Watsonia

roe a

-. 2

. .

: .
. sa

ae ena
3 1

‘ - a cag ous. ot a NP ele eget ge ges geet 8 ct el ee ae ead
ef Ode cre be. or Ste Bt ake wie .f é ears a =
28 tor ete ans Ope 8
sites * Shao oe pteig Blea
ene see : se :
. -* r arte Sa ef grees sete
oa Bo i.ryea” SF. ot ore + feet e cree > OLS 92» fseP
‘ he at tte - septs go ere tse
. red + be 8s .
oe & me shat sBets® gre . oe ee ee ee ee
7 . este 8 se ots ‘

Pe tad see
foie

Fase >
tee

wcgtetres =
tay tte

Pe ee ee
Sree a Gece RTE Seger He
abled owbrai gs.

et cet ht :
byte koe ge tee free
oes

tie
fy teres
wer acid sete lees

[ere ‘sete
Pe et eae sos

no

HARVARD UNIVERSITY

LIBRARY

OF THE

GRAY HERBARIUM

Digitized by the Internet Archive
in 2014

https://archive.org/details/watsonial/Sbota

WATSONIA

Botanical Society of the British Isles.

Patron: Her Majesty Queen Elizabeth the Queen Mother

Applications for membership should be addressed to the Hon. General Secretary,
c/o Department of Botany, British Museum (Natural History), Cromwell Road, London,
SW7 SBD, from whom copies of the Society’s Prospectus may be obtained.

Officers for 1988-89

Elected at the Annual General Meeting, 7th May 1988
President, Professor C. A. Stace

Vice-Presidents, Dr H. J. M. Bowen, Dr F. H. Perring, Dr A. J. Richards, .
Dr N. K. B. Robson

Honorary General Secretary, Mrs M. Briggs
Honorary Treasurer, Mr M. Walpole

Back issues of Watsonia are handled by Messrs Wm Dawson & Sons Limited,
Cannon House, Folkestone, Kent, to whom orders for all issues prior to Volume 17
part 1 should be sent.

Recent issues (Vol. 17 part 1 onwards) are available from the Hon. Treasurer of the
B.S.B.I., 68 Outwoods Road, Loughborough, Leicestershire.

Watsonia, 17, 217-245 (1988) 17

The Aizoaceae naturalized in the British Isles

Cale. PRO LN

Institute of Terrestrial Ecology, Monks Wood Experimental Station, Abbots Ripton,
Huntingdon, PEI7 2LS

and
Po. Sel

Botany School, Downing Street, Cambridge, CB2 3EA

ABSTRACT

An introduction is given to the Aizoaceae with particular emphasis on the subfamilies Mesembryanthemoideae
and Ruschioideae. It is followed by a detailed account of the complicated floral parts and fruits of the species of
these groups. A conspectus of the subfamilies and genera naturalized in the British Isles is followed by a key to
the species. Accounts of the nomenclature, typification, morphology, distribution, habitat and reproductive
biology of Aptenia cordifolia (L.fil.) Schwantes, Ruschia caroli (L. Bolus) Schwantes, Lampranthus roseus
(Willd.) Schwantes, L. falciformis (Haw.) N.E.Br., Oscularia deltoides (L.) Schwantes, Disphyma crassifolium
(L.) L. Bolus, Drosanthemum floribundum (Haw.) Schwantes, Erepsia heteropetala (Haw.) Schwantes,
Carpobrotus edulis (L.) N.E.Br. var. edulis, var. rubescens Druce, var. chrysophthalmus C. D. Preston & P. D.
Sell, var. nov., and C, glaucescens (Haw.) Schwantes, follow. An Appendix by R. D. T’Ons gives the
chromosome numbers.

~

INTRODUCTION

The Aizoaceae are a large family, with their centre of distribution in South Africa (Rowley 1978).
Two of its component subfamilies, the Mesembryanthemoideae and the Ruschioideae, are
sometimes removed as a separate family, the Mesembryanthemaceae. All the species in these two
subfamilies were once included in the genus Mesembryanthemum L., but have since been divided
into as many as 125 genera which contain about 2000 species (Rowley 1978).

South African Aizoaceae have long been cultivated in British gardens. A detailed account of the
botanical exploration of South Africa is given by Gunn & Codd (1981). Although Bartolomeu Dias
rounded the Cape of Good Hope in 1488, it was not until 1605 that any plant specimen from South
Africa was described in print. Live bulbs were collected from the beginning of the 17th century and
increasing numbers of plants from South Africa, including some ‘mesembs’, were introduced by the
Dutch in the second half of that century. James Petiver received live material of the Aizoaceae from
the Temple Garden, London, in 1706 (Edwards 1968). The collection grown by James Sherard at
Eltham in Kent was illustrated by Dillenius (1732) in the important Hortus Elthamensis, which
provides an almost complete coverage of the ‘mesembs’ then known in cultivation (Gunn & Codd
1981). There is at Oxford, in the library of the Department of Plant Sciences, a copy coloured by
Dillenius himself. We have also seen a similarly coloured copy in a private collection.

Two botanists are pre-eminent in the interpretation of early ‘mesemb’ names. Adrian Hardy
Haworth (b. 1768, d. 1833) described 206 species in his Synopsis (Haworth 1812). His descriptions
were based on living plants, including many grown by himself as well as some in the collections at
Kew. His herbarium was bought by H. B. Fielding, who threw most of the specimens away.
Fortunately, a great many coloured drawings were made by George Bond and Thomas Duncannon
between 1822 and 1835 of the plants Haworth had seen at Kew (Brown 1920), and these are
available for selection as neotypes. In the facsimile of all Haworth’s works on succulent plants,
Stearn (1965) has written an informative introduction.

218 C.D. PRESTON AND P. D. SELE

Prince Salm-Dyck (b. 1773, d. 1861) was an enthusiastic cultivator and student of succulent
plants. His most important work is the Monographia Generum Aloes. et Mesembryanthemi,
published in seven parts (Salm-Dyck 1836-1863). It is a collection of lithographed and partly hand-
coloured plates accompanied by Latin descriptions. A detailed account of the work has been
published by Stearn (1938).

A number of ‘mesembs’ have become naturalized by the sea in the British Isles, and these are
described in this paper. We regard species as naturalized if they are established in semi-natural
vegetation. It is not necessary that a species should reproduce by seed to satisfy this criterion,
although some of the naturalized species certainly do so. The extent to which the individual species
are naturalized is discussed in the accounts that follow.

The localities where ‘mesembs’ are naturalized are all coastal. The species’ tolerance of salt-laden
gale-force winds enables them to grow very close to the sea, where the microclimate is most
equable. In such positions naturalized colonies often escape low winter temperatures which severely
damage or kill plants of the same species growing in gardens a short distance inland. Even if
extensively damaged, colonies can regrow from vegetative parts or seed (Preston 1988).

Lousley (1973) regarded the naturalized Aizoaceae as “plants dependent on a run of mild winters
and deliberate planting’. We consider that this underestimates their ability to persist and spread.
Carpobrotus edulis, in particular, is now a characteristic feature of the seaside landscape in many
parts of south-western England. Betjeman (1945) mentions it growing with stonecrop on a Cornish
cliff in his poem Sunday Afternoon Service in St Enodoc Church, Cornwall and Rowse (1941),
describing Henry VIII’s castle at St Mawes, says “‘it remains unchanged, the walls of its three-lobed
batteries but weathered to a silver grey, the grass slopes covered with mesembryanthemum and
rock-plants down to the edge of the water’’. Flowers of Carpobrotus and plants of Lampranthus are
featured on picture postcards and sometimes protected by wild-flower lovers. This protection is
unnecessary: Carpobrotus competes so effectively with native species that its continued spread is a
source of anxiety to conservationists (Hopkins 1979; Frost 1987).

TAXONOMY OF THE AIZOACEAE

Commenting on attempts to identify one naturalized species, J. P. M. Brenan (in litt. to L. J.
Margetts 1970) wrote that ‘‘one would have to be a magician to give itaname. . . the whole family is
in a state of chaos’’. Four factors have contributed to this unhappy state of affairs:
i. Many of the species were originally described from material cultivated in European gardens. The
precise origin of the cultivated stock was not known, so the names are difficult to relate to plants in
their native habitats. Often type specimens were not preserved, or have not survived.
ii. Herbarium specimens of these succulent plants are difficult to identify, as leaf shape and flower
colour are destroyed when the plant is pressed and dried. Pressed specimens can often be identified
only to generic level. Even if type specimens exist they are often of little use. Good coloured
illustrations, or photographs, are more useful than specimens. For this reason we have selected
illustrations as lectotypes when they are available.

Recently Fuller & Barbe (1981) have described a method of pressing succulent plants, including
Carpobrotus edulis, using a microwave oven. We have not experimented with this technique.
iii. Although many enthusiasts have studied the ‘mesembs’, their efforts have been devoted to the
description of more and more new species. H. M. L. Bolus alone described 1445 species but “‘she did
not regard all these as true new species” (Herre 1971). Descriptions of new taxa have rarely been
accompanied by a key or a conspectus to the species in the genus, and few generic monographs are
available.
iv. Very little is published about the ecology, reproductive biology and cytology of ‘mesemb’
species. Within genera or closely related groups of species the taxa seem to be either geographically
or ecologically isolated, and it is not known if they are interfertile. In the absence of this
information, it is impossible to interpret the described taxa biologically.

Lousley’s (1971, 1973) studies of the Aizoaceae naturalized in the Isles of Scilly are the most
valuable published accounts of the British species. McClintock’s (1975) and le Sueur’s (1984) studies

NATURALIZED AIZOACEAE 219

of the Channel Island species are also useful. Recently Kent (1982) listed the species reported from
the British Isles.

METHODS OF STUDY

Initially we collected living material of the Aizoaceae naturalized in the British Isles and grew this in
a greenhouse at Cambridge University Botanic Garden. Descriptions were first prepared from these
cultivated plants, then compared in the field with wild populations. It was necessary to modify the
descriptions of cultivated material both because plants grown in the greenhouse lacked the vigour of
plants in the wild, and because of the greater environmental variation present in wild populations.
The cultivated plant material was used to count the chromosome numbers of the naturalized species
(see Appendix).

Some herbarium specimens, and a larger number of colour photographs, have been deposited in
CGE. Full details of all records available to us have been deposited at the Biological Records
Centre, Monks Wood Experimental Station.

We are conscious that our work is only a preliminary contribution to the elucidation of the
taxonomy and nomenclature of the species naturalized in Britain. In particular, further work on the
British species of Lampranthus is needed. However, we believe that progress is more likely to result
from studies of the taxa in their native habitats rather than from more intensive studies of the species
in Britain.

MORPHOLOGY OF FLOWERS AND FRUIT

Some authors treat the outer perianth segments as sepals, while others consider them to be the
divided part of the calyx. The latter view is taken here and they are regarded as calyx lobes. They are
very unequal in size and some of them usually have wide scarious flaps which cover the petals in bud.
The petals and stamens are usually situated between the point where the segments join the
remainder of the calyx and the top of the half-inferior ovary. The whole arrangement is similar to
the hypanthium in the Rosaceae.

The inner perianth segments are usually considered to be derived from stamens. They are here
called petals and include those inner ones that are often narrower and smaller than the outer. Only
when they are in the form of stamens, but without anthers, are they called staminodes. There is
sometimes a graded sequence between stamens and petals, both in form and colour. In Aptenia
cordifolia the petals are united at the base to form a short corolla tube; the stamens arise from this
tube. The nectaries are sunk in a hollow around the rim of the ovary. In the other species naturalized
in Britain the petals are free, and the nectaries form a slight crest around the rim of the ovary.

The capsules of the Aizoaceae are extremely complicated in structure. An illustrated survey is
given by Schwantes (1957). In most species growing in the British Isles the capsule is hygrochastic,
i.e. it opens when it becomes moist and closes when it dries out again. In the simplest case (Aptenia)
the capsule has axile placentation with open loculi, so that raindrops can easily wash out the seeds.
In the tribe Ruschieae the capsules are much more complex and the placentation is parietal. There
are five loculi and partition walls. The lid of the capsule is opened and closed by the expansion of the
epidermal lining of the loculus when it becomes wet. The cells of expanding tissue have a
trhomboidal shape, the inner shorter axis elongating considerably while the outer longer axis
remains almost unchanged. The result is a pronounced elongation of the expanding tissue in one
direction while the remaining tissues of the capsule do not change appreciably, thus causing the
valves to unfold outwards. In some species a membrane covers the loculi so that the seeds are not
exposed when the capsules open. The membranes are elastic and are never found in the middle of
the loculus. The division is always brought about by the valve wings which are formed by the
splitting of the false septum which projects into the loculus from above. The valve wings always
remain firmly attached to the valve. When the capsule opens for the first time they act like the rip-
cords of a parachute and bring about the exact division of the parts of the covering membrane. The
membrane never covers the loculus completely, as a portion of the endocarp on the roof of the

220 C2 DAPRESTONJAND:F. 7D; SELL

loculus is required for the formation of the expanding mechanism. The- opening remaining is
sometimes reduced to two minute openings by a swelling of the placenta (tubercle).

SYNOPSIS OF THE CLASSIFICATION OF THE BRITISH GENERA

In the synopsis below we follow the classification proposed by Schwantes (1971) for the
Mesembryanthemaceae. Unless otherwise stated, the number of species per genus is taken from
Herre (1971). Critical revision of the larger genera would almost certainly reduce the number of
recognized species considerably.

SUBFAMILY MESEMBRYANTHEMOIDEAE
Nectaries sunken in a hollow. Placentation axile.

Tribe Aptenieae
Petals soft, not rigid.

Aptenia N.E.Br. in Gdnrs’ Chron., ser. 3, 78: 412 (1925). Holotype species: A. cordifolia (L. fil.)
Schwantes, cf. J. Bot., Lond. 66: 139 (1928). Number of species: 2.

SUBFAMILY RUSCHIOIDEAE
Nectaries forming a slight crest round the rim of the ovary. Placentation parietal.

Tribe Ruschieae
Fruit a dry hygrochastic capsule with numerous seeds.

Ruschia Schwantes in Z. SukkulKde 2: 186 (1926). Holotype species: R. rupicola (Engler)
Schwantes. Number of species: c. 350.

Lampranthus N.E.Br. in Gdnrs’ Chron., ser. 3, 87: 71 (1930). Holotype species: L. multiradiatus
(Jacq.) N.E.Br. Number of species: 178 according to Herre (1971), 56 according to Glen (1980).

Oscularia Schwantes in Mollers dtsch. GGrt.-Ztg 42: 187 (1927). Holotype species: O. deltoides (L.)
Schwantes. Number of species: 1 (Glen 1986).

Disphyma N.E.Br. in Gdnrs’ Chron., ser. 3, 78: 433 (1925). Lectotype species: D. crassifolium (L.)
L. Bolus, chosen by L. Bolus in FI. Pl. S. Afr. 7: t. 276 (1927). Number of species: 5 or 1 (see
below).

Drosanthemum Schwantes in Z. SukkulKde 3: 14, 29 (1927). Holotype species: D. hispidum (L.)
Schwantes. Number of species: c. 95.

Erepsia N.E.Br. in Gdnrs’ Chron., ser. 3, 78: 433 (1925). Lectotype species: E. inclaudens (Haw.)
Schwantes, chosen by N. E. Brown in E. P. Phillips, Gen. S. Afr. Fl. Pl. 248 (1926). Number of
species: 38.

Tribe Carpobroteae
Fruit fleshy, indehiscent, the outer wall becoming tough and leathery, the seeds embedded in
mucilage.

Carpobrotus N.E.Br. in Gdnrs’ Chron., ser. 3, 78: 433 (1925). Lectotype species: C. edulis (L.)
N.E.Br., chosen by N.E. Brown in E.P. Phillips, Gen. S. Afr. Fl. Pl. 249 (1926). Number of
species: 29.

KEY TO SPECIES NATURALIZED IN THE BRITISH ISLES
1. Leaves flat, ovate; petals connate at base, forming a short tube; calyx lobes and stigmas 4
Weddbie vev db TOA GORA AEGON Le V osgldel ae Pitney i ath heed pea & Veet af renee 1. Aptenia cordifolia

1. Leaves triangular, cylindrical or rarely narrowly obovoid in section; petals free; calyx lobes
(4—)5=6; stigmmpag-S OF DOA—14% CE Fa 0 oe thins ued ied « o PRU Esa Rss SONS eb pe rend Pee Peete 2

“

NATURALIZED AIZOACEAE 221

Daihkeavesc memate pantveast/oneot the angles! is. ela vis cea a/5. 220i) Oa ca uletek Ok adds 3
- vildaver notmentote dmithe anglet: tcl lvzicnsie resin. sheet Renee ise ke desea te As 4
3. Leaves broadly triangular-obovoid in outline, deltate in section, less than twice as long as

broad; petals exceeding calyx lobes; stigmas narrowly ovoid with curved apex .................

py cbscuas eetcnie bun Veet Caews lneskepenVidsad eens a) cst alk eee peek Abi ae Ad. ea 5. Oscularia deltoides

3. Leaves asymmetrically oblanceolate in outline, triangular in section, at least twice as long as
broad; petals shorter than calyx lobes; stigmas small, roundish in outline .......................

as cee rebass: Alea op. DeidrateaeE ITA RNNe ym LL AEG RI laes ads a od Males A 8. Erepsia heteropetala

4. Stems with numerous hairs; leaves cylindrical or narrowly obovoid, covered with glass-like
FOUETOIOS. tures. penne cts). clos eet). Sons. ae 7. Drosanthemum floribundum

4. Stems glabrous; leaves + triangular in section, not covered with glass-like tubercles ...... 5
5. Leaves in groups of 3-10, narrowly linear-lanceolate or linear-oblanceolate in outline, with a
row of almost flat papillae on the angles; stigmas ending in a long awn ......................008

Wo ade eth tee eee eae. Bes At Sa 6. Disphyma crassifolium
5. Leaves not obviously in groups of 3—10 and some obviously decussate, without flat papillae on
the angles; stigmas plumose or long-acute but not ending in a long awn ...................... 6

6. Leaves 1-5—5-0 mm wide; flowers 10—50 mm in diameter; stigmas 5; fruit a dehiscent capsule .

COP e eee HEHEEEHEEHEHE HEHEHE EEE HEE EEE EEE HEHEHE HEHEHE HEHEHE HEHEHE EEE EEE EE HEHEHE

6. Leaves 5-15 mm wide; flowers 35-90 mm in diameter; stigmas 10-14; fruit fleshy and

indenisee nt (Carpabroms spp. )i03. 14 2esenreateee saad Twine uLane sua es 9

7. Leaves 15-70 x 2—5 mm; stigmas long-acute at apex ................. cee eeee 2. Ruschia caroli
7. Leaves 6-24 X 1-5—4-5 mm; stigmas plumose (Lampranthus spp.) ............cecceceeceeeeeees 8
8 Most stems and branches erect; leaves 10—16 X 2-0—2-5 mm in cultivation, rather dark bluish-
Sece me trates close wees eer. ete. Apel (hes ae. elon 3. Lampranthus roseus

8. Most stems and branches prostrate; leaves 6-13 X 1-5—2-5 mm in cultivation, taken as a whole
shorter and fatter, markedly pruinose ....................ccceeeeeeees 4. Lampranthus falciformis

Oo Petals yellow tiiged Pink, WHEN OlGy sad. 25... saccoe sss caccuas 9a. Carpobrotus edulis var. edulis
Oo) Petals mainly purples oranee-pinksonreddish-purple 242. ...2.c506cbe ss ec cncsecceeccseeteusseasedes 10
10. Petals pale to deep purple throughout .................. 9b. Carpobrotus edulis var. rubescens
10. Petals purple, orange-pink or reddish purple with obvious white or yellow bases ......... 11

ii>9(Petals pale-orange-pink, pale’ seddish purple or palé purpleis..: .9222.341 val, ATR
Behe a ates eM et 9a X b. Carpobrotus edulis inter var. edulis et var. rubescens
PATE AitES CL COPTIC ype «308 oe al Renia eld oraiend ss Mende cdee cosine vn sSeveducssvsscetsecscanesees 12
12. Leaves (20—)35—125 x 5—15 mm; base of petals yellow; filaments yellow with whitish bases
shohliaimerenes Sote).in. sisite eon, aed. A. 9c. Carpobrotus edulis var. chrysophthalmus
12. Leaves 20-70 x 5-8 mm; base of petals white; filaments whitish, sometimes with a little
ellowhakapexdics.tegr naw nett, ACO!) eigert. ss) ange 10. Carpobrotus glaucescens

ACCOUNT OF SPECIES

1. Aptenia cordifolia (L. fil.) Schwantes in Gartenflora 77: 69 (Feb. 1928); N.E.Br. in J. Bot., Lond.
66: 139 (May 1928).

Mesembryanthemum cordifolium L. fil., Suppl. 260 (1781). Described from the Cape of Good
Hope. NeotyPe: Salm-Dyck, Monogr. Aloes Mesembr. 61: fig. 1 (1842), designated here.

Litocarpus cordifolius (L. fil.) L. Bolus in Fi. Pl. S. Africa 7: sub. t. 261 (1927).

Vernacular names: Heart-leaf Ice-plant, Heart-leaf Mesembryanthemum.

Illustrations: Herre, Gen. Mesembr. 79 (1971); Salm-Dyck, Monogr. Aloes Mesembr. 61: fig. 1
(1842), pro Mesembryanthemum cordifolium; Bichard & McClintock, Wild Fl. Channel Is. no. 87
(1975).

Description: Rather succulent, short-lived perennial, but not infrequently behaving as an annual.
Stems up to 60 cm, pale yellowish-green, sometimes tinted brownish and angled, prostrate, much-
branched, covered with dense greyish papillae; internodes 1-2—9 cm. Leaves 10-35 x 10-30 mm,

222 GC. DS PRESTON-AND:P.4Di SELL

green, decussate, flat, more or less ovate, obtuse to subacute at apex, entire, more or less cordate,
truncate or cuneate at base, covered with greyish papillae; petiole up to 15 mm. Flowers solitary, 5—
16 mm in diameter, terminal at the end of branches or axillary; peduncles up to 13 mm, sometimes 5-
angled. Calyx green, covered with greyish papillae; lobes 4, the 2 outer 10-15 x 4-13 mm, ovate to
narrowly ovate, subobtuse to acute at apex, slightly to distinctly keeled and curved outwards, entire,
the inner 5—10 x 2—6 mm, lanceolate or linear-lanceolate, obtuse to acute at apex, entire, curved
outwards. Petals numerous, 3—7 mm, reddish-pink, linear or linear-oblanceolate, obtuse at apex,
united at the base into a short tube. Stamens numerous, arising from the corolla tube in a mass;
filaments 2—3 mm, whitish, glabrous; anthers cream. Staminodes numerous, 2—3 mm, whitish,
forming a ring between the stamens and the corolla tube. Stigmas 4, greenish, small, sessile.
Nectaries sunken in a hollow. Capsule 12—15 x 8-10 mm, green, 4-locular, obovoid, valves broader
than long, with the apex so abruptly thickened that the basal termination of the thickening is quite
vertical, middle of inner surface of the 4 valves with a bright yellow keel reaching from the central
column to the end of the flat part, each keel consisting of 2 parallel ridges, without marginal wings or
flaps, with 4 narrow strips arranged like a star between the central column and the point of origin of
keels; loculi roofs and tubercle absent; placentation axile. Seeds c. 40, 1-0—1-2 X 1-0-1-2 mm, dark
reddish-brown, subrotund or broadly ovate, compressed, tuberculate, with funicles up to 1 mm.
2n = 18.

Nomenclature and typification: There are no specimens in the Linnaean herbarium (LINN) and no
additional references in the protologue. The name has always been applied to the same taxon and
the Salm-Dyck illustration is chosen here as the neotype, and reproduced as Fig. 1.

Taxonomy and variation: Le Sueur (1984) has noted that in Jersey a single plant at Grouville
differed from a large population at St Brelade in colour, leaf shape and flower size. Otherwise little
variation is recorded in naturalized populations except in plant size. A variegated form is sometimes
grown in gardens.

Reproductive biology: Flowers from May to August. Reproduces by seed. Poorly grown plants are
said to behave as annuals, dying after they have produced seed, but we have no observations on the
reproductive biology of the species in the British Isles. Some, but not all, populations in Jersey
failed to survive the severe 1962-63 winter (le Sueur 1984).

Distribution and habitat: Naturalized on banks and stone wails on Bryher, St Martin’s and St Mary’s,
Isles of Scilly, and on Guernsey and Jersey. A large population in Jersey is naturalized under
coniferous trees. Recorded as a casual on rubbish tips in W. Kent and Herts.; it perhaps occurs
elsewhere in such localities. According to Higgins (1956), Aptenia was first introduced into British
gardens in 1774. It was recorded in Guernsey in 1928, but there were no subsequent records until
naturalized populations were discovered in Jersey in the early 1950s and on St Mary’s, Isles of Scilly,
in 1956. Although it is thoroughly established in some localities, Aptenia has not invaded semi-
natural habitats to the same extent that Carpobrotus and Disphyma have done. It is a native of the
eastern coastal districts of Cape Province, South Africa (Herre & Friedrich 1959).

2. Ruschia caroli (L. Bolus) Schwantes in Z. SukkulKde 3: 20 (1927).

Mesembryanthemum caroli L. Bolus in Ann. Bolus Herb. 3: 129 (1922). Described from living
plants in the Kirstenbosch Botanic Garden and in Dr C. Juritz’s garden at Three Anchor Bay,
Cape Town, South Africa. Not typified.

Description: Low, spreading shrub, up to 60 cm. Stems up to 60 cm, much branched from the base
upwards, greyish-brown or reddish-brown with darker brown elliptical spots, straight and ascending
or decumbent, irregularly striate or ridged, glabrous; internodes 1-3 cm. Leaves 15—70 x 2-5 mm,
rather unequal in size, more or less erect, bluish-green, faintly pink at apex, loosely decussate with
each pair with a connate base enclosed in a brown sheath, linear in outline, 3-angled, sometimes
recurved, obtuse or subacute at apex, sessile, covered with green spots and smaller white dots when
young, becoming irregularly muriform when mature. Flowers 1—4, 10-30 mm in diameter, terminal

~“

NATURALIZED AIZOACEAE Zee

\itt

FicureE 1. Aptenia cordifolia (L.fil.) Schwantes, illustrated as Mesembryanthemum cordifolium in Salm-Dyck’s
Monographia Generum Aloes et Mesembryanthemi 61: fig. 1 (1842). The coloured plate is the neotype of M.
cordifolium L. fil.

224 C. D. PRESTON AND P. D. SELL

at the end of branches; peduncles 10-25 mm. Calyx pale green; lobes 5, 3-5—4-0 x 2-0-2-5 mm,
triangular-ovate, 2 with a pale narrow margin and subacute at apex, 3 with a broad scarious flap and
a reddish subacute apex. Petals c. 30, 4-5—10-0 x 1-2 mm, free, purplish, with a darker central
longitudinal line and slightly whitish at base, narrowly oblanceolate, obtuse and entire at apex.
Stamens numerous, some outer without anthers; filaments 3-3-5 mm, purplish with white bases,
with numerous, pale, slightly clavate hairs at base; anthers yellow. Stigmas 5, green, narrowly
triangular-ovoid with a long acute apex and a few hairs near the base. Nectary glands united to form
a crenulate ring round the ovary. Capsule not seen in British material, but said to have much
diverging valve keels; loculi roofs present with a large tubercle; placentation parietal. We cannot
trace a description of the seeds. 2n = 18.

This description is based solely on plants from a naturalized population cultivated in Cambridge.
We have not had an opportunity to compare it with plants in the field.

Taxonomy and variation: Little variation has been recorded in our naturalized populations, but in
South Africa the species seems to produce much larger leaves than in Britain.

Reproductive biology: Flowers in April and May. We have no observations on the reproductive
biology of the species in the British Isles.

Distribution and habitat: Only recorded from the Isles of Scilly, where it is naturalized on walls on St
Mary’s and Tresco. The only record from semi-natural habitats is a note by Lousley on a specimen in
RNG that Ruschia occurs “‘in ‘wild’ places or rocks” at Carn Friars, St Mary’s, Isles of Scilly. Lousley
(1971) says that there is a specimen in K sent from Tresco Abbey in 1897, nearly 30 years before the
species was described. The first collection from a naturalized population was made in 1959. Ruschia
caroli is a native of Cape Province, South Africa.

3. Lampranthus roseus (Willd.) Schwantes in Reprium nov. Spec. Regni veg. 43: 230 (1938).
Mesembryanthemum roseum Willd., Enum. Pl. Horti Berol. 535 (1809). Described from the Cape
of Good Hope. Not typified.

Vernacular name: Rosy Dew-plant.

Illustrations: Marloth, Fl. South Africa 1: pl. 53A (1913); Salm-Dyck, Monogr. Aloes Mesembr. 29:
fig. 4 (1849), pro Mesembryanthemum roseum, Polunin, Flowers Med. t. 12 (1965).

Description: Low evergreen shrub. Stems up to 20 cm, decumbent to erect, 4-angled, glabrous,
brownish or purplish, much branched, the branches short, ascending or patent; internodes 0-3-—3-5
cm, sometimes with narrow wings on the angles. Leaves 10-16 X 2—2-5 mm, decussate just under
the flowers, becoming more numerous and dense lower down the fertile stems and branches and on
the sterile stems and axillary shoots, the terminal pairs on axillary sterile shoots often pressed
together giving a flattened appearance, dark glaucous green, mostly triangular in section with fairly
acute angles, more or less acute at apex and sometimes red-tipped, rounded at the sessile base with
each pair touching in their upper part and together encircling the stem, covered with dense pellucid
dots. Flowers 1-3 together, terminal or at the end of lateral shoots, 20-50 mm in diameter;
peduncles 12—20 mm. Calyx with dense pellucid dots; lobes 5, all red-tipped; 2 of the lobes 5—7 mm,
triangular-lanceolate, acute at apex, green (darker towards apex) with a scarious margin up to 0-5
mm wide in the lower half; remaining 3 lobes with a green part similar to the other 2 but with a deep
rusty orange, subapical, often asymmetrical appendage up to 3 mm wide on the inner side and
extending to the base where it is joined to that of the next lobe, the true apex of the lobe reflexed.
Petals numerous (over 100), very variable in length, 6-5—18-0 * 0-4—2-0 mm, free, shining light
purple or salmon pink sometimes on same plant, inner short and narrowly linear becoming
gradually larger to outer which are narrowly linear-oblanceolate, obtuse at apex, broader ones
slightly emarginate, with pale lines on back. Stamens numerous (100-150), more or less as long as
stigmas; filaments 3—4 mm, pale greenish-yellow; anthers yellow. Stigmas 5, greenish, outline more
or less triangular and formed from 3 plumose rays. Nectary glands united to form a crenulate ring

NATURALIZED AIZOACEAE 225

round the ovary. Capsule 6-7 x 7—8 mm, 5-locular, each valve with diverging keels and a
membranous wing; loculi roofs present, without a tubercle; placentation parietal. Seeds numerous,
1-0 x 1-2 mm, pale brown, obovoid, with shallow, obtuse tubercles. 2n = 36.

The above description is based on material cultivated at Cambridge from Bec du Nez, Guernsey,
and Lizard Head.

Taxonomy and variation: We find the taxonomy of Lampranthus difficult and unsatisfactory. The
two species accepted here looked quite different when grown side by side in the Cambridge Botanic
Garden, but their differences were very difficult to put into words. In the wild differences tend to be
masked by phenotypic variation. Thus the plant of L. roseus illustrated by Bichard & McClintock
(1975) seems to have the habit and leaves of L. falciformis. L. roseus shows considerable variation
in flower colour, purple and pink flowers occasionally occurring on the same plant.

Reproductive biology: Most Lampranthus species are thought to be self-sterile. L. roseus flowers
from late May to July. Seedlings have been seen at Portelet Bay, Jersey, at Beaucette and Bec du
Nez, Guernsey, and at Poldhu Cove. Some plants in the Lizard Point population were killed by frost
in the 1978-9 winter and all Lampranthus plants in Jersey were killed in the 1962-3 winter, but
reappeared from seed. Seed collected at Lizard Point, Cornwall and Bec du Nez, Guernsey
germinated freely.

Distribution and habitat: Naturalized on sea cliffs, coastal rocks and walls in several localities in
Guernsey and also collected from Alderney, Herm and Sark. In W. Cornwall naturalized on cliffs at
Lizard Point and Poldhu Cove. A Lampranthus naturalized on cliffs by Beau Port and on an
artificial scree of stone blocks at Portelet Bay, Jersey, is probably also referable to this species. The
earliest collection was made in Sark in 1924. Native of South Africa.

4. Lampranthus falciformis (Haw.) N.E.Br. in Gdnrs’ Chron., ser. 3, 87: 212 (1930).

Mesembryanthemum falciforme Haw., Syn. Pl. Succ. 299 (1812). Described from a specimen grown
in a London garden. NeoryPe: coloured illustration made between 1822 and 1835 from the plants
on which Haworth worked (K), designated here.

Vernacular name: Sickle-leaved Dew-plant.

Illustrations: Salm-Dyck, Monogr. Aloes Mesembr. 29: fig. 1 (1836), pro Mesembryanthemum
falciforme (reproduced here as Fig. 2).

Description: Low evergreen shrub. Stems up to 80 cm, mostly spreading and prostrate, but a few
erect, 4-angled, glabrous, greyish or brownish, much-branched, the branches short, ascending or
patent; internodes 0-2—1-7 cm, sometimes with narrow wings on angles. Leaves 6-13 x 1-5—2-5
mm, decussate just under the flowers, becoming more numerous and dense lower down the fertile
stems and branches and on the sterile stems and axillary shoots, the terminal pairs on the axillary
shoots often pressed together giving a flattened appearance, pale bluish-green with thick waxy
covering, half obovoid, flattened on one surface and rounded and keeled on the other making them
more or less triangular in section, more or less acute and Sometimes with a red dot at apex, semi-
connate at the sessile base, covered with dense pellucid dots. Flowers 1—2 together, terminal or at
the end of lateral shoots, 35-45 mm in diameter; peduncles 5—20 mm. Calyx with dense pellucid
dots; lobes 5, all red-tipped; 2 of the lobes 5.0—10.5 mm, triangular-lanceolate, acute at apex, green
with a scarious margin up to 0-5 mm wide in the lower half; remaining 3 lobes with a green part
similar to the other 2, but with a subapical, often asymmetrical appendage up to 3 mm wide on the
inner side and extending to the base where it is joined to that of the next lobe, the true apex of the
lobes reflexed. Petals numerous, very variable in length, 6-5—21-0 x 0-4—2-5 mm, free, pale pink,
the inner very pale, or purple, narrowly linear, becoming gradually larger to outer which are
narrowly linear-oblanceolate, obtuse at apex, broader ones slightly emarginate, with pale lines on
back. Stamens numerous, more or less as long as stigmas; filaments 3—4 mm, whitish, with a tuft of
whitish hairs at base; anthers pale yellow. Stigmas 5, greenish, outline more or less triangular and

226 C. D. PRESTON AND P. D. SELL

-~

i |

1 |e
Wa 7
7,
"i g

wy

aM

FiGure 2. Lampranthus falciformis (Haw.) N.E.Br., illustrated as Mesembryanthemum falciforme in Salm-
Dyck’s Monographia Generum Aloes et Mesembryanthemi 29: fig. 1 (1836).

NATURALIZED AIZOACEAE zat

formed from 3 plumose rays. Nectary glands united to form a crenulate ring. Capsule 6-7 x 7-8
mm, 5-locular, each valve with diverging keels and a membranous wing. Seeds numerous, 1-0—1-2
mm, pale brown, obovoid, with shallow, obtuse tubercles.

The above description is based on plants cultivated at Cambridge from Pembrokeshire and St
Mary’s, Isles of Scilly.

Nomenclature and typification: The obvious neotype is the coloured drawing at Kew made from
plants on which Haworth worked.

Taxonomy and variation: The only variation noted is in flower colour which varies from pale pink to
purple.

Reproductive biology: Most Lampranthus species are thought to be self-sterile. L. falciformis
flowers from May to July. It is very susceptible to frosts and there may be little or no flowering after a
bad winter, but some parts of the plant usually survive and soon regenerate vegetatively. At The
Gann, Pembrokeshire, it normally regenerates abundantly from seed. It seems to be immune from
grazing by cattle and rabbits (Davis 1978).

Distribution and habitat: Supposedly introduced into British gardens in 1805 (Higgins 1956). There
are specimens in K from Tresco Abbey Gardens in the Isles of Scilly collected in 1895. It is now
naturalized on St Mary’s, Tresco, St Martin’s, St Helen’s and Round Island in those islands, growing
on coastal rocks, cliffs and stone walls near the sea in shallow soil on ledges and in crevices. Both St
Helen’s (where it is particularly abundant) and Round Island are uninhabited and the Lampranthus
was presumably carried there by sea birds. It was collected on Lundy (K) as naturalized in 1908 and
1911. It was planted in a long disused quarry at The Gann, St Ishmael, Pembrokeshire in 1919 and 30
or more plants still occur on the almost perpendicular face of a natural rock outcrop of rhyolite. In
the 1920s or early 1930s it was introduced from this quarry to Dale Point, 2 km south of The Gann,
where a few plants were still present in 1978 (Davis 1978): Plants from Three Castle Head, W. Cork,
and on sandy cliffs at Roches Point, E. Cork, are probably referable to this species. Native of South
Africa.

5. Oscularia deltoides (L.) Schwantes in Mollers dtsch. Gart.-Ztg 43: 187 (1927).

Mesembryanthemum deltoides L., Sp. Pl. 482 (1753). Described from Africa. Lectotype: Dillenius,
Hort. Eltham. 255, t. 195, fig. 246 (1732), pro Mesembryanthemum deltoides, & dorso, & lateribus
muricatus , minus, in the copy in the Department of Plant Sciences, Oxford University, coloured
by Dillenius; designated here.

Lampranthus deltoides (L.) Glen ex D. O. Wijnands, Bot. Commelins 148 (1983), comb. invalid.,
basion. non cit.

Lampranthus deltoides (L.) Glen in Bothalia 16: 55 (1986).

Vernacular name: Deltoid-leaved Dew-plant.

Illustrations: Salm-Dyck, Monogr. Aloes Mesembr. 30: fig. 3 (1840), pro Mesembryanthemum
muricatum; Herre, Gen. Mesembr. 245 (1971); Lousley, Fi. Scilly 216, fig. 10 (1971).

Description: Low, sprawling shrub. Stems up to 50 cm, ascending or decumbent, glabrous, when
young bluish-green, heavily tinged reddish, soon becoming reddish- or orange-brown, older growth
woody, terete or slightly angled, glabrous, with short, erect or patent branches; internodes 0-2—2:-5.
cm, often with narrow wings on the angles. Leaves 6—18 x 3—10 mm, unequal or more or less equal
in size, pruinose with a pinkish-red apical point and apices of teeth, waxy, in decussate, clustered
pairs with each pair slightly connate at base, smooth, 3-angled, widest towards the shortly acute
apex, narrowed towards the sessile base, dentate-apiculate on the angles. Flowers 1—3 together, 12-
18 mm in diameter, terminal at the ends of branches, almond-scented; peduncles 2-3 mm. Calyx
yellowish-green; lobes 5, 3-0—3-5 x 2—3 mm, triangular-ovate, 2 with a pale, narrow margin and up
to 4 pinkish-red points at the apex, 3 with a broad scarious flap and a subapical, pinkish-red tipped

228 CzDiUPRESTON:AND?P.;DrSELL

point. Petals 40-50, 4-6 x 0-5-1-:5 mm, free, purplish, paler towards the base, narrowly
oblanceolate, obtuse and entire at apex. Stamens numerous, some outer without anthers; filaments
2-3 mm, white with rather pale simple eglandular hairs at base; anthers yellow. Stigmas 5, green,
narrowly ovoid with a curved apex. Nectary glands forming a crenulate ring round the ovary but in 5
distinct groups, dark green. Capsule 6—7 x 5—6 mm, 5-locular, obovoid; each valve with expanding
narrowly winged keels diverging from the base and half as long as the valves; loculi roofs present,
without a tubercle; placentation parietal. Seeds 1-0—1-2 mm, nearly black, ovoid or pyriform with a
nipple, minutely tuberculate, on a funicle up to 2 mm. 2n = 18.

The above description of flowers and fruit is based on plants from a naturalized population
cultivated in Cambridge. We have not had an opportunity to compare it with plants in the field.

Nomenclature and typification: Mesembryanthemum deltoides was described by Linnaeus (1753) in
Species Plantarum. His diagnosis Mesembryanthemum caulescens, foliis deltoidibus triquetris
dentatis is taken from the Hortus Cliffortianus (Linnaeus 1738) without change. There are no
specimens in the herbarium of the Hortus Cliffortianus (BM), nor in the Linnaean Herbarium in
London (LINN). Two synonyms follow the diagnosis: Mesembr. deltoides & dorso & lateribus
muricatis Dill. elth. 255. t. 195. f. 246 and Ficus aizoides africana erecta folio triangulari breviusculo,
fimbriato, floribus roseis odoratis Volk. hesp. 223. t. 224. f. 5. They are also cited in the Hortus
Cliffortianus. Both illustrations are referable to the plant usually called Oscularia deltoides, as is a
specimen in Dillenius’ herbarium (OXF). Glen (1986) cited the illustration in Hortus Elthamensis as
the iconotype and the specimen at OXF as the typotype, but did not select either as the lectotype.
More information can be obtained from the coloured illustration in Dillenius’ own copy of his work
than from the shrunken herbarium specimen so we select it as the lectotype. An uncoloured version
of this illustration, which is an excellent portrait of the plant, is reproduced here as Fig. 3.

Taxonomy and variation: British plants show little variation. Lousley (1971) questioned whether the
allied Oscularia caulescens (Miller) Schwantes occurs as a naturalized species in the Isles of Scilly,
but all the British plants seem to be O. deltoides. In fact, Glen (1986) considers that the three names
published in the genus Oscularia belong to the same species. Glen (1980) also thinks that Oscularia
and Lampranthus should be united and proposes that Lampranthus (1930) be conserved over
Oscularia (1927) as it is a much larger genus. The Committee for Spermatophyta (Taxon 32: 282
(1983)) recommended the acceptance of this proposal. Despite the fact that Glen (1980) says that his
monograph of Lampranthus is in press, it has not yet been published. He did not justify his transfer
of Oscularia to Lampranthus (Glen 1986), but merely cited an unpublished thesis in support of this
move. We are thus unable to assess his proposals and for the time being retain Oscularia as a
separate genus.

Reproductive biology: Flowers in April. Oscularia is a sprawling shrub, which can spread to cover
large areas of walls. Seedlings have been observed by G. D. Rowley (in litt. 1988) on walls on
Tresco, Isles of Scilly, and by J. R. Palmer on a wall in Guernsey (D. McClintock in litt. 1987).

Distribution and habitat: Naturalized in the Isles of Scilly on St Martin’s, St Mary’s and Tresco. It is
cultivated but not naturalized in the Channel Islands. It was introduced into British gardens before
1732, but not noted as a naturalized plant until Lousley collected it in Tresco in 1953. When Lousley
(1971) compiled his Flora, it was only known from walls. In 1982 it was found growing in small
quantity on a coastal rock outcrop, associated with Sedum anglicum, where it was still present in
1987. It will be interesting to see if it continues to spread into semi-natural habitats. It is a native of
the south-western part of Cape Province, South Africa (Herre 1971).

6. Disphyma crassifolium (L.) L. Bolus, F/. Pl. S. Africa 7: t. 276 (1927).
Mesembryanthemum crassifolium L., Sp. Pl. 484 (1753). Described from Africa. LECTOTYPE:
Dillenius, Hort. Eltham. 266, t. 201, fig. 257 (1732), pro Mesembryanthemum crassifolium repens,

229

NATURALIZED AIZOACEAE

"| sapioyap wnwayjunduquasay Jo 3d410199] dy} SI snIUdTIG Aq pomNojoo uOHRASNTE OY, “(ZELT SMUSTTIG) 9pZ “BY “S6T
‘COZ SISUaWUDYI]Y SNI4OH] UL Snunu ‘spon snqialy] 3 ‘OSsop 2 ‘saployap WNWaYyIUDKiquiasa- Se payeIsn[! ‘sayueMYIS ("T) Saployap DIAvINISO “€ AANOIA

“LI0U 0 OU
PORIILIMU LVMYILAIDE ) ko ty ee 7 Op

PL ay ge Lip ettd CH
IRIE ag

230 C2D/PRESTON AND |P. DS SEEL -

flore purpureo, in the copy in the Department of Plant Sciences, Oxford University, coloured by
Dillenius; designated here.

Vernacular name: Purple Dew-plant

Illustrations: Salm-Dyck, Monogr. Aloes Mesembr. 18: fig. 3 (1836), pro Mesembryanthemum
crassifolium; Lousley, Fl. Scilly 217, fig. 10 (1971).

Description: Succulent shrub. Stems up to 1 m, trailing or hanging, often rooting at the nodes,
glabrous, yellowish-green, tinted reddish or brownish to pale brown, rather woody especially at the
base; internodes 1-5—4-0 cm. Leaves in groups of 3-10, 12-42 x 3-9 mm, translucent yellowish-
green, faintly or heavily tinted pinkish, or reddish, usually blotched red, narrowly linear-
oblanceolate or linear-lanceolate in outline, triangular in section with obscure angles, rounded to a
minute point at apex, sessile, when in opposite pairs slightly connate at base, with a row of almost
flat papillae on the angles and covered with small pellucid dots. Flowers solitary, 30-50 mm in-
diameter, axillary; peduncles 4-40 mm. Calyx pale yellowish-green, sometimes tinted reddish at
base of lobes, covered with small pellucid dots; lobes 5, the 2 longest 8-15 X 4-7 mm, linear or
oblong, subacute at apex, the 3 shortest 6-5—10-5 x 4-0—5-5 mm, shortly oblong, obtuse at apex,
with a scarious margin (except at apex) up to 5 mm wide. Petals numerous (c. 50-60), 15—24 x 1-2
mm, free, on inside purple with white bases, on outside white at base and along central line with
purple margins, narrowly linear-oblanceolate, obtuse at apex. Stamens numerous, longer than the
styles; filaments 3-10 mm, white, with a tuft of white simple eglandular hairs at base; anthers pale
yellow. Stigmas 5, greenish, the basal 1-2 mm with a plumose inner surface, the apex with a long
awn up to 4mm. Nectary glands forming a crenulate ring round the ovary. Capsule c. 10 X 10mm, 5-
locular, pale yellowish, obovoid, each valve with a diverging keel with oblong wings; loculi roofs
present, with a pale bifid tubercle; placentation parietal. Seeds 0-8—0-9 x 0-7—0-8 mm, chestnut
brown, obovoid, surface irregular but smooth, on funicles c. 1 mm. 2n = 36.

Nomenclature and typification: Mesembryanthemum crassifolium was described by Linnaeus (1753)
in Species Plantarum. His diagnosis Mesembryanthemum caule repente semi-cylindraceo, foliis
semicylindricis laevibus connatis apice triquetris is taken direct from the Hortus Cliffortianus
(Linnaeus 1738) except that the words semicylindraceo and repente are transposed. There are no
specimens in the herbarium of the Hortus Cliffortianus (BM), nor in the Linnaean Herbarium in
London (LINN). Two synonyms follow the diagnosis: Mesembr. crassifolium repens, flore purpureo
Dill. elth. 266. t. 201. f. 257. and Ficoides africana reptans, folio triangulari, viridi, flore saturate
purpureo. Bradl. Succ. 4. p. 16. t. 38. They are also cited in the Hortus Cliffortianus. Both
illustrations are referable to the plant usually called Disphyma crassifolium. The illustration in the
first of these, in the copy coloured by Dillenius, is designated as the lectotype. An uncoloured
version is reproduced here as Fig. 4.

Taxonomy and variation: The living plants from the Isles of Scilly, mainland Cornwall, Sussex,
Anglesey and Jersey that we have studied show little variation. The variation in peduncle length,
from 4 to 40 mm, is phenotypic. Flowers on the exposed edge of a clone have short peduncles
whereas those arising amongst the longer leaves in the centre of the same clone have longer
peduncles.

Following Chinnock’s (1971) revision of the genus in Australasia five species were recognized,
two native to South Africa and three to Australasia. More recent work suggests that these might all
be conspecific (Venning 1984b). If this is so, however, some infraspecific taxa surely need to be
recognized. All naturalized British material belongs to D. crassifolium (L.) L. Bolus sensu stricto.
The population on Tresco, Isles of Scilly, reported as D. australe by McClintock (1964), is similar to
the other populations we have examined. A hybrid between Disphyma australe (Aiton) J. M. Black
and Carpobrotus edulis has been recorded from New Zealand (Chinnock 1972). Hybrids between
Carpobrotus and Disphyma could occur in Britain where the two genera grow together, but have not
yet been recorded.

“T unyofissps9 unWaYylUuDdiquiasa-Z JO 9d4}0}99] oY) SI sntuayiq Aq pamojoo uonensnqyt UL (ZELT sntusyfiq) LSz

Q ‘BY “1OZ 1 “997 SIsuauMYNY snjsopzy ut Oaandund a4oy ‘suadas winyofissp49 UNWIYIUDAIQuasaW Sk poyersnyyi ‘snjog’'y (7) wnyofissyia vuikydsig “> aan
’ fi 2 vp
-oameaimd at BY SUNMLAA Ut YO Ste MUNUIAYUD Male taf :
i L’ | : C 3
}

1S 2 \ in
ny ' 4 . B| R
} fs § i} 4 : >. ; ~\: j

NATURALIZED AIZOACEAE

232 C+D PRESTON AND P)’D. SEEL

Reproductive biology: Flowers in May and June. Viable seed is produced in Britain but we have not
seen seedlings in the wild. Colonies appear to have become established from planted material and
from stems discarded from cliff-top gardens onto the cliffs below. In cultivation the species roots
easily from cuttings. Effective reproduction of established colonies is mainly or entirely by
vegetative spread.

Disphyma is less susceptible to frost damage than other naturalized ‘mesembs’. In Jersey it
survived the 1962-63 winter “with ease” (le Sueur 1984). At Lizard Point, W. Cornwall, patches of
Carpobrotus edulis were severely damaged by frost in the 1978-79 winter, but adjacent patches of
Disphyma were unaffected. In the 1981—82 winter Carpobrotus at Cooden, E. Sussex, was damaged
but Disphyma was not (K. E. Bull in litt. 1982).

Distribution and habitat: Naturalized on St Mary’s and Tresco, Isles of Scilly, in W. Cornwall from
Kenneggy east to Coverack on the south coast and at Constantine Bay on the north coast (Margetts
& David 1981), and on Jersey, Channel Islands. It occurs on walls, sea cliffs, sand dunes and (in
Jersey) on a bank under pine trees and in salt marsh conditions behind a sea wall. It is also
naturalized on a sea cliff and coastal rocks north of Bwa Du on the west coast of Holy Island,
Anglesey, and on a damp, sheltered cliff at Cooden, E. Sussex. It has been found elsewhere as a
casual introduced with wool shoddy.

Disphyma was first recorded as a naturalized plant by J. E. Lousley in 1936 at New Grimsby,
Tresco, when it was even then said to have been present for as long as anyone could remember
(Lousley 1971) and where it still (1987) survives. It was noted on cliffs at the Lizard peninsula in 1951
and collected at Prussia Cove, Marazion, in 1954. Disphyma is now, except for Carpobrotus edulis,
the most thoroughly naturalized of all ‘mesembs’ and in places (e.g. Praa Sands and Lizard Point,
W. Cornwall) it occurs in abundance. It is not quite as effective a competitor as C. edulis and it tends
to be naturalized in greatest quantity on steep cliffs or shallow soil, where the cover of other
vegetation is reduced. Disphyma crassifolium sensu stricto is native to Cape Province, os Africa;
other (possibly conspecific) taxa are native to Australia and New Zealand.

7. Drosanthemum floribundum (Haw.) Schwantes in Z. SukkulKde 3: 29 (1927).

Mesembryanthemum floribundum Haw., Misc. Nat. 100 (1803). Described from plants grown in a
London garden. NrotyPe: Salm-Dyck, Monogr. Aloes Mesembr. 51: fig. 7 (1840), designated
here.

Mesembryanthemum candens Haw., Rev. PI. Succ. 186 (1821). NEotyPe: coloured drawing of a
vegetative plant (K), designated here.

D. candens (Haw.) Schwantes in Z. SukkulKde 3: 29 (1927).

Vernacular name: Pale Dew-plant

Illustrations: Salm-Dyck, Monogr. Aloes Mesembr. 51: fig. 7 (1840), pro Mesembryanthemum
floribundum.

Description: Low sprawling shrub forming dense prostrate to hanging patches. Stems up to 60 cm,
trailing, pale brown, with numerous, very short and short, thick, pale, usually deflexed, broad-
based simple eglandular hairs; branches usually weak, short and patent; internodes 0-5—3-0 cm.
Leaves 6-30 x 1-0-—3-5 mm, fleshy, decussate, dull pale to medium green, cylindrical or narrowly
obovoid, deltate-ovate in section, sometimes curved, more or less obtuse at apex, sessile, covered
with rounded, glass-like tubercles. Flowers 18-27 mm in diameter, solitary at the end of branches or
axillary; peduncles 2:-5—50-0 mm. Calyx hemispherical to slightly elongated, covered with glass-like
tubercles; lobes 5 (—6), 3-6 mm, lanceolate or linear-lanceolate, obtuse at apex, 3(—4) with broad
scarious flaps, 2 without flaps. Petals 25-35, 5-12 x 0-5—2-0 mm, free, shining white with a purple
apex when young, becoming tinted lilac and finally lilac with age, narrowly linear-oblanceolate,
more or less obtuse at apex, the inner much shorter than outer and sometimes bifid at apex. Stamens
30-70, from a little shorter to a little longer than styles; filaments 4-5 mm, whitish, with a group of
sub-basal pale simple eglandular hairs, becoming lilac after anthesis and curving outwards to expose
the stigmas and nectary glands; anthers pale yellow. Stigmas 5, linear in outline, curved outwards,

a

NATURALIZED AIZOACEAE 233

brownish-yellow, sometimes tinged purplish at apex. Nectary glands forming a ring round the ovary
but in disjunct groups. Capsule 2-5-5 x 4—6 mm, 5-locular, obovoid, each valve with 2 contiguous
expanding keels on inner face, each keel with a membranous wing; loculi-roofs present;
placentation parietal. Seeds c. 1 x 0.5 mm, brown, with ribs and tubercles. 2n = 36.

Nomenclature and typification: The specimens in Haworth’s herbarium have mostly been destroyed.
There are no coloured drawings of Mesembryanthemum floribundum at Kew based on Haworth’s
material. The only illustration cited with the original diagnosis, M. pilosum micans flore purpureo
pallidiore Dill. Elth. t. 214. f. 280, forms the basis for and is here designated as the lectotype of
another species, Mesembryanthemum torquatum Haw., Rev. Pl. Succ. 187 (1821). Thus there is no
original type material of M. floribundum. Fortunately the description is a good one and is in accord
with the plant we have described here. It is also in agreement with Salm-Dyck’s illustration
(reproduced here as Fig. 5) which we designate as the neotype.

Mesembryanthemum candens Haw. was described from vegetative material. There is a coloured
drawing of such a specimen at Kew, which does not in our opinion differ from M. floribundum, and
which we designate as the neotype. Another drawing at Kew shows a single pink flower which,
although not certainly the same as the vegetative plant, can be used as a guide, as it was probably
made from the same clone at a later date. It also does not differ from M. floribundum and we
conclude that the name should be placed into the synonymy of that species.

The lectotype illustration of Mesembryanthemum torquatum Haw. (Drosanthemum torquatum
(Haw.) Schwantes in Z. SukkulKde 3: 30 (1927)) looks rather like a pot-grown D. floribundum, but
we have seen no plants with such long peduncles.

Taxonomy and variation: Drosanthemum floribundum shows little variation in Britain. We have
studied living plants from the Isles of Scilly, the Lizard peninsula (W. Cornwall) and Guernsey and
all undoubtedly belong to a single taxon. The naturalized plant is referred to D. candens (Haw.)
Schwantes by Tutin (1962, 1964) but to D. floribundum by recent authors. D. candens is said to
differ from D. floribundum in its narrower calyx, paler pink or nearly white petals and stigmas not
exceeding the stamens (Adamson & Bolus 1950; Fernandes 1972). As described above, the flower
colour and the position of the stigmas relative to the stamens of the British plants varies with age and
photographs in CGE show white and pink flowers on the same plant.

Reproductive biology: Flowers from May to July. As its specific epithet implies, patches of
Drosanthemum usually flower freely. We have noticed flies, bees (including Bombus cf. lucorum
and B. terrestris) and ants visiting the flowers. Drosanthemum is more attractive to bees than any
other ‘mesemb’ naturalized in Britain, and in sunny weather they are frequent visitors. Seeds are
ripened in Britain, but we do not know if they are viable. We have not seen any seedlings in the wild
and effective reproduction is by vegetative spread. In cultivation cuttings of Drosanthemum root
less readily than those of Carpobrotus or Disphyma.

Drosanthemum is more susceptible to frost than Carpobrotus or Disphyma. Some colonies have
not survived severe winters, e.g. a well-established clone on a wall at Cadgwith Cove, W. Cornwall,
which was covered by a mass of flowers in July 1978, did not survive the severe 1978—9 winter.

_

Distribution and habitat: Naturalized on coastal walls, banks, rock outcrops and cliffs on St Mary’s
and Tresco, Isles of Scilly, at Rinsey Head and the Lizard peninsula, W. Cornwall, and on
Guernsey. The largest and most vigorous plants form patches hanging down vertical walls or cliffs.
Patches on less steep banks are smaller and less vigorous, and Drosanthemum is apparently less able
than Carpobrotus or Disphyma to spread vegetatively in such habitats. There is a single record from
Jersey, where it was killed in the 1962-63 winter (le Sueur 1984), and a plant collected as
Mesembryanthemum candens at Newquay, W. Cornwall, in 1921 (BM) was probably this species. It
is also recorded as a casual introduced with wool shoddy.

Specimens of Drosanthemum were collected at St Mary’s in 1875 (BM) and 1885 (OXF). These
were presumably cultivated plants. The first records of naturalized plants appear to be the 1921
collection from sand hills near Newquay mentioned above and a plant collected by T. G. Tutin in
1948 from a cliff at Cadgwith, Lizard peninsula (LTR). None of the pressed specimens can be

234 C. D. PRESTON AND P. D. SELL

\W

Ficure 5. Drosanthemum floribundum (Haw.) Schwantes, illustrated as Mesembryanthemum floribundum in
Salm-Dyck’s Monographia Generum Aloes et Mesembryanthemi 51: fig. 7 (1840). The coloured plate is the
neotype of M. floribundum Haw.

NATURALIZED AIZOACEAE 235

identified as D. floribundum with certainty but all probably represent the same species as that
currently naturalized and we have confirmed the identity of the Cadgwith plant in the field.

Drosanthemum is less thoroughly naturalized than Carpobrotus , Disphyma or Lampranthus. It is
probably limited by its frost sensitivity and low competitive ability. It does occur, however, in
natural habitats, although often only in small quantity, and will probably persist as a member of the
British flora. It is a native of South Africa (Herre & Friedrich 1961).

8. Erepsia heteropetala (Haw.) Schwantes in Gartenflora 77: 68 (1928).

Mesembryanthemum heteropetalum Haw., Misc. Nat. 67 (1803). Described from plants cultivated in
a London garden. NeotypPe: Salm-Dyck, Monogr. Aloes Mesembr. 21: fig. 2 (1840), pro
Mesembryanthemum heteropetalum, designated here.

Illustrations: Lousley, Fl. Scilly 217, fig. 10 (2) (1971); Salm-Dyck, Monogr. Aloes Mesembr. 21: fig.
2 (1840), pro Mesembryanthemum heteropetalum.

Description: Low, glabrous, fleshy shrub. Stems up to 35 cm, erect or decumbent, glabrous,
yellowish-green or tinted brownish, often spotted purplish, slightly flattened with decussate ridges
or wings; internodes up to 4-5 cm. Leaves 16-30 x 6-13 mm, with bluish-green bloom on faces and
yellowish-green angles and covered with dense pellucid dots, unmarked when young, becoming
tinted or spotted purplish and eventually turning purple with age, decussate, asymmetrically
oblanceolate in outline, triangular in section, shortly acute at apex, sessile, irregularly dentate on
one angle, often minutely denticulate on the others. Flowers solitary or 2—3 together, terminal at the
end of branches and side shoots, 10-15 mm in diameter; peduncle up to 15(—25) mm. Calyx bluish-
green, hemispherical, covered in pellucid dots, sometimes spotted or tinted purplish; lobes 5, erect;
2 longest lobes 11-15 mm, broadly triangular-ovate, abruptly acute to more or less obtuse at apex,
entire, both usually keeled on the back, the keel continuing to the base of calyx and entire or
irregularly dentate below, rarely only 1 keeled; 3 shortest lobes 8—15 mm, triangular-ovate, acute at
apex, entire, not keeled, 2 with a scarious often purplish-tinted margin on both sides and one with a
similar margin on one side. Petals numerous, in several irregular series, 4-11 X 0-5—0-8 mm, free,
pale purple, linear-oblanceolate, obtuse and sometimes bifid at apex, outer petals more or less
spreading, inner erect or inflexed. Staminodes numerous, 2-:0—4-5 x 0-1—0-2 mm, whitish, linear,
more or less obtuse at apex, inner inflexed and covering the stamens. Stamens numerous, inflexed;
filaments 0-5—2-0 mm, white; anthers pale yellow. Stigmas 5—6, very small, pale greenish, roundish
to almost globular in outline. Nectary glands minute and scarcely visible. Capsule 10-15 x 10-15
mm, 5-locular, each valve with two expanding keels which diverge towards the apex and end in
broad awn-tipped membranes; loculi roofed with stiff wings, but without tubercles at the openings;
placentation parietal. Seeds 1-3—1-5 x 0-8—1 mm, chestnut-brown, ovoid, flattened, tuberculate, on
a funicle up to c. 1 mm. 2n = 18.

Nomenclature and typification: Haworth’s original description of Mesembryanthemum heteropeta-
lum is a good one and fits our plant except that the petals are said to be white. There is a coloured
drawing of a plant worked on by Haworth at Kew but as it is only a vegetative shoot we have not
selected it as a neotype, preferring instead Salm-Dyck’s illustration of a flowering plant. This is
reproduced here as Fig. 6.

Taxonomy and variation: There is very little variation within the only population naturalized in
Britain.

Reproductive biology: Flowers in May and June. The morphology of the flower suggests that it is
self-pollinating. We have not seen flying insects visit the flowers (although they are visited by ants)
and Erepsia is certainly much less frequently visited by insects than Carpobrotus or Drosanthemum.
Reproduction in the wild is by seed: viable seed is produced and seedlings have been observed. We
have no evidence of any vegetative reproduction.

236 C. D. PRESTON AND P. D. SELL

FiGuRE 6. Erepsia heteropetala (Haw.) Schwantes, illustrated as Mesembryanthemum heteropetalum in Salm-
Dyck’s Monographia Generum Aloes et Mesembryanthemi 21: fig. 2 (1840). The coloured plate is the neotype of
M. heteropetalum Haw.

|

NATURALIZED AIZOACEAE 237

Distribution and habitat: Only naturalized in one locality, Buzza Hill, St Mary’s, Isles of Scilly,
where it grows in shallow soil on the edge and face of a disused quarry and on a nearby cliff slope. It
was first collected by J. E. Lousley in 1957, and was still present in 1987. In 1983 the population
consisted of approximately 30 plants. Erepsia heteropetala, unlike the more showy ‘mesembs’, is not
widely grown in coastal gardens in south-western England. It is a native of South Africa.

9. Carpobrotus edulis (L.) N.E.Br. in E. P. Phillips, Gen. S. Afr. Fl. Pl. 249 (1926).

Mesembryanthemum acinaciforme var. flavum L., Sp. Pl. 485 (1753). Lecrotype: Dillenius, Hort.
Eltham. 284, t. 212, fig. 272 (1732), pro Mesembryanthemum falcatum majus, flore amplo luteo,
chosen by Blake in Contr. Queensland Herb. 7: 17 (1969); the copy in the Department of Plant
Sciences, Oxford University, is coloured by Dillenius.

_ M. edule L., Syst. Nat., 10th ed., 1060 (1759), nom. nov. pro M. acinaciforme var. flavum L.

Vernacular names: Hottentot Fig, Kaffir Fig, Sally-my-handsome.
Illustrations: Ross-Craig, Draw. Br. Pl. 11: 39 (1958).

Description: Prostrate or hanging, succulent perennial forming large mats. Stems up to at least 3 m,
trailing or hanging, sometimes rooting at the nodes, fleshy, angled, glabrous, slightly bluish-green
turning pink, pinkish-purple or orange with age; branches axillary; internodes 1-5—7-0 cm. Leaves
(20—)40-140 x 5-15 mm, decussate, each pair slightly connate at base, slightly bluish-green and
waxy, turning pinkish-purple and finally orange with age, covered with pellucid dots, linear or
narrowly oblong in outline, triangular in section, straight to more or less curved, obtuse to acute at
apex, sessile. Flowers terminal or at the end of side shoots, solitary, 40-90 mm in diameter;
peduncles 6—60 mm. Calyx yellowish-green, sometimes tinged purplish, covered with pellucid dots;
lobes 4—6, the 2 longest 15-65 x 8—17 mm, lanceolate to linear-lanceolate in outline, more or less
acute at apex, flat on inner surface, keeled on outer surface, the keel continuing down the undivided
calyx, the 2—4 short lobes 10—50 x 6—15 mm, ovate to lanceolate in outline with a curved, acute apex
and a brownish scarious flap on one or both sides up to 10 mm wide, that on the shortest of the
shorter lobes sometimes almost circular, exceeding the apex and adnate to its inner surface, one of
the longest lobes sometimes with a flap and one of the shorter lobes sometimes without a flap. Petals
65-150, 25-45 x 1-5-3-5 mm, free, yellow or pale or deep purple, pale reddish-purple or orange-
pink, sometimes with a yellow base, narrowly linear or narrowly linear-oblanceolate, entire or
slightly toothed at apex. Stamens numerous; filaments 5-10 mm, yellow or brownish-orange,
sometimes with a whitish base, inner with a tuft of whitish or purplish hairs at base; anthers yellow.
Staminodes resembling petals sometimes present between stamens and petals. Stigmas 10-14,
greenish-yellow, curved upwards and outwards, plumose on the inner surface. Nectary glands
united to form a crenulate ring round the ovary. Fruit adnate to the calyx, 1-4—3-0 x 1-4—2-7 cm,
obovoid; seeds 1-1-5 x 0-8-1 mm, dark brown when ripe, obovate in outline, flat, minutely and
faintly reticulate, on a funicle 2-3 mm.

a. var. edulis

Illustrations: Curtis’s bot. Mag. 144: t. 8783 (1918), pro parte; Sj6gren, Acores Flores t. 73 (1984).

Description: Leaves (20—)40-140 x 5-14 mm. Flowers 40—90 mm in diameter; petals pure yellow
when freshly opened, becoming pink-tinged with age. Fruits 2-5—3-0 x 2-5—2-7 cm. 2n = 18.

b. var. rubescens Druce in Rep. botl Soc. Exch. Club Br. Isl. 7: 771 (1926). Not typified (see below).
Mesembryanthemum virescens auct., non Haw.

M. aequilaterum auct., non Haw.

C. deliciosus auct., non L. Bolus.

Illustrations: Curtis’s bot. Mag. 144: t. 8783 (1918), pro parte; Bichard & McClintock, Wild FI.

_ Channel Is. no. 67 (1975).

238 C: Dz: PRESTON: AND P. D. SELL

Description: Leaves (20—)40-100 x 5-10 mm. Flowers 45-75 mm in diameter; petals purple with
little or no sign of yellow at their bases. Fruits 2-5—3-0 x 2-5-2-7 cm. 2n = 18.

c. var. chrysophthalmus C. D. Preston & P. D. Sell, var. nov. HoLorype: Plant collected as living
material at the sea front, Porthloo, St Mary’s, Isles of Scilly, 12 January 1978, C. D. Preston,
cultivated at the Botanic Garden, Cambridge, and pressed on 8 June 1979 as P. D. Sell 79/22
(CGE).

C. acinaciformis sensu Lousley, Fl. Scilly 145 (1971).

Illustration: Lousley, Fl. Scilly dustcover (1971) ut C. edulis.

Description: Folia (20—)35—125 mm longa, 5—15 mm lata. Flores 45-70 mm in diametro; petala
purpurea basi area aurei; fructus 1-4—2-5 x 1-4—-1-8 cm.

Leaves (20—)35-125 x 5-15 mm. Flowers 45-70 mm in diameter; petals purple with a distinct area
of yellow at their bases, which forms a ring as you look into the flower from above. Fruit 1-4—2-5 x
1-4—1-8 cm.

Nomenclature and typification: Linnaeus (1753) described Mesembryanthemum acinaciforme vat.
flavum based on Mesembr. falcatum majus, flore amplo luteo. Dill. elth. 284. t. 212. f. 272. and
Ficoides s. Ficus aizoides africana major procumbens, triangulari folio, fructu maximo eduli. Herm.
lugd. 244. t. 245. Both illustrations are referable to what is now known as Carpobrotus edulis. We
are in agreement with Blake (1969) who selected the Dillenius plate as the lectotype, but in
particular we cite the coloured copy at Oxford. An uncoloured version is reproduced here as Fig. 7.
Linnaeus (1759) gave M. acinaciforme var. flavum a new name, M. edule. The type of M. edule is
thus the type of M. acinaciforme var. flavum.

The purple-flowered variant, var. rubescens, was described by Druce in 1926. No specimens are
cited in the protologue, and none of the specimens in Druce’s herbarium (OXF) is labelled var.
rubescens or described as C. edulis with red- or purple-flowers. G. H. Douglas sent plants to Druce
in 1924, stating ‘‘the flowers are of two colours’’. In the poorly dried specimens one can no longer see
any colour, but it may be that chemical analysis could distinguish them. At the moment we are
unable to lectotypify var. rubescens. M. aequilaterum Haw., M. virescens Haw. and Carpobrotus
deliciosus (L. Bolus) L. Bolus have all been wrongly applied to this taxon. C. aequilaterus (Haw.)
N.E. Br. is asmaller-flowered Australian species more akin to C. glaucescens, but without the white
bases to the petals (cf. Blake 1969). For C. virescens (Haw.) Schwantes see under C. glaucescens.
Lousley (1971) thought that specimens sent in 1895 from Tresco-Abbey for a drawing in Curtis’s
botanical Magazine (t. 8783) were referable to C. deliciosus (L. Bolus) L. Bolus. The pink-flowered
specimen on this plate, however, seems to us to be indistinguishable from C. edulis var. rubescens,
and furthermore the text states that the plants illustrated were collected at Caerthillian Valley,
which is on the Lizard peninsula, mainland Cornwall (cf. Hutchinson 1917).

The third variety, var. chrysophthalmus, has the bases of the purple petals yellow, so that when
looking into the open flowers one can see a golden ring round the base. This seems to be the plant
Lousley (1971) called C. acinaciformis, although he did not mention the flower character.
Mesembryanthemum acinaciforme was described by Linnaeus (1753). The diagnosis Mesembryan-
themum foliis acinaciformibus connatis: angulo carinali scabris, ramis angulatis is taken verbatim
from the Hortus Cliffortianus (Linnaeus 1738). There are no specimens in the herbarium of the
Hortus Cliffortianus (BM), or in the Linnaean Herbarium in London (LINN). The diagnosis is
followed by var. purpureum based on Mesembr. acinaciforme, flore amplissimo purpureo Dill. elth.
282. t. 211. f. 270 & t. 212. f. 271. This is also given in the Hortus Cliffortianus, but without a varietal
name. It is clearly what Linnaeus considered to be the type variety. The plate t. 211 fig. 270 in the
Oxford coloured copy is therefore designated as the lectotype of M. acinaciforme and shows a large-
flowered, large-leaved plant which is not matched by any material from the British Isles which we
have seen. Jt was apparently cultivated in the Isles of Scilly as the illustration in Curtis's bot. Mag. t.
5539 (1865) seems to be correctly named. We cannot find any described taxon which matches our
plant with yellow centres to the flowers and have therefore described it as new, although we have no
suggestions as to its geographical origin or derivation. The holotype is taken from the clone we have

NATURALIZED AIZOACEAE 239

a le Cfe nb) yanlhe 7eLe? Nola ie 1 MALES ,
“fore v anyslo Lileo

Figure 7. Carpobrotus edulis (L.) N.E.Br. var. edulis, illustrated as Mesembryanthemum falcatum majus, flore
amplo luteo in Hortus Elthamensis 284, t. 212, fig. 272 (Dillenius_1732). The illustration is the lectotype of
Mesembryanthemum edule L.

most studied, and there are colour photographs in CGE of the plant we pressed (Sell colour film 297/
ay Sl & 32).

Taxonomy and variation: Variation within var. edulis is described by le Sueur (1984) from Jersey,
where neighbouring clones differ in leaf and flower colour. Var. rubescens differs only in its smaller
purple flowers (though vegetatively it never seems to reach the maximum. size of the yellow-
flowered plant). Both these varieties are widely naturalized and often grow together. Intermediate
plants with pale purple, pale reddish-purple or pale orange-pink flowers occur, and look as though
they have been derived from hybridization between yellow- and purple-flowered plants. As their
origin is obscure we have treated the yellow- and purple-flowered plants as varieties of the same
species under the only names that seem to be available for them.

240 Ci Dz PRESTONIAND: RSDOSELL

Var. chrysophthalmus has slightly smaller flowers, like var. rubescens but with the base of the
petals yellow. It also has smaller fruits. The only published illustration we are aware of is on the
dustcover of Lousley (1971), which most libraries will have thrown away. We can find no name for
it. We originally thought it was a distinct species, as species go in ‘mesembs’, but have finally decided
to name it as a variety of C. edulis as it is just possible that it arose from hybridization between the
yellow and purple variants.

Reproductive biology: The main flowering period is from May to July. The flowers are visited by
Coleoptera, Diptera, Hymenoptera and Lepidoptera, and snails can sometimes be seen sitting in
the open flowers. Viable seed is produced and seedlings have been seen, usually growing in areas of
bare soil around the fringe of naturalized clones.

Carpobrotus edulis spreads mainly by vegetative means. In the Isles of Scilly and the Channel
Islands it has been planted for dune stabilization or for ornament, and elsewhere colonies often
originate from stems thrown out of gardens. New colonies become established from fragments
collected by gulls as nesting material, and in this way Carpobrotus has spread even to small
uninhabited islands. Fragments of C. edulis root readily in cultivation: one potted up in an unheated
greenhouse produced roots up to 22 cm long in 18 days.

Large areas of C. edulis overwhelm native vegetation, forming mats in which virtually no other
species grow. These mats acidify the underlying soil down to a depth of at least 8-12 cm (Frost
1987). The spread of Carpobrotus is accelerated by cliff fires, which scarcely affect the succulent
‘mesemb’ but destroy the surrounding vegetation. In severe winters C. edulis is badly damaged by
frost. However, stems may survive and produce new shoots even in areas in which all the existing
foliage has been killed. A thick layer of litter accumulates under mats of Carpobrotus and this must
help insulate stems from low temperatures, and certainly inhibits the colonization by other
flowering plants of areas in which Carpobrotus has been frost-killed. Cattle sometimes eat the leaves
of Carpobrotus, but only a few colonies grow in localities where such grazing is likely to occur. Frost
(1987) reports that it is eaten “with relish” by pet rabbits.

Distribution and habitat: C. edulis is primarily a species of sea cliffs and sand dunes, but it also grows
on coastal banks, rocks and walls. Its distribution is mapped in Fig. 8. Var. edulis and/or var.
rubescens is recorded from the following vice-counties: Channel Islands, 1—4, 9-11, 14, 15, 19, 41,
45, 49, 52, 60, 71, H1, HS, H20, H21, H38. The population in v.c. 74 mapped in the Atlas of the
British flora (Perring & Walters 1962) is C. glaucescens and confirmation of the identity of the plant
recorded on shingle in v.c. 19 by Jermyn (1974) is desirable in view of the presence of C. glaucescens
on the adjacent Suffolk coast. Var. chrysophthalmus is naturalized on Bryher and St Mary’s in the
Isles of Scilly and at St Brelade and St Ouen’s Bay in Jersey. Lousley (1971) records “C.
acinaciformis” — the name he apparently applied to var. chrysophthalmus — from St Martin’s and
Tresco. C. D. P. and P. Clough (formerly of Tresco Abbey Gardens) have searched for it on Tresco
without success.

Carpobrotus edulis was introduced into British gardens about 1690. The history of its naturaliza-
tion is difficult to reconstruct. For many years botanists did not record naturalized populations of
this and other ‘mesembs’, and collectors of herbarium specimens sometimes fail to state whether or
not they were collected from naturalized populations. C. edulis was recorded as ‘‘semi-wild” in
Guernsey in 1886 and reported again in 1893 (McClintock 1975). A yellow-flowered ‘mesemb’,
presumably C. edulis var. edulis, grew freely in great abundance at St Aubin’s Bay, Jersey by 1886
(le Sueur 1984). In the Isles of Scilly it was planted in the mid-nineteenth century and by 1921 it was
common on most of the islands (Lousley 1971). On mainland Cornwall ‘mesembs’ were in
cultivation at Falmouth by 1871 (Baker 1871) and at The Lizard by c. 1880 (Frost 1987). Although it
was not listed amongst the alien plants seen near Penzance between 1885 and 1887 by Glasson
(1889), C. edulis was ‘“‘abundant”’ in the Lizard peninsula in 1905 and “‘quite naturalised” in 1909. It
was “‘rapidly establishing itself” at East Pentire, Newquay, in 1907 and by 1909 it had been recorded
in several Cornish localities (Vigurs 1908; Davey 1909). Eight years later Hutchinson (1917) could
describe it as ‘‘now so thoroughly naturalised and so characteristic a feature of the sea-slopes of
Cornwall”. In Devon it was collected at Torquay (v.c. 3) in 1885 and Saunton Sands (v.c. 4) in 1919.
It is a native of South Africa.

NATURALIZED AIZOACEAE 241

“s g
| ° es
| +
| C 0
| F oe
- a
| L on
a aes |
| te 4
it iesrcae
eae ee he
|
[a ==
|
| |
'
| |
1 !
|
| !
Mi SSS ls |
| |
a ee eee ee |

—

Ficure 8. The distribution of Carpobrotus edulis in the British Isles. The records are plotted in the 10-km squares
of the British and Irish national grids; the Channel Islands are shown as an inset with the records plotted on the
UTM grid.

10. Carpobrotus glaucescens (Haw.) Schwantes in Gartenflora 77: 69 (Feb. 1928); N.E. Br. in J.
| Bot., Lond. 66: 324 (Nov. 1928).
| Mesembryanthemum glaucescens Haw., Syn. Pl. Succ. 236 (1812). Described from plants from
Nova Hollandia grown in a London garden. NeEoryPe: cultivated plant originating from Botany
Bay, S. T. Blake 19677 (BRI), vide Blake in Contr. Queensland Herb. 7: 23 (1969).
Mesembryanthemum virescens Haw., Syn. Pl. Succ. 236 (1812). Described from plants from Nova
Hollandia grown in a London garden. NeotyPe: King George Sound, West Australia, Sept. 1959,
S. T. Blake 20910 (BRI), vide Blake in Contr. Queensland Herb. 7: 25 (1969).

242 C.D? PRESTON AND P2 Dy SELL

Mesembryanthemum abbreviatum Haw. in Phil. Mag. 68: 329 (1826). Described from a plant from
Australasia grown at Kew. NgotypPe: Salm-Dyck, Mongr. Aloes Mesembr. 5: fig. 7 (1849), vide
Blake in Contr. Queensland Herb. 7: 25 (1969).

Mesembryanthemum edule var. virescens (Haw.) Moss, Cambr. Brit. Fl. 2: 151 (1914), quoad
basionym, exclud. descript.

C. virescens (Haw.) Schwantes in Gartenflora 77: 69 (1928).

C. abbreviatus (Haw.) Schwantes in Gartenflora 77: 68 (1928).

Vernacular name: Angular Pigface is the name used for this species in Australia.
Illustrations: Bichard & McClintock, Wild Fl. Channel Is. no. 68 (1975).

Description: As C. edulis except: Internodes 0-5—3-0 cm. Leaves 20-70 x 5—8 mm, bluish-green and
waxy, more or less acute at apex. Flowers 35—45(—60) mm in diameter; sessile or with a peduncle up
to 10 mm. Calyx bluish-green and waxy; lobes 5, the 2 longest 15—20 x 10-12 mm, the 3 shortest 5—
13 x 3-15 mm, ovate in outline with a brownish scarious flap on one or both sides up to 8 mm wide.
Petals 55—80, 17-25 x 1-0—1-2 mm, purple with obvious white bases, narrowly linear, with a few
teeth at the apex. Stamens numerous; filaments 4-6 mm, white, sometimes yellowish towards apex,

inner sometimes with tufts of white simple eglandular hairs at base. Stigmas 10-12. Fruit 1- a5 1-7 X
1-2—1-5 cm; seeds 1-0-—1-2 X 0-8-1-0 mm. 2n = 18.

Nomenclature and typification: This has been satisfactorily dealt with by Blake (1969). There are in
fact coloured drawings of both C. glaucescens and C. virescens at Kew not seen by Blake. We can
detect no obvious difference between the drawings. C. chilensis (Molina) N.E.Br., the name used
for the Suffolk populations of this plant by Simpson (1982), is a synonym of C. aequilaterus, a
species which lacks white bases to the petals (Blake 1969).

Taxonomy and variation: Plants naturalized in the British Isles show little variation. Blake (1969)
separated C. glaucescens from C. virescens mainly on stamen number and ascribed to them different
geographical ranges, C. glaucescens being confined to coastal areas of eastern Australia and C.
virescens to the south-west. Venning (1984a) retained the two species, but said that dried material
was difficult to name and that not enough fresh material was available to review their status. In the
few naturalized British plants in which we have counted the stamens the number is around 300 which
is intermediate between the two. We therefore believe that they are not distinct species, but that
there may be slight differences which could allow the upkeep of geographical races.

Reproductive biology: The main flowering period is from May to July. Flowers have, however, been
seen as early as February and C. glaucescens shows a greater tendency than C. edulis to flower
outside the main flowering period. Effective reproduction is by vegetative growth and spread. In
cultivation vegetative fragments root readily. C. glaucescens is a smaller plant than C. edulis, and
does not possess as great a competitive vigour as that species.

Distribution and habitat: Naturalized in the Channel Islands (Alderney, Guernsey, Herm and
Jersey), Suffolk and Wigtownshire. On Herm it is the only naturalized Carpobrotus but in Guernsey
and Jersey it is much less frequent than C. edulis. In Suffolk it is recorded from two localities near
Felixstowe: it is the C. chilensis of Simpson (1982). In Wigtownshire it grows around the Logan Fish
Pond at Port Logan. Most naturalized colonies are on sea cliffs, but it also grows on walls and
roadside verges. In Australia C. glaucescens is recommended for planting as a stabilizer of sand
dunes (Beach Protection Authority of Queensland 1981) but it has not been recorded from this
habitat in Britain. It is cultivated in the Isles of Scilly but has not become naturalized there.

C. glaucescens has been known for about 25 years in Jersey (le Sueur 1984). It was first noticed in
Herm in 1947, in Guernsey in 1970 and in Suffolk in 1974. Although first officially recorded in
Wigtownshire in 1955 (as C. edulis), it is said by local people to have grown there for many years
before that. It is a native of the eastern coast of Australia (Venning 1984a).

NATURALIZED AIZOACEAE 243

ACKNOWLEDGMENTS

Mrs F. le Sueur (Jersey), D. McClintock and Mrs P. Ryan (Guernsey), P. Clough and Mrs R.
Parslow (Scilly), L. J. Margetts (Cornwall), K. E. Bull (Sussex), R. H. Roberts (Anglesey), E. F.
Greenwood (Lancashire), Dr H. A. Lang (Wigtownshire) and P. Hackney (Co. Down) have kindly
helped us by providing information on the distribution of ‘mesembs’ or by sending living material.
We have benefitted from discussion with J. R. Akeroyd, currently revising the account of Aizoaceae
in Flora Europaea. C. Jarvis checked his records to see if the Linnaean taxa had been lectotypified,
Miss S. K. Marner answered queries about books and specimens at Oxford, Mrs D. M. Greene
plotted the distribution map and Miss J. M. Abblitt turned our initial palimpsest into a typescript.
A. O. Chater, L. C. Frost, D. McClintock and G. D. Rowley kindly commented on a draft of the
paper, and R. A. Finch checked the Appendix.

REFERENCES

Apamson, R. S. & Botus, L. (1950). Aizoaceae, in ADAMSON, R. S. & SALTER, T. M. Flora of the Cape
Peninsula, pp. 363-390. Cape Town and Johannesburg.

Baker, J. G. (1871). On the botany of the Lizard peninsula. J. Bot., Lond. 9: 353-358.

BEACH PROTECTION AUTHORITY OF QUEENSLAND (1981). Coastal dune management. Brisbane.

BETJEMAN, J. (1945). New bats in old belfries. London.

BICHARD, J. D. & McC.intock, D. (1975). Wild flowers of the Channel Islands. London.

BLAKE, S. T. (1969). A revision of Carpobrotus and Sarcozona in Australia, genera allied to Mesembryanthe-
mum (Aizoaceae). Contr. Queensland Herb. 7: 1-65.

Brown, N. E. (1920). New and old species of Mesembryanthemum, with critical notes. J. Linn. Soc. (Bot.) 45:
53-140. x

CHINNOCK, R. J. (1971). Studies in Disphyma — a genus related to Mesembryanthemum. 1. A revision of
Disphyma australe (Ait.) J. M. Black. N. Z. J. Bot. 9: 331-344.

CHINNOCK, R. J. (1972). Natural hybrids between Disphyma and Carpobrotus (Aizoaceae) in New Zealand.
N. Z. J. Bot. 10: 615-625.

Davey, F. H. (1909). Flora of Cornwall, p. 204. Penryn.

Davis, T. A. W. (1978). Lampranthus falciformis. B.S.B.I. News 19: 17.

DILLeNtus, J. J. (1732). Hortus Elthamensis. London.

Epwarps, P. I. (1968). Sloane manuscript 5286; an important source for Decade 9, ‘Herbarium Capense’ of
Petiver’s Gazophylacium Naturae and Artis. J. S. Afr. Bot. 34: 243-253.

FERNANDES, R. B. (1972). Mesembryanthemaceae. Anu. Soc. broteriana 38: 127-136.

Frost, L. C., ed. (1987). The alien Hottentot-fig (Carpobrotus edulis) in Britain — a threat to the native flora and
its conservation control. University of Bristol Lizard Project Report no. 22.

Futter, T. C. & Barse, G. D. (1981). A microwave-oven method for drying succulent plant specimens. Taxon
30: 867.

Giasson, W. A. (1889). On the occurrence of foreign plants in West Cornwall. Rep. Trans. Penzance nat. Hist.
Antiq. Soc. 1888-1889: 62-69.

GLEN, H. F. (1980). Proposal to conserve the generic name 2405 Lampranthus N.E. Br. (1930) against Oscularia
Schwantes (1927) (Mesembryanthemaceae). Taxon 29: 693-694.

GLEN, H. F. (1986). A new combination in Lampranthus. Bothalia 16: 5S.

Gunn, M. & Copp, L. E. (1981). Botanical exploration of southern Africa. Cape Town.

Haworth, A. H. (1812). Synopsis plantarum succulentarum. London.

Herre, H. (1971). The genera of the Mesembryanthemaceae. Cape Town.

Herre, H. & Friepricu, H. C. (1959). Die geographische Verbreitung der Ficoidaceae - Mesembryanthemoi-
deae in Siidafrika. 1. Teil: Aptenieae. Mitt. bot. StSamml., Miinch. 3: 44-70. .

Herre, H. & Friepricu, H. C. (1961). Die geographische Verbreitung der Ficoidaceae - Mesembryanthemoi-
deae in Siidafrika. 2. Teil: Ruschieae-Delosperminae. Mitt. bot. StSamml., Miinch. 4: 37-58.

Hicains, V. (1956). Mesembryanthemum, in CHITTENDEN, F. J., ed. The Royal Horticultural Society dictionary
of gardening 3: 1290-1294. Oxford.

Hopkins, J. J. (1979). The alien Carpobrotus edulis — a threat to the Lizard flora. The Lizard 6(3): 14-15.

Hutcuinson, J. (1917). The Lizard on a morning in June. Gdnrs’ Chron., ser. 3, 62: 31.

JERMYN, S. T. (1974). Flora of Essex, p. 72. Colchester. ~_

KENT, D. H. (1982). The native and introduced succulents of the British Isles. The Xerophyte 5: 65-72.

Le Sueur, F. (1984). Flora of Jersey, pp. 28-29. Jersey.

LINNAEUS, C. (1738). Hortus Cliffortianus. Amsterdam.

244 C.D PRESTON‘AND’*P/D: SEEL

LINNAEUS, C. (1753). Species Plantarum. Stockholm.

LINNAEUS, C. (1759). Systema Naturae, 10th ed. Stockholm.

LousLey, J. E. (1971). The flora of the Isles of Scilly, pp. 32-34, 142-148. Newton Abbot.

LousLey, J. E. (1973). ‘Mesembryanthemums’ established in the Isles of Scilly, in GREEN, P. S. ed. Plants: wild
and cultivated, pp. 83-85. Hampton.

Marcetrs, L.J. & Davin, R. W. (1981). A review of the Cornish flora 1980, p. 55. Redruth.

McC.uintock, D. (1964). Exotics 1963. Wild Flower Mag. 340: 24-29.

McCuntock, D. (1975). The wild flowers of Guernsey, pp. 84-85. London.

PERRING, F. H. & WALTERS, S. M., eds (1962). Atlas of the British flora, p. 80. London.

Preston, C. D. (1988). Hottentot Fig, Carpobrotus edulis (L.) N.E.Br., in Guernsey. Rep. Trans. Soc. guernés.
22: 296-302.

Row ey, G. D. (1978). Aizoaceae, in HEYwoop, V. H., ed. Flowering plants of the world, pp. 65-67. Oxford.

Rowse, A. L. (1941). Tudor Cornwall, p. 17. London.

SaLM-Dyck, J. M. F. A. H. (1836-1863). Monographia Generum Aloes et Mesembryanthemi. Bonn.

SCHWANTES, G. (1927). Zur Systematik der Mesembrianthemen. Z. SukkulKde 3: 14-30, 105-106.

SCHWANTES G. (1957). Flowering stones and mid-day flowers. London.

ScHWANTES, G. (1971). The classification of the Sst UE in HeRRE, H. The genera of the
Mesembryanthemaceae, pp. 3—6. Cape Town.

Simpson, F. W. (1982). Simpson’s Flora of Suffolk, pp. 116-117. Ipswich.

STEARN, W. T. (1938). An annotated index to Salm-Dyck’s ““Monographia Generum Aloes et Mesembryan-
themi’’. Cactus J. 7: 34-44, 66-85.

STEARN, W. T. (1965). Adrian Henry Haworth 1768-1833 botanist, entomologist and gardener: his life and
publications, in HAwortTH, A. H. Complete works on succulent plants 1: 5-80. London.

Tutin, T. G. (1962). Aizoaceae, in CLAPHAM, A. R., Tutin, T. G. & Warsurc, E. F. Flora of the British Isles,
2nd ed., p. 270. Cambridge.

Tutin, T. G. (1964). Mesembryanthemum L., in TuTin, T. G. et al., eds. Flora Europaea 1: 113. Cambridge.

VENNING, J. (1984a). Carpobrotus, in GeorGE, A. S., ed. Flora of Australia 4: 25-27. Canberra.

VENNING, J. (1984b). Disphyma, in GeorcE, A. S., ed. Flora of Australia 4: 36. Canberra.

Vicurs, C. C. (1908). Trifolium striatum, Linn., ? var. erectum, Leight. Rep. botl Soc. Exch. Club Br. Isl. 2:
282-283.

(Accepted May 1988)

APPENDIX: CHROMOSOME COUNTS OF AIZOACEAE NATURALIZED IN THE
BRITISH ISLES

R. D. PONS

Chromosome counts were made of specimens collected by C. D. Preston and Mrs F. le Sueur on
which the above account was based. They were grown in pots in the University Botanic Garden,
Cambridge, from which root tips were easily procured. Voucher specimens of flowers and
sometimes fruits were pressed and placed in CGE.

The root tips were placed in 0-05% colchicine in corked tubes for 2 h at room temperature,
followed by fixation in Carnoy’s solution for 2h. They were then put in Im HCl and hydrolysed,
uncovered, for 6 min in an oven at 60°C, and stained in Feulgen’s solution for 1-2 h. The root tips
were then mounted on a slide with propionic orcein, squashed, and the chromosomes counted. The
slides were made permanent by using liquid nitrogen to facilitate removal of the coverslip, passing
the slide into 90% alcohol followed by absolute alcohol and xylene for 1 min each, and finally
mounting in dammar xylene or Euparal. Occasionally root tips were stored in 70% alcohol
overnight after fixation in Carnoy’s solution, but this was less satisfactory than if the complete
process was carried out on the day of collection.

Aptenia cordifolia (L.fil.) Schwantes: St Brelade’s Bay, Jersey, GR WV/589.486, Mrs F. le Sueur;
2n = 18. Grouville, Jersey, GR WV/706.499, Mrs F. le Sueur; 2n = 18.

Carpobrotus edulis (L.) N.E.Br. var. edulis: near Kempt Tower, St Ouen’s Bay, Jersey, GR WV/
56.52,’Mrs F. le Sueur; 2n = 18.

NATURALIZED AIZOACEAE 245

Carpobrotus edulis (L.) N.E.Br. var. rubescens Druce: N. end of Rocquaine Bay, Guernsey, GR
WV/25.78, C. D. Preston; 2n = 18.

Carpobrotus glaucescens (Haw.) Schwantes: south coast of Alderney, between Vallée des
Gaudulons and Val l’Emauve, GR WA/564.060, C. D. Preston; 2n = 18.

Disphyma crassifolium (L.) L. Bolus: ‘White Hut’, St Ouen’s Bay, Jersey, GR WV/565.516, C. D.
Preston; 2n = 36.

Drosanthemum floribundum (Haw.) Schwantes: Newman House, St Mary’s, Isles of Scilly, GR 00/
898.108, C. D. Preston; 2n = 36.

Erepsia heteropetala (Haw.) Schwantes: Buzza Hill, St Mary’s, Isles of Scilly, GR 00/905.103, C. D.
Preston; 2n = 18.

Lampranthus roseus (Willd.) Schwantes: Le Gouffre, Guernsey, GR WV/294.743, C. D. Preston;

2n = 36.
Oscularia deltoides (L.) Schwantes: ‘Juliet’s Garden’, Porthloo, St Mary’s, Isles of Scilly, GR 00/
909.115, C. D. Preston; 2n = 18.

Ruschia caroli (L. Bolus) Schwantes: Old Town Bay, St Mary’s, Isles of Scilly, GR 00/90.10, C. D.
Preston; 2n = 18.

oN
7

-

ogee
ae : CExst +. Mh, & Wate. *% m cig 5) -Ates of Pia ri
; a aie - Ps We cS metinnth FT anh id sof
_— eh: 3 ew <a beaA Fae ie Re. a* sabre omi4 — tBu
= : aa 4 a y : he
4 * a io i Se ——

; ie 25 fa - ¢ i. _- fs of v7 Mi + ace! rene ; Attia atom na, =

s
ms 4
+ : *: iy a \ ; ta he ie z tL Seen ra. &

rt 5-4 * 27 ‘ as s - rem ri
. a 7. Mead 3 ie. © “Me Lert Ue DiS bare :
i*o Me a Mel 2 vin if ofa nOd aon 2 ene
. . oF £8 LSB eR -£

: { t<i ir” hob peat ue 32, Gi ORG
at ’= ;
2 2 ‘ healed = a a
3 >, j |
7 = wits. i as 7

as ) "shee rae ght. aki
o ‘ 2 ; = - = U a m ie A
P— - Po f naw a . d f a Vis! mn ac ictorwe ieee ape a

: ed J ‘ ait é p 29 ty pie B Dec ten ot ghdside i:s- 30. Ls

~

: it y —
= dated re. anes y
a 2 7 Poi
ry 7 -
— = ms a
" " = = ~ a -~
A c ofa & >
r i
. = ~ = —
7 i a
- , me
: : die - <P ae > MI
?
7 > é
’ rf”
- ( oo
‘3 7 nh
‘
— We ‘
I -
. ” =
¥ >
— Get
4
*
.
> my ~~
rn
oe)
‘ 7 f. i
wi ‘
4
] ‘ j a
A % -
- rastii re:
s
;
* a <
“ -
. e
—~
*
4 Z :
_ s
« zr
7 ,
_
7

Watsonia, 17, 247-264 (1988) 247

Taxonomy, morphology and distribution of Atriplex hybrids in
the British Isles

/

P.M. TASCHEREAU

Institute for Resource & Environmental Studies, Dalhousie University, Halifax, Nova Scotia,
Canada, B3H 3E2

ABSTRACT

A study based on field, culture and experimental work delineates six hybrids in the genus Atriplex indigenous to
the British Isles. Detailed morphological descriptions are given and distribution maps and illustrations are
provided for A. glabriuscula Edmondston x A. longipes Drejer; A glabriuscula x A. praecox Hilphers; A.
glabriuscula X A. prostrata Boucher ex DC.; the variable A. longipes x A. prostrata (= A. X gustafssoniana
Taschereau, hybr. nov., which includes var. gustafssoniana and var. kattegatensis (Turesson) Taschereau, comb.
et stat. nov.); A. littoralis L. x A. prostrata (= A. X hulmeana Taschereau, hybr. nov.); and A. littoralis x A.
patula L. Hybrid derivatives involving A. /ongipes, widespread on the coasts of the British Isles, are responsible
for many of the identification problems in the A. prostrata group.

~

INTRODUCTION

Jones (1975a) summarized the literature reports of Atriplex hybrids in the British Isles and recorded
the presence of seven natural hybrids. As Jones herself noted, however, some of the hybrids
reported on the basis of intermediate specimens were doubtful, and as I have explained elsewhere
(Taschereau 1986), some of these reports are almost certainly wrong.

Hybrids in Atriplex are particularly difficult to recognize without a foundation of experimental
work and field studies. One reason for this is the extensive variation, both genetic and phenotypic,
exhibited by so many of the species. Genetically distinct morphs, differing in such characters as leaf
shape, habit or colour, are known in several species (Taschereau 1985a). Morphological and habit
changes occur also in response to environmental factors. This phenotypic plasticity of Atriplex has
long been recognized as complicating the identification of many species. As long ago as 1860, for
example, C. C. Babington wrote that, “‘Atriplexes on fat land are amongst the most undeterminable
of plants’ (A. M. Babington 1897).

Another complicating factor in Atriplex hybrid and species recognition has been the widespread
occurrence of hybrid derivatives. Like their facultatively autogamous parents, hybrid derivatives
perpetuate themselves by selfing. Unlike many first or second generation hybrids, they are highly
fertile, often range well beyond the parent populations, and may occupy an ecological niche
different from that of either of the parents. Unaware of their hybrid nature, taxonomists have dealt
with these species-like entities in various ways. Atriplex longipes X prostrata, for example, has given
rise to anumber of more or less distinctive morphs, some of which are very widespread. Babington’s .
(1841) A. rosea (non L.) is, in part, a derivative of this hybrid (see, for instance, the sheet in E
labelled “A. rosea, Little Hampton Sussex, 1837 com. L. W. Borrer’’, in Babington’s handwriting
and printed “‘from Charles C. Babington’’); Westerlund’s (1861) type collection of A. patula var.
bracteata is also a derivative of this hybrid, as is Turesson’s (1925) A. longipes subsp. kattegatense.

Except for A. laciniata L., all the Atriplex species indigenous to the British Isles are known to
hybridize. With the experimental studies of Turesson (1925), Hulme (1957), Gustafsson (1972,
1973a, 1973b, 1976) and my own studies (Taschereau 1985b, 1986), it is possible to distinguish many
of the Atriplex hybrids in the British flora and to identify most of them with reasonable certainty.
This paper describes and illustrates six hybrids and one hybrid variant, gives their distribution, and
discusses their taxonomy.

248 P. M. TASCHEREAU
MATERIALS AND METHODS

The field, cultivation and experimental work upon which this study is based have been set forth in
detail in Taschereau (1985a, b, 1986). Data from field studies were supplemented by specimens sent
to me between 1977 and 1978 by participants in the B.S.B.I. Atriplex Survey.

Material from the following herbaria was studied: ABD, BM, C, CGE, DBN, E, K, LD, LIV (incl.
herb. Barbara Hulme), LIVU, MANCH, NMW, OXF, S, SLBI, TCD (abbreviations according to
Kent & Allen (1984) and Holmgren et al. (1981)). I have annotated the entire holdings of the
following herbaria: ABD, CGE, DBN, E, LIV, LIVU, TCD. Also, approximately half of the large
holdings of NMW have been annotated by me.

My collections, including those of the B.S.B.I. Survey, are deposited in MANCH, except for type
specimens as indicated in the text. A duplicate set of Gustafsson’s (1976) hybrid collections used for
reference is in my possession (herb. Taschereau). Each dot on the distribution maps of the hybrids is
supported by one or more specimens filed in MANCH.

ATRIPLEX LONGIPES: A KEY TAXON

The occurrence of A. /ongipes in the flora of the British Isles was confirmed only in 1977, although its
presence was earlier suspected (Hulme in Aellen 1964; Jones 1975b). Taschereau (1985b) indicated
that A. /ongipes is the key to understanding much of the confusing variation within the A. prostrata
group. Atriplex longipes hybridizes with A. prostrata and with A. glabriuscula; species that only
rarely hybridize with each other. The hybrid derivatives involving A. longipes are widely distributed
on the coasts of the British Isles and occur also in inland salt marshes. They are usually highly fertile
and often species-like in their behaviour. Morphologically, they may resemble one or the other
parent, or be distinct from either parent. They may occur in the same habitat as one or the other
parent species, or occupy a different ecological niche. This situation, and some of the frequent
identification problems it has given rise to, are summarized in Fig. 1.

Many of the long-recognized taxonomic problems i in the A. prostrata group in Britain have been
caused by the presence of hybrid derivatives involving A. longipes. There yet remains, however, a
residue of plants, especially from the northern and north-western coasts of Scotland and the coasts
of Shetland, which have not been satisfactorily identified except to group.

DESCRIPTIONS

1. A. glabriuscula Edmondston x A. longipes Drejer (Fig. 2)

Most plants resembling A. glabriuscula in habit and general morphology. Prostrate or ascending.
Lower leaves triangular, or rhombic to ovate-lanceolate; base cuneate, obtuse or truncate.
Bracteoles rhombic or ovate-lanceolate, herbaceous at the apex, frequently large (10—20 mm long)
and foliose, some of the axillary ones stalked, margins united almost up to the middle or only near
the base, thickened towards the base by a moderately to strongly developed silvery brown spongy
tissue, dorsal surface often strongly reticulate-veined in the lower part.

Habitat and Distribution. Exposed coastal beaches, frequently with A. glabriuscula. Hybrid
derivatives between A. glabriuscula and A. longipes are frequent on the coasts of northern Scotland
and north-western England. Reported by Gustafsson (1976) as rare in western Scandinavia with one
record from Bodo, Norway. The distribution in the British Isles is shown in Fig. 3.

Variation and Biosystematics. Hybrid derivatives, although often occurring in the same habitat as A.
glabriuscula, are well-established and independent of the parent taxa. The hybrid was artificially
synthesized by Gustafsson (1973a). Seed germination of F;, artificial hybrids was only between 10%
and 30%, although the pollen stainability of F, plants was between 80% and 100% (Gustafsson
1973a).

-“

249

ATRIPLEX HYBRIDS IN THE BRITISH ISLES

‘sadisuo] “py BUIA[OAUL SOANRALIOP PLIQhH “[ ANNI

BTreqroy uy BMosntaqets “Vy
se petjTyueptstu ATquenbery °g ‘eTmosntaqe[s *Y se patsfyuapysyfu ATyusenbery

Plnosnyaqely “VY — sjzuered oyuy jo auo

ATuO Jo 4eyTqey oyy UF sMo000 ATyUaNbary °G syusered ey} JO

euo TayytTe JO 4yeytqey ouy UT M900 ATJUsNbery

quered d rk b
Uo Jo ooUesQe ey} UT aMo00 ATqUeNberT *y s}uarIed yyOQ JO BoUesqLe O44 UF IMo50 ATZUENbaeT

exe} yUuated ey3 usemnqeq AVOTOYUdIOM txe} yUeTed ey}, UsaMzZoq
uy OJETpsMIE{UT ore AT[VUOTEBDDO °¢ / ABojToydiow uf opeTpourozuTF ere ATJUsnbary

Bodyauo, *y AT Tensn

Btnosnyaqets °“y ATTN
e Feast pot Sl ‘qguoivud suo ATuO eTquasezr AT Teuotsess9

‘yuoiud euo ATuO sTquesezr ATyuanberzy °z

quored r0yzfeu eTquesexz ATT eUoTEBdDON *1 queried Ioyzyeu eTquesezr ATyuenbery

BoATpUAFIOG PFAQAT

erey ATOA BaATYBATIAG pTaqAl

Blnosmypaqetys
v.

¥}01}8010

hy

aivy ATIOA spTaqAll
Areuypog

vyTreqiay ut BoedyouoT °V zo *eqeryso0ad *y -

250 P.M. TASCHEREAU

Ficure 2. A. glabriuscula x A. longipes.

The name A. glabriuscula f. pedicellata Abromeit has been applied to variants of this hybrid with
large stalked bracteoles (Aellen 1960; Gustafsson 1976).

Diagnostic Characters. The hybrid is most frequently confused with A. glabriuscula, from which it
can be distinguished by the stalks present on some of the bracteoles in the leaf axils, and the
tendency of many bracteoles to develop large foliose tips and thickened, veiny basal portions.

2. A. glabriuscula Edmonston x A. praecox Hiilphers (Fig. 4)
Most plants resembling A. praecox in leaf outline but similar to A. glabriuscula in bracteole

morphology. Prostrate-spreading or ascending. Foliage mostly reddish or less commonly green,
resembling A. praecox. Lower leaves lanceolate-triangular or ovate-lanceolate; base cuneate to

“

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 251

Ficure 3. Distribution of A. glabriuscula x A. longipes in the British Isles.

obtuse. Upper leaves lanceolate to linear. Bracteoles rhombic, sessile or slightly stalked (stalks c.
0-5 mm long), margins united almost up to the middle, thick-spongy towards the base, dorsal surface
smooth and without prominent veins.

Habitat and Distribution. Very rare in north-western Scotland and Shetland where it occurs with A.
praecox and A. glabriuscula at the margins of protected sea inlets. Reported by Gustafsson (1976) as
rare in the northernmost parts of Norway. The distribution in the British Isles is shown in Pips -y.

Variation and Biosystematics. The hybrid was artificially synthesized by Gustafsson (1973a). In the
artificial hybrids, pollen fertility and seed germination were greatly reduced. In two crosses, one had

252 P.M. TASCHEREAU

Ficure 4. A. glabriuscula X A. praecox. a) Habit, b) Bracteoles (not to scale), c) Leaf variation.

73% and the other 67% stainable pollen, and only 40% to 50% of the seeds germinated (Gustafsson
1973a).

Diagnostic Characters. The small, often reddish, leaves resembling those of A. praecox, in
combination with the thick-spongy bracteole characteristic of A. glabriuscula, will distinguish this
hybrid.

3. A. glabriuscula Edmondston x A. prostrata Boucher ex DC. (Fig. 6)

Plants generally intermediate in morphology between the parent species; ascending. Lower leaves
triangular as in the parents; base subcordate, truncate to obtuse. Inflorescence spiciform as in A.
prostrata. Bracteoles sessile, smaller than A. glabriuscula and in more densely packed glomerules,
thick-spongy, margins united almost up to the middle, dorsal surface bi-tuberculate or smooth,
without prominent veins. Seeds mostly less than 3-0 mm wide, radicle variously positioned and
directed.

Habitat and Distribution. Middle beach zone on coastal beaches with the parent species. Rare from
southern England in v.c. 15 (Badmin 1978) to south-western Scotland in v.c. 74 and on the eastern

“

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 253

oO fe2)

b

SS SS Se Cer a Cee Cr el Cay Cay a a Cy Ca Wey Can Ca Caney Ci) Ua Oo

Ficure 5. Distribution of Arriplex hybrids in the British Isles: O A. glabriuscula x A. praecox; @ A. glabriuscula
x A. prostrata;~)-A. littoralis x A. prostrata; X A. littoralis x A. patula.

coast of England in v.c. 66. Reported by Gustafsson (1976) as relatively rare in the western parts of
Scandinavia. The distribution in the British Isles is shown in Fig. 5.

Variation and Biosystematics. The hybrid was artificially synthesized by Gustafsson (1973a). Most of
the artificial hybrids had a high pollen stainability (four out of seven crosses had greater than 90%
stainable grains), but seed germination was poor. Seeds from two of the crosses did not germinate
and in the other seven crosses seed germination varied between 1% and 40% (Gustafsson 1973a).

254 P.M. TASCHEREAU

Ay Can
Ga-
Do

RIG Oy
i — ‘AY 2
WAS QF SBE CES
= MIRA Sika
Oe

SS
bey.
Sos
»
© oe y
oF

SS
aa - J LA
) LOE ats
~~ thas = Aat BeOS
By SKE GER? eS
cag > SN GSA Rie
A <I ;
‘
SAG
oa: pe pte

8mm
—_—

4cm
—

Figure 6. A. glabriuscula X A. prostrata. a) Habit, b) Bracteoles and seed.

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 255

FiGuRE 7. A. X gustafssoniana var. gustafssoniana. a) Habit, b) Variation in bracteoles.

256 P.M. TASCHEREAU

Germination of seeds from spontaneous hybrids in nature was also low (less than 40%) (Gustafsson
1973b).

From identified material in herbaria, it is clear that hybrid derivatives of A. longipes x A.
prostrata and A. longipes X A. glabriuscula have occasionally been mistaken for A. glabriuscula x
A. prostrata. This hybrid is rare; it is not common as suggested by Moss & Wilmott (1914) and by
Jones (1975a), and earlier records should be re-examined.

Diagnostic Characters. The size, morphology and arrangement of the bracteoles are most useful in
distinguishing the hybrid from its parents: the bracteoles are thick-spongy and frequently bi-
tuberculate as in A. glabriuscula, but generally much smaller and arranged in somewhat dense,
leafless, spiciform inflorescences similar to those of A. prostrata.

&
~

4. Atriplex < gustafssoniana Taschereau, hybr. nov.
Ho totypus: Snettisham, N. W. Norfolk, v.c. 28. Margin of Phragmites australis stand in open area
of saltmarsh by the coast, 18 September 1976, Taschereau & Libbey 76-12 (DAO).

Hybrida ex A. longipes Drejer et A. prostrata Boucher ex DC. orta. Folia inferiora elongato-
triangularia vel ovato-lanceolata, basi truncata vel cuneata. Quidam bracteoli in axillo foliorum |
superiorum magni, tenuiter herbacei, petiolati, angulis lateralibus acutis saepe bene evolutis,
superficie plusminusve forte reticulo-venata basin versus.

i) var. gustafssoniana (Fig. 7)

Atriplex patula var. bracteata Westerlund, Sveriges Atriplices 57 (1861). LEctotyPE: Sweden, Skane;
sheet labelled ‘“‘Atriplex patula Lin. — bracteata Wester., Skane, C.A.W.” (S); chosen by
Taschereau in Watsonia 15: 204 (1985).

Erect, ascending or prostrate, with long-spreading lower branches. Lower leaves elongate-
triangular in erect forms, or ovate-lanceolate in some prostrate forms, usually much longer than
wide (length/width ratio 1-5—2-5); base cuneate to truncate. Bracteoles 4—9 mm long, triangular,
sessile or stalked but usually some with stalks 0-5—5-0 mm long present in the axils of the upper
leaves and branches; thin-herbaceous, sometimes becoming foliose; margins united at the base,
lateral angles pointed and often strongly developed; dorsal surface often strongly reticulate-veined
toward the base.

Habitat and Distribution. Coastal sand and shingle beaches, often remote from the parent species.
In coastal, estuarine and inland salt marshes, frequently in stands of Agropyron pungens or Juncus
maritimus and at the margins of Phragmites australis stands. One of the commonest Atriplex taxa on
the coasts and in estuaries around the British Isles and occasional in inland saltmarsh vegetation at
Nantwich, Cheshire and Preesall, Lancashire. Reported by Gustafsson (1976) as common on the
western coast of Sweden northwards to the southern parts of Norway, and as relatively rare in
Denmark and the Baltic area. Most of the Atriplex populations along the Swedish western coast
investigated by Gustafsson (1973b) included individuals of hybrid origin. The distribution in the
British Isles is shown in Fig. 8.

Variation and Biosystematics. The cultivated progeny of most wild A. longipes xX A. prostrata
derivatives exhibited a mixture of parental characters. Many plants, however, showed a degree of
segregation that gave some indication of their parentage (Taschereau 1986). In the artificial hybrids
(Gustafsson 1973a) pollen stainability was high (16 of 22 crosses had greater than 80% stainable
pollen grains), but seed germination was greatly reduced. Germination varied from 0% to 60% but
in 20 out of 23 crosses less than 30% germinated. In the wild hybrids examined by Gustafsson
(1973b) both pollen fertility and seed germination were reduced. The F; hybrids in many of the
artificial crosses were morphologically similar to A. prostrata even when A. longipes was used as the
female parent (Gustaffson 1973a). A similar situation has been observed in wild British plants
(Taschereau 1985b). In parts of Scandinavia where grazing has disturbed the habitat, introgressive
hybridization between A. longipes and A. prostrata has given rise to a more or less continual gene

“

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 257

Figure 8. Distribution of A. x gustafssoniana in the British Isles: @ var. gustafssoniana, -@-var. kattegatensis.

flow between these species (Gustafsson 1974). This seems to be the situation in most parts of Britain
where, according to Adam (1978), undisturbed estuarine salt marsh vegetation (the habitat of A.
longipes) is virtually non-existent.

Hybrid derivatives are well established and independent of the parent species. They are
extremely variable and almost every combination of parental characters has been observed in the
field. On the basis of size, habit and habitat our plants can be broadly placed into two groups: 1) Salt
marsh variants. Large + succulent plants, erect, spreading to straggling, up to 1 m high; stems
strongly angled; leaves large, elongate-triangular, with cuneate to truncate base; occurring in tall
Salt marsh vegetation or at the margins of such communities. 2) Beach variants. Small non-succulent

~

258 P.M. TASCHEREAU

or more or less succulent plants, prostrate, decumbent or weakly erect; lower branches wide-
spreading and often longer than the central axis; angularity of stems variable; occurring on +
exposed coastal beaches and at the margins of or in sparsely covered or open areas of coastal salt
marsh vegetation.

Diagnostic Characters. In the axils of some of the upper leaves occur large thin-herbaceous, stalked
bracteoles with lateral angles pointed and often strongly developed and the dorsal surface more or
less strongly reticulate-veined toward the base. The occurrence of these characteristic bracteoles in
combination with elongate-triangular or ovate-lanceolate lower leaves at least some of which have a
more or less truncate base is diagnostic of this hybrid.

Nomenclatural Note. The type specimen of Atriplex patula var. bracteata Westerlund is a hybrid
derivative between A. longipes and A. prostrata. Plants such as this are part of the myriad of beach
variants comprising the hybrid derivatives described here under var. gustafssoniana. While I have
myself seen specimens along the Swedish coast at Malm6 that are similar to the type, the characters
distinguishing such plants are, in my view, not constant enough to merit further formal taxonomic
recognition. In addition, there has been considerable confusion regarding the application of the
name var. bracteata Westerlund since Aellen (1960) applied it (as f. bracteata) to plants of A. patula
L. s.s. with foliose bracteoles (ennatia), and Fernald (1950) applied the name A. patula var.
bracteata to plants described as very similar to typical A. patula ‘“‘but with fruiting bracts 1-1-5 cm
long” occurring on marshes of Cape Breton, Nova Scotia. Fernald’s specimen from Cape Breton
(GH!) representing his var. bracteata is, like Westerlund’s type, a hybrid derivative in the A.
prostrata group. Furthermore, Westerlund’s (1861) description is not sufficiently detailed to exclude
certain variants of A. patula sens. str., and his statement, “Hab. rara ad vias circa Lund et in litore
marino ad Malmo Scaniae”’ would seem to include the ruderal A. patula (‘‘ad vias circa Lund’),
along with the obligate halophytic derivatives of A. longipes X A. prostrata (“‘in litore marino ad
Malm6 Scaniae’’).

ii) var. kattegatensis (Turesson) Taschereau, comb. et stat. nov. (Fig. 9)

A. longipes subsp. kattegatense Turesson in Lunds Univ. Arsskr., N.F. Adv. 2, 24 (4): 4 (1925).
LectotyPE: Sweden, Bohuslan, Tanums socken, Otteron, August 1880, H. Thedenius (S); the
specimen annotated “‘A. longipes kattegatense Turess. nov. subsp.”’, to the middle-left, chosen by
Gustafsson in Opera Botanica 39: 29, 31 (1976).

Plants 10—30 cm, prostrate to decumbent, spreading with the lower branches longer than the central
axis. Stems sub-angular, striate, green and stramineous striped or reddish. Branches opposite to
sub-opposite in the lower region of the central axis but often barely separated by extremely short
internodes. Foliage bluish-green, frequently reddish, succulent; mature lower and upper leaves
glabrous; juvenile leaves glabrous above, sparsely farinose on the undersurface. Lower leaves 2-0-
3-5 cm long, 1-0—2-5 cm wide, with a pair of outpointing to forward-curving basal lobes; margins
irregularly serrate or entire; apex acute; base cuneate. Upper leaves smaller, linear-lanceolate,
without or with basal lobes; margins entire or weakly toothed. Inflorescence axillary to the base of
the plant and shortly terminal, leafy to the tip. Bracteoles 3-5—5-0 mm long, at least some shortly
stalked (stalks 0-5—1-0 mm long), rhombic to triangular-lanceolate or ovate-lanceolate; apex acute
or acuminate; base cuneate; margins entire, united at the base or somewhat higher up to the lateral
angles; lateral angles pointed or rounded, not strongly developed; mostly herbaceous and thin but
sometimes slightly thickened toward the base by the presence of spongy tissue; dorsal surface
smooth or with a few short, weak appendages; venation obscure or somewhat prominent towards
the base. Seeds polymorphic, black or brown, orbicular, 1-5—2-5 mm wide; radicle sub-basal,
obliquely up-pointing.

Habit and Distribution. Exposed coastal beaches close to the salt water where the vegetation is very
sparse; often with A. praecox on the shores of somewhat less exposed sea inlets. Frequent in
northern Scotland and occasional in Shetland. Reported by Gustafsson (1973b) as common on the
western coast of Sweden. The distribution in the British Isles is shown in Fig. 8.

“

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 259

FicurE 9. A. X gustafssoniana var. kattegatensis. a) Habit, b) Bracteoles and seed.

Variation and Biosystematics. Gustafsson’s (1973b) experimental work with this taxon indicates it is
probably a more or less well established product of hybrid origin. Some of the cultivated offspring of —
var. kattegatensis raised by Gustafsson (1973b) from seeds collected in nature showed considerable
morphological variation. He observed segregates similar to A. longipes and to A. prostrata as well as
intermediate forms. Some offspring, however, did not differ from the parent and showed no
indication of segregation. Likewise, plants cultivated from var. kattegatensis collected at Ullapool,
v.c. 105, did not segregate (Taschereau 1986). These non-segregating forms may represent ancient

260 P.M. TASCHEREAU

FiGurE 10. A. X hulmeana. a) Habit, b) Leaf variation, c) Bracteoles and seed.

“

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 261

or historical hybrids that have arisen by stabilization of an F, or later segregant adapted to the
specialized exposed habitat that is colonized by neither of the parents. In the British Isles, var.
kattegatensis appears to have primarily a northern distribution and specimens from southern
regions such as those from v.c. 15 (Badmin 1978) require further study.

Hybrids. None have been identified from nature. Gustafsson (1973b) has made experimental
hybrids between var. kattegatensis and A. longipes, A. praecox and A. prostrata. Pollen fertility in
the majority of crosses was greater than 80% but hybrids with reduced fertility were common.
Morphological variation in the F; and F, progenies of the experimental hybrids was considerable
and difficult to categorize because of the free recombination of morphological characters that
occurred. '

5. Atriplex < hulmeana Taschereau, hybr. nov. (Fig. 10)
Ho ortypus: Wolferton, Norfolk, England, 18 September 1976, Taschereau & Libbey 76-24 (DAO).

Hybrida ex Atriplex littoralis L. et A. prostrata Boucher ex DC. orta. Folia primaria succulenta,
triangulari-lanceolata, cum lobis basalibus patentibus vel falcatis, basi cuneata. Bracteolae
triangulares vel ovati-triangulares, crassae, spongiosae, basibus conjunctae.

The hybrid is named after Barbara Hulme, British botanist and first person to produce experimental
hybrids in Atriplex by controlled crossing (Hulme 1957, 1958). Her elegant work provided an
experimental basis for understanding the relationship of A. patula to A. littoralis and to members of
the Hastata complex (A. prostrata group) in Britain.

Erect, very robust plants with thick stems and ascending branches. Foliage dark green, glabrous,
very succulent. Lower leaves triangular-lanceolate with a pair of strongly developed out-pointing to
forward-curving basal lobes; length at least twice the width; margins irregularly sinuate-dentate,
attenuate to the apex; base strongly cuneate. Upper leaves lanceolate, without basal lobes or with a
single lobe developing on one side only, entire or with few teeth. Bracteoles 5—10 mm long, sessile,
broadly triangular to ovate-triangular; apex acute; base truncate to cuneate; margins denticulate or
with only a few teeth, united at the base; lateral angles rounded and weakly developed; thick-
spongy; dorsal surface strongly bi-tuberculate or smooth; venation obscure. About 70-80% of the
bracteoles contain well-developed seed. Both large-brown and small-black seed types are frequent
in the same plant. Segregants showing a wide range of character combinations sometimes occur
amongst the F, plants described above. Many of the segregants are largely sterile and often weak-
stemmed, but some are vegetatively very vigorous and develop into quite prominent plants. The
following three variants appeared in cultivation and are sufficiently common in the field to be
mentioned.

1. Patula-leaf variant. Plants with leaves like A. patula that possess the falcate basal lobes so
characteristic of this species.

2. Serrata-leaf variant. Plants with markedly sinuate-dentate leaves like extremes of plants formerly
called A. littoralis var. serrata.

3. Gigas variant. Plants up to 1 m high with thick stems and gigantic leaves with ovate-lanceolate
irregularly lobed lamina up to 15 cm long and 4 cm wide.

Habitat and Distribution. Disturbed habitats where the parent species are present in abundance:
disturbed banks of estuaries recently dredged, earthen sea walls less than three years old, salt
marshes disturbed by rabbit grazing. Occasional on the eastern and western coasts of England in
v.cc. 28, 58, 60, 66. Jones (1975a) states it has been recorded as rare in Denmark. The distribution in
the British Isles is shown in Fig. 5.

Variation and Biosystematics. See Taschereau (1986) for a detailed discussion. Jones’ (1975a) report ~
of the putative hybrid A. glabriuscula Edmondston x A. littoralis L. probably refers to A. X
hulmeana as discussed in Taschereau (1986).

Diagnostic Characters. The characteristic shape of the large succulent lower leaves in combination
with the spongy-thick bracteoles is diagnostic of this hybrid.

262 P.M. TASCHEREAU

uns

Figure 11. A. littoralis x A. patula.

6. Atriplex littoralis L. x A. patula L. (Fig. 11)

Erect, up to 0-5 m high. Lower leaves up to about 6 cm long and 3 cm wide, ovate-lanceolate to
rhombic with a pair of forward-curving basal lobes; succulent, glabrous, dark glossy-green; margins
irregularly serrate to sinuate-dentate; base strongly cuneate. Upper leaves smaller, similar to the
lower ones or lanceolate to linear with or without basal lobes, entire or with a few teeth. Bracteoles
rhombic, sessile; margins denticulate, united up to the middle; thick-spongy, tuberculate; occurring

“

ATRIPLEX HYBRIDS IN THE BRITISH ISLES 263

in dense terminal inflorescences. The majority of bracteoles are empty and collapsed. Dispersed
between these sterile bracteoles, however, are occasional larger, well-formed, filled out bracteoles
that contain seed. From one to five of the larger fertile bracteoles occur in the inflorescenses of most
branches.

Habitat and Distribution. Disturbed weedy ground by the coast where both the parents are growing
together. Known only from one locality in Scotland at Leith Docks, Edinburgh, v.c. 83. Reported
by Turesson (1925) from Lomma and Torekov along Ore Sund in south-western Sweden. The
distribution in the British Isles is shown in Fig. 5.

Variation and Biosystematics. This hybrid has been experimentally synthesized three times: by
Turesson (1925), by Hulme (1957) and by Taschereau (1986), but its occurrence in nature is very
rare. It has been looked for on numerous occasions in locations where the parents were growing
together but found only once. A published record from Findhorn, v.c. 95, Scotland (Webster 1979)
is a mistake. Subsequent to publication of this record, I changed my original identification of the
hybrid specimen to A. littoralis x A. prostrata. For a detailed discussion of the biosystematics see
Taschereau (1986).

Diagnostic Characters. The green, densely spiciform inflorescence consists mainly of small empty
uniform bracteoles, but on almost every inflorescence branch, scattered among the green sterile
bracteoles, one or two large grey-black seed-containing bracteoles stand out. The plants resemble
A. littoralis in habit, but the lower leaves are similar to those of A. patula. The bracteoles are
spongy-thick as in A. /ittoralis but united almost to the middle as in A. patula. In early November,
the bright dark green colour of the hybrids makes them stand out strongly from the surrounding
vegetation. The parent species by this time are mostly brown and dying.

ACKNOWLEDGMENTS

I am grateful to Mr R. C. Palmer and to the curator of Oxford University Herbarium for
photographs of Atriplex glabriuscula X praecox from Shetland, upon which the illustration of this
hybrid is based. I thank Miss C. Bird who provided the Latin translation of my diagnoses, and Mr B.
U. Borluk who prepared most of the drawings.

REFERENCES

Apa, P. (1978). Geographical variation in British saltmarsh vegetation. J. Ecol. 66: 339-366.

AELLEN, P. (1960). Atriplex L. in RECHINGER, K. H., ed. Hegi Illustrierte Flora von Mitteleuropa, 2nd ed., 3:
664-693. Miinchen.

AELLEN, P. (1964). Atriplex L. in TuTin, T. G. et al. eds. Flora Europaea 1: 95-97. Cambridge.

BaBINGTON, A. M. (1897). Memoirs, journal and botanical correspondence of Charles Cardale Babington, p. 347.
Cambridge.

BaBINGTON, C. C. (1841). Monograph of the British Atripliceae. Trans. Proc. bot. Soc. Edinb. 1: 1-17.

BapMIN, J. (1978). Records of Atriplex spp. from Shellness, Isles of Sheppey, Kent. Trans. Kent Field Club 6:
167-170.

FERNALD, M. L. (1950). Gray’s Manual of Botany, 8th ed., p. 597. New York.

Gustarsson, M. (1972). Distribution and effects of paracentric inversions in populations of Atriplex longipes.
Hereditas 71: 173-194.

GustaFsson, M. (1973a). Evolutionary trends in the Atriplex triangularis group of Scandinavia, 1. Hybrid
sterility and chromosomal differentiation. Bot. Notiser 126: 345-392.

Gustarsson, M. (1973b). Evolutionary trends in the Atriplex triangularis group of Scandinavia, 2. Spontaneous ~
hybridization in relation to reproductive isolation. Bot. Notiser 126: 398-416.

Gustarsson, M. (1974). Evolutionary trends in the Atriplex triangularis group of Scandinavia, 3. The effects of
population size and introgression on chromosomal differentiation. Bot. Notiser 127: 125-148.

Gustarsson, M. (1976). Evolutionary trends in the Atriplex prostrata group of Scandinavia, 4. Taxonomy and
morphological variation. Op. bot. Soc. bot. Lund 39: 1-63.

Hoimcren, P. K., KEUKEN, W. & SCHOFIELD, E. (1981). Index herbariorum, 7th ed. Utrecht.

264 P.M. TASCHEREAU

Hume, B. A. (1957). Studies on some British species of Atriplex. Ph.D. thesis, University of Edinburgh.

Hung, B. A. (1958). Artificial hybrids in the genus Atriplex. Proc. bot. Soc. Br. Isl. 3: 94.

Jones, E. M. (1975a). Atriplex, in Stace, C. A., ed. Hybridization and the flora of the British Isles, pp. 185-186.
London.

Jones, E. M. (1975b). Taxonomic studies of the genus Atriplex (Chenopodiaceae) in Britain. Watsonia 10: 233-
Zi.

Kent, D. H. & ALLEN, D. E. (1984). British and Irish herbaria. London.

Moss, C. E. & Witmotrt, A. J. (1914). Atriplex L. in Moss, C. E. The Cambridge British flora 2: 168-182, plates
172-188. Cambridge.

TASCHEREAU, P. M. (1977). Atriplex praecox Hilphers: a species new to the British Isles. Watsonia 11: 195-198.

TASCHEREAU, P. M. (1985a). Taxonomy of Atriplex species indigenous to the British Isles. Watsonia 15: 183-209.

TASCHEREAU, P. M. (1985b). Field studies, cultivation experiments and the taxonomy of Atriplex longipes Drejer
in the British Isles. Watsonia 15: 211-219.

TASCHEREAU, P. M. (1986). Hybridization in the genus Atriplex section Teutliopsis (Chenopodiaceae). Watsonia
16: 153-162.

Turesson, G. (1925). Studies in the genus Atriplex. Lunds Univ. Arsskr., N. F. Avd. 2., 21(4): 1-15.

WessTER, M. McC. (1979). Plant record. Watsonia 12: 350.

WESTERLUND, C. A. (1861). Bidrag till kinnedomen af Sveriges Atriplices. Lund.

(Accepied March 1988)

Watsonia, 17, 265-272 (1988) 265

Hybridization between Polygonum mite Schrank, P. minus
Huds. and P. hydropiper L. in Northern Ireland with
comments on their distinction

J.A..N, PARNELL

‘

School of Botany, Trinity College, Dublin 2, Ireland
and
D. A. SIMPSON

The Herbarium, Royal Botanic Gardens, Kew, Surrey, TW9 3AE

ABSTRACT

Polygonum mite Schrank, P. minus Huds. and P. hydropiper L., in Polygonum sect. Persicaria (Miller) DC.
occur together around Lough Neagh, Northern Ireland. Initial observations of these species had suggested the
presence of intermediate plants of possible hybrid origin. Populations from two sites at the Lough were
examined for hybridity using multivariate analyses, and by measurement of pollen fertility. Most plants could be
assigned to one of the three species, but a few were almost certainly hybrids between P. mite and P. minus (= P.
x wilmsii G. Beck). P. hydropiper did not appear to hybridize with the other species, although P. mite and P.
hydropiper may approach each other in their vegetative morphology. The characters which best distinguish the
three species are described and discussed. 5

INTRODUCTION

Polygonum mite Schrank, P. minus Huds. and P. hydropiper L. are closely related species within
Polygonum sect. Persicaria (Miller) DC. Of the three, P. mite is considered to be the least common,
occurring mainly in south-eastern England and Northern Ireland (Perring & Walters 1976). P.
minus is rather rare, although scattered throughout the British Isles, while P. hydropiper is common
and widely distributed (Lousley & Kent 1981).

Timson (1965, 1966, 1975) carried out a series of studies on hybridization in sect. Persicaria. He
found that most species are rarely visited by insects, produce very sticky pollen and are usually self-
pollinated before the flower opens. He concluded that interspecific hybridization in sect. Persicaria
would be rare, but theoretically possible; both P. mite and P. minus are tetraploids with 2n = 40.
Timson also suggested that hybridization could occur between P. mite and P. hydropiper (2n = 20),
the former species being an autopolyploid derivative of P. hydropiper. Nevertheless Lousley &
Kent (1981) noted that hybrids between each of the three species occur only rarely in Great Britain.

Webb (1984) carried out a survey of P. mite in Ireland, including an examination of all extant
sheets of Irish material labelled P. mite at TCD, DBN and BEL. He concluded that all records for
this species prior to 1969 were erroneous, being either P. minus, P. hydropiper or P. persicaria L.
Furthermore he concluded that there is only one area in Ireland where P. mite could be found, this
being the shores of Lough Neagh. Webb and J. R. Akeroyd visited this area in 1979 and found what
they thought might be a complex hybrid swarm, which included P. mite and P. minus. P. hydropiper
also occurs around the lake, although Webb did not regard it as part of the swarm. In the late
summer of 1984, we visited a number of sites around Lough Neagh, and our observations suggested
that some material was intermediate in morphology between P. mite, P. minus and P. hydropiper.
However it was known that all three species show a wide range of phenotypic variation. Therefore

266 J. A. N. PARNELL AND D. A. SIMPSON

we decided to investigate these populations in more detail, to establish whether any true hybrids
were present.

MATERIALS AND METHODS

Plants were obtained from two sites on the shores of Lough Neagh, these being near Newport
Trench harbour, Tyrone, v.c. H36 (GR H298.377) and Sands Bay, Co. Antrim, v.c. H39 (GR
J312.372). At each site material of the three species was gathered randomly, and a total of 100 plants
was collected and pressed. Voucher material has been deposited in TCD. 17 morphological
characters (Table 1) were scored from this material, three observations of each character being
made from each plant. Mean values per plant were then calculated for each character, and these
data were analysed by Principal Components Analysis (PCA) using the BMDP4M PCA program.
The data were also subjected to a Stepwise Discriminant Analysis (DSC) using the SPSS-X
statistical package, to test the strength of disinctions between the species and to indicate their best
distinguishing features (Parnell 1987).

TABLE 1. CHARACTERS USED IN NUMERICAL ANALYSES

. Leaf length

. Leaf width (maximum)

. Distance between leaf base and widest point on leaf

. Leaf venation (recorded on a scale of 1-2, where 1 = venation inconspicuous, 2 = venation conspicuous)

. Mid-stem ochrea tooth length

. Flower spike posture (recorded on a scale of 1-3, where 1= erect, 2 = slightly nodding, 3 = nodding)

. Spike length

. Number of axillary flowers

. Perianth length

10. Perianth width (maximum)

11. Perianth colour (recorded on a scale of 1-3, where 1 = red, 2 = pale pink, 3 = pink-green- an)

12. Perianth gland density

13. Nut length

14. Nut width (maximum)

15. Distance between nut base and widest point on nut

16. Degree of shininess of nut surface (recorded on a scale of 1-3, where 1 = shiny, 2 = somewhat shiny, 3 =
dull)

17. Degree of roughness of nut surface (recorded on a scale of 1-3, where 1 = smooth, 2 = slightly rough, 3 =

very rough)

OMOrnNANDRWN EH

Pollen fertility, as indicated by stainability tests, was determined using selected material of the
species and their possible hybrids. Three to four ripe anthers from each specimen were dissected on
a microscope slide and stained in Cotton Blue/Lactophenol for 30 minutes. The numbers of stained
and unstained grains were then counted and percentage stainability calculated.

RESULTS

Most of the plants from both populations could be visually assigned to one of the three species (Figs.
1—3), and these findings were confirmed by both PCA (Fig. 4) and DSC. In the PCA, plants visually
assigned to P. mite made up the largest group of points on the scatter plot, with smaller groups
corresponding to P. minus and P. hydropiper. Principal components 1 and 2 accounted for 54% and
28% of the total variation respectively. The DSC distinguished each species at a 99-9+ % level of
confidence (all F>17, P<0-001), and the characters picked out by this analysis that would best
separate the species are given in Table 2. Using a combination of these it was possible to assign
herbarium specimens to a taxon with little difficulty.

A few plants were found at both sites which seemed to be intermediate between P. mite and P.

“

POLYGONUM HYBRIDS IN N. IRELAND 267

a ie ol

FicurE 1. Polygonum mite Schrank from Lough Neagh, Northern Ireland, August 1984, Parnell & Simpson 73
(TCD). Scale bar = 20 mm.

TABLE 2. CHARACTERS WHICH BEST SEPARATE POLYGONUM MITE, P. MINUS AND P.
HYDROPIPER (MEASUREMENTS ARE MEAN VALUES + 95% CONFIDENCE LIMITS)

Character P. mite P. minus P. hydropiper

1. Nut length (mm) 3-03+0-06 2-18+0-12 3-21+0-12

2. Nut width (mm) 1-83+0-43 1-12+0-24 2-20+0-07

3. Nut shininess very shiny somewhat shiny dull

4. Perianth gland shape flat — raised

5. Perianth glands, 12 0 83

approx. no. per flower

6. Perianth length (mm) 3-66+0-06 2-48+0-33 4-10+0-15

7. Flower colour pale pink to purplish- usually reddish pale pink to greenish-
pink white

8. Spike slightly nodding erect nodding

9. Mid-stem ochrea tooth 1-76+0-09 1-67+0-26 0:30+0-16

length (mm)

J. A. N. PARNELL AND D. A. SIMPSON

268

“WU QZ = eq BBS (GOL) 62 vosduag x jjausng “pel isndny ‘purjoly
UIOYION ‘yYseoN YsnoyT wo “J sadidospdky wnuosdjog *¢ AANA

:

7
|

POLYGONUM HYBRIDS INN. IRELAND 269

Principal component 2

a =e —1 0 1 2 3
Principal component 1

Ficure 4. Principal components analysis on all plants sampled. @ = P. mite, O = P. minus, @ = P. hydropiper,
® = P. mite x P. minus.

minus. In particular these plants had fewer flowers than either mite or minus and produced fewer
nutlets. One plant (specimen no. 71) was also found which appeared to be intermediate between P.
mite and P. hydropiper. On the PCA scatter plot five points were located between the mite and
minus groups and these corresponded with the visually intermediate material. One point was also
located between the hydropiper and mite groups, which again corresponded with the visually
intermediate plant.

The results of the pollen stainability tests are shown in Table 3. Plants which were not considered
to be intermediate showed pollen stainability levels in the range 87-0—99-3%. Those plants which
were intermediate between P. mite and P. minus showed a significantly reduced pollen stainability
in the range 76-1—-82-4% (P. mite (t=6-41) and P. minus (t=9-84), both P<0-001). However
specimen no. 71, the intermediate between P. mite and P. hydropiper, at 92:7% did not have
significantly reduced pollen stainability (cf. P. mite (t=0-76) and P. hydropiper (t=0-71), both not
significantly different).

DISCUSSION
Our initial observations showed that as well as the three species studied, P. persicaria L. and P.

lapathifolium L. were also present at both sampling sites. Webb (1984) drew attention to the
difficulty in separating the two latter species from the others, particularly P. mite, and suggested that

270 J. A. N. PARNELL AND D. A. SIMPSON

TABLE 3. PERCENTAGE POLLEN STAINABILITY OF POLYGONUM SPECIMENS

FROM LOUGH NEAGH

Specimen no. % stainability Identity
38 88-7 mite
40 89-2 mite
72 94-2 mite
73 92-2 mite
75 87-0 mite
28 94-4 minus
29 97-4 minus
5 92-0 minus
34 92-2 minus
45 98-1 minus
58 93-6 minus ~

19 82-3 mite X minus
30 80-7 mite X minus
35 78-9 mite X minus
49 76-1 mite X minus
D20 85-7 mite X minus

6 99-2 hydropiper

8 94-7 hydropiper
21 93-7 hydropiper
22 90-7 hydropiper
23 98-0 hydropiper
71 92-7 hydropiper

P. persicaria was also part of the hybrid swarm at Lough Neagh. Furthermore Timson (1966) had
indicated that hybrids could occur between P. persicaria and P. mite or P. minus. However, we
found P. persicaria and P. lapathifolium to be readily distinguishable from the others, including P.
mite; the denser spike and small black patches on the leaves of P. persicaria and the possession of
both these features, together with pedicel glands in P. lapathifolium, were very distinctive. In
addition we found no evidence of hybridization between P. persicaria or P. lapathifolium with the
other species, and for this reason we concentrated on P. mite, P. minus and P. hydropiper.

Five of the plants we sampled fall between pure P. mite and pure P. minus, and these
intermediates are almost certainly the hybrid P. mite x P. minus (= P. X wilmsii G. Beck; Fig. 5).
Do hybrids with P. hydropiper also occur? The plant which was morphologically intermediate
between P. mite and P. hydropiper did not have reduced pollen stainability and also produced a
similar number of flowers and nutlets to P. hydropiper. This is almost certainly not a hybrid, because
an F,; hybrid would be triploid (2n=30) and, presumably, highly sterile. Therefore it would seem
that Webb (1984) was correct when he indicated the potential presence of hybridization in sect.
Persicaria around Lough Neagh, but with hybrids only occurring between P. mite and P. minus, and
not all species present.

Out of the three species studied, P. hydropiper is supposedly the easiest to recognize. Webb
(1984) and others (Tutin 1987; Lousley & Kent 1981) have indicated that the possession of glands on
the perianth of P. hydropiper is, apart from its taste, one of the key characters allowing its
separation from both P. minus and P. mite. Surprisingly we found that virtually all of the P. mite
plants also had perianth glands, and initially we thought these were similar to those on P.
hydropiper. This led us to believe at first that such plants might have been hybrids. However on
closer examination we found the glands to be smaller, fewer in number and almost flat compared
with those on P. hydropiper. Similar glands were also found on the perianth of British and European
material of P. mite, despite the fact that these glands are not mentioned in standard works on
Polygonum (Tutin 1987; Lousley & Kent 1981). However the differences in gland structure and
number are further key characters, identified by DSC, separating P. mite from P. hydropiper. We
also found that some of the other characters used to separate all three species, such as nut size and

=“

POLYGONUM HYBRIDS INN. [IRELAND 271

Figure 5. Polygonum mite Schrank x P. minus Huds. from Lough Neagh, Northern Ireland, August 1984,
Parnell & Simpson D20 (TCD). Scale bar = 20 mm.

leaf length:breadth ratio are rather more variable than previously thought (Webb 1984 and pers.
comm.) Both these latter characters are useful for distinguishing P. minus from both P. hydropiper
and P. mite, but not for separating the last two from each other. This is particularly emphasized by
the leaf length:breadth ratio, so that the leaves of P. minus are up to 8-5 times as long as broad
whereas those of the other two species are up to 4-5—4-8 times as long as broad. The leaf shapes and
sizes of P. hydropiper and P. mite may thus be very similar. Therefore, as is often the case, it is best
to use a combination of the characters given in Table 2 to delimit the species, rather than rely on
only a single character.

ACKNOWLEDGMENT

We thank Professor D. A. Webb for his help and encouragement during this work.

272 J. A.N. PARNELL AND D. A. SIMPSON
REFERENCES r

Lous Ley, J. C. & Kent, D. H. (1981). Docks and Knotweeds of the British Isles. London.

PARNELL, J. A. N. (1987). Variation in Jasione montana L. (Campanulaceae) and related species in Europe and
North Africa. Watsonia 16: 249-267.

Tiwmson, J. (1965). A study of hybridization in Polygonum section Persicaria. J. Linn. Soc. Bot. 59: 155-161.

Timson, J. (1966). Polygonum hydropiper L., in Biologial Flora of the British Isles. J. Ecol. 54: 815-821.

Timson, J. (1975). Polygonum L. sect. Persicaria (Mill.) DC., in Stace, C. A., ed. Hybridization and the flora of
the British Isles: 274-277. London.

TuTin, T. G. (1987). Polygonum L., in CLapHaM, A. R., Tutin, T. G. & Moore, D. M. Flora of the British Isles,
3rd ed., pp. 300-303. Cambridge.

WEBB, D. A. (1984). Polygonum mite Schrank in Ireland. Jr. Nat. J. 21: 283-286.

3 (Accepted May 1988)

|

Watsonia, 17, 273-277 (1988) 273

Two hybrids of Equisetum sylvaticum L.
new to the British flora

CoN, BPAGE

Royal Botanic Garden, Edinburgh, EH3 5LR

/
ABSTRACT

Two hybrids of Equisetum sylvaticum are reported, which are new to the British flora. These are E. x bowmanii

‘C.N. Page, hybr. nov. (E. sylvaticum L. x E. telmateia Ehrh.), from Hampshire, and E. x mildeanum Rothm.

(E. pratense Ehrh. X E. sylvaticum L.) from Perthshire.

INTRODUCTION

No hybrids involving Wood Horsetail (Equisetum sylvaticum L.) have been previously recorded
from the British Isles. This paper reports the occurrence of two such hybrids new to the British flora,
one of which is unknown elsewhere and is described here for the first time. These are E. X bowmanii
C. N. Page, hybr. nov. (E. sylvaticum L. Xx E. telmateia Ehrh.), from Hampshire, and FE. x
mildeanum Rothm. (E. pratense Ehrh. X E. sylvaticum L.) from Perthshire. Herbarium material of
each has been deposited in the herbarium of the Royal Botanic Gardens, Edinburgh (E).

EQUISETUM X BOWMANII

Equisetum < bowmanii C. N. Page, hybr. nov. (Equisetum sylvaticum L. X E. telmateia Ehrh.)
TYPUS: Hampshire, New Forest, Minstead, near A337 Minstead Road, July 1986, R. P. Bowman
s.n., det. R. P. Bowman & C. N. Page (Holotypus: E).

Caules 30-50 cm alti, 2-5—5-0 mm diam., fere albi, 8—14 sulcati; vaginae (dentibus exclusis) 5—7
mm, pallide viridi-griseae; dentes tot quot sulci, 2—costati, tenues, acuti, marginibus latis pallide
brunneis scariosis lateraliter inter se adhaerentibus: cavitas centralis 3-3 caulis diam. Ramuli
patentes, tenues regulariter verticillati, iterum ramosi, plerumque tetragoni vivide sed pallide
virides; costae biangulatae; vaginae pallidae, dentibus triangularibus-acuminatis patulis minute

nigro-apiculatis. Strobilus c. 1-2 cm longus; sporae abortivae.

Stems 30—50 cm or more, erect, thick, 2-5—5-0 mm in diameter; main stem internodes somewhat
fleshy, mostly very pale green, nearly white in the lower part of the stem, minutely rough; grooves
8-14, numerous, fairly deep, ridges + rounded, sheaths long, 5-0—7-0 mm (excluding teeth), +
appressed, pale greenish-grey, sometimes blackish above and below; teeth numerous, as many as
the grooves, 2-ribbed, of variable length, mostly 4-0—6-0 mm long, slender, acute, their central
portions deep brown, mostly straight, their margins broad, pale brown and scarious; the teeth often
adhering laterally by their margins in 2s—4s for their full length; central hollow about 3—2 diameter
of stem. Branches numerous, long (up to c. 7-5 cm), in regular and numerous whorls, spreading
widely all around the shoot and drooping at their tips, very slender, mostly 4-angled, bright pale
green, almost all bearing sparse to numerous long secondary branches of similar form and colour;
branch sheaths pale greenish-brown, teeth triangular-acuminate, green and dark-tipped, somewhat
spreading; branchlet ochreolae conspicuous, chaffy, pale golden brown; branch internode with
narrowly grooved (biangulate) ridges and shallowly rounded furrows.

Cones, when present, borne from the tops of shoots of generally similar form to the vegetative

274 ¢. NwPAGE

ones but the shoots thicker (mostly 5 mm diam.), with larger, looser, leafier sheaths to 2 cm long; the
cones ovoid, c. 1-2 cm long, dark, hardly rising out of the uppermost sheath, their sporangia poorly-
filled, the spores abortive.

The vegetative and cone-bearing shoots of this hybrid are both strikingly intermediate in size and
morphology between those of E. telmateia and E. sylvaticum. The vegetative shoots differ from
those of E. sylvaticum in their slightly larger size, and thicker and more succulent stems, and
especially in the long, narrow, dark-centred teeth to their sheaths, the paleness of their main stem
internodes (especially on the lower parts of the stem), and branches with + biangulate internodal
ridges and of bright pale green colour, forming the dominant colour of the plant. They differ from E.
telmateia in the much more slender main stem, with long nodding tip, the slenderness of the
branches which are themselves drooping at the tips and copiously branched, and roughness of the
main stem internodes and the depth of their grooves, and the conspicuous brown scarious margins to
the main stem teeth which are broad and tend to adhere laterally with each other, forming broad
lobes. The stem in section has a central hollow, and the number of carinal and vallecular canals
(about 8-14) is intermediate between those of the parents. The length of the first internode of the
branches is also intermediate between the short, first internode of E. telmateia and the long, first
internode of E. sylvaticum.

The intermediate form of the fertile (cone-bearing) shoots is also striking. These are semi-
dimorphic, with thick, somewhat succulent shoots, shorter than those of E. sylvaticum, and much
larger chaffier and looser sheaths than in E. sylvaticum, sparse whorls of branched green branches,
and a cone like that of E. telmateia in shape but similar to that of E. sylvaticum in size.

The sporangia of the cone of E. X bowmanii contain numerous but poorly-formed spores, which
appear to be 100% abortive. These include about 95% of very small, generally irregularly rounded,
misshapen or wholly shrivelled, nearly colourless spores, each with a shrunken and nearly colourless
cytoplasm and only very rudimentary and scarcely-formed elaters; and about 5% of varyingly larger
spores of imperfectly rounded form, but with varyingly pale green-coloured chloroplasts, and fairly
well-formed but inoperative elaters.

The spore evidence thus wholly supports the conclusion that this plant is of hybrid origin, whilst
its very distinctive morphology, which is almost exactly intermediate between that of two native
horsetail species in a large number of distinct characters, enables its parentage to be deduced with a
very high degree of certainty. These features also readily distinguish this hybrid from both parents
and from E. X mildeanum (see below), which shares E. sylvaticum as a common parent, as well as
from other hybrids in Equisetum (Page 1972, 1982).

DISTRIBUTION AND ECOLOGY

This hybrid was found by R. P. Bowman in July 1986 in a single colony spreading over heathland
road banks, grass verges and in adjacent short turf on wet clay soil near the A337 Minstead Road,
Minstead, New Forest, S. Hants., v.c. 11 (GR 41/296.105). Plants occur in open sites where they are
more slender and rather stunted and subject to grazing by ponies, except where they emerge in the
shelter of low Rubus fruticosus, Ulex, Prunus spinosa and Pteridium aquilinum, with some Quercus
robur seedlings. Other associates include Mentha aquatica, Cirsium palustre, Senecio aquaticus,
Betonica officinalis, Prunella vulgare, Galium palustre, Lysimachia nemorum and Juncus conglo-
meratus. Equisetum arvense occurs nearby on the same road banks. On the periphery of the site is a
light cover of medium-aged Quercus robur, Fagus sylvatica and Pinus sylvestris, with an open
understorey of stunted Jlex aquifolium. It is possible that a few trees were earlier removed during
the construction of the new section of road. The site dips slightly to the west and the road ditches
drain into a main ditch, in turn draining an open grazed sward. The altitude of the site is
approximately 35 m. On the southern side of the road, shoots of the hybrid occur along
approximately 30 m of verge, whilst on the northern side, they extend along at least 50 m of road
bank. The clay exposure is that of the Barton Clay of the Eocene.

There are four recorded sites for Equisetum sylvaticum in the area, all situated along the same
watercourse: Fleet Water, Minstead, on flushed slope in deciduous woodland; below mill pond
outfall (GR 41/295.100), 500 m distant from the hybrid site; Fleet Water, Newtown, Minstead, side
of damp ride in bushy plantation (GR 41/278.107), 1800 m distant; Stoney Cross, wet flush in

EQUISETUM HYBRIDS NEW TO BRITAIN 275

pasture (GR 41/260.117), 3650 m distant; and near Bartley Water, Busketts Lawn Inclosure, damp
sides of ditch draining into stream (GR 41/314.103), 1750 m distant.

Recorded but unlocalized sites for Equisetum telmateia exist at Minstead (GR 41/26.10) and
Canterton (GR 41/26.12), both about 3 km away, west and north-west of the hybrid site. In Busketts
Lawn Inclosure (GR 41/314.103), there is a local population of E. telmateia along ditch banks on
clay only 10 m from the very small population of E. sylvaticum described above. The habitat of this
site is beech and oak plantation, but the presence of Sorbus torminalis is indicative of a former
ancient woodland. Here E. telmateia grows mainly where the clay is exposed, whilst E. sylvaticum
prefers the damp slightly flushed ground, where some Sphagnum is present.

EQUISETUM X MILDEANUM

Equisetum < mildeanum Rothm. (E. pratense Ehrh. x E. sylvaticum L.).

Stems 10-35 cm, erect, slender, 0-8—1-5 mm in diameter; main stem internodes very pale green,
conspicuously and minutely rough with perpendicular setose spicules; grooves 8-15, numerous,
fairly deep, the ridges angular; sheaths long, 3-0—4-5 mm (excluding teeth), somewhat loose, pale
greenish-grey, sometimes pale brown to blackish above; teeth numerous, as many as the grooves, 1-
ribbed, 2-0—3-0 mm long, slender, acute, their central portions deep brown above, grey-green at the
base, straight or, when paired, bent laterally mostly only at the base, their margins scarious, mostly
pale brown, narrow near the tips of the teeth, but broadening rapidly downwards, adjacent teeth
mostly adhering laterally by their margins in 2s—3s for their full length; central hollow about 3
diameter of stem. Branches numerous, up to c. 5-5 cm long, whorled but becoming somewhat
asymmetrically arranged to give the shoot an often slightly bilaterally compressed appearance,
spreading and drooping at their tips, very slender, all 3-angled, mid-green, each usually bearing a
small number of regular, short, spreading secondary branches of similar form and colour; branch
sheaths pale green, teeth triangular-acuminate, green throughout, + straight or only slightly
spreading; branchlet ochreolae chaffy, pale brown; branch internodes with very prominent acute,
uniangulate ridges and deep, flat-sided V-shaped furrows each with a narrow, lengthwise basal
channel.

Vegetative shoots of this hybrid are strikingly intermediate in morphology between those of E.
pratense and E. sylvaticum. They differ from E. pratense most clearly in having, in most specimens,
secondarily branched branches throughout all or much of the length of the shoots, somewhat looser
main stem sheaths which bear fewer teeth with broader, whitish to pale brown margins which also
typically adhere together laterally by their margins in 2s and 3s. It differs from its other parent, E.
sylvaticum, principally in the more bilaterally compressed (and hence not quite radially symmetric)
habit of the overall shoot, its shorter, somewhat fewer, branchlets to the branches, the paler
ochreolae at the branchlet bases, the generally non-spreading and concolorously green habit of the
branch node teeth, the more numerous, more acute angles of the main stem internodes, and the
straighter, darker-centred, more acutely tapering teeth of the main stem sheaths. Its sheaths are also
generally tighter and typically somewhat less cup-shaped than are those of E. sylvaticum. Cone-
bearing shoots have not, as yet, been found, but overall, the slender, pale green, somewhat
compressed shoots of E. X mildeanum, combined with main stem sheaths with long dark teeth
adhering together by broad scarious margins, and branches which are themselves mostly sparsely
but regularly secondarily branched, leave little doubt about its identity, and produce a horsetail of
distinctive and delicate appearance. The combination of these characters distinguishes this hybrid
from both parents and from E. X bowmanii, which shares E. sylvaticum as acommon parent, as well
as from other hybrids in Equisetum (Page 1972, 1982).

DISTRIBUTION AND ECOLOGY IN BRITAIN

Three British stations for E. < mildeanum are known:

Perthshire, Ben Lawers, between large tumbled boulders on west-facing, slightly flushed damp,
grassy slope, c. 2050 ft (625 m), with both parents; July 1985, C. N. Page & D. Marden, det. C.N.
Page.

Perthshire, Caenlochan, in scree, 2900 ft (860 m); 10 July 1986, H. McHaffie, conf. C. N. Page.

276 C.N. PAGE

Perthshire, Glenshee: Glas Choire, 26 August 1986, E. Birse, comm. P. I. Rothwell, conf. C. N.
Page.

All three montane locations are in sites where both parents occur in the general vicinity or in the
immediate neighbourhood of the hybrids. All are in fairly base-flushed (and in one site, wet)
habitats over mica-schist rocks, and two of the sites are amongst boulders or on scree. In its Glas
Choire locality, the rather small shoots of the plant grow with Carex nigra, C. echinata and
Sphagnum recurvum. In the other two localities, plants grow in sites where, in addition to both
parents, other pteridophytes of upland, basic screes are frequent, including Dryopteris expansa,
Polystichum aculeatum and P. lonchitis.

DISCUSSION

Both hybrids reported here have Equisetum sylvaticum as one parent, and are the first two hybrids
involving this parent known in the Bnitish flora.

Equisetum X bowmanii is a hybrid hitherto unknown anywhere, and its two, rather dissimilar-
looking parents are, elsewhere, often widely separated geographically. In Europe as a whole, E.
sylvaticum is generally northern and montane, and E. telmateia southern and lowland (Jalas &
Suominen 1972). The ranges of two parent species thus mostly overlap only in middle latitudes in
Europe, where they are often further separated by habitat differences (Page 1982, 1988). Even in
Britain, their geographical overlap is not large (Jermy et al. 1978), and the New Forest locality for
this hybrid is in one of the few regions of Britain where the two parents seem to occur near to each
other in some quantity.

Here, the habitat of E. X bowmanii in many ways represents an abutment of habitats each more
typical of the parent species locally: E. telmateia on damp clay banks and E. sylvaticum in damp acid
heathy woodland. Further, the bulk of the habitat of E. x bowmanii occurs along a section of road
which was constructed anew as part of a road realignment 13 years before (1973) the date of the first
find of this plant, with the area of the hybrid closely coincident with that of roadside clay originally
disturbed. It is not known whether the hybrid was present before the road widening event, but it
seems probable that its formation may be a direct result of the soil disturbance resulting from this
activity, the road construction creating new damp clay verge sites initially free of competition from
established vegetation and suitable for gametophyte growth. It is by no means impossible that the
whole of the present population area, including plants on both sides of the road, is a single clone.

The only other hybrid involving E. telmateia in the British Isles is E. x font-queri Rothm. (E.
telmateia Ehrh. X E. palustre L.) known from the Isle of Skye (Page 1973) and from two lowland
English sites (Roberts & Page 1979; Page & Busby 1985). The damp, clay substrate of E. x
bowmanii and its roadside situation is strongly reminiscent of some of the sites for E. X font-queri in
Skye. The only other known hybrid involving E. sylvaticum in the British Isles (or, indeed
anywhere) is E. X mildeanum Rothm. (E. sylvaticum L. xX E. pratense Ehrh.), reported and
described here. E. X bowmanit is thus significant not only in its unusual appearance and in being a
hybrid between two species of very different morphology and ecology, but also in having two
parents neither of which seems to enter into many natural hybrid combinations.

Equisetum X mildeanum was described by Rothmaler (1944) from a number of localities in
central and northern Europe, viz. ““Hassia, Holsatia, Saxonia, Prussia, Silesia, Rossia’’. I have
examined and confirmed the identity of Rothmaler’s type (at Jena, J), and the match of these British
specimens to it. Through central and northern Europe and Asia, the two parents of this hybrid have,
indeed, very widely sympatric ranges, and their ecology is such that the two species probably
frequently meet. It is of interest that all three known Bnitish stations for E. x mildeanum reported
here are in Perthshire, thus all within the sympatric portions of their parental ranges (cf. Page &
Barker 1985), and indeed, Scotland is probably the only part of the British Isles in which E. pratense
occurs in any great abundance. Only one other hybrid involving E. pratense in its parentage is
known in the literature. This is E. X montellii Hiitonen (E. arvense L. X E. pratense Ehrh.),
reported from Finland, Sweden and the Canadian arctic (Duckett & Page 1975), and which could
also yet be found in the British Isles. These two finds now bring to eight the total number of native

“

EQUISETUM HYBRIDS NEW TO BRITAIN 277

British horsetail hybrids (Page 1982), and update the information and further emphasize the
ecological conclusions drawn in Page & Barker (1985).

ACKNOWLEDGMENTS

Grateful thanks must go to the various discoverers and communicators of the plants reported here:
E. Birse, R. P. Bowman, D. Marden and H. McHaffie. It is also a pleasure to acknowledge valuable
comments on the draft of the manuscript made by R. P. Bowman, A. C. Jermy, J. McNeill and B. L.
Burtt, and also to thank the latter for preparing the Latin description.

REFERENCES

-Ducketr, J. G. & Pace, C. N. (1975). Equisetum, in Stace, C. A., ed. Hybridization and the flora of the British
Isles, pp. 99-103. London.

JALas, J. & SUOMINEN, J. (1972). Atlas Florae Europaeae 1. Helsinki.

Jermy, A. C., ARNOLD, H. R., FARRELL, L. & PERRING, F. H. (1978). Atlas of ferns of the British Isles. London.

Pace, C. N. (1973). Two hybrids in Equisetum new to the British flora. Watsonia 9: 229-237.

Pace, C. N. (1982). The ferns of Britain and Ireland. Cambridge.

Pace, C. N. (1988). Ferns. Their habitats in the landscape of Britain and Ireland. London.

Pace, C. N. & BARKER, M. (1985). Ecology and geography of hybridization in British and Irish horsetails. Proc.
Roy. Soc. Edinb. 86B: 265-272.

Pace, C. N. & Bussy, A. R. (1985). Equisetum xX font-queri in Shropshire. Pteridologist 1: 72.

Roserts, R. H. & Pace, C. N. (1979). A second British record for Equisetum xX font-queri, and its addition to
the English flora. Fern Gaz. 12: 61-62.

ROTHMALER, W. (1944). Pteridophyten-Studien, 1. Feddes Repert. 54: 55-82.

(Accepted May 1988)

~

ca ~ rep
ea

Sie: tevin: 7

wie 8 Sbnes *

, TH lt ha!

et Sie

+ rae es

mi 0 sm he

re so J sity,

J
. ay
i r
yw i’ . 5 -£ in? eo
s 4 : gts! 3 tee } ei 4 elp-2
: ® —<s 5 a) i
xy bath ERO AER ov EBRR ebb
ror plein aan seshagsed oma eee ps
4 70 EI -_= a am vs wily x ‘ ue
J => > at ie 7 )
sieal bro nGess, \o papal’ raw a
- = A . :
GPO RRC SATIS eaters
=} | mE are
>t ; di
‘ee ae ‘ Muersatied- QUE
Ss aaah
S - : :
= . —i2 2
oe. i a ie ew
5 r, : ; oF
, =f eat 6
hx pint ote
- _——
; a VL SOR
° =f hm iS
. “——~ ra
- ‘ :
& °
i
%
i] ' £
= Hi : .
e 4 EB prs
< =
; civ us -*%
. = 2 “+ :
Ske 7 "fil
J ;
‘ -2, (he Seo
J win) Thi
te x a fos £ Biv
- , te VL thats r pater otal 2
é ‘ : a
; wc :
‘ ’ ¢ ,
| i
. +
F RCA ¢
- f th é . a“
« oe
f »
ha oe ,
_ a _ _ ye’

Watsonia, 17, 279-288 (1988) 179

The spread of Epilobium ciliatum Raf. in the British Isles

C.D, PRESTON

Institute of Terrestrial Ecology, Monks Wood Experimental Station, Abbots Ripton,
Huntingdon, PE17 2LS

ABSTRACT

Epilobium ciliatum Raf. (E. adenocaulon Hausskn.) was first collected in Britain at Cropston Reservoir,
Leicester, in 1891, and by 1894 it was well-established there. It was gathered in a further five vice-counties before
1930, but not correctly identified until the 1930s when G. M. Ash realized that it was naturalized in Surrey.
Thereafter it spread rapidly, reaching Wales by 1942, Scotland by 1957 and Ireland by 1958. This spread is
illustrated by a series of distribution maps; some of the problems it poses are discussed, and the possible reasons
for the success of E. ciliatum as a weed are briefly reviewed.

INTRODUCTION

Of the many thousands of plant species which have been deliberately or accidentally introduced into
the British Isles, relatively few have become widely naturalized. Some of these, notably certain
arable weeds, were introduced in prehistoric times, so long ago that they are difficult to distinguish
from native species. Silene alba, probably introduced with crops in the Neolithic (Baker 1947;
Godwin 1976), provides one such example. Other species spread in historic times. These include
plants, such as Acer pseudoplatanus, which were so widespread before systematic recording began
that their spread cannot be reconstructed from botanical records. Such records have, however, been
used to study the spread of those naturalized aliens which were introduced more recently. These
include Cardaria draba (Scurfield 1962), Elodea spp. (Simpson 1984), Polygonum and Reynoutria
spp. (Conolly 1977), Senecio squalidus (Kent 1956 et seg.) and Veronica filiformis (Bangerter &
Kent 1957, 1962). This paper documents the spread in the British Isles of Epilobium ciliatum Raf.
(E. adenocaulon Hausskn.), American Willow-herb, which is now one of the more widespread
weeds of alien origin.

Epilobium ciliatum is native to North America, from Newfoundland and Alaska south to Baja
California, Sonora, New Mexico and Virginia. In this area the name E. ciliatum covers a complex
which “‘is far from resolved taxonomically” (Raven & Raven 1976). It is an established alien on the
European mainland and in Australasia: the early records from these areas are reviewed by Lawalrée
& Reichling (1961) and Raven & Raven (1976) respectively.

The biology of E. ciliatum has been studied by Myerscough & Whitehead (1966, 1967). The seeds
can retain viability for several years. Seeds which germinate in autumn give rise to plants which
Overwinter as rosettes and produce flowering shoots in spring in response to increasing day length.
Seeds germinating in early summer can flower within 5-5 weeks. The flowers are automatically self-
pollinated, and the fruit capsules dehisce about 4 weeks after pollination. The seeds possess a tuft of
silky hairs, and are wind-dispersed. Plants perennate by means of leafy rosettes, which develop at
the base of the stem and produce their own adventitious root systems independent of the parent
plant. Despite this ability to perennate, Raven & Raven (1976) have noted that E. ciliatum often
behaves as an annual, relying on its abundant seed production for survival.

SOURCE OF RECORDS

This study is based on records held by the Biolegical Records Centre, Monks Wood Experimental
Station. In order to ensure that these are as complete as possible I have extracted records from the

280 C. D. PRESTON

following herbaria: ABRN, BM, CGE, E, GL, K, LIV, LTR, NMW, RNG, OXF. I have also
searched appropriate literature sources, including the Proceedings of the Botanical Society and
Exchange Club of the British Isles (from 1935), the B.S.B.I. Year Book, the Proceedings of the
B.S.B.1., Watsonia and many county and local Floras published up to the end of 1986.

THE SPREAD OF EPILOBIUM CILIATUM

In describing the spread of Epilobium ciliatum it is convenient to consider four arbitrarily chosen
time spans: 1891-1929, 1930-1949, 1950-1969 and 1970-1986.

1891-1929
The presence of Epilobium ciliatum in Britain was not announced until 1934, but subsequent
investigation has unearthed a number of earlier herbarium specimens which were not correctly
identified at the time of their collection. All the specimens cited below have been determined by
G. M. Ash.

The species was first collected by the Rev. T. A. Preston on 20 July 1891 at Cropston Reservoir,
Leics., v.c. 55 (BM). Preston collected the plant again in 1894 (BM, CGE, NMW). Both gatherings
were named as the hybrid E. obscurum X E. roseum, an identification “‘confirmed” by the Rev. E.
S. Marshall, then the British expert on the genus. Fortunately a detailed description of the habitat of
E. ciliatum at Cropston in 1894 is available (Mott 1895): ““On the muddy shores left dry by the
receding water of Cropstone Reservoir . . . sprang up vast beds of plants, sometimes a belt of one
genus followed lower down by a belt of a quite different genus. One of these was an Epilobium belt.
The species were hirsutum, parviflorum, montanum, obscurum and roseum, but hybrid forms
occupied by far the largest area, the individuals of the pure type being comparatively rare. Among
the hybrids, what appears to be obscurum xX roseum |i.e. E. ciliatum| was perhaps most
conspicuous. This belt being rather high on the sloping banks has probably been exposed for at least
two summers, so that it is possible the hybridizing may have taken place on the spot.” A. R.
Horwood collected E. ciliatum at Cropston in 1905 (CGE, NMW), noting that it “continues to
flourish on the silty ground . . . growing in great profusion, and reaching a height of 3—4 feet’’.

S. H. Bickham and R. F. Towndrow collected E. ciliatum at a second locality on 31 July 1905, a
timber yard at Malvern Link, Worcs., v.c. 37 (CGE). This collection was again named E. obscurum
x E. roseum, with the assent of E. S. Marshall. The next record was also from a timber yard, at
Woodchester, W. Gloucs., v.c. 34, where H. J. Riddelsdell found it in 1920 (NMW, OXF). In 1921
J. Fraser gathered it on heaths north of Woking (K) and in 1927 E. C. Wallace found it at Ballards
Plantation near Croydon (RNG); both localities are in Surrey, v.c. 17. Two more collections were
made in 1928, from Coleman’s Moor, Berks., v.c. 22 (G. C. Druce, OXF) and from a wood near
Meopham Green, Kent, v.c. 16(A. R. Horwood, K). This wood may have been Ryarsh Wood, asin
1929 Horwood collected E. ciliatum there. It seems to have been well-established, as there are 15
plants on the seven sheets at K gathered by Horwood in 1929.

Records of E. ciliatum made up to 1929 are plotted in Fig. 1.

1930-1949

G. M. Ash first collected Epilobium ciliatum in 1931. By the time that the plant was identified and its
discovery published, it was clear that the species was well-established in Surrey. During the period
1930-1934 there are records of E. ciliatum from seven 10-km squares in that county; it was also
recorded from five of the six neighbouring vice-counties, v. cc. 12, 13, 14, 16 and 22. Ash (1934)
described it as abundant in south-western Surrey, and “‘as common as any other Epilobium about
Witley, Godalming and Eashing”’. It grew in a range of habitats: ““damp woods, copses and along
stream-sides far from houses, as well as on railway-banks and in gardens, timber yards and waste
places” (Ash & Sandwith 1935).

During the next five years, 1935-1939, E. ciliatum was recorded from a further six vice-counties in
south-eastern England, v. cc. 11, 15, 18, 20, 23 and 24. In addition it was found further west, in v. cc.
6, 33 and 34, and further north, in v. cc. 31 and 39. G. M. Ash and J. F. G. Chapple also revisited the
original British locality, Cropston Reservoir, and found that E. ciliatum still persisted there and also
occurred at the nearby Swithland Reservoir.

=“

SPREAD OF EPILOBIUM CILIATUM

Ficure 2. Records of Epilobium ciliatum up to 1949.

Ficure 1. Records of Epilobium ciliatum up to 1929. Records are plotted in

10-km squares of the British and Irish national grids.

281

C. D. PRESTON

6961 01 dn wnjoy1d wnigojidq jo sp1000y ‘'p AUN

a a "y
t : the A ose

OHHH $FF646G644-
stess.
33°

ee
ee
J
(Bee
e-
e-
@it
ns

SOPHO SSS —HFFF6F464

Fete esesseen

re ¢

e
| e Or,
S03" °3, Hes .

e i ~
a : } | {
eo |e e |
“ + | ele i | ated ‘a
e > oa, »
x Q3 ae baa | |
\ | 6 , a
‘ os V 2 fod f |
\ / |
e K | > /<
e J |
\ | 5 wt \ | ¢ Senn
\e 2 ' ;
\@ oie poo
' | f)

oot a tu)

Povere re rer rr eum ak

ws

‘6S61 01 dn wanna wmigopidy Jo sp10d2y “¢ AANOLY

ve. i) oe 5
ePed 7 2:
~. ae * dol o 6
LY Ol De opera | ‘
ee a ” a iD
ote f “ 3) e a “E, = _—. a)

r)
° e cas 2
iti + ques <,
FE) aeage | g i
e e as es j Sobek
e|8 2 = <
e
Sa a
° ay NS “a ( pry \,
: ; t ®
:" 3
° ) ee
wv) ; 7 ( AN .
LJ ¢ ?
\ ti ae J
~e ua iS ) Aa
JLo ON K ?
‘ ~ : >
e / il 94 \¢ >
Z ( 1 Ny h oe
AO Vii
oh WHA
~* Way |
¥. a ; pig a
f eS “5 +0) oot a =a | 0
; nue eae ry
rpg oo! 0
2, 6. ao
| ty
=S Oe :
( Se . bf he &-
Se BU ae Oy
\ a a
e\ oF we
beb—tigs |

SPREAD OF EPILOBIUM CILIATUM 283

1950 - 1969

FH] 1930 - 1949
[]] 1970 - 1986

ise

BB 1891 - 1929

counties from which it has been recorded. Unshaded vice-counties are those

Ficure 6. The date of the first record of Epilobium ciliatum in the vice-
in which the species has not yet been found.

o* Ses Sesesstesee

Ses AS)
oe ee. Ki
ri : ® >
$ Tt —& e
é e ©. 6553
es oe
ee ee
rs / e
- VE e e ©
i e
e |
e@ee0e
e eo
ee
s | @
e e |

ee
° Se
an
2 20 e808 eocceces @ es oo
: eee eeccecccs & eecese ee cess
So e ecco @ @ eeeee eoosces

ee eecee
seeesteoesse
eee

FicureE 5. Records of Epilobium ciliatum up to 1986.

284 C.D. PRESTON

E. ciliatum continued to spread during the next decade, 1940-1949. It was first recorded in Wales
at Rookwood Hospital, Llandaff, v.c. 41, in 1942 and by 1947 it was established in Cardiff (Wallace
1949). All records up to 1949 are plotted in Fig. 2. The concentration of records around Surrey is still
apparent, and in 1946 J. P. M. Brenan commented that it was “‘as yet rare in Oxfordshire” (Brenan
1948). The westernmost records mapped in Fig. 2 are in Dorset, v.c. 9, and Glamorgan, v.c. 41, but
a specimen collected at Torrington in Devon, v.c. 4, in 1935 (J. F. G. Chapple et al., OXF) was
determined as E. ciliatum x E. obscurum by G. M. Ash (Hall 1936). Ash commented: ‘“‘on the
whole I consider the plants very convincing and | shall want much argument before I believe that
Ep. adenocaulon is not to be found near Torrington’”’.

There is no evidence that the spread of E. ciliatum was favoured by the wartime bombing of
London and other major cities. Lousley (1944, 1946) does not list it as a colonist of the bombed sites
in central Londen which he investigated in 1942 and 1944, nor is it referred to by Salisbury (1945).
Lousley did, however, collect it from three bombed sites in 1945. These were the first records of the
species from Middlesex, v.c. 21 — surprisingly, as it was known from all the surrounding counties by
1937.

Epilobium species hybridize readily, and by 1949 hybrids of E. ciliatum with all the native lowland
species had been recorded (Stace 1975).

1950-1969

The dramatic spread of E. ciliatum in this period is shown by the accompanying maps (Figs 2—4). At
the start it was still a relatively local species, albeit frequent in some areas of south-eastern England.
By 1969 it was frequent throughout most of southern and midland England and in many areas of
Wales, and there were scattered records in northern England and Scotland. The first Scottish record
was made in 1957 by Dr M. Dunn, by a wooded stream at Kenley, Fife, v.c. 85; the second was of a
population in “overgrown open woodland” on the banks of a stream at the Raith estate, Fife, where
it was discovered by G. H. Ballantyne in 1959. Between 1950 and 1969 it was also found on several
islands, including Lundy (1952), the Isle of Man (1958), Anglesey (1964) and Jersey, Channel
Islands (1968). In 1958 it was collected by J. G. and C. M. Dony at Arklow, Co. Wicklow, v.c. H20,
the first Irish record.

1970-1986

In Great Britain the spread of E. ciliatum continued during this period (Fig. 5). The species was
recorded for the first time in many Scottish vice-counties, and reached the Hebridean islands of Skye
(1976), Ulva (1982) and Tiree (1982).

Although it was first recorded in Ireland in 1958, there were few records before 1980. The hybrid
E. ciliatum X palustre was collected in Co. Down, v.c. H38, in 1969 and there is a 1971 field record of
E. ciliatum itself on an island in Lough Neagh, Co. Antrim, v.c. H39. In 1978 E. ciliatum was found
in a garden at Aughrim, Co. Wicklow, v.c. H20, where it grew with E. lanceolatum, and in a nearby
nursery (Walters 1979). In the next few years there were numerous records from all parts of Ireland,
and in some areas the species spread explosively (Doogue et al. 1985).

In south-eastern England, where it was first recognized as a naturalized plant, E. ciliatum was
very abundant by the 1970s. In Surrey it was ‘“‘almost ubiquitous” and Lousley (1976) described it as
“by far our commonest willowherb’”’. It was ‘‘the commonest willow-herb in London” (Burton 1983)
and ‘‘much the commoner of the small willowherbs” in Kent (Philp 1982); in both these areas, as in
Bedfordshire (Dony 1976), it was recorded in 70-78% of tetrads (2 xX 2 km squares). In
Hertfordshire (Dony 1967) and Sussex (Hall 1980) it is known in 62% and 64% of tetrads, but
further north and west the percentage is smaller: 47% in Devon (Ivimey-Cook 1984), 43% in
Rutland (Messenger 1971) and only 22% in Shropshire (Sinker et al. 1985).

ERRONEOUS RECORDS
In compiling the records of Epilobium ciliatum I have rejected a number as erroneous. The
following published records are sufficiently important for the reason for their rejection to be put on
record.

Cadbury et al. (1971) report as the first Warwickshire record a specimen in herb. J. E. Bagnall
collected by Bagnall in a wood near Gannaway Gate in 1889. If substantiated this would be the first
British record, but the specimen cannot be traced in Bagnall’s herbarium at BIRA, nor are the

SPREAD OF EPILOBIUM CILIATUM 285

details known to Mrs P. Copson, the vice-county recorder. It seems likely that the reference is an
error introduced in the compilation of the Flora. The earliest acceptable record from Warwickshire
known to me is dated 1952.

The first Sussex record is given by Wolley-Dod (1937) as Lurgashall, 1930, det. G. M. Ash, but
Ash & Sandwith (1935) cite ‘““Lurgashall, 1934” as the only Sussex record. There are no Lurgashall
specimens in BM, where both herb. Wolley-Dod and herb. Ash are lodged. I prefer to follow Ash &
Sandwith, and to regard Wolley-Dod’s date as a misprint.

Epilobium ciliatum was reported from the dried up bed of a loch at Gallanach Dunes, Coll, v.c.
103 (Harrison et al. 1941) and from the Bornish-Stoneybridge area, South Vist, v.c. 110 (Harrison
1941). Harrison’s (1941) comment that EF. ciliatum “is an American species and may not be a
colonist in the Hebrides” suggested that it might be native at these sites. However these records
have not been accepted by other botanists (e.g. Perring & Walters 1962).

DISCUSSION

HISTORY OF SPREAD IN THE BRITISH ISLES

The records of Epilobium ciliatum in Britain and Ireland are summarized above. The first question
that needs to be considered is whether or not these accurately reflect its spread in the British Isles.
The genus Epilobium is somewhat difficult and is sometimes ignored by field botanists. The smaller
species cannot be distinguished without close scrutiny, and unusual plants can be explained away as
hybrids. If E. ciliatum was overlooked for a considerable period or recognized by only a few
botanists, the picture of its spread presented above might be misleading.

Until the early 1930s botanists were unaware that EF. ciliatum occurred in Britain. A few
specimens collected before then have been found, but it would be unwise to deduce too much from
this fragmentary evidence. E. ciliatum was well-established at Cropston Reservoir by 1894. It is
difficult to say whether it was established or casual at the other localities where it was collected
before 1930, and the date when it became established in Surrey is not known. As Ash & Sandwith
(1935) pointed out, the Rev. E. S. Marshall lived in the very area of Surrey in which E. ciliatum was
well-established by the 1930s. Marshall was successively curate at Witley (1885-1890) and Vicar of
Milford (1890-1900), and it was in this period that he carried out most of his work on Epilobium
(Britten 1920). Although Marshall failed to recognize the pressed specimens of E. ciliatum that were
submitted to him, it seems rather less likely that he would have overlooked the living plant and
highly probable that he would have at least collected it in mistake for some other species or hybrid.
It is tolerably certain, therefore, that E. ciliatum became established in Surrey between 1900 and
1930.

The first detailed account of E. ciliatum in Britain was published in 1935. Even after this there
must have been some delay between the appearance of the plant in an area and its recognition by
botanists, and no doubt this delay was greater in some regions than in others. Even the map of the
current distribution (Fig. 5) shows some areas where the scarcity of records probably reflects a lack
of recorders rather than any real scarcity of the plant. With these qualifications, it seems likely that
the records available do provide a reasonably accurate picture of the spread in Great Britain of
E. ciliatum after 1935. The extent to which Irish records reflect its spread there is discussed by
Doogue et al. (1985).

A number of other questions can be asked about the origin of E. ciliatum in Britain. Did it arrive
directly from North America? Were the scattered early records the results of separate introduc-
tions? Is the fact that two of its first three localities were timber yards significant, indicating that it
may have been introduced or spread with timber?

Raven & Raven (1976) pointed out that E. ciliatum in Britain and New Zealand appears to be
identical, despite the fact that they are drawn from a highly variable complex in N. America, and .
suggest that the New Zealand plant might have been introduced from Britain. E. ciliatum was first
recorded in New Zealand in 1896. By 1905 it seems to have been widespread in North Island, and it
was first recorded in South Island in 1912. Thus the species was found in Great Britain five years
before it was collected in New Zealand, but appears to have spread more rapidly in New Zealand.
The possibility that it was introduced from New Zealand to Britain, rather than vice versa, cannot
be ruled out.

286 C. D. PRESTON

This raises perhaps the most mysterious question about the spread of E. ciliatum in Britain, first
posed by Ash (1953) who wrote “Why has it made such rapid progress in recent years whilst
remaining more or less static near Leicester for 50 years?’ Although E. ciliatum was well-
established and persistent at Cropston Reservoir there is no evidence to suggest that it spread
beyond the immediate neighbourhood. It did not become a weed in T. G. Tutin’s Leicester garden
until 1965 (Tutin 1973). The distribution maps suggest that the populations near Cropston were
engulfed by the advancing front of E. ciliatum from further south. It is tempting to suggest that the
plants at Cropston had a different genotype from those which spread so rapidly, but this is only
speculation.

After the initial scattered records, the spread of FE. ciliatum in Britain appears to have been
continuous rather than discontinuous. This is suggested by Fig. 6, which illustrates the date of the
first record from each vice-county. This continuous spread presumably reflects a natural expansion
of an established population, rather than introduction by man at different sites which then acted as
foci for secondary expansion. The spread of E. ciliatum contrasts with that of Senecio squalidus,
described by Salisbury (1961) as “markedly discontinuous with respect to the more remote
infections, whilst at the same time spreading locally around each new station’”’.

POSSIBLE REASONS FOR THE SUCCESS OF E. CILIATUM AS A WEED

Myerscough & Whitehead (1966, 1967) compared the germination behaviour and subsequent
growth of Epilobium ciliatum with that of the native species E. montanum. Both species possess
many of the attributes of successful weeds: considerable phenotypic plasticity which enables them to
grow in a range of environmental conditions, efficient vegetative and sexual reproduction and seeds
with the capacity to survive in a dormant state. However, the minimum period from germination to
flowering is shorter in E. ciliatum, 5-5 weeks compared to 7, and its relative growth rate is greater.
The overwintering rosettes of E. ciliatum are larger than those of E. montanum, which probably
allows it to make more growth during the winter although it renders the plant more susceptible to
adverse weather conditions. Flowering and fruiting plants of E. ciliatum are less leafy than those of
E. montanum, and the green stems and capsules make a greater contribution to photosynthetic
assimilation. Raven & Raven (1976) stress the ““enormous seed production” of E. ciliatum as a
factor in its spread in Europe and Australasia, but quantitative comparisons with other species have
not been made.

It has recently been suggested that the evolution of resistance to the widely used persistent
triazine herbicides such as simazine might be one factor responsible for the predominance of E.
ciliatum in urban areas. S. M. Walters (pers. comm.) found that it was the only species which
flourished on soil in an area of Grantchester churchyard, Cambs., which had been subjected to
excessive applications of simazine, and resistance to pre-emergence treatments has been demon-
strated experimentally (Bailey et al. 1982).

ACKNOWLEDGMENTS

I am grateful to the vice-county recorders and other B.S.B.I. members who have helped by
providing records of E. ciliatum, to T. D. Pennington for examining a number of specimens, to Mrs
D. M. Greene for much assistance including the production of the 10-km square distribution maps
and to J. Clark for drawing the vice-county map. S. M. Walters and D. Briggs pointed out the
possible importance of herbicide resistance in the E. ciliatum story, and J. R. Akeroyd, M. D.
Hooper and T. C. G. Rich kindly commented on a draft of this paper.

REFERENCES

Asu, G. M. (1934). Epilobium adenocaulon Hausskn. Rep. Watson bot. Exch. Club 4: 218-219.
Asu, G. M. (1953). Epilobium adenocaulon in Britain (Exhibit), in Loustey, J. E., ed. The changing flora of
Britain, p. 168. Arbroath.

-

SPREAD OF EPILOBIUM CILIATUM 287

Asu, G. M. & SANpwitH, N. Y. (1935). Epilobium adenocaulon Hausskn. in Britain. J. Bot., Lond. 73: 177-
184.

Baitey, J. A., RicHarpson, W. G. & Jones, A. G. (1982). The response of willowherbs (Epilobium
adenocaulon and E. obscurum) to pre- or post-emergence herbicides. Proc. Br. Crop Prot. Conf. Weeds
1982: 299-305.

Baker, H. G. (1947). Melandrium (Roehling em.) Fries, in Biological Flora of the British Isles. J. Ecol. 35: 271-
292.

BANGERTER, E. B. & KENT, D. H. (1957). Veronica filiformis Sm. in the British Isles. Proc. bot. Soc. Br. Isl. 2:
197-217.

BANGERTER, E. B. & KENT, D. H. (1962). Further notes on Veronica filiformis. Proc. bot. Soc. Br. Isl. 4: 384-
397.

BrENAN, J. P. M. (1948). September 6-9, the neighbourhood of Oxford. Rep. botl Soc. Exch. Club Br. Isl. 13:
223-230.

Britten, J. (1920). In memory of Edward Shearburn Marshall (1858-1919). J. Bot., Lond. 58: 1-11.

Burton, R. M. (1983). Flora of the London area, p. 81. London.

Capsury, D. A., Hawkes, J. G. & READETT, R. C. (1971). A computer-mapped Flora, p. 171. London.

. Cono ty, A. P. (1977). The distribution and history in the British Isles of some alien species of Polygonum and

Reynoutria. Watsonia 11: 291-311.

Dony, J. G. (1967). Flora of Hertfordshire, p. 74. Hitchin.

Dony, J. G. (1976). Bedfordshire plant atlas, p. 57. Luton.

Doocug, D., Ketty, D. L. & Wyse Jackson, P. S. (1985). The progress of Epilobium ciliatum Rafin. (E.
adenocaulon Hausskn.) in Ireland, with some notes on its hybrids. Jr. Nat. J. 21: 444-446.

Gopwin, H. (1975). The history of the British flora, 2nd ed. Cambridge.

HA.t, P. C. (1980). Sussex plant atlas, p. 91. Brighton.

HA_t, P. M. (1936). Epilobium adenocaulon Hausskn. X obscurum Schreb. Rep. botl Soc. Exch. Club Br. Isl.
bE: ,29.

Harrison, J. W. H. (1941). A preliminary flora of the Outer Hebrides. Proc. Univ. Durham phil. Soc. 10: 228-
Z13.

Harrison, J. W. H., Harrison, H. H., CLark, W. A. & Cooke, R. B. (1941). The flora of Coll, Tiree and
Gunna (v.-c. 110B). Proc. Univ. Durham phil. Soc. 10: 274-308.

IvimEy-CooK, R. B. (1984). Atlas of the Devon flora, p. 114. Exeter.

KENT, D. H. (1956). Senecio squalidus L. in the British Isles — 1. Early records (to 1877). Proc. bot. Soc. Br. Isl.
2: 115-118.

LAWALREE, A. & REICHLING, G. (1961). Epilobium adenocaulon Hausskn., au Grand-Duché de Luxembourg,
en Belgique et en Allemagne occidentale. (Service de la carte des groupements végétaux, Musée d’histoire
naturelle, no. 41). Luxembourg.

LousLey, J. E. (1944). The pioneer flora of bombed sites in central London. Rep. botl Soc. Exch. Club Br. Isl.
12: 528-531.

LousLey, J. E. (1946). The flora of bombed sites in the City of London in 1944. Rep. botl Soc. Exch. Club Br. Isl.
12: 875-883.

Lous_ey, J. E. (1976). Flora of Surrey, pp. 208-209. Newton Abbot.

MESSENGER, [K.]G. (1971). Flora of Rutland, p. 64. Leicester.

Morr, F. T. (1895). E. obscurum X roseum. Rep. botl Soc. Exch. Club Br. Isl. 1: 449.

MyerscouGH, P. J. & WHITEHEAD, F. H. (1966). Comparative biology of Tussilago farfara L., Chamaenerion
angustifolium (L.) Scop., Epilobium montanum L. and Epilobium adenocaulon Hausskn. I. General
biology and germination. New Phytol. 65: 192-210.

MyerscouGH, P. J. & WHITEHEAD, F. H. (1967). Comparative biology of Tussilago farfara L., Chamaenerion
angustifolium (L.) Scop., Epilobium montanum L., and Epilobium adenocaulon Hausskn. II. Growth and
ecology. New Phytol. 66: 785-823.

PERRING, F. H. & WALTERS, S. M., eds (1962). Atlas of the British yore: p. 145. London.

Pup, E. G. (1982). Atlas of the Kent flora, p. 94.

RAVEN, P. H. & Raven, T. E. (1976). The genus Epilobium nteaeac) in Australasia: a systematic and
evolutionary study. Bull. N. Z. Dep. scient. ind. Res. 216. 321 pp. Christchurch.

SALISBURY, E. J. (1945). The flora of bombed areas. Proc. R. Instn Gt Br. 32: 435-455.

SALISBURY, E. J. (1961). Weeds and aliens. London.

ScuRFIELD, G. (1962). Cardaria draba (L.) Desv., in Biological Flora of the British Isles. J. Ecol. 50: 489-499.

Simpson, D. A. (1984). A short history of the introduction and spread of Elodea Michx in the British Isles.
Watsonia 15: 1-9.

SINKER, C. A., PACKHAM, J. R., TRUEMAN, I. C., OSWALD, P. H., PERRING, F. H. & Prestwoop, W. V. (1985).
Ecological Flora of the Shropshire region, p. 224. Shrewsbury.

Stace, C. A. (1975). Epilobium L., in Stace, C. A., ed. Hybridization and the flora of the British Isles, pp. 246-
265. London.

288 C. D. PRESTON

TutTin, T. G. (1973). Weeds of a Leicester garden. Watsonia 9: 263-267.

WaLLAce, E. C., comp. (1949). Plant records. Watsonia 1: 37-61.

Watters, S. M. (1979). Epilobium lanceolatum Seb. & Mauri — a plant to look for in your garden. Watsonia 12:
399.

Wo tey-Dop, A. H., ed. (1937). Flora of Sussex, p. 191. Hastings.

(Accepted April 1988)

Watsonia, 17, 289-299 (1988) 289

The taxonomic relationships and typification of Festuca
brevipila Tracey and F. lemanii Bastard (Poaceae)

M. J. WILKINSON and C. A. STACE

Botany Department, The University, Leicester, LE] 7RH

ABSTRACT

_ Festuca brevipila Tracey (= F. trachyphylla (Hack.) Kraj.) and F. lemanii Bast. have long been a source of
confusion for British botanists. The aim of this paper is to clarify the current taxonomic situation by
distinguishing the two species on the basis of their morphology and anatomy, by reviewing the taxonomic history
and nomenclature and by examining their taxonomic status. Some data are also presented on the geographical
distribution and ecology of each species.

INTRODUCTION

The subject of this paper is two species of grass that have for long been collectively known in the
British literature as Festuca longifolia Thuill. It was shown conclusively by Auquier (1973) that this
application is wrong, and that F. /ongifolia is the correct name for the glaucous fescue found in East
Anglia, to which the names F. caesia Smith and F. glauca Lam. had been applied. F. caesia correctly
refers to that taxon (though a later synonym), but F. glauca is another misapplication, the true F.
glauca not occurring in the wild in the British Isles.

The correct names of the two species to be discussed are F. brevipila Tracey and F. lemanii
Bastard. The misapplication of the name F. longifolia to these seems to date from Howarth (1925),
who was followed by Hubbard (1954), Dandy (1958) and Tutin (1952, 1962). These authors
included both F. brevipila and F. lemanii as a single species, although the former is probably more
closely related to F. valesiaca Schleicher ex Gaudin and the latter to F. ovina L. The drawing in
Hubbard (1954) for the most part could refer to either species, but the leaf section is definitely that
of F. lemanii.

The long list of differential characteristics which separate F. /emanii and the only slightly less
variable F. brevipila (Table 1) suggests that the species are very well separated. However, Hackel

TABLE 1. CHARACTERS WHICH SEPARATE F. LEMANII AND F. BREVIPILA

Character F. brevipila F. lemanii

Panicle usually nodding usually erect

Awn length (mm) 1-4-2-5 0-3-1-8

Palea length (mm) 3-8—4-7 ' 4-0-5-2

Leaf-blade scabridity usually scabrid or scabrid distally usually scabridulous distally

Number of adaxial leaf-blade 4-6 2-4

furrows

Number of leaf-blade veins (S—)7-9 5-7

Margins of leaf-blades infolded infolded or not

Sclerenchyma in leaf-blades usually in three islets (sometimes usually in ring or broken ring
semi-connected)

Prickle length on adaxial surface of | (2-7—)5—13(-15) (10-5—)23-—46(-63-5)

leaf-blade (um)

Stomatal length (leaf) (um) (37-5—)41-50 38—42(—46)

Stomatal length (lemma) (um) 31-5-37°5 29-33:°5

290 M. J. WILKINSON AND C. A. STACE

(1882, p. 87) noted the existence of individuals intermediate between F. ovina subsp. eu-ovina var.
vulgaris subvar. firmula Hack. (= F. lemanii) and F. ovina subsp. eu-ovina var. duriuscula subvar.
trachyphylla Hack. (= F. brevipila). The present study has also uncovered a number of intermediate
specimens, although completely intermediate individuals are rather rare and the taxa are generally
easily separated when all differential characters are taken into consideration. We therefore consider
them to be best recognised as distinct species.

FESTUCA LEMANII BASTARD

F. lemanii Bastard, Essai Fl. Maine et Loire 36 (1809). TYPE: France. Chinon. In sabulosis aridis.
E. H. Tourlet No. 12530 — ex herb. Hackel (Neotype: W, designated here).

F. cinerea sensi Desvaux, Fl. Anjou 61 (1827); sensu Moisan, Fl. Nantaise 185 (1839); non Vill.
(1787).

F. duriuscula sensu Moisan, Fl. Nantaise 185 (1839); sensu Guépin, Fl. Maine et Loire, 3rd ed., 50
(1845); sensu Boreau in Mém. Soc. Acad. Maine et Loire 6: 181 (1859); sensu Lloyd, Fl. Ouest
France, 3rd ed., 372 (1876); sensu Corbiére, Nouv. Fl. Normandie 645 (1893); sensu Desvaux, FI.
Anjou 61 (1927); sensu Huon in Bot. Rhed., sér. A, 9: 187 (1970); non L. (1753).

F. glauca sensu Moisan, Fl. Nantaise 185 (1839); non Lam. (1787).

F. duriuscula L. var. hirsuta sensu Hardouin, Renou & le Clerc, Catal. Pl. vasc. Calvados 307
(1848); sensu Tourlet, Catal. rais. Pl. vasc. Indre et Loire 583 (1908); non (Host) Gaudin (1828).

F. ovina subsp. eu-ovina var. vulgaris Koch subvar. firmula Hack. f. lemanii (Bast.) Hack., Monogr.
Fest. Europ. 87 (1882); F. ovina subsp. ovina var. firmula (Hack.) Hegi f. lemanii (Bast. ) Hack. in
Stohr, Wiss. Z. Martin-Luther-Univ. Halle-Wittenberg 4: 405 (1955).

F. hirsuta sensu Corbiére, Fl. Normandie 645 (1893); non Host (1802).

F. ovina subsp. eu-ovina var. lemanii (Bast.) Aschers & Graebn., Syn. Mitteleur. Fl. 2: 468 (1900).

F. longifolia sensu Howarth in J. Linn. Soc. (Bot.) 47: 35 (1925), pro parte; sensu Hubbard, Grasses
111 (1954), pro parte; sensu Tutin, F/. Br. Isl., 2nd ed., 1129 (1962), pro parte; non Thuill. (1800).

F. ovina var. firmula (Hack.) Richt. subvar. /emanii (Bast.) Kraj. in Acta Bot. Bohem. 9:189, 193
(1955).

F. cinerea Vill. var. trachyphylla sensu Stohr in Wiss. Z. Martin-Luther-Univ. Halle-Wittenberg 4:
732 (1960); sensu Auquier in Bull. Jard. Bot. nat. Belg. 39: 97-118 (1969); non (Hack.) Stohr.

TAXONOMIC HISTORY

The type specimen of F. lemanii was described by Bastard as being collected from the area around
Brissac and Saumur (Maine et Loire, France). The herbarium of Bastard is said to be mainly in
Angers (ANG) with other specimens in Paris (P) and Manchester (MANCH). M. Kerguélen (in litt.
1984) examined the collections in ANG and P but failed to find either a type specimen or an
authentic specimen which may become a neotype. Auquier (1974) reported having requested such
material from MANCH and ANG but without success. M.J.W. visited MANCH but also failed to
find a suitable specimen. In the apparent absence of a type specimen or of an authentic specimen to
serve as a neotype, Kerguélen (in litt. 1984) suggested that an alternative neotype might be sought.

Hackel’s (1882) concept of the epithet ‘/emanii’ came from the examination of herbarium material
collected from areas around Chinon (Indre et Loire, France) by E. H. Tourlet. Auquier (1974)
examined one such specimen from Hackel’s herbarium (W, No. 12530) collected by Tourlet from
“Chinon in sabulosis aridis’’, and stated that it agrees well in the majority of characters with plants
of the thermophilous calcareous grasslands in the region (and by implication with F. lemanii). He
later suggested, in a note attached to the specimen, that this plant may provide a convenient neotype
for F. lemanii. Kerguélen (in litt. 1984) agreed with this view.

We have observed F. lemanii growing on both calcareous and acidic soils and a detailed
comparison of the two ecodemes covering 126 characters has revealed few differences between
them. Traditionally important characters such as spikelet length, awn length and leaf-blade
diameter were not found to differ between the two ecodemes, but minor differences were observed
in both panicle length and culm height (Fig. 1). These differences are considered by us to be too
small to warrant taxonomic distinction, since both characters were found to be highly unreliable
under cultivation. Therefore, we feel that it is not inappropriate to use a specimen collected from

“

TAXONOMY OF F. BREVIPILA AND F. LEMANII 291

5 | |
/ 0
Breath wide 0g 735 Luo 0-5 0-6 0-7 0-8 0-9

6 7 8 leaf diameter

(mm)
Spikelet length (mm)

B

5
0
0-2 0-6 1:0 1:4 1:8 2:2
Awn length (mm)

0
0-5 06 0-7 0-8 0-9 1:0

B 5 leaf diameter
Becxas lahiag coe api eSA | (mm)

0-2 0-6 1-0 1:4 1:8
Awn length (mm) ~

A

ea ana )
A
5
0
10 20 30 40 50
Height (cm) 0

2G 4 62060 <-TUSBUBO10 At
Panicle length (cm)

B - B

! 5
JL
ir 2092) 30 4Oe 6d, 60 7008

DIZ AL Be Sty Ov8. N19 F110 641
Panicle length (cm)
Height (cm)

Ficure 1. Variation in spikelet length, awn length, leaf-diameter, height and panicle length in two ecodemes of
| F. lemanii, A. calcifuge, B. calcicole. Vertical axes = no. of plants.

292 M. J. WILKINSON AND C. A. STACE

acid sands as a neotype for F. /Jemanii, even though the original plant was collected from an area of
predominantly calcareous soils.

The Tourlet specimen was obtained on loan from W and examined. The characteristics of this
specimen agree reasonably well with the descriptions of F. lemanii given by Auquier (1974),
Markgraf-Dannenberg (1980) and Kerguélen (1983) (Table 2). Accordingly, we designate this
specimen as the neotype of F. lemanii.

Hackel (1882, p. 87) included F. /emanii under F. ovina var. vulgaris subvar. firmula. Plants
referable to F. /emaniti were distinguished from others in the subvariety only by their hairy spikelets.
However, the modern concept of F. /emanii encompasses plants with both hairy and glabrous
spikelets and so is more accurately referable to the whole of Hackel’s subvar. firmula.

‘ TABLE 2. CHARACTERISTICS OF F. LEMANII

Neotype Markgraf-
Character (W-N.12530) Auquier (1974) Kerguélen (1983) Dannenberg (1980)
Panicle length (4-6—)5-7(-7-5) (2—)5-7(-11) (2-)5-7(-11) (2—)4-7(-11)

(cm)
Spikelet length (6-3—)6-6(—7-2) (5-6—)6-5-7-2(-8-3) (5-6—)6-5-7-2(-8-3) (5-6-)6-5—7-2(-8-3)
(mm)
Lemma length (3-8—)4-1(-4-5) (3-6—-)4-2-4-6(-5-5) (3-6-)4-2-4-6(-5-5) (3-6—)4-2-4-6(-5-5)
(mm)

Awn length (0-7—)1-5(-2-3) (0-2—)0-8—1-6(—2-8) — <2-75

(mm)

Upper glume (3-6—)3-8(-—4-3) (2-8—)3-5—4-0(-5:-2) — (2-8—)3-5—4-0(-5-2)
length (mm)

Leaf-blade scabridulous in usually scabridulous scabrid only at tip usually scabrid distally
scabridity distal third distally or at tip

Leaf-blade (0-59—)0-62 (0-43—)0-62—0-79 (0-45—)0-6—0-8(—1-14) (0-43—)0-62-0-79
diameter (—0-66) (-1-14) (-1-14)

No. of veins 5(-7) (5—)7(-9) (5—)7(-9) (5—)7(-9)

No. of grooves 2(-4) 2-4(-6) (2—)3-4 (2-)3-4
Sclerenchyma much brokenmng continuous or broken continuous ring continuous or broken
distribution ring ring

MORPHOLOGICAL VARIATION

Plants fairly densely to loosely tufted; vegetative shoots all intravaginal and retaining or shedding
old leaves. Culms (19—)28—66 cm, erect, (0-32—)0-38—0-5(—0-55)mm wide below the first panicle
node, generally smooth or scabridulous below the panicle, glabrous or hairy, grooved or weakly
grooved distally; nodes 2—3, the uppermost not pruinose, visible beyond the subtending sheath or
not, and reaching (3—)12—32(—37)% of the height of the plant.

Leaves green, usually not glaucous, subpruinose or not. Leaf-sheaths (1-3—)2-1—4-4(—4-7)cm,
fused for (0—)15—36(—41)% of their length, smooth or scabridulous distally, glabrous or hairy;
auricles short, usually minutely ciliate (rarely glabrous). Ligules short, usually minutely ciliate
(rarely glabrous). Leaf-blades lank to robust, (4—)6—11(-—17)cm, not acutely pointed, (9—)11—56(-
79)% on each plant curved at the tip, usually scabrid at least in the upper half, occasionally
scabridulous in the distal third or smooth, glabrous or hairy at the base.

Panicles usually erect but occasionally slightly nodding, 3-7—7-8(—8.6)cm, with (8—)16—24(—34)
spikelets. Panicle branches subpruinose or not pruinose, not narrowing below the spikelets, smooth
to scabrid, covered in hairs or prickles, rarely subglabrous; lst and 2nd panicle nodes (1-1—)1-3-
2-7(—3-2)cm apart. Spikelets subpruinose or not, (6-1—)6-5—7-5(—8-5)cm, with (2—)3—6(—8) florets
(plus one sterile floret); pedicels (0-62—)1-45—2-35(—2-5)mm. Glumes unequal; the lower subulate
to narrowly lanceolate, (2-15—)2-4—3-2(—3-52) x 0-58—0-82 mm, usually glabrous, but sometimes
with a scabrid tip or rarely laxly hairy, 1-veined, with a ciliate margin; the upper narrowly lanceolate
to lanceolate, (3-4—)3-5—4-6(—4-8) x (0-89—)1-05—1-2(—1-3)mm, usually glabrous or with a scabrid

“

TAXONOMY OF F. BREVIPILA AND F. LEMANII 293

distal third, rarely laxly hairy, 3-veined with a ciliate margin or rarely glabrous. Lemmas lanceolate,
(4-05—)4-1—4-85(—5-35) (excluding awn) x 1-5—1-83(—1-93)mm, green but often purple at the apex,
with awns (0-3—)0-6—1-6(—1-9)mm long, 5-veined, scabrid in the distal half or hairy. Paleas linear-
lanceolate, (3-9—)4-1—4-6(—5-2) x 0-68—0-9 mm. Anthers yellow or purple, 1-83—2-Smm. Pollen
grains oblong, brown, 2-3—3-1 x 0-65—1-0mm.

Leaf-blade anatomy (Fig. 2): outline of leaf-blades usually oval or V-oval, occasionally V-form or
circular-elliptical; margins usually not infolded; leaf-blade diameter (0-43—)0-59-0-85(—0-95)mm,
diameter/thickness ratio (1-87—)2-1—2-4(—2-9); veins 5 (—7); sclerenchyma forming a complete or
broken ring which may be thickened slightly at the margins and midrib, sometimes forming a much
broken ring; 2—4 grooves and 1—4 ridges on the adaxial surface; adaxial midrib (0-09—)0-1—0-14(-
0-15)mm wide and 0-035—0-08mm deep; adaxial epidermal cells usually of fairly uniform size, rarely
individualized, lacking bulliform cells but with prickles (10-5—)23—46(—63-5)um. Leaf-blade adaxial
epidermis: stomata (31—)38—41-5(—45-8)um, usually predominantly solitary or with an accompany-
ing prickle-cell, (1-)8—74(—91)% solitary, (0—)23-—88(—99)% with an accompanying silica-cell;

prickles usually predominantly solitary, (28—)35—79(—100)% solitary, (0O—)11—50(-—57)% forming

short rows of 2 or more prickle-cells, 0—15(—65)% with an accompanying silica-cell.
Leaf-blade abaxial epidermis: stomata absent; long-cells with sinuous walls, (75—)85—133(-
143) um; silica-cells and cork-cells present; prickles usually present.

FESTUCA BREVIPILA TRACEY

F. brevipila Tracey, Plant Syst. Evol. 128: 287—292 (1977). TYPE: Nieder6ésterreich, Waldviertel,

Ampliendorf (nérdlich von Schrems), in der Nahe des Houses Nr. 55; in sehr flachgriindigen Rasen

iiber Granit. 20 Mai 1977, A. Weber (Holotype: WU).

F. ovina L. var. glaucescens Link, Hort. Berol. 2: 266 (1833), nom. ambig. sec. Stohr (1960, p. 398).

F. ovina subsp. eu-ovina var. duriuscula (L.) Koch subvar. trachyphylla Hack., Monogr. Fest.
Europ. 91 (1882) (Lectotype: Prenzlau, Grantzow, ex herb. Hackel (W), designated here).

F. ovina subsp. eu-ovina var. duriuscula subvar. pubescens Hack., Monogr. Fest. Europ. 91 (1882).

F. duriuscula var. trachyphylla (Hack.) Richt., Pl. Eur. 1: 94 (1890).

F. duriuscula subsp. trachyphylla (Hack.) Rohlena in Vestn. Kral. ces. spol. Nauk 24: 4 (1900), sec.
Rauschert (1960, p. 270).

F. ovina var. trachyphylla (Hack.) Druce, List Br. Pl. 83 (1908).

F. longifolia var. trachyphylla (Hack.) Howarth in J. Linn. Soc. (Bot.) 47: 35 (1925).

F. trachyphylla (Hack.) Kraj. in Acta Bot. Bohem. 9: 190, tab. 2, fig. 5, 6 (1930); non Hackel ex
Druce (1915).

F. ovina subsp. ovina var. duriuscula subvar. trachyphylla (Hack.) Maire, Fl. Afrique Nord 3: 126
(1955).

F. longifolia sensu Hubbard, Grasses 111 (1958), pro parte; sensu Tutin, F/. Br. Is/., 2nd ed., 1129
(1962), pro parte; non Thuill. (1800).

F. duvalii (St. Yves) Stohr in Wiss. Z. Martin-Luther-Univ. Halle-Wittenberg 4: 232 (1955), pro
parte quoad descr., excl. typ.

F. cinerea var. trachyphylla (Hack.) Stohr in Wiss. Z. Martin-Luther-Univ. Halle-Wittenberg 9: 395,
402 (1960).

F. stricta Host subsp. trachyphylla (Hack.) Patzke in Osterr. Bot. Zeitschr. 108: 506 (1961).

TAXONOMIC HISTORY

Auquier (1973, pp. 270-272) pointed out that considerable confusion had arisen over Hackel’s
Original concept of subvar. trachyphylla. Some authors had applied the epithet trachyphylla to
fescues with a continuous ring of sclerenchyma (Stohr 1955, 1960; Korneck 1961; Huon 1970;
Bidault 1966, 1968), whereas others applied the name to fescues with three islets of sclerenchyma
tissue (e.g. Howarth 1925, 1948; Krajina 1930). Auquier (1973) outlined the principal arguments for
each view. Evidence in favour of using the epithet to describe specimens with a continuous ring of
sclerenchyma is as follows:

i) Hackel (1882, pp. 90) placed subvar. trachyphylla under var. duriuscula (L. ) Koch, which he
considered to have a continuous ring of sclerenchyma only rarely interrupted.

294 M. J. WILKINSON AND C. A. STACE

05mm

Ficure 2 Leaf-blade sections of F. lemanii and F. brevipila.

1. F. lemanii, Seven Sisters, nr River Wye, Herefordshire, v.c 36 (LTR).
2. F. lemanii, Monks Dale, nr Tideswell, Derbyshire, v.c. 57 (LTR).

3. F. lemanii, Jezainville, Meurth-et-Moselle, France (LTR).

4. F. lemanii, (Neotype) Chinon, France (W).

5. F. brevipila, Lakenheath Warren, W. Suffolk, v.c. 26 (LTR).

6. F. brevipila, Wangford, W. Suffolk, v.c. 26 (LTR).

7. F. brevipila, nr Brandon, Suffolk (LTR).

8. F. brevipila, (Holotype), Niederosterreich, Austria (WU).

ii) Hackel determined and annotated specimens with a continuous ring of sclerenchyma under this
name (cf., for example, St Yves (1913, pp. 63-64)).
iii) Hackel (1882) established a close affinity between subvar. trachyphylla and subvar. firmula of
var. vulgaris Koch, the latter typically having a continuous ring of sclerenchyma.
iv) In Hackel’s herbarium (W) specimens with a continuous ring of sclerenchyma are often classed
under the epithet “‘trachyphylla’’.

Evidence in favour of the epithet being used to describe specimens with discontinuous
sclerenchyma is as follows:
1) Hackel said of subvar. trachyphylla “‘vulgaris in Germania boreali ... non vidi e Gallia,
Hispanica”’. Auquier (1973) pointed out that this distribution agrees well with that of fescues with
discontinuous sclerenchyma but differs strongly from those with a continuous ring.
ii) Hackel (1882, p. 105), in proposing a new taxon (F. ovina subsp. sulcata Hack. var. genuina
Hack. subvar. typica Hack.) that typically has three sclerenchyma islets (cf. p. 104: “‘fasciculis
sclerenchymaticis 3 crassiusculis’’) and five veins, distinguished it from subvar. trachyphylla as
follows: “‘Laminae rarissime 7 nerves inveniuntur et planta tum a F. ovina duriuscula trachyphylla

=“

TAXONOMY OF F. BREVIPILA AND F. LEMANII 295

vix distinguenda”’. Auquier (1973) concluded from the last point that it seemed that Hackel
considered subvar. trachyphylla to have a discontinuous sclerenchyma as well.

Specimens of subvar. trachyphylla were obtained on loan from Hackel’s herbarium in Vienna (W)
in order to search for a possible lectotype. This search produced one specimen of relevance (W, No.
977). The main label of this specimen contains a description of the collection site signed by
Grantzow (of which only the locality Prenzlau can be diciphered readily) and the determination “‘F.
ov. var. duriuscula subv. trachyphylla” signed by Hackel. A pencil note contains measurements of
the lemma, awn and leaf-diameter and the date 2.5. [18] 65. This predates Hackel’s monograph
(1882) in which the taxon is described. A note added by P. Auquier places Prenzlau 60 km north of
Berlin, but a later note by C.A.S. corrects this to 90 km N.N.E. of Berlin in Brandenberg (E.
Germany). Morphologically this specimen agrees fairly well with the modern concept of F.
trachyphylla (F. brevipila) (Table 3) and it is accordingly selected as the lectotype.

- TABLE 3. CHARACTERISTICS OF THE LECTOTYPE OF F. OVINA SUBVAR. TRACHYPHYLLA

COMPARED WITH THOSE OF F. BREVIPILA

F. ovina subvar. trachyphylla

Character F. brevipila (W-No.977)
Panicle length (cm) (3-S5—)4-2-8-8(-9-5) (4-S—)5-8(-6-8)
Spikelet length (mm) (6-1—)6-4—8-0(-8-5) (6:9-)7-2(-7-5)
Lemma length (mm) (3-9—)4-2-5-1(-5-5) (3-9-)4-2(-4-4)
Awn length (mm) (1-:2—)1-4—2-3(-2-6) (0-3—)1-6(—2:-2)
Leaf-blade scabridity scabrid or scabrid distally scabrid distally
Leaf-blade shape in section V-form or V-oval V-form or V-oval
No. of leaf-blade veins (5—)7-9 ih

No. of adaxial leaf-blade furrows 4(-6) 4

Prickle length (um) (2-7—)5—13(-15) (7-S—)11-6(-15)

Krajina (1930) was the first author to recognise subvar. trachyphylla Hackel at the species level.
However, Hackel (ex Druce 1915) had previously used the name to describe a totally different taxon
from South America. The one character (prickles on the glumes) used to distinguish it from F.
dumetorum Phillippi non L. is sufficient to validate the name. Therefore, the later homonym F.
trachyphylla (Hackel.) Kraj. must be rejected. This had earlier been realized by Dandy (1958) when
erroneously placing F. trachyphylla as a synonym of F. longifolia, as pointed out to us by P. J. O.
Trist (in litt. 1984). M. Kerguélen (in litt. 1984) drew our attention to the fact that Pils (1984) quoted
F. brevipila Tracey as a synonym of F. stricta Host subsp. trachyphylla (Hack.) Patzke. Tracey
(1977) distinguished F. brevipila from F. trachyphylla (Hack.) Kraj. by its shorter culms and leaves,
its very shortened hairs on the leaves, its more variable stomatal length, its shortened panicles, and
its relatively shorter spikelet lengths only encompassing the lower part of the range found in F.
trachyphylla. The holotype and the isotype of F. brevipila were received on loan from WU and
compared with the lectotype of F. trachyphylla (Hack.) Kraj. (Table 4). These specimens bear a
remarkable resemblance to one another in all! of Tracey’s diagnostic characters, including the
variability of stomatal length, and there seems little justification for their separation taxonomically.
This being so, F. brevipila becomes the oldest legitimate name for the taxon at the species level and
so replaces the illegitimate F. trachyphylla (Hack.) Kraj.

MORPHOLOGICAL VARIATION
Plants densely to loosely tufted; vegetative shoots all intravaginal and retaining or shedding old
leaves. Culms (9--)30—47(-—72)cm, erect, (0-3—)0-38—0-62(—0-74)mm wide below the first panicle
node, smooth or scabrid distally, grooved; nodes 2(—3), the uppermost not pruinose, visible beyond
subtending sheath or not, and reaching (8-5—)11—20(—32)% of the height of the plant.

Leaves green, occasionally glaucous, usually subpruinose. Leaf-sheaths (1-3—)1-7—4-2(—4-9)cm,
fused for (6-5—)13-—32(—39)% of their length, smooth or scabridulous in distal third, usually laxly
hairy but sometimes glabrous; auricles short, minutely ciliate. Leaf-blades robust to fairly robust,

296 M. J. WILKINSON AND C. A. STACE

TABLE 4. COMPARISON OF TWO TYPE SPECIMENS OF F. BREVIPILA AND LECTOTYPE OF F.
OVINA SUBVAR. TRACHYPHYLLA

F. ovina subvar.

F. brevipila F. brevipila trachyphylla
Character holotype isotype lectotype
1. Diagnostic characters
Culm height (cm) (16—)30(-36) (25—)32(-46) (20—)26(-32)
Panicle length (cm) (3-5—)6-0(-7-0) (4-5—)6-5(-7-5) (4-5—)6-0(-7-0)
No. of spikelets per panicle (15-)21(-31) (13—)21(-34) (15—)20(-24)
Spikelet length (mm) (7-0—)7-6(-8-2) (6-3—)6-8(-7-1) (6:9—)7-2(-7-5)
Leaf-blade length (cm) (2-5—)5-5(-7-5) (7-0—)9-0(-13-0) (4-5—)6-0(-7-5)
Stomatal length (um) (42-5—)46-9(-SO) (42-S—)45(—48-8) (33-75—)38(—42-5)
2. Other important characters
Lemma length (mm) (4-2—)4-9(-5-3) (4-1—)4-4(-4-6) ~ (3-9-)4-2(-4-4)
Awn length (mm) (0-6—)1-9(-3-0) (1-1-)1-7(-2:3) (0-3—)1-6(-—2:-2)
Sclerenchyma in leaf-blade 3 islets 3 islets 3 islets
No. of leaf-blade veins 7(-8) 7-8 ‘|
No. of leaf-blade furrows 4(-6) 4 4

(3-7—)5-5—15(-—19-2)cm, (6—)14—54(—89)% on each plant curved at the tip, scabrid at least in the
upper third, usually hairy at base, but sometimes glabrous.

Panicles erect to nodding, (3-5—)4-2—8-8(—9-5) cm, with (10—)14—28(-39) spikelets. Panicle
branches subpruinose or not pruinose, not narrowing below the spikelets, smooth to scabrid,
covered in hairs or prickles, rarely subglabrous; 1st and 2nd panicle nodes (0-89—)1-3—2-6(—4-3)cm
apart. Spikelets subpruinose or not, (6-1—)6-4—8-0(—8-5)mm, with (2-)3-6(-—7) florets (plus one
sterile floret); pedicels (1-2)1-5—2-6(2-8)mm. Glumes unequal; the lower subulate to narrowly
lanceolate, (2—)2-5—3-1(—3-3) x (0-45—)0-58—0-8(—0-88)mm, usually glabrous or scabrid at tip but
rarely weakly hairy, 1-veined, with a ciliate or serrate margin; the upper narrowly lanceolate to
lanceolate, (3-3—)3-6—4-5(—4-7) x (1—)1-2—1-45(—1-6)mm, usually scabrid near apex but occasio-
nally hairy or glabrous, 3-veined, with a ciliate margin. Lemmas lanceolate, (3-9—)4-2—5-1(-5-5)
(excluding awn) X (1-5—)1-6—2-1(—2-5)mm, green but sometimes purple at apex, with awns (1-2-
)1-45-2-3(—2-6)mm, 5-veined, usually hairy or scabrid on distal half, but occasionally glabrous.
Paleas linear-lanceolate, (3-8—)4-1-5-0(—5-3) x (0-68—)0-8—0-9(-1-0)mm. Anthers yellow or
purple, (1-9—)2-1—2-9 mm.

Leaf-blade anatomy (Fig. 2): outline of leaf-blades usually V-form but occasionally V-oval;
margins not infolded; leaf-blade diameter (0-59—)0-74—0-93(—1-1)mm; diameter/thickness ratio
2-15—2-5(—2-75); veins (S—)7—9; sclerenchyma usually forming three tailing islets at midrib and
margins, 1—4 cells thick, often with smaller islets opposite veins (1 cell thick), rarely forming a
complete ring much thickened at margins and midrib; 4(—6) grooves and 3-5 ridges on the adaxial
surface; adaxial midrib (0-85—)0-9-1-6(—1-8)mm wide and (0-03-—)0-04—0-07mm deep; adaxial
epidermal cells of fairly uniform size, usually lacking bulliform cells but occasionally with small
bulliform cells in grooves, with prickles (2-7—)5—13(—15)um.

Leaf-blade adaxial epidermis: stomata (37-8—)40-3—48-3(—50-5)um, predominantly solitary, 52-
80(—100)% solitary, (0O—)3—16(—29)% with an accompanying prickle-cell, (OQ—)3—16(—29)% with an
accompanying silica-cell; prickles predominantly solitary, (44—)53—97(-—100)% solitary, 0-19(-
53)% forming short rows of 2 or more prickle-cells, 0—35(—50%) with an accompanying silica-cell.

Leaf-blade abaxial epidermis: stomata absent; long-cells (101—)112—200(—212)um; silica-cells
and cork-cells present; prickles present or absent.

DISTRIBUTION

At present, we have relatively little information concerning the distribution of either F. brevipila or
F. lemanii.

“

TAXONOMY OF F. BREVIPILA AND F. LEMANII 297

3 4
ee a 6)

oO - F. lemanii

e - F. brevipila

Kilometres

Miles

CHANNEL ISLANDS
PLOTTED ON
UTM GRID

0)

5 6
Ficure 3. Geographical distribution of Festuca lemanii Bastard (O) and F. brevipila Tracey (@).

In Britain, F. brevipila appears to be most common in southern and central England, extending to
South-West Yorkshire (v.c. 63) and westwards to Salop (v.c. 39). We have also collected a single
specimen from a car-park in Jersey. Records have been made from the following vice-counties: 17,
19, 20, 21, 22, 25, 26, 28, 29, 30, 39, 55, 63. A distribution map is given in Fig. 3.

It is highly unlikely that F. brevipila is native to the British Isles. It has probably been introduced
during the past 200 years as a commercial turf-grass species for roadside verges and railway
embankments, and it is still available commercially as several cultivars of ‘Hard Fescue’. It does not
grow in its natural habitat in the British Isles. It is native to central Europe, but is also introduced in
France and in much of Scandinavia.

~

298 M.J. WILKINSON AND C. A. STACE

Our records of F. lemanii are even less complete. Specimens have been observed from scattered
sites in England, Scotland and Wales. Kirkcudbright, Scotland (v.c. 73) marks its northern limit and
Glamorgan, Wales (v.c. 41) its western limit. Unlocalised specimens suggest that it may also be most
common in southern and eastern England. It has been recorded with certainty from vice-counties
19, 37, 41, 57 and 73. A distribution map is given in Fig. 3.

F. lemanii is commonly found growing in more natural habitats, together with F. ovina subsp.
ovina, subsp. hirtula (Hack. ex Travis) M. Wilkinson and/or subsp. ophioliticola (Kerguélen) M.
Wilkinson, and it is likely to be native in the British Isles. It has a wide range of habitat preferences
and has been found growing on both acidic and calcareous soils.

In Europe, F. /emanii extends through Belgium, France and Spain (Markgraf-Dannenberg 1980).

»!
~

ACKNOWLEDGMENTS

We are extremely grateful to Dr M. Kerguélen for his valuable help concerning the taxonomy and
nomenclature of both species. We are also indebted to P. J. O. Trist and P. D. Sell for their helpful
advice and comments concerning the nomenclature of F. trachyphylla. Thanks are also due to the
curator of W for supplying valuable type material and to the curator of MANCH for allowing us to
search for a neotype of F. lemanii.

REFERENCES

AUQUIER, P. (1973). Qu’est-ce que le Festuca caesia Sm. (Poaceae)? Lejeunia, nov. sér., 70: 1-12.

AUQUIER, P. (1974). Biosystématique, taxonomie et nomenclature du groupe de Festuca ovina L. s.1. (Poaceae)
en Belgique et.dans quelques régions voisines. Ph.D. Thesis, Liége.

BIDAULT, M. (1966). Remarques sur le Festuca ovina L. var. glauca et var. duriuscula des Alpes Maritimes. Bull.
Soc. bot. Fr. 113: 173-183.

BiDaAuLt, M. (1968). Essai de taxonomie expérimentale et numérique sur Fesiuca ovina L. s.1. dans le Sud-Est
de la France (Thése, Paris). Rev. Cytol. Biol. vég. 31: 217-356.

Danpy, J. E. (1958). List of British vascular plants, p. 158. London.

Druce, G. C. (1915). Plant notes, etc., for 1914. Rep. botl Soc. Exch. Club Br. Isl. 4: 1- 48.

HackEL, E. (1882). Monographia Festucarum Europaearum. Kassel & Berlin.

Howarth, W. O. (1925). On the occurrence and distribution of Festuca ovina L., sensu ampliss., in Britain. J.
Linn. Soc. (Bot.) 47: 29-39.

Howarth, W. O. (1948). A synopsis of the British Fescues. Rep. botl Soc. Exch. Club Br. Isl. 13: 338-346.

HussarpD, C. E. (1954). Grasses. A guide to their structure, identification, uses and distribution in the British Isles.
Harmondsworth.

Huon, A. (1970). Les Fétuques de l’ouest de la France. Recherches de biosystématique et de biogéographie. Ph.D.
Thesis, Rennes.

KERGUELEN, M. (1983). Les Graminées de France au travers de ‘‘Flora Europaea”’ et de la ‘‘Flore’”’ du CNRS.
Lejeunia, nov. sér., 110: 14-61.

Korneck, D. (1961). Uber einige in Baden und Elsass vorkommende Schafschwingel (Festuca ovina L. s. lat.).
Mitt. Bad. Landesver. Naturk. 8: 187-196.

KrasINA, V. (1930). In Domin, K., Schedae ad floram ceschoslovenicam exsiccatam, centuria II (Genus
Festucam V. Krajina exposuit). Acta. Bot. Bohem. 9: 184-220.

MARKGRAF-DANNENBERG, I. (1980). Festuca L., in Tutin, T. G. et al., eds. Flora Europaea 5: 125-153.
Cambridge.

Pits, G. (1984). Systematics, karyology and distribution of the Festuca valesiaca group (Poaceae) in Austria and
the South Tyrol. Phyton, Horn 24: 35-77.

RAUSCHERT, S. (1960). Studien iiber Systematik und Verbreitung der thiiringischen Sippen der Festuca ovina L.
s. lat. Repert. Spec. nov. Regnum Veg. 63: 251-283.

St-Yves, A. (1913). Les Festuca de la section Eu-Festuca et leurs variations dans les Alpes Maritimes. Ann.
Cons. Jard. bot. Genéve 17: 1-218.

Stour, G. (1955). Der Formenkreis der Festuca ovina L. in Mitteldeutschen Trockengebiet. Wiss. Z. Martin-
Luther-Univ. Halle-Wittenb. 4: 729-746.

Stour, G. (1960). Gliederung der Festuca-ovina-Gruppe in Mitteldeutschland unter Einschluss einiger
benachtbarter Formen. Wiss. Z. Martin-Luther-Univ. Halle-Wittenb. 9: 393-414.

TAXONOMY OF F. BREVIPILA AND F. LEMANII 299

Tracey, R. (1977). Drei neue Arten des Festuca ovina-Formenkreises (Poaceae) aus dem Osten Osterreichs.
Plant Syst. Evol. 128: 287-292.

TutTin, T. G. (1952). Festuca L., in CLAPHAM, A. R., TuTIn, T. G. & WARBURG, E. F. Flora of the British Isles,
pp. 1421-1427. Cambridge.

Tutin, T. G. (1962). Festuca L., in CLAPHAM, A. R., Tutin, T. G. & WARBURG, E. F. Flora of the British Isles,
2nd ed., pp. 1125-1130. Cambridge.

(Accepted July 1988)

9

. danuins aaa
fond tan |
ma, S18 iis Sar ee
eo > ee abd ree ban
oc com ty Saal gion eke ap in FX ar
4 nde x - Trew i ag ae

s Bri Tales, i ne

pe hag chenh tess A emai om ath
In Ruel, Teageeifenea oe

. he ste chy spetiet ii eM Ketgsblen tor bis talpabie &
ae poids: ti a Doth soactes, We ave ign indebted @ F, 30. Tr
atin od tenryeedé boacpraing the Romentiatape of E tranciyy :
Sapston of 4 ee = ere Ug v valugbte iis BAGS ind to the Saat

MORrGS: fv iSceee bot #, lever A eee

ak ad -RRMIINCES <>

” ues? (P¥7S): Qc'en-ce gat le Fen rene esecls-Srn, | (podioaa ti Leditaeie alts
okenter, 3 39S Rice vetting puciigete af raleresutange ds grou ca
on Delete cr iion ¢ccloas ites womens, PRD. Thema. Liege, &
Teeny, M. PPE. Romaine mur le Fevancd die £ var hess at Vist aber ita bee

. . ee
> 4 ~~ rey | a ‘ =
; ; “4 ‘ -~
pata (ite, Bg SS Sei oy. caval BEC ea ct cree cpt ssi Reta Sm A
I afae ties i: new. f ye Rint, ep, 34> AX T= 308, 7 a bra: a
= Set. f, yesic A tigen; 6. 158i Outow, = Ae:

i, . C. -bPFES Phat Stas. ot payer Reo. Soe Sse Leck cub Broth ee

{ Te he ‘4 1fite ent a ec aril’ Pare Cpa Seria. Karl & é ME, - wy
ere Mes EY Vt. oc ote eat Hay batinn OF Raters ovina Ly aehe ee

rus 12 : oe ~» 7 a ae 3 Te \
iw apes Wy 4 22O8E}, Al densa. <2 the Britah Rescoes: ig hind Sod -Exelt lub Be it. 334
Hemeex! oa % on hd eeir ores, hier: cm Mcigie, \Udez csi
8 Baie ; .
~<a: a 4 ne bo is Ponce Rewharhclins wis irs op btedi
” - ei ne e
ve - o jain’ ce Pega Meyer ite Plame Enropatal’: “det
éaet ie te, “8 wy ,
fe 4 ¢ “a, a wey Sat > nav > t=! F rh: vara Sanne Schobistwingd Rica
5: Sail wives Perot. © mM, 2 a4
winds VOTH in (ete ey a me) ai ten euiegpis, bhi
reo e's Obes ek. Dox. Faken’ 9 16220. = Ott
00, bf aaa. 1, CR Medics m Tin, 0, > ey Aa flo, &
<1 ; .
iF : YY: £9ca tri a, Mare ot) i 5 pags wey ifve res wa hia HOTN

2 ry. ae nnditex Overceebath, writ Verknoitiang er OF nga gue
’ aia) ‘Ls hia nad i Sef 74 Ot 4a ‘ ean tv ages Auge
: 12. ea a ae en ae cUdarytechen oct ak:

rs ls - i | il na Tipe. iH Misnekiewenchtata cent
ori ‘i erne hey oy Evie a end > Vosieg 4,

Watsonia, 17, 301-308 (1988) 301

Alopecurus X plettkei Mattfeld in Britain

P- JO; FRIST

Glovers, High St, Balsham, Cambridgeshire CB1 6DJ

/

and
M. J. WILKINSON

Department of Botany, The University, Leicester LEI 7RH

ABSTRACT

The hybrid Alopecurus X plettkei Mattfeld (Alopecurus bulbosus Gouan xX A. geniculatus L.) is reported new to
Britain. Its variation, distribution, habitat and growth performance are discussed.

INTRODUCTION

Alopecurus X plettkei Mattfeld, the sterile hybrid between A. bulbosus Gouan and A. geniculatus
L., is known from Belgium, Germany and the Netherlands (Hubbard 1975). It was first found in
Britain by Brian Wurzell on 27 May 1977, and confirmed by C. A. Stace, from marshes west of
Glynde Reach, near Lewes, E. Sussex, v.c. 14, GR 51/465.094. It has subsequently been found at
several sites in S. Hants. (v.c. 11), W. Kent (v.c. 16) and E. Suffolk (v.c. 25). The following account
describes the morphology, anatomy, phenology, cytology and ecology of the British populations.

DESCRIPTION OF ALOPECURUS X PLETTKEI

Perennial, erect and densely tufted or sometimes forming mats of decumbent spreading culms,
compactly tufted or occasionally solitary, frequently branched from nodal growth, with some
internodes arcuate. Culms (15—)18—77(—93) cm, weak, with (2—)4—-13 nodes at uneven intervals,
from 5—15 cm below the panicle, over the entire length of the culm; lowest internode 1—33 x 1-6(-
7-5) mm, very swollen to slightly swollen, extremely variable in shape, with a smooth or ridged
surface, sometimes with one or two slightly bulbous swellings but generally straight sided. Basal
leaves (8—) 15—20 (—38) cm; culm leaves hairless, 1-5—15 cm X 0-7—3-3 mm, finely pointed, flat or
infolded, minutely scabrid on margins and veins; sheaths smooth, the uppermost often inflated;
ligule 1-8—4-5 mm, broadly obtuse, membranous.

Panicles 1-4—4-5 (—6-3) cm X 2-5-5 mm; dense, narrowly cylindrical, spike-like, bluntly pointed.
Pedicels 0-4—1 mm. Spikelets oblong, 2-5-3 mm, flattened, 1-flowered, falling entirely at maturity.
Glumes equal, 2-5-3 mm, free to the base, narrowly oblong, broadly obtuse, exceeding lemma by
(0-05—)0-1—0-4(—0-5) mm, 3-veined with uneven-length hairs on the keel and margins. Lemma 2-3
mm, keeled, smooth, narrowly oblong, membranous, 4—veined, sometimes weakly hairy in distal
quarter, awned on the back c. 1 mm from the base; awns 3-5-6 mm, protruding 1-0—2-8 (—3-5) mm
beyond the glumes. Anthers 1-1-5 mm, indehiscent, containing imperfect pollen 0(—4)% stainable
in Muntzing’s aceto-carmine, 20-28 um in diameter.

Leaves in section: adaxial and abaxial epidermises with cells of irregular size and shape, with
small bulliform-cells on the adaxial surface between veins; small islets of sub-epidermal scleren-
chyma usually occur in association with the veins.

Abaxial epidermis with solitary stomata 35—49 um long, in continuous or semi-continuous rows,

302

P. J. O. TRIST AND M. J. WILKINSON

47—95 um apart in continuous rows; short-cells (other than stomata) (0—)1—4(—6) per 100 um?, 27—
73% being hook-cells or prickles; long-cells (60—)100-—220 um.
Adaxial epidermis with fewer stomata than abaxial epidermis but with (2—)4—10 non-stomatal

short-cells per 100 um*; long-cells 90-380 um long.
A comparison of A. X plettkei and its parental species is given in Table 1.

TABLE 1. A COMPARISON OF A. BULBOSUS, A. X PLETTKEI AND A. GENICULATUS

Character A. bulbosus A. X plettkei A. geniculatus

Culm 7 erect erect to spreading spreading, geniculate

Culm length (cm) 11-29 (15—)18-77(-93) 15-38

No. of branches per culm 0 0-3(-7) 0-3(-S)

No. of nodes per culm 1—4(-S) (2—)4-13 2-6(-8)

Rooting culm absent absent or present absent or present

“Basal internode swelling (1-)2-4-5(-6-0) (O—)0-5—2-2 (—0-5—)0-1-0(-1-5)

(mm) |

Basal internode length (cm) 0-4-2 0-6-33 10-35

Twin’ basal swollen inter- present or absent usually absent absent

nodes

Cauline leaf width (mm) 0-5-3-5 0-7—4-5 ~ 0-7-5-0

Cauline leaf infolded usually flat flat

Ligule (mm) 1-6—4-0 1-8—4-5 2:0-5-5

Panicle length (cm) 1-3-4-4(-5-4) 1-4-6-3 (2-0—)2-6=5-0

Spikelet length (mm) 2:5—3-3 2:5-3-0 2-0-3-0

Glumes lanceolate, acute. Short lanceolate, acute to lanceolate, obtuse.
hairs on keel and _ obtuse. Uneven long Silky hairs on keel and
margins and short silky hairs on — shorter appressed hairs

keel and margins

from middle to base of
keel

Total awn length (mm) (3-4—)4-0-5-3(-6-5) (3-0—)4-5-5-5 3-0—4-5(-5-0)

Awns exceeding glumes by 3— exceeding glumes by exceeding glumes by
4mm 1-0-3-5 mm 0-8—1-7 mm

Lemma hairy at tip hairs at tip present usually no hairs at tip

(often few) or absent

Anther length (mm) 1-2-—2-2 1-0-1-5 0-8-1-7

Distance between glumes 0-2-0-7(-0-9) (O—)0-1-0-4(-0-5) (O—)0-1—0-35(—0-55)

and lemmas (mm)

Chromosome no. 14, 21, 28 21 28

“Difference between basal internode width and stem width.
> Adjacent basal internodes swollen.

The hybrid is very variable morphologically and anatomically and cannot be reliably identified
with any one character. Indeed, even pollen sterility can break down as a diagnostic character;
specimens of both A. bulbosus and A. geniculatus have been recorded with highly sterile pollen
(<1% stainable with Muntzing’s aceto-carmine).

All hybrids were observed to contain at least some characteristics from both parents, although all
were invariably much more similar to one parent than to the other. Whilst scatter diagrams can be
used to show some degree of intermediacy (Fig. 1), overall intermediacy is best represented on a
percentage hybrid index. Representatives of A. bulbosus, A. X plettkei and A. geniculatus were
compared on a percentage hybrid index according to the scheme given in Appendix 1. A histogram
constructed from the index (Fig. 2) shows that the individual plants fall into two relatively broad but
distinct groups, corresponding to the index range of the two parental species. The score of the
hybrids falls at the margins of the two parents and is in no case entirely intermediate between them.
Hence, there are both ‘A. geniculatus-like’ and ‘A. bulbosus-like’ representatives of A. X plettkei,
but apparently no nearly intermediate representatives.

ALOPECURUS xX PLETTKEI IN BRITAIN 303

400

Fs

e my | A. bulbosus
< 300 >

fo)

Cc

- O=- 4

:

2 daalertg Hb

S 200 “bus

- ) 1 2 3 4 S)
Basal node swelling (mm)

Ficure 1. Scatter diagram showing specimens of A. bulbosus, A. geniculatus and some representatives of the
hybrid between them, A. X plettkei (H1-5).
KEY
O <25% of short-cells are prickles/hook-cells (abaxial surface)
O- >25% of short-cells are prickles/hook-cells (abaxial surface)

© glumes obtuse O rooting nodes absent
© — glumes obtuse to acute rooting nodes present
glumes acute O _ tetraploid or diploid
O © leaves flat @ triploid

-O leaves mostly infolded

Hybrid localities: H1, GR 63/505.074; H2, GR 50/514.994; H3, GR 41/362.144; H4, GR 51/465.094; H5, GR 40/
170.918

The decumbent growth habit of the ‘A. geniculatus-like’ hybrid makes it unlike either of its two
parents. The panicle is often hidden under leaves or just turning up beyond the leaf-tips. The lengths
of the stem, panicle, cauline and basal leaves and the widths of the panicle and cauline leaves are
extremely variable. On open ground, decumbent stems readily develop from 4—5 nodes, producing
branches which in total produce 15-50 leaves on a stem. This nodal branching is seen in A.
geniculatus but with less leaf production. This hybrid is not a robust plant: its stems, unlike those of
either parent, need support from surrounding vegetation.

Two features are significant for field recognition: the number of nodes and the number of leaves
per culm, which both far exceed those of either of the parents.

The ‘A. bulbosus-like’ hybrids are less distinctive in the field but may be identified by their
intermediate internode swelling (0-5-2 mm), their variable glume-tip shape (obtuse to acute), their
lemmas which are only weakly hairy in the distal third, their fairly erect culms, and their ‘A.
geniculatus-like’ awn length (3-4-4 mm).

304 P.J.O. TRIST AND M. J. WILKINSON

-

A. bulbosus A. geniculatus

5
4
o 3
c 3 5
®
=
oO
£2 ,
HH HH
12 h34

_
ae

0 10 20 30 40 50 60 7:0 80 90. - 100
Percent hybrid index

FiGure 2. Percent hybrid index showing relationship between A. bulbosus, A. geniculatus and A. X plettkei (H)
(see Fig. 1 for localities of the hybrids).

CHROMOSOME NUMBER

Given the close affinity shown by representatives of A. X plettkei to its parents, chromosome
number can provide strong additional evidence for identification purposes.

Several authors have reported a chromosome complement of 2n=14 for A. bulbosus (e.g. Maude
1939; Sieber & Murray 1979, 1980) and of 2n=28 for A. geniculatus (e.g. Kattermann 1930; Sieber
& Murray 1979). Sieber & Murray (1979) investigated chromosome numbers in Alopecurus species
at several stations on the south coast of England. For A. bulbosus they recorded that, in addition to
diploids, triploids and tetraploids were found growing together in a mixed population from
Seasalter, E. Kent, v.c. 15. Of the eight plants studied three were diploid, three were triploid and
two were tetraploid. The chromosome numbers were 2n=14, 21, and 28. In a subsequent work
(Sieber & Murray 1980) it was concluded on the basis of pairing behaviour during meiosis that the
triploid and tetraploid plants were autopolyploids.

The same authors (Sieber & Murray 1979) found 2n=28 in natural populations of A. geniculatus
from Buxton, Derbys., v.c. 57. Their observations on the meiotic and mitotic behaviour of two
tetraploid species of Alopecurus agree with those of Johnsson (1941) and support the suggestion that
A. pratensis is an allotetraploid and A. geniculatus is an autotetraploid.

Dr P. E. Brandham of Kew has recently confirmed that material of A. x plettkei on the Stanpit
South Marsh, Christchurch, S. Hants., v.c. 11, (Sept. 1985) is triploid (2n=21). We have also
confirmed the following hybrids to be triploid: two specimens from Stanpit South Marsh,
Christchurch, S. Hants, GR 40/170.918 and three specimens from Glynde Reach, nr. Lewes, E.
Sussex, v.c. 14, GR 51/465.094. It is of some importance to remember that because specimens of A.
bulbosus exist with chromosome numbers of 2n=21 and 2n=28, there may be specimens of A. X
plettkei derived from such plants with chromosome numbers 2n=21-—28. Therefore it is possible that
there are tetraploid representatives of A. bulbosus and A. X plettkei as well as of A. geniculatus.
Furthermore, although the possession of 21 chromosomes is strong additional evidence of hybrid
status it is not a guarantee since triploid representatives of A. bulbosus also occur.

PHENOLOGY
In A. plettkei the anthers are indehiscent and produce very little pollen; that produced is largely

sterile. Presumably reproduction is almost entirely by vegetative means by growth from swollen

“

ALOPECURUS Xx PLETTKEI IN BRITAIN 305

stem bases (particularly in the ‘A. bulbosus-like’ hybrids) and by rooting at culm nodes (particularly
in the ‘A. geniculatus-like’ hybrids). In cultivation, the decumbent culms of the ‘A. geniculatus-like’
hybrids will develop 4—S nodes which are quick to root and produce leaves 10—35 cm long within two
weeks. This is also seen in the wild, where propagation has taken place on bare ground.

In cultivation, the autumn leaf growth of the hybrids survives a fairly severe winter up to mid-
February. This growth then withers and is quickly followed by spring growth with culms appearing
before the end of April. Panicles break from the sheath throughout May but their progress is slow;
stems with partial panicle exsertion of + 2-5 cm in mid-May can take a further 15 days for the panicle
to break fully to a length of 4-5—5-8 cm. At this stage the plants are light green in colour with a
slightly glaucous sheath as in A. bulbosus. All flowering is completed before the end of June, when
the leaves start to wither. By early July new growth is developing, as the panicles of previous growth
are breaking up.

This hybrid is vigorous and responds to management by grazing and cutting by extending its
growing season. At Glynde Reach, E. Sussex, in 1986 the marsh had been cut for hay in July and on

-8th August there was fresh A. X plettkei growth with panicles only half exserted. At Nursling N.R.,
S. Hants., v.c. 11, in late July, in an area grazed by cattle earlier in the year, A. X plettkei had
regrown, with decumbent culms up to 77 cm long and with a considerable bulk of basal leaves. To
consider the hybrid’s capacity for propagation, three basal internodes of 2 cm in length were put into
autumn cultivation, with subsequent production of 93 culms. Single internodes autumn-planted will
produce 4—13 tillers by February.

THE DISTRIBUTION OF ALOPECURUS X PLETTKEI IN BRITAIN

1. South Hampshire, v.c. 11: marshes subject to occasional tidal overflow. Lower Test Nature
Reserve, Nursling, GR 41/368.145 and 41/362.144, R. P. Bowman, 1978 (CGE, K, herb. R.P.B.,
herb. P.J.O.T.). The Furlongs, Redbridge, GR 41/364.137, R.P. Bowman, 1978 (CGE, K, herb.
R.P.B., herb. P.J.O.T.). Unprotected coastal marshes, Stanpit South Marsh, Christchurch, GR
40/170.918, S. R. Davey, 1980 (CGE, herb. R.P.B., herb. P.J.O.T., herb. R.M. Walls). Derelict
marsh subject to tidal inflow, Keyhaven, GR 40/305.916, R.P. Bowman, 1986 (herb. R.P.B.).

2. East Sussex, v.c. 14: managed grazing marshes, south-east of Decoy Wood and adjacent to
Glynde Reach, near Lewes, GR 51/465.094, B. Wurzell, 1977 (CGE, LTR, herb. P.J.O.T., herb.
B.W.). Disturbed river wall-berm, east bank of R. Cuckmere, near Exceat bridge, Seaford, GR
50/514.994, Lady R. FitzGerald, 1987 (herb. P.J.O.T., herb. R.F.).

3. West Kent, v.c. 16: wet track in rough grazing, upland of marsh level, Higham marshes, Church
Street, near Higham, east of Gravesend, GR 51/713.742, Lady R. FitzGerald, 1987 (herb.
P.J.O.T., herb. R.F.).

4. East Suffolk, v.c. 25: managed grazing marshes, Castle Farm, Burgh Castle, GR 63/473.058, R.
P. Libbey, 1980 (LTR). Humberstone marshes, south of Breydon Water near Great Yarmouth,
GR 63/505.074, P.J.O. Trist, 1982 (herb. P.J.O.T.).

Specimens collected from Lewes and from Burgh Castle resemble A. bulbosus more closely than

A. geniculatus. Specimens collected from the remaining sites show a stronger resemblance to A.

geniculatus.

ECOLOGY

The habitat of Alopecurus X plettkei is coastal marsh that is inadequately protected from tidal water
Owing to an absence or a breakdown in sea defences. These areas are occasionally shallowly flooded
for a few hours at periods of high spring and autumn tides, when tidal water runs up the ditches and
overflows on the land. A saline deposit is built up in the soil, more particularly in the marsh
depressions and in the furrows on marshes that were at one time under cultivation.

A. X plettkei thrives in a soil texture of organic loam with additions of alluvium. It has a lower salt
tolerance than A. bulbosus and a preference for an open position at or slightly above marsh level. It
will colonise the bare margin of a depression, but not the depression itself, showing its preference
for drainage, as opposed to A. bulbosus which will thrive in the water-logged bed of a depression. At

306 P.J.O. TRIST AND M. J. WILKINSON

Stanpit South Marsh, Christchurch, where there are old grips, and rills in adjacent saltings, the
hybrid colonises the margin and not the bed of grips that support Salicornia spp., an indication of a
high salt level (too high for A. bulbosus).

At Keyhaven in a small 0-5 ha marsh, A. xX p/lettkei has been found only in two small colonies at
the back of the marsh, just above the general level, so that a tide lapping near the area would rapidly
run off. Here A. bulbosus is widespread but thinly spaced among Glaux maritima, Puccinellia
maritima and some Aster tripolium in a soil with 0-5% NaCl. During the warm months of the year it
is likely that this percentage is higher and that A. bulbosus is here at its limit of salt tolerance. At The
Furlongs, Redbridge, there is a similar demonstration of the difference in saline tolerance of A.
bulbosus and A. xX plettkei. The sward, subject to fairly frequent tidal over-spills, is almost entirely a
simple mixture of Puccinellia maritima and Agrostis stolonifera with some Aster tripolium. Soil
samples taken contained a range of 0-4—0-5% NaCl, but the presence of Aster tripolium suggested a
higher percentage in June—July. A. bulbosus was thinly distributed and only occasionally small,
weak plants of A. x plettkei were found on low elevations of old ditch spoil where drainage would be
rapid following tides.

The marshes on the Lower Test N.R. at Nursling have probably never oa disturbed. The
surface levels are therefore uneven, with isolated areas of halophytes demonstrating saline deposits
from occasional flooding. In an unimproved marsh, Eleocharis uniglumis, Festuca rubra, F.
arundinacea, Agrostis stolonifera and Trifolium fragiferum are common associates, but, where A.
bulbosus is present in lower levels, these plants are supplemented by Glaux maritima, Juncus
gerardit, Plantago maritima, Ranunculus sceleratus and Triglochin maritima. A. X plettkei would be
found in open areas of Agrostis- -Festuca but not, or rarely, in the lower and wetter areas of A.
bulbosus and its associates.

On the Lower Test N.R. the limits of variation of salt content of soil samples taken in August at 15
cm depth were 0-08—0-48% NaCl. This range will vary with the temperature of the season when
samples are taken, and seasonal factors diluting sea water. Both at Redbridge by Southampton
Water and at Stanpit South Marsh by Christchurch, for instance, there is a complicated double tide
which increases salinity on the one hand, and an outflow of fresh-water from the R. Test, Avon and
Stour which diminishes salinity on the other.

The hybrid was abundant on the site where it was first recorded in 1977 (Glynde Reach, v.c. 14)
and occupied wet areas over a large part of G-.R. 51/46.09. No Alopecurus bulbosus was present in
1977; it was last recorded near the river bank in 1962 by B. Wurzell, C. A. Stace and others on a
London Natural History Society field day. The site would have been in the area of a now derelict
ditch below the high bank containing the river. Over-spills of saline water no longer occur.
Sometime prior to 1975, a tidal flap and pump were installed on The Reach at the point just above
the A27 road, preventing land flooding. The former wet state of the hybrid site is still witnessed by
small grips crossing the marsh and connecting to boundary ditches. The marsh is now dry with a
diminishing amount of Alopecurus X plettkei. A soil sample taken in August 1986 at 15 cm from the
bed of a grip, where only a few plants of the hybrid were seen, contained 0-01% NaCl. There was no
plant indicative of salinity in the sward or in the river which rejoiced with Nuphar lutea on its
surface.

A. X plettkei would not make a constituent of a grass-seed mixture. It produces long decumbent
stems which are weak and lie over the ground and would therefore be smothered by tall marsh
grasses such as Festuca arundinacea, Agrostis stolonifera and Elymus repens. The A. geniculatus
parent of the hybrid is common on most coastal marshes, but its other parent, A. bulbosus, is still
comparatively rare, in spite of the fact that it has been removed from the Red Data lists (Perring &
Farrell 1983). This hybrid is able to continue to occur where A. bulbosus is extinct, as on the marshes
by Glynde Reach, but from observations of this taxon in cultivation we consider its perennial life is
limited.

IS A. X PLETTKEI A THREAT TO THE SURVIVAL OF A. BULBOSUS?

It has been suggested that the spreading growth of A. X plettkei may have a smothering affect on A.
bulbosus and may pose a threat to its survival. This is unlikely. The growth period of the two taxa
does not overlap. A. bulbosus is fully developed and flowering by the last week in May to the first

-“

ALOPECURUS xX PLETTKEI IN BRITAIN 307

week in June, and, before the end of July, leaves and culms have died back when cattle-treading
soon disperses the plant remains so that A. bulbosus is difficult to trace until the following year,
except by exploration for the swollen stem bases (Trist 1981). A. x plettkei has an extended growing
season and the plants which over-winter die back in July only to be immediately replaced by new
growth which continues into September. Thus A. bulbosus has completed its growth in early
summer while strong growth of the hybrid continues throughout July and August.

There is a significant difference in the habitat of the two taxa. The hybrid is seen on average mid-
marsh level and not below, and preferably in a position of better drainage which is slightly above the
average level. A. bulbosus prefers a reasonably constant moisture supply and has a tolerance to
salinity of the order of 0-4—0-5% NaCl. It is found on the higher margin of saltings and often in the
low areas and furrow depressions of marsh adjacent to tidal rivers, which gives an indication of
tolerance or even preference for salinity. Although it can be cultivated for several years in the
absence of salinity and a constant moisture supply, it eventually reverts to a poor state and dies. A.
x plettkei behaves as a perennial but it is likely that it is short-lived.

ACKNOWLEDGMENTS

We would like to thank Dr R. Page, Conservation Officer of the Hampshire and Isle of Wight
Naturalists Trust, for permission to study A. X plettkei on the Lower Test Nature Reserve. Thanks
are also due to Marian Reed, R. M. Walls, Paul Bowman, and Rosemary FitzGerald for help in the
field; Dr B. Davies (of M.A.F.E.) for arranging salt analyses; Dr P. Brandham of Kew for a hybrid
chromosome count; and Mrs Sue Ogden for drawing the Figures. We are grateful for the help and
advice of Professor C. A. Stace and C. D. Preston.

REFERENCES

HuBBARD, C. E. (1975). Alopecurus L., in Stace, C. A., ed. Hybridization and the flora of the British Isles, p.
585. London.

Jounsson, H. (1941). Cytological studies in the genus Alopecurus. Acta Univ. Lund 37: 1-43.

KATTERMANN, G. (1930). Chromosomenuntersuchungen bei Gramineen. Planta 12: 19-37.

MaupE, P. F. (1939). The Merton Catalogue. A list of the chromosome numerals of species of British flowering
plants. New Phytol. 38: 1-31.

PERRING, F. H. & FