M.)]. D. Hymenoptera. Apzdae: 15. *Apis mellifica Z. ¥, skg. (H. M., Kn.);
16. *Bombus agrorum F. ¥ and 9, not infrequent, skg. (H. M., Kn.); 17. *B. der-
hamellus A. ¥ (Fohr, Kn.); 18. *B. lapidarius Z. 9, ¥ and 4, skg. (H.M., Kn.);_ 19.
*B. sylvarum LZ. ¥, do. (H.M.); 20. *B. terrester Z. 9 and ¥, do. (H.M., Kn.);
21. Chelostoma nigricorne (Vy/. 9, do. (Budd.); 22. *Cilissa melanura Ny/. 9 and 4,
skg. and po-cltg. (H.M., Budd.); 23. [Halictus cylindricus /. 9, po-cltg. (H. M.)];
24. [H. leucopus K. 9, skg. (Budd.)]; 25. [H. leucozonius Schr. 9 and 4, do.
(Budd.)]; 26. [H. minutissimus X’. 9, do. (H. M.)]; 27. [H. morio F. 9, do. (Budd.)];
28. Megachile centuncularis Z. 6, do. (H. M.); 29. *M. fasciata Sm. 6, do. (H.M.) ;
30. *Osmia adunca Panz. 4, do. (Budd.); 31. *Saropoda rotundata Pz. é and 9, not
infrequent, skg. (H.M.). E. Lepidoptera. (a) Geometridae: 32. Timandra amata
L., skg. (Budd.). (6) Rhopalocera: 33. Pieris rapae Z., freq., skg. (H.M., Kn.);
34. Rhodocera rhamni Z., do. (H.M.). F. Thysanoptera. 35. [Thrips (H. M.)].

The following were observed by the authorities, and for the localities named.—

Loew (Silesia) a bee (Bombus agrorum F. 9, skg.), a Microlepidopterid
(undetermined sp., skg.), and a butterfly (Pieris brassicae Z., skg.) (‘ Beitrage,’
p. 32). Alfken (Bremen) a Curculionid (Nanophyes lythri #., in vast numbers),
and 3 bees (Bombus derhamellus A. ¥; 3B. distinguendus Jor. 9, skg.; and
Halictus calceatus Scop. 4). Wiistnei (Schleswig) the bee Macropis labiata Pz.
Friese (Baden (B.), Alsace (A.), Fiume (F.), Mecklenburg (M.), Switzerland (S.),
and Hungary (H.)) the bees—1. Epeoloides caecutiens /. (F. and M., very rare S.);
2. Eucera basalis Mor. (F., according to Korlevié); 3. E. salicariae Zep., not un-
common (A., S., H., Tyrol); 4. Melitta (Cilissa) melanura /Vy/. (B., F., M., and H.,
occasional, A. rare). Réssler (Wiesbaden) the Lepidoptera Earias chlorana Z.. and
Orthosia lota C7. Von Fricken (Westphalia and E. Prussia) the Curculionid Nano-
phyes lythri #. Schletterer (Tyrol) the bees—1. Bombus variabilis Schmzedekn. ;
2. Halictus maculatus Sm.; Melitta (Cilissa) melanura /Vy/. (also seen there by
von Dalla Torre). Redtenbacher (Vienna) the Curculionid Nanophyes lythni /.
Ducke (Aquileja (A.) and Austrian Silesia (S.)) the bees—1. Eucera (Macrocera)
dentata Klug. $ (A.); 2. E. (M.) salicariae Zep. 9 and 8 (A.); Melitta (Cilissa)
melanura /Vy/. (S.). H. de Vries (Netherlands) a humble-bee, Bombus terrester
ZL. ¥ (Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel, 1875). MacLeod (Flanders)

LYTHRARIEAE 439

6 humble-bees, 5 hover-flies, and 5 Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, vi,

1894, p. 393). Thomson (Sweden) the rare bee Melitta (Cilissa) melanura JVy/.

Scott-Elliot (Dumfriesshire) Apis, 2 humble-bees, a hover-fly, and a Lepidopterid

(‘Flora of Dumfriesshire,’ p. 68). Loew (Berlin Botanic Garden) 2 hover-flies, skg.

(Melanostoma mellina Z., and Syrphus pyrastri Z.), and a bee (Apis mellifica Z. ¥,

skg.); and, on the var. angustifolia, 2 bees, skg. (Apis, and Bombus agrorum J. 6).
Heinsius noticed the following, all skg., in Holland.—

A. Hymenoptera. (a) Apidae: 1.*Apis mellifica Z. ¥; 2.*Bombus agrorum
F.& and ¥; 3. B. cognatus S/eph.; 4. *B. terrester Z. 8; 5. *Melitta (Cilissa)
melanura /Vy/.9; 6. Heriades nigricornis My. 6; 7. Melecta luctuosa Scop.; 8.
*Psithyrus campestris Pz.; 9. *P. vestalis Fourcr.é. B. Lepidoptera. (2) Rhopa-
locera: 10. Lycaena icarus Qoff.d; 11. Pieris napi 2.6; 12. P. rapae 2.8; 13.
Polyommatus dorilis H/n. 6; 14. Papilio machaon Z.; 15. Rhodocera rhamni Z.
and 9. (6) Moctu:dae: 16. Euclidia glyphica Z. C. Diptera. (a) Musczdae: 17. Prosena
siberita #. 9. (0) Syrphidae: 18. Helophilus pendulus Z. 9; 19. Rhingia campestris
Mg.%; 20. Syritta pipiens Z.—Of these visitors only the humble-bees regularly effect
all the legitimate unions, while the others leave some of the anthers and stigmas
untouched, especially those of the long stamens and styles.

1035. L. Hyssopifolia L. (Schulz, ‘ Beitrage,’ I, p. 38.)—The small lilac
flowers of this species are arranged in terminal spikes, and Schulz describes them
as being slightly protogynous. Since the anthers are at the same level as the
stigmas, and very near them, automatic self-pollination regularly takes place, should
crossing not be effected by insects, as it sometimes is,

Visirors.—I observed the following in the Kiel Botanic Garden (‘ Bloemenbiol.
Bijdragen ’).—

A. Hymenoptera. Apzdac: all skg.; 1. Apis mellifica Z. %; 2. Bombus

lapidarius Z. ¥; 3. B. terrester Z. ¥. B. Diptera. Syrphidae: 4. Eristalis tenax Z.,
skg. and po-dvg. C. Lepidoptera. Rhopalocera: 5. Pieris rapae L., skg.

297. Peplis L.
Very small, inconspicuous, rose-coloured flowers, with exposed nectar.

1036. P. Portula L. (Henslow, Trans. Linn. Soc. (Bot.), London, 2. Ser.,
1875, p. 363; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 303-4; Koehne,
Bot. Jahrb., Leipzig, vi, 1885, p. 39; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
p- 73-}—In the flowers of this species, according to Koehne, the petals are only
6 mm. long, fugacious, and often absent altogether. The five or six stamens do
not project beyond the calyx. The stigma is almost sessile, so that autogamy
by the fall of pollen regularly takes place in the tiny blossoms. Submerged flowers
remain closed, and, since they contain air, are pseudo-cleistogamously fertilized as
the result of automatic self-pollination.

MacLeod describes a small scantily secreting nectary at the base of the ovary.
During anthesis the flower is wide open, and its six stamens are curved slightly
inwards; but the most anterior and the most posterior stamen, owing to a lateral
compression of the flower, do not spread out so much as the other four. Hence
it follows that the anthers of these two stamens almost always touch the pistil,
so that automatic self-pollination is inevitable. When the flower closes, all six
anthers are pressed against the stigma.

440 ANGIOSPERMAE—DICOTYLEDONES

According to Willis and Burkill (‘Fls. and Insects in Gt. Britain,’ I, p. 266),
the small inconspicuous sessile flowers are 3 mm. in diameter. The stigma matures
a little before the anthers, so that for a time crossing may result from the visits
of insects. Automatic self-pollination, however, regularly takes place, for the stamens
bend inwards and dust the stigma. All the flowers produce seeds.

VisiTors.—None were observed by Willis and Burkill.

298. Cuphea P. Br. ‘

1037. C. purpurea Hort. (=C. procumbens Oréeg.).—Gaertner describes this
species as self-sterile.

1038. C. micropetala H. B. et K.(=C. eminens Planch. ef Linden). (Kerner,
‘Nat. Hist. Pl.,’ Eng. Ed. 1, Il, pp. 235, 346.)—Kerner states that the protandrous
flowers of this species have their openings laterally directed. The anthers dehisce
on the side which faces upwards away from the stigma; the pollen is therefore
likely to be carried away by nectar-sucking insects, and to effect cross-pollination.
Some days afterwards, the style, which has increased its length by 11 mm., brings
the stigma into the line of access to the nectar, so that insect visitors dusted with
pollen must bring about crossing. Failing insect-visits, autogamy takes place by
the longest stamen curving up to the stigma, to which it applies its pollen-covered
surface.

The obliquely placed ovary bars the way to the nectar secreted in the sacculation
at the base of the corolla-tube, leaving only two narrow passages through which
pollinating insects have to thrust their proboscis. Creeping insects (ants) are pre-
vented from entering the corolla-tube by adhesive bristles on the margin of the
calyx, such forms being useless, or even injurious to the flowers.

299. Punica L.

Odourless and nectarless pollen flowers, bright red in colour; homogamous
or protandrous.

1039. P. Granatum L. (Schulz, ‘ Beitrige,’ pp. 72-3.)—In this species both
calyx and corolla contribute to the conspicuousness of the flower. The former is
coral-red, very thick-walled, 26-30 mm. long and 20-25 mm. broad. The brilliantly
red, delicate, very fugacious petals are 20-30 mm. long and 1o-z0 mm. broad.
The numerous stamens, the filaments of which are orange-red in colour, are bent
inwards, and therefore block the entrance to the flower. The style is very short,
and may be either receptive during the dehiscence of the anthers or become so
after their pollen is shed. Autogamy by means of pollen that remains in the flower,
is possible in both cases.

VisiTors.—Schulz observed (South Tyrol) beetles belonging to the genera
Cetonia and Trichodes, dvg. the flowers, and often effecting cross- as well as self-
pollination.

ONAGRARIEAE 441

XLI. ORDER ONAGRARIEAE JUSS.
300. Epilobium L.

Flowers red, more rarely white, frequently arranged in large conspicuous
racemose inflorescences; protandrous, homogamous, or protogynous; with con-
cealed nectar secreted by the upper surface of the ovary. The pollen-grains are
usually bound together by threads of viscin.

1040. E. angustifolium L. (=E. spicatum Zam., and Chamaenerion angusti-
folium Scop.). (Sprengel, ‘Entd. Geh.’; Herm. Miiller, ‘Fertilisation,’ pp. 261-2,
‘Weit. Beob.,’ II, p. 237; Lindman, ‘Bidrag till Kanned. om Skandin. Fjellvaxt.
Blomn. o. Befrukt.’; Warming, ‘Bestévningsmaade af nogle grénlandsk. Blomster,’
Pp. 32-3; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 291-2; Kerner,
Nat. Hist. PL,’ Eng. Ed. 1, 11; Schulz, ‘ Beitrage,’ II, p. 73; Knuth, ‘ Bloemenbiol.
Bijdragen’; Loew, ‘ Bliitenbiol. Floristik,’ p. 394.)—Kerner says that the purple-red,
rarely white flowers of this species, open between 6 and 7 a.m., and are protandrous
to an extent that precludes self-pollination, as Sprengel long ago observed. The nectar
secreted by the fleshy green top of the ovary is protected from rain, for the expanded
bases of the filaments converge to form a hollow cone, which grasps the base
of the style. Where the style leaves the cone it is covered with hairs which prevent
rain-drops from entering, while the proboscis of an insect can easily be pushed
through to the nectar.

In younger flowers, the stamens (which are covered with pollen-grains bound
together by threads of viscin) serve as the only possible alighting-place for insects,
for the style is still short, and the stigmas apposed. In older flowers the dehisced
stamens curve downwards, while the now elongated style with its four diverging
recurved stigmas forms the only alighting-place, so that insects coming from younger
flowers must pollinate older ones.

The flower mechanism does not always conform to the above description.
Warming states that the var. Jecostyla is slightly protogynous along the Isortokfjord,
and therefore capable of self-pollination. According to Schulz, the flowers in the
lowlands are more distinctly protandrous than in the mountains (Tyrol), where
the stigmas are mature in some cases before the pollen is entirely shed. Kerner
says that the style, short at first, elongates after twenty-four hours, and the widely
diverging stigmas finally recurve so as to touch the pollen-covered anthers: hence
automatic self-pollination can be effected. Kerner found plants with normal flowers
in only a few places. In shady localities the flowers dry up and fall off; an attempt
is also made to overcome the disadvantages of unfavourable habitats of this kind
by the development of long runners.

VisiTors.—Sprengel observed various humble-bees. 1 myself noticed 5 nect-
skg. bees.—

1. Apis mellifica Z.¥; 2. Bombus agrorum F.9; 3. B. hortorum Z. 9 and ¥;
4. B. lapidarius Z. 9, ¥ and $; 5. B. terrester Z. 9 and ¥.

Herm. Miiller (H.M.) and Buddeberg (Budd.) give the following list.—

A. Coleoptera. Cerambycidae: 1. Strangalia melanura Z., nect-lkg. (H. M.).
B. Diptera. (a) Emprdae: 2. Empis livida Z., freq., skg. (H.M.); 3. E. rustica
Faill., do. (H.M.). (6) Strattomyidae: 4. Chrysomyia polita Z., skg. (Budd.).

442 ANGIOSPERMAE—DICOTYLEDONES

(c) Syrphidae: 5. Syrphus ribesii Z., po-dvg. (H. M.). C. Hymenoptera. (a)
Apidae: 6. Apis mellifica Z. ¥, exceedingly common, skg. and po-cltg. (H.M.);
7. Bombus agrorum F. 9, ¥ and 4, very common, skg. (H.M.); 8. B. confusus
Schenck 9, freq., skg. (H. M.); 9. B. lapidarius Z. g and ¥, do. (H.M.); 10. B.
pratorum Z. 9, § and 6, do. (H.M.); 11. B. terrester Z. 9, ¥ and @, do. (H. M);
12. Halictus flavipes / 9, skg. (Budd.); 13. H. malachurus &. 9, do. (Budd.);
14. H. nitidus Schenck 9, do. (Budd.); 15. Megachile versicolor Sm. 9, do. (H. M.,
Thuringia) ; 16. Nomada jacobaeae Pz. 9 (H.M.); 17. N. roberjeotiana Pz. 9, skg.
(H.M.); 18. Psithyrus campestris Pz. 4, skg. (H.M.); 19. Sphecodes gibbus ZL. 9,
do. (H.M.). (4) Sphegidae: 20. Ammophila sabulosa Z., skg. (H. M.); 21. Cerceris
labiata F’., do. (H. M.); 22. Crabro alatus Pz., do. (H.M.); 23.C. cribrarius Z. 4, do.
(H.M.). (c) Zenthredinidae: 24. Allantus scrophulariae Z., skg. (H.M.). D. Lepi-
doptera. 25. Ino statices Z., do. (H.M.); 26. Zyganea filipendulae Z. do. (H. M.).

Loew observed the following. —

(2) In the Berlin Botanic Garden, the bees—r. Apis mellifica Z. ¥, skg.; 2,
Bombus rajellus A. 8, do.; 3. Chelostoma nigricorne My/. 8, do. (4) In the Riesen-

gebirge, the humble-bee Bombus agrorum /. ¥, skg. (c) In Silesia, the butterfly
Hesperia comma ZL.

Fic. 149. Epzlobtum Fletschert, Hochst. (after Herm. Miller). A. Young protandrous flower seen
directly from the front (x 24). 8. Middle of the same ( - 7). C. A homogamous flower, after removal
of a sepal and two petals (= 24). D. Style of a protogynous flower, in which an anther is begin-
ning to dehisce (x 7). £. Style of a flower, on the anthers of which a small quantity of pollen still
remains (x 7). a, anther; /, filament; gv, style; sd, nectar-cover; s¢. stigma.

The following were observed by the authorities, and in the localities stated.—

Alfken (Bremen) the bees—1. Bombus agrorum /.; 2. B. arenicola Zés.;
3. B. derhamellus K.; 4. B. distinguendus Jfor.; 5. B. jonellus A.; 6. B. lapidarius
L.; 7. B. proteus Gers’. 9; 8. B. terrester Z.; 9. Halictus calceatus Scop. 9; 10.
Macropis labiata J. 9 andé; 11. Megachile centuncularis Z.; 12. Podalirius
furcatus Pz. 8. Verhoeff (Norderney) the Capsid Calocoris chenopodii Fa//. 9 and 8.
Krieger (Leipzig) the bees—1. Bombus agrorum /. 4; 2. Eriades nigricornis Wy. ;
3. Halictus smeathmanellus X.; 4. Prosopis confusa Vy/. Hoffer (Steiermark) the
bees—1. Bombus distinguendus JA/or.é; 2. B. hypnorum Z. ¥; 3. Psithyrus vestalis
fourcr. $. Redtenbacher (Vienna) the Chrysomelid Adoxus obscurus Z. Schmiede-
knecht the bees—1. Andrena fumipennis Schmiedekn.; 2. Bombus distinguendus

ONAGRARIEAE 443

Mor.3; 3. B. hypnorum Z.8; 4. B. jonellus K.5; 5. B. mastrucatus Gers. 6;
6. B. pratorum Z. $; 7. B. soroénsis /. d; 8. B. terrester Z. 8; 9. Psithyrus vestalis
Fourcr.%. Frey-Gessner (Switzerland) the humble-bees Bombus pratorum ZL. 9, ¥
and 6, and B. scrimshiranus X. (=B. jonellus X.) 6. Herm. Miiller (Alps) a beetle,
5 flies, 11 Hymenoptera, and a Lepidopterid (‘Alpenblumen,’ p. 209). Scott-Elliot
(Dumfriesshire) 2 humble-bees and a wasp, both freq. (‘Flora of Dumfriesshire,’

Pp. 64).

1041. E. Dodonaei (=E. rosmarinifolium aenke, and E. angustissimum
Weber). (Schulz, ‘ Beitrage,’ II, p. 73.)—Schulz says that the flowers of this species
usually display well-marked protandry, for the four stigmas only spread out when
the anthers have lost all their pollen. Sometimes, however, they begin to do so
before the pollen is entirely shed, and automatic self-pollination is then possible.

Visrrors.—Schulz observed nect-skg. and po-cltg. bees, skg. Lepidoptera, and
po-dvg. Diptera.

Herm. Miiller noticed 4 bees and 2 Lepidoptera in the Alps (‘ Alpenblumen,’
p. 211). Loew saw Apis, skg., in the Berlin Botanic Garden.

1042. E. Fleischeri Hochst. (=E. denticulatum Ulend.). (Herm. Miller,
‘Fertilisation,’ p. 262, ‘Alpenblumen,’ pp. 209-11.)—-The flowers of this species
agree in many respects as regard mechanism with those of E. angustifolium, but
may be protandrous, homogamous, or protogynous. In all three cases, however,
crossing is favoured by the fact that either the diverging stigmatic branches or the
stamens serve as the most convenient alighting-place. Automatic self-pollination
always takes place should insect-visits fail.

Visitors.—Herm. Miiller observed a Syr-
phid, 13 bees, 2 fossorial wasps, and 4 Lepi-
doptera. Loew noticed a Syrphid (Syrphus
balteatus Deg.) and a bee (Halictus minutis-
simus A’. 9) in the Berlin Botanic Garden.

1043. E. hirsutum L. (=E. grandiflorum
Weber). (Herm. Miiller, ‘Fertilisation,’ pp. 263-4;
Schulz, ‘Beitrage,’ I, pp. 35-6.)—The mechanism
of the large, dark-purple flowers of this species
varies in different localities, while Hermann
Miiller recognized only homogamous flowers FIG. 150. Epilobium hirsutum, L. (after
25-30 mm. in diameter. Schulz describes He™-. eee hee cus pie!
three forms borne on different stocks.—1. stamens; sé, stigma.

Large: these are zygomorphous and markedly

protandrous, with long styles curved in such a way that self-pollination is
prevented. 2. Medium: these are rarer (at Halle and in North Thuringia) than
the previous kind. They are less distinctly zygomorphous, and are usually slightly
protandrous, rarely almost homogamous. The style is straight; and should insect-
visits fail the stigmas curve back in such a way as to touch the anthers of the
longest stamens, so that self-pollination is possible. 3. Sma//: these are still smaller
than the medium ones, and homogamous. As the stigmas are at the same level
as the longest stamens, automatic self-pollination is inevitable.

444 ANGIOSPERMAE—DICOTYLEDONES

Besides the hermaphrodite stocks there are some female ones. Their flowers
possess stamens, but the anthers do not dehisce. Schulz observed gynomonoecism,
more rarely gynodioecism.

Kerner says that the flower mechanism does not differ essentially from that
of E. angustifolium.

Vistrors.—I observed the following at Gliicksburg (‘ Bloemenbiol. Bijdragen ’).—

A. Diptera. Syrphidae: 1. Eristalis tenax Z., po-dvg. B. Hymenoptera.
Apidae: 2. Apis mellifica Z. §, very freq., skg. and po-cltg.; 3. Bombus agrorum
F. 9, skg. and po-cltg.; 4. B. sylvarum Z. 9, do.; 5. B. terrester Z.9, do. C. Lepi-
doptera. 6. Pieris sp., freq., skg.

MacLeod saw Apis, a Syrphid, a Muscid, and a Lepidopterid in Flanders
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 294, 380).

In Dumfriesshire, a humble-bee and a short-tongued bee were recorded (Scott-
Elliot, ‘Flora of Dumfriesshire,’ p. 69).

1044. E. parviflorum Schreb. (Schulz, ‘ Beitrage,’ I, pp. 36-7; Herm. Miller,
‘Fertilisation,’ pp. 262-3.)—In this species again, according to Schulz, the length
of the stamens and style varies, as
does their order of development.
In most cases the anthers are at
the same level as the end of the
style, or even project beyond it,
so that, the flowers being homo-
gamous, automatic self-pollination
is inevitable. More rarely the
style projects beyond the anthers,
and sometimes the stigmas mature
a little before they dehisce, but
the flowers are often homogamous.
Here again self-pollination fre-
quently takes place. It may be
Fic.151. L£pslobfun parviforum, Schreb. (after Herm. effected, according to Kerner, as

Miiller). Flower seen from the side, after removal of the early as the first day of anthesis.
greater part of the ovary, two petals and most of a sepal. a!

and a*, anthers of long and short stamens; sé, stigma. Hermann Miiller states that

the flowers are homogamous, the

anthers of the four short stamens being at a lower level than the stigmas, and
serving for cross-pollination, while those of the four long ones are at the same level
and bring about autogamy. When insects visit the flowers, they usually first touch
the stigmas, which occupy the middle of the flower, and therefore usually effect
cross-pollination. (Cf Fig. 151.)

The rather small pale-red flowers are solitary, and do not receive many visits.

VisiTors.—I have only observed the honey-bee, skg. and po-cltg. Herm. Miiller
only noticed the beetle Meligethes, and a butterfly (Pieris rapae Z., freq., skg.).
MacLeod saw Pieris sp. in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p- 298).

1045. E. montanum L. (Schulz, ‘ Beitrage,’ I, p. 37.)—Schulz describes the
flowers of this species as homogamous. Since the anthers of the long stamens

ONAGRARIEAE 445

are usually on a level with the stigmas, autogamy is the rule, and Kerner says it
may take place as early as the first day of anthesis. It is excluded in flowers where
the long stamens do not reach the stigmas,

Visttors.—These are few. Herm. Miiller observed flies (Anthomyia 9, po-cltg.),
and a butterfly (Pieris napi Z., skg.) (‘ Weit. Beob.,’ II, p. 237).

Schletterer records the humble-bee Bombus pomorum for the Tyrol. MacLeod
saw a hover-fly and a beetle in Flanders (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
p- 298).

In Dumfriesshire, 2 Muscids and 2 Syrphids were recorded (Scott-Elliot, ‘ Flora
of Dumfriesshire, p. 65).:

1046. E. collinum C. C. Gmel.—In this species again, according to Kerner,
automatic self-pollination may take place on the first day of anthesis, the stamens
being long enough for the anthers to touch the stigmas.

Fic. 152. Epélobinm alsintfolium, Vill. (after Herm. Miller). A. Young flower, shortly after
opening: the upper parts of the sepals and petals have been removed. B. Older flower, after removal
of two sepals and petals (x 7). a! and a’, anthers of long and short stamens; gr, style; #, nectar:
ov, ovary; ~, petal; fo, pollen; s, sepal; sd, nectar-cover; sé, stigma.

Visttors.—Herm. Miiller saw 2 short-tongued bees in the Alps (‘ Alpenblumen,’
p- 213).

1047. E. roseum Retz.—Schulz says that automatic self-pollination is usually
inevitable in the homogamous flowers of this species, the anthers of the long stamens
reaching the level of the non-divergent stigmas and coming into contact with them.

Visitors.— MacLeod saw the butterfly Pieris napi Z. in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 296).

446 ANGIOSPERMAE—DICOTYLEDONES

In Dumfriesshire, 2 hover-flies and a Lepidopterid were recorded (Scott-Elliot,
‘Flora of Dumfriesshire,’ p. 66).

1048. E. alpinum L. (Axell, ‘Om Anord. for Fanerog. Vaxt. Befrukt.,’
pp. 18, 109.)—-Axell states that the flowers of this species are autocarpous as the
result of automatic self-pollination, should insect-visits fail. Warming describes
plants brought from Greenland and cultivated in Copenhagen as extremely fertile
after automatic self-pollination.

1049. E. alsinifolium Vill. (=E. origanifolium Zam.). (Herm. Miller, ‘ Fer-
tilisation,’ p. 263, ‘Alpenblumen,’ pp. 211-13; Lindman, ‘ Bidrag till Kanned. om
Skandin. Fjellvaxt. Blomn. o. Befrukt.’; Schulz, ‘Beitrage,’ I, p. 37.)—The flowers
of this species are always capable of automatic self-pollination in the Alps, but
being slightly protogynous may also be cross-pollinated if visited early by insects.
The flower mechanism is the same in the Scandinavian Highlands, but, according
to Lindman, the flowers are homogamous there. Schulz found them to be slightly
protogynous in the Riesengebirge, and also adapted for automatic self-pollination,
the anthers being applied closely to the stigmas. Owing to the tubular form of the
flower, the nectar is especially accessible to Lepidoptera, though a proboscis of
6-7 mm. long is required to reach it.

Visttors.—Herm. Miiller saw a butterfly (Argynnis) and a hover-fly (Syrphus).

1050. E. adnatum Griseb. (=E. tetragonum Z.). (MacLeod, Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 296-7.)—This species was studied during very hot
weather (2. 7.’94), and the three following stages made out in the development
of the flowers. —

1. In the still closed flower, the tips of the petals project about o-25 mm.
beyond the calyx. The four long (antisepalous) stamens reach half-way up the
style, and their anthers have already shed most of their pollen-grains. These are
loosely united into tetrads, and indeed some of them have already sent out pollen-
tubes into the pistil. The anthers of two of the four short (antipetalous) stamens
are beginning to dehisce.

2. The petals now project 2 mm. from the bud. The four short stamens have
elongated and discharged most of their pollen upon the lower part of the style.
Many pollen-grains have sent out long tubes into the pistil. There is a considerable
quantity of nectar in the base of the flower.

3. The flower is quite open. The four long stamens have elongated to such
an extent that their anthers project beyond the style, while the four short ones are
half its length. All the anthers are brown and empty, and the style is brownish
in colour; indeed it usually begins to assume this hue in the bud. The flower
closes towards the end of anthesis, and the anthers are pressed against the style.

When the weather is not so warm, the development of the reproductive organs
appears to be retarded. In any case automatic self-pollination is inevitable;
crossing, though not impossible, is very unlikely to take place.

Visttors.—None have so far been observed.

1051. E. roseum Schreb. (MacLeod, op. cit., vi, 1894, pp. 295-6.) When
anthesis begins in this species the eight anthers have already dehisced, those of

ONAGRARIEAE 447

the four long (antisepalous) stamens being (in some flowers) at the same level as,
but remote from, the stigmas, which by this time have become receptive. The
anthers of the four short stamens are 4-1 mm. below the stigmas, though not so
far away from them as the others. Automatic self-pollination is therefore im-
possible. Insects can effect cross- and self-pollination with equal facility.

In other flowers, the anthers of the four long stamens cling to the style after
discharging their pollen. The filaments elongate later on, and, as the anthers are
not set free from the style, are subjected to a stress which forces them to bend
inwards. They sometimes remain in this condition till the end of anthesis, so that
crossing by insects is almost impossible. In other cases they gradually free them-
selves from the style, on which the greater part of the pollen is left behind. When
the flower closes the anthers are pressed against the stigmas, so that autogamy
is inevitable.

VisiTors.—MacLeod saw the butterfly Pieris napi Z., skg.

1052. E. latifolium L. (Warming, ‘Bestévningsmaade af nogle grénlandsk.
Blomster,’ p. 143.)—This species is native to high northern latitudes. Warming
states that the large flowers vary from slightly protandrous to slightly protogynous.
The remarkably short style is bent down, bringing the stigmas below the anthers,
so that autogamy may result from the fall of pollen. Should this take place, however,
it does not result in autocarpy. There is a large amount of vegetative reproduction
by means of offsets.

Several sub-species have been found in Greenland, where blossoms of pure
white colour are rare. Vanhdffen states that as a rule the flowers are feebly
protandrous, but two stocks bearing strongly protandrous ones were observed
(Abromeit, ‘ Bot. Ergeb. von Drygalski’s Grénlandsexped.,’ p. 13).

There would appear to be insect visitors in Greenland, for a hybrid (=E.
ambiguum /7zes) has been found at Disko between this species and E. angusti-
folium.

1053. E. lactifolium Haussk. (Abromeit, op. cit., pp. 12-13.)—This arctic
species ranges south to Scandinavia and Lapland. It bears small whitish flowers,
and reproduces vegetatively by means of short offsets arranged in a rosette-like
manner. It has been observed in Greenland by Vanhdéffen and K. Rosenvinge.

gol. Lopezia Cav.

Flowers markedly protandrous, often with an explosive stamen.

According to Delpino (‘Ult. oss., II, pp. 124-6), there are two pseudo-
nectaries at the knee-like bend of the two upper petals. These shine like drops
of nectar, but are dry in reality (cf Parnassia, p. 413). The true secreting nectary
is at the base of the two stamens, of which one is sterile and specialized. (Vide
infra.)

1054. L. coronata Andr. (Hildebrand, Bot. Ztg., Leipzig, xxiv, 1866, p. 76.)—
One of the original two stamens in this species is modified into a stalked spoon-
shaped staminode, the two halves of which at first surround the anther of the normal
stamen, and project horizontally from the flower. As the stalk of the spoon is

448 ANGIOSPERMAE—DICOTYLEDONES

in a state of downward and the filament in a state of upward tension, an insect
alighting upon the staminode in order to reach the pseudo-nectaries (vzde supra),
will cause these opposed strains to become effective. The spoon will suddenly bend
downwards, while the stamen immediately above it will explode upwards, at the
same time applying its pollen to the under-side of the visitor's body. The stamen
afterwards bends out of the flower, while the style goes on growing, and projects
from the flower to serve as an alighting-place. Automatic self-pollination is excluded.

Visttors.—Herm. Miiller frequently observed the house-fly (Musca domestica
£.) and the common gnat (Culex pipiens Z.) effect pollination in his room, and the
honey-bee do so at an open window (‘ Fertilisation,’ p. 265).

1055. L, racemosa Cav.—According to Ogle’s account (Pop. Sci. Rev., London,
viii, 1869, p. 271), the flower mechanism of this species is the same as that of
L. coronata.

1056. L. miniata DC. (Hildebrand, op. cit., xxiv, 1866, pp. 478-9.)—In this
species the fertile stamen is not explosive, and lies above the spoon-shaped staminode.

302. Oenothera L.

Protandrous lepidopterid flowers. Honey is secreted by and concealed in the
base of the calyx-tube. Kerner states that the peduncle is curved so as to make
the entrance to the flower lateral. He also describes the pollen-grains as being
bound together by threads of viscin. Many of the species are night flowerers
(vide infra).

1057. O. biennis L. (Sprengel, ‘Entd. Geh.,’ pp. 217-23; Herm. Miiller,
‘Fertilisation, p. 264; Kerner, ‘ Nat. Hist. PL,’ Eng. Ed. 1, II; Knuth, ‘BI. u. Insekt.
a. d. nordfr. Ins.,’ p. 151, ‘Bloemenbiol. Bijdragen.’)—Sprengel long ago thoroughly
described the flower mechanism of this species, which is native to Virginia. The
large bright yellow flowers are devoid of nectar-guides. They open in the evening,
and are then most fragrant, and would therefore appear to be especially adapted
to crepuscular and nocturnal Lepidoptera. They are, however, owing to their
bright yellow petals, also conspicuous during the day, when they are visited by
long-tongued bees, and must consequently be placed in the transition class, LmH.
The secretion of the smooth, yellow nectary in the base of the calyx-tube is
protected by fine woolly hairs. It flows into the upper naked part of the tube,
and adheres to the style, which is here pressed against the lower wall of the
tube. The individual flowers remain in bloom for two nights. Kerner says that
they open shortly before 6 p.m., and close twenty-four hours later. When the
flower opens, the anthers shed their pollen, but at this stage the four stigmas are
still apposed. On the morning of the second day the stigmas begin to diverge,
and are fully developed the second night, while the stamens are now shrivelled.
Kerner says that the edge of the corolla does not serve as an alighting-place for
insect visitors, but is only attractive. When a visitor inserts its proboscis into
a flower, it touches the anthers with its head, and at the beginning of anthesis
the stigmas, owing to a lateral curvature of the style, are out of the line of access
to the nectar. But, according to Kerner’s account, the style straightens half an
hour later, and the stigmas spread out, so that crossing may be effected by insects

ONAGRARIEAE 449

already dusted with pollen. As the four stigmas roll back so as ultimately to touch
the anthers before these have lost all their pollen, autogamy takes place if insect-
visits fail.

Visirors—On the North Frisian Islands I only noticed po-cltg. or po-dvg.
insects (Apis, Bombus terrester Z., Eristalis, Scatophaga). I also observed skg.
insects at Kiel, i.e. Macroglossa stellatarum JZ. (in the dusk), and Bombus hortorum
L. 9 (in the forenoon). Loew saw Apis, po-cltg., in the Berlin Botanic Garden.

Herm. Miiller gives the following list.—

A. Diptera. Syrphidac: 1. Eristalis arbustorum JZ., very freq., po-dvg. ;
2. E. nemorum Z., do.; 3. E. tenax Z., do. B. Hymenoptera. Apidae: 4. Apis
mellifica Z. ¥, skg. and po-cltg.; 5. Bombus agrorum F. 9, skg.; 6. B. lapidarius
L. 9, do.; 7. B. sylvarum Z. 9, do.; 8. Colletes daviesanus X. 9, po-cltg.; 9. Pan-
urgus calcaratus Scop. 9 and é&. C. Lepidoptera. Sphingidae: 10. Macroglossa
stellatarum Z., skg.

Redtenbacher records the Elaterid Corymbites sulphuripennis Germ. for Austria.

1058. O. muricata L. (=O. biennis Z., according to the Jndex Kewensis).—
Kerner states that the flower mechanism in this North American species agrees with
that of O. biennis.

1059. O. biennis L., x O. muricata L.—

Visirors.—Heinsius saw the following in Holland (Bot. Jaarb. Dodonaea,
Ghent, iv, 1892, p. 115).—

Three humble-bees, skg. and effecting pollination (Bombus cognatus Sveph. 9 ;
B. hortorum Z. 9, 6 and ¥; B. rajellus A. 9). A small bee (Halictus leucozonius

Schr.) skg., and occasionally effecting pollination. Empids (Empis hyalipennis Fa//.
é and 9; E. pennaria /a//. 9) and a Syrphid (Eristalis nemorum JZ. 4), do.

1060. O. Lamarkiana Ser. (=O. biennis Z., according to the Jndex Kewensts).
(Stadler, ‘ Beitrage zur Kennt. de Nekt. u. Biol. d. Bl.,’ Berlin, 1886.)—The flower
mechanism of this species agrees with that of O. biennis. The very fragrant Lepi-
dopterid flowers are protandrous. Stadler says that the nectary lines the base of
the corolla-tube, and, like the inner surface of two-thirds of the tube, is beset
with unicellular barrier-hairs. The upper part of the tube is lined with a felting
of hairs. Nectar is secreted so abundantly that it usually ascends to a height of

.5 mm. The pollen-grains are bound together by threads of viscin, and remain
hanging between the anthers. Self-pollination is rendered impossible by this peculiar
fixing arrangement. Each pollen-grain has rounded poles, from which two or
more small bunches of threads project. These interweave with those of neighbouring
grains, and thus the pollen, ‘entangled as in a spider’s web into strings and clumps,’
remains on or between the anthers, and cannot be scattered by the wind nor by
the action of gravity.

VisiTors.—Heinsius observed the following in Holland (Bot. Jaarb. Dodonaea,
Ghent, iv, 1892, pp. 113-15).—

Four humble-bees, skg. (Bombus agrorum /. 9 and ¥; B. cognatus Sveph. ;
LB. hortorum Z. ¥: B. lapidarius Z. 9 and ¥), and 3 hover-flies, po-dvg. (Eristalis
horticola Z. 9; E. intricarius Z. 9; Pelecocera tricincta J/g. $).

1061. O. missouriensis Sims. (Knuth, ‘ Bloemenbiol. Bijdragen.’) — The
North American species cultivated under this name in our gardens bears moth

WANIS. IL G gs

450 ANGIOSPERMAE—DICOTYLEDONES

flowers. They exhale a strong odour of citrons in the evening, but are odourless
during the day. The calyx-tube is more than 1o cm. long, so that none of our
native hawk-moths possess a proboscis long enough to reach the bottom of it.
Even the proboscis of Sphinx convolvuli Z. only exceptionally attains a length
of 8cm. The flowers are homogamous. The four stigmas project about 15 mm.
beyond the tip of the anthers, so that alighting insects first touch the former and
then the latter, thus constantly effecting cross-pollination.

Visitors.—Hitchcock records the hawk-moth Deilephila lineata /. (Bull. Torrey
Bot. Cl., New York, xx, 1893, p. 362).

1062. O. grandiflora Ait—Kerner states that this North American species
bears night flowers (¢4 O. biennis, p. 448). When they open the petals rapidly
separate, and spread out in half an hour.

VistTors.— Loew observed the following in the Berlin Botanic Garden.—

A Chrysomelid (Haltica oleracea Z.) resting in the mouth of the corolla-tube,
and a bee (Apis mellifica Z. ¥), po-cltg., loading itself with long threads of pollen,
which hung down from its legs and impeded flight.

1063. O. speciosa Nutt.—The beautifully fragrant Lepidopterid flowers of
this species are white in colour, turning red when they are over. Wolfensberger
(Mitt. Schweiz. Entomol. Ges., Schaffhausen, vii, 1884, pp. 5-7) states that various
hawk-moths (Deilephila elpenor Z., D. porcellus Z.) are captured by them, the
proboscis being held fast by inwardly directed barrier-hairs lining the corolla-tube.
But Glaser (Ent. Nachr., Berlin, xiv, 1888, pp. 53-5) found (at Mannheim) that
hawk-moths went to sleep with the proboscis sunk in the flowers, thus giving the
appearance of captivity. He denies the existence of arrangements able to hold
fast the proboscis.

1064. O. Simsiana Ser. (Willkomm, ‘ Nachtbl. u. ihr Leben.’)— This Mexican
species bears night flowers.

Visirors.—The flowers are pollinated by beetles in the Prague Botanic Garden.

303. Godetia Spach.

1065. G. Lindleyana Spach.—Comes describes this species as protandrous
and self-fertile (‘Ult. stud.’).

1066. G. Cavanillesii Spach.— This species is native to Central Chili.
Philippi (Bot. Ztg., Leipzig, xxviii, 1870, pp. 104-6) says that it produces cleisto-
gamous flowers in spring.

304. Circaea Tourn.
Homogamous hover-fly flowers, in the bases of which nectar is secreted.

1067. C. lutetiana L. (Herm. Miiller, ‘ Fertilisation,’ pp. 265-7.)—The small
flowers of this species are white, often with a reddish tinge, and are arranged in
loose racemes. Hermann Miiiller describes the flower mechanism as being very
similar to that of Veronica Chamaedrys (vol. 1, p. 136). The two stamens project
one on either side from the vertically pendulous flower. Between them is the style,

ONAGRARIEAE 451

which projects rather further from the flower, and bears a terminal capitate stigma.
‘These three organs serve as alighting-rods, by which an insect must support itself
while getting at the nectar, which is secreted in the base of the flower by a ring
surrounding the beginning of the style. As the latter is somewhat lower and
longer than the stamens, insect visitors mostly alight upon it, and will effect crossing
if they are dusted with poHen from another flower. They apply this to the bilobed
stigma with the under-side of their body. The visitor next proceeds to get nearer
the nectar, using its fore-legs to seize the stamens by their thin easily yielding
bases, and pulling them inwards and downwards. The dehiscing anthers are thus
brought into contact with the under-side of the
body, dusting it anew with pollen.

An insect not infrequently alights upon
one of the two stamens. As this bends
downwards from the weight, its base and
the style are at once seized by the fore-
legs. In this way the stigma is usually
brought into contact with the under-side
of the visitor, and crossing results if a visit
has previously been paid to another flower,
for the stigma and anther touch opposite
sides of the body.

If insect-visits fail, automatic self-pol-
lination is usually excluded, for it but
rarely happens that the anthers and stigma
come into direct contact when the flower
fades.

Visitons.—These are exclusively Diptera. yuki: ‘ShoxGi et ean fan Herm
especially Syrphids. I observed the following, va": % sepali ¢ petal; 4, stamen ¢, stigma.
all skg. and po-dvg.—

(a) Muscidae: 1. Lucilia cornicina #.; 2. Musca domestica Z.; 3. Scatophagu
stercoraria Z. (6) Syrphidae: 4. Ascia podagrica /.; 5. Eristalis nemorum Z.;
6. Melanostoma mellina Z.; 7. Syrphus sp.

Herm. Miiller gives a similar list.—

(a) Afuscrdae: 1. Anthomyia sp.; 2. Musca domestica 4. (4) Syrphidae: 3.
Ascia podagrica /.; 4. Bacha elongata #.; 5. Melanostoma mellina Z.

MacLeod saw an Andrena and a hover-ily in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 299).

In Dumfriesshire, a hover-fly was recorded (Scott-Elliot, ‘Flora of Dumfries-
shire,’ p. 67).

1068. C. alpina L.—The flower mechanism is the same as that of the last
species. Kerner states that automatic self-pollination takes place towards the end
of anthesis, one or both anthers being applied to the stigma.

VisiTors.—These are again chiefly hover-flies, which behave as described for
the last species. I observed the medium-sized forms Melanostoma mellina Z. and
Eristalis sp.

Gg 2

452 ANGIOSPERMAE—DICOTYLEDONES

1069. C. intermedia Ehrh.—

VisiTors.—These are the same hover-flies as in C. alpina. This lends support
to the view that this species is a hybrid between C. lutetiana and C. alpina which
has become fixed. Loew saw Thrips in the Berlin Botanic Garden.

305. Isnardia L.

1070. I. palustris L. (=Ludwigia palustris Z.).—In this species, according
to Vaucher (‘ Hist. physiolog. des pl. d’Europe,’ II, p. 338), the anthers of the green
inconspicuous flowers are inclined towards the stigma at the beginning of anthesis,
and the shrivelled anthers together with the style soon fall off. Monoecious stocks
also occur (var. paludosa Rabenhorst).

306. Gaura L.
1071. G. biennis L.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syrphus ribesii Z., po-dvg. B. Hymeno-
ptera. (a) Apzdae: 2. Apis mellifica L. ¥, po-cltg. (0) Vespidae: 3. Odynerus
parietum Z., var. renimacula Lep. 9.

307. Fuchsia L.

1072. F. sp.—Gaertner states that the species of Fuchsia are self-fertile—

VisiTors.—Schneider records the humble-bees Bombus pratorum Z. ¥ and B.
terrester Z. ¥§ for gardens in Arctic Norway (Mus. Aars. Tromsg, 1894).

308. Trapa L.

Small inconspicuous whitish flowers. Caspary describes the nectary as a
glandular ring surrounding the middle of the ovary (‘ De nectariis’).

1073. T. natans L.—For plants of this species living in Lake Maggiore
Gibelli and Buscalioni determined that anthesis extends from the end of June to
ihe beginning of September, reaching its maximum in August. The flowers normally
expand an hour or half an hour before sunrise, and remain open only a few hours.
On warm and dry days the peduncles begin to curve carpotropically after 5-6
hours; on dull cloudy days this takes place later. The flowers almost always
open in the air, rarely under water. Some of the closed submerged ones possess
dehisced anthers and pollen-covered stigmas, so that they must be described as
hydro-cleistogamous. Usually, however, these flowers open if the plant is taken
out of the water. Such opening is effected by the elongating stamens, which press
against the petals, and push them apart. The higher temperature of the air also
helps to bring this about. After the flowers have opened in the air autogamy
usually follows.

VistTors.—Gibelli advanced the view in 1891 that pollination is presumably
effected by the larvae of the water-bug Mesovelia furcata Afuls ef Rey, but he and
Buscalioni subsequently (1893) came to the conclusion that the presence of the
larvae in the flowers is merely accidental, for these little creatures are not in the
smallest degree adapted to the mechanism. Nor did these observers regard

PASSIFLOREAE 453

the weevils occasionally met with in the flowers as regular agents of pollination,
though they might sometimes bring this about.

1074. T. verbanensis De Not. Cat. Sem. Ort. Bot. Roma (1875).—The
flowers of this species possess the same mechanism as T’. natans.

XLII. ORDER LOASEAF JUSS.

309. Caiophora Presl.

1075. C. lateritia Benth. (Delpino, ‘ Altri appar. dicog. recent. oss.’)—Delpino
describes the flowers of this species as markedly protandrous. In the first stage
of anthesis the five anthers open in succession, and occupy the middle of the flower;
they afterwards bend back towards the petals. In the second stage the stigma
matures, and occupies the position previously taken up by the anthers.

VistTors.— These appear to be bees.

XLIV. ORDER TURNERACEAE H.B. et K.

Urban states that about eight-ninths of all the known species of this order are
dimorphously heterostylous (Verh. bot. Ver., Berlin, xxiv, 1882).

XLV. ORDER PASSIFLOREAE JUSS.

(including PapayaceaE Juss.)

310. Passiflora L.

Protandrous, humble-bee (and humming-bird) flowers; with nectar secreted by
a fleshy ring in the base of the calyx, and protected by three nectar-covers.

1076. P. caerulea L. (Sprengel, ‘Entd. Geh., pp. 160-5; Herm. Miiller,
- Fertilisation, pp. 267-8; Warnstorf, Schr. natw. Ver., Wernigerode, xi, 1896,
pp. 3-4.)—The beautiful flowers of this species are large and very conspicuous.
The petals are white, and so is the inner surface of the calyx. There are differently
coloured concentric rings which serve as nectar-guides. They consist of a large
outer circlet of rays, a small inner one, and the outer nectar-cover. As the nectar
reservoir possesses but one ring-like aperture, visitors must go right round the
reservoir to get all the nectar. Larger insects only can effect crossing, and it is
easy for them to go round owing to the large outer circlet of rays; they move
from ray to ray as if on the spokes of a wheel, thrusting their proboscis into the
reservoir as they do so.

In the first stage of anthesis, a large insect (such as a humble-bee) when
sucking the nectar, receives pollen on its back from the downwardly dehiscing
anthers. In the second stage the styles have curved downwards to such an extent
that the now receptive stigmas are lower than the empty anthers. It follows that
older flowers are fertilized by pollen from younger ones.

Warnstorf gives a similar account.—Anthesis lasts one day. When the bud
opens the five anthers have already dehisced and are directed outwards, in line
with the thick stiff filaments. As the flower fully expands, each anther rotates
through an angle of 180° in the vertical plane of its filament, so as to bring its

454 ANGIOSPERMAE—DICOTYLEDONES

pollen-covered lobes towards the interior of the flower. There is next a second
rotation of g0° into a horizontal plane cutting the first at right angles, so that
the anther finally comes to be at right angles to the tip of its filament with its
dehisced surface facing downwards. The three purple-flecked styles with their
green capitate stigmas curve upwards, and project about ro mm. beyond the anthers.
Autogamy would seem to be excluded under such circumstances, yet it is possible
that the stigmas and anthers may be brought into contact when the flower closes
at the end of the single day’s anthesis. This is the more probable as Warnstor!
saw a fully formed fruit in a greenhouse. Here then, there is apparently a case
in which an obviously chasmogamous flower is only self-pollinated after it has
closed. The pollen-grains are golden-yellow in colour, adhesive, conico-tetrahedral,
with low anastomosing ridges or folds, 63-75 » in diameter.

Visirors.—Delpino has observed humble-bees and the carpenter bee Xylocopa
violacea (‘Sugli appar. d. fecondaz. nelle piante autocarp.,’ p. 31; Hildebrand, Dot.
Ztg., Leipzig, xxv, 1867, p. 284).

1077. P. princeps Lodd. (=P. racemosa Bro/.).—According to Delpino (* Ul.
O8S.,, pp. 170, 172), the corolla-tube is divided into three chambers by circlets of
rays, the lowest containing the nectar, which only discriminating visitors can reach.

Visirors.—Delpino supposes these to be humming-birds. Fritz Miiller (Herm.
Miller, ‘Feftilisation,’ p. 268) actually observed them in Brazil on flowers of
species of Passiflora. He is of opinion that the latticed work in the upper part
of the corolla-tube is not for the purpose of keeping out unbidden guests, but for
trapping small insects to serve as food for the humming-birds, in return for which
the latter effect pollination.

3u.. Papaya Tourn.

1078. P. carica L. (=Carica Papaya Z.)—According to Baillon (Bull. soc.
linn., Paris, i, 1887), the species is usually dioecious, but often monoecious under
cultivation. A plant reared in a greenhouse from Bourbon seed produced flowers
which were all male. When transplanted to the open air, the terminal flowers of
numerous inflorescences were female, and after these were fertilized the originally
male plant set a number of good rapidly growing fruits.

NLVI. ORDER CUCURBITACEAE JUSS.

Lirerature.— Knuth, ‘ Grundriss d. Bliitenbiol.,’ p. 55.

Species monoecious, or more frequently dioecious. The male flowers are larger
than the female ones, so that insects usually pay their first visits to the former.
The nectar is secreted in the bottom of a naked fleshy cup formed by the fusion
of the lower parts of calyx and corolla. On the stamens of many species there
are numerous glands, which, when their tips are broken off, serve to moisten the
pollen-grains and make them sticky, according to Halsted.

Arcangeli (Atti del congresso bot. internaz., 1892, pp. 441-54, Genova, 1393)
describes the flower mechanism, and especially the nectaries of various Cucurbitaceac,
i.e. Cucurbita maxima Duch., C. Pepo L., Lagenaria vulgaris Ser., Cucumis Melo Z.,
Benincasa, Ecballion, Momordica, and Trichosanthes. Bees are particularly active

CUCURBITACEAE 455

as pollinators; in the case of Benincasa cerifera, a humble-bee was also observed.
Lagenaria appears to be visited by crepuscular insects (presumably hawk-moths).

The nectaries consist of a layer of secretory tissue about 1 mm. thick, and
are provided with water-stomata. The nectar results from the conversion of starch
into sugar (glucose) by means of the protoplasm or a special ferment (Solla, Justs
bot. Jahresber., Leipzig, xxi (1893), 1896, p. 335).

312. Bryonia L.

Flowers monoecious or dioecious, greenish-yellow in colour; with concealed

nectar, secreted as above described. Two pairs of filaments are fused, the fifth
is free.

1079. B. dioica Jacq. (Herm. Miiller, ‘Fertilisation,’ pp. 268-70, ‘ Weit.
Beob.,’ I], p. 210; Ludwig, Verh. bot. Ver., Berlin, xxvi, 1885, p. 21; Schmiede-

knecht, ‘ Apidae europ.,’ I, p. 665; Knuth, ‘ Bloemenbiol. Bijdragen.’)—The female
flowers of this species are only

half as large as the male ones. Nee SQ
In the latter, according to Hermann 'e Gis
Miiller, the filaments spring from ff ow oe

the edge of the cup formed by
the fusion of the bases of calyx
and corolla, and they incline
together so as completely to
cover the cup. Three narrow
lateral passages covered by long
hairs lead between the filaments
to this nectary, and there is a
fourth access from above, running
down between the upper ends of Fic. 154. Bryonza dioica, L. (after Herm. Miiller), (1) and

(2) Anthers of male flowers. (3) A male flower in longitudinal
the filaments. The anthers dehisce section; enlarged. (4) The same, seen from above; further

: enlarged. a, filament; 4, colourless spherules; 2, nectary ;
by long narrow clefts, curved in f, pollen. The lateral nectar-passages are indicated by arrows.
such a way that for most of their

length they are turned towards the lateral passages, while their uppermost parts face
directly upwards. An insect probing for nectar will therefore be dusted with pollen
either on the lower side of its body, or on both sides of its head, and in subsequent
visits to female flowers will effect pollination. The female flowers are only half
the size of the male ones, and are therefore usually visited later. The style projects
from the middle of the nectar-cup, and divides into three diverging branches. The
ends of these are very broad, lobed, and set with projecting points, so that an
alighting insect is obliged to touch them. Pollen is carried by visitors to a con-
siderable distance. F. Ludwig, for example, observed the crossing of a female
plant by pollen from a male plant 40 m. away.

Visirors.—The chief one is the bee Andrena florea /.9 and @, which visits
the flowers of this species almost exclusively (Herm. Miiller, Schmiedeknecht),—
‘It is by far the commonest visitor, and appears to obtain all it requires in
the way of flower food from this plant alone’ (Herm. Miiller). Schletterer

456 ANGIOSPERMAE—DICOTYLEDONES

also records it for Pola as the almost exclusive pollinator of the species, ‘con-
stantly enjoying a dip into the flowers.’ I have not hitherto succeeded in capturing
this rare bee, although white bryony is common in the neighbourhood of Kiel,
and I have repeatedly watched the flowers. Of other visitors only those that
suck nectar require mention as pollinating agents. For those which devour or
collect pollen usually visit none but male flowers, and only occasionally go by
mistake to one of the smaller female ones.

Herm. Miiller (H.M.), Buddeberg (Budd.), and myself (Kn.) have observed
the following.—

A. Coleoptera. Zelephoridae: 1. Dasytes sp., only on @ flowers, po-dvg.
(H. M.). B. Diptera. (a) Empzdae: 2. Empis livida 9, skg.(H.M.). (4) Syrphidae :
3. Ascia podagrica /., po-dvg. (H.M.); 4. Eristalis tenax Z, do. (Kn.); 5. Rhingia
rostrata L., do. (Kn.); 6. Syrphus balteatus Deg., do. (H.M.). C. Hymenoptera.
(a) Apidae: 7. Andrena florea &. 9 and 6, skg. and po-cltg. (H.M.); 8. A. fulvicrus
&. 3, skg. (H.M.); 9. A. nigroaenea A. 9 and 6, do. (H.M.); ro. Apis mellifica Z. ¥,
po-cltg. (H.M., Kn.); 11. Coelioxys simplex MWy/. 9, skg. (H.M.); 12. Halictus
cylindricus /. 9, do. (Budd.); 13. H. morio /. 4, do. (Budd.); 14. H. sexnotatus
K. 9, po-cltg. (H.M.); 15. H. sexstrigatus Schenck 9, do. (H.M.). (4) Sphegidae:
16. Ammophila sabulosa Z., skg. (H.M).; 17. Gorytes mystaceus Z., do. (H. M.).
(c) Vespedae: 18. Eumenes pomiformis /. 4, skg. (H.M.); 19. Odynerus parietum
1. 9, do. (H.M.).

Schiner saw the Muscid Orellia wiedemanni jJ/g. in Austria; Schletterer the
fossorial wasp Pemphredon unicolor /. at Pola; and Handlirsch the fossorial wasp
Gorytes mystaceus L.

1080. B. alba L. (Sprengel, ‘Entd. Geh.,’ pp. 435-6.) —This species is monoe-
cious. The flower mechanism is the same as that of B. dioica. Hildebrand (Bot.
Ztg., Leipzig, li, 1893, p. 30) says that at first purely male inflorescences appear,
and ultimately purely female ones. In intermediate stages individual flowers may
develop female instead of male organs, and vice versa.

Schmiedeknecht (Thuringia), Friese (Alsace, Hungary, and Switzerland), Saunders
and Smith (England), all record Andrena florea /. as the exclusive visitor.

Schenck observed the following bees in Nassau.—

1. Andrena cingulata #.; 2. A. florea #.; 3. A. fucata Sm.; 4. A. labialis ALS;
5. A. labiata Schenck; 6. Halictus morio #.; 7. H. sexnotatus A’ 9.

313. Sicyos L.

Flowers monoecious, greenish-white in colour; with exposed nectar, secreted
by a central disk.

1081. S. angulata L. (Knuth, ‘ Bliitenbiol. Herbstbeob.’)—This species climbs
to a height of several metres, and bears inconspicuous greenish-white blossoms.
The male flowers are arranged in corymbs which gradually develop into racemes.
There are from ten to twenty flowers in each inflorescence, but as only one (rarely
two) matures at a time, the flowering period is prolonged. When the anthers have
shed their pollen the flower closes again and soon drops off. The male flower
is about 1 cm. in diameter, of which approximately one-third is taken up by a large
central secretory disk, while the remaining two-thirds are occupied by the five
whitish petals, that are traversed by green veins. From the centre of the disk

CUCURBITACEAE 457

the column of filaments rises to a height of 1 mm. It is 2 mm. in diameter, and
bears at its apex the united and convoluted anthers, as a rounded mass. Dehiscence
takes place in the mature bud.

The female flowers are considerably smaller than the male ones, and are
arranged in heads of fifteen to twenty flowers, these too being less conspicuous
than the male inflorescences. The latter are therefore (as Sprengel emphasized
for Bryonia alba Z.) usually the first to be noticed and visited by insects, and it
is not till they are done with that the less easily observed female flowers receive
attention. All the female flowers of an inflorescence mature simultaneously. It
is therefore possible or even probable that all may be crossed with the pollen
brought at a single insect-visit. The diameter of the female flower is only 4-5 mm.
In the middle of the five petals, which resemble those of the male flowers, the style
projects to a height of 2 mm. from a small secretory disk. It terminates in three
capitate stigmas. These occupy the entrance to the flower, so that a nectar-seeking
insect must inevitably touch and pollinate them if it has previously visited a male
tlower. The large number of visitors to these inconspicuous greenish flowers, as
well as their strong action (like those of Bryonia dioica Z.) on photographic plates,
suggested to me that they may possess means of attraction invisible to human
eyes though perceptible to those of insects. I expressed the view that the flower
pigment of Sicyos (and also of Bryonia) emits ultra-violet rays; but it is also
possible that the marked effect upon photographic plates is due to reflection of
light by the numerous glands which cover the flowers (cf Vol. I, pp. 87-8; and
Nos. 1742, 1743, 1747, in the Bibliography, Vol. I, pp. 285-6).

Visirors.—I observed the following, all freq., skg.—a bee (Apis mellifica Z.),
a wasp (Vespa vulgaris Z.), and 6 flies (Eristalis nemorum Z.; Lucilia caesar Z.,
Onesia sepulcralis Z.; Sarcophaga carnaria Z.; Sepsis cynipsea Z.; Syrphus
ribesii Z.).

314. Cucumis L.

Large yellow monoecious flowers, with the same mechanism as Bryonia.

‘I'wo pairs of filaments are united, the fifth is free. The anthers converge.

1082, C. sativus L. (Sprengel, ‘Entd. Geh.,’ p. 435.)—In this species again
the male flowers are much larger than the female ones, and are therefore usually
visited first.

According to the researches of F. Noll, this species, like the fig and Mespilus
yermanica, var. apyrena Koch, can produce ripe but seedless fruits without pollination
(parthenocarpy). (Verh, nathist. Ver., Bonn, 1902.)

Visitors.—I only observed the honey-bee, skg., at Kiel (‘Bloemenbiol. Bij-
dragen’).

Sickmann records the fossorial wasp Crabro brevis 7. d. Z. for Osnabriick,
very common.

1083. C. Melo L. (Aubert, J. soc. horticul. France, Paris, Ser. 3, iii, 1881,
p. 233-)—Aubert states that the first female flowers of this species open five to six
days after the first male ones, and that the latter are much more numerous. The
results of artificial fertilization are seen in two to three days, and the first ripe fruits
appear after seven to eight weeks.

458 ANGIOSPERMAE—DICOTYLEDONES

315. Cucurbita Juss.
Very large yolk-yellow monoecious flowers; with nectar secreted as in Bryonia.
1084. C. Pepo L. (Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The

male flowers of this species mature somewhat earlier than the female ones, and
their stalks are rather longer, so that they are first visited.

Warnstorf states that the calyx and corolla of the male flowers are uniled at
the base, where they support a cup-shaped disk with a tumid edge. Abundant
nectar is secreted. This cup is completely covered by the column of stamens, and
is only accessible by means of two to four openings at the bases of the filaments.
The anthers are extrorse. The pollen-grains are very large, yellow, spherical,
prickly; covered with a thin layer of oil, and therefore extremely adhesive; on
an average 163 w in diameter.

Visitors.—I observed the honey-bee, very freq., skg., at Kiel; 2 or even 3
bees were often to be seen in one flower.

316. Sechium P. Br.

1085. S. edule Sw.—Arcangeli states that in this species there are two nectaries
in the base of both female and male flowers. In the latter they are in the form
of small narrow, inconspicuous pockets; in the former they are larger and morc
conspicuous. The explanation of this is perhaps that insect visitors can only find
nectar in the female flowers, while in the male ones they may also obtain pollen.

317. Ecballium A. Rich.

1086. E. Elaterium A. Rich.—According to Hildebrand (Bot. Ztg., Leipzig,
li, 1893), this species shows the most varied arrangements of the male and female
inflorescences, In autumn the last flowers to appear are solitary and female, which
can be fertilized by the pollen of earlier male ones.

XLVI. ORDER CACTEAL DC.

Hansgirg states that the numerous filaments of many cactus flowers are almost
equally sensitive on all sides to mechanical stimulation, curving inwards from the
corolla towards the stigmas in response, e.g. in Opuntia Ficus-indica A/7/7., O. Engel-
manni Sa/m-Dyck, O. Camanchica Lugelm. e¢ Bigel., and O. Rafinesquii Engel.

318. Opuntia Tour.

Flowers mostly large, feebly protogynous, and devoid of nectar.

1087. O. vulgaris Mill. (=Cactus Opuntia Z.). (Schulz, ‘Beitrage, 11,
p- 80.)—The flowers of this species, which is cultivated in South Europe, are
nectarless, and Schulz gives their diameter as 30-40 mm. The outermost perianth
leaves are greenish-yellow, while the inner ones are of a brilliant sulphur-yellow
colour. In dull weather and at night they bend somewhat inwards. The stigmas
are receptive at the beginning of anthesis. Before the flowers expand the filaments
are curved inwards, but afterwards become more or less erect. The anthers are

UMBELLIFERAE 459

at first turned outwards, but subsequently assume an oblique or horizontal position ;
they are rarely turned inwards. The filaments are somewhat sensitive. They
incline inwards on being touched by insects. or even spontaneously, and thickly
cover the stigmas with pollen, so that autogamy regularly takes place. It is always
effective.

VisiTors.—Schulz records numerous flies, bees, and beetles for Bozen, but
especially the beetle Trichodes apiarius Z., which devours pollen, stamens, and
even petals. He found it in almost every flower, as many as 5-10 in some.

1088. O. nana Vis. (Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, Il, pp. 213, 219.)—
This species is cultivated in the South Tyrol and in Dalmatia. Kerner describes
the flower as having but a short anthesis, opening between g and 10 a.m. and
losing its petals the following day. It is, however, slightly protogynous, the stigma
being receptive a few hours before the anthers dehisce. Towards the end of
anthesis, automatic self-pollination is effected, the outer anthers coming into contact
with the stigma, which is in the form of a sinuous thickening on the end of the style.

XLVI. ORDER UA/BELLIFERAE JUSS.

LiTeraTure.—Sprengel, ‘ Entd. Geh.,’ pp. 153-9; Herm. Miller, ‘ Fertilisation,’
pp. 270-1; Drude, in Engler and Prantl, ‘D. nat. Pflanzenfam.,’ III, 8, pp. 88
et seq.; Knuth, ‘Flora d. Prov. Schleswig-Holstein, &c.,’ p. 326, ‘Bl. u. Insekt. a. d.
nordfr. Ins., pp. 75-6, ‘Grundriss d. Bliitenbiol.,’ pp. 59-60.

The small flowers are aggregated into inflorescences of considerable size, which
are usually compound, often radially symmetrical umbels, less frequently capitula.
Insects can therefore see them from a distance. An additional means of attraction
is the aromatic odour, often very strong, which characterizes many species. Nectar
is secreted by an epigynous disk, and lies freely exposed in the middle of the
flower in umbellate species. In capitulate ones (Eryngium and the like) it is
concealed in the base of a tube formed by the erect petals. The flowers of most
Umbelliferae consequently belong to class E, but some to class S. As most species
are protandrous, cross-pollination by means of insects is possible. The number
and diversity of visits are in proportion to the conspicuousness of the inflorescences.
Protogyny is rare (Echinophora spinosa Z.); homogamy occurs here and there.

Kerner (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, IJ, p. 311) describes the genera Eryngium
and Hacquetia as protogynous, because the stamens are at first curved inward like
hooks, and the anthers are still closed, when the already sticky shining stigmas
project far out of the bud. He also (loc. cit.) describes as protogynous the species
of Aethusa, Astrantia, Caucalis, Pachypleurum, Scandix, and Turgenia. I cannot,
however, adopt Kerner’s view, but agree with the following remarks made by
Kirchner on the subject (‘Die Bliiten der Umbelliferen’).—This contention of
Kerner’s may well be called in question, for it rests on insufficient evidence, and
most of it is in contradiction to the observations of other careful observers, as may
be seen by consulting the following papers for particular genera.—Aethusa : Sprengel,
‘Entd. Geh., p. 153; Schulz, ‘ Beitrige,’ II, p. 84. Astrantia: Herm. Miiller,
‘Fertilisation,’ pp. 272-3; Schulz, op. cit., 1, p. 41. Caucalis: Schulz, op. cit., 1,
p. 59. Eryngium: Herm. Miller, op. cit, pp. 271-2; Schulz, op. cit., I, p. 42:

460 ANGIOSPERMAE—DICOTYLEDONES

Knuth, Bot. Centralbl., Cassel, xl, 1889, p. 273. Pachypleurum: Herm. Miiller,
‘Alpenblumen,’ p. 120. Sanicula: Herm. Miiller, ‘Weit. Beob.,’ I, p. 303; Schulz,
op. cit, I, p. 40. Scandix: Henslow, Trans. Linn. Soc. (Bot.), Ser. 2, i, 1880,
p. 365; Schulz, op. cit. I, p. 61. Turgenia: Schulz, op. cit., I, p. 60. As regards
Astrantia major Z., Eryngium campestre Z., and Sanicula europaea Z., it is expressly
stated by Schulz that the styles project from the hermaphrodite flowers at an early
stage, giving the appearance of protogyny, although the stigmas are still immature.
Kerner goes too far in his conception of protogyny, when, for example (op. cit.,
p- 312), he asserts that the Rosaceae and Cruciferae are exclusively protogynous,
and (p. 310) even speaks of protogyny when the anthers dehisce ten to fifteen minutes
after the flower opens.

Kirchner (op. cit.) discusses protogyny among the Umbelliferae. The first
accounts were given by A. F. Foerste (Bot. Gaz., Chicago (Ill.), vii, 1882, pp. 70-1),
and W. Trelease (loc. cit.), and have reference to Erigenia bulbosa Mutt The
protogyny of this species was subsequently confirmed by C. Robertson (op. cit., xii,
1888, p. 193), and that of four other North American Umbelliferae was also
established, i.e. Sanicula marilandica Z., Zizia aurea A’och, Pimpinella integerrima
Benth. et Hook. f., and Polytaenia Nuttallii DC. During the autumn of 1891
Kirchner had the good fortune to observe on the Lido near Venice a case of distinct
protogyny in one of the European Umbelliferae, Echinophora spinosa Z., the stigmas
of which are mature before any of the anthers dehisce.

In protandrous hermaphrodite flowers the stamens ripen successively. When
the flower opens, one anther springs up so as to occupy the middle of the flower, its
filament being curved. When it has shed its pollen the filament bends back towards
the corolla, a second stamen assuming the position of the first, and so on. The
styles usually begin to grow when all the stamens of a flower, or even of an umbel,
are withered; they then diverge so that the terminal stigmas occupy the middle of
the flower.

In correspondence with the exposed position of the nectar in most species, the
large majority of guests are short-tongued insects (flies, beetles, wasps, and some
hees). Lepidoptera with their long proboscis are only seen occasionally as visitors.
but seek out more frequently the flowers of those Umbelliferae which belong to
class S. The more specialized bees (honey-bees, humble-bees, and the like) also
resort in greater numbers to flowers of the latter class, and generally only collect
pollen from those of class E; sometimes, however, they lick the nectar of such
flowers. Many species are andromonoecious.

Warnstorf makes the following remarks (Verh. bot. Ver., Berlin, xxxviii, 1896.)—
A definite line of adaptation is noticeable in our native Umbelliferae as regards
flower pollination. In order to appreciate this fully, all the inflorescences of a plant
or branch have to be considered. It then appears that in by far the greater number
of cases, the primary umbel only bears hermaphrodite flowers in its umbellules, there
being rarely a few male flowers in the centre, while very rarely all the flowers have
become female by the degeneration of the anthers. The last condition is associated
with the presence of relatively long styles. In most smaller secondary umbels,
as a general rule, the marginal flowers of the umbellules are hermaphrodite only,
while the central ones are male: more rarely all are hermaphrodite, as in the

UMBELLIFERAE 461

umbellules of the primary umbel. The umbellules of tertiary umbels display a still
further diminution of hermaphrodite flowers, and an increase of male ones: either
a few marginal hermaphrodite flowers still remain, or these have entirely disappeared,
so that the entire umbel is often purely male. If the well-marked protandry of
Umbelliferae promotes cross-pollination in a high degree, this is further accentuated
by the peculiar distribution of the sexes. Warnstorf found that the anthers of our
native species are neither introrse nor extrorse, but dehisce laterally. As, however,
the two outer anther-valves curve back towards one another, while the two inner ones
retain their original position, the shed pollen faces outwards.

The protandry varies greatly in degree. Beketow (‘U. d. Proterand. d.
Umbellif.,”” 1890) found it to be most strongly marked in Anthriscus sylvestris
Hoffm. and Carum Carvi Z., the flowers appearing at first purely male, and
afterwards purely female. The first (terminal) umbel, borne by the primary
axis, is here very feebly developed; the lateral umbels are much larger, and by
elongation of the secondary axis attain a higher level. The terminal umbel is
purely female at the time when the lateral ones are still male; the female flowers
being at a lower level than the male ones, their geitonogamous pollination is
ensured.

Beketow states that protandry is much feebler in Heracleum Sphondylium Z.,
Aegopodium Podagraria Z., and Angelica sylvestris Z. In these species the primary
umbel is larger and at a higher level than the lateral ones. It is possible that this
relation between the degree of protandry and the development and position of the
various umbels is of more general application. (Rothert, Bot. Centralbl., Cassel, xlv,
1891, p. 381.)

Schulz summarizes his observations somewhat as follows (‘Beitrage,’ II,
pp. 90-1).—

Male flowers are generally associated with hermaphrodite ones; they are
either borne on the same or on different stocks. In the former case, they are
distributed in one or other of the two following ways.—

1. Hermaphrodite and male flowers are associated either in all the umbels
{mixed umbels), or only in some, usually those of a higher order. In the latter
case, umbels of a lower order are as a rule entirely hermaphrodite. In mixed
umbels both kinds of flower are usually present in all the umbellules; purely male
umbellules are not common in most species, and as a rule only occur in the centre of
umbels of a higher order. Now and then, however, individual umbels are purely
male; in some species, such as Oenanthe fistulosa Z., those of the highest order are
always so. Most of the species investigated by Schulz belong to this group. The
umbellules either bear,—

(2) marginal hermaphrodite and central male flowers, except in certain genera
where the terminal flower, when present, is usually hermaphrodite; or,

(4) the hermaphrodite flowers at the periphery, in the centre (Sanicula europaea
L.), or an intermediate zone between marginal and central male flowers (Astrantia
major Z.).

2. Hermaphrodite and male flowers are but very seldom associated in the same
umbel; only umbels of a higher (or even the highest) order are male. Sometimes,
however, all the umbels of a plant are completely hermaphrodite ; more rarely flowers

462 ANGIOSPERMAE—DICOTYLEDONES

of both kinds are associated in the same umbel, either in the same or in different
umbellules. To this group belong, for example, Eryngium campestre Z. and
Laserpitium latifolium Z.

The two kinds of flower are only found on separate stocks in the case of
‘Trinia glauca Rezchd. Even in this species, however, plants bearing both kinds
sometimes occur—the hermaphrodite flowers being generally in the minority —
either in all the umbellules of all or some umbels, or only in some umbellules,
as a rule of all the umbels. This variation is either associated with male and
hermaphrodite plants, or with male ones only, or it may be quite sporadic. In
the case of Trinia, hermaphrodite flowers may also be replaced by female. Female
flowers are much rarer than male ones in this order, and it appears that they are
not constantly present in any one species. They have been observed in the following
cases besides that of Trinia glauca—-Eryngium campestre Z., Pimpinella magna Z.,
P. Saxifraga Z., and Daucus Carota Z. In Eryngium campestre and the two species
of Pimpinella, the female flowers either occur alone or together with neuter ones, but
are never associated on the same stock with hermaphrodite or male flowers. The
same is the case as a rule for Daucus Carota; but in this species female flowers now
and then occur at the periphery of umbellules bearing marginal hermaphrodite flowers
and central male ones. Not infrequently the male flowers of the four species
mentioned are replaced by neuter ones. though throughout the summer they only
possess umbels bearing hermaphrodite flowers, or hermaphrodite and male ones,
or ultimately only male ones. Neuter flowers are of sporadic occurrence and in
some localities they appear to be wanting altogether. This also applies to the
case of Orlaya grandiflora “offm. The hermaphrodite flowers are protandrous in
most species, and in some protandry is so well marked that the styles and stigmas do
not attain their full development till after the stamens and petals have fallen off.
A number of Umbelliferae, however, bear flowers which are homogamous or but
slightly protandrous. These are almost exclusively species that owing to the small
number and size of the flowers, and their indistinct white or greenish-white colour,
attract but few insect visitors, e.g. Aethusa Cynapium Z., Caucaloides daucoides Z.,
Torilis infesta o/h, Scandix Pecten-Veneris Z., and Anthriscus vulgaris Bernh.
In the favourite habitats of these species, among corn or dense brushwood, there are
but few insects that visit flowers. It is more remarkable that Anethum graveolens Z.
is a homogamous species, for though its flowers are small and produce but little
nectar, their brilliant yellow colour makes them very conspicuous, and they possess
a powerful aromatic odour. To the foregoing autogamous species I would add
Helosciadium inundatum Aoch (cf. Knuth, ‘Bl. u, Insekt. a. d. nordfr. Ins.,’ p. 78) and
Hydrocotyle vulgaris Z.

Drude distinguishes the following ways of distribution of the sexes, and order of
maturation of stamens and carpels (in Engler and Prantl, ‘D. nat. Pflanzenfam.,’ III,

8, pp. 89-91).—
A. Flowers monomorphous, all hermaphrodite (except in the feebly developed
umbels of higher order).

1. Flowers almost homogamous, the two sexes developing in rapid succession ;
e.g. Hydrocotyle vulgaris Z., Anethum, Aethusa, and others.

UMBELLIFERAE 463

2. Flowers strongly protandrous (those of the ultimate lateral umbels being
male by reduction). The commonest case.
B. Flowers pleomorphous (¥ and 6) in the primary umbels.
3- Here belong the common cases of andromonoecism ; e.g. Astrantia major
/.., Chaerophyllum aromaticum Z., Scandix Pecten-Veneris Z., Torilis Anthriscus
Bernh., and so forth.
4. Well-marked monoecism ; e.g. Echinophora.
5. Well-marked dioecism ; e. g. Arctopus.
C. Flowers of the primary umbels with uniformly reduced male organs; those of
the lateral umbels, on the contrary, purely male.

6. Here belong the rare cases of trimonoecism, or of monoecious polygamy ;
e.g. Ferula.

4

FiG.155. Gettonoganzy of Chaerophylluim aromaticuim, L.(after Kerner). A. The hermaphrodite
lowers are open; the pseudo-hermaphrodite pollen flowers are still closed. B. The hermaphrodite
lowers have lost their stamers; the pseudo-hcrmaphrodite pollen flowers have opened and are shedding
their pollen upon the stigmas of the former.

To division A must be added the instance of protogyny discovered by Kirchner
in Echinophora spinosa Z.

Drude (op. cit.) says that if all the different cases are reviewed and compared,
it will be recognized that the umbels of this order display a tendency to ensure
crossing by preponderating development of female organs in the first flowers to
open, and their reduction in those which bloom later. This tendency is seen

464 ANGIOSPERMAE—DICOTYLEDONES

even in the third type (C), where hermaphrodite and female flowers are associated,
for here the primary umbels possess but few male flowers, while in those which
bloom last scarcely any others are present.

Something has already been said about the geitonogamy that so frequently
takes place among Umbelliferae (Vol. I, pp. 41-2). Kerner describes some very
diverse arrangements of the kind (‘ Nat. Hist. Pl,’ Eng. Ed. 1, II, pp. 323-5). In
the crowded capitula of Eryngium and Hacquetia, when the filaments elongate
their pollen-covered anthers come into contact with the mature stigmas of
neighbouring flowers, in consequence of the divergence of the styles. In
Sanicula, Astrantia, and Laserpitium, there is a deviation from this form of
geitonogamy owing to the fact that pollen flowers as well as hermaphrodite ones
are present; but here again the same thing happens, because the elongating
styles bend over and entangle themselves with adjacent flowers, so that their
stigmas can take up pollen. The converse is true for Pachypleurum, where
the stamens ultimately radiate in an almost stellate manner, and touch the mature
stigmas of neighbouring flowers. The relations are similar in the case of Siler;
while in Athamanta, Meum, and Chaerophyllum, hermaphrodite and pollen flowers
are associated. After the stamens of the former have dehisced and fallen off, the
pollen flowers ripen their anthers, and drop their pollen upon the still receptive
stigmas of the originally hermaphrodite flowers. (Cf. Fig. 155 2.)

For Anthriscus, Foeniculum, Coriandrum, Sium, and Ferulago, Kerner describes
inflorescences of two kinds. The umbels that blossom first mostly contain her-
maphrodite flowers, with a few scattered pollen flowers, while those which do so later
contain only the latter kind. After the stamens of the protandrous hermaphrodite
flowers have shed their pollen and fallen off,
the stigmas become receptive and remain so
for some days. Meanwhile the lateral axes
bearing male inflorescences continue growing,
so that ultimately their umbels lie above the
mature stigmas of the hermaphrodite flowers.
These stigmas are consequently pollinated by
the shower of pollen from the dehiscing anthers
of the pollen flowers.

319. Hydrocotyle L.

Flowers small, white in colour, and arranged
in imperfect umbels ; with exposed nectar.

Fic. 156. Aydrocotyle vulgaris, L. (after . iiller. ‘ Weit.
Drude). 4. Flower at the close of pollina- 1089. H. vulgaris L. (Herm. Miller,

tion: the stamens have fallen off, and the styles Beob.,’ J, Pp- 302-33 MacLeod, Bot. Jaarb. Do-
are erect (x 5) &. Stamens with dehisced , -XK he Bl
anthers; stigmas still immature (> 5). C. donaea, Ghent, vi, 1894, p. 257; Knuth, -uU.
Dried pollen-grains swollen by immersion in Insekt. a. d. nordfr. Ins.,’ p. 76 : Warnstorf, Verh.
water (x 400), 4, bract. pana 7 7 2 ‘

bot. Ver., Berlin, xxxviii, 1896.)—The flowers of
this species are extremely inconspicuous, and only 3-5 of them are found in an

umbellule. Hermann Miiiller says they are so feebly protandrous that automatic

UMBELLIFERAE 465

self-pollination can take place, failing insect-visits. The anthers first dehisce in
slow succession, but the stigmas mature so early that the last stamen is covered
with pollen when they have become receptive. The stigmas come automatically
into contact with the pollen so as to bring about autogamy, which is completely
effective. I observed it in the North Frisian Islands. Warnstorf says that occasional
flowers are often purely female by degeneration of the anthers. The pollen-grains
are pale yellow in colour, irregular, either bipyramidal or in the form of a pyramid
with convex base, about 25 « long and 18 p» broad.

1090. H. americana L.—
Visirors.—Henslow records minute Muscids for Kew Gardens.

320. Sanicula L.

Flowers white, andromonoecious, in small sub-globose umbels; with exposed
nectar.

1og1. S. europaea L. (Herm. Miiller, ‘Weit. Beob., I, p. 303; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 257-9; Kirchner, ‘Flora v. Stuttgart,’
P- 3753 Schulz, ‘ Beitrige,’ II, pp. 81-2; Kerner,‘ Nat. Hist. Pl.,’ Eng. Ed., 1, II,
pp. 323-4; Francke, ‘ Beitrage.’)— Hermann Miller states that there are 1-3 pro-
tandrous hermaphrodite flowers in each of the small umbels of this species, surrounded
by 10-20 purely male ones that mature later. The nectary of the small, bright-
reddish flower is a groove bounded by an annular ridge; nectar is secreted in
tolerable abundance.

According to Schulz the male flowers may also occur in the middle of the
inflorescence, and mature before the hermaphrodite ones.

Other investigators, e.g. Kerner and Francke, describe the flowers as proto-
gynous, so that the mechanism would appear to vary in different districts. Kerner
confirms Hermann Miiller’s statement that there are central hermaphrodite flowers
which first mature. The stigmas of these flowers at this stage can only be
xenogamously pollinated by insect agency. The filaments subsequently elongate
so as to bring the anthers to the same level as the stigmas. As, however, the
styles are upright, and the filaments directed obliquely outwards, the anthers and
stigmas do not come into contact. Though automatic self-pollination is thus
rendered impossible, yet, after the stamens have fallen off, there may be automatic
geitonogamy of the originally hermaphrodite flowers, for the styles diverge so as to
approach the anthers of adjacent flowers in the same inflorescence.

Visitors.— Herm. Miiller noticed small flies and small beetles (Meligethes).
MacLeod saw 2 short-tongued bees and an Empid in Flanders (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 259).

In Dumfriesshire, a Vespid, 2 Muscids, and a Syrphid were recorded (Scott-
Elliot, ‘ Flora of Dumfriesshire,’ p. 74).

321. Astrantia L.

Flowers white or reddish in colour, arranged in simple umbels; the concealed
nectar is secreted by an epigynous disk. The petals serve as a nectar-cover, being
DAVIS, 1 Hh

466 ANGIOSPERMAE—DICOTYLEDONES

erect and curved inwards. Andromonoecious or androdioecious; hermaphrodite
flowers protandrous.

1og2. A. major L. (Herm. Miiller, ‘ Fertilisation,’ pp. 272-3; ‘Alpenblumen,’
p- 116; Knuth, ‘Bliitenbiol. Herbstbeob.’; Schulz, ‘ Beitrage, II, p. 90; Kerner,
‘Nat. Hist. Pl,’ Eng. Ed. 1, I, pp. 323-4; Ricca, Atti Soc. ital. sc. nat., Milano, xiv,
1871; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—Hermann Miiller states
that the white or reddish flowers of this species are not in a continuous plane, as in
most Umbelliferae, but conspicuousness of the inflorescences is increased by the
broad whitish involucral bracts. Besides protandrous hermaphrodite flowers, each
umbel includes numerous marginal and central male ones, and as these usually

mature late, they serve to pollinate the last developed stigmas of the hermaphrodite
flowers.

Fic. 157. As/rantia major, L. (after Herm. Miller). (1) Male flower at the beginning of anthesis;
one stamen is erect, but its anther is still unripe; the four other stamens are still bent down within the flower.
(2) Male flower, with erect stamens, the anthers of two have dehisced. (3) Hermaphrodite flower at the
beginning of anthesis; two stamens are erect, but their anthers have not dehisced; the others are still bent
down within the flower. The styles project from the flower, but their stigmas are immature. (4) Her-
maphrodite flower in the second (female) stage; the stamens have all fallen off; the styles have elongated
and their stigmas are mature. a, stamen bent down within the flower; a’, erect stamen; @”, stamen
with dehisced anther; 0z, ovary; ~, petal; s, sepal; s/, immature stigma; s¢’, mature stigma.

The male flowers are distributed andromonoeciously or androdioeciously.
Schulz says that the male flowers always exceed the hermaphrodite ones in
number. Purely female inflorescences are rare. According to Schulz, protandry
is so well marked that the stigmas do not become receptive till after the anthers
have dehisced. Kerner describes the hermaphrodite flowers as being protogynous,
and, like those of Sanicula europaea, geitonogamously fertilized by the pollen of
adjacent male ones.

Warnstorf says that primary umbels include both hermaphrodite and male
flowers; while secondary ones either contain a few hermaphrodite flowers and
numerous male ones, or else the latter only, purely male umbels being the last

UMBELLIFERAE 467

to develop. The pollen-grains are white in colour, ellipsoidal, tuberculated, 63 p

long and 25 » broad.
Vistrors.—Herm. Miiller (H. M.) and myself (Kn.) observed the following.—

A. Coleoptera. Dermestidae: 1. Anthrenus pimpinellae 7. (H. M.). B. Di-
ptera. (a) Muscidae: 2. Lucilia caesar L. (Kn.); 3. L. cornicina Z., nect-lkg. (H. M.,
Kn.); 4. Miltogramma punctata Mg. (H. M.); 5. Onesia sepulcralis, Afg., freq.
(Kn.); 6. Pollenia rudis /. (Kn.); 7. Sarcophaga carnaria Z.(Kn.); 8. Scatophaga
merdaria Z.(Kn.); 9. S. stercoraria Z., very numerous (Kn.). (4) Syrphidae: 10.
Eristalis arbustorum Z., po-dvg. and nect-Ikg. (H. M., Kn.); 11. E. nemorum Z.,
common (Kn.); 12. Helophilus floreus Z. (Kn.); 13. Melanostoma gracilis Mg.
(Kn.): 14. Syritta pipiens Z.(Kn.); 15. Syrphus ribesii Z.(Kn.). C. Hemiptera.
16. Lygus (Orthops) kalonii Z.(Kn.). D. Hymenoptera. (a) Apidae: 17. An-
drena albicrus X. 6, skg. (H. M.); 18. Bombus lapidarius Z. (Kn.); 19. B. terrester.
L. (Kkn.); 20. Prosopis armillata Vy/. 3, skg. (H. M.); 21. P. signata Pz. 8, do.

a

mitts ime,

Fic. 158. As/rantia minor, L. (after Herm, Miller), 4d. Mermaphrodite Hower in the first (males
stage. &. Upper part of a hermaphrodite flower in the second (female) stage. C. Male flower with
vestiges of ovary and styles. D. Male flower without any such vestiges. Z. Fieshy disk (with 10 nectar-
pits! on the ovary of a male flower. 2, 3, 4, 5, stamens; 97, style; ov, ovary; 7, petal; s, sepal; sé, stigma.

(H. M.). (8) Sphegidae: 22, Cerceris arenaria £.(Kn.); 23. Oxybelus uniglumis
£.(Kn.). (c) Vesprdae: 24. Odynerus parietum Z.(Kn.); 25. Vespa sylvestris Scop.
(Kn.). E. Lepidoptera. hopalocera: 26. Pieris sp., skg. (Kn.); 27. Vanessa

atalanta Z., do. (Kn.).
Herm. Miiller saw 7 beetles, 3 flies, 2 Hymenoptera, and a Lepidopterid in the

Alps. F. F. Kohl noticed 3 true wasps in the Tyrol (Odynerus parietum Z.;
O. trifasciatus /.; O. simplex /.); and MacLeod 3 Hymenoptera, 3 beetles, and 2
Muscids in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 417-18).
Loew observed the following in the Berlin Botanic Garden. —
A. Diptera. (a) Muscidac: 1.Graphomyia maculata Scop.; 2. Lucilia caesar LZ.
(6) Syrphidae: 3. Eristalis nemorum Z.; 4. Syritta pipiens 2. B. Hymenoptera.
(a) Apidae: 5. Apis mellifica Z.,ske. (4) Sphegidae: 6. Cerccris variabilis ; and on
Hh 2

468 ANGIOSPERMAE—DICOTYLEDONES

the var. zufermedia a Syrphid (Syrphus balteatus Deg.): also on the var. znvolucrata
Koch.—A. Coleoptera. Coccinellidae: 1. Coccinella quatuordecimpunctata L.,
nect-Ikg. B. Diptera. Syrphidae: 2. Eristalis nemorum Z. C. Hymenoptera.
(2) Apidae: 3. Prosopis sp. 9, po-clig. (4) Sphegidae: 4. Oxybelus uniglumis
L. ¥.

1093. A. minor L. (Herm. Miller, ‘ Alpenblumen,’ pp. 114-16.)—The her-
maphrodite flowers of this species are protandrous ; there are also transitions between
andromonoecism and androdioecism. (Cf. Fig. 158.)

Visitors. Herm. Miiller saw a few Muscids.

1094. A. helleborifolia Salisb.—

Visitors.—Loew saw a Muscid (Anthomyia sp.).

1095. A. neglecta Koch et Bouché.—

Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalisnemorum Z.; 2. E.tenax Z. B. Hymeno-
ptera. (a) Agidae: 3. Sphecodes gibbus LZ. 9, skg. (4) Sphegidae: 4. Crabro
spinicollis 7.-Sch. 9; 5. Oxybelus sericatus Gerst.$; 6. Philanthus triangulum #” 9.
C. Lepidoptera. Rhopalocera: Vanessa urticae L., skg.

1 =

os,
?

FiG.159. Eryngium maritimuim, L. (1-4, enlarged four times from photographs. 5-7, drawn from
nature.) (1) Trifurcate bracteole. (2) Late flower-bud; the filaments are still incurved. (3) Flower in the
first (male) stage; the anthers have all dehisced; the stigma is not yet mature. (4) Flower towards the
end of the second (female) stage; the petals and stamens have fallen off; thestigmais mature. (5) Nectar-
secreting disk (x 8). (6) Petal from within (x 5). (7) The same from the side. a@. stamen; , base of
style; 2, nectary; ov, ovary; /, petal; s, stamen; s¢, stigma.

322. Eryngium L.

Flowers belong to class S; arranged in dense capitate umbels; whitish or
amethyst-blue in colour; protandrous; with concealed nectar, secreted by an
epigynous ro-rayed disk, and protected by the inwardly turned tips of the erect
petals. The stiff and extremely sharp involucral bracts and sepals serve as a
further protection, as do also the rigid spinous-toothed foliage-leaves. Conspicuous-
ness is brought about not only by the petals, but in some cases by the involucre
as well, and even the peduncle may contribute to this end (E. maritimum Z. and
E. amethystinum Z.).

1096. E, maritimum L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 76-8, 155 ;
Loew, ‘ Bliitenbiol. Floristik, p. 390.)—The protandrous flowers of this species are

UMBELLIFERAE 469

crowded together into conspicuous dark-blue capitate umbels. These are surrounded
by a spinose involucre, which makes it almost impossible for soft-skinned animals like
snails or caterpillars to creep up from below, and enter the umbel as unbidden guests.
The involucre is supplemented by trifurcate bracteoles, situated at the base of each
flower, and also by the five sharply pointed sepals.

The filaments are incurved in the bud, so that the anthers are enclosed in the
corolla, which is about 4 mm, long. At this time the bluish colour of the foliage-
leaves is not yet well marked, but the whole plant is whitish, and therefore not so
conspicuous as later on, when the filaments have elongated and the flowers have
entered upon their first (male) stage.

Meanwhile, the 1o-rayed disk has begun to secrete. The petals are incurved at
the tip, and closely apposed, only leaving a space for the passage of the filaments.
This firm approximation of the parts of the flower, and the deep situation of the
nectar, makes sucking impossible for any but strong insects with a proboscis at least
3-4 mm. long; and in consequence of this large or middle-sized insects are almost
the only visitors observed. These get dusted with pollen from the anthers, which
project about 3 mm. beyond the corolla, and when visiting other flowers in the
second stage necessarily pollinate the stigmas, which in such flowers are at a
corresponding level. In the second stage of anthesis the anthers have dropped
off, and the long stigmatic branches project a long way out of the flower. Self-
pollination is therefore prevented.

Visitors.—I observed the following, all skg., on the shores of Kiel Bay,
and in Sylt—

A. Diptera. 1. Syrphus ribesii Z.; 2. S. umbellatarum /. B. Hymenoptera.
3. Apis mellifica Z.; 4. Bombus lapidarius Z. C. Lepidoptera. 5. Lycaena

semiargus oé/.; 6. Polyommatus phlaeas Z.; 7. Vanessa atalanta Z.; 8. V.
urticae LZ.

MacLeod observed Apis and other bees, Vespa, and hover-flies, in Belgium.

Loew noticed the following in Mecklenburg (‘ Beitrage,’ p. 41).—

A. Diptera. Muscidae: 1. Sarcophaga carnaria Z. B. Hymenoptera.
(a) Apidae: 2. Bombus distinguendus AZor. 9, ¥, and 6, skg., ¥ and 9, also po-cltg. ;
3. B. soroénsis #. 9, skg. (6) Sphegidae: 4. Ammophila sabulosa Z., skg.;
5. Cerceris arenaria Z. 6, do.

1097. E. campestre L. (Herm. Miiller, ‘ Fertilisation, pp. 271-2; Kerner,
‘Nat. Hist. Pl.’ Eng. Ed. 1, Il, pp. 278, 311, 323; Schulz, ‘ Beitrige, I, p. 42; Knuth,
‘Bloemenbiol. Bijdragen.’)—Hermann Miiller describes the flower mechanism of this
andromonoecious species as closely resembling that of E. maritimum. The her-
maphrodite flowers are protandrous. The nectary here again is a depression
surrounded by a ten-lobed ridge, and secreting abundant nectar. This is completely
concealed at some depth, for the nectary is surrounded by the five stiff erect petals,
which are about 3 mm. long, and sharply incurved at their tips. Outside all are the
five rigid awn-like sepals, acuminate in shape and projecting considerably beyond the
petals. The sepals, together with the equally stiff and spinose involucral bracts, form
an effective protection against unbidden guests. As first pointed out by Schulz, the
styles project early from the flower, so that, as in the last species, there is apparent
protogyny. It is probably for this reason that Kerner actually describes Eryngium as

470 aANGIOSPERMAE—DICOTYLEDONES

protogynous (op. cit., pp. 311, 323). He agrees (op. cit., p. 278), however, with
Hermann Miller and others in stating that at the beginning of anthesis only pollen-
covered anthers project from all the flowers; and later only styles with receptive stigmas.

Schulz states that umbels
of orders 1 to 3 usually bear
only hermaphrodite flowers,
while those of the fourth
order chiefly contain male
ones.

Vistrors.—I observed the
following po-dvg. hover-flies ;
and skg. bees and butterflies
in the Botanic Garden of the
Kiel Ober-Realschule.—

A. Diptera. Syrphidae :
1. Eristalis nemorum JZ.;
2. Syrphus ribesii Z. B.
Hymenoptera. <Aprdae:
3. Apis mellifica Z. ¥; 4.
Bombus lapidarius Z. 9; 5.
B. terrester Z. 9 and ¥. C.
Lepidoptera. Rhopalocera:
6. Pieris rapae Z.; 7. Vanessa

Fic. 160. Eryngium campestre, 1. (after Herm. Maller).
(1) Flower in the first (male) stage. (2) Flower in the second (female) atalanta i.

stage. (3) The same; after removal of sepals, petals, and styles. ss .
(4) Petal seen from the inner side. a’, mature anther; 7, nectary; ov, Herm. Miller gives the

ovary; ~, petal; s, sepal; s/, immature stigma; s¢’, mature stigma. following list. —

A. Diptera. (a) Afuseidac: all skg.: 1. Species of Anthomyia; 2. Echinomyia
fera L.; 3. Lucilia caesar Z.; 4. Sarcophaga carnaria L. (2) Syrphidae: all freq.:
5. Eristalis arbustorum Z.; 6. E. nemorum Z.; 7. E. tenax Z.; 8. Helophilus
floreus Z. B. Hymenoptera. All skg. (a) Aprdae: 9. Andrena rosae Pz. 9;
ro. Apis mellifica Z. ¥; 11. Halictus cylindricus #. 6; 12. H. longulus Sm. é;
13. Nomada robjeotiana Ps. 9. (0) Chrystdidae: 14. Chrysis sp. (c) Sphegidac :
15. Ammophila sabulosa Z.. freq.; 16. Cerceris albofasciata Ross?, one; 17. C.
Jabiata /”, freq. ; 18. C. variabilis ScArv., not infrequent ; 19. Philanthus ‘riangulum J*;
20. Salius versicolor Scop. F.9. (d) Scolitdae: 21. Tiphia femorata F. (e) Vespidae :
22, Odynerus parietum /. 9: 23. Polistes gallica £., exceedingly common; 24. P.
biglumis Z., do.

The following are recorded by the observers, and for the localities stated.—

Krieger (Leipzig), the Chrysidid Hedychrum nobile Scop. and the fossorial wasp
Ammophila affinis A. Friese (Hungary), on the authority of Mocsary, the bee Nomia
femoralis Pall., freq. F. I’. Kohl (Tyrol) the Chrysidid Chrysis ruulans Ov. Réssler
(Wiesbaden) the moth Agrotis vestigialis Ro//. Schiner (Austria), a Syrphid (Merodon
analis AZg.), and 8 Muscids (1. Anthomyia albescens Ze//.; 2. Cnephalia bucephala
Ug.; 3. Melania bifasciata A7g.; 4. M. volvulus /.; 5. Ocyptera brassicaria Deg. ;
6. Sarcophaga grisea AZg.; 7. Sarcophila latipons /a//.; 8. S. meigeni Schzn.),
Plateau (Belgium), the honey-bee and 3 Syrphids (Syritta; Eristalis arbustorum J. ;
i. tenax J.).

1098. E. bourgati Gouan.—The flowers of this species are blue in colour,
and belong to class S.

UMBELLIFERAE 4q1

VisiTors.—MacLeod saw 2 humble-bees, an Eristalis, and 2 Muscids in the
Pyrenees.

1099. E. alpinum L.—Christ states that the involucre of this species expands
at sunrise, and closes at sunset.

1100. E. amethystinum L.—

Visitors.—Von Dalla Torre saw the wasp Vespa norwegica F. in the Tyrol,

where F, F. Kohl noticed 3 wasps (Eumenes pomiformis F.; Odynerus dantici
Rosst; Polistes gallica L.).

1101. E. giganteum 47ze).—

Visitors.—Loew saw a humble-bee in the Berlin Botanic Garden (Bombus
terrester Z., skg.).

i102. E. planum L.—
Visirors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (2) Aluscidae: 1. Lucilia caesar ZL. (5) Syrphidac: 2. Eristalis

tenax Z.; 3. Syritta pipiens Z.; 4. Syrphus corollae #. B. Hymenoptera.
Apidae: 5. Apis mellifica Z. ¥, skg.

323. Conium L.

Flowers white in colour, arranged in compound umbels; with exposed nectar. (All
the remaining genera of the order to be considered present these oecological characters,
except that the colour of the flowers may be yellow, greenish, or in some cases red.)

1103. C. maculatum
L. (Herm. Miiller, ‘ Fer-
tilisation, pp. 274-5, ‘ Weit.
Beob.,’ I, p. 311; Knuth,
‘Bl. u. Insekt. a. d. nordfr.
Ins.,’ pp. 79, 156.)—This
species will serve to illus-
trate the protandrous

flower mechanism which
h : Umbel Fic. 161. Condum macula/um, L. (after Herm. Miller). (1) Flower
characterizes most Umbel- at the beginning of the first (male) stage. (2) Flower in the middle of the

liferae. Several hundred same stage. (3) Flower in the second (female) stage. a'-a°, stamens;
2, nectary; sé, stigma.

small white flowers are
aggregated into a large compound radiating umbel, which makes the plant very
conspicuous ; this inflorescence is borne on a stem which may be as much as a metre
high. The flowers are distinctly protandrous, and when the bud opens the stamens
are at first horizontal in position, and their anthers are unripe. They occupy the
intervals between the petals. They then successively become erect, so that the
upwardly dehisced anthers lie above the still immature stigmas. When an anther
has dehisced, its filament resumes the horizontal position, and another stamen
takes its place. All the stamens are completely withered as a rule before the stigmas
mature. The styles are then r mm. long, so that they are in the same position as
that occupied by the anthers during the first (male) stage of anthesis. Numerous

472 ANGIOSPERMAE—DICOTYLEDONES

insects (flies, beetles, bees) resort to the nectar secreted by the epigynous disk, and
consequently effect cross-pollination.

Kerner says that the flowers exhale a delicate odour of nectar, but the plant as
a whole possesses a disagreeable, mouse-like smell.

Visttors.—I have seen Apis and several other bees, hover-flies (3), Muscidae
(4), and the beetle Meligethes, in the North Frisian Islands.

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list.—

A. Coleoptera. (a) Dermestidae: 1. Anthrenus pimpinellae /. (H. M.).
(b) Mitedulidae: 2. Meligethes, freq. (H. M.). (c) Scarabaeidae: 3. Trichius
fasciatus Z.(H.M.). (d@) Zelephoridae: 4. Cantharis fulva Scop., nect-lkg. (H. M.);
B. Diptera. (a) Dolichopodidae: 5. Gymnopternus germanus Wied., skg. (H. M.);
(2) Muscidae: 6. Species of Anthomyia (H. M.); 7. Aricia vagans Fa//. (Budd.);
8. Calliphora vomitoria Z. (H. M.); 9. Cyrtoneura curvipes Macg., skg. (H. M.);
10. Lucilia cornicina #. (H. M.); 11. Musca corvina #., skg. (H. M.); 12. M.
domestica Z. (H. M.); 13. Phasia analis /. (Budd.); 14. Scatophaga stercoraria LZ.
(H. M.); 15. Sepsis sp. (H.M.). (c) Stratiomyzdae: 16. Chrysomyia formosa Scop.,
skg. (H. M.); 17. Sargus cuprarius Z.(H.M.). (d@) Syrphzdae: 18. Chrysogaster
coemeteriorum Z., skg. (H. M.); 19. Eristalis arbustorum Z. (H.M.); 20. E.
nemorum Z. (H.M.); 21. Helophilus floreus Z. (Budd.); 22. Syritta pipiens Z.
(Budd.); 23. Syrphus ribesii Z., skg. (H. M.). C. Hemiptera. 24. Graphosoma
lineatum Z., skg. (Budd.). D. Hymenoptera. (a) Apzdae: 25. Andrena lepida
Schenck 8 (H. M.). (6) Jchneumonidae: 26. Several species. (¢) Sphegzdae :
27. Crabro fossorius Z. 9, nect-lkg. (Budd.); 28. C. subterraneus /. 6 (Budd.);
29. Gorytes campestris A/#//. nect-lkg. (Budd.); 30. Pompilus gibbus /. 9 (H. M.).
(d) TYenthredinidae: 31. Hylotoma cyaneocrocea forst., nect-lkg. (Budd.); 32. H.
segmentaria Pz., do. (Budd.); 33. Nematus vittatus Z. (H.M.); 34. Species of
Tenthredo (undetermined) (H.M.). E. Neuroptera. 35. Panorpis communis Z.,
nect-lkg. (H. M.).

Alfken and Leege observed the following in Juist.—

A. Coleoptera. (a) Alleculidae: 1. Cteniopus sulphureus Z., one. (6) Cocctnelli-
dae: 2. Coccinella septempunctata Z.,freq.; 3.C.undecimpunctataZ.do. B. Diptera.
(a2) Strahiomyzdae: 4. Chrysomyia formosa Scop. (6) Syrphidae: 5. Syrphus balteatus
Deg., freq. (c) Muscidae: 6. Lucilia caesar £., very common. C. Hymenoptera.
(a) Vespidae: 7. Odynerus parietum Z., several. (4) Sphegidae: 8. Crabro Cros-
socerus) wesmaéli v7. d. Z.9, do. (c¢) Scolztdae: 9. Tiphia femorata F., rare.
D. Neuroptera. /lanipennia: 10. Chrysopa abbreviata Curt., skg.

In Dumfriesshire a humble-bee, a wasp, a saw-fly, a Muscid, a hover-fly, and
a beetle were recorded (Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 82).

Friese, on the authority of Korlevic, gives the bee Andrena figurata 7or., rare,
for Fiume.

324. Smyrnium L.
1104. S. Olusatrum L.—The flowers of this species are yellowish-green in
colour.

Visirors.—Schletterer, at Pola, noticed the Braconid, Bracon urinator /., and
2 Evaniids (Gasteruption granulithora Zoarn.; G. rugulosum A2.).

Plateau observed 2 Muscids (Calliphora vomitoria Z.; Scatophaga sp.) in the
Ghent Botanic Garden.

UMBELLIFERAE 473

325. Pleurospermum Hofim.

1105. P. austriacum Hoffm.—Schulz (‘ Beitrage,’ II, p. 90) has observed this
species in the Riesengebirge, bearing only markedly protandrous hermaphrodite
flowers. Self-pollination is rare, occurring exceptionally when the stamens happen

to be still upright and some pollen remains in their anthers at the time when the
stigmas mature.

326. Cicuta L.

106. C. virosa L.—Warnstorf states that the umbellules of the primary
umbels bear hermaphrodite flowers only; those of the secondary umbels marginal
hermaphrodite and central male flowers; and those of tertiary umbels male flowers.

Visitors.—I noticed a Syrphid (Eristalis tenax Z.) and a Muscid (Lucilia
caesar Z.), both skg. (‘ Bloemenbiol. Bijdragen’).

In Dumfriesshire a humble-bee (freq.), a wasp (freq.), a fossorial wasp, and
numerous flies were recorded (Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 75).

327. Apium L.

1107. A. graveolens L. Kirchner states that the small whitish flowers of this
species are self-fertile, perhaps in consequence of imperfect protandry.

Vistrors.—I have observed a Muscid (Scatophaga sp.) and 2 Syrphids (Syritta
pipiens Z.; Syrphus sp.).

328. Petroselinum Hoffm.

1108. P. sativum Hoffm. Schulz says that the yellowish-green flowers of this
species are distinctly protandrous at Bozen. Henslow describes them as homogamous.
Schulz also observed andromonoecism. According to Warnstorf, the umbellules of
the primary umbels bear hermaphrodite flowers; those of the secondary and tertiary
ones, marginal hermaphrodite and central male flowers.

Visitors.—Herm. Miller (H. M.) at Lippstadt, and Buddeberg (Budd.) in Nassau
observed the following.—

A. Diptera. (a) Aluscidae: 1. Cyrtoneura simplex Loew (H.M.); 2. Lucilia
cornicina /. (H.M.); 3. Sarcophaga carnaria Z.(H.M.). (4) Syrphidae: 4. Chei-
losia sp. (Budd.); 5. Eristalis arbustorum Z. (H.M.); 6. E. sepulcralis Z. (H. M.);
7. Helophilus floreus Z. (H.M.); 8. Syritta pipiens Z. (H.M.); 9. Xanthogramma
citrofasciata Deg.(H.M.). B. Hymenoptera. (a) Afzdae: 10. Andrena minutula
XK. 9 (Budd.) ; 11. A. parvula X. 9 (Budd.); 12. Halictus morio F. 9, nect-lkg. (H.M.);
13. H. nitidus Schenck 9, do. (H. M.); 14. Prosopis communis /Vy/. (Budd.); 15. P.
sinuata Schenck 9, nect-lkg. (H. M., Budd.) ; 16. Sphecodes gibbus Z. 9 and 4 (H. M.,
Budd.); 17. Stelis breviuscula (Vy/. 8 (Budd.). (6) Chaletdidae: 18. Leucaspis dor-
sigera /., nect-lkg. (Budd.). (c) Zvanitdae: 19. Foenus sp. (Budd.). (d) Sphegidae :
20. Crabro clypeatus Schred. 9, nect-lkg. (Budd.). (¢) Vespedae: 21. Odynerus parietum
L. & (Budd.); 22. Polistes gallica Z., nect-lkg. (Budd.).

Schletterer records the bee Halictus levis K. for the Tyrol; and Plateau the
honey-bee, a Syrphid (Eristalis tenax Z.), and 2 Muscids (Musca domestica Z. ;
Lucilia caesar Z.) for Belgium.

474 ANGIOSPERMAE—DICOTYLEDONES

329. Trinia Hoffm.

1109. T. glauca Reichb. (=T. vulgaris DC.). (Henslow, ‘Origin of Floral
Structures,’ p. 227; Schulz, ‘ Beitrage,’ II, pp. go, 91, 189.)—Henslow and Schulz
describe this species as frequently dioecious, though undoubtedly androdioecious at
times. The latter states that female flowers sometimes replace hermaphrodite ones.

330. Helosciadium Koch.

u10. H. inundatum Koch (= Apium inundatum Rech). f.). (Knuth, ‘ Bl. u.
Insekt. a. d. nordfr. Ins.,’ p. 78.)—The flowers of this species are only 2 mm. in
diameter. The plants observed by me in the island of Fohr can be effectively
self-pollinated, for protandry is ill marked.

111. H. nodiflorum Koch (= Apium nodiflorum Reich. /.).—The flowers of
this species are distinctly protandrous. Secretion of nectar stops when the anthers
shrivel and begins again when the stigmas mature. It ceases altogether when the
petals fall off.

Visirors.—MacLeod saw 2 Muscids and a Neuropterid in Flanders (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 259-61).

331. Falcaria Riv.

m2. F. vulgaris Bernh. (=F. Rivini Hos/, F. sioides Aschers., and Sium
Falearia Z.). (Schulz, ‘ Beitrage, II, p. 190.)—Schulz describes this species as
andromonoecious, with markedly protandrous hermaphrodite flowers. The primary
umbels usually contain hermaphrodite flowers only; secondary ones now and then
bear 1-3 central male flowers, which owing to their position are the first to develop.
The rather small tertiary umbels flower late and are purely male.

Warnstorf describes the umbellules of the primary umbels as hermaphrodite ;
those of the secondary ones as bearing marginal hermaphrodite and central male

flowers, or male flowers only. It rarely happens that the umbels bear hermaphrodite
flowers.

VisiToRs.—Warnstorf observed flies and beetles.

332. Ammi L.

113. A. majus L. Schulz (‘ Beitrage’) describes this species as being andro-
monoecious, with protandrous hermaphrodite flowers of white colour.

Vistrors.—Schletterer observed an Ichneumonid (Trachynotus foliator 7.) and
a Scoliid (Tiphia minuta v. d. Z.) at Pola.

333. Aegopodium Knaut.

11144. A. Podagraria L. (Herm. Miiller, ‘Fertilisation,’ pp. 266-7, ‘ Weit.
Beob., I, p. 303, ‘Alpenblumen,’ p. 116; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
pp. 78, 155; MacLeod, Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 211-54; Loew,
‘ Bliitenbiol. Floristik,’ pp. 388, 292.)—Warnstorf says that at Ruppin the umbellules
of the primary umbels bear hermaphrodite flowers, those of the secondary umbels

UMBELLIFERAE +75

marginal hermaphrodite and central male ones. MacLeod states that in Flanders the
flowers of primary umbels are hermaphrodite (or in weaker plants hermaphrodite and
male), those belonging to umbels of other orders hermaphrodite on strong plants, and
male on weaker ones.

Visirors.—I observed 4 hover-flies, 3 Muscids, and a Lepidopterid in the island
of Fohr.—

Loew observed the following in Brunswick (B.), Silesia (S.), the Harz region (H:)
and the Riesengebirge (R.) (‘ Beitrage,’ p. 46).—

A. Coleoptera. (a) Cerambycidae: 1. Callidium violaceum Z. (S.); 2. Gs
livida F. Sor 3. L. sanguinolenta F. (S.); 4. Pachyta octomaculata /” (S.) ;
virginea Z. (S.); 6. P. quadrimaculata Z. (S.); 7. Strangalia arcuata Pz. Fy
(8) Ocdemer idae : 8. Chrysanthia viridis Schmidt (S.); 9. Oedemera virescens JL. (S.).
(c) Scarabaerdae: 10. Hoplia philanthus Sz/z. (B.). (2) Telephoridae: 11. Dasytes
niger F., nect-lkg. (S.);_ 12. Dictyoptera rubens Cy//. (S.); 13. Cantharis melanura
F.(S.). B. Diptera. (a) Asilidae: 14. Dioctria flavipes Mg. (B.). (4) Conoprdae :
15. Conops quadrifasciatus Deg. (B.); 16. Sicus ferrugineus Z. (B.). (¢) Muscidae:
17. Echinomyia grossa LZ. (B.); 18. Graphomyia maculata Scop. (B.); 19. Lasiops
apicalis AZg. (B.); 20. Macquartia chalybeata A@g. (B.); 21. M. nitida Zeét. (B.);
22. Nemoraea erythura J/g. (B.); 23. N. pellucida Ag. (B.); 24. Siphona cristata
Fabr. (B.); 25. Zophomyia tremula Scop. (B.). (d) Pipunculidae: 26. Pipunculus
tufipes JZg.(S.). (e) Stratiomyidae: 27. Chrysomyia formosa Scop. (B.); 28. Odon-
tomyia hydroleon Z. (B.); 29. Sargus infuscatus AZg. (B.). (/) Syrphidae: 30.
Brachyopa ferruginea fad. (S.); 31. Cheilosia variabilis AZg. (B.); 32. Chrysochlamys
ee Scop. (B.); 33. Chrysogaster coemeteriorum Z.(B.); 34. Chrysotoxum festivum

L.(B.); 35. Pipiza geniculata J/g. (B.); 36. Platycheirus albimanus /. (S.); 37.
Syrphus balteatus Deg. (B., 5.); 38 S. corollae /. (B.); 39. 5. glaucius Z. (B.);
40. S. grossulariae Mg. (S.) ; ql. S. laternarius Af//. (B.); 42. 5. lineola Zeds. (B.) ;

43. 5. pyrastri Z. (B.); 44. S. ‘bedi ZL. (B.); 45. Volucella inanis Z. (R.);
40. V. pellucens Z. (B.). C. Hemiptera. 47. Graphosoma lineatum JZ. (B.).
D. Hymenoptera. (a) Sphegidae: 48. Cerceris arenaria Z. 4 (S.); 49. Crabro
cribrarius Z. 6 (R.); 50. Gorytes mystaceus /. (B.); 51. Mutilla melanocephala 7.
& (S.); 52. Passaloecus corniger Schuck. (H.). (0) Zenthredinidae: 53. Allantus
bicinctus /. (B.); 54. Dolerus pratensis /. (B.); 55. Eriocampa ovata Z. (B.);
56. Hylotoma ustulata Z. (B.); 57. Tenthredo flava Poda (B.); 58. T. livida L.
(B.). (ce) Vespidae: 59. Eumenes coarctata Z., skg. (B.); 60. Polistes gallica Z.(R.);
61. Vespa austriaca Pz. 9, skg.(S.); 62. Odynerus gracilis Bradlé (S.); 63. O. sinua-
tus &., skg. (B.). E. Lepidoptera. Rhopalocera. 64. Argynnis paphia Z. (R.).
Also in Mecklenburg (‘ Beitrage,’ p. 35).—A. Coleoptera. (a) TZelephoridae:
1. Cantharis fulvicollis /.; 2. C. nigricans Afil/. (6) Oedemeridae: 3. Oedemera
podagrariae Z. B. Diptera. (a) Muscidae: 4. Graphomyia maculata Scop., skg.
(6) Syrphidae: 5. Tropidia milesiformis Fa//, skg. C. Hymenoptera. Sphegidae:
6. Crabro peltarius Schred. 9.

Alfken noticed the following at Bremen.—

A. Diptera. Syrphidac: 1. Helophilus floreus Z., very common. B. Hy-
menoptera. (a) Apidac: 2. Andrena nitida Fourcr. 9, rare; 3. A. parvula X. 9,
rare; 4. A. proxima A’, 9, not infrequent; 5. Prosopis communis WVy/. 9 and 4,
freq.; 6. P. confusa Wy7. 8, nect-dvg. (4) Yenthredinddae: 7. Hemichroa alni L.
C. Coleoptera. Scarabaeidae: 8. Cetonia aurata Z., freq.; 9. Guorinnus nobilis
L., rare.

I saw 3 Muscids (Coelopa frigida /a//., skg. and po-dvg.; Fucellia fucorum

476 ANGIOSPERMAE—DICOTYLEDONES

Fall., do.; Lucilia caesar Z., do.) in Helgoland (Bot. Jaarb. Dodonaea, Ghent, viii,
1896, p. 34).
Sickmann observed the following at Osnabrtick.—

Hymenoptera. (a) Mutillidae: 1. Methoca ichneumonides Zir. a 6;
2. Mutilla melanocephala #.. (0) Scolitdae: 3. Tiphia minuta v. d. Z., not freq.
(c) Sphegidae: 4. Agenia hyrcana F, not freq.; 5. Ammophila hirsuta Scop., not
rare; 6. Ceropales maculatus #., freq.; 7. Crabro albilabris #., very common ;
8. C. brevis v. d. Z., do.; 9. C. chrysostoma Lep., freq.; 10. C. clavipes £. Dahlo.,
not freq.; 11. C. fuscitasis A. Sch., rare; 12. C. lituratus Pz., do.; 13. C. peltarius
Schreb., very common; 14. C. planifrons Zhoms., very rare; 15. C. podagricus
7. d. L., freq.; 16. C. scutellatus Schev., fairly freq.; 17. C. sexcinctus /., freq. ;
18. C. spinicollis #. Sch., freq.; 19. C. vagabundus Pz., do.; 20. C. vagus L.,
do.; 21. C. wesmaéli v. d. Z., not freq.; 22. Dinetus gultatus #., very common ;
23. Dolichurus corniculus Spzv., rare; 24. Gorytes campestris J7/ull., do.; 25. G.
mystaceus Z., freq.; 26. G. quadrifasciatus /., do.; 27. Miscophus bicolor /ur., not
infrequent; 28. Nysson maculatus /., fairly freq.; 29. N. spinosus /orst., do.;
30. Oxybelus uniglumis Z., freq.; 31. Passaloecus brevicornis A. Mor., rare ;
32. Pompilus nigerrimus Scop., freg.; 33. P. spissus Schyédie, freq.; 34. P. gibbus
#., very common; 35. Psen atratus Pz., freq.; 36. P. concolor DaAlé., infrequent.;
37. Pseudagenia carbonaria Scopf., very common; 38. Salius hyalinatus /., freq. ;
39. S. notatus Rossz, do.; 40. S. sepicola Sm., do.; 41. Trypoxylon attenuatum
Sm., rare; 42. T. clavicerum Lep.; 43. T. figulus Z., very common.

The following were recorded by the observers, and for the localities stated.—

Schmiedeknecht (Thuringia) the bee Andrena combinata Chr. Krieger (Leipzig),
4 bees (Colletes daviesanus K.; Prosopis communis /Vy/.; P. hyalinata Sm.; P. pictipes
WVyl.). MacLeod (Flanders) Apis, 5 short-tongued bees, 28 other short-tongued
Hymenoptera, 12 hover-flies, 16 other flies, 12 beetles, a Lepidopterid, and a
Neuropterid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 262-4). H. de Vries
(Netherlands) the bee Andrena trimmerana A. 9 (Ned. Kruidk., Arch., Nijmegen,
2 Ser., 2 Deel, 1875). Scott-Elliot (Dumfriesshire) a fossorial wasp, a Vespid,
2 Ichneumonids and 2 Muscids (‘Flora of Dumfriesshire,’ p. 76).

Herm. Miiller (H. M.), Buddeberg (Budd.), and Borgstette (Borg.) give the
following list.—

A. Coleoptera. (a) Cerambycidae: 1. Grammoptera ruficornis /. (H. M.);
2. Leptura livida #. (H. M.); 3. Pachyta octomaculata /. (Borg.). (6) Cistelidae:
4. Cistela murina Z. (H. M.). (c) Curculionidae: 5. Spermophagus cardui Svv.
(H. M.). (d) Cleridae: 6. Trichodes apiarius Z. (H. M.). (e) Dermestidae :
7. Anthrenus pimpinellae /. (H.M.). (/) laterrdae: 8. Agriotes aterrimus Z.
(H.M.); 9. Athous niger Z. (H. M.); to, Lacon murinus Z.(H.M.). (g) Mordelli-
dae: 11. Anaspis rufilabris Gy/?. (H. M.); 12. A. frontalis Z. (H. M.); 13. Mordella
aculeata Z., very common (H.M.); 14. M. fasciata #. (H. M.). (2) Nitdulidae :
15. Cychramus luteus O/v.(H.M., Borg.). (2) Oedemeridae: 16. Oedemera virescens
£.(H.M.). (&) Scarabaetdae: 17. Cetonia aurata Z.(H. M.); 18. Hoplia argentea
Poda (H. M., in the Alps); 19. Phyllopertha horticola Z. (H.M.); 20. Trichius
fasciatus Z.(H.M.). (2) Zelephoridae: 21. Dasytes flavipes #.(H.M.); 22. Malachius
bipustulatus Z. (H. M.); 23. Telephorus fuscus Z.(H. M.). (m) Zrixagidae: 24.
Trixagus fumatus /.(H.M.). B. Diptera. (a) Bombylidae: 25. Anthrax flava AZ.
(H.M.). (6) Dolichopodidae: 26. Gymnopternus chaerophylli AZ.(H.M.). (c) Zm-
pidae: 27. Empis livida Z. (H. M.); 28. E. punctata #. (H. M.). (@) Muscidae:
29. Species of Anthomyia (H. M.); 30. Aricia obscurata A/g. (H. M.); 31. Echino-
myia fera Z.(H. M.); 32. Lucilia cornicina F. (H. M.); 33. L. sylvarum J/g. (H.M.);
34. Musca corvina /. (H.M.); 35. Sarcophaga albiceps 4/g.(H.M.); 36. Scatophaga

UMBELLIFERAE 477

stercoraria Z.(H.M.); 37.S.merdaria #.(H.M.); 38. Sepsis,freq.(H.M.) ; 39. Zopho-
myia tremula Scop.(H.M.). (e) Stratiomyzdae: 40. Chrysomyia formosa Scop. (H.M.);
41. Sargus cuprarius Z. (H.M.); 42. Stratiomys chamaeleon Deg. (H.M.). (f) Syr-
phidae: 43. Chrysogaster chalybeata A7g.(H.M.); 44. C. coemeteriorum Z. (H.M.);
45. C. viduata Z.(H.M.); 46. Eristalis arbustorum Z.(H.M.); 47. E. nemorum Z.
(H. M.); 48. E. tenax Z. (H. M.); 49. Helophilus floreus Z., freq. (H. M.);
50. Melithreptus taeniatus 17g. (H.M.) ; 51. Pipizella virens #.(H.M.); 52. Syritta
pipiens Z., freq. (H. M.); 53. Syrphus nitidicollis M/g.(H.M.); 54. S. pyrastri Z.
(H.M.); 55. S. ribesii Z.(H.M.); 56. Volucella pellucens Z.(Borg.). (g) Therevidae :
57. Thereva anilis Z.(H.M.). (4) Zipulidae: 58. Pachyrhina crocata LZ. (H. M.);
59. P. histrio #. (H.M.). C. Hymenoptera. (a) Apzdae: 60. Andrena albicans
Mull, skg. (H. M.); 61. A. albicrus K. 9 (H. M.); 62. A. dorsata A. 9, po-cltg.
(H. M.); 63. A. fucata Sm. 9, skg. (H. M.); 64. A. fulvago Chr. 9, po-cltg. (H. M.);
65. A. helvola Z. 9 and 4, po-cltg. and skg. (H. M.); 66. A. parvula A. 9 and é (H. M.);
67. A. pilipes #. 6, skg. (H. M.); 68. A. proxima X. 9, skg. and po-cltg. (H. M.);
69. Apis mellifica Z. ¥, do. (H. M.); 70. Halictus albipes F. 9, skg. (H. M.);
71. H. cylindricus F. 9, do. (H. M.); 72. H. minutus A. 9, do. (H. M.); 73. Prosopis
clypearis Schenck 6, do.(H.M.); 74. P.communis Vyl.é, do. (H.M.). (0) Chrystdidae :
75. Hedychrum lucidulum /. 4, in great numbers (H. M.). (c) Evanzidae: 76. Foenus
affectator #. (H. M.); 77. F. jaculator /.(H.M.). (d) Lchneumonrdae : 78. Numerous
sp. (H. M.). (e) Sphegidae: 79. Cerceris variabilis Schr. 9 and 6, rare (H. M.);
80. Crabro interruptefasciatus Re/z. $ (H. M.); 81. C. lapidarius Pz. 9 (H. M.);
82. C. sexcinctus #. 6 (H. M.); 83. C. vagus Z. 9 (H. M.); 84. Gorytes campestris
Atul. 9 and , not infrequent (H. M.); 85. G. laticinctus Schuck. 9 (H. M.);
86. Myrmosa melanocephala /. 9 (H. M.); 87. Oxybelus lineatus /. é, numerous ;
88. O. bellus Dahld. $ (H. M.); 89. O. bipunctatus O7. §; go. O. uniglumis L., very
numerous (H. M.); 91. Philanthus triangulum /. (H.M.). (e) Pomprlidae: g2.
Pompilus minutus Dahld.9 (H.M.); 93. P. nigerrimus Scop. 9 (Borg.); 94. P. spissus
Schjodie 9 (H.M.). (f°) Tenthredinidae: all nect-lkg.: 95. Abia sericea Z. (H. M.);
96. Allantus arcuatus /ors?,, freq.(H.M.); 97. Hylotoma caeruleopennis Re/z. (H. M.) ;
98. H. melanochroa Gme/. (H. M.); 99. H. rosae Z. (H. M.); roo. H. ustulata Z.
(H.M.); ror. Selandria serva /., freq. (H. M.); 102. Tenthredo atra Z. (Budd.) ;
103. T. bifasciata K7/g. (= Allantus rossii Pz.) (H.M.); 104. T. flavicornis #. (H. M.);
105. T. sp. (H.M.). (g) Vespidae: 106. Odynerus elegans Wsm. 9, nect-lkg. (H. M.);
107. O. quinquefasciatus /.9, do. D. Lepidoptera. Rhopalocera: 108. Pieris napi
L., skg.(H.M.). E. Neuroptera. 109. Panorpa communis Z. (H. M.).

115. A. alpestre Ledb.—

Visirors.—Loew observed a Muscid (Chloria demandata /.) and a bee (Andrena
fasciata, Wesm. 9), skg. and po-cltg., in the Berlin Botanic Garden.

334. Carum Rupp.

m6. C. Carvi L.—Beketow (Justs bot. Jahresber., Cassel, xviii (1890), 1892,
p- 464) points out that protandry in this species is so marked that the primary umbel
is purely female at the time when the flowers of the lateral ones are in the
male stage.

Warnstorf (Schr. natw. Ver., Weringerode, xi, 1896) states that the primary
umbel bears either hermaphrodite flowers, or ones that have become purely female
by the degeneration of the pollen-grains in the white anthers, borne on relatively
short filaments. Should the primary umbel be purely female, the others are
usually hermaphrodite; though an entire stock may often be female, as a result
of degeneration of the anthers. At Ruppin, therefore, the species is gynodioecious.

478 ANGIOSPERMAE—DICOTYLEDONES

The pollen-grains are white, biscuit-shaped, somewhat contracted in the middle,
with 3 longitudinal furrows, about 30 » long and 12 » broad.

VistTors.—I observed (Helgoland, 5. 6. ’97) a Muscid (Lucilia caesar Z.),
Syrphids (sp. of Eristalis and Syrphus), and beetles (Cantharis).

Lindman noticed ripe fruits on the Dovrefjeld, and records several species of
flies and a bee, as visitors.

Herm. Miller observed the following (‘ Fertilisation,’ pp. 275-6; ‘ Weit. Beob..,’
I, p. 304).—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia atra Lazch., nect-lkg.
(6) Chrysomelidae: 2. Crioceris duodecimpunctata Z. (c) Curculionidae: 3. Bruchus
sp., numerous; 4. Phyllobius oblongus Z. (d@) Zelephoridae: 5. Anthocomus fasciatus
L.; 6. Dasytes flavipes /., nect-lkg.; 7. Malachius bipustulatus Z.; 8. Telephorus
fuscus Z., nect-lkg.; 9. T. lividus Z., do.; 10. T. pellucidus /, do.; 11. T. rusticus
Fall. (e) Alordellidae, all nect-lkg.: 12. Anaspis rufilabris Gy//.; 13. Mordella
pumila Gy//.; 14. M. pusilla De. (f) Staphylinidae: 15. Tachinus fimetarius Grv.,
nect-lkg.; 16. Tachyporus solutus Zr.,do. B. Diptera. (a) Bibconidae: 17. Bibio
hortulanus Z. (4) Empidae: 18. Empis stercorea £., skg. (c) Muscidae: 19. Aricia
incana Wredem.; 20. Cyrtoneura hortorum Fa//. 9; 21. Echinomyia fera Z.; 22.
Gymnosoma rotundata Z.; 23. Species of Lucilia; 24. Pyrellia aenea Zeé/.;
25. Sarcophaga carnaria Z.; 26. S. albiceps A/g.; 27. Scatophaga merdaria /” ;
28. Zophomyia tremula Scop. (d@) S/ratiomyidae: 29. Chrysomyia formosa Scop. ;
30. Stratiomys longicornis Scop. (e) Syrphidae: 31. Chrysotoxum festivum JZ. ;
32. Eristalis aeneus Scop.; 33. E. arbustorum Z.; 34. E. horticola Deg.; 35.
Helophilus floreus Z., very common; 36. H. pendulus Z.; 37. Melanostoma mellina
L.; 38. Melithreptus taeniatus A/g.; 39. Pipizella virens /.; 40. Platycheirus pel-
tatus AZg.; 41. Pyrophaena sp., skg.; 42. Syritta pipiens Z.; 43. Syrphus ribesii Z.,
skg. (f) Tipulidae: 44. Tipula sp., nect-lkg. C. Hemiptera. 45. One small
Capsus. D. Hymenoptera. (a) Afzdae: all skg.: 46. Andrena albicans Afi//.
9 and 4, skg.; 47. A. fulvicrus A. 9; 48. A. minutula A. 9; 49. A. nana XK. 6;
50. A. nigroaenea A.9; 51. A. parvula A“, skg.and po-cltg.; 52. Halictus albipes
F. 9, po-clig.; 53. H. maculatus Sm. 9, skg., repeatedly; 54. H. sexnotatus
A. 9, po-cltg.; 55. Prosopis brevicornis My/. 6; 56. P. communis JVyl. 6.
(4) Formicidae: 57. Formica fusca £. ¥; 58. Lasius niger Z. ¥; 59. Myrmica
rugulosa Wyl. ¥; 60. M. levinodis My/. ¥. (c) Ichneumonidac: 61. Numerous
species. (d) Pleromalidae: 62. An undetermined sp., nect-lkg. (e€) Sphegidae :
63. Pemphredon unicolor /., several; 64. Crabro lapidarius Pz. 9; 65. C. scu-
tellatus Schev. 6; 66. C. vagabundus Ps. 9; 67. Gorytes campestris AZzi/. 3.
(S) Tenthredinidae: 68. Athalia spinarum /.; 69. Cephus niger Harr.; 70. Abia
sericea L.; 71. Dolerus pratensis L.; 72. Hylotoma caerulescens /.; 73. H. enodis
L.; 74. H. femoralis Alve.; 75. H. rosarum Al/g.; 76. Selandria serva F.; 77.
Tenthredo bifasciata K7g.; 78. Allantus vespa Re/z. E. Lepidoptera. (a) Zinecdac:
79. Adela sp., skg. F. Neuroptera. Planipennia: 80. Sialis lutaria Z.

Verhoeff observed the following in Norderney.—

A. Diptera. (a) Zmpidae: 1. Hilara quadrivittata Afg. (6) Dolichopodidae :
2. Dolichopus aeneus Deg. (c) Muscidae: 3. Anthomyia sp. 6 (2); 4. A. sp., freq.;
5. A. triquetra Wredem. one é; 6. Aricia incana Wiedem.; 7. A. obscurata AZg. one 3 ;
8. Cyrtoneura hortorum /a//. one g and one $; 9. Hylemyia conica Wiedem.one $; 10.
Limnophora quadrimaculata Fa//. 9 andé; 11. Lucilia caesar L., freq.; 12. Myospila
meditabunda F.; 13. Onesia floralis R.-D.; 14. Psila villosula Mg.; 15. Sarcophaga
sp. 9; 16. Scatophaga stercoraria Z. one 9. (@) Syrphidac: 17. Pipizella virens F,

UMBELLIFERAE 479

one 6; 18, Syritta pipiens Z. one 9. (e) Zherevvdae: 19. Thereva anilis Z. one 6.

(f) Tipuledae : 20. Pachyrhina scurra A/g.one 9. B. Hymenoptera. Zenthredinidae :
21. Nematus sp.

The following were recorded by the observers, and for the localities stated.—

Alfken (Bremen) 6 Tenthredinids.—1. Allantus temulus Scog.; 2. Arge enodis Z.;
3. A. ustulata Z.; 4. Dolerus fissus W/g.; 5. Macrophya quadrimaculata #.; 6. Pachy-
protasis rapae Z. Herm. Miiller (Alps) 7 Diptera, 4 Hymenoptera, and 6 Lepidoptera
(‘Alpenblumen,’ p. 116). Schletterer and von Dalla Torre (Tyrol) the bee Prosopis
borealis Vy/. Kohl the fossorial wasp Crabro scutellatus Schev. MacLeod (Pyrenees)
a bee and 2 flies (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 413-14). Scott-Elliot
(Dumfriesshire) a saw-fly, 2 Ichneumonids, 5 Muscids, and several Dolichopodids
(‘Flora of Dumfriesshire,’ p. 76).

335. Pimpinella L.

117. P. magna L. (Schulz, ‘ Beitrage, I, p. 43, I, pp. 82-4, 91, 190.)—
Schulz, for Germany and the Tyrol, describes this species as andromonoecious, with
protandrous hermaphrodite flowers. Gelmi and Schulz, in the South Tyrol, observed
a purely female form with partly or completely degenerate pollen: the central flowers

Fic. 162. Prapinella magna, L., var. B rosea, Koch (after Herm. Miller), .4. Flower in the first
(male) stage. #8. Pistil of the same. C. Flower in the second (female) stage. D. Pistilof the same.
(=< 7.) 1-5, stamens; 7, nectary; ov, ovary; s/, stigma.

of the umbels were neuter, instead of being male as usual. Gelmi states that the
styles are always shorter than the ovary in the hermaphrodite flowers, but longer in the
purely female ones. The same is true for P. saxifraga. At Ruppin, according to
Warnstorf, the umbellules of primary umbels bear hermaphrodite flowers; those of
secondary umbels marginal hermaphrodite and central male ones; tertiary umbels
male flowers only.

Visirors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Central Germany), 2 bees (Andrena parvula A. 9, skg. and
po-cltg.; A. rosae Pz. 9, skg.) (‘ Fertilisation, p. 277); in the Alps a Cerambycid
beetle (Pachyta quadrimaculata Z.) (‘Alpenblumen,’ p. 116). Sickmann (Osnabriick),
2 fossorial wasps (Crabro dives Zef., rare; Mellinus sabulosus &.). MacLeod
(Pyrenees), a beetle (Bot Jaarb. Dodonaea, Ghent, iii, 1891, p. 413).

In the sub-Alpine region this species usually bears rose-red flowers (var. 8 rosea
Koch, = the species Pimpinella rubra /Zopfe), and the styles and stigmas appear to be
developed to such an extent in the first (male) stage, that they might be regarded as
mature, but as a matter of fact they do not attain their full size till later.

480 ANGIOSPERMAE—DICOTYLEDONES

Visttors.—The following were recorded by the observers and for the localities
stated.—

Herm. Miiller, 7 flies, 2 Hymenoptera, and a Lepidopterid. von Dalla Torre and
Schletterer (Tyrol), 3 bees (Halictus major Wy/.; H. tetrazonius K/ug.; H. zonulus
Sm.). MacLeod (Pyrenees), a saw-fly, a beetle, and 6 flies (loc. cit.).

1118. P. saxifraga L. (Loew, ‘ Bliitenbiol. Floristik,’ pp. 379, 389, 393 ; Herm.
Miiller, ‘Fertilisation,’ pp. 277-8, ‘ Weit. Beob.,’ I, p. 304; Knuth, ‘BI. u. Insekt. a. d.
nordfr. Ins., p. 155 ; Schulz, ‘ Beitrage,’ I, p. 44, 11, pp. 84, 91, 190.)—Both Gelmi and
Schulz state that this species resembles the chief form of P. magnum as regards flower
mechanism. It is mostly andromonoecious, but now and then gynodioecious. (Cf.
the last species.)

At Ruppin, according to Warnstorf, the umbellules of primary umbels bear
hermaphrodite flowers; those of the secondary umbels marginal hermaphrodite
and central male ones.

Visirors.—Herm. Miiller in Westphalia and Buddeberg (Budd.) in Nassau
observed the following.—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida /., nect-lkg.; 2. Pachyta
octomaculata F", freq., licking sap. (6) Coccinellidae: 3. Coccinella septempunctata
£., creeping over the flowers. (c) Chrysomelidae: 4. Clytra scopolina Z. (d)
Telephoridae: 5. Dasytes flavipes /.; 6. Telephorus melanurus A. B. Diptera.
(a) Astlidae: 7. Isopogon brevirostris AZg. (6) Conopidae: 8. Conops quadrifasciata
Deg. (c) Syrphidae: 9g. Eristalis horticola A/g.; 10, Syrphus nitidicollis J/g. ;
11. S. pyrastn Z. (d) Tabanidae: 12. Chrysops caecutiens Z.; 13. Tabanus nicans
fg. ©. Hymenoptera, (a) Af:dae: 14. Andrena fulvescens Sm. 9; 15. A.
parvula A., skg. and po-cltg.; 16. Sphecodes gibbus Z., skg. (06) Jchneumonidae :
numerous sps. (c) TZenthredinidae: 17. Abia sericea Z.; 18. Allantus arcuatus
frorst., freq. (H. M., Budd.); 19. A. temulus Scop.; 20. Hylotoma rosae Z.;
21, Selandria serva /. D. Neuroptera. Plannipennia: 22. Panorpa communis Z.

The following were recorded by the observers, and for the localities stated.—

Lindman (Dovrefjeld), a saw-fly. Schletterer (Tyrol), the Scoliid Tiphia femo-
rata F. Alfken (Bremen):—A. Diptera. Syrphidae: 1. Chrysotoxum festivum L.
B. Hymenoptera. (a) Apidae: 2. Prosopis communis Wy/. 9 and 6. (4) Sphegidae :
3. Ceropales maculatus /#., very common; 4. Crabro brevis v. d. Z. 9 and 4.
(¢) Lenthredinidae: 5. Allantus arcuatus Forst.; 6. Athalia glabricollis /hs. (Juist),
A. Coleoptera. Alleculidae: 1. Cteniopus sulphureus L., freq., po-dvg. B. Diptera.
(a) Muscidae: 2. Lucilia caesar Z.; 3. Nemoraea radicum F,, very common, po-dve. ;
4. Sarcophaga albiceps J/g. (6) Syrphidae: 5. Eristalis tenax Z. Sickmann
(Osnabriick): Hymenoptera. Sphegidae: 1. Crabro lituratus Pz.; 2. Gorytes
tumidus Pz., infrequent: 3. Mellinus sabulosus #., rare. MacLeod (Flanders) 12
short-tongued Hymenoptera, 10 hover-flies, 14 other flies, and Panorpa (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 265-6); (Pyrenees) 8 short-tongued Hymenoptera,
4 beetles, and ro flies (op. cit., iii, 1891, pp. 412, 414). Scott-Elliot (Dumfriesshire),
a saw-fly and numerous flies (‘ Flora of Dumfriesshire,’ p. 47).

I observed the following in Thuringia (‘ Bliitenbiol. Beob. in Thiiringen,
P: 32).—
A. Coleoptera. 1. Judolia cerambyciformis Schr., freq.; 2. Leptura livida F.;
. Strangalia melanura Z.; 4. Trichius fasciatus Z. B. Diptera. (a) Muscidae:
5. Aricia serva AZg. (6) Syrphidae: 6. Syrphus lineola Zef//.; 7. Volucella pellucens Z.
C. Lepidoptera. 8. Zygaena pilosellae L's.

vive

UMBELLIFERAE 481
Loew (Riesengebirge) (‘ Beitrage,’ p. 48).—

A. Coleoptera. Cerambycidae: 1. Strangalia nigra Z. B. Diptera. (c)
‘Muscidae: 2. Meigenia floralis Ag. (2) Syrphidae: 3. Cheilosia oestracea LZ. ;
4. Eristalis rupium #. C. Hymenoptera. /chneumonidae: 5. Undetermined sp.
(Mecklenburg) (op. cit., p. 37): Hymenoptera. (a) Apidae: 1. Prosopis annu-
laris Sm. 9; 2.P. sp. (4) Chrystdidae: 3. Chrysis saussurei Chevr. (c) Vespidae:
4. Odynerus spinipes Z.; 5. Pterocheilus phaleratus Pz.

Willis records the following for the neighbourhood of the south coast of Scotland
(‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera. (a) Telephoridae: 1. Rhagonycha fulva Scop., skg. (0) Mete-
dulidae : 2. Epuraea melina Zr., skg.; 3. Meligethes sp.,do. B. Diptera. (a) Lmprdae:
4. Ramphomyia tenuirostris /a//., skg. (6) Muscidae: allskg.: 5. Anthomyia radicum
L., freq.; 6. A. sp.; 7. Hyetodesia incana W.; 8. Lucilia caesar Z.; 9. Morellia
curvipes Jfcg.; 10. Phorbia floccosa Afcg.; 11. Anthomyia radicum Z.; 12. Themira
minor /fal., freq.; 13. Lasiops cunctans Afg. (c) Phoridae: 14. Phora sp., skg.
(2) Syrphidae: all skg.: 15. Cheilosia sp.; 16. Chrysogaster splendida Jfg.;
17. Eristalis aeneus Svop., freg.; 18. E. horticola Deg., do.; 19. E. tenax Z.;
20. Orthoneura nobilis Fa//.; 21. Sphaerophoria scripta Z.; 22. Syritta pipiens Z. ;
23. Syrphus ribesii Z., freq. (¢) Zipulidae: 24. Boletina sp., skg.; 25. Sceptonia
nigra J/g., do. (/) Mycetophilidae: 26. Sciara sp., freq., skg. C. Hemiptera.
27. Anthocoris sp.,skg. D. Hymenoptera. Jchneumonidae: 28. Nine undetermined
sp. E. Lepidoptera. Rhopaloccra: 29. Pieris napi L., skg.

1119. P. peregrina L.—

Visitors.—Schletterer observed the following Hymenoptera at Pola.—

(2) LEvanitdae: 1. Gasteruption granulithorax Zourn. (6) Ichneumonidae:
2. Angitia armillata Gr.; 3. Linoceras macrobatas Gr., var. geniculata Archd. ;
4. Mesoleius cruralis Gr. (c) Zenthredintdae: 5. Cephus (Philoecus) parreyssini
Spin.

336. Berula Hoffm.

120. B. angustifolia Mert. et Koch (=Sium erectum Auds.).— All the umbels
of this species bear hermaphrodite flowers; some of those in the middle of the
umbellules frequently possess three styles (Warnstorf, Verh. bot. Ver., Berlin, xxxviii,
1896).

337. Sium L.

121. S. latifolium L. (Herm. Miiller, ‘ Fertilisation,’ p. 276, ‘ Weit. Beob.,’ J,
p- 304; Schulz, * Beitrage, I, p. 44, II, p. 190; Warnstorf, Verh. bot. Ver., Berlin,
xxxvili, 1896; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 325 ; Knuth, ‘Bliitenbiol. Beob.
a. d. Ins. Riigen,’ ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 155.)—Schulz describes this species
as andromonoecious, with markedly protandrous hermaphrodite flowers; the umbels
of higher order are chiefly or entirely male. According to Kerner, the hermaphrodite
flowers open first, and are distinctly protandrous; their stigmas are dusted, after the
anthers have dropped off, by the crumbling masses of pollen which fall from the male
flowers of the later developed lateral umbels. Geitonogamy is thus automatically
effected. Warnstorf says that at Ruppin the umbellules of the primary and secondary
umbels bear hermaphrodite flowers; those of tertiary umbels hermaphrodite and
central male ones, or male flowers only.

DAVIS. 1 11

482 ANGIOSPERMAE—DICOTYLEDONES

Visivors.—The following were recorded by the observers, and for the localities
stated.—

H. de Vries (Netherlands), the honey bee. MacLeod (Flanders), 3 short-tongued
Hymenoptera, 4 hover-flies, 4 other Diptera, and a beetle (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 267). Knuth (North Frisian Islands), 2 hover-flies and a Muscid :
(Riigen),—A. Coleoptera. TZelephoridae: 1. Cantharis fulva Scop. B. Diptera.
(a) Afuscidae: 2. Aricia sp.; 3. Graphomyia maculata Scop. 9. (¢) Straliomyzdae :
4. Odontomyia viridula /.; 5. Stratiomys furcata #. 9. C. Hymenoptera.
Sphegidae: 6. Gorytes quadrifasciatus #. 6. Alfken (Bremen), the Ichneumonid
Amblyteles laminatorius #. Kohl (Tyrol), the fossorial wasp Crabro scutellatus Schev.

Herm. Miiller gives the following list.—

A. Coleoptera. (a) Cocctnellidac: 1. Coccinella quatuordecimpunctata L.,
nect-lkg. (0) Mordellidae: 2. Mordella fasciata F. (c) Scarabaeidae: 3. Trichius
fasciatus LZ. (d) Lilephoridae: 4. Telephorus melanurus #. B. Diptera. (a) Dolr-
chopodrdae: 5. Dolichopus aeneus Deg. (4) Empidac: 6. Empis sp. (c) AMusctdae :
7. Aricia incana W7ed., numerous; 8. Calliphora vomitoria Z.; 9. Cyrtoneura simplex
Loew; 10. Lucilia caesar Z.; 13. L.cornicina /.; 12. L. sylvarum A/g.; 13. Mesem-
brina meridiana Z., skg.; 14. Musca corvina #.; 15. Ocyptera brassicaria 2” ;
16. Sepsis sp.; 17. Tetanocera ferruginea /a//.; 18. Tephritis pantherina /a//.,
nect-lkg., 2 individuals. (d@) Syrphrdae: 19. Eristalis aeneus Scop. ; 20. E. arbustorum
£.; 21. E. nemorum Z.; 22. Helophilus floreus Z.; 23. Syritta pipiens /.; 24.
Syrphus ribesii Z. (e) Stratiomyz¢dae: 25. Stratiomys riparia J/g. C. Hemiptera.
26. One small sp. of Anthocoris. D. Hymenoptera. (a) Apidae: 27. Prosopis
variegata /, nect-lkg. (4) Jchneumonidae: 28. Numerous species. (¢) Sphegtdae:
29. Crabro dives H.-Sch.d; 30. C. lapidarius Pz. é and 9, freq.; 31. C. scutellatus
Schev. 8; 32. C. vagus Z.d; 33. Gorytes quadrifasciatus #. 3, skg.; 34. Oxybelus
uniglumis Z., do. (d) Zenthredinidae: 35. Allantus arcuatus Fors?.; 36. Athalia
rosae 7.; 37. Selandria serva £”.

338. Conopodium Koch.
1122. C. denudatum Koch.—This is a white-flowered species.

Visirors.—MacLeod (Pyrenecs) observed 5 Hymenoptera, 2 beetles, 17 flies,
and 2 Lepidoptera.

339. Bupleurum L.

1123. B. stellatum L. (Herm. Miller,
‘ Alpenblumen,’ pp. 117-18.)—The protandry
of this species is so well marked that the
whole umbel is at first entirely male, after-
wards entirely female. Self-pollination is
therefore excluded.

Visttors.— Herm. Miller noticed 8 Diptera
and 3 Hymenoptera.

1124. B. ranunculoides L. (Kirchner.
FG. 163. Bupleurum stellatum (after Herm.

‘ Rees >
Miller). +f. Flower in the first (male) stage. Beitrage, Ys 3t-)—The orange-yellow blos
fi. Ditto, in the second (female) stage. (*7-) soms of this species, which later on turn to
a'-a", anthers; gv, style; 2, nectary ; ov, ovary ; z re
p, petal; sf, ae : sa a brownish-yellow colour, are distinctly pro-

tandrous at Zermatt. Only hermaphrodite flowers have been observed.

UMBELLIFERAE 483

1125. B. longifolium L.—Schulz (‘Beitrage,’ I, p. 46) states that in this species
the petals are completely involute in the bud (as in all other German species of the
genus), so that the nectaries are exposed. The anthers with their incurved filaments
are also uncovered in the bud. The petals usually remain in this condition during
anthesis. It is only after the anthers have dehisced that the styles develop, often very
slowly (as in other species), so that a considerable time often clapses between the
beginning of anthesis, and the accomplishment of pollination.

Only hermaphrodite flowers have been observed.

1126. B. tenuissimum L.—According to Schulz (loc. cit.) the flower
mechanism of this species agrees with that of B. longifolium.

1127. B. falcatum L.—As B. tenuissimum. The flowers are dull yellow in
colour.

Visirors.—Herm. Miiller observed the following in Thuringia.—

A.Coleoptera. A/ordellidae: 1. Mordella pumila CyZ/., nect-lkg., very numerous.
B. Diptera. (a) Bombyhidae: 2. Anthrax flava, A/g.,skg. (6) Muscidae: 3. Gymno-
soma rotundata Z., nect-lIkg., one. (c) Syrphidae: 4. Eristalis arbustorum, L., skg. ;
5. Pipizella annulata A/acg., do.; 6. Syritta pipiens /.., very numerous, skg. and
po-dve. C. Hymenoptera. (a) Apidae: 7. Halictus interruptus Pz., 6 skg.
(6) Ichneumonidae: 8. Various sp., skg. (c) Zenthredinidae: gy. Hylotoma rosae L.,
skg. (d) Vesprdae: 10. Polistes gallica Z., skg.; 11. P. biglumis Z., do.

MacLeod (Pyrenees) saw 2 Hymenoptera (Bot. Jaarb. Dodonaea, Ghent, ii,
1891, p. 412).

1128. B. rotundifolium L.—The flowers of this species are yellow, and their
nectar may be seen with the naked eye as a glistening surface.

Visttors.—Herm. Miiller gives the following list for Thuringia (‘ Weit. Beob.,’ I,
p- 304).—

A. Diptera. (a) Muscidae: 1. Sp.of Anthomyia; 2. Gymnosoma rotundata ZL. ;
3. Ulidia erythrophthalma .I/g., skg. (4) Strattomyidae: 4. Chrysomyia formosa Scop.
B. Coleoptera. Curculionidae: 5. Bruchus olivaceus Germ., nect-lkg.; 6. Spermo-
phagus cardui Szev., do. C. Hymenoptera. (a) Jchneumontdae: 7. Various sp.
(4) Tenthredinidae: 8. A yellow sp. (c) Sphegidae: g. Tiphia minuta vz. d@. L., skg.
D. Lepidoptera. ro. Lycaena bellargus Ro//., skg.

340. Oenanthe L.

u2g. O. fistulosa Z.—According to Schulz (‘ Beitrage, 1, pp. 47-8), this
species is andromonoecious, and the hermaphrodite flowers feebly protandrous.
Purely male stocks occur here and there. The male flowers are usually at the
margin, rarely in the middle of the umbellules.

Visitors.—Herm. Miiller gives the following list (‘ Fertilisation,’ p. 281).—

A. Coleoptera. Scarabaerdae: 1. Trichius fasciatus Z. B. Diptera. (a)
Emprdae: 2. Empis livida Z.; 3. E. rustica Fall. (6) Leptidae: 4. Atherix ibis
FI. (c) Afuscidae: 5. Sp. of Lucilia. (d@) Stratiomyida : 6. Stratiomys chamaeleon
Deg. (e) Syrphidae: 4. Eristalis arbustorum Z.; 8. E. nemorum £.; 9g. E. sepul-
cralis Z.; 10. Syritta pipiens £. C. Hymenoptera. Afgzdac: 11. Heriades
truncorum JL. 9, skg.; 12. Macropis labiata /. 6, do.; 13. Prosopis sp. Ad.

MacLeod (Flanders) observed a hover-fly (Bot. Jaarb. Dodonaea, Ghent, vi,
1894, p. 270). Schletterer (Pola) saw the Iclineumonid Tryphon rutilator Gr.

1i2

484 ANGIOSPERMAE—DICOTYLEDONES

1130. O. aquatica Poir. (=O. Phellandrium Zam., and Fhellandrium aquaticum
L.).—Schulz (‘ Beitrage,’ II, p. 190) states that this species is andromonoecious with
protandrous hermaphrodite flowers. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896)
describes the plants growing in the neighbourhood of Ruppin as being andromonoe-
cious. The umbellules of primary umbels bear hermaphrodite flowers, or there may
be a few marginal male flowers; those of the secondary umbels may bear herma-
phrodite flowers only, or marginal male and central hermaphrodite ones, or male
flowers exclusively : those of the tertiary umbels are purely male.

VistTors.—Herm. Miiller observed the following.—

A. Coleoptera. (a) Cerambycidac: 1. Leptura livida Z., numerous, skg.,
po-dvg. (4) Chrysomelidae: 2. Prasocuris phellandrii Z., dvg. the flowers. (¢) Coced-
nellidae: 3. Coccidula rufa Hés/, nect-lkg. (d@) Llateridae: 4. Adrastus palleus J. £’.,
skg. B. Diptera. (a) Afuseidae: 5. Aricia vagans Fall., skg.; 6. Cyrtoneura
curvipes A/acg. (as determined by Herr Winnerz), do.; 7. Lucilia cornicina /.
(0) Alycetophilidae: 8. Sciara thomae L. (c) Stratiomyrdae: 9. Odontomyia viridula
£. (da) Syrphidae: 10. Eristalis arbustorum Z.; 11. Syritta pipiens Z. and others.
C. Hymenoptera. (a) Apidae: 12. Prosopis variegata /. $; 13. Sphecodes
gibbus Z. 6. (2) Lchneumonidae: 14. Various sp. (c) Scolitdae: 15. ‘Viphia
tuficornis Kig. (d@) Sphegidac: 16. Oxybelus bipunctatus O/. 9; 17. Pompilus
trivialis Dahli. 9; 18. P. viaticus L. (e) Lenthredintdae: 19. Athalia rosae L. ;
20. Tenthredo sp. D. Lepidoptera. 21. Vanessa c-album JZ.

The following were recorded by the observers, and at the places stated.—

Knuth (Foéhr) (‘ Weit. Beob. ii. B]. u. Insekt. a. d. nordifr. Ins.,’ p. 234), 2 skg. hover-
flies (Eristalis sp. and Syrphus sp.) and Muscids (Musca domestica Z.. Sarcophaga
carnaria Z.,and Scatophaga stercoraria). Alfken(Bremen),—A. Diptera. Syrphidae:
1. Chrysotoxum festivum /,. B. Hymenoptera. (a) Afzdac: 2. Andrena parvula
K. 9. (0) Sphegidae: 3. Ceropales maculatus /.; 4. Crabro brevis v. d@ Z. 9 and 6.
(c) Vespidae: 5. Odynerus parietum Z. 9. MacLeod (Flanders), 6 short-tongued
Hymenoptera, 8 hover-flies, 3 Muscids, a beetle, and a Lepidopterid (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 269).

1131. O. peucedanifolia Pollich.—

Visttors.—MacLeod (Flanders) observed a Tenthredinid and 3 Muscids (Bot.
Jaarb. Dodonaea, Ghent, vi, 1894, p. 27):

1132. O. crocata L.—

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Scott-Elliot (Dumfriesshire), a humble-bee, a short-tongued bee, a wasp, 5 hover-
flies, and 4 Muscids (‘ Flora of Dumfriesshire,’ p. 78). Loew (Berlin Botanic Garden),
2 hover-flies (Syritta pipiens Z. and Syrphus ribesii Z.), skg.

341. Aethusa L.

1133. A. Cynapium L. (Herm. Miiller, ‘Weit. Beob., I, p. 305; Schulz,
‘Beitrage, II, pp. 84, go-1; Kerner, ‘ Nat. Hist. Pl.” Eng. Ed. 1, Il, pp. 342, 344;
Warnstorf. Verh. bot. Ver., Berlin, xxxv, 1895; Knuth, ‘Bl. tu. Insekt. u. d. nordfr.
Ins.,’ p. 155.)—According to Schulz, the hermaphrodite flowers of this species are
slightly protandrous or homogamous, but Kerner describes them as protogynous.
Automatic self-pollination regularly takes place, as a result of the filaments bending
inwards.

UMBELLIFERAE 485

Warnstorf says that at Ruppin either all the umbels are purely hermaphrodite,
or the tertiary ones bear marginal hermaphrodite and central male flowers.

Vistrors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (North Frisian Islands), 5 Syrphids, 2 Muscids, and a beetle. Buddeberg
(Nassau),—A. Diptera. Syrphidae: 1. Ascia podagrica /., po-dvg., very numerous ;
2. Helophilus floreus Z., nect-lkg. and po-dvg.; 3. Paragus cinctus Scheer ef Egg.,
nect-lkg. B. Hymenoptera. All nect-lkg. (a) Apidae: 4. Prosopis communis
Ny. 9; 5. P. obscurata Schenck (=P. punctulatissima Sm.) $; 6. P. signata Pz. $;
7. P. sinuata Schenck 8. (6) Sphegidae: 8. Crabro vexillatus Pz. 6. 9. Pompilus
concinnus Dah/b. 9. (c) Tenthredinidae: 10. Allantus temulus Scop. L. MacLeod
(Flanders) 4 hover-flies, 2 Muscids, and Trombidium (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, pp. 271, 280). Sickmann (Osnabriick) the fossorial wasp Pemphredon
lugubris Zé.

342. Foeniculum Tourn.

1134. F. vulgare Mill. (=. capillaceum Gv/vd., F. officinale A//., and Anethum
Foeniculum Z.)—According to Schulz (‘ Beitrage,’ II, pp. 84, 190), the small yellow
flowers of this species are andromonoecious, the hermaphrodite ones being markedly
protandrous. Kerner (‘Nat. Hist. Pl.,’ Eng. Ed. 1, Il, p. 325) says that these
protandrous flowers are the first to bloom, and after their stamens have dropped off
are geitonogamously crossed by crumbling masses of pollen, which fall from the later
maturing male flowers of adjacent lateral umbels.

Visirors.—These were recorded by the observers, and for the localities stated.—

Loew (Berlin Botanic Garden), a true wasp, Eumenes coarctatus Z. Kohl
(Tyrol), 4 Chrysidids (1. Chrysis scutellaris #.; 2. C. distinguenda Spzw.; 3. Stilbum
cyanurum Fors/., var. calens #.; 4. Hedychrum roseum foss7) and 10 true wasps
(1. Vespa germanica /.; 2. V. holsatica 7; 3. Polistes gallica Z.; 4. Eumenes
pomiformis /.; 5. E. unguiculata V7//.; 6. Odynerus sinuatus /.; 7. O. bifasciatus L. ;
8. O. parvulus Lep.; 9. O. bidentatus Zep.; 10. O. modestus Sazss.). Handlirsch the
fossorial wasp Gorytes pleuripunctata Cos/a. Schletterer and von Dalla Torre (Tyrol),
4 bees (1. Halictus albipes #.; 2. H. costulatus Avrchd. 8; 3. H. sexcinctus &.;
4. H. vulpinus Wy).

343. Seseli L.

1135. S. Hippomarathrum Jacq.—Schulz (‘ Beitrage,’ I, p. 49) states that the
umbels of this species develop very slowly. Only protandrous hermaphrodite flowers
have been observed. Some tertiary umbels do not set fruits.

1136. S. annuum L.—In this species again, Schulz (loc. cit.) only observed
protandrous hermaphrodite flowers.

344. Libanotis Riv.

1137. L. montana Crantz (=Athamanta Libanotis Z., and Seseli Libanotis
Koch).—Schulz (‘ Beitrige,’ I, p. 49) found only hermaphrodite flowers in primary
and secondary umbels. Tertiary ones (not always present) are often purely male.
According to Kerner (‘ Nat. Hist. Pl.’ Eng. Ed. 1, II, p. 511), the petals are white
in the lowlands, but reddish-violet on the under-side in plants growing on Alpine
heights. In Hohenzollern the flowers are tinged with yellow, and rather fragrant.

486 ANGIOSPERMAE—DICOTYLEDONES

Visitors. —X. Rieber, quoted by Kirchner, gives the following list for the
neighbourhood of Haigerloch in Hohenzollern (Jahreshefte Ver. Natk., Stuttgart.
xlvili, 1892).—

A. Coleoptera. (a) Cerambycidac: 1. Leptura testacea Z.; 2. Molorchus
minor Z.; 3. Strangalia bifasciata AZw//.; 4. S. melanura Z.; 5. S. quadrifasciata Z.
(4) Scarabaeidae: 6. Cetonia aurata Z. B. Diptera. 7. 33 different flies, the species
of which cannot be given with certainty. C. Hemiptera. (Carnivorous forms not
seeking for pollen or nectar): Penfatomidae: 8. Carpocoris nigricornis #.; 9. Eury-
dema festivum Z.; 10. E. oleraceum £.; 11. Eurygaster hottentotta A.-Sch. ;
12. Graphosoma lineatum Z.; 13. Falomena prasina £.; 14. Tropicoris rufipes Z.
D. Hymenoptera. (a) dApidac: 15. Andrena hattorfiana F.; 16. Coelioxys
rufescens Lep.; 17. Nomada lineola Pz.; 18. N.ochrostoma A. (0) Lehneumonidae:
19. Amblyteles negatorius #.; 20. A. palliatorius Gr.; 21. Caenocryptus bimaculatus
Grav.; 22. Ichneumonsarcitorius L. (c) Zenthredinidae: 23. Allantus arcuatus Fors?.;
24. A. schaefferi Alg.; 25. A. vespa QRe/z.; 26. Macrophyia albicinecta Schr. ;
27. M. militaris K/g.; 28. M. diversipes Schr.; 29. Tenthredo fagi Ps.; 30. T.
flava Poda; 31. T. dispar Klg. (d) Vespidae: 32. Odynerus parietum Z.;
33. Polistes gallica 7. E. Lepidoptera. Rhopalocera: 34. Argynnis paphia /. ;
35. Limenitis sibylla /.; 36. Melanargia galatea Z.; 37. Syrichthus alveus /7/2. ;
38. Thecla quercus Z.; 39. Vanessa io L.

Kohl (Tyrol) observed the true wasp Odynerus parietum Z., and also the
fossorial wasp Crabro rhaeticus Asch. e/ Archd.

MacLeod (Pyrenees) noticed 5 short-tongued Hymenoptera, a Lepidopterid,
3 beetles, and 9 flies (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 412).

345. Cnidium Cusson.

1138. C. venosum Koch (Sescli venosum /offm. and Selinum sylvestre Z.).—
Schulz (‘ Beitrige,’ I, p. 49) describes this species as andromonoecious. In some
places all the flowers are hermaphrodite, in others the secondary umbels usually bear
some male flowers or, rarely, may be purely male.

346. Athamanta L.

1139. A. cretensis L.—According to Kerner (‘ Nat. Hist. Pl.” Eng. Ed. 1, U,
p. 296), there is a central hermaphrodite flower in every umbel of this species,
surrounded by pseudo-hermaphrodite male flowers, and these again by hermaphrodite
ones.

347. Silaus Bernh.

1140. S. pratensis Bess. (=S. flavescens Bernh., Peucedanum silaus L., and
Seseli pratense Cran/z).—Schulz (‘Beitrage,’ I, p. 49) observed only hermaphrodite
flowers in this species.

VisiTors.— The following were recorded by the observers, and for the localities
stated.— -

Herm. Miiller (‘ Fertilisation,’ p. 282), a bee (Halictus longulus Sm. 4, skg.),
a Pompilid (Pompilus viaticus /. 8, nect-lkg.), and a saw-fly (Allantus nothus A’7g.).
Krieger (Leipzig), the fossorial wasp Mellinus sabulosus #. 9. Kohl (Tyrol), 2 true
wasps (Odynerus parietum Z., and O. trifasciatus /.).

UMBELLIFERAE 487

348. Meum Adans.

1141. M. athamanticum Jacq.—Schulz (‘Beitrage, II, pp. 84-5, 190) describes
this species as andromonoecious, with markedly protandrous hermaphrodite flowers.

1142. M. Mutellina Gaertn. (=Phellandrium Mutellina Z.). (Ricca, Atti Soc.

ital. sc. nat., Milano, xiv, 1871; Herm. Miller, ‘Alpenblumen,’ pp. 116-20; Kerner,
‘Nat. Hist. Pl,’ Eng. Ed. 1, II,

p- 296; Schréter, ‘Beitrage zur
Kennt. schweiz. Bliitenpfl.’)— The
flowers of this species are rose-red
or dark carmine-red in colour,
and smell like honey. Schréter
states that there are male flowers

as well as the markedly protan- Bs
ee Fic. 104. Afcum Mutellina, Gaertn. (after Herm. Miiller).
drou : + :
> hermaphrodite ones ‘The A. Flower in the first (male) stage. B. Ditto, in the second
two kinds are usually associated (female) stage. 1-5, stamens; 77, filament; 1, nectary; 07,

ovary ; s/, stigma.
on the same stocks, but a few , .

plants are purely male. Kerner describes the umbels as possessing an intermediate
zone of pseudo-hermaphrodite male flowers.

Visitors.—Herm. Miiller observed 5 beetles, 32 Diptera, 5 Hymenoptera, and
y Lepidoptera in the Alps.

349. Pachypleurum Ledeb.

1143. P. alpinum Ledeb.—Ekstam describes the flowers of this species in Nova
Zemlia as being markedly protandrous, sometimes protogynous-homogamous ; their
odour resembles that of elder, and their diameter is 1-5-2 mm.

Visirors.—Flies have been observed by Ekstam in Nova Zemlia.

350. Crithmum L.

1144. C. maritimum L. (Kirchner, ‘D. Bliiten d. Umbelliferen.’)—Kirchner
states that in this species the protandry characteristic of Umbelliferae is extremely
well marked. ‘The small flowers (only about 2 mm. in diameter) possess yellowish-
white petals, which are permanently involute. The stamens, that also are at first
incurved, diverge as usual during dehiscence, after which they wither and fall off with
the petals. The two styles now develop, though in the male stage no trace of them
was recognizable; they are very short. An umbel does not as a rule pass into the
female stage till all the stamens and petals have dropped off, so that insect-visits
necessarily bring about crossing between different umbels. Owing to the white
colour of the disks, umbels in the female stage appear of a whitish-green, and are
less conspicuous than during the preceding male stage.

Visrrors.—The following were recorded by the observers, and for the localities
stated.—

Kirchner (on the Lido near Venice), flies. Plateau (Ghent Botanic Garden),
Musca domestica /., and a smal] bug (Miris sp.).

488 ANGIOSPERMAE—DICOTYLEDONES

351. Gaya Gaud.

1145. G. simplex Gaud.—Schulz (‘ Beitrage’) describes this species as andro-
monoecious, with protandrous hermaphrodite flowers. According to Hermann
Miiller (‘ Alpenblumen,’ p. 120), the flower
mechanism agrees with that of Meum
Mutellina,

Visirors.—Herm. Miiller saw 8 Diptera.

352. Conioselinum Fisch.

m46. C. tartaricum Hoffm. (=C.
Fischeri Wimm. ef Grab., and Selinum
Fic. 165. Gaya simplex, Gand. (after Herm. Gmelini Bray).— Schulz (‘ Beitrage,’ II,
Due ee ee ee p. 190) describes this species as andromo-
a ea noecious, with protandrous hermaphrodite

flowers.

Loew (Berlin Botanic Garden) observed the following.—

A. Diptera. (a) Afuscidae: 1. Anthomyia sp., skg.; 2. Chloria demandata F” ;
3. Pyrellia cadaverina Z.; 4. Sarcophaga carnaria Z. (6) Syrphidae: 5. Eristalis
arbustorum Z., skg.; 6. E. nemorum /..; 7. E. tenax Z., skg.; 8. Helophilus floreus
£., do.; 9. Syritta pipiens Z.,do. B. Hymenoptera. Sphegidae: 10. Oxybelus
bipunctatus O/zv., 9 and 6.

353. Levisticum Riv.

1147. L. officinale Koch (=Ligusticum Levisticum Z.).—

Vistrors.—The following were recorded by the observers, and for the localities
stated. —

Loew (Berlin Botanic Garden),—A. Diptera. (a) Afusczdae: 1. Anthomyia
sp. skg. (6) Syrphidac: 2. Helophilus trivittatus #.; 3. Syrphus pyrastri Z.,
skg. B. Hymenoptera. (a) Apidae: 4. Apis mellifica Z. ¥, skg. and po-cltg.
(6) Sphegidae: 5. Crabro cribrarius Z.é. von Dalla Torre (Tyrol), the humble-bee
Bombus terrester £. Kohl (Tyrol), the true wasp Ancistrocerus parietum L.

354. Ligusticum L.
1148. L. pyrenaicum Gouan.—

Visttors.—The following were recorded by the observers, and for the localities
stated.—

MacLeod (Pyrenees), 17 short-tongued Hymenoptera, 16 Syrphids, 26 Muscids
and Empids. Loew (Berlin Botanic Garden), a hover-fly (Eristalis arbustorum Z.).

1149. L. commutatum Regel.—

Visi1rors.—Loew saw a humble-bee (Bombus terrester Z. ¥), skg., in the Berlin
Botanic Garden.

355. Selinum L.

1150. S. pyrenaeum Gouan (=Angelica pyrenaea Spreng.).— This species bears
flowers of a greenish colour.

UMBELLIFERAE 489

Vistrors.—MacLeod (Pyrenees) saw a beetle and 2 flies.

51. S. Carvifolia L. (= Angelica Carvifolia 22/.).—Schulz (‘ Beitrage,’ I, p. 49,
II, p. 190) describes this species as andromonoecious, with markedly protandrous
hermaphrodite flowers, many of which do not attain maturity. The secondary
umbels are often purely male. According to Warnstorf (Verh. bot. Ver., Berlin,
XXXvill, 1896), the umbellules of primary umbels bear hermaphrodite flowers at
Ruppin ; secondary ones marginal hermaphrodite and central male flowers.

Visirors.—The following were recorded by the observers, and for the localities
stated. —

Loew (Switzerland) (‘ Beitrage,’ p. 56), Ichneumonids (undetermined sp.) and
a saw-fly (Tenthredo sp.). Sickmann (Osnabriick), the parasitic fossorial wasp
Ceropales maculatus 7”.

356. Ostericum Hoffm.

1152. O. palustre Bess. (=O. pratense Hoffm., and Angelica pratensis Bzvd.).—
Schulz describes this species as andromonoecious, with protandrous hermaphrodite
flowers.

357. Angelica L.

1153. A. sylvestris L. (Schulz, ‘ Beitrige,’ 1, p. 50.)—Warnstorf (Verh. bot.
Ver., Berlin, xxxviii, 1896) states that the umbellules of primary and secondary umbels
bear hermaphrodite flowers at Ruppin; tertiary ones marginal hermaphrodite and
central male flowers. Sometimes all the umbels are female by degeneration of the
anthers.

Vistrors.—Herm. Miiller gives the following list (‘ Fertilisation, p. 282; ‘ Weit.
Beob.,’ I, p. 305).—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella septempunctata Z., skg.;
2. C. quattuordecimpunctata Z., do. (0) Dermestidae: 3. Anthrenus pimpinellae /.
(c) Nitidulidae: 4. Meligethes, freq. (d@) Scarabaeidae: 5. Trichius fasciatus L., nect-
Ikg. (e) Zelephoridae: 6. Telephorus melanurus #. B. Diptera. (a) Muscrdac:
7. Echinomyia fera Z.; 8. Lucilia sylvarum Z.; 9. Mesembrina meridiana JZ. ;
10, Sarcophaga sp.; 11. Scatophaga merdaria #.; 12. S. stercoraria Z.; 13. Tachina
larvarum LZ. (0) Syrphidae: 14. Eristalis pertinax Scop.; 15. Helophilus floreus Z.:
16. Pipizella virens /.; 17. Syritta pipiens Z. C. Hymenoptera. (a) Aprdae:
18. Andrena pilipes #9; 19. Sp., skg. (6) Evaniidae: 20. Foenus affectator /.
(c) Ichneumonidae: 21. Various species. (d) Sphegidae: 22. Crabro lapidarius Pz.,
é and 9, freq.; 23. Philanthus triangulum /. (e) Zenthredinidac: 24. Athalia rosae L.;
25. Species of Tenthredo. (/) Vesprdae: 26. Odynerus debilitatus Sauss.; 27. Vespa
rufa Z.¥,skg. D. Lepidoptera. 28. Argynnis paphia Z. (skg.?). E.Neuroptera.
2g. Panorpa communis Z., nect-lkg.—And (in the Alps), 4 beetles, a Muscid, and
2 wasps (‘ Alpenblumen,’ p. 120).

Alfken observed the following at Bremen.—

A. Diptera. (a) Zherevidae: 1. Thereva nobilitata /, skg. (4) Aluscrdac:
2. Cyrtoneura hortorum Fal, freq.; 3. Frontina laeta Mg.; 4. Graphomyia maculata
Scop., freq.; 5. Nemoraea radicum /.,do.; 6. Onesia sepulcralis Mg.,do. (c) Syrphidac:
7. Cheilosia variabilis Pz.; 8. Chrysotoxum bicinctum Z. 9; 9. Eristalis intricarius Z.,
very common; ro. Sericomyia borealis Fa//.; 11. Syrphus balteatus Deg., do.; 12. S.

490 ANGIOSPERMAE—DICOTYLEDONES

corollae 7”, do.; 13. S. pyrastri Z., freq.; 14. Volucella bombylans Z. B. Hymeno-
ptera. («) Zchneumonidae: 15. Amblyteles occisorius /.; 16. Banchus falcator .,
rare; 17. Metopius micratorius Gr.; 18. Ophion ramidulus Gr.; 19. Phygadenon
cephalotes Gr. (4) Sphegidae: 20. Crabro cribrarius Z. 9 and é, very common ;
a1. C, fuscitarsis H.-Sch. 9, very rare; 22. C. vagus LZ. 9, very common. (c) Zenthre-
dinidae: 23. Abia sericea L., rare. (7) Vespidae: 24. Odynerus parietum Z. 4, very
common; 25. O. sinuatus /. 4, rare.

Loew gives the following for the Riesengebirge (‘ Beitrage,’ p. 47).—

A. Coleoptera. (a) Cerambycidae: 1. Pachyta octomaculata F.; 2. P. quadri-
maculata Z.; 3. Strangalia armata Wdst. (0) Scarabaetdac: 4. Trichius fasciatus Z.
B. Diptera. (a) Afuscidae: 5. Echinomyia fera £.; 6. E. grossa Z. (6) Ayceto-
philidae: 7. Sciara thomae L. (c) Syrphidae: 8. Eristalis nemorum Z.; 9. Syrphus
cinctellus Ze//.$; 10. S. glaucius Z.; 11. Volucella pelluceus Z, C. Hymenoptera.
(a) Apidae: 12. Psithyrus rupestris /. 9, skg. (6) Zenthredinidae : 13. Rhogogastera
viridis Z. (c) Vespidae: 14. Vespa rufa 7. ¥, skg. D. Neuroptera. 15. Panorpa
communis £. And for Switzerland (op. cit., p. 55),—A. Diptera. TZadanidae:
1, Tabanus infuscatus Zw. (?). B. Hymenoptera. Sphegidac: 2. Crabro cri-
brarius £. 9; 3. Gorytes campestris J/iii/.

The following were recorded by the observers, and for the localities
stated.—

Handlirsch, 3 fossorial wasps (1. Gorytes bicinctus Ross’; 2. G. quadrifasciatus
F.; 3. G. quinquecinctus 7). Von Dalla Torre (Tyrol), the true wasp Leionotus
minutus /, Sickmann (Osnabriick),—A. Hymenoptera. (a) Sphegidae: 1. Cero-
pales maculatus F., occasional; 2. Crabro cribrarius Z., very common; 3. C. dives
Hf.-Sch., vare; 4. C. sexcinctus v.d. L., freq.; 5. C. vagus Z., very common; 6.Gorytes
bicinctus Ross?,aé; 7. G. laticinctus Schuck., infrequent; 8. G. quadrifasciatus / ;
yg. G. quinquecinctus 7°, rare; 10. Mellinus sabulosus #.; 11. Mimesa atra Pz.;
12. Pemphredon unicolor #. H. de Vries (Netherlands) (Ned. Kruidk. Arch.,
Nijmegen, 2. Ser, 2. Deel, 1875), a bee (Apis mellifica Z. ¥), a humble-bee
(Bombus terrester Z. ¥), a parasitic humble-bec (Psithyrus vestalis /ourcr. 9), a true
wasp (Vespa germanica /. §), and a fossorial wasp (Crabro vagus /. 9). MacLeod
(Flanders), Apis, 9 short-tongued Hymenoptera, 10 hover-flies, 11 other flies, 2
beetles, and Panorpa (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 273-4, 380);
(Pyrenees), 8 Hymenoptera, 5 beetles, and 4 flies (op. cit., iii, 1891, pp. 407-8).
Lindman (Dovrefjeld), numerous flies, and several bees.

Willis records the following for the neighbourhood of the south coast of
Scotland (‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Diptera. (a) Chironomidac: 1. Chironomus sp., skg.; 2. C. (Cricotopus)
tremulus Z., do. (0) Afuscrdae: 3. Anthomyia radicum Z., skg.; 4. A. sp., do.;
5. Caricea tigrina 2, do.; 6. Aricia incana Wed., do.; 7. A. lucorum a7, freq.,
skg.; 8. Lucilia caesar Z., do.; 9. L. sericata .J/g., do.; 10. Cyrtoneura curvipes
Mcqg., do.; 11. Mydaea sp.; 12. Myobia inanis Fall; 13. Sarcophaga sp.; 14.
Scatophaga stercoraria Z.; 15. Spilogaster communis R.-D. (c) Alycetophilidae :
16. Glaphyroptera fasciola M/g.; 17. Sceptonia nigra Afg. (@) Phoridae: 18. Phora
sp. (e) Syrphidac: 19. Cheilosia oestracea .; 20. Eristalis horticola Deg.; 21. F.
pertinax Scop., freq.; 22. Platycheirus peltatus J/g. B. Hemiptera. 23. Antho-
coris sp.; 24. Calocoris fulvomaculatus Deg. C. Hymenoptera. (a) Apidae:
all skg.: 25. Bombus terrester Z.; 26. Halictus rubicundus Cér.; 27. Prosopis
brevicornis Wy/.; all skg. (6) Lchncumonidae: 28. Eight undetermined species.
(¢) Lenthredinédae: 29. Selandria serva F. (d) Vespidae: 30. Vespa sylvestris Scop.,
freq. D. Lepidoptera. Rhopalocera: 31. Polyommatus phlaeas Z., skg.

UMBELLIFERAE 491

358. Archangelica Hoffm.

1154. A. officinalis Hoffm. (= Angelica Archangelica L.).—The hermaphrodite
flowers of this species are markedly protandrous in the Tyrol and Central Germany
(Schulz, ‘ Beitrage,’ 1, p. 190); also in Greenland (Warming). In the former
regions the species is andromonoecious, and the secondary umbels are partly male,
the tertiary oncs entirely so.

Plateau observed the following insects as common visitors.—

Bees (Apis; Odynerus quadratus Pz.), a Chrysidid (Chrysis ignita Z.), and
Muscids (Calliphora; Musca; Lucilia).

359- Peucedanum U..

1155. P. Cervaria Cusson.—Schulz (‘ Beitrage,’ II, pp. 50-1) states that this
species is andromonoecious, with protandrous hermaphrodite flowers ; sometimes these
latter preponderate, sometimes the male flowers do so. The secondary umbels may
bear male flowers only, hermaphrodite ones only, or both. When tertiary umbels
occur they are purely male.

Visivors.—Herm. Miiller observed the following in Thuringia (‘ Fertilisation,’
p. 282).—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia bifasciata AZdéller. (6) Chryso-
melidae: 2. Clythra scopolina Z. B. Diptera. (a) Bombylizdae: 3. Anthrax maura /..
(4) Muscidae: 4. Gymnosoma rotundata Z., very numerous; 5. Phasia analis &., one ;
6. P. crassipennis /, freq. C. Hymenoptera. (a) Apidae: 7. Andrena minutula
X. 9, numerous, po-clig.; 8. Halictus leucozonius Sc/r. 6 and 9, skg. and po-cltg. ;
g. H. quadricincius /. 9, skg.; 10. Megachile lagopoda /. 9, on one occasion, skg.
(4) Chrystdidae: 11. Hedychrum lucidulum #6 and 9. (c) Sphegrdae: all nect-lkg. ;
12, Ammophila sabulosa Z.; 13. Ceropales maculata /. 9; 14. C. variegata F.
g and 6; 15. Crabro cribrarius Z. 9 and 6, freq.; 16. C. vagus Z.9; 17. Nysson
maculatus /. 9; 18. Pompilus viaticus Z. 6; 19. Priocnemis bipunctatus /. 9;
20. P. obtusiventris Schjédte 9; 21. Psammophila viatica 2.8; 22. Tachysphex nitidus
Spin. Q; 23. T. pectinipes . 9; 24. Tiphia femorata /., very numerous. (d) Ves-
pidae: 25. Polistes biglumis Z.; 26. P. gallica L.

Kohl records the fossorial wasp Crabro cribrarius Z. 9 and 6, freq., for the Tyrol.

Loew saw the following in the Berlin Botanic Garden.—

A. Diptera. (a) A/uscrdae: 1. Graphomyia maculata Scop. (4) Syrphidac :
2. Eristalis arbustorum /., skg.; 3. E. nemorum “Z., do.; 4. E. tenax Z., do.
B. Hymenoptera. (c) Apidae: 5. Prosopis sp. 9, skg. (4) Venthredinidac:
6. Allantus viennensis Pz.

156. P. Oreoselinum Moench (= Athamanta Oreoselinum £.).—Schulz
(‘ Beitrage, I, p. 52) describes this species as andromonoecious, with markedly
protandrous hermaphrodite flowers, and, as a rule, a relatively large number of male
ones. The primary umbels usually contain hermaphrodite flowers, rarely male ones
only. In the latter case secondary umbels are exclusively hermaphrodite. Tertiary
umbels, when present, are male. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896)
found the primary umbels to be hermaphrodite at Ruppin, while the others were
almost always male, their umbellules only now and then possessing a few marginal
hermaphrodite flowers.

492 ANGIOSPERMAE—DICOTYLEDONES

Visrtors. —Loew observed the following in Brandenburg (B.) and Mecklenburg
(M.) (‘ Beitrage,’ p. 37).—

A. Coleoptera. (a) A//eculidac: 1. Cteniopus sulphureus Z. (6) Oedemeridae :
2. Oedemera flavescens Z. § (B.); 3. O. flavipes /. 8 (B.); 4. O.lurida AZarsh. (B.) ;
5. O. subulata Otv. 9 (B.); 6. O. podagrariae LZ. (B.); 7. O. virescens LZ. (B.).
(c) Telephoridae: 8. Dasytes flavipes / (B.); 9. Rhagonycha melanura f. (B.).
B. Diptera. (a) Afuscrdac: 10. Cynomyia mortuorum Z£. (M.); 11. Exorista
lucorum A/g. (M.); 12. Olivieria lateralis /. (M.). (0) Syrphrdae: 13. Eumerus
ovatus Zw. 9 (M.). C. Hymenoptera. (a) Apidae: 14. Colletes daviesanus X. 9,
skg. (M.); 15. C. fodiens A. 9, do. (M.). (0) Lehneumonidae: 16. Undetermined
species (B.). (c) Scolitdae: 17. Tiphia minuta 7. d. L. 9 (B.). (d) Vesprdae:
18, Odynerus trifasciatus 7. (M.). Also in Switzerland (op. cit., p. 56),—A. Coleo-
ptera. (a) Cleridae: 1. Trichodes apiarius Z. (¢) Scarabaeidae: 2. Cetonia
aurata Z., var. lucidula; 3. Hoplia praticola Du/7. B. Diptera. (a) ALuscedae :
4. Ocyptera brassicaria #. (0) Stratiomjidae: 5. Stratiomys chamaeleon Deg. ;
6. S. longicornis Scop. (c) Syrphidae: 7. Syrphus diaphanus Ze//.(?). (¢) Tabanzdae :
8. Tabanus bromius Z.; 9. T. infuscatus Zw. C. Hymenoptera. TZenthredinidae:
to. Allantus viduus Ross.; 11. Hylotoma berberidis Sc/r.; 12. Tenthredo sp.

Herm. Miller (Kitzingen) saw the hawk-moth Zygaena meliloti £sp., skg. or
attempting to do so. Réssler (Wiesbaden) noticed a Tineid : Chauliodus iniquellus
Wek.

1157. P. officinale L.—Schulz (‘ Beitrige,’ II, p. 190) describes this species as
andromonoecious, with protandrous hermaphrodite flowers.

1158. P. venetum Koch.—According to Schulz (‘ Beitrage,’ II, pp. 85, 90), the
hermaphrodite flowers of this species are strongly protandrous. He did not observe
any purely male ones.

1159. P. alsaticum L.—Schulz (‘ Beitrage, I], p. 190) describes this species as
andromonoecious, with protandrous hermaphrodite flowers.

1160. P. palustre Moench (=Selinum palustre /., and Thysselinum palustre
Floffm.).—As P. alsaticum (Schulz, loc. cit.).

Visrtors.—Herm. Miiller observed the following visitors at Lippstadt (‘ Weit.
Beob.,’ I, p. 306).—

A. Diptera. (2) Bidbionidae: 1. Dilophus vulgaris A7Zg. freq. (0) ALuscidae :
2. Aricia sp.; 3. Sepsis sp. (c) Syrphidae: 4. Eristalis arbustorum Z., nect-lkg. ;
5. Helophilus floreus Z. B. Coleoptera. TZi/ephoridae: 6. Dasytes flavipes /.,
nect-lkg.; 7. Telephorus melanurus /.,do. C. Hymenoptera. (a) /chneumonidae :
8. Various species. (d) Sphegidac: 9. Crabro brevis v. d. Z., in great numbers, skg.
(c) Apidae: 10. Prosopis clypearis Schenck 6, skg.

Loew noticed the following in Silesia (‘ Beitrage,’ p. 30).-—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia armata “dst. (0) Tele-
phoridae: 2. Dasytes flavipes #., nect-lkg. (¢) Metidulidac: 3. Meligethes sp.
B. Diptera. Syrphidae: 4. Eristalis arbustorum Z., skg. C. Lepidoptera.
Rhopalocera: 5. Argynnis aglaia Z., skg.; 6. A. pandora S. 1”, do.; 7. A. paphia
/.., do.

1161. P. Ruthenicum Biel.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Coccrnellidae: 1. Coccinella bipunctata Z., nect-lkg.; 2. C.
septempunctata Z., nect-lkg. B. Diptera. (a) Afuscidae: 3. Anthomyia sp., skg. ;

UMBELLIITERAE 493

4. Chloria demandata /.; 5. Pyrellia cadavorina L.; 6. Sarcophaga carnaria ZL.
(4) Syrphidae: 47. Eristalis arbustorum /.; 8. E. nemorum Z., skg.; 9. Helophilus
floreus Z., skg.; 10. Syritta pipiens Z., skg. C. Hymenoptera. (a) Apidae:
11. Prosopis armillata 4y/. 9, skg.; 12. P. sp. 9, skg. (0) Sphegidae: 13. Crabro
vexillatus Pz. 9; 14. Oxybelus bipunctatus O//7. 9 and 6.

360. Tommasinia Bertol.

162. T. verticillaris Bertol. (= Peucedanum verticillare Spreng.).—
Visitors.—Loew observed the following in the Berlin Botanic Garden (‘ Bliiten-
biol. Floristik,’ p. 242).—

A. Coleoptera. TZelephoridae: 1. Anthocomus equestris /, nect-lkg. B.
Diptera. Syrphidae: 2. Eristalis nemorum Z.,skg. C. Hymenoptera. <Aprdac:
3. Apis mellifica Z. ¥, skg. and po-cltg.

361. Ferulago Koch.

1163. F. monticola Boiss. et Heldr. (=Ferula monticola Vym.),—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Coccznellidae: 1. Coccinella septempunctata Z. (6) Derme-
stidae: 2. Anthrenus scrophulariae /., nect-lkg. B. Diptera. Brbzonidae: 3. Bibio

hortulans Z. 9, nect-lkg. ©. Hymenoptera. enthredinidae: 4. Hylotoma
berberidis Schr. 9.

1164. F. sylvatica Reichb.—

Vistrors.—Loew observed the hover-fly Syritta pipiens Z., skg., in the Berlin
Botanic Garden.

362. Imperatoria I.

165. I. Ostruthium L. (=Peucedanum Ostruthium och). According to
Schulz (‘Beitrage,’ II, p. 190), this species is andromonoecious, with protandrous
hermaphrodite flowers. Hermann Miiller says that the flower mechanism resembles
that of Gaya.

Visitors.—Loew observed in the Alps (Heuthal) the fly Tabanus borealis /. &
(‘ Blitenbiol. Floristik,’ p. 396). Herm. Miiller, in the Alps (foot of Piz Alv), noticed
9 beetles, 11 Diptera, 7 Hymenoptera, a Lepidopterid, and a Neuropterid (‘ Alpen-
blumen,’ p. 121).

Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Dermestidae: 1. Anthrenus scrophulariae Z., nect-lkg.
B. Diptera. (a) Muscidae: 2. Lucilia caesar LZ. (0) Syrphidae: 3. Eristalis

nemorum Z., skg.; 4. Syritta pipiens Z. C. Hymenoptera. Afrdae: 4. Andrena
schrankella yl. 9, skg. and po-cltg.

363. Anethum Tourn.

1166. A. graveolens L. (=Peucedanum graveolens Benth. ef Hook. f.).—The
small yellow flowers of this species contain but little nectar, though they possess
a strong odour. Schulz (‘Beitrige, 11, pp. 85, 90, 91) describes them as homo-
gamous and hermaphrodite, and says that they are visited by many insects, chiefly
flies and Hymenoptera, rarely beetles. According to Warnstorf (Verh. bot. Ver.,

494 ANGIOSPERMAE—DICOTYLEDONES

Berlin, xxxviii, 1896), the umbellules of the primary umbels bear hermaphrodite
flowers at Ruppin; those of the secondary and tertiary ones marginal hermaphrodite
and central male flowcrs.

Visttors.—Herm. Miiller gives the following list (‘ Fertilisation,’ p. 283).—

A. Diptera. (a) Bomébylitdae: 1. Anthrax maura 4. (d) Jluscidae:
2. Cyrtoneura curvipes AZacy., and C. simplex Lw. (both determined by Herr
Winnertz); 3. Gymnosoma rotundata Z., freq.; 4. Lucilia cornicina /.; 5. Musca
corvina #.; 6. Sepsis, freq. (c) Stratiomyidae: 7. Chrysomyia formosa Scop., skg. ;
(¢) Syrphidae: all nect-lkg.: 8. Cheilosia scutellata /a//.; 9. Eristalis arbustorum L.;
1o, E. nemorum /.; 11. E. sepulcralis Z.; 12. E. tenax £.; 13. Syritta pipiens Z.;
14. Syrphus pyrastri £. (ev) Zipulidac: 15. Tipula sp. B. Hymenoptera:
(a) Apedae: 16. Andrena dorsata K. 9, po-cltg.; 17. A. parvula AY 9, do.;
18. Prosopis armillata Vy/. é (Borgstette, ‘Vecklenburg); 19. P. communis /Vy/.
g and $ (Borgstette, Tecklenburg); 20. P. sinuata Schenck 9 and é; 21. Sphecodes
gibbus Z. é and 9, frey. (0) Chrystdidaec: 22. Chrysis bidentata 4.9; 23. C. ignita
L. 9; 24. Hedychrum lucidulum #. 9 and 4, not infrequent. (¢) Lvandzdae:
25. Foenus affectator /.; 26. F. jaculator #. (d@) formicidav: 27. Not infrequent.
(c) Zchneumonidac: 28. Numerous species. (/) Scolitdae: 29. Tiphia femorata
L.9. (g) Sphegidae: 30. Cemonus unicolor #. 9; 31. Crabro denticrus #.-Sch. ;
32. C. podagricus #.-Sch. 9; 33. C. sexcinctus 7. 6; 34. C. vexillatus Pz. 9;
3n. C. wesmaeli vz. d. 4.8; 36. Mutilla melanocephala #.; 37. Oxybelus uniglumis
/., freq.; 38. Pompilus cinctellus Sp7v. 9; 39. P. neglectus Dahlb. 9; 40. Psen
atratus Pz. 9 and $; 41. Tachytes pectinipes Z. 9; 42. Trypoxylon clavicerum
Lep. 9. (hk) Tenthredinidae: 43. Several species of Tenthredo. (/) Vespidae:
44. Eumenes pomiformis #.6; 45. Odynerus debilitatus Savss.; 46. O parietum Z.;
47- Polistes gallica Z.

Sickmann (Osnabriick) saw the fossorial wasp Crabro lituratus Pz., rare.

According to Marshall (in André, ‘Spéc. des hym. d’Eur., IV, p. 563), the
Braconid Agathis umbellatarum JVees is especially attracted by this species.

Loew gives the following for Silesia (‘ Beitrage,’ pp. 28-g).—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida /., nect-lkg. () Sca-
rabacidae: 2. Cetonia aurata Z., nect-lkg. (c) TZelephoridae: 3. Rhagonycha
melanura /., nect-Ikg. (d) Witidulidae: 4. Meligethes sp. (e) Sz/phidae: 5. Necro-
phorus vespillo Z. B. Diptera. (a) Muscidae: all skg.: 6. Anthomyia sp.;
7. Gymnosoma rotundata 7.; 8. Lucilia caesar Z.; 9. Phasia analis /.; 10. P.
crassipennis #. (0) Mycetophihdac: 11. Sciara thomae ZL. (c) Sfratiomytdae: 12.
Chrysomyia formosa Scop., skg.; 13. Stratiomys chamaeleon Deg., do. (¢) Syrphidae :
all skg.: 14. Eristalis nemorum Z.; 15. Helophilus floreus Z.; 16. Melithreptus
scriptus Z.; 17. Syritta pipiens Z. C. Hymenoptera. (a) Aprdae: 18. Andrena
gwynana &., var. bicolor F. 9, skg.; 19. A. lucens /mh. 9, skg. and poc-cltg.;
20. A. pilipes #. 3, skg.; 21. A. propinqua Schenck 9, do.; 22. A. tibialis A. 6 (?);
23. Apis mellifica ZL. §, skg.; 24. Halictus sexnotatus K. 9, do.; 25. Sphecodes
gibbus Z. 8, do. (6) Chrysididar: 26. Chrysis viridula Z. (c) Jchneumonidae :
27. Undetermined species. (¢) Sphegidae: all skg.: 28. Cerceris arenaria L.,
29. Crabro albilabris 71; 30. C. subterraneus /.; 31. C. vexillatus Pz.; 32.
Oxybelus lineatus #. 9; 33. O. mucronatus /. 6 and 9; 34. O. pulchellus
Gorst.é; 35. O. uniglumis £.; 36. Pompilus viaticus Z. (e) Scolz?dae: 347. Tiphia
femorata /. 9, skg. (/) Zenthredinidae: 38. Hylotoma ciliaris Z., var. corrusca
Gadd.; 39. Tenthredo sp. (g) Vespzdae: all skg.: 40. Odynerus parietum Z.;
41. Polistes gallica £.; 42. Vespa germanica /.§¥. D. Lepidoptera. Rhopaloccra:
both skg.: 43. Epinephele janira 4.; 44. Polyommatus virgaureae /..

UMBELLIFERAE 495

364. Pastinaca L.

1167. P. sativa L. (=Peuccdanum sativum Benth. ef Hook. f.).—According
to Schulz (‘Beitrage,’ II, pp. 85, 93, 190), the yellow flowers of this species are
distributed andromonoeciously, and the hermaphrodite ones are protandrous. The
primary umbels either bear only hermaphrodite flowers, or may possess a central
male one, or even several such. Secondary umbels often bear marginal hermaphrodite
and central male flowers; sometimes hermaphrodite ones only. Umbels of higher
order include numerous male flowers. Purely male umbels are rare. Warnstorf
says that at Ruppin the umbellules of primary umbels are hermaphrodite; secondary

ones bear marginal hermaphrodite and central male flowers ; tertiary ones are almost
exclusively male.

Hermann Miller states that the yellow flowers are not readily visited by beetles,
and Kerner says that they are especially attractive to dung-flies.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list (Herm. Miiller, ‘ Fertilisation,’ p. 284, ‘Weit. Beob.,’ I, p. 306).—

A. Diptera. (a) Bombylidae: 1. Anthrax flava Ag. (H.M.). (0) Muscidae :
2. Dexia rustica / (H. M.); 3. Lucilia sylvarum J/g. (H. M.); 4. Onesia sepulcralis
Mg. (H.M.); 5. Sarcophaga carnaria 1. (H. M.). (c) Syrphidae: 6. Chrysotoxum
bicinctum Z. (H. M.); 7. Syritta pipiens 7., po-dvg. (H. M., Budd.). B. Hymeno-
ptera. (a) Lchneumonidae: 8. Numerous species (H. M.). (6) Scolitdae: 9. Tiphia
femorata /. (H.M.). (c) Sphegzdae: 10. Crabro sexcinctus #.$(H.M.); 11. Mutilla
europaea LZ. 9 (H. M.); 12. M. melanocephala #. & (Budd.). (d) Zenthredinidac :
13. Several species of Tenthredo (H. M.). (e) Vespidae: 14. Odynerus parietum /.
8 (H.M.); 15. Polistes biglumis Z. (H. M.); 16. P. gallica Z. (H. M.).

The following were recorded by the observers, and for the localities stated. —

Alfken (Bremen), a Muscid (Nemoraea erythrura JZg.), a bee (Andrena austriaca
Pz. 9), and a saw-fly (Allantus omissus /érs/.). Friese (Central Saalthal),—Hymeno-
ptera. (a) [chneumonidae: 1. Amblyteles fossorius (Miill.) Wesm.; 2. A. fuscipennis
Wesm.; 3. A. sputator (F.) Wesm.; 4. Exenterus apiarius (Gr.) Zhs.; 5. Exoechus
gravipes Gr.; 6. Ichneumon similatorius (F.) Zhs.; 7. Tryphon elongator Gr.
(6) Afutiliidae: 8. Mutilla rufipes /., var. nigra Rossz. (c) Sphegidae: 9g. Salius
hyalinatus #.; 10. S. versicolor Scop. (d@) Vespidae: Polistes gallica . Schiner
(Austria),—Diptera. (a) Conoprdae: 1. Conops capitatus Loew. (0) Muscidae:
2. Alophora hemiptera /.; 3. Frontina laeta Afg.; 4. Germaria ruficeps Fall. ;
5. Nemoraea radicum /.; 6. Phorocera punicata AZg. (c) Syrphidae: 7. Chryso-
toxum bicinctum Z.; 8. C. elegans Loex'; 9. Eumerus sinuatus /oew; to. Syrphus
cinctellus Ze//.; 11. S. cinctus Fa//. F. F. Kohl (Tyrol), 2 ruby-wasps (Chrysis
analis Spz., and Hedychrum rutilans Dahl.) and 3 true wasps (1. Odynerus parietum
/.., var. renimacula Lep.; 2. O. parvulus Lep.; 3. O. rossi Lep.). Schletterer (Tyrol),
the Scoliid Tiphia femorata #. Loew (Mecklenburg), Anthomyia sp. ; (Brandenburg),
the bee Halictus cylindricus /. 3, skg.; (Steiermark), a Sphegid (Crabro sp.). Warnstorf
(Brandenburg), bees, species not stated. MacLeod (Flanders), a hover-fly, 2 Muscids,
and a short-tongued Neuropterid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 275).

1168. P. opaca Bernh.—Schulz (‘ Beitrage,’ II, p. 180) describes this species as
andromonoecious, with protandrous hermaphrodite flowers.

365. Heracleum L.

169. H. Sphondylium L.—The odorous flowers of this species are usually
white and actinomorphous, but Kirchner says that they are sometimes irregular, and

496 ANGIOSPERMAE—DICOTYLEDONES

greenish, yellowish, or reddish in colour. According to Ricca and Schulz, they
are always hermaphrodite and markedly protandrous. Warnstorf (Verh. bot. Ver.,
Berlin, xxxvii, 1895) states that in Brandenburg this common umbellifer is extremely
variable as regards the colour and form of the petals. (Cf Schr. natw. Ver.,
Weringerode, vii, 1892, pp. 64-6.) The umbellules of primary umbels bear her-
maphrodite flowers; those of secondary ones marginal hermaphrodite and central
male flowers ; tertiary ones are nearly all male. Sometimes all the flowers of all the
umbels are female by degeneration of the anthers, which are then bent back on
the ovary, and contain vestigial pollen-grains only about 25 mw long and 12-13
broad, the ordinary fertile ones being about double the size. At Ruppin therefore
the species is andromonoecious and gynodioecious (Warnstorf ).

Visitors.—The gigantic inflorescences are visited by innumerable short-tongued
insects.

Herm. Miiller (H. M.), Buddeberg (Budd.), and Borgstette (Borg.) have observed
the following (Herm. Miller, ‘ Fertilisation,’ pp. 284-5, ‘ Weit. Beob.,’ I, p. 306).—

A. Coleoptera. (a) Cerambycidae: 1. Leptura maculicornis Deg., freq. (H. M.);
2. L. testacea £.(Budd.); 3. Pachyta octomaculata #. (H. M., Budd.); 4. Stenocorus
mordax Deg. (H.M.); 5. Strangalia armata Hés/. (Budd.); 6. 5. attenuata Z.(Budd.) ;
7.5. melanura Z., very freq., nect-Ikg. (H. M.); 8. S. nigra Z.(H.M.). (6) Chryso-
melidae: 9. Cryptocephalus sericeus Z.(H.M.). (¢) Clertdae: 10. Trichodes apiarius Z.
(H.AI.). (@) Coccinelidae: 11. Exochomus aurietus Scrzéa (H.M.). (e) Dermestidae :
12. Anthrenus pimpinellae /.(H.M.). (/) Evlaterrdac: 13. Agriotes ustulatus Schaller
(H. M.); 14. Corymbites holosericeus Ofv.(H. M.); 15. C. purpureus Poda (H. M.).
(g) Mordeliidae: 16. Mordella fasciata /, nect-lkg. (H. M.). (2) Mitedulidae: 17.
Meligethes, freq. (H. M.); 18. Thalycra fervida Gy/2, (H. M.). (2) Ocdemeridae :
19. (Qedemera virescens ZL. (H. M.). (4) Scarabacidac: 20. Cetonia aurata L., very
freq. (H. M.); 21. Hoplia philanthus Sw/z., do. (H. M.); 22. Oxythyrea funesta Poda,
freq. (H. M.); 23. O. hirta Poda (Budd.); 24. Trichius fasciatus Z. freq. (H. M_).
(2) Telephoridae: 25. Telephorus fuscus Z. (H. M.); 26. T. lividus Z. (H. M.);
27. T. melanurus /., very numerous (H. M.). B. Diptera. (@) Aszidae: 28. Dioc-
tria reinhardi Wred., freq. (H. M.). (8) Brdbcontdae: 29. Dilophus vulgaris JAZe.,
9 freq., 4 infreq. (H. M.). (c) Bomébyliidae: 30. Anthrax flava 47g. (H. M., Borg.) ;
31. A. hottentotta Z. (Budd.). (d@) Conopidae: 32. Myopa occulta Mg. (H. M.);
33. Zodion cinereum /,, nect-lkg. (H.M.). (ec) Empidae: 34. Empis livida Z.(H.M.).
(/) Muscidae: 35. Calliphora erythrocephala J/g. (H. M.); 36. C. vomitoria Z.
(H. M.); 37. Cynomyia mortuorum Z., nect-lkg. (Budd.); 38. Echinomyia fera Z.
(H. M.); 39. E. grossa Z.(H.M.); 40. E. lurida /. (Budd.); 41. E. magnicornis
Zell. (H. M., Budd.); 42. Exorista vulgaris #@//. (H. M.); 43. Graphomyia maculata
Scop. (H.M.): 44. Lucilia caesar Z. (H.M.); 45. L. cornicina 7. (H. M.); 46. L.
sericata J/g. (H. M.); 47. L. sylvarum JZg.(H. M.); 48. Mesembrina meridiana Z.
(H. M.); 49. Musca corvina #. (H. M.); 50. Nemoraea sp. (H. M.); 51. Onesia
floralis Ftod.-Desv. (H. M.); 52. O. sepulcralis AZg. (H. M.); 53. Phasia analis 2.
(H. M.); 54. Pollenia vespillo #. (H. M.); 55. Pyrellia aenea Ze//. (H. M.);
56. Sarcophaga carnaria Z., freq. (H. M.); 57. S. haemarrhoa J~%g. (H. M.);
58. Scatophaga merdaria #, freq. (H. M.); 59. Sepsis cynipsea Z., do. (H. M.);
60. Tachina erucarum Fond. (H.M.). (g) Alycetophilidae: 61. Platyura sp. (H. M.).
(2) Syrphidae: 62. Ascia lanceolata Mg. (H. M.); 63. A. podagrica 7. (H. M.);
64. Cheilosia scutellata Fa//.(H.M.); 65. C. oestracea L., freq. (H. M.); 66. Chryso-
gaster viduata Z. (H. M.); 67. Chrysotoxum bicinctum Z., rare (H. M.); 68. C.
festivum Z. (Borg.); 69. Fristalis aeneus Scop. (H.M.); 70. E. arbustorum Z.
(H. M.); 71. E. horticola A/g. nect-lkg. (H. M.); 72. E. nemorum JZ. (H. M.);

UMBELLIFERAE 497

73. E. pertinax Scop.(H.M.); 74. E. sepulcralis Z.(H.M.); 75. E. tenax Z.(H.M.);
76. Helophilus floreus Z., freq. (H. M.); 77. Melanostoma mellina Z. (H. M.);
78. Melithreptus menthastri Z. (H. M.); 79. Pipizella annulata Macg. (H. M.);
80. P. virens 7. (H. M.); 81. Syritta pipiens Z. (H. M.); 82. Syrphus balteatus
Deg. (H. M.); 83. S. glaucius Z. (H. M.); 84. S. pyrastri Z. (H. M.); 85. S. ribesii
L.(H.M.); 86. Volucella pellucens Z., nect-Ikg. (Budd.); 87. Xylota florum £,,
nect-lkg. (H.M.). (7) Zabanidae: 88. Tabanus micans Mg. (Budd.); 89. T. rusticus
LZ. (H.M.). (4) Tipulidae: go. Pachyrhina histrio /. (H. M.). C. Hemiptera.
gt. Several bugs (H. M.). D. Hymenoptera. (a) Apidae: 92. Andrena argentata
Sm. 9, po-cltg. (H. M.) ; 93. A. coitana X. 9, nect-Ikg. (H.M.); 94. A. fucata Sm. 9,
skg. and po-cltg. (H. M.); 95. A. nana X. 9, skg. (H. M.); 96. A. nitida K. 9, one
(Budd.); 97. A. rosae Pz. 9, freq. (H. M.); 98. A. tibialis A. 9, one (Budd.);
99. Apis mellifica ¥, skg. and po-cltg. (H. M.); 100. Bombus terrester Z. 9, po-cltg.
(H. M.); ror. Halictus cylindricus /. 9, do. (H. M.); 102. H. flavipes #. 9 (H. M.);
103. H. leucopus A. 6 (H. M.); 104. H. lugubris X. 9, in large numbers (H. M.);
105. H. tetrazonius Klg. 9 (Budd.); 106. Megachile centuncularis Z. 9, po-cltg.
(H.M.); 107. Nomada ferruginata X. 9, skg. (H. M.); 108. Prosopis armillata My/.
9 (H.M.); 109. Sphecodes gibbus Z. 6, skg. (H.M.). (6) Evandidae: 110. Foenus
sp., nect-lkg. (Budd.). (c) Jchneumonidae: 111. Numerous species (H. M.). (d) Sphe-
gidae: 112. Cerceris quadrifasciata Pz. (H.M.); 113. Ceropales maculatus /’., not
infrequent (H. M.); 114. Crabro cribrarius Z. 9 and 6 (H. M.); 115. C. lapi-
darius Pz. 9 and 4, in great numbers (H. M.); 116. C. vagus Z. 9 and 6 (H. M.);
117. Dinetus pictus /. 9 and 3, in great numbers (H. M.); 118. Gorytes campestris
AMill.g and é(H.M.); 119. G. quadrifasciatus /.$(H.M.); 120. G. quinquecinctus /.

and 6, freq. (H. M.); 121. Mimesa bicolor Jur. (H. M.); 122. M. unicolor v. d. Z.
aa M.). (¢) Mutilidae: 123. Myrmosa melanocephala /. 6 (H.M.). (/) Sphegidae :
124. Nysson maculatus /. 9 (H. M.); 125. N. spinosus forst., nect-lkg. (H. M.);
126. Oxybelus uniglumis Z., freq. (H.M.); 127. Philanthus triangulum /. 9 (H. M.).
(g) Pompilidae: 128. Pompilus neglectus Dahld. (H.M.); 129. P. pictinipes v. d. L.
é (H. M.); 130. P. viaticus Z. 6 (H. M.); 131. Salius exaltatus 7. (H. M.).
(A) Scolitdae: 132. Tiphia femorata #., numerous (H.M.). (¢) TZenthredinidae :
133. Abia sericea Z., not infrequent, nect-lkg. (H. M.); 134. Allantus albicornis /.
9 (H. M.); 135. A. bicinctus Z. (Budd.); 136. A. marginellus Ade. (Budd.);
137. A. nothus K/g., not infrequent (H. M.); 138. A. tricinctus #. (H. M.);
139. Athalia annulata /. (H. M.); 140. A. rosae Z. (H. M.); 141. Hylotoma
caerulescens /. (H. M.); 142. H. enodis Z. (H. M.); 143. H. femoralis A7Zg.
(H. M.); 144. H. rosarum Aig. (H. M.); 145. H. ustulata Z.(H.M.); 146. H.
vulgaris Ade. (H. M.); 147. Macrophya rufipes Z. (H. M.); 148. M. rustica LZ.
(Budd.) ; 149. Tenthredo bifasciata A7g. (Allantus rossii Pz.) freq. (H. M.); 150. T.
sp. (H. M.). (4) Vespidae: 151. Odynerus bifasciatus Z., 9 and 6 (H. M.); 152. O.
gazella Pz. 8 (H.M.); 153. O. parietum Z., numerous (H. M.); 154. O. sinuatus
(H. M.); 135. O. trifasciatus /. 9 (H. M.); 156. Vespa germanica /. 6 and §, freq.
(H. M.); 157. V. rufa Z. 9 (H. M.); 158. V. sylvestris Scop. 6 (H. M.); 159. V.
vulgaris Z. ¥ (H. M.). E. Lepidoptera. (a) Rhopalocera: 160. Thecla betulae
L., skg. (H. M.). (4) Zinerdae: 161. Hyponomeuta sp. (H. M.); 162. Nemotois
scabiosellus Scop. 9, skg. (Budd.). F. Neuroptera. Planipennia: 163. Panorpa
communis Z., nect-lkg., in great numbers (H. M.).

Alfken records the following for Bremen.—

A. Coleoptera. 1. Aromia moschata Z.; 2. Cetonia aurata Z. B. Diptera.
(a) Muscidae: 3. Exorista vulgaris Fall.; 4. Musca domestica Z.; 5. Oliviera
lateralis #.; 6. Pollenia vespillo #.; 7. Trypeta wintheoni MZ. (6) Syrphidae:
8. Arctophila mussitans #.; 9. Ascia lanceolata Mg.; 10. A. podagrica F.;
11. Bacha elongata /.; 12. Chrysotoxum bicinctum Z.; 13. C. festivuam Z.;
14. Helophilus floreus Z.; 15. Merodon albifrons M/g.; 16. Syrphus glaucius /.;
17. S. pyrastri Z.; 18. Volucella bombylans Z.; 19. Xylota segnis Z. C. Hymeno-

DAVIS. II kK k

498 ANGIOSPERMAE—DICOTYLEDONES

ptera. (a) Aprdae: 20. Andrena austriaca Pz. 9; 21. A. flavipes Pz. 9, 2nd gen. ;
22. A. parvula A. 9 and 6; 23. Eriades nigricornis Vy. 9; 24. Halictus calceatus
Scop. 8; 25. Prosopis communis WVy/.9; 26. P. pictipes My/.5; 247. P. punctatissima
Sm. (6) Ichneumonidae: 28. Exyston cinctulus Gr.; 29. Glypta incisa Gr. ;
30. Ichneumon gradarius [Vesm.; 31. Stylocryptus vagabundus #. (c) Sphegidae :
32. Crabro brevis v. d. Z.9 and s; 33. C. subterraneus /. 9 and 6; 34. C. vaga-
’ bundus Pz. 9. (d) Zenthredinidae: 35. Allantus arcuatus Forst.; 36. A. omissus
Forst.; 37. A. vespa Refz.; 38. Athalia glabricollis Zhs.; 39. A. spinarum F.;
40. Entodecta pumila A7g.; 41. Selandria cinereipes K/g. (e) Vespidae: 42. Odynerus
clavipennis Zhs. 9; 43. O. oviventris Wesm. 9; 44. O. parietum Z. 9 and 6.

Loew gives the following for Silesia (‘ Beitrage,’ p. 29).—

A. Coleoptera. Ze/ephoridae: 1. Anthocomus fasciatus Z., nect-lkg.; 2. Axi-
notarsus pulicarius /.,do. B. Diptera. (a) Empidae: 3. Rhamphomyia umbripennis
Mg., skg. (6) Muscidae: 4. Metopia leucocephala Rossz, skg.; 5. Olivieria lateralis 7;
6. Tachina agilis A/g. (c) Mycetophilidae: 7. Sciara thomae ZL. (d) Stratiomyrdae :
8. Stratiomys chamaeleon Deg., skg.; 9. S. equestris AZg., do.; 10. S. furcata F., do.
(e) Syrphidae: 11.Cheilosia mutabilis Fa//.; 12. Helophilus floreus Z.,skg; 13. Syrphus
balteatus Deg.,do.; 14. S. seleniticus AZg.,do.; 15. S. umbellatarum #.,do. C. Hymeno-
ptera. (a) Chrystdidae: 16. Cleptes semiauratus Z. (0) Sphegrdae: 17. Cerceris
labiata #. 3, skg.; 18. C. nasuta Lér. 6, do.; 19. Crabro patellatus Pz., do.; 20. C.
vexillatus Pz. 6, do.; 21. Philanthus triangulum / @, do. (c) TZenthredinidae:
22. Dolerus pratensis Z.; 23. Hylotoma enodis Z., skg.; 24. H. ustulata Z., do.
(d) Vespidae: 25. Polistes gallica Z., skg.

Loew observed the following in Brandenburg (‘ Beitrage,’ p. 36).—

A. Coleoptera. Ceramdycidae: 1. Leptura testacea Z. 9 and 6. B. Diptera.
(a) Muscidae: 2. Gymnosoma rotundata Z.; 3. Phasia crassipennis / C. Hymeno-
ptera. (a) Apidae: 4. Halictus leucozonius Schr. 9, skg. (6) Scoliddae: 5. Tiphia
femorata 9 and 6. (c) Sphegzdae: 6. Crabro albilabris F. 9, do.; 7. C. subterraneus
F. 9, skg.; 8. Hoplisius quadrifasciatus #., do.; 9. Mellinus arvensis Z. 9; 109.
Pompilus quadripunctatus #. (d) TZenthredinrdae: 11. Allantus scrophulariae LZ.
(e) Vesprdae: 12. Vespa germanica F. 9, skg.

Loew records the following for the Riesengebirge (R.), Silesia (S.), and Glatz
(G.) (‘ Beitrige,’ p. 48).—

A.Coleoptera. (a) Cerambycidae: 1. Clytus arietis Z. (S.); 2. C. mysticus Z.
(S.); 3. Leptura testacea Z. 9 and $ (S.); 4. Strangalia annularis /. (R.); 5. S.
bifasciata Afill. 9 and & (R.). (4) Scarabaeidae: 6. Trichius fasciatus Z. (R.).
(c) Lelephoridae: 7. Cantharis alpina Payk. (S.). B. Diptera. (a) Bibionzdae:
8. Bibio pomonae /. (R.). (4) Conopidae: 9. Conops quadrifasciatus Deg. (G.).
(c) Muscidae: 10. Gymnosoma rotundata Z. (G.); 11. Leucostoma analis Ag. (G.).
(¢) Pipunculidae: 12. Pipunculus ruralis AZg. (G.). (e) Syrphidae: 13. Cheilosia
oestracea LZ. (S.); 14. Chrysotoxum octomaculatum Curv. (S.); 18. Syrphus glaucius
£. (S.). C. Hymenoptera. (a) Apidae: 16. Halictus albipes /. 4, skg. (G.);
17. H. morio #. 9, do. (G.). (6) Chrysididae: 18. Chrysis ignita Z. (G.).
(c) Sphegidae: 19. Crabro cribrarius Z. 9 and 8 (S.); 20. Mellinus arvensis Z. (G.)
21. M. sabulosus F. (S.).

Sickmann gives the following list for Osnabriick.—

Hymenoptera. (a) Sphegidae: 1. Ammophila sabulosa Z., fairly freq. ;
2. Calicurgus fasciatellus Sfz7., not freq.; 3. Crabro alatus Pz., do.; 4. C. cetratus
Shuck. do.; 5. C. chrysostomus Zep., very common; 6. C. cribrarius Z., do.;
7. C. dives A.-Sch., rare; 8. C. exiguus v. d. L., fairly freq.; 9. C. gonager Lep.,
very rare; 10. C. guttatus vd. Z. 9; 11. C. larvatus Wesm.9; 12. C. lituratus
fz., rare; 13. C. podagricus 7. d. Z., freq.; 14. C. sexcinctus /., very common ;

’

UMBELLIFERAE 499

15. C. spinicollis H-Sch., freq.; 16. C. subterraneus #, fairly freq.; 17. C. varius
Lep., freq.; 18. Dahlbomia atra F.; 19. Gorytes laticinctus Shuck., infrequent ;
20. C. mystaceus L., freq.; 21. G. quadrifasciatus /, do.; 22. Mellinus sabulosus F.,
do.; 23. Mimesa bicolor Jur.; 24. M. dahlbomi Wesm., rare; 25. M. equestris F.,
very common; 26. Nysson maculatus /, fairly freq.; 27. Oxybelus uniglumis Z..,
very common; 28. Pemphredon unicolor F, freq.; 29. Pompilus abnormis Dahio.,
rare; 30. P. nigerrimus Scop., freq.; 31. P. trivialis Dahlb.; 32. Psen atratus P2z.,
very common; 33. Pseudagenia carbonaria Scop., do.; 34. Salius exaltatus 7, do. ;
35- S. notatus Lep., freq.; 36. S. obtusiventris Schjéd/e, very rare; 37. Trypoxylon
attenuatus Sm., rare. (6) JMusilidae: 38. Myrmosa melanocephala F. é.

Friese observed the following.—

In Baden, 2 bees (Andrena austriaca Pz. 9, and Halictus minutus Schenck
(=!H. rugulosus Schenck) 9). In Thuringia.—Hymenoptera. (a) Apidae:
1. Andrena austriaca Pz.; 2. A. coitana K. (5) Ichneumonidae: 3. Amblyteles
(Ctenichneumon) funereus /ourcr.; 4. A. (Protichneumon) fucipennis Wesm. ;
5. Metopius micratorius Gr. (c) Mutillidae: 6. Mutilla rufipes /, var. nigra
Rosst. (d) Sphegidae: 7. Crabro alatus Pz.; 8. C. cribrarius Z.; g. C. lituratus
Pz.; 10. Oxybelus nigripes Ov.; 11. Pemphredon lugens Dah/d.; 12. Pompilus
quadripunctatus # (e) Zenthredinidae: 13. Allantus marginellus /; 14. A. vespa
Retz. (f) Vespidae: 15. Discoelius zonalis Pz.; 16. Odynerus crassicornis Pz. ;
17. O. sinuatus #; 18. Vespa austriaca Pz.

Schenck gives the following for Nassau.—

Hymenoptera. (a) Apzdae: 1. Andrena austriaca Pz.; 2. A. nana X.;
3. Halictus interruptus Pz.; 4. Prosopis trimaculata Schenck. (6) Mutillidae:
5. Myrmosa melanocephala © 9 and 8. (c) Scolttdae: 6. Tiphia femorata F ;
4. T. minuta v.d. LZ. (d) Sphegidae: 8. Ceropales maculatus #.; 9. C. variegatus
F.; 10. Dahlbomia atra Pz.; 11. Gorytes levis Z/r.; 12. G. mystaceus Z.; 13. G.
quadrifasciatus #.; 14. G. quinquecinctus #.; 15. Pompilus anceps Sm.; 16. P.
trivialis Dahld.; 17. Pompilus unicolor Spzz.; 18. Psen atratus Pz.; 19. Tachysphex
pectinipes Z.

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), 11 beetles, 5 flies, and 5 Hymenoptera (‘ Alpenblumen,’
pp. 121-2). Schmiedeknecht (Thuringia),—4 bees—1. Andrena austriaca Pz.;
2. A. dubitata Schenck, 2nd gen.; 3. A. fulvicrus A. (=A. flavipes Pz.), do.;
4. Nomada obtusifrons My Krieger (Leipzig), 2 bees (Andrena austriaca Pz., and
A. denticulata X.), and 2 Sphegids (Mellinus sabulosus /. 9, and Mimesa atra F.).
Rossler (Wiesbaden) the moth Grapholitha aurana /., group aurantiana ollar.
Schiner (Austria), 2 Muscids—1. Frontina laeta M/g.; 2. Homalomyia pretiosa Schin.
Von Fricken (Westphalia and Eastern Prussia), the Scarabaeid Hoplia philanthus Sw/z.,
and the Cerambycid Acmaeops collaris Z. von Dalla Torre (Tyrol), the ruby-wasp
Chrysis austriaca #., the true wasp Odynerus minutus 7, and 3 bees—r1. Eriades
campanularum &.3,; 2. Halictus morio /.; 3. Osmia leucomelaena K.é. Schletterer
(Tyrol), 2 bees—1. Halictus morio /.; 2. Nomada succincta Pz.: (Pola) the Scoliid
Tiphia femorata ¥. Kohl (Tyrol), 2 true wasps—1. Odynerus spiricornis Spzm. ;
2. Eumenes arbustorum Pz., var. dimidiata Brud/.; the ruby-wasp Ellampus caeruleus
Dahid.; and 2 fossorial wasps—1. Crabro cribrarius Z.; 2. C. scutellatus Schev.
Handlirsch, 3 fossorial wasps—1. Gorytes bilunulatus Cos/a (on the authority of
Schmiedeknecht); 2. G. quadrifasciatus #.; 3. G. quinquecinctus #. Redtenbacher
(Vienna), the Cerambycid Callimus cyaneus #. Knuth (Helgoland).—A. Diptera.
Muscidae: 1. Coelopa frigida Fall.; 2. C. pilipes Aal.; 3. Lucilia caesar ZL. ;
4. Olivieria lateralis #.; 5. Scatella sp.; 6. Scatophaga stercoraria Z.; 7. Medium-
sized and minute sp. B. Hymenoptera. Vespidae: 8. One of the wasps, which
escaped. MacLeod (Flanders), 3 short-tongued bees, a saw-fly, a true wasp, an

Kk 2

500 ANGIOSPERMAE—DICOTYLEDONES

Ichneumonid, 6 hover-flies, 4 Muscids, a Lepidopterid, and 2 beetles (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 275-7, 380). Heinsius (Holland), 2 Muscids—
1. Lucilia cornicina #.; 2. Scatophaga stercoraria Z. 9 and 6; and the hover-fly
Eristalis tenax Z. 9 (Bot. Jaarb. Dodonaea, Ghent, iv, 1892, p. 59). Scott-Elliot
(Dumfriesshire), Apis, a humble-bee, a wasp, 4 Muscids, and 4 hover-flies (‘ Flora‘of
Dumfriesshire,’ p. 80).

Loew observed the following in Switzerland (‘ Beitrige,’ p. 56).—

A. Coleoptera. Cerambycidae: 1. Leptura maculicornis Deg.; 2. Pachyta
lamed Z.; 3. P. quadrimaculata Z.; 4. P. virginea Z.; 5. Strangalia armata AZds?.
B. Diptera. (a) Muscidae: 6. Echinomyia fera ZL.; 7. Hydrotaea dentipes / 3;
8. Mesembrina meridiana Z.; 9. M. mystacea Z. (4) Syrphidae: 10. Eristalis
rupium /.; 11. Melithreptus pictus 47g.; 12. Syritta pipiens Z. C. Hymenoptera.
Sphegidae: 13. Gorytes sp.; 14. Myrmosa melanocephala /. &.

H. de Vries noticed the following in the Netherlands (Ned. Kruidk. Arch.,
Nijmegen, 2. ser., 2. Deel, 1875).—

Hymenoptera. (a) Apzdae: 1. Apis mellifica Z.Y; 2. Bombus agrorum F. ¥ ;
3. B. subterraneus Z. ¥; 4. B. terrester Z. ¥ and 6; 5. Halictus cylindricus /. 6.
(4) Sphegidae: 6. Crabro cribrarius Z.9. (c) Zenthredinidae: 7. Allantus tricinctus
f (d) Vespidae: 8. Vespa germanica F. ¥.

1170. H. pyrenaicum Lam. (=H. montanum Schleich.).—The flowers of this
species are white.

Vistrors.—MacLeod (Pyrenees) observed 6 short-tongued Hymenoptera, 4
beetles, and 26 Diptera (8 Syrphids and 16 Muscids).

1171. H. sibiricum L.—Lindman says that the flowers of this species possess
a strong urinous odour.

Visrrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Cistelzdae: 1. Cistela sulphurea Z. (6) Scarabaeidae :
2. Cetonia aurata Z., dvg. the flowers; 3. Phyllopertha horticola Z.,do. B. Diptera.
Syrphidae: 4. Eristalis arbustorum Z., skg.; 5. E.nemorum Z., do. C. Hymenoptera.
Apidae: 6. Apis mellifica Z. 9, skg. and po-cltg.

Lindman saw numerous Diptera and Hymenoptera on the Dovrefjeld.

1172. H. dissectum Ledeb. (=H. lanatum AZichx.).—

VisiTors.—Loew saw the bee Andrena schrankella Wy/. 9, skg. and po-dvg., in
the Berlin Botanic Garden.

1173. H. pubescens Bieb.—

Visirors.—Loew saw the following in the Berlin Botanic Garden. —

A. Coleoptera. Scaradaerdae: 1. Cetonia aurata £., devouring the flowers.
B. Hymenoptera. Afrdac: 2. Apis mellifica /. ¥, skg. and po-cltg. ; also (on the
var. V}2/helmsz? Fisch. et Lall.), the Muscid, 3. Pollenia rudis /.

366. Tordylium Tourn.

1174. T. maximum L.— Schulz (‘ Beitrage,’ II, p. 190) describes this species as
andromonoecious, with protandrous hermaphrodite flowers.

1175. T. apulum L.—
Visttors.—Schletterer observed the following at Pola.—

UMBELLIFERAE 501

Hymenoptera, (a) Apidae: 1. Andrena carbonaria Z.; 2. A. parvula &.;
3. A. taraxaci Grr., infreq.; 4. Halictus calceatus Scop.; 5. H. levigatus K. 9;
6. H. minutus K.; 7. H. moris #.; 8. H. quadrinotatus X.; 9. H. variipes Aor.
(3) Braconidae: 10, Bracon urinator #. (c) Chrysididae: 11. Chrysis angustifrons
Ab.; 12. C. inaequalis Dahid.; 13. Ellampus auratus Z.; 14. Hedychrum longicolle
Ab. (d) Evanitdae: 15. Gasteruption granulithorax Zourn.; 16. G. terrestre Tourn.
(e) Ichneumonidae: 17. Amblyteles armatorius For's/. (=A. fasciatorius /.); 18. Angitia
armillata Gr.; 19. Anilasta notata Gr.; 20. Cryptius hellenicus Schmzedekn.; 21. C.
viduatorius #.; 22. Hoplocryptus heliophilus Zschek.; 23. Ichneumon bilunulatus Gr.;
24. I. finitimus Z7schd.; 25. I. scanthorius Forst.; 26. Omorga mutabilis Agr. ;
27. Pimpla instigator #.; 28. P. roberator #.; 29. Trychosis plebeia Zschek., with
the vars. nigricornis Archd. and nigritarsis Archd. (/) Pompilidae: 30. Pompilus
minutus Dahlé.(=P. cellularis Dahld.); 31. P. sexmaculatus Spzv.; 32. P. viaticus Z.;
33. Salius fuscus #.; 34. S. parvulus Dahl. (g) Scoliidae: 35. Tiphia minvta v.d.Z.;
36. T. morio #.; 37. T.femorata #. (A) Sphegidae: 38. Cerceris quadrifasciata Pz.;
39. Crabro clypeatus Z.; 40. C. meridionalis Costa; 41. Diodontus minutus /. 6;
42. Gorytes pleuripunctatus Cosfa. (7) Zenthredinidae: 43. Allantus fasciatus Scop.;
44. A. viduus Ross?; 45. Amasis laeta #.; 46. Arge cyaneocrocea Férst.; 47. A.
melanochroa Gmel.; 48. Athalia annulata #.; 49. A. glabricollis Zhs., freq.; 50. A.
spinarum /.; 51. A. rosae Z., var. cordata Lep.; 52. Macrophya rustica Z.
(4) Vespidae: 53. Polistes gallica LZ.

367. Siler Crantz.

1176. S. trilobum Crantz (=Laserpitium aquilegiaefolium /acg.).— Schulz
(‘ Beitrage,’ II, pp. 85-6, 190) describes this species as andromonoecious, with
protandrous hermaphrodite flowers. The male flowers are numerous, and situated
in the middle of the umbellules. According to Kerner, the direction of the styles
and the position of the stigmas remain unchanged, but the slender filaments elongate
and curve in such a way that the pollen is applied to the stigmas of neighbouring
flowers (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 324).

Visitors.—Loew observed the following in Steiermark (‘ Beitrige,’ p. 48).—

A. Coleoptera. 1. Anoncodes rufiventris Scop.; 2. Chrysanthia viridissima L.;
3. Oxythyrea stictica Z.; 4. Strangalia armata Hés/. B. Diptera. Muscidae:
5. Clytia pellucens Fa//.; 6. Echinomyia ferox Pz. C. Hemiptera. 7. Nabis
sp.; 8. Graphosoma lineatum Z.; 9. Undetermined sp. And in the Berlin Botanic
Garden—A. Diptera. Syrphidae: 1. Eristalis nemorum Z., skg.; 2. Syritta
pipiens Z., do.; 3. Syrphus ribesii 2. B. Hymenoptera. (a) Apzdae: 4. Andrena
tibialis K. 9, skg. and po-cltg.; 5. Apis mellifica Z.¥§, do.; 6. Prosopis communis
Nyl. 9, skg. (4) Vesprdae: 7. Odynerus parietum LZ.

368. Laserpitium L.

1177. L. latifolium L.—The flowers of this species are white or rarely reddish
in colour. Schulz (‘Beitrage,’ I], pp. 90, 94, 190) describes them as andro-
monoeciously distributed, those which are hermaphrodite being markedly protandrous.
The primary umbels usually bear hermaphrodite flowers only, those of higher order
chiefly male ones. According to Kerner (‘Nat. Hist. Pl.,’ Eng. Ed. 1, I, p. 296),
there are short-stalked pseudo-hermaphrodite male flowers in all the umbellules,
surrounded by long-stalked, truly hermaphrodite ones.

1178. L. prutenicum L.—Schulz states that the yellowish-white flowers of this
species are all hermaphrodite, and markedly protandrous (‘Beitrage,’ II, p. 190).

502 ANGIOSPERMAE—DICOTYLEDONES

1179. L. hirsutum (=L. Panax Gouan)—Hermann Miiller describes the
white flowers of this species as protandrous (‘Alpenblumen,’ p. 122).

Visrtors.—Herm. Miiller observed a beetle, 23 Diptera (including 17 Muscids),
7 Hymenoptera, and 3 Lepidoptera, in the Alps.

369. Daucus L.

80. D. Carota L.—Schulz (‘Beitrage,’ II, pp. 86-9, 91, 93, 190) describes
this species as andromonoecious, with white flowers, of which the hermaphrodite ones
are markedly protandrous. For Holland monoecism has also been recorded, and in
Central Germany purely female flowers have been observed. The marginal blossoms,
as in many Umbelliferae, are much bigger than the others, their outer petals being
specially enlarged. Schulz was able to distinguish two varieties in the Tyrol and
Central Germany. The commoner bears white hermaphrodite and male flowers in
the same umbellule ; the latter being central and more numerous in the umbels of
higher order. The second rarer variety often bears flowers tinged with green or
ted; their umbellules either bear female flowers only, or these and neuter ones. In
many cases the anthers contain normal pollen-grains, often mixed with smaller
irregular ones. They rarely dehisce, and remain in the position which they occupied
in the bud. The terminal flowers are either female or neuter.

According to Warnstorf, the umbellules of primary umbels bear hermaphrodite
flowers at Ruppin; secondary ones marginal hermaphrodite and, as a rule, a few
central male flowers; tertiary umbels are almost exclusively male. The terminal
flower in the umbellules of secondary umbels is commonly hermaphrodite ; rarely
all the umbels are female by degeneration of the anthers.

A very remarkable feature is the occurrence of an enlarged actinomorphous
central flower of a purple-red colour. In rare cases several may be present, even as.
many as 5 to 10. This flower is sometimes wanting, e.g., according to Buchenau
(‘Flora d. ostfr. Ins.,’ p. 143), in some parts of the East Frisian Islands. It is not
often met with in the North Frisian Islands (Knuth, ‘Flora d. nordfr. Ins.,’ p. 67).
Schulz states that a terminal umbellule occurs in at most 3-5 % of all the umbels,
and only a small fraction of this number possess one or more of the purple-red
flowers (Bot. Centralbl., Cassel, li, 1892, p. 12). Kronfeld describes these flowers
as cleistogamous and fertile; he regards them as due to inherited gall-formation.
(Cf. Justs bot. Jahresber., Leipzig, xx, (1892) 1894, p. 491.)

The umbellules of plants observed by Beijerinck (‘Gynodioeciae bei Daucus
Carota’) at Wageningen in Holland possessed central male and marginal female
flowers, often with a single terminal hermaphrodite one as well. In plants examined
by Staes (Bot. Jaarb. Dodonaea, Ghent, i, 1889, p. 124) in Belgium (Ghent and
Blankenberg) the marginal flowers were hermaphrodite instead of female. The
form with reddish blossoms considered by Beijerinck to be physiologically female
not infrequently bears hermaphrodite flowers, and can therefore be fertilized
independently of the white-flowered variety. Schulz has not observed the forms
described by Beijerinck, either in Central Germany or in the Tyrol.

The two different forms of umbels of this species are described as follows
by Beijerinck (op. cit.) and Staes (op. cit.), for Holland (Wageningen) and Belgium
(Ghent and the Blankenberg dunes) respectively.—

UMBELLIFERAE

5°3

1. Ferm with white flowers.

At Wageningen, according to Bei-
jerinck, the marginal flowers of all the
umbellules are either devoid of stamens,
or if these are present they drop off
before their anthers dehisce.

These flowers are therefore always
female.

At Ghent and Blankenburg, ac-
cording to Staes, the stamens may be
vestigial, but are usually fertile.

These flowers may therefore be fe-
male, but are often hermaphrodite.

2. Form with red or greenish-red flowers.

At Wageningen the stamens are
often more or less metamorphosed into
petals; the anthers never dehisce.

The whole umbel is female.

At Wageningen this form can only
reproduce with the aid of the white-
flowered one.

At Ghent and Blankenberg the
stamens, when not degenerate,
possess anthers that dehisce.

The umbel may be female (owing
to degenerate stamens, or anthers which
do not dehisce) ;
hermaphrodite.

At Ghent and Blankenberg the two
forms can reproduce independently of
each other.

often

it is often, however,

Deichmann (Bot. Centralbl., Cassel, xlix, 1892, p. 271) calls attention to the
fact that in Denmark, owing to frequent crossing of the cultivated variety with the
wild form the former displays an unwelcome reversion.

Visitors.—I observed the following at Gliicksburg (‘ Bloemenbiol. Bijdragen ’).—

A. Coleoptera. (a) Coccznellidae: 1. Coccinella septempunctata Z. (6) Zele-
phortdae: 2. Cantharis fusca Z. B. Diptera. (a) Muserdae: 3. Lucilia caesar L.
(6) Syrphidae: 4. Syrphus balteatus Deg. C. Hymenoptera. Afrdae: 5. Apis
mellifica Z. $; 6. Bombus terrester Z. 9. All skg., or po-cltg., or po-dvg.: the
last running over the inflorescences with great rapidity.

In Schleswig-Holstein I saw 4 hover-flies, a Muscid, a humble-bee, and a
fossorial wasp (‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 155); in Helgoland 3 Muscids—
1. Coelopa frigida Fal/.; 2. Fucellia fucorum Fa//.; 3. Scatophaga stercoraria L.
(BI. u. Insekt. a. Helgoland,’ p. 35).

Sickmann records the following for Osnabriick.—

Hymenoptera. (a) Sphegidae: 1. Astata minor ofl. freq.; 2. Cerceris
labiata &, do.; 3. C. quinquefasciata Rossz, do.; 4. Ceropales maculatus /., do.;
5. Crabro alatus Pz.; 6. C. albilabris #., very common; 7. C. armatus v. d@. L.
3; 8. C. brevis v.d. Z., freq.; 9. C. clypeatus Schreb.; 10. C. cribrarius Z., very
common; 11. C. distinguendus A. M/or.; 12. C. elongatus v. d. L., freq.; 13. C.
exiguus v. d. L., fairly freq.; 14. C. palmarius Schred., infreq.; 15. C. peltarius
Schreb., very common; 16. C. pygmaeus zv. d. Z., rare; 17. C. scutellatus Schev.;
18. C. sexcinctus #, freq.; 19. C. vagabundus Pz., do.; 20. C. wesmaéli v. d. L.,
infreq.; 21. Gorytes fallax Handi.9; 22. G. quadrifasciatus #.; 23. G. quinque-
cinctus /., rare; 24. Mellinus sabulosus /., freq.; 25. Mimesa equestris /., very
common; 26. Oxybelus bipunctatus Olv., freq.; 27. O. nigripes Odv., infreq. ;
28. O. uniglumis Z., very common; 29. Pemphredon shuckardi A. JZor., freq. ;
30. Pompilus pectinipes 7. ¢. Z., var. campestris Wesm., freq.; 31. P. viaticus Z.,
very common; 32. P. wesmaéli Zhms., infreq.; 33. Psen atratus Pz., very common;

504. ANGIOSPERMAE—DICOTYLEDONES

34. Pseudagenia carbonaria Scop., do.; 35. Salius affinis v. d. Z., rare; 36. S.
exaltatus /., very common; 37. S. notatus Zep., freq.; 38. Trypoxylon figulus Z.,
do. (4) Scoliidae: 39. Tiphia femorata &, very common; 40. T. minuta z. d. Z.,
infreq. (c) Mutillidae: 41. Myrmosa melanocephala F. é.

Friese noticed the following in the middle Saalthal._—

Hymenoptera. (a) Apidae: 1. Andrena convexiuscula K.; 2. Bombus
terrester Z. (0) Chrysididae: 3. Chrysis callimorpha AZocs.; 4. C. fulgida L.;
5. C. inaequalis Dahib.; 6. Splendidula Ross?; 7. C. succincta Z.; 8. C. viridula Z. ;
g. Cleptes nitidulus ¥.; 10. Ellampus scutellaris Pz.; 11. Holopyga curvata Fors.
(c) Ichneumonidae: 12. Amblyteles oratorius (7.) Wesm.; 13. A. (Ctenichneumon)
repentinus (Gr.) Zhs.; 14. Hellwigia elegans Gr.; 15. Ichneumon leucamelas (/-)
Wesm.; 16. I. (Protichneumon) similatorius (4) Zhs.; 17. Lissonota maculatoria
Ross? ; 18. Microcryptus curvus (Gr.) Tho. (d) Mutillidae: 19. Mutilla rufipes F.,
var. nigra Ross. (e) Scolivdae: 20. Scolia quadripunctata #. (/) Sphegidae: 21.
Astata boops Schr.; 22. Didineis lunicornis /.; 23. Gorytes levis L/r.; 24. Nysson
maculatus #. (g) Tenthredintdae: 25. Allantus marginellus #.; 26. Cladius
pectinicornis Mourcr.; 27. Cyphona furcata V7/l.

Burkill records the following for the east coast of Scotland (‘Fls. and Insects in
Gt. Britain,’ Part I).—

A. Coleoptera. (a) Chrysomehdae: 1. Crepidodera ferruginea Scop. (6) Witt-
dulidae: 2. Cercus rufilabris Zir.; 3. Meligethes picipes Sturm. (c) Staphylinidae :
4. Tachyporus obtusus Z. B. Diptera. (a) Muscrdae: 5. Anthomyia brevicornis
Zit.; 6. A. radicum Z., very common; 7. Calliphora erythrocephala JZg., skg. ;
8. C. vomitoria Z., do.; 9. Hydrellia griseola Fadll.; 10. Lucilia cornicina F.;
11. L. sylvarum AZ/g.; 12. L. splendida MZg.; 13. Morellia sp.; 14. Oscinis frit Z.;
15. Pollenia rudis #.; 16. Sarcophaga, 2 sps.; 17. Drosophila graminum Feil. ;
18. Scatophaga stercoraria Z., skg.; 19. Sepsis cynipsea LZ. (0) Phoridae: 20. Phora
sp. (c) Syrphidae: 21. Eristalis arbustorum Z., skg.; 22. E. pertinax Scop., do. ;
23. E. tenax Z., do.; 24. Melanostoma scalare #.; 25. M. barbifrons Fail. ;
26. Paragus sp.; 27. Platycheirus albimanus #.; 28. Sphaerophoria scripta ZL. ;
29. Syritta pipiens Z., skg.; 30. Syrphus ribesii L., do. (d) Chtronomidac :
3r. Ceratopogon niger Winn. (e) Psychodidae: 32. Pericoma sp. (/) ALpeeto-
philidae: 33. Sciara sp., freq. ©. Hymenoptera. (a) Apidae: 34. Bombus
hortorum Z., one, skg. (3) Formicidae: 35. Formica fusca Z., skg.; 36. Myrmica
rubra L., do. (¢) Lchneumonidae: 37. 25 undetermined sps. (@) Sphegrdae:
38. Priocnemis pusillus Schjidle. (e) Tenthredinidae: 39. Allantus arcuatus
Forst., skg.

Herm. Miiller gives the following list (‘ Fertilisation,’ pp. 285-6; ‘ Weit. Beob.,’
I, p. 307) for himself (H. M.) and Buddeberg (Budd.).—

A. Coleoptera. (a) Cerambyctdae: 1. Strangalia armata Hdst. (H. M.);
2. S. bifasciata AZiiller (H. M.). (6) Clertdae: 3. Trichodes apiarius Z., nect-lkg.,
(Budd.). (¢) Coceznelidae: 4. Coccinella mutabilis Scrzba, nect-lkg. (Budd.); 5. C.
quinquepunctata Z., do. (H.M.). (d) Curculionidae: 6. Spermophagus cardui Sve.
(H. M.). (e) Dermestidae: 7. Anthrenus pimpinellae /. (H. M.). (/) Elatertdac:
8. Agriotes gallicus Zac. (H. M.); 9. A. sputator Z. (H. M.); 10. A. ustulatus
Schall. (H. M.). (g) Scarabaeidae: 11. Trichius fasciatus Z.(H.M.). (A) Tele-
phoridae: 12. Danacea pallipes Pz. (H. M.); 13. Telephorus melanurus /, in
copula, nect-lkg. (H. M.). (¢) M@Jordellidae: 14. Mordella aculeata Z. (H. M.);
15. M. fasciata #. (H. M.). B. Diptera. (a) Bomébyliidae: 16. Anthrax flava Ag.
(H. M.). (6) Aluscrdae: 17. Gymnosoma rotundata Z. (H. M.); 18. Species of
Lucilia (H. M.); 19. Phasia crassipennis /. (Budd.); 20. Sarcophaga albiceps A/g.
(H. M.); 21. Species of Sepsis (H. M.). (c) Stratiomyidae: 22. Stratiomys

UMBELLIFERAE 505

chamaeleon Deg., freq. (H. M.); 23. S. riparia Afg., do. (H. M.). (d@) Syrphrdae:
24. Ascia podagrica #. (H. M.); 25. Cheilosia barbata Loew, skg. (H. M.); 26. C.
soror Ze/t.(H.M.); 24. C. variabilis Pz., skg. (Budd.); 28. Chrysogaster viduata Z.
(H. M.); 29. Eristalis arbustorum Z. (H. M.); 30. E. sepulcralis Z. (H. M.);
31. Helophilus floreus Z. (H. M.); 32. Melithreptus scriptus Z.(H.M.); 33. Pipiza
funebris /.(H. M.); 34. Pipizella annulata Jfacg. (H. M.); 35. Syritta pipiens Z.
(H. M.); 36. Syrphus pyrastri Z.(H. M.). C. Hemiptera, 37. Graphosoma
nigrolineata /. freq. (H. M.). D. Hymenoptera. (a) Apidae: 38. Andrena nana
K. 9, skg. (H. M.); 39. A. parvula A. (H. M.); 40. Halictus albipes #. $ (H. M.);
41. H. levis A. 6 (H. M.); 42. H. interruptus Pz. 9 (H. M.); 43. Nomada lateralis
Pz. 9 (H. M.); 44. Prosopis sinuata Schenck & (H.M.); 45. P. variegata F. 6
(H. M.); 46. Specodes gibbus Z. (H. M.). (2) Chrystdidae: 47. Hedychrum
lucidulum /. 6 and 9, freq. (H. M.). (c) Jchneumonidae: 48. Various sp. (H. M.).
(2) Sphegidae: 49. Cerceris variabilis Schr. 9 (H. M.); 50. Ceropales maculatus F”.
(H. M.). (e) Alutlidae: 51. Mutilla europaea LZ. 8 (H. M.); 52. Oxybelus
bipunctatus Ov. (H.M.); 53. O. uniglumis Z., freq. (H.M.); 54. Pompilus
intermedius Schenck (H.M.); 55. P. neglectus Daklé. (H.M.); 56. P. niger
(H.M.); 57. P. viaticus Z. § (H.M.); 58. Priocnemis obtusiventris Schjéd/e (H. M.).
(f) Scolitdae: 59. Tiphia femorata #., numerous (H. M.). (g) Zenthredinidae :
60. Allantus nothus A7vg. (H. M., Budd.); 61. Athalia rosae Z. (H. M.); 62.
Hylotoma femoralis A/g. (H. M.); 63. H. rosarum A7g., nect-lkg. (Budd.);
64. H. ustulata Z. (H. M.); 65. Selandria serva #. (H. M.). (h) Vesprdae:
66. Odynerus sinuatus /. 9 (H. M.). E.. Lepidoptera. (a) Rhopalocera:
67. Hesperia lineola O., skg. (H. M.): 68. Spilothyrus alceae Zp. (Budd.).
(2) Zenerdae: 69. Nemotois Hézn. sp.,skg.(H.M.). F.Neuroptera. Planipennia:
470. Hemerobius (H. M.).

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), a beetle and 2 Lepidoptera (‘ Alpenblumen,’” p. 122).
Wiistnei (island of Alsen), the bee Halictus nitidiusculus A. Alfken (Bremen), 3 bees
(1. Andrena austriaca Pz. 9 and é; 2. A. hattorfiana #6; 3. A. parvula A. 9), and 3
saw-flies (1. Allantus omissus Férs/.; 2. Poecilostoma luteolum A7g.; 3. Tenthredo
coryli Pz.). Krieger (Zwickau), the rare fossorial wasp Nysson dimidiatus /wr.
Schmiedeknecht (Thuringia), 4 bees (1. Andrena austriaca Pz.; 2. A. combinata
Chr.; 3. A.lucens /mf.; 4. A.nana .). Loew in Brandenburg (B.) and Mecklenburg
(M.) (‘ Beitrage,’ p. 36)—A. Diptera. Afuscrdae: 1. Xysta cana Afg., skg. (M.).
B. Hymenoptera. (a) Jchneumonidae: 2. Undetermined sp. (M.). (2) Sphegidae :
3. Cerceris interrupta Pz. 9, skg.(B.); 4. Mellinus sabulosus F., do. (M.): in Silesia, the
hover-fly Eristalis horticola Deg., skg.: in Steiermark (op. cit., p. 48), a Muscid (Phasia
analis #.) and a bee (Andrena parvula K. 9, po-cltg.): in Switzerland (op. cit., p.55).—
A.Coleoptera. (a) Cerambycidae: 1.Leptura sanguinolenta Z. (0) Cleridae: Trichodes
apiariusZ. B.Diptera. (2) Stratiomyzdae: 3.Stratiomys longicornis Scop.9, var. (0) Syr-
phidae: 4. Cheilosiaimpressa Lw.; 5. Syrphus lasiophthalmus Ze//.; 6.5. umbellatarum
F. (c) Tabanidae: 7. Tabanus auripilus M7g., var. aterrimus J/g. 9; 8. T. infuscatus Zw.
Schenck (Nassau).—Hymenoptera. <Apzdae: 1. Andrena austriaca Pz.; 2. A. nana
K.; 3. Prosopis variegata /. (0) Sphegidae: 4. Ceropales maculatus /; 5. C.variegatus
¥.; 6. Gorytes levis Z/r.; 7. Tachysphex pectinipes Z. (c) Mutillidae: 8. Mutilla
rufipes /., var. nigrita Pz. F. F. Kohl (Tyrol), the true wasp Odynerus parietum Z..,
and the fossorial wasp Crabro arbitrarius Z. Handlirsch, 2 fossorial wasps (Gorytes
levis Zir., and G. quadrifasciatus #.). Schiner (Austria), a Syrphid (Cheilosia
impressa Zw.), and 6 Muscids (1. Alophora hemiptera #.; 2. Clairvillia ocypterina
R.-D.; 3. Germaria ruficeps /.; 4. Miltogramma ruficornis MZ/g.; 5. Plesina nigri-
squama Ze/f.; 6. Siphona geniculata Deg.). MacLeod (Flanders), 9 short-tongued
Hymenoptera, 5 Syrphids, 3 other flies, 2 beetles, and a Lepidopterid (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, pp. 278-9). H. de Vries (Netherlands), the fossorial wasp

506 ANGIOSPERMAE--DICOTYLEDONES

Ceropales maculatus /, and the Scoliid Tiphia femorata #6 and 9 (Ned. Kruidk.
Arch., Nijmegen, 2. Ser., 2. Deel, 1875). Heinsius (Holland), a Stratiomyid (Stratiomys
furcata #. 9), and 4 beetles (1. Agriotes obscurus Z.; 2. Cistela sulphurea Z.;
3- Coccinella septempunctata Z.; 4. Cantharis fulva Scop. (=Telephorus melanurus
F-)) (Bot. Jaarb. Dodonaea, Ghent, iv, 1892, p. 59). MacLeod (Pyrenees), 12
Hymenoptera, 4 beetles, and 7 flies (op. cit., iii, 1891, p. 407). Saunders (England),
the Scoliid Tiphia femorata /.

370. Orlaya Hoftm.

u81. O. glandifiora Hoffm. (=Daucus grandiflorus Scop.). (Herm. Miiller,
‘Weit. Beob.,’ I, pp. 307-10; Schulz, ‘ Beitrage,” II, pp. 86, 91-2, 190.)—This
white-flowered species is andromonoecious, according to Schulz, but Hermann Miller
says that it may also be gynomonoecious. The hermaphrodite flowers are homo-
gamous. The central flowers of the umbellules are male, with vestigial ovaries, but
neither styles nor stigmas; their petals are small and incurved. The marginal
flowers are hermaphrodite, or sometimes (Herm. Miiller) fertile female ones; the
external petal is enlarged. Sometimes the reproductive organs of male, female, and
hermaphrodite flowers are completely abortive. In each flower at the margin of an
entire umbel the external petal is enlarged into a deeply bifid expansion over
1 cm. long.

Although the plant grows among corn, it is very conspicuous, owing to the
large marginal flowers, and the inflorescences are visited by many insects, which
effect the crossing of different umbels and frequently of different stocks, if they
alight on the margins of the umbels. Schulz says that automatic self-pollination
is possible only for a short time in the homogamous hermaphrodite flowers, as the
stamens quickly bend outwards.

Vistrors.—Schulz (Tyrol) noticed numerous flies and beetles, more rarely small
Hymenoptera.

Herm. Miiller observed the following in Thuringia.—

A. Diptera. (a) Bombylidac: 1. Ploas grisia F, skg. (0) Empidae: 2. Empis
livida Z., skg. (c) SyrpAddae: 3. Syritta pipiens Z., freq. (d) Muscidae: 4. Species
of Anthomyia; 5. Gymnosoma rotundata Z., skg.; 6. Ocyptera brassicaria F",
do.; 7. Ulidia erythrophthalma J/y., in great numbers, skg. B. Coleoptera.
(a) Zelephoridae: 8. Wanacea pallipes Pz., nect-lkg.; 9. Dasytes subaeneus Schh.
(6) Mordellidae: 10. Mordella fasciata #., numerous, nect-lkg. (c) Curculionidae :
11. Spermophagus cardui S/ev. («/) Cerambycrdae: 12. Strangalia bifasciata AZi//.,
nect-lkg. C. Hymenoptera. (a) Formicidae: 13. Several sps. (6) Apidae:
14. Halictus maculatus Sm. 9, po-clig. D. Lepidoptera. hopalocera: 15.
Coenonympha pamphilus Z., skg.

F. F. Kohl (Tyrol) records 3 ruby-wasps (1. Chrysis rutilans Olv.; 2. C.
scutellaris #.; 3. Hedychrum regium /.), and ro true wasps (1. Polistes gallica
“L.; 2. Eumenes pomiformis /.; 3. E. coarctata Z.; 4. E. unguiculata V772.;
5. Ancistrocerus parietum Z.; 6. Leionotus simplex /.; 7. L. dantici ossz;
8. L. parvulus Lep.; 9. L. chevrieranus Sauss.; 1o. L. tarsatus Sauss.

Schletterer observed the following at Pola and in the Tyrol

Hymenoptera. (a) Apidac: 1. Andrena aeneiventris Mor. (T.); 2. Ceratina
cucurbitina Moss?; 3. Halictus villosulus A.; 4. Prosopis clypearis Schenck;
5. P. hyalinata Sm. (=P. corvina Forst.); 6. P. variegata F. (6) Chryszdidae:

UMBELLIFERAE 507

7. Chrysis cuprea Ross?; 8. C. refulgens Spen.; 9. C. viridula Z.; 10. Ellampus
auratus Z.; 11. Stilbum cyanurum /%rs/. (=S. calens F.). (c) Evanitdae: 12.
Gasteruption granulithorax Zourn. (d) Ichneumonidae: 13. Colpagnathus celerator
Gr.; 14. Ichneumon xanthorius Fors’. (e) Pompilidae: 15. Pompilus tripunctatus
Dahib.; 16. P. viaticus Z.; 17. Pseudagenia carbonaria Scop.; 18. Salius fuscus F.
(0) Scolttdae: 19. Scolia insubrica Scop.; 20. S. quadripunctata /; 21. Tiphia
morio /. (g) Sphegidae: 22. Cerceris emarginata Pz.; 23. C. quadrifasciata Pz.
(A) Tenthredinidae: 24. Amasis laeta F. (7) Vespidac: 25. Polistes gallica L.

371. Caucalis L.

1182, C. daucoides L. (Schulz, ‘ Beitrage,’ IJ, pp. 91, 94, 190; Kerner, ‘Nat.
Hist. Pl.,” Eng. Ed. 1, II, pp. 296, 311, 342.)—According to Schulz, this white-flowered
species is andromonoecious, with homogamous or more rarely slightly protandrous
hermaphrodite flowers. The male flowers are usually in the middle of the umbels
and umbellules. The terminal umbel generally possesses the largest number of
hermaphrodite flowers, and self-pollination may easily take place.

Kerner describes the flowers as protogynous. He says that the central
umbellules contain only pseudo-hermaphrodite male flowers, while each of the
others possesses 4-7 of these and two really hermaphrodite flowers. Self-pollination
is brought about by an incurving of the bent filaments.

Visirors.—Herm. Miiller saw the bug Graphosoma nigrolineatum /. in
Thuringia (‘ Weit. Beob.,’ I, p. 306).

Schletterer observed the following at Pola.—

Hymenoptera. (@) Chrysididae: 1. Chrysis succincta LZ. (4) Evanirdae:

2. Gasteruption kriechbaumeri Sch/ett. (c) Ichneumonidae: 3. Mesoleius cruralis Gr.
(d@) Tenthredinidae: 4. Cephus variegatus Svein.

372. Turgenia Hoffm.

1183. T. latifolia Hoffm. (=Tordylium latifolium, and Caucalis latifolia Z.).—
According to Schulz (‘Beitrage,’ I], pp. 92, 191), this species is andromonoecious,
with homogamous hermaphrodite flowers. The distribution of sexes is the same as
in Caucalis.

Kerner (‘ Nat. Hist. Pl.,’ Eng. Ed. 1, I, pp. 296, 342) also describes the flowers
as protogynous. He says that in each umbellule there are 6-9 pseudo-hermaphrodite
pollen flowers in the centre, and 5-8 actinomorphous hermaphrodite ones externally.
Self-pollination is effected as in Caucalis daucoides.

373- Torilis Adans.

1184. T. Anthriscus Bernh.(=Tordylium Anthriscus Z.,and Caucalis Anthriscus
Huds.).— According to Schulz (‘Beitrage,’ I, p. 60), the flowers of this species,
which are white, often tinged with red, are distributed andromonoeciously, the
hermaphrodite ones being markedly protandrous. The short-stalked male flowers
occupy the centre of all the umbellules, and in umbels of a higher order the
number of male flowers increases. Tertiary and quaternary umbels occasionally
include these only. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) says that at

508 ANGIOSPERMAE—DICOTYLEDONES

Ruppin the umbellules of primary umbels bear marginal hermaphrodite and central
male flowers; secondary ones possess a few marginal hermaphrodite and numerous
central male flowers; tertiary umbels are completely or almost completely male.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) observed the following
(Herm. Miller, ‘ Fertilisation,’ p. 286; ‘Weit. Beob., I, p. 307; ‘Alpenblumen,’
p. 122).—

A. Coleoptera. (a) Alalacodermata : 1. Trichodes apiarius Z., nect-lkg. (H. M.).
B. Diptera. (a) Dolichopodidae: 2. Gymnopternus germanus Wredem., nect-lkg.
(H.M.). (8) Aluscidae: 3. Gymnosoma rotundata Z., in large numbers (H. M.).
(c) Syrphidae: 4. Ascia podagrica #., nect-lkg. (H. M.). C. Hymenoptera.
(2) Apidae: 5. Prosopis variegata #. 6 (H. M.). (0) Sphegzdae: 6. Cerceris
quinquefasciata Ross? 6, nect-lkg. (Budd.); 7. Ceropales maculata #. & and 9,
numerous (H. M.); 8. Crabro cribrarius Z. $ (H. M.); 9. C. sp. (H. M., Alps),
nect-lkg.; 10. C. vagus Z. 9 (H.M.); 11. Oxybelus bellicosus Ol. (H.M.); 12. O.
uniglumis Z., numerous (H.M.). (c) Zenthredinidae: 13. Tenthredo notha £7.
(Budd.). (d) Vespzdae: 14. Odynerus parietum Z. (H. M.). D. Lepidoptera.
15. Pieris rapae Z. (H.M.).

Willis (‘Fls. and Insects in Gt. Britain,’ Part I) gives the following for the
neighbourhood of the south coast of Scotland, all skg.—

A. Diptera. (a) Muscidae: 1. Agromyza flaveola Fal/.; 2. Anthomyia
tadicum Z.; 3. Hylemyia strigosa #.; 4. Phorbia floccosa A/acg.; 5. Stomoxys
calcitrans Z. (0) Syrphidae: 6. Platycheirus albimanus #. 3B. Hemiptera.
4. Anthocoris sp. C. Hymenoptera. (a) Apidae: 8. Halictus sp. (2) lehneu-
monidae: 9. 4 undetermined species. D. Lepidoptera. (a) Mocturdae: 10. Plusia
gamma ZL. (6) Khopalocera: 11. Epinephele janira LZ. (c) Muicrolepidoplera:
12. Simaéthis fabriciana Sveph.

The following were recorded by the observers, and for the localities stated.—

Loew (Mecklenburg), the true wasp Vespa sylvestris Scop. 6, skg. (‘ Beitrage,’
p- 38). Alfken (Bremen), A. Coleoptera. (a) Chrysomelidac: 1. Lema duodecim-
punctata Z. (4) Telephoridae: 2. Malachius aeneus Z. B. Hymenoptera.
(a) Zenthredinidae: 3. Allantus temulus Scop.; 4. Arge enodis LZ.; 5. A. ustulata Z.;
6. Dolerus fissus H/g.; 7. Macrophya quadrimaculata /.; 8. Pachyprotasis rapae Z.
Sickmann (Osnabriick), the parasitic fossorial wasp Ceropales maculatus #. MacLeod
(Flanders), 3 short-tongued Hymenoptera, 5 hover-flies, 4 Muscids, and a Lepidopterid
(Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 279-80).

1185. T. nodosa Gaertn. (=Caucalis fallax Borss. ef Balansa).—

VisiTors.—Schletterer (Pola) observed a ruby-wasp (Ellampus auratus Z.),
2 ichneumon flies (Acaenitis fulvicornis Gr., and Anisobus sp.), and a saw-fly (Argae
rosae L.).

1186. T. infesta Roth (=Caucalis infesta Tes/.).— According to Schulz
(‘Beitrage,’ 11, pp. 91, 191), this species is andromonoecious, with homogamous
or slightly protandrous hermaphrodite flowers, in which self-pollination can easily
take place.

Visirors—MacLeod (Pyrenees) observed a Muscid (Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 407); Schletterer (Pola) saw the ichneumon fly Glypta pictipes
Taschenb.

UMBELLIFERAE 509

374. Scandix L.

1187. S. Pecten-Veneris L. (Schulz, ‘Beitrage,’ II, pp. 91, 94, 191;
Kirchner, ‘Flora v. Stuttgart,’ p. 394; MacLeod, Bot. Jaarb. Dodonaea, Ghent,
vi, 1894, pp. 280-2; Kerner, ‘ Nat. Hist. Pl., Eng. Ed. 1, II, p. 342; Knuth,
‘Bloemenbiol. Bijdragen.’)—According to Schulz, Kirchner, and MacLeod, the
small white flowers of this species are distributed andromonoeciously, and the
hermaphrodite ones are homogamous or slightly protandrous. The long-stalked
male flowers possess no trace of a pistil; they are usually in the middle of the
umbellules, but the primary umbels often contain hermaphrodite flowers only,
while the tertiary ones are frequently composed of nothing but male flowers;
the number of the latter is generally greater in umbels of high order. Warnstorf
(Schr. natw. Ver., Wernigerode, xi, 1896) states that in Brandenburg all the umbels
are hermaphrodite to begin with, but become more or less female by the partial
or complete degeneration of the anthers. The anthers are greenish-yellow in
colour; the pollen-grains white, ovoid, not constricted in the middle, with 3
longitudinal grooves, about 13 y broad and 30 4 long.

Fic. 166. Anthriscus sylvesiris, Hoffm. (after Herm. Miller). (1) Flower in the first (purely
male) stage. a, immature anthers, hanging out of the flower; a’, mature anthers, projecting obliquely
upwards. The styles are not yet visible. (2) Flower in the second (purely female) stage. The
stamens have dropped off, the styles have developed, and their stigmas (s/) are mature. 7, nectary;
ov, ovary ; #, inner petal; #’, outer petal.

Self-pollination readily takes place, and Kerner says it is brought about by
incurving of the filaments, resulting in the anthers being applied to the stigmas.
The same authority describes the flowers as protogynous.

Visitors.—On the island of Fehmarn (adjoining ‘Land Oldenburg’) I only
noticed a hover-fly (Eristalis tenax Z.), po-dvg.; MacLeod (Flanders) saw a fossorial
wasp and 3 Diptera (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 282).

375. Anthriscus Bernh.

1188. A. sylvestris Hoffm.— Warming, Kirchener, Kerner, Schulz, and
MacLeod describe this white-flowered species as andromonoecious, with markedly
protandrous hermaphrodite flowers. The inner flowers of the umbellules are male,
the outer ones hermaphrodite, and Schulz says that the number of the former
increases in umbels of higher order. MacLeod (Bot. Jaarb. Dodonaea, Ghent, vi,
1894) gives a full account of these relations. According to Schréter (Justs bot.

510 ANGIOSPERMAE—DICOTYLEDONES

Jahresber., Leipzig, xvii, (1889) 1891, p. 557), the whole plant is protandrous, all
the flowers of the same stock being at the same time first male, then neuter, and
finally female. Kerner, on the other hand, observed the same kind of geitonogamy
as in Sium and Foeniculum (¢. v.).

VistTors.—Herm. Miiller gives the following list for himself (H. M.), Buddeberg
(Budd.), and Borgstette (Borg.) (‘Fertilisation, pp. 280-1, ‘Weit. Beob.,’ I,

p. 310).—

A. Coleoptera. (a) Cerambycidae: 1. Clytus arietis Z.(H.M.); 2. Leptura
livida #. (Borg.); 3. Grammoptera ruficornis #. (H. M.); 4. Pachyta collaris Z.
(H.M.); 5. P. octomaculata /. (H.M.). (6) Crsteldae: 6. Cistela murina Z.
(H.M.). (¢) Cleridae : 7. Trichodes apiarius Z., freq., nect-lkg. (H. M.). (d) Cocez-
nellidae: 8. Coccinella quattuordecimpunctata Z., nect-lkg. (H.M.); 9. C. sep-
tempunctata Z., do. (H. M.). (e) Curculionidae: 10. Bruchus sp., numerous.
(/) Dermestidae: 11. Anthrenus claviger £7., freq., nect-lkg. (H. M.); 12. A.
scrophulariae Z., do. (H. M.); 13. Tiresias serra &, do. (H. M). (g) Llateridae :

14. Athous niger Z. (H. M.); 15. Corymbites quercus O2. (H.M.); 16. Lacon
murinus Z., several (H. M.); 17. Synaptus filiformis #. (H. M.). (2) Zelephoridae :
18. Anthocomus fasciatus /., freq., nect-lkg. (H.M.); 19. Axinotarsus publicarius 7”,
nect-lkg. (H. M.); 20. Malachius aeneus Z. (H.M.); 21. M. bipustulatus Z. (H. M.);
22. Telephorus fuscus Z. (H. M.); 23. T. lividus Z.(H.M.); 24. T. rusticus Yad/.
(H. M.). (¢) Mordellidae: 25. Mordella fasciata F. (H. M.); 26. M. pumila Gyl.
(H.M.). (4) Mittdulidae: 27. Epuraea sp. (H. M.); 28. Meligethes sp. (H. M.).
B. Diptera. (a) B2dionddae: 29. Bibio hortulanus Z.(H.M.). (6) Cheronomidae :
30. Ceratopogon sp., skg. (H. M.). (c) Emprdae: 31. Empis punctata /. (H. M.);
32. E. stercorea Z. (H.M.). (d) Muscidae: 33. Echinomyia fera Z. (H. M.);
34. Graphomyia maculata Scop. (H. M.); 35. Lucilia sericata Me. (H. M.);
36. Musca corvina #. (H. M.); 37. Platystoma seminationis #. (H. M.); 38. Psila
fimetaria Z. (H. M.); 39. Sarcophaga sp. (H. M.); 40. Scatophaga merdaria F.
(H. M.); 41. 5S. stercoraria Z., numerous (H. M.); 42. Sepsis sp. (H. M.);
43. Zophomyia tremula Scop. (H.M.). (e) Stratiomyzdae: 44. Nemotelus pantherinus
L.(H.M.); 45. Stratiomys chamaeleon Deg. (H. M.). (/) Syrphtdae: 46. Ascia
podagrica #. (H. M.); 47. Eristalis arbustorum Z. (H. M.); 48. E. pertinax Scop.
(H.M.); 49. Helophilus floreus Z.(H.M.); 50. Melithreptus pictus 2g. (H. M.);
51. M. scriptus Z. (H. M.); 52. Syritta pipiens Z. (H. M.); 53. Syrphus corollae /.
(H. M.); 54. S. ribesii Z. (H. M.); 55. Xylota lenta Mg. (Borg.) (g) Tipulidae:
56. Pachyrhina crocata Z. (H. M.); 57. P. pratensis Z. (H. M.). C. Hemiptera.
58. Systellonotus triquitatus Z., skg. (H.M.). D. Hymenoptera. (a) Apidae:
59. Andrena collinsonana A. 9 (H. M.); 60. A. dorsata KX. 9, po-cltg. (H. M.);
61. A, fucata Sm. 9 (H.M.); 62. A. parvula X., skg. and po-cltg. (H.M.); 63. Apis
mellifica Z. ¥, po-cltg. (H. M.); 64. Chelostoma campanularum &. 9 and 4, nect-lkg.
(H. M.); 65. Colletes daviesanus X. 4, skg. (H. M.); 66. Halictus smeathmanellus -
K. 9 (H. M.); 67. Prosopis annularis Sm. 9, nect-lkg. (H. M.). (6) Braconidae:
68. Microgaster sp., nect-Ikg. (H.M.). (c) Cynipidae : “69. Eucoda subnebulosa Gir.
(teste Schenck) 9 (H. M.). (d) Formicidae: 70. Various species. (e) Ichneumonidae :
71. do. (f) Sphegzdae: 72. Crabro cephalotes #. $ (H. M.); 73. C. sexcinctus &. 3
(H. M.); 74. Gorytes laticinctus Zep. 9 (H. M.); 75. Pompilus neglectus Dahid. 9
(H. M.); 76. P. viaticus Z.9 (H.M.); 77. Psen atratus Pz. 9, nect-lkg. (H. M.).
(g) Tenthredinidae: 78. Abia sericea Z., in large numbers (H. M.); 79. Allantus
nothus A7/e. (H. M.); 80. Amauronematus vittatus Zep. (H. M.); 81. Athalia
annullata #. (H.M.); 82. A. rosae Z.(H.M.); 83. Dolerus fissus Avg. (H. M.);
84. Hylotoma femoralis K/g. (H. M.); 85. H. rosarum A7g., nect-lkg. (H. M.);-
86. Macrophya neglecta Ale. (H.M.); 87. M. rustica Z.(H.M.); 88. Pachyprotasis
rapae A7. (H.M.); 89. Pteronus myosotidis #. (H.M.); go. Selandria serva F.

UMBELLIFERAE 511

(H.M.); 91. Tenthredo sp. (H.M.). (4) Vespidae: 92. Odynerus elegans #7.-Sch.
9 (Borg.). E. Lepidoptera. (a) Rhopalocera: 93. Thecla betulae Z. (Budd.).
(4) Zortricidae: 94. Grapholitha compositella #., skg. (H. M.). F. Neuroptera.
Planipennia: 95. Hemerobius sp. (H. M.); 96. Panorpa communis JZ., nect-lkg.
(H.M.); 97. Sialis lutaria Z. (H. M.).

Alfken records the following for Bremen.—

A. Diptera. (a) Muscidae: 1. Platystoma seminationis 7. (b) Bzbronidae:
2. Bibio marci Z.; 3. Dilophus vulgaris Mg. (c) Syrphidae: 4. Eristalis sepulcralis
L.; 5. Rhingia rostrata Z.; 6. Xylota ignava Pz. B. Hymenoptera. (a) Apidae:
7. Andrena albicans Jill. 9; 8. A. chrysosceles K. 9, rare; g. A. labialis X. 8;
to. A. nitida Fourcr. 9 po-cltg., 6 skg., rare; 11. A. parvula A’, very common,
Q po-cltg. and skg., 8 skg.; 12. A. proxima K., freq., 9 skg. and po-cltg., 5 skg.
(6) Ichneumonidae: 13. Alomya ovator #. 9 and é; 14. Ichneumon extensorius Z. ;
15. I. fabricator #.; 16. Tryphon trochanteratus Her. (c) Sphegidae: 17. Crabro
chrysostomus Lep. 6, freq.; 18. C. nigritus Zep. , rare; 19. C. planifrons Zhms. 9,
rare; 20. C. vagabundus Pz, @, not infreqg.; 21. Psen concolor Dahlb. 9, do.
(2) Zenthredinidae: 22. Athalia glabricollis ZAs.; 23. A. lugens A/e.; 24. A. rosae
£.; 25. A.spinarum /.; 26. Poecilostoma luteola K/g.; 27. Pteronus myosotidis /” ;
28. Selandria serva #.; 29. Tenthredo atra Z.; 30. Tenthredopsis gibberosa Anw.
(e) Vespidae: 31. Odynerus oviventris Wesm. 8, infreq.; 32. O. parietum Z. 9 and 6,
freq.; 33. O spinipes Z. 4, do.

Verhoeff saw the following in Norderney.—

A. Coleoptera. (2) Elateridae: 1. Athous haemorrhoidalis #. (0) Zelephoridae :
2. Cantharis fusca Z.; 3. Dasytes plumbeus AZa//. (c) Mordellidae: 4. Anaspis flava LZ.
(d) Nitdulidae: 5. Brachypterus gravidus ///.; 6. Epurea aestiva Z.; 7. Meligethes
aeneus Z.; 8. M. coracinus Sturm. (e) Curculionidae: 9. Phyllobius urticae Deg.
B. Diptera. (a) Brbionidae: 10. Bibio sp. () Chironomidae: 11. Chironomus sp.
(c) Dolichopodidae: 12. Dolichopus aeneus Deg.; 13. D. brevipennis J/g. (2) Emprdae :
14. Empis stercorea Z.; 15. Hilaria quadrivittata 17g.; 16. Platypalpus flavipalpis 4/7.
(e) Afuscrdae: 17. Anthomyia muscaria Ze/f.; 18. A. sp.; 19. Aricia incana Weedem. ;
20. Calliphora erythrocephala A/g.; 21. Chlorops sp.; 22. Cynomyia mortuorum Z.
$; 23. Cyrtoneura hortorum Fa//. $; 24. Dryomyza anilis Fa//.; 25. Hylemyia
conica Wredem.,&; 26. Lucilia caesar Z.; 27. Musca domestica Z.; 28. Myospila
meditabunda #. $ and 9; 29. Nemopoda sp.; 30. Onesia floralis R—D.& and 9;
31. Psila villosula @%/g.; 32. Sapromyza rorida Fal/.; 33. Scatophaga lutaria /;
34. S. stercoraria Z.$; 35. Sepsis cynipsea Z.; 36. Spilogaster duplicata Mg. 3;
37. S. vespertina Fall. 3. (f) Syrphidae: 38. Eristalis arbustorum Z.; 39. Helo-
philus pendulus Z.; 40. Platycheirus albimanus F. 9; 41. Syritta pipiens Z. ;
»42. Syrphus corollae /. 9 and 6. (g) Yherevidae: 43. Thereva anilis Z., a 6.
(h) Tipulidae: 44. Ptychoptera contaminata Z. C. Hymenoptera. (a) Chalc:-
didae: 45. Torymus sp. (6) Formicrdae: 46. Lasius niger LZ. (c) Zenthredintdae :
47. Pteronus monticola Zhs.

Loew noticed the following in Brandenburg (B.), Riigen (R.), and Mecklenburg
(M.) (‘ Beitrage,’ p. 35).—

A. Coleoptera. (a) Cerambycidae: 1. Pachyta collaris Z. (B.). (6) C&ryso-
melidae: 2, Crioceris duodecimpunctata Z. (B.). (c) Dermestidae: 3. Anthrenus
scrophulariae LZ. (B.). (d@) Scarabaeidae: 4. Cetonia floricola Hést. (=C. metallica
£.) (B.). (e) Telephoridae: 5. Rhagonycha testacea L. (B.); 6. Telephorus fulvicollis
f. (B.); 7. T. fuscus Z. (B.); 8. T. obscurus Z. (B.): 9. T. rufus Z. (B.);
10. T. rusticus Fall. (B.). B. Diptera. (a) Bzdionzdae: 11. Bibio hortulanus
fF. 9 (B.). (8) Dolichopodidae: 12. Dolichopus sp. (B.). (¢) Zmpidae: 13. Empis
fallax Egg. (B.). (d) Syrphidae: 14. Melanostoma hyalinata /a//, skg. (R.).
C. Hymenoptera. (a) Sphegidae: 15. Crabro cetratus Shuck. 9, skg. (B.);

512 ANGIOSPERMAE—DICOTYLEDONES

16. C. vagus Z. (B.); 17. Gorytes. campestris Aull. 9 and é, skg. (B.); 18. Nysson
interruptus /. (B.); 19. Oxybelus uniglumis Z. 9 and é (B.); 20. Pemphredon
rugifer Dahib. (B.). (6) Tenthredinidae; 21. Rhogogastera viridis Z. (M.); 22.
Tenthredo livida Z.(M.). In Silesia (S.), Hesse (H.), Riesengebirge (R.), and the
Harz (Hr.) (op. cit., pp. 29, 47)-—A. Coleoptera. (a) Dermestidae: 1. Anthrenus
scrophulariae Z. nect-lkg. (S.); 2. Byturus fumatus #, do. (S.). (4) Zéelephoridae :
3. Rhagonycha melanura F’. (S.). B. Diptera. (a) Aszlidae: 4. Dioctria atricapilla
Me. (H.). (8) Muscidae: 5. Miltogramma germari Mg. (Hr.). (c) Syrphidae:
6. Chrysotoxum fasciolatum Deg. (S.); 7. Microdon devius Z. (H.); 8. Spilomyia
diophthalma Z., skg. (S.); 9. Volucella pellucens Z. (R.). (d@) Zabanidae: 10.
Tabanus micans JZe. (H.). C. Hymenoptera. (a) Sphegidae: 11. Crabro
wesmaéli v. d. Z. 9 and 4, skg. (S.). (4) Vespidae: 12. Vespa rufa L. ¥, skg. (S.);
13. V. sylvestris Scop. %, skg. (S.). In Switzerland (S.) and the Tyrol (T.),—
A. Coleoptera. (a) Cerambycidae: 1. Leptura sanguinolenta Z. (T.); 2. Oxymirus
cursor Z. (T.); 3. Pachyta collaris Z. (T.); 4. P. octomaculata #. (T.); 5. P.
quadrimaculata Z. (T.); 6. Strangalia armata Hfést. (T.); 7. S. attenuata Z.
(T.); 8. S. melanura Z.(T.); 9. Toxotus meridianus £. (T.). (6) Scarabaeidae:
10. Hoplia praticola Duff. (T.); 11. Trichius fasciatus Z.(T.). (c) Zelephoridae:
12. Malachius bipustulatus Z. (T.); 13. Rhagonycha terminalis Red/. (T.).
B. Diptera. (a) Stratiomyidae: 14. Odontomyia viridula F. (T.). (6) Syrphidae :
15. Cheilosia decidua Zgg. (?) (T.); 16. C. pigra Loew. (?) (T.); 17. Syrphus vittiger
Zett.(T.). C. Hymenoptera. () Chrysididae: 18. Chrysis ignita Z. (=C. angustula
Schenck) (T.). (6) Tenthredinidae: 19. Allantus albicornis #. (T.); 20. Tenthredo
flavicornis #. (T.). (c) Vespidae: 21. Leionotus simplex F. (T.); 22. Polistes
gallica Z. (S.).

The following were recorded by the observers, and for the localities stated.—

Knuth (Riigen), the Cerambycid Strangalia maculata Poda, and the bug Calocoris
norvegicus Gmel. Friese (Mecklenburg), the cuckoo bee Nomada guttulata Schenck,
infreq.; (Thuringia), 4 saw-flies—zx. Allantus fasciatus Scop.; 2. A. koehleri A7g.;
3. A. marginellus #.; 4. A. temulus Scop. Sickmann (Osnabriick), the fossorial
wasp Gorytes quadrifasciatus /., occasional. Schmiedeknecht (Thuringia), the bee
Andrena chrysosceles A. F.F. Kohl, the fossorial wasps Crabro cribrarius Z. and
C. scutellatus Schev. Handlirsch, the fossorial wasp Gorytes quadrifasciatus 7.
Schletterer (Pola), 2 bees (Halictus minutus A’, and Prosopis clypearis Schenck),
and 2 ichneumon-flies (Amblyteles armatorius Forsé., and Pimpla examinator /.).
H. de Vries (Netherlands), the saw-fly Dolerus haematodes Schr. (Ned. Kruidk.
Arch., Nijmegen, 2. Ser., 2. Deel, 1875). MacLeod (Flanders), 2 short-tongued
Hymenoptera, 13 Diptera, and 3 beetles (Bot. Jaarb. Dodonaea, Ghent, vi, 1894,
pp. 285-6). Burkill (Yorkshire coast), the Muscid Sepsis nigripes J/g., skg.
(‘Fertlsn. of Spring Fls.’).

1189. A. nitida Garcke.—
Visitors. — Loew (Altvatergebirge, Silesia) observed an ichneumon-fly of

undetermined species, and 2 saw-flies—Pamphilius hortorum A7g. and Tenthredopsis
scutellaris F*.

t1g0. A. Cerefolium Hoffm.—Warnstorf (Schr. natw. Ver., Wernigerode, xi,
1896) says that in this species the primary umbels bear hermaphrodite flowers, with
vestigial anthers here and there; secondary ones usually possess only pseudo-
hermaphrodite pollen flowers ; rarely some of the marginal flowers are hermaphrodite,
more frequently the anthers are partly degenerate. The pollen-grains are white,
smooth, ovoid, with 3 furrows and a central constriction and band, about 15 » broad
and up to 35 » long.

UMBELLIFERAE 513

VisiTors.—Herm. Miiller observed the following (‘ Fertilisation,’ p. 281).—

A. Coleoptera. (a) Cerambycidae: 1. Grammoptera ruficornis F., nect-Ikg.
(4) Dermestidae: 2. Anthrenus pimpinellae ¥., freq.; 3. A. scrophulariae Z., do.
(c) Mordellidae: 4. Anaspis frontalis Z. (d) Witidulidae: 5. Meligethes, very common,
nect-lkg. (ce) Zelephoridae: 6. Anthocomus fasciatus Z.; 7. Malachius aeneus Z.
B. Diptera. (a) Brbionidae: 8. Bibio hortulanus Z. (4) Muscidae: 9. Anthomyia
radicum Z.; ro, Cyrtoneura simplex Z.; 11. Exorista vulgaris Fad//.; 12. Gymnosoma
rotundata Z.; 13. Sarcophaga dissimilis A/g.; 14. S. haemorrhoa A/g.; 15. Sepsis
sp., nect-lkg. (c) Syrphidae: 16. Eristalis arbustorum Z.; 17. E. nemorum Z.;
18. Syritta pipiens Z. C. Hymenoptera. (a) Apidae: 19. Apis mellifica Z. ¥,
po-cltg.; 20. Prosopis armillata Myl. 9; 21. P. communis Wyl. 8. (4) Formicidae :
22. Several species. (c) Ichneumonidae: 23. Numerous species. (d) Sphegidae :
24. Oxybelus uniglumis Z., freq.; 25. P. pectinipes Z.d; 26. P. spissus Schyédte.

The following were recorded by the observers, and for the localities stated.—

Schenck (Nassau), the bee Andrena proxima XA. 9, very freq. MacLeod
(Flanders), a Muscid and a beetle (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 286).
Schletterer (Pola), the ichneumon Limneria (Angilia) fenestralis (Hgr.) Zhs.

igi. A, vulgaris Bernh. (=Scandix Anthriscus Z.).—Schulz (‘ Beitrage,’ IT,
pp. 89-91, 94) describes the small greenish-white flowers of this species as her-
maphrodite and homogamous. Self-pollination inevitably results from incurving of
the stamens: it is effective.

VistTors.—Flies, occasional and casual.

376. Chaerophyllum L.

1192. C.temulum L. (Schulz, ‘Beitrage,’ I, p.62; Herm. Miller, ‘ Fertilisation,’
p. 279, ‘ Weit. Beob.,’ I, p. 310; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 155.)—
Schulz describes this white-flowered species as andromonoecious, with markedly
protandrous hermaphrodite flowers. In most umbels there are marginal hermaphro-
dite flowers, and also a central one; the rest are male. The number of the latter
increases in umbels of higher order: tertiary ones are usually completely male, and
sometimes even the inner umbellules of secondary ones may be so.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list.—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida Z. (H. M.); 2. Obrium
brunneum /., nect-lkg. (Budd.); 3. Pachyta octomaculata /. (Budd., Borgstette).
(6) Dermestidae: 4. Anthrenus pimpinellae /.(H.M.); 5. A.scrophulariae Z. (H. M.).
(c) Mordellddae: 6. Anaspis rafilabris Gy//., nect-lkg. (H. M.). (¢@) Mittdulidae: 7.
Epuraea aestiva Z., nect-lkg. (H. M.); 8. Meligethes aeneus /, do. B. Diptera.
(a) Muscidae: 9. Gymnosoma rotundata Z.(H.M.). (6) Stratzomyrdae: 10. Chrysomyia
formosa Scop. (H.M.). (c) Syrphrdae: 11. Bacha elongata /. (H. M.); 12. Cheilosia
scutellata /a//.(H.M.); 13. C.sp., po-dvg. (Budd.) ; 14. Chrysogaster coemeteriorum
ZL. (H.M.); 15. Eristalis nemorum Z. (H. M.); 16. Helophilus floreus Z., skg. and
po-dvg. (H. M., Budd.); 17. Melanostoma mellina Z. (H. M.); 18. Melithreptus
scriptus Z. (H.M.); 19. Syritta pipiens Z.(H.M.). C. Hymenoptera. (a) Apzdae:
20. Andrena parvula A. 9, skg. (H. M.); 21. Apis mellifica Z. ¥, po-cltg. (H. M.);
22. Prosopis armillata Vy/. 9 (H. M.); 23. P.communis Vy/.é(H.M.). (4) Formicedae:
24. Several sp. (c) /chneumonidae: 25. Numerous sp. (d) Sphegidae: 26. Crabro
dives A-Sch. 3, nect-lkg. (H. M.); 27. Oxybelus uniglumis Z. freq. (H. M.);
28. Pompilus pectinipes v. ¢. Z. 3 (H. M.); 29. P. spissus Schjédte (H. M.).
(e) Tenthredinidae: 30. Hylotoma caerulescens, /, nect-lkg. (H. M.).

DAVIS. 11 Ll

514 ANGIOSPERMAE—DICOTYLEDONES

The following were recorded by the observers, and for the localities stated. —

Knuth (Féhr), 2 Syrphids and a Pseudoneuropterid. Alfken (Bremen).—
A. Diptera. Syrphidae: 1. Chrysogaster coemeteriorum Z. B. Hymenoptera.
(a) Apidae: 2. Andrena shawella K. 8. (6) Ichneumonidae: 3. Campoplex oxyacanthae
Bote. Schenck (Nassau)—Hymenoptera. (a) Apidae: 1. Andrena proxima K. 9,
very common. (4) Sphegidae: 2. Diodontus minutus /.; 3. Tachysphex nitidus Spz7.
(c) Sapyeidae: 4. Sapyga clavicornis Z. F. F. Kohl (Tyrol), the true wasp Leionotus
dufourianus Sawss., and the fossorial wasp Crabro cribrarius Z., 9 and 6. MacLeod
(Flanders), 6 Syrphids, a Stratiomyid, a gnat, 7 short-tongued Hymenoptera, and
6 beetles (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 287).

1193. C. bulbosum L.—Kirchner (‘Flora v. Stuttgart,’ p. 396) describes this
species as andromonoecious, with a similar distribution of sexes to that of C. temulum.
The quaternary umbels, which develop late, are almost always completely male.

Visirors.—Loew (Berlin Botanic Garden) observed the bee Prosopis armillata
Ny/. 9, skg.

1194. C. aureum L.—According to Schulz (‘ Beitrage,’ I], p. 191), this species
is also andromonoecious, with protandrous hermaphrodite flowers.

VisiTors.— Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Dermestidae: 1. Anthrenus scrophulariae Z., nect-lkg.
(6) Scarabaeidae: 2. Cetonia aurata Z. B. Diptera. (a) Bidbionzdae: 3. Bibio
hortulanus Z. 9, skg. (6) Aluserdae: 4. Graphomyia maculata Scop. (c) Stratio-
myidae: 5. Stratiomys longicornis Scop. (d) Syrphidae: 6. Eristalis nemorum L.,
skg.; 7. Helophilus floreus Z.; 8. Melanostoma mellina Z., skg.; 9. Platycheirus
scutatus Afg., do.; 10. Syritta pipiens Z., do.; 11. Syrphus ribesii Z., do.

The following were recorded by the observers, and for the localities stated.—

MacLeod (Pyrenees), 7 short-tongued Hymenoptera, 5 beetles, 6 Syrphids, and
17 other Diptera (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 415-16). von Dalla
Torre (Innsbruck Botanic Garden), the bee Prosopis annulata Z. 9 and 6.

1195. C. aromaticum L.—Schulz (‘Beitrége’) describes this species again as
andromonoecious, with protandrous hermaphrodite flowers. According to Kerner
(‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 295), each umbellule contains a central and 3-5
marginal hermaphrodite flowers, the intermediate space being occupied by about
20 pseudo-hermaphrodite male ones. The hermaphrodite flowers develop earlier than
the male ones, their anthers having fallen off and their stigmas matured before
dehiscence takes place in the latter. The stigmas come into the line of fall of the
pollen of the male flowers, and are thus geitonogamously crossed. (Cf Fig. 155,
p- 463, and Vol. I, pp. 41-2.)

1196. C. hirsutum L.— Schulz (‘Beitrage, I], p. 191) describes this species
also as andromonoecious, with protandrous hermaphrodite flowers.

Visirors.—Herm. Miiller observed the following (‘ Fertilisation,’ pp. 278-9).—

A. Coleoptera. (a) Llateridac: 1. Agriotes gallicus Lac. (6) Oedemeridae:
2. Oedemera flavescens L. B. Diptera. (a) Syrphrdae: 3. Eristalis pertinax Scop.
C. Hymenoptera. (a) Apidae: 4. Sphecodes ephippius Z. (6) Chryszdidac:
5. Chrysis ignita Z. (c) L£vaniidae: 6. Foenus affectator #. (d) Sphegzdac:
7. Crabro subterraneus #.5; 8. Pompilus pectinipes v. d. Z. (e) Lenthredinidae:
y. Athalia rosae Z.; 10. Hylotoma enodis Z., in large numbers; 11. H. segmentaria
Pz.; 12. Allantus arcuatus Fors/.; 13. A. rossii Pz.; 14. Tenthredo sp.

UMBELLIFERAE 515

Loew saw the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella septempunctata Z., nect-lkg.
(2) Scarabaeidae: 2. Cetonia aurata L. (c) Telephoridae: 3. Malachius bipustulatus Z.,
nect-lkg. B. Diptera. (2) Bibzonidae: 4. Bibio hortulanus Z. 6, skg. (3) Muserdae :
5. Graphomyia maculata Scop.; 6. Onesia floralis R.-D. (c) Syrphidae: 7. Eristalis
sepulcralis Z., skg. C. Hymenoptera. Apzdae: 8. Andrena tibialis A. 6, skg.;
g. Apis mellifica Z., skg. and po-cltg.

1197. C. villarsii Koch (=C. hirsutus 17//.).—<According to Schulz (‘Beitrage,’
II, pp. 89-90, 191), this species is andromonoecious, with markedly protandrous
hermaphrodite flowers. The primary umbels usually bear marginal hermaphrodite
and central male flowers; secondary ones are chiefly male.

Vistrors.— Herm. Miiller observed 9 beetles, 23 Diptera, 4 Hymenoptera,
3 Lepidoptera, and a Neuropterid in the Alps (‘ Alpenblumen,’ p. 123).

Loew, also for the Alps (Pontresina) records the following (‘ Beitrage,’ p. 55).—

A. Coleoptera. (a) Cerambycidac: 1. Callidium violaceum Z.; 2. Strangalia
melanura Z.; 3. Tetropium luridum Z. (6) TZelephoridae: 4. Dasytes alpigradus
Kiesw.; 5. Rhagonycha nigripes Red/.; 6. R. denticollis Schumm. B. Diptera.
(a) Bombyliidae: 7. Anthrax paniscus Rossz. (6) Syrphidae: 8. Chrysotoxum
vernale Lw.; 9. Eristalis tenax £.; 10. Volucella bombylans A. (c) Zabanidac :
11. Tabanus borealis #. 6. C. Hymenoptera. TZenthredinidae: 12. Tenthredo
sp. D. Lepidoptera. (a) Geometridae: 13. Odezia atrata LZ. (6) Noctuidae:
14. Undetermined sp.

377- Echinophora Tourn.

1198. E. spinosa L. (Kirchner, ‘D. Bliiten d. Umbelliferen.’)—This white-
flowered species, which resembles a thistle in habit, is common on the Lido at Venice.
The umbels are flat or somewhat convex; each umbellule includes about twelve
flowers, but marginal ones have more and central ones fewer. Only the middle
flower of an umbellule is hermaphrodite, all the others being male, with no styles
and merely a vestigial ovary, on which is a translucent annular disk, that secretes
nectar. Before the anthers dehisce, the filaments of all the flowers are curved
inwards, but they afterwards diverge, and their anthers open. ‘The white petals are
deeply bilobed, and there is an inwardly directed appendage in the middle of the
incision. The petals of the inner flowers of the whole umbel, and of each umbellule,
are very small; those of the marginal flowers (especially at the periphery of each
umbel) are larger and spreading. The two stigmas of a central flower have become
receptive before any of the stamens of the same umbellule have erected themselves :
the filaments of the male flowers diverge in centripetal order. The styles and stigmas
of an hermaphrodite flower persist till all the anthers of the same umbellule have
withered. This early development and long persistence of the stigmas undoubtedly
secure the crossing of the relatively few hermaphrodite flowers, in which automatic
self-pollination is excluded by the relative positions of the reproductive organs. It
would also be superfluous, as the flowers are visited by many insects.

Visttors.—Flies, hover-flies, bees, and several Lepidoptera (Lycaena, Zygaena,
and a micro-Lepidopterid) have been observed.
Lle

516 ANGIOSPERMAE—DICOTYLEDONES

378. Myrrhis L.

1199. M. odorata Scop. (Herm. Miiller, ‘Fertilisation,’ p. 278, ‘ Weit. Beob.,’
I, p. 311; Schulz, ‘Beitrage,’ II, p. r91.)—Schulz describes this species as andro-
monecious, with protandrous hermaphrodite flowers. Herm. Miiller says that the
last flowers which appear are purely male, their small petals falling off without ovary,
styles, or stigmas having developed. These flowers consequently provide pollen for
fertilizing the youngest hermaphrodite ones.

Vistrors.—Herm. Miiller (H. M.) at Lippstadt, and Borgstette (Borg.) at Teck-
lenburg, observed the following. —

A. Coleoptera. (a) Cerambycidae: 1. Grammoptera ruficornis /., in large
numbers (H.M.). (0) C&rysomelidac: 2. Galeruca calmariensis Z. (Borg.). (c)
Dermestidae: 3. Anthrenus scrophulariae Z., in very large numbers, nect-lkg. (H. M.).
(d) Mordellidae: 4. Anaspis frontalis Z., nect-lkg. (H.M.). 5. Mordellistena pumila
Cyil., occasional, nect-lkg. (H. M.). (e) Wrtvdulidae: 6. Epuraea sp., freq. (H. M.);
7. Meligethes aeneus /., occasional, nect-lkg. (H. M.). 3B. Diptera. (a) Boméby-
Ziidae: 8. Bombylius maior
L. (Borg.). (6) Emprdae :
9. Empis punctata /., skg.
(H. M.); 10. G. stercorea
L., freq., skg. (H. M.); 11.
i. tessellata F. (Borg.); 12.
E. vernalis Mg. & (H. M.);
13. Platypalpus candicans
Fall. (H. M.); 14. Rham-
phomyia umbripennis AZg. 9
(H.M.). (c) Mluscrdae: 15.
Anthomyia aterrima AZ. and
other sps. (H. M.); 16.
Calobata cothurnata Pz., in
large numbers (H. M.);

17. Chlorops hypostigma

‘ Fic. 167. Myrrhts Series ise Herm. ils . qa) is Meg., freq. (H. M.). 18.
ower, as it appears at the end of anthesis. (2) Ditto, after it has . : °

faded. (3) Hermaphrodite flower in the last (purely female) stage. Coenosia intermedia Fall.

a, dehisced anthers; a’, undehisced anther; 7, nectary; ov, ovary; (H. M.); 19. Cordylura pu-
A, petal; sé, stigma. bera Z. (H. M.); 20. Dryo-

myza flaveola /. (H. M.);
21. Nemopoda cylindrica #. (H. M.); 22. N. stercoraria &.-D. (H. M.); 23.
Piophila casei Z. (H.M.); 24. Psila fimetaria Z., in large numbers (H. M.); 25.
Scatophaga lutaria #. (H. M.); 26. Sepsis sps., in large numbers (H.M.). (d)
Syrphidae: 27. Xylota femorata Z. (Borg.). (e) Tipulidae: 28. Species of Tipula
(H. M.). ©. Hymenoptera. (a) Apidae: 29. Halictus maculatus Sm. (Borg.).
(0) Formicidae: 30. Lasius brunneus Z/r. ¥ and other species of ants (H. M.).
(c) Lchneumonrdae: 31. Several sps. (Borg.). (d) Yenthredinidae: 32. Allantus
temulus Scop., nect-lkg. (H. M.); 33. Athalia rosae 1. (H. M.); 34. Rhogogastera

’

viridis Z., nect-skg. (H. M.); 35. Tenthredo flavicornis /, do. (H. M.).
The following were recorded by the observers, and for the localities stated.—

MacLeod (Pyrenees), 4 short-tongued Hymenoptera, a beetle, and 5 flies (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 417). Scott-Elliot (Dumfriesshire), a saw-fly,
2 Empids and 6 Muscids (‘ Flora of Dumfriesshire,’ p. 80). Loew (Berlin Botanic
Garden).—A. Coleoptera. Scaradbacidae: 1. Cetonia aurata Z., dvg. the flowers.
B. Diptera. (a) Afuscidae: 2. Chloria demandata F.; 3. Lucilia caesar Z.; 4.
Scatophaga merdaria 7. (8) Syrphidae: 5. Eristalis nemorum Z., skg.

UMBELLIFERAE 517

379. Molopospermum Koch.
1200. M. Peloponnesiacum Koch.—
VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (2) Dermestidae: 1. Anthrenus scrophulariae Z., nect-lkg.
(6) Scarabaeidae: 2, Cetonia aurata Z., dvg. the flowers. (c) Lelephoridae: 3. Tele-
phorus fuscus Z., nect-lkg. B. Diptera. (a) Bidionidae: 4. Bibio hortulanus Z.,
skg. (6) Aluscidae: 5. Echinomyia fera L., skg.; 6. Scatophaga merdaria
(c) Stratiomyidae: 7. Stratiomys longicornis Scop. (2) Syrphidae: 8. Eristalis
nemorum Z., skg. C. Hymenoptera. (a) Apidae: 9. Andrena tibialis X. 9, skg.

and po-cltg.; ro. Apis mellifica 4. ¥, ditto. (6) Zenthredintdae: 11. Hylotoma
rosae L. 8.

380. Prangos Lindl.

1201. P. ferulacea Lindl.—
Vistrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella bipunctata Z., nect-lkg.
(2) Curculionidae: 2, Ceutorhynchidius floralis Pays. (c) Dermestidae: 3. Anthrenus
scrophulariae Z., nect-lkg. B. Diptera. (a) A/uscidae: 4. Graphomyia maculata
Scop.; 5. Lucilia caesar 2. (4) Syrphidae: 6. Eristalis arbustorum Z.; 7. Helo-
philus floreus Z. C. Hymenoptera. (a) Formicidae: 8, Lasius niger Z., nect-
Ikg. (4) Jchneumonidae: 9. Campoplex sp.

XLIX. ORDER ARALIACEAE JUSS.
381. Hedera Tourn.

Flowers greenish in colour; protandrous or homogamous ; with exposed nectar,
secreted by a disk surrounding the styles.

1202. H. Helix L. (Delpino, ‘Altri appar. dicog. recent. oss.,’ p. 52; Herm.
Miiller, ‘Weit. Beob.,’ I, pp. 301-2; Knuth, ‘ Blitenbiol. Herbstbeob.’; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, vi, 1894, pp. 255-6; Kirchner, ‘Flora v. Stuttgart,’
p- 398; Macchiati, Bot. Centralbl., Cassel, xxi, 1885, p. 7; Wittrock, op. cit.,
xxvi, 1886, p. 124.)—In Schleswig-Holstein the ivy is one of the latest plants to
blossom, and (in 1890) I observed the first buds open on Nov. 1, from which date
flowering went on until mid-December. About twenty green flowers, with stalks
1-1°5 cm. long, are aggregated into a hemispherical umbel, and the/ faint almost
putrefactive odour attracts various minute and larger flies, as well as Hymenoptera.
The individual flowers are protandrous. There are five (rarely six) recurved petals,
surrounding a nectar-secreting yellowish-green disk. In the middle of this rises
the short (scarcely 1 mm.) style with its stigma. The five (rarely six) inwardly
inclined stamens are situated at the margin of the disk; the filaments are 2-3 mm.
long. The anthers are bright yellow in colour, turning to a brownish-yellow and
quickly dropping off after dehiscence. The stigma is now mature, and the disk
secretes nectar more actively, thus compensating for the diminution in conspicuous-
ness resulting from the loss of the anthers. The under-sides of insects visiting
flowers in the first (male) stage get dusted with pollen, which is transferred to the
stigmas of those in the second (female) stage.

Delpino also describes the flowers as protandrous, and he observed flies effecting

518 ANGIOSPERMAE—DICOTYLEDONES

pollination. Hermann Miiller and Kirchner, on the contrary, found them to be
homogamous. The insect visitors of these flowers, however, effect cross-pollination,
for they alight upon the central stigma, and do not touch the radiating stamens till
afterwards. Hermann Miiller and Kirchner add that nectar is secreted so abundantly
that, if not removed by insects, it covers the nectary with a white sugary crust after
the flower has faded. Wittrock calls attention to the fact that the ivy but rarely
blooms in Central Sweden, the most northerly point for this being in Sédermanland
(58° 54’ N. lat.). The same writer mentions that the ivy flowering every October in
a Stockholm greenhouse never produces fruit, probably owing to the absence of the
insects necessary for pollination. Automatic self-pollination would therefore seem to
be ineffective.

Visitors.—Burkill and Willis observed the following near Cambridge (‘ Fls.
and Insects in Gt. Britain.’).—

A. Diptera. (a) Muscidae: 1. Anthomyia. 2 sp.; 2. Aricia lucorum Fa//. ;
3. Calliphora erythrocephala AZg., very common, skg.; 4. Chloropisca ornata MZg.;
5. Hydrellia griseola Fa//.; 6. Limnophora sp.; 7. Lucilia sp., skg.; 8. Onesia
sepulcralis Mg., ske.; 9. Phytomyza sp.; 10. Aricia lardaria 7; 11. Pollenia
tudis &,, freq., skg.; 12. Drosophila graminum fa//.; 13. Scatophaga stercoraria L.,
skg.; 14. Siphona geniculata Deg., do.; 15. Trichophthicus cunctans Zy. (6) Syr-
phidae: 16. Eristalis tenax Z., freq., skg. (c) ALycetophilidac: 17. Bolitophila fusca
Meg.; 18. Metriocnemus sp.; 19. Orthocladius sp.; 20. Sciara sp. B. Hymeno-
ptera. (a) Ichneumonidae: 21. Five undetermined species. (0) Vespedaec: 22.
Vespa vulgaris Z., freq., skg. C. Lepidoptera. Zoréricrdav: Tortrix sp.

The following were recorded by the observers, and for the localities stated.—

Knuth, the wasp Vespa vulgaris Z., the Muscid Aricia lardaria #., and the
earwig Forficula auricularia. Herm. Miiller, 3 Muscids—1. Calliphora erythroce-
phala M@g.; 2. Echinomyia fera Z., freq.; 3. Lucilia cornicina #., do. Plateau
(Belgium), the wasp Vespa germanica Z., in countless numbers; 2 hover-flies (Eris-
talis, Helophilus); and a Muscid (Calliphora vomitoria Z., freq.). Schletterer (Pola),
the Scoliid Scolia hirta Schr. (Oct. 10), and (also in October) 3 true wasps—
1. Eumenes mediterranea A7vchb.; 2. Polistes gallica 7.; 3. Vespa germanica /.

L. ORDER CORNACEAE DC.
382. Cornus L.

Flowers homogamous; with exposed nectar, secreted by a ring surrounding the
style.

1203. C. sanguinea L. (Herm. Miller, ‘ Fertilisation, pp. 287-8, ‘ Weit.
Beob.,’ I, p. 301; Kirchner, ‘Flora v. Stuttgart,’ p. 399; Knuth, ‘Bloemenbiol.
Bijdragen,’ ‘ Bliitenbiol. Beob. a. d. Ins. Riigen.’)—Hermann Miller says that in this
species the stamens and stigmas develop simultaneously. The anthers are introrse,
and at the same level as the stigma, though some distance from it. The larger
insects that alight on the inflorescence, or on a single flower, will therefore, when
licking the nectary, usually touch an anther or two with one side of their heads, and
the stigma with the other. Should they creep further on the same inflorescence, or
visit another, cross-pollination will be favoured. Smaller flies and beetles, on the
contrary, owing to their erratic movements, will sometimes effect cross- sometimes

CORNACEAE 519

self-pollination. Hermann Miiller states that automatic geitonogamy may now and
then take place, for some of the stigmas are touched by the anthers of the diverging
stamens of neighbouring flowers. (Cf ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, pp. 325-6.)
The flowers possess the odour of trimethylamide. Warnstorf describes the pollen-
grains, when examined in water, as large, rounded, opaque, with granular proto-
plasm, 63 to 75 pw in diameter.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel), several skg. or po-dvg. hover-flies (1. Eristalis tenax Z.; 2. E.
arbustorum Z.; 3. Syrphus balteatus Deg.; 4. Volucella pellucens Z.) and Muscids
(1. Lucilia caesar Z.; 2. L. cornicina #.): (Rigen), the butterfly Argynnis paphia Z.,
skg. for a short time. Krieger (Leipzig), the true wasp Odynerus spinipes /., and the
bee Andrena labiata Schenck (=A. schencki Afor.). Schmiedeknecht (Nassau),
the bee Andrena carbonaria Z. von Dalla Torre (Tyrol), the bee Andrena albicans
Mill. 9. MacLeod (Flanders), an Empis, and the beetle Meligethes (Bot. Jaarb.
Dodonaea, Ghent, vi, 1894, p. 256).

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list.—

A. Coleoptera. (a) Crrambycidae: 1. Clytus arietis /.. (Budd.); 2. Grammo-
ptera tabacicolor Deg. (H. M.); 3. Leptura livida /. (H.M.); 4. Pachyta octomacu-
lata #. (Budd.); 5, Strangalia armata Wés/.(H. M., Budd.); 6. S. atra Larch. (H.M.);
7. S. attenuata Z.(H.M.). (4) Curculion{dae: 8. Otiorhynchus picipes /. (H. M.).
(c) Dermestidae: 9. Byturus fumatus & (H.M.) (@) #lateridae: 10. Athous niger
Z. (H. M.); 11. Dolopius marginatus Z. (H.M.). (e) Zelephoridae: 12. Telephorus
pellucidus / (H. M.).
(/) Netidulidae: 13.
Meligethes (H. M.);
14. Thalycra sericea
Sturm. (AH. M.). B.
Diptera. (a) Lmpidae:
15. Empis livida JZ.
(H. M.). (2) (?). 16.
A tiny gnat in very
large numbers (H. M.).
(c) Syrphidae: 17.
Eristalis | arbustorum
L., po-cltg. (H. M.);
18. E. nemorum Z.,

o-cltg. (H. M. 5 9. Fic. 168. Cornus sanguinea, L. (after Herm. Miller). (1) Flower
f

Volucella pellucens J. seen from above. (2) Do., from the side. a@, stamen,; 7, nectary; ov,

(Budd ) Cc Hyme ovary; f, petal; s, sepal; s/, stigma.

noptera. Sphegidae :
20. Pompilus sp. (H.M.). All these visitors were licking the fleshy disk.

1204. C.MasL. (Sprengel, ‘ Entd. Geh.,’ p. 85 ; Schulz, ‘ Beitrage,’ II, p. rgt.)
—Schulz says that the yellow flowers of this species are hermaphrodite and homoga-
mous; but they have also been described as dioecious, or polygamous. The
mechanism of the hermaphrodite flowers agrees with that of C. sanguinea. Accord-
ing to Kerner (loc. cit.), geitonogamy may also take place as in that species.
Warnstorf describes the pollen-grains as pale-yellow in colour, ellipsoidal to ovoid,
almost smooth, 37 » long and 23-25 » broad.

520 ANGIOSPERMAE—DICOTYLEDONES

1205. C. florida L.—Kerner states that this species displays the same kind of
geitonogamy as C. Mas.

1206. C. suecica L. (Knuth, ‘Bloemenbiol. Bijdragen.’)—I had the oppor-
tunity of examining the flower mechanism of this species in the Dravitholz, between
Tondern and Liigumkloster (Central Schleswig), at the beginning of July, 1891.
The four yellowish bracts, streaked with reddish veins, play the part of petals, so
that a pseudo-flower with a diameter of nearly 2 cm. is formed. Each bract is
1 cm. long; one pair of opposite bracts are somewhat broader than the others,
i.e. 8 mm. as compared with 6. From the middle of this pseudo-flower spring
about twenty truly hermaphrodite flowers, arranged in an umbel. They are red in
colour, only 2 mm. long, and borne upon peduncles of the same length. Sepals and
petals are recurved. The style of each flower, with its stigma, is 1 mm. high, while
the four diverging stamens are 2 mm. in length. An insect alighting on the umbel
must therefore first touch the stigmas, and then the anthers, so that cross-pollination
is effected when a visit is made to a second flower.

I could not decide whether the flowers were protandrous or homogamous with
persistent stigmas, for it was only towards the end of anthesis that I was able to
examine them; most of the anthers had dropped off, while the stigmas were still
receptive.

Failing insect-visits, automatic geitonogamy is possible by the spreading of the
stamens so as to touch the neighbouring flowers.

VistTors.—I observed a few hover-flies—Eristalis arbustorum Z., and Helophilus
pendulus Z., po-dvg.

LI. ORDER CAPRIFOLIACEAE JUSS.

LireraTure.—Hermann Miiller, ‘ Fertilisation,’ pp. 289-99; Knuth, ‘Grundriss
d. Bliitenbiol.,’ pp. 61-2.

As Hermann Miiller has remarked, the members of this order are extremely
variable in regard to their flower mechanisms. Lonicera Caprifolium possesses
a corolla-tube as much as 30 mm. long, and its nectar is therefore only available
to hawk-moths with the most elongated proboscis; L. Periclymenum, with a corolla-
tube about 20 mm. in length, gives access to long-tongued bees as well as to
hawk-moths; L. caerulea is a humble-bee flower; L. nigra is a bee-flower; while
the nectar of L. tatarica and L. Xylosteum (corolla-tube only 3-7 mm. long) is
accessible to bees and also to certain flies; Symphoricarpos is regarded by Hermann
Miiller as a wasp-flower (I observed chiefly bees, but also hover-flies as visitors), and
Lonicera alpigena is similar; the funnel-shaped corolla of Linnaea also allows insects
with a moderately short proboscis to reach the nectar; Viburnum possesses fully
exposed nectar, and is therefore pollinated by short-tongued insects (flies and
beetles), which also visit to some extent the nectarless species of Sambucus; Adoxa,
finally, with its perfectly flat exposed layer of nectar, attracts minute insects of various
orders (flies, Neuroptera, beetles). Automatic self-pollination is chiefly rendered
possible in species which receive the smallest number of insect-visits; cross-

CAPRIFOLIACEAE 521

pollination is ensured when such visits take place. Our most important native
types of the order are therefore distributed as follows among the flower classes.—
Po, Sambucus;
E, Vibumum, Adoxa ;
C, Symphoricarpos, Linnaea, Lonicera alpigena, L. tatarica, L. Xylosteum ;
H, Lonicera caerulea, L. nigra ;
Lm, Lonicera Periclymenum, L. Caprifolium.

° 383. Adoxa L.

Flowers inconspicuous, greenish; homogamous or protogynous; with exposed
nectar, secreted by a fleshy ring at the base of the stamens.

1207. A. Moschatellina L. (Herm. Miiller, ‘Fertilisation,’ pp. 289-90; Ricca,
Atti Soc. ital. sc. nat., Milano, xiii, 1870; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v,
1893, p- 389 ; Knuth, ‘ Bloemenbiol. Bijdragen’; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1,
II, pp. 334-5; Kirchner, ‘Flora v. Stuttgart,’ p. 668.)—The flowers of this species

Fic. 169. Adoxa Moschatellina, L. (after Herm. Miller). (1) Terminal flower, seen from above
(x 3}). (2) Ditto, seen from below. (3) Lateral flower, opened and spread out, the style pressed
downwards; seen from the front. (4) Ditto, seen from behind. (5) Part of a flower with two (divided)
stamens (x7). (6) Style of a terminal flower, seen from the side. a, immature half-anther ; a’, ditto,
mature; 2, nectary; ov, ovary; #, petal of the terminal flower; #', upper petal of a lateral flower; °,
lower petal of ditto; ', lateral petal of ditto; s, sepal; s¢, stigma.

are arranged in a cuboidal head, and exhale a faint musky odour. According to
Hermann Miller, the stamens of the terminal tetramerous flower are directed upwards,
while those of the four lateral pentamerous flowers are directed outwards. The
anthers are at the same level as the stigmas, which develop simultaneously. When
nectar-licking or pollen-devouring insects creep over the blossoms, they sometimes
touch the anthers and sometimes the stigmas with their feet or proboscis; and
chiefly effect cross-pollination. Automatic self-pollination is possible in the lateral
flowers, by the fall of pollen on the edges of the stigmas; in the terminal ones this
only takes place when the plant is bent over by the wind. Kerner describes the
flowers as protogynous, and says that the anthers are at first remote from the
stigmas; the filaments later on inclining inwards, so as to bring about self-pollination.

522 ANGIOSPERMAE—DICOTYLEDONES

Warnstorf (Verh. bot. Ver., Berlin, xxxvii, 1895) says that at Ruppin the flowers
are slightly protogynous to homogamous; the terminal flower (with eight or rarely
ten stamens) opens first, then the pairs of opposite lateral flowers in succession.
Sometimes the terminal and one or two lateral flowers are degenerate. The 4-7
stigmatic branches, and also the anthers, persist for several days; the latter are
at a lower level than the stigmas, so that autogamy is scarcely possible. The
pollen-grains are bright yellow in colour, ovoid, slightly tuberculated, very variable in
size, up to 37 « long and 18 » broad.

Visrrors. — Warnstorf observed small beetles, and Herm. Miiller gives the
following list.—

A. Coleoptera. Curcultonidae: 1. Apion columbinum Germ., nect-lkg.
B. Diptera. (a) Cecidomyidae: 2. Various sps. nect-lkg. (6) Muscidae:
3. Borborus niger Afg., nect-lkg. (c) MLycetophilidae: 4. Various species, from
14to4mm.long. (d) Srmulitdae: 5. Simulia sp. C. Hymenoptera. (a) /chneu-
monidae: 6, Pezomachus Grav., 2 sps. (6) Preromalidae: 7. Eulophus 6; 8. Seven
other species.

Burkill observed the following on the Yorkshire coast (‘Tertlsn. of Spring
Fls.’).—

A. Coleoptera, Curculionidae: 1. Apion apricans Hés/., ske. B. Diptera.
(a) Alusctdae: 2. Scatophaga stercoraria L.; 3. Sepsis nigripes AZe. (6) ALycelo-
philidae: 4. Exechia sp.; 5. Sciara sp. and 3 other species. (c) RAyphidae:
6. Rhyphus sp. (d@) Syrphidae: 7. Melanostoma quadrimaculata Verra/. C. Hy-
menoptera. Jchneumontdae: 8. Pezomachus sp. and one other small Ichneumon.
D. Thysanoptera. 9. Thrips sp., all skg.

MacLeod saw a Neuropterid, a fly, and a beetle in Flanders (Bot. Jaarb.
Dodonaea, Ghent, v, 1893, p. 389).

384. Ebulum Garcke.

Flowers reddish-white in colour, odorous, arranged in umbellate panicles ; with
exposed nectar, secreted on the top of the ovary.

1208. E. humile Garcke (=Sambucus Ebulus Z.).—According to Bonnier (z7de
Herm. Miiller, ‘Alpenblumen,’ p. 392), the white flowers of this species, which are
reddish externally, secrete exposed nectar. Kirchner (‘Flora v. Stuttgart,’ p. 670)
states that the diameter of the corolla is 8 mm., and that the stamens project almost
vertically from it.

Visittors.—The following were recorded by the observers, and for the localities
stated.—

Herm, Miiller (Switzerland), Apis, 2 sp. of Bombus, and a Volucella. Borgstette
(Herm. Miiller, ‘Weit. Beob., III, p. 76) (Central Germany), 2 flies (Leptis vitripennis
Mg., and Aricia sp.). von Dalla Torre and Schletterer (Tyrol), 2 bees—Nomada
ferruginata A. 9, and Sphecodes gibbus Z. 4. MacLeod (Pyrenees), 4 Muscids (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 346). Scott-Elliot (Dumfriesshire), a humble-bee,
several flies and Lepidoptera (‘ Flora of Dumfriesshire, p. 81), Loew (Berlin Botanic
Garden), the Muscid Lucilia caesar Z., and 2 Syrphids—Eristalis nemorum Z., and
Helophilus floreus Z.

CAPRIFOLIACEAE 523

385. Sambucus Tourn.

Flowers whitish in colour, often odorous; homogamous or protogynous ; arranged
in large cymes.

120g. S. nigra L. (Herm. Miiller, ‘Fertilisation,’ p. 290, ‘ Weit. Beob.,’ III,
p. 76; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 369; Knuth, ‘Bl. u.
Insekt. a. d. nordfr. Ins.,”’ pp. 80, 156; Kirchner, ‘Flora v. Stuttgart,’ p. 669;
Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—The homogamous yellowish-white
flowers of this species are nectarless, but strongly odorous. They are crowded
together into large flat inflorescences, arranged in tiers, so as to be very conspicuous.
In spite of this, insect visitors are but few; perhaps the pungent odour is repugnant
to many of them, or perhaps the booty is too small. The stamens diverge widely,
while the stigmas are sessile on the ovary in the base of the flower. Insects creeping
over the inflorescences, while devouring or collecting pollen, effect cross- and self-
pollination with equal facility; the
latter also readily takes place auto-
matically, for the stigmas lie in the
line of fall of the pollen.

Warnstorf states that the stamens
spread outwards in the course of
anthesis, and may possibly effect : :

. A ‘ : Fic. 170. Sambucus nigra, L. (after Herm. Miller).
geitonogamy. Self-pollination is also (1) Flower seen from the front. (2) Ditto, seen obliquely
rendered difficult by extrorse dehis- ae front. (3) Ditto, seen obliquely from behind.
cence of the anthers. The pollen- .
grains are pale-yellow in colour, small, ellipsoidal, densely tuberculated, up to
31» long and 15-16 » broad.

According to Kirchner, there are nectar-secreting glands on the petioles ; and
these attract ants, which serve to protect the plant against animals creeping up from
the ground.

VisiTors.—Herm. Miiller observed a beetle in the Alps, and the following in
Central and South Germany.—

A. Coleoptera. Scaradaetdae: all gnawing the petals and other parts of the
flowers: 1. Cetonia aurata Z.; 2. Gnorimus nobilis Z.; 3. Oxythyrea stictica L. ;
4. Phyllopertha horticola Z.; 5. Trichius fasciatus Z. B. Diptera. (a) Stratio-
mytdae: 6. Sargus cuprarius Z., po-dvg. (4) Syrphidae: all po-dvg.: 7. Eristalis
arbustorum Z.; 8. E. horticola Deg.; 9. E. nemorum Z.; tro. E. tenax Z.;
rr. Volucella pellucens Z. C. Hymenoptera. (a) Tenthredinidae: 12. Allantus
nothus A7g.

The following were recorded by the observers, and for the localities stated.—

Knuth (Féhr), 2 po-dvg. hover-flies—Eristalis tenax Z., and Syrphus ribesii Z. ;
(Helgoland), the Muscid Lucilia caesar Z. creeping over the umbels, and effecting
geitonogamy. F. F. Kohl (Tyrol), the ruby-wasp Ellampus aeneus /. Rossler
(Wiesbaden), the moth Botys sambucalis ScAzf. MacLeod (Pyrenees), the beetle
Cetonia aurata Z. (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 346).

1210. S. racemosa L. (Kirchner, ‘Flora v. Stuttgart,’ p. 670; Schulz,
‘Beitrage,’ II, pp. 94-5; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, II, pp. 200, 326.)—
Kirchner describes the flowers of this species as being protogynous, with persistent

524 ANGIOSPERMAE—DICOTYLEDONES

stigmas, in Wurtemberg; but in the South Tyrol, according to Schulz, there is
variation between protogyny, homogamy, and slight protandry. Kerner says that
their odour suggests herring-brine (trimethylamide), while Kirchner describes it as
meal-like. The latter also states that, after the flowers have opened, the tips of the
petals sometimes bend right back, and the stamens diverge so much as to be almost
in one plane; their anthers, however, are still unripe, while the three short stigmas
are fully mature. The tips of the petals then grow to some extent, and assume
a yellowish colour; the anthers dehisce downwards and outwards, the stigmas
remaining receptive. All the flowers of an inflorescence, at a given time, are in about
the same stage of anthesis. As the inflorescences are of more inconspicuous greenish
colour in the first (female) than in the second (hermaphrodite) stage, flowers in the
latter condition are usually first visited by insects, which afterwards transfer the pollen
to those in the female stage. In the second stage automatic self- and cross-pollination
are both possible, for the numerous flowers are turned in all directions. Kerner
describes geitonogamy as taking place in the later stages of anthesis by elongation
and bending of the filaments, so that pollen is applied to the stigmas of neighbouring
flowers.

Visitors.—Redtenbacher (Vienna) noticed 2 Cerambycid beetles — Leptura
virens £., and Strangalia quadrifasciata L.

1211. S. australis Cham. et Schlecht.—K. Miller describes this species as
gynodioecious (Ber. D. bot. Ges., Berlin, ii, 1884).

386. Viburnum L.

Flowers white in colour, possessing an odour of amide; arranged in umbellate
cymes ; homogamous; with exposed to half-concealed nectar, secreted in a flat layer
on the upper surface of the
ovary, immediately below

—e the stigmas, in the base of
the flower.

1212, V. Opulus L.
(Sprengel, ‘Entd. Geh.,’ p.
159; Herm. Miller, ‘ Fertili-
sation, pp. 291-2, ‘ Weit.
Beob.,’ III, pp. 75-6; Knuth,
‘Bloemenbiol. Bijdragen.’)—
Sprengel clearly explained
how the large marginal
neuter flowers serve to render

Fic.171. Vrburnus Opulus, L. (after Herm. Miiller). (1) Mar- x .
ginal flower seen from above, showing the vestiges of anthers and the whole inflorescence more

stigma (X 23). (2) Fertile flower, soon after opening, seen obliquely conspicuous in this species.
from above (x 43). (3) Ditto, after removal of the anterior petals and

stamens (x 44). Hermann Miiller de-

scribes the hermaphrodite
flowers as homogamous. The diverging stamens project from them, and their anthers
are covered with pollen all round, affording abundant booty to pollen-collecting
bees, while the flat layer of nectar only attracts flies and other short-tongued insects.

'

CAPRIFOLIACEAE 525

As these visitors creep over the inflorescences they chiefly effect cross-pollination,
though autogamy is also often brought about. The latter may also take place

automatically, for in many flowers the stigmas are vertically below an anther.
Kerner states that geitonogamy occurs.

VisiTors.—Herm. Miiller gives the following list.—

A. Coleoptera. (a) Axzsotomidae: 1. Anisotoma obesa Schmidt, nect-lkg. (?).
(4) Elatertdae: 2. Athous vittatus #.; 3. Cryptophypnus pulchellus Z. (c) Scara-
baeidae: 4. Oxythyrea stictica Z., freq., dvg. the delicate parts of the flowers; 5.
Phyllopertha horticola Z., devouring the petals and other parts of the flowers.
6. Trichius fasciatus Z., freq., do. (Borgstette). (d) Mitidulidae: 7. Meligethes,
freq. B. Diptera. (a) Lmpidae: 8. Empis tessellata F., skg. (Budd.). (4) Musct-
dae: g. Echinomyia fera LZ. (c) Syrphidae: all freq., skg. and po-dvg.: 10. Eris-
talis arbustorum Z.; 11. E. nemorum Z.; 12. E. sepulcralis Z.; 13. E. tenax Z.;
14. Helophilus floreus Z.; 15. H. pendulus Z. C. Hymenoptera. 16. Halictus
sexnotatus £., po-cltg.

The following were recorded by the observers, and for the localities stated.—

Sprengel mentions Meligethes (‘flower-beetle’) and Phyllopertha horticola Z.
(‘smaller cockchafer’). Knuth, only the humble-bee Bombus terrester Z. 9, vainly
seeking for nectar in the marginal flowers. Alfken (Bremen), 3 bees—1. Andrena
albicans AZzil/. 9, skg. and po-cltg.; 2. A. labialis K. 6; 3. A. tibialis K. 9, po-cltg.
von Fricken (Westphalia), the Scarabaeid Trichius abdominalis Ae. MacLeod
(Flanders), 2 hover-flies, an Empid, a beetle, and a Lepidopterid (Bot. Jaarb.

Dodonaea, Ghent, vi, 1894, p. 373). Scott-Elliot (Dumfriesshire), 2 hover-flies
(‘Flora of Dumfriesshire,’ p. 84).

1213. V. Lantana L. (Kirchner, ‘ Flora v. Stuttgart,’ p. 671; Schulz, ‘ Beitrage,’
Il, p. 95; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 126.)—Kirchner says that the
mechanism of the white flowers of this species resembles that of V. Opulus, but
automatic self-pollination is rendered still more easy, for the anthers are placed
almost vertically over the stigma ; less nectar is also secreted. Schulz describes the
flowers as protogynous, with persistent stigmas. The stamens are at first curved
inwards, but afterwards incline outwards over the edge of the expanded corolla,
so that automatic self-pollination does not usually take place. Kerner states that
geitonogamy is effected automatically.

Visirors.—The following were recorded by the observers, and for the localities
stated.—

Schulz, numerous flies, Hymenoptera, and beetles. Loew (Berlin Botanic
Garden), the Bibionid Bibio laniger AZg. 6, skg. F. F. Kohl (Tyrol), the true

wasp Leionotus rossii Zep. von Dalla Torre and Schletterer (Tyrol), the humble-
bee Bombus pomorum Pz.

387. Weigela Thunb.

Flowers red to white in colour, infundibuliform to campanulate, belonging to
class Hb; with nectar secreted by a green swelling between the base of the style and
the corolla.

1214. W. rosea Lindl. (=Diervilla florida Sved. ef Zucc.). (Herm. Miiler,
‘Weit. Beob.,’ III, pp. 73-4; Knuth, ‘Bloemenbiol. Bijdragen.’)—Hermann Miller
states that in the flowers of this species the base of the corolla is a narrow tube

526 ANGIOSPERMAE—DICOTYLEDONES

12mm. long and 2-3 mm. in diameter, which suddenly expands to two or three
times that width; a total length of about 27 mm. being attained. The throat is
8-10 mm. in diameter, so that a bee of the size of Osmia rufa Z. 9 (which Miiller
observed as a particularly frequent po-cltg. or nect-skg. guest) has room to creep
right in, and reach the nectar with its extended proboscis; larger humble-bees,
on the contrary, cannot enter the flower. When the above-named bee creeps
into the entrance of the flower, it first touches the 2-5 lobed stigma, which
projects beyond the stamens, and dusts it with pollen brought from other flowers.
The bee then encounters the anthers, which surround the throat of the corolla,
and its hairy covering takes up a fresh supply of pollen. Stadler describes the
flowers as protogynous, but says that automatic self-pollination is not excluded
(‘ Beitrage’).

The corollas remain persistent for a considerable time, and even assume a darker
rose-red colour than when the stigma and anthers were mature. The oecological
meaning of this is the same as in Ribes sanguineum and R. aureum. C/ p. 419, and
Vol. I, p. 86.

Visitors.—Besides Osmia rufa Z. 9, Herm. Miiller observed two other bees
(Halictus leucopus X. 9, and H. sexnotatus X. 9), both creeping entirely into the
flower: also a po-dvg. beetle (Dasytes sp.).

Alfken saw the bee Prosopis hyalinata Sm., asleep in the early morning in the
hollow of the flower.

I have frequently noticed in my garden the humble-bee Bombus agrorum /. 9,
pushing the front part of its body into the flower, and inserting its proboscis far
enough to suck the nectar. Though cross-pollination is regularly effected in this
way, I have never remarked the setting of fruit. I have also seen the same
humble-bee as a visitor in Mecklenburg and Pomerania.

388. Diervilla Tourn.

1215. D. japonica Thunb.—Stadler (‘ Beitrige’) describes the flowers of this
species as protogynous; so is the var. amadzl’s (Francke, ‘ Beitrage’).

1216. D. canadensis Willd. (=D. trifolia Moench). (Francke, op. cit.; Loew,
‘ Bliitenbiol. Beitrage,’ I, pp. 61, 63; W. J. Behrens, ‘D. Nekt. d. Bliiten.’)—This
species is protogynous, and possesses a nectary beset with long clavate hairs.

1217. D. floribunda Sieb. et Zucc.—This and other related species agree with
D. canadensis.

389. Aucuba Thunb.

1218. A, japonica Thunb.—
Visitors.—Plateau (Belgium) noticed po-dvg. Muscids (Calliphora vomitoria Z.,
and Musca domestica Z.).

390. Symphoricarpos Dill.

Flowers reddish in colour, bell-shaped, homogamous; with concealed nectar,
secreted, according to Delpino, by a papillose gibbosity on one side of the corolla.
Bonnier says that all the parts of the flower are very rich in sugar. He does

CAPRIFOLIACEAE 527

not consider the swelling at the base of the style as a nectary; but Hermann
Miiller, on the other hand, describes it as such. The swelling in question appears
to me to possess the characters of a nectary in so marked a form that I concur with
the latter view.

1219. S. racemosus Michx. (Herm. Miiller, ‘Fertilisation,’ pp. 292-3, ‘ Weit.
Beob.,’ III, p. 78; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p- 81, ‘ Bliitenbesucher,’
I, p. 16, ‘Bliitenbiol. Beob. in Thiiringen,’ ‘Bloemenbiol. Bijdragen’; MacLeod,
Bot. Centralbl., Cassel, xxix, 1887, p. 119; Loew, ‘ Bliitenbiol. Floristik,’ p. 250, and
bibliography.)—The slightly odorous pendulous flowers of this species obviously
belong to class Hw. They are 7-8 mm. long and 5 mm. broad, and therefore,
according to Hermann Miiller, conveniently receive the head of a wasp (5 mm.
broad, 2-24 mm. thick). Wasps actually appear to be the most frequent visitors
and pollinators, but, in my opinion, the size of the heads of other guests, such as
hover-flies and bees, corresponds equally well to the internal dimensions of the
flowers.

Where the pendulous bell-shaped flower is broadest, its inner surface is closely
beset with numerous long hairs. These stretch
from the five lobes of the corolla to the middle of
the bell, and thus not only effectively protect the
abundant nectar from the rain, but also prevent
it from running out.

The five epipetalous stamens spring from
about the middle of the bell. They converge so
that the introrse anthers are placed at the entrance
of the flower, in the lowest part of the lining of Fic. 172. Symphoricarpos racemosa,
hairs. Immediately above the anthers, where the one Sees
bell narrows, the simultaneously maturing stigma is tinal section (x 24).
situated.

When a nectar-seeking insect, with a head corresponding to the size of the bell,
pushes this into the flower, it first of all comes into contact with the five anthers,
getting covered with pollen, and then one side of it touches the stigma. Hermann
Miller says, however, that little or no pollen remains on the insect’s head by the time
the stigma is reached, partiy because this is but slightly sticky, partly because any
grains that happen to adhere are brushed off again by the thick coating of hairs which
lines the corolla. It is only when withdrawing from the bell that the head of the
insect, now wet with nectar, becomes thoroughly covered with pollen, part of which
is transferred to the stigma of the next flower visited. Cross-pollination is thus
effected. In the absence of insect visitors, automatic self-pollination is probably
always prevented, owing to the pendulous character of the flower, and the relative
positions of anthers and stigma.

VisiTors.—I have seen the humble-bee Bombus agrorum /., and the true wasp
Vespa saxonica /, skg., in Thuringia: in Schleswig-Holstein, Mecklenburg, and
Pomerania, in spite of careful observation of the flowers, I have never seen a wasp
visiting them, but only nect-skg. bees (Apis, Bombus terrester Z.) and hover-flies
(Eristalis sp., Syrphus ribesii Z., Syritta pipiens Z., po-dvg.): also (20. 7.’97), at

528 ANGIOSPERMAE—DICOTYLEDONES

Heringsdorf in the island of Usedom, Apis mellifica Z. ¥, skg., and Bombus lapidarius
L£.%, do., together with numerous nect-skg. sp. of Vespa, especially V. vulgaris Z.,
V. media Reéz., and V. sylvestris Scop.

Herm. Miller (H. M.) and Buddeberg (Budd.) give the following list.—

A. Diptera. Syrphidae: 1. Helophilus floreus Z., skg. (?) (H. M.).
B. Hymenoptera. (a) Apidae: 2. Apis mellifica Z. ¥, freq. skg., (H. M.);
3. Bombus agrorum F/. ¥, skg..(H. M.); 4. B. pratorum Z. ¥, do. (H. M.);
5. Eucera longicornis Z. 6, do. (H. M.); 6. Halictus sexnotatus X. 9, freq.,
skg. and po-dvg. (H. M., Budd.); 7. H. smeathmanellus A. 9, skg. (Budd.);
8. Megachile centuncularis Z. 8, do. (H. M.). (0) Sphegidae: g. Ammophila
sabulosa Z., do. (H.M.). (c) Vespédae: 10, Eumenes pomiformis F. (Budd.), do. ;
11. Odynerus sp., perforating the flowers from outside (H. M.); 12. Polistes
diadema Z/r. (H.M.); 13. P. gallica Z.; 14. Vespa media Re/z. (H.M.); 15. V.
rufa 7. (H.M.); 16. V. saxonica #. (H. M.); 17. V. sylvestris Scop. 9, skg. (H. M.,
Budd.).

Loew noticed the following in Brandenburg (‘ Beitrage,’ p. 42).—

A. Diptera. (a) Muscidae: 1. Lauxania aenea Fal/. (6) Syrphidae: all
skg.; 2. Eristalis arbustorum Z.; 3. Helophiius floreus Z.; 4. Syritta pipiens Z.;
5. Syrphus balteatus Deg.; 6. S. corollae #. B. Hymenoptera. (a) Apidae:
all skg.; 7. Halictus cylindricus 7.6; 8. H. malachurus A. 6. (4) Vesprdae: all
skg.; 9. Eumenes pomiformis 7; ro. Odynerus parietum Z. 11. O. parietum Z.
(=O. renimacula Zep. 9); 12. Vespa sylvestris Scop. ¥.

The following were recorded by the observers, and for the localities stated —

Alfken (Bremen), a Cerambycid (Judolia cerambyciformis Schr., freq.), 2 true
wasps (Vespa media Re/z. ¥, and V. sylvestris Scop.), and 11 bees—1. Andrena
convexiuscula A. 6; 2. Bombus agrorum /. ¥; 3. B. derhamellus K. 9 and
4. B. hortorum Z. ¥; 5. B. jonellus A. ¥ and 6, skg.; 6. B. lucorum Z. ¥ and 3;
7. B. muscorum #. 6; 8. B. pratorum Z. ¥ and 8; 9. B. terrester Z. 8;
ro. Podalirius parietinus #9; 11. Psithyrus vestalis Fourcr. §. Schmiedeknecht
(Thuringia), the bee Andrena combinata Cfr. von Dalla Torre (Tyrol), the bee
Halictus morio /. 9, very freq. MacLeod (Flanders), numerous moths, skg. (Bot.
Centralbl., Cassel, xxix, 1887). C. Schréder (Rendsburg, 9 and ro p. m.), numerous
Noctuids of the genera Agrotis, Mamestra, and Plusia, skg.

391. Linnaea Gronov.

Flowers white in colour, homogamous, provided with nectar-guides on their inner
surface; with entirely concealed nectar, secreted by a thickened part of the receptacle
between the bases of the shorter filaments.

1220. L. borealis L. (Herm. Miiller, ‘Alpenblumen,’ pp. 393-4; Loew,
‘ Bliitenbiol. Floristik,’ p. 249; Lindman, ‘ Bidrag till Kanned. om Skandin. Fjellvaxt.
Blomn. o. Befrukt.’)—The nectar in this species is protected from rain by the position
of the flower, which hangs obliquely downwards. The hairs on the inner surface
of the corolla perhaps serve as a protection against the smaller creeping insects.
Cross-pollination is favoured by the fact that the freely secreting bilobed stigma
projects far beyond the anthers, which mature simultaneously. Hermann Miiller
considers that automatic self-pollination is only possible in flowers that are unusually
pendulous. Perhaps, however, pollen which remains clinging to the hairs lining the
corolla may sometimes reach the stigma.

CAPRIFOLIACEAE . 529

In plants from Tegel, near Berlin, Loew (‘ Bliitenbiol. Floristik.,’ p. 250) observed
but a very small quantity of pollen, which was probably functionless, for the flowers
never became fertilized. Loew suggests that a fungus parasitic in the anthers may
have been the cause of this.

Kerner describes the flowers as funnel-shaped, 10-12 mm. long, and possessing
an odour of vanilla. He considers that they permit the access of tolerably short-
tongued insects.

Visitors.— Herm. Miiller (Alps) observed 3 flies and a Lepidopterid; and

Fic. 173. Leznaea borealis, L. (after Herm. Miiller). A. Flower seen from the side, after removal
of the right half of calyx and corolla (x 7). B. Lower half of the corolla, with the epipetalous stamens
andthenectary. 47, bract; ca, calyx; co, corolla; 2, nectary; ov, ovary; s¢, stigma.

Loew (Brandenburg) noticed the Dolichopodid Neurigona quadrifasciata /., skg. (?)
(‘ Beitrige,’ p. 44).

392. Lonicera L.

Flowers belonging to classes H or L; homogamous, protogynous, or pro-
tandrous; with concealed nectar, secreted by the receptacle or in a pouch of the
corolla.

1221. L. Periclymenum. (Herm. Miiller, ‘ Fertilisation,’ pp. 295-7, ‘ Weil.
Beob.,’ III, p. 75; Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892; MacLeod,
op cit., v, 1893, pp. 390-1; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. go, 156,
‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 234-5, Bot. Centralbl., Cassel,
Ix, 1894, pp. 41-4; ‘Bl. u. Insekt. a. Helgoland’; Warnstorf, Schr. natw. Ver.,
Ver., Wernigerode, xi, 1896.)—The protandrous flowers of this species belong to
class Lm. Hermann Miiller describes them as homogamous, and gives the following
illustration (Fig. 174), which applies to both L. Periclymenum and L. Caprifolium,
except that in the latter the corolla-tube is 5-8 mm. longer.

What is represented in the accompanying illustration (Fig. 174) can be always
observed in full daylight. I gave a similar description of the flowers of L. Periclymenum,
after investigating them at midday, and observing visits of the diurnal hawk-moth
Macroglossa stellatarum Z. (‘ Bl. u. Insekt. a. d. nordfr. Ins,’ p. 80).

At the end of July, 1894, I studied the flower mechanism of the species (adapted
to long-tongued nocturnal Sphingids) at Nieblum in the island of Féhr. This

DAVIS. Wl Mm

530 ANGIOSPERMAE—DICOTYLEDONES

climbing shrub had been cultivated there, and grown very luxuriantly, the bloom
being remarkably fine during that year. I found very noticeable deviations from
the description of Hermann Miller.

The buds are vertical, and the anthers dehisce within them between 6 and
7 p.m.; the stigma becoming receptive at the same time. But automatic self-
pollination cannot take place, because the stigma projects 2 mm. beyond the longest
stamens. (Cf Fig. 175, 1.)

The first flowers open about 7 o'clock; by 8 most of the flowers have already
closed. The lower lip of the corolla first separates from the upper one, then the
stamens successively protrude from under the latter, while the end of the style is
still held fast by its hood-like tip. More rarely the style protrudes before the
stamens. The flower now gradually sinks, passing from the vertical to the horizontal
position. This rotation through go° is complete by the time the style and stamens
have escaped from the upper lip, and the former curves down between the stamens
till it rests on the horizontal or at first slightly curved under-lip.

Fic. 174. Lonicera Caprtfolium, L. (after Herm. Miiller). (1) Flower, natural size, seen
from the side. (2) Ditto, seen from the front. The stigma projects beyond the anthers, and is
therefore first touched by visitors, so that it receives pollen from other flowers.

At the same time a strong odour is exhaled; this is very faint during the
day. Hover-flies (Syrphus sp.) at once make their appearance, and settle on the
anthers to devour the pollen, though frequently they alight on the stigma, and must
therefore occasionally effect crossing. The versatile anthers are placed in front
of the entrance to the flower, with their dehisced sides facing upwards or outwards,
so that the pollen which covers them must be rubbed off by the under-surface of every
hawk-moth probing for nectar. This is secreted by the ovary and lodged in the
corolla-tube (about 25 mm. long), which at this stage is straight. The style projects
some 25 mm., but, as already described, is curved downwards to such an extent at this
stage that contact of hawk-moths with the stigma is impossible. (Cf Fig. 175, 2.)

On the following morning the flowers present quite a different appearance,
though they are still of a whitish colour. The anthers, if insect-visits have taken
place, are destitute of pollen, and the style has changed its position. It is moving
upwards, and now lies between or slightly below or above the stamens. This
upward movement is completed by the time new buds are opening, i.e. between

CAPRIFOLIACEAE 531

7 and 8 p.m.: the filaments have meanwhile curved downwards, and their anthers
shrivelled so much as to look like little withered hooks.

The flower has now reached its second, purely female, stage. The style
projects forwards above the stamens, which, as already stated, are now curved
downwards, and its tip has bent a little upwards. It follows that the stigma now
dominates the entrance to the flower (Fig. 175, 3), and a visiting hawk-moth is
sure to touch it with the under-side of its body, effecting cross-pollination if
a flower in the first stage has previously been visited.

In the course of the day several other changes have taken place. The
upper- and under-lips have rolled up more or less, so that the surface , rendering
the flower conspicuous has become increasingly smaller. At the same time there

1

OS

Fic. 175. Lonticera Periclymenum, L. (from nature; x 1). (1) Bud shortly before opening: the
stigma is already mature, and the stamens have dehisced, but automatic self-pollination is prevented by the
erect position and the greater length of the style. (2) Flower on the first evening: the pollen-covered
anthers are in front of its entrance, the style is bent downwards so much that the stigma is not liable to be
touched by hawk-moths: the upper- and under-lips are but slightly curved, and are white in colour.
(3) Ditto, on the second evening, the style has curved upwards so much that the stigma is in front of the
entrance to the flower; on the other hand, the stamens have curved downwards and the anthers are
shrivelled ; the upper- and under-lips are rolled back so as to occupy less space, and have become yellow.
a, anther; s, stigma.

has been a gradual change of colour, the corolla, originally white inside and red
outside, being now bright yellow. This change is completed by the evening, so
that just before the fresh buds open there are no more pure white flowers to
be seen.

The meaning of these facts in relation to pollination is obvious. Hawk-moths
are attracted from a distance by the fragrance of the flowers, when nearer by the
conspicuous inflorescences, and when quite close by the more clearly visible blossoms,
i.e. the brighter and larger red-and-white ones in the first stage. The moths there-
fore visit these first, and afterwards go to the duller yellow flowers in the second
stage, which are also smaller, owing to the rolling up of the ends of the petals.
The pollen of the former is thus transferred to the latter. Although the two stages
are so sharply outlined against the clear evening sky that it is possible to distinguish
them at once, yet I was not able to confirm the above explanation by direct
observation, for the movements of the hawk-moths which visit the flowers (Sphinx
ligustri Z., and S. convolvuli Z.) are so extremely rapid, and their approach is

Mm 2

532 ANGIOSPERMAE—DICOTYLEDONES

so noiseless, that I found it impossible to discriminate which flowers were first
visited by them.

The flowers in the second stage, with curved corolla-tubes, assume a darker
tint in the course of the next few days, ultimately becoming a dirty orange-brown
colour. The tips of the petals roll up more and more, and the fragrance exhaled
in the evening gets less and less, but the position of the stamens and style remains
the same. A certain amount of nectar is still secreted, and the stigma continues
receptive for a time. Although the visits of hawk-moths become less frequent,
the possibility of cross-pollination remains for some days.

Warnstorf states that inside the basal half of the corolla-tube, in line with the narrow
under-lip, there is a longitudinal yellow ridge, indicated externally by a groove. The
surface of this ridge is beset with small sessile glands secreting abundant nectar, the little
drops of which collect in the base of the tube. An insect must possess a proboscis
at least 15 mm. long in order to reach the beginning of this store. The style with
its capitate stigma usually projects about 28 mm. from the corolla. Self-pollination
is rendered very difficult, if not quite impossible, in such flowers. There are,
however, others in which the stigma only projects about 1 mm. beyond the anthers,
and here it is obvious that autogamy is greatly facilitated should insect-visits
fail. The anthers dehisce in 30-40 minutes after the flower has opened. The
pollen-grains are white in colour, sticky and coherent, tetrahedral, rendered opaque
by numerous short spinose warts, and 88-100 yz in diameter.

Kerner and Warnstorf give the same hour for the opening of the flowers as
myself; the former also states that the fragrance is strongest from 6 p.m. till
midnight. He further describes the subsequent curving of the corolla-tube, adding
that in this way direct contact between stigma and anthers, and therefore automatic
self-pollination, may be brought about. I have not myself seen this; in the flowers in
the island of Féhr it is almost impossible, the stigma projects so far beyond the
anthers. My own observations in Helgoland prove that the mechanism is not
everywhere the same. Here the buds are quite horizontal; except that in free
inflorescences not crowded by others they are at first vertical, subsequently becoming
inclined. Anthers and stigmas mature simultaneously, and three of the stamens are
as long as the style, so that their pollen-covered anthers must touch the stigma,
automatic self-pollination being therefore inevitable. The two other stamens are
shorter by the length of an anther, and consequently serve for cross-pollination only.
The secretion of nectar is so copious that the corolla-tube is often half-filled, giving
access to Lepidoptera with a short proboscis. Even long-tongued humble-bees can
obtain part of the spoil, at the same time effecting cross-pollination.

Hermann Miiller observed a reduction of the corolla-tube from 22-5 mm. in
length to 6 mm. in plants growing under ‘unnatural conditions of life,’ being
subjected to the drip from a roof, which apparently killed them.

Visitors.—In Fdéhr, besides the above-named legitimate pollinators (Sphinx
convolvuli and S. ligustri), I observed other Sphingids (Macroglossa stellatarum Z..,
Deilephila elpenor Z., Smerinthus ocellatus Z.), a Noctuid (Plusia), and po-dvg. hover-
flies (Syrphus, Eristalis, Rhingia, Syritta). I have also seen the humble-bee Bombus
hortorum Z. 9, skg. legitimately and effecting pollination. This insect cannot,
indeed, get all the nectar, but obtains a considerable part of it. On the island of

CAPRIFOLIACEAE 533

Amrum I specially observed Plusia gamma L., freq., which also sucked legitimately
and effected pollination. This moth was so zealous about the matter that I was
able to remove it from the flowers with my fingers.

Herm. Miiller only observed the humble-bee Bombus hortorum Z. 9, but
supposed visits to be made by nocturnal hawk-moths as well. He makes the
following remarks about the humble-bee mentioned :—‘ The bee wasted considerable
time in obtaining a convenient position for sucking, and it crawled from the broad
upper-lip to the mouth of the tube without touching the stigma and anthers; and the
amount of honey it obtained must have been small, for after visiting a few flowers it
flew away, though the plants were in full bloom. Bees, therefore, are only accidental
visitors, which have had no influence in developing the special characters of the
flower’ (‘ Fertilisation,’ pp. 295-7).

The following were recorded by the observers, and for the localities stated—

Knuth (Helgoland) 2 hawk-moths (Deilephila galii Ro//., and Macroglossa
stellatarium £), the Noctuid Plusia gamma Z., and small Noctuids (Bot. Jaarb.
Dodonea, Ghent, viii, 1896, p. 44). Heinsius (Holland), the humble-bee Bombus
hortorum JZ. 4, freq., skg., and the po-dvg. hover-fly Melanostoma hyalinata Fail. 9
(op. cit. iv, 1892, pp. 115-16). MacLeod (Belgium) noticed two humble-bees, skg.
(Bombus hortorum Z., and B. agrorum /.); also the diurnal hawk-moth Macro-
glossa stellatarum Z., skg. Scott-Elliot (Dumfriesshire), 3 Lepidoptera (‘Flora of
Dumfriesshire,’ p. 84). Willis observed (in the neighbourhood of the south coast
of Scotland) the humble-bee Bombus hortorum J. 9, freq., skg. (‘Fls. and Insects
in Gt. Britain, Part I).

1222. L. caprifolium L. (Herm. Miiller, ‘ Fertilisation,’ pp. 293-5; Kirchner,
‘Flora v. Stuttgart,’ p.672; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II, pp. 201, 208, 211.)
—This species bears flowers adapted to nocturnal hawk-moths. The mechanism
(Fig. 174) agrees with that of L. Periclymenum (Herm. Miiller, Kirchner, Kerner),
but the corolla-tube is about 30 mm. long, so that only those Sphingids with a
proboscis of corresponding length can exhaust the nectar. Kerner describes the
flowers as slightly protogynous, while Hermann Miiller considers them to be homoga-
mous. According to Kirchner, the corolla in the unfertilized stage is white or
reddish-white internally and is tinged with pink externally, but later on assumes a bright
yellow colour. Kerner says that, as in the last species, the fragrance is most
pronounced between 6 p.m. and midnight, and that the flowers open in a few
minutes, anthesis lasting three days. This would indicate that the mechanism of
L. Caprifolium agrees with the description I have given above for L. Periclymenum,
but I have not yet had an opportunity of confirming this opinion, to which I first
gave expression in 1894, by actual observation. Kerner says that automatic self-
pollination takes place when insect-visits fail.

Visirors.—Herm. Miiller observed the following Lepidoptera (the figures in
parenthesis give the length of proboscis in mm.).—

(a) Sphingidae: 1. Sphinx convolvuli Z. (65-80); 2. S. ligustri Z. (37-42);
3. S. pinastri Z. (28-33); all these 3 can completely exhaust the nectar; 4. Deile-
phila elpenor Z. (20-24); 5. D. porcellus Z. (20); these 2 can get most of the
nectar; 6. Smerinthus tiliae Z. (3), vainly trying to suck. (4) Moctucdac: 7. Dian-
thoecia capsincola A. (23-5), as 4 and 5; 8. Cucullia umbratica Z. 9 (18-22),
do.; 9. Plusia gamma Z. (15), getting some nectar. (c) Bombycrdae: 10. Dasy-
chira pudibunda ZL. (0), as 6.

534 ANGIOSPERMAE—DICOTYLEDONES

MacLeod (Belgium) saw a hawk-moth (Deilephila sp.).

1223. L. tatarica L. (Herm. Miiller, ‘Fertilisation,’ p. 297.)—This Siberian
species, cultivated in our public gardens and so forth, bears flowers with concealed
nectar. Hermann Miiller says that the corolla-tube is 6-7 mm. long, and secretes
nectar in a shallow pouch at its base. The bright red flowers are homogamous; the
anthers project a little beyond the stigma. Insects, while probing for nectar, touch
the stigma with one side of their heads and the pollen-covered anthers with the
other. When visits are repeated crossing is favoured, though, of course, self-pollina-
tion may also be effected. The latter may also take place automatically, for it is not
unusual to find flowers in which the stigma touches one or two of the anthers.

Visirors.—The following were recorded by the observers, and for the localities
stated —

Herm. Miiller (H. M.) (‘ Weit. Beob.,’ p. 235) and myself (Kn.) (‘ Bloemenbiol.
Bijdragen’).— A. Diptera. Syrphidae: 1. Rhingia rostrata Z. (Kn., H. M.), very
common, skg. and po-dvg. B. Hymenoptera. Apidae: 2. Andrena albicans

Fic. 176. Lonzcera nigre, L. (after Herm. Miller). 4. A pair of flowers, seen from the front.
B. Lower part of a flower in longitudinal section (x 7). 2, nectary.

fill. 9 (H. M.), vainly trying to suck ; 3. Apis mellifica Z.¥ (Kn., H. M.), freq. skg. ;

4. Megachile centuncularis Z. 6 (H. M.), skg. Alfken (Bremen), 3 humble-bees—
1. Bombus derhamellus A. 9; 2. B. sylvarum Z.9; 3. B.lucorum Z.gand ¥. Mora-
witz (St. Petersburg), 2 leaf-cutting bees, freq.—Megachile willughbiella A‘, and

M. circumcincta A.

1224. L. Xylosteum L. (Herm. Miller, ‘ Fertilisation, pp. 297-8; Kerner,
‘Nat. Hist. Pl.” Eng. Ed. 1, II, p. 278.)—Hermann Miller describes the yellowish-
white flowers of this species as homogamous. The corolla-tube is only 3-4 mm.
long, so that the concealed nectar contained in a shallow pouch, and protected at the
base of the tube with hairs, is accessible even to short-tongued insects. The stamens
and style project far out of the flower, and the former diverge widely from the latter.
It follows that insect visitors touch the stamens and stigma with opposite sides of their
heads, and regularly effect cross-pollination. Failing such visitors, automatic self-

CAPRIFOLIACEAE 535

pollination may take place by the fall of pollen on the stigma. Humble-bees are the
only insects which regularly effect crossing, their behaviour being as above described.
Hermann Miiller states that the honey-bee and flies sometimes do not touch the
stigma at all.

Kerner describes the flowers as protogynous, and says that the stigma is at first
in the line of access to the nectar. The style subsequently bends downwards, and
the anthers take up the position previously occupied by the stigma.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller, bees, skg. (Apis mellifica Z.%; Bombus agrorum J. §; B. pra-
torum Z. ¥), and a few hover-flies (Empis opaca /”,, freq., skg.; Rhingia rostrata Z.,
skg. and po-dvg.). Schmiedeknecht (Thuringia), the humble-bee Bombus distin-
guendus Afor.9. Réssler (Wiesbaden), the moth Grapholitha albersana HW. Schlet-
terer and von Dalla Torre (Tyrol), the humble-bee Bombus pomorum Z., ?.

1225. L. nigra L. (Herm. Miller, ‘Fertilisation, p. 297, ‘Alpenblumen,’

Pp- 394-5-)—The flowers of this species are homogamous, and belong to class Hb.
The nectar is protected

against rain by numerous
hairs lining the corolla-
tube. The stigma projects
furthest from the flower,
so that it is first touched
by insect visitors, which
necessarily effect cross-
pollination.. As the style
curves downwards, auto-
matic self-pollination re-
sults from the fall of pollen
should insect-visits fail.
(CA Fig. 176.)

Visitors. — Herm.
Miiller saw the bees Apis
and Halictus sp.; Ricca
observed numerous hum-

E Fic. 177. Lonicera caerulea, L. (after Herm. Miller). 4. A pair of
ble-bees, bees, and flies pendulous flowers. 2. A flower in longitudinal sections (x 4). a, anther;

(Atti Soc. ital. sc. nat., A, nectar; 2, nectary; ov, ovary ; Sd, protecting hairs; s¢, stigma.
Milano, xiv, 1871).

1226. L. caerulea L. (Hildebrand, ‘D. Geschlechts-Vert. b. d. Pfl.,’ p. 18 ;
Ricca, Atti Soc. ital. sc. nat., Milano, xiv. 1871; Herm. Miiller, ‘ Alpenblumen,’
pp. 397-8.)—This species bears yellowish-white pendulous humble-bee flowers,
described by Hildebrand as homogamous, by Ricca as protogynous. Hermann
Miiller says that the corolla-tube is about 10 mm. long, and that the nectar is most
easily sucked by long-tongued bees, especially humble-bees. When these probe the
flowers they first touch the stigma, and then the anthers, thus regularly effecting
cross-pollination. In obliquely hanging blossoms automatic self-pollination can
readily take place by the fall of pollen.

530 ANGIOSPERMAE—DICOTYLEDONES

Visitors.—Herm. Miiller observed 12 Hymenoptera (including 5 species of
humble-bees), 3 Syrphidae, 2 beetles, and 3 Lepidoptera.
Ricca noticed Bombus lapidarius Z., even at an altitude of 2,000-2,500 mm.

1227. L. alpigena L. (Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, II, p. 278;
Schulz, ‘ Beitrige,’ II, pp. 95-7 ; Herm. Miiller, ‘ Fertilisation,’ p. 298, ‘Alpenblumen,’
pp. 395-7-)—This species bears reddish-brown flowers belonging to class Hw,
which are visited and pollinated by bees, humble-bees, and especially wasps. The
pouch of the corolla-tube secretes very abundant nectar, protected by many hairs.
The lower lip, directed obliquely downwards, forms a convenient platform for visitors.
These must first encounter and touch the stigma, and then the anthers, so that
cross-pollination is necessarily effected.

Fic. 178. Lonzcera alpigena, L. (after Herm. Miiller). A pair of flowers soon after opening,
seen from the front (x 4). The one on the right side possesses a supernumerary stamen, but no

corresponding corolla-lobe.
While Hermann Miiller describes the flowers as homogamous, Kerner says
they are protogynous, and that crossing is only possible at first, self-pollination being
inevitable later on as the result of contact between stigma and anthers.
Vistrors.—Herm. Miller observed 9 Hymenoptera (including great numbers of

2 species of wasps), 2 Syrphidae, 2 Lepidoptera, and 2 beetles. Schulz specially
noticed the diurnal hawk-moth Macroglossa stellatarum Z.

1228. L. iberica Bieb—Cultivated plants of this Caucasian species were
investigated by Kirchner in Wurtemberg (‘ Beitrage,’ pp. 62-3). The bright yellow
flowers are slightly protogynous; the lower part of the corolla-tube rises abruptly
to a height of ro mm., and its upper part is 3 mm. long. The under-lip (10 mm.
in length) is downwardly reflexed, while the upper one becomes erect and spreads
cut. Insects probing for the nectar secreted in the base of the flower first touch the

CAPRIFOLIACEAE 537

stigma, which projects beyond the dehisced anthers for about 1-2 mm, so that cross-
pollination is favoured.

Visitors.—Kirchner observed Apis and the humble-bee Bombus lapidarius Z.

1229. L. implexa Ait.—

Visitors.—Schletterer (Pola) observed the ichneumon fiy Gravenhorstia picta
Gore (=Anomalon fasciatum Gir).

1230. L. etrusca Santi.—

VisiTors.—Schletterer (Pola) observed a humble-bee (Bombus argillaceus Scop.),

a Braconid (Bracon (Vipio) castrator /.), and a true wasp (Eumenes mediterranea
Krchb.).

LI]. ORDER RUBIACEAE DC.

LiteraTure.—Knuth, ‘B]. u. Insekt. a. d. nordfr. Ins.,’ p. 81, ‘Grundriss d.
Bliitenbiol.,’ p. 63; Schumann, ‘ Rubiaceae,’ in Engler and Prantl, ‘D. nat. Pflan-
zenfam., IV, 4, pp. 8-13.

Our native Rubiaceae are mostly small plants, bearing flowers of white or
yellow colour, rarely red or blue, and only made conspicuous by being associated in
crowded racemose inflorescences. Nectar is usually sparingly secreted by a fleshy
disk on the ovary. Some foreign species, on the other hand (e.g. of Manettia,
according to Fritz Miiller), possess deeply concealed nectar accessible only to hawk-
moths with a long proboscis or humming-birds with elongated beaks. Among
German species only Asperula taurina and A. azurea, perhaps also Sherardia
arvensis, belong to flower class L, while the other species of Asperula belong to
C, and those of Galium to E.

Some foreign forms are dimorphous, e.g. species of Hedyotis (according to
Treviranus); Borreria, Faramea, and Manettia (Fritz Miller); Mitchella, Knoxia,
and Cinchona (Darwin); Chasalia, Nertera, Ophiorrhiza, and Luculia (Kuhn).
Darwin (‘ Diff. Forms of Fls.’) enumerates 17 genera with dimorphous flowers.

Fritz. Miller has described the flower mechanism of Posoqueria (Martha)
fragrans, native to Brazil (Bot. Ztg., Leipzig, xxiv, 1866, p. 129; xxv, 1867; p. 80).

393. Sherardia Dill.

Flowers bright-violet in colour: perhaps belonging to class L, for the nectar is
secreted by a fleshy disk surrounding the base of the style, and concealed in a narrow
tube, so that it is most easily accessible to small Lepidoptera.

1231. S. arvensis L. (Herm. Miiller,* Weit. Beob.’ II], pp. 71-2; MacLeod,
Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 385-6; Meehan, Bull. Torrey Bot. Cl.,
New York, xiv, 1887, pp. 238-9 ; Schulz, ‘ Beitrage,’ I, p. 64; Kirchner, ‘ Beitrage,’
p- 61.)—Hermann Miiller describes this species as gynodioecious. The hermaphrodite
flowers are somewhat larger than the female ones. The former are imperfectly
protandrous, for the stamens with dehisced anthers curve out of the flower before the
stigmas are fully mature. Not infrequently, however, the stigmas are completely
receptive while the pollen-covered anthers are still on the same level, so that

538 ANGIOSPERMAE—DICOTYLEDONES

automatic self-pollination easily takes place. Schulz says that the corolla-tube of
hermaphrodite flowers is 23-33 mm. long. Autumn flowers are self-fertilized, and do
not open. Schulz also observed gynomonoecism. Kirchner found the hermaphrodite
flowers to be homogamous.

Visitors.—Herm. Miiller supposed small Lepidoptera to be the pollinators of
the flowers, but these have not yet been observed, and the visitors of the inconspicuous
flowers are very few in number. Kirchner was fortunate enough to observe the
following.—

A. Diptera. (ca) Syrphidae: 1. Eristalis tenax Z., freq.; 2. Platycheirus
scutatus AZe., do. (4) Afusctdae: 3. Siphona cristata #.; 4. Caenosia sp.; 5.

Chlorops sp. B. Hymenoptera. Agzdae: 6. Bombus agrorum &. C. Hemiptera.
7. Calocoris seticornis /.

394. Asperula L.

Flowers white, reddish, yellow, or blue in colour; arranged in cymose
inflorescences ; belonging to class C, or more rarely to L.

1232. A.cynanchica L. (Herm. Miller, ‘Fertilisation,’ pp. 302-3, ‘Weit. Beob.,’
III, pp. 72-3; Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, Il, p. 201; Schulz, ‘ Beitrage,.’
I, p. 65; Loew, ‘ Bliitenbiol. Floristik.,’ p. 394.)—The flowers are white or reddish

Fic. 179. Asperula cynanchica, L. (after Herm. Miiller). (1) Flower with pure white, smooth
petals, seen from above (» 7). (2) Ditto, after removal of half the corolla; seen from the side.
(3) Flower with rough petals ornamented with red lines; seen from the side. (4) Ditto, seen from

above. a, anther; 2, nectary; ov, ovary; sv, stigma.

in colour, and Kerner says that they smell like vanilla. Hermann Miiller describes
them as homogamous. The abundant nectar is concealed at the bottom of a
corolla-tube 2 mm. long. In the middle of this are situated the two closely apposed
capitate stigmas, while the converging anthers occupy the entrance to the flower.
Insect visitors are likely to effect cross-pollination, for they usually touch the pollen
and stigma with opposite sides of their proboscis. Automatic self-pollination can
easily take place by the fall of pollen on the stigma. Hermann Miiller was able to
distinguish two different forms of flower in Thuringia; one with smooth, white,
moderately obtuse corolla-lobes, the other with these lobes rough on the upper
surface, ending in sharp recurved tips, and each marked with three red lines. Warnstorf
describes the pollen-grains, examined in water, as yellow in colour, small, spheroidal,
delicately striated, transparent, and about 25 p in diameter.

RUBIACEAE 939

Willis observed gynodioecism in England (Proc. Phil. Soc., Cambridge, ix,
1893).

Visttors.—Herm. Miiller observed the following in Thuringia.—

A. Coleoptera. (a) Elateridae: 1. Agriotes ustulatus Schall., resting inactively
on the flowers. (4) Ze/ephoridae: 2. Danacea pallipes Panz., resting inactively on the

flowers; 3. Dasytes subaeneus Schh., skg.(?); 4. Ebaeus thoracicus Ov. B. Diptera.
(a) Bombyliidae: 5. Systoechus sulphureus AZ2z., skg. (6) E’mpzdae: 6. Empis livida L.,

Fic. 180. Asperiula taurina, L. (after Herm. Miller). .4. Hermaphrodite flower, seen from the side.
B. Pistil and nectary of the same flower. C. Another hermaphrodite flower with the stigmas more
exserted. 2D. A third ditto, with long stylar branches. Z. Half-withered flower, with the branches of
the style even longer. F. Vestigial pistil of a male flower. G. A trimerous male flower (x 7).
m, nectary; ov, ovary; s/, stigma.

freq., skg.; 7. Rhamphomyia sp., eagerly skg., in very large numbers. (¢) AZuscddae :
8. Siphona geniculata Deg., freq., skg.; 9. Ulidia erythrophthalma JZg., do. (@) Stratto-
myidae: 10. Nemotelus pantherinus Z., skg. (e) Svrphidae: 11. Syritta pipiens Z.,
hovering and skg. C. Hymenoptera. Afidac: 12. Bombus agrorum /. 4%,
casually trying to suck. D. Lepidoptera. (a) Geometridae: 13. Minoa murinata
Scop., skg. (6) Rhopalocera: 14. Coenonympha arcania L., skg.

Loew noticed the bee Exoprosopa picta J/g., skg., in Steiermark (‘ Beitrage,’
p-. 51). MacLeod saw 2 flies in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii,
1891, p. 345).

540 ANGIOSPERMAE—DICOTYLEDONES

1233. A. montana Waldst. et Kit. (=A. cynanchica, according to the /ndex
Kewensts).—Kirchner (‘ Beitrage,’ pp. 59-60) says that in the canton Valais the
corolla-tube of this species is 4-53 mm. long. The flowers are homogamous,
with styles of varying length and persistent stigmas. In the long-styled form
self-pollination is hindered, but in the short-styled one it easily takes place.

1234. A. glauca Bess. (=A. galioides Bzed.).—Schulz (‘ Beitrage’) says that
the white or reddish-white odorous flowers of this species are homogamous or slightly
protandrous. As the anthers usually remain above the middle of the flower till all
the pollen is shed, automatic self-pollination is inevitable.

Vis1Tors.—Schulz states that these are numerous small insects of the orders
Diptera, Hymenoptera, and Lepidoptera; but in many cases these undoubtedly only
effect self-pollination.

1235. A, odorata L. (Sprengel, ‘Entd. Geh.,’ p. 84 ; Herm. Miller, ‘ Fertili-
sation, p. 304, ‘ Weit. Beob.,’ HI, p. 73; Knuth, ‘Bloemenbiol. Bijdragen.’)—
Hermann Miiller says that the white flowers of this species, which smell of cumarin,
possess the same mechanism as A. cynanchica. Warnstorf describes the pollen-
grains as white in colour, ellipsoidal, smooth, about 25 p» long and 12-15 p» broad.

Vistrors.—Herm. Miiller (H. M.) and myself (Kn.) have observed the
following.—

A. Coleoptera. (a) Cerambyctdae: 1. Grammoptera levis #. (H. M.), not
infrequent, po-dvg. (0) Zelephoridae: 2. Dasytes sp. (H. M.). (¢) Mordeliidae :
3. Anaspis frontalis Z. (H. M.), freq. (d@) Mitedulidae: 4. Meligethes (H. M.), freq.
B. Diptera. (a) Empidae: 5. Empis tessellata /. (H. M.), one, skg. (0) ALusczdae:
6. Siphona geniculata Deg. (H. M.), freq. skg. (c) Syrphidae: 47. Eristalis nemo-
rum Z. (H. M.), skg.; 8. Rhingia rostrata Z., one, skg. (Kn.); 9. Syritta pipiens
L., freq., skg. (H. M., Kn.). C. Hymenoptera. <Apzdae: 10. Apis mellifica Z. ¥,
freq., skg. (H. M., Kn.); 11. Halictus cylindricus /. 9, skg. (Kn.). D. Lepidoptera.
Microlepidoptera: 12. Elachista sp., skg. (H. M.).

1236. A. taurina L. (Herm. Miller, ‘Fertilisation, pp. 303-4, ‘ Alpen-
blumen,’ pp. 390-2.)—The white colour of the corolla, and its long (9~11 mm.)
narrow tube, indicate that this species is visited by moths. It is andromonoecious, and
the hermaphrodite flowers are markedly protandrous (Fig. 180). According to
Kerner (‘ Nat. Hist. P].,’ Eng. Ed. 1, I, pp. 325-6), the stigmas, owing to curving of
the stylar branches, are later on brought so close to the anthers of the neighbouring
male flowers that geitonogamy is automatically effected.

Visirors.—Herm, Miiller did not observe the actual pollinators, but saw a few
casual visitors (a Bombylius, an Empis, an Echinomyia, a Syritta, and 2 beetles).

Loew (Berlin Botanic Garden) noticed a Syrphid (Melithreptus scriptus Z.,
po-dvg.), and 2 bees (Andrena nitida Fourcr. 9, po-cltg.; Prosopis communis JVy7. 6,
settling on the stamens and po-dvg.).

1237. A. azurea Jaub. et Spach. (=A. orientalis Bozss. ef Hohen.). (Herm.
Miiller, ‘ Weit. Beob.,’ III, p. 73.)—Hermann Mfiiller says that the flowers of this species
conceal their nectar in a tube as long and narrow as that of A. taurina, but their
blue colour indicates that they are pollinated by butterflies, not by moths.

1238. A, tinctoria L. (Herm. Miiller, ‘Weit. Beob.,’ III, p. 72; Schulz,
‘ Beitrage,’ I, p. 65.)—Hermann Miiller says that in the homogamous flowers of this

RUBIACEAE 541

species the corolla-tube is scarcely 2 mm. long: the anthers occupy its throat, and
the two capitate stigmas are situated a little below the middle of it. Nectar-sucking
insects touch the stigmas and anthers with opposite sides of their proboscis, and
therefore usually effect geitonogamy or xenogamy. The stamens converge towards
the end of anthesis, so that pollen can fall upon the stigmas, and effect automatic

self-pollination as a last resort if insect-visits have failed. Schulz also observed
protandry.

Vistrors.—Herm. Miller (Thuringia) saw a Muscid (Ulidia erythrophthalma
Mg., skg.), several small ichneumon flies, and a small Gelechid moth, skg. Roéssler
records the Pyralid Orobena limbata Z., for Wiesbaden.

1239. A. stylosa Boiss.—

VisiTors.—Loew saw Apis, skg., in the Berlin Botanic Garden.

‘1240. A. scoparia Hook. f., and 1241. A. pusilla Hook. f. (=A. Gunnii
Hook. f.).—These two Tasmanian species are described by Treviranus as dimorphous
(Bot. Ztg., Leipzig, xxi, 1863, p. 6).

395. Rubia L.
Flowers small, greenish ; homogamous, with exposed nectar.

1242. R. tinctorum L. (Kirchner, ‘ Beitrage,’ p. 69.)—Aithough the flowers
of this species are united into cymose inflorescences, they are not at all conspicuous,
owing to their small size and green colour. Kirchner gives the diameter of the flat
corolla as 5 mm. When the flowers open, the almost sessile anthers have already
dehisced. The two rounded stigmas are borne on such short styles that they are
situated about the level of the lower part of the anthers. They remain in this
position for some time after the anthers shrivel up, and continue to be receptive.
Automatic self-pollination is easily effected, and regularly takes place. Kirchner,
however, observed insect visitors (small sucking Hymenoptera and flies), by which
crossing may be brought about. Nectar is secreted in the base of the saucer-shaped
corolla-tube (only half a mm. deep), and is accessible to all comers.

396. Galium L.
Herm. Miller, ‘ Fertilization,’ pp. 300-2.

Flowers white or yellow in colour; arranged in cymose inflorescences: with
exposed nectar. Hermann Miller says that the transfer of pollen to the stigmas
is chiefly effected by the feet and only to a lesser extent by the proboscis of insect
visitors that creep about the inflorescences. Probably automatic geitonogamy of the
small crowded flowers is possible in all species by the fall of pollen upon the stigmas
of flowers at a lower level.

1243. G. cruciata Scop. (Darwin, ‘Different Forms of Fs.,’ p. 286; Kirch-
ner, ‘Flora v. Stuttgart,’ p. 666, ‘ Neue Beob.,’ p. 65.)—The inconspicuous greenish-
yellow flowers of this species smell like honey, and are arranged in scanty
inflorescences (helicoid cymes). Darwin and Kirchner describe them as andro-
monoeciously distributed, the lower ones being male, and the upper ones hermaphro-
dite. Schulz (‘ Beitrage,’ I, p. 66) investigated very numerous plants from different

542 ANGIOSPERMAE—DICOTYLEDONES

localities, and only occasionally observed such relations. He found, on the contrary,
that the first flowers to open in each main and lateral axis are hermaphrodite, while
the later ones are mostly male. The hermaphrodite flowers are more or less dis-
tinctly protandrous, rarely homogamous. Self-pollination is therefore rare or
impossible, despite the centripetal movement of the stamens.

Visitors.—Schulz observed bees, and Schletterer (Pola) noticed the saw-fly
Athalia rosae Z., var. cordata Lep.

In Dumfriesshire, 7 hover-flies and several Dolichopodids were recorded
(Scott-Elliot, ‘Flora of Dumfriesshire,’ p. 84).

1244. G. Mollugo L. (Herm. Miiller, ‘Fertilisation,’ pp. 300-1, ‘ Weit. Beob.,’
III, pp. 69-70; Knuth ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ ‘ Bliitenbiol. Beob. a. d.
Ins. Riigen’; Schulz, ‘Beitrage, I, p. 67; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1,
II, p. 344.)—The small white flowers of this species are arranged in conspicuous
inflorescences. Hermann Muller describes them as protandrous. Like all our native
Rubiaceae, they secrete nectar on a disk above the ovary, and surrounding the base
of the style. The quantity is very small, and forms only a thin layer.

Fic. 181. Galinm Mollugo, L. (after Herm. Miller). (1) Young flower, with erect stamens and
styles. (2) Older flower with stamens curved out of it, and styles diverging. (3) Middle of the flower,
seen from above, more highly magnified. a, stigma; 4, fleshy disk of the ovary.

In younger flowers the stamens are erect, with anthers pollen-covered all over,
while the two capitate stigmas are still closely apposed, but already receptive. The
stamens subsequently spread out, and finally bend quite out of the flower; the two
styles also diverge. In this second stage, therefore, cross-pollination is likely to
result from the visits of insects. Schulz states that autumn plants in particular are
frequently homogamous, and as in these the anthers are at first situated above the
stigmas, self-pollination may take place. I have never observed the incurving of
the stamens leading to autogamy that is described by Kemer as occurring in this
species, G. infestum, and G. tricorne.

Visitors.—I observed the following Diptera, skg., in the island of Sylt (‘ Weit.
Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 235).—

1. Coenosia tigrina #.; 2. Dolichopus aeneus Deg.; 3. Hylemyia sp.; 4. H.
variata /.; 5. Sargus cuprarius £.; 6. Scatophaga stercoraria Z.; 47. Spilogaster
communis &.-D.; 8. S. duplaris Zef#.; 9. 5. duplicata AZg.; 10. Stomoxys stimulans
Mg.9; 11. Thereva nobilitata /.; and in Riigen the beetle Cantharis fulva Scop.,
po-dvg.

Herm. Miiller gives the following list.—

A. Coleoptera. Ocdemeridae: 1. Oedemera podagrariae Z., po-dvg. (Thurin-
gia). B. Diptera. (a) Bombylidae: 2. Anthrax flava Afg., not infrequent, nect-
Ike. (Thuringia); 3. Systoechus sulphureus J/ck., skg., probably piercing the

RUBIACEAE 543

nectary (nect-lkg., Thuringia). (4) Muscidae: 4. Musca corvina F.; 5. Scato-
phaga merdaria /., skg. (c) Stratiomyidae: 6. Odontomyia viridula /., not infre-
quent, nect-lkg. (d@) Syrphidae: 7. Melithreptus sp., po-dvg. (Buddeberg) ; 8. Mero-
don aeneus A/g., po-dvg. (Thuringia); 9. Syritta pipiens Z., freq., skg. and po-dvg.;
10. Syrphus ribesii Z., do. (e) Tipulidae: 11. Pachyrhina crocata Z., skg.
C. Hymenoptera. Sphegzdae: 12. Ammophila sabulosa Z. 9.

The following were recorded by the observers, and for the localities stated—

Alfken and Leege (Juist), the butterfly Pieris napi Z., and the Noctuid Plusia
gamma Z£. Verhoeff (Norderney).—A. Coleoptera. Zelephoridae: 1. Dolicho-
soma lineare Ross?. B. Diptera. (a) Dolichopodidae: 2. Dolichopus aeneus Deg.
(4) Empidae: 3. Hilara quadrivittata AZe. (c) Afuscidae: 4. Anthomyia sp. 5.
Aricia incana Wred.; 6. Cynomyia mortuorum Z., skg.; 7. Hydrotaea sp.,1 6; 8.
Lucilia caesar Z., skg.; 9. Miltogramma sp.; 10. Sarcophaga striata /., skg.; 11.
Sepsis cynipsea Z. (d) Syrphidae: 12. Eumerus sabulosus Fal/. 9; 13. Melithrep-
tus menthastri Z. 9, skg.; 14. Platycheirus clypeatus J/g.; 15. Syritta pipiens Z.,
skg. (e) Zherevidae: 16. Thereva anilis Z.,ske. C. Hymenoptera. (a) Formi-
ctdac: 17. Myrmica rubra Z.; 18. M. rugulosa Ny/. a ¥. Arachnida. Zromér-
ditdae: 19. Rhyncholopus phalangioides Deg. Verhoeff describes the last-mentioned
mite as a constant visitor (Ent. Nachr., Berlin, xviii, 1892). Loew (Silesia)
(‘ Beitrage,’ p. 29), an Oedemerid (Chrysanthia viridis Schmzd/), and a hover-fly
(Melithreptus scriptus Z., skg.). MacLeod (Flanders), 3 flies and a beetle (Bot.
Jaarb. Dodonaea, Ghent, v, 1893, p. 386); (Pyrenees), 4 Muscids and 2 Syrphids

(op. cit., p. 345). Scott-Elliott (Dumfriesshire), an Empid, and several other Diptera
(‘Flora of Dumfriesshire,’ p. 86).

1245. G. sylvaticum L. (Kirchner, ‘Flora v. Stuttgart’; Schulz, ‘ Beitrige,’
I, p. 67.)—According to Kirchner and Schulz, the small white flowers of this species
are protandrous, and their mechanism agrees with that ef G. Mollugo; the stamens,
however, do not bend outwards in the same way, but remain incurved, so that auto-
matic self-pollination can easily take place.

VisiTors.—The following were recorded by the observers, and for the localities
stated.—

Knuth, the po-dvg. hover-fly Syritta pipiens Z. (‘Bloemenbiol. Bijdragen’).
Herm. Miiller (Bavarian Oberpfalz) (‘Weit. Beob.,’ III, p. 69)—A. Coleoptera.
(2) Cerambyctdae: 1. Leptura testacea Z. 4, dvg. the anthers. (0) Lyczdae: 2. Dic-
tyoptera sanguinea Scop., resting inactively on the flowers. (c) Oedemeridae: 3.
Oedemera flavescens Z., its mouth busy with the anthers. B. Diptera. (a)
Muscidae: 4. Sarcophaga sp., freq., skg. (6) Syrphidae: 5. Melithreptus men-
thastri Z., skg. Loew (Switzerland) (‘ Beitrage,’ p. 56)—2 Bombyliids, skg.:
(Anthrax maura Z., and Argyromoeba sinuata Fa//.), and the Syrphid Melanostoma
barbifrons F.

1246. G. sylvestre Pollich. (Herm. Miiller, ‘Alpenblumen,’ pp. 389-90;
Schulz, ‘ Beitrage,’ I, p. 67; Knuth, ‘ Bloemenbiol, Bijdragen.’)—The white flowers
of this species agree essentially as regards their mechanism with those of G. Mollugo,
but the stamens after dehiscing do not curve so far outwards. They vary between
protandry and homogamy; in the latter case automatic self-pollination easily takes
place. Hermann Miiller and Schulz state that in the Alps the flowers are larger
(5-7 mm. in diameter) than in the lowlands.

Visitors.—Herm. Miiller observed 2 Syrphids and 12 Lepidoptera. Schulz

saw flies, beetles, small bees, and small Lepidoptera (especially Noctuids). I only
noticed the hover-fly Syritta pipiens Z., po-dvg., at Kiel.

544 ANGIOSPERMAE—DICOTYLEDONES

MacLeod (Pyrenees) observed a variety of this species (possibly G. Lapeyrou-
sianum) to he visited by a beetle, 3 Muscids, and 3 Syrphids. He describes it as
belonging to flower class C, while the other species of Galium belong to E.

1247. G. verum L.xG. Mollugo L. (=G. ochroleucum Wo/f). (Knuth,
‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.” p. 235.)—In the island of Sylt
(2. 7.93) I have seen numerous insects visiting in succession the flowers of
G. verum Z. and G. Mollugo Z., and bringing about an interchange of pollen.
G. ochroleucum Wolf, which grows along with these two species, proves that this
crossing is effective, for it is a hybrid between them.

Visitors.—I observed the following Diptera, ske.—

1. Coenosia tigrina #.; 2. Dolichopus aeneus Deg.; 3. Hylemyia sp. 9.
4. H. variata #.; 5. Spilogaster communis 2.-D.; 6. S, duplaris Ze#.; 7. S.
duplicata Afg.; 8. Stomoxys stimulans J/g. 9.

1248. G. verum L. (Herm. Miiller, ‘Fertilisation, p. 301, ‘ Weit. Beob.’ III,
p- 70; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 387; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ pp. 82-3, ‘ Bli-
tenbiol. Beob, a. d. Ins. Riigen,’ ‘ Weit.
Beob. ti. Bl. u. Insekt. a. d. nordfr. Ins.,’
p. 235; Schulz, ‘Beitrage, I, p. 67.)—
The plants of this species that I examined
in the island of R6m possessed the fol-
lowing flower mechanism.—The buds are
odourless, but when the corolla expands
a very strong smell of cumarin is exhaled.
(Kerner compares it to that of honey.)
The flowers are only 4 mm. in diameter,
but being crowded into dense inflorescences
their intense yellow colour makes them
conspicuous from a distance. They are
markedly protandrous. In the first stage
Fic. 182. Galtvunt sylvestre, Pollich (after 2

Herm. Miller). 4. Young flower (x 7). 8, Of anthesis the four stamens bend back so
fone fe ae far that the lower parts of their filaments
(x 16). 2, nectary; ov, ovary ; s/, stigma. lie between the lobes of the flat expanded
corolla. At the same time the upper
parts of the filaments are curved upwards, so that the dehisced anthers are in the
way of chance visitors. After these have completely or partly shed their pollen,
the two styles, so far united, separate, grow a little, and raise the mature stigmas

almost to the level occupied by the anthers during the first stage of anthesis.
Autogamy occasionally results from the bending over of the filaments, till the
anthers touch the stigmas. Automatic geitonogamy is frequently brought about by the
fall of pollen from a higher flower on to the stigmas of a lower one. Lastly, the
inflorescences are so crowded, and different stocks grow so close together, that
xenogamy may possibly be effected by the wind. Hermann Miller describes
a remarkable difference in the size of flowers from different plants, and indicates
this in his illustration. (Cf Fig. 183.) I have not noticed this in Rém, nor

RUBIACEAE 545

any of the North Frisian Islands, where the plant grows in enormous numbers
on the dunes. Schulz, however, observed considerable variations as to the size
of flowers at Halle and in Thuringia, the extreme forms being connected by
a large series of intermediate stages. He further remarked variations between
strong protandry and complete homogamy. In the latter case automatic self-
pollination is possible; but is prevented at a later stage by the stamens bending
quite out of the flowers.

Visrrors.—Hermann Miiller gives the following list (T.=Thuringia; O.=
Bavarian Oberfals).—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia bifasciata fill, dvg. the
anthers (T.). (6) Llateridae: 2. Agriotes gallicus Lac. (T.). (¢) Mordellidae:
3. Mordella aculeata Z.(T.); 4. M. fasciata #. (T.). (d) Oedemeridae: 5. Oedemera
podagrariae Z., po-dvg. (T.). (e) Scarabaeidae: 6. Cetonia aurata Z., dvg. the
flowers (T.). B. Diptera. (a) Bombyhidae: 7. Anthrax flava AZg., nect-lkg. (O.).

Fic. 183. Galium verum, L. (after Herm. Miiller). (1) Young flower, from a plant with very small
blossoms. The anthers are covered with pollen, and the stigmas are immature (x 7). (2) Older flower
of the same plant with faded anthers, curved out of the corolla, and mature stigmas. (3) Flower of
a plant with large blossoms, in the middle of anthesis: older than 1, younger than 2(x 7). (4) Ditto,
seen from the side.

(6) Conopidae: 8. Conops flavipes Z., nect-lkg. (¢) Afuscidae: 9. Ulidia erythro-
phthalma J/g. freq., nect-lkg. (T.). (d@) Syrphidae: 10. Eristalis arbustorum Z.,
po-dvg. (T.). ©. Hymenoptera. (a) Aprdae: 11. Halictus cylindricus /. 6,
nect-lkg. (O.); 12. Prosopis sp. 4, nect-lkg. (O.). (4) Chrystdidae: 13. Holopyga
ovata Dahib., nect-lkg. (T.). (c) Zenthredinidae: 14. Pachyprotasis rapae X.,
‘ nect-lkg. D. Lepidoptera. Sphingidae: 15. Macroglossa stellatarum Z., vainly
searching for nectar (T.); 16. Zygaena lonicerae Esp. (7Z.).

The following were recorded by the observers, and for the localities stated —

Knuth (Sylt and Foéhr), numerous skg. Diptera.—1. Coenosia tigrina F.;
2. Dolichopus aeneus Deg.; 3. Hylemyia sp.; 4. H. variata #.; 5. Musca sp.;
6. Spilogaster communis &.-D.; 7. S. duplaris Ze//.; 8. S. duplicata MMg.;
g. Stomoxys stimulans J/g.; the hover-fly Syritta pipiens Z., po-dvg., and the
butterfly Epinephele janira Z., trying to suck: (Riigen), the beetle Strangalia
melanura Z., po-dvg., and the hover-fly Syritta pipiens Z., do.: (Helgoland),
Muscids, all skg.—1. Coelopa frigida Fa//.; 2. Lucilia caesar Z.; 3. Scatophaga
stercoraria Z.; 4. small undetermined sp. (Bot. Jaarb. Dodonaea, Ghent, viii, 1896,
p. 34). Alfken (Bremen), the Chrysomelid Agelastica halensis Z., freq. MacLeod
(Pyrenees), 2 Muscids (op. cit., iii, 1891, p. 345). Scott-Elliott (Dumfriesshire),
2 hover-flies and 4 Muscids (‘Flora of Dumfriesshire,’ p. 85).

1249. G. boreale L. (Axell, ‘Om Anord. for Fanerog. vaxt. Befrukt.,’ p. 97;
Herm. Miller, ‘ Fertilisation,’ pp. 301-2, ‘ Weit. Beob.,’ III, p. 70, ‘ Alpenblumen,’
p. 390; Schulz, ‘Beitrage,’ I, pp. 66—-7.)—This species agrees with G. sylvestre as
regards secretion of nectar, its slight protandry, and the relative positions of stamens
and pistil. It follows that crossing is likely to be effected by insect-visits; falling

DAVIS. I Nn

546 ANGIOSPERMAE—DICOTYLEDONES

these, automatic self-pollination is possible. As in G. Mollugo, the withered stamens
hang out of the flower.

A. Schulz, confirming the earlier observations of Axell, describes the flowers as
more or less distinctly protandrous, but he found them to be homogamous in the
Riesengebirge. In the last case automatic self-pollination is possible; later on,
however, this is prevented by an outward curving of the stamens, although in very
many flowers the stigmas are quite receptive at the time when the anthers are
dehiscing. Warnstorf (Verh. bot. Ver., Berlin, xxxvii, 1895) describes the flowers as
homogamous or protogynous (op. cit., xxxviii, 1896).

VisiTors.—Herm. Miiller observed a hover-fly and a Lepidopterid in the Alps,
and the following in Westphalia and Thuringia.—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia bifasciata JZull., dvg. the
anthers. (4) Chrysomelidae: 2. Luperus flavipes Z. (c) Dermestidae: 3. Anthrenus
claviger Zr., nect-lkg. (d@) Aordellidae: 4. Mordella aculeata Z., in large numbers,
nect-lkg. B. Diptera. (a) Muscidae: 5. Ulidia erythrophthalma Afg. (Thuringia).
(4) Syrphidae: 6. Tropidia milesiformis Fa//., nect-lkg. C. Hymenoptera.
(a) Apidae: 7. Prosopis brevicornis Myl. 8, skg. (6) Zenthredinidae: 8. Tarpa

cephalotes #., making erratic visits. D. Lepidoptera. Aficrolepidoptera: 9. A
small moth, skg.

1250. G. palustre L. (Axell, ‘Om Anord. for Fanerog. Vaxt. Befrukt.,’
p- 97; Kirchner, ‘Flora v. Stuttgart,’ p. 664.)—Axell was the first to call attention
to the protandry of this species. According to Kirchner, there is the same possibility
of automatic self-pollination as in G. sylvaticum.

Visitors.—The following were recorded by the observers, and for the localities
stated. —

Verhoeff (Norderney), the Empid Hilara quadrivittata J4/g., and the Muscid
Sepsis cynipsea Z. MacLeod (Flanders), a hover-fly, a Siricid, an ichneumon fly, and
a beetle (Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 488). Scott-Elliot (Dumfriesshire),
2 hover-flies and 2 Muscids (‘ Flora of Dumfriesshire,’ p. 85).

1251. G. uliginosum L. (Axell, ‘Om Anord. for Fanerog. Vaxt. Befrukt.,’
p- 973 MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 387; Kirchner, ‘ Flora
v. Stuttgart,’ p. 665; Schulz, ‘ Beitrige,’ I, p. 66.)—Axell was the first to recognize
the protandry of this species. Kirchner says that the white flowers have the same
mechanism as those of G. Mollugo. In autumn, according to Schulz, flowers are
produced which do not open, but are fertilized cleistogamously.

Lindman describes the plants of the Dovrefjeld as first protandrous, then
homogamous. The flowers possess an odour similar to that of G. verum; at
first the pollen-covered anthers incline together above the middle of the flower,
while the stigma is still immature, though it becomes receptive before all the
pollen is shed. Even after the anthers have withered, they hang down loosely
from their filaments, which remain curved inwards. The style then grows to
such an extent that the stigmas occupy the position previously taken up by the
anthers. Automatic self-pollination easily takes place during the homogamous
stage.

1252. G. Aparine L. (MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893,
1. 388; Kirchner, ‘Flora v. Stuitgart,’ p. 665; Knuth, ‘Bloemenbiol. Bijdragen.’)—

RUBIACEAE 547

Kirchner describes the small white inconspicuous flowers of this species as pro-
tandrous. The stamens, however, do not bend back out of the flower, so that
when the stigmas mature and diverge later on, they always touch the anthers,
which, though dry, still carry pollen, so that automatic self-pollination is constantly
secured. Darwin says that this is followed by the setting of fruit.

Visitors.—I observed the po-dvg. hover-fly Syritta pipiens Z.

In Dumfriesshire a wasp, an ichneumon fly, and a Muscid were recorded
(Scott-Elliot, ‘ Flora of Dumfriesshire,’ p. 87).

1253. G. purpureum L. (Schulz, ‘Beitrage,’ Il], p. 97.)—Schulz states that
the stamens are erect in the dark brownish-red flowers of this species, so that the
anthers lie almost immediately above the stigmas, which mature simultaneously.
By the fall of pollen autogamy is inevitably brought about.

Visitors.—T hese are hover-flies, with small wasps and other little Hymenoptera,
which usually effect self-pollination, though sometimes they undoubtedly bring about
crossing.

1254. G. tricorne Stokes. (Herm. Miiller, ‘Weit. Beob.,’ III, pp. 70-1.)—
Herm. Miiller states that nectar is abundantly secreted in this species, but
the solitary white and yellowish-white flowers are so small that they do not
Teceive many visitors. Automatic self-pollination therefore regularly takes place:
the stamens do not bend outwards, but their anthers remain above the stigma,
which matures simultaneously. Kerner says that automatic self-pollination is
brought about by the anthers touching the stigmas as a result of incurving of
the filaments.

Visirors.— Herm. Miiller only observed a Muscid (Anthomyia), nect-lkg.

1255. G. lucidum All. (=G. corrudaefolium V7//.). (Schulz, ‘ Beitrage,’ II,
pp. 97-8.)—Schulz states that the flower mechanism of this species resembles that of
G. Mollugo. Protandry more or less marked. Self-pollination is usually prevented
by outward curving of the stamens. The styles elongate considerably during
anthesis.

VisiTors.—Schulz observed many small insects (flies, small Hymenoptera,
beetles), which frequently effected self-pollination.

1256. G. rubrum L. (Schulz, op. cit.)—The flowers of this species are red or
dark red in colour, and Schulz describes them as homogamous. Since the stamens
bend outwards automatic self-pollination is not readily effected; but it is the only
important means of fertilization, for Schulz, in spite of repeated observations in
favourable weather, only saw 2 hover-flies visiting the flowers.

Visitors.— Vide supra.

1257. G. rubioides L. (Kirchner, ‘ Beitrage, p. 61.)—The plants of this
species examined by Kirchner in the Berne Botanic Garden were protandrous.
The flowers are white and shaped like a shallow funnel: the stamens are at first
erect, but later on bend outwards, and their anthers drop off. It is only when this
has taken place that the styles elongate and the stigmas diverge.

1258. G. helveticum Weigel. (Schulz, ‘ Beitrage,’ Il, p. 99.)—The whitish-
yellow or greenish-yellow flowers of this species spread out into a star of about
Nn2

548 ANGIOSPERMAE—DICOTYLEDONES

34-5 mm. in diameter. Schulz states that the stamens usually remain almost erect
throughout anthesis. The anthers are therefore situated above the stigmas, which
mature simultaneously, so that automatic self-pollination is inevitable. For the same
reason, insect visitors probably always effect self-pollination.

Visitors.—Schulz observed numerous flies, beetles, small bees, and small
Lepidoptera in the Alps.

1259. G. saxatile L. (Herm. Miiller, ‘Weit. Beob.,’ III, p. 69; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.’ p. 83, ‘Bliitenbiol. Beob. a. d. Ins. Riigen.’)—
The plants of this species observed by me in the North Frisian Islands and at
Kiel were protandrous, their flower mechanism agreeing in the main with that
of G. Mollugo. The anthers dehisce before the styles diverge. In young
flowers the stamens are erect, but as the styles elongate they separate so widely
that they come to lie between the lobes of the corolla, while the stigmas assume
the position at first occupied by the anthers. Automatic geitonogamy can easily
take place, partly by the styles curving outwards till the stigmas touch the anthers
of neighbouring flowers, partly by the fall of pollen upon the stigmas of adjacent
blossoms of the same stock.

VistTors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller, the Cerambycid Leptura livida #., dvg. the flowers, and the
Syrphid Syritta pipiens Z., freq., skg. and po-dvg. Knuth (Riigen), the Syrphid
Eristalis sepulcralis Z., skg. and po-cltg. Scott-Elliot (Dumfriesshire), 3 hover-flies
(‘Flora of Dumfriesshire,’ p. 86).

1260. G. persicum DC. (=G. coronatum S7d¢h. ef Sm.).—

VisiTors. —Loew (Berlin Botanic Garden) saw an undetermined Pyralid.

397. Lxora L.

1261. I, salicifolia DC. (=I. fulgens Roxd.).—According to Willis (Proc.
Phil. Soc., Cambridge, viii, 1892), the flowers of this species secrete nectar at the
bottom of a long corolla-tube, so as to be accessible only to long-tongued insects.
The pollen is shed on the still immature stigma. The style serves as an alighting
place, and in the first stage of anthesis affords pollen to visitors, while in the second
stage it terminates in the mature stigma. ,

1262. I. coccinea L.—As the last species.

398. Phyllis L.

1263. P. Nobla L. This species is native to the Canary Islands. It is
markedly anemophilous (Delpino, Malpighia, Genova, iii, 1889).

399. Crucianella L.

1264. C. stylosa Trin. (=L. Asperula ciliata Poche/). (Francke, Inaug.-
Diss., Halle, 1883; Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, II, pp. 265, 267.)—
Francke states that the pollen is shed while the flower is still a bud, and is

RUBIACEAE 549

subsequently pushed out by the elongating style. The stigma then matures.
Kerner describes the flower mechanism as follows. The long thin serpentiform
style bears a thick stigma, which is held fast by the anthers, and gets covered
with pollen. By elongation of the style, the stigma is raised to the dome-like
top of the still closed flower. When insects alight the petals suddenly open, and
the stigma springs out, dusting the visitors with pollen from below. The style
with its maturing stigma now projects well out of the flower so that the latter
‘is first touched by the hymenopterous or dipterous visitors. In this way crossing
is necessarily effected. Failing insect-visits, the flower explodes, and its scattered
pollen is wafted to the stigmas of neighbouring blossoms.

Visitors.— Vide supra.

1265. C. angustifolia L.— The small flowers of this species are greenish-
yellow in colour.

Visirors. — Plateau observed the beetle Cassida nobilis Z., and a bee
(Andrena sp.).

400. Coffea L.

1266. C. arabica L.—According to Bernoulli (Bot. Ztg., Leipzig, xxviii, 1869,
p- 17), only small purely female fertile flowers are present at the beginning of
anthesis. Ernst states that the hermaphrodite flowers are protandrous.
Visitors.—Bourdillon chiefly observed Lepidoptera (Nature, London, xxxvi,
1887).
4o1. Nertera Banks et Soland.

1267. N. depressa Banks et Soland.—Francke describes this species as
protogynous, autogamy being excluded (Inaug.-Diss., Halle, 1883).

402. Rondeletia L.

1268. R. strigosa Hemsl.—According to Penzig (Malpighia, Genova, viii,
1894, pp. 466-75), there are crowded yellow granules on the cup-shaped part of
the corolla of this species, which closely mimic pollen-grains, and serve to attract
insects.

LIT. ORDER VALERIANAE DC.
Lirerature.—Knuth, ‘Grundriss d. Bliitenbiol.,’ p. 63.

The flowers are aggregated into dense cymes, so that though usually small
they are rendered conspicuous. The nectar is almost always secreted and concealed
in a pouch or spur of the corolla-tube, so that most of the species belong to the
flower class S, but those of the genus Centranthus are obviously members of L.
Cross-pollination is secured by dichogamy, more rarely by dioecism (Valeriana
dioica). Homogamy is exemplified among the smaller flowers.

403. Valeriana L.

Flowers whitish in colour, and arranged in corymbose capitate or paniculate
cymes ; protandrous or protogynous; with concealed nectar, secreted and concealed
above the origin of the corolla-tube in a small pouch with a green fleshy base.

550 ANGIOSPERMAE—DICOTYLEDONES

1269. V. officinalis L.—The small whitish or flesh-red, very fragrant flowers
of this species are rendered conspicuous by aggregation into large inflorescences.
Sprengel (‘Entd. Geh.,’ pp. 63-5) described them as protandrous, and this has been
confirmed by Ricca (Atti Soc. ital. sc. nat., Milano, xiv, 1871) and Herm. Miiller
(‘ Fertilisation,’ pp. 306-7, ‘ Alpenblumen,’ pp. 469-70).

Five purple lines, which become paler in old flowers, serve as nectar-guides.
The corolla-tube is 4-5 mm. long, and about 4 mm. above its base there is
a pouch containing the nectary, while above this the inner surface of the tube is
beset with hairs.

In the first stage of anthesis the pollen-covered anthers project from the
flower; in the second stage the 3 diverging stigmas also protrude.

Insects visiting the younger flowers will therefore cover their feet and the
under-surface of their bodies with pollen, and transfer it to the stigmas of older
flowers. As the stamens are curved outwards in the second stage, automatic
self-pollination is prevented.

According to Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896), the stamens do
not project from the flower simulta-
neously, but in succession, subse-
quently curving outwards, so that
the extrorse anthers easily pollinate
neighbouring flowers in the female
stage. The pollen-grains are white
in colour, densely covered with spinose
tubercles, ellipsoidal, usually truncated
at one pole, up to 75 m» long and

44 p broad.
Fic. 184. Valertana officinalts, L. (from nature, °
enlarged). A, Flower in the first (male) stage: the The same authority states that
pollen-covered anthers (a2) are above the corolla, the by degeneration of the anthers purely

immature stigmas (s) are still curved to the side. 2. ; :
Flower in the second (female) stage: the anthers (a) have female stocks are constituted, which.

shed their pollen, and are curved to the side, while the : 4

mature stigmas (s) project from the corolla. 2, nectary. are recognizable from a distance by
their smaller and more crowded

flowers (Schr. natw. Ver., Wernigerode, xi, 1896). The species is therefore

gynodioecious.

Visttors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list for Central Germany (Herm. Miller, ‘ Fertilisation,’ p. 307, ‘ Weit. Beob.,’ ITI,

p. 98).—

A. Coleoptera. iateridac: 1. Adrastus pallens F. £7., inactive (H. M.).
B. Diptera. (a) Conoprdae: 2. Conops quadrifasciatus Deg., skg.(H. M.); 3. C.
scutellatus Mg., do. (H. M.); 4. Sicus ferrugineus Z., do. (H. M.). (6) Lmpzdae :
5. Empis livida Z., in very large numbers, skg. (H. M.); 6. E. rustica Z., do.
(H.M.). (¢) Muscidae: 7. Anthomyia sp., po-dvg. (Budd.); 8. Calliphora erythro-
cephala J7Zg., freq., skg. (H. M.); 9. C. vomitoria Z., do. (H. M.); 10. Echinomyia
fera Z., skg. (H. M.); 11. Lucilia cornicina &., freq., skg. (H. M.); 12. Musca
domestica Z., do. (H. M.); 13. Onesia floralis R.-D., do., (H.M.); 14. Sarcophaga
carnaria Z., do. (H. M.). (d¢) Syrphidac: 15. Chrysotoxum festivum Z., skg. and
po-dvg. (H. M.); 16. Eristalis arbustorum Z., do. (H. M., Budd.); 17. E. horticola
Deg., do. (H.M.); 18. E. nemorum Z., do. (H. M.); 19. E. sepulcralis Z., do.

VALERIANAE 551

(H. M.);_ 20. E. tenax Z., skg. (H. M.); 21. Helophilus floreus Z., freq., skg. and
po-dvg. (H. M.); 22. H. pendulus Z., do. (H. M.); 23. Syritta pipiens Z., do.
(H. M.); 24. Volucella bombylans Z., do. (H. M.); 25. V. inanis Z., skg. (H. M.);
26. V. pellucens Z., skg. and po-dvg. (H.M.). (e) Zadanddae: 27. Tabanus luridus
(H.M.). C. Hemiptera. 28. Pentatoma sp., skg.(H. M.). D. Hymenoptera.
(2) Apzdae: 29. Apis mellifica Z., ¥, freq. (H. M.); 30. Bombus pratorum LZ. ¥, skg.
(H. M.); 31. Chelostoma nigricorne Wy/. 8, do. (Budd.); 32. Halictus malachurus
K. 9 (Budd.); 33. Small sp. of Halictus 9 and , skg. (H. M.); 34. Sphecodes
gibbus Z., do. (Budd.). (6) Sphegzdae: 35. Crabro vexillatus Pz. 9 (H. M.).
E. Lepidoptera. 36. Epinephele hyperanthus Z., skg. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), a beetle, 16 flies, 6 Hymenoptera, and 5 Lepidoptera
(‘Alpenblumen,’ p. 469). Loew, the Muscid Spilogaster angelicae Scop. (‘ Bliitenbiol.
Floristik,’ p. 398). MacLeod (Flanders), 4 Syrphids, a Muscid, a Lepidopterid, and
a beetle (Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 392); (Pyrenees) a Lepidopterid,
3 Muscids, and 2 Syrphids (op. cit., ili, 1891, p. 346). Réssler (Wiesbaden), the
butterfly Limenitis camilla S. V¥ Schenck (Nassau), the fossorial wasp Gorytes
mystaceus Z. Lindman (Dovrefjeld), several saw-flies and flies, a humble-bee, and
2 Nitidulids. Knuth (Kiel) (‘Bloemenbiol. Bijdragen’)—A. Diptera. Syrphidae:
1. Syrphus balteatus Deg., po-dvg.; 2. Eristalis tenax Z., po-dvg. B. Lepidoptera.
Rhopalocera: 3. Pieris sp., skg.: (Riigen)—Diptera. (a) Muscidae: 1. Aricia sp.,
skg. (0) Stratiomyidae: 2. Odontomyia viridula #., freq., skg. (c) Syrphidae :
3. Eristalis pertinax Scop. 9, skg. Loew (Berlin Botanic Garden), the hover-fly
Eristalis nemorum Z., skg., and the honey-bee, do.: and on the var. altissima Alchx.
—the scarabaeid Cetonia aurata Z., dvg. the flowers, and 2 bees—r. Andrena
albicans Jill. 9, skg. and po-dvg.; 2. Apis mellifica Z. 9, skg.

1270. V. dioica L.—Sprengel (‘Entd. Geh.,’ pp. 65-7), and subsequently
Hermann Miller (‘ Fertilisation,’ p. 307), state that, the male flowers being consider-
ably larger than the female ones, the former are almost always visited first by insects,
so that crossing is effected. The corolla-tube of the male flower expands upwards
like a funnel, and is about 3 mm. long, while that of the female flower is only 1 mm.
in length: the nectar is consequently accessible even to the shortest-tongued insects.
Kerner says that the pseudo-hermaphrodite female flowers open 3-5 days earlier
than the pseudo-hermaphrodite male ones. According to Hermann Miiller the
unisexual flowers are of different sizes, and the vestiges of the organs of the other
sex are variable. Some male flowers have a very large corolla, and no trace of
a pistil; others, a somewhat smaller corolla, and a vestigial pistil. There are also
female flowers with large corolla and small pistil; others with large pistil and very
small corolla. In rare cases hermaphrodite flowers also occur.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel) the honey-bee, skg. Herm. Miiller, ditto, and also the bee
Andrena albicans J7/z//. 9, 2 hover-flies (1. Eristalis arbustorum Z., skg.; 2. Rhingia
rostrata Z., po-dvg.), Tipula, a butterfly (Pieris napi Z., skg.), and the beetle Meli-
gethes in great numbers. MacLeod (Flanders), 2 Muscids (Bot. Jaarb. Dodonaea,
Ghent, v, 1893, p. 392).

1271. V. montana L. (Herm. Miller, ‘ Alpenblumen,’ pp. 470-1; Schulz,
‘Beitrage,’ II, pp. 100, ro1—2, 192.)—This species is gynodioecious in the canton
Graubiinden ; Schulz describes it as trimonoecious to trioecious in the Tyrol. Some

552 ANGIOSPERMAE—DICOTYLEDONES

stocks bear large markedly protandrous hermaphrodite flowers ; others small purely
female ones. The latter possess stamens which present little external evidence of
degeneration, but their anthers do not contain a single well-developed pollen-grain
(Fig. 185).

Visirors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Alps), 2 beetles, 35 Diptera, 3 Hymenoptera, and a Lepi-
dopterid. MacLeod (Pyrenees), the hover-fly Syrphus pyrastri Z. (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891,
p- 347), and Schletterer
(Tyrol), the humble-bee
Bombus terrester Z.

1272. V. saxatilis L.
(Schulz, ‘ Beitrige,’ II,
pp. 102-3, 193.)— Schulz
describes this species as
trimonoecious to trioecious,
and states that the female
flowers are much smaller
than the male and herma-
phrodite ones.

Visitors.—Schulz ob-
served small and medium-
sized flies.

1273. V. supina L.—
Kerner states that this
species is gynodioecious.

1274. V. saliunca
All.—As the last species.

1275. V. tripteris L.
(Herm. Miiller, ‘Alpen-
blumen,’ pp. 471-3.)—
This species is dioecious

Fic. 185. Valertana montana, L. (after Herm. Miller). 4. Female in the canton Graubiinden )

flower with small corolla. 8. Large hermaphrodite flower in the frst Schulz describes it as gyno-
(male) stage. C. Ditto, in the second (female) stage (x 7). ca, calyx; 2 : -
7, nectary ; 2’, supplementary do. ; ov, ovary; sé, stigma. dioecious and androdioe-

cious with protandrous
hermaphrodite flowers in the Tyrol. Kerner states that the pseudo-hermaphrodite
female flowers open (as in V. dioica) 3-5 days before the male ones. Here again
there are small-flowered and large-flowered stocks. The latter are purely male,
and though besides three projecting stamens their flowers possess a style, it
remains within the corolla. (Fig. 186.) Massalingo says that the plants on
Monte Baldo are either micrandrous female, or macrandrous hermaphrodite, as in
V. montana (Boll. Soc. bot. ital., Firenze, 1896).

VALERIANAE 553

Visitors.—Herm. Miiller observed 17 Diptera, a beetle, a bee, and 3 Lepi-
doptera.

1276. V. cordifolia Hoeck (=Phyllactis cordifolia Wedd.)—Ricca describes
this species as markedly protandrous (Atti Soc. ital. sc. nat., Milano, xiv, 1871).

1277. V. capitata Pall—Ekstam says that the flowers of this species are
5-8 mm. in diameter in Nova Zemlia. They smell like heliotrope, and are either
markedly protandrous or homogamous.

VisiTors.— These are flies, including Sarcophaga atriceps Ze//.

1278. V. exaltata Mik.—

Visrrors.—Loew observed the Muscid Cynomyia mortuorum Z. in the Berlin
Botanic Garden.—

Fic. 186. Valertana tripteris, L. (after Herm. Miller). A. Female flower with small corolla
x 7). B. Stigma of ditto (x 80). C. Male flower with large corolla(x 7). D. Ditto in longitudinal
section; ovary and calyx left out (x 7). @, anther; ca, calyx; gv, style; #, nectary; ov, ovary;
sd, nectar-cover ; Sé, stigma.
1279. V. alliariaefolia Vahl.—
Visirors.—Loew observed the following in the Berlin Botanic Garden

A. Coleoptera. Scarabaeidae: 1. Cetonia aurata Z., dvg. the flowers.
B. Diptera. (a) Muscidae: 2. Echinomyia fera LZ. (4) Syrphidae: 3. Eristalis
tenax Z. C. Hymenoptera. Apidae: 4. Bombus terrester Z. ¥, skg.

1280. V. asarifolia Dufr.—

Vistrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Scatophaga merdaria F. (4) Syrphidae:
2. Syritta pipiens Z. B. Hymenoptera. Aprdae: 3. Andrena sp. 9, skg. and

554 ANGIOSPERMAE—DICOTYLEDONES’

po-cltg.; 4. Osmia fulviventris Pz. , skg. C. Lepidoptera. Rhopalocera:
5. Pieris brassicae Z., skg.

1281. V. Phu L.—

Visrtors.—Loew observed the following in the Berlin Botanic Garden.—

A, Coleoptera. Scarabaeidae: 1. Rhizotrogus solstitialis Z., dvg. the flowers.
B. Diptera. Syrphidae: 2. Helophilus floreus Z. C. Hymenoptera. Afzdae:
3. Osmia rufa Z. 9, po-cltg. D. Lepidoptera. Rhopalocera: 4. Pieris brassicae
L., skg.

404. Centranthus DC.

Red or white Lepidopterid flowers; markedly protandrous (Delpino, ‘Ult. oss.,’
p. 127); with nectar secreted in a spur at the base of the corolla-tube.

1282. C. ruber DC.—Schulz (‘Beitrage,’ II, pp. 103-4) describes the flowers
of this species as zygomorphous, like those of the other Valerianeae. One corolla-
lobe forms the upper lip, while the four others make up the lower one. The narrow
corolla-tube, which Kerner describes as longitudinally divided into two by a thin
membrane, is 8-10 mm. long, with a spur of 6-7 mm. The anther of the single
stamen, which is situated right or left of the upper lip, is introrse, but during
dehiscence it assumes an oblique or even horizontal position. At first the style
projects but little from the corolla-tube, but after the anthers have dehisced it
elongates, and protrudes 5~6 mm. beyond the corolla. Automatic self-pollination is
thus prevented.

Visttors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (garden plants in Helgoland), the diurnal hawk-moth Macroglossa
stellatarum Z. Loew (Bellagio), ditto. Schulz (Bozen), chiefly butterflies—1.
Papilio podalirius Z.; 2. P. machaon Z.; 3. Parnassius apollo Z.; 4. Pieris
brassicae Z.; 5. P. rapae Z. Mattei (Genoa), butterflies and Zygaenidae. Loew
(Berlin Botanic Garden)—A. Coleoptera. Scarabaeidae: 1. Cetonia aurata L.,
dvg. the flowers. B. Diptera. SyrpAzdae: 2. Syritta pipiens Z., po-dvg. (?);
3. Syrphus luniger JZg., casual. C. Lepidoptera. Rhopalocera: 4. Vanessa
urticae L., skg.

1283. C. angustifolius DC.—

VistTors.—The following were observed by Loew in the Berlin Botanic Garden.

A. Coleoptera. Scarabaeidae: 1. Cetonia aurata Z., dvg. the flowers. B.
Diptera. Syrphidae: 2. Eristalis sepulcralis Z., casual; 3. E. tenax Z., do.
C. Lepidoptera. Rhopalocera: 4. Pieris brassicae Z., skg.; 5. Vanessa urticae
L., do.

405. Valerianella Tourn.

Flowers bluish-white in colour; homogamous or protogynous; in dense cymes,
but inconspicuous owing to their small size; nectar secreted and concealed in an
expansion at the base of the corolla-tube.

1284. V. olitoria Pollich, (Herm. Miller, ‘Fertilisation, p. 308, ‘ Weit.
Beob.,’ III, p. 98; Kirchner, ‘Flora v. Stuttgart, p. 6753; MacLeod, Bot. Jaarb.
Dodonaea, Ghent, v, 1893, pp. 392-3 3 Knuth, ‘ Bloemenbiol. Bijdragen’ ; Warnstorf,
Verh. bot. Ver., Berlin, xxxviii, 1896.)—Hermann Miiller states that the corolla of
the minute flowers of this species consists of a tube, the lower part of which is about

VALERIANAE 555

4 mm. long and barely 1 mm. wide. It then suddenly expands to a diameter of
about 3 mm., and ends in a limb which is usually 5- or 6-lobed, and about 2 mm.
across. In the base of the expansion tiny drops of nectar are secreted. The
flowers are homogamous. Soon after they open the three stamens project straight
out of the corolla, their anthers being covered with pollen all over ; the simultaneously
matured stigma is at a lower level, and some pollen-grains that have fallen from the
anthers adhere to it. The style gradually elongates, bringing the stigma to the level
of the anthers. Automatic self-pollination is therefore inevitable, but crossing may
also take place as the result of insect-visits.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list.—

A. Coleoptera. (a) Chrysomelidae: 1. Lema cyanella ZL. (H.M.). (6) Ela-
tertdae; 2. Limonius cylindricus Payk. (H. M.). (c) Witidulidae: 3. Meligethes,
very numerous, po-dvg. (H. M.). (d) Staphylinidae: 4. Philanthus sp. (H. M.).
B. Diptera. (a) Empidae: 5. Cyrtoma spuria Fall. (H.M.); 6. Empis pennipes
L., freq., skg. (H. M.); 7. E. trigramma J/g., in large numbers, skg. (H. M.) ;
8. Hilara sp., do. (H. M.). (4) Lonchopteridae: 9. Lonchoptera punctum JZ.
(H. M.). (c) Muscidae: 10. Aricia incana Wredem., freq., skg. (H. M.); 11. Lucilta
sp., freq. (H. M.); 12. Onesia sepulcralis 4g. (H.M.); 13. Pollenia vespillo /.,
nect-lkg. (H. M.); 14. Psila fimetaria Z., skg. (H. M.); 15. Scatophaga stercoraria
L., 9 and 4, in large numbers, skg. (H. M.); 16. Sepsis sp. (H. M.); 17. Siphona
geniculata Deg., skg. (H. M.). (d) Syrphidae: 18. Ascia podagrica /., very com-
mon, skg. and po-dvg. (H. M.); 19. Syritta pipiens Z., do. (H.M.). (e) Bebeontdae :
20. Dilophus sp. (H. M.). (/) Mycetophilidae: 21. Sciara sp. (H. M.). C.
Hemiptera. 22. Eurydema oleraceum Z., skg. (H. M.). D. Hymenoptera.
Apidae: all skg.: 23. Andrena albicans J7Zw//. 9 (Budd.); 24. A. collinsonana K. 9
(Budd.); 25. A. convexiuscula &. 9 (Budd.); 26. A. gwynana X. 9 (Budd.); 27. A.
nitida Mourcr, 9 (Budd.); 28. A. parvula &. 9 (Budd.); 29. A. Smithella K. ¢
(Budd.); 30. Halictus politus Schenck, 9 (Budd.); 31. Nomada ruficornis Z. (=N.
signata Jur.) $ (Budd.); 32. N. sp. (Budd.); 33. Sphecodes gibbus Z. 6 (Budd.).
E. Lepidoptera. (a) Nocturdae: 34. Euclidia mi Z., casual, skg. (H. M.). (4)
Rhopalocera: 35. Polyommatus dorilis W/n. (H. M.).

The following were recorded by the observers, and for the localities stated. —

Knuth, the beetle Meligethes, and some flies (skg.)—1. Lucilia caesar Z.; 2.
Syritta pipiens Z.; 3. Syrphus ribesii Z. Alfken (Bremen), the humble-bee
Bombus muscorum /. 9, skg. Réossler (Wiesbaden), the Tineid moth Adela rufi-
frontella 77.

1285. V. auricula DC. (=V. rimosa Bas/. and V. dentata Lozse/)—Kirchner
states that the flower mechanism of this species agrees essentially with that of
V. olitoria. Kerner describes the flowers as protogynous. Automatic self-pollination
is at first excluded by downward curving of the style at the time when the anthers
dehisce ; but this subsequently bends upwards, and autogamy is then possible.

Visirors.—Herm. Miiller only saw the small bee Halictus longulus Sm., skg.

1286. V. carinata Loisel—Kerner states that the flower mechanism of this
species is similar to that of V. auricula.

1287. V. dentata Pollich (=V. Morisonii DC.).—Hermann Miiller states the
mechanism of this species is similar to that of V. olitoria. Kerner describes
the flowers as protogynous.

556 ANGIOSPERMAE—DICOTYLEDONES

LIV. ORDER DIPSACEAE DC.

LiteraturE.—Knuth, ‘Grundriss d. Bliitenbiol.,’ pp. 63-4.

The small flowers are united into large capitula, so that they are very con-
spicuous and receive numerous insect-visits. Cross-pollination is secured in our
native species by marked protandry of the hermaphrodite flowers. Nectar is secreted
by the surface of the ovary, and is concealed in the base of the corolla-tube. All the
species consequently belong to flower class S$. Gynodioecism is common.

406. Morina Tourn.

Slightly protogynous flowers, which open in the twilight, and are therefore
adapted to the visits of crepuscular and nocturnal Lepidoptera.

1288. M. persica L. (Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, I, p. 351.)}—Kerner
states that in this species protogyny only lasts for half-an-hour, but this is sufficient to
permit of crossing taking place at first. As soon as the limb of the corolla expands,
the thick swollen stigma becomes visible just above the passage leading to the nectar,
its receptive surface facing downwards. The two anthers behind it are still immature,
so that the proboscis of an insect, if already dusted with pollen, must effect crossing
while probing for nectar.

1289. M. elegans Fisch. et Avé-Lall. (=M. longifolia Wal/.). (Hildebrand,
Bot. Ztg., Leipzig, xxvii, 1869, pp. 488—-91.)—Hildebrand describes the flowers of
this species as homogamous, but the stigma projects beyond the anthers, so that
insects usually touch it first, cross-pollination being thus favoured. The stigma
subsequently bends downwards and comes into contact with the anthers, and
automatic self-pollination may therefore take place at this stage.

407. Dipsacus L.

Flowers whitish or lilac in colour; markedly protandrous; aggregated into
ovoid or globular heads. Spinescent bracts prevent insect visitors from creeping
over the inflorescences, and these touch the anthers and stigmas with their heads,
not their feet.

1290. D. sylvestris Mill. (Herm. Miller, ‘Fertilisation,’ pp. 308-9, ‘ Weit.
Beob.,’ III, p. 76; Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892, p. 81; Knuth,
‘Bloemenbiol. Bijdragen’ ; Loew, ‘ Bliitenbiol. Floristik,’ p. 390.)—Hermann Miiller
describes the corolla-tube of the lilac flowers of this species as 10-11 mm. long. In
the first stage of anthesis the dehisced anthers project; in the second stage usually
‘only one branch of the style protrudes, the other being vestigial as a rule. Hermann
Miiller explains this last peculiarity as an adaptation to the visits of humble-bees.
When one of these thrusts its head into a flower, one branch of the style would get
in the way of the other were both equally developed. A more complete pollination
of the entire stigmatic surface of one branch by the head of the humble-bee is
possible when the other branch is absent. Kirchner says that anthesis begins at
an intermediate zone of the capitulum, and extends both inwards and outwards
from this.

DIPSACEAE | 557

Kirchner mentions an interesting arrangement for securing safety from wingless
insects that might creep up from the ground, i.e. the collection of rain water in
the hollows formed by the union of the bases of the cauline leaves. The same
arrangement is found also in D. laciniatus and D. fullonum.

Visitors.—Herm. Miiller observed the following in Westphalia, all skg.—

A. Diptera. Syrphidae: 1. Volucella pellucens Z. B. Hymenoptera.
Apidae: 2. Bombus agrorum /. 9 and ¥; 3. B. lapidarius Z. 9, ¥ and 4, freq.;
4. Crocisa scutellaris 7. g 5. Halictus tetragonius A/g.é; 6. H. sexcinctus /. 6;
7. Megachile lagopoda Z. 9 and 5; 8. M. maritima A. 9 and 6; 9. Psithyrus
rupestris /. 9.

Heinsius gives the following for Holland.—

A. Diptera. Syrphidae: 1. Eristalis pertinax Scop. 9. B. Hymenoptera.
Apidae: 2. Bombus agrorum F. 8; 3. B. rajellus K. &; 4. Megachile maritima
K.%; 5. Psithyrus campestris Pz. 3; 6. P. vestalis Fourcr.d. C. Lepidoptera.
Rhopalocera: 4. Pieris rapae Z.$; 8. Rhodocera rhamni Z. 8 and 9g; 9. Vanessa
io Z.; ro. V. urticae Z.

The following were recorded by the observers, and for the localities stated.—

Loew (Brandenburg), the humble-bee Bombus cognatus S/ep. 4, skg. (‘ Beitrage,”
p-. 40). Schletterer (Tyrol), 2 humble-bees—Bombus terrester Z. and B. arenicola
Thoms. von Dalla Torre (Tyrol), the humble-bee Bombus muscorum 7. é. MacLeod
(Flanders), 3 humble-bees, a hover-fly, and a Lepidopterid (Bot. Jaarb. Dodonaea,
Ghent, vi, 1894, p. 373). Knuth (Kiel Botanic Garden), 2 skg. and po-cltg. humble-
bees—Bombus lapidarius Z. ¥, and B. terrester Z. 9 and ¥.

1291. D. fullonum L. (Kirchner, ‘Flora v. Stuttgart,’ pp. 678-9.)—Kirchner
describes the whitish protandrous flowers of this species as possessing a corolla
12-14 mm. long, scarcely 1 mm. thick in its lower part, but gradually expanding
-upwards like a funnel. Anthesis begins (as in D. sylvestris) in an intermediate zone,
from which it extends upwards and downwards. After the flower has opened, the
stamens with their lilac-coloured anthers project 5-6 mm. beyond the corolla. When
the anthers have withered, the style (at first enclosed in the corolla) elongates, so that
the stigmas are situated 2-4 mm. beyond the stamens. One of the two stigmatic
branches is often vestigial.

Visttors.—The following were recorded by the observers, and for the localities

stated.—

Kirchner (Wurtemberg), humble-bees and small Nitidulids. Knuth (Kiel
Botanic Garden), 2 humble-bees—Bombus lapidarius Z. and B. terrester Z., skg.
(‘Bloemenbiol. Bijdragen’). FF. F. Kohl (Tyrol), the true wasp Ancistrocerus
parietum Z.

1292. D. laciniatus L. (Kirchner, ‘Beitrige,’ p. 63.)—The flowers of this
species are pale lilac, almost white in colour. Kirchner describes them as markedly
protandrous, with a corolla about 10 mm. long, from which the diverging stamens
project 5 mm. After the anthers have dropped off the style grows out of the
corolla-tube, and protrudes as much as 4-5 mm. Here again one of the two
stigmatic branches is usually vestigial; if both are fully formed they are recurved.
Anthesis takes place as in the last two species.

558 ANGIOSPERMAE—DICOTYLEDONES

408. Cephalaria Schrad.

1293. C. alpina Schrad.—

Visirors.—Loew (Berlin Botanic Garden) observed 2 humble-bees—Bombus
hortorum Z. 9, skg., and B. terrester Z. ¥, skg. and po-cltg.

1294. C. radiata Griseb. et Schenk.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis nemorum Z.; 2. E. tenax Z.; 3. Syrphus
albostriatus Fal/.; 4. S. balteatus Deg.; 5. S.ribesiiZ. B. Hymenoptera. Ajzdae:

6. Apis mellifica Z. ¥, skg. and po-cltg.; 7. Bombus hypnorum Z. ¥, skg.; 8. Halictus
cylindricus Z. é, do.

1295. C. uralensis Roem et Schult. (=C. corniculata Roem et Schult.).—
VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syrphus ribesii Z.; 2. Volucella pellucens Z., skg.
B. Hymenoptera. Apidae: 3. Bombus agrorum /. 4, skg.; 4. B. terrester L. 9,
do.; 5. Psithyrus campestris Pz., var. rossiellus K. 6, do. Also on the var. crefacea
3 bees—r. Bombus terrester Z. 6, skg.; 2. Prosopis communis (yi. 9; 3. Psithyrus
rupestris #. 4, skg.

409. Knautia L.

Flowers lilac or white in colour; arranged in conspicuous flat hemispherical
inflorescences, belonging to class S. Gynodioecious, with protandrous hermaphrodite
flowers.

1296. K. arvensis Coult. (=Scabiosa arvensis Z., and Trichera arvensis
Schrad.). (Sprengel, ‘Entd. Geh.,’
p- 84; Herm. Miiller, ‘ Fertilisa-
tion,’ pp. 309-13, ‘Alpenblumen,’
p- 399,‘ Weit. Beob., III, pp. 76-7;
Schulz, ‘Beitrage, II, pp. 173,
192; Knuth, ‘Bl. u. Insekt. a.
d. nordfr. Ins.,’ pp. 83-4, 156-7,
‘Weit. Beob. ii. Bl. u. Insekt. a. d.
nordfr. Ins.,’ p. 235, ‘ Bliitenbiol.
Beob. in Thiiringen,’ ‘ Bliitenbiol.
Beob. a. d. Ins. Riigen,’ ‘ Bloem-
enbiol. Bijdragen,’ and other
memoirs; Loew, ‘ Bltitenbiol.

ot ate ty
Fic. 187. Avazutia arvensis, Coult. (after Herm. Miller). Floristik, PP. 399, 394, 398.)—

(1) Hermaphrodite flower iu the first (male) stage, after removal In thi j ,

of the corolla-lobes (x 3$). (2) Ditto, in the second (female) B SPSCics about fifty flowers

stage. (3) Female flower, after removal of the limb of the are aggregated into a head, the
corolla. a, stigma; 4, stamens still enclosed in the flower; : f hi h ae
¢, ripe stamens; d, exhausted stamens; e, vestigial stamens. conspicuousness of which 1s in-

creased, as Hermann Miller
points out, by increase in size of the corollas towards the periphery. The corolla-
tubes are 4-9 mm. long, and expand above into funnels, to an extent proportional
to the size of the florets. Nectar is secreted on the upper surface of the ovary, and
protected from rain by the hairs lining the corolla-tube: it is readily accessible even

DIPSACEAE 559

to short-tongued insects. The pollen is also easily secured. In the first (male)
stage the stamens of the hermaphrodite flowers project 4-5 mm. from the corolla,
and the anthers turn their pollen-covered sides upwards. The stamens mature
successively, so that this first stage lasts for several days. After all the stamens have
completed their cycle of development, the anthers drop off and the filaments shrivel.
The style, so far concealed in the mouth of the corolla, now grows to such an extent
that the maturing stigma projects from the flower as far as the stamens did at first.
Although the individual flowers develop centripetally, the elongation of the styles and
the maturation of the stigmas only begin when all the stamens of the head have
withered, so that the entire inflorescence is at first purely male, and then purely
female. Insect visitors, therefore, when creeping over a capitulum either get dusted
with abundant pollen or else pollinate numerous stigmas during a single visit.
Automatic self-pollination is not entirely excluded, for as the styles elongate some
stigmas come into contact with the anthers of their florets.

Warnstorf describes the pollen-grains (examined in water) as almost spherical,
white, quite opaque, with 3 large germinating processes, up to 137 y in diameter.

Besides stocks with hermaphrodite flowers, there are often some with female
ones, especially at the beginning of the flowering season. The stamens of these are
more or less reduced. The diameter of the inflorescence is usually as great as in
hermaphrodite stocks, but in some female plants it is scarcely 2 cm.

MacLeod (Flanders) and Charles Darwin (Kent) noticed gynodioecism. Willis
says that in Cambridgeshire female stocks are commoner than hermaphrodite ones.

Besides plants with female flowers, Lindman observed on the Dovrefjeld
a variety with shorter styles, vestigial stamens, and enlarged actinomorphous
corollas (var. zsantha L. M. Neumann). This modification is apparently due to
a fungus.

Visitors.—The most important one is a bee, Andrena hattorfiana #., which
devotes itself almost exclusively to Knautia. It is everywhere associated with this
species. I observed it in Thuringia, Sylt, Féhr, Schleswig-Holstein, Mecklenburg,
and also in Riigen, where I captured 6 individuals one warm forenoon.

I also observed the following in Schleswig-Holstein (S.-H.) and Riigen (R.).—

A. Coleoptera. All po-dvg.: (a) Cerambyczdae: 1. Strangalia melanura LZ.
(R.). (8) Curculionidae: 2. Miarus campanulae Z. (S.-H.). (¢c) itédulidae:
3. Meligethes aeneus /, (S.-H.). B. Diptera. (a) Conopidae: 4. Sicus ferrugineus
L., very common, po-dvg. (R.). (6) Lmprdae: 5. Empis livida Z., po-dvg. (S.-H.);
6. E. opaca ¥. do. (S.-H.); 7. E. tessellata F. do. (S.-H.). (c) Muscidae: 8. Aricia
incana Wied. po-dvg. (S.-H.); 9. Dexia canina /. do. (R.); 10. Small Muscids, do.
(S.-H.). (2) Syrphidae : all skg. and po-dvg.: 11. Eristalis anthophorinus Ze/s. 6 and 9
(R.); 12. E. arbustorum JZ. (S.-H.); 13. E. horticola Deg. (S.-H., R.); 14. E.
intricarius Z. (S.-H., R.); 15. E. nemorum JZ. (S.-H.); 16. E. pertinax Scop.
(S.-H.); 17. E. rupium /. (S.-H.); 18. E. sepulcralis Z. (R.); 19. E. tenax Z.
(S.-H., R.); 20. Helophilus pendulus Z. (S.-H.); 21. H. trivittatus # (S.-H.);
22. Sericomyia borealis Fad/. § (R.); 23. Syritta pipiens Z. (S.-H.); 24. Syrphus
pyrastri Z.9(R.); 25. S. ribesii Z. (S.-H.); 26. Volucella bombylans Z. 9 and 4,
and the var. plumata JZ. (R., S.-H.). (e) Zabanidae: 24. Haematopota pluvialis Z.
(R.). C. Hemiptera. 28. Calocoris roseomaculatus Deg. (S.-H.). D. Hymeno-
ptera. Apidae: 29. Andrena gwynana X. (S.-H.); 30. Apis mellifica Z. ¥, very
common (S.-H., R.); 31. Bombus agrorum /. ¥ (S.-H., R.); 32. B. distinguendus
Mor. (S.-H.); 33. B. hortorum JZ. (S.-H.); 34. B. lapidarius Z. § and 9 (S.-H., R.);

560 ANGIOSPERMAE—DICOTYLEDONES

35- B. pratorum Z. (S.-H.); 36. B. rajellus K. (S.-H.); 37. B. terrester Z. (S.-H.);
38. Dasypoda plumipes Pz. (S.-H.); 39. D. thomsoni ScAdet/t. 9 (R.); 40. Halictus
fulvicornis K. (S.-H.); 41. Megachile centuncularis Z. 9 (R.); 42. Nomada armata
H-Sch., one (R.); 43. Psithyrus vestalis Fourcr. (S.-H.). All skg., 29-42 also
po-cltg. E, Lepidoptera. All skg.: (2) Moctutdae: 44. Plusia gamma Z. (S.-H.).
(3) Rhopalocera: 45. Argynnis aglaja Z. (S.-H.); 46. A. ino Z. (S.-H.); 47. A.
paphia Z., very numerous (R.); 48. Epinephele janira Z. (S.-H., R.); 49. Hesperis
lineola Ochs. (S.-H.); 50. Lycaena semiargus o//. (S.-H.); 51. Pieris sp. (R.);
52. Polyommatus phloeas Z. (S.-H.); 53. Satyrus semele Z. (S.-H.); 54. Vanessa
atlanta Z. (R.); 55. V. urticae Z. (R., S.-H.). (c) Sphingidae: 56. Ino statices
Esp. (R.); 57. Zygaena filipendulae Z. (S.-H.): 58. Z. 2 sp. (R.).

I observed the following in Thuringia (‘ Bliitenbiol. Beob. in Thiiringen,’ p. 38).—

A. Coleoptera. 1. Judolia cerambyciformis Schr.; 2. Meligethes sp.; 3.
Strangalia melanura Z.; 4. Trichius fasciatus Z., freq., po-dvg. B. Diptera.
(2) Empidae: 5. Empis tessellata Z., skg. and po-dvg. (4) Muscidae: 6. Aricia
basalis Je’, freq.; 7. Homalomyia scalaris #, skg. (c) Syrphidae: all skg. and
po-dvg.; 8. Eristalis pertinax Scop. d; 9. Syrphus annulipes Ze/#. 9; 10. Volucella
bombylans Z., var. plumata J/g., freq,; 11. V. pellucens Z. C. Hymenoptera.
Apidae: all skg. except 20: 12. Bombus agrorum /. 9; 13. B. hypnorum JZ. $;
14. B, lapidarius Z. ¥; 15. B. soroénsis /., var. proteus Gers¢. ¥; 16. B. terrester
L.¥; 17. Psithyrus barbutellus A. 3; 18. P. quadricolor Leg.=P. luctuosus Hoffer. ;
19. P. vestalis Fourcr. $, freq.; 20. Halictus cylindricus / 9, tarsi thickly covered
with violet pollen. D. Lepidoptera. All skg.: 21. Argynnis paphia Z.; 22.
Epinephele janira Z.; 23. Ino statices Z., freq.; 24. Zygaena pilosellae Lf. ;
25. Z. trifolii Lsp.

Alfken observed the following at Bremen.—

A. Diptera. (a) Bomébyliidae: 1. Anthrax paniscus Ross?, very common,
skg.; 2. Exoprosopis capucina /., not infrequent, skg. (6) Conopidae: 3. Conops
quadrifasciatus Deg., not infrequent; 4. Physocephala rufipes /., rare, skg.;
5. Sicus ferrugineus Z., very common, skg. (c) Muscidae: 6. Echinomyia tessellata
¥F., very common. (d) Syrphidae: 7. Eristalis anthrophorinus Ze/t., rare; 8. E. arbus-
torum Z.; 9. E. intricarius Z., 9 and 4, very common, skg.; 10. E. pertinax Scop.;
11. E. sepulcralis Z.; 12. E. tenax Z., 9 and 4, very common, skg. ; 13. Helophilus
pendulus Z.; 14. H. trivittatus #.; 15. Platycheirus peltatus M/.; 16. Volucella
bombylans Z.; 17. V. pellucens Z. (¢) Zadantdae: 18. Tabanus rusticus Z.
B. Hymenoptera. (a) Apidae: 19. Andrena gwynana X. 9, 2nd generation,
Tare, po-cltg.; 20. A. hattorfiana /., freq., 9, skg. and po-cltg., $ skg.; 21. A. margi-
nata #., rare, 9 skg. and po-dvg., § skg.; 22. Bombus agrorum /, ¥ and 8;
23. B. arenicola Thoms. 9, ¥ and $; 24. B. derhamellus X. 9, ¥ and 5; 25. B.
hortorum Z. 9, skg.; 26. B. lapidarius Z. 4, skg.; 27. B. lucorum Z.¥ and $;
28. B. muscorum /. 9 and é; 29. B. pratorum Z. 4, skg.; 30. B. proteus Gerst.
g and ¥, freq., skg. and po-dvg., 6 freq., skg.; 31. B. ruderatus /. 4, skg.;
32. B. sylvarum Z. 4, skg.; 33. B. terrester Z. 6, very common, skg.; 34. Coeli-
oxys acuminata WVyl.$; 35. C. conoidea ///. 9 and 4, rare, skg.; 36. C. rufescens
Lep. 9, skg.; 37. Epeolus variegatus Z. 9, rare; 38. Eriades nigricornis Vy. 9;
39. Halictus calceatus Scop. 9; 40. H. leucozonius Schr. 9; 41. H. zonulus Sm. 9;
42. Megachile centuncularis Z.5; 43. M. circumcincta K. 3; 44. Melitta leporina
Pz.é; 45. Nomada armata H.-Sch., very rare, skg.; 46. N. jacobaeae Pz. 9 and 4,
rare, skg.; 47. Psithyrus barbutellus A. 9 and 4, not infrequent, skg.; 48. P. cam-
pestris Pz. 9 and 3, skg.; 49. P. rupestris /. 9 and , do.; 50. P. vestalis Fourcr.
g and 6, do.; 51. Stelis phaeoptera K. (4) Sphegidae: 52. Ammophila campestris
Lir., skg.; 53. A. sabulosa Z. 9 and 4, freq., skg.; 54. Crabro scutellatus Schev. ;
55. C. subterraneus / 9 and 6, not infrequent. (c) Vesprdae: 56. Odynerus oviventris
Wesm. 8; 57. O. parietum Z. 6.

DIPSACEAE 561

Loew gives the following. —

I. For Brandenburg (‘Beitrige,’ p. 40).—Hymenoptera. (a) Apidae: 1.
Andrena hattorfiana F. 9, skg. (6) Sphegidae: 2. Bembex rostrata LZ. 9, skg.;
3. Tachytes obsoletus Rossi 9, skg. II. For Silesia (op. cit., p. 32).—A. Coleoptera.
(a) Cerambycidae: 1. Leptura maculicornis Deg.; 2. Strangalia bifasciata J/dU/.
(6) Nitidulidae: 3. Meligethes sp. (c) Ocdemeridae: 4. Oedemera flavipes Z. 6.
B. Diptera. (a) Empidae: 5. Empis sp. skg. (5) Syrphidae: all skg.: 6. Meli-
threptus scriptus Z.; 7. Syrphus ribesii Z.; 8. Volucella bombylans Z.; 9. V.
pellucens Z. C. Hymenoptera. Apidae: all skg.: 10. Apis mellifica Z. 9;
11. Bombus agrorum #. 9; 12. Macropis labiata Pz. d; 13. Megachile argentata /.
$; 14. Nomada jacobaeae Pz. 9; 15. Psithyrus campestris Pz.9. D. Lepidoptera.
Rhopalocera: all skg.: 16. Argynnis paphia Z.; 17. Epinephele janira Z.; 18.
Rhodocera rhamni Z.; 19. Vanessa urticae Z. III. Further for Silesia (op. cit.,
p. 26).—A. Diptera. Conopidae: all skg.: 1. Myopa fasciata 1Zg.; 2. Physocephala
vittata #. 3; 3. Zodion cinereum #. B. Hymenoptera. (a) Apidae: all skg.:
4. Andrena hattorfiana /.9; 5. Anthophora furcata Pz. 5; 6. Apis mellifica LZ. ¥;
7. Coelioxys octodentata (Z. Duf.) Zep. 6; 8. Crocisa histrio #. $; 9. Dasypoda
hirtipes #. 6 and 9, the latter also po-cltg.; 1o. Halictus leucozonius Schr. 9;
11. Nomada jacobaeae Pz. 5; 12. Psithyrus campestris Pz. 9. (6) Sphegidae:
13. Bembex rostrata /., skg. C. Lepidoptera. hopalocera: 14. Pieris brassicae
L., skg. IV. For the Riesengebirge (R.) and Silesia (S.) (op. cit. p. 50).—
A. Diptera. (a) Asztidae: 1. Dioctria flavipes Mg. (S.). (6) Syrphidae:
2. Syrphus nitidicollis A/g. (S.). B. Hymenoptera. Afzdae: 3. Andrena
convexiuscula K. $, skg. (S.). CC. Lepidoptera, Sphingidae: 4. Zygaena
achilleae Zsp. (R.); 5. Z. minor S.V. (R.). V. For Switzerland (op. cit., p. 59).—

2 Syrphids (Eristalis jugorum Zgg., and Volucella pellucens Z.), and a moth
(Zygaena lonicerae £sf.).

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list for North
and Central Germany.—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida F. (H. M.); 2. Pachyta
octomaculata #. (H. M.); 3. Strangalia armata Zést. (H. M., Siebengebirge) ;
4. S. atra Larch, (H. M., Siebengebirge); 5. S. attenuata Z. (H. M.); 6. S. melanura
L.(H.M.); 7. Toxotus meridianus Z. (H. M., Siebengebirge). These beetles dvg.
pollen and anthers; also, especially S. attenuata, skg. (4) CArysomelidae: all dvg.
the flowers: 8. Cryptocephalus sericeus Z. (H. M.). (c) Scarabaeidae: 9. Hoplia
philanthus Sw/z., rare (H. M.); 10. Trichius fasciatus Z., very common. (d) Zéle-
phoridae: 11. Malachius bipustulatus Z., dvg. the anthers (H. M.). (e) MMetedudidae:
12. Meligethes, freq., po-dvg. (H. M.). (/) Phalacridae: 13. Olibrus bicolor £.,
po-dvg. (H.M.). B. Diptera. (a) Conopidae: 14. Sicus ferrugineus Z., numerous,
skg. (H. M.). (6) Zmpidae: 15. Empis livida Z., extremely common, skg. (H. M.);
16. E. tessellata #., do. (H. M.). (¢) Musctdae: 17. Echinomyia tessellata /., skg.
(H. M.); 18. Micropalpus fulgens A/g., skg. (H. M.); 19. Ocyptera cylindrica /., do.
(H. M.); 20. Prosena siberita /., freq., skg. (H. M., Liebenau). (d) Syrphedae: all
freq. and persistent, skg. and po-dvg.: 21. Eristalis arbustorum Z. (H.M.); 22. E.
intricarius Z. (H. M.); 23. E. memorum Z. (H.M.); 24. E. tenax Z. (H. M.);
25. Pipiza festiva Mg. (H. M., Lippstadt); 26. Rhingia rostrata Z. (H. M.);
24. Syrphus ribesii Z. (H. M.); 28. Volucella bombylans Z.(H. M.); 29. V.
pellucens Z., nect-lkg. (H. M.); 30. V. plumata Z. (H. M.). C. Hymenoptera.
Numbers in parentheses indicate length of proboscis in mm.: (a) Apidae: 31.
Andrena gwynana X., po-cltg.; 32. A. hattorfiana #. 6 and 9 (6-7), skg. and
po-cltg. (H. M.); 33. Apis mellifica Z., ¥ (6), freq., skg., more rarely po-cltg. ;
34. Bombus agrorum F”. 9, only skg. (H. M.); 35. B. hortorum Z. 4, ¥ and g, skg.
(H. M.); 36. B. hypnorum Z. 6, do. (H. M.); 37. B. lapidarius Z. ¥, do. (H. M.);
38. B. pratorum Z. 9, ¥ and é, do. (H. M.); 39. B. rajellus A. $, do. (H. M.);

DAVIS. IT 0 oO

562 ANGIOSPERMAE—DICOTYLEDONES

40. B. sylvarum Z. 9 and ¥, do. (H. M.); 41. B. terrester Z. 9 and 6, do. (H. M.);
42. B. tristis Sed). Y, do. (Budd.); 43. Ceratina callosa F. 8, do. (Budd.);
44. C. cyanea A. 9 and 4, do. (H. M., Lippstadt); 45. Coelioxys conoidea ///.
g, do. (H. M.); 46. C. quadridentata Z. 3 and 9, freq., skg. (H. M.); 47. Diphysis
serratulae Pz. $ and 9, very common, skg. (H. M.); 48. Halictus albipes /. g
(H. M.); 49. H. cylindricus #. 9 and $ (H. M.); 50. H. leucozonius Schr.
(H. M.); 51. H. lugubris K. 9, skg. (Budd.); 52. H. malachurus &. 9, do.
(Budd.), and po-cltg. in the Oberpfalz; 53. H. quadricinctus / 9 (Budd.);
54. H. quadristrigatus Ziv. 9 (Budd.); 55. H. sexcinctus /. 9 (Budd.); 56. H.
sexnotatus X. 9 (H. M.); 57. H. xanthopus K. 9 (Budd.), the species of Halictus
both skg. and po-cltg.; 58. Heriades truncorum Z. 4, skg. (H.M.); 59. Megachile
centuncularis Z. 5, do. (H. M.); 60. M. circumcincta AK. 9 and 4, do. (H. M.);
61. M. maritima X, and g, in large numbers, skg. (H. M.); 62. M. willughbiella
K. 3%, skg. (H.. M.); 63. Nomada armata H.-Sch. 9, do. (H. M.); 64. N. fabriciana
L. 9, do. (H. M.); 65. N. jacobaeae Pz. 9, do. (H. M., Budd.); 66. N. lineola Pz.
g and &, do. (H.M.); 67. Osmia aenea Z. $, do. (H. M., Budd.); 68. O. fulviventris
Pz.9, po-cltg. (H. M.); 69. Prosopis signata Pz. 9 and $, Budd.); 70. Psithyrus
barbutellus K. $ and 9, skg. (H. M.); 71. P. campestris Pz. 9 and 6, do. (H. M.):
72. P. rupestris Z. 9, do. (H.M.); 73. P. vestalis ourcr. 9, do.(H.M.); 74. Stelis
aterrima Pz. 6, do.(Budd.); 75. S. breviuscula Wy/. 3, do.(H.M.). (4) Ichneumonidae:
76. One small sp., creeping far into the flowers (Budd.). y, Sphegidae: 77. Bembex
rostrata Z. skg. (H. M.); 78. Mimesa bicolor Jur. 6 (Budd.); 79. Philanthus
triangulum /. 3, skg. (Budd.); 80. Psammophila affinis A. 9, do. (H. M.);
81. P. viatica Deg. 6, do. (H. M.). (d) Vespidae: 82. Odynerus parietum ZL. 9,
skg. (H. M.). D. Lepidoptera. (2) Microlepidoptera: 83. Nemotois scabiosellus
Scop. 9 (Budd.). (4) Moctutdae: 84. Euclidia glyphica Z. (H. M.); 85. Mamestra
serena $.V. 9 (Thuringia). (c) Rhopalocera: 86. Argynnis latonia Z., skg. (H. M.,
Oberpfalz); 87. A. niobe Z., do. (H. M., Oberpfalz) ; 88. Colias hyale Z., freq. (H. M..
Thuringia); 89. Hesperia comma Z., skg. (H. M., Fichtelgebirge; Budd., Lausitz) ;
go. H. lineola Ochs. (H. M.); 91. Papilio machaon ZL. (Budd.); 92. Pieris napi Z.,
skg. (Liebenau) ; 93. Epinephele janira Z.(H. M.); 94. Erebia aethiops Zsp. (H. M.);
95- Vanessa urticae ZL. (d¢) Sphingidae: 96. Ino statices Z.; 97. Zygaena carniolica
Scop.; 98. Z. filipendulae Z.; 99. Z. minos S.V.; Buddeberg says that these Zygaenae
are almost entirely limited to this species and Carduus crispus; roo. Z. lonicerae £sp.,
freq. (H. M., Thuringia). (e) Ziezdae: 101. Adela sp., very common, sometimes as
many as 4 on a head.

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), 4 beetles, g flies, 10 Hymenoptera (including Andrena
hattorfiana /.), and 23 Lepidoptera (‘Alpenblumen,’ pp. 399-400). Wiistnei, 2 bees—
Andrena hattorfiana #. (Alsen, Eutin, Husum), and Nomada armata #.-Sch. (Alsen).
Schmiedeknecht (Thuringia)—Hymenoptera, Agzdae: 1. Andrena hattorfiana F. ;
2. Bombus derhamellus X.5; 3. B. pratorum Z.4; 4. Psithyrus barbutellus X. 6;
5. P. globosus Z£v. 9; 6. P. quadricolor Zep. $; 7. Nomada armata #7.-Sch. Hoffer
(Steiermark), the cuckoo-bee Psithyrus barbutellus A.8. Schletterer (Tyrol), the bees
Andrena marginata /. and Eucera cinerea Zep. Friese in Alsace (A.), Baden (B.),
Hungary (H.), and Mecklenburg (M.), the following bees.——1. Andrena hattorfiana /.
(A., B., M.); 2. A. marginata #. (M.); 3. Coelioxys acuminata Wy/., not rare (M.);
4. C.conoidea Ji. (H., M.); 5. C. mandibularis JVy/., rare (M.); 6. Dasypoda argen-
tata Pz., occasional (M.); 7. D. Thomsoni Sch/e//. 8 occasional, 9 very rare (M.);
8. Nomada armata #.-Sch., occasional (M.); 9. Stelis aterrima Pz., one 9 (A.);
10. Eucera pollinosa Zep., several (H.). H. de Vries (Netherlands), a bee (Halictus
cylindricus #. 9), and a humble-bee (Bombus agrorum F. 8) (Ned. Kruidk. Arch.,
Nijmegen, 2. ser., 2. deel, 1875). MacLeod (Pyrenees), 4 humble-bees, 4 Lepi-
dopterids, and 4 flies (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 347). Saunders (S.)

DIPSACEAE 563

and Smith (Sm.) (England), 5 bees:—1. Andrena hattorfiana /. (S., Sm.); 2. A.
marginata J. (S.,Sm.); 3. Nomada armata H.-Sch.(Sm.); 4. N. jacobeae Pz. (Sm.);
5. Osmia spinulosa A. (Sm.). Marquard (extreme W. of Cornwall), the bee Andrena
hattorfiana /.

1297. K. sylvatica Duby (=Scabiosa sylvatica Z.; and Trichera sylvatica
Schrad.).—Kirchner (‘Flora v. Stuttgart,’ p. 680) says that the mechanism of the
reddish-blue flowers of this species entirely agrees with that of K. arvensis, but
female stocks are very rare.

Visttors.—The following are recorded by the authorities, and for the localities
stated.—

Loew (Steiermark) (‘Beitrage,’ p. 50).—A. Diptera. (a) Conopidae: 1.Occemyia
atra &.,skg.; 2. Sicus ferrugineus Z., do. (4) Syrphidae: 3. Cheilosia personata Loew. ;
4. Rhingia rostrata M7/g., skg. B. Hymenoptera. Apzdae: 5. Andrena hattorfiana
F. 9, po-cltg.; 6. Ceratina cyanea K. 9, skg.; 7. Halictus zonulus Sm. 9, po-cltg. ;
8. Psithyrus barbutellus A. 9, skg.: also (Switzerland), the bee Physocephala rufipes
F. (op. cit., p. 59). Ricca, chiefly Lepidoptera (Atti Soc. ital. sc. nat., Milano, xiv,
1871). Kirchner, chiefly beetles, bees, and Lepidoptera. Herm. Miiller (Alps), 3 bees,
2 flies, 3 Lepidoptera, and a beetle (‘ Alpenblumen,’ p. 400).

4io. Succisa Neck.
Gynodioecious, with protandrous hermaphrodite florets; blue in colour, rarely
white ; aggregated into hemispherical heads ; belonging to class S.
1298. S. pratensis Moench (=Scabiosa succisa Z.). (Sprengel, ‘Entd. Geh..,’
p. 84; Herm. Miller, ‘ Fertilisation,’ pp. 313-14, ‘ Weit. Beob.,’ III, p. 77; Magnus,
Sitzber. Ges. natf. Freunde, Berlin, 1881 ; Schulz, ‘ Beitrage,’ IT, p. 192; Knuth, ‘Bl.

Fic. 188. Succtsa pratensts, Moench. (after Herm. Miller). (1) Flower before the anthers have
dehisced (after removal of the involucel). (2) Ditto, after the anthers have dehisced. (3) Ditto,
in the female stage.

u. Insekt. a. d. nordfr. Ins., pp. 84, 157, ‘ Bloemenbiol. Bijdragen,’ p. 31 (43), ‘ Weit.
Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 235.)—In this species, according to
Hermann Miiller, 50-80 tolerably similar florets are aggregated into a hemispherical
head. Nectar is secreted on the surface of the ovary (as in all Dipsaceae, and is
sheltered in the smooth narrow base of the corolla-tube (3 mm. long), which is lined
with hairs above and expands upwards to a diameter of 2 mm. The outermost of
the four (rarely five) corolla-lobes is the largest.
002

564 ANGIOSPERMAE—DICOTYLEDONES

The stamens are curved inwards in the bud, but when the floret opens they
extend one after the other, and their anthers dehisce, while the style has so far
attained to scarcely half its length. When the anthers have withered, the style
elongates to its full length, and the stigma becomes receptive. Automatic self-
pollination is consequently impossible.

Besides stocks bearing hermaphrodite florets, there are also somewhat smaller
female ones. Such florets possess stamens reduced to a greatly varying extent,
and they are not infrequently double. In some localities female stocks are rare ;
elsewhere they are commoner; Magnus says they number about 10 % at Homburg,
while Schulz states that they may even reach about 30 % at Brunswick and Halle.
Turner (Nature, London, xl, 1889) says the florets are trimorphous. Warnstorf
(Verh. bot. Ver., Berlin, xxxviii, 1896) describes the pollen-grains as very large,
white, spherical, adherent owing to the presence of short spinose tubercles, up to
93 » in diameter.

Visitors.—Herm. Miiller (H. M.) and myself (Kn.) observed the following in
North and Central Germany.—

A. Coleoptera. Chrysomelidae: 1. Cryptocephalus sericeus Z., gnawing the
florets (H. M.). B. Diptera. (a) Bombylizdae: 2. Exoprosopa capucina F£,, freq.
in July (H. M.). (4) Empidae: 3. Empis livida Z., very numerous, skg. (H. M.).
(c) Muscidae: 4. Lucilia species (H. M.); 5. L. cornicina & (H.M.). (d) Syrphidae :
6. Eristalis arbustorum Z., freq., skg. and po-dvg. (Kn., H.M.); 7. E. intricarius Z.
(H. M.); 8. E. nemorum Z., freq., skg. and po-dvg. (H. M.); 9. E. tenax Z., do.
(Kn., H. M.); 10. Helophilus pendulus Z., do. (Kn., H. M.); 11. Rhingia rostrata
L., skg. (Kn., H. M.); 12. Syrphus balteatus Deg., po-dvg. (Kn.); 13. S. pyrastri Z.,
skg. and po-dvg. (H. M.); 14. S. ribesii Z., do. (Kn.); 15. Volucella plumata Mz.,
skg. (H. M.). C. Hymenoptera. Apzdae: 16. Andrena cetii Schr. 9, po-dvg.
(H. M.); 17. A. convexiuscula X. § (H.M.); 18. Apis mellifica Z. ¥, freq., skg. and
po-dvg. (H. M., Kn.); 19. Bombus agrorum /. ¥ and 4, very common, skg. (H. M.) ;
20. B. lapidarius Z. ¥ and $, do. (Kn., H. M.); 21. B. muscorum /. ¥ and 6, do.
(H. M.); 22. B. pratorum Z., do. (H. M.); 23. B. sylvarum Z. 9, ¥% and 4, do.
(H. M.); 24. B. terrester Z. ¥ and 3, do.(H. M.); 25. Halictus cylindricus /. 3,
in large numbers (H. M.); 26. H. leucozonius Schr. 3, do. (H. M.); 27. H.
rubicundus Chr. 9, po-cltg. (H. M.); 28. H. zonulus Sm. 9, skg. (H.M.); 29.
Psithyrus rupestris Z. 3, do. (H. M.); 30. P. vestalis Pourcr. 9 and 3, do. (H. M.).
D. Lepidoptera. (a) Pyraidae: 31. Botys purpuralis Z., skg. (H. M.). (3) Woc-
tuidae: 32. Plusia gamma L., freq., skg.(Kn., H.M.). (c) Rhopalocera: 33. Epinephele
janira Z., skg. (Kn., H.M.); 34. Pieris brassicae Z., do. (Kn.); 35. P. rapae Z.,
numerous, skg.(Kn., H.M.); 36. Polyommatus phlaeas Z., do.(H.M.). (d) Sphingidae:
37. Zygaena filipendulae Z., very common, skg. (Kn.).

Alfken and Héppner (H.) record the following for Bremen.—

A. Hymenoptera. Afzdae: 1. Andrena marginata /.9 and é; 2. Bombus
agrorum /. 9, ¥ and 8; 3. B. arenicola Zhs. 9, ¥ and 5; 4. B. derhamellus X.
9, ¥ and 4, skg.; 5. B. hortorum Z., var. nigricans, Schmiedekn. §; 6. B. jonellus KX.
%andé; 7. B. lapidarius Z. 9, § and 6; 8. B. terrester Z. ¥ and 4; 9. B. lucorum L.
9; 10. B. proteus Gersé. 9, ¥ and é; 11. B. muscorum F.9, é and ¥; 12. B. sylvarum
£.¥% and $; 13. B. variabilis Schmredekn. 9, ¥§ and $; 14. Halictus calceatus Scop.
9g and é; 15. H. leucopus K. 9 and 6; 16. H. leucozonius K. 8; 17. H. rubicundus
Chr. 9; 18. H. zonulus Sm. 9 and $; 19. Psithyrus barbutellus AK. 9 and 3;
20. P. campestris Pz. 9 andé; 21. P. rupestris 7.9 and é; 22. P. vestalis Fourcr.
g and 6 B. Diptera. Syrphidae: 23. Arctophila mussitans #. 9. and 6, very
common, skg. (H.); 24. Eristalis tenax Z.

DIPSACEAE 565

Willis records the following for the south coast of Scotland (‘Fls. and Insects in
Gt. Britain,’ Part I).—

A. Diptera. (a) Muscidae: 1. Anthomyia sp., skg.; 2. Mydaea sp., do.
(4) Syrphidae: 3. Eristalis intricarius Z., skg.; 4. E. tenax Z., do.; 5. Helophilus
pendulus Z., do.; 6. Melanostoma scalare F., freq., po-dvg.; 7. Syrphus balteatus
Deg., skg. and po-dvg. B. Hymenoptera. Apzdae: 8. Bombus agrorum /,, freq.,
skg.; 9. B. pratorum Z., do.; ro. B. terrester Z., do. ; 11. Halictus cylindricus /, do. ;
12. H. rubicundus Chr., skg.; 13. Psithyrus campestris Pz. do. C. Lepidoptera.
Rhopalocera: 14. Pieris napi Z., skg.

Burkill saw the following on the east coast of Scotland (‘Fls. and Insects in Gt.
Britain,’ Part I).—

A. Coleoptera. (a) Chrysomelidae: 1. Crepidodera ferruginea Scop., skg.
(6) Nitidulidae: 2. Meligethes picipes Sturm, skg.; 3. M. viridescens /, do.
B. Diptera. (2) Muscidae: 4. Anthomyia brevicornis Zeff., skg.; 5. A. radicum
£., po-dvg.; 6. Calliphora erythrocephala Jfg., skg.; 7. Lucilia cornicina /., do. ;
8. Scatophaga stercoraria Mg., do.; 9. Siphona geniculata Deg., do. (5) Syrphidae :
ro. Eristalis tenax Z., skg.; 11. Melanostoma scalare /.; 12. Platycheirus manicatus
Mg.; 13. Sphaerophoria scripta Z.; 14. Syrphus balteatus Deg.; 15. S. ribesii Z.,
skg. C. Hymenoptera. (2) Apzdae: 16. Bombus agrorum /., very common, skg.
and po-cltg.; 17. B. hortorum Z., skg.; 18. B. lapidarius Z., skg. and po-cltg.
(6) Formicidae: 19. Myrmica rubra Z., skg. D. Lepidoptera. Noctucdae:
20. Plusia gamma Z., skg.

Burkill and Willis record the following for Mid-Wales (‘Fls. and Insects in
Gt. Britain,’ Part I).—

A. Coleoptera. Mitdulidae: 1. Meligethes viridescens /., freq. B. Diptera.
(2) Empidae: 2. Pachymeria palparis Zgg.; 3. Rhamphomyia sp., po-dvg. (4) Mus-
cidae: 4. Anthomyia sp., skg.; 5. Hyetodisia incana Wred., freq.; 6. Hylemyia lasciva
Zett.; 7. H. strigosa £”, freq.; 8. Lucilia cornicina &, po-dvg.; 9. Scatophaga stercoraria
L., skg.; 10. Siphona geniculata Deg., do.; 11. Trichophticus cunctans Mg., freq.
(c) Syrphidae: 12. Eristalis horticola Deg., skg.; 13. E. intricarius Z.; 14. E. pertinax
Scop., skg.; 15. E. rupium /, do.; 16. E. tenax Z., very common, skg.; 17. Helo-
philus pendulus Z., skg.; 18. Melanostoma scalare /.; 19. Platycheirus manicatus
Mg., skg.; 20. Sericomyia borealis Fad/., do.; 21. Volucella pellucens Z., do.
C. Hymenoptera. Apidae: 22. Bombus agrorum Z., freq., skg.; 23. B. hortorum
L., skg.; 24. B. lapidarius Z., freq., skg.; 25. B. pratorum Z., skg.; 26. B. scrim-
shiranus Kzréy, do.; 27. B. terrester Z., do. D. Lepidoptera. (a) Noctuidae:
28. Charaeas graminis Z., skg.; 29. Luperina haworthii Cur/., do.; 30. Plusia
gamma Z., do. (4) Rhopalocera: all skg.: 31. Coenonympha pamphilus Z.; 32.
Lycaena icarus Ro//.; 33. Pieris rapae Z.; 34. Polyommatus phlaeas Z.; 35. Vanessa
atalanta Z.; 36. V. c-album Z.; 37. V. urticae Z., freq. E. Thysanoptera.
38. Thrips sp., freq.

The following were recorded by the observers and for the localities stated.—

Sickmann (Osnabriick), the fossorial wasp Gorytes quadrifasciatus /. 9. Friese
for Hungary (H.) and Mecklenburg (M.), the following bees :—1. Andrena bimaculata
X., 2nd generation (M.); 2. A. marginata /. (M.) (according to Konow); 3. A.
nigriceps ., very rare (M.); 4. Epeolus variegatus Z., occasional (M.); 5. Nomada
jacobaeae Pz., not infrequent (M.); 6. N. roberjeotiana Pz., do. (M.); 7. N. solida-
ginis Pz., do. (M.); 8. Prosopis dilatata X., not rare (H.), occasional (M.). Réssler
(Wiesbaden), 2 Tineids (Nemotois cupriacellus /7én., and N. minimellus Zeéz),
Handlirsch, the fossorial wasp Gorytes quadrifasciatus #. MacLeod (Flanders),
3 long-tongued and 3 short-tongued bees, a Vespid, 9 hover-flies, a Muscid, and
6 Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 395-6). H. de Vries
(Netherlands), 3 humble-bees—1. Bombus agrorum 7. §; 2. B. lapidarius Z. ¥;

566 ANGIOSPERMAE—DICOTYLEDONES

3. B. subterraneus Z. § (Ned. Kruidk. Arch., Nijmegen, 2. ser., 2. deel, 1875).
Heinsius (Holland)—A. Hymenoptera. Apzdae: 1. Bombus rajellus K. 3.
B. Lepidoptera. (a) Rhopalocera: 2. Epinephele janira Z. 9; 3. Pieris napi Z. 5;
4. Polyommatus dorilis H/n. 6 and 9. (6) Sphingidae: 5. Zygaena filipendulae L.
3 and 9, freq.; 6. Z. trifolii Zsp.

1299. S. australis Reichb.—

Visitors.—Loew observed a Muscid (Echinomyia fera Z.) and a Syrphid
(Eristalis tenax Z.) in the Berlin Botanic Garden.

411. Scabiosa L.

Flowers lilac, reddish, white, or rarely yellow in colour; aggregated into
conspicuous heads: belonging to class S. Stocks gynodioecious, with protandrous
hermaphrodite flowers.

1300. S. Columbaria L. (Sprengel, ‘Entd. Geh.,’ pp. 82-4; Herm. Miiler,
‘Fertilisation,’ p. 315, ‘Alpenblumen,’ pp. 400-1 ; Knuth, ‘ Bloemenbiol. Bijdragen.’)—
Ludwig observed gynomonoecism more frequently than gynodioecism in this species.
Sprengel has given an admirable account of the mechanism of the protandrous
hermaphrodite flowers. It agrees with that of Knautia arvensis, except that there are
5 instead of 4 corolla-lobes. Each head contains 70-80 florets, those of the ray being
large and zygomorphous. Sprengel, in describing this species, makes special reference to
the oecological significance of these large marginal florets. Hermann Miiller measured
the florets, and found that the marginal ones possess corolla-tubes 6 mm. long, with
throats 2-24 mm. broad, the outer corolla-lobes being 7-8 mm. long, the lateral ones
6 mm., and the inner ones 2-3 mm. Next to these ray-florets come disk-florets with
corolla-tubes 5 mm. long and throats 2 mm. broad, their lobes being only 3 mm.,
2 mm., and 14 mm., in length respectively. The central florets of the head possess
corolla-tubes only 4 mm. long and 14 mm. broad, with lobes 14 mm. in length.
Hermann Miller adds that, owing to the small size of these, and the slight increase
in size from the middle to the periphery of the head, there is room for a much larger
number of florets in the same area in this species than in Knautia arvensis.

The female flowers, which are most numerous at the beginning of the flowering
season, also agree with those of Knautia arvensis, but they appear to occur only in
certain localities, for Hermann Miiller never found them at Lippstadt, where, however,
the plant is scarce.

VistTors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel Botanic Garden)—A. Hymenoptera. Apidae: 1. Apis mellifica
L. ¥, skg. and po-cltg.; 2. Bombus agrorum /. 9, do.; 3. B. lapidarius Z., do.
B. Diptera. Syrphidae: 4. Rhingia rostrata Z., skg. and po-cltg. C. Lepidoptera.
Rhopalocera: all skg.: 5. Vanessa atalanta Z.; 6. V. io Z.; 7. V.urticae Z. Herm.
Miiller (Alps), 6 Diptera, 7 Hymenoptera, 27 Lepidoptera; and (Westphalia) the
following :—A. Diptera. (2) Conopidae: 1. Sicus ferrugineus L., skg. (6) Syrphidae :
2. Eristalis nemorum Z., freq., skg. and po-dvg.; 3. E. tenax Z., do.; 4. Helophilus
trivittatus /., do. B. Hymenoptera. 5. Apis mellifica Z. ¥, freq., skg.; 6. Bombus
lapidarius Z., 4, very numerous, skg. Loew (Brandenburg) (‘ Beitrage,’ p. 40), 2 bees
(Andrena schencki Jor. 9, po-cltg.; and Halictus sexcinctus /. 4, skg.) ; also (Berlin
Botanic Garden).—A. Diptera. Syrphidae: 1. Helophilus floreus Z.; 2. H. tri-

DIPSACEAE 567

vittatus #.; 3. Syrphus ribesii Z. B. Hymenoptera. Aprdae: 4. Bombus agrorum
£. 3, skg.; 5. B. pratorum Z. 9,do. C. Lepidoptera. Rhopalocera: 6. Argynnis
latonia Z., skg.: 7. Colias rhamni Z., do. Wiistnei (Holstein), the bee Andrena cetii
Schr. Schmiedeknecht (Thuringia), 3 bees—r1. Andrena cetii Schr. (=A. marginata
¥); 2. Nomada brevicornis Mocs.; 3. Psithyrus barbutella X.é. Schenck (Nassau),
the bee Andrena marginata #. Schletterer and von Dalla Torre (Tyrol) 4 bees—
1. Bombus mesomelas Gerst.5; 2. B. soroénsis F.$; 3. Podalirius parietinus /- ;
4. Psithyrus campestris Pz. 9. Frey (Ziirich), the Tineid Nemotois minimellus
Zeit, MacLeod (Pyrenees), 11 Hymenoptera (including Andrena hattorfiana F.),
20 Lepidoptera, 4 beetles, and 13 Diptera (Bot. Jaarb. Dodonaea, Ghent, iii, 1891,
PP- 347-440).

1301. S. ochroleuca L. (as a species).—Schulz (‘ Beitrige,’ II, p. 192) describes
this yellow-flowered variety of S. columbaria as gynomonoecious with protandrous
hermaphrodite flowers. Comes (‘ Ult. stud.’) also speaks of the flowers as protandrous
and self-fertile.

Visirors.—The following were recorded by the observers and for the localities
stated.—

Friese (Hungary) (on the authority of Mocsary), 3 bees—1. Dasypoda argentata
Pz., and the var. braccata Zv., not infreq.; 2. Eucera pollinosa Zep., freq.; 3. E.
scabiosae Mocs.: also (Thuringia) the bee Cilissa haemorrhoidalis #. Loew (Steier-
mark), 2 bees—z. Andrena cetii Schr. 9, po-cltg.; 2. Halictus cylindricus F. 9, skg.:
also (Berlin Botanic Garden).—A. Diptera. Syrphidae: 1. Eristalis intricarius J. ;
2. E.nemorum Z.; 3. E. tenax Z.; 4. Pipiza festiva 1Zg.; 5. Syrphus balteatus Deg. ;
6. S. ribesii Z.; 7. Volucella pellucens Z., skg. B. Hymenoptera. Afzdae: 8.
Apis mellifica Z. ¥, skg. C. Lepidoptera. (a) NMoctuzdae: 2. Plusia gamma Z., skg.
(4) Rhopalocera: 10. Pieris brassicae Z., skg.; 11. P. rapae Z., do.

1302. S. suaveolens Desf.—Schulz (‘ Beitrage,’ I, pp. 67-8, II, p. 192) states
that this species is gynomonoecious, much more rarely gynodioecious, with markedly
protandrous hermaphrodite flowers. At first the strongly zygomorphous florets of the
two outer circlets come into bloom. Next come the innermost florets of the head,
and lastly those of the intermediate zone. Schulz asserts that the species is very
injuriously affected by this way of flowering, for in many cases the stigmas of the
two outer circlets of florets fully develop before the anthers of the next circlet have
dehisced. And as the stigmas and anthers of the florets of two adjacent circlets are
very close together, pollination of the outer by the inner ones, sometimes by means
of insects, is easily possible.

Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) states that plants growing on
sterile sandy ground under pine-trees near the Altruppin shooting-box, bear either
only protandrous hermaphrodite florets, or else some of the zygomorphous marginal
ones are female by degeneration of the anthers. This species is therefore gyno-
monoecious in the locality mentioned. The anthers are introrse, but during
dehiscence they turn through an angle of go°, lying horizontally at the end of
the long filaments which project far out of the florets. The pollen consequently
faces upwards. The pollen-grains are white, tuberculated, spheroidal or ellipsoidal,
as much as 112 p long and 88 » broad.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

568 ANGIOSPERMAE—DICOTYLEDONES

Loew (Berlin Botanic Garden), the Muscid Echinomyia fera Z. Krieger
(Leipzig), 4 bees—r. Andrena marginata F. (=A. cetii Schr.); 2. Nomada
jacobaeae Pz.; 3. N. roberjeotiana Pz.; 4. N. solidaginis Pz.

1303. S. lucida Vill—According to Schulz (‘Beitrage,’ I, p. 65, II, p. 192),
this species agrees with the last as regards the protandry of the hermaphrodite flowers,
gynomonoecism, and the course of anthesis.

Visrrors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Alps), 4 Lepidoptera and a hover-fly (‘ Alpenblumen,’ p. 401).
Loew (Berlin Botanic Garden).—A. Diptera. Syrphidae: 1. Helophilus pendulus LZ.
B. Hymenoptera. Apidae: 2. Bombus agrorum &. 4, skg.; 3. B. lapidarius LZ. 3,
do.; 4. Heriades truncorum JZ. 4, po-cltg. C. Lepidoptera. Rhopalocera: 5.

Argynnis aglaja Z.
1304. S. gramuntia L.—Schulz describes this species as being gynomonoecious
with protandrous hermaphrodite flowers at Bozen (‘Beitrage,’ II, p. 192).
VisiTors.—Schletterer (Pola) saw the bee Halictus calceatus Scop.
1305. S. Dallaportae Heldr.—

Vistrors—Loew observed a hover-fly (Helophilus trivittatus /), a bee (Apis
mellifica Z. ¥, skg.), and a butterfly (Rhodocera rhamni Z., skg.), in the Berlin
Botanic Garden.

1306. S. daucoides Desf.—

Visitors. —Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis aeneus Scop.; 2. E. intricarius Z.;
3. Helophilus trivittatus F.; 4. Pipiza festiva Me. B. Hymenoptera. Apzdae:
5. Apis mellifica Z. ¥, skg.; 6. Bombus lapidarius Z. 4, do. ; 7. B. terrester Z. 6, do. ;
8. Stelis aterrima Pz. 9, do. C. Lepidoptera. Rhopalocera: g. Pieris brassicae Z.,
skg.; 10. Vanessa urticae Z., do.

1307. S. Hladnikiana Host.—

Vistrors.—Loew observed 2 hover-flies (Syrphus balteatus Deg., and Volucella
pellucens Z., skg.), and a butterfly (Pieris brassicae Z., skg.), in the Berlin Botanic
Garden.

1308. S. ucranica L.—

VisiTors.—Loew saw a hover-fly (Syrphus corollae #.) in the Berlin Botanic
Garden.

1309. S. atropurpurea L. (=S. maritima Z.).—

VisiTors.— Plateau (Belgium) observed bees (Apis, Bombus hypnorum .Z.,
Megachile ericetorum Zeg.), hover-flies (Eristalis tenax Z., Syrphus sp.), and butter-
flies (Vanessa c-album Z., Pieris napi Z.). These insects visited with about equal
frequency the purple, red, rose-coloured, and white capitula.—For the butterflies,—
Vanessa io Z., Pieris brassicae Z., and P. napi Z., this also applied to the larger
capitula (4-5 cm. diameter) of a sub-species.

LV. ORDER COMPOSITAE Adans.

Sprengel long ago called attention to the advantageous oecological peculiarities
of the Compositae (‘Entd. Geh.,’ p. 365). Even before his time, Kélreuter had

COMPOSITAE 569

observed the sensitiveness of the filaments of certain species (of Centaurea, Ono-
pordon, Cichorium, and Hieracium) (‘3. Fortsetzung, p. 199, Leipzig, 1766).
Hildebrand’ has made the mechanism of the style the subject of a thorough
investigation (‘U. d. Geschlechtsverhaltn. b. d. Compositen, Nova Acta Leop.,
Halle, 1869). Delpino in 1870 discussed the oecological peculiarities of the

Fic. 189. Styles and Stigmas of Composttae (from nature). (1) Enlarged tip of the style of Bidens,
with branches still almost apposed; on the outside are prominent sweeping-hairs, which diminish in size
towards the apex. (2) Ditto, with the branches of the style spread out; their inner surface is closely beset
with stigmatic papillae. (3) Enlarged tip of the style of Aster, still closed. The conical apex is densely

beset with prominent sweeping-hairs, and below these are seen the stigmatic papillae. (4) Enlarged tip
of the style of a disk-floret of Chrysanthemum segetum Z., in the first (male) stage. (5) Ditto, in the
second (female) stage, greatly enlarged. (6) Enlarged tip of the style of a ray-floret of the same plant,

with diverging stigmatic branches. (7) Tip of the style of a disk-floret of the same plant, seen from
within: in the centre is a stylar groove filled with pollen-grains: greatly enlarged.

flowers of Compositae, and Hermann Miller in 1873 gave a review of the order
as regards pollination mechanisms (‘ Fertilisation,’ pp. 315-18).
The many small flowers are aggregated into a head, which is closely

1 He here deals with the following species.—Taraxacum officinale (pp. 7 et seq., Taf. I,
Figs. 1-7); Cichorium intybus (Taf. I, Figs. 8-10); Vernonia scaberrima (p. 14); Cacalia
sonchifolia (p. 15, Taf. I, Figs. 11-13); Eupatorium riparium and E. cannabinum (pp. 16-17,
Taf. I, Figs. 14-19); Liatris spicata (pp. 17-19, Taf. 1, Figs. 20-5); Dahlia variabilis (pp. 19-20,
Taf. I, Figs. 26-9); Bidens tripartita (Taf. I, Figs. 30-1); Agathaea coelestis (pp. 20-1, Taf. II,
Figs. 1-6); Solidago virgaurea (pp. 22-3, Taf. IT, Figs. 7-10); Bellis perennis (pp. 23-4, Taf. IT,
Figs. 11-15); Telekia speciosa (pp. 24-5, Taf. I, Figs. 16-17); Doronicum macrophyllum
(pp. 25-6, Taf. II, Figs. 13-28); Senecio populifolius (pp. 27-8, Taf. II, Figs. 29-36); Gaillardia
lanceolata (pp. 28-9, Taf. ITI, Figs. 1-3); Silphium doronicifolium (pp. 29-31, Taf. III, Figs. 4-9) ;
Calendula arvensis (pp. 31-3, Taf. III, Figs. 10-17); C. officinalis (p. 33, Taf. III, Figs. 18-20) ;
Melampodium divaricatum (pp. 33-4, Taf. III, Figs. 21-5); Madaria elegans (pp. 34-5, Taf. IV,
Figs. 26-7) ; Petasites officinalis (pp. 35-40, Taf. IV, Figs. 1-19); Gnaphalium dioicium (pp. 40-2,
Taf. III, Figs. 26-32) ; Gazania rigens and G. speciosa (pp. 42-4, Taf. IV, Figs. 20-5); Crypto-
stemma hypochondriacam (pp. 44-5, Taf. VI, Figs. 23-5); Arctotis acaulis (p. 45, Taf. VI,
Figs. 21-2); Lappa minor and other sp. (p. 46, Taf. V, Fig. 32); Echinops sphaerocephalus
(pp. 46-8, Taf. VI, Figs. 1-3); Xeranthemum annuum (pp. 48-50, Taf. V, Figs. 24-30);
Centaurea montana (pp. 50-6, Taf. V, Figs. 1-23); C. scabiosa (pp. 56-7); C. dealbata (pp. 59-60,
Taf. VI, Figs. 6-9); Cnicus benedictus (pp. 57-8, Taf. V, Fig. 31); Amberoa Lippii (Taf. VI,
Fig. 4-5) ; Jurinea alata (pp. 58-9) ; Silybum Marianum (pp. 60-2, Taf. VI, Figs. 10-20).

57° ANGIOSPERMAE—DICOTYLEDONES

surrounded in the bud by an involucre, usually made up of several series of
bracts. This subsequently serves as an adequate protection against creeping
animals, as well as to hold the inflorescence together. The whole is rendered
conspicuous by the crowding of the flowers, the effect being heightened by
outward curving of all the corollas, or by the production of the limb of each
into a long outer lobe. Finally, as in most cases, the marginal flowers have
lost their stamens, and even their pistils as well, being converted into long
radiating tongues that greatly add to the size of the head. These ray-florets
are often of a different colour from the disk-florets, so as to still further increase
the conspicuousness of the inflorescences. In a few cases (Carlina) the inner
bracts assume this function.

Another result of crowding is that numerous flowers of the same inflorescence
are simultaneously pollinated by insects creeping over them in search of nectar, or
collecting or devouring pollen. For in the first stage of anthesis the pollen-grains,
in the second stage the stigmatic papillae, are so far above the general level that
insects must rub against them. It is thus highly probable that crossing will be
effected; but in many cases automatic self-pollination takes place in the absence
of insect visitors, for the branches of the style bend back, applying to the stigmatic
papillae pollen still clinging to the sweeping hairs.

Nectar is secreted by a ridge surrounding the base of the style. It is so
abundant as to rise in the corolla-tube, and is protected from rain by the filaments
which converge above it. It is accessible both to long-tongued and short-tongued
insects. The Compositae are therefore typical examples of flower class S.

In the markedly protandrous hermaphrodite flowers pollen is discharged even in
the bud into the anther-cylinder that surrounds the style with its apposed stigmas.
As the style elongates it sweeps the pollen before it out of the cylinder, so that the
grains accumulate above the opening of the flower. This is effected with the help of
hairs or papillae which beset the surface of the style, and which have a characteristic
form and arrangement in different genera. (Cf Fig. 189.) Pollen adheres to
the ventral surface of insect visitors, this being rendered the more certain
because the filaments contract when touched by the proboscis as it is thrust
towards the nectar, so that the anther-tube often sinks several millimetres, and
the contained pollen is pressed out. After this has been accomplished the
stylar branches diverge, and the stigmatic papillae, which usually stud their inner
surfaces, become receptive.

Another advantageous peculiarity of the Compositae, remarked by Sprengel,
is that the heads close in unfavourable weather.

A short account (taken from Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, IJ, p. 318)
has already been given of the geitonogamy common in Compositae (Vol. I, pp. 41-2).
(Also cf Fig. 191.)

A. Tubuliflorae Less.
Disk-florets not ligulate.

1. Sus-oRDER CORYMBIFERAE Juss.

Florets either all tubular or, more commonly, ray-florets ligulate. Style not
thickened at the tip, and devoid of a terminal circlet of sweeping hairs.

COMPOSITAE 571

(2) Tribe Eupatorotdeae Less.

Style of the hermaphrodite florets cylindrical and two-branched. Stylar
branches }-terete or somewhat clavate, pubescent above.

412. Eupatorium L.

Flowers protandrous, but few in each of the heads, which are arranged in
dense terminal corymbs. The stylar branches are as long as the corolla-tube,
and beset for their lowest fourth with a row of stigmatic papillae on each side ;
above this densely clothed with sweeping-hairs. Flower class SL.

1310. E. cannabinum L. (Herm. Miller, ‘Fertilisation, pp. 318-20,
‘Alpenblumen,’ p. 450, ‘ Weit. Beob.,’ III, p. 92; MacLeod, Bot. Jaarb. Dodonaea,

FIG. 190. Eupatorium bi L. (after Herm. Miiller). (1) A head with four florets, in the first
(male) stage. (2) A single floret in the second (female) stage. From 2 to b each branch of the style is
beset with a line of stigmatic papillae on each side; from 2 toc it is covered with sweeping-hairs.
Ghent, iii, 1891, v, 1893, vi, 1894; Hildebrand, *U. d. Geschlechtsverhalt. b. d.
Compositen, pp. 16-17, Taf. I, Figs. 14-19; Kerner, ‘ Nat. Hist. Pl,’ Eng. Ed. 1,
II; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896; Knuth, ‘ Bloemenbiol.
Bijdragen,’ ‘ Blitenbiol. Herbst-Beob.’)—In this species, according to Hermann
Miiller, each head usually contains only five, or even only four, dull-red florets.
As, however, there are generally several hundred such heads arranged in dense

572 ANGIOSPERMAE—DICOTYLEDONES

corymbose inflorescences, these are very conspicuous, especially as the greatly
protruding stylar branches are white, while the margins of the involucral bracts
are reddish in colour. The corolla-tube is 24 mm. long, ending in a bell
scarcely 2 mm. in length, from which the stylar branches project for about 5 mm.

In the first stage of anthesis the lowest part of the stylar branches (beset with
stigmatic papillae) remains in the corolla-tube, while their upper three-fourths
(covered with sweeping-hairs) project freely, and diverge so widely that insect
visitors touch them, and rub off the pollen-grains clinging to the sweeping-hairs.
In the second stage of anthesis the lower parts of the stylar branches project
from the bell, so that insects probing for nectar must come into contact with
the stigmatic papillae. Warnstorf adds that the anther-cylinders do not project
out of the florets, and that the sweeping-hairs are thick, bluntly conical papillae,
sometimes two-celled, delicately striated, and projecting horizontally. The pollen-
grains are white, roundish to ellipsoidal,
spinose, on an average 25 » in diameter.

If insect visitors are sufficiently
numerous to remove the pollen from
the sweeping-hairs before the stigmas
project, cross-pollination is assured. If,
however, pollen still clings to the sweep-
ing-hairs when the lower parts of the
stylar branches protrude, self-pollination
is equally possible as the result of insect-
visits. Automatic self-pollination cannot
take place in the total absence of
visitors, but geitonogamy undoubtedly
may, for the stylar branches spread
out so far that they occasionally touch
the stigmas of adjacent florets. Kerner
(Nat. Hist. Pl.” Eng. Ed. 1, II, p. 319)
calls particular attention to this possibility
(Fig. 191).

Visirors.— Of these Lepidoptera
are the chief observed.

Herm. Miiller (Alps) noticed 6
Diptera, 4 Hymenoptera, and 6 Lepi-

Fic. 191. Gettonogamy of Eupatortum can-
nabinum, L. (after Kerner). Fertilization of neigh- doptera. He (H. M.) and Buddeberg

bouring florets is effected by means of pollen adhering (Budd.) give the following list for Central

to the stylar branches.
Germany.—

A. Diptera. (a) Muscidae: 1. Dexia canina &. (H. M.); 2. Echinomyia fera
L. (H.M.); 3. Lucilia albiceps WZ. (H.M.). (2) Syrphidae: 4. Eristalis arbustorum
L. freq., po-dvg. (H. M.); 5. E. nemorum Z., do. (H. M.); 6. E. tenax Z., do.
(H. M.). B. Hymenoptera. Apidae: 7. Apis mellifica Z. ¥, skg. (H. M.);
8. Psithyrus vestalis Fourcr. 6, skg. (H.M.). C. Lepidoptera. (a) Bombycidae:
g. Callimorpha dominula Z., skg. (Budd.). (6) Rhopalocera: all skg.: 10. Argynnis
paphia Z., freq. (H. M., Budd.); 11. Erebia medusa S.V. (H. M.); 12. Hesperia

COMPOSITAE 573

lineola O. (H. M.); 13. Lycaena sp. (H. M.); 14. Melanargia galatea Z. (H. M.);
15. Pararge egeria Z. (H.M.); 16. Pieris rapae Z. (H.M.); 17. Thecla quercus Z.
(H. M.); 18. Vanessa io Z., freq. (H. M., Budd.). D. Neuroptera. 19. Panorpa
communis Z., skg. (H. M.).

Burkill observed the following on the east coast of Scotland (‘Fls. and Insects
in Great Britain,’ Part I).—

A. Coleoptera. NWitidulidae: 1. Meligethes picipes S‘urm, po-dvg. B. Di-
ptera. (a) Muscidae: 2. Anthomyia brevicornis Ze//., po-dvg.; 3. A. radicum Z.,
skg. and po-dvg.; 4. Calliphora erythrocephala M/g., skg.; 5. Lucilia cornicina F.,
do. ; 6. Onesia sepulcralis AZg.; 7. Scatophaga stercoraria Z.; 8. Siphona geniculata
Deg. (6) Syrphidae: g. Eristalis horticola Deg., skg.; 10. E. pertinax Scop., do. ;
11. E. tenax Z., do.; 12. Platycheirus manicatus /g.; 13. Sphaerophoria scripta
L.; 14. Syritta pipiens Z.; 15. Syrphus ribesii Z. (c) Phorzdae: 16. Phora sp.
C. Hymenoptera. (a) Apidae: 17. Bombus lapidarius Z., skg. (6) Formicidae:
18. Myrmica rubra Z., creeping over the inflorescences. D. Lepidoptera.
Rhopalocera: 19. Vanessa urticae Z., skg.

The following were recorded by the observers, and for the localities stated._—

Knuth (Gliicksburg), 2 butterflies, skg. (Vanessa io Z., and Pieris napi Z.),
and a hover-fly (Eristalis tenax Z.). The same at Kiel, also Apis. von Dalla Torre
(Tyrol), a true wasp (Vespa norvegica #.), 2 bee (Halictus leucopus KX. 9), a
parasitic humble-bee (Psithyrus campestris Pz. 6), and 2 humble-bees (Bombus
muscorum /. ¥). F. F. Kohl (Tyrol), the ruby-wasp Cleptes semiauratus Z.
Schletterer (Tyrol), a bee (Halictus leucopus X.), a parasitic humble-bee (Psithyrus
campestris Pz.), and 3 humble-bees (Bombus pomorum P2z.; B. soroénsis Z.;
B. variabilis Schmiedekn.). Gerstacker (Kreuth), the commonest parasitic humble-
bee Psithyrus vestalis #ourcr. Alfken (Juist), the parasitic humble-bee Psithyrus
rupestris #. 8. Schiner (Austria), the Tabanid Silvius vituli # Frey (Switzerland),
the Muscid Callimorpha hera Z., and the moth Tortrix inopiana Haw. MacLeod
(Flanders), Apis and a Muscid (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 373);
also 4 hover-flies, a Muscid, and 3 Lepidoptera (op. cit., v, 1893, p. 410);
{Pyrenees), 3 Lepidoptera and a Muscid (op. cit., iii, 1891, p. 359).

1gu1. E. riparium Regel.—Hildebrand (op. cit.) states that the mechanism of
this species agrees with that of E. cannabinum.

1312. E. ageratoides L.—

Visrtors.—Loew observed the following visitors in the Berlin Botanic Garden.—

Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Echinomyia fera Z. (2) Syr-

phidae: 3. Eristalis arbustorum Z.; 4. E. nemorum Z.; 5. Syritta pipiens Z.;
6. Syrphus corollae /.

1313. E. purpureum L.—

VisrTors.—Loew observed the following visitors in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z.; 2. E. tenax Z.; 3.
Helophilus floreus Z.; 4. Melithreptus scriptus Z.; 5. Syritta pipiens Z.; 6.
Syrphus ribesii Z., ]kg. and po-dvg. in the styles of some florets. B. Hymenoptera.
Apidae: 7. Apis mellifica Z. ¥, skg.; 8. Bombus terrester Z. 9 and 4, do. ; 9. Psithyrus
rupestris /. 6, do.; 10. P. vestalis Fourcr. 6. C. Lepidoptera. Rhopalocera:
11. Pieris brassicae Z., skg.; 12. Vanessa c-album Z., do.

413. Vernonia Schreb.
1314. V. fasciculata Michx.—

Visttors.—Loew observed the following in the Berlin Botanic Garden.—
A. Diptera. Syrphidae: 1. Syrphus balteatus Deg. B. Hymenoptera.

574 ANGIOSPERMAE—DICOTYLEDONES

Apidae: 2. Apis mellifica Z. ¥, skg.; 3. Bombus lapidarius Z. , do.; 4. B. terrester
L£. 3, do.; 5. Psithyrus vestalis Fourcr. $, do. C. Lepidoptera. Rhopalocera:
6. Pieris brassicae L., do.

1315. V. praealta Ell_—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Hymenoptera. Apidae: 1. Apis mellifica Z. ¥, skg.; 2. Bombus
terrester Z. $, do.; 3. Psithyrus vestalis Fourcr. $, do. B. Lepidoptera.
Rhopalocera: 4. Pieris brassicae Z., skg.

FIG. 192. Adenostyles alpina, Bluff et Fing (after Herm. Miller). 4. A head with four florets; the two
in the middle are in the first (male) stage, the two outer ones in the second (female) stage (x 7). 2. Single

floret (x 7). C. Part of a stylar branch (x 80). 4, sweeping-hairs; #0, pollen clinging to these; sé,
stigmas.

414. Adenostyles Cass.

Flowers protandrous, but few in each of the heads, which are arranged in dense
corymbs. The whole outer surface of the style is thickly covered with small capitate
sweeping-hairs. These are short-stalked, and their heads are glandular. The inner
surfaces of both stylar branches are closely beset with minute stigmatic papillae.
Kerner states that some species possess no sweeping-hairs. Flower class SL.

COMPOSITAE 575

1316. A. alpina Bluff et Fing (=A. viridis Cass., and Cacalia alpina Z.).
(Herm. Miiller, ‘Alpenblumen,’ pp. 450-2.)—As in Eupatorium cannabinum, each
head in this species contains only 4-5 florets; the corolla consists of a tube
about 3 mm. long, with a slightly longer bell. When the two stylar branches
diverge they often rupture the anther-cylinder, and ultimately bend back so far
that the stigmatic papillae touch the sweeping-hairs, so that if any pollen-grains
still cling to these automatic self-pollination may be effected.

Visttors.—The following were recorded by the observers, and for the localities
stated.—
Herm. Miiller (Alps), chiefly Lepidoptera (21), less frequently beetles (14) and

Hymenoptera (4). Schletterer and von Dalla Torre (Tyrol) record the humble-bee
Bombus terrester.

1317. A. albida Cass. (=A. albifrons Rezchd.). (Herm. Miiller, ‘ Alpenblumen,.’
p. 452.)—Kerner states that there are no sweeping-hairs in this species, so that the
pollen is pressed out by
the ends of the stylar
branches.

Visttors.—The fol-
lowing were recorded by
the observers, and for the
localities stated.—

Herm. Miiller (Alps), »
only Diptera (Echino-
myia and Eristalis), though
probably the chief visitors
are Lepidoptera as in
the last species. Loew
(Altvatergebirge), a Chry-
somelid (Chrysomela ca-

caliae Schr., subsp. sene- FIG. 193. Homogyne alpina, Cass. (after Herm. Miller). 4. Disk-

toni floret in the first (male) stage ; pappus omitted. B. Ditto, in the second
res Schumm.), . and a female) stage. C. D. &, Ray-florets with reduced corolla limb and long
Pyralid (undetermined sp., _ protruding style.

skg.).
1318. A. hybrida DC. (=A. candidissima Cass.). (Herm. Miller, ‘Alpen-
blumen,’ p. 452.)—
Visttors.—Herm. Miiller only observed the hover-fly Eristalis tenax Z., skg.
and po-dvg.
415. Homogyne Cass.

Gynomonoccious, with protandrous hermaphrodite florets, many of which are
aggregated into a head, Ray-florets female, filiform; disk-florets hermaphrodite,
tubular. Stylar branches of the latter beset with sweeping-hairs externally, and
stigmatic papillae internally. Flower class SL.

1319. H. alpina Cass. (=Tussilago alpina Z.). (Herm. Miiller, ‘Alpenblumen,’
pp. 452-43 Ricca, Atti Soc. ital. sc. nat., Milano, xiii, 1870; Kerner, ‘ Nat. Hist. Pl.,’

576 ANGIOSPERMAE—DICOTYLEDONES

Eng. Ed. 1, II.)—In this species the ray-florets are female, nectarless, with a more
or less reduced corolla-limb. The disk-florets are hermaphrodite and distinctly
protandrous, so that, according to Hermann Miiller, automatic self-pollination is
completely excluded or almost so. Kerner states that geitonogamy is possible,
by curving outwards of the pollen-covered stylar branches.

Visrrors.—Ricca observed flies; Herm. Miiller chiefly Lepidoptera (28), also
5 flies and a humble-bee.

416. Tussilago L.

Monoecious. Disk-florets male, with vestigial pistils; the ray-florets female ;
multiseriate, ligulate. The stylar branches of the female florets possess stigmatic
papillae on their inner surfaces, and (useless) sweeping-hairs externally and apically.
In male flowers these branches remain united almost to their tips; externally and
above they are covered with short sweeping-hairs.

1320. T. Farfara L. (Sprengel, ‘Entd. Geh.” pp. 374-6; Herm. Miller,
‘Fertilisation,’ pp. 333-4, ‘Alpenblumen,’ pp. 454-5; Knuth, ‘ Bl. u. Insekt. a. d.
nordfr. Ins.,’ pp. 85, 157; Kerner, ‘ Nat. Hist. Pl.’ Eng. Ed. 1, II.)—In this species
30-40 golden-yellow purely male disk-florets are surrounded by about 300 purely
female ray-florets, similarly coloured, and multiseriate. The head expands in the
sunshine into a surface 20-25 mm. broad, and closes at night or during dull weather.
The male florets possess an ovary containing a vestigial ovule, and there is a yellow
nectar-ring at the base of the style. The pollen is pushed out of the anther-cylinder
by the sweeping-hairs.

The nectarless female ray-florets possess a corolla-tube 3 mm. long, with a
narrow linear limb 6-8 mm. in length, and directed outwards. The style projects
2-3 mm. from the floret, and is divided at its end into two stylar branches about
% mm. long, and papillose internally. The stigmas of the ray-florets become
receptive a considerable time before the pollen is swept out of the male ones.
There is therefore constant crossing of different stocks when insect-visits are
sufficiently numerous. Owing to the unisexual character of the florets, automatic
self-pollination is necessarily excluded. Kerner states that automatic geitonogamy
takes place when the ligulate ray-florets close about 5-6 p.m. In doing so, they
bend over the disk-florets in such a way that they touch the pollen-masses which
have been swept out of the anther-cylinders of the male florets. The pollen-grains
adhere to the ligulate florets, and when the head opens again next morning some
of them slide down to the respective stigmas. Warnstorf describes the pollen-grains
as golden-yellow in colour, rounded to ellipsoidal, densely spinose, about 44 » long
and 37 p» broad.

According to Burkill (‘ Fertlsn. of Spring Fls.’), the heads that blossom on the
cliffs of the Yorkshire coast contain about 200 to 300 female florets and about
40 male ones, and are 20-36 mm. in diameter; while those on the lower slopes
of the shore only attain a diameter of about 15 mm. During anthesis the corolla-
tubes of the disk-florets lengthen about 1 mm.; the ligulate florets simultaneously
increase in size, and their receptacles broaden. The head, therefore, becomes
somewhat more conspicuous as it passes from the first to the last stage. As

COMPOSITAE 577

the heads become older, and the growth of the corollas ceases, the power of
closing at night is gradually lost; this is well known to depend upon differential
growth of the ligulate florets. Female florets are consequently better protected
against unfavourable weather than male ones. After the female florets have been
fertilized they still retain their fresh appearance, and it is only when the male

florets, which open much later, have discharged their pollen, that the heads wither.

Visrtors.—Alfken and Héppner (H.) observed the following at Bremen.—

A. Diptera. (a) Conopidae: 1. Myopa sp. (4) Muscidae: 2. Anthomyia sp. ;
3. Lucilia caesar Z., skg. and po-cltg.; 4. Musca domestica Z.; 5. Pollenia vespillo
Ff’, freq., skg. (c) Syrphidae: 6. Cheilosia sp.; 7. Eristalis tenax Z., very common, skg.
B. Hymenoptera. (a) Apidae: 8. Andrena albicans Jil. 5, not infrequent, skg. ;
g. A. albicrus X. 4, rare, skg.; 10. A. apicata Sm. 6 and 9 (H.); 11. A. clarkella X.
9, not infrequent, skg.; 12: A. flavipes Pz. 9, freq., skg. and po-dvg., é freq., skg.;
13. A. gwynana X., not infrequent, 9, skg. and po-cltg., § skg.; 14. A. lapponica Ze/t.
$; 15. A. morawitzi Zhs. 9 and %, skg. (H.); 16. A. nigroaenea K. 9 and 3 wee
17. A. parvula X., freq., 9 skg. and po-cltg., 6 skg.; 18. A. praecox Scop. 9 and 3;
1g. A, rufitarsis Ze“, 9 and; 20. A. thoracica F’, rare, 9 skg. and po-cltg., & skg.;
21. A. tibialis A. é (H.); 22. A. tibialis K. 9, rare; 23. A. varians K. 9 and 4, rare;
24. Apis mellifica Z., §, very common, skg. and po-dvg.; 25. Bombus agrorum F. 9
(H.); 26. B. jonellus A. 9, freq.; 27. B. terrester Z. 9, do.; 28. Nomada bifida Zhs.
6, one; 29. N. borealis Zef//. 9 and 4, not infrequent, skg.; 30. N. fabriciana Z. 9,
one, skg. 6, freq., skg.; 31. N. flavoguttata K., var. héppneri Adffen 9 and ¥
(5) Sphegidae: 32. Ammophila sabulosa ZL. 4, freq., skg.; 33. A. hirsuta Scop.
g and 4, skg. C. Lepidoptera. (a) Nymphalidae: 34. Vanessa antiopa L., skg.
(H.); 35. V. io Z., freq., skg.; 36. V. urticae Z., do. (4) Pieridae: 37. Rhopalocera
rhamni Z., freq., skg.

Burkill gives the following list for the Yorkshire coast (‘Fertlsn. of Spring
Fis.’).—

A. Araneidae. 1. Xysticus pini Hahn, freq., lying in wait for flies.
B. Coleoptera. 2. Meligethes picipes S/rm., po-dvg.; 3. Omalium florale PayA. ;
4. Thyamis fuscicollis Foudr. C, Diptera. (a) Muscidae: 5. Calliphora (Onesia)
cognata Mg.; 6. C. erythrocephala Me., skg.; 7. C. (Onesia) sepulcralis JZg., do. ;
8. C. vomitoria Z., do.; 9. Actora aestuum J/g., po-dvg.; 10. Coelopa sp.; 11.
Ephydra sp.; 12. Helomyza sp., skg. and po-dvg.; 13. Hylemyia sp.; 14. Lasiops
sp., Skg.; 15. Lucilia cornicina &., skg. and po-dvg.; 16. Phorbia sp.; 17. Pollenia
tudis #.; 18. Scatophaga stercoraria Z., skg. and po-dvg.; 19. Sepsis nigripes Mg.,
do. (6) Phorzdae: 20. Phora sp. (c) Syrphidae: 21. Eristalis horticola Deg.; 22. E.
pertinax Scop., skg. and po-dvg.; 23. Melanostoma quadrimaculata Verral, skg. ;
24. Platycheirus sp.; 25. Syrphus lasiophthalmus Ze//.; 26. S. macularis Ze//. (d)
Cecidomyidae: 27. Cecidomyia sp. D. Hymenoptera. (a) Apidae: 28. Andrena
clarkella K. 9 and 4, skg.; 29. A. gwynana K. 9 and $; 30. A. nigroaenea K. 9 and
3; 31. Apis mellifica Z., skg. and po-cltg.; 32. Bombus agrorum /., skg.; 33. B.
terrester £., do. (4) Formicidae: 34. Formica fusca Z. (c) Ichneumonidae: 35.
Ichneumon sp. E. Lepidoptera. Rhopalocera: 36. Vanessa urticae Z. skg.
F. Thysanoptera. 37. Thrips sp.

The following were recorded by the observers, and for the localities stated.—

Knuth (Kiel), the honey-bee, very freq., skg. and po-cltg. Herm. Miiller
(Westphalia), 4 bees—1. Apis mellifica Z., skg. and po-cltg.; 2. Andrena fulvicrus
K. 9, skg.; 3. A. gwynana X. 9, skg. and po-dvg.; 4. A. parvula X. 9, do.; flies
(Bombylius major Z., skg., and Eristalis tenax Z., po-dvg.); and the beetle
Meligethes, freq., po-dvg.; also (Alps), 21 flies, 3 bees, an ant, and 2 Lepidoptera.
Wiistnei (Alsen), 2 bees (Andrena tibialis A., and A. ruficrus JVy/.), MacLeod

DAVIS, Il P p

578 ANGIOSPERMAE—DICOTYLEDONES

(Flanders), Apis, a short-tongued bee, and 3 flies (Bot. Jaarb. Dodonaea, Ghent, vi,
1894, p- 373). Friese (Fiume), the bee Andrena taraxaci Gzr. Schneider (Arctic
Norway), 2 humble-bees—Bombus hypnorum Z., and B. terrester Z. (Troms@ Mus.
Aarsh., 1894).

417. Petasites L.

Dioecious-polygamous. Male florets with a bell-shaped regularly five-toothed
limb to the corolla: female florets filiform, with the corolla-limb obliquely truncated.
The stylar branches, especially in male florets, covered externally with sweeping-
hairs; papillose on their inner surfaces in female florets. Kerner points out that
sexually differing stocks also differ from one another in appearance. In one kind
of plant there are numerous pseudo-hermaphrodite pollen-florets in the disk, and
a small number of female flowers round the margin; in the other kind the relative
numbers are reversed. .

1321. P. officinalis Moench (=P. vulgaris Des~, Tussilago Petasites Z., the
so-called hermaphrodite plant, and T. hybrida Z., the female plant). (Hildebrand,
‘U. d. Geschlechtsverhilt. b. d. Compositen,’ pp. 35-40, Taf. IV, Figs. 1-19; Kirchner,
«Flora v. Stuttgart,’ p. 690; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, I1; Herm. Miiller,
‘Weit. Beob.,’ III, p. 92; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 411;
Knuth, ‘ Bloemenbiol. Bijdragen’ ; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—
The florets of this species are of a dull-purple colour, rarely pale-red, or nearly
white. The plant occurs in two forms quite different from one another in appearance.
Kerner says that in one of these there are numerous pseudo-hermaphrodite pollen-
florets in the disk, and a smaller number of purely female ray-florets; in the other
form the relative numbers are reversed. Male stocks possess shorter peduncles
and crowded inflorescences; the 22-38 florets in a head, according to Kirchner,
are either all similar and nectariferous, or there may be as many as three herma-
phrodite florets among them. The ovule is usually vestigial; the style presents
a club-shaped somewhat compressed thickening under its branches, which are beset
with sweeping-hairs. The stylar branches curve away a little from each other,
and are covered externally with short sweeping-hairs; they are devoid of stigmatic
papillae internally. The corolla of the male florets is tubular below, expanding
above into a bell with five recurved lobes.

For the female plants, Kirchner describes the inflorescences as taller but less
dense. Each head contains about 140 florets, of which 1-3 central ones are male.
The female florets are nectarless and show no trace of stamens. Their corolla is
in the form of a long narrow tube, which is produced into two lips, one narrow
and the other broader. The style is filiform and smooth; its two branches are
covered with short hairs externally, and beset with stigmatic papillae internally.
The 1-3 male florets in the centre possess a bifurcated style, but slightly if at all
thickened, and covered with sweeping-hairs. The annular nectary secretes abundantly ;
the anthers are vestigial and produce no pollen.

Burkill observed only male inflorescences on the Yorkshire coast (‘Fertlsn.
of Spring Fis.’). At Neu-Ruppin (Brandenburg) the species also only occurs,
according to Warnstorf, in one form bearing pseudo-hermaphrodite infertile pollen-
florets. The stylar branches, which are densely covered with papillose sweeping-

COMPOSITAE 579

hairs, remain permanently apposed, and project far out of the florets, their only
function being to brush the pollen-masses out of the anther-cylinder. Here, there-
fore, we have a third different kind of stock; the two others have already been
described.

The pollen-grains are white, roundish to ellipsoidal, up to 37 » long and 31 p
broad, with long spines adapted to secure adhesion to the stigmatic papillae.

Fic. 194. Fetasttes albus, Gaertn. (after Herm. Miller). A. Sexual floret from a female head
(ov, ruptured ovary with projecting ovule. 4’. Upperpartofacorolla. 2. Nectar-secreting floret of
the female head. C. Ditto, split longitudinally. D. Sexual nectar-secreting floret from a male head.
E. Part ofa stylar branch from the same floret. F. Degenerate floret from a male head. (4-D and
F x7; & x80.) a, anther-cylinder; co, corolla; fi, filament; gr, style; 7, nectary; ov. ovary.

Visrtors.—The following were recorded by the observers and for the localities

stated,—

Knuth (Kiel), the honey-bee. Warnstorf (Ruppin), do. Wiistnei (Alsen), the
humble-bee Bombus terrester Z. Alfken (Bremen), 2 humble-bees—1. Bombus
lucorum Z. 9; 2. B. pratorum ZL. 9. MacLeod (Belgium), 5 bees, 3 Lepidoptera,
and small flies. Burkill (Yorkshire coast), a midge (Chironomus sp.), a bug (Hetero-
cordylus sp.), and 2 bees (Andrena gwynana X. 9, and Bombus terrester L.), skg.

Pp2

580 ANGIOSPERMAE—DICOTYLEDONES

1322. P. albus Gaertn. (=Tussilago alba Z., the hermaphrodite plant, and
T. ramosa Hoppe, the female plant). (Herm. Miller, ‘Fertilisation, pp. 334-5,
‘Alpenblumen,’ pp. 455-9.)—This species is dioecious, with four different forms
of floret. In the heads of the female stocks there are both nectar-secreting and
nectarless sexual florets. In the heads of the male stocks there are often only
nectar-secreting and pollen-producing florets, but about as frequent as these are
heads which also include one or two florets producing neither nectar nor pollen.

Visitors.—Herm. Miiller (Alps) observed 6 flies, Lepidoptera (only 2 species,
but very numerous individuals), 2 beetles.

1323. P. fragrans Pres].—

Visitors. — Burkill observed the following Diptera on the Yorkshire coast
(‘Fertlsn. of Spring Fls.’).—

(2) Muscidae: 1. Lucilia cornicina F.; 2. Onesia cognata Mg., po-dvg.; 3. O.
sepulcralis AZg., do. (6) Syrphidae: 4. Eristalis pertinax Scop.; 5. Melanostoma
quadrimaculata Verral.

1324. P. frigida Fries.—According to the observations of Lindman on the
Dovrefjeld, this northern species is dioecious, though to a lesser degree than P. albus.
The florets are odourless. In male heads the ray-florets are purely female: the
reddish stylar branches project from the large but few male florets, and have not
only the function of sweeping out pollen, but also serve to attract insects. The
female heads are smaller than the male: their disk-florets possess vestigial stamens,
but their ray-florets are purely female. Ekstam gives the diameter of heads examined
by him in Nova Zemlia as about 10 mm. According to Andersson and Hesselman
(‘Bidrag til Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ pp. 11-12),
separate sexual forms are found in Spitzbergen, as on the Dovrefjeld. The male
heads are 12-5-16 mm. long; the ligulate corolla-lobe of the ray-florets is strongly
developed, and projects beyond the style. The female heads are only 11-14 mm.
long: the ray-florets possess a shorter recurved corolla-lobe. The florets are bright
reddish, the male ones being somewhat more strongly coloured. The fragrance
is tolerably well marked (cf Ekstam, ‘Bliitenbiol. Beob. a. Spitzbergen,’ p. 6).
Ripe fruits are but rarely formed in Spitzbergen, though De Geer (according to
Nathorst) observed some on August 30, 1882.

Visirors.—Ekstam saw a medium-sized fly in Nova Zemlia, but no visitors
in Spitzbergen.

(4) Tribe Asterordeac Less.

Stylar branches linear, pointed, almost flat externally, otherwise as in the
last tribe.

418. Aster Tourn.

Ray-florets uniseriate, female, usually differently coloured from the yellow disk-
florets. Stylar branches broadened, covered with sweeping-hairs above, and beset
by stigmatic papillae laterally and internally: their tips almost always inclined
together in hermaphrodite florets.

Kerner states that the stylar branches of the ray-florets come into contact
with the pollen-masses thrust out from the disk-florets. He also says that automatic

COMPOSITAE 561

self-pollination occurs in the latter, by crossing of the stylar branches, but I have
never noticed this.

1325. A. alpinus L. (Herm. Miiller, ‘Alpenblumen,’ pp. 447-8 ; Kerner, ‘ Nat.
Hist. Pl.’ Eng. Ed. 1, II; Loew, ‘ Bliitenbiol. Floristik,’ p. 397; Knuth, ‘Bloemen-
biol. Bijdragen.’)—This species is gynomonoecious, with protandrous {hermaphrodite
florets. The 50-150 yellow disk-florets are surrounded by 24—40 violet ray-florets,
making up a terminal head 32-45 mm. in diameter. The style with its two diverging
branches projects 2-3 mm. out of the ray-florets. From the bells of the disk-florets
pollen is first extruded, and then the stylar branches, which curve together above,
make their appearance. Kerner states that the stigmas of the female florets became
receptive several days before pollen is produced by the neighbouring hermaphrodite
florets.

Visttors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miller (Alps), 2 beetles, 2 bees, 36 Lepidoptera, and g flies. Knuth
(Andermatt, July, 1878), 3 hover-flies (Eristalis tenax Z., Helophilus trivittatus J,

Melanostoma mellina Z., skg. and po-dvg.), and numerous Lepidoptera. Loew
(Switzerland), a hover-fly (Cheilosia caerulescens J/g.), and a butterfly (Lycaena sp.).

1326. A. Tripolium L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 86-7,
157, ‘Weit. Beob. ti. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 235; MacLeod, Bot.
Centralbl., Cassel, xxix, 1887.)—The heads of this species are about 20 mm. in
diameter, each containing 20-30 bright violet female ray-florets, with a corolla-lobe

l

Fic. 195. Aster Tripolium, L. (from nature). (1) Floret in the first (male) stage: the pollen is
being extruded from the tip of the anther-cylinder; the lobes of the corolla are spread out. (2) Ditto, in
the second (female) stage; the projecting style is provided with apical sweeping-hairs, and below these are
stigmatic papillae ; the anther-cylinder has retracted into the corolla, the lobes of which now point
upwards. (3) Greatly enlarged tip of the style of a floret inthe female stage; above are the sweeping-
hairs, and below the stigmatic papillae. (4) A stylar branch seen from the outside; above and in the
middle are sweeping-hairs; below and laterally stigmatic papillae. (5) Ditto, from within.

about 11 mm. long and 23 mm. broad, and an equal number of yellow tubular
hermaphrodite disk-florets. Owing to the colour contrast presented by the ray-_
and disk-florets, as well as the crowding together of numerous heads, the plant
is rendered very conspicuous. The corolla of the disk-florets is contracted below
like a stalk for a length of 4 mm., expanding above into a bell 2 mm. long. The
pollen is swept out by the rhombic tips of the stylar branches, which are provided
with sweeping-hairs directed obliquely upwards. When the anther-cylinder is empty,
the stylar branches project 2 mm. from the inflorescence. Below the collecting-

582 ANGIOSPERMAE— DICOTYLEDONES

hairs these branches are beset with stigmatic papillae internally, and on a longi-
tudinal outer fold; they remain apposed at their tips. Insects visiting flowers in
the first stage therefore get their under-surfaces covered with pollen, and will carry
this to the stigmas of heads in the second stage. Automatic self-pollination is possible
should insect-visits fail, for some pollen always remains between the stylar branches.

After fertilization the disk-florets assume a discoloured orange hue, and ulti-
mately become brown. Now and then heads devoid of a ray occur. MacLeod
says that these contain ten rather large disk-florets.

Visrrors.—I observed the following at Kiel (‘ Bl. u. Insekt. a. d. nordfr. Ins.,’
p- 1573 ‘Weit. Beob. ii. BI. u. Insekt. a. d. nordfr. Ins.,’ p. 235).—

A. Diptera. (c) Muscidae: all skg.: 1. Anthomyia sp. 9; 2. Aricia obscurata
Mg.; 3- Dolichopus sp.9; 4. D.sp.6; 5. Lucilia caesar Z.; 6. L. sp.; 7. Musca
corvina Fal/.; 8. Platycephala planifrons #.; 9. Pollenia rudis #.; 10. Scatophaga
litorea Fal/.; 11. S. merdaria #.; 12. S. stercoraria Z.; 13. Siphona cristata
F.; 14. Very small sp. (4) Syrphidae: 15. Melithreptus taeniatus Ag. 9, skg.;
16. Syrphus corollae #. 9, do. B. Hymenoptera. Apidae: 17. Apis mellifica,
L. ¥, skg. and po-cltg.; 18. Bombus lapidarius Z., do.

Willis gives the following for the neighbourhood of the south coast of Scotland
(‘ Fis. and Insects in Gt. Britain,’ Part 1).—

A. Coleoptera. Mitidulidae: 1. Meligethes aeneus /”, freq., skg. and po-dvg.
B. Diptera. (a) Muscidae: 2. Anthomyia radicum Z., freq., skg. and po-dvg.; 3. A.
sp., skg.; 4. Hyetodesia incana W., skg. and po-dvg.; 5. Lucilia cornicina /., skg. ;
6. Onesia sepulcralis M/g., do. ; 7. Scatophaga stercoraria Z., freq., skg. ; 8. Tephritis
vespertina Loew, skg. (6) Syrphidae: 9g. Eristalis aeneus Scop., freq., skg.; 10. E.
horticola Deg., skg.; 11. E. tenax Z., freq., skg.; 12. Platycheirus manicatus JZg.,
skg. C. Hymenoptera. Apzdae: all skg.: 13. Apis mellifica Z.; 14. Bombus
agrorum Z.; 15. B. lapidarius Z.; 16. B. pratorum Z.; 17. B. terresterZ. D. Lepi-
doptera. Rhopalocera: 18. Polyommatus phlaeas Z.

The following were recorded by the authorities, and for the localities stated.—

Leege (Juist), the Noctuid Hydroecia nictitans Bk. MacLeod (Belgium), Apis
and some of the smaller bees. Scott-Elliot (Dumfriesshire), 5 humble-bees, 6 hover-
flies, 12 Muscids, a Lepidopterid, and the beetle Meligethes (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 90).

1327. A. Amellus L. (Herm. Miiller, ‘ Fertilisation, p. 322; Kirchner, ‘ Bei-
trage,’ pp. 63-4; Loew, ‘ Bliitenbiol. Floristik,’ p. 258.)—The vanilla-scented florets
of this species are aggregated into heads of about 35 mm. diameter. Kirchner states
that each of these contains twenty lilac-coloured female ray-florets, and twice as
many golden-yellow hermaphrodite disk-florets. The former possess a corolla-tube
2mm. long, with a tongue about 13 mm. in length; the two branches of the blue
style are divergent. The corolla-tube of the disk-florets is 2}—3 mm. long, expanding
into a bell 3 mm. in length. The style grows out of the anther-cylinder for about
3 mm. above the bell. Its branches are so curved that they turn their concave
stigmatic surfaces towards each other, and meet at the tip; their outer sides subse-
quently become more strongly convex so that the branches curve past each other.

VisiTors.—Herm. Miiller noticed a hover-fly (Eristalis arbustorum Z., po-dvg.)
in Thuringia.

COMPOSITAE 583

Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Echinomyia fera Z. (4) Syr-
phidae: 3. Eristalis arbustorum Z.; 4. E. nemorum Z.; 5. Syrphus corollae F.
B. Hymenoptera. (a) Apzdae: 6. Halictus cylindricus /. 9, skg. (4) Sphegedae :
7. Ammophila sabulosa Z. (c) Vespzdae: 8. Vespa germanica &. C. Lepidoptera.
Rhopalocera: 9. Pieris brassicae Z., skg. Also on the var. dessarabicus Bernh.
A. Diptera. (2) Muscidae: 1. Sepsis annulipes JZg., resting on a ray-floret. (4)
Syrphidae: 2. Eristalis nemorum Z.; 3. Helophilus floreus Z.; 4. Syritta pipiens Z. ;
5. Syrphus ribesii Z. B. Hymenoptera. (2) Apzdae: 6. Bombus terrester L. 9,
skg. (4) Vespidae: 7. Vespa germanica #. C. Lepidoptera. Rhopalocera: 8. Epi-
nephele janira Z., skg.; 9. Vanessa urticae Z., do.

1328. A. Novae-Angliae L. (Knuth, ‘Bliitenbiol. Herbstbeob.’)—In this
species the branching stem attains a height of 14 m., and bears numerous faintly
fragrant heads 34 cm. in diameter. About 4 this breadth is occupied by a hundred
or so yellow disk-florets, and the rest by 80-90 blue ray-florets, which are usually
in several series. The ray-florets are about 2 cm. long and 13 mm. broad. In
the evening and during wet weather these close in so as to cover the disk-florets.
The flower mechanism is the same as in the other species of Aster: the tips of
the stylar branches as they grow through the anther-cylinder brush out the pollen
by means of small sweeping-hairs, and subsequently project so far that their stigmatic
papillae are exposed.

This species is among the latest of the Asters to flower among those I have
observed: even on October 16 (1891) there were numerous heads in the bud-
condition, as well as others that had completely faded. On that day all the visitors
mentioned below were found on the flowers, and the under-sides of all were covered
with pollen.

Visitors.—Knuth records the following.—

A. Hymenoptera. Afzdae: all skg.: 1. Apis mellifica Z., very common
(a few seen as late as October 23); 2. Bombus lapidarius Z.; 3. B. terrester Z. ;
4.B. sp. B. Lepidoptera. Rhopalocera: all skg.: 5. Vanessa io Z.; 6. V. atalanta
L.; 7. Argynnis sp. C. Diptera. (a) Syrphidae: all skg. and po-dvg.: 8. Eristalis
tenax Z., very common (a few seen as late as October 23); 9. E. arbustorum Z. ;
10. Helophilus pendulus Z.; 11. Syritta pipiens Z. (6) Musczdae: 12. Onesia sepul-
cralis M/g.; 13. Sarcophaga sp.; 14. Lucilia cornicina /, freq.; 15. Scatophaga
stercoraria Z., freq. (a few seen as late as October 23); 16. S. merdaria Z.; 17.
Calliphora erythrocephala Mg.; 18. Pollenia rudis /.

1329. A. chinensis L. (=Callistephus chinensis Vees)—The flower mechanism
of this species essentially agrees with that of previous ones.

VistTorS.—The following were recorded by the observers and for the localities
stated.—

Herm. Miiller (Lippstadt), a bee (Coelioxys simplex /Vy/. 9, skg.), a butterfly
(Vanessa urticae Z., skg.), and two hover-flies (Eristalis arbustorum Z., and E.
nemorum Z., skg. and po-dvg.). Schletterer and von Dalla Torre (Tyrol), the bee
Coelioxys elongata Zep. 9. Macchiati made the following noteworthy observations
in Sardinia, Calabria, and Piedmont (Nuovo Giorn. bot. Ital., xvi, 1884; Justs bot.
Jahresber., xii, (1884) 1886, pp. 663-4).—Before the heads expand, an aphis (Aphis
capsellae Kal/enbach) is often found living on the floral axes, where it is ‘ milked’
by many ants. When the plant flowers in autumn, a new generation of aphides
appears, i.e. the winged females, which fly to the open heads. The ants are unable
to follow them there, for the sticky involucral bracts prove an insurmountable

584 ANGIOSPERMAE—DICOTYLEDONES

barrier. This is of some significance in connexion with pollination, since the ants
would frighten away pollinating insects, while the aphides with their sweet excretion
help to attract them, and thus play the part of ‘ living nectaries.’

1330. A. salicifolius Scholl. (=A. salignus Willd., and A. paniculatus Zam.).
(Knuth, ‘ Bliitenbiol. Herbstbeob.’)—The tall much-branched stem bearing numerous
heads serves to render the plants of this species conspicuous from a distance,
attracting a larger number of species of insects than in almost any other aytumnal
form. Each head contains 20-30 ray-florets (15 mm. long) with blue corolla-lobes
(ro mm. long and 2 mm. broad), and 30-40 yellow disk-florets 9 mm. in length,
of which the ovary takes up 2 mm., the contracted part of the corolla-tube 4 mm.,
the nectar-containing bell 2 mm. (this is 4 mm. in diameter), and the corolla-lobes
1mm. The flower mechanism agrees completely with those of Aster Tripolium Z.
and A. Amellus Z., except that the diameter of the bell is larger, and the nectar
is therefore readily accessible even to insects with a stout proboscis or thick tongue.
Ludwig states that the disk-florets of the older heads assume a bright red colour, as
do those of A. parviflorus Mees (Justs bot. Jahresber., Leipzig, xiv, (1886) 1888, p. 806).

Visttors.—I observed the following. —

A. Diptera. (a) Musczdae: all skg.: 1. Anthomyia sp.; 2. Calliphora vomi-
toria Z.; 3. Lucilia caesar Z.; 4. L. cornicina #.; 5. Pollenia vespillo #.; 6. Sepsis
cynipsea ZL. (5) Syrphidae: all skg. and po-dvg.: 7. Eristalis arbustorum Z.; 8. E.
nemorum Z.; 9g. Helophilus floreus Z.; 10. H. pendulus Z.; 11. Melanostoma
gracilis 2g. B. Hymenoptera. 12. Bombus terrester Z., skg. C. Lepidoptera.
13. Vanessa io Z., skg.

Loew saw a hover-fly (Helophilus pendulus Z. 9 and 4) in the Berlin Botanic
Garden.

1331. A. abbreviatus Nees.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Musctdae: 1. Anthomyia sp.; 2. Echinomyia fera Z.;
3. Sarcophaga carnaria LZ. (4) Syrphidae: 4. Eristalis nemorum Z.; 5. Syritta
pipiens Z.; 6. Syrphus ribesii Z. B. Hymenoptera. <Apzdae: 4. Apis mellifica
L. ¥, skg. and po-dvg.

1332. A, azureus Lindl.— :

Visitors.—Loew observed the following Diptera in the Berlin Botanic Garden.—

(2) Muscidae: 1. Lucilia caesar Z. (2) Syrphidae: 2. Eristalis aeneus Scop.;
3. E. nemorum Z.; 4. Syrilta pipiens Z.

1333. A. brumalis Nees.—

VisiTors.—Loew observed a bee (Halictus leucozonius Schr. 4, skg.) in the
Berlin Botanic Garden.

1334. A. concinnus Willd.—
Vis1Tors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Echinomyia fera Z.
(2) Syrphidae: 3.Eristalis aeneus Scop.; 4. E. tenax Z.; 5. Melanostoma mellina Z.;
6. Syritta pipiens Z. B. Hymenoptera. (a) Apidae: 7. Apis mellifica Z. ¥, skg.
and po-cltg. (4) Sphegidae: 8. Ammophila sabulosa Z.

1335. A. floribundus Willd.—

Vistrors.—Loew observed the following in the Berlin Botanic Garden.—

COMPOSITAE 585

A. Diptera. (a) Muscidae: 1. Anthomyia sp. (5) Syrphidae: 2. Eristalis
arbustorum Z.; 3. E. nemorum Z.; 4. E. tenax Z.; 5. Syritta pipiens Z.;
6. Syrphus luniger MZ. B. Lepidoptera. Rhopalocera: 7. Pieris brassicae
L., skg.

1336. A. laevis L.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (2) Muscidae: 1. Echinomyia fera Z.; 2. Lucilia caesar LZ.
(6) Syrphidae: 3. Eristalis arbustoram Z.; 4. E. nemorum Z.; 5. Helophilus
floreus Z. B. Hymenoptera. Sphegidae: 6. Oxybelus uniglumis Z.

1337. A. lanceolatus Willd.—

Visttors.—Loew observed 4 Syrphids (1. Eristalis arbustorum Z.; 2. E.
nemorum L.; 3. E. tenax Z.; 4. Syrphus ribesii Z.) and a bee (Prosopis communis
Wyl. 9, skg.) in the Berlin Botanic Garden,

1338. A. Lindleyanus Torr. et Gray.—

Visttors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Echinomyia fera Z.; 2. Pyrellia cadaverina
L.; 3. Sarcophaga carnaria Z. (3) Stratiomyidae: 4. Chrysomyia formosa Scop.
(c) Syrphidae: 5. Eristalis arbustorum Z.; 6. E. nemorum Z.; 7. E. tenax Z.;
8. Helophilus floreus Z.; 9. H. pendulus Z.; 10. H. trivittatus #.; 11. Syritta
pipiens Z. B. Hymenoptera. (a) Apidae: 12. Halictus rubicundus Car. 4, skg.
(4) Vespidae: 13. Vespa crabro Z.¥. C. Lepidoptera. Rhopalocera: 14. Pieris
brassicae Z., skg.

1339. A. Novi Belgii L._—

Visitors.—Loew observed a Muscid (Anthomyia sp.) and a Syrphid (Syritta
pipiens Z.) in the Berlin Botanic Garden.

1340. A, paniculatus Ait.—
VistTors.—Loew observed 2 Muscids (Anthomyia sp. and Echinomyia fera L.)
and a Syrphid (Syritta pipiens Z.) in the Berlin Botanic Garden.

1341. A. paniculatus Ait., var. pubescens.—
Visirors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Pyrellia cadaverina Z.; 2. Sarcophaga carnaria
L. (0) Syrphidae: 3. Eristalis nemorum Z.; 4. Syrphus ribesii Z. B. Hymeno-
ptera. (a) Apidae: 5. Apis mellifica Z. ¥, skg. and po-cltg.; 6. Bombus terrester
L. 8, skg. (4) Sphegidae: 7. Ammophila subulosa Z. (c) Vespidae: 8. Vespa crabro
L. %.

1342. A. phlogifolius Muhl.—

VisiTors.—Loew observed 2 Syrphids (Eristalis nemorum Z., and Syritta pipiens
L.) in the Berlin Botanic Garden.

1343. A. prenanthoides Muhl.—

VisiTors.—Loew observed a Syrphid (Melithreptus menthastri Z.), a bee
(Halictus cylindricus /. , skg.), and a wasp (Vespa germanica /.), in the Berlin
Botanic Garden.

1344. A. sagittifolius Wedem.—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Calliphora vomitoria L.;
3. Lucilia caesar Z.; 4. Sarcophaga carnaria Z. (4) Syrphidae: 5. Eristalis aeneus

586 ANGIOSPERMAE—DICOTYLEDONES

Scop.; 6. E.nemorum Z.; 7. E. tenax Z.; 8. Helophilus floreus Z.; 9, Melithreptus
scriptus Z.; 10. Syritta pipiens Z. B. Hymenoptera. (a) Apidae: 11. Apis
mellifica Z. ¥; 12. Bombus terrester Z. ¥, skg.; 13. Sphecodes gibbus L. 9, do.
(2) Sphegidae: 14. Ammophila sabulosa LZ.

1345. A. sparsiflorus Michx.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Echinomyia fera Z.;
3. Sarcophaga carnaria Z. (6) Syrphidae: 4. Eristalis aeneus Scop.; 5. E. nemorum
£.; 6. Helophilus floreus Z.; 7. Syritta pipiens Z. B. Hymenoptera. Apzdae:
8. Apis mellifica Z. ¥, skg. and po-cltg.; 9. Bombus terrester Z. ¥, skg.; 10.
Halictus sexnotatus XK. 9, do.

1346. A. squarrosulus Nees.—

VistTors.— Loew observed a Muscid (Echinomyia fera Z.), 2 Syrphids (Eristalis
arbustorum Z., and Helophilus floreus Z.), and a wasp (Odynerus parietum LZ. 9 and
4), in the Berlin Botanic Garden.

419. Biotia DC.
1347. B. commixta DC.—
Vistrors.—Loew observed the following Diptera in the Berlin Botanic Garden.—

(2) Muscidae: 1. Anthomyia sp.; 2. Echinomyia fera Z.; 3. Lucilia caesar Z. ;
4. Pyrellia cadaverina Z. (6) Syrphidae: 5. Eristalis arbustorum Z.; 6. E. nemorum
L.; 7. Helophilus floreus Z.; 8. Syritta pipiens Z.; 9. Syrphus ribesii Z.

1348. B. corymbosa DC.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Echinomyia fera Z.; 2. Pyrellia cadaverina ZL.
(6) Syrphidae: 3. Eristalis arbustorum Z.; 4. E. nemorum Z. B. Hymenoptera.
(a) Apidae: 5. Halictus cylindricus #. 4, skg. (4) Spheg¢dae: 6. Ammophila sabulosa
L. (c) Vespidae: 7. Vespa crabro L. ¥.

1349. B. macrophylla DC.—

Visirors.—Loew observed a Muscid (Echinomyia fera Z.), and a butterfly
(Polyommatus phlaeas Z., skg.), in the Berlin Botanic Garden.

1350. B. Schreberi DC.—

VistTors.—'Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Echinomyia fera Z. (6) Syrphidae: 2. Eristalis
nemorum Z.; 3. E.tenax Z. B. Lepidoptera. Rhopalocera: 4. Pieris brassicae
L., skg.; 5. Vanessa urticae Z., do.

420. Galatella Cass.
1351. G. dracunculoides Nees.—

Visitors.—Loew observed 3 Syrphids (1. Eristalis aeneus Scop.; 2. E. nemorum
L.; 3. Helophilus trivittatus #.) in the Berlin Botanic Garden.

1352. G. hyssopifolia Nees.—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Muscidae: 1. Lucilia caesar Z. B. Hymenoptera. (a) Aprdae:
2. Apis mellifica Z. ¥, skg.; 3. Prosopis communis /Vy/. 9, skg. (6) Sphegidae:

COMPOSITAE 587

4. Ammophila sabulosa Z.; 5. Oxybelus quattuordecimnotatus Jur. $; 6. O. uni-
glumis Z. 3.

1353. G. punctata Nees.—

VisiTors.— Loew observed 2 bugs (Aphanus lynceus /., and Eurydema oleraceum
L.), and a butterfly (Pieris brassicae Z., skg.), in the Berlin Botanic Garden.

421. Chrysocoma L.

Ray-florets neuter or absent : flower mechanism otherwise like that of Aster.

1354. C. Linosyris L. (=Aster Linosyris Bernh., Linosyris vulgaris Cass.,
and Galatella Linosyris Rechd. £.).—According to Hermann Miller (‘ Fertilisation,’
pp. 322-4) the golden-yellow heads of this species are arranged in flat-topped
inflorescences. Miiller says that all the florets are alike. The protandrous florets
which are in the first (male) stage expand their corolla-lobes, and are therefore more
conspicuous than those in the second (female) stage, when these lobes are erect.
The stylar branches (13 mm.
long) are beset to about their
centre with a marginal row of
stigmatic papillae on either side,
above which they are expanded
and covered externally and
laterally by sweeping -hairs.
Their tips do not diverge, but
remain permanently apposed ;
otherwise they are separate
as in Aster. It follows that
insects get covered with pollen
on their under-sides, when
visiting florets in the first stage,
and convey this to the stigmas
of florets in the second stage.
As all the florets of a head are
in the same stage, it follows
that many of them may be
pollinated during a single visit. Fi. 106.

Chrysocoma Linosyris, L. (after Herm. Miller).
he (1) Floret in the first (male) stage. (2) Ditto, in the second (female)
Visitors.—Herm. Miiller stage. a, anthers; 7, sweeping-hairs; 1, stigmatic papillae; ov,

observed the following.— ovary ; A, pollen.

A. Diptera. (a) Muscidae: 1. Ocyptera cylindrica /, skg. (3) Syrphidae:
2. Eristalis arbustorum Z., very common, skg. and po-dvg.; 3. E. nemorum Z., do.;
4. Syritta pipiens Z., do. B. Hymenoptera. Apzdae: 5. Halictus albipes &. 4,
very numerous, skg.; 6. H. cylindricus /. 4, freq., skg.; 7. H. flavipes /. 3, skg.;
8. H. nitidiusculus £. 4, in large numbers, skg. C. Lepidoptera. (a) Mocturdae:
9. Plusia gamma Z., skg. (6) Rhopalocera: 10. Lycaena alsus W.V., skg.;
11. Polyommatus dorilis H/n., do.

Loew records the following for the Berlin Botanic Garden—

A. Coleoptera. Coccinellidae: 1. Coccinella impustulata Z. B. Diptera.
(a) Muscidae: 2. Anthomyia sp.; 3. Chloria demandata #.; 4. Echinomyia fera Z. ;

588 ANGIOSPERMAE—DICOTYLEDONES

5- Onesia sepulcralis Mg.; 6. Pyrellia cadaverina Z. (6) Syrphidae: 7. Eristalis
nemorum Z.; 8. E. sepulcralis Z.; 9. E. tenax Z.; 10. Melanostoma mellina Z. ;
11. Melithreptus scriptus Z.; 12. Syrphus balteatus Deg.; 13. S. pyrastri Z.;
14. S. ribesii Z.

422. Bellidiastrum Cass.

Ray-florets white, female; disk-florets yellow, hermaphrodite. The stylar
branches of the disk-florets usually apposed above, covered with sweeping-hairs above,
both externally and internally, and beset with stigmatic papillae on their outer
margins below. Stylar branches of the female ray-florets devoid of sweeping-
hairs, divergent, beset with marginal and apical stigmatic papillae.

1355. B. michelii Cass. (=Doronicum Bellidiastrum Z., Arnica Bellidiastra
All., and Aster Bellidiastrus Scop.), (Herm. Miiller, ‘Alpenblumen,’ pp. 449-5°-)—
This species is gynomonoecious, with protandrous hermaphrodite florets. A head
usually contains considerably more than 100 yellow disk-florets and 40-50 white
ray-florets, making up a surface 30 or more mm. in breadth. The development
of the florets progresses slowly in centrifugal order, so that there is never more
than a narrow ring of disk-florets in bloom. Kerner states that the stigmas of
the female florets become receptive several days before the pollen is ripe in the
neighbouring hermaphrodite ones (Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, I, p. 312).

VisiTors.—Herm. Miiller observed 5 beetles, 20 flies, 2 bees, and 14 Lepidoptera.

423. Bellis L.

Ray-florets white, uniserial, female; disk-florets yellow, bell-shaped, hermaphro-
dite. Stylar branches of the latter short, broadly ovoid, covered externally as far as
their broadest part with sweeping-hairs, beneath which on the outer margin on either
side is a short line of stigmatic papillae. Stylar branches of the female ray-florets
elongated and devoid of sweeping-hairs: stigmatic papillae more numerous than
in the hermaphrodite florets.

1356. B. perennis L. (Sprengel, ‘Entd. Geh., p. 377; Hildebrand, ‘U. d.
Geschlechtsverhalt. b. d. Compositen,’ pp. 23-4, Taf. II, Figs. 11-15; Herm.
Miller, ‘Fertilisation,’ pp. 321-2, ‘ Weit. Beob.,’ III, p. 92; Knuth, ‘BI. u. Insekt.
a. d. nordfr. Ins.” pp. 87, 157; Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II.}—This
species 1s gynomonoecious. Hermann Miiller states that the golden-yellow
hermaphrodite disk-florets are only 1-2 mm. long; the white female ray-florets,
often tinged with red, possess tongues about 5 mm. in length. The diameter
of the heads averages about 16 mm., but some are considerably larger or smaller.
On the North Frisian Islands I observed a breadth of 10 mm. and even less.
The sweeping-hairs on the elongating styles of the disk-florets partly drive the
pollen before them, partly hold it fast, and thus present it to insect visitors.
Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896) says that the stigmas of the
female ray-florets become receptive before the pollen of the disk-florets is ripe.
The latter develop centripetally. The pollen-grains are pale yellow in colour,
rounded, spinose, 21-5 » in diameter. After fertilization has been effected, the stylar
branches retract into the bell. During dull weather and at night the heads close.

COMPOSITAE 589

Visitors.—Herm. Miiller saw 4 Muscids, 2 hover-flies, and 2 Lepidoptera in
the Alps. He gives the following for Westphalia.—

A. Coleoptera. (a) Cerambyctdae: 1. Leptura livida Z., po-dvg. (3) Mit-
dulidae: 2. Meligethes, po-dvg. (c) Ocedemeridae: 3. Oedemera virescens L.
(2) Phalacridae: 4. Olibrus sp. B. Diptera. (a) Empzdae: 5. Empis livida Z.,
very common, skg.; 6. E. opaca /, skg. (4) Muscidae: 47. Lucilia cornicina F&.,
numerous, po-dvg.; 8. Musca corvina /., do.; 9. Scatophaga merdaria /., do.;
10. S. stercoraria Z., do.; 11. Zophomyia tremula Scop., po-dvg. (c) Syrphidae:
12. Ascia podagrica /., po-dvg.; 13. Eristalis arbustorum Z., very common, po-dvg.;
14. E. pertinax Scop, do.; 15. E. sepulcralis Z., do.; 16. E. tenax Z., do.;
17. Melithreptus scriptus Z., po-dvg.; 18. Rhingia rostrata Z., very common, po-
dvg.; 19. Syritta pipiens Z., do., and skg. CC. Hymenoptera. (a) Apidae:
zo. Andrena nitida Fourcr. 9, casually skg.; 21. A. parvula KX. 9, po-cltg.;
22. Apis mellifica Z. ¥, in large numbers, po-cltg.; 23. Halictus albipes /. 9, skg. ;
24. H. cylindricus /. 9, do.; 25 H. minutissimus X. 9, in large numbers, po-cltg. ;
26. Nomada flavoguttata K. , skg.; 27. N. lineola Pz. 4, do.; 28. Osmia rufa Z. 9,
skg. and po-cltg.; 29. Sphecodes gibbus Z. 9, skg. (4) Formicidae: 30. Myrmica
levinodis /Vy/., trying to suck. D. Lepidoptera. (2) Rhopalocera: 31. Coenonympha
pamphilus Z., skg.; 32. Polyommatus dorilis /n., casually skg. (6) Zinerdae:
33- Adela violella Zr. 8, skg.

Burkill observed the following on the Yorkshire coast (‘ Fertlsn. of Spring Fls.’).—

A. Araneidae. 1. Xysticus pini Hahn, lying in wait for prey. B. Coleoptera.
(a) Curculionidae: 2. Apion striatum K. (0) Mitedulidae: 3. Meligethes sp. C.
Diptera. (a) Brbionidae: 4. Dilophus albipennis Ae. (6) Muscidae: 5. An
Ephydrid; 6. Helomyza sp.; 7. Lucilia cornicina #., skg. and po-dvg.; 8. Onesia
cognata Mg., do.; 9. Pollenia rudis; 10. Scatophaga stercoraria Z., skg. and po-dvg. ;
11. Sepsis nigripes Mg., po-dvg. (c) Syrphidae: 12. Eristalis pertinax Scop., po-dvg. ;
13. Melanostoma quadrimaculatum Verral; 14. Syrphus lasiophthalmus Ze//, D.
Hymenoptera. (a) Apidae: 15. Andrena clarkella X.é, skg.; 16. A. gwynana K.9;
17. Bombus terrester L. (4) Ichneumonidae: 18. Ichneumon sp. E. Lepidoptera.
Rhopalocera: 1g. Pieris rapae Z., skg.; 20. Vanessa urticae Z., do.

The following were recorded by the authorities, and for the localities stated.—

Knuth (North Frisian Islands), Apis, 4 hover-flies, 4 Muscids, a butterfly, and the
beetle Meligethes (‘Bliitenbiol. Beob. a. d. nordfr.Ins.,’ p.157); (Helgoland), Diptera.
All skg.: (a) Muscidae: 1. Coelopa frigida Fall.; 2. Homalomyia scalaris &. 4;
3. Lucilia caesar Z.; 4. Minute species. (4) Syrphzdae: 5. Eristalis tenax ZL. (Bot.
Jaarb. Dodonaea, Ghent, viii, 1896, p. 38); (Thuringia), a hover-fly (Melithreptus sp.,
skg.), and 2 Muscids, skg. (Anthomyia sp., and a small sp.). Alfken (Bremen).—
A. Diptera. Muscidae: 1. Lucilia caesar Z., very common; 2. Musca domestica Z.,
one; 3. Pollenia rudis /.; 4. P. vespillo #, rare. B. Hymenoptera. Apzdae:
5. Andrena parvula XK. 9 and 4, freq., skg.; 6. Halictus morio /. 9, skg. and po-cltg.;
7. H. nitidiusculus A. 9, freq., skg.; 8. Nomada flavoguttata A., var. héppneri
Alfken 9 and %, do.; 9. Sphecodes ephippius Z.; 10. S. sp. Loew (Silesia),
a hover-fly (Syritta pipiens Z., skg.), and a bee (Trypetes truncorum Z. 9, po-cltg.)
(‘Beitrage,’ pp. 30-1). Verhoeff (Norderney).— A. Diptera. (a) Empidae: 1.
Hilara quadrivittata J/g. (6) Muscidae: 2. Anthomyia sp.; 3. Onesia floralis 2.-D.
B. Lepidoptera. /uerzdae: 4. Pieris brassicae Z., skg. Krieger (Leipzig), the bee
Halictus nitidiusculus &. Schletterer (Pola), the following Apidae: 5 bees—
1. Andrena carbonaria Z.; 2. A. parvula K.; 3. A. thoracica #.; 4. Halictus
calceatus Scop.; 5. H.malachurus X. Schmiedeknecht (Florence), on the authority of
Piccioli, gives the bee Andrena florentina Mgr. MacLeod (Pyrenees), a bee, a Lepi-
dopterid, 3 hover-flies, and 6 Muscids (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 360) ;
(Flanders), Apis, 9 species of Andrena, 6 species of Halictus, 4 other short-tongued

590 ANGIOSPERMAE—DICOTYLEDONES

bees, 5 hover-flies, 10 Muscids, 5 Lepidoptera, 3 beetles (op. cit., v, 1893, pp. 412-14).
H. de Vries (Netherlands), the bee Halictus leucozonius Schr. (Ned. Kruidk. Arch.,
Nijmegen, 2 ser., 2 deel, 1875). Scott-Elliot (Dumfriesshire), a hover-fly, and 2
Muscids (‘Flora of Dumfriesshire,’ p. 91).

424. Stenactis Cass.

Ray-florets slender, whitish, female, biseriate. Disk-florets bell-shaped yellow,
hermaphrodite. Stylar branches as in Aster.

1357. S. annua Nees (=S. bellidiflora A. &r., Aster annuus Z., and Erigeron
annuus Pers.). (Kirchner, ‘ Beitrage,’ pp. 64—5.)—Kirchner says that in this species
there are about roo ray-florets and numerous disk-florets, together forming a head
15-20 mm. in diameter. The tongues of the ray-florets are 5-6 mm. long; at
the beginning of anthesis they are somewhat recurved, subsequently they spread
out horizontally, and finally, when flowering is over, become erect—as they also
do in the evening. The diameter of the disk-florets is 5-6 mm., their corolla is
24mm. long, and the style projects beyond it another} mm. The stylar branches
are concave towards each other until anthesis is over.

425. Diplopappus Cass.

1358. D. amygdalinus Hook.—

VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella impustulata Z. (0) Curcu-
lonidae: 2. Apion miniatum Germ. B. Diptera. (a) Muscrdae: 3. Anthomyia sp. ;
4. Chloria demandata /.; 5. Echinomyia fera Z.; 6. Graphomyia maculata Scop. ;
4. Pyrellia cadaverina Z. (6) Syrphidae: 8. Eristalis nemorum Z.; 9. E. tenax L.;
10. Helophilus floreus Z.; 11. Syritta pipiens Z,; 12. Syrphus balteatus Deg.; 13. S.
corollae #.; 14. S. ribesii Z. C. Hymenoptera. (a) Apzdae: 15. Apis mellifica
vey p skg. and po-cltg.; 16. Bombus terrester Z. 9, skg.; 17. Halictus cylindricus
F.%, skg.; 18. Prosopis communis /Vy/. 9 and 4, very numerous, skg.; 19. P. con-
fusa Wyl. 9, skg.; 20. Sphecodes ephippius LZ. 6, skg. (4) Ichneumonidae: 21. Foenus
sp. (c) Sphegrdae: 22. Cerceris arenaria Z.9; 23. Crabro vexillatus Pz.9; 24.
Oxybelus bipunctatus Ol’v.9 and 6; 25. O. quattuordecimnotatus Jur.é; 26. O.
uniglumis Z.9 and é. (d) Vespidae: 27. Eumenes coarctatus Z.; 28. Odynerus
parietum Z. 9 and é; 29. O. parietum Z., var. renimacula Zep. 9; 30. O. trifasciatus
Ff. 9; 31. Vespa crabro Z.¥. D. Lepidoptera. Rhopalocera: 32. Vanessa urticae
L., skg.

F, F. Kohl saw 2 true wasps (Eumenes pomiformis /., and E. coarctatus Z.) in
the Tyrol.

426. Erigeron L.

Ray-florets multiseriate, female; either all ligulate or the inner ones tubular.
Disk-florets hermaphrodite, tubular. The outer side of the stylar branches are beset
with sweeping-hairs in the hermaphrodite florets, but devoid of them in female
ones. These branches only diverge towards the end of anthesis, but are not
recurved. Kerner says that automatic self-pollination is possible during later stages
by contraction of the stylar branches, which are thus brought into contact with
the pollen of the same flower.

COMPOSITAE 591

1359. E. canadensis L. (Kirchner, ‘ Beitrage,’ p. 65.)—In this species the
heads are only 5 mm. long, 3 mm. thick, and 3 mm. broad above. The female
ray-florets are very numerous. Kirchner gives their length as 3 mm., the
slender filiform whitish tongue, which stands erect, measuring less than 1 mm.
The disk-florets (3 mm. long) are all hermaphrodite, slender, tubular, and coloured
yellow above. Automatic self-pollination does not appear to take place, despite the
inconspicuousness of the heads (but c/ Kerner’s statement, quoted above).

Visirors.—Schenck noticed the bee Halictus pauxillus Schenck in Nassau.

1360. E. alpinus L. (Herm. Miiller, ‘Alpenblumen,’ pp. 445-7.)—This species
is gynomonoecious, with two forms of female florets. The yellow disk is 5-7 mm.
in diameter, and surrounded by a ray of slender reddish-lilac florets, with tongues
5 mm. long. In these heads there are three kinds of florets:—(z) fertile female
ray-florets, with tongues which enhance conspicuousness; (2) fertile female florets,
devoid of tongues, between the margin and centre of the disk, serving only for
the production of fruits; (3) hermaphrodite florets in the middle of the head, pro-
ducing nectar and pollen, and with stigmas adapted for crossing, and probably,
in the absence of insect visitors, for. automatic self-pollination. Kerner says that
the stigmas of the female florets become receptive a few days before pollen is
produced in the hermaphrodite ones of the same inflorescence.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller, a fly and 2 Lepidoptera. MacLeod (Pyrenees), a Lepidopterid
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 359). Lindman (Dovrefjeld), a Lepido-
pterid.

1361. E. compositus Pursh. (Abromeit, ‘Bot. Ergeb. von Drygalski’s Groén-
landsexped.,’ pp. 65-7.)—This species is native to Greenland, arctic America, and
the highest parts of the Rocky Mountains. Abromeit says that there are three
varieties, in which the rose-coloured to pale-violet ray-florets are, respectively, 12 mm.
long (var. grandiflorus Hook.), 6-7 mm. long (var. drevzradzafus), and absent (var.
discoideus A. Gr.).

1362. E. acris L. (Kirchner, ‘Beitrage,’ p. 65.)—In this species the heads
are 8-10 mm. in diameter. Kirchner states that the female florets agree with
Hermann Miiller’s description of E. alpinus. The 30—40 female ray-florets possess
a tube 3-4 mm. long, and a slender lilac-coloured tongue of the same length. The
outer part of the disk is occupied by a great number of other female florets, white
in colour, and devoid of tongues. In the middle of the head there are 6-12 or
more yellow hermaphrodite florets, of which the stylar branches ultimately diverge.
When flowering is over the hermaphrodite florets assume a dirty dark-red colour.

1363. E. uniflorus L. (Herm. Miiller, ‘ Alpenblumen,’ p. 447.)—This species
is gynomonoecious, with only one kind of female floret. The yellow disk is not
more than 3-4 mm. in diameter; but the white or bright-red tongues of the
numerous ray-florets extend this to a surface 8-15 mm. across.

The species is widely distributed in the arctic and boreal regions. Abromeit
states that there are several sub-species in Greenland (e.g. E. pulchellus /7rzes, and
E. eriocephalus /. Vai), these being distinguished by the purple-red colour of the

.

592 ANGIOSPERMAE—DICOTYLEDONES

involucre, the character of the investment of hairs, and so forth (‘ Bot. Ergeb. von
Drygalski’s Grénlandsexped.,’ pp. 67-8). Andersson and Hesselman say that in
Spitzbergen the flowering season lasts from the beginning of July till the first half
of September, normal pollen being produced (‘ Bidrag till Kanned. om Spetsbergens
o. Beeren Eil. Karlvaxtflora,’ p. 10). Ekstam gives the diameter of the heads as
10-13 mm., and describes the florets as odourless, those of the ray being white
or violet in colour (‘Bliitenbiol. Beob. a. Spitzbergen,’ p. 6). The setting of fruits
was observed by Kjellman in 1872-3 (according to Ekstam, op. cit.), and by
T. M. Fries on Sept. 9, 1868 (Andersson and Hesselman, op. cit.).

Visttors.—H. Miiller observed a beetle, a fly, a Hymenopterid, and ro Lepi-
doptera. Ekstam (20. 7.’97) noticed an undetermined fly in Spitzbergen.

1364. E. speciosus DC.—

Visttors.—Loew saw the hover-fly Eristalis arbustorum Z. in the Berlin Botanic
Garden.

1365. E. Villarsii Bell. (Kirchner, ‘Beitrage,’ p. 66.)—According to Kirchner,
who investigated this species at Zermatt, the heads agree essentially in their mechanism
with those of E. alpinus.—About 1oo female ray-florets, with their lilac-coloured
tongues (3 mm. long), extend the diameter of the open inflorescence to about
15 mm. Within these come tubular female disk-florets in one or more series,
while the middle of the disk is occupied by hermaphrodite florets, varying in number,
and sometimes reduced to one.

427. Solidago L.

The mechanism of the disk-florets is similar to that described for Chrysocoma.
Kerner states that ray-florets, as in Aster, are geitonogamously pollinated by the
disk-florets.

1366. S. Virgaurea L. (Hildebrand, ‘U. d. Geschlechtsverhilt. u. d. Com-
positen,’ pp. 22-3, Taf. II, Figs. 7-10; Warnstorf, Verh. bot. Ver., Berlin, xxxviii,
1896; Herm. Miiller, ‘ Fertilisation,’ pp. 320-1; Knuth, ‘ Bliitenbiol. Herbstbeob.,’
‘Bloemenbiol. Bijdragen.’)—This species is gynomonoecious. The tongues of the
golden-yellow female ray-florets are 5-7 mm. long; the stylar branches are almost
devoid of sweeping-hairs, and are beset marginally and internally with stigmatic
papillae. The diameter of the heads is 14-19 mm. Warnstorf states that the
stigmas of the female ray-florets, and those of the outer row of hermaphrodite
disk-florets become mature almost simultaneously. In the absence of insect visitors,
pollen from the little heaps on the upper parts of the stylar branches may easily
reach the margins of their lower stigmatic regions, and effect self-pollination. The
pollen-grains are yellow in colour, rounded or ellipsoidal, covered with coarse
spinose tubercles, up to 31 » long and 23 p broad.

Visitors.—I observed the following.—

A. Diptera. (a) Muscidae: 1. Lucilia caesar Z., skg.; 2. Musca domestica Z.,
do. (4) Syrphidae: all po-dvg.: 3. Eristalis arbustorum Z.; 4. E. nemorum Z.;
5. E. pertinax Scop.; 6. E. tenax Z.; 7. Syritta pipiens Z.; 8. Syrphus sp.
B. Hymenoptera. Afidae: g. Apis mellifica Z. ¥, skg. and po-cltg.; 10. Bombus
lapidarius Z. 9, do.; 11. B. terrester Z. ¥, do. C. Lepidoptera. Rhopalocera:
12. Epinephele janira Z., skg.

COMPOSITAE 593

Herm. Miller gives the following.—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z., freq., po-dvg.; 2. E.nemorum
ZL.,do. B. Hymenoptera. Apidae: 3. Andrena denticulata K. 9 and 4, po-cltg.
and skg. (Borgstette); 4. Apis mellifica Z. 8, freq., skg.; 5. Bombus terrester L. 6,
skg.; 6. Psithyrus campestris Z. 3, skg.; 7. P. rupestris Z. 8, skg. C. Lepidoptera.
Rhopalocera: 8. Thecla ilicis Zsp., skg. Also (Alps), a beetle, 22 flies, 6 bees,
a true wasp, and 27 Lepidoptera (‘ Alpenblumen,’ pp. 444-5).

Alfken observed the following bees at Bremen.—

1. Andrena denticulata K. 8; 2. A. gwynana K. 9, 2nd gen.; 3. Bombus
agrorum #.$; 4. B. derhamellus A.é; 5. B. lapidarius Z.é; 6. Halictus flavipes
F.8; 7. H. leucozonius Schr. 6.

Schletterer gives the following bees for the Tyrol.—

1. Andrena parvula X.; 2. Bombus mastrucatus Gers/.; 3. B. terrester L. ;
4. Halictus albipes /.; 5. Psithyrus campestris Pz.; 6. P. globosus £v.; 7. P.
rupestris #.; 8. P. vestalis Pourcr.

The following were recorded by the observers, and for the localities stated.—

Schmiedeknecht (Thuringia), 2 bees (Bombus hypnorum Z. 4, and B. lapidarius
£.%). Hoffer (Steiermark), 2 bees (Bombus lapidarius Z. 9, 9, and 4, and B. hypno-
rum Z. 4). Friese (Baden), 3 bees—1. Halictus calceatus Scop. 3, not infrequent ;
2. H. flavipes /. 4, freq.; 3. Nomada solidaginis Pz. do. Schenck (Nassau), 5
bees—1. Bombus confusus Schenck 9 and$; 2. Halictus calceatus Scop.d; 3. H.
flavipes #. $; 4. H. rubicundus Cér. 3; 5. H. tetrazonius Aly. 6. MacLeod
(Pyrenees), a Syrphid and 3 Muscids (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 359);
(Flanders), Apis, 3 humble-bees, 5 short-tongued bees, an Ichneumonid, a fossorial
wasp, 9 hover-flies, 9 Muscids, and 6 Lepidoptera (op. cit., v, 1893, pp. 414-15).
Lindman (Dovrefjeld), flies, humble-bees and a Lepidopterid. Scott-Elliot (Dumfries-
shire), a parasitic humble-bee and several flies (‘ Flora of Dumfriesshire,’ p. 91).

1367. S. canadensis (L. Herm. Miller, ‘Fertilisation,’ p. 321, ‘ Weit. Beob.,’
III, p. 292; Knuth, ‘ Bloemenbiol. Bijdragen.—

Visttors.—Herm. Miiller observed the following.—

A. Coleoptera. Phalacridae: 1. Phalacrus corruscus Panz., occasional.
B. Diptera. (a) Muscidae: 2. Calliphora erythrocephala M/g.; 3. Lucilia caesar Z. ;
4. L. cornicina #.; 5. Musca corvina /.; 6. M. domestica Z.; 7. Sarcophaga
carnaria Z., po-dvg.; 8. Numerous small species. (6) Syrphidae: 9. Cheilosia
scutellata Fad/.; 10. Eristalis arbustorum Z., freq., po-dvg.; 11. E. nemorum Z., do. ;
12. E. pertinax Scop.; 13. E.tenax Z.; 14. Helophilus floreus Z.; 15. H. pendulus Z. ;
16. Syritta pipiens Z., freq., po-dvg. C. Hymenoptera. (a) Ag:dae: 17. Halictus
cylindricus #. 6, numerous; 18. H. zonulus Sm. 9 and 6, very numerous, skg.,
po-dvg. and po-cltg.; 19. Sphecodes gibbus Z. 9 and 4, very numerous, skg. and
po-dvg. () Formicidae: 20. Formica fusca Z. ¥, very numerous. (c) Sphegidae:
21. Ammophila sabulosa L. 9, skg.; 22. Pompilus niger /. 9, skg. D. Neuroptera.
23. Panorpa communis Z. in large numbers.

The following were recorded by the observers, and for the localities stated.—

Loew (Berlin Botanic Garden), 3 Muscids (1. Anthomyia sp.; 2. Echinomyia
fera Z.; 3. Pyrellia cadaverina Z.) and 3 Syrphids (1. Eristalis arbustorum Z.;
2. E. nemorum Z.; 3. Syrphus ribesii Z.). Knuth, 3 Syrphids, skg. and po-dvg.
(1. Eristalis arbustorum Z.; 2. Helophilus pendulus Z.; 3. Syritta pipiens Z.).
Alfken, the bee Andrena denticulata ., very numerous on one occasion.

DAVIS. 11 eq

594 ANGIOSPERMAE—DICOTYLEDONES

1368. S. ambigua Ait—
Vistrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Echinomyia fera Z.;
3. Pyrellia cadaverina Z. (6) Syrphidae: 4. Eristalis nemorum Z.; 5. Syritta
pipiens Z. B. Hymenoptera. <Afgidac: 6. Bombus terrester Z. 9, skg.

1369. S. bicolor L—

Visttors.—Loew observed 2 Muscids (Calliphora erythrocephala JMg., and
Lucilia caesar Z.) and a Syrphid (Syrphus balteatus Deg.) in the Berlin Botanic
Garden.

1370. S. caesia L.—

Vistrors.—Loew observed 2 Muscids (Anthomyia sp. and Onesia sepulcralis
Mg.) and a Sphegid (Oxybelus uniglumis Z. $) in the Berlin Botanic Garden.

1371. S. carinata Schrad.—

Visrtrors.—Loew saw the butterfly Vanessa c-album JZ., skg., in the Berlin
Botanic Garden.

1372. S. Drummondii Torr. et Gray.—

VistTors.—Loew saw a Sphegid (Ammophila sabulosa Z.) and 2 Vespids
(Eumenes coarctatus Z., and Odynerus parietum Z.), in the Berlin Botanic Garden.

1373. S. fragrans Willd.—
Visttors.—Loew observed the following in the Berlin Botanic_Garden.—

A. Coleoptera. Jagritdae: 1. Lagria hirta Z. B. Diptera. (2) Muscidae:
2. Calliphora vomitoria Z.; 3. Echinomyia fera Z.; 4. Onesia sepulcralis J/g. ;
5. Pyrellia cadaverina Z.; 6. Sarcophaga carnaria Z.; 7. Sarcophila latifrons Fadl.
(4) Syrphidae: 8. Eristalis aeneus Scop.; 9. E. arbustorum Z.; 10. E. nemorum Z.;
11. E. tenax Z.; 12. Helophilus floreus Z.; 13. H. trivittatus #.; 14. Syritta
pipiens Z. C. Hymenoptera. (a) Apzdae: 15. Apis mellifica Z. ¥, skg. and po-cltg. ;
16. Bombus terrester Z. §, skg.; 17. Halictus cylindricus /. 9, skg. (6) Sphegidae :
18. Ammophila sabulosa Z. (c) Vesprdae: 19. Vespa crabro Z.¥; 20. V. germanica /’.

1374. S. glabra Desf.—

VisiTors.— Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1, Anthomyia sp.; 2. Chloria demandata F.;
3. Echinomyia fera Z.; 4. Lucilia caesar Z.; 5. Pyrellia cadaverina Z.; 6. Sar-
cophaga carnaria Z. (4) Syrphrdae: 4. Eristalis arbustorum Z.; 8. E. nemorum Z.;
9. Helophilus floreus Z.; 10. Syritta pipiens Z.; 11. Syrphus balteatus Deg.
B. Hymenoptera. Apidae: all skg.: 12. Halictus cylindricus /. 6; 13. Prosopis
armillata Vy/. 9; 14. P. communis Vy/.9; 15. Sphecodes ephippius Z. 8. (5) Sphegidae :
16. Ammophila sabulosa Z. 9 and 6; 17. Cerceris variabilis Schr.g; 18. Crabro

vexillatus Pz. 9; 19. Oxybelus bipunctatus Oliv.9; 20. O. uniglumis Z. (c) Vespidae :
21. Odynerus parietum Z.; 22. Vespa germanica F.

1375. S. graminifolia Ell.—

Visirors.— Loew observed the Straphylinid beetle Xantholinus linearis Ov. in
the Berlin Botanic Garden.

1376. S. juncea Ait—

Visttors.—Loew observed the Bibionid fly Dilophus vulgaris Zg. in the Berlin
Botanic Garden.—

COMPOSITAE 595

1377. S. lateriflora Ait.—
VisiToRs.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Scarabaerdae: 1.Cetonia aurataZ. B. Diptera. (a2) Muscidae:
2. Calliphora erythrocephala M/g.; 3. Echinomyia fera Z.; 4. Lucilia caesar Z.
(4) Syrphidae: 5. Syritta pipiens Z. C. Hymenoptera. (a) Apidae: 6. Bombus
terrester Z.5, skg.; 7. Prosopis communis Wyi. 9, skg. (5) Sphegidae: 8. Ammophila
sabulosa Z. (c) Vespidae: 9. Vespa germanica F. D. Lepidoptera. Rhopalocera:
10. Epinephile janira Z.

1378. S. latifolia L.—

VisiTors.—Loew observed a Muscid (Anthomyia sp.) in the Berlin Botanic
Garden.

1379. S. lithospermifolia Willd.—
VistTors.—Loew observed a Muscid (Anthomyia sp.) and a Vespid (Vespa
germanica /’) in the Berlin Botanic Garden.

1380. S. livida Willd.—
VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Muscidae: 1. Anthomyia sp.; 2. Lucilia caesar Z.; 3. Sarcophaga
carnaria Z. B. Hymenoptera. (a) Agzdae: 4. Halictus cylindricus ¥. 3, skg. ;
5. H. rubicundus Cér., 9, skg. and po-clig.; 6. Prosopis communis WVyi. 6, skg.
(6) Sphegidae: 7. Crabro vexillatus Pz. 9; 8. Oxybelus uniglumis Z. 6.

1381. S. missouriensis Nutt.—

VistTors.—Loew observed 4 Syrphids in the Berlin Botanic Garden—1. Eristalis
arbustorum Z.; 2. E. nemorum Z.; 3. E. tenax Z.; 4. Helophilus floreus Z.

1382. S. ohiohensis Riddell.—

Visrrors.—Loew observed a Muscid (Anthomyia sp.) in the Berlin Botanic
Garden.

1383. S. Ridellii Frank.—

Vistrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Spilogaster urbana JZ.
(4) Syrphidae: 3. Syritta pipiens Z.; 4. Syrphus balteatus Deg. B. Hymenoptera.
(a) Apidae: 5. Halictus cylindricus #8, skg.; 6. Prosopis communis MVy/. 4, skg.
(8) Vespidae: 7. Vespa germanica /.

1384. S. rigida L.—

Visrrors.— Loew observed the following Diptera in the Berlin Botanic Garden.—

(a) Muscidae: 1. Anthomyia sp. (4) Syrphidae: 2. Eristalis arbustorum Z. ;
3. E.nemorum Z.; 4. Helophilus floreus Z.; 5. Syritta pipiens Z.

1385. S. ulmifolia Muhl.—

Vistrors.—Loew observed a Muscid (Anthomyia sp.) and a Vespid (Vespa
germanica /.) in the Berlin Botanic Garden.

428. Micropus L.
Kerner states that there are purely female florets in addition to pseudo-
hermaphrodite pollen-florets, but none which are truly hermaphrodite.
Qqz2z

596 ANGIOSPERMAE—DICOTYLEDONES

429. Telekia Baumg.

Clavate end of the style with sweeping-hairs; stigmatic papillae occupying
a longitudinal groove on the inner side of each branch.

1386. T. speciosa Baumg. (Hildebrand, ‘U. d. Geschlechtsverhalt. b. d.
Compositen,’ pp. 24-5, Taf. II, Figs. 16-17; Kerner, ‘ Nat. Hist. Pl.,’ Eng. Ed. 1, II,
pp. 191, 322.)—In this species the disk-florets are at first yellow in colour, but
afterwards turn brown. Kerner states that the common receptacle is flat to begin
with, but becomes raised in the course of anthesis, so that the receptive stigmas of
the outer florets are brought into the line of fall of the pollen of the inner ones.

430. Buphthalmum L.

1387. B. salicifolium L. (=B. grandiflorum Z.).—As No. 1386.
Visttors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miller (Alps), 3 flies, 6 Hymenoptera, and 6 Lepidoptera (‘Alpenblumen,’
p- 444). Schiner (Austria), the Bombyliid Exoprosopa cleomene £gg., and the
Conopid Myopa variegata J7g., occasional.

431. Dahlia Cav.

The stylar branches are covered externally with sweeping-hairs, extending from
about half-way down to their tips, and longest in the middle. Stigmatic papillae
in two marginal rows.

1388. D. variabilis Desf. (Hildebrand, ‘U. d. Geschlechtsverhalt. b. d.
Compositen,’ pp. 19-20, Taf. I, Figs. 26-9.)—The female ray-florets of this species
usually possess vestigial stamens.

VisiTors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (on unfertilized heads in the Botanic Garden of the Kiel Ober-Realschule ;
10. 9.97), numerous honey-bees, which while nect-skg. covered their under-sides with
pollen (‘Bliitenbiol. Notizen’). Alfken (Bremen), 4 humble-bees—1. Bombus
agrorum /. 9 and $; 2. B. lapidarius Z. $; 3. B. ruderatus #4; 4. B. terrester LZ. d.

1389. D. Cervantesii Lag.—

VisiTors.—Loew observed a hover-fly (Syrphus balteatus Deg.) and a humble-
bee (Bombus terrester Z. 4), both skg., in the Berlin Botanic Garden.

432. Silphium L.

1390. S. Asteriscus L.—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis tenax Z.; 2. Syritta pipiens Z. B.
Hymenoptera. Apzdae: 3. Apis mellifica Z. ¥, skg. and po-cltg.; 4. Bombus
terrester Z. 6, skg.; 5. Halictus leucozonius Schr. 9, do.; 6. H. sexnotatus K.
9, do. C. Lepidoptera. Rhopalocera: 4. Rhodocera rhamni Z.; 8. Pieris
brassicae Z., skg.

1391. S. connatum L.—

Visttors.—Loew observed a hover-fly (Eristalis nemorum Z.), a humble-bee
(Bombus terrester Z. 5, skg.), and a butterfly (Pieris brassicae Z., skg.) in the Berlin
Botanic Garden.

COMPOSITAE 597
1392. S. dentatum Ell.—

VisrTors.—Loew observed the hover-fly Eristalis tenax Z. in the Berlin Botanic
Garden,

1393. S. erythrocaulon Bernh.—
VistTors.—Loew observed the following in the Berlin Botanic Garden.

A. Diptera. Syrphidae: 1. Eristalis nemorum Z.; 2. E. tenax Z.; 3. Meli-
threptus scriptus Z, B. Hymenoptera. Apzdae: 4. Apis mellifica Z. ¥, skg. and
po-cltg.; 5. Bombus terrester Z. 4, skg.

1394. S. gummiferum El]_—

Vistrors.—Loew observed a hover-fly (Eristalis tenax Z.) and a humble-bee
(Bombus terrester Z. 4, skg.) in the Berlin Botanic Garden.

1395. S. perfoliatum L.—Sprengel says that insects creeping into the florets
of this species ‘ push the anthers in front of them into the flower,’ so that the pollen-
covered stylar brushes are caused to protrude (‘ Entd. Geh.,’ pp. 383-4).

Visitors.—Loew observed a hover-fly (Eristalis tenax Z,) and a butterfly (Pieris
brassicae Z., skg.) in the Berlin Botanic Garden.

1396. S. terebinthinaceum L.—

Visttors.—Loew observed 2 hover-flies (Eristalis tenax Z., and Syrphus ribesii
Z.) in the Berlin Botanic Garden.

1397. S. trifoliatum L.—
VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Chrysomelidae: 1. Cassida nebulosa Z. B. Hymenoptera.

2. Apis mellifica Z. §, skg. and po-cltg.; 3. Bombus terrester Z. $, skg.; 4. Psithyrus
vestalis Pourcr. $, do.

433- Inula L.

Ray-florets uniseriate, female; disk-florets hermaphrodite.

Kerner says that many of the Inulas growing side by side in the Black Sea
region—I. Oculus-Christi Z., I. ensifolia Z., I. germanica Z., I. salicina Z. and
so on—during the summer ‘blossom in definite succession, so that one species
always begins to fade when another is in its prime. Each capitulum of these
Inulas consists of tongue-shaped pseudo-hermaphrodite florets on the circumference
and tubular hermaphrodite florets in the centre. The former unfold earlier than the
latter, and for each of these species there is a certain period, if only two days, when
the pollen, brought by insects to the stigmas of the pistillate flowers in the circum-
ference, can only have been obtained from another species, since their own pollen is
not obtainable.’

1398. I. hirta L. (Herm. Miiller, ‘Weit. Beob.,’ III, p. 91, ‘Nat. Hist. Pl.,’
Eng. Ed. 1, I, p. 316.)—Hermann Miller states that in this species about 200
dark-yellow tubular florets, expanding into narrow bells, make up a disk 13-15 mm.
in diameter. They are surrounded by about 50 golden-yellow ray-florets with tongues
15 mm. long. The star-like head thus constituted is 40-45 mm. in diameter. The
tube of the disk-florets is 3-34 mm. in length; the bell 2 mm. long and 1 mm. broad.
As the nectar rises as far as the bell, it is accessible even to very short-tongued

598 ANGIOSPERMAE—DICOTYLEDONES

insects. The stylar branches project 1 mm. from the disk-florets, and diverge at an
angle of 45-60°. :

Vistrors.—Herm. Miiller observed the following in Thuringia.—

A. Coleoptera. Cerambycidae: 1. Strangalia bifasciata Jil/., gnawing the
anthers. B. Diptera. (a) Empidae: 2. Empis sp., skg. (6) Muscedae: 3. Aricia
sp., skg. C. Hymenoptera. (a) Apidae: 4. Coelioxys conoidea Ji. é, skg.;
5. Megachile centuncularis Z., 4, do.; 6. Nomada ruficornis Z. 9, do.; 7. Osmia
spinulosa X. 9, extremely numerous, actively po-cltg.; 8. Stelis breviuscula WVy/. 9,
skg. (5) Zenthredinidae: 9. Tarpa cephalotes /, freq., skg. D. Lepidoptera.
Rhopalocera: 10. Coenonympha pamphilus Z., skg.; 11. Melitaea athalia Feo/z., very
common, skg.; 12. Thecla ilicis Zsp., skg.

1399. I. Helenium L. (Herm. Miller, ‘ Weit. Beob.,’ III, p. 91.)—
Visitors.—Buddeberg observed the following in Nassau.—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z., po-dvg.; 2. Volucella
inanis Z.,do. B.Hymenoptera. Apidae: 3. Anthidium manicatum Z., 6, skg. (?) ;
4. Andrena minutula X. 9, do.; 5. Chelostoma nigricorne JVy/. 6, do.; 6. Coelioxys
rufescens Zep. 9 and %, do.; 7. Epeolus variegatus Z., do.; 8. Halictus leucopus £.
3, do.; 9. H. sexcinctus /. 9 and 4, po-cltg. and skg.; 10. H. tetrazonius Al. 9,
do.; 11. Osmia claviventris Zhoms. 9, do.; 12. Stelis aterrima Pz. 9 and 4, very
numerous, skg.; 13. S. phaeoptera X. 9, occasional, skg.

Handlirsch, on the authority of Assmuss, records the fossorial wasp, Alyson
fuscatus Pz.

1400. I. britannica L. (Herm. Miiller, ‘ Weit. Beob.,’ III, p. 92.)—
Visrtors.—Buddeberg observed the following in Nassau.—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z., po-dvg. B. Hymeno-
ptera. Aprdae: 2. Anthidium manicatum ZL. 3, skg.; 3. Epeolus variegatus L.
g and 6, do.; 4. Panurgus calcaratus Scop, 9 and 4, skg. and po-dvg.

Alfken (Bremen) noticed the parasitic humble-bee Psithyrus vestalis /ourcr. 9,
skg. Loew (Berlin Botanic Garden) records the bug Eurygaster maura Z.

1401. I. salicina L. (Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—In
this species the stigmas of the female ray-florets are still receptive when the outer
hermaphrodite disk-florets open, so that geitonogamy without the help of insects
is easily possible. The stylar branches of the tubular florets are flattened, and
somewhat enlarged apically, so that as long as they remain apposed in the anther-
cylinder they form a clavate terminal expansion by which the pollen is pushed out.
Sweeping-hairs are only present in very small numbers at the extreme tip, and
owing to the peculiar construction of the stylar branches are almost superfluous.
Internally these branches are beset with very short stigmatic papillae, and ultimately
diverge to an angle as much as go°, but without becoming reflexed. Since they
are relatively long, the pollen of younger florets can easily reach the stigmas of
older ones, and thus effect geitonogamy in the absence of insect visitors. The
pollen-grains are yellow in colour, polyhedral, with marginal spinose tubercles, about
22 ms in diameter.

VisiTors.—Loew (Brunswick) observed the moth Zygaena onobrychis S. V.,
skg. (‘ Beitrage,’ p. 50).

COMPOSITAE 599
1402. I, thapsoides DC.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis arbustorum Z.; 2. E. nemorum Z.;
3. Syrphus ribesii Z. B. Hemiptera. 4. Aelia acuminata Z.

1403. I. viscosa Ait. (=Cupularia viscosa Godr. e¢ Gren.).—

Visitors.—Delpino observed Pieris, Vanessa, and other butterflies (‘ Ult. oss.,’
Atti soc. ital. sc. nat., xvi, 1873).

1404. I. ensifolia L.—

Visitors.—Schiner (Austria) observed the fly Myopites inulae v. Roser.

1405. I, Conyza DC. (=Conyza squarrosa Z.).—

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Schenck (Nassau), the leaf-cutting bee Megachile centuncularis Z. Schiner
(Austria), the fly Tephritis zelleri Zoew. Schletterer and von Dalla Torre (Tyrol),
the bee Halictus sexcinctus 7.

434. Pulicaria Gaertn.

Ray-florets uniseriate, female; disk-florets tubular, hermaphrodite. Stylar
branches beset with stigmatic papillae over their entire inner surfaces; the upper
third of their outer surface covered with sweeping-hairs directed obliquely upwards.

1406. P. dysenterica Gaertn. (=Inula dysenterica Z.). (Herm. Miller,
‘Fertilisation,’ pp. 324-6, ‘ Weit. Beob.,’ III, p. 90; Giard, Bot. Jaarb. Dodonaea,
Ghent, ii, 1890, pp. 334-7; Knuth, ‘Bloemenbiol. Bijdragen.’)—This species is
gynomonoecious. Upwards of 600 yellow disk-florets are surrounded by about 100
ray-florets of the same colour. Hermann Miller states that the corolla-tube is about
4 mm. long. Only the stylar branches (about 4 mm. in length) project from the
anther-cylinder. These branches spread out horizontally and become recurved,
so that the stigmas occupy the place where pollen was to be found in the first (male)
stage. Hence insect visitors creeping about on the head pollinate simultaneously
numerous florets in the female stage. The triangular valves forming the upper
ends of the anthers are fringed with hairs, which are much longer and thicker than
the sweeping-hairs: they hold the pollen pressed out of the anther-cylinder.

At Boulogne (Pas-de-Calais) Giard (1877) found several plants bearing abnormal
heads, some female and devoid of a ray, others male with an imperfect one. The
florets contained either vestigial stamens or vestigial pistils, and were degenerate in
other respects. For ten years in succession Giard removed all the normal stocks
from this station, and so converted the originally gynomonoecious plant into a purely
dioecious one.

Visrrors.—I only saw the butterfly Vanessa urticae Z., skg.

Herm. Miller observed the following.—

A. Coleoptera. Chrysomelidae: 1. Cassida murraea Z., not infrequent, creeping

over the heads. B. Diptera. Syrphzdae: 2. Eristalis arbustorum Z., very common,
po-dvg.; 3. E. sepulcralis Z.,do.; 4. Melithreptus scriptus Z., po-dvg.; 5. Syritta

600 ANGIOSPERMAE—DICOTYLEDONES

pipiens Z., do. C. Hymenoptera. Apidae: 6. Halictus albipes F. 4, skg.;
4. H. cylindricus /. 8, do.; 8. H. longulus Sm. 8, do.; 9. H. maculatus Sm. 8, do. ;
10. H. nitidus Schenck 5, do.; 11. Heriades truncorum Z. 9 and 4, very numerous.
D. Lepidoptera. Rhopalocera. 12. Hesperia thaumas H/n., skg.; 13. Lycaena
sp.; 14. Polyommatus dorilis H/n.

MacLeod (Flanders) noticed the bee Halictus, 6 hover-flies, and 3 Muscids
(Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 416).

Burkill observed the following on the East coast of Scotland (‘Fls. and Insects
in Gt. Britain, Part 1).—

A. Coleoptera. MNitidulidac: 1. Meligethes aeneus /, dvg. and almost com-
pletely covered with pollen; 2. M. obscurus Z7.; 3. M. picipes Sturm.; 4. M.
viridescens /., as M. aeneus. B. Diptera. (a) Musczdae: 5. Anthomyia brevi-
cornis Ze//., po-dvg.; 6. A. radicum Z., very common, po-dvg.; 7. Calliphora
erythrocephala M/g.; 8. Coelopa sp., po-dvg.; 9. Drymeia hamata fall., skg. and
po-dvg.; 10. Hylemyia strigosa #.; 11. Lucilia cornicina #.; 12. Morellia sp.;
13. Phorbia lactucae Bouché, po-dvg.; 14. Scatophaga stercoraria Z.; 15. Siphona
geniculata Deg., freq., skg. and po-dvg. (4) SyrpAcdae: 16. Eristalis arbustorum Z. ;
17. E. pertinax Scop., skg.; 18. E. tenax Z., do.; 19. Platycheirus albimanus /.;
20. P. manicatus M/g.; 21. Sphaerophoria scripta Z.; 22. Syritta pipiens Z.; 23.
Syrphus ribesii Z. C. Hymenoptera. (a) Apidae: 24. Bombus lapidarius Z..,
skg. (4) Ichneumonidae: 25. 2 undetermined species. D. Lepidoptera. Jficro-
lepidoptera: 26. Plutella xylostella Z., skg.; 27. Simaéthis fabriciana Szeph., do.
E. Thysanoptera. 28. Thrips sp.

(c) Tribe Senectonideae Less.
Styles of the hermaphrodite florets cylindrical, with linear penicillate branches
fulcrate at the tip.
435. Xanthium L.

Male and female florets in different heads on the same stock. Male florets
with vestigial styles (Kirchner). Female florets open much earlier than male ones
(Kerner).

436. Helenium L.
1407. H. autumnale L.—
Vis1Tors.— Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis nemorum Z.; 2. E. tenax Z.; 3. Helo-
philus floreus Z.; 4. Syrphus balteatus Deg. B. Hymenoptera. Apzdae: 5. Apis
mellifica Z. ¥, skg. and po-cltg.; 6. Halictus cylindricus /. 4, skg.; 7. H. rubicundus
Chr. 9, skg. and po-cltg.; 8. Heriades truncorum Z. 9, po-cltg. C. Lepidoptera.
Rhopalocera: g. Pieris brassicae Z., skg.; 10. P. rapae L., skg.

1408. H. californicum Link—

Vistrors.—Loew noticed Apis, skg. and po-cltg., in the Berlin Botanic Garden.
1409. H. decurrens Vatke.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis tenax Z.; 2. Syritta pipiens Z. B.
Hymenoptera. Apzdae: 3. Apis mellifica Z. ¥, skg. and po-cltg.; 4. Heriades
truncorum JL. 9, po-cltg.

COMPOSITAE 601

437. Bidens L.

Ray-florets sometimes absent in German species; ligulate, neuter, coloured like
the hermaphrodite tubular disk-florets. Stylar branches with lancet-shaped tips,

covered externally by strong sweeping-hairs, and beset with numerous stigmatic
papillae internally.

1410. B. tripartita L. (Hildebrand, ‘U. d. Geschlechtsverhalt. b. d. Compositen,’
p. 67, Taf. I, Figs. 30-1; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 416-17;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 88, 157.)—Ray-florets are almost always
wanting in this species. The diameter of the head is at most 1 cm. The sweeping-
hairs at the tip of the style are moderately long, the next below are shorter, and the
lowest are the longest. They brush out the 1. 2.
pollen from the anther-cylinder, which after- ¢s
wards becomes completely retracted into the A /
corolla-tube. The stylar branches then unfold WZ
their papillose inner sides; the corolla-lobes
so far spread out at the same time resume
a nearly vertical position; and the pappus
bristles, with deflexed prickles on their outer
sides, spread out so that the diameter of the FIG. 197. Bidens tripartita, L. (from
: nature). (1) Tipof style, with sweeping-hairs,
head is ultimately increased to 24 cm. The inopeningfloret. (2) Ditto, of floret in female
florets are at first yellow, but towards the end of —*t2g¢: stylar branches recurved, and beset with

: i stigmatic papillae internally. (« 20.)
anthesis assume an inconspicuous brown colour.

Vistrors.—MacLeod (Belgium) saw.2 bees (Bombus and Andrena).
I noticed at Kiel 3 hover-flies (1. Melithreptus taeniatus 4Zg.; 2. Platycheirus
manicatus M/v.; 3. Syrphus balteatus Deg.), and a bug (Calocoris bipunctatus f°).

wu. B. cernua L. (Herm. Miiller, ‘ Weit. Beob.,’ III, p. 88; MacLeod,
op. cit.; Knuth, ‘Bloemenbiol. Bijdragen,’ ‘ Bliitenbiol. Herbstbeob.’)—In this
species about 100 yellow florets are aggregated into a head. Hermann Miller
states that each disk-floret has a corolla-tube about 14 mm. long, with a bell
of about the same length and 1 mm. in width. In the first stage of anthesis
the pollen-covered anther-cylinder projects about 1 mm. from the bell, and in the
second stage the stylar branches (also 1 mm. long) diverge. The mechanism of
these branches agrees with that described for B. tripartita. The stigmatic papillae
are so broad that pollen-grains of the same flower may readily remain adhering
on the margin, so that automatic self-pollination is possible, as it is also in the last
species. There are three varieties.—

(a) discordea Wimm.; ray-florets absent ;

(8) radiata DC. (=Coreopsis Bidens LZ.) ; large ray-florets ;

(c) minima L. (=B. minima Z.); plant low (stem only 4-10 cm. high);
usually only one small head.

Vistrors.—Herm. Miiller saw only the honey-bee at Lippstadt.
At Kiel I observed a humble-bee (Bombus terrester Z. 9, skg.) and a Muscid
(Lucilia cornicina F., po-dvg.).

602 ANGIOSPERMAE—DICOTYLEDONES

438. Actinomeris Nutt.
1412. A. helianthoides Nutt.—

Visitors.—Loew saw the bee Bombus lapidarius Z. 4, skg., in the Berlin Botanic
Garden.

439. Boltonia L’Hérit.
1413. B. glastifolia L’ Hérit—
Visttors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp.; 2. Lucilia sylvarum Mg.;
3. Pyrellia cadaverina Z. (4) Syrphidae; 4. Eristalis arbustorum Z.; 5. E. nemorum
L.; 6. E. tenax Z.; 7. Helophilus floreus Z.; 8. Melithreptus scriptus Z.; 9. Syritta
pipiens Z.; 10. Syrphus ribesii Z. B. Hymenoptera. (c) Apidae: 11. Apis
mellifica Z., skg. (6) Vespidae: 12. Vespa germanica F. C. Lepidoptera.
Rhopalocera: 13. Vanessa urticae L., skg.

440. Helianthus L.
Ray-florets neuter; disk-florets hermaphrodite.

1414. H. annuus L. (Sprengel, ‘Entd. Geh.,’ pp. 378-80.) In this species

the diameter of the heads is as much as 3m. The ray-florets are yellow in colour,
the disk-florets are brown.

Visttors.—Alfken observed the following bees at Bremen.—

1. Bombus agrorum ¥.; 2. B. hortorum Z. 9, skg.; 3. B. hypnorum L. 3;
4. B. lapidarius Z. 8; 5. B. pomorum Pz. $; 6. B. ruderatus F. 4, skg.; 7. B.
sylvarum LZ. 9 and é, do.; 8. B. terrester Z. ¥; 9. Coelioxys acuminata MVyi. 9, do. ;
10. Megachile circumcincta Z. 9; 11. Psithyrus campestris Pz. ; 12. P. rupestris
fF. %. Also (Bozen), the bee Xylocopa violacea Z. 9, occasional, skg.; and the
Pentatomid Carpocoris baccarum L., freq., skg. The former was seen flying from
one head to another as late as 10 p.m.

The following were recorded by the observers, and for the localities stated.—

Knuth, 3 bees, skg. and po-cltg. (1. Apis mellifica Z. §; 2. Bombus lapidarius
L.; 3. B. terrester Z. 9 and ¥), 2 flies, po-dvg. (Eristalis sp. and Pollenia rudis 7),
2 Hemiptera (Calocoris bipunctatus #. and Lygus pabulinus Z.), and an earwig
(Forficula auricularia Z., dvg. the florets). Sprengel also records the earwig as
a common visitor. Sickmann (Osnabriick), the Pompilid Pseudagenia carbonaria
Scop., rare. Schletterer (Tyrol), the bee Halictus alternans J//.

1415. H. multiflorus L.—In this species the pollen projects from the anther-
cylinder during the first stage of anthesis, and the stylar branches during the second
stage.

2 VisiTors.—The following were recorded by the observers named.—

Delpino the bee Heriades truncorum Z, Herm. Miiller, 2 bees (Megachile
centuncularis Z., po-cltg., and Halictus zonulus Sm. 9, skg. and po-dvg., and
3 hover-flies), skg. and po-dvg.'(1. Eristalis tenax Z.; 2. Syrphus pyrasti L. ;
3. S. ribesii Z.).

14416. H. atrorubens L.—

Visrtors—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis nemorum Z.; 2. E. tenax L.; 3.

Syrphus balteatus Deg. B. Hymenoptera. Apidae: 4. Apis mellifica LZ. ¥,

skg. and po-cltg.; 5. Bombus terrester Z. 8, skg.; 6. Psithyrus vestalis Fourcr.
8, do.

COMPOSITAE 603

1417. H. decapetalus Darl.—
VistTors.—Loew observed the following in the Berlin Botanic Garden.—

_A. Diptera. Syrphidae: 1. Syrphus balteatus Deg. B. Hymenoptera.
Apidae: 2. Apis mellifica LZ. ¥, skg. and po-cltg.; 3. Bombus pratorum JZ. 4, skg.;
4. B. terrester Z. ¥, do.

1418. H. divaricatus /.—
VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalisnemorum’Z.; 2. Melithreptus scriptus Z. ;

3. Syritta pipiens Z.; 4. Syrphus corollae #/ B. Hymenoptera. Apidae: 5. Apis

cane L.%, skg. and po-cltg. C. Lepidoptera. Rhopalocera: 6. Pieris brassicae
., Skg.

1419. H. lactiflorus Pers.—

VistTors.— Loew observed the hover-fly Eristalis tenax Z. in the Berlin Botanic
Garden.

1420. H. Maximiliani Schrad.—

Visitors.—Loew observed 2 bees (Apis mellifica Z. ¥, skg. and po-cltg.;
Bombus terrester Z. ¥, skg.) in the Berlin Botanic Garden.

1421. H. mollis Lam.—

VistTors.—Loew observed 2 bees, skg. (Bombus pratorum L. 4, and B. terrester
£. 9 and @) in the Berlin Botanic Garden.—

1422. H. trachelifolius Mill.

Visirors.—Loew observed 2 hover-flies (Eristalis tenax Z. and Helophilus
trivittatus #.) in the Berlin Botanic Garden.

441. Echinacea Moench.

1423. E. purpurea Moench.—

Vistrors.—Loew observed the humble-bee Bombus terrester Z. 8, skg., in the
Berlin Botanic Garden.

442. Heliopsis Pers.

1424. H. laevis Pers.—

VistTors.—Loew observed 3 bees (1. Bombus pratorum Z. 4, skg.; 2. B. terrester
L. , do. ; 3. Halictus sexnotatus K. 9, do.) in the Berlin Botanic Garden.

1425. H. scabra Dun.—

VisiTors.—Loew observed a Muscid (Calliphera vomitoria Z.) and a humble-bee
(Bombus terrester Z, ¥, skg.) in the Berlin Botanic Garden.

1426. H. patula Wender.—

Vistrors.—Schletterer (Tyrol) records the humble-bee Bombus confusus Schenck ;

and von Dalla Torre (Innsbruck Botanic Garden) the bee Trypetes truncorum Z. 9

and é.
443. Chrysostemma Less.

1427. C. tripteris Less.—

VisiTors.—Loew observed a hover-fly (Eristalis tenax Z.) and a butterfly (Pieris
brassicae Z., skg.) in the Berlin Botanic Garden.

604 ANGIOSPERMAE—DICOTYLEDONES

444. Coreopsis L.
1428. C. auriculata L.—

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Echinomyia fera Z.; 2. Graphomyia maculata
Scop. (6) Syrphidae: 3. Eristalis arbustorum Z.; 4. Syritta pipiens Z.; 5. Syrphus
luniger A/g.; 6. S. ribesii Z. B. Hymenoptera. Apidae: 4. Halictus cylindricus
F. 3, skg.

1429. C. lanceolata L.—

Visrrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Echinomyia fera Z. (5) Syrphidae: 2. Eristalis
nemorum Z.; 3. E. tenax Z.; 4. Melithreptus scriptus Z.; 5. Syritta pipiens Z. ;
6. Syrphus balteatus Deg.; 7. S. corollae / B. Lepidoptera. (2) Noctuidae:
8. Plusia triplasia Z., skg. (6) Rhopalocera: 9. Pieris brassicae Z., skg.

445. Rudbeckia L.

Ray-florets elongated, ligulate, neuter; disk-florets hermaphrodite.

1430. R. laciniata L. (Knuth, ‘Bliitenbiol. Herbstbeob.,’ ‘Bloemenbiol.
Bijdragen.’)—

Visttors.—I observed 2 humble-bees, skg. (Bombus lapidarius Z., and B. terrester
L£.), a hover-fly (Syritta pipiens Z.), and a Muscid (Pollenia vespillo /) in the Botanic
Garden of the Kiel Ober-Realschule.

Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis tenax Z.; 2. Helophilus floreus Z. ;
3. Syritta pipiens Z.; 4. Syrphus balteatus Deg.; 5. S. corollae #. B. Hymeno-
ptera. Apidae: 6. Apis mellifica Z. ¥, skg. and po-cltg. ; 7. Bombus terrester Z. g,

skg.; 8. Chelostoma nigricorne WVy/. 9, po-cltg.; 9. Coelioxys elongata Zep. 9, skg. ;
10. Heriadas truncorum Z. 9, po-cltg.; 11. Megachile centuncularis Z., 9, do.

1431. R. speciosa Wender.—

Visirors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Pyrellia cadaverina Z. (6) Syrphidae:
2. Helophilus floreus Z.; 3. Syritta pipiens Z. B. Hymenoptera. (a) Apzdae:
4. Apis mellifica Z. ¥, po-cltg.; 5. Halictus sexnotatus K. 9, skg. (6) Vespzdae:
6. Eumenes coarctatus Z. C. Lepidoptera. Rhopalocera: 47. Pieris brassicae
L., skg.

446. Filago L.

Ray-florets filiform, in two or more series, female; disk-florets tubular
hermaphrodite.

1432. F. minima Fries. (Knuth, ‘ Bloemenbiol. Bijdragen.’)—The heads of this
species are yellowish-white in colour.

?

Vistrors.—I saw a hover-fly (Melanostoma mellina Z.).

447. Antennaria Gaertn.

Dioecious. Styles of the male florets without stigmatic papillae, but densely
covered with sweeping-hairs above; styles of the female florets with few hairs, but
possessing a line of stigmatic papillae internally on each side.

COMPOSITAE 605,

1433. A. dioica Gaertn. (=Gnaphalium dioicum Z.). (Hildebrand, ‘U. d. Ge-
schlechtsverhilt. b. d. Compositen,’ pp. 40-2, Taf. III, Figs. 26-32; Herm. Miller,
‘Alpenblumen,’ p. 436; Lindmann, ‘Bidrag till Kanned. om Skandin. Fjelivaxt.
Blomn. o. Befrukt.’; Kerner, ‘Nat. Hist. Pl.,’ Eng. Ed. 1, IL; Kirchner, ‘ Flora
v. Stuttgart,’ p. 703; Knuth, ‘Bloemenbiol. Bijdragen.’)}—This species is dioecious.
The white or rose-coloured involucral bracts render the heads conspicuous. The
female florets are filiform, the male ones tubular; both kinds contain nectar.
Hildebrand states that the ovary of male florets contains no ovule, while the
style divides into two short blunt branches, devoid of stigmatic papillae; on the
other hand, the whole upper part of the style is covered with sweeping-hairs, those
at the tip being longest. The corolla, tubular below, ends in a bell with somewhat
recurved lobes. ‘The filaments are irritable, and curve when touched. In this way
the anther-cylinder is retracted, so that pollen protrudes from its upper end. The
corolla-tube of female flowers is long and narrow. The style projects beyond it,
and the branches of this are externally covered with short hairs at their tips only,
while their inner surfaces possess a line of stigmatic papillae on either side.

Visirors.—The following were recorded by the observers, and for the localities
stated.—

Lindmann (Dovrefjeld), a few Lepidoptera. Herm. Miiller (Alps), a hover-fly,
a fossorial wasp, and 9 Lepidoptera. Knuth (Tondern), a po-dvg. hover-fly (Eristalis

tenax Z.) and a butterfly, skg. MacLeod (Pyrenees), 2 Muscids (Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 363).

1434. A. margaritacea R. Br. (=Gnaphalium marginatum Z.).—

Vistrors.—I observed (12.9.’97), 2 po-dvg. hover-flies (Eristalis tenax Z., and
E. intricarius Z.), 2 beetles (Coccinella quinquepunctata Z., and C. quattuordecim-
punctata Z.), and Thrips.

1435. A. alpina Gaertn. (=Gnaphalium alpinum Z.).—This northern species
is dioecious. According to Hartman (‘ Handbok i. Skand. Flora,’ p. 7), Laestadius
found male plants in 1842. Vahl, Lange, and Warming do not know of any such,
but on the contrary suppose the species to be parthenogenetic, as it has been
observed to set fruits in various localities,

The specimens collected by Dr. von Drygalski’s Greenland expedition were
all female (Abromeit, ‘ Bot. Ergeb. von Drygalski’s Grénlandsexped.,’ p. 65).

According to Juel (‘ Vergleich. Untersuch. ti. typische u. parthenogenet. Fortpfl.
b. d. Gatt. Antennaria,’ pp. 12-13), male stocks are extremely rare in Lapland and
Norway: their florets usually contain a small quantity of functionless pollen or
none at all. Reproduction is parthenogenetic. Juel (op. cit., p. 14) considers that
the existence of male stocks is due to atavism: but he also thinks it possible
that A. alpina may really be a hybrid, perhaps the result of crossing between
A. dioica Gaertn. and A. monocephala DC.

448. Gnaphalium L.
Ray-florets female, filiform, in several series; disk-florets hermaphrodite and
tubular; the tip of the style covered with tufted sweeping-hairs.

1436. G. Leontopodium L. (=Leontopodium alpinum Cass.). (Herm. Miiller,
‘Alpenblumen,’ pp. 434-6; Kerner, ‘Nat. Hist. Pl,’ Eng. Ed. 1, I1; MacLeod,

606 ANGIOSPERMAE—DICOTYLEDONES

‘Pyreneénbl.,’ p. 363.)—This species is monoecious. Twenty to thirty male disk-
florets with a considerably larger number of female ray-florets are aggregated into
a head 4 mm. in diameter. Conspicuousness is enhanced by the cauline leaves,
which are covered with a thick white hairy coat and surround the corymb of tiny
heads to form a whitish star of 20 to 40 or 50 mm. in diameter.

The ray-florets possess a narrow corolla-tube 24-3 mm. in length, and
secreting no nectar. The style, of which the branches are closely beset with
stigmatic papillae internally, projects 1 mm. from it. The style is covered
externally with short sweeping-hairs for some distance below the point where it
divides. In male flowers the style does not bifurcate, and therefore possesses
no trace of stigmatic papillae. It is in the form of a cylindrical rod covered
with papillose sweeping-hairs at its end, and serving to brush out the pollen
from the anther-cylinder. These pseudo-hermaphrodite male florets possess a

Fic. 198. Gnaphalium Leontopodium, L. (after Herm. Miller). 4. Group of seven heade (nat.
size). B. Female ray-floret without the pappus (x 7). C. Male disk-floret, do. (x 7). D. End of
the style of a male floret, which acts as a brush ( x 80). £. Do. of a female floret (x 80). / sweeping-
hairs; gr, style; ov, ovary ; fo, pollen-grains; s¢, stigmatic papillae.

corolla-tube about 2 mm. in length, expanding into a bell scarcely 1 mm. long,
from which the anthers and style project. Nectar is secreted at the base of the
style. Schroter describes nectar-florets, which resemble the male ones. They
possess a vestigial rudimentary style with quite short sweeping-hairs, but no
stamens (Ber. Schweiz. Bot. Ges., v, 1895, p. 5).

Kerner states that the stigmas of the female florets become receptive several
days before the pollen of the neighbouring pseudo-hermaphrodite male florets is shed.

According to MacLeod, the species appears in the Pyrenees in the sub-alpine
and lowest mountain region, where it possesses a considerably different habit. It is
there more vigorous; the heads are more numerous, and more loosely aggregated ;
and the woolly leaves which surround the entire inflorescence are relatively longer.

Visttors.—These are very few. MacLeod saw a Muscid: Herm. Miiller
observed a beetle, a Muscid, and Thrips.

COMPOSITAE 607

1437. G. luteo-album L. (Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—
Warnstorf says that the numerous narrow, tubular, female ray-florets of this species
mature their stigmatic branches before the few (8-10) hermaphrodite disk-florets open ;
and that they elongate after pollination so as once more to completely enclose the
stigmas. Geitonogamy is only possible between the inner female florets and the outer
hermaphrodite ones. The pollen-grains are yellow in colour, rounded, with long
spines, on an average 25 p in diameter.

Vistrors.—Herm. Miiller observed the following at Lippstadt (‘ Fertilisation,’
P- 324).—

A. Diptera. Muscidae: 1. Lucilia, in large numbers; 2. Pollenia rudis /.,
po-dvg. Syrphidae: 3. Melanostoma mellina Z., po-dvg.; 4. Melithreptus scriptus
Z., do. B. Hymenoptera. (a) Apidae: 5. Halictus quadrinotatus Schenck 8 and 9,
skg.; 6. Sphecodes gibbus Z., 6 and 9, different varieties, also S. ephippius Z., skg.
(6) Sphegidae: 7. Ceropales maculatus F., skg.; 8. Pompilus viaticus Z., do.

1438. G. sylvaticum L. (Knuth, ‘ Bl. u. Insekt. a. d. nordfr. Ins.,’p.89; Kirchner,
.‘ Beitrage,’ p. 66.)\—The elongated heads of this species are arranged in a spike,
and are but slightly conspicuous. Kirchner describes them as 5-6 mm. long, with a
diameter above of only 13-2 mm. Each contains 60-70 female florets, and a few
(usually only 3-4) hermaphrodite ones. In both of these the corolla is 4 mm. in
length, and there is no difference in the form of the pistil: it contains one ovule, and
there is an annular nectary at the base of the style. The style of the female florets is
glabrous, and its two curved stigmatic branches, which are thin and moderately long,
spread out above the corolla-tube. The upper end of the style of hermaphrodite
florets is covered with sweeping-hairs. These brush out the pollen, and the two
Sstylar branches then diverge, so that the stigmatic papillae of their inner surfaces
become visible. ;

As I observed in the North Frisian Islands, either automatic self-pollination or
geitonogamy may now occur, for the spinose pollen-grains still clinging to the
sweeping-hairs may fall upon the stigmas of the same or neighbouring florets.

Visttors.—A Muscid was recorded in Dumfriesshire (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 92).

1439. G. uliginosum L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.” p. 89;
Kirchner, ‘ Beitrage,’ p. 66.)—The heads of this species are of a rounded ovoid
shape, 14-2 mm. in diameter above, and aggregated into crowded leafy clusters.
The individual florets are also only 13-2 mm. long. Kirchner counted rather more
than 100 female florets, and usually 6 hermaphrodite ones in each head: I have
observed only about 30. The flower mechanism otherwise agrees with that of the
last species.

According to Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896), the tips of the
involucral bracts of the small heads project somewhat beyond the female ray-florets,
pressing their stigmatic branches inwards towards the pollen of the few hermaphrodite
florets, so as to effect geitonogamy. The pollen-grains are pale yellow in colour,
ellipsoidal, with spinose tubercles, about 25 « long and 19 p broad.

Visttors.—Herm. Miiller observed the bee Sphecodes ephippius Z., skg.

608 ANGIOSPERMAE—DICOTYLEDONES

449. Helichrysum Vaill.

Ray-florets female, filiform, uniseriate ; disk-florets hermaphrodite, tubular.

1440. H. arenarium Moench (=Gnaphalium arenarium Z.).—The citron-
coloured involucral bracts make the heads of this species conspicuous.

According to Warnstorf (Verh. bot. Ver., Berlin, xxxiii, 1896), only the variety
without filiform female ray-florets occurs at Neu-Ruppin; the variety with orange-red
involucral bracts is not rare. The tip of each stylar branch possesses a bunch of
papillose sweeping-hairs, below which the stigmatic papillae are situated on the inner
surface. At first the style projects but little or not at all from the anther-cylinder, but
later on the diverging stigmatic branches project between the anthers so as to ensure
geitonogamy. The pollen-grains are golden-yellow in colour, rounded or ellipsoidal,
spinose, as much as 31 p» long and 23 pw broad.

Visirors.—Herm. Miiller (Brandenburg) repeatedly observed the beetle Cocci-
nella quattuordecimpunctata Z., resting on the heads (‘ Weit. Beob.,’ III, p. 89).

1441. H. bracteatum Andr. (Knuth, ‘Bliitenbiol. Herbstbeob.,’ ‘ Bliitenbiol,
Notizen.’)—

Vis1Tors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel Botanic Garden), a true wasp (Vespa vulgaris Z.), 2 hover-flies
(Eristalis arbustorum Z., and Helophilus pendulus Z.), 2 beetles (Coccinella quinque-
punctata Z., and C. quattuordecimpunctata Z., especially the latter), and earwigs

(Forficula, dvg. the florets). Schletterer (Tyrol), the parasitic humble-bee Psithyrus
tupestris /., and the carpenter-bee Xylocopa violacea Z.

1442. H. angustifolium DC.—
VisiTors.—Schletterer observed the following at Pola.—

Hymenoptera. (a) Apzdae: 1. Eriades truncorum Z.; 2. Eucera interrupta
Baer.; 3. Halictus leucozonius Schr.; 4. H. quadricinctus /.; 5. H. scabiosae Rossz’;
6. Megachile muraria Z. (6) Scolitdae: 7. Scolia hirta Schr.; 8. S. insubrica Scop.

450. Artemisia L.

Varies from class An to class Po. As Kirchner mentions (‘Beitrage,’ p. 67),
Delpino has shown that the group of Compositae constituted by him under the name
of Artemisiaceae, is anemophilous (‘Studi’sopra un lignaggio anemofilo delle Composte,
&c.,’ Firenze, 1871). The florets have become inconspicuous; the heads are often
pendulous; there is no nectar; the pollen is dry, and scattered without the agency of
animals,

Delpino describes various degrees of anemophily: the initial stage is found in
the genus Artemisia, and the adaptation is complete in the closely related genus
Oligosporus Cass. as well as in the groups Iveae and Ambrosieae.

In the genus Artemisia (including Oligosporus) Delpino distinguishes the
following stages.—Artemisia (sens. sfrict.) embraces species with heads containing
hermaphrodite as well as female florets; the stigmatic branches of the former
diverge in the usual way, and ultimately become recurved. Oligosporus is purely
monoecious: in each head there are purely male florets along with female ones:

COMPOSITAE 609

the ovaries of the former are vestigial, and the papillose stigmatic branches do not
diverge.

Delpino divides the genus Artemisia (sens. sirict.) into three sub-genera.—
1. Absinthium Zourn. (A. Absinthium Z., and A. camphorata V2i/.), representing
early stages of anemophily, i. e. with dust-like pollen that scatters itself, but with the
stigmas of the female florets still short, brightly coloured corollas to the hermaphrodite
florets, and heads that are not always pendulous. 2. Evartemia De/p. With better
marked adaptations to anemophily, i.e. prominent feathery stigmas in the female
florets, inconspicuous brownish corollas, and ovoid or ventricose pendulous heads.
3. Seriphidium Bess. With further specialized adaptations to anemophily, especially
heads containing a small number of hermaphrodite homogamous florets.

Kirchner rightly maintains that, between the normal Senecionidae and the species
of Artemisia belonging to Delpino’s sub-genus Absinthium, it would be more correct
to introduce another group, iucluding those species of Artemisia (e.g. A. glacialis Z.,
A. Mutellina Vi2/., and A. spicata Wud.) which may more accurately be described as
entomophilous than anemophilous. Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896)
describes as follows the flower mechanism of A. Absinthum, A. vulgaris, and A.
campestris.—The florets of the small heads are all tubular; the female ray-florets
possess a corolla-tube that is somewhat wider below and narrows gradually above,
passing into a 5-partite limb not sharply defined. The stylar branches are developed
before the central hermaphrodite florets are mature. In A. vulgaris and A. campestris
they are relatively long, devoid of apical thickenings and actual sweeping-hairs, but
closely beset with stigmatic papillae. Ata later stage they diverge and curve more
or less outwards, so as to come into contact with the pollen produced by their own
capitulum. The yellow or reddish hermaphrodite florets are bell-shaped above,
and the corolla-lobes expand when the pollen is ripe, so that the five awl-shaped
appendages on the upper part of the anther cylinder protrude. The two stigmatic
branches lie close together, and are expanded above, forming a funnel-shaped apical
depression covered all round with long papillose sweeping-hairs. They are still
within the anther-cylinder, but have already brushed out the pollen. This lodges
only for a short time between the appendages of the anthers, for owing to its small
size and the absence of any means of adhesion it is soon carried away by the wind.
The position of the head is also exceedingly favourable to anemophily. The stylar
branches now project beyond the limb of the corolla, and curve outwards from one
another. They turn their inner stigmatic surfaces upwards, and the appendages of
the anther-cylinder retract themselves into the corolla. The pollen-grains are small,
yellowish in colour, rounded or ellipsoidal, tuberculated, on an average 25 » long and
18 » broad.

1443. A. glacialis L.—Kirchner (‘Beitrage,’ pp. 67-8) says that at Zermatt
the individual florets of this species are very small, but the inflorescences are by no
means inconspicuous, for the corollas are of a golden-yellow colour. About 30-40
florets are aggregated into an erect head 4-6 mm. in diameter, and usually 5-7 of
these are crowded together. The species is gynomonoecious, with protandrous
hermaphrodite florets. In each head there are female ray-florets. Pollen is brushed
out of the central hermaphrodite florets by the two apposed stylar branches, which

are somewhat expanded above and covered with sweeping-hairs, while below they
DAVIS. II RT

610 ANGIOSPERMAE—DICOTYLEDONES

are beset internally with stigmatic papillae. These branches are relatively broad and
long, and ultimately bend outwards from one another till they become reflexed. The
yellow pollen is not mealy and not readily dispersed, but clumps of it remain clinging
to the top of the anther-cylinder. The florets are nectarless, and the annular nectary
at the base of the style is absent.

1444. A. borealis Pall. (Abromeit, ‘Bot. Ergeb. von Drygalski’s Grénlands-
exped.,’ pp. 64-5.)—This very variable species was several times collected by
Drs. von Drygalski and Vanhéffen in Greenland. The colour of the involucral
bracts and corolla-lobes of male florets is often purple, but may be golden-yellow or
greenish-white. In many cases the heads are almost purely male or predominatingly
female, and either protogynous or protandrous.

1445. A. Mutellina Vill—This species, like A. glacialis, was investigated at
Zermatt by Kirchner (‘ Beitrage,’ p. 69). It, too, belongs to class Po, with transition
to class An. The inflorescences are smaller, and their racemose arrangement also
tends to make them less conspicuous. A head usually consists of 8-16 golden-
yellow florets, of which the 5-8 marginal ones are female. Their stigmas mature
before the pollen of the hermaphrodite florets of the same head is available, so that

they are usually crossed by pollen from older heads, geitonogamously according to
Kerner.

1446. A. Absinthium L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 89-90.)—
This species belongs to class Po with transition to class An. The numerous, almost
spherical heads (about 4 mm. in diameter) are closely crowded on the many switch-like
branches that are easily moved by the wind. The stem is more than a metre high.
The plant is rendered tolerably conspicuous by the yellow colour of the tiny florets,
of which about 50 are aggregated into a head. Each of them, including the
ovary, is only 2 mm. long. The stylar branches of the central female florets possess
a few sweeping-hairs at their tips, and when the pollen has been removed they roll
up into two circular coils that project beyond the corolla, and present their stigmatic
surfaces to alighting insects, by which they may be pollinated. Pollen-devouring
insects are now and then attracted to the plant by sight, though it is only moderaiely
conspicuous, and by its aromatic odour.

VisiTors.— The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel), the hover-fly Syrphus ribesii Z., po-dvg. (‘ Bloemenbiol. Bijdragen’).
Réssler (Wiesbaden), the moth Grapholitha pupillana C7. Schletterer (Tyrol), the
bee Andrena combinata Chr.

1447. A. Dracunculus L.—

VistTors. — Borgstette (Nassau) observed a po-dvg. hover-fly (Melanostoma
mellina Z.).

1448. A. maritima L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 90.)—
This species possesses an aromatic odour. The short-stalked ovoid heads are
easily moved by the wind.

1449. A. vulgaris L. (Knuth, loc. cit.}—The very small ovoid heads are

6 mm. long and 3 mm. broad. Each contains about 20 florets 4 mm. in length.
MacLeod gives a figure (Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 420).

COMPOSITAE 611

1450. A. campestris L.—

Visitors.—Réssler records the following Lepidoptera for Wiesbaden, but
makes no reference to the object of their visits —

1. Conchylis dipoltella #2.; 2. Crambus alpinellus H%.; 3. Eurycreon turbidalis
Tr.; 4. Grapholitha lactaena 77.

451. Cotula L.

Heads golden-yellow in colour, solitary. Ray-florets female, barren, with inflated
corolla-tube : disk-florets hermaphrodite, with 4-toothed limb to corolla.

1451. C. coronopifolia L. (Roth, Bot. Jahrb., Leipzig, v, 1884.)—Roth, at
Liitjenburg (Holstein), repeatedly endeavoured to observe insects pollinating the
florets, but never succeeded in doing so. The heads, though crowded together, are
not at all conspicuous; besides which they are odourless, and appear to produce
scarcely any nectar. Roth suggests that perhaps some suitable insect, absent from
our area, may be found in California, which would account for the rapid spread of
the species there, and its almost stationary condition in the far North of Europe.
This idea does not appear to me to be supported by the facts, for nectar is secreted
at no great depth, and therefore accessible to most of our flower-visiting insects. The
case is one of insufficient observation.

452. Ammobium R. Br.
1452. A. alatum R. Br.—

Visttors.—I observed the beetle Coccinella quattuordecimpunctata Z., occasional,
in the Kiel Botanic Garden.

453. Achillea L.

Gynomonoecious. Heads small. Ray-florets female, white (rarely rose-coloured),
with rounded tongue, style devoid of sweeping-hairs. Disk-florets usually yellowish
and hermaphrodite ; style with diverging sweeping-hairs at the tip, and two bands of
stigmatic papillae, separated by a median space, internally.

1453. A. Millefolium L. (Herm. Miiller, ‘Fertilisation,’ pp. 325-7, ‘ Alpen-
blumen, p. 428, ‘ Weit. Beob.,’ III, p. 84 ; Lindman, ‘ Bidrag till Kanned. om Skandin.
Fjellvaxt. Blomn. o. Befrukt.’; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. go,
157-8, ‘Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 236; Verhoeff, ‘Bl. u.
Insekt. a. d. Ins. Norderney’ ; Heinsius, Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 363,
v, 1893, pp. 421-3; Loew, ‘Bliitenbiol. Floristik,’ pp. 390, 395.)—In this species
a large number of heads (often over roo) are arranged in a corymbose manner, thus
giving a continuous flat surface. Not only is the plant thus rendered conspicuous,
but a single visitor may pollinate numerous florets. Insects can pass from one head
to another, using the contiguous ligulate ray-florets as bridges, without having to fly.

Hermann Miiller states that a head contains about 20 disk-florets, with corolla-
tubes hardly 2 mm. long, and expanding into nectar-containing bells about 1 mm.
in length, The two stylar branches are closely apposed when the floret opens, and
ate situated in the lowest part of the anther-cylinder, through which they grow,
pushing the pollen before them by means of the sweeping-hairs. They then diverge,

RT 2

612 ANGIOSPERMAE—DICOTYLEDONES

turning their stigmatic surfaces upwards, and protrude a little from the bell, while the
empty anthers become somewhat retracted. The disk-florets are usually surrounded
by 5 ray-florets with large corolla-limbs and no stamens ; these increase the diameter
of the head to 9-10 mm.
Failing insect-visits, automatic self-pollination is effected in the disk-florets by
the fall of pollen from the sweeping-hairs on the outspread stigmatic papillae.
Gynodioecism is also said to occur occasionally.

Fic. 199. Achillea Millefolium, L. (after Herm. Miller). (1) A single head, seen from above :
4, stigmas of purely female ray-florets; 4, do. of disk-florets in the second stage; c, anther-cylinder of
disk-floret in the first stage; d, bud nearly ready to open. (2) Single disk-floret in the act of opening,
seen in longitudinal section: ¢, ovary; /, nectary; g, style; 4, corolla-tube; 7, filaments; 4, stylar
branches; /, tip of the style, with sweeping-hairs; 2, anther-cylinder; 7, throat of corolla; 9, valves
closing the anther-cylinder. (3) Disk-floret (with bract) somewhat more advanced; pollen is issuing
from the anther-cylinder. (4) Older disk-floret; the diverging stylar branches protrude, and the
anther-cylinder is retracted intothe corolla. (5) Tip of the style of a disk-floret in the first (male) stage :
4’, stigmatic papillae ; 7, sweeping-hairs ; #, pollen-grains.

Chrysanthemum Leucanthemum, L. (6) Disk-floret in the first (male) stage: #, pollen. (7) Do.,
in the second (female) stage. (8) Tip of a stylar branch, seen from within (x 60): &', stigmatic
papillae; 2, sweeping-hairs.

Vistrors.—I observed the following in Schleswig-Holstein (S.-H.), and on the
higher part of Helgoland (H.).—

A. Coleoptera. All po-dvg. (a) Curculionidae: 1. Apion marchicum Wés¢.
(S.-H.). (8) Mitedulidae: 2. Meligethes sp. (S.-H.). (c) Telephoridae: 3. Cantharis
fulva Scop.(S.-H.). B. Diptera. (2) Alusczdae: all skg.: 4. Anthomyia sp. (S.-H.);
5. Aricia incana Wred. 9 (S.-H.) ; 6. Dolichopus plumipes Scop. (S.-H.); 7. Leucostoma
aenescens Ze#/. (S.-H.); 8. Lucilia caesar Z. (S.-H.); 9. Musca corvina / (S.-H.);
10. Nemotelus uliginosus Z. (S.-H.); 11. Oliviera lateralis #, skg. (S.-H.);
12. Onesia sp. (S.-H.); 13. Pollenia sp. (S.-H.); 14. Sarcophaga carnaria Z.
(S.-H.); 15. 5S. striata Fadr. (S.-H.); 16. Scatophaga lutaria #. (S.-H.); 17. S.
merdaria #. (S.-H.); 18. S. stercoraria Z. (S.-H. and H.); 19. Spilogaster carbo-

COMPOSITAE 613

nella Ze/t. (S.-H.). (6) Syrphidae: all skg. and po-dvg.: 20. Eristalis arbus-
torum Z. (S.-H.); 21. E. pertinax Scop. (S.-H.); 22. E. tenax Z. (S.-H.); 23.
Helophilus pendulus Z. (S.-H.); 24. Syritta pipiens Z. (S.-H.). C. Lepidoptera.
All skg. (a) Rhopalocera: 25. Epinephele janira Z. (S.-H.); 26. Pieris sp. (S.-H.) ;
27. Polyommatus phlaeas Z. (S.-H.). (6) Sphingidae: 28. Zygaena filipendulae Z.
(S.-H.). D. Orthoptera. 29. Forficula auricularia Z., gnawing the florets (H.).

Loew gives the following for Brandenburg (B.) and Mecklenburg (M.) (‘ Beitrage,”
P- 39)—

A. Diptera. (a) Conopidae: 1. Zodion cinereum /. (M.). (6) Stratiomyzdae :
2. Nemotelus uliginosus Z. 9 and é (M.). (c) Syrphidae: 3. Eristalis aeneus Scop.
(B.); 4. E. sepulcralis Z.(B.). B. Hymenoptera. (a) Apidae: 5. Colletes fodiens
K. 3, po-cltg. (M.); 6. Prosopis dilatata K. 3, skg. (M.). (6) Ichneumonidae :
7. Undetermined sp. (M.). (c) Sphegidae: 8. Oxybelus bellus Dahid. (M.).

And in Silesia (op. cit., pp. 25-6).—

A. Diptera. (a) Muscidae: 1. Cistogaster globosa ., skg.; 2. Gymnosoma
rotundata Z., do.; 3. Ocyptera brassicaria &., do. (0) Syrphidae: 4. Eristalis
intricarius Z., skg.; 5. E. tenax Z., do. B. Hymenoptera. (a) Apidae: 6.
Cilissa tricincta K. 3, skg.; 7. Coelioxys octodentata Lep. $, do. (4) Chrysididae :
8. Hedychrum lucidulum #. (c) Sphegidae: 9. Cerceris nasuta Dahlb., skg.
C. Lepidoptera. 10. Polyommatus virgaureae Z.

To these he adds (op. cit., p. 49) A. Diptera. , Syrphidae: 1. Chrysogaster
coemeteriorum Z. B. Hymenoptera. Tenthredinidae: 2. Tenthredo sp. Also
(op. cit., p. 30)—A. Coleoptera. Cerambycidae: 1. Leptura testacea Z., nect-lkg.
B. Diptera. (a) Stratiomyidae: 2. Odontomyia irridula &, skg. (6) Syrphidae:
3. Syritta pipiens Z., skg.; 4. Volucella bombylans Z., do, C. Lepidoptera.
Rhopalocera: all skg.: 5. Argynnis aglaja Z., skg.; 6. A. pandora S.V.; 7. Coeno-
nympha arcania LZ.

Willis observed the following in the neighbourhood of the south coast of
Scotland (‘Fls. and Insects in Gt. Britain,’ Part I)—

A. Coleoptera. (a) Nitdulidae: 1. Cercus rufilabris Z/r., very common,
po-dvg. (4) Staphylintdae: 2. Quedius boops Grav., po-dvg. 8B. Diptera.
(a) Chironomidae: 3. Cricolopus sp., po-dvg. (4) Afuscidae: 4. Anthomyia radicum
L., freq., skg. and po-dvg.; 5. A. sp.; 6. Hydrellia griseola Fadl, po-dvg.;
4. Hyetodesia incana W., skg.; 8. Lucilia sericata Mg., skg. and po-dvg.;
g. Oliviera lateralis /’, skg.; 10. Phorbia floccosa MMcg., po-dvg.; 11. Scatophaga
stercoraria Z., do.; 12. Spilogaster communis Dev., skg. (c) Syrphidae: 13. Eristalis
pertinax Scop., po-dvg.; 14. E. tenax Z., skg.; 15. Sphaerophoria scripta Z., freq.,
skg.; 16. Syritta pipiens Z., skg.; 17. Syrphus balteatus Deg., do. C. Hemiptera.
18. Anthocoris sp., freq., skg.; 19. Calocaris bipunctatus /, skg.; 20. C. fulvo-
maculatus Deg., freq., skg. D. Lepidoptera. All skg. (a) Microlepidoptera:
21. Chorentis myllerana /.; 22. Simaéthis fabriciana Steph. (0) Noctuidae: 23.
Hydroecia nictitans (Z.) Bkh. (c) Rhopalocera: 24. Pieris napi Z.; 25. P. rapae L.;
26. Polyommatus phlaeae Z.

Herm. Miiller (H. M.) and Buddeberg (Budd.) give. the following list for
Achillea Millefolium and A. Ptarmica in Westphalia, Thuringia, and Nassau.—

A. Coleoptera. (2) Buprestidae: 1. Anthaxia millefolii #. (Budd); 2. A.
nitidula Z. (H. M., Thuringia; Budd.). (6) Cerambycidae: 3. Leptura livida F.,
po-dvg. (H. M.); 4. L. testacea Z., do. (H. M.); 5. Strangalia bifasciata AZduil.
(H. M., Thuringia); 6. S. melanura Z., po-dvg. (H. M., Bavarian Oberpfalz).
(c) Chrysomelidae: 4. Cryptocephalus sericeus Z., gnawing the florets (H. M.).
(d) Coccinellidae: 8. Coccinella mutabilis Scr7d., freq., resting on the heads (H. M.) ;

614 ANGIOSPERMAE—DICOTYLEDONES

9. C. septempunctata Z., do. (H. M.);_ 10. Exochomus auritus Scr7d., freq. (H. M.).
(¢) Elateridae: 11. Agriotes gallicus Zac., po-dvg. (H. M., Thur.); 12. A. ustulatus
Schail., do. (H. M., Thuringia and the Oberpfalz). (/) Mordellidae: 13. Mordella
fasciata F. (H.M.). (g) Ocdemeridae: 14. Oedemera podagrariae Z., po-dvg. (H. M.,
Thuringia). (4) Scarabaerdae: 15. Cetonia aurata Z., gnawing the florets (H. M.,
Thuringia). (2) TZelephoridae: 16. Telephorus melanurus #, as 15 (H. M.).
B. Diptera. (a) Bombyhidae: 17. Exoprosopa capucina #., in large numbers
(H. M.). (4) Conopidae: 18. Conops flavipes Z., freq., skg. (H. M.); 19. E.
scutellatus JMZg., skg. (H. M., Oberpfalz and Fichtelgebirge): 20. Physocephala
vittata #., freq., skg. (H. M.). (c) Empzdae: 21. Empis livida Z., freq. (H. M.).
(d) Muscidae: 22. Aricia vagans Fall. (H. M.); 23. Echinomyia ferox P2z., skg.
(H. M., Oberpfalz); 24. E. tessellata /, do. (H. M.); 25. Gonia capitata Deg., do.
(H. M.); 26. Gymnosoma rotundata Pz., do. (H. M., Budd.); 27. Ocyptera
cylindrica F., do. (H. M.); 28. Phasia crassipennis #. (H. M., Thuringia; Budd.) ;
29. Scatophaga stercoraria Z., po-dvg. (Budd.); 30. Trypeta pantherina Fal.
(Budd.); 31. Ulidia erythrophthalma J/g., very numerous (H. M., Thuringia).
(e) Syrphidae: 32. Chrysotoxum bicinctum Z., po-dvg. (H. M., Oberpfalz); 33.
Eristalis arbustorum Z., freq., skg. and po-dvg. (H. M.); 34. E. horticola Deg.,
po-dvg. (Budd.); 35. E. nemorum Z., freq., skg. and po-dvg. (H. M.); 36. E.
sepulcralis Z., do. (H. M.); 37. E. tenax Z., do. (H. M.); 38. Eumerus sabulonum
Fall. (H. M.); 39. Helophilus floreus Z., po-dvg. (H. M.); 40. Melithreptus scriptus
Z. (H. M.); 41. M. taeniatus 47/. (H. M.); 42. Paragus bicolor Z., pq-dvg.
(Budd.); 43. Syritta pipiens Z., freq., skg. and po-dvg. (H. M.); 44. Syrphus ribesii
L. (H. M., Oberpfalz); 45. Volucella bombylans Z.(H.M.); 46. V. pellucens Z.
ee M., Almetral). (/) Stratiomyidae: 47. Odontomyia viridula F., freq. (H. M.).
g&) Tabanidae: 48. Tabanus rusticus Z., freq. (H. M.). C. Hymenoptera.
(a) Apidae: 49. Andrena albicans A/ull. é, skg.(H.M.); 50. A. argentata Sm. 4,
do. (H. M.); 51. A. chrysosceles K. 9 (H. M.); 52. A. denticulata A. 8, skg.
(H. M.); 53. A. dorsata K. 9 and 4, in large numbers, skg. and po-cltg. (H. M.);
54. A. fulvicrus X. 9 and 4, do. (H.M.); 55. A. fuscipes K. é(H.M.); 56. A. lepida
Schenck 8, skg. (H. M.); 5%. A. nana X. 4, do. (H. M.); 58. A. nigripes X. 9, do.
(H. M.); 59. A. pilipes # 8, do. (H. M.); 60. A. schrankella My7. é (Budd.); 61.
Chelostoma nigricorne JVy/, 8, skg. (H. M.); 62. Colletes daviesanus X. 9 and 4, very
common, skg. and po-cltg. (H. M., Oberpfalz; Budd.); 63. C. fodiens K. 9 and 4,
very common, po-cltg., po-dvg. and skg.; 64. Halictus cylindricus #. 9 and 4,
po-cltg. and skg. (H. M.); 65. H. interruptus Pz. 9, po-cltg. (H. M., Thuringia) ;
66. H. leucozonius Schr. 9, do. (H. M.); 67. H. maculatus Sm., do. (H. M.);
68. H. morio F. 9 and 4, po-cltg. and skg. (H. M., Budd.) ; 69. H. quadricinctus #.
g and 6, do. (H. M., Budd.); 70. H. rubicundus Chr. 8, skg. (H. M.); 71. H.
smeathmanellus £. 9, do. (Budd.); 72. H. villosulus A. 9, skg. and po-cltg. (H. M.,
Budd.); 73. Heriades truncorum Z. 9 and 64, do. (H. M.); 74. Nomada ruficornis
L. 9, skg. (H. M.); 75. N. zonata Pz. 9, do. (H. M.); 76. Osmia leucomelaena £&.
9, po-cltg. (H. M.); 77. O. spinulosa X. 9, do. (H.M.); 78. Prosopis pictipes Wy/.
9 and 4, skg. and seizing pollen with its mouth-parts (H. M.); 79. P. confusa My/.
g and é (Budd.); 80. P. variegata / 9 and 4, very numerous, as P. pictipes (H. M.,
Budd.); 81. Rhophites quinquespinosus Sprx. 4, freq., skg. (H. M., Oberpfalz) ;
82. Sphecodes gibbus Z. (and a var.) 9 and 6, skg. (H. M., Budd.); 83. Stelis
breviuscula Vy). 9 and @, do. (H. M.). (8) CArysididae: 84. Hedychrum lucidulum
¥. 9 and 4, in large numbers (H. M.). (c) £vaniidae: 85. Foenus sp. (Budd.).
(2) Sphegidae: 86. Ammophila sabulosa Z. (H.M.); 87. Cerceris arenaria Z.,
not infrequent (H. M.); 88. C. labiata 7”, freq. (H. M.); 89. C. variabilis Schr., very
common (H.M.); go. Ceropales maculatus /,, in large numbers (H. M.); 91. Crabro
alatus Pz. 9 and 6, freq. (H.M.); 92. C. subterraneus /. 9 (H. M.); 93. C. vexillatus
Pz. & (Budd.); 94. Dinetus pictus #. (H. M.); 95. Lindenius albilabris F., in large
numbers (H. M.); 96. Oxybelus bellus DaA/é., numerous (H. M.); 97. O. nigripes

COMPOSITAE 615

Oliv. 9 (H. M.); 98. O. trispinosus F., freq. (H. M.); 99. O. uniglumis Z., do.
(H. M.); roo, Philanthus triangulum F. 9 and @, do. (H. M.); 101. Pompilus
chalybeatus Schjédte 9 (H. M.); 102. P. plumbeus F. 9 and $ (H.M.); 103. P.
rufipes Z. 9 and 6 (H. M.); 104. P. trivalis Dahib. 8 (H. M.); 105. P. viaticus Z.
$ (H. M.). (¢) Tenthredinidae: 106. Allantus nothus K/g., freq. (H. M., Budd.);
107. A. scrophulariae Z. (H. M.); 108. Athalia rosae Z., pairing on the heads
(H.M., Budd.); rog. Several undetermined sp. of Tenthredo (H.M.). (/) Vespidae :
110. Odynerus parietum Z. 9 (H. M., Budd.); 111. O. sinuatus /.9 (H.M.); 112. O.
spinipes Z. 9 (Budd.); 113. Pterocheilus phaleratus Z/r.9 (H.M.). D. Lepidoptera.
(2) Pyralidae: 114. Botys purpuralis Z., skg. (H. M.). (6) Rhopalocera: 115.
Coenonympha arcania Z., skg. (H. M., Thuringia); 116. Epinephile janira Z., do.
(H.M.); 117. Hesperia lineola O., do. (H. M., Oberpfalz); 118. H. sylvanus Zsp.,
do. (H. M.); 119. Lycaena aegon Schn. (H. M.); 120. L. icarus Rof, skg. (H. M.,
Oberpfalz) ; 121. Melanargia galatea Z. do. (Budd.); 122. Pieris napi Z. (H. M.);
123. P. rapae Z., skg. (H. M.); 124. Polyommatus phlaeas Z. (H. M.); 125.

Coenonympha pamphilus Z. (H. M.). (c) Zinerdac: 126. Pleurota schlaegeriella
Zell., skg. (Budd.).

Herm. Miiller also saw 4 flies, 2 Hymenoptera, and 24 Lepidoptera in the Alps.

The following were recorded by the observers, and for the localities stated.—

Lindman (Dovrefjeld), several Lepidoptera, a humble-bee, and a fly. Wiistnei
(Alsen), the ruby-wasp Hedychrum nobile Scop. Verhoeff (Norderney).—A. Diptera.
(a) Bibtonidae : 1. Dilophus vulgaris M/g., not infrequent. (6) Muscidae: 2. Aricia incana
Wiedem.; 3. Calliphora erythrocephala Mg. 9; 4. Cynomyia mortuorum Z., not
infrequent ; 5. Cyrtoneura simplex Zoew ; 6. Lucilia latifrons Schzx., very common ;
7. Scatophaga stercoraria Z.; 8. Stomoxys calcitrans Z. (c) Syrphidae: 9. Eristalis
arbustorum Z., very common; ro. Platycheirus manicatus Mg. 9; 11. Syrphus
balteatus Deg., occasional; 12. S. corollae #. B. Hymenoptera. (a) Apidae:
13. Prosopis communis Wy/.; 14. Sphecodes cirsii Verh. 8, occasional. (6) Vesprdae:
15. Odynerus parietum Z., occasional. Alfken (Juist)—Diptera. (2) Muscidae:
1. Lucilia caesar Z.; 2. Sarcophaga carnaria Z. (6) Syrphidae: 3. Eristalis arbus-
torum Z.; 4. Syritta pipiens Z. Schmiedeknecht (Thuringia), the bee Andrena
flavipes Pz., 2nd gen. Krieger (Leipzig)—Hymenoptera. (a) Aprdae: 1. Andrena
flavipes Pz.; 2. Halictus maculatus Sm.; 3. Nomada roberjeotiana Pz.; 4. N. soli-
daginis Pz.; 5. Prosopis communis Wy/.; 6. P. nigrita #. (6) Chrysididae: 7.
Chrysis neglecta Shuck.; 8. Hedychrum nobile Scop. (c) Sphegidae: 9g. Cerceris
labiata F.; ro. C. quinquefasciata Rossz. (d) Vespidae: 11. Eumenes coarctatus L.
Sickmann (Osnabriick), the parasitic fossorial wasp Ceropales maculatus F., once.
Friese (Mecklenburg), 2 bees (Colletes impunctatus /Vy/. and Prosopis variegata F.) ;
(Alsace), the bee Colletes daviesanus A.; (Thuringia), the fossorial wasp Dinetus
guttatus #. Gerstiacker (Berlin), the fossorial wasp Oxybelus quattuordecemnotatus
Jur. Alfken (Bremen).—A., Diptera. Syrphidae: 1. Eristalis anthophorinus Ze?z.,
not infrequent. B. Hymenoptera. (a) Apidae: 2. Colletes daviesanus X. 9 and é,
rare; 3.C. marginatus Z.9 and 4, do.; 4. Eriades truncorum Z. 9, do.; 5. Megachile
argentata /. 8, do.; 6. Stelis breviuscula, not infrequent. (4) Sphegidae: 7. Crabro
quadrimaculatus Sim., rare. C. Coleoptera. (a) Cerambycidae: 8. Leptura livida
L.; g. Stenopterus rufus Z. (4) Silphidae: 10. Necrophorus vespillo Z. Réssler
(Wiesbaden), the Tineid Adela tombacinella H.-Sch. Schenck (Nassau), 6 bees—
1. Halictus albipes #.; 2. H. morio #.; 3. Nomada furva Pz. ; 4. Prosopis bipunctata
F.; 5. P. brevicornis Ny/.; 6. P. nigrita # Schiner (Austria), 8 Muscids—1.
Besseria melanura A/g.; 2. Gymnosoma nitens /g.; 3. Lauxania cylindricornis /. ;
4. Metopia argentata Afacg.; 5. Miltogramma ruficornis JZg.; 6. Saltella scutellaris
Fall.; 7. Tephritis flavipennis Zoew; 8. Urophora stigma Loew. F.F. Kohl (Tyrol),
the ruby-wasp Hedychrum regium adr. Heinsius (Holland), the hawk-moth
Ino statices Z., and the butterfly Lycaena aegon Schn. 6. MacLeod (Flanders),

616 ANGIOSPERMAE—DICOTYLEDONES

6 Hymenoptera, 5 hover-flies, 7 other Diptera, 4 Lepidoptera, and 3 beetles
(Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 423); (Pyrenees), 2 Hymenoptera, 4 beetles,
and 11 flies (op. cit., iii, 1891, p. 362). Smith (England), the bees Colletes daviesanus
K. and C. marginatus Z. Saunders (England), 4 bees—z1. Colletes picistigma
Thoms.; 2. Prosopis cornuta Sm.; 3. P. dilatata K.; 4. P. masoni Saund. Scott-
Elliot (Dumfriesshire), numerous Hymenoptera, flies, Lepidoptera, and beetles (‘ Flora
of Dumfriesshire,’ p. 95).

1454. A. Ptarmica L. (Herm. Miller, ‘Fertilisation,’ pp. 327-9; Knuth, ‘ Bl.
u. Insekt. a. d. nordfr. Ins.,’ pp. 90, 158, ‘ Weit. Beob. ti. BI. u. Insekt. a. d. nordfr.
Ins.,’ p. 236 ; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893.)—The heads of this
species are larger than in A. Millefolium, but not so many of them are grouped:
together, and the two species are about equally conspicuous. Hermann Miiller says
that they occur in equal abundance, and flower simultaneously, in the same localities

Fic. 200. Achillea moschata, Jacg. (after Herm. Miller). 4. A head in the course of anthesis:
a, stylar branches of ray-florets; 5, do. of outer disk-florets in second (female) stage; c, inner disk-florets,
in first (male) stage. B. A single disk-floret in the first (male) stage: a, anther-cylinder ; g/, throat of
corolla; ov, ovary; fo, pollen; », corolla-tube. C. Do. in the second (female) stage. Lettering as for
B; gr, stylar branches. D. End of one of the stylar branches. , sweeping-hairs ; 22, stigmatic papillae.
E. Style with its branches rolled back. (4, B, C, £, x 7; D, x 80.)
in Westphalia, where they are visited with equal frequency by the same insects,
especially by species of Prosopis, which are chiefly attracted by the odour of the
plant. In other places, e. g. Schleswig-Holstein, A. Ptarmica is much less abundant
than A. Millefolium, so that the insect visitors are also much less numerous.
Hermann Miiller describes the heads of A. Ptarmica as containing 80-100 disk-
florets scarcely 2% mm. in length, which make up a surface 6-7 mm. in diameter.
This is extended to 15-18 mm. by means of 8-12 ray-florets, of which the tongues
are 4-6 mm. long, and about the same breadth.

COMPOSITAE 617

Vistrors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (North Frisian Islands), a Lepidopterid, 3 Diptera, and a beetle; (Thu-
ringia), the po-dvg. beetle Cetonia aurata Z. MacLeod (Belgium), 2 flies (Bot.
Jaarb. Dodonaea, Ghent, v, 1893, p. 424). Loew (Berlin Botanic Garden), a Muscid
(Echinomyia fera Z.) and 2 Syrphids (Eristalis nemorum Z., and Syritta pipiens Z.).
Scott-Elliot (Dumfriesshire), a hover-fly and a Muscid (‘Flora of Dumfriesshire,’
p- 95). For the list given by Herm. Miiller, see pp. 613-13.

1455. A. moschata Jacq. (Herm. Miiller, ‘Alpenblumen,’ pp. 426-8.)—
This species is gynomonoecious (in the Upper Engadine), with protandrous her-
maphrodite florets. The diameter of the yellow disk is 3-5 mm., that of the entire
head is 10-14 mm. Numerous heads are aggregated into a corymb. The 20-25,
hermaphrodite disk-florets mature successively in centripetal order. Failing insect-
visitors, automatic self-pollination easily takes place, partly by the fall of pollen from
the papillose sweeping-hairs upon the expanded inner (stigmatic) surfaces of the stylar
branches, partly by the rolling back of these until the stigmatic papillae touch the
undivided part of the style. The plant is odorous.

VisiTors.—Herm. Miiller observed a beetle, g flies, 2 bees, and 10 Lepidoptera.

1456. A. nana L. (Herm. Miiller, ‘Alpenblumen,’ p. 428.)—The flower
mechanism agrees with that of A. moschata, but there are only 6-9 heads in the
corymb, which is 12-20 mm. in diameter. There are about 20 disk-florets, and 7-10
ray-florets.

Visirors.—Herm. Miiller observed 10 species of Diptera.

1457. A. atrata L. (Herm. Miiller, ‘Alpenblumen,’ pp. 428-9.)—The flower
mechanism of this species agrees with that of A. moschata. About 50 disk-florets
and 9-12 ray-florets make up a head 12-18 mm. in diameter, and 3-8 heads are
aggregated into a corymb.

Visirors.—Herm. Miiller observed 5 beetles, g flies, a Hymenopterid, and 2
Lepidoptera.

1458. A. macrophylla L. (Herm. Miiller, ‘Alpenblumen,’ pp. 429-30.)—
This species is gynomonoecious, with protandrous hermaphrodite florets. Each
head is 10 mm. in diameter, and consists of about 20 disk-florets, and usually 5 ray-
florets: 6-12 heads are aggregated into a dense corymb 25-40 mm. in breadth.
Automatic self-pollination is easily possible.

1459. A. coronopifolia Willd.—

Visirors.—Loew observed the fossorial wasp Dinetus pictus /”. 6 in the Berlin
Botanic Garden.

1460. A. dentifera DC.—

Visitors.—Loew observed the hover-fly Eristalis arbustorum Z. in the Berlin
Botanic Garden.

1461. A. filipendulina Lam.—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Coccinellidae: 1. Coccinella quattuordecimpunctata Z.
B. Diptera. Syrphidae: 2. Syritta pipiens Z. C. Hemiptera. 3. Calocoris sp.;
4. Corizus parumpunctatus Sc/z//. D. Hymenoptera. Vespidae: 5. Eumenes
coarctatus ZL.

618 ANGIOSPERMAE—DICOTYLEDONES

1462. A. grandifolia Frivald—

Visttors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp. (4) Syrphidae: 2. Eristalis
arbustorum Z.; 3. E. nemorum Z.; 4. Helophilus floreus Z.; 5. Syritra pipiens Z.

B. Hymenoptera. Sphegidae: 6. Eumenes coarctatus Z. C. Lepidoptera.
Rhopalocera: 7. Pieris brassicae L., skg.

1463. A. nobilis L—

Visrrors.— Loew observed 2 hover-flies (Eristalis arbustorum Z., and E. nemorum
L.) and a bee (Sphecodes gibbus Z. 9, skg.) in the Berlin Botanic Garden.

1464. A. tanacetifolia All., var. dentifera DC.—

Visirors.— Loew observed the hover-fly Syritta pipiens Z. in the Berlin Botanic
Garden,

454. Anthemis Mich.

Ray-fiorets white or yellow in colour, somewhat elongated. Heads larger than
in Achillea, but structure of the style the same.

1465. A. arvensis L. (Herm. Miller, ‘Fertilisation, p. 329; Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.’ pp. 90, 158.)—Hermann Miller describes the heads of, this
species as being 21-7 mm. in diameter, of which 5-7 mm. are taken up by the
yellow disk. The number of florets in the latter reaches several hundreds, while
Ludwig states that there are usually 5, 8, or 13 ray-florets. The sweeping-hairs
of the female florets are considerably shorter than those of the hermaphrodite
ones. (Cf Fig. 202, 8.)

Automatic self-pollination is always possible, as in Achillea Millefolium.

VistTors.— Herm. Miiller gives the following list.—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida Z. (6) Curculionidae:
2. Bruchus sp. (c) £lateridae: 3. Athous niger Z. B. Diptera. (a) Muscidae:
4. Echinomyia tessellata #., po-dvg.; 5. Scatophaga merdaria /., do.; 6. S. sterco-
raria Z., do. (4) Strattomyidae: 7. Nemotelus pantherinus Z., extremely numerous.
(c) Syrphidae: all po-dvg.: 8. Eristalis arbustorum Z.; 9. E. nemorum Z.;
10. E. sepulcralis Z.; 11. E. tenax Z. C. Hymenoptera. (a) Apidae: 12.
Andrena fulvicrus X. 9, skg. and po-dvg.; 13. A. minutula X.d; 14. A. nana K. 9,
skg.; 15. A. nigroaenea X. 9, skg. and po-cltg.; 16. A. schrankella JVy/. 9, do.;
17. Apis mellifica Z. ¥, skg.; 18. Colletes daviesanus X. 9 and 4, freq., skg. and
po-cltg. ; 19. Halictus lucidulus Schenck 9; 20. H. nitidiusculus A. 9. (4) Sphegidae :
21. Cerceris variabilis Schr. $; 22. Crabro alatus Pz. 9 and 6; 23. C. cribrarius Z. 9.
(c) Tenthredinidae: 24. Allantus nothus AZ.

The following were recorded by the observers, and for the localities stated.—

Knuth (Schleswig-Holstein), 2 bees and 5 flies; (Helgoland), the Muscid Lucilia
caesar Z. Wiistnei (Alsen), the bee Colletes daviesanus A. Schmiedeknecht (Thu-
ringia), the bee Osmia montivaga Mor. Redtenbacher (Austria), the Cistelid Podonta
nigritia #. Schletterer (Pola), 8 bees—1. Andrena convexiuscula X., var. fuscata
A; 2. A. cyanescens Vyl.; 3. A. nana &.; 4. Eriades truncorum Z.; 5. Halictus
calceatus Scop.; 6. H. fasciatellus Schenck; 7. H. tetrazonius Klug; 8. Prosopis
hyalinata Smzth, var. corvina Fors/. Kohl (Tyrol), the very common sphegid Crabro
cribrarius Z.

COMPOSITAE 619

1466. A. tinctoria L. (Herm. Miller, ‘Fertilisation,’ p. 329, ‘Weit. Beob.,
III, pp. 86-8 ; Knuth, ‘ Bloemenbiol. Bijdragen.’)—The diameter of the golden yellow
disk is 12-18 mm. in this species. Hermann Miiller describes it as consisting of
from 300 to over 500 tubular hermaphrodite florets, surrounded by 30-35 female
ray-florets, which are usually of the same colour. The entire head presents a circular
surface 25-40 mm. across. The ray-florets open first : their stylar branches diverge
and roll back to some extent. The disk-flowers next mature in centripetal order, in
successive zones: their mechanism agrees with that of the disk-florets of Achillea
Millefolium. As the corolla-tubes of the disk-florets are only 2 mm. in length, and
the bells (into which the nectar ascends) are only 1 mm. long, the nectar is accessible

even to insects with the shortest form of proboscis, provided this is of suitable
thickness.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list. —

A.Coleoptera. (a) Buprestidae: 1. Anthaxia nitidula Z.(Budd.). (0) Ceram-
bycidae: 2. Strangalia bifasciata A/#l/. 9 and %, po-dvg. (H. M., Thuringia).
(c) Chrysomelidae: 3. Cryptocephalus sericeus Z., gnawing the anthers (H. M.,
Thuringia). (d@) Llateridae: 4. Agriotes gallicus Zac. (H. M.). (¢) Mordellidae:
5. Mordella aculeata Z. (H. M.); 6. M. fasciata #. (H. M.). (f°) Ovdemeridae :
7. Oedemera flavescens Z., po-dvg. (H. M., Thuringia). B. Diptera. (2) Bombylitdae :
8. Exoprosopa capucina /. (Budd.). (4) Conopidac: 9. Myopa sp., skg. (H. M.).
(c) Muscidae: 10. Anthomyia sp., po-dvg. (Budd.); 11. Aricia sp., po-dvg. (H. M.,
Thuringia); 12. Gymnosoma rotundata Z. (H. M.); 13. Ocyptera brassicariae #.,
skg. (H. M., Thuringia); 14. Ulidia erythrophthalma A/., very common (H. M.,
Thuringia). (d) Syrphidae: 15. Eristalis arbustorum Z., po-dvg. (H. M.); 16.
Helophilus floreus Z., po-dvg. (H. M., Thuringia); 17. Melithreptus taeniatus JZg.,
po-dvg. (H. M.); 18. Syritta pipiens Z., skg. and po-dvg. (H. M., Budd.).
C. Hemiptera. 19. Calocoris chenopodii Fa//., skg. (H. M., Thuringia). D.
Hymenoptera. (a) Apidac: 20. Colletes daviesanus A. 9 skg. and po-cltg.
(H. M., Thuringia), 5 skg. (Budd.); 21. C. marginatus Z. 4, skg. (H. M.);
22. Halictus maculatus Sm. 6 skg., 9 po-cltg. (H. M., Budd.); 23. Heriades
truncorum Z. 9, po-cltg. (H. M.); 24. Osmia spinulosa A. 6, skg. (Budd.);
25. Prosopis propinqua /Vy/. 4, skg. (Budd.); 26. Rhophites quinquespinosus Spzn.
8, skg. (Budd.). (4) Jchneumonidae: 24. Several (H. M.). (¢) Zenthredinidae:
28. Tarpa cephalotes /., very common (H.M., Thuringia). (d) Vespidae: 29. Vespa
rufa L. ¥, settling but flying off again at once (H. M., Thuringia). E. Lepidoptera.
(a) Rhopalocera: 30. Epinephele janira Z., skg. (H. M., Thuringia); 31. Lycaena
corydon Scop., skg. (H. M., Thuringia); 32. Melanargia galatea Z., skg. (H. M.,
Thuringia) ; 33. Thecla ilicis £sp., skg. (H. M., Thuringia; Budd.). (4) Sphingzdae :
34. Zygaena achilleae Zsp., skg. (H. M., Thuringia). (¢) Zenerdae: 35. Nemotois
dumeriliellus Dup., skg. (Budd.).

The following were recorded by the observers, and for the localities stated. —

Knuth (on cultivated plants), some skg. and po-cltg. bees (Apis, and Halictus
cylindricus #.), po-dvg. hover-flies (Eristalis arbustorum Z., E. tenax Z., E. nemorum
Z.,and Helophilus pendulus Z.), and skg. Muscids (Lucilia caesar Z., L. cornicina #.,
and Scatophaga stercoraria Z.). Delpino the Bombyliid Lometia beelzebub F. (‘Ult.
oss.,’ Soc. ital. sc. nat., Milano, xvii). Loew (Berlin Botanic Garden).—A. Coleoptera.
Coccinellidae: 1.Coccinella impustulata Z. ; 2.C.quattuordecimpunctataZ. B. Diptera.
(2) Muscidae: 3. Anthomyia sp. (4) Syrphrdae: 4. Eristalisnemorum Z.; 5. Helophilus
floreus Z.; 6. Syritta pipiens Z.; 7. Syrphus ribesii Z. C. Lepidoptera. Rhopalocera:
8. Pieris brassicae Z., skg.

620 ANGIOSPERMAE—DICOTYLEDONES

1467. A. rigescens Willd.—

Visrrors.—Loew observed a Muscid (Anthomyia sp.) and 3 Syrphids (1. Eristalis
tenax Z.; 2. Melanostoma mellina Z.; 3. Syritta pipiens Z.) in the Berlin Botanic
Garden.

1468. A. Cotula L.—Ludwig states that the white ray-florets are neuter and
usually 8 or 13 in number.

455. Matricaria L.

Ray-florets ligulate, white in colour, sometimes wanting ; disk-florets hermaphro-
dite, yellow. Structure of style as in Achillea (cf Fig. 202, 10).

1469. M. Chamomilla L. (Ogle, Pop. Sci. Rev., London, ix, 1870, pp. 1604 ;
Herm. Miller, ‘ Fertilisation,’ p. 332, ‘ Weit. Beob.,’ III, p. 86; Knuth, ‘BI. u. Insekt.
a. d. nordfr. Ins.,’ pp. 91, 158; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893.)—
The diameter of the entire head in this species is 18-24 mm., that of the disk 6-8 mm.
Hermann Miller points out that as the florets mature in centripetal order, the common
receptacle rises up into a cylinder terminated by a blunt cone. The florets which
are over cover the cylinder, and the closed ones the cone, while the region actually in
flower comes between the two. The last is naturally the part first touched by
alighting insects, so that visitors constantly settle on the best place for obtaining
booty, and for effecting pollination. In other respects the flower mechanism agrees
with that of Anthemis arvensis.

VisiTors.—These are more especially flies, while the strong odour of the heads
is disagreeable to bees, with the exception of species of Prosopis.
Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list.—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida Z., not infrequent
(H. M.); 2. Strangalia attenuata Z., do. (H. M.). (6) Mitdulidae: 3. Meligethes,
freq. (H. M.). B. Diptera. (a) Empidae: 4. Empis livida Z., skg. (H. M.).
(6) Muscidae: 5. Lucilia cornicina #. (H. M.); 6, Pollenia vespillo #., po-dvg.
(H. M.); 7. Sarcophaga carnaria Z., freq., po-dvg. (H. M.); 8. S. haemorrhoa
Mg., po-dvg. (H. M.); 9. Spilogaster nigrita al’. (H.M.). (¢) Stratomyidae: 10.
Nemotelus pantherinus Z., very common, skg. (H. M.). (d@) Syrphidae: 11. Eristalis
arbustorum Z., very common, po-dvg. (H. M.); 12. E. nemorum Z., do. (H. M.);
13. E. sepulcralis Z., do. (H.M.); 14. Syritta pipiens Z., do. (H.M.). C. Hymeno-
ptera. (a) Apidae: 15. Colletes daviesanus X. 4, in large numbers, skg. (Budd.) ;
16. Halictus nitidus Schenck 3, skg. (Budd,); 17. Prosopis signata Pz. 9 and 4,
freq., alternately settling and flying away (H. M., Budd.); 18. Sphecodes gibbus Z.
g and é(H.M.). (4) Sphegedae: 19. Oxybelus uniglumis Z., freq. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Knuth (North Frisian Islands), Apis, 9 flies, a Lepidopterid, and a beetle.
Schletterer (Pola), the dasygastrid bee Eriades trancorum Z. MacLeod (Flanders),
a humble-bee, 2 Lepidoptera, and 7 beetles (Bot. Jaarb. Dodonaea, Ghent, v,
1893, p. 424). Saunders (England), the two bees Colletes daviesanus A. and C.
picistigma Zhoms. Krieger (Leipzig), the very common bee Colletes daviesanus X.

1470. M. inodora L. (=Chrysanthemum inodorum Z.). (Knuth, ‘BI. u.
Insekt. a. d. nordfr. Ins.,’ pp. 91, 158; Herm. Miiller, ‘ Fertilisation,’ p. 331, ‘ Weit.
Beob.,’ III, p. 86.)—The flower mechanism of this species essentially resembles that
of Anthemis arvensis. Ludwig states that the almost odourless heads usually contain
13 or 21 ray-florets.

COMPOSITAE 621

According to Kerner, the common receptacle is at first only slightly curved, but
subsequently rises up so as to bring the stigmas of the outer tubular florets into the
line of fall of pollen from the inner ones, so that geitonogamy takes place.

Visitors.— Willis observed the following in the neighbourhood of the south
coast of Scotland (‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera. Curculionidae: 1. Anthonomus rubi Zerést., freq. B.
Diptera. (a) Bibionidae: 2. Scatopse brevicornis Mg., freq., skg. (4) Muscidae:
3. Anthomyia radicum Z., very common, skg. and po-dvg.; 4. A. sp., po-dvg.;
5. Hydrellia griseola Fal/., skg. and po-dvg.; 6. Oscinis frit Z., freq. po-dvg.;
7. Drosophila graminum Fall, freq., skg.; 8. Spilogaster communis Dsv., skg. ;
9. Themira minor al., freq., skg. (c) Mycetophilidae: 10. Sciara sp., freq., skg.
(d) Syrphidae: 11. Ascia podagrica F., freq., skg.; 12. Eristalis pertinax Scop.,
skg.; 13. E. tenax Z., do.: 14. Sphaerophoria scripta Z., do. C. Hemiptera.
15. Calocoris bipunctatus /., freq.; 16. C. fulvomaculatus Deg., do. D. Hymeno-
ptera. (a) Apzdae: all skg.: 17. Bombus lapidarius Z.; 18. Halictus cylindricus F. ;
19. H. rubicundus Chr.; 20. Prosopis brevicornis Vy/.; 21. Sphecodes affinis Hag.,
freq. (4) Vespidae: 22. Odynerus pictus Curz., freq., skg. E. Lepidoptera. All
skg.: (a) Microlepidopiera: 23. Choreutis myllerana #.; 24. Simaéthis fabricana
Steph. (6) Rhopalocera: 253. Polyommatus phlaeas L., freq.

The following were recorded by the observers, and for the localities stated.—

Herm. Miller, a Chrysidid (Hedychrum lucidulum #. ) and a Muscid (Ulidia
erythrophthalma J@g.). Knuth (Schleswig-Holstein)—A. Diptera. (a) Dolicho-
podidae: 1. Gymnopternus nobilitatus Z. (0) Alusczdae: all skg.: 2. Lucilia
caesar Z.; 3. L. cornicina #.; 4. Pollenia rudis #.; 5. Scatophaga merdaria /. ;
6. small Muscids. (4) Syrphidae: all skg. and po-dvg.; 7. Eristalis arbustorum Z. ;
8. E. sp.; 9. E. tenax Z. B. Hymenoptera. Apidae: 10. Colletes daviesanus /”.,
skg. Also (Helgoland), 4 Muscids, all skg. (1. Coelopa frigida Fa//.; 2. Fucellia
fucorum Fal/.; 3. Lucilia caesar Z.; 4. Olivieria lateralis #.) (Bot. Jaarb. Dodonaea,
Ghent, viii, 1896, p. 40). Scott-Elliot (Dumfriesshire), numerous bees, flies, and
Lepidoptera (‘Flora of Dumfriesshire,’ p. 94).

I investigated an interesting variety that occurs on the sea-shore at Kiel,
M. maritima Z. (as a species) (=Chrysanthemum maritimum Pers.). It is very
conspicuous owing to the numerous large heads that are borne on the diffusely
branched ascending stem. The plant exhales a faint odour of chamomile (intensified
by rubbing) which leads one to suppose that flies are the pollinating agents. The
white tongues of the 20-30 female ray-florets are (as in the type form) rather more
than 1 cm. long, and about 4 mm. broad at the end. These florets surround
several hundred tubular yellow disk-florets, presenting a surface over 1 cm. broad,
so that the total diameter of the head is about 34 cm. The corolla of the disk-
florets is about 2 mm. in length, including a bell (scarcely 1 mm. long) containing
a little nectar. It is white in colour with yellow teeth.

The small florets are male during the first stage of anthesis: pollen is then
pressed out of the tip of the anther-cylinder by upgrowth of the apposed stylar
branches, and covers the surface of the floret. In. the second (female) stage the
stylar branches diverge so as to place their receptive inner surfaces in the position
previously occupied by the pollen. It follows that insects creeping about on the
surface of the head touch either the pollen or the stigmatic papillae, and effect
the crossing of a large number of florets at one time. Should insect-visits fail,

622 ANGIOSPERMAE—DICOTYLEDONES

the stylar branches gradually bend round till their stigmatic surfaces touch any
pollen that may remain, and are thus automatically self-pollinated.

The florets develop in centripetal order (as in all Compositae), so that in the
disk of a fully mature head they may be observed in the following stages, beginning
at the outside,—faded, female stage, male stage, not yet open. As flowering
progresses inwards, the at first sharply arched common receptacle swells up and
becomes globular, so that florets which have yet to shed their pollen or are in
the act of doing so are at a considerably higher level than those that have faded.
It follows that alighting insects settle only on the former, and they avoid the faded
florets situated on the sloping surface below. The arrangement is favourable to
geitonogamy, as in the type form.

Vistrors.—I observed the hover-fly Eristalis arbustorum Z., and the Muscid
Scatophaga merdaria /., both skg.

1471. M. discoidea DC. (=Matricaria suaveolens Buch., and Chrysanthemum
suaveolens Aschers.).—In this species there are no ray-florets. The mechanism of
the yellow disk-florets appears to be the same as in the last two species. Warnstorf
(Verh. bot. Ver., Berlin, xxxviii, 1896) describes it as follows——Many very small
hermaphrodite tubular florets about 4 mm. long are aggregated into an almost
spherical yellowish-green head. They turn green after the pollen is matured,
and become larger, so that the head increases considerably in size. The pollen
is not raised above the corollas by the stylar branches, but lies between their lobes,
which protect it from being carried away. Owing to the smallness of the flowers,
geitonogamy inevitably results from the divergence of the stylar branches. The
pollen-grains are yellow in colour, polyhedral, with spinose tubercles, about 25 p»
in diameter.

Vistrors.—I observed the hover-fly Syritta pipiens Z., skg. and po-dvg., and
the Muscid Scatophaga stercoraria Z., skg.

456. Tanacetum Tourn.

Ray-florets are wanting in some species, when present ligulate, and white in
colour. Otherwise like Matricaria.

1472. T. vulgare L. (=Chrysanthemum Tanacetum V7s.). (Herm. Miller,
‘Fertilisation,’ pp. 232-3; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 91.)—The heads of this species are in one plane,
as in Achillea, conferring the advantage that insect visitors can creep over the whole
surface without having to use their wings, and a very large number of florets can be
pollinated at the same time. The crowding together of numerous heads, though
these are small and rayless, makes the plant so conspicuous that it is very easily seen,
and therefore visited by a large number of insects.

Hermann Miiller states that each head contains several hundred yellow florets.
The bell of the corolla is only 1 mm. deep. The style has the same structure as
in Achillea: it bears a capitate bunch of diverging sweeping-hairs at the tip of its
branches, by which pollen is swept out of the anther-cylinder during the first stage of
anthesis. In the second stage the stylar branches, which are beset internally with

COMPOSITAE 623

stigmatic papillae, diverge in such a way as to stand at the same level as that
previously occupied by the pollen masses.

Visitors.—Herm. Miiller gives the following list.—

A. Coleoptera. Coccinellidae: 1. Coccinella bipunctata Z.; 2. C. quinque-
punctata Z. B. Diptera. (2) Muscidae: 3. Sarcophaga carnaria ZL. (6) Sératio-
myidae: 4. Odontomyia viridula ¥., freq. (c) Syrphidae: 5. Eristalis arbustorum Z.,
freq., po-dvg.; 6. E. nemorum Z., do.; 7. Melithreptus taeniatus J/g., po-dvg.;
8. Syritta pipiens Z., very numerous, skg. and po-dvg.; 9. Syrphus ribesii Z.,
freq., po-dvg. C. Hemiptera. 10. Several sp. of bugs. D. Hymenoptera.
(2) Apidae: 11. Andrena denticulata K. 9, po-cltg.; 12. A. fulvicrus X. 3, skg.;
13. Apis mellifica Z. ¥, do.; 14. Colletes daviesanus A’. é and 9, extraordinarily freq.,
skg. and po-cltg.; 15. C. fodiens K. 8 and 9, very common, skg. and po-cltg. ;
16. Halictus maculatus Sm. $ and 9, do.; 17. Sphecodes gibbus L. 8 and 9,
different varieties (including ephippius Z.), skg., and dusting itself with pollen.
(5) Sphegidae: 18. Crabro sp.; 19. Dinetus pictus /. 9 and 4, in large numbers;
20. Mellinus arvensis Z. (c) Vespedae: 21. Odynerus parietum Z.$. E,. Lepido-
ptera. (2) Pyralidae: 22. Botys purpuralis Z., skg. (5) Moctuidae: 23. Hadena
didyma Zp. 3, skg. (c) Rhopalocera: 24. Polyommatus dorilis H/n., skg.; 25. P.
phlaeas Z., skg.; 26. Vanessa atalanta Z., skg. F. Neuroptera. 27. Panorpa
communis Z., freq.

The following were recorded by the observers, and for the localities stated.—

Knuth, Apis, 2 humble-bees, a short-tongued bee, a Tenthredinid, 3 Lepidoptera,
5 hover-flies, 6 Muscidae, and a beetle (‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 158-9).
Wiistnei (Alsen), 4 bees (1. Colletes daviesanus A.; 2. Halictus albipes /ér.;
3. H. morio /ér.; 4. H. nitidiusculus X.). Friese (Mecklenburg), 5 bees (1. Colletes
daviesanus X., very common; 2. C. fodiens X., freq.; 3. C. marginatus Sm., rare ;
4. Epeolus productus Zzs., not infrequent; 5. E. variegatus Z.): also (Thuringia)
the bee Colletes daviesanus A. Alfken (Juist), a Muscid (Lucilia caesar Z.), a Capsid
(Calocoris norvegicus Gmel., very common, skg.), and a bee (Colletes daviesanus X.
9, occasional. Sickmann (Osnabriick), 3 Sphegids (1. Ceropales maculatus F.,
occasional; 2. Dioretus pictus /., freq.; 3. Mellinus sabulosus /., freq. Alfken
(Bremen).— A. Diptera. (a) Muscidae: 1. Gymnosoma rotundata Z., rare.
(6) Syrphidae: 2. Eristalis anthophorinus Zeét., very rare; 3. Melanostoma mellina
L., freq. B. Hymenoptera. (a) Apzdae: 4. Bombus derhamellus XK. ¥; 5. B.
muscorum /. 9; 6. B. terrester L. 3, skg.; 7. Colletes daviesanus XK. 9 very common,
skg. and po-cltg., 5 very common, skg. (this is the most frequent visitor among the
bees); 8. C. fodiens K., freq., 9 skg. and po-cltg., skg.; 9. C. picistigma Zzhs.,
rare, 9 skg. and po-cltg., 6 skg.; 10. Epeolus variegatus Z. 9 and 4, freq., skg.;
1x. Eriades nigricomis JVy/. 9, rare; 12. E. truncorum Z. 9 and 4, freq.;
13. Halictus rubicundus Cér. 9, rare; 14. Melitta leporina Pz. §; 15. Psithyrus
barbutellus A. 9; 16. Stelis breviuscula My/. 9 and 6, freq. (4) Sphegidae:
17. Ammophila affinis X. 9, not infrequent, skg.; 18. Diodontus tristis v. d. LZ. 9.
(c) Zenthredinidae: 19. Athalia glabricollis Zhs.; 20. A. spinarum F.; 21. Dolerus
pratensis Fall. Schenck (Nassau), the following bees—r. Colletes fodiens £.;
2. Halictus minutulus Schenck; 3. H. nitidiusculus K.; 4. H. pauxillus Schenck 8;
5. Nomada rhenana Mor.; 6. N. ruficornis Z.; 7. Prosopis bipunctata /.;
8. P. nigrita #. MacLeod (Flanders), Apis, 4 short-tongued bees, 4 hover-flies,
6 Muscids, and a Lepidopterid (Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 426-7):
also a Muscid, a Vespid, and a beetle (op. cit., vi, 1894, p. 374). Schneider (on
imported plants in Arctic Norway), the humble-bee Bombus terrester Z. $ and 9, by
the hundred (1890), also B. lapponicus Z. and B. scrimshiranus AX. (Tromse Mus.
Aarsh., 1894). Smith (England), 3 bees (1. Colletes daviesanus K.; 2. C. fodiens K.;
3. Epeolus variegatus Z.). Loew (Berlin Botanic Garden), the Muscid Tephritis
elongatula Zw., and undetermined hemipterous larvae.

624 ANGIOSPERMAE—DICOTYLEDONES

1473. T. corymbosum Sch. Bip. (=Chrysanthemum corymbosum Z., and
Pyrethrum corymbosum Scop.).—Kerner made comparative cultivations of this
species in the Vienna Botanic Garden and on the Blaser in the Tyrol, and observed
that the lowland specimens developed larger heads (26 mm. in diameter), and larger
ray-florets (tongues 8 mm. long and 4 mm. broad) than the highland ones, the heads
of which were only 20 mm. in diameter with ray-florets 7 mm. long.

Vistrors.—Herm. Miiller observed the following in Thuringia.—

A.Coleoptera. (a) Buprestidae: 1. Anthaxia nitidula Z. (6) Cerambycidae :
2, Strangalia bifasciata Aw. 9 and %, freq.; 3. S. melanura Z., both po-dvg.
({c) Curculionidae: 4. Spermophagus cardui S¥ev. (d) Mordellidae: 5. Mordella
aculeata Z. (e) Oedemeridae: 6. Oedemera marginata F.; 7. O. virescens L.,
po-dvg. (/) Telephoridae: 8. Danacea pallipes Pz.; 9. Dasytes flavipes fF.
B. Diptera. (a) Bombyliidae: 10. Anthrax morio LZ. (4) Empzdae: 11. Empis
livida L., freq., skg. (c) Muscedae: 12. Aricia sp.; 13. Ulidia erythrophthalma A/g.,
in very large numbers. (d) Stratiomyidae: 14. Nemotelus pantherinus Z., skg.
C. Hemiptera. 15. Capsus sp., skg.; 16. Phytocoris ulmi Z., skg. D. Hymeno-
ptera. (a) Apidae: 17. Halictus maculatus Sm. 9, freq., skg. and po-cltg.;
18. Prosopis confusa Wy/. 6; 19. P. variegata #. 9 and 4, skg. and po-dvg.,
also in copuld on the heads. (4) Sphegidae: 20. Cerceris variabilis Schr. 9.
(c) Chrystdidae: 21. Hedychrum lucidulum F. $. (d) Zenthredinidae: 22. Tarpa
cephalotes F., skg. (?). E. Lepidoptera. (a) Rhopalocera: 23. Melitaea athalia
Esp., skg.; 24. Thecla spini S.V., do. (6) Sphingidae: 25. Zygaena sp., do.

Loew noticed the Sphegid Dinetus pictus ¥. $ in the Berlin Botanic Garden.

1474. T. Parthenium Sch. Bip. (=Chrysanthemum Parthenium Bernh.,
Matricaria Parthenia Z., and Pyrethrum Parthenium Sm.). (Knuth, ‘BL u. Insekt.
a. d. nordfr. Ins., pp. 93, 159-)—In this species the style of the white female ray-
florets has no apical sweeping-hairs, but the yellow disk-florets possess them just like
Matricaria Chamomilla, though the stylar branches do not diverge so much in the
second stage of anthesis (cf Fig. 202, 9). As Kirchner remarks (‘Flora v.
Stuttgart,’ p. 711), they do not project at all from the corolla after the anther-
cylinder has become retracted.

The following were recorded by the observers, and for the localities stated._—

Herm. Miiller (H. M.) and Buddeberg (Budd.), (Herm. Miiller, ‘ Fertilisation,’
p- 331, ‘ Weit. Beob.,’ III, p. 96) the bee Halictus smeathmanellus K. 9, skg. (H. M.,
Budd.), an Evaniid (Foenus sp.), skg. (H. M., Budd.), and the Sphingid Sesia tipuli-
formis C/., skg. (H. M.). Schletterer (Pola), 3 bees (1. Halictus levigatus A. 9;
2. H. patellatus M/or.; 3. Andrena carbonaria Z.).

1475. T. alpinum Sch. Bip. (=Chrysanthemum alpinum Z.). (Herm. Miller,
‘Alpenblumen,’ pp. 430-2.)—This species is gynomonoecious with protandrous
hermaphrodite florets. The yellow disk-florets greatly exceed 100 in number, and
cover a circular area 10 mm. in diameter, which is extended by the white ray-florets
(about 30 in number) to 30-4 mm. The florets develop in centripetal order. The
possibility of self-pollination is ensured.

Visitors.—Herm. Miiller observed 3 beetles, 35 flies, 4 Hymenoptera, and
4 Lepidoptera.

Loew saw a Muscid (Anthomyia sp.) on the Piz Umbrail (‘ Bliitenbiol. Floristik.’
P- 397):

COMPOSITAE 625

1476. T. atratum Sch. Bip. (=Chrysanthemum atratum /acg., and C. coronopi-
folium V2i/.). (Herm. Miiller, ‘Alpenblumen,’ p. 432.)—

The flower mechanism of this species agrees completely with that of the
last one.

Visttors.—Herm. Miiller observed 7 flies.

1477. T. macrophyllum Sch. Bip. (=Chrysanthemum macrophyllum Walds?.
e¢ Kit, and Pyrethrum macrophyllum Wil/d.).—

Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (ca)
Dermestidae: 1. Anthrenus
scrophulariae Z. (4) Sca-
rabaedae: 2. Cetonia
aurata Z., freq. B. Di-
ptera. Muscidae: 3. Lu-
cilia caesar Z. C. Hy-
menoptera. (2) Ap:dae:
4. Apis mellifica Z. ¥, skg.
and po-cltg.; 5. Prosopis
communis JVy/, 9, skg.
(4) Vespidae: 6. Odynerus
parietum Z.

1478. T. partheni-
folium Sch. Bip. var.
pulverulentum (=Chrys-
anthemum praealtum Ven.
and Pyrethrum partheni-
folium Wiild.).—

Vistrors.—Loew ob-
served a Muscid (Ocyptera
brassicaria i), a Syrphid Fic. 201. Yanacetum alpinum, Sch. Bip. (after Herm. Miller).
(Syritta pipiens L.), anda A. A head (natural size). 2. Disk-floret in the first (male) stage (x 7).

iad C. Do. in the second (female) stage (x 7). D. Do. in section; the
bug (Heria es truncorum ovary omitted(x 7). &. Ray-floret(x 7). F. Pollen-grain. a, anther-

oe 9) in the Berlin Botanic cylinder; ca, calyx; co, corolla; /, filaments; 7, nectary; ov, ovary;
oe 9, pollen ; s¢, stylar branches.
arden.

1479. T. Meyerianum Sch. Bip. (=Pyrethrum tanacetoides DC.).—

VisiTors.—Loew saw a beetle (Coccinella bipunctata Z.) in the Berlin Botanic
Garden.

457- Chrysanthemum L.

Ray-florets yellow or white in colour; otherwise as Anthemis.

1480. C. segetum L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.’ pp. 91-3,
159.)—The diameter of the golden-yellow heads of this species is 4~5 cm., of
which a third is occupied by the disk. The 12-16 ray-florets are female: the
stigma projects a little from the corolla-tube, which is 4 mm. long. The tongue
is almost at right angles to this tube; its length is 13-2 cm. and its breadth

DAVIS. ss

626 ANGIOSPERMAE—DICOTYLEDONES

almost 1cm. There are about 300 disk-florets. Each of these is 6-7 mm. long,
of which 2 mm. is taken up by the ovary, 23 by the corolla-tube, and 2 mm. by the
bell of the corolla. In the first stage of anthesis the apposed stylar branches,
covered with pollen, project from the bell; in the second stage the stigmatic
inner surfaces of these branches protrude slightly. The diameter of the spinose
pollen-grains is somewhat less than that of the median groove that traverses
the stigmatic surface. The stylar branches of the ray-florets bear shorter sweeping-
hairs than those of the disk-florets. It is noteworthy that in this species the
upper surface of the bell of the corolla in both kinds of floret is beset with
countless microscopic papillae. Should insect-visits fail the pollen falls automatically
on the expanded stigmatic surfaces. Warnstorf (Verh. bot. Ver., Berlin, xxxvii,
1896) describes the pollen-grains as yellow in colour, usually ellipsoidal, closely
spinose, on an average 37-5 u long and 30 p broad.

9 10

FIG. 202. 1-7. Chrysanthemum segetum, L. (from nature). (1) Disk-floret in the first (male)
stage; pollen is issuing from the anther-cylinder. (2) Do, in the second (female) stage. (3) Female
ray-floret. (4) Tip of the style of a disk-floret in the first (male) stage (with branches apposed), enlarged.
p, stigmatic papillae. (5) Do. in the second (female) stage (with branches curving outwards), greatly
enlarged. (6) Tip of the style of a ray-floret with diverging branches, enlarged. (7) Tip of stylar
branch of a disk-floret, seen from within; in the middle is a groove mixed with pollen-grains; greatly
enlarged.

8. Anthemts arvensis, L. (8) Tip of the style of a disk-floret with diverging and strongly recurved
branches.

9. Tanacetum Parthentum, L. (9) As 8, but branches much less divergent; a few pollen-grains
are seen.

10. Matricaria Chamomilla, L. (10) As 8, but branches less divergent.

Visitors.—I observed the following in the island of Féhr and at Kiel—

A. Diptera, (a) Muscidae: all skg.: 1. Lucilia caesar Z.; 2. L. cornicina F. ;
3. Scatophaga merdaria Z.; 4. Sepsis cynipsea LZ. (6) Syrphidae: all skg. and
po-dvg.; gs. Eristalis arbustorum Z.; 6. E. nemorum Z.; 7. E. tenax L.;
8. Helophilus pendulus Z.; g. Syritta pipiens Z. B. Hemiptera. Capszdae:
10. Calocoris roseomaculatus Deg.; 11. Lygus pabulinus Z.; 12. L. pratensis 7.
<. Lepidoptera. Rhopalocera: 13. Vanessa io L., skg.

COMPOSITAE 627

The following were recorded by the observers, and for the localities stated.—

Buddeberg observed the Sphegid Sapyga decemguttata Jur. 4, skg. (according
to Herm. Miller, ‘ Weit. Beob.,’ III, p. 86). Alfken (Bremen), the po-dvg. Muscid
Pyrellia cadaverina Z. Scott-Elliot (Dumfriesshire), a parasitic humble-bee, 2
hover-flies, and 4 Muscids (‘Flora of Dumfriesshire,’ p. 94).

1481. C. Leucanthemum L. (=Leucanthemum vulgare Zam., Tanacetum
Leucanthemum Sch. B:p.). (Herm. Miiller, ‘ Fertilisation,’ pp. 329-31, ‘ Weit. Beob.,’
III, p. 85, ‘Alpenblumen,’ p. 432; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893;
Loew, ‘ Bliitenbiol. Floristik,’ pp. 394, 397; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
Pp. 93, 195; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—In this species the
diameter of the yellow disk is 12-15 mm.; it is surrounded by a border of white
ray-florets as broad or broader, so that the total width of the head is 40 mm. or more.
Hermann Miiller gives the length of the corolla of the 400-500 tubular florets
as scarcely 3 mm.: the tongue of the usually 13 or 21 ray-florets (according to
Ludwig) is 14-18 mm. long and 3-6 mm. broad. Nectar rises into the bell of the
disk-florets, which is barely 1 mm. long. In the first stage of anthesis these florets
present pollen, in the second stage their diverging stigmas to insects, so that these
will necessarily effect numerous crossings during a single visit. The stylar mechanism
is the same as in the last species (cf Fig. 199, 6-8). There is also a similar possibility
of automatic self-pollination : if the pollen has not been removed by insects from the
sweeping-hairs it will fall from them upon the papillose stigmatic surfaces of the
diverging stylar branches, so as to effect autogamy. Warnstorf describes the pollen-
grains as yellow in colour, polyhedral, with strong spinose tubercles, 25-31 mm. in
diameter.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list
for Westphalia and Nassau.—

A. Coleoptera. (a) Cerambycidae: 1. Leptura livida #, very numerous
(H. M.); 2. L. testacea Z. (H. M., Fichtelgebirge); 3. Pachyta octomaculata F.
(H. M.); 4. Strangalia armata ds, po-cltg. (H. M. and Budd., Thuringia) ;
5. S.atra / (H.M.); 6. S. attenuata Z. (H.M.); 7. S. melanura Z., freq. (H.M.).
(6) Chrysomelidae: 8. Clytra quadripunctata Z.(H.M., Kitzingen). (c) Dermestidae :
g. Anthrenus pimpinellae /., po-dvg.(H.M.). (d) Elateridae: 10. Agriotes ustulatus
Schall., po-dvg. (H. M., Thuringia); 11. Athous niger Z. (H. M.). (e) Mordellidae:
12. Mordella aculeata Z., freq. (H.M.); 13. M. fasciata #. (H. M.). (/) WVete-
dulidae: 14. Meligethes, very common (H.M.). (g) Ocdemeridae: 15. Oedemera
podagrariae Z., po-dvg. (H.M., Thuringia). (4) Scarabaerdae: 16. Cetonia aurata
L. (H.M., Sauerland) ; 17. Gnorimus nobilis Z. (H.M.); 18. Trichius fasciatus Z.,
freq. (H. M.). (2) Zelephoridae: 19. Dasytes flavipes /. (H.M.); 20. Malachius
aeneus Z.(H.M.). (4) Cleridae: 21. Trichodes apiarius Z,(H.M.). B. Diptera.
(a) Bombyliidae: 22. Bombylius canescens Mzkan, skg. (H.M., Budd.). (4) Conoprdae :
23. Conops flavipes Z., skg. (H. M.); 24. Sicus ferrugineus Z., skg. (H. M.).
(c) Empidae: 25. Empis rustica #., skg. (H. M.). (d@) Muscidae: 26. Echinomyia
tessellata F.(H.M.); 27. Lucilia cornicina /. (H.M.); 28. L. sylvarum J/g. (H. M.);
29. Macquartia praefica Afg. (H. M.); 30. Musca corvina #. (H. M.); 31. Pollenia
vespillo #., po-dvg. and skg. (H. M.); 32. Pyrellia aenea Ze/t. (H.M.); 33. Scato-
phaga stercoraria Z., skg. (H.M.); 34. Sepsis sp., do. (H.M.). (e) Stratiomy:dae:
35. Nemotelus pantherinus Z., extremely numerous, skg. (H. M.); 36. Odontomyia
viridula ¥., very common, skg. (H.M.). (/) Syrphidae: 37. Cheilosia fraterna AZg.
(H.M.); 38. Eristalis aeneus Scop., very common, po-dvg. (H. M.); 39. E. arbustorum

$$ 2

628 ANGIOSPERMAE—DICOTYLEDONES

L., do. (H. M.); 40. E. horticola Deg., do. (H. M., Sauerland); 41. E. nemorum Z.,
do. (H. M.); 42. E. sepulcralis Z., do. (H. M.); 43. Helophilus floreus Z., po-dvg.
(H. ay 44. H. pendulus Z., do. (H.M.); 45. Melithreptus taeniatus J/g., po-dvg.
(H. M.); 46. Paragus bicolor /., po-dvg. (H. M., Budd.); 47. Pipiza lugubris F.
(H.M.); 48. Syritta pipiens Z., skg. (H. M.); 49. Syrphus nitidicollis AZg., po-dvg.
(H. M.); 50. Volucella pellucens Z. (H. M., Sauerland). C. Hymenoptera.
(2) Apidae: 51. Andrena nigroaenea K. 9, po-cltg. (H. M.); 52. A. schrankella
Nyl. 8, skg. (H. M., Budd.); 53. A. xanthura XK. 9, skg. (H. M.); 54. Bombus
terrester Z. ¥, skg. (H. M.); 55. Colletes daviesanus XK. 9 and 4, very common,
po-cltg. and skg.(H.M.); 56. Halictus albipes F. 4, skg. (H.M.); 57. H.cylindricus /.
g and 4, very common, skg. and po-cltg. (H. M.); 58. H.leucozonius Schr. 9, po-cltg.
(H.M.); 59. H. lugubris X. 9, do. (H. M., Budd.); 60. H. maculatus Sm. 9 and 6,
numerous, skg. and po-cltg. (H. M.); 61. H. rubicundus Chr. 9, po-cltg. (H. M.);
62. H. villosulus A. 9 and 4, po-cltg. and skg. (H. M., Budd.); 63. Prosopis
communis Wy/. 9 (H. M.); 64. Sphecodes gibbus Z. 9 and 4, all the vars. including
ephippius Z. (H. M., Budd.). (4) Ichneumonidae: 65. Several (H.M.). (c) Sphegidae:
66. Cerceris variabilis Schv. (H. M.); 67. Crabro cephalotes /. 9 (H. M.); 68. C.
cribrarius Z. 6, in large numbers (H. M.); 69. C. dives H—Sch. & (H. M.,
Budd.) ; 70. Oxybelus trispinosus /. (H. mM: 41. O. uniglumis Z., freq. (H. M.).
(ad) Tenthredinidae: 72. Abia sericea L. (H.M.); 73. Allantus nothus Kig., skg.
(H.M.); 74. A. scrophulariae Z.(H.M.); 75. Several undetermined sp. of Tenthre-
dinids(H.M.) D. Lepidoptera. (a) Woctucdae : 76. Anarta myrtilli Z., skg. (H.M.).
(6) Rhopalocera: 77. Epinephele janira Z., do. (H. M.); 78. Hesperia thaumas
Hyn., do. (H. M.); 79. Melitaea athalia Z’sp., do. (H. M.); 80. Pieris napi Z., do.
(H. M.); 81. Polyommatus phlaeas Z., do. (H. M.); 82. Syrichthus alveolus /Z2.,
do. (H. M.). (c) Sphingidae: 83. Ino statices /., freq. (H. M.).

The following were recorded by the observers, and for the localities stated—

Knuth (North Frisian Islands), Apis, 2 Bombus, one other bee, 7 Syrphids,
4 Muscids, and 2 Lepidoptera: also (Thuringia) the following (‘ Bliitenbiol. Beob. in
Thiiringen’)— A. Coleoptera. All po-dvg.: 1. Judolia cerambyciformis Schr. ;
2. Leptura maculicornis Deg.; 3. Strangalia melanura Z.; 4. Trichius fasciatus Z.,
freq. B. Diptera. (a) Musczdae: all skg.: 5. Anthomyia sp.; 6. Aricia basalis
Zett.; 7. Hydrotaea sp. (6) Syrphidae: all skg. and po-dvg.: 8. Melithreptus sp. ;
9. Syrphus annulipes Ze//, 9; 10. Volucella pellucens Z. C. Lepidoptera.
11. Epinephele janira Z., skg.; 12. Zygaena trifolii Zsp., do.: also (Helgoland)
2 Muscids, po-dvg. (Lucilia caesar Z. and Scatophaga stercoraria Z.) (Bot. Jaarb.
Dodonaea, Ghent, viii, 1896, p. 40). Alfken (Bremen), a bee (Eriades truncorum Z.
4, skg.) and a bug (Calocoris roseomaculatus Deg., very common, skg.). Krieger
(Leipzig), a bee (Halictus zonulus Sm.), a Sapygid (Sapyga clavicornis Z.), and a
Sphegid (Cerceris labiata #.). Warnstorf (Brandenburg), numerous small Staphylinid
beetles, transferring pollen. Loew (Silesia), the beetle Meligethes sp. (‘ Beitrage,’
p. 31): and (Riesengebirge) a Conopid (Conops quadrifasciatus Deg.), a bee (Prosopis
armillata Vy/. 9, skg,), and a butterfly (Melanargia galatea Z., skg.). Rossler (Wies-
baden), the Tineid Butalis laminella A.-Sch. von Fricken (Westphalia and E. Prussia),
the Chrysomelid Cryptocephalus vittatus /., freq. Loew (Switzerland), the Empid
Empis tessellata #. Herm. Miiller (Switzerland), 6 beetles, 20 flies, a bug, 7
Hymenoptera, and 34 Lepidoptera (‘ Alpenblumen,’ pp. 432-4). Delpino (Florence),
the Bombyliid Lomatia beelzebub /. (‘Ult. oss.,’ Atti Soc. ital. sc. nat., Milano,
xvii). Schletterer (Tyrol), the carpenter-bee Xylocopa violacea Z. Kohl (Tyrol),
the fossorial wasp Crabro cribrarius Z. MacLeod (Pyrenees), 3 Hymenoptera, a
Lepidopterid, 3 beetles, and 21 flies (Bot. Jaarb. Dodonaea, Ghent, iii, 1891,
pp. 360-2); also (Flanders) 5 Hymenoptera, 6 hover-flies, 14 other flies, 6 Lepi-
doptera, and 7 beetles (op. cit., v, 1893, pp. 425-6). Scott-Elliot (Dumfriesshire),
4 Muscids (‘Flora of Dumfriesshire,’ p. 93).

COMPOSITAE 629

Ludwig found a slug (Limax laevis Aiiler) in hundreds on the heads in wet
weather. (Cf Vol. I, pp. 79-80.)

Clessin also observed (according to von Jhering) a slug (Limax brunneus
Drap.) as an occasional agent of cross-pollination in Rio Grande do Sul.

458. Doronicum Tourn.

Disk-florets hermaphrodite ; stylar branches, immediately below their extreme
tip, surrounded by a circlet of sweeping-hairs directed obliquely upwards, the outer-
most being the longest; their inner surface completely beset with stigmatic papillae.
Ray-florets ligulate, female, the outer ones almost devoid of sweeping-hairs.

1482. D. macrophyllum Fisch. (Hildebrand, ‘U. d. Geschlechtsverhilt. b. d.
Compositen,’ pp. 25-6, Taf. II, Figs. 18-28.)—Hildebrand states that in this Persian
species there is not such a perfectly adapted arrangement of sweeping-hairs as in
most other Compositae, less because of the structure of the style itself than from the
fact that when the pollen is shed its tip is already some distance above the base
of the anther-cylinder. The ray-florets possess vestiges of the five stamens, and
their nectaries are as well developed as those of the disk-florets.

Visitors.—Loew observed an Elaterid (Limonius cylindricus Pays.) and a bee
{Halictus cylindricus /. 9, po-cltg.) in the Berlin Botanic Garden.

1483. D. Pardalianches L.—Hildebrand (loc. cit.) states that this species
resembles the last one. Here again the ray-florets possess vestigial stamens, and
their nectaries are as well developed as those of the hermaphrodite florets. According
to Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896), the pollen-grains are golden-
yellow in colour, spherical to ellipsoidal, with long spinose tubercles, 25-31 p»
in diameter.

VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Dermestidae: 1. Anthrenus scrophulariae Z. (6) Mite
dulidae: 2. Meligethes aeneus ¥., numerous. B. Diptera. (a) Muscidae: 3.
Anthomyia sp.; 4. Lucilia caesar Z. (4) Syrphidae: 5. Eristalis arbustorum Z.;
6. E. nemorum Z.; 7. Helophilus floreus Z.; 8. H. pendulus Z.; 9. Platycheirus
albimanus /. 6; 10. Syritta pipiens Z. C. Hymenoptera. Apidae: 11. Chelostoma
nigricorne JVyi. 4, skg.; 12. Heriades truncorum JZ. 9, po-cltg.; 13. Osmia fulvi-
ventris Pz. 9, do.

1484. D. cordatum Sch. Bip. (Warnstorf, Verh. bot. Ver., Berlin, xxxviii,
1896.)—In this species the ray-florets are female; their stigmas mature before those
of the hermaphrodite disk-florets. The latter develop in centripetal order, and the
stamens of the outermost ones are usually vestigial. The pollen-grains are dark-
yellow in colour, rounded to ellipsoidal, with long spines, 30-7 » in diameter.

1485. D. austriacum Jacq.—
VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A.Coleoptera. (2) Buprestidae: 1. Anthaxia quadripunctata LZ. (5) Witidulidae :
2. Meligethes sp. B. Diptera. (a) Muscidae: 3. Lucilia caesar Z. (4) Syrphedae :
4. Eristalis arbustorum Z.; 5. E. nemorum Z.; 6. Helophilus floreus Z.; 7. Syritta
pipiens Z. C. Hymenoptera. (a) Afidae: 8. Andrena fasciata Wesm. 3, skg.;
9. Halictus cylindricus F. 9, po-cltg.; 310. H. leucozonius Schr. 9, do.; 11. H.

630 ANGIOSPERMAE—DICOTYLEDONES

sexnotatus X. 9, do.; 12. Heriades truncorum Z. é skg., 9 po-cltg.; 13. Megachile
centuncularis Z. 3, skg.; 14. Osmia fulviventris Pz. 9, po-cltg.; 15. Prosopis
armillata Vy/. 9 and 6, skg.; 16. P. communis /Vyi. $, do.; 17. Sphecodes
ephippius Z. 9, do.; 18. S. gibbus Z. 9, do.; 19. Stelis aterrima Pz. 9, do.;
20. S. phaeoptera K.9, do. (6) Zenthredinidae: 21. Cephus sp.é. D. Lepidoptera.
Rhopalocera: 22. Pieris brassicae L., skg.

1486. D. caucasicum Bieb.—
Visitors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. (a) Coccinellidae: 1. Coccinella bipunctata Z. (6) Mitidulidae :
2. Meligethes sp. B. Diptera. SyrpAzdae: 3. Eristalis aeneus Scop. C. Hymeno-
ptera. Apidae: 4. Andrena nitida Fourcr. 9, skg. and po-cltg.; 5. Apis mellifica
L. ¥, skg. and po-cltg.; 6. Halictus nitidiusculus X. 9, po-cltg.

1487. D. plantagineum L.—

Visttors.—Loew observed a Dermestid (Anthrenus scrophulariae Z.), a bee
(Halictus leucozonius Schr. 9, po-cltg.), and a Sphegid (Cerceris variabilis Schr.
9) in the Berlin Botanic Garden.

459. Aronicum Neck.
The tip of the style in disk-florets closely: beset externally with long pointed
sweeping-hairs, internally (including the swollen margin) with stigmatic papillae.
Style of ray-florets similar, but sweeping-hairs shorter.

1488. A. Clusii Koch (including A. glaciale Rezchd.). (Herm. Miiller. ‘ Alpen-
blumen,’ pp. 437-8.)—-This species is gynomonoecious with protandrous hermaphrodite
florets. The orange-yellow heads possess a disk 14-20 mm. in diameter, which is
extended to 40-60 mm. by the ray-florets.

In the variety A. glaciale Rezchd. (=the species Doronicum glaciale Nym.),
according to Kerner (‘Nat. Hist. Pl. Eng. Ed. 1, II, p. 322), the receptive stigmas
of the outer florets are brought into the line of fall of the pollen of the inner ones by
gradual elevation of the common receptacle, so that automatic geitonogomy is effected.
Kerner adds that the stigmas of the ray-florets are mature several days before the
hermaphrodite florets of the same head open.

Visitors.—Herm. Miiller observed 11 flies, a bee, and 5 Lepidoptera.

1489. A. scorpioides DC.—Kermer says that this species behaves like the last
in regard to geitonogamy.

Visitors.—MacLeod (Pyrenees) observed a Lepidopterid and 2 Muscids (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 360).

460. Ligularia Cass.
1490. L. macrophylla DC. (=Senecio Ledebourii Sch. Bip.).—
Visttors.—Loew observed the bee Megachile centuncularis Z. 9, po-cltg., in the
Berlin Botanic Garden.
1491. L. speciosa Fisch. et Mey. (=Senecio Ligularia Hook.).—

VisiTors.—Loew observed the bee Coelioxys elongata Zep. 9, skg., in the Berlin
Botanic Garden.

COMPOSITAE 631

461. Arnica Rupp.

Ray-florets, female, ligulate; disk-florets, hermaphrodite, tubular. Style of the
latter covered on their whole outer surface, including the somewhat expanded tip,
with stiff sweeping-hairs directed obliquely upwards; closely beset internally with
stigmatic papillae. Styles of the ray-florets devoid of sweeping-hairs.

14492. A. montana L. (Herm. Miiller, ‘Alpenblumen,’ p. 436; Warnstorf,
Verh. bot. Ver., Berlin, xxxvii, 1896; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
PP. 93, 159-60.)—In this species the diameter of the orange-coloured heads is
7 cm. in the North Frisian Islands, almost a third of this being taken up by the
disk. Each of the 50-90 disk-florets possesses a corolla-tube 4 mm. long, expanding
into a bell 5 mm. deep, with teeth 1 mm. in length. The pollen pressed out of the
anther-cylinder projects from the bell during the first stage of anthesis; during the
second the style with its branches, which gradually roll back in a circle. There are
about 20 ray-florets with a corolla-tube 5 mm. deep, and a tongue 2-2} cm. long
and 5-7 broad. The style projects from the former, its branches being at first
appased, but subsequently diverging and displaying the stigmatic papillae.

Hermann Miller made similar observations in the Alps. He also found 50-90
disk-florets, making up a surface about 20 mm. in diameter, and about 20 ray-florets
increasing the breadth of the head to 60-70 mm. Warnstorf adds that the rounded
yellow pollen-grains are closely covered with spinose tubercles, and measure 31
in diameter on an average. When they are thrust out of the anther-cylinder of
a disk-floret by the apposed stylar branches they fall upon the teeth of the corolla,
which are beset at the edge with large blunt papillae that hold them fast. Soon after
the two long stylar branches have emerged from the anther-cylinder they diverge,
and roll back so that they not only reach the pollen-grains of the same floret
with their inner (stigmatic) surfaces, but frequently bring these into contact with
the pollen of the neighbouring florets. Automatic self-pollination and geitonogamy
are therefore both possible.

’ Kerner also states that autogamy ultimately takes place by the rolling back of
the stylar branches.

Visttors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (North Frisian Islands and at Tondern), 3 bees, 5 Lepidoptera, 10 flies,
and a beetle. Herm. Miiller (Alps), 3 beetles, 5 flies, 5 Hymenoptera, and 34
Lepidoptera. Loew (Switzerland), a Muscid (Spilogaster duplicata J/g.), and a
Syrphid (Cheilosia antiqua J/g.) (‘Beitrage, p. 58). Kriechbaumer (Alps), the
parasitic humble-bee Psithyrus quadricolor Zep. 6. Schiner (Austria), the Muscid
Tephritis arnicae Z. Schletterer (Tyrol), the Alpine humble-bee Bombus alticola
Krchb., and the parasitic humble-bee Psithyrus quadricolor Zep. The former was
also seen (Tyrol) by von Dalla Torre.

1493. A. alpina Olin et Landau.—This species is native to Greenland, arctic
America, Siberia, Spitzbergen, and Lapland. It. is slenderer and more delicate than
A. montana. The involucral bracts are often purple-red at their tips or entirely so ;
the bright-yellow ray-florets are generally twice as long as the involucre, but are
sometimes only of the same length (Abromeit, ‘Bot. Ergeb. von Drygalski’s Grén-
landsexped.,’ pp. 69-70). In Spitzbergen the species blooms from the beginning

632 ANGIOSPERMAE—DICOTYLEDONES

of September, and fruiting specimens have been observed (by Nathorst, 17.8.’82.
The pollen is abundant, and normally developed (Andersson and Hesselman,
‘ Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 10).

1494. A. Chamissonis Less.—

Visrrors.—Loew observed 2 Syrphids (Eristalis aeneus Scop., and E. tenax L.)

and 2 bees (Heriades truncorum ZL. 9, po-cltg.; and Megachile centuncularis Z. 9, do.)
in the Berlin Botanic Garden.

462. Cacalia L.

1495. C. hastata L.—The heads of this species exhale a strong odour of
honey.

Visrtors.—Knuth observed the following (12.9. ’97) in the Kiel Botanic Garden ;
all (except 5 and 7) very freq., and all skg. (‘ Bliitenbiol. Notizen’).—

A. Diptera. Syrphidae: 1. Eristalis intricarius Z.; 2. E.tenax Z.; 3. Syrphus
corollae F.; 4.S. ribesii Z.; 5. Volucella pellucens Z., occasional. B. Hymenoptera.
Apidae: 6. Apis mellifica Z. §; 7. Bombus agrorum F. 9, occasional ; 8. B. lapidarius

ZL. 8: 9. B. terrester Z. 9. C. Lepidoptera. Rhopalocera: 10. Vanessa io L.;
11. V. urticae Z.

463. Senecio L.

LrreraTure.—Hildebrand, ‘ U. d. Geschlechtsverhilt. b. d. Compositen,’ pp. 27-8,
Taf. II, Figs. 29-36 (S. populifolius).

Ray-florets female, ligulate, yellow in colour, sometimes wanting; disk-florets
hermaphrodite, tubular, yellow. Stylar branches with an apical bundle of sweeping-
hairs, quite covered with stigmatic papillae internally and marginally.

1496. S. vulgaris L. (Herm. Miller, ‘ Fertilisation,’ p. 336, ‘Weit. Beob.,’
III, p. 90; Knuth,‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 94, ‘Bloemenbiol. Bijdragen.’)—
Ligulate ray-florets are wanting in this species. Hermann Miiller says that there are
60-80 florets in a single head, with corolla-tubes 33-4 mm. in length and bells
1-13 mm. long. Nectar ascends into the bell and is therefore very easily accessible.
Owing to the small size of the heads (only 4 mm. in diameter) and the fact that the
plant is not very conspicuous, insect-visits are few. The pollen-grains brushed out of
the anther-cylinder by the sweeping-hairs which form a tuft at the end of the stylar
branches, partly remain clinging to the marginal papillae when the branches diverge,
and partly fall on their stigmatic inner surfaces. It follows that automatic self-
pollination regularly takes place, and this is certainly effective. Bateson found,
however, that plants resulting from cross-pollination were larger and more fertile than
those produced by autogamy.

VisiTors.—The following were recorded by the observers, and for the localities
stated. —

Knuth, on one occasion, a hover-fly (Melanostoma mellina Z.) po-dvg. Herm.
Miiller, during a period of 15 years, not infrequently saw a hover-fly (Syritta pipiens Z.)
and a bug (Pyrrhocoris apterus Z., skg.). Buddeberg (Nassau), 2 bees (Halictus
morio F. 9, skg.; and Heriades truncorum JZ. é, do.), Verhoeff (Norderney), the
hover-fly Syritta pipiens Z. MacLeod (Flanders), 4 Hymenoptera, 5 Syrphidae,
4 Muscidae, and a Lepidopterid (Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 427-8,
vi, 1894, p. 374). Scott-Elliot (Dumfriesshire), a Muscid (‘Flora of Dumfriesshire,’
p-97). Burkill (Yorkshire coast),a Muscid (Anthomyia sp.) (‘ Fertlsn. of Spring Fls.’).

COMPOSITAE 633

1497. S. viscosus L.—Kerner says that the stylar branches in this species
ultimately roll back in a semicircle so far that their stigmatic papillae touch the pollen
which remains clinging to the elongated pappus-hairs of the same floret.

VisiTors.—Buddeberg observed the bee Panurgus calcaratus Scop. 9 skg. and
po-cltg., 5 skg. (Herm. Miiller, ‘ Weit. Beob.,’ III, p. go).

1498. S. sylvaticus L. (Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ p. 94.}—
The diameter of the yellow heads of this species is only 5 mm. The few ray-florets
(9-12) possess very small tongues, which roll up in dry weather. The disk-florets
number about 40, and are 8 mm. in length, including the ovary, which is 2 mm.
long. The stylar branches, only the tips of which are covered with sweeping-hairs,
roll back in a semicircle during the second stage of anthesis, so that the pollen-grains
still clinging to the collecting-hairs fall upon the exposed stigmatic papillae, thus
effecting automatic self-pollination if crossing has not already taken place.

Visirors.—Buddeberg (Oberpfalz) saw 2 po-dvg. flies (Echinomyia magnicornis
Zett., and Melithreptus scriptus Z.) (Herm. Miiller, ‘ Weit. Beob.,’ III, p. 90).

Sickmann (Osnabriick) records the fossorial wasp Mellinus arvensis Z.

1499. S. Doronicum L. (Herm. Miller, ‘Alpenblumen,’ pp. 438-40.)—In
this species 100-200 female florets form an orange-yellow disk 10-20 mm. in
diameter. The ray-florets, about 20 in number, extend this to a star 36-58 mm.
broad. The male stage of the disk-florets lasts only for a very short time, for
the outermost row spread out their stylar branches before those adjoining them have
opened. The female stage persists longer, the stigmas of the outermost florets being
still receptive, when the central ones are just entering on this stage. Kerner says that
geitonogamy is possible, as in Aronicum glaciale, by the swelling up of the common
receptacle (cf p. 630).

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller, a beetle, 14 flies, 4 Hymenoptera, and 39 Lepidoptera. Loew
(Switzerland), Merodon cinereus /. ; also (Berlin Botanic Garden) 3 bees—1. Heriades
truncorum L. 9, po-cltg.; 2. Osmia fulviventris Pz. 9, do.; 3. Stelis phaeoptera K.
3, skg.

1500. S. paludosus L.—

Visitors.— Heinsius (Wageningen) observed the hover-fly Eristalis horticola
Deg. , and the beetle Meligethes aeneus /.

1501. S.nemorensis L. (Herm. Miller, ‘Weit. Beob.,’ ITI, p.go, ‘Alpenblumen,’
Pp. 440-1.)—

The heads in this species are made up of 10-13 disk-florets and 5-6 ray-
florets, and their total diameter is only 4-6 mm. As, however, 20-30 or even more
heads are aggregated into a loose corymb, the plant is moderately conspicuous.
Kerner states that automatic self-pollination may be effected by the rolling back of
the stylar branches.

Visttors.—Herm. Miiller observed 4 flies, 4 bees, and 10 Lepidoptera in the
Alps; and the following in Central Germany.—

A. Diptera. (2) Conopidae: 1.Conops scutellatus Mg.,skg. (6) Lepiidae: 2. Leptis
tringaria Z., skg. (c) Muscidae: 3. Aricia sp.; 4. Echinomyia fera Z., skg. (?).

634 ANGIOSPERMAE—DICOTYLEDONES

(2) Syrphidae: 5. Eristalis pertinax Mg., po-dvg.; 6. Volucella inanis Z., do.;
4. Xylota sp.,do. B. Hymenoptera. (a) Apidae: 8. Bombus hypnorum Z. 4,
skg.; 9. B. muscorum /. 9 and ¥, do.; 10. B. pratorum Z. 9 and ¥, do.; 11. B.
terrester Z. $ (Thuringia); 12. Halictus cylindricus /. 4, skg.; 13. H. lucidus
Schenck 9 and é, do.; 14. Psithyrus quadricolor Zep. $, do,; 15. P. vestalis Fourcr.
$, do. (4) Vesprdae: 16. Vespa rufa Z. ¥. C. Lepidoptera. Rhopalocera:
17. Erebia ligea Z., skg.

Frey saw the Tortricid moth Grapholitha hepaticana 77. in the Alps. Loew
observed the hover-fly Cheilosia canicularis Pz. in the Riesengebirge (‘Beitrage,’
p. 51), and the following in the Berlin Botanic Garden.—

A. Coleoptera. Coccinelitdae: 1. Coccinella quattuordecimpunctata Z. B.
Diptera. (a) Muscidae: 2. Echinomyia fera Z.; 3. E. tessellata #.; 4. Pyrellia
cadaverina Z.; 5. Sarcophaga albiceps M/g.; 6. S. carnaria LZ. (5) Syrphidae:

7. Eristalis aeneus Scop.; 8. E. arbustorum Z.; 9. E. nemorum Z.; 10. E. sepulcralis
L.; 11. E. tenax Z.; 12. Syritta pipiens Z.; 13. Syrphus balteatus Deg.; 14. S.

FIG. 203. Senecto carniolicus, Willd. (after Herm. Miller). 4. Normal ray-floret (pappus omitted).
B-E. Various other forms of ray-floret. Cc’. Anther-cylinder of C. D’. Anthers and style of D.
¥. Normal disk-floret (ovary omitted).

cinctellus Ze//.; 15. S. ribesii Z. C. Hymenoptera. (a) Apidae: 16. Apis
mellifica LZ. ¥, skg. and po-cltg.; 17. Bombus terrester Z. 8, skg.; 18. Halictus sp.
8, do.; 19. Heriades truncorum Z. 9, po-cltg.; 20. Prosopis armillata Myl. 3,
skg.; 21. Stelis phaeoptera K., do. (6) Sphegidae: 22. Ammophila sabulosa Z.
D. Lepidoptera. Rhopalocera: 23. Pieris brassicae L., skg.

1502. S. saracenicus L. (Knuth, ‘ Bliitenbiol. Herbstbeob.’)—
VisiTors.—I observed Apis, Vanessa io Z., and 2 hover-flies (Eristalis nemorum
L., and Syritta pipiens Z.), all skg., in the Kiel Botanic Garden.

1503. S. Fuchsii C. C. Gmel.—Kerner states that autogamy may be effected
in this species in the same way as in S. nemorensis.

COMPOSITAE 635

Vistrors.—Loew saw the butterfly Polyommatus phlaeas Es skg., in the Berlin
Botanic Garden.

1504. S. carniolicus Willd. (Herm. Miiller, ‘Alpenblumen,’ pp. 441-2.)—
In this species the diameter of the heads is 20-30 mm., and 3-10 of them are
aggregated into a corymb. Each head consists of 5-10 disk-florets, and usually
3-5 ray-florets, though the latter may be entirely wanting. The ray-florets are
remarkable as showing transitions to the tubular disk-florets (<f Fig. 203).

Visrrors.—Herm. Miiller observed a beetle, 3 flies, and 2 Lepidoptera.

1505. S. cordatus Koch (=S. alpinus Scop., and Cineraria cordifolia Gowan).
(Herm. Miiller, ‘Alpenblumen,’ p. 442.)—In this species there are 150-200 disk-florets,
making up a surface 12-18 mm. in diameter, which is extended by more than 20
ray-florets to form a star 50-60 mm. broad. The flower mechanism agrees with
that of S. Doronicum, as in which the common receptacle swells up and renders
geitonogamy possible (Kerner).

Visttors.—Herm. Miiller observed 2 flies and 2 Lepidoptera; and von Dalla
Torre saw the Alpine humble-bee Bombus alticola Arzechd. in the Tyrol.

1506. S. abrotanifolius L. (Herm. Miiller, ‘Alpenblumen,’ pp. 442-3.)—In
this species the heads are 25-35 mm. in diameter, and as many are aggregated to
form a continuous surface the plant is very conspicuous. The disk is 8-10 mm. broad,
and contains 60-80 florets. The stylar branches of these diverge so widely in the
second stage of anthesis that they almost touch the upper end of the anther-cylinder,
but Hermann Miiller never observed automatic self-pollination.

Visitors.—Herm. Miiller observed 2 beetles, 7 flies, 18 Lepidoptera, and
a Hemipterid. It is noteworthy that the orange-red flowers are specially sought
out by red butterflies (cf Crepis aurea and Hieracium aurantiacum).

1507. S. nebrodensis L.—Kerner’s investigations on the Blaser in the Tyrol
prove that this usually annual species may continue to live on if its seeds are not able
to ripen during the first year.

Vistrors.—Herm. Miiller saw 8 flies, 5 bees, and 11 Lepidoptera in the Alps
(‘Alpenblumen,’ p. 444).

Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (a) Muscidae: 1. Anthomyia sp. (6) Syrphidae: 2. Eristalis

arbustorum Z.; 3. E. nemorum Z.; 4. Syritta pipiens Z. B. Hemiptera. 5.
Pyrrhocoris apterus Z. C. Hymenoptera. Afidae: 6. Halictus nitidiusculus X.

9, po-cltg. :

1508. S. macrophyllus Bieb.—

Visirors.—Loew observed the following in the Berlin Botanic Garden.—

A. Coleoptera. Coccinellidae: 1. Coccinella bipunctata Z. B. Diptera.
(a) Muscidae: 2. Echinomyia fera Z.; 3. Pyrellia cadaverina Z.; 4. Sarcophaga
camaria Z. (é) Syrphidae: 5. Eristalis arbustorum Z.; 6. E. nemorum Z.;
7. Helophilus floreus Z.; 8. H. trivittatus /.; 9. Syritta pipiens Z. C. Hymeno-
ptera. Apidae: 10. Apis mellifica Z., ¥, skg. and po-cltg.; 11. Halictus cylindricus
F. 9, skg.; 12. Psithyrus vestalis Mourcr. 6, do.

1509. S. JacobaeaL. (Herm. Miller, ‘Fertilisation,’ pp. 335-6, ‘ Weit. Beob.,’
III, p. 89; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.” pp. 94, 160; MacLeod, Bot.

636 ANGIOSPERMAE—DICOTYLEDONES

Jaarb. Dodonaea, Ghent, iii, 1891, v, 1893.)—This is gynomonoecious like the other
species. Hermann Miiller states that the disk is 7-10 mm. in diameter, and
consists of 60-80 florets with corolla-tubes 24-3 mm. long, and bells of the
same length. The 12-15 ray-florets increase the breadth of the head to about three
times that of the disk. Since the heads are aggregated into a moderately dense
cyme, the plant is conspicuous, and insect visitors are correspondingly numerous.

Visitors—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list
for Westphalia and Nassau.—

A. Coleoptera. Ocdemeridae: 1. Oedemera virescens Z., po-dvg. (H. M.).
B. Diptera. (a) Conopidae: 2. Zodion cinereum F., skg. (Budd.). (d) Empzdae:
3. Empis livida Z., very common, skg. (H. M.). (c) Muscidae: 4. Aricia incana
Wredem., skg. (H. M.); 5. Gymnosoma rotundata Z. (Budd.); 6. Lucilia sp., skg.
(H. M.); 7. Oliviera lateralis Pz., very numerous, skg. (H. M.); 8. Onesia floralis
R.-D., skg. (H. M.); 9. O. sepulcralis M/g., do. (H. M.); 10. Phasia analis 7.
(Budd.); 11. P. crassipennis /. (Budd.); 12. Pollenia rudis &, skg. (H. M.).
(d) Mycetophilidae: 13. Sciara thomae Z., skg. (H.M.). (¢) Stratiomyidae: 14.
Odontomyia viridula #., very common, skg. and po-dvg. (H. M.). (/) Syrphrdae:
15. Ascia podagrica F., do. (H. M.); 16. Cheilosia barbata Loew, do. (Budd.);
17. C,. praecox Zef/., very common (Borgstette, Tecklenburg); 18. C. soror Ze/t.
(H. M.); 19. Eristalis aeneus Scop., very common, skg. and po-dvg. (H. M.);
20. E. arbustorum Z., do. (H. M., Budd.); 21. E. nemorum JZ. do. (H. M.);
22. E. sepulcralis Z., do. (H. M.); 23. E. tenax Z., do. (H. M.); 24. Paragus
tibialis Fal/., skg. and po-dvg. (Budd.); 25. Syritta pipiens Z., very common, do.
(H. M., Budd.). C. Hymenoptera. (a2) Apidae: 26. Andrena denticulata X. 9,
po-cltg. (H. M.); 27. A. dorsata K., do. (Budd.); 28. A. fulvicrus X. 9, do. (H. M.);
29. Apis mellifica Z., do. (H. M.); 30. Bombus lapidarius Z. ¥ and 4, skg. and
po-cltg. (H. M.); 31. B. pratorum Z. ¥ and 4, do. (H. M.); 32. Halictus albipes /.
é, skg. (H. M.); 33. H. cylindricus / 9 and 4, do. (H.M., Budd.); 34. H. longulus
Sm. %, po-cltg. (Budd.); 35. H. maculatus Sm. 8, do. (H. M.); 36. H. malachurus
K. 9, po-cltg. (Budd.); 37. H. nitidus Schenck 8, skg. (H.M.); 38. H. villosulus K.
9, skg. and po-cltg. (Budd.); 39. H. zonulus Sm. 9, po-cltg. (Budd.); 40. Heriades
truncorum Z., 9 and 4, skg. and po-cltg. (H. M., Budd.); 41. Nomada ferruginata XK.
9, skg. (H. M.); 42. N. furva Pz. 9, do. (H. M.); 43. N. jacobaeae Pz. 4, in large
numbers, skg. (Budd.); 44. N. varia Pz. 9, do. (H. M.); 45. N. zonata Pz. 9, skg.
(H. M.); 46. Osmia spinulosa K. 9, po-cltg. (H. M., Thiir.) ; 47. Psithyrus campestris
Pz.%, skg.(H.M.). (6) Zenthredinidae: 48. Tarpa cephalotes /. (H. M., Thuringia).
D. Hemiptera. 49. Capsus sp., skg. (H. M.). E. Lepidoptera. (a) Rhopalocera:
50. Epinephele hyperanthus Z., skg. (H. M.); 51. Melitaea athalia Z., do. (Budd.);
52. Polyommatus phlaeas Z., do. (H. M.). (4) Sphingidae: 53. Sesia asiliformis
Rott., skg. (H. M., Thuringia).

Willis observed the following in the neighbourhood of the south coast of
Scotland (‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera. (a) Cryptophagidae: 1. Antherophagus nigricornis F., po-dvg.
(4) Mitidulidae: 2. Meligethes sp., freq., po-dvg. B. Diptera. (a) Brbonidae:
3. Bibio pomonae F., freq., skg.; 4. Dilophus sp., skg. (6) Jusctdae: 5. Antho-
myia radicum JZ., freq., skg. and po-dvg.; 6. Calliphora erythrocephala J/g., skg.;
7. Hyetodesia incana Wied., do.; 8. Lucilia caesar L., freq., skg.; 9. L. sericata Mg.,
do.; ro. Morellia sp., skg.; 11. Mydaea sp., freq., skg. and po-dvg.; 12. Oliviera
lateralis J”, freq., skg.; 13. Phytomyza geniculata Macg., po-dvg.; 14. Sarcophaga
carnaria Z., skg.; 15. Scatophaga stercoraria Z., freq., skg.; 16. Trichophthicus
cunctans JMg., skg. (c) Syrphidae: 14. Aretophila mussitans F”, skg.; 18. Cheilosia
sp.,do.; 19. Eristalis aeneus Scop., freq., skg. and po-dvg.; 20. E. horticola Deg., do. ;

COMPOSITAE 637

21. E. pertinax Scop., do.; 22. Helophilus pendulus Z., skg.; 23. Sphaerophorea
scripta Z., skg. and po-dvg.; 24. Syrphus balteatus Deg., po-dvg.; 25. S. ribesii Z.,
po-dvg.; 26.S.topiarius M/g.,skg. C. Hemiptera. 27. Acocephalussp.; 28. Antho-
coris sp., freq. ; 29. Calocoris bipunctatus /., freq. D. Hymenoptera. (a) Apidae:
30. Andrena nigriceps K., skg.; 31. Apis mellifica Z., freq., skg.; 32. Bombus
agrorum f”, do.; 33. B. cognatus Sveph., skg.; 34. B. hortorum Z., freq., skg. ;
35- B. lapidarius Z., skg.; 36. B. pratorum Z., freq., skg.; 37. Halictus albipes
&., do.; 38. H. rubicundus Chr., do.; 39. Psithyrus quadricolor Lep., do.
(6) Formicidae: 40. Myrmica rubra L., do. (c) Ichneumonidae: 41. Several sp.
(2) Vespidae: 42. Odynerus pictus Curt, skg. E. Lepidoptera. ll skg.:
(a) Microlepidoptera: 43. Crambus sp., freq.; 44. Choreutis myllerana F.; 45.
Plutella cruciferarum Ze/t.; 46. Simaéthis fabriciana Sveph., freq. (6) Noctuidae:
47. Charaeas graminis Z. (c) Rhopalocera: 48. Epinephele janira Z.; 49. Pieris
rapae Z.; 50. Polyommatus phlaeas Z., freq.; all skg.

Alfken and Hoppner (H.) saw the following bees at Bremen.—

1. Andrena convexiuscula X. 9, rare; 2. A. denticulata K.9; 3. A. fuscipes
K. (H.); 4. Coelioxys elongata Zep. 9, skg.; 5. Colletes picistigma Zhs. 9 (H.);
6. Dufourea halictula Vy. 9 (H.); 7. Eriades nigricornis My/. 9 and 5; 8. E.
truncorum ZL. 9; 9. Halictus calceatus Scop. 9 and 4, very common, skg. and
po-cltg.; ro. H. flavipes /., very common, skg. and po-dvg.; 11. H. morio F., not
infrequent; 12. Nomada jacobaeae Pz. 9 and 4, freq., skg.; 13. Stelis breviuscula
Nyl.3. (6) Sphegidae: 14. Crabro (Entomognathus) brevis v. d. LZ. 9 and 3.

Schenck gives the following bees for Nassau.—

1. Andrena carbonaria £.; 2. A. denticulata K.; 3. A. flavipes Pz., 2nd gen.;
4. Colletes fodiens K.; 5. Epeolus variegatus Z.; 6. Eriades truncorum Z.;
7. Halictus albipes 7.8; 8. H. calceatus Scop.; 9. H. flavipes #.; 10. H. levigatus
K.8; 11. H. nitidiusculus X.; 12. H. pauxillus Schenck 9 and $; 13. H. rubicundus
Chr.é; 14. H. tetrazonius Aig. 8; 15. Nomada furva Pz.; 16. N. jacobaeae Pz. ;
17. N. lineola Pz.; 18. N. rhenana MZor.; 19. N. roberjeotiana Pz.; 20. N. rufi-
cornis Z.; 21. N. sexfasciata Pz.9 and 4; 22. N. solidaginis Pz.; 23. N. zonata
P2z.; 24. Osmia spinulosa K.

The following were recorded by the observers, and for the localities stated.—

Knuth (North Frisian Islands and Kiel), Apis, 2 Bombus, 4 Syrphids, and
3 Muscids. Gerstacker (Berlin), 2 fossorial wasps (Oxybelus lineatus Dahid., and
O. sericatus Zep.). Wiistnei (Alsen), 2 bees (Andrena listerella X., in this species
only, and Nomada roberjeotiana Pz.). Alfken (Juist), a fossorial wasp (Mellinus
arvensis Z., rare), a butterfly (Hipparchia semele Z.), and a Noctuid moth (Plusia
gamma Z.). Sickmann (Osnabriick), the following fossorial wasps,—1. Crabro
cribrarius £., very common; 2. C. scutellatus Schev.; 3. C. sexcinctus /, rare;
4.C. vagus Z.; 5. C. wesmaéli v. d. Z., infrequent; 6. Gorytes mystaceus Z.., freq. ;
4. G. quadrifasciatus /., freq.; 8. Pompilus viaticus Z., very common; g. Salius
exaltatus /, do.; 10. S. notatus Zep., freq. Schmiedeknecht (Thuringia), the
following bees—1. Andrena listerella K.; 2. A. nigripes K.; 3. Nomada brevicornis
Mocs.; 4. N. fabriciana Z., 2nd gen.; 5. N. ferruginata K.; 6. N. fucata Pz., 2nd
en.; 7. N. jacobaeae Pz.; 8. N. rhenana Jor.; 9. N. roberjeotiana Pz. Friese
(Baden (B.) and Mecklenburg (M.)), 4 bees—1. Halictus calceatus Scop. (B.) , not
infreq.; 2. Nomada fucata Pz. (M.), 2nd gen.; 3. N. jacobaeae Pz. (M.), not
infreq.; 4. N. solidaginis Pz. (M.), occasional. Réssler (Wiesbaden), 2 Tortricid
moths (Grapholitha hepaticana Zy., and G. trigeminana Syeph.). Schletterer and
von Dalla Torre (Tyrol), 3 humble-bees (1. Bombus hortorum Z. 9 and ¥; 2. B.
mastrucatus Gers/. J; 3. B. soroénsis #. ¥) and a bee (Andrena collinsona X. 6).
Loew (Brandenburg), the bee Nomada jacobaeae Pz. 9, skg. (‘Beitrage,’ p. 40) ;
also Silesia (op. cit., p. 32)—Hymenoptera. (a) Apzdae: 1. Andrena fulvicrus

638 ANGIOSPERMAE—DICOTYLEDONES

K. é and 9, skg., 9 also po-cltg. (4) Chrystdtdae: 2. Hedychrum lucidulum F.
(c) Sphegidae: 3. Ammophila sabulosa Z. 9, skg.; 4. Crabro cribrarius Z. 9, skg.;
5. Psammophila viatica Deg. 9, skg. H. de Vries (Netherlands), the bee Colletes
fodiens K., and the humble-bee Bombus terrester Z. (Ned. Kruidk. Arch., Nijmegen,
2. Ser., 2. Deel, 1875). MacLeod (Flanders), 5 hover-flies and 2 Lepidoptera
(Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 427): also (Pyrenees), a bee, a hover-fly,
and 2 Muscids (op. cit., iii, 1891, p. 360). Scott-Elliot (Dumfriesshire), 2 humble-
bees, 3 short-tongued bees, several flies, and a beetle (‘Flora of Dumfriesshire,’
p- 98). Saunders (S.) and Smith (Sm.) (England), 7 bees (1. Andrena denticulata
K.(S.); 2. A. tridentata K. (S., Sm.); 3. Colletes fodiens X. (S., Sm.); 4. Halictus
calceatus Scop. § (Sm.); 5. Nomada jacobaeae Pz. (S., Sm.); 6. N. roberjeotiana
Pz. (S., Sm.); 7. N. solidaginis Pz. (S., Sm.)), and a fossorial wasp (Oxybelus
mucronatus /’,).

1510. S. vernalis Waldst. et Kit—

Visirors.—Loew observed a beetle (Meligethes sp.) and a Muscid (Onesia
floralis #.-D.) in Brandenburg (‘ Beitrage,’ p. 40).

15. S. vulgaris L. x S. vernalis Waldst. et Kit. (Warnstorf, Schr. natw. Ver.,
Weringerode, xi, 1896.)—In this hybrid the diameter of the heads is about ro-12 mm.
(in S. vernalis 22-25 mm.); 8-12 small ray-florets (in S. vernalis 12-13), with a
semi-tubular tongue about 4 mm. long, spoon-shaped and concave above, 3-toothed,
with prominent marginal papillae. In S. vernalis the tongue is flat from its base,
8-10 mm. in length, slightly emarginate at its rounded tip, and not papillose at the
edges. The pollen-grains of the hybrid are golden-yellow in colour, very irregular,
and variable in size. They are roundish to ellipsoidal, closely beset with spinose
tubercles, 23-24 » in diameter. Those of S. vernalis are regular, on an average
37 «long and 24 « broad. The tongues of the ray-florets do not roll up towards
evening in the case of the hybrid, or only partly so, while those of S. vernalis all roll
back strongly. After the pollen has been shed the disk-flowers of the hybrid are
very soon exceeded in length by the pappus.

Warnstorf considers that the ray-florets of this hybrid are undoubtedly meta-
morphosed 5-toothed disk-florets, which represent an intermediate stage between
the two kinds of floret. This also appears from the fact that a fourth tooth is
sometimes visible immediately above the tubular part of the tongue, though as
a rule two teeth fuse to form the tube, while the elongated spoon-shaped region
possesses three teeth at its truncate end.

1512. S. erucifolius L—According to Kirchner (‘ Beitrage,’ p. 70), the flower
mechanism of this species is the same as in S. Jacobaea. The total diameter of the ©
head is about 30 mm., that of the disk ro mm. The number of ray-florets is 12-14.
The stylar branches of the disk-florets curve downwards to such an extent when they
begin to fade at the end of the second stage of anthesis that they touch the undivided
part of the style.

Visttors.—Kirchner observed the hover-fly Eristalis tenax Z., and Schenck the
following bees in Nassau.—

1. Epeolus variegatus Z.; 2. Nomada jacobaeae Pz.; 3. N. roberjeotiana Pz. ;
4. N. ruficornis Z.; 5. N. sexfasciata Pz.; 6. N. solidaginis Pz.; 7. N. zonata Pz.

1513. S. uniflorus All.—Kirchner (‘ Beitrage,’ p. 70) says that at Zermatt the
diameter of the heads of this species is 30 mm. The number of ray-florets is 12-15.

COMPOSITAE 639

The mechanism of the numerous disk-florets agrees with that of related species:
their stylar branches do not curve back for more than a semicircle towards the end
of anthesis, so that automatic self-pollination cannot take place.

1514. S. aquaticus Hill.—

VisiTors.—Heinsius (Holland) observed 2 Muscids (Lucilia cornicina /. 9, and
Scatophaga stercoraria Z. and 9) and a Syrphid (Eristalis tenax Z.).

In Dumfriesshire, a hover-fly and 5 Muscids were recorded (Scott-Elliot, ‘ Flora
of Dumfriesshire,’ p. 98).

464. Calendula L.

Ray-florets ligulate, female; stylar branches almost smooth externally, with
a stigmatic line on each margin internally. Disk-florets tubular and male; end
of the style conical and covered with sweeping-hairs, but no trace of stigmatic
papillae. Kerner states that the stylar branches of the ray-florets curve till they
touch the pollen of the disk-florets.

1515. C. arvensis L. (Hildebrand, ‘ U. d. Geschlechtsverhilt. b. d. Compositen,’
pp. 31-3, Taf. III, Figs. 1o-17.)}—This species is monoecious. Hildebrand says that
the lower sweeping-hairs of the disk-florets are longer than the upper ones. Linnaeus
observed at Upsala that the heads open at g a.m., closing again about noon.

Visitors. —Schletterer observed 2 bees (Andrena parvula X., and Halictus
calceatus Scop.) at Pola.

1516. C. officinalis L. (Hildebrand, op. cit., p. 33, Taf. III, Figs. 18-20.}—
Hildebrand states that this species agrees in mechanism with the last one, but the tip
of the style thickens abruptly and is covered with sweeping-hairs of nearly uniform
length. According to Kerner, the stigmas of the female ray-florets become receptive
before pollen is pressed out of the anther-cylinders of the male disk-florets. He
noticed at Innsbruck that the heads open between 9 and 1o a.m., closing again
between 4 and 5 p.m.

Visttors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel Botanic Garden), Apis, a humble-bee (Bombus sylvarum Z.),
3 hover-flies (1. Eristalis arbustorum Z.; 2. E. tenax Z.; 3. Syrphus ribesii Z.),
and a Muscid (Calliphora erythrocephala Mg.): all skg. (‘ Bliitenbiol. Bijdragen’).
Wiistnei (Alsen), 3 bees (1. Megachile centuncularis Z.; 2. M. circumcincta .;
3. Coelioxys acuminata /Vy/.).

2. Sus-oRDER CyNaREAE Less.

Stylar branches of the hermaphrodite florets usually combined into a pubescent
bifid cylinder with a ring of hairs or swelling at their base. Otherwise as
CoRYMBIFERAE.

465. Echinops L.

Stylar branches pubescent externally, with a ring of long sweeping-hairs at their
base ; papillose internally.

640 ANGIOSPERMAE—DICOTYLEDONES

1517. E. sphaerocephalus L. (Hildebrand, ‘U. d. Geschlechtsverhilt. b. d.
Compositen,’ pp. 46-8, Taf. VI, Figs. 1-3; Herm. Miiller, ‘ Fertilisation,’ pp. 336-7,
‘Weit. Beob., III, p.79; Knuth, ‘Bloemenbiol. Bijdragen,’ ‘Bliitenbiol. Herbstbeob.’)—
Hermann Miller says that the nectar rises into the bell after filling the corolla-tube
(3-6 mm. long), which is almost completely occupied by the style. The bell is, split
almost to its base into five linear lobes, and the nectar is therefore accessible even
to insects with a very short proboscis. After the style protrudes from the anther-
cylinder, its branches still remain apposed for a time, so that the pollen may be
removed by insect visitors before the stigmatic surfaces are spread out.

Fic. 204. Echtnops sphaerocephalus, L. (after Herm. Miller). (1) Floret at the end of the first

male) stage. (2) Do. in the second (female) stage. (3) Longitudinal section of the style and ot
the corolla-tube which surrounds it. @, filaments; 4, anthers; c, brush of sweeping-hairs; d, stigma;
é, nectary.

VisiTors.—I observed (Kiel Botanic Garden), Apis, skg., 2 humble-bees (Bombus
lapidarius Z. ¥ and 9, skg., and B. terrester Z. 9, do.), 2 butterflies (Pieris sp., and
Vanessa io Z.), 2 hover-flies (Eristalis sp., and Syritta pipiens Z.), and 2 Muscids
(Lucilia cornicina /., and Pollenia rudis ¥.). Loew saw Bombus terrester L. 9, skg.,
in the Berlin Botanic Garden.

Herm. Miiller (H. M.) and Buddeberg (Budd.) record the following for Thuringia
and Nassau.—

Hymenoptera. (a) Apzdae: 1. Bombus lapidarius Z. ¥, skg. (H. M.);
2. B. muscorum F. ¥, do. (H. M.); 3. B. sylvarum Z. §, do. (H.M.); 4. B.
variabilis Schmzedekn. ¥, do. (Budd.); 5. Halictus cylindricus #. 9 and 6, very
numerous, skg. (Budd.); 6. H. interruptus Pz. 4, skg. (Budd.); 7. H. maculatus
Sm. 9, do. (Budd.); 8. H. minutissimus X. 9, do. (Budd.); 9. H. morio F&. 9,
do. (Budd.); 10. H. quadricinctus #. 9 and 6, do. (H. M.); 11. H. rubicundus Cér.
$, do. (H. M.); 12. Prosopis communis /Vy/. 9, do.(Budd.). (6) Vespidae: 13. Polistes
gallica Z.(H.M.); 14. P. diadema Zyr., freq., skg. (H. M.).

COMPOSITAE 641

1518. E. Ritro L. (Sprengel, ‘Entd. Geh.,’ pp. 384-5; Warnstorf, Verh. bot.
Ver., Berlin, xxxviii, 1896.)—The heads of this beautiful species are 5-6 cm. in
diameter. The amethystine colour of those of the first order and the abundant
supply of nectar cause many insects to be attracted. The florets develop in
succession from the middle of the spherical heads downwards. The limb of the
corolla in the tubular florets is divided almost to its base into 5 slender, bright blue
lobes that spread out like a star above, while their white under-parts bulge outwards
to form an ovoid or spherical nectar reservoir. This is partly covered and protected
against rain by the hairy margins of the lobes. The blue tip of the style is densely
covered with small hairs adapted to hold for a considerable time the pollen pushed out
from the anther-cylinder, while the circlet of longer hairs immediately below the stylar
branches serves to remove it from the cylinder. The stigmatic surfaces on the inner
sides of the apposed stylar branches are quite undeveloped at this stage, and it is only
several days later when they have become receptive that the branches diverge. The
pollen has meanwhile been removed by insects or the wind, and the corolla-lobes
bend upwards and erect themselves. It follows that insects can only effect crossing,
transferring pollen from florets in the first (male) stage to others in the second
(female) stage. The pollen-grains are white in colour, with low spinose tubercles,
and of two kinds, rounded and about 56 » in diameter, or ellipsoidal, and 88 » long
and 50 p» broad.

1519. E. bannaticus Rochel.—
VisiTors.— Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syrphus albostriatus Fal/.; 2. S. cinctellus Ze#t. ;
3. S. corollae # B. Hymenoptera. (a) Apidae: 4. Apis mellifica Z. §, skg. and
po-cltg.; 5. Bombus terrester Z. 6, skg. (6) Sphegidae: 6, Philanthus triangulum
fF. 3.

1520. E. exaltatus Schrad.—
VistTors.—Loew observed the following in the Berlin Botanic Garden.—

A.Coleoptera. Scarabaeidae: 1.Cetonia aurataZ. B. Diptera. (2) Muscidae:
2. Chloria demandata #. (6) Syrphidae: 3. Eristalis nemorum ZL.; 4. Syrphus
balteatus Deg.; 5. S. corollae /.; 6. S. pyrastri Z. C. Hymenoptera. Apzdae:
4. Apis mellifica Z. 9, skg. and po-cltg.; 8. Bombus hypnorum Z. ¥, skg.; 9. B.
terrester Z. $ and ¥, skg. D. Lepidoptera. Rhopalocera: 10. Colias rhamni
L., skg.

466. Cirsium Adans.

Florets all tubular, hermaphrodite or dioecious. Stylar branches remain almost
or entirely closed; covered with small pointed sweeping-hairs externally, with a ring
of longer hairs at their base; their margins beset with stigmatic papillae.

1521. C. arvense Scop. (=Cnicus arvensis /offm.). (Herm. Miiller, ‘ Fer-
tilisation,’ pp. 340-2, ‘Weit. Beob.,’ III, p. 81, ‘Alpenblumen,’ p. 422; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 94-5, 160, ‘Weit. Beob. ii. BI. u. Insekt.
a. d. nordfr. Ins., p. 236, ‘Bl. u. Insekt. a. d. Halligen,’ ‘ Bloemenbiol. Bijdragen’ ;
Verhoeff, ‘B). u. Insekt. a. d. Ins. Norderney’; MacLeod, Bot. Jaarb. Dodonaea,
Ghent, v, 1893; Heinsius, op. cit., iv, 1892; Loew, ‘Bliitenbiol. Floristik,’
PP: 399, 394).—

DAVIS. Il T t

642 ANGIOSPERMAE—DICOTYLEDONES

This species is gynodioecious. Hermann Miiller states that there are more
than 100 lilac-coloured florets in the hermaphrodite heads. The corolla-tubes
are 8-12 mm. long, passing into bells 1-13 mm. deep, with 5 linear slightly
spreading lobes 4-5 mm. in length. Below, next the involucre, the head is
scarcely 8 mm. in diameter, but above, where the corolla-lobes diverge, it is
zo mm. or more. As the plant usually bears numerous heads, it is very
conspicuous, and the number of insect-visits is considerable. Visitors are the
more numerous owing to the fact that the nectar rises as high as the bell of
the corolla, so as to be accessible even to insects with a very short proboscis.
The stylar branches are almost 2 mm. in length, and remain apposed even

Fic. 205. Ctrstum arvense, Scop. (after Herm. Miller). (1) Floret in the first (male) stage.
(2) Pollen-grain in surface view (x 400). (3) Do. in optical section. (4) Floret in the second (female)
stage: pollen still clinging to the style. (5) Uppermost part of the style (x 88). a, anther-cylinder ;
b, long sweeping-hairs; c, short do.; d, stigmatic papillae; ¢, pollen.

during the second stage of anthesis, only their margins, which are beset with stig-
matic papillae, protruding. In the first stage a large quantity of pollen is swept out
of the anther-cylinders. The sticky pollen-grains are beset with pointed out-
growths, and readily adhere to the hairy under-sides of insects creeping over the
inflorescence: in favourable weather they are soon brushed away by visitors, so
that crossing is assured. But if insects do not pay their visits till after the stigmatic
papillae protrude, self-pollination may also occur. If such visits entirely fail,
self-pollination may also be automatically effected by the fall of pollen-grains from
the sweeping-hairs upon the stigmas. MacLeod states that the species is gyno-
dioecious on the Blankenberg dunes, and Warnstorf says this is also the case at

COMPOSITAE 643

Neu-Ruppin. Kerner describes the distribution of the sexes very aptly: pseudo-
hermaphrodite fruiting- and pollen-flowers upon different stocks.

Vistrors.—I observed the following in Schleswig-Holstein (S.-H.), Riigen (R.),
the dunes of Helgoland (H.), and Thuringia (T.).—

A. Coleoptera. (a) Cerambycidae: 1. Pachyta virginea Z.(T.). (5) Cocer-
nellidae: 2. Coccinella septempunctata Z. (H.). (c) Scarabaeidae: 3. Trichius
fasciatus Z., very common, po-dvg. or resting inertly on the inflorescences (T.).
(d) Telephoridae: 4. Psilothrix cyanea OL, freq. (H.). B. Diptera. (a) AMusctdae:
all skg.: 5. Anthomyia sp. (T.); 6. Aricia incana Wied. (S.-H.); 7. Calliphora
vomitoria Z. (H.); 8. Coelopa frigida Fall. (H.); 9. Lucilia caesar Z. (S.-H. and
H.); ro. Nemotelus uliginosus Z. (S.-H.); 11. Rivellia syngenesiae Fr. (S.-H.);
12. Sarcophaga carnaria L. (S.-H.); 13. Scatophaga merdaria Z. sae 14. S.
stercoraria LZ. (S.-H.). (4) Syrphidae: all skg.: 15. Eristalis aeneus Scop. 6 (S.-H.);
16. E. arbustorum JZ. (S.-H.); 17. E. intricarius Z. é (S.-H.); 18. E. nemorum Z.
(S.-H.); 19. E. pertinax Scop. (S.-H.); 20. E. tenax Z. (S.-H.); 21. Helophilus
pendulus Z. (S.-H.); 22. Melithreptus taeniatus J/g. (S.-H.); 23. Syritta pipiens Z.
(S.-H.); 24. Syrphus arcuatus Fall. 9 (H.); 25. S. ribesii Z. (S.-H.); 26. Volu-
cella bombylans Z., var. plumata Mg. (T.); 27. V. pellucens Z., freq. (H.)-
C. Hymenoptera. (a) Apidae: all but 35 skg. and po-cltg.: 28. Anthophora
quadrimaculata For. 9 (S.-H.); 29. Apis mellifica Z. (S.-H. and R.); 30. Bombus
agrorum /.9(T.); 31. B. cognatus Sveph. §(T.); 32. B. lapidarius Z. (S.-H. and
T.); 33. B. sorognsis /., var. proteus Gers. §(T.); 34. B. terrester Z. (S.-H. and
T.); 35. Psithyrus quadricolor Lep., only skg. (T.). (6) Sphegidae: 36. Ammophila
sabulosa Z., freq. (R.). (c) Zenthredinidae: 347. Allantus nothus lg. (S.-H.).
(d) Vespidae: 38. Odynerus parietinus Z.(S.-H.); 39. O. trifasciatus /. 6 (S.-H.);
40. Vespa vulgaris Z.(S.-H.). D. Lepidoptera. Allskg. (a) Moctudae: 41. Plusia
gamma L. (S.-H.). (4) Rhopalocera: 42. Argynnis adippe Z. (T.); 43. Epinephele
janira Z.(T. and S.-H.); 44. Pieris brassicae Z. (S.-H.); 45. P. napi Z. (S.-H.);
46. P. rapae Z. (S.-H.); 47. P. sp. (T.); 48. Polyommatus phlaeas Z. (S.-H.);
49. Vanessa urticae Z. (S.-H.). (c) Sphingidae: 50. Zygaena filipendulae Z. (S.-H.).

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list for
Westphalia (W.), Nassau (N.), Thuringia (T.), and the Bavarian Oberpfalz (O.).—

A. Coleoptera. (a) Carabidae: 1. Levia crux-minor Z. (H. M., O.).
(3) Cerambycidae: 2. Leptura testacea £., po-dvg. (H. M., O.); 3. Strangahia
melanura Z., po-dvg. (H. M., O.). (c) Chrysomehdae: 4. Cryptocephalus sericeus
Z., resting inactively on the florets (H. M., O.). (d) Clerzdae: 5. Trichodes
apiarius Z. (H. M., O.). (e) Curculionidae: 6. Bruchus sp. (H. M.); 7. Larinus
jaceae Z. (H.M., T.); 8. L. obtusus Sch. (H. M., O.). (/) ilateridae: 9.
Agriotes gallicus Zac. (H. M., T.); 10. A. ustulatus Schall, (H. M., T.);
11. Corymbites holosericeus Ov. (H. M., O.). (g) Lycidae: 12. Dictyoptera
sanguinea Scop. (H. M., O.). (A) Mordellidae: 13. Mordella aculeata Z., freq.
(H. M.); 14. M. fasciata #., do. (H.M.). (2) Oedemeridae: 15. Oedemera poda-
grariae L., po-dvg. (H. M., Kitzingen). (4) Scaradaecdae: 16. Cetonia aurata L.,
dvg. the florets (H. M., O.); 17. Trichius fasciatus Z. (H. M.). ,(2) Zelephoridae
18. Telephorus melanurus /., extremely numerous, inserting its head into the
corolla-bells (H. M.). B. Diptera. (a) Bomédylidae: 19. Anthrax flava Mg.(H.M.,
T.). ‘(6) Conopidae: 20. Conops flavipes Z. (H. M.); 21. C. quadrifasciatus Deg.,
skg. (H. M., O.); 22. Physocephala rufipes /., skg.(H. M., O.and W.). (c) Empidae:
23. Empis livida Z., skg. (H. M.). (d) Muscidae: 24. Lucilia cornicina F., freq., skg.
(H. M.); 25. L. sericata Mg. (H.M.); 26. Musca cornicina F., freq., skg. (H. M.);
27. Ocyptera brassicaria /., do. (H. M.); 28. O. cylindrica #, do. (H. M.):
29. Olivieria lateralis #., skg. (H. M.); 30. Onesia floralis R-—D. (H. M.);
31. Platystoma seminationis /. (H. M.); 32. Sarcophaga carnaria Z., skg. (H. M.).

Tt 2

644 ANGIOSPERMAE—DICOTYLEDONES

(ce) Stratiomyidae: 33. Odontomyia viridula #., not infrequent, skg. (H. M.).
(/) Syrphidae: 34. Cheilosia oestracea Z. (H. M., Fichtelgebirge); 35. Eristalis
aeneus Scop., freq., skg. (H. M.); 36. E. arbustorum Z., do. (H. M.); 37. E. intri-
carius Z., do. (H.M.); 38. E. nemorum Z. (H. M., W. and O.); 39. E. sepulcralis
Z., do. (H. M.); 40. E. tenax Z., do. (H. M.); 41. Melithreptus taeniatus JZ., do.
(H.M.); 42. Syritta pipiens Z., freq. (H. M.); 43. Syrphus sp., freq., skg. (H. M.);
44. Volucella inanis Z., po-dvg. (H. M., Fichtelgebirge); 45. V. pellucens Z., po-dvg.
(H. M., Fichtelgebirge); 46. V. bombylans Z., var. plumata Z., po-dvg. (H. M.,
Fichtelgebirge). (g) ZYabanidae: 47. Tabanus bromius Z. (H. M., O.); 48. T.
rusticus Z. (H.M., T.). C. Hymenoptera. (a) Apzdae: 49. Andrena dorsata K.
g and 4, in large numbers, skg. (H. M.); 50. A. fulvicrus K. 9, skg. (H. M.);
51. A. gwynana &. 9 and 4, freq., skg. (H. M.); 52. A. nana K. 6, skg. (H. M.);
53. A. pilipes F. 8, skg. (H. M.); 54. A. bimaculata X,, var. vitrea Sm. 6 (H. M.,
Cassel); 55. Apis mellifica Z., in very large numbers, skg. and occasionally po-cltg.
(H. M.); 56. Bombus hortorum Z. 6, skg. (H. M.); 57. B. lapidarius Z. ¥, skg.
(H. M.); 58. Cilissa leporina Pz. $, freq., skg. (H. M.); 59. Dasypoda hirtipes /
g and 4, skg. and po-cltg., 6 freq. (H. M.); 60. Halictus albipes /. 6, skg. (H. M.,
Budd.); 61. H. cylindricus /. 9 and 4, freq., skg. (H. M.); 62. H. flavipes /. 9,
skg. (H. M.); 63. H. longulus Sm. 6, do. (H. M.); 64. H. maculatus Sm. 9, do.
(H. M., Budd.); 65. H. minutus A. 9, do. (H. M.); 66. H. nitidiusculus KX. 6, do.
(H. M.); 67. H. nitidus Schenck 5, do. (H. M.); 68. H. rubicundus Chr. 6, do.
(H. M.); 69. H. tarsatus Schenck 9, do. (H.M.); 70. Heriades truncorum Z. 9,
do. (H. M.); 71. Macropis labiata Pz. § (H. M., O.); 72. Nomada jacobaeae Pz.
9 and 4, freq., skg. (H. M.); 73. N. lineola Pz. 9 and 6, do. (H. M.); 74. N.
fabriciana Z., var. nigrita Schenck 4, skg. (H. M.); 75. N. roberjeotiana Pz.
9 and 6, do. (H. M.); 76. N. solidaginis Pz. 9 and 4, skg. (H. M.); 77. Prosopis
communis JVy/. 9, freq., skg. (H. M.); -78. P. confusa Wy/. 9 and 4, skg. (H. M.);
79. P. sinuata Schenck 3, do. (H. M.); 80. P. variegata /. 9 and 6, freq., skg. (H. M.,
Budd.); 81. P. sp. 6, skg. (H. M., O.); 82. Sphecodes gibbus Z. 9 and 4, do.;
represented by different varieties including ephippius Z.(H.M.). (4) CAryszdzdae:
83. Hedychrum lucidulum /. 9, skg. (H. M.). (c) Jchneumonidae: 84. Several sp.
(H. M.). (d) Sphegzdae: 85. Ammophila sabulosa Z., skg. (H. M.); 86. Bembex
rostrata Z. 9, do. (H. M.); 87. Cerceris arenaria Z. 9 and 4, not infrequent, skg.
(H.M.); 88. C. nasuta Dehld. 9 and 4, freq., skg. (H. M., Budd.); 89. C. variabilis
Schr. 9 and 4, very common, skg. (H. M.); go. Crabro alatus Pz. 9 and 4, very
numerous (H. M.); gr. C. cribrarius Z. 9 and 4, freq., skg. (H. M., W. and O.);
g2. C. vagus ZL. é, skg. (H. M., O.); 93. Dinetus pictus /. 9 and 4, do. (H. M.);
94. Gorytes quinquecinctus F., freq., skg. (H. M., O.); 95. Lindenius albilabris /.
g and 4, skg. (H. M.); 96. Oxybelus trispinosus /. 9, do. (H. M.); 97. O. uniglumis
£. gand , freq., skg. (H. M.); 98. Philanthus triangulum F. 8, skg. (H.M., W. and O.);
99. Salius sanguinolentus /., skg. (H. M.). (e) Zenthredinidae: 100. Allantus nothus
Kilg., skg. (H. M.); 101. Several undetermined species, do. (H. M.). (/) Vespidae :
102. Eumenes pomiformis /. 9 (H. M., O.); 103. Polistes diadema Zér. (H. M., O.).
D. Lepidoptera. All skg. (a) Mocturdae: 104. Hydroecia nictitans Beh., var.
erythrostigma Haw. (H. M.). (6) Rhopalocera: 105. Epinephele hyperanthus Z.
(H. M., Fichtelgebirge); 106. E. janira Z. (H. M., Fichtelgebirge and W.); 107.
Hesperia lineola O. (H. M., O.); 108. H. sylvanus £5. (H. M.); 109. Pieris
brassicae Z., freq. (H. M.); 110. Rhodocera rhamni Z. (H.M.); 111. Thecla
rubi Z. (H. M.); 112. Vanessa urticae Z. (H.M.). (c) Sphingidae: 113. Ino
statices Z. (H. M., Fichtelgebirge); 114. Zygaena carniolica Scop. (H. M., T.);
115. Z. minos S.- V. (H. M., Fichtelgebirge).

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), 4 beetles, 6 flies, 8 Hymenoptera, and 15 Lepidoptera.
Alfken (Juist)—A. Coleoptera. Coccznelltdae: 1. Coccinella septempunctata Z.

COMPOSITAE %, 645

B. Diptera. (a) Afuscidae: 2. Calliphora vomitoria Z.; 3. Cynomyia mortuorum
L.; 4. Lucilia caesar Z.; 5. Nemoraea radicum F. (6) Syrphidae: 6. Eristalis
arbustorum Z.; 7. E. tenax Z., very common; 8. Syrphus pyrastri Z. C. Hy-
menoptera. (a) Apzdae: 9. Bombus arenicola Zs. §, once; 10. B. hortorum Z. 9;
11. B. lapidarius Z. ¥, freq.; 12. B. ruderatus /. 5; 13. B. terrester Z. ¥, freq.
(4) Sphegidae: 14. Oxybelus mucronatus /.; 15. O. uniglumis Z., rare. D.
Lepidoptera. (a) Nymphalidae: 16. Argynnis aglaia Z.; 17. A. niobe J.
(6) Lycaentdae: 18. Lycaena icaris Roff.; 19. Polyommatus phlaeas Z. Verhoeff
(Norderney).—A. Diptera. (a) Muscidae: 1. Calliphora erythrocephala J/g. 9, not
infrequent; 2. Cynomyia mortuorum Z. 9 and 4; 3. Lucilia latifrons Schzz., very
common; 4. Sarcophaga carnaria Z.; 5. Scatophaga stercoraria Z. 9; 6. Stomoxys
calcitrans Z.9. (4) Syrphidae: 7. Eristalis arbustorum Z. 9, freq.; 8. E. tenax Z.
9, occasional; 9. Helophilus pendulus Z. 9; 10. Platycheirus manicatus J/g. 9,
occasional; 11. Syritta pipiens Z., occasional; 12. Syrphus balteatus Deg. 6;
13. S. pyrastri Z.8. B. Hymenoptera. (a) Apidae: 14. Bombus sylvarum ZL. 4,
occasional ; 15. Psithyrus vestalis Mourcr.8; 16. Sphecodes cirsii Verh. 6, occasional.
(6) Formicidae: 17. Formica fusca Z. %, freq. (c) Vespidae: 18. Odynerus
parietum Z. 9 and é. Alfken (Bremen), a Muscid (Aricia basalis Ze//.) and 11
bees—1. Andrena flavipes Pz. 9; 2. A. nigriceps A. 9; 3. Bombus agrorum F.
é and ¥; 4. B. derhamellus X.9 and é; 5. B. lapidarius Z.¥; 6. B. lucorum Z. 6;
7. B. proteus Gerst. $; 8. Osmia solskyi Jor. 9; 9. Prosopis hyalinata Sm. 9;
ro. Psithyrus rupestris /”. 9, skg.; 11. P. vestalis Fourcr. $. Sickmann (Osnabriick),
the fossorial wasp Passaloecus brevicornis A. Jfor., rare; (Wellingholthausen), do.,
and Cerceris arenaria Z. Schmiedeknecht (Thuringia), the bee Andrena fumipennis
Schmiedekn. $, and the humble-bee Bombus terrester Z. $. Schenck (Nassau), the
Sphegid Cerceris rybiensis Z., and 5 bees—1. Andrena austriaca Pz.; 2. A. florea F.;
3. Coelioxys conoidea fi/.; 4. Macropis labiata #.; 5. ‘Prosopis hyalinata Sy.
Loew (Silesia), the following (‘ Beitrage,’ p. 31).—A. Coleoptera. (a) Scarabaezdae :
1. Cetonia aurata Z., dvg. the florets. (4) Zelephoridae: 2. Rhagonycha.melanura /.
B. Diptera. (a) Conopidae: 3. Conops quadrifasciatus Deg.d and 9, skg. (4) Muscidae :
4. Nemoraea pellucida Mg., skg.; 5. N. strenua J/g., do. (c) Stratiomyzdae: 6.
Odontomyia hydroleon Z., skg.; 7. O. viridula #., do. (d@) Syrphidae: all skg.:
8. Eristalis intricarius Z.; 9. E. nemorum Z.; 10. Syritta pipiens Z.; 11. Volucella
bombylans Z. C. Hymenoptera. (a) Apidae: 12. Apis mellifica Z. ¥, skg.
(5) Chrysididae: 13. Hedychrum lucidulum Dahld.; 14. Holopyga amoenula Dahld.
(c) Scolitdae: 15. Scolia bicincta Ross.9 andé,skg. D. Lepidoptera. Rhopalocera :
all skg.: 16. Hesperia comma Z., skg.; 17. Melanargia galatea Z., skg.; 18.
Epinephele janira Z., skg.; 19. Pieris brassicae Z., skg.; (Brandenburg) (op. cit.,
p. 39),—Diptera. Syrphidae: 1. Eristalis arbustorum Z.; 2. E. nemorum Z.;
3. E. tenax Z.: (Brunswick) (op. cit., p. 50),—Diptera. Syrphidae: 1. Volucella
bombylans Z.; 2. V. pellucens Z.,skg. Kohl, the fossorial wasp Crabro cribrarius Z.
Schletterer (Tyrol), the humble-bee Bombus lapidarius Z.; (Pola), the bee Osmia
fulviventris Pz. Schiner (Austria), the Muscid Trypeta ruficauda F. Heinsius
(Holland).—A. Diptera. (a) Empidae: 1. Empis livida Z. and 9, very numerous.
(6) Muscidae: 2. Scatophaga stercoraria Z. 9. B. Hymenoptera. (a) Apidae:
3. Apis mellifica Z.¥%; 4. Macropis labiata #. 8; 5. Psithyrus quadricolor Leg. 6.
C. Lepidoptera. Rhopalocera: 6. Vanessa io L., persistently skg. H. de Vries
(Netherlands), 2 humble-bees (Bombus subterraneus Z., and B. terrester Z.) (Ned.
Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel, 1875. MacLeod (Flanders), 7 long-
tongued bees, 6 short-tongued bees, a saw-fly, a ruby-wasp, 7 true wasps, 15 hover-flies,
12 other Diptera, 6 Lepidoptera, and 2 beetles (Bot. Jaarb. Dodonaea, Ghent, v, 1893,
pp. 407-8): (Pyrenees), a humble-bee and a hover-fly (op. cit., ili, 1891, p. 350).
Smith (England), the bee Macropis labiata /. Scott-Elliot (Dumfriesshire), a humble-
bee, a saw-fly, a short-tongued bee, and several other bees (‘ Flora of Dumfriesshire,’
p- 100).

646 ANGIOSPERMAE—DICOTYLEDONES

1522, C. lanceolatum Hill. (=Cnicus lanceolatus Wel/d., and Carduus lanceo-
latus Z.). (Herm. Miiller, ‘ Fertilisation,’ p. 343, ‘ Weit. Beob.,’ III, p. 82; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 94, 169, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr.
Ins.,’ p. 236; Heinsius, Bot. Jaarb. Dodonaea, Ghent, iv, 1892; MacLeod, op. cit.,
ili, 1891, v, 1893; Loew, ‘ Bliitenbiol. Floristik,’ p. 390; Warnstorf, Verh. bot. Ver.,
Berlin, xxxviii, 1896.)—The florets of this species are bright purple in colour.
Hermann Miller states that their mechanism is similar to that of C. arvense.
The nectar, however, is less accessible, for there is a bell 4-6 mm. deep at the
end of the corolla-tube of 16-18 mm., so that to reach the nectar in the base of the
latter a much longer proboscis is required. Warnstorf gives somewhat different
measurements :—corolla-tube about 23 mm. long, and its limb, which possesses two
deep incisions and three shallower ones, 11 mm.; the style with its branches projects
8 mm. beyond the retracted anther-cylinder, so that its total length is 42 mm.
The pollen-grains are white in colour, spherical or ellipsoidal, coarsely spinose, about
56 » in diameter. The species is consequently chiefly adapted to the visits of
long-tongued bees. -

Visttors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list
of visitors for Westphalia and Nassau.—

A. Diptera. (a) Conopidae: 1. Physocephala rufipes /, skg. (H. M_.).
(4) Syrphidae: 2. Eristalis arbustorum Z., very common, po-dvg. and skg. (H. M.);
3. E. nemorum Z., do. (H. M.); 4. E. tenax Z., do. (H. M.). B. Hymenoptera.
(a) Apidae: 5. Apis mellifica Z. ¥, freq., skg. (H.M.); 6. Bombus agrorum /.
¥ and 4, do. (H. M.); 7. B. lapidarius Z. ¥ and @, do. (H. M.); 8. B. terrester Z.
3 and 4, do. (H. M.); 9. Halictus cylindricus /. 9 po-cltg., 6 vainly trying to suck
(Budd.); 10. H. maculatus Sm. 9, po-cltg. (Budd.); 11. H. malachurus &. 9, do.
(Budd.); 12. H. tetrazonius K/g. 9, do. (Budd.); 13. H. zonulus Sym. 4, vainly trying
to suck (Budd.); 14. Megachile maritima X. 9, po-cltg. (H. M.); 15. Psithyrus
campestris Pz. 6, skg.(H.M.); 16. Stelis aterrima Pz. 9, skg.(Budd.). (6) Vesprdae:
17. Polistes gallica Z., freq., skg.(?) (H. M.); 18. P. diadema Z/r., do. (H. M.).
C. Lepidoptera. hopalocera: 19. Hesperia sp., skg. (H. M.); 20. Pieris brassicae
L., freq., skg. (H. M.); 21. P. napi Z., skg. (Budd.).

The following were recorded by the observers, and for the localities stated.—

Loew (Mecklenburg), the bee Megachile lagopoda Z., po-cltg. (‘ Beitrage,’ p. 40).
Knuth (Schleswig-Holstein)—A. Diptera. (a) Muscidae: all po-dvg.: 1. Lucilia
caesar L. (6) Syrphidae: 2. Eristalis arbustorum Z.; 3. E.nemorum Z.; 4. E. tenax L.
B. Hymenoptera. Apzdae: all po-cltg. and skg.: 5. Apis mellifica Z.; 6. Bombus
cognatus S/eph. $; 7. B. lapidarius Z.; 8. B. terrester Z. C. Lepidoptera.
Rhopalocera: 9. Pieris brassicae Z., skg.: (Helgoland) (Bot. Jaarb. Dodonaea,
Ghent, viii, 1896, p. 40)—A. Coleoptera. TZelephoridae: 1. Psilothrix cyanea Ol.
B. Diptera. Muscidae: 2. Coelopa frigida Fall., po-dvg.; 3. Scatophaga sterco-
raria £., do. C. Lepidoptera. Rhopalocera: 4. Pieris brassicae Z., skg. D.
Orthoptera. 5. Forficula auricularia Z., dvg. the florets. Alfken (Juist), a bee
(Bombus hortorum ZL. 6, freq.), a Sphegid (Ammophila sabulosa Z.), a butterfly
(Pieris brassicae Z.), and a moth (Plusia gamma Z.): (Bremen), 6 bees—1. Bombus
agrorum /. $; 2. B. hortorum Z. 9; 3. B. sylvarum Z. ¥; 4. B. terrester Z. 5;
5. Megachile centuncularis Z. 9 and 6; 6. Psithyrus rupestris #. 9, skg. Herm.
Miiller (Alps), a beetle, 7 bees, and 8 Lepidoptera (‘Alpenblumen,’ pp. 425-6).
Schenck (Nassau), the parasitic bee Coelioxys conoidea Z//. MacLeod (Pyrenees),
3 long-tongued bees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 349): (Flanders),
Apis, 9 humble-bees, 3 hover-flies, and a Lepidopterid (op. cit., v, 1893, p. 404).

COMPOSITAE 647

H. de Vries (Netherlands), the humble-bee Bombus agrorum F. 4, and the parasitic
humble-bee Psithyrus vestalis Fourcr. (Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel,
1875). Heinsius (Holland), a hover-fly (Eristalis horticola Deg. 9), a bee (Halictus
leucozonius Schr. 9), and 2 butterflies (Epinephele janira Z., and Vanessa urticae L.).
Scott-Elliot (Dumfriesshire), 2 humble-bees, another long-tongued bee, an Empid,
3 hover-flies, 2 Muscids, and a Lepidopterid (‘Flora of Dumfriesshire,’ p. 100).

152g. C. palustre Scop. (=Cnicus palustris Wii/d.). (Herm. Miiller, ‘ Fertilisa-
tion,’ pp. 343-4, ‘ Weit. Beob.,’ III, pp. 82-3, ‘Alpenblumen,’ p. 425; Heinsius, Bot.
Jaarb. Dodonaea, Ghent, iv, 1892; Loew, ‘Bliitenbiol. Floristik,’ p. 394.}—This
species is gynodioecious. Hermann Miiller says that the purple-red florets are
intermediate between those of the two last as regards accessibility of nectar (bell
of corolla 23 mm.), and, consequently, number of insect visitors. Female stocks are
rarer than hermaphrodite ones. Warnstorf (Ver. bot. Ver., Berlin, xxxviii, 1896)
adds that in the middle of the upper part of every outer involucral bract there is
a swelling which secretes a sticky substance (of unknown function) during anthesis.
The tube of the corolla is about 7 mm. long, as also is its limb, beyond which
the style projects 4-5 mm. The upper parts of the filaments are hairy. The
pollen-grains are white in colour, spheroidal, coarsely spinose, on an average
52 » in diameter.

VisiTors.—Herm. Miiller observed the following in Central and South Germany.—

A. Coleoptera. (a) Cerambycidae: 1. Strangalia melanura Z., freq. (Sauerland).
(6) Elateridae: 2. Agriotes ustulatus Schall. (Sauerland). B. Diptera. (a) Conopidae :
3. Conops quadrifasciatus Deg., occasional, skg.; 4. C. scutellatus A/g., freq., skg. ;
5. Sicus ferrugineus Z., skg. (6) Musczdae: 6. Echinomyia fera Z. (c) Syrphidae:
4. Eristalis tenax Z., skg. and po-dvg.; 8. Rhingia rostrata Z.; 9. Syrphus ribesii Z. ;
10. S. tricinctus /a//. po-dvg.; 11. Volucella bombylans Z., do.; 12. V. inanis Z.,
skg. and po-dvg.; 13. V. pellucens Z., do. C. Hymenoptera, (a) Apidae:
14. Andrena coitana K. 9, skg.; 15. A. denticulata KX. 9, do.; 16. A. gwynana K.
3, do.; 17. Apis mellifica Z. Y, very numerous, skg. and po-cltg.; 18. Bombus
lapidarius Z. ¥ and 4, po-cltg. and skg.; 19. B. pratorum Z. 4, skg.; 20. B. rajellus
K. %, do.; 21. Halictus cylindricus #. 9 and 4, very numerous, po-cltg. and skg.;
22. H.sp. 4, skg.; 23. Heriades truncorum Z. é, do.; 24. Megachile centuncularis
F.%, do.; 25. M. maritima K.3, do.; 26. Psithyrus quadricolor Zep. 4, freq., skg. ;
24. P. vestalis Fourcr. 9, skg. (6) Sphegidae: 28. Cerceris labiata F’. 3, vainly trying
to suck; 29. Lindenius albilabris #. D. Lepidoptera. (a) Noctuidae: 30. Plusia
gamma Z., not infrequent, skg. (4) Rhopalocera: 31. Argynnis paphia Z., persistently
skg.; 32. Epinephele hyperanthus Z., skg.; 33. E. janira Z., do.; 34. Erebia ligea
L., freq., skg.; 35. Hesperia silvanus Esp.; 36. Pieris brassicae Z., numerous, skg: ;
37. P. rapae Z., numerous; 38. Vanessa urticae Z., in large numbers, skg.
(c) Sphingidae: 39. Zygaena minos S.-V,, skg.

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), 5 bees and 6 Lepidoptera. Loew (Brunswick) (‘ Beitrage,’
p. 50), 3 Chrysomelid beetles (1. Cryptocephalus bipunctatus Z.; 2. C. moraei Z.;
3. C. vittatus /.) and a Muscid (Herina frondeocentiae Z.): (Silesia) (op. cit., p. 31),
a hover-fly (Eristalis intricarius Z., skg.) and 2 hawk-moths (Zygaena achilleae Esp.,
skg.; and Z. minos S.-V., do.). Alfken (Bremen), 3 bees—1. Bombus agrorum F. 4 ;
2. B. distinguendus Mor. 9 and ¥; 3. Halictus zonulus Sm.9. Schmiedeknecht (Thu-
ringia), 3 bees—1. Bombus hypnorum Z. 8; 2. B. pratorum LZ. ¥ and $; 3. Psithyrus
quadricolor Zep., freq. Hoffer (Steiermark), the parasitic bee Psithyrus quadricolor
Lep.8. Schiner (Austria), 3 Muscids—1. Trypeta ruficauda /.; 2. T. winthemi A@.;
3. Urophora stigma Loew. Schletterer (Tyrol), 3 bees—1. Megachile ligniseca X. (also

648 ANGIOSPERMAE—DICOTYLEDONES

by von Dalla Torre); 2. M. pacifica Pz.; 3. Psithyrus quadricolor Zep. Heinsius
(Holland).—A. Diptera. (a) Empidae: 1. Empis livida LZ. 9. (4) Syrphidae: 2.
Volucella bombylans Z.é. B. Hymenoptera. Apidae: 3. Bombus agrorum F. 9;
4. B. scrimshiranus K. ¥. C. Lepidoptera. Rtopalocera: 5. Vanessa urticae £.
H. de Vries (Netherlands), the bees Bombus agrorum F. 6 and Apis mellifica LZ. ¥
(Ned.- Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel, 1875). MacLeod (Flanders),
13 long-tongued bees, 4 short-tongued bees, a fossorial wasp, 8 hover-flies, 2 Empids,
and 7 Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 404-5): Scott-Elliot
(Dumfriesshire), 2 humble-bees, a true wasp, and a hover-fly (‘ Flora of Dumfriesshire,
p. too).

1524. C. eriophorum Scop. (=Cnicus eriophorus Ro/h, and Carduus eriophorus
L.).—MacLeod says that the nectar is very deeply concealed in the purple florets of
this species (‘ Pyreneénbl.,’ pp. 349-50). The tube of the corolla is 20 mm. long,
and the bell 9 mm., with 5 lobes 4.5 in length. One of the five incisions between the
lobes is about 2 mm. deeper than the other four, enabling a humble-bee to thrust its
head for 1-2 mm. into the bell, and to reach the base of this if its proboscis is
4-8 mm. long. The deeply concealed nectar is only accessible to long-tongued bees
and Lepidoptera.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

MacLeod (Pyrenees), only humble-bees (6 species). Herm. Miiller (Thuringia),
the long-tongued bee Megachile lagopoda Z. 9 po-cltg. and skg., 5 skg.); (Alps),
2 humble-bees and 2 Lepidoptera (‘Alpenblumen,’ p. 425). Schiner (Austria),
2 Muscids—1z. Trypeta acuticornis Zvew ; 2. Urophora eriolepidis Loew, very freq.

1525. C. heterophyllum Hill (=Cnicus heterophyllus Rosh, and Carduus
heterophyllus Z.). (Herm. Miiller, ‘Alpenblumen,’ pp. 424-5.)—In each head
of this species there are 200-300 florets with red bells 8 mm. long. During the
first stage of anthesis pollen is swept out of the anther-cylinder; during the second
the stylar branches diverge a little at the end, and the papillose edges of their inner
surfaces swell up to some extent. Failing insect-visits, automatic self-pollination is
possible, for the stigmatic margins of the stylar branches project till they touch the
pollen that remains clinging to the sweeping-hairs.

Visirors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller, the humble-bee Bombus mesomelas Gers?., skg. and po-cltg.
Scott-Elliot (Dumfriesshire), 2 humble-bees and a hover-fly (‘Flora of Dumfriesshire,’
p- 101). Schneider (Arctic Norway), chiefly the humble-bee Bombus agrorum /.
(Tromse Mus. Aarsh., 1894). Loew (Berlin Botanic Garden)—A. Coleoptera.
Telephoridae: 1. Dasytes flavipes /, freq. B. Hymenoptera. <Apzdae: 2. Apis
mellifica Z. ¥, skg.; 3. Bombus hortorum Z. ¥, persistently skg.; 4. Osmia fulviventris
Pz. 9, po-cltg.

1526. C. acaule Wigg. (=Cnicus acaulis Weld, and Carduus acaulis Z.).
(Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896, pp. 38-9.)—Warnstorf describes
the flower mechanism of this species for Neu-Ruppin, where the purple-red florets
are always hermaphrodite. The tube of the corolla is 20-22 mm. long, and its limb
as much as 15 mm. The latter is cleft by 3 incisions about 5-6 mm. deep, and 2 of
about 10 mm., into 5 slender erect lobes, forming a sort of cap to the floret. The
anther-cylinder possesses basal hair-like appendages, and when pollen is pressed out

COMPOSITAE 649

of it from below by the apposed stylar branches, projects somewhat beyond the limb
of the corolla. But when the style has completely elongated, the cylinder is withdrawn
between the lobes of the corolla by contraction of the filaments. The outer surface
of the stylar branches is densely covered with very short sweeping-hairs, which are
scarcely to be seen even with the aid of a lens, except at the base of the branches,
where they are somewhat larger. In the second stage of anthesis the margins of the
inner surfaces of the stylar branches, which are beset with papillae, bend somewhat
outwards, enabling insects to effect cross-pollination, or rendering autogamy possible
should some pollen still cling to the sweeping-hairs. The pollen-grains are white in
colour, roundish to ellipsoidal, coarsely spinose, as much as 63 p in diameter.

In other localities the species is gynodioecious.

According to Ljungstrém (Bot. Not., Lund, 1884), the female stocks in Sweden
bear smaller heads than the hermaphrodite ones.

Visitors.—The following were recorded by the observers and for the localities
stated.—

Herm.. Miiller (Alps), 7 bees. and 6 Lepidoptera (‘Alpenblumen,’ p. 422).

Réssler (Wiesbaden), the Tineid moth Depressaria incarnatella Zell. MacLeod
(Pyrenees), 2 humble-bees (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 349-50).

_ 1527. C. rivulare Link (=Cnicus rivularis Wlid.).—
Vistrors.—Hoffer records Bombus lapidarius Z. & for Steiermark.

1528, C. oleraceum Scop. (=Cnicus oleraceus Z.). (Herm. Miiller, ‘ Fertilisa-
tion,’ p. 343; Knuth, ‘ Bliitenbiol. Herbstbeob.,’ ‘ Bliitenbiol. Floristik,’ pp. 260, 397 ;
Warnstorf, Verh, bot. Ver., Berlin, xxxviii, 1896.)—Warnstorf observed only her-
maphrodite florets at Neu-Ruppin. The tube of the corolla is about 15 mm. long,
and its limb 6-7 mm. The style projects 7-8 mm. from the tube. The anther-
cylinder possesses hair-like appendages at its base, and the upper halves of the
filaments are hairy. The pollen-grains are white in colour, roundish-ellipsoidal,
coarsely spinose, up to 62 mu in diameter.

E. Kohne (Verh. bot. Ver., Berlin, xxviii, 1886, pp. 6-7) has published an
interesting observation. He found in Pomerania that the yellowish-white heads were
visited by very large numbers of the brimstone butterfly (Rhodocera Rhamni Z.),
which in colour and form of wings presents a certain agreement with the pale-yellow
upwardly-directed tips of the involucral bracts. This would seem to be a case of
special protective resemblance (cf Vol. I, p. 144).

Visrrors.—Herm. Miiller (H. M.) in Westphalia and myself (Kn.) in Schleswig-
Holstein observed the following.—

A. Hymenoptera. Apidae: all skg.: 1. Apis mellifica Z. § (Kn., H. M.);
2. Bombus agrorum /. (Kn.); 3. B. lapidarius Z. (Kn.); 4. B. terrester Z. § and é
(H. M.); 5. Psithyrus vestalis Fourcr. (Kn.). B. Lepidoptera. (a) Rhopalocera:
6. Pieris sp. (Kn.) skg. (6) Moctu:dae: 7. Euclidia glyphica Z., skg. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Herm. Miller (Alps), 3 humble-bees (‘ Alpenblumen,’ p. 424). Loew (Alps),
the Muscid Spilographa neigenii Zoew. Wiistnei (Alsen), the humble-bee Bombus
latreillellus A. Alfken (Bremen), 2 skg. humble-bees—Bombus arenicola Zhs. ¥,
and B. proteus Gers/. 6. Schmiedeknecht (Thuringia), 2 bees—Psithyrus barbutellus
K. %, and Osmia solskyi A/or.9. Hoffer (Steiermark), 2 bees—Bombus lapidarius 7.

650 ANGIOSPERMAE—DICOTYLEDONES

é, and Psithyrus barbutellus X. 4, very freq. Schletterer and von Dalla Torre (Tyrol),
the bee Melecta luctuosa Scop. Loew (Berlin Botanic Garden), on the var. ama-
rantinum, a beetle (Cetonia aurata L., dvg. the florets), 2 hover-flies (Syrphus balteatus
Deg., and S. corollae .), and 2 bees (Bombus pratorum Z. 4, and Psithyrus vestalis
Fourcr. 8).

1529. C. oleraceum Scop. x C. acaule Wigg. (=C. decoloratum Koch).
(Warnstorf, Verh. bot. Ver., Berlin, xxxiii, 1896.)—This hybrid is gynodioecious.
The flower mechanism is like that of C. acaule. The limb of the corolla is white or
pale lilac in colour, and its tube about 10 mm. long. The stylar branches are white.
The pollen-grains are white, rounded, spinose, about 50 «in diameter.

VisiTors.—Loew observed a bee (Psithyrus vestalis Fourcr. 4, skg.) and a butter-
fly (Pieris brassicae Z., freq., skg.) in the Berlin Botanic Garden.

1530. C. acaule Wigg. x C. oleraceum Scop.—

VistTors.—Loew observed 2 bees (Bombus terrester Z. 4, skg.; and Psithyrus
campestris Pz. 4, do.) in the Berlin Botanic Garden.

1531. C. oleraceum Scop. x C. palustre Scop. (=C. hybridum Koch and
C. lacteum Koch).—In this hybrid the leaves are not decurrent, and there are no
glandular swellings on the involucral bracts. Warnstorf describes the pollen-grains as
white in colour, spheroidal or ellipsoidal, and varying greatly in size (from 37 » to
56 » in diameter) (Verh. bot. Ver., Berlin, xxxviii, 1896).

1532. C. spinosissimum Scop. (Herm. Miiller, ‘Alpenblumen,’ pp. 423-4.)
—This species is very prickly, and bears a number of yellowish-white heads, the
conspicuousness of which is increased by the involucral bracts being of the same
colour. The corolla-tube is 8-9 mm. in length, and ends in a bell 4-5 mm.
deep, with 5 lobes about 5 mm. long. The flower mechanism resembles that of
C. heterophyllum.

Vistrors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Alps), 6 beetles, 6 flies, 15 Hymenoptera, and 14 Lepidoptera.
Loew (Heuthal), a Hesperid (Hesperia comma Z., skg.), a Noctuid (Agrotis occelina
S.-V.), and a hawk-moth (Zygaena exulans Achw. ef Rein.). von Dalla Torre
(Otzthaler Alps), the humble-bee Bombus mastrucatus Gers/. 9. Schmiedeknecht
(teste Morawitz) and von Dalla Torre and Schletterer (Tyrol), the bee Osmia confusa
Mor. The last-named also records 3 humble-bees—1. Bombus alticola Archd. ¥

(po-cltg. in the heaviest rain); 2. B. hortorum ZL. 9; 3. the parasitic species Psithyrus
globosus £7.

1533. C. ochroleucum All. (=Cnicus ochroleucus Spreng.).—
VisiTors.—Herm. Miiller observed 4 humble-bees and a butterfly in the Alps
(‘Alpenblumen,’ p. 425).

1534. C. monspessulanum Hill (=Cnicus monspessulanus Roth).—MacLeod
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, pp. 350-1) says that the stem of this species
bears 3-4 purple-coloured heads, each 25-30 mm. in diameter. The tube of the
corolla is 7-8 mm. in length, and the bell 6-7 mm. deep, with lobes 3-4 mm. long.
A proboscis 6 mm. in length is therefore required to reach the nectar.

Visitors.—MacLeod (Pyrenees) observed 4 humble-bees, 9 Lepidoptera, and

COMPOSITAE 651

3 Syrphidae. Loew noticed the butterfly Pieris brassicae Z., skg., in the Berlin
Botanic Garden.

1535. C. glabrum DC. (=Cnicus spinosissimus Z.(?)).—The heads of this
species are yellowish-white in colour.

Vistrors.—MacLeod (Pyrenees) saw the humble-bee Bombus hortorum Z. ¥
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 352).

1536. C. serrulatum Bieb. (=Cnicus serrulatus B7eb).—
Visttors.—Loew observed the following in the Berlin Botanic Garden.—

A. Hymenoptera. (a) Apidae: 1. Psithyrus rupestris #.,skg. (5) Sphegidae :
2. Ammophila sabulosa Z. B. Lepidoptera. Rhopalocera: 3. Rhodocera rhamni
L., skg.; 4. Pieris brassicae Z., do.: and on the var. ucranicum Bess., 2 bees
(Psithyrus campestris Pz. $, skg.; and P. vestalis Pourcr. 8).

467. Kentrophyllum Neck.
1537. K. lanatum DC. et Duby (=Carthamus lanatus Z.).—
Vistrors.—Schletterer observed the rare bee Halictus quadrinotatus K.

468. Silybum Vaill.

Stylar branches covered externally by short sweeping-hairs, a ring of longer,
usually somewhat oblique ones, at their base. They diverge only at the tip, during
the second stage of anthesis.

1538. S. Marianum Gaertn. (=Carduus Marianus Z.). (Sprengel, ‘Entd. Geh.,’
pp. 371-2.)—The florets of this species are purple-red in colour. Hildebrand states
that the pollen-grains are sometimes imperfect in heads that mature early (U. d.
Geschlechtsverhilt. b. d. Compositen,’ pp. 60-—2).—

Visttors.—Buddeberg saw the following bees in Nassau (Herm. Miiller, ‘ Weit.
Beob.,’ III, p. 81).—

1. Chelostoma nigricorne /Vy/. 6, skg.; 2. Halictus sexcinctus /. 9, skg. and
po-cltg.; 3. H. tetrazonius Aig. 9, skg.; 4. Megachile fasciata Sm. 8, skg.;
5. Osmia adunca Panz.%, do.; 6. O. fulviventris Pz. 9, do.; 7. Stelis phaeoptera
K. 3, do.

469. Carduus L.

Florets tubular and hermaphrodite. Stylar branches diverge only at the tip;
externally a ring of sweeping-hairs at their base; margins of inner surfaces (which
appear at a later stage) beset with stigmatic papillae.

1539. C. crispus L. (Herm. Miller, ‘Fertilisation,’ pp. 338-9, ‘ Weit. Beob.,’
Ill, p. 83; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 95, 161; Heinsius, Bot.
Jaarb. Dodonaea, Ghent, iv, 1892; Loew, ‘ Bliitenbiol. Floristik,’ p. 395.)—Hermann
Miller says that in this species 35-80 bright purple florets are aggregated into a head
which is scarcely 10 mm. in diameter below, but they curve outwards in such a way
that a red surface 25-30 mm. across is presented above. The corolla of each floret
expands into a ventricose bell, 23-3 mm. long, with linear slightly diverging lobes
4-53 mm, in length. The flower mechanism agrees in other respects with that
of the hermaphrodite florets of Cirsium arvense, but, owing to the depth of the

652 ANGIOSPERMAE—DICOTYLEDONES

bell, the number of insects to which the nectar is accessible is necessarily more
limited.

Purely female stocks were observed by Ljungstrom in Sweden (Bot. Not., Lund,
p. 675).

VistTors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list.—

A. Diptera. (a) Empidae: 1. Empislivida Z., freq., skg.(Budd.). (2) Muscedae:
2. Cynomyia mortuorum Z., skg.(Budd.). (c) SyrpAidae: 3. Eristalis arbustorum Z.,
skg. and po-dvg. (Budd.); 4. E. tenax Z., freq., skg. and po-dvg. (H. M.). B. Hy-
menoptera. Apidae: 5. Andrena gwynana K. 9, skg. (Budd.); 6. Apis mellifica
L., ¥, freq., skg. (Budd.); 7. Bombus agrorum #., do. (H. M. and Budd.); 8. B.
lapidarius Z. ¥ and é, do. (H. M. and Budd.); 9. B. terrester Z. 9, skg. (H. M. and
Budd.) ; 10. Chelostoma nigricorne WVy/. 8, do. (Budd.); 11. Coelioxys conoidea 7.
9, do. (Budd.); 12. Halictus albipes /. 4, do. (Budd.); 13. H. cylindricus /. 6 and
9, do. (H. M. and Budd.); 14. H. leucozonius Schr. 9, do. (Budd.) ; 15. H. sexnotatus
XK. 9, do. (Budd.); 16. Megachile lagopoda K. é and 9, do. (Budd.); 17. Osmia
fulviventris Pz. 9, po-cltg. (H.M.); 18, Psithyrus barbutellus X. é, skg. (Budd.); 19.
Stelis aterrima Pz. 9, do. (H. M. and Budd.). C. Lepidoptera. (2) Pyralidae:
20. Eurycreon verticalis Z., skg. (H. M.). (6) Rhopalocera: 21. Hesperia comma L.,
skg. (Budd.); 22. Melanargia galatea L., freq., skg. (Budd.); 23. Pieris napi Z., skg.
(H. M.); 24. P.rapae Z.(H.M.). (c) Sphingidae: 25. Zygaena carniolica Scop., freq.
(Budd.); 26. Z. filipendulae Z., do. (Budd.); 27. Z. minos S.-V., do. (Budd.).

The following were recorded by the observers, and for the localities stated. —

Knuth (Fohr), 4 bees, 3 Lepidoptera, and 2 hover-flies. Loew (Brandenburg),
the Conopid Conops quadrifasciatus Deg., skg. (‘ Beitrage,’ p. 39). Wiistnei (‘ Marsch ’
of Schleswig- Holstein), the humble-bee Bombus cullumanus (X.) Z/s.; Alfken
(Bremen), the bee Megachile centuncularis Z. 9, skg. Schmiedeknecht (Thuringia),
2 humble-bees—Bombus soroénsis /. 5, and B. confusus Schenck, H. de Vries
(Netherlands), 3 humble-bees (1. Bombus hypnorum Z.é; 2. B. subterraneus Z. 6;
3. B. terrester Z. $), and 2 parasitic humble-bees (Psithyrus rupestris /. 6, and
P. vestalis ourcr.$). Heinsius (Holland).—A. Diptera. (a) Empidae: 1. Empis
livida Z. $ and 9. (4) Muscidae: 2. Scatophaga stercoraria Z. 3. (c) Syrphidae:
3. Melanostoma mellina Z.é B. Hymenoptera. Ajidae: 4. Bombus pomorum
Pz.8; 5. B. terrester Z. 6; 6. Halictus cylindricus #. 4; 7. H. flavipes /. 4.

1540. C. glaucus Baumg.—

VistTors.—Loew observed the butterfly Melanargia galatea Z., skg., in Steiermark
(‘ Beitrage,’ p. 49).

1541. C. acanthoides L. (Herm. Miiller, ‘Fertilisation,’ pp. 339-40, ‘ Weit.
Beob.,’ III, p. 83, ‘Alpenblumen,’ p. 417; Knuth, ‘ Bloemenbiol. Bijdragen,’
‘Bltitenbiol. Herbstbeob.’)—The heads of this species are also bright purple in
colour, and more conspicuous than those of C. crispus, for (according to Hermann
Miiller) the linear lobes of the bells are 7-8 mm. long. As too the nectar-containing
bell is somewhat wider, but less deep (2 mm.) than in C. crispus, the number of
visitors is greater. In other respects the flower mechanism, including the securing of
cross-pollination by insect-visits and the possibility of automatic self-pollination
should these fail, entirely agrees with those of C. crispus and Cirsium arvense.

VisiTors.—Herm. Miiller gives the following list for Central Germany.—

A. Coleoptera. (a) Chrysomelidae; 1. Cryptocephalus sericeus Z. (0) Cur-
cultonidae: 2. Larinus jaceae F.; 3. Spermophagus cardui Sfev., in very large
numbers in the florets. (c) £ateridae: 4. Corymbites holosericeus Ov. (d) Scara-
baeidae: 5. Trichius fasciatus Z. B. Diptera. (2) Conopidae: 6. Conops scutellatus

COMPOSITAE 653

Mg., skg.; 7. Physocephala rufipes &., do. (4) Syrphidae: 8. Eristalis arbustorum
L., skg. C. Hemiptera. 9. Anthocoris sp. D. Hymenoptera. (a) Apidae:
10, Bombus agrorum F. ¥, skg.; 11. B. lapidarius Z. ¥, do.; 12. B. pratorum Z. 4,
do.; 13. B. sylvarum Z. ¥ and 9, do.; 14. Cilissa tricincta K.9, do.; 15. Chelo-
stoma campanularum ZL. 9 and 4, skg. and po-dvg.; 16. Dasypoda hirtipes /. 9, skg.;
17. Halictus albipes /” 4, freq., skg.; 18. H. cylindricus /. 4, skg.; 19. H. interruptus
P2.%, do.; 20. H. leucozonius Schr. and 9, skg. and po-cltg.; 21. H. longulus Sm.
é and 9, skg.; 22. H. lucidulus Schenck 9, do.; 23. H. maculatus Sm. 8 and 9, do.;
24. H. minutus X. , do.; 25. H. nitidiusculus X. § and 9; 26. H. quadricinctus
F. $ and 9, very common, skg.; 27. H. quadrinotatus K. 4, occasional, skg.;
28. H. rubicundus Cfr. and 9, in large numbers, skg.; 29. H. smeathmanellus X.
9, skg.; 30. Heriades truncorum Z. 9 and 4, skg. and po-cltg.; 31. Megachile
centuncularis Z. 4, skg.; 32. M. lagopoda Z. 9 and 4, skg. and po-cltg.; 33. M.
versicolor Sm. 9, skg.; 34. Osmia aenea Z. %, do.; 35. O. auruleata Pz. 9, skg. and
po-cltg.; 36. O. fulviventris /. 9, do.; 37. Prosopis punctulatissima Sm. 9, skg. ;
38. Stelis aterrima Pz. 9 and 4, in large numbers, skg.; 39. S. breviuscula Wy/. 9,
skg.; 40. S. phaeoptera A. 9, not infrequent, skg. (4) Sphegidae: 41. Cerceris
variabilis Schr. 9, skg. E. Lepido-
ptera. (a) Noctuidae: 42. Plusia
gamma L., skg. (6) Rhopalocera:
43. Argynnis aglaja Z., skg.; 44.
Epinephele janira Z., do.; 45. Pieris
brassicae Z., do. (¢) Sphengidae: 46.
Zygaena carniolica Scop., skg.

The following were recorded by
the observers, and for the localities
stated.—

Knuth (Kiel Botanic Garden), 2
humble-bees (Bombus lapidarius LZ. 9
and ¥, skg., and B. terrester Z., do.)
and 2 butterflies (Pieris brassicae Z.,
skg., and Vanessa io Z., do.). Herm.
Miiller (Alps), 4 humble-bees, 3 Lepi-
doptera, anda beetle. Schmiedeknecht
(Tyrol), the bee Osmia confusa Jor.
(teste Morawitz). Schletterer (Tyrol),
the two dasygastrid bees Osmia confusa
Mor., and Megachile willughbiella &.
Schiner(Austria), the Muscid Oxyphora
miliaria Schr.

1542. C. defloratus L. (Herm.
Miiller, ‘Alpenblumen,’ pp. 418-22.)—
In this species the heads contain 100-

200 purple florets, and are only 20 mm.

ya , Fic. 206. Carduns defloratus, L. (after Herm. Miiller),
below, but 25-39 mm. above, where A. Upper part of the style (x 17). B. Lower part of the
the corollas bend outwards. The tube corolla-tube, cut open (x 7). _C. Longitudinal section

. : through the anther-cylinder shortly before the opening of
of the corolla is 7-8 mm. long, and its the floret (x 7). D. Lowest part of the corolla-tube, cut

i inear open. E. Lowest part of the style (x 7). co, corolla-
bell about Ramat deep, with 5 linea tube; 7, sweeping-hairs; 7, filaments; gv, style and its
diverging lobes 6-7 mm. in length. branches; 2, nectary; oz, ovary; fo, pollen; sé, stigmas.
On both sides of the lowest lobe the
bell is cleft to a depth of 3 mm., so that insects with a proboscis 3 mm. long can

reach the nectar which rises into the bell.

654 ANGIOSPERMAE—DICOTYLEDONES

In the first stage of anthesis the florets are covered by an abundance of bluish-
coloured pollen; in the second stage the stylar branches diverge somewhat at the
tip, and their papillose stigmatic edges swell outwards. The second stage lasts
considerably longer than the first. Should insect-visits fail, automatic self-pollination
may be effected by the further swelling of the stigmatic margins of the stylar branches,
until they touch the pollen that remains clinging to the floret. But autogamy is
scarcely likely to take place, for Hermann Miller observed a very large number
of insect visitors (not less than 103 species) in Switzerland, i.e. 8 beetles, 10 flies,
31 Hymenoptera, and 54 Lepidoptera.

Visitors.—Loew observed the humble-bee Bombus terrester Z. 3, skg. in the
Berlin Botanic Garden, and the following in Switzerland (‘ Beitrage,’ p. 58).—

A. Diptera. (a) Bombylitdae: 1. Argyromoeba sinuata Fall.,skg. (4) Empidae:
2. Empis tessellata #. (c) Tabanidae: 3. Tabanus bromius Z. B. Hymenoptera.
Apidae: 4. Halictus quadricinctus / 9; 5. Osmia villosa Schenck 9 and 4, skg.,
g also po-cltg. C. Lepidoptera. (a) Rhopalocera: 6. Parnassius delius Esp.
(4) Sphingidae: 7. Zygaena exulans Hehw. et. Rein.

1543. C. Personata Jacq. (=Arctium Personata Z.). (Herm. Miiller, ‘Alpen-
blumen,’ pp. 417-18.)—In this species there are about 6 burr-like purple-red heads
at the end of the stem, each measuring 30-40 mm. in diameter. Each of the 150-200
florets in a head possesses a corolla-tube 7-9 mm. in length, and a bell (ventricose
below) about 3 mm. long. The filaments are highly irritable. In other respects the

flower mechanism agrees entirely with that of C. defloratus, the filaments of which,
however, are less or not at all irritable.

Visitors. — Herm. Miiller noticed 2 beetles, 6 flies, 3 humble-bees, and 6
Lepidoptera.

Loew observed 2 bees in the Berlin Botanic Garden—1z. Apis mellifica Z. ¥,
skg.; 2. Osmia fulviventris Pz. 9, po-cltg.

1544. C. nutans L. (Sprengel, ‘Entd. Geh,,’ pp. 370-1; Herm. Miller,
‘Fertilisation,’ p. 340, ‘Weit. Beob.,’ II], pp. 83-4; Loew, ‘Bliitenbiol. Floristik,’
p- 390; Kirchner, ‘Flora v. Stuttgart,’ p. 390.)—Kirchner says that in this species
several hundred purple-red florets are aggregated into an odorous head, which
presents a surface about 40 mm. in diameter. The corolla-tube of each floret
is 10 mm. and the bell 5 mm. long, while the corolla-lobes vary from 5 to 8 mm. in
length. In other respects the flower mechanism agrees with that of C. arvense.

Visitors.—Herm. Miiller gives the following list for Central Germany.—

A. Diptera. Syrphidae: 1. Eristalis tenax Z., po-dvg.; 2. Syrphus ribesii Z.,
do. B. Hymenoptera. Apzdae: 3. Apis mellifica Z., freq. skg.; 4. Bombus
hortorum Z. 6, skg.; 5. B. hypnorum Z. ¥, do.; 6. B. pratorum Z. ¥ and 4, do.;
4. B. sylvarum Z. 9 and ¥, do.; 8. Psithyrus vestalis /ourcr., do.; 9. Halictus
cylindricus F. 3, do.; 10. H. leucozonius Schr. 9, po-cltg.; 11. H. malachurus X. 9,
skg.; 12. H. quadrinotatus X. 6, do. (Thuringia); 13. H. sexcinctus /. 9, skg. and
po-dvg.; 14. H. zonulus Sm. 9, skg. (Thuringia). C. Lepidoptera. (a) Rhopa-
locera: 15. Argynnis aglaja Z., in large numbers, skg.; 16. A. paphia Z., skg.; 17.
Epinephele janira Z., do. (Thuringia); 18. Hesperia lineola O., skg. (6) Sphingzdae :
1g. Zygaena lonicerae £'sp., skg.

The following were recorded by the observers, and for the localities stated.—

COMPOSITAE 655

Loew (Brandenburg), the bee Megachile lagopoda Z. 9, po-cltg. (‘ Beitrage,’

Pp: 39); (Silesia) Parnopes grandior Pall., skg. (op. cit., p. 31). Schmiedeknecht
Thuringia), 2 parasitic humble-bees—Psithyrus globosus Zv. , and P. rupestris
#.%. Alfken (Bremen), 10 bees—1. Bombus arenicola Zhs. $; 2. B. distinguendus
Mor. ¥; 3. B. lapidarius Z. 4; 4. B. ruderatus F. 9, skg.; 5. B. terrester Z. 3;
6. Halictus calceatus Scop. 9; 7. Osmia solskyi Mor. 9; 8. Psithyrus barbutellus
K. 9 and 8; 9. P. campestris Pz. 8; 10. P. rupestris F. 8 Schiner (Austria),
3 Muscids—1. Oxyphora miliaria Schr.; 2. Urophora solstitialis Z.; 3. U. stylata /.
Schletterer (Pola ; and for the Tyrol=T.).—Hymenoptera. (a) Azdae: 1. Andrena
florea #., po-cltg.; 2. Bombus hypnorum Z. (T.); 3. B. mesomelas Gerst. (T.);
4. B. terrester Z.(T.); 5. Ceratina nigroaenea Gers/.; 6. Halictus levigatus K. 3;
7. H. morbillosus A7rchd.; 8. H. quadricinctus F.; 9. H. scabiosae Ross7; 10. Osmia
pli Rosst; 11. Psithyrus rupestris #.(T.). (4) Scolivdae: 12. Scolia insubrica
OSSt.

1545. C. medius Gouan.—According to MacLeod (‘ Pyreneénbl.,’ pp. 352-4),
the diameter of the purple-red heads of this species is 30 mm. in the Pyrenees. The
corolla-tube is ro—11 mm. long, and the ventricose nectar-containing bell is 4-53 mm.
deep. Insects with a proboscis 4-5 mm. in length are therefore able to suck the
nectar, while beetles, with short-tongued flies and Hymenoptera, may be found
devouring the pollen.

VistTors.—MacLeod observed 14 Hymenoptera, 16 Lepidoptera, 6 beetles,
3 Syrphids, and 9 Muscids.

1546. C. carlinoides Gouan.—MacLeod states that the purple heads of this
species are 25—30 mm. in diameter in the Pyrenees (Bot. Jaarb. Dodonaea, Ghent, iii,
1891). The corolla-tube is 7-8 mm. long, and the bell 3-4 mm. deep. As in the
last species, most of the visitors are long-tongued insects.

VisiTors.—MacLeod observed 14 Hymenoptera, 2 Lepidoptera, a beetle, a
Muscid, and an Empid.

1547. C. pycnocephalus L.—
Visirors.—Schletterer observed the following at Pola.—

Hymenoptera. (a) Apidae: 1. Anthidium septemdentatum Z/r.; 2. Andrena
lucens /mh. ; 3. Ceratina cucurbitina Rossz; 4. Osmia fulviventris Pz.; 5. O. spinolae
Schenck; 6. Prosopis hyalinata Sm., var. subquadrata F. (4) Chrysididae: 7. Holo-
pyga amoenula Dahib. (c) Sphegidae: 8. Pemphredon unicolor F.; 9. Tachytes
obsoleta Rossz.

470. Onopordon L.

Florets, hermaphrodite, tubular. Stylar branches non-divergent; rows of
stigmatic papillae on their outer margins; at their base a ring of moderately
short sweeping-hairs directed obliquely upwards.

1548. O. acanthium L. (Herm. Miller, ‘Fertilisation,’ pp. 344-5, ‘ Weit.
Beob.,’ III, p. 81, ‘Alpenblumen,’ p. 417; Kerner, ‘Nat. Hist. PL,’ Eng. Ed. 1, II,
p. 252; Knuth, ‘Bliitenbiol. Herbstbeob.,’ ‘Bliitenbiol. biol. a. d. Ins. Riigen,’
‘Bloemenbiol. Bijdragen.’}—The bright purple florets of this species, according
to Hermann Miller, possess a corolla-tube 10-12 mm. in length, and a bell
3-4 mm. deep, with 5 linear non-divergent lobes 6-8 mm. long. The nectar
ascends as far as the bell. In the first stage of anthesis the pollen which has

656 ANGIOSPERMAE—DICOTYLEDONES

been swept out covers the florets; in the second stage the stigma projects 5—7 mm.
beyond the lobes of the corolla, gnd the stigmatic papillae are turned prominently
outwards. As regards the irritability of the filaments in this species, Kerner says
that as in other Compositae the pollen is protected from rain and dew by the
anther-cylinder, but by the retraction of this the upper end of the style and the
pollen covering it are exposed.

Visttors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list for Central Germany.—

A. Coleoptera. Coccznellidae: 1. Coccinella mutabilis Scrié., vainly seeking
for nectar (H. M.). B. Hemiptera. 2. Capsus, 2 different sp., skg. (H. M.);
3. Lygaeus equestris Z., skg. (H. M., Thuringia). C. Hymenoptera. (a) Apidae:
4. Andrena schrankella WVy/. 9 (H. M.); 5. Bombus lapidarius Z. ¥, skg. (H. M.);
6. Psithyrus rupestris /. 9, do. (H. M.); 7. Bombus terrester Z. 9, do. (H. M.);
8. Coelioxys conoidea J//. 9, do. (H. M.); 9. Halictus cylindricus #. 9 (Budd.) ;
10. H. leucozonius Schr. 9, skg. (H. M., Thuringia); 11. H. maculatus Sm., po-cltg.
(H. M., Thuringia); 12. H. quadricinctus F. 9, skg. (Budd.); 13. H. quadristrigatus
Lir. 9, do. (H. M.); 14. H. sexcinctus #. § (Budd.); 15. H. tetrazonius Aig. 9
(Budd.); 16. Megachile lagopoda Z. 9 and 4, po-cltg. and skg. (H. M., Budd.) ;
17. M. ligniseca A. 9, do. (Budd.); 18. Osmia aurulenta Pz. 9, do. (H. M.,
Thuringia); 19. O. fulviventris Pz. 9, freq. skg. and po-cltg. (Budd.); 20. Saro-
poda bimaculata Pz. 9, skg. (H. M.); 24. Stelis aterrima Pz. 9 and 4, do. (H. M.,
Budd.); 22. S. phaeoptera A. 9, do. (Budd.). (6) Sphegidae: 23. Psammophila
affinis K. 9, skg. (H. M.). D. Lepidoptera. (a) Rhopalocera: all skg.: 24.
Hesperia sylvanus £sp. (H. M., Thuringia); 25. Melanargia galatea Z. (H. M.);
26. Vanessa cardui Z. (H. M., Thuringia) ; 27. V. urticae Z.(H.M.). (4) Sphingidae:
28. Macroglossa stellatarum Z., skg. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Herm. Miiller (Alps), a humble-bee. Knuth (Kiel Botanic Garden), 4 skg.
humble-bees (1. Bombus agrorum /.; 2. B. hortorum Z.; 3. B. lapidarius Z. ;
4. B. terrester Z.), 4 skg. butterflies (1. Pieris napi Z.; 2. Vanessa atalanta Z. ;
3. V. io Z.; 4. V. urticae Z.), and a Muscid (Calliphora erythrocephala J/.):
(Riigen), a butterfly (Pieris sp.). Gerstacker (Berlin), the bee Osmia fulviventris
Pz.9. Schmiedeknecht (Thuringia), the bee Osmia solskyi AZor.9. Friese (Hungary),
the bee Eucera nigrifacies Zep. Schiner (Austria), the Muscid Tephritis postica Loew.
von Dalla Torre (Tyrol), 5 bees,—1. Bombus hypnorum Z.$; 2. B. muscorum /.$;
3. Andrena cetii Schr. 9; 4. Halictus sexcinctus Fr.9; 5. Stelis phaeoptera KX. 9.
Schletterer (Tyrol), 4 bees,—1. Andrena marginata #.; 2. Bombus hypnorum Z.;
3. Halictus sexcinctus /.; 4. Stelis phaeoptera A.

471. Lappa Rupp.

Florets tubular, hermaphrodite. Stylar branches very short, beset internally with
stigmatic papillae, externally with short pointed sweeping-hairs directed obliquely
upwards, extending below the bifurcation, and ending beneath in a ring of longer
sweeping-hairs.

1549. L. minor DC. (Hildebrand, ‘ U. d. Geschlechtsverhalt b. d. Compositen,.’
p. 46, Taf. V, Fig. 32; Herm. Miller, ‘ Fertilisation,’ p. 338, ‘ Weit. Beob., II, p. 84;
Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 161.)—In this species the heads are rather
small, usually only about the size of a hazel-nut, but they are rendered more conspicuous
by the reddish colour of the inner involucral bracts. The corolla is coloured purple

COMPOSITAE 657

above, and Herm, Miiller states that its bell is 3 mm. long. The lobes of the corolla
are erect, triangular, and only 1 mm. in length. During the first stage of anthesis the
pollen is discharged from the anther-cylinder; in the second stage the style projects
from its tip to 1-2 mm. below the ring of longer sweeping-hairs, and its branches
diverge completely. These are beset with stigmatic papillae on their inner surface.

Vistrors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (North Frisian Islands and Kiel), Apis and 2 Lepidoptera (Pieris sp.,
and Plusia gamma Z.), skg. Herm. Miiller (Westphalia), 2 bees (Bombus agrorum
F. ¥, skg.; and Halictus longulus Sm. ¥, do.). Buddeberg (Nassau), 2 bees
(Halictus cylindricus ¥. 9 and %, skg., and Stelis aterrima Pz. 9 and @, do.), and
a fossorial wasp (Ammophila sabulosa Z. 9, skg.). Alfken (Bremen), the humble-bee
Bombus proteus Gers#.8. MacLeod (Pyrenees), a humble-bee (Bot. Jaarb. Dodonaea,
Ghent, iii, 1891, p. 359).

1550. L. tomentosa Lam.—

Visttors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller, a Noctuid (Plusia gamma Z., skg.) and 5 bees—1. Apis, skg. and
po-clig.; 2. Bombus agrorum #. ¥ and 4, skg.; 3. B. sylvarum Z. ¥, do.; 4. Psithyrus
campestris Pz. , do.; 5. Megachile centuncularis Z. 9, do. (‘ Fertilisation,’ p. 338).
Loew (Berlin Botanic Garden), the last-named bee: (Switzerland), the Muscid
Trypeta tussilaginis ’” (‘ Beitrage,’ p. 59). H.de Vries (Netherlands), 2 humble-bees
(Bombus agrorum F. ¥, and B. subterraneus Z. 6) and a parasitic humble-bee
(Psithyrus campestris Pz. 5) (Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel,
1875).

1551. L. major Gaertn.—

Vistrors.—Herm. Miiller (Alps) noticed Apis, 3 humble-bees, and 3 Lepidoptera
(‘Alpenblumen,’ p. 426).

472. Carlina L.

Florets hermaphrodite. The whitish inner bracts of the involucre take the
place of ray-florets, but, besides adding to the conspicuousness of the head, they
serve aS a protection, closing over the florets in wet weather. The style bears
numerous sweeping-hairs on the outer side of its tip, which is shaped: like an
inverted club, and these hairs extend below the cleft. The stylar branches are
very short and remain almost closed, but allow a row of stigmatic papillae to
protrude along their outer edges where they touch. The outer involucral bracts
are spinose, forming an effective protection against animals that might creep up
to the flowers.

1552. C. acaulis L. (Herm. Miiller, ‘Alpenblumen,’ pp. 414-15; Knuth,
‘ Bliitenbiol. Herbstbeob.,’ ‘ Bloemenbiol. Bijdragen.’)—In this species several hundred
similar florets make up the inconspicuous head, which lies upon the ground, and is
20-40 mm. in diameter. The 60-80 inner involhucral bracts are dry, rigid, band-shaped,
and of a glistening white colour. They measure 35-40 mm. in length and 23-3 mm.
in breadth, and give to the head the appearance of a shining star 75-80 mm. in
diameter. The corolla-tube of each floret is 4-5 mm. long, and the bell s~6 mm.
The short, blunt stylar branches are scarcely 1 mm. in length, and bear a circlet of

DAVIS. II uu

658 ANGIOSPERMAE—DICOTYLEDONES

long sweeping-hairs on the outside, below the short ones that extend below the cleft.
Kerner says that the heads open at Innsbruck between 7 and 8 a.m., closing again
between 6 and 7 p.m.

Visttors.—Herm. Miiller gives the following list for Thuringia (‘ Fertilisation,’
P- 338).—

A. Coleoptera. Curculionidae: 1. Larinus senilis /., larvae and pupae are to
be found in the common receptacle, the perfect beetles on the flowering capitula and
other parts of the plant. B. Hymenoptera. Apidae: 2. Bombus agrorum F. 6,
very numerous, skg.; 3. B. confusus Schenck 8, do.; 4. B. lapidarius Z. 6, do.;
5. B. muscorum F. é, do.; 6. B. sylvarum Z. 4, do.; 7. B. terrester Z. 6, do.;
species of Halictus, especially 8. Halictus cylindricus /. 4; and g. H. quadricinctus
#.8; 10. Psithyrus rupestris Z. 6.

The following were recorded by the observers, and for the localities stated.—

Knuth (Bernese Oberland), the humble-bee Bombus lapidarius Z., skg.; (Kiel
Botanic Garden), 3 bees (1. Bombus hortorum Z.; 2. B. terrester Z.; 3. Apis),
a butterfly (Vanessa io Z.), and a hover-fly (Eristalis arbustorum Z.); all skg.
Herm. Miiller (Alps), 3 humble-bees and 2 Lepidoptera. Redtenbacher (Austria),
the weevil Larinus senilis #. Hoffer (Steiermark), the humble-bee Bombus pomorum
P2., var. elegans Sezdl. (=B. mesomelas Gersi.). Schmiedeknecht (Thuringia), 2 bees
(Bombus pomorum P2., and Psithyrus rupestris /.). MacLeod (Pyrenees), a humble-
bee, an ant, and a Muscid (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 358).

1553. C. acanthifolia All—The yellow heads of this species are rendered
more conspicuous by the golden-yellow involucre.

Visitors.— MacLeod observed a humble-bee in the Pyrenees (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, pp. 358-9).

1554. C. vulgaris L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.’ p. 161, ‘ Weit.
Beob. ti. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 236; MacLeod, Bot. Jaarb. Dodonaea,
Ghent, v, 1893, p. 402; Warnstorf, Verh. bot. Ver., Berlin, xxxviii, 1896.)—In the
island of Sylt this species usually bears only one head (rarely 2-5), 40 mm. in
diameter. This consists of several hundred florets, and is surrounded by several
series of strong thorny protective involucral bracts, within which is a circlet of
straw-coloured bracts 20 mm. long and 13-2 mm. broad. These are membranous
and not prickly. They play the part of a ray so well that the heads, which
but for them would be inconspicuous, are visible at a great distance. In dull
weather, and when it is dark, these bracts bend inwards and upwards, so as to
form a protective cover. Beneath this again is a second conical roof, formed by
the bristle-like chaffy bracteoles, which project several mm. beyond the florets.

The pollen, as in all Compositae, is shed into the anther-cylinder, while the
florets are still closed, and is brushed out by the stiff sweeping-hairs, which are
directed obliquely upwards. When the anther-cylinder is empty, the laterally placed
stigmatic papillae make their appearance.

Warnstorf adds the following particulars.—The limb of the corolla is dark-
violet at the time of anthesis, and the edges of its teeth are beset with simple
and irregularly branched short hairs, adapted to hold fast the pollen as it falls.
The florets are about 10-11 mm, long, and the anther-cylinder projects for as much
as 3 mm. By contraction of the filaments the latter is subsequently completely
withdrawn into the corolla, so that the dirty-yellow pollen-grains are completely

COMPOSITAE 659

exposed. The stylar branches are short, covered externally with sweeping-hairs
(which are somewhat longer beneath the cleft) and beset with stigmatic papillae
internally. At a late stage they diverge at an acute angle. The inner straw-
coloured linear involucral bracts here play the part of the ray-florets of other
Compositae, and periodically bring about opening and closing, so that the stigmas
of the older florets are brought into contact with the pollen of neighbouring
younger ones. Geitonogamy is thus provided for, should insect-visits fail. The
pollen-grains are roundish, with low spinose tubercles, and are about 50 p» in
diameter.

Kerner says that the heads open about 7-8 a.m., closing again 12 hours
later.

Visirors.—Herm. Miller (H. M.) (‘ Fertilisation,’ p. 338, ‘Weit. Beob.,’ II,
p- 79) and Buddeberg (Budd.) give the following list for Thuringia and Nassau.—

Hymenoptera. All skg. (a) Apidae: 1. Bombus lapidarius Z. $ (H. M.);
2. B. terrester Z.$(H. M.); 3. B. tristis Sed. § (H. M., Schwiebus); 4. Coelioxys
acuminata JVy/. 9 (H. M.); 5. C. quadridentata Z. 9 (H. M.); 6. Halictus cylindricus
F. 3, very common (H. M.); 7. H. quadricinctus #. 6 (H. M., Budd.), freq. ;
8. Megachile circumcincta K. 9 (H. M.); 9. M. lagopoda Z. 8, once (H. M.).
(6) Sphegidae: 10. Ceropales maculatus /. 9 (H. M.).

The following were recorded by the observers, and for the localities stated. —

Knuth, 3 sucking humble-bees (1. Bombus derhamellus X.; 2. B. lapidarius Z. ;
3. B. terrester Z.), a po-dvg. hover-fly (Syrphus balteatus Deg. 3), and 2 po-dvg.
Muscids (Olivieria lateralis #., and Anthomyia sp. 9). Schletterer (Tyrol), the
humble-bee Bombus derhamellus A. MacLeod (Pyrenees), 3 humble-bees and
a Lepidopterid (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 358).

473. Saussurea DC.

Florets hermaphrodite. Outer surface of the stylar branches completely covered
with long pointed sweeping-hairs, which are longest below: inner surface beset with
stigmatic papillae. These branches diverge and roll back.

1555. S. alpina DC. (Herm. Miiller, ‘Alpenblumen,’ pp. 413-14.)—In this
species each head is made up of 11-17 florets, with a white corolla-tube 7-8 mm.
and a violet bell 2 mm. in length. From 5-9 such heads are closely aggregated.
The florets are protandrous, and smell like violets or vanilla.

Visitors.—Hermann Miller thinks these are probably po-dvg. flies, together
with po-cltg. and nect-skg. bees; but he only saw a single hover-fly. Lindman
observed flies, a saw-fly, and a humble-bee on the Dovrefjeld.

1556. S. albescens Hook. et Thoms.—

Visrrors.—Loew observed 3 bees (1. Apis mellifica Z. ¥, skg.; 2. Bombus
hortorum Z. %, do.; 3. Halictus nitidiusculus X. 8, do.) and an undetermined
Chalcidid in the Berlin Botanic Garden.

474. Jurinea Cass.

1557. J. mollis Reichb. (v. Wettstein, ‘ Compos. d. ést-ungar. Flora.’)—In this
species the stomata of the young unopened heads secrete nectar which entices ants—
usually Camponotus sylvaticus O“v., var. aethiops La/z., more rarely Aphenogaster

uu2

660 ANGIOSPERMAE—DICOTYLEDONES

structor Za/z., at Vienna and Budapest. The ants keep off injurious insects. The
secretion ceases at the beginning of anthesis.

1558. J. alata—According to Hildebrand (‘W. d. Geschlechtsverhilt. b. d.
Compositen,’ pp. 58-9), the style is similar in structure to that of Centaurea montana.

VisiTors.—Loew observed the po-cltg. bee Osmia fulviventris Pz. 9 in the
Berlin Botanic Garden.

475. Alfredia Cass.
1559. A. cernua Cass.—

VistTors.—Loew observed the humble-bee Bombus terrester Z. 9, po-cltg., in
the Berlin Botanic Garden.

476. Rhaponticum Hall.
1560. R. pulchrum Fisch. et Mey. (=Centaurea pulchra DC.).—

VistTors.—Loew observed 3 bees (1. Dasypoda hirtipes #. 4, skg.; 2. Osmia
fulviventris Pz. 9, po-cltg.; 3. Stelis aterrima Pz. 9, skg.) and a wasp (Odynerus
parietum Z.) in the Berlin Botanic Garden.

477. Serratula Dill.
Florets hermaphrodite or dioecious.

1561. S. tinctoria L. (Kirchner, ‘Flora v. Stuttgart,’ p. 727; Herm. Miiller,
‘Fertilisation,’ pp. 345-6 ; Knuth, ‘ Bliitenbiol. Herbstbeob.’)—This species is gyno-
dioecious, and bears purple-red florets. Kirchner says there are transition-forms
between the female and hermaphrodite ones.

VisiTors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Thuringia), a humble-bee (Bombus agrorum /. 9 and 4, skg.)
and a butterfly (Colias hyale Z., freq., skg.). Knuth (Kiel Botanic Garden), 6 hover-
flies, po-dvg. and skg. (1. Eristalis horticola Deg. ; 2. E. pertinax Scop. ; 3. Platycheirus
sp.; 4. Syritta pipiens Z.; 5. Syrphus ribesii Z.; 6. S. umbellatarum Z.), and 2

butterflies, skg. (Pieris sp., and Vanessa io Z.). Schiner (Austria), the Muscid
Trypeta ruficauda F.

1562. S. quinquefolia Bieb.—

Vistrors.—Loew observed 2 humble-bees (Bombus agrorum /. 6, and B. terrester
L. 9, skg.) in the Berlin Botanic Garden.

1563. S. lycopifolia Vill., and 1564. S. centauroides Bieb. (=S. radiata
Bieb.). (Von Wettstein, ‘Compos. d. ést.-ungar. Flora.’)—Wettstein says that in
these species nectar is discharged in abundance from the stomata on the involucral
bracts of the young unopened heads, serving to attract ants that keep injurious insects
from the buds. (C/ Jurinea mollis.)

VisiTors.—Wettstein observed 4 ants on S. lycopifolia (1. Formica exsecta JVy/. ;
2. F. rufibarbis #.; 3. Lasius niger Z.; 4. Myrmica lobicornis Vy/.), and one (Lasius
alienus /érst.) on S. centauroides.

COMPOSITAE 661

478. Cnicus L.

- Style of the hermaphrodite florets very similar to that of Centaurea (montana).

1565. C. benedictus L. (Hildebrand, ‘U. d. Geschlechtsverhalt. b. d. Com-
positen, pp. 57-8, Taf. V, Fig. 31.}—Although the anthers appear normal in this

species, the first florets produce no pollen.

The neuter ray-florets are so small as to

be hardly visible beside the disk-florets.

479. Centaurea L.

Ray-florets neuter, tubular, radiating. Disk-florets hermaphrodite.

very irritable.
directed obliquely upwards; above
this short hairs on the style ; stig-
matic papillae on the inner sur-
faces of the branches. Kerner
states that the pollen is concealed
in the anther-cylinder until insects
visit the florets, being thus pro-
tected from rain and dew. When
the proboscis of a nectar-seeking
insect stimulates the filaments, they
contract so that the crumbling
pollen is carried off by the visitor
as soon as it is swept out. After
removal of the pollen, only cross-
fertilization is possible for a short
time; the stylar branches then
roll back in such a way that the
stigmatic papillae touch the pollen
still clinging to the sweeping-hairs,
thus effecting self-pollination. I
have not myself observed this
rolling back of the branches. In
some species—C. alpina (Wett-
stein); C. montana, in the Apen-
nines (Delpino), but not at Vienna
(Wettstein)—nectar is secreted by
the involucral bracts of the bud,
as in Serratula lycopifolia, S. cen-
tauroides, and Jurineamollis. Wett-
stein observed the ant Campo-

Filaments

Below the short broad stylar branches a ring of sweeping-hairs

FIG. 207.
(1) Disk-floret in the first (male) stage: the style is still con-
cealed in the anther-cylinder (cba) borne upon the filaments (4):
ab is the uppermost part of the cylinder formed by the appen-

Centaurea Cyanus, L. (after J. MacLeod).

dages of the anthers. (2) Uppermost part (a5) of (1) in
longitudinal section: the two stylar branches are still apposed ;
at their base is a ring of hairs, above which is the pollen.
(3) Style in the second stage, with diverging branches; the ring
of hairs is laden with pollen-grains, a few of which adhere to
the papillose inner surfaces of the branches. (4) Disk-floret
in the second (female) stage, half schematic: the style protrudes
from the tip of the anther-cylinder. a5, appendages of the
anthers; dc, anthers; de, corolla-tube (the nectary (not figured) is
situated ate); 4, filaments; 4, lobes of the corolla; sv, style ;
wh, ring of sweeping-hairs.

notus sylvaticus Oliv., var. aethiops Za/z., on C, alpina in Istria.

1566. C. Jacea L. (Herm. Miiller, ‘Fertilisation,’ pp. 346-9, ‘ Weit. Beob.,’
III, pp. 79-80, ‘ Alpenblumen,’ p. 415; Loew, ‘ Bliitenbiol. Floristik, pp. 390, 393,
397; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 401-2 ; Knuth, ‘ Bliiten-

biol. Herbstbeob.,” ‘Bloemenbiol. Bijdragen.’)—This species is trioecious.

The

662 ANGIOSPERMAE—DICOTYLEDONES

florets are usually of a bright purple colour, and, according to Hermann Miller,
from 60 to over 100 are found in one head. The capitulum is compressed below to
a diameter of 8-10 mm., but spreads out above to form a surface 20-30 mm. across.
The sterile ray-florets only serve to render the heads conspicuous, and are modified
into large outwardly directed funnels. The disk-florets are hermaphrodite; the
corolla-tube is 7-10 mm. and the bell 3-42 mm. long. There are 5 linear
corolla-lobes, 5 mm. in length. The filaments are hairy and irritable; they
bend when touched by the proboscis of an insect visitor, pulling down the anther-
cylinder. The result is that the annular stylar brush sweeps out the pollen contained
in the cylinder. At a later stage the style grows out from this, and the papillose
inner surfaces of the stylar branches of the style separate to some extent. Automatic
self-pollination is therefore excluded, according to Hermann Miller. (Vide supra for
Kerner’s view.) But self-pollination may be brought about by insects when the
stigmatic papillae are exposed, if the pollen has not been completely removed.
When visitors are numerous, however, no pollen is left on the head during the second
stage of anthesis, so that crossing is necessarily effected. Warnstorf describes the
pollen-grains as white in colour, ellipsoidal, furrowed, and covered with low spinose
tubercles, about 56 « long and 30 » broad.

Hermann Miller also observed plants bearing male and female heads with enlarged
ray-florets, while MacLeod (in Belgium) saw some rayless female heads besides the
hermaphrodite ones, with neuter ray-florets. The male heads are of a paler colour ;
their ray-florets are greatly enlarged ; the nectary is vestigial, and the stylar branches
are permanently apposed. The florets of the female heads are darker and smaller,
with shrunken anthers devoid of pollen.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list for Central Germany.—

A. Diptera. (a) Conopidae: 1. Conops flavipes Z., skg. (H. M.); 2. Physo-
cephala vittata /., do. (H. M.); 3. Sicus ferrugineus Z., do. (Budd.). () Empzdae:
4. Empis livida Z., skg. (H. M., Budd.); 5. E. rustica /, do. (H.M.). (c) Syrphidae:
6. Eristalis intricarius Z., skg. (H. M.); 7. E. tenax Z., po-dvg., and inserting its long
proboscis into the bells of the individual florets (H. M.); 8. Helophilus pendulus Z..,
skg. (H. M.); 9. Rhingia rostrata Z., do. (H. M.); 10. Syrphus balteatus Deg.,
po-dvg. ae B. Hymenoptera. (ca) Apidae: 11. Anthidium strigatum Pz.,
po-cltg. (H. M., Thuringia); 12. Andrena pilipes F. 9, po-cltg. (H. M.); 13. Apis
mellifica Z. ¥, freq., skg. and po-cltg.; 14. Bombus agrorum F. §, skg. (H. M.);
15. B. lapidarius Z. ¥, do.(H.M.); 16. B. pratorum Z., do.(H.M.); 17. B. sylvarum
L. 9 and ¥, do. (H. M.); 18. Dasypoda hirtipes F. 4, in large numbers, skg. (H. M.);
19. Halictus albipes F. 4, skg. (H. M.); 20. H. cylindricus &. 9 and 4, very common,
skg. and pq-cltg. (H. M.); 21. H. interruptus Pz. 5, skg.(H.M.); 22. H. leucozonius
Schr. 9 and 4, skg. and po-cltg. (H. M.); 23. H. longulus Sm. 9 and é(H.M.); 24.
H. lucidulus Schenck 9, skg. and po-cltg. (H. M.); 25. H. maculatus Sm. 9 and 6, do.
(H. M.); 26. H. malachurus &. 9, do. (Budd.); 27. H. minutus X. 4, skg. (H. M.);
28. H. nitidiusculus A. § and 9, po-cltg. and skg. (H. M.); 29. H. quadricinctus
#. 9 and 4, freq., skg. and po-cltg. (H. M., Budd.); 30. H. rubicundus Chr.
g and 6, skg. and po-cltg. (H. M.); 31. H. sexcinctus F. 9, skg. (Budd.);
32. H. smeathmanellus X. 9, skg. and po-cltg. (H. M.); 33. H. tetrazonius A7g.
g and 4, skg. (Budd.); 34. H. villosulus A. 9, skg. and po-cltg. (Budd.); 35. H.
zonulus Sm. 6, skg. (H. M.); 36. Megachile centuncularis Z. 9 and 4, po-cltg. and
skg. (H. M., Budd.); 37. M. lagopoda Z. 9, po-cltg. (H. M., Thuringia); 38. Osmia

COMPOSITAE 663

spinulosa X. 9, po-dvg. (H. M., Thuringia); 39. Psithyrus barbutellus X. 4, skg.
H. M.); 40. P. campestris Pz. $, do. (H. M.); 41. P. rupestris #. 9 and 4, do.
H. M.); 42. P. quadricolor Zep. 6, do.(H.M.); 43. Saropoda bimaculata Z. 6, do.
(H. M., Liebenau near Schwiebus). (6) Sphegidae: 44. Ammophila sabulosa Z. 9, skg.
A M.). (c) Vespidae: 45. Pollistes gallica Z.(H.M., Thuringia). C, Lepidoptera.
a) Noctuidae: 46. Plusia gamma L.(H.M.) (6) Rhopalocera: 47. Coenonympha
pamphilus Z.(H.M.); 48. Colias hyale Z.(H.M.); 49. Epinephele janira Z. (H. M.);
50. Hesperia thaumas A/n. (H. M.); 51. Lycaena coridon Scop., skg. (H. M.,
Thuringia); 52. L. sp. (H. M.); 53. Melanargia galatea Z., in large numbers, skg.
(H. M.); 54. Pararge megaera Z. (H. M.); 55. Pieris brassicae Z. (H. M.);
56. P. napi Z. (H. M.); 57. Polyommatus phlaeas Z. (H.M.). (c) Sphingidae:
58. Zygaena carniolica Scop. (H. M., Thuringia); 59. Z. lonicerae Esp. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Knuth (Gliicksburg), the butterfly Pieris napi Z., very freq., skg.: (Kiel), all skg.,
3 bees (1. Apis mellifica Z.; 2. Bombus lapidarius Z.; 3. Psithyrus rupestris /.),
4 hover-flies (1. Eristalis nemorum Z.; 2. Helophilus hybridus Zoew; 3. H. pendulus
L.; 4. Rhingia campestris Z.), and 4 Lepidoptera (1. Lycaena sp.; 2. Pieris sp. ;
3. Plusia gamma Z.; 4. Vanessa io Z.). Herm. Miiller (Alps), 3 bees and, 3 Lepi-
doptera. Réssler (Wiesbaden), the moth Grapholitha hohenwarthiana Z7., and the
hawk-moth Zygaena meliloti Zsp. Friese (Thuringia), the bee Halictus sexcinctus /.
Loew (Brandenburg), the hover-fly Eristalis sepulcralis Z. (‘Beitrage,’ p. 39):
(Switzerland), the bee Bombus pascuorum Scop. ¥, po-cltg., and the hawk-moth
Zygaena carniolica Scop. (op. cit., p. 58): (Steiermark), 2 bees, po-cltg.—Halictus
zonulus Sm. 9, and Megachile melanopyga Costa 9 (op. cit., p. 49). Schletterer
(Tyrol), 3 bees—r. Andrena lucens /mh.; 2. Bombus pascuorum Scop.; 3. Halictus
albipes #. von Dalla Torre (Tyrol), the ruby-wasp Chrysis analis Spen. Hoffer
(Steiermark), the parasitic humble-bee Psithyrus barbutellus X.d. MacLeod (Flanders),
7 long-tongued bees, 5 short-tongued bees, 6 hover-flies,a Muscid, an Empid, and 13
Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 401-2). H. de Vries (Nether-
lands), 2 bees (Megachile argentata /. 9, and M. spinulosa K. 9) and 2 humble-bees
(Bombus subterraneus L. 4, and B. terrester Z. 5and ¥) (Ned. Kruidk. Arch., Nijmegen,
2. Ser., 2. Deel, 1875).

1567. C. nigra L.—The florets of this species are bluish-red in colour, and
Kirchner (‘ Beitrage,’ p. 70) states that their mechanism agrees with that of C. Jacea.
As, however, enlarged ray-florets are never present, the top of the head is only about
25 mm. broad. Each head contains over 100 florets, with a corolla-tube ro mm.
and a bell 4-5 mm. long, the 5 corolla-lobes being of the same length.

VisiTors.— Willis observed the following in the neighbourhood of the south coast
of Scotland (‘Fls. and Insects in Gt. Britain,’ Part I).—

A.Coleoptera. (a) Mitzdulidae : 1. Meligethes viridescens /., very common, po-
dvg. (5) Scarabaeidae: 2. Crepidodera ferruginea Scop. B. Diptera. (a) Muscidae:
3. Anthomyia radicum Z., po-dvg.; 4. A. sp., freq.; 5. Hylemyia strigosa J, skg. ;
6. Trichophthicus cunctans Mg., do. (4) Syrphidae: 4. Eristalis aeneus Scop., freq.,
skg.; 8. E. pertinax Scop., po-dvg.; 9. E. tenax Z., skg.; 10. Platycheirus albimanus
Me., po-dvg.; 11. P. manicatus J/g., do.; 12. Rhingia rostrata Z., freq., skg. and
po-dvg.; 13. Sphaerophoria scripta Z., skg.; 14. Syrphus balteatus Deg., freq., po-dvg.
C. Hemiptera, 15. Anthocoris sp.; 16. Calocoris bipunctatus /.; 17. C. fulyo-
maculatus Deg. D. Hymenoptera, Afzdae: all skg.: 18. Anthidium manicatum--
L.; 19. Apis mellifica Z.; 20. Bombus agrorum /.; 21. B. hortorum Z., freq. ;
22. B. lapidarius Z.; 23. B. pratorum Z.; 24. B. scrimshiranus X., freq.; 25. B.
terrester Z. E. Lepidoptera. (a) Microlepidoptera: 26. Crambus sp. (4) Rhopa-
locera: all skg.: 27. Argynnis aglaia Z.; 28. A. sp.; 29. Epinephele janira Z.;

664 ANGIOSPERMAE—DICOTYLEDONES

30. Pieris napi Z., freq.; 31. P. rapae Z.; 32. Polyommatus phlaeas Z., freq. ;
33- Vanessa urticae Z.
The following were recorded by the observers, and for the localities stated.—

Kirchner, humble-bees and a hover-fly (Eristalis tenax Z.). Heinsius (Holland),
a Conopid (Sicus ferrugineus Z. 3), 3 bees (1. Bombus agrorum /. 8; 2. Coelioxys
conica Z.9; 3. Megachile centuncularis Z. 9), and a butterfly (Pieris brassicae Z. 4).
MacLeod (Pyrenees), 12 Hymenoptera, 18 Lepidoptera, a beetle, and 6 flies (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, pp. 356-7). Scott-Elliot (Dumfriesshire), Apis,
4 humble-bees, another long-tongued bee, 5 hover-flies, 3 Muscids, and 5 Lepidoptera
(‘Flora of Dumfriesshire,’ p. 101). E.D. Marquard (Cornwall), the bee Andrena
denticulata K. Saunders (England), the bee Rophites quinquespinosus S#z7.

1568. C. montana L. (Hildebrand, ‘ U.d. Geschlechtsverhiilt. b. d. Compositen,’
pp. 50-6, Taf. V, Figs. 1-23.)—In this species the funnel-shaped ray-florets are
neuter, while the disk-florets are hermaphrodite. Hildebrand says that the longer
sweeping-hairs on the style of the latter are not at the same level, but form an arch
beneath the stylar branches. As the style grows up it sweeps the pollen before it,
first into the empty cone formed by the teeth of the anthers, then out of the apex of
this cone. When an insect visits the florets, the filaments contract when touched
by its proboscis, so that still larger masses of pollen protrude, and adhere to the
ventral surface of the visitor. The style now protrudes, and the papillose inner
surfaces of its branches become exposed. With reference to the secretion of
nectar by the involucral bracts, vzde p. 661.

Visrtors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel Botanic Garden), the humble-bee Bombus hortorum JZ., skg.:
(Iserlohn, in Westphalia), the humble-bee B. lapidarius Z. 9, skg. (‘Bloemenbiol.
Bijdragen’). Schenck (Nassau), 4 bees (1. Coelioxys quadridentata Z.; 2. Megachile
centuncularis Z.; 3. M. ericetorum Zep.; 4. Stelis phaeoptera K.). Schletterer and
von Dalla Torre (Tyrol), the humble-bee Bombus pratorum Z. Loew (Berlin Botanic
Garden), 2 bees (Apis mellifica Z. ¥, skg., and Osmia fulviventris Pz. 4, do.).

1569. C. axillaris Willd—Kirchner (‘ Beitrage,’ p. 71) had an opportunity of
examining the flower mechanism of this species on the Simplon, and found it agreed
with that of C. montana. The 9-12 neuter ray-florets possess such greatly enlarged
corollas that they project 25-35 mm. from the involucre. The corolla-tubes of the
disk-florets are g mm., the bells 3 mm., and the corolla-lobes 7 mm. in length. The
anther-cylinder is 7 mm. long, and projects from the mouth of the floret. The
style grows 4 mm. beyond this, and in the second stage of anthesis its branches
become recurved.

VisiTors.—Kirchner observed humble-bees; and Loew saw the honey-bee in
the Berlin Botanic Garden.

1570. C. Phrygia L.—

Visirors.—Hermann Miiller noticed 2 Lepidoptera in the Alps (‘Alpenblumen,’
p- 415); and Loew saw the parasitic humble-bee Psithyrus vestalis Fourcr. 4, skg., in
the Berlin Botanic Garden.

1571. C. Mureti Jord. (=C. maculosa auct. pro parte, and C. caerulescens
Willd.).—

COMPOSITAE 665

Visitors.—Herm, Miiller saw a humble-bee and 3 Lepidoptera in the Alps
(‘ Alpenblumen,’ p. 4158). :

1572. C. Cyanus L. (Sprengel, ‘Entd. Geh.,’ pp. 371-3; Herm. Mtiller,
‘ Fertilisation,’ pp. 350-1, ‘ Weit. Beob.,’ II], pp. 80-1; Knuth, ‘BI. u. Insekt. a. d.
nordfr. Ins.,’ pp. 96, 161; MacLeod, Bot. Jaarb. Dodonaea, Ghent, v, 1893,
pp. 398-401.)—In this species the ray-florets are usually 8 in number, and Ludvig
says that they only serve to attract insects. They are sterile, and modified into
large outwardly directed funnels, so as to’ increase the diameter of the blue surface
of the head from 2-5 cm., rendering it conspicuous from all sides. (C/ Plateau’s
observations, Vol. I, p. 209.) Hermann Miller describes the disk-florets as having
a corolla-tube 5-6 mm. in length, which expands into a bell only 3 mm. deep,
with 5 linear lobes: nectar ascends to the base of this bell. The disk-florets
are not very numerous; they do not form a flat surface, but the anther-cylinders
project at intervals. The filaments are highly irritable. (Cf Fig. 207, p. 661.)

VisiTors.—Herm. Miiller gives the following list for Westphalia and Thuringia.—

A. Diptera. (ca) Empidae: 1. Empis livida Z., freq., skg. (4) Syrphidae:
2. Eristalis arbustorum Z., po-dvg.; 3. Helophilus pendulus Z., do.; 4. Melithreptus
scriptus Z., do.; 5. Rhingia rostrata Z., skg. B. Hymenoptera. (a) Afidae:
6. Apis mellifica Z. §, freq., skg. and po-cltg.; 7. Bombus lapidarius Z. ¥, skg. ;
8. B. sylvarum Z. ¥, do.; 9. Halictus tetrazonius K7g. 9, do.; 10. Megachile
maritima X. $, do.; 11. Saropoda bimaculata Pz. 9, persistently skg. and po-cltg. ;
12. Stelis breviuscula Wy/. 9, skg. (4) Spheg¢dae: 13. Psammophila affinis K., skg.
C. Lepidoptera. (a) Noctuidae: 14. Plusia gamma Z., skg. (6) Rhopalocera:
15. Lycaena aegon S.-V. 4, skg.; 16. L. damon S.-V,, skg.

The following were recorded by the observers, and for the localities stated.—

Knuth (North Frisian Islands and at Kiel), Apis, 3 humble-bees, a Lepidopterid,
and 6 hover-flies: (Riigen), all skg., a hover-fly (Volucella bombylans Z.) and 3 bees
(1. Apis mellifica Z.; 2. Bombus agrorum F.; 3. B. lapidarius Z. é and ¥). Alfken
(Bremen), 2 bees (Apis, freq., skg., and Megachile maritima K. 9, skg.). Krieger
(Leipzig), the bee Halictus smeathmanellus K. Réssler (Wiesbaden), the Noctuid
moth Chariclea delphinii Z. Friese, in Alsace (A.), Mecklenburg (M.), Thuringia
(T.), and Hungary (H.), 3 bees—1. Eucera hungarica Friese, not rare (H.);
2. Osmia claviventris Zhoms. § (A., M., T., H.); 3. Osmia papaveris Z/r. (M. and T.,
occasional; H.). Loew (Silesia), an Asilid (Dioctria flavipes J/g.), a Muscid
(Anthomyia sp.), and a bee (Apis mellifica Z. ¥, po-cltg.) (‘Beitrage,’ p. 31).
MacLeod (Flanders), Apis, 2 humble-bees, an Halictus, 6 hover-flies, an Empis,
and 2 Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, v, 1893, pp. 400-1). H. de Vries
(Netherlands), the bee Apis mellifica Z. § (Ned. Kruidk. Arch., Nijmegen, 2. Ser..
2. Deel, 1875).

1573. C. Scabiosa L. (Hildebrand, ‘U.d. Geschlechtsverhalt. b. d. Compositen,’
pp. 56-7; Herm. Miller, ‘Fertilisation,’ pp. 349-50, ‘Weit. Beob.,’ III, p. 80,
“Alpenblumen,’ p. 416; Ljungstrém, Bot. Not. Lund, 1884; Loew, ‘ Bliitenbiol. Flo-
ristik,’ p. 393; Knuth, ‘Bliitenbiol. Beob. a. d. Ins. Riigen,’ ‘ Bloemenbiol. Bijdragen,’
‘Bliitenbiol. Herbstbeob.’}—The florets of this species are usually dull purple in
colour, and Hermann Miller states that their mechanism corresponds on the whole
to that of the hermaphrodite florets of C. Jacea. The ray-florets, however, are
neuter, devoid of bells, and considerably larger, their corolla-tubes being 16-22 mm.
in length. The nectar is even more easily accessible, for the corolla-tubes of the

666 ANGIOSPERMAE—DICOTYLEDONES

disk-florets are 11-12 mm. long, and the bells 33-4 mm. deep. Warnstorf describes
the pollen-grains as resembling those of C. Jacea, but as much as 75 mp long and
44 p broad. Purely female stocks with vestigial stamens in the disk-florets were
observed by Ljungstrém in Sweden.

Visitors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following
list for Westphalia, Thuringia, and Nassau.—

A. Coleoptera. Chrysomelidae: 1. Cryptocephalus sericeus Z., resting inac-
tively on the florets (H. M.). B. Diptera. (a) Empidae: 2. Empis sp., freq., skg-
(H.M., Thuringia). (5) Musctdae: 3. Trypeta cornuta #. (H. M.). (c) Syrphidae:
4. Eristalis horticola Deg., po-dvg. (Budd.); 5. E. nemorum Z. (H.M.). C. Hemi-
ptera. 6. Capsus sp., skg. (H. M.). D. Hymenoptera. Apzdae: 7. Anthidium
manicatum Z. 9 and @, po-cltg. and skg. (H. M.); 8. Apis mellifica Z. ¥, numerous,
skg. (H. M.); 9. Bombus agrorum /. é and 9, skg. and po-dvg. (H. M.); 10. B.
confusus Schenck 8, skg. (H. M.); 11. B. lapidarius Z. 8, do. (H. M.); 12. B.
sylvarum Z. 4, do. (H. M.); 13. B. terrester Z. 6 (H. M.); 14. Coelioxys conoidea
ill. 8, freq., skg. (H. M.); 15. Halictus maculatus Sm. 9, po-cltg. (H. M.);
16. H. quadricinctus #. 9 and é, very common, skg.(H. M.); 17. Megachile
argentata /. 3, skg. (H. M., Strassburg); 18. M. ligniseca &. 8, do. (H. M.);
19. Osmia aenea Z. and 9, po-cltg. and skg. (H. M.); 20. O. rufa Z. 9, do.
(H. M., Strassburg); 21. O. spinulosa K. 9, po-cltg. (H. M.); 22. Psithyrus
rupestris /. 4, skg. (H. M.). E. Lepidoptera. (a) Rhopalocera: 23. Epinephele
janira Z. (H.M.); 24. Lycaena corydon Scop., skg. (H.M.); 25. Melanargia
galatea Z., in large numbers, skg. (H.M.); 26. Melitaea athalia 2g. (H. M.).
(6) Sphingidae: 27. Zygaena carniolica Scop. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Knuth (Gliicksburg), 2 bees (Apis, and Bombus lapidarius Z. 9 and 4) and a hover-
fly (Eristalis tenax Z.), skg.: (Kiel), a parasitic humble-bee (Psithyrus vestalis Fourer.),
3 hover-flies (1. Eristalis sp.; 2. Helophilus pendulus Z.; 3. Platycheirus peltatus A@.),
and 3 Lepidoptera (1. Lycaena sp.; 2. Plusia gamma Z.; 3. Vanessa io Z.), all skg.,
also the beetle Meligethes: (Riigen), 3 bees, all skg. (1. Bombus lapidarius Z. 9;
2. B. sylvarum Z. ¥; 3. B. terrester Z.¥). Schmiedeknecht (Thuringia), the humble-
bee Bombus derhamellus AX. 6. Schiner (Austria), 2 Muscids (Trypeta cornuta
£., and T. tussilaginis #.) and a Bombyliid (Phthiria gaedii M/g.). Schletterer
and von Dalla Torre (Tyrol), 6 bees—z. Andrena eximia Sm.; 2. A. propinqua
Schenck; 3. Halictus leucopus K.6; 4. H. sexcinctus #. 5; 5. H. sexnotatus X. 9;
6. Osmia spinulosa K.3. Herm. Miiller (Switzerland), 2 beetles, 2 hover-flies, 12 bees,
and 2 Lepidoptera. MacLeod (Pyrenees), 4 humble-bees, and 2 Lepidoptera (Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, p. 358). Loew (Steiermark), 3 bees (1. Bombus
pratorum Z. 4, skg.; 2. B. variabilis Schmzedekn. ¥, po-cltg.; 3. Megachile melanopyga
Costa 9, do.) (‘ Beitrage,’ p. 49): (Berlin Botanic Garden),—A. Hymenoptera.
(a) Apidae: 1. Apis mellifica Z., skg.; 2. Bombus terrester Z. 5, do. (6) Vespidae:
3. Eumenes coarctatus Z. B. Lepidoptera. Rhopalocera: all skg.: 4. Argynnis
latonia Z.; 5. Pieris brassicae Z.; 6. Vanessa urticae Z.; and on the var. spenulosa,—
A. Diptera. Syrphidae: 1. Syritta pipiens Z.; 2. Syrphus balteatus Deg.
B. Hymenoptera. Afzdae: 3. Psithyrus rupestris /. 4, skg.

1574. C. nervosa Willd. (Herm. Miiller, ‘Alpenblumen,’ pp. 415-16.)—The
red heads of this species are 60-70 mm. in diameter. There are about 20 neuter
ray-florets, modified into tubes 22 mm. long, with 5 lobes 15-20 mm. in length.
There are about 100 disk-florets, each with a corolla-tube 8-g mm. and a bell
5 mm. long. The flower mechanism agrees in other respects with that of C. Cyanus.

Visirors.—Herm. Miiller observed 5 humble-bees and 14 Lepidoptera in the Alps.

COMPOSITAE 667

1575. C. Calcitrapa Z. (Knuth, ‘ Bloemenbiol. Bijdragen.’)—

Vistrors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel Botanic Garden), the humble-bee Bombus lapidarius Z. 9, skg.
Friese (Istria), the parasitic bee Crocisa major Jor. Schletterer (Pola), a Sphegid
(Crabro clypeatus Z.), and 4 bees (1. Crocisa major Mor.; 2. Halictus leucozonius
Schr.9; 3. H. scabiosae Ross?; 4. Megachile apicalis Spzz.).

1576. C. rhenana Bor. (=C. paniculata Jacg., and C. maculosa auct., non
Lam.).—

VistTors.—Loew observed the following in Silesia (‘ Beitrage,’ pp. 26—7).—

A. Coleoptera. Chrysomelidae: 1. Cryptocephalus sericeus Z. B. Diptera.
(a) Bombylitdae: 2. Bombylius minor Z., skg.; 3. Systoechus sulphureus A/zkaz., do.
(4) Conopidae: all skg.: 4. Myopa fasciata Mg.; 5. Physocephala nigra Deg. 9;
6. P. truncata Zw. $; 7. P. vittata ¥. 8. (c) Stratomyzdae: 8. Odontomyia hydroleon
L., skg. (d) Syrphidae: all skg.: 9. Eristalis tenax Z.; 10. Syrphus lineola Zevz. ;
11. S. pyrastri Z.; 12. Volucella bombylans Z. C. Hymenoptera. (a) Apidae:
13. Andrena pilipes /. 9, po-clig. and skg.; 14. Bombus rajellus X. ¥, po-cltg. ;
15. B. variabilis Schmzedekn. Y, do.; 16. Coelioxys punctata Zep. 9 and 4, skg. ;
17. Dasypoda hirtipes #. 9 and 4, do., 9 actively po-cltg.; 18. Eucera (Tetralonia)
pollinosa Zep. 9, skg.; 19. Halictus leucozonius Schr. 9, do.; 20. H. quadristrigatus
Lir. 9, do.; 21. Megachile argentata F. 6, do.; 22. M. fasciata Sm. 6, do.; 23. M.
maritima X. 9, po-cltg.; 24. M. octosignata Vy/. 9, do.; 25. Nomada jacobaeae Ps.
g and 4, skg. ; 26. Osmia solskyi Mor.9, do.; 27. Psithyrus rupestris /. 9, do.; 28.
Saropoda rotundata /. 9, do. (4) Sphegidae: 29. Bembex rostrata &, 9 and 4, skg.
D. Lepidoptera. (a) Noctuidae: 30. Acronycta aceris Z., skg.; 31. Plusia gamma
L., do. (5) Rhopalocera: all skg.: 32. Argynnis aglaia Z.; 33. Melanargia galatea
L.; 34. Papilio machaon Z.; 35. Pieris brassicae Z.; 36. P. daplidice Z.; 37.
Vanessa cardui Z.; 38. V. urticae LZ. (c) Sphingidae: 39. Ino statices Z., skg.
Also (Tyrol), 3 Cerambycid beetles (1. Clytus ornatus #.; 2. C. plebeius /;
3. Mylabris floralis Pa/?.).

The following were recorded by the observers, and for the localities stated.—

Gersticker (Chiavanna and Meran), 3 bees—1. Ceratina cucurbitina Rossd,
freq., po-cltg.; 2. C. gravidula Gers? 9, po-cltg.; 3. Megachile melanopyga Coséa 9,
po-cltg. Schletterer and von Dalla Torre (Tyrol), the same bees.

1577. C. arenaria Bieb.—

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Schletterer (Tyrol), 4 bees—1. Anthidium manicatum Z.; 2. A. septemdentatum
Lir.; 3. Epeolus tristis Sm. (=E. luctuosus £v.); 4. Melitta leporina Pz. Friese
(Alps), the bee Eucera dentata A7g.: (Hungary), the bee E. graja Ev. Alfken
(Bozen), the Pentatomid Carpoceris nigricornis /, and 10 bees—1. Anthidium
laterale Z/r., in vast numbers, one on almost every capitulum, 9 skg. and po-cltg.,
$ skg.; 2. A. manicatum Z., less freq., 9 and $; 3. A. septemspinosum Z., rare;
4. Andrena carbonaria Z. 9, very common, skg. and po-cltg.; 5. Eriades crenulata
Nyl. 9, freq., po-cltg.; 6. Eucera dentata Klug. é, rare; 7. Megachile apicalis Spr7.,
exceedingly common, 9 skg. and po-cltg., skg.; 8. M. lagopoda Z. 9, skg. and
po-cltg. ; 9. M. pilicrus Aor. 9; 10. Xylocopa violacea Z., occasional.

1578. C. Bibersteinii Schur.—
VistTors.—Friese (Hungary) observed the following bees.—

668 ANGIOSPERMAE—DICOTYLEDONES

Two parasitic species; 1. Ammobates vinctus Gerst., freq.; 2. Pasites minutus
Mocs.; and 8 po-cltg. forms; 3. Camptopoeum frontale F,, freq.; 4. Eucera graja
£v.: 5. Lithurgus chrysurus Fonsc.; 6. L. fuscipennis Lep.; 7. Osmia bidentata
Mor.; 8. O. dives Mocs.; 9. O. spinulosa X.; 10. Podalirius bimaculatus Pz.

1579. C. vallesiaca Jord.—

VisiTors.—Friese (Switzerland) observed the bee Podalirius bimaculatus Pz.,
and the rare parasitic bee Stelis frey-gessneri Friese.

1580. C. solstitialis L—

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Friese (Hungary), 5 bees—1. Lythurgus chrysurus Fonsc.; 2. L. fuscipennis
Lep.; 3. Osmia bidentata Mor.; 4. O. dives Mocs., occasional; 5. O. spinulosa X.
Schletterer (Pola), a Sphegid (Crabro clypeatus Z.) and 3 bees—1. Crocisa major
Mor.; 2. Halictus calceatus Scop.; 3. Lithurgus chrysurus Fonsc.

1581. C. amara L.—

VisiTors.— Schletterer observed the small bee Podalirius bimaculatus Pz. at Pola.

1582. C. nigrescens Willd.—

VisiTors.—Schletterer records 3 bees for the Tyrol—1. Anthidium oblongatum
Lir.; 2. Andrena nitida Fourcr.; 3. Halictus calceatus Scop.

1583. C. argentea L.—

VisiTors.—Loew observed 2 hover-flies (Eristalis arbustorum Z., and E. nemorum
L.) in the Berlin Botanic Garden.

1584. C. astrachanica Spreng.—

Vistrors.—Loew observed a humble-bee (Bombus terrester Z. $, skg.) and
a butterfly (Pieris brassicae Z., skg.) in the Berlin Botanic Garden.
1585. C. atropurpurea Waldst. et Kit.—

VistTors.—Loew (Berlin Botanic Garden) observed 2 bees—1. Bombus pratorum
L.%, skg.; 2. B. terrester Z. 4, do. (also on the var. ochroleuca).

1586. C. calocephala Willd—

Vis1Tors.—Loew (Berlin Botanic Garden) observed a beetle (Cetonia aurata Z.),
2 hover-flies (Eristalis tenax Z., and Syritta pipiens Z.), and a butterfly (Vanessa
urticae Z., skg.).

1587. C. conglomerata C. A. Mey.—

Visitors.—Loew (Berlin Botanic Garden) observed a humble-bee (Bombus
terrester Z. 6, skg.) and a butterfly (Epinephele janira Z., skg.).

1588. C. dealbata Willd. —

VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Syrphus pyrastri Z. B. Hymenoptera. Apidae:
2. Apis mellifica Z. ¥, skg. and po-cltg.; 3. Megachile centuncularis Z. 9, po-cltg. ;
4. Osmia fulviventris Pz. 9, do.; Prosopis communis JVy/. 6, skg.

1589. C. Endresii Hochst. et Steud.—

Vistrors.—Loew observed a hover-fly (Syrphus corollae /.) in the Berlin Botanic
Garden.

COMPOSITAE 669

1590. C. Fischeri Willd— ,
Visitors.—Loew observed the following bees in the Berlin Botanic Garden.—

1. Anthidium manicatum Z. 9, po-dvg.; 2. Apis mellifica Z. ¥, skg.; 3. Mega-
chile lagopoda Z. 8, do.; 4. Osmia fulviventris Pz. $, do.; 5. O. papaveris Z/r. é,
do.; 6. Stelis phaeoptera K. 9, do.: also the wasp Odynerus parietum Z. 9 and é.

1591. C. Fontanesii Spach.—

Vistrors.—Loew observed .a hover-fly (Syrphus balteatus Deg.) in the Berlin
Botanic Garden.

1592. C. leucolepis DC.—

VisiTors.—Loew observed a humble-bee (Bombus pratorum Z. 8, skg.) in the
Berlin Botanic Garden.

1593. C. microptilon Godr. et Gren.—

VisiTors.—Loew (Berlin Botanic Garden) observed a hover-fly (Syrphus ribesii
L.) and a bee (Halictus cylindricus &. 8, skg.).

1594. C. ochroleuca Pushk.—

Visitors.—Loew observed a bee (Osmia fulviventris Pz. 6, skg.) in the Berlin
Botanic Garden.—

1595. C. orientalis L.—
VisiTors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. SyrpArdae: 1, Eristalis arbustorum Z.; 2. Syrphus balteatus Deg.
B. Hymenoptera. Apidae: 3. Apis mellifica Z. ¥, skg.; 4. Psithyrus rupestris /.
$, do.; 5. P. vestalis Mourcr. 8, do. C, Lepidoptera. Rfopalocera: 6. Vanessa
urticae Z., skg.

1596. C. rigidifolia Bess.—

Vistrors.—Loew (Berlin Botanic Garden) observed a hover-fly (Eristalis intri-
carius Z.) and a bee (Psithyrus rupestris /. 4, skg.).
1597. C. rupestris L.—

Visrtors.—Loew (Berlin Botanic Garden) observed a beetle (Cetonia aurata Z.)
and a bee (Bombus terrester Z. 6, skg.).

1598. C. ruthenica Lam.—

Vistrors.—Loew (Berlin Botanic Garden) observed 2 hover-flies (Syritta pipiens
Z., and Syrphus corollae F.) and 2 bees (Bombus terrester Z. 5, skg., and Megachile
lagopoda ZL. 9, po-cltg.

1599. C. salicifolia Bieb.—

Visirors.—Loew (Berlin Botanic Garden) observed a humble-bee (Bombus

terrester Z. §. skg.) and 2 butterflies (Argynnis latonia Z., skg., and Pararge
megaera J.).

1600. C. salonitana Vis.—

Visrrors.—Loew (Berlin Botanic Garden) observed a bee (Osmia fulviventris Pz.
9, po-cltg.) and a butterfly (Pieris brassicae Z., skg.).

1601. C. stereophylla Bess.—

Visrrors.—Loew observed the parasitic humble-bee Psithyrus vestalis Fourcr. 3,
skg., in the Berlin Botanic Garden. :

670 ANGIOSPERMAE—DICOTYLEDONES

.480, Xeranthemum Tourn.

Short conical tip of the style of the hermaphrodite disk-florets covered as far
down as the cleft with sweeping-hairs directed obliquely upwards; inner surfaces
of the stylar branches beset with stigmatic papillae. There are neither stigmatic
papillae nor sweeping-hairs on the styles of the neuter ray-florets.

1602, X. annuum L. (Sprengel, ‘Entd. Geh.,’ p. 371; Hildebrand, ‘U. d.
Geschlechtsverhilt. b. d. Compositen,’ pp. 48-50, Taf. V, Figs. 24-30.)—Sprengel
regarded the ray-florets of this species as female, but Hildebrand says that their
ovary, though fairly well developed, never contains an ovule. In the second stage of
anthesis the stylar branches diverge and expose their papillose stigmatic surfaces to
insects.

B. Liguliflorae Less. (=Cichoriaceae Juss.)

Florets all ligulate and hermaphrodite ; style unjointed, its filiform branches are
covered with soft hairs, and become reflexed. Kirchner observed that the species of
this group present a marked agreement as regards their flower mechanism. The
corolla is a tube closed above, before the florets open; it is split on the inner
side and produced into a ligulate expansion. When the style grows through the
anther-cylinder, it does not push the pollen before it, but becomes covered with
this on its outer surface, which is beset with sweeping-hairs. With regard to insect-
visits Hermann Miiller remarks that as the anther-cylinders usually project for
several mm., and the styles as much further, most of the visitors creep about between
the ends of the latter rather than over them, so as to transfer pollen with their sides
rather than their ventral surfaces. The simultaneous pollination of numerous florets
therefore takes place to a less extent than among the Senecionidae and Asteroideae,
where heaps of pollen are succeeded by the stigmatic surfaces at the same level. On
the other hand, the florets of the Ligulatae can simultaneously receive pollen brought
by insect visitors and dust these with their own pollen.

The yellow-flowered species of the group are visited with special eagerness by
bees of the genus Panurgus.

481. Catananche L.

1603. C. lutea L. (Murbeck, Vet.-Ak. Ofvers., Stockholm, lviii, 1901-2.) —
Murbeck states that in this Mediterranean species subterranean heads are borne
in the axils of the outer leaves of the rosette. These only contain 1-3 cleistogamous
florets. j

482. Lampsana Tourn.

Florets yellow. Style covered externally with sweeping-hairs, which extend
along way down, and closely beset with stigmatic papillae internally.

1604. L. communis L. (=Lapsana communis Z.). (Herm. Miller, ‘ Fer-
tilisation,’ pp. 351-2, ‘Weit. Beob.,’ III, pp. 97-8; Kirchner, ‘Flora v. Stuttgart,’
P- 733; MacLeod, Bot. Jaarb. Dodonaea, Ghent, iii, 1891, v, 1893; Knuth, ‘BI.
u. Insekt. a. d. nordfr. Ins.,’ pp. 96, 191 ; Warnstorf, Verh. bot. Ver., Berlin, xxxviii,
1896.)—In this species the heads are solitary, and not very conspicuous. Hermann

COMPOSITAE 671

Miiller says that they contain only 8-17 florets with corolla-tubes 14-23 mm. and
ligules 4-6 mm. long, the whole diameter being only 8-10 mm. The style projects
13-2 mm. from the anther-cylinder, which in turn exceeds the corolla-tube by
2-3 mm. The stylar branches are only 3 mm. long. During the second stage
of anthesis, they diverge widely in such a way that automatic self-pollination is
regularly effected, if insect visitors have not previously removed the pollen. Kerner
states that the heads open between 6 and 7 a.m. in favourable weather, closing again
as early as 10 or 11 o'clock in the forenoon; in unfavourable weather they remain
completely closed. According to Warnstorf, they open between 6 and 7 a.m. at
Neu-Ruppin, and close again between 3 and 4 p.m. The pollen-grains are yellow
in colour, and polyhedral; they bear spinose tubercles on their edges, and average
31 » in diameter.

queen following were recorded by the observers, and for the localities
stated.—

Knuth (North Frisian Islands) 3 hover-flies. Herm. Miiller (Westphalia),
3 po-dvg. hover-flies (1 Eristalis arbustorum Z.; 2. E. nemorum Z.; 3. E. sepulcralis
L.); Buddeberg (Nassau), a hover-fly (Ascia podagrica #., po.-dvg.), and 3 bees
(1. Halictus leucozonicus Schr., po-cltg.; 2. H.morio /. 4, skg.; 3. H. smeathmanellus
&. 9, do.). MacLeod (Flanders), a hover-fly, 3 Muscids, and a Lepidopterid (Bot.
Jaarb. Dodonaea, Ghent, v, 1893, p. 428); (Pyrenees), a Muscid (op. cit., iii, 1891,
Pp. 364). Scott-Elliot (Dumfriesshire), a short-tongued bee, a hover-fly, and 3 Muscids
(‘Flora of Dumfriesshire,’ p. 108).

483. Aposeris Neck.

1605. A. foetida Less. (=Hyposeris foetida Z., and Lampsana foetida /. W.
Schmidt). (Briquet, ‘Etudes de biol. flor. dans les Alpes occident.’)—Briquet says
that in this species the head contains 10-25 yellow florets, and has a diameter
of 25-30 mm. Each floret possesses a ligule 13-15 mm. and a corolla-tube
2-23 mm. in length. The anther-cylinder is about 4 mm. long, and the style
projects as far beyond it, ultimately rolling back its branches so far that they
can be self-pollinated.

Visirors.—Few in number. Kirchner says they are beetles, Diptera, and
even humble-bees, which effect both cross- and self-pollination.

484. Arnoseris Gaertn.

Heads small and yellow. Style (below the cleft) covered with short sweeping-
hairs that stand out horizontally; inner surfaces
of the stylar branches beset with stigmatic papillae.

1606. A. minima Dum. (=A. pusilla
Gaertn., and Hyoseris minima Z.). (Knuth, ‘BI.
u. Insekt. a. d. nordfr. Ins.,’ pp. 96, 161; ‘ Weit.
Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 229.)—
The heads of this species examined by me in the
island of Féhr were 8 mm. in diameter. They _ Fic.208. Arnoseris minima, Dum.

. (from nature, enlarged). (1) Stylar branches
consisted of 20-5 florets, each 6 mm. in length; (x) closed; g, stylar brush. (2) Do., di-
the ligule was 3 mm. long and 1X mm. broad. VYeTeing; 4 Pollen-grains on the brush.

In the second stage of anthesis the stylar branches diverge in a crescentic manner.

672 ANGIOSPERMAE—DICOTYLEDONES

Visirors.—Knuth (Féhr) saw 2 hover-flies and a minute Muscid; MacLeod
(Flanders), a small fly (Bot. Jaarb. Dodonaea, Ghent, vi, 1894, p. 374)

485. Hyoseris L.
1607. H. radiata L.—
VistTors.—Delpino observed the bee Megachile centuncularis Z. (‘Ult. oss.,’
p. 125).

486. Anandria Siegesb.

As Linnaeus had already recognized there are cleistogamous flowers in species
of this genus (H. von Mohl, Bot. Ztg., Leipzig, xxi, 1863).

487. Cichorium L.

Florets blue in colour, rarely red or White. The style is covered externally with
strong sweeping-hairs directed obliquely upwards and extending for some distance
below the cleft: inner surfaces of the branches beset with stigmatic papillae.

1608. C. Intybus L. (Hildebrand, ‘U. d. Geschlechtsverhilt. b. d. Com-
positen,’ p. 10, Taf. I, Figs. 8-10; Herm. Miller, ‘Fertilisation,’ p. 351, ‘ Weit.
Beob.,’ II, p. 97; Loew, ‘Bliitenbiol. Floristik,’ p. 390; Knuth, ‘Bloemenbiol.
Bijdragen,’ ‘ Bliitenbiol. Herbstbeob.’; Warnstorf, Verh. bot. Ver., Berlin, xxxviii,
1896.)—In this species the heads display in the sunshine a disk about 30 mm.
broad, and usually blue in colour. They contain relatively few florets, but the
ligules are 12-14 mm. in length, though the corolla-tubes are only 3 mm. long.
In the second stage of anthesis the stylar branches are twisted into a spiral of
1-2 coils, so that the inner stigmatic surfaces come into contact with the pollen-
grains that remain among the sweeping-hairs, and automatic self-pollination is
thus effected in the absence of insect visitors.

The capitula, according to Linnaeus, open about 5 o'clock a.m. at Upsala,
closing again at ro o'clock in the forenoon; at Innsbruck, according to Kerner,
they only open about 6-7 a.m., and close again at 2-3 p.m. Warnstorf says
that they open at Neu-Ruppin between 6 and 7 a.m. The pollen-grains are
white in colour, polyhedral, with spinose tubercles on their edges, and their
average diameter is 46 p.

Visirors.—Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list
for Westphalia and Nassau.—

A. Coleoptera. Télephoridac: 1. Malachius bipustulatus Z., freq., po-dvg.
(H. M.). B. Diptera. (@) Conopidae: 2. Sicus ferrugineus Z., skg. (H. M.).
(6) Syrphidae: 3. Eristalis tenax Z., skg. and po-dvg. (H. M.); 4. Syritta pipiens
L£. do. (H. M.). C. Hymenoptera. Afrdae: 5. Andrena fulvicrus X. 9, po-cltg.
(H.M.); 6. Apis mellifica Z. ¥, freq., skg. (H.M.); 7. Chelostoma campanularum Z.
9 (Budd.); 8. Dasypoda hirtipes#. 4, skg.(H.M.); 9. Halictus albipes/. 6, do. (H.M.) ;
10. H. cylindricus /. 9, do. (H.M.); 11. H. interruptus Pz. 9, do. (H. M., Thuringia) ;
12. H. leucozonius Schr. 6, do. (Budd.); 13. H. longulus Sm. 9, do. (H. M.); 14. H.
nitidiusculus K. , do. (H.M.); 15. H. quadricinctus #. 4, do. (H.M.); 16. H. rubi-
cundus Chr. 5, do. (H. M.); 17. H. smeathmanellus X. 9, do. (H. M., Bavarian
Oberpfalz); 18. H. tetrazonius K/g. 9, do. (H. M., Bavarian Oberpfalz); 19. Osmia
adunca Pz. 4, do. (H.M., Kitzingen) ; 20. O. spinulosa X. 9, not infrequent, skg. and

COMPOSITAE 673

po-cltg. (H. M., Thuringia) ; 21. Prosopis nigrita F. 8, in large numbers, skg. (H. M.,
Bavarian Oberpfalz). D. Lepidoptera. Rhopalocera: 22. Colias hyale Z., skg.
(H. M., Thuringia).

The following were recorded by the observers, and for the localities stated.—

Knuth (Kiel), 6 skg. and po-dvg. hover-flies—r1. Eristalis sp.; 2. Melanostoma
sp.; 3- Platycheirus podagratus Ze/#., 4. Syrphus balteatus Deg. 3; 5. S. ribesii Z. ;
6. S. umbellatarum #. Alfken (Bremen), 4 bees—1. Eriades nigricornis Wy. 9;
2. Halictus calceatus Scop. 8; 3. H. flavipes Pz. 9; 4. H. morio ¥.9. Schenck
(Nassau), 4 bees—r. Anthidium punctatum Z/r.; 2. Halictus leucozonius Schr. ;
3. H. lucidulus Schenck; 4. H. sexnotatulus Wy/. (=H. quadrifasciatus Schenck).
Friese, the bee Panurgus banksianus K.: (Alsace), the bee Dasypoda plumipes Pz.
Loew (Brandenburg) (‘Beitrage,’ p. 39), a hover-fly (Eristalis sepulcralis Z.) and
3 bees (1. Dasypoda hirtipes #. 8, skg.; 2. Halictus cylindricus #. 3, do.; 3. H.
sexnotatus X. 4, do.).

1609. C. Endivia L. (Knuth, ‘Bliitenbiol. Herbstbeob.”)—The flower
mechanism of this species agrees with that of C. Intybus. The blue florets
develop to a great size and spread out their ligules to attract insects, while
the reproductive organs form a circle in the middle of the inflorescence. In
both species the ligules are about 2 cm. long and 6—7 mm. broad. In C. Endivia
the diameter of the head is 4-5 cm.; while in C. Intybus it is somewhat smaller
as a rule. The number of florets in an inflorescence is 20-30 in the former
species, 12-20 in the latter.

Vistrors.—Knuth observed a hover-fly (Eristalis sp.) and a butterfly (Pieris sp.).

488. Thrincia Roth.

1610. T. hirta Roth (=Leontodon hirtus L.),—

Vistrors.—Herm. Miiller gives the following list for Westphalia and the Bavarian
Oberpfalz (‘ Fertilisation,’ pp. 358-9, ‘ Weit. Beob.,’ III, p. 97).—

A. Coleoptera. Aupresiidae: 1. Anthaxia quadripunctata Z., 1” copuldé on the
florets. B. Diptera. Syrphidac: 2. Eristalis arbustorum Z., freq., skg. and po-dvg. ;
3. E. sepulcralis Z., po-dvg.; 4. E. tenax Z., freq., skg. and po-dvg.; 5. Syrphus
balteatus Deg., do. C. Hymenoptera. (a) Aprdae: 6. Andrena denticulata X. 9,
skg. and po-dvg. (Sld., Thiiringia); 7. A. fulvago Chr. 9, po-cltg. (Thuringia); 8.
A. fulvescens Sm. 9, do. (Thuringia); 9. A. fulvicrus A. 9, po-cltg.; 10. Bombus
confusus Schenck 8, skg.; 11. B. tristis Sed. 6, do. (Liebenau near Schwiebus) ;
12. Cilissa melanura MWVyl. 9; 13. Dasypoda hirtipes #. %, skg. (Liebenau near
Schwiebus) ; 14. Dufourea vulgaris Schenck 9 and 6, po-cltg. and skg.; 15. Halictus
cylindricus #. 9 and 4, do.; 16. H. flavipes #. 4, skg.; 17. H. leucozonius Schr.
g and 6, very numerous, skg. and po-cltg. (Thuringia); 18. H. lugubris X. 4, skg. ;
19. H. maculatus Sm. %, do.; 20. H. sexcinctus /. 9, skg. and po-cltg.; 21. H.
smeathmanellus XX. 9, po-cltg.; 22. H. villosulus X. 9, do.; 23. Panurgus calcaratus
Scop. 9 and 4, freq., po-cltg. and skg. (4) Sphegidae: 24. Cerceris variabilis Schr. 9,
skg. D. Lepidoptera. (a) Noctuidae: 25. Plusia gamma L., skg. (6) Rhopalocera:
26. Pieris napi Z., skg.

The following were recorded by the observers, and for the localities stated.—

Alfken and Héppner (Bremen), 10 bees—1. Bombus agrorum #9; 2. B. areni-
cola Zhs. ¥; 3. B. lapidarius Z. ¥; 4. B. sylvarum Z. §; 5. B. variabilis Schmiedekn.
¥; 6. Halictus calceatus Scop. 9 and; 7. H. flavipes /. 9 and $; 8. H. leucozonius
Schr. 9 and 6; 9. H. rubicundus Chr. 9 and $; 10. ‘H. zonulus Sm. 9 and 4.
MacLeod (Flanders), 2 humble-bees, 5 hover-flics, a Muscid, and 3 Lepidoptera
(Bot. Jaarb. Dodonaea, Ghent, v, r893, pp. 428-9).

DAVIS. II x x

674 ANGIOSPERMAE—DICOTYLEDONES

489. Leontodon L.

Florets yellow. Styles rather closely covered outside with sharp sweeping-hairs
extending far above the cleft; inner surfaces of the stylar branches—which sometimes
do not entirely separate—densely beset with stigmatic papillae. Kerner states that
geitonogamy by pollen from the inner florets is automatically effected by the spreading
of the stylar branches. He further remarks that automatic self-pollination may also

FIG. 209. Leontodon autumnalis, L.(after Herm. Miller).
(1) Floret in the second (female) stage, after removal of
calyx and ovary (x 7). (2) End of style of do. (x 35).
a, sweeping-hairs ; 6, stigmatic papillae; c, pollen-grains.

be brought about by gradual elonga-
tion of the ligulate corolla, which
thus carries up the pollen adhering
to it.

16u. L.autumnalis L. (Herm.
Miiller, ‘ Fertilisation,’ pp. 356-8,
‘Weit. Beob.,’ III, pp. 96-7; Knuth,
‘Bl. u. Insekt. a. d. nordfr. Ins.,’
pp. 96-7, 161, ‘Bl. u. Insekt. a. d.
Halligen,’ p. 37, ‘Weit. Beob. ti. BI.
u. Insekt. a. d. nordfr. Ins.,’ pp. 236-7,
‘Bliitenbiol. Beob. a. d. Ins. Riigen,’
‘Bloemenbiol. Bijdragen’; Lindman,
‘Bidrag till Kéanned. om Skandin.
Fjellvaxt Blomn. o. Befrukt.’; Verhoef,
‘Bl. u. Insekt. a. d. Ins. Norderney’;
MacLeod, Bot. Jaarb. Dodonaea,
Ghent, v, 1893.)—The heads of this
species expand in the sunshine to
form a golden-yellow disk 20-30 mm.
broad, but in rainy weather they
contract to a diameter of 5 mm.
Hermann Miiller says that each head
contains 40-70 florets, with corolla-
tubes 23-5 mm. and ligules 7-12 mm.
long. The style projects 3-4 mm.
beyond the tube, into the wider part
of which the nectar rises. If insect-
visitors are sufficiently numerous at
the right time, the pollen is removed
from the sweeping-hairs before the
stigmatic surfaces unfold; at a later
stage insects may effect self-pollination.

The latter may also take place automatically should insect-visits fail, by the stigmatic

surfaces coming into contact with the pollen.

Vistrors.—Knuth observed the following in Schleswig-Holstein (S.-H.), Riigen
(R.), and the dunes and high land of Helgoland (H.).—

A. Coleoptera. (a) Chrysomelidae: 1. Cryptocephalus sericeus Z., po-dvg.(R.).
(3) Telephoridae: 2. Psilothrix cyanea O/., po-dvg.(H.). B. Diptera. (a) Muscidae:

COMPOSITAE 675

3. Anthomyia sp. (S.-H.); 4. Aricia incana Wied. (S.-H.); 5. Coleopa frigida Fad/.
(H.).; 6. Lucilia caesar Z. (H.); 7. Sarcophaga sp. (S.-H.) ; 8. Scatophaga stercoraria
£.(S.-H.) ; 9. Small Muscids (S.-H.). (4) Syrphidae: all skg. and po-dvg.: ro. Eristalis
arbustorum JZ. (S.-H.); 11. E. sp. (S.-H.); 12. E. tenax Z. (S.-H.); 13. Helophilus
pendulus Z. (S.-H.); 14. H. trivittatus &. 9 (S.-H.); 15. Syrphus balteatus Deg. (S.-H.) ;
16. S. ribesii Z. (SH. and R.); 17. Volucella bombylans Z. (S.-H.). C. Hy-
menoptera. Apidae: all skg. or po-cltg.: 18. Apis mellifica Z. (S.-H.); 19.
Bombus agrorum /”. (S.-H.); 20. B. derhamellus X. 9 (S.-H.); 21. B. lapidarius
LZ. (S.-H.); 22. Colletus daviesanus KX. (S.-H.); 23. Dasypoda plumipes Pz.
(S.-H.); 24. Panurgus banksianus X. (S.-H.); 25. P. calcaratus Scop. (=P.
lobatus /.) (S.-H. and R.). D. Lepidoptera. All skg. (a) Hesperidae: 26.
Hesperia lineola O.(R.). (4) Microlepidoptera: 27. Tortrix sp. (S.-H.). (c) Moc-
tuidae: 28. Plusia gamma ZL. (S.-H.). (@) Rhopalocera: 29. Argynnis adippe
L.(R.); 30. A. paphia Z.(R.); 31. Epinephele janira Z. (S.-H.); 32. Pieris sp.
(S.-H.); 33. Polyommatus phlaeas Z. (S.-H.). (e) Sphengidae: 34. Zygaena
filipendula Z. (S.-H.); Z. 2 sp. (R.). And in Thuringia (‘Bliitenbiol. Beob. in
Thiiringen,’ p. 36).—A. Diptera. (a) Muscidae: all skg. and po-dvg.: 1. Aricia basalis
Zett., freq.; 2. A.serva Mg. (4) Syrphidae: 3. Eristalis pertinax Scop. 9; 4. Syrphus
annulipes Ze//. 9; 5. S. ribesii Z. 6. B. Hymenoptera. <Apzdae: all skg. or
po-cltg.: 6. Bombus lapidarius ¥; 7. B. soroénsis F., var. proteus Gers/.¥; 8. B.
terrester Z.9; 9. Halictus leucozonius Schr. 9; 10. Psithyrus vestalis Fourcr. 3.
C. Lepidoptera. All skg.: 11. Argynnis adippe Z.; 12. A. paphia Z.; 13.
Epinephele janira Z.; 14. Pieris sp.; 15. Vanessa urticae Z.

Alfken observed the following at Bremen.—

A. Diptera. (2) Bombylidae: 1. Exoprosopis capucina Z. (4) Syrphidae:
2. Eristalis anthophorinus Ze//.; 3. E. intricarius Z.; 4. Helophilus pendulus JZ. ;
5. Melanostoma mellina Z. B. Hymenoptera. (a) Afzdae: 6. Andrena argentata
Sm.9; 7. A. combinata Chr.d; 8. A. denticulata K. 9, freq., skg. and po-cltg. ;
g. A. flavipes K., 2nd gen., freq., 9 skg. and po-cltg., 6 skg.; 10. A. marginata J. 6 ;
11, A. parvula &. 9; 12. A. propinqua Schenck, 2nd gen. 9 and $; 13. A. shawella K.
gand $; 14. A. tarsata Vyl.9 and 6; 15. Bombus agrorum /. ¥; 16. B. derhamellus
&. 9, po-cltg.; 17. B. distinguendus Mor. ¥ and $; 18. B. lapidarius Z. ¥ and $;
19. B. lucorum LZ. ¥; 20. B. muscorum /. ¥; 21. B. proteus Gerst. ¥ and $;
22. B. sylvarum Z.; 23. Coelioxys acuminata Wy/.d; 24. Colletes daviesanus &.
g and; 25. Dasypoda plumipes P2z., freq., 9 skg. and po-cltg., 6 skg. ; 26. Dufourea
vulgaris Schenck, very common, 9 skg. and po-cltg.; 27. Epeolus variegatus
ZL. 8; a8. Eriades truncorum Z. 9, po-cltg.; 29. Halictoides inermis Wyi. 6;
30. Halictus calceatus Scop. 9 and 6, very common; 31. H. brevicornis Schenck 9, rare ;
32. H. flavipes / 9 and 4, freq.; 33. H. levis A. 9 and 6; 34. H. leucopus X. 9;
35. H. leucozonius Schr., very common, 9 skg. and po-cltg., skg.; 36. H. malachurus
K.é; 37. H. minutissimus X. 6; 38. H. minutus X. 6; 39. H. punctulatus A.
g and 4, very common; 40. H. rubicundus CAr.$; 41. H. quadrinotatulus Schenck g3
42. H. semipunctulatus Schenck 9; 43. H. sexnotatus WVy/.9; 44. H. tumulorum Z. 4;
45. H. zonulus Sm. 9; 46. Megachile maritima X. 9, po-cltg.; 47. Melitta leporina Pz.,
rare, 9 po-cltg.,6 skg.; 48. Nomada fuscicornis Vy. 9 and 8; 49. N. solidaginis Pz. 9,
skg.; 50. Osmia solskyi Mor.9; 51. Panurgus banksianus X. 9 and 6, very common ;
52. P. calcaratus Scop. 9 and 4, do.; 53. Psithyrus barbutellus K.3; 54. P. rupestris
F. 3; 85. P. vestalis fourcr. 6; 56. Trachusa serratulae Pz. 4. (6) Sphegidae:
57. Crabro albilabris 7.9; 58. C. palmarius Schred. 9; 59: Diodontus tristis v. d. Z. 8.

Herm. Miiller gives the following list for Westphalia and the Bavarian Oberpfalz.—

A. Diptera. (a) Bombylitdae: 1. Systoechus sulphureus A/z.,skg. (6) Conopidae:

2. Sicus ferrugineus Z., skg. (¢c) Muscidae: 3. Sarcophaga carnaria Z.,skg. (d) Syr-

phidae: 4. Eristalis arbustorum Z., very common, skg. and po-dvg.; 5. E. sepulcralis
XX2

676 ANGIOSPERMAE—DICOTYLEDONES

L., do.; 6. E. tenax Z., do.; 7. Melithreptus taeniatus /g.,do.; 8. Syrphus balteatus
Deg., do.; 9. S. nitidicollis M%g., do.; 10. S. pyrastri Z., freq., do.; 11. Volucella
bombylans Z.,do. B. Hymenoptera. (a) Apidae: 12. Andrena fulvicrus KX. 9,
po-cltg.; 13. Apis mellifica Z. ¥, skg.; 14. Bombus agrorum Z. 9, do.; 15. B. lapi-
darius Z. ¥, do.; 16. Dasypoda hirtipes #. 9, po-cltg.; 17. Diphysis serratulae Pz,
g and 4, occasional, skg.; 18. Dufourea vulgaris Schenck 9 and 4, po-cltg. and skg. ;
19. Halictus cylindricus #. 8, skg.; 20. H. leucopus A. 9, skg. and po-cltg. ;
21. H. leucozonius Schr. 9, do.; 22. H. longulus Sm. 9, do.; 23. H. maculatus
Sm. , skg.; 24. H. morio F. 3, do.; 25. H. smeathmanellus K. é, do.; 26. H.
villosulus K. 9, skg. and po-cltg.; 27. Panurgus banksianus &. 9, po-cltg.; 28.
P. calcaratus Scop., skg. and po-cltg., often resting inactively among the florets;
29. Prosopis armillata yl. $, skg.; 30. Sphecodes gibbus Z. 9 and %, skg. and
po-dvg. (4) Sphegidae: 31. Pompilus viaticus Z., skg. C. Lepidoptera. (2) Moc-
turdae: 32. Plusia gamma L., skg. (6) Rhopalocera: 33. Argynnis aglaia L., freq.,
skg.; 34. Colias hyale Z., skg. (Thuringia).

H. de Vries observed the following 11 bees in the Netherlands (Ned. Kruidk.
Arch., Nijmegen, 2 Ser., 2. Deel, 1875).—

1. Bombus subterraneus Z. ¥; 2. B. terrester Z.¥; 3. Chelostoma florisomne
L.&; 4. Halictus cylindricus #8; 5. H. leucozonius Schr. 9; 6. H. nitidiusculus
K.é; 7. H. villosulus X. 9; 8. H. zonulus Sm. 9; 9. Nomada fabriciana LZ. 6;
ro. Panurgus banksianus K. 6; 11. Prosopis communis JVyi. 9. :

Willis noticed the following in the neighbourhood of the south coast of Scotland
(‘Fls. and Insects in Gt. Britain,’ Part I).—

A. Coleoptera. Curculionidae: 1. Sitona puncticollis S%ph. B. Diptera.
(a) Muscidae: 2. Anthomyia radicum Z.; 3. A. sp.; 4. Hydrellia griseola Fadl. ;
5. Tricophthicus cunctans A/g. (6) Mycetophilidae: 6. Sciara sp. (c) Syrphidae:
7. Brachyopa bicolor Fall.; 8. Eristalis aeneus Scop.; 9. E. pertinax Scop.;
10. E. tenax Z.; 11. Platycheirus manicatus M/g.; 12. Sericomyia borealis Fall. ;
13. Sphaerophoria scripta Z.; 14. Syrphus ribesii Z.; 15. S.sp. C. Hemiptera.
16. Acocephalus sp.; 17. Calocoris bipunctatus /.; 18. C. fulvomaculatus Deg. ;
19. Miris levigatus Z. D. Hymenoptera. (a) Apidae: 20. Bombus agrorum
F.; 21. B. terrester Z.; 22. Halictus rubicundus Chr. (6) Ichneumonidae: 23.
Ichneumon sp. E. Lepidoptera. (2) Microlepidoptera: 24. Crambus sp.;
25. Simaéthis fabriciana Sveph. (0) Rhopalocera: 26. Lycaena icarus Roét.

The following were recorded by the observers, and for the localities stated—

Sickmann (Osnabriick), the fossorial wasp Mellinus arvensis Z. Verhoeff
(Norderney), 2 hover-flies (Melithreptus menthastri Z. 9, and Syrphus ribesii Z. 9).
von Dalla Torre (Innsbruck Botanic Garden), the bee Halictus leucozonius Schr. 8.
MacLeod (Flanders), 3 humble-bees, 2 sp. of Halictus, 6 hover-flies, and 3 Lepidoptera
(Bot. Jaarb. Dodonaea, Ghent, v, 1893, p. 429). Scott-Elliot (Dumfriesshire), 2
humble-bees, numerous bees and flies (‘ Flora of Dumfriesshire,’ p. 102).

1612. L. hastilis L. (Kirchner, ‘Flora v. Stuttgart,’ p. 735.)—Kirchner
states that the heads of this species contain from 4o to over 80 florets, and
spread out in the forenoon in sunny weather into a golden-yellow disk of 20-25 mm:
broad. The corolla-tube is 4-6 mm. and the ligule 8-12 mm. long. The anther-
cylinder projects 3-4 mm. from the corolla-tube, and the style extends 4-5 mm.
further. The stylar branches are 2 mm. long, and finally roll back to form a spiral
of 13 turns. Warnstorf describes the pollen-grains as pure yellow in colour,
polyhedral, and spinose. measuring about 37-5 m in diameter.

COMPOSITAE 677

VistTors.—Herm. Miiller gives the following list for Westphalia (‘ Fertilisation,’
P- 358).—

A. Diptera. (a) Bombylitdae: 1. Systoechus sulphureus J/7., skg. (Sauerland).
(5) Conopidae: 2. Sicus ferrugineus Z., skg. (c) Syrphidae: 3. Cheilosia sp.;
4. Eristalis arbustorum Z., very common, po-dvg. and skg.; 5. E. horticola Deg.,
po-dvg. and skg.; 6. Melithreptus taeniatus MZ/g., do.; 7. Sericomyia lappona Z.,
skg. (Sauerland); 8. Volucella pellucens Z., in large numbers, skg. (Sauerland).
B. Hymenoptera. (a) Apzdae: 9. Andrena coitana KX. 9 and 4, skg. and po-cltg. ;
10. A. fulvescens Sm. 9, do., dusting itself with large quantities of pollen; rt.
Psithyrus barbutellus K. 9, skg.; 12. Bombus pratorum JZ. 9, skg. and po-cltg. ;
13. Halictus albipes /. 9, po-cltg.; 14. A. cylindricus /. 9, freq., po-cltg.; 15. H.
leucozonius Schr. 9, po-cltg.; 16. H. smeathmanellus X. 9, do.; 17. H. villosulus
K. 9, do. (6) Yenthredinidae: 18. Tenthredo sp. skg. C. Lepidoptera.
Rhopalocera: 19. Hesperia sylvanus Zsp., skg.

The following were recorded by the observers, and for the localities stated.—

MacLeod (Pyrenees), 9 bees, 7 Lepidoptera, 3 beetles, and 7 Muscids. Frey-
Gessner (Switzerland), the rare dasygastrid bee Eriadis grandis My/. Schenck
(Nassau), the bee Osmia leucomelaena A. Loew (Riigen), the bee Panurgus lobatus
fF. %, skg. (‘ Beitrage,’ p. 40): Silesia, the hover-fly Pipiza noctiluca Z. (?), and the
bee Anthidium strigatum Pz. (op. cit., p. 32); Riesengebirge (R.) and Silesia (S.),
the hover-fly Cheilosia canicularis Pz., skg. (R.), and the bee Andrena shawella K. 9,
skg. (S.) (op. cit., p. 51): Switzerland (S.) and the Tyrol (T.) (op. cit, p. 59)—
A. Diptera. Syrphidae: 1. Cheilosia antiqua Mg.(S.); 2. Merodon cinereus F.
(S.); 3. Syrphus confusus Zgg.(?) (S.). B. Hymenoptera. <Apzdae: 4. Andrena
proxima A. 9(T.). C. Lepidoptera. Rhopalocera: 5. Argynnis pales S.-V. (S.);
6. Colias phicomene £sp. (S.); 7. Erebia medea S\-V.(S.); 8. Melitaea parthenie
Bkh(S.): (Berlin Botanic Garden).—A. Diptera. Syrphidae: 1. Eristalis arbustorum
L.; 2. E. nemorum Z.; 3. E. tenax Z.; 4. Syrphus balteatus Deg.; 5. S. ribesii Z.
B. Hemiptera. 6. Strachia oleracea Z. C. Hymenoptera. Apidae: 7. Halictus
villosulus K. 9, skg. D. Lepidoptera. Rhopalocera: 8. Pieris brassicae L., skg. ;
g. Vanessa urticae Z., do. Herm. Miiller (Alps, L. hastilis, L. pyrenaeus, and other
sp.), 6 beetles, 21 flies, 29 Hymenoptera, and 43 Lepidoptera (‘ Alpenblumen,’
pp. 466-8). Scott-Elliot (Dumfriesshire), Apis, a humble-bee, and a Muscid (‘Flora
of Dumfriesshire,’ p. 102).

1613. L. asper Poir.—

Vistrors.—Loew (Berlin Botanic Garden) observed a hover-fly (Syrphus ribesii
Z.) and 2 bees (Halictus villosulus K. 9, po-cltg., and Heriades truncorum Z, 9, do.).

1614. L. crispus Vill.—

VistTors.—Loew (Berlin Botanic Garden) observed 2 bees (Halictus villosulus
K. 9, po-cltg., and Psithyrus campestris Pz. 8, skg.).

1615. L. pyrenaicus Gouan (?).—

Visttors.—MacLeod observed a hover-fly and a Muscid in the Pyrenees.

490. Picris L.

Florets yellow. Sweeping-hairs and stigmatic papillae as in Leontodon.—

1616. P. hieracioides L. (Herm. Miiller, ‘Fertilisation, pp. 352-3, ‘ Weit.
Beob.,’ III, p. 96.}—Hermann Miller states that in this species there are 44 to 75
florets in a head, and the marginal ones are larger than those in the centre. In
sunny weather the capitulum spreads out into a golden-yellow disk 24-36 mm.

678 ANGIOSPERMAE—DICOTYLEDONES

broad, but in dull weather it contracts to a diameter of 7 mm, As numerous
inflorescences are associated on a stem about a metre in height, the plant is very
conspicuous. The corolla-tube is 4-6 mm. and the ligule 8-12 mm. long. The
anther-cylinder projects 5 mm. beyond the corolla-tube, and the style 24-33 mm.
further. The stylar branches are 2 mm. long. When they diverge they often touch
each other, thus effecting automatic self-pollination.

Vistrors.—Herm. Miiller (H. M.) in Westphalia and Buddeberg (Budd.) in
Nassau observed the following. —

A, Diptera. (a) Empidae: 1. Empis livida Z., very numerous, skg. (Budd.).
(3) Syrphidae: 2. Chrysogaster viduata Z., skg. and po-dvg. (H. M.); 3. Eristalis
arbustorum Z., very common, do. (H. M.); 4. E. nemorum Z., do. (H. M.); 5. E.
sepulcralis Z., do. (H. M.); 6. E. tenax Z., do. (H. M.); 7. Melithreptus scriptus
L., skg. and po-dvg.(H.M.); 8. M. taeniatus M/y., do. (H.M.); 9. Syrphus balteatus
Deg., do. (H. M.). B. Hymenoptera. (a) Apidae: 10. Dasypoda hirtipes /.
g and 4, skg. and po-cltg. (Budd.); 11. Halictus albipes /. 4, skg. (H.M.); 12. H.
cylindricus /. 4, skg. (H. M., Budd.); 13. H. leucozonius Schr. 9, skg. and po-cltg.
(H. M.); 14. H. longulus Sm. 9, do., skg. (H. M.); 15. H. maculatus Sm. 9, do.
(H. M., Budd.); 16. H. minutus X. 9, do., 6 skg. (H4M.) ; 17. H. nitidiusculus X. 8,
skg. (H. M.); 18. H. quadricinctus #. 4, do. (H. M.); 19. H. rubicundus Car. 4, do.
(H. M.); 20. H. sexnotatus &., 9 skg. and po-cltg., & skg. (H. M.); 21. H.
smeathmanellus X. 9, do. (H. M.); 22. H. zonulus Sm. 9, do.(H. M.); 23. Heriades
truncorum Z. 9, po-cltg. (H. M.); 24. Osmia leucomelaena &. 9, skg. (H. M.,
Thuringia); 25. O. spinulosa X. 9, po-cltg. (H. M., Thuringia); 26. Panurgus
calcaratus Scop. 9 and 4, freq., skg. and po-cltg. (H. M., Budd.); 27. Dufourea
vulgaris Schenck, 9 freq., 6 scarce (H. M., Thuringia). (6) Sphegidae: 28. Crabro
sexcinctus #. 9 (H. M.). (c) Vespidae: 29. Vespa sylvestris Scop. ¥, pushing its
head deep into the florets (H.M.). C. Lepidoptera. Rhopalocera: 30. Pieris
brassicae Z., very common, skg. (H. M.); 31. P. rapae Z., do. (H.M.); 32. Epinephele
janira Z., skg. (H. M.).

The following were recorded by the observers, and for the localities stated—

Loew (Berlin Botanic Garden), the butterfly Argynnis latonia Z., skg. the flowers.
Friese, 7 bees in Baden (B.), Alsace (A.), Mecklenburg (M.), and Hungary (H.)—
1. Andrena bimaculata X. (H.), znd gen., not infreq. ; 2. Dasypoda plumipes Pz. (A.),
freq.; 3. Dufourea vulgaris Schenck (B.), not infrequent (H.); 4. Eriades truncorum Z.
(M.), freq. (H.); 5. Osmia villosa Schenck, not infrequent (B.); 6. Panurgus banksianus
XK. (B.); 7. P. calcaratus Scop. (B.). Schenck (Nassau), 11 bees—1. Dasypoda
plumipes Pz.; 2. Dufourea vulgaris Schenck; 3. Eriades truncorum Z.; 4. Halictus
albipes #. 4; 5. H. levigatus X.8; 6. H. levis K.; 7. H. villosulus X. (=H. punc-
tulatus X.); 8. Macropis labiata #; 9. Panurgus calcaratus Scof.; 10. Stelis
breviuscula WVy/.; 11. S. ornatula Klug.,a9. Schletterer (Tyrol), the bee Halictus
albipes #. H. de Vries (Netherlands), 5 bees (Ned. Kruidk. Arch., Nijmegen, 2. Ser.,
2. Deel, 1875)—-1. Bombus agrorum F. 6; 2. B. terrester Z. ¥; 3. Halictus
cylindricus #.8; 4. H. leucozonius Schr. 8; 5. Osmia spinulosa K. 9. MacLeod
(Pyrenees), a humble-bee, 2 species of Panurgus, 2 beetles, and 3 flies (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, p. 366).

491. Helminthia Juss.
As Picris.

1617. H. echioides Gaertn.—According to Kirchner (‘Beitrage,’ pp. 71-2), the
golden-yellow heads of this species spread out in the sun to form a disk 20 mm.

COMPOSITAE 679

broad. Towards the end of anthesis the stylar branches roll back into a spiral
of 13 to 13 turns, so that the papillose stigmatic surfaces easily come into contact
with the pollen still clinging to the sweeping-hairs.

Visirors.—Sprengel observed the honey-bee (‘Entd. Geh.,’ p. 367).

492. Tragopogon L.

Florets yellow in colour, more rarely violet; hermaphrodite and ligulate. Style
with sweeping-hairs on its outer surface, and stigmatic papillae on the inner surfaces
of its branches, which roll up at a later stage. Kerner states that the stylar branches
of the outer florets roll back when they diverge, so as to come into contact with the
pollen of the inner ones. The florets of the outer whorl are exactly opposite the
spaces between the florets of the adjacent inner whorl, and this favours geitonogamy,
for one of the stylar branches will touch the pollen-covered style of the floret on
the right, while the other will touch the style of that on the left.

1618. T. pratensis L.—According to Kirchner (‘Flora v. Stuttgart,’ p. 737),
the heads contain 20-50 golden-yellow florets, and when the weather is sunny they
spread out in the forenoon to a diameter of 60 mm.; in the afternoon and during
dull weather they are closed. Linnaeus states that they open 3-5 a.m. at Upsala,
closing again between 8 and 10 a.m. The corolla-tube of the marginal florets is
6—7 mm. and the ligules as much as 30 mm. long. The central florets are smaller,
with a corolla-tube of 5 mm. and a ligule of 7 mm. The stylar branches are 3 mm.
long, and bend so far backwards that they make a spiral of several turns. Automatic
self-pollination is thus effected if some pollen still clings to the sweeping-hairs.
Warnstorf describes the pollen-grains as golden-yellow in colour, polyhedral, beset
with spinose tubercles, up to 56 » in diameter.

Ohana ait as following were recorded by the observers, and for the localities
stated.—

Knuth (Schleswig-Holstein).—A. Coleoptera. Nitidulidae: 1. Meligethes,
freq., po-dvg. B. Diptera. (a) Muscidae: all po-dvg.: 2. Calliphora erythro-
cephala Mg.; 3. Scatophaga merdaria &. (4) Syrphidae: all po-dvg.: 4. Melithreptus
taeniatus Mg.; 5. Syrphus balteatus Deg. C. Hymenoptera, Afidae: 6. Andrena
sp., po-cltg.; 7. Bombus agrorum /., skg.; 8. Halictus morio /,, po-cltg. D. Lepi-
doptera. Rhopalocera: 9. Pieris rapae Z., skg. Schiner (Austria), the Muscid
Trypeta falcata Scop. Scott-Elliot (Dumfriesshire), a Muscid (‘ Flora of Dumfriesshire,’

p- 102).

1619. T. orientalis L., and 1620. T. floccosus Waldst. et Kit.—In these
species, according to Kerner, the heads open between 6 and 7 a.m. at Innsbruck,
closing again between ro and 11 a.m.

VistTors.—The following were recorded by the observers, and for the localities
stated.—

Loew (Switzerland), on T. orientalis the Muscid Spilogaster angelicae Scop.
{‘Beitrage,’ p. 60). Schletterer (Tyrol), the bee Halictus calceatus Scop. Loew
(Berlin Botanic Garden), on T. floccosus—A. Diptera. (a) Muscidae: 1. Antho-
myia sp. (4) Syrphidae: 2. Eristalis nemorum Z.; 3. E. tenax Z.; 4. Syritta
pipiens Z.; 5. Syrphus balteatus Dey. B. Hymenoptera. (a) Apzdae: 6. Halictus
cylindricus /. 9, po-cltg.; 7. Osmia fulviventris Pz. 9, do. (4) Vespidae: 8. Odynerus
parietum Z.

680 ANGIOSPERMAE—DICOTYLEDONES

1621. T. major Jacq.—
Visttors.—Schiner saw the Muscid Trypeta falcata Scop. in Austria.

493- Urospermum Scop.

1622. U. Dalechampii F. W. Schmidt (=Tragopogon Dalechampii Z.).—
Visrtors.—Schletterer observed. the bee Halictus calceatus Scop. at Pola.

494. Scorzonera L.

Florets yellow in colour, more rarely lilac or rose-red. Structure of the style as
in Tragopogon, and Kerner describes geitonogamy similar to what is found in that
genus, the stylar branches of the outer florets diverging and rolling back so as
to come into contact with the pollen of the inner ones.

1623. S. humilis L. (Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 97.)—
Should insect-visits fail, automatic self-pollination is possible by ee back of
the stylar branches.

Vistrors.—I observed the following visitors on garden plants in the Kiel
Botanic Gardens (‘ Bloemenbiol. Bijdragen’).

A.Coleoptera. Nitidulidae: 1. Meligethes, freq. B. Diptera. (a) Muscidae:
2. Lucilia cornica /., po-dvg. (4) Syrphzdae: 3. Eristalis tenax Z., po-dvg.; 4. Syrphus
balteatus Deg., do. C. Hymenoptera. Apzdae: 5. Halictus cylindricus F., skg.;

6. Panurgus sp., po-cltg. D. Lepidoptera. Rhopalocera: 7. Pieris brassicae L.,
freq., skg.; 8. P. rapae Z., do.

1624. S. hispanica L.—

Visttors.—Loew (Berlin Botanic Garden) observed a beetle (Anthaxia quadri-
punctata Z.) and a bee (Osmia fulviventris Pz. 9, po-cltg.), on the var. glastifolia
Willd.

1625. S. parviflora Jacq.—

VisiTors.—Loew (Berlin Botanic Garden) observed a hover-fly (Eristalis nemorum
£.) and 2 bees (Halictus cylindricus /. 9, po-cltg., and Osmia fulviventris Pz. 9, do.).

495. Hypochoeris L.

Florets yellow, and otherwise resembling those of Scorzonera, except that the
stylar branches do not roll back so much. Kerner states that automatic self-

pollination is brought about by gradual elongation of the ligules, the pollen adhering
to which is thus brought into contact with the stigmas.

1626. H. radicata L. (Sprengel, ‘Entd. Geh.,’ pp. 369-70; Kirchner, ‘ Flora
v. Stuttgart,’ p. 739; Herm. Miiller, ‘Fertilisation,’ p. 356, ‘Weit. Beob.,’ III,
p- 97, ‘Alpenblumen,’ p. 469; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’ pp. 97,
161-2, ‘ Weit. Beob. ii. Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 237; MacLeod, Bot.
Jaarb. Dodonaea, Ghent, iii, 1891, v, 1893; Verhoeff, ‘BI. u. Insekt. a. d. Ins.
Norderney’; Loew, ‘ Bliitenbiol. Floristik,’ p. 394.}—In this species the heads spread
out in the forenoon in sunny weather to a diameter of 20-30 mm. Kirchner says
that they contain from 50 to over 100 florets, with ligules 9-12 mm. in length. The

COMPOSITAE 681

corolla-tube is 5~8 mm. long, and the anther-cylinder projects 4-5 mm. from it,
while the style extends 5-6 mm. further. The stylar branches are 1 mm. long
and do not curve far enough back to render automatic self-pollination possible.
Bees belonging to species of Panurgus are particularly common visitors.

Vistrors.—Knuth observed the following in the North Frisian Islands. —

A. Diptera. (a) Empidae: 1. Empis livida Z., skg. (6) Muscidae: 2. Antho-
myia sp. 9; 3. Coenosia sp.; 4. Trypeta sp.; 5. One small Muscid. (c) Syrphidae :
all skg. and po-dvg.: 6. Chrysotoxum festivum Z.; 7. Eristalis arbustorum Z.;
8. E. tenax Z.; 9. Helophilus pendulus Z. B. Hymenoptera. Aprdae: all po-cltg.
and skg.: 10. Apis mellifica Z.; 11. Colletes daviesanus K.; 12. Dasypoda plumipes
Lir.; 13. Panurgus ater Zir.; 14. P.lobatus/’. C.Lepidoptera. Rhopalocera: all
skg.: 15. Coenonympha pamphilus Z.; 16. Epinephele janira Z.; 17. Polyommatus
phlaeas L.

Alfken observed the following bees at Bremen.—

1. Andrena albicrus K.9; 2. A. humilis Jmh. 9; 3. A. tarsata MWVyl. 9;
4. Bombus lapidarius Z. ¥; 5. Dasypoda plumipes Pz. 9 and 8; 6. Dufourea
vulgaris Schenck 9 and é; 7. Eriades florisomnis Z. 6; 8. E. truncorum ZL. 9;
9. Halictus calceatus Scop. 9; 10. H. leucozonius Schr.9; 11. H. morio &.
g and 6; 12. H. nitidiusculus K. 9; 13. H. punctulatus KX. 9; 14. H. tomentosus
Schenck 9; 15. H. tumulorum Z. 9; 16. H. zonulus Sm. 9; 17. Megachile cen-
tuncularis K. $ and 9; 18. Nomada brevicornis Mocs. 9; 19. Nomada fuscicornis
Nyl. 9; 20. Osmia solskyi Mor.9; 21. Panurgus banksianus K. 9 and 4; 22. P.
calcaratus Scop. 9 and é; 23. Prosopis communis Wy/.$; 24. P. confusa Vy. 9.

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list for West-
phalia and Nassau.—

A. Diptera. (a) Conopidae: 1. Sicus ferrugineus Z.,skg.(H.M.). (6) Muscidae :
2. Demoticus plebeius Fa//., skg. (H. M.); 3. Ocyptera brassicariae /, do. (Budd.).
(¢) Syrphidae: 4. Eristalis arbustorum Z. (H. M.); 5. E. nemorum Z. (H. M.);
6. E. sepulcralis Z., po-dvg. (H. M.); 7. Pipiza funebris Zg., po-dvg. (Budd.).
B. Hymenoptera. (a) Apzdae: 8. Andrena denticulata X. 9 and 4, po-cltg. and skg.
(H. M., Thuringia, Borgstette); 9. A. fulvago Chr. 9, po-cltg. (H. M., Thuringia) ;
to. A. fulvescens Sm. 9, skg. and po-cltg. (H.M., Thuringia and the Bavarian
Oberpfalz); 11. A. xanthura X. 9, skg. (H. M., Budd.); 12. Apis mellifica Z. 9,
po-cltg. (H. M.); 13. Bombus lapidarius Z. ¥, skg. (H. M.); 14. Colletes daviesanus
K. 9 and 6, numerous, po-cltg. and skg. (H. M.); 15. Dasypoda hirtipes /. 9, freq.,
po-cltg. (H. M., Budd.); 16. Diphysis serratulae Pz. 6, skg. (H. M.); 17. Halictus
brevicornis Schenck 3, do. (H. M.); 18. H. cylindricus #. 9 and 4, po-cltg. and skg.
(H. M.); 19. H. flavipes #. 6 (H. M.); 20. H. leucozonius Schr. 9 and 4, po-cltg.
and skg. (H. M.); 21. H. lugubris K.é (H. M.); 22. H. malachurus X. 9, po-cltg.
(H. M.); 23. H. quadricinctus /. 9, skg. and po-cltg. (Budd.); 24. H. rubicundus
Chr. 9, po-cltg. (H. M.); 25. H. sexcinctus #. 9, skg. and po-cltg. (Budd.);
26. H. sexstrigatus Schenck 9, po-cltg. (H. M.); 27. H. villosulus . 9, do. (H. M.);
28. Panurgus banksianus &. 9 and 4, skg. and po-cltg. (H. M.); 29. P. calcaratus
Scop. 9 and 4, freq., skg. and po-cltg. (H. M., Budd.); 30. Dufourea vulgaris
Schenck 9, skg. and po-cltg. (H. M.); 31. Sphecodes gibbus Z. 9 and 6 (H. M.).
(6) Sphegidae: 32. Lindenius albilabris #, skg. (Budd.). C. Lepidoptera.
Rhopalocera: 33. Rhodocera rhamni Z., skg. (H. M., Fichtelgebirge).

Loew observed the following —

In Silesia (‘ Beitrage,’ p. 31).—A. Coleoptera. (a) Chrysomelidae: 1. Crypto-
cephalus hypochoeridis Z. 9 and &. (6) Oedemeridae: 2. Oedemera flavipes /. $;
3. O. virescens Z. B. Diptera. Syrphidae: 4. Cheilosia sp., po-dvg.; 5. Meli-

682 ANGIOSPERMAE—DICOTYLEDONES

threptus scriptus Z.; 6. Syrphus balteatus Deg. CC. Hymenoptera. Apidae:
7. Andrena nana X. 9, po-cltg.; 8. Diphysis serratulae Pz. 9, do.; 9. Halictus
cylindricus /. 9, do.; 10. H. leucozonius Schr. 9, do. D. Lepidoptera. Rhopa-
locera: 11. Vanessa urticae Z., skg. Also in Silesia (S.), especially at Glatz (G.)
(op. cit, p. 50).—A. Diptera. (a) Buibionidae: 1. Bibio pomonae &. (S.).
(4) Syrphidae: 2. Cheilosia variabilis Pz. (S.). B. Hymenoptera. Apidae:
3. Andrena convexiuscula X. 9, po-cltg. (S.); 4. A. fulvescens Sm. 9, do. (S.);
5. Dufourea vulgaris Schenck 9 and 6 skg., 9 also po-cltg. (G.); 6. Halictus flavipes
£. 9, po-cltg. (S.); 7. H. malachurus &. 9 (S.); 8. H. punctulatus X. 9 (S.);
g. H. smeathmanellus X. 9 (S.); 10. H. xanthopus X. 9, po-cltg. (S.); 11. Panurgus
banksianus KX. 9 and 6, skg., po-cltg. (S.); 12. Prosopis sp. (S.).

The following were recorded by the observers, and for the localities stated.—

A. Diptera. (ac) Muscidae: 1. Onesia floralis RD. (6) Syrphidae: 2. Chryso-
gaster sp., occasional ; 3. Platycheirus albimanus /.9; 4. Syrphus corollae /. 9 and é.
B. Hymenoptera. Apzdac: 5. Bombus lapidarius Z. ¥ and 4, freq., skg. Alfken
(Juist), the humble-bee Bombus muscorum F. ¥. Krieger (Leipzig), the bees Halictus
rubicundus Cz, and Panurgus banksianus A. Herm. Miiller (Alps), 4 bees, 2 hover-
flies, and a beetle. MacLeod (Flanders), 9 short-tongued bees (among them 2 species
of Panurgus), 8 hover-flies, 4 Muscids, and 3 Lepidoptera (Bot. Jaarb. Dodonaea,
Ghent, v, 1893, pp. 430-1): (Pyrenees), 7 bees (among them 2 species of Panurgus),
5 beetles, 5 Lepidoptera, and 11 flies (op. cit., iii, 1891, pp. 364-5). Scott-Elliot
(Dumfriesshire), 2 Muscids (‘ Flora of Dumfriesshire,’ p. 103).

1627. H. glabra L. (Herm. Miiller, ‘Fertilisation,’ p. 356.)—
Visitors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Westphalia), 5 bees—1. Andrena fulvescens Sm. 9, po-cltg.;
2. Dufourea vulgaris Schenck 9, po-cltg. and skg.; 3. Halictus cylindricus /. 9, po-cltg. ;
4. H. nitidiusculus X. 9 (Borgstette, Tecklenburg); 5. Sphecodes gibbus Z. 9 and
8, skg. and covering itself with pollen. H. de Vries (Netherlands), the humble-bee
Bombus subterraneus Z. ¥ (Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel, 1875).

496. Achyrophorus Adans.
As Hypochoeris.

1628. A. maculatus Scop. (=Hypochoeris maculata Z.),—Linnaeus says
that the heads of this species open at 6 a.m. at Upsala, and close between
4 and 5 p.m.; according to Kerner, they open between 7 and 8 a.m. at
Innsbruck, closing again about 6 or 7 p.m.

Visitors. — Loew (Berlin Botanic Garden) observed the bee Osmia fulviventris
Pz. 9, po-cltg.

1629. A. uniflorus F. W. Schmidt (=Hypochoeris helvetica Wulf, and
Hypochoeris uniflora Vil/.).—In this species, according to Hermann Miller
(‘Alpenblumen,’ p. 468), the stylar branches bend gradually so far outwards
and backwards that the stigmatic papillae on their inner surfaces come into
contact with any pollen that may have remained among the sweeping-hairs, so
that in the absence of insect visitors automatic self-pollination may be effected.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

COMPOSITAE 683

Herm. Miiller (Alps), 3 beetles, 2 flies, 4 bees, and 14 Lepidoptera. - Loew
(Alps), the hawk-moth Zygaena exulens Hochw. ef Rein: (Altvatergebirge), 11
hover-flies, skg.—1. Cheilosia canicularis Pz.; 2. C. sp.; 3. Didea intermedia
Loew; 4. Platycheirus manicatus Mg.; 5. Sericomyia borealis Fal/.; 6. Syrphus
annulipes Ze/f.; 4. S. cinctellus Zef.; 8. S. corollae /; 9. S. lunulatus MW/g.;
ro. S. pyrastri Z.; 11. S. topiarius WZ.

497. Taraxacum L.

Florets yellow. Style covered externally with sweeping-hairs that extend
considerably below the cleft, and are directed obliquely upwards; stylar branches
beset with stigmatic papillae internally, and rolling up to a marked degree.
Kerner says that in the outer florets these branches diverge and roll back to such
an extent that they come into contact with the pollen of the inner ones.

Amang the species native to Denmark, Raunkjer (Bot. Tids., Kjébenhavn, xxv,
1903, pp. 109-40) found some plants in which the anthers were always devoid
of pollen, and which were therefore purely female (T. paludosum Scop., T. Osten-
feldii Raunkj., T. speciosum Raunkj., and T. decipiens Raunkj.). He cut off with
a razor the upper half of an unopened capitulum of such a plant, leaving only
the lower part of the corolla-tubes, filaments, and styles, together with the ovaries.
In spite of this severe treatment the last developed into perfect fruits. The same
experiment was tried with the same result in hermaphrodite species (T. vulgare Zam.,
T. intermedium Rauniy., and T. Gelertii Raunkj.), including the S. European form
T. obovatum DC., and the Pamir species T. glaucanthum DC. As in these cases
no germinating pollen-grains could be found on the stigmas, Raunkjzr regarded the
species named as parthenogenetic.

1630. T. officinale Wigg. (=Leontodon Taraxacum Z.). (Hildebrand, ‘U. d.
Geschlechtsverhalt. b. d. Compositen,’ pp. 7-13, Taf. I, Figs. 1-7; Herm. Miiller,
‘Fertilisation,’ pp. 359-61, ‘Weit. Beob.,’ III, pp. 94-5, ‘Alpenblumen,’ p. 464;
Loew, ‘ Bliitenbiol. Floristik,’ pp. 390, 394, 398; Lindman, ‘ Bidrag till Kanned. om
Skandin. Fjellvaxt. Blomn. o. Befrukt.’; Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, II,
pp. 214-18, 319; H. de Vries, Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel,
1875; MacLeod, Bot. Jaarb. Dodonaea, Ghent, iii, 1891, v, 1893; Knuth, ‘BI.
u. Insekt. a. d. nordfr. Ins.,’ pp. 97-8, 162, ‘Bl. u. Insekt. a. Helgoland’; Warnstorf,
Verh. bot. Ver., Berlin, xxxviii, 1896, xxxix, 1897, xl, 1897; Benecke, Ber. D. bot.
Ges., Berlin, ii, 1884, p. 192.)—In this species the yellow capitula are 30-50 mm.
broad when expanded in the sunshine: on the Dovrefjeld Lindman observed
abnormally large inflorescences of a bright yellow-red colour, with a diameter of as
much as 60 mm., and possessing greatly enlarged marginal florets. At night and
during dull weather the heads remain closed. Linnaeus says that they open at
5-6 a.m. at Upsala, closing again as early as 8-10 am. At Innsbruck, according to
Kerner, they open at 6-7 a.m., and close at 2-3 p.m. Benecke states that when
the heads open, the outer involucral bracts first fold back as a result of the more
vigorous growth of their inner surfaces. The inner involucral bracts are only
passively turned outwards by the opening of the florets, and this holds for the
first expansion of the head, as well as for every time it subsequently opens during the
morning ; in the evening these bracts close in again owing to their elasticity.

684. ANGIOSPERMAE—DICOTYLEDONES

Hermann Miller states that a head consists of 100-200 florets with corolla-tubes
3-7 mm. and ligules 7-15 mm. in length. The anther-cylinder projects 24-5 mm.
from the corolla-tube, and the style extends 3-5 mm. further. In the second stage of
anthesis, the stylar branches, which are 14-2 mm. long, bend outwards and roll
back so far as to form a spiral of 13 turns. If, therefore, the pollen has not all
been removed by insect visitors, automatic self-pollination necessarily takes place.
According to Kerner, automatic geitonogamy also takes place, the stylar branches
of the outer florets diverging and rolling back so as to come into contact with
the pollen of the inner ones. Hansgirg also observed pseudo-cleistogamous florets.
Warnstorf describes the pollen-grains as golden-yellow in colour, polyhedral, spinose,
about 37 » in diameter.

Visitors.—Alfken observed the following at Bremen.—

A. Diptera. (a) Bibionidae: 1. Bibio marci Z., not infrequent. (4) Empzdae:
2. Empis ciliata Z., freq.; 3. E. opaca F., excessively freq., skg. (c) Syrphidae:
4. Eristalis arbustorum ., freq., skg.; 5. E. intricarius Z., do.; 6. E. pertinax Scop.,
do.; 7. E. sepulcralis Z., not infrequent, skg.; 8. E. tenax Z., very common, skg. ;
g. Helophilus pendulus Z., do.; 10. Leucozona lucorum Z., rare, skg.; 11. Sericomyia
borealis Fall, freq., skg.; 12. Spilomyia vespiformis Z., very rare; 13. Syrphus
tricinctus /all., skg.; 14. S. venustus JZ., very common, skg. B. Hymenoptera.
(a) Apidae: 18. Andrena albicans Mull., very common, 9 skg. and po-cltg., 6 skg.
This is the commonest pollinator: with obvious satisfaction it lies on its side and
probes the inflorescence. 16. A. albicrus X. 9 and 4, not infrequent ; 17. A. apicata
Sm. 9 and 4, rare, it chiefly visits willows; 18. A. argentata Sm. 9, rare; 19. A.
carbonaria Z., 9 and 8, do.; 20. A. chrysosceles K. 9 and 4, do.; 21. A. cineraria
L. 9, freq., skg. and po-cltg., 6 very common, skg.; 22. A. cingulata /. 9 and 6, very
rare; 23. A, convexiuscula X. 9 and 4, rare; 24. A. dorsata K.d, once; 25. A. extri-
cata Sm. 9 not infrequent, skg. and po-cltg., very common, skg.; 26. A. flavipes Pz.
9 extremely freq., skg. and po-cltg., 6 do., skg.; 27. A. fucata Sm. 9, very rare;
28. A. fulvaga Chr.5; 29. A. gwynana X. 9 not infrequent, skg. and po-cltg., 4
not infrequent, skg.; 30. A. helvola Z.9 and 4, very rare; 31. A. humilis JmA. 9 and 4,
mostly stylopized individuals; 32. A. labialis K. 9 and 4, rare; 33. A. labiata Schenck
9, very rare, skg. and po-cltg.; 34. A. morawitzi Zhs. 9 and 6, not infrequent; 35. A.
nigroaenea X. 9 very common, skg. and po-cltg., 8 very common, skg.; 36. A. nitida
Fourcr. 9 very common, skg. and po-cltg., § very common, skg.; 37. A. ovina X7g.
@ infrequent, skg. and po-cltg., & rare, skg.; 38. A. parvula K. 9 very common, skg.
and po-cltg., 4 not frequent, skg.; 39. A. praecox Scop. 9, not freq., skg. and po-cltg.,
chiefly visits willows, 6 rare; 40. A. propinqua Schenck; 41. A. proxima X. 9 and 4,
rare; 42. A. rufitarsis Zeé#. 9 freq., skg. and po-cltg., $ very rare; 43. A. thoracica F.
9 rare, skg. and po-cltg., rare, skg.; 44. A. tibialis K., rare, 9 skg. and po-cltg., 6
skg.; 45. A. trimmerana K., do., 9 do., $ do.; 46. A. varians K. 9 and 4, rare;
47. Bombus agrorum /. 9, not infreq. skg. and po-cltg.; 48. B. arenicola Zhs. 9,
rare; 49. B. derhamellus &. 9, freq., skg. and po-cltg.; 50. B. distinguendus Zor. 9,
not infreq., skg. and po-cltg.; 51. B. hortorum Z. 9 rare, é rare, skg. in plants
flowering in autumn; 52. B. jonellus X. 9, rare, skg. and po-cltg., a characteristic
visitor of willows; 53. B. lapidarius Z. 9, very common, skg. and po-cltg.; 54. B.
lucorum Z. 9, do.; 55. B. muscorum /. 9 and 9, not infreq., skg. and po-cltg.;
56. B. pratorum Z. 9, skg.; 57. B. ruderatus /. 9, rare, skg. and po-cltg.; 58. B.
sylvarum JZ. 9, not infreq., skg. and po-cltg.; 59. B. terrester Z., very common, skg.
and po-cltg.; 60. Colletes cunicularius Z, 9, rare, po-cltg.; 61. Eriades florisomnis Z.
g and $; 62. Eucera difficilis (Duf) Pér.é; 63. Halictus calceatus Scop., and var.
elegans Zep. 9, rare, skg. and po-cltg.; 64. H. flavipes &. 9, freq., skg. and po-cltg. ;
65. H. levis A. 9, not infrequent, skg. and po-cltg.; 66. H. malachurus X., very rare ;

COMPOSITAE 685

67. H. morio ¥. 9, not infrequent, skg. and po-cltg.; 68. H. nitidiusculus X. 9, very
common, do. ; 69. H. punctatissimus Schenck 9, freq., do. ; 70. H. punctulatus X. 9, very
common, do.; 71. H. quadrinotatus K. 9, very rare; 72. H. quadrinotatulus Schenck,
once; 73. H. rubicundus Chr. 9, very common, skg. and po-cltg., the spring females
confine their visits almost exclusively to this species; 74. H. sexnotatulus JVy/. 9, rare,
skg..and po-cltg. ; 75. H. zonulus Sm. 9, do.; 76. Melecta armata Pz. 9, rare, skg. ;
77. Nomada alternata XK. 9 and 4, freq., skg.; 78. N. bifida Z%s. 9 and 4, very freq.,
skg.; 79. N. borealis Ze/t, 9 and 4, not infrequent, skg.; 80. N. fabriciana Z. 9 and 6,
very rare; 81. N. flavoguttata K. 9 and 4, not infrequent, skg.; 82. N fucata Pz.
g and 4, not infrequent, skg. ; 83. N. lathburiana X. 9 and 4, rare, skg.; 84. N. lineola
Pz. 9 and 4, rare, skg.; 85. N. ruficornis Z. 9 and 4, do.; 86. N. succincta Z. 9 and 4,
freq., skg.; 87. N. xanthosticta K. 9 and 4, rare, skg.; 88. Osmia caerulescens Z.
9 and 6, rare; 89. O. rufa Z. 9 and 4, rare; go. O. solskyi Mor., not infrequent, skg.;
91. Podalirius acervorum Z. 9, once; 92. Psithyrus barbutellus K. 9, freq., skg.;
93. P. campestris Pz. 9, do.; 94. P. quadricolor Zep. 9, once; 95. P. rupestris F. 9,
not infreq., skg.; 96. P. vestalis Fourcr. 9, very common, skg. (5) Zenthredinidae:
97. Dolerus pratensis Fal/.; 98. Emphytus cinctus Z.; 99. Pachyprotasis rapae L.
(c) Vespidae: 100. Odynerus callosus Zhs.

Friese observed the following bees in Alsace (A.), Baden (B.), Hungary (H.),
and Mecklenburg (M.); also at Fiume (F.) and Trieste (T.).—

1. Andrena albicans J/zl/, (M.), not infreq.; 2. A. carbonaria Z. (M.), do. ;
3. A. cineraria Z., 1st gen., not infreq. (M.), 2nd gen., 1 9 (H.); 4. A. extricata Sm,
ve not infreq.; 5. A. flavipes Pz. (M.), freq. (B., A.); 6. A. fucata Pz., not infreq.
A.); 7. A. gwynana K., freq. (M.); 8. A. humilis Zm., freq. (T., H.); 9. A.
taraxaci Gir., freq. (T., H.); 10. A. tibialis H., freq. (M.); 11. A. thoracica &, 1 9,
(A.); 12. Halictus rubicundus Chr., freq. (M.); 13. H. xanthopus X., occasional
(M.); 14. Nomada lathburiana &., occasional (M.); 15. N. succincta Pz., freq.
(M.); 16. N. trispinosa Schmiedekn., freq. (H.); 17. N. zonata Pz. (F., H.);
18. Osmia rufohirta Zep. 9, freq., po-cltg. (H.): also in Thuringia—r1. Andrena
cineraria Z.; 2. Halictus rufocinctus (Szck.) My; 3. H. xanthopus X.

Loew observed the following.—

In Brandenburg (‘Beitrige,’ p. 40).—A. Coleoptera. (a) Buprestidae: 1. An-
thaxia nitidula Z. (6) Mitzdulidae: 2. Meligethes sp. B. Diptera. (2) Stratiomyzdae :
3. Cheilosia praecox Ze//.; 4. Helophilus pendulus Z. C. Hymenoptera. Apidae:
5. Andrena albicans Aull. 9, po-cltg.; 6. A. albicrus K. 9, do.; 7. A. cineraria Z.
9, do.; 8, A. combinata Chr. 9, do.; 9. A. nigroaenea KX. 9, do.; 10. A. pilipes
Ff. 9; 11. A. ventralis Jmh. 9, po-cltg.; 12. Chelostoma maxillosum JZ. @, skg.;
13. Dasypoda hirtipes & 9, po-cltg.; 14. Halictus cylindricus & 9, skg.; 15. H.
levis K. 9, po-cltg.; 16. H. maculatus Sm. 9, po-cltg.; 17. H. minutus K. 9, do.;
18. H. punctulatus K. 9,do.; 19. H. quadristrigatus Z/r. 9 and 6, skg., 9 also po-cltg. ;
20. H. sexcinctus F. 3, skg.; 21. Nomada fucata Pz. 4, do.; 22. Sphecodes gibbus
L., do.; 23. Trypetes truncorum Z. 9, po-cltg. In Hesse (op. cit., p. 51).—Andrena
chrysosceles A. 9, po-cltg. In Switzerland (op. cit., p. 59).—A. Coleoptera.
Chrysomelidae: 1. Cryptocephalus hypochoeridis Z. B. Diptera. Syrphidae:
2. Cheilosia canicularis Pz.; 3. C. plumulifera Loew (?); 4. Eristalis nemorum JZ. ;
5. Merodon cinereus #.; 6. Syrphus lineola Ze/.; 7. S. vittiger Zest. (?); 8. Xylota
triangularis Ze, C. Hymenoptera. Zenthredinidae: 9. Tarpa spissicornis A/g.

Burkill observed the following on the Yorkshire coast (‘ Fertlsn. of Spring Fls.’).—
A. Coleoptera. (a) Curculionidae: 1. Apion sp. (6) Mitidulidae: 2. Meligethes
picipes S/urm, skg. and po-dvg. B. Diptera. (a) Bidiondae: 3, Dilophus albipennis
Mg., po-dvg. (6) Muscidae: 4. Calliphora erythrocephala AZg.; 5. Helomyza sp.,
skg.; 6. Lucilia cornicina /, do.; 7. Scatophaga stercoraria Z., skg. and po-dvg. ;

686 ANGIOSPERMAE—DICOTYLEDONES

8. Sepsis nigripes A7/g., skg.; 9. Stomoxys sp., po-dvg. (c) Syrphidae: 10. Eristalis
arbustorum Z., skg.; 11. E. pertinax Scop., do. C. Hymenoptera. (a) Apidae:
12. Andrena clarkella K.; 13. A. gwynana X., skg.; 14. Apis mellifica Z., rare, skg. ;
15. Bombus agrorum /., skg.; 16. B. terrester Z., do.; 17. Nomada borealis Zeéz.,
once, skg. (4) Jchneumonidae: 18. Ichneumon sp., skg. D. Lepidoptera. Rhofa-
locera: 19. Pieris rapae Z., do., skg.; 20. Vanessa urticae Z.,do. E. Thysanoptera.
21, Thrips sp.

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list for
Westphalia and Nassau.—

A. Coleoptera. (c) Buprestidae: 1. Anthaxia nitidula Z.(H.M.). (4) Chry-
somelidae: 2. Gastrophysa polygoni Z. zm copula on the flowers (H. M.).
(c) Coccinellidae: 3. Coccinella septempunctata Z., vainly trying to suck (H. M_.).
(d) Elateridae: 4. Corymbites haematodes /, sinking its head deeply into the
florets (H. M.); 5. Limonius cylindricus Pays., do. (H. M.). (e) Mitidulidae :
6. Meligethes, in extremely large numbers in the florets (H. M.). (/) Zelephoridae:
7. Malachius bipustulatus Z., po-dvg. (H. M.); 8. M. elegans Ok. 3, do. (H. M.).
B. Diptera. (2) Empidae: 9. Empis livida Z., freq., skg. (H.M.); 10. E. opaca F.,
skg. (H. M.); 11. E. punctata /., in large numbers, skg. (H. M.). (6) Muscidae :
12. Cyrtoneura hortorum Fall. 9, skg. and po-dvg. (H. M.); 13. Onesia floralis
R.-D., freq. (H. M.); 14. Pollenia vespillo /, po-dvg. (H. M.); 15. Sarcophaga
carnaria Z., do. (H. M.); 16. Scatophaga merdaria /, freq., skg. and po-dvg.
(H. M.); 17. S. stercoraria Z., do. (H. M.). (c) Syrphcdae: 18. Ascia lanceolata
MMg., skg. (H.M.); 19. A. podagrica /., freq., po-dvg. (H.M.); 20. Cheilosia chloris
iMg., po-dvg. (H. M.); 21. C. vernalis Fal/., do. (H.M.); 22. Eristalis arbustorum Z.,
freq., skg. and po-dvg. (H. M.); 23. E. intricarius Z., do. (H. M.); 24. E. nemorum
L.,do.(H.M.); 25. E. pertinax Scop., do. (H.M.); 26. E. sepulcralis Z., do. (H. ray
27. E, tenax Z., do. (H.M.); 28. Melithreptus menthastri Z., po-dvg. (H. M.);
29. M. taeniatus Mg., do. (H. M.); 30. Rhingia rostrata LZ. (H. M.); 31. Syrphus
nitidicollis M/g., po-dvg. (H. M.); 32. S. pyrastri Z., do. (H. M.). C. Hymenoptera.

1(a) Apidae: 33. Andrena albicans JMuil. 9, skg. and po-dvg., 6 (H. M.); 34. A.
albicrus X., do. (H. M.); 35. A. argentata Sm. 3, freq., skg. (H. M.); 36. A. atri-
ceps KX. 9 skg. and po-cltg., skg. (H. M.); 37. A. cineraria Z., do. (H. M.);
38. A. cingulata A. 6, skg. (H. M.); 39. A. connectens X. 9, skg. and po-cltg.
(H. M.); 40. A. convexiuscula X. 9, do. (H. M.); 41. A. dorsata K. 9 do., $ skg.
(H. M.); 42. A. fasciata Wesm., do. (H. M.); 43. A. flavipes Pz., very common,
9 skg. and po-cltg., § skg. (H. M.); 44. A. fulva Schr. 9, skg. and po-cltg. (H. M.);
45. A. fulvescens Sm. 9, do. (H.M.); 46. A. gwynana X., do. (H.M.); 47. A. helvola
L., do. (H. M.); 48. A. mixta Schenck 9, skg. and po-cltg. (a var. of 47) (H. M.);
49. A. nigroaenea X. 9, skg. and po-cltg. (H. M.); 50. A. nitida Fourcr. 9 do., 8 skg.
(H.M.); 51. A. parvula X,, do., freq. (H.M.); 52.A. pratensis WVy/. 9, skg. and po-cltg.
(H. M.); 53. A. smithella K. 9 do. & skg. (H. M.); 54. A. trimmerana XK. 9, do.
(H. M.); 55. A. varians X, 9, do., not infreq. (H. M.); 56. Apis mellifica Z. ¥, very
numerous, skg. and po-cltg. (H. M.); 57. Bombus agrorum ¥. 9, skg. (H. M.);
58. B. confusus Schenck 9, do. (H. M.); 59. B. lapidarius Z. 9, do. (H. M.); 60. B.
syivarum JL, 9, do. (H. M.); 61. B. terrester Z. 9, do. (H. M.); 62. Halictus albipes
F, 9, freq., skg. and po-cltg. (H. M.); 63. H. cylindricus /. 9, do. (H. M.); 64. H.
flavipes /. 9, skg. and po-cltg. (H. M.); 65. H. leucopus X. 9, do. (H. M.);
66. H. leucozonius Schr. 9, freq., do. (Budd.); 67. H. longulus Sm. 9, skg. and
po-cltg. (H. M.); 68. H. lucidulus Schenck 9, do. (H. M.); 69. H. maculatus Sm.,
9, do. (H. M.); 70. H. malachurus &. 9, skg. (H. M.); 71. H. minutissimus X. 9,
skg. and po-cltg. (H. M.); 72. H. morio F. 9, do. (H. M.); 73. H. nitidiusculus A.
9, freq., do. (H. M.); 74. H. nitidus Schenck 9, skg. and po-cltg. (H. M.); 75. H.
rubicundus Chr. 9, do. (H. M.); 76. H. sexnotatus X. 9, freq., do. (H. M.);
77. H. sexsignatus Schenck 9, do. (H. M.); 78. H. smeathmanellus X. 9, occasional,

COMPOSITAE 687

skg. and po-cltg. (Budd.); 79. H. villosolus K. 9, skg. and po-cltg. (H. M.);
80. H. zonulus Sm. 9, do. (H. M.); 81. Megachile centuncularis Z. 5, skg. (Budd.) ;
82. Nomada alternata X. 9, do. (H.M.); 83. N. flavoguttata K. 4, do. (H. M.);
84. N. lathburiana K. 9, do. (H. M.); 85. N. lineola Pz. 9, do. (H. M.);
86. N. ruficornis Z. 9 and 3, very numerous, skg. (H. M.); 87. N. ruficornis Z.,
var. signata Jur. 9 and 4, skg. (H. M.); 88. N. succincta Pz. 9 and 4, do. (H. M.);
89. N. varia Pz. 9 and 4, freq., skg. (H. M.); 90. Osmia aenea Z. 4, skg. (Budd.);
gt. O. aurulenta Pz. 9, do. (H. M., Thiir.); 92. O. fusca Chr. 9, po-cltg. (H. M.);
93- O rufa LZ. 3, skg. i M.); 94. Psithyrus barbutellus X. 9, do. (H. M.); 95. B.
vestalis Fourcr. 9, do. (H.M.); 96. Sphecodes gibbus Z. 9, skg. and po-cltg. (H. M.) ;
97. Stelis aterrima Pz. 8, skg. (Budd.); 98. S. minuta Zep. 8, do. (Budd.). (6) For-
micidae: 99. Formica congerens JVyi. ¥, freq., skg. (H.M.); 1oo. Lasius niger Z. ¥,
freq.(H.M.). (c) Sphegedae : 101. Oxybelus uniglumis Z., probing the florets deeply
(H.M.). (d) Tenthredinidae: 102. Cephus pallidipes Kig. (Budd.); 103. C., a small
sp., numerous (H. M.). (e) Vespzdae: 104. Odynerus parietum Z. 8 (H. M.).
D. Hemiptera. 105. Pyrrhocoris apterus Z., freq., skg.(H.M.). E. Lepidoptera.
Rhopalocera: all skg.: 106. Parage megaera Z.(H. M.); 107. Pieris brassicae LZ.
(H.M.); 108. P. napi Z. (H. M.); 109. Rhodocera rhamni Z.(H.M.); 110. Syrich-
thus alveolus #. (H. M.); 111. Vanessa io Z., freq. (H. M.); 112. V. urticae Z.,
freq. (H.M.). F. Thysanoptera. 113. Thrips, freq. (H. M.).

The following were recorded by the observers, and for the localities stated.—

Knuth (North Frisian Islands), Apis, 3 humble-bees, 8 hover-flies,2 Muscids, 4 Lepi-
doptera, and the beetle Meligethes: (Helgoland)—A. Coleoptera. TZélephoridae:
1. Psilotrix cyanea O/.,very common. B. Diptera. (a) Musczdae: skg. and po-dvg.:
2. Coelopa frigida Fad/., very common ; 3. Lucilia caesar Z., common ; 4. Scatophaga
stercoraria Z., freq. (6) Syrphidae: 5. Helophilus pendulus Z., one 9, skg. and
po-dvg. C. Lepidoptera. Rhopalocera: 6. Pieris brassicae Z., occasional, skg.
D. Hymenoptera. Apzdae: 7. Andrena labialis K. 9, skg.; 8. Eucera difficilis
(Duf) Per. 3, freq., skg. Verhoeff (Norderney).—A. Diptera. (a) Brbzonidae:
1. Dilophus vulgaris Mg. one 9 and one 6. (6) Lmprdae: 2. Hilara quadrivittata Me.
one 6. (c) Muscedae: 3. Calliphora erythrocephala JZg., skg.; 4. Cynomyia mortuorum
Z. two $; 5. Cyrtoneura hortorum /a//. one $; 6. Hylemyia variata Fall. one 3;
4. Lucilia caesar Z. 9 and é; 8. Limnophora protuberans Zef/. 9 and é; 9. Micro-
peza sp.; 10. Myospila meditabunda #. (d) Syrphidae: 11. Helophilus pendulus Z.;
12. H. trivittatus #. one 9; 13. Rhingia rostrata Z. B. Hymenoptera. Apidae:
14. Bombus lapidarius Z. 9, skg.; 15. B. terrester Z. 9 and ¥, do.; 16. Colletes
cunicularius Z. 9, skg. and po-cltg.; 17. Psithyrus vestalis Kourcr.9. C. Lepidoptera.
Pieridae: 18. Pieris brassicae Z., skg.; 19. P. napiZ.,do. Alfken (Juist), 2 humble-
bees (Bombus hortorum Z. é, and B. ruderatus /.), Wiistnei (Alsen), 5 bees—r. Cilissa
tricincta K.; 2. Andrena tibialis K.; 3. A. chrysosceles X.; 4. Halictus cylindricus
F.8; 5. Nomada fabriciana Z.: (Flensburg), the bee Andrena labialis K. Krieger
(Leipzig), 5 bees—1. Andrena albicans JMil/.; 2. A. flavipes Pz.; 3. A. gwynana
K.; 4. A. nigroaenea K.; 5. Halictus calceatus Scop. Schmiedeknecht (Thuringia),
8 bees—r. Andrena extricata Sm.; 2. A. flavipes Pz.; 3. A. humilis Jmh.; 4. A.
nitida Fourcr.; 5. A. parvula X.; 6. A. tibialis K.; 7. Bombus hypnorum JZ. 9;
8. Psithyrus quadricolor Zep. 9. Schletterer records 8 bees for the Tyrol (T.) and
Pola (P.)—1. Andrena congruens Schmiedekn. (T.); 2. A. convexiuscula XK. (P.);
3. Bombus hortorum Z. (T.); 4. Dasypoda hirtipes Pz. (T.); 5. Halictus calceatus
Scop.; 6. H. malachurus XK. (P.); 7. H. morio ¥. (P.); 8. H. vulpinus yd. (T.).
Herm. Miiller (Alps), 9 beetles, 26 flies, 30 Hymenoptera, and 35 Lepidoptera.
Redtenbacher (Austria), 2 beetles (Anthaxia nitidula Z., and Malachius gracilis 17W.).
Schenck (Nassau), 13 bees—1. Andrena chrysosceles K.; 2. A. extricata Sm;
3. A. flavipes Pz.; 4. A. fulvago Chr.; 5. A. nitida Fourcr.; 6. A. proxima KX. 3;
4, A. tibialis A.; 8. A. trimmerana X.; 9. Halictus flavipes /.; 10. H. levigatus K. ;

688 ANGIOSPERMAE—DICOTYLEDONES

11. H. rubicundus Chr. 9; 12. H. rufocinctus (Sich.) Nyl.; 13. H. tetrazonius A7g. 9.
Schiner (Austria), the Muscid Tephritis conjuncta Zoew. H. de Vries (Netherlands),
12 bees (Ned. Kruid. Arch., Nijmegen, 2. Ser., 2. Deel, 1875)—1. Andrena albicans
Mill.9; 2. A. albicrus K.9; 3. A. fasciata Wesm. 3; 4. A. nigroaenea X. 9;
5. Apis mellifica Z. ¥; 6. Chelostoma florisomne Z. 9; 7. Halictus cylindricus
F. 9; 8. H. leucozonius Schr. 9; 9. H. rubicundus Chr. 9; 10. H. seladonius /. 9;
11. H. villosulus X. 9; 412. Psithyrus vestalis Fourcr. 9. MacLeod (Pyrenees),
9 Hymenoptera, 4 Lepidoptera, a beetle, and 10 flies (Bot. Jaarb. Dodonaea, Ghent,
iii, 1891, pp. 366-7): (Flanders), Apis, 4 humble-bees, 19 short-tongued bees,
a Siricid, 3 hover-flies, 2 Empids, 8 Muscids, 2 Lepidoptera, and 2 beetles (op. cit.,
v, 1893, p. 432-3). Schneider (Arctic Norway) (Tromse Mus. Aarsh., 1894);
8 bees—r. Bombus alpinus Z. 3 2. B. hypnorum L. ¥; 3. B. lapponicus LZ.
Yandé; 4. B. pratorum LZ. ¥ and 6; 5. B. scrimshiranus X. ¥ and é; 6. B. terrester
£.% and 8; 7. Psithyrus quadricolor Zep. 3; 8. P. vestalis Fourcr. 8. Warming
(Greenland), the butterfly Colias boothi #.-Sch. (=C. hecla Lef). Lindman
(Dovrefjeld), numerous flies, several Lepidoptera, and a humble-bee. Scott-Elliot
(Dumfriesshire), 4 short-tongued bees, an Empid, and 3 Muscids (‘Flora of Dum-
friesshire,’ p. 104). Smith (England), 4 bees—1. Andrena albicans A/ill.; 2. A.
angustior K.; 3. A. nitida Mourcr.; 4. A. nigroaenea AK. Saunders (England),
3 bees—x. Andrena extricata Sm.; 2. A. humilis Zmh.; 3. A. labialis K.

1631. T. croceum Dahlstedt.—This is a sub-species of T. officinale defined by
the author, and includes all the plants collected by Andersson and Hesselman in
Spitzbergen (‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora’). It
also occurs in Iceland, and alpine Norway and Sweden. The heads attain a diameter
of 26 mm. and close in bright sunshine between 5 and 6 pm. The florets are
orange-yellow in colour. With one exception the specimens collected in Spitzbergen
were markedly female, so that the species is apparently apogamous.

1632. T. salinum Poll.—

Vistrors.—Loew observed 2 hover-flies (Eristalis nemorum Z., and Syrphus
ribesii Z.) and a bee (Halictus zonulus Sm. 8, skg.) in the Berlin Botanic Garden.

1633. T. phymatocarpum J. Vahl.—Ekstam states that in this species the
diameter of the head is 35 mm. in Nova Zemlia. The florets are white and bright
violet in colour, and faintly odorous. Autogamy or geitonogamy may be effected by
spiral movements of the stylar branches. According to Ekstam (‘ Bliitenbiol. Beob.
a. Spitzbergen,’ pp. 6-7), the capitula are white or bright violet in Spitzbergen, with
a scarcely perceptible odour. At the beginning of July they are already partly over,
and set fully mature fruits during the same month. Andersson and Hesselman
(‘Bidrag till Kanned. om Spetsbergens o. Beeren Eil. Karlvaxtflora,’ p. 15) found
reduced anthers devoid of pollen in all the specimens they examined, so that the
species is apparently apogamous. Cf Raunkjaer’s experiments, p. 683.

Vistrors.—A small spider and a medium-sized fly were observed in Nova
Zemilia.

498. Chondrilla L.

Florets yellow. The stylar branches roll back into a semicircle. Kerner says
that geitonogamy is brought about in the same way as in Taraxacum.

1634. C. juncea L.—Kirchner says (‘ Beitrage,’ p. 72) that each head in this
species includes only 7-12 (usually 11) golden-yellow florets, and expands to

COMPOSITAE 689

a diameter of 17 mm. All the florets of a head develop simultaneously. The
anther-cylinder projects 3-4 mm. from the corolla-tube, and the style 3 mm. beyond
this. The stylar branches only roll back far enough to form a semicircle, so that
automatic self-pollination is excluded. On the other hand, according to Kerner, the
stylar branches of the outer florets diverge far enough to touch the pollen of the
inner ones, thus automatically effecting geitonogamy. Warnstorf states (Verh. bot.
Ver., Berlin, xxxviii, 1896) that the florets open at Neu-Ruppin about 10 a.m., closing
again between 2 and 3 p.m. The pollen-grains are golden-yellow in colour, spherical,
densely covered with spinose tubercles, irregular in size, up to 50 » in diameter.

Visrrors.—Schletterer records the bee Dasypoda plumipes Pz. for the Tyrol,
and observed the fossorial wasp Notogonia pompiliformis KoA/ at Pola.

499. Prenanthes Vaill.

Florets purple-red. Style entirely covered with pointed sweeping-hairs directed
obliquely upwards, and extending considerably below the cleft: stylar branches
beset internally with stigmatic papillae, and rolling far back. Kerner says that
geitonogamy is brought about by the intertwining of the stylar branches of neigh-
bouring florets when these begin to fade, so that the stigmas become dusted
with any pollen that may remain.

1635. P. purpurea L.—In this species, according to Hermann Miller (‘ Weit.
Beob.,’ III, pp. 98-6), the heads consist of only 4-6 florets. The involucre is
12-14 mm. long, but only 2 mm. broad; the purple-red florets projecting from
it possess ligules ro mm. long and 3-4 mm, broad, so that the heads are moderately
conspicuous. At a later stage the style projects 7 mm. beyond the anther-cylinder,
which is 5-6 mm. long and scarcely 3 mm. broad. Finally, the stylar branches
(3 mm. in length) diverge and roll back into a spiral of 14 or 2 turns, so that in the
absence of insect visitors automatic self-pollination is effected.

VistTors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Bavarian Oberpfalz), a beetle (Agrilus coeruleus Rossz), a Muscid
(Sarcophaga carnaria Z., po-dvg.), and 2 bees (Apis, freq., skg. and po-cltg.), and
Andrena denticulata K. 9). Hoffer (Steiermark), the humble-bee Bombus hypnorum
L.%. Schletterer and von Dalla Torre (Tyrol), 2 humble-bees (Bombus confusus
Schenck, and B. mastrucatus Gers/. 9). MacLeod (Pyrenees), a Muscid (Bot. Jaarb.
Dodonaea, Ghent, iii, 1891, p. 368). Loew (Berlin Botanic Garden), Apis, skg.

500. Lactuca L.

Florets yellow, rarely lilac in colour. Structure of the style as in Prenanthes.
Kerner says that geitonogamy takes place when the florets begin to fade. He also
states that the milky juice of many of the species is a protection against destructive
animals.

1636. L. Scariola L. (Kirchner, ‘ Beitrage,’ p. 72.)—Kirchner says that the
diameter of the yellow heads, which consist of 20 florets, is about 20 mm. when
fully open. Even before complete expansion, the pollen-covered style grows out

DAVIS. te Y y

690 ANGIOSPERMAE—DICOTYLEDONES

through the anther-cylinder, and soon afterwards the stylar branches diverge, lying about
13 mm. above the anther-cylinder and 5 mm. above the entrance to the corolla-tube.
Towards the end of anthesis these branches roll back into a spiral of 14 turns, so
that automatic self-pollination necessarily takes place if any pollen remains among the
sweeping-hairs. Kerner says that the heads open about 8-9 a.m. at Innsbruck,
closing again about 3-4 p.m.

Visirors.—Kirchner observed a small bee.

1637. L. sativa L. (=L. Scariola Z. according to the Index Kewensts).
(Kirchner, ‘ Beitrage,’ p. 73.)—Kirchner says that the flower mechanism is very
similar to that of L. Scariola. Each head contains 10-16 yellow florets which
develop simultaneously. The ligules are 11 mm. long, and are directed obliquely
outwards, so that the diameter of the expanded head is about 15 mm. The anther-
cylinder projects 4 mm. beyond the corolla-tube, which is 44 mm. in length. About
2 mm. above the anther-cylinder the stylar branches diverge: they ultimately roll
back into a complete circle, so that automatic self-pollination necessarily results from
contact of the stigmatic papillae with the pollen-grains clinging to the sweeping-hairs.

According to Linnaeus, the heads open about 7 a.m. at Upsala, closing again
about 10 a.m. Kerner gives the corresponding times for Innsbruck as 8-9 a.m. and
1-2 p.m.

VisiTors.—Kirchner observed various flies.

1638. L. muralis Less. (=Prenanthes muralis Z.). (Kirchner, ‘ Beitrage,’
p- 73-)—Kirchner says that the heads of this species contain only 5, sometimes
only 4, bright yellow florets, with ligules that spread out horizontally, or sometimes
bend a little downwards, so that the inflorescence attains a diameter of 13-14 mm.
The florets of each head develop simultaneously. Even before complete expansion
the styles project from the anther-cylinders. They grow so far in an oblique
direction that their tips stand about 5 mm. above the level of the head. The stylar
branches at first curve outwards from each other, and subsequently roll downwards,
though not far enough to touch the pollen remaining among the sweeping-hairs.

Warnstorf (Verh. bot. Ver., Berlin, xxxviii, 1896), however, observed the stylar
branches rolling back till they came into contact with the pollen, so that autogamy
necessarily took place. Geitonogamy, on the other hand, is only possible when the
head is closed. The pollen-grains are yellow in colour, polyhedral, with spinose
tubercles on their edges, from 40 yz to 43 p in diameter.

At Neu-Ruppin the heads open between 6 and 7 a.m., closing again between
4 and 5 p.m. Kerner says that they open at Innsbruck between 7 and 8 a.m., and
close again about 2 or 3 p.m.

VisiTors.—The following were recorded by the observers, and for the localities
stated.—

Kirchner (Stuttgart), 2 flies, a small bee, and the beetle Meligethes. Herm.
Miiller (Fichtelgebirge), a Muscid (Echinomyia grossa Z., po-dvg.), and a bee
(Halictus albipes /., skg.) (‘ Weit. Beob.,’ III, p. 96). MacLeod (Pyrenees), a bee
of the genus Panurgus (Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 367).

1639. L. perennis L. (Herm. Miiller, ‘Alpenblumen,’ pp. 463-4.)—In this
species each head contains about 16 florets, which spread out their ligules (16-18 mm.

COMPOSITAE 691

long) in the midday sun, to give a violet star about 40 mm. in diameter. The
stylar branches ultimately roll back, often so far that their stigmatic inner surfaces
touch their outer surfaces, to which some pollen often remains clinging. As in
the case of the last species, the florets of each head develop simultaneously, being
first male and then female, so that insect visitors necessarily effect cross-pollination.

Kerner says that at Innsbruck the heads open between 6 and 7 a.m., closing again
about 5 or 6 p.m.

Visitors.— The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller-(Alps), a fly and a beetle. Loew (Berlin Botanic Garden).—
A. Coleoptera. Telephoridae: 1. Dasytes flavipes F. B. Diptera. Syrphidae:

2. Syrphus luniger /g. C. Hymenoptera. Apidae: 3. Chelostoma nigricorne
Ny. 3, skg.

1640. L. viminea J. et C. Pres]._—

VisiTors.—Loew (Berlin Botanic Garden) observed a beetle (Anthaxia quadri-
punctata Z.) and 3 bees (1. Megachile centuncularis Z. 4, skg.; 2. Prosopis armillata
WNyi. 9, do.; 3. Stelis aterrima Pz. 9, do.).

501. Mulgedium Cass.

Florets blue. Style covered externally with very sharp thorn-like sweeping-
hairs, extending far down its undivided part, but denser on its branches. The inner
surfaces of the latter beset with stigmatic papillae. Kerner states that the stylar
branches diverge so far when the florets fade as to touch the outer surfaces of
adjacent branches, and as they are still covered with pollen geitonogamy is effected.

1641. M. alpinum Less. (=Sonchus alpinus Z., and Lactuca alpina Benth.
et Hook. f.). (Herm. Miiller, ‘ Alpenblumen,’ pp. 459-60.)—In this species about
20 florets are aggregated into a head which in the closed condition is only
4 mm. in diameter, but 20-30 mm. when expanded in the sunshine. The
stylar branches are 2 mm. long; they diverge in the second stage of anthesis,
but never roll back far enough to effect automatic self-pollination.

Visitors.—The following were recorded by the observers, and for the localities
stated. —

Herm. Miiller (Vosges), Apis,a humble-bee, a Lepidopterid, and a beetle. Lindman
(Dovrefjeld), a humble-bee. Loew (Berlin Botanic Garden), 6 bees—1. Bombus
lapidarius Z. 9 and ¥, skg.; 2. Chelostoma nigricorne JVyi. 6, do.; 3. Halictus
sexcinctus F. 9, do.; 4. Osmia fulviventris Pz. 6 and 9, do.; 5. Osmia rufa Z. 9,
po-cltg.; 6. Stelis phaeoptera A. 3, skg.

1642. M. macrophyllum DC. (=Lactuca macrophylla A. Gray).—

Visirors.—Loew (Berlin Botanic Garden) observed a hover-fly (Melanostoma
mellina Z.) and a bee (Chelostoma nigricorne Wy/. 6, skg.).

1643. M. prenanthoides DC.—

Vistrors.—Loew (Berlin Botanic Garden) observed 2 hover-flies (Didea inter-
media Loew, skg., and Syrphus balteatus Deg., licking the style).
Yy2

692 ANGIOSPERMAE—DICOTYLEDONES

1644. M. Plumieri DC. (=Sonchus Plumieri Z., and Lactuca Plumieri Grer.
et Godr.).—Kerner says that the heads of this species open at Innsbruck about
6-7 a.m., closing again about 8-9 p.m.

502. Sonchus Tourn.

Florets yellow. Style covered externally with sweeping-hairs directed obliquely
upwards: stylar branches beset with stigmatic papillae internally.

1645. S. oleraceus L. (Kirchner, ‘Flora v. Stuttgart,’ p. 745; Herm. Miller,
‘Fertilisation,’ p. 361; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.” pp. 98, 162.)—
In this species the expanded heads are about 20 mm. in diameter, and Kirchner
says they contain about 120 bright-yellow florets, the marginal ones being tinged
with reddish-grey on the outside. The corolla-tube is ro mm. and the ligule
6 mm. long; the anther-cylinder is orange-yellow, while the style and its branches
(scarcely 1 mm. long) are covered with blackish sweeping-hairs. The branches
ultimately diverge and bend round into a semicircle. Linnaeus says that the
heads open at Upsala at 5 a.m., closing again between 11 a.m. and 12 noon.
Kerner gives the corresponding times for Innsbruck as 6-7 a.m. and 1-2 p.m.

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Schleswig-Holstein), 2 humble-bee, 2 hover-flies, and a Lepidopterid-
Herm. Miiller (Westphalia), 3 hover-flies (1. Eristalis arbustorum Z.; 2. Syrphus
arcuatus Fal/.; 3. S. balteatus Deg.: all skg. and po.-dvg.) and a butterfly (Pieris
brassicae Z., skg.). Schletterer (Pola), the small bee Halictus villosulus K. Scott-
Elliot (Dumfriesshire), 2 hover-flies and 3 Muscids (‘ Flora of Dumfriesshire,’ p. 103).

1646. S. arvensis L. (Kirchner, ‘Flora v. Stuttgart, p. 745.)—Kirchner says
that in this species the heads contain over 200 golden-yellow florets, and spread out
to a diameter of 40-50 mm. The corolla-tube is 8-12 mm. and the ligule 8-14 mm.
in length. The stylar branches ultimately roll back so far as to make a spiral of
three turns, and of course automatic self-pollination is thereby effected.

The marginal florets close in dull weather. Linnaeus says that the heads

open at Upsala in clear weather about 6-7 a.m., closing again at roa.m. The times
given by Kerner are about 7-8 a.m. and 12-1 p.m.

Visitors.—Knuth observed the following.—

In Schleswig-Holstein (‘ Bl. u. Insekt. a. d. nordfr. Ins.,’ p. 162; Weit. Beob. ii.
Bl. u. Insekt. a. d. nordfr. Ins., p. 237).—A. Diptera. Syrphzdae: all skg.: 1. Eri-
stalis arbustorum Z.; 2. E.tenax Z.; 3. Syrphus balteatus Deg.; 4. S. pyrastri Z.;
5. S. ribesii Z. B. Hymenoptera. Apidae: 6. Apis mellifica Z., skg. and
po-cltg.; 7. Dasypoda plumipes Pz., po-cltg. C. Lepidoptera. Rhopalocera:
8. Pieris napi Z.,skg.; 9. P. rapae Z.,do. In Helgoland (‘ Bl. u. Insekt. a. Helgo-
land’), the Muscid Lucilia caesar Z., common, and small Muscids.

Alfken gives the following.—

In Juist—A. Diptera. (a) Muscidae: 1. Cynomyia mortuorum Z. (6) Syr-
phidae: 2. Eristalis arbustorum Z. B. Hymenoptera. (a) Apidae: 3. Bombus
lapidarius Z. ¥; 4. B. muscorum 7.4; 5. B. terrester Z.; 6. Dasypoda plumipes Pz.
9, freq., po-cltg. (6) Sphegidae: 7. Oxybelus mucronatus /., very common; 8. O.
uniglumis Z., rare. C. Lepidoptera. Pvzerzdae: 9. Pieris brassicae Z. At Bremen,
4 bees—1. Bombus lapidarius Z.¥; 2. B. muscorum /.8; 3. B. terrester Z. ¥;
4. Dasypoda plumipes Pz. 9 and .

COMPOSITAE 693

Herm. Miiller gives the following list for Westphalia (‘ Fertilisation,’ p. 361).—

A. Coleoptera, (a) Curculionidae: 1. Spermophagus cardui Siev., in large
numbers. (6) Telephoridae: 2. Malachius sp., po-dvg. B. Diptera. (2) Conopidae :
3. Sicus ferrugineus Z., skg. (4) Syrphidae: 4. Cheilosia sp., po-dvg.; 5. Eristalis
arbustorum Z., skg. and po-dvg., freq.; 6. E. tenax Z., do. C. Hymenoptera.
Apidae: 7. Apis mellifica Z. ¥, very common, skg. and po-cltg.; 8. Bombus sp.,
skg.; 9. Halictus flavipes /. 9, po-cltg.; 10. H. lugubris X. , skg.; 11. H. quadri-
-cinctus J. 9, po-cltg.; 12. H. rubicundus Cér. 9, po-cltg. and skg.; 13. Megachile
centuncularis Z. 9, do.; 14. Nomada varia Pz. 9, skg.; 15. Osmia spinulosa X. 9,
po-cltg. and skg. (Thuringia); 16. Panurgus banksianus X. 9 and 4, not very freq. ;
17. P. calcaratus Scop. 9 and 6, very common, skg. and po-cltg. D. Lepidoptera.
Rhopalocera: 18. Hesperia sp., skg.

The following were recorded by the observers, and for the localities stated.—

H. de Vries (Netherlands), a humble-bee (Ned. Kruidk. Arch., Nijmegen, 2. Ser.,
2. Deel, 1875). MacLeod (Flanders), 2 hover-flies (Bot. Jaarb. Dodonaea, Ghent, v,

1893, p. 434). Scott-Elliot (Dumfriesshire), several flies (Scott-Elliot, ‘Flora of
Dumfriesshire,’ p. 103).

1647. 5. asper Hill.—The florets of this species are yellow in colour.

VisiTors.—The following were recorded by the observers, and for the localities
stated —

Herm. Miiller (‘ Weit. Beob.,’ ITI, p. 96).— A. Diptera. Muscidae: 1. Anthomyia
sp., po-dvg. B. Hymenoptera. Apidae: 2. Chelostoma campanularum J. 4, skg. ;
3. Coelioxys rufescens Lep. ,do.; 4. Halictus morio /. 4, do.; 5. H. smeathma-
nellus X. 9, do.; 6. Prosopis armillata My/. $, do.; 7. Stelis aterrima Pz. 9, do.
Schletterer (Pola), the bee Osmia rufohirta Z#. MacLeod (Flanders), 2 short-
tongued bees, 3 hover-flies, 2 Muscids and a Lepidopterid (Bot. Jaarb. Dodonaea,

Ghent, v, 1893, p. 433). Alfken (Juist), the bee Dasypoda plumipes Pz.d. Verhoeff
{Norderney), the humble-bee Bombus lapidarius Z. é.

503. Crepis L.

Florets yellow. The whole outer surface of the stylar branches covered with
spinose sweeping-hairs, which extend down the margins of the style below the cleft,
as far as the anther-cylinder. The inner surfaces of the branches are beset with
stigmatic papillae up to their edges. Kerner says that the stylar branches of the
outer florets diverge and roll back so as to touch the pollen of the inner ones. In
addition to this geitonogamy, automatic self-pollination also takes place. by elongation
of the ligules and the consequent raising of the pollen adhering to them, until it is
brought into contact with the stigmatic papillae.

1648. C. biennis L. (Kirchner, ‘Flora v. Stuttgart,’ p. 747.)—Kirchner says
that the golden-yellow florets of this species are aggregated into heads of which the
upper surface is 35-40 mm. broad when expanded. The corolla-tube is 5 mm. and
the ligule 12-16 mm. long. The stylar branches ultimately roll back into a spiral
of two turns, so that in the absence of insect visitors automatic self-pollination
necessarily takes places.

Visrrors.—Herm. Miiller observed the following (‘Fettilisation,’ p. 353, ‘ Weit.
Beob.,’ III, pp. 93-4).—

A. Coleoptera. (a) Chrysomelidae: 1. Cryptocephalus sericeus Z., dvg. the
anthers (Thuringia). (4) Wetidulidae: 2. Meligethes, in large numbers. B. Diptera.

694 ANGIOSPERMAE—DICOTYLEDONES

(a) Musctdae: 3. Gonia capitata Fall., skg. (Thuringia). (5) Syrphzdae: 4. Cheilosia
chrysocoma Jg., po-dvg.; 5. Eristalis arbustorum Z., very common, skg. and
po-dvg.; 6. E. nemorum Z., do.; 7. E. sepulcralis Z., do.; 8. E. tenax Z., do. ;
g. Syritta pipiens Z., do. ; 10. Syrphus sp., po-dvg. C. Hymenoptera. (a) Aprdae :
11. Andrena denticulata X. 9 and 4, po-cltg. and skg.; 12. A. dorsata K. 9, po-cltg. ;
13. A. fulvago Chr. 9, do. (Thuringia); 14. A. fulvescens Sm. 9, do. (Thuringia) ;
15. A. parvula X. 4, skg.; 16. A. zonalis K. 8, do. (Thuringia); 17. Apis mellifica
L. ¥, do.; 18. Chelostoma campanularum X. 9 and 4, very numerous, po-cltg. and
skg.; 19. Dasypoda hirtipes /. 4, freq., still resting on the florets in the evening ;
20. Halictus albipes &. 9, skg. and po-cltg.; 21. H. cylindricus &. 9 do., é skg.,
freq.; 22. H. flavipes #. 3, skg.; 23. H. leucopus &. 9, po-cltg. (Buddeburg,
Nassau); 24. H. leucozonius Schr. 9 skg. and po-cltg., $ freq. skg.; 25. H..
longulus Sm. 9, do.; 26. H. lugubris K. 3, skg.; 27. H. maculatus Sm. 9, skg. and
po-cltg.; 28. H. nitidus Schenck 9, do.; 29. H. quadricinctus F. 4, freq., skg.; 30.
H. rubicundus Chr., $, skg.; 31. H. sexcinctus /. 9, do. (Thuringia); 32. H. villo-
sulus K. 9, very numerous, skg. and po-cltg. (Thuringia); 33. H. zonulus Sm., 9,
po-cltg. (Thuringia); 34. Heriades truncorum Z. 9 and 4, very numerous, skg. and
po-cltg.; 35. Osmia spinulosa X. 9 and 4, do. (Thuringia); 36. Panurgus banksianus XK.
9 and 4, freq., po-cltg. and skg., rolling about among the florets; 37. P. calcaratus
Scop. 9 and 3, do.; 38. Dufourea vulgaris Schenck 9 and 4, very numerous (Thuringia);
39. Stelis breviuscula (Vy/. 6, skg. (Thuringia); 40. S. phaeoptera X., do. (Thu-
ringia). (6) Zenthredinidae: 41. Tarpa cephalotes /., freq., skg. (Thuringia).
D. Lepidoptera. (a) Rhopalocera: all skg.: 42. Argynnis latonia Z. (Bavarian
Oberpfalz); 43. Epinephele janira Z. (Thuringia); 44. Lycaena sp. (Thuringia) ;
45. Melitaea athalia Zsp. (Thuringia); 46. Thecla sp. (Thuringia). (4) Sphingzdae :
47. Zygaena lonicerae £sp., skg. (Thuringia).

The following were recorded by the observers, and for the localities stated.—

Loew (Berlin Botanic Garden), 2 bees (Dasypoda hirtipes /. 4, several, skg., and
Osmia fulviventris Pz. 9, po-cltg.). Alfken (Bremen).—A. Diptera. Syrphidae:
all freq.: 1. Eristalis arbustorum Z; 2. E. nemorum Z.; 3. E. pertinax Scop.; 4. E-
sepulcralis Z.; 5. Syrphus pyrastri Z.; 6. Volucella bombylans Z., var. bombylans
Mfg. B. Hymenoptera. Apidae: 7. Andrena albicans Juli. 9; 8. A. flavipes.
Pz.9; 9. A. parvula X.9; 10. Bombus hortorum Z.%; 11. B. lapidarius L. ¥;
12. Eriades truncorum Z.9; 13. Eucera difficilis (Du), Pér. 6; 14. Halictus leuco-
zonius Schr. 9, freq., po-cltg.; 15. Osmia solskyi Aor. 9, one, po-cltg.; 16. Panur-
gus banksianus X. 9 and é; 17. Psithyrus rupestris /. 9, skg. Schmiedeknecht
(Thuringia), 2 bees (Andrena flessae Pz., and A. humilis Jm4.). Schenck (Nassau),
2 bees (Andrena fulvago Chr., and Osmia caerulescens Z. 8).

1649. C. virens L.—

Visttors.—Herm. Miiller (H. M.) (‘ Fertilisation,’ p. 353, ‘ Weit. Beob.,’ III, p. 94)
and Buddeberg (Budd.) give the following list for Westphalia and Nassau.—

A. Coleoptera. Mordellidae: 1. Mordella fasciata F. (H.M.). B. Diptera.
a Conopidae: 2. Occemyia atra /., skg. (H.M.); 3. Sicus ferrugineus Z., do.
H. M.). (6) Syrphrdae: 4. Cheilosia chrysocoma J/g., po-dvg. (Borgstette) ; 5.
Eristalis tenax Z., do. (H.M.); 6. Melithreptus scriptus Z., do. (H. M.); 7. M. taenia-
tus Mg., do. (H.M.); 8. Syrphus arcuatus Fall, do. (H. M); 9. S. balteatus Deg.,
do. (H. M.); ro. S. ribesii Z., do. (H. M.). C. Hymenoptera. Apidae: 11.
Andrena denticulata A’, 9, po-cltg. (H. M., Borgstette); 12. A. dorsata K. 4, skg.
(H.M.); 13. A. fulvago Chr. 9, skg. and po-cltg. (Budd.); 14. A. xanthura . 9, po-
cltg. (Budd.); 15. Chelostoma campanularum JZ. 9, skg. (Budd.); 16. Dasypoda
hirtipes /. 8, not infrequent, skg. (H. M., Budd.); 17. Dufourea vulgaris Schenck
9 and 6, po-cltg. and skg. (H. M., Bavarian Oberpfalz); 18. Halictus cylindricus 7.
9, po-cltg. (H. M.); 19. H. lucidus Schenck 9, skg. (Budd.); 20. H. minutus K. 9,

COMPOSITAE 695

po-cltg. (H. M.); 21. H. morio #. $ (Budd.); 22. H. smeathmanellus X. 9, skg.
and po-cltg (Budd.); 23. H.villosulus X. 9, po-cltg. (H.M.); 24. H. zonulus Sm. 9,
do. (Budd.); 25. Panurgus banksianus K. 9 and 4, rare (H.M.); 26. P. calcaratus
Scop. 9 and 4, freq., skg. and po-cltg., rolling about among the florets (H. M., Budd.);
24. Prosopis propinqua /Vy/. 9, skg. (Budd.); 28. Stelis aterrima Pz. 9, do. (Budd.).

D. Lepidoptera. Rhopalocera: 29. Pieris rapae Z., skg. (H. M., Bavarian
Oberpfalz.)

The following were recorded by the observers, and for the localities stated.—

H. de Vries (Netherlands), 2 humble-bees (Bombus subterraneus Z. ¥, and B.
terrester Z. Y) (Ned. Kruidk. Arch. Nijmegen, 2. Ser., 2. Deel, 1875). MacLeod
(Flanders), a humble-bee, 6 short-tongued bees (among them 2 species of Panurgus),
8 hover-flies, 4 Muscids, and 6 Lepidoptera (Bot. Jaarb. Dodonaea, Ghent, v, 1893,
p- 434): (Pyrenees), 4 Hymenoptera (among them a species of Panurgus), a beetle,
and 6 flies (op. cit., iii, 1891, p. 368). Scott-Elliot (Dumfriesshire), 3 short-tongued
bees, a saw-fly, several flies, and a beetle (‘Flora of Dumfriesshire,’ p. 104). Alfken
(Bremen), 12 bees—1. Andrena denticulata K.9; 2. A. fucata Sm. 9; 3. Dasypoda
plumipes Pz. 9; 4. Eriades truncorum Z.9; 5. Halictus leucozonius Schr. 9; 6. H.
punctatissimus Schenck 9; 7. H. punctulatus K.9; 8. Nomada flavoguttata K. 9
and $; 9. N. fuscicornis My/.9; 10. Osmia solskyi Afor.9; 11. Panurgus bank-
sianus K. 9 andé; 12. P. calcaratus Scop. 9 and 6.

1650. C. tectorum L.—Warnstorf describes the flower mechanism of this
species as follows (Verh. bot. Ver., Berlin, xxxvii, 1896).—

The long stylar branches are covered with spinose sweeping-hairs, all of which
project at right angles, and not only push out from the anther-cylinder, but also hold fast
the polyhedral pollen-grains. These are covered with minute drops of oil, and beset
with spinose warts on their edges. In the second stage of anthesis the stylar branches
roll spirally downwards, and thus come into contact with the pollen-grains that still
remain among the sweeping-hairs, so that if crossing has not been effected by insects,
autogamy is rendered possible.

Visirors.— Herm. Miiller (H. M.) (‘ Fertilisation,’ p. 353, ‘ Weit. Beob., III,
p. 94) in Westphalia, Buddeberg (Budd.) in Nassau, and Borgstette (Borg.) in
Tecklenburg observed the following. —

A. Diptera. Syrphidae: 1. Cheilosia chrysocoma Mg., po-dvg. (Borg.);
2. Eristalis sepulcralis Z., do. (Budd.). B. Hymenoptera. (a) Apidae: 3. Andrena
chrysopyga Schenck, po-dvg. (H. M., Thuringia); 4. A. denticulata #. 9 and 6
(Borg.); 5. A. fulvicrus X. 9, po-cltg. (H. M.); 6. Halictus malachurus £. 9, skg.
and po-cltg. (Budd.); 7. H. quadricinctus /. 4, freq. (H. M.); 8. H. rubicundus
Car. 8, skg. (H. M.); 9. H. villosulus X. 9, po-cltg. (H. M.); 10. Heriades trun-
corum L. é, skg. (H.M.); 11. Osmia spinulosa X. 9, freq., po-cltg. (H. M.);
12. Dufourea vulgaris Schenck 9 and 6 (H.M.). (6) Sphegidae: 13. Pompilus
viaticus Z. 9, skg. (H. M.).

Loew records a bee (Halictus punctulatus K. 3, skg.) and a butterfly (Poly-
ommatus virgaureae Z., skg.) for Silesia (‘Beitrage,’ p. 31); and a bee (Halictus
vulpinus WVy/. 9, po-cltg.) for Switzerland (op. cit., p. 58).

1651. C. pulchra L.—Kerner states that the heads of this species open at
Innsbruck about 6-7 a.m., closing again about 9-10 a.m.

1652. C. Jacquini Tausch (=Hieracium chondrilloides Z.).—Linnaeus says
that the heads of this species open at Upsala about 9 a.m., and close again
about 1 p.m.

696 ANGIOSPERMAE—DICOTYLEDONES

1653. C. rubra L.—This South European species opens its heads at Innsbruck,
according to Kerner, about 7-8 a.m., closing them again about 6-7 p.m.

Visirors.—Schletterer and von Dalla Torre (Tyrol) observed 3 bees—xz. An-
drena albicrus K. 9 and 6; 2. Halictus sexnotatus K. 9; 3. Prosopis annulata Z.
g and 6.

1654. C. aurea Reichb. (=Hieracium aureum Scop., Leontodon aurea Z.).
(Herm. Miiller, ‘ Alpenblumen,’ pp. 462—3.)—In this species there are usually over 100
florets in the head, which expands in the sunshine to a diameter of 35-60 mm.
The anther-cylinder projects as much as 6-7 mm. from the corolla-tube, and in
the second stage of anthesis the style extends 5% mm. further. The stylar
branches are 3 mm. long, and curve away from each other, though only in
individual cases to such an extent as to render automatic self-pollination possible.

Vistrors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Alps), 2 beetles, 4 flies, 3 Hymenoptera, and 19 Lepidoptera.
Loew (Switzerland) (‘ Beitrage,’ p. 58), 2 Bombyliid (Nemeophila plantaginis Z.),
a Noctuid moth (Agrotis ocellina S.-V.), and a butterfly (Argynnis selene S.-V.).
von Dalla Torre (Tyrol), the humble-bee Bombus mastrucatus Gers#. Schletterer
(Tyrol), 3 bees—1. Bombus mastrucatus Gerst.; 2. Halictus levis K. (=H. fulvicornis
XK.) ; 3. H. smeathmanellus KX.

It is noteworthy that the orange-red florets are chiefly visited by red butterflies
(Argynnis, Melitaea, species of Polyommatus). (Cf Senecio abrotanifolius and
Hieracium aurantiacum, and the note in Vol. I, p. 144.)

1655. C. paludosa Moench (=Hieracium paludosum Z.),—

Visirors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Alps), a Lepidopterid. Buddeberg (Nassau) (Herm. Miller, ‘ Weit.
Beob.,’ ITI, p. 94), 6 bees—1. Andrena fulvago Chr. 9, skg.; 2. Halictus leucozonius
Schr.9; 3. H. quadricinctus F. 9, skg. and po-cltg.; 4. H. tetrazonius K/g., do.;
5. Osmia aenea ZL. 3, skg.; 6. O. rufa F. 9, do. MacLeod (Pyrenees), a Muscid
(Bot. Jaarb. Dodonaea, Ghent, iii, 1891, p. 369). Scott-Elliott (Dumfriesshire),
a saw-fly, 4 hover-flies, and a Muscid (‘Flora of Dumfriesshire,’ p. 105).

1656. C. montana Tausch.—

Visitors.—Loew observed 2 bees (Chelostoma nigricorne My/. 8, skg., and
Osmia fulviventris Pz. 9, po-cltg.) in the Berlin Botanic Garden.

1657. C. rigida Waldst. et Kit.—

Visirors.—Loew observed 2 hover-flies (Eristalis nemorum Z., and E. tenax Z.)
and a bee (Megachile centuncularis Z. 9, po-cltg.) in the Berlin Botanic Garden.

1658. C. sibirica L.—

Vistrors.—Loew observed the bee Stelis phaeoptera X. 9, skg., in the Berlin
Botanic Garden.

1659. C. succisaefolia Tausch.—

VistTors.—Loew observed the bee Osmia fulviventris Pz. 9, po-cltg., in the
Berlin Botanic Garden.

COMPOSITAE 697

1660. C. albida Vill.—The florets of this species are of a yellowish colour.

Visitors.—MacLeod (Pyrenees) observed 4 bees, a Lepidopterid, 4 beetles,
a Syrphid, and 2 Muscids.

1661. C. grandiflora Tausch. (Kerner, ‘Nat. Hist. Pl.’ Eng. Ed. 1, IL.)—
Kerner has observed that when the heads of this species close in the evening they are
used as quarters for the night by some of the smaller beetles (Cryptocephalus, Meli-
gethes) and bees (Panurgus ursinus Z/r.), for at that time their interior is warmer
than the surrounding air. These insects leave their refuge at sunrise, and being
dusted with pollen transfer this to other flowers which they visit.

Kerner has also noticed that autogamy is effected by the shrivelling and spiral
twisting of the stylar branches, which are thus brought into contact with the pollen of
their own florets.

504. Hieracium L.

Florets usually bright-yellow to golden-yellow in colour, rarely orange. The
part of the style which projects from the anther-cylinder is completely covered with
pointed spinose sweeping-hairs, while the inner surfaces of the stylar branches are
beset with stigmatic papillae. Kerner states that geitonogamy takes place as in
Crepis, and that as in that genus automatic self-pollination is rendered possible
by subsequent elongation of the corolla.

Ostenfeld and Raunkjer (Bot. Tids., Kjébenhavn, xxv, 1902, pp. 409-13)
castrated the flowers of a number of species by the method used in the case of
Taraxacum (cf 683), and obtained parthenogenetically formed fruits, though no
germinating pollen-grains were observed on the stigmas. In the case of one
species (H. hyparcticum A/mg.) plants were raised from such fruits.

1662. H. Pilosella L. (Herm. Miiller, ‘Fertilisation, pp. 355-6, ‘ Weit.
Beob.,’ p. 93, ‘Alpenblumen, p. 460; Knuth, ‘BI. u. Insekt. a. d. nordfr. Ins.,’
pp- 98, 162, ‘ Bloemenbiol. Bijdragen’; de Vries, Ned. Kruidk. Arch., Nijmegen, 2. Ser.,
2. Deel, 1875; MacLeod, Bot. Jaarb. Dodonaea, Ghent, iii, 1891, v, 1893; Loew,
‘ Bliitenbiol. Floristik,’ pp. 390-8.)—Cf Fig. 210. Hermann Miller states that the
heads of this species contain 42-64 florets of a bright sulphur-yellow colour,
the marginal ones being usually streaked with red. During sunny weather the
heads expand (according to Linnaeus from 7 a.m. to 3 p.m.) to a diameter of
20 mm. They are closed in the evening and at night, also during dull weather.
The florets increase in size from the centre to the periphery; their corolla-tube
is 3-6 mm. and their ligule from 4-8 mm. long. In the first stage of anthesis
the sweeping-hairs of the style brush the accumulated pollen out of the anther-
cylinder, after which the stylar branches curve back to such an extent that automatic
self-pollination becomes possible.

VisiTors.—Alfken and Hoppner (H.) observed the following at Bremen.—

A. Coleoptera. Buprestidae: 1. Anthaxia quadripunctata Z., freq. B. Di-
ptera. Syrphidae: 2. Cheilosia soror Ze/t., skg.; 3. Eristalis tenax L., freq., skg.;
4. Helophilus trivittatus ¥., very common, skg. C. Hymenoptera. (2) Apidae:
5. Andrena albicans A/u#l/. 9 and 4, rare; 6. A. albicrus K. 6, freq., skg.; 7. A.
argentata Sm. 9, rare, po-cltg.; 8. A. chrysopyga Schenck é, rare; 9. A. convexiuscula
K. 9 and 3, do.; 10. A. fulvago Cfr., very rare, 9 skg., po-cltg., 3 skg.; 11. A.

698 ANGIOSPERMAE—DICOTYLEDONES

fulvida Schenck 9, very rare; 12. A. humilis Jm., the commonest visitor, exceedingly
numerous, 9 po-cltg. and skg., 6 skg.: they roll about among the florets as bees of their
kind usually do; 13. A. labialis A. 4, rare; 14. A. parvula X. 9, rare, skg. and
po-cltg.; 15. A. praecox Scop. 9, rare; 16. A. proxima X. 9 and 4, rare, skg.;
17. A. xanthura K. 9 and 4, rare; 18. Bombus variabilis Schmiilekn. 9 (H.);
1g. Eriades florisomnis Z. 9 and $; 20. Halictus calceatus Scop. 9, very common,
skg. and po-cltg.; 21. H. flavipes /” 9, do.; 22. H. leucozonius Schr. 9, an extremely
common visitor, skg. and po-cltg.; 23. H. minutus KX. 9, rare; 24. H. nitidiusculus X.
9, freq., po-cltg. and skg.; 25. H. punctatissimus Schenck 9, not infrequent; 26. H.
punctulatus K. 9; if one of the small black bees belonging to the genus Halictus is
taken on Hieracium Pilosella it is usually one of this species; extremely numerous
(often 4 or 5 on one head), skg.; 27. H. rubicundus Chr., freq., skg. and po-cltg. ;
28. H. zonulus Sm., rare, po-cltg. and skg.; 29. Megachile circumcincta K. 8;
30. Nomada bifida Zhs., rare; 31. N. ferruginata A. é, very rare; 32. N. flavo-
guttata K. 3, freq., skg.; 33. N. ochrostoma KX. 9 and 4, rare, skg.; 34. Osmia
claviventris Zhs. 9, once; 35. O. solskyi Mor. 9 and 4, rare; 36. Panurgus
banksianus &. 9, skg. and po-cltg., d skg. (6) Zenthredinidae: 37. Cephus nigrinus
Ths. 9 and é, skg.

Herm. Miiller (H. M.) and Buddeberg (Budd.) give the following list for
Westphalia and Nassau. —

A. Coleoptera. (a) Buprestidae: 1. Anthaxia nitidula Z. (Budd.). (6) Ceram-
bycidae: 2. Leptura livida Z. (H. M.). (c) Chrysomelidae: 3. Cryptocephalus
moraei Z., freq. (H. M., Bavarian Oberpfalz); 4. C. sericeus Z., do. (H. M.).
(2) Ocdemeridae: 5. Oedemera livida Marsh. po-dvg. (Budd.). B. Diptera.
(2) Bombyliidae: 6. Bombylius canescens J%. (Sauerland), skg. (H. M.).
(6) Conopidae: 7. Sicus ferrugineus Z., skg. (H. M., Bavarian Oberpfalz).
(c) Syrphidae: 8. Helophilus floreus Z., po-dvg. (H. M.). C. Hymenoptera.
(a) Apidae: 9. Andrena cyanescens Wy. 9, skg. and po-cltg. (H. M.); 10. A.
fulvago Chr. 9, in large numbers, skg. and po-cltg. (H.M., Budd.) ; 11. A. fulvescens
Sm. 9, po-citg. and skg. (H. M.); 12. Ceratina callosa #. 9, skg. (Budd.); 13. C.
cyanea X. é, occasional, do., skg. (H. M., Budd.) ; 14. Diphysis serratulae Pz. 5, do.
(H. M.); 15. Halictus cylindricus /. 9, skg. (H. M., Bavarian Oberpfalz); 16. H.
leucopus K. 9, skg. and po-cltg. (Budd.); 17. H. leucozonius Schr. 9, po-cltg.
(H. M., Bavarian Oberpfalz); 18. H. maculatus Sm. 9, skg. and po-cltg. (H. M.,
Bavarian Oberpfalz); 19. H. nitidus Schenck 9, do. (H. M., Bavarian Oberpfalz) ;
20. H. tetrazonius Aig. 9, skg. (Budd.); ar. H. villosulus X., skg. and po-cltg.
(H. M., Budd.); 22. Nomada fabriciana Z., 9, skg. (H. M.); 23. Osmia aenea Z.
8, do. (Budd.); 24. Panurgus banksianus A. 6 and 9, skg. and po-cltg. (H. M.,
Bavarian Oberpfalz and Thuringia); 25. P. calcaratus Scop. 9 and 4, freq., po-cltg.
and skg. (H. M.); 26. Prosopis armillata My/. 9, skg. and po-cltg. (H. M.);
27. Sphecodes gibbus Z. 9, skg. (H. M., Bavarian Oberpfalz). (6) Zenthredinidae :
28. Cephus, small species, innumerable (H.M.). D. Lepidoptera. (a) Nocturdae:
29. Euclidia mi Z., skg. (H. M.). (4) Rhopalocera: 30. Lycaena argiolus L., skg.
(H. M.); 31. Pieris brassicae Z., do. (H. M.); 32. Polyommatus dorilis A/x., do.
(Budd.).

The following were recorded by the observers, and for the localities stated.—

Knuth (North Frisian Islands and at Kiel), a bee (Panurgus sp.), a butterfly
(Pieris sp.), 5 hover-flies, and small Muscids: (Helgoland), small Muscids : (Thuringia)
(‘Bliitenbiol. Beob. in Thiiringen,’ p. 38).—A.Coleoptera. 1.Cryptocephalus sericeus
L. B.Diptera. Syrphidae: 2. Syrphus balteatus Deg., po-dvg. C. Hymenoptera.
3. Bombus soroénsis /., var. proteus Gers/., skg., 9 also po-cltg. D. Lepidoptera.
4. Pieris sp., skg. Friese (Thuringia), the bee Andrena polita Sm. Verhoeff
(Norderney), a Scarabaeid beetle (Phyllopertha horticola Z., po-dvg.), an Empid
(Hilara quadrivittata AZg.), and 2 Muscids (Anthomyia sp., and Cyrtoneura hortorum

COMPOSITAE 699

fall.). Schmiedeknecht (Thuringia), 2 bees (Andrena fulvago Chr. 9, and A. humilis
Imh.). Friese in Alsace (A.), Baden (B.), Hungary (H.), and Mecklenburg (M.), 3 bees
—1. Andrena fulvago Chr. (B., M.), rare; 2. A. humilis ZmA. (A., B., H. rare, M. more
freq.) ; 3. Nomada alboguttata H.-Sch.,rare(H.,M.). Loew in Anhalt (A.) and Branden-
burg (B.) (‘ Beitrige,’ p. 40), 6 bees—r. Andrena albicans J/wll. 9, po-cltg.(A.); 2. A.
fulvescens Sm. 8, skg. (A.); 3. A. ventralis Zh. 9 and 6, do.(A.); 4. Halictus leuco-
zonius Schr.9, po-cltg.(B.); 5. H. quadricinctus F. 9, po-cltg.(A.); 6. H. sexcinctus/* 9,
po-cltg. (B.): (Silesia)—A, Coleoptera. Chrysomelidae: 1. Cryptocephalus sericeus
L. B.Diptera. (a) Muscidae: 2. Echinomyia tessellata F. (3) Syrphidae: 3. Chryso-
toxum octomaculatum Cur/.,skg. C.Hymenoptera. Apidae: 4. Dasypoda hirtipes
F’. 9, po-cltg.; 5. Panurgus lobatus F. 3 and 9, @ po-cltg.; 6. Prosopis communis
Nyt. ; 7. P. sinuata Schenck 8. D. Lepidoptera. Rhopalocera: 8. Rhodocera
thamni Z., skg. Schletterer for the Tyrol (T.) and Pola (P.), 7 bees—1. Andrena
marginata #. (=A. cetii Schr.); 2. Dufourea vulgaris Schenck; 3. Halictoides
dentiventris Vy/.; 4. Halictus fasciatellus Schenck (P.); 5. H. flavipes /.; 6. H.
longulus Sm.; 7. H. minutus X. von Dalla Torre (Tyrol), the first 3 of these bees.
Loew (Switzerland) (‘Beitrage,’ p. 59), a beetle (Chrysochus pretiosus /.), and
2 bees (Andrena fulvago Chr. 9, po-cltg., and Panurgus banksianus X. 4, skg.).
Herm. Miller (Switzerland), a beetle, a butterfly, and 2 short-tongued bees.
MacLeod (Pyrenees), 2 species of the bee genus Panurgus, an ant, a Lepidopterid,
6 beetles, a Syrphid, and 6 Muscids (Bot. Jaarb. Dodonaea, Ghent, iii, 1891,
Pp. 369-70). H. de Vries (Netherlands) (Ned. Kruidk. Arch., Nijmegen, 2. Ser.,
2. Deel, 1875), 5 bees—1. Apis mellifica Z. 9; 2. Halictus cylindricus F. 9;
3. H. leucozonius Schr.9; 4. H. villosulus KX. 9; 5. Nomada ruficornis Z. 9.
Smith (England), 6 bees—1. Andrena albicrus K.; 2. A. fulvago Chr.; 3. A.
humilis /m’. (=A. fulvescens Sm.); 4. Dasypoda plumipes Pz.; 5. Epeolus varie-
gatus Z.; 6. Panurgus calcaratus Scop. Saunders (England), the bee Andrena
angustior A. E. D. Marquard (Cornwall), the bee Andrena fulvescens Sm.

1663. H. Auricula L.—Linnaeus says that in this species the heads open at
Upsala about 8 a.m., and close again about 2 p.m.

VisiTors.—The following were recorded by the observers, and for the localities
stated.—

Herm. Miiller (Alps), 2 beetles, 3 Muscids, a bee, and 8 Lepidoptera (‘ Alpen-
blumen,’ pp. 460-1). Loew (Berlin Botanic Garden), a Muscid (Anthomyia sp.).

1664. H. aurantiacum L.—Hermann Miller has pointed out (‘Alpenblumen,’
p- 461) that the orange-red florets of this species, like the similarly coloured ones
of Senecio abrotanifolius and Crepis aurea, are visited with zest by red butterflies
(f. Vol. I, p. 144). Kerner states that the heads open at Innsbruck about 6-7 a.m.,
and close again 3-4 p.m.

VistTors.—Herm. Miiller (Alps) observed butterflies (3 sp. of Argynnis, one of
Melitaea, and one of Polyommatus), and a hawk-moth (Zygaena sp.).

1665. H. villosum Jacq. (Herm. Miiller, ‘Alpenblumen,’ pp. 461-2.)

Vistrors.—Herm. Miiller saw 3 Hymenoptera, 2 Lepidoptera, and 3 flies in
the Alps.

1666. H. granduliferum Hoppe. (Herm. Miiller, ‘Alpenblumen,’ p. 462.)

Visirors.—Hermann Miller saw 2 flies in the Alps.

1667. H. albidum Vill. (=H. intybaceum Wui/.).—

Vistrors.—Herm. Miiller saw a humble-bee and a Lepidopterid in the Alps
(op. cit.).

JOO ANGIOSPERMAE—DICOTYLEDONES

1668. H. staticifolium Vill. (Herm. Miiller, ‘Alpenblumen,’ p. 461.)—

Visirors.—Herm. Miiller saw 13 flies, 7 bees, and 19 Lepidoptera in the Alps.
Schletterer records the bee Andrena propinqua Schenck for the Tyrol.

1669. H. laevigatum Willd.—

VisiTors.— Loew observed the beetle Cryptocephalus sericeus Z. in Brandenburg
(‘ Beitrige,’ p. 40).

1670. H. vulgatum Fries.—

Visitors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Kiel), a hover-fly (Helophilus pendulus Z., skg. and po-dvg.), a Muscid
(Musca sp.), and a Tortricid moth (Tortrix sp.). Herm. Miiller (‘ Fertilisation,’
p- 356, ‘ Weit. Beob.,’ III, p. 93).—A. Diptera. Syrphidae: 1. Eristalis tenax L.,
po-dvg. (Bavarian Oberpfalz). B. Hymenoptera. Apzdae: 2. Andrena coitana XK.
é, skg.; 3. A. denticulata K.é,do.; 4. A. fulvescens Sm. 9, po-cltg.; 5. Bombus
rajellus K. 9, skg.; 6. B. sylvarum Z. 9, do.; 7. B. terrester Z. 9, do.; 8. Halictus
cylindricus /. 9 and 4, freq., po-cltg. and skg.; 9. Panurgus calcaratus Scop. 9 and é,
freq., po-cltg. and skg. C. Lepidoptera. Rfopalocera: all skg.: 10. Epinephele
hyperanthus Z. (Bavarian Oberpfalz); 11. E. janira Z. (Bavarian Oberpfalz) ;
12. Erebia ligea Z. (Fichtelgebirge); 13. Lycaena icarus Roff.; 14. Melitaea
athalia Hsp. (Thuringia). MacLeod (Flanders), a Bombus, 5 short-tongued bees,
a Siricid, 14 hover-flies, 6 Muscids, 4 Lepidoptera, and 2 beetles (Bot. Jaarb.
Dodonaea, Ghent, v, 1893, pp. 435-7). H. de Vries (Netherlands), the bee
Chelostoma florisomne Z. $ (Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel, 1875).
Loew (Berlin Botanic Garden), a hover-fly (Eristalis nemorum Z.) and 2 bees
(Megachile centuncularis Z. 9, po-cltg., and Osmia fulviventris Pz. 9, do.).

1671. H. murorum L.—Linnaeus states that the heads of this species open at
Upsala about 9 a.m., and close again about 1 p.m.

Visttors.—The following were recorded by the observers, and for the localities
stated.—

Knuth (Thuringia), a beetle (Cryptocephalus sericeus Z.), 2 bees (Bombus
soroénsis /., var. proteus Gers, skg., and Halictus punctulatus X. 9, skg. and
po-cltg.), and a butterfly (Pieris napi Z., skg.) (‘ Bloemenbiol. Bijdragen’): (Lauter-
berg in the Harz, 9.10.’97), the hover-fly Eristalis rupium /., po-dvg. Loew
(Alps), the short-tongued bee Andrena fulvago Cér. 9, po-cltg. (‘ Beitrage,’ p. 58).
Borgstette (Tecklenburg), the bee Andrena listerella A. 9, po-cltg. Buddeberg
(Nassau), 2 bees (Halictus albipes /. 4, skg., and H. tetrazonius Ag. 9, do.) (Herm.
Miiller, ‘ Weit. Beob.,’ p. 93). Alfken (Bremen), 9 bees—z. Andrena denticulata
K. &; 2. Dasypoda plumipes Pz. 9 and 4; 3. Dufourea halictula Wy. 9; 4. D.
vulgaris Schenck 9 and $; 5. Halictus flavipes #. 5; 6. H. leucozonius Schr. 6;
4. Panurgus banksianus £. 9 and 6; 8. P. calcaratus Scop. 9 and 8; 9. Prosopis
communis /Vy/. 9. Schletterer (Tyrol), the humble-bee Bombus soroénsis #. Loew
(Berlin Botanic Garden), 2 bees (Coelioxys rufescens Lep. 9, skg., and Osmia fulvi-
ventris Pz. 9, po-cltg.).

1672. H. umbellatum L. (Herm. Miller, ‘Fertilisation,’ pp. 354-5, ‘ Weit.
Beob.,’ II, p. 92; Knuth, ‘Bl. u. Insekt. a. d. nordfr. Ins.,’ pp. 98, 163; Verhoeff,
‘Bl. u. Insekt. a. d. Ins. Norderney’; de Vries, Ned. Kruidk. Arch., Nijmegen,
2. Ser., 2. Deel, 1875.)—The expanded heads of this species are 25 mm. in
diameter. Hermann Miiller says that the golden-yellow florets possess corolla-

COMPOSITAE 7or

tubes 3-5 mm. and ligules 8-16 mm. long. The style grows through the
anther-cylinder, sweeping before it the pollen which adheres to its spinose
sweeping-hairs, for 6 mm., 23 mm. of this being taken up by its branches. In
the second stage of anthesis these branches gradually diverge and bend back
until the stigmatic papillae touch the pollen clinging to the sweeping-hairs, so
that automatic self-pol-
lination must take place
if insect-visits fail.
Linnaeus says that
the heads open at Upsala
about 6 a.m., and close
again about 5 p.m.

Visitors. — Herm.
Miller gives the following
list for Lippstadt.—

A. Coleoptera. 1.
Coccinella quinquepunc-
tata Z. B. Diptera. (a)
Conopidae: 2. Occemyia
atra #., skg.; 3. Sicus
ferrugineus Z., do. (0)
Syrphidae: 4. Eristalis
arbustorum Z., very com-
mon, po-dvg.and skg.; 5.
E. nemorum Z., skg.; 6.
E. tenax Z., very common,
po-dvg. and skg.; 7. Syr-
phus balteatus Deg., do. ;
By 3, tbesi me po-dyg. Fic. 210. Hieractum umbellatum, L. (after Herm. Miller). (1) Floret
C. Hymenoptera. (2) in the second stage (x 7). (2) The stylar branches, still further recurved,
Apidae : 9. Apis mellifica so that autogamy is ultimately possible. (3) The part a6 of the left stylar
LS: ¥, freq., skg. and po- branch of (1) (x 60). _ (4) Floret in which self-pollination is taking place
cltg. ; 10. Bombus lapida- (x 7). ¢, stigmatic papillae; d, sweeping-hairs ; ¢, pollen-grains; /, anther-

i cylinder ; g, filaments; 4, style; z, corolla-tube; 4, unsymmetrical limb of
rius Z. ¥, skg. ; 11. Coe- the corolla; 4, pappus; ms, ovary.

lioxys conoidea 7/7. 9, do.;

12. C. simplex yl. 9, do.; 13. Dasypoda hirtipes #. 9, skg. and po-cltg.;
14. Halictus cylindricus #. 3, skg.; 15. H. leucozonius Schr. 9 and 4, po-cltg.
and skg.; 16. H. villosulus X. 9 and 4, do.; 17. H. zonulus Sm. 9; 18. Megachile
argentata F’. 9, skg.; 19. M. willughbiella A. 4, do.; 20. Panurgus calcaratus Scop.
g and 4, po-cltg. and skg.; 21. Sphecodes gibbus Z. 4, skg. (6) Chrystdidae :
22. Hedychrum lucidulum /. 6 D. Lepidoptera. Rhopalocera: all skg.:
23. Hesperia sp.; 24. Lycaena icarus Roff.; 25. Pieris napi Z., freq.; 26. P-
rapae Z.; 27. Polyommatus dorilis A//n.; 28. Pararge megaera L., do.; 29.
Vanessa urticae Z., not infrequent.

Alfken observed the following.—

In Juist.—A. Diptera. (a) Muscidae: 1. Cynomyia mortuorum Z.; 2.
Echinomyia tessellata Z.; 3. Lucilia caesar Z. (4) Syrphidae: 4. Eristalis tenax L.;
5. Melithreptus sp.; 6. Platycheirus manicatus 4g. B. Hymenoptera. Apidae:
7. Bombus distinguendus A/or.; 8. B. hortorum Z.é; 9. B. lapidarius Z. ¥ and 3;
ro. B. muscorum /. ¥ andé; 11. B. terrester Z.d; 12. Dasypoda plumipes Pz. 9,

7jo2 ANGIOSPERMAE—DICOTYLEDONES

very common, often 3 or 4 in a head, skg. and po-cltg. 6 skg. (6) Sphegidac: 13.
Ammophila sabulosa Z.; 14. Oxybelus mucronatus #. C. Lepidoptera. (2) Zy-
caenidae: 15. Lycaena icarus Roff.; 16. Polyommatus phlaeas LZ. (6) Pueridae:
17. Pieris brassicae Z.; 18. P. napi Z. (c) Satyridae: 19. Epinephele janira L.
In Bremen, 9 bees—1z. Andrena gwynana &. 4, 2nd gen.; 2. A. humilis Jm. 9;
3. Bombus distinguendus Jor. ¥; 4. B. hortorum Z. 9; 5. B. muscorum /. ¥;
6. Dasypoda plumipes Pz. 9 andé; 7. Melitta leporina Pz. 6; 8. Panurgus banksianus
K.gandé; 9. P. calcaratus Scop. 9 and 6.

Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis aeneus Scop.; 2. E. arbustorum Z.;
3. E.nemorum Z.; 4. E. tenax Z.; 5. Helophilus floreus Z.; 6. Syritta pipiens Z. ;
7. Syrphus albostriatus Fal. B. Hymenoptera. (a) Apzdae: 8. Apis mellifica L.
¥, skg. and po-cltg.; 9. Dasypoda hirtipes /. 8, do.;_ 10. Halictus leucozonius Schr.
g and @, skg. (6) Sphegidae: 11. Ammophila sabulosa Z. C. Lepidoptera.
Rhopalocera: all skg.; 12. Rhodocera rhamni Z.; 13. Pieris brassicae Z.; 14. P.
rapae L.

The following were recorded by the observers, and for the localities stated.—

Knuth (North Frisian Islands), Apis, a Bombus, a Panurgus, a hover-fly, and
a butterfly. Verhoeff (Norderney).—A. Diptera. Syrphidae: 1. Eristalis arbustorum
L. 3, freq.; 2. Syrphus corollae /. 4, do.; 3. S. nitidicollis AZg. 9, occasional.
B. Hymenoptera. Apidae: 4. Bombus lapidarius Z. ¥ and 4, freq.; 5. B. terrester
L.%; 6. Psithyrus rupestris /. 3. C. Lepidoptera. Mymphalidae: 7. Argynnis
latonia Z., occasional. H. de Vries (Netherlands), the humble-bee Bombus sub-
terraneus L. ¥§ (Ned. Kruidk. Arch., Nijmegen, 2. Ser., 2. Deel, 1875).

1673. H. australe Fries. —

Visitors.—Loew (Berlin Botanic Garden) observed 3 bees—1. Apis mellifica
Z. ¥, skg. and po-cltg.; 2. Bombus agrorum /. 6, skg.; 3. Halictus leucozonius
Schr. 9, do.

1674. H. boreale Fries.—

Visitrors.—Loew observed the bee Prosopis armillata /Vy/. 9, skg., in the Berlin
Botanic Garden.

1675. H. brevifolium Tausch.—

Visrrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. (ac) Muscidae: 1. Anthomyia sp. (4) Syrphtdae: 2. Helophilus
floreus Z. B. Hymenoptera. Apidae: 3. Apis mellifica Z. ¥, skg. and po-cltg.;
4. Bombus terrester Z. 9, skg.; 5. Halictus cylindricus 7. 8, do.; 6. Halictus
nitidiusculus X. 8, do.; 7. Panurgus calcaratus Scop. 9.

1676. H. bupleuroides Gmel.—

Visirors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis tenax Z.; 2. Helophilus pendulus Z.
B. Hymenoptera. Apidae: 3. Apis mellifica Z. ¥, skg. and po-cltg.; 4. Heriades
truncorum Z. $, skg.; 5. Prosopis sp. 9, do.

1677. H. crinitum Sibth. et Sm.—

Vistrors.—Loew observed the following in the Berlin Botanic Garden.—

A. Diptera. Syrphidae: 1. Eristalis nemorum Z.; 2. E. tenax Z.; 3. Helo-
philus floreus Z.; 4. Syritta pipiens Z. B. Hymenoptera. Apidae: 5. Apis
mellifica Z. ¥, skg. and po-cltg.; 6. Halictus cylindricus F. 6, skg.; 7. H. leucozonius
Schr. 9, do.

COMPOSITAE 703

1678. H. cymosum L.—

VisiTors.—Loew observed the hover-fly Syrphus balteatus Deg. in the Berlin
_ Botanic Garden.

1679. H. echioides Lumnitz.—

Vistrors.—Loew observed the bee Osmia fulviventris Pz. 9, po-cltg., in the
Berlin Botanic Garden.

1680. H. foliosum Waldst. et Kit.—

VisiTors.—Loew observed a hover-fly (Pipiza festiva Mfg.) and a bee (Chelo-
stoma campanularum &’” 9, po-cltg.) in the Berlin Botanic Garden.
1681. H. hirsutum Bernh.—

VisiTors.—Loew observed a hover-fly (Syritta pipiens Z.), a bee (Psithyrus

vestalis Fourcr. $, skg.), and a butterfly (Vanessa io Z., skg.) in the Berlin Botanic
Garden.

1682. H. porphyritae F. W. Schultz.—

Visitors.—Loew observed the bee Stelis aterrima Pz. 9, skg., in the Berlin
Botanic Garden.

1683. H. pratense Tausch.—

Vistrors.—Loew observed the hover-fly Syrphus balteatus Deg. in the Berlin
Botanic Garden.

1684. H. pulmonarioides Vill.—

Visrtors.—Loew observed the bee Apis mellifica Z. ¥, skg. and po-cltg., in the
Berlin Botanic Garden.

1685. H. Retzii Griseb.—

Visitors.—Loew observed the bee Osmia fulviventris Pz. 9, po-cltg., in the
Berlin Botanic Garden.

1686. H. virosum Pall.—

VisiTors.—Loew observed a hover-fly (Eristalis tenax Z.) and 3 bees (1. Apis

mellifica Z. ¥, skg. and po-cltg.; 2. Halictus cylindricus F. 8, skg.; 3. H.
nitidiusculus A. 9, skg.) in the Berlin Botanic Garden.

LVI. ORDER S7VYLIDIEAE R. Br.

Delpino (‘Ult. oss.,” pp. 125-6) examined herbarium specimens of species
belonging to this order, and describes them as markedly protandrous, so that
they are apparently adapted to insect-visits.

OXFORD
PRINTED AT THE CLARENDON PRESS

BY HORACE HART, M.A.
PRINTER TO THE UNIVERSITY

sniaeesss
eneicinne
foten en
ies San

chests stmeeriy ene
ee peste
wins rife ro.
slash mln nha id ai we
oe fecapninimregagmlsvanersaeeenia
eer nian

5 Spemeaaey
pei treatin atinsesetarn
See

foot ant

Spi)

war LER rat Moms
Siptetcteieteted et t-inurtiernctaniqaian tinted ea ati ap iy
elder ymoeser ng ousny ges eRe pay eens ems itlsTmiosid ac yeletniat niall
oe eiehe ecmeres aly

Sarit arta

OS EAA eM A Ades ef ati mig tte oA pitted
Sofitel seanaiciewesenlcinratesatencieay allsias
shitrcreh

ert

feat
bee i eed nO
Haining cat
el pat

shes ee cleda tues le

ap
ai jarnaees
pe ree

igmat atl ee

Ax iuetetcaret adel not Ld atiin
fe a Postmen wire i

Caan

Paste
sie arent Tete

Pistauadiceivielet eee

Mie

chiertiterneesehei cette

EUEM eC ATAG zelah i
Peebev onthe

ise
[ii itiered ont i
tele

se ees 1
i HE et re
Sea im ileal sens

eerie aaa teeta

Tir eewaUianetal ct dedisssi cat noel
een tassel bet ehet ett deeceld

sey

nie Petntrigwiccu nt aicayngs ctaeetutans ge ets ap et

featiteeel The siieed
rHIeee geen
Titel Liha an cote ibe dele vedi
{heathseunderpe ely

FHA H F
Later trashebutarinyaenia ata tas
Pealict evs tele gael!

ena ation tu

(utepelparaeidate persons 5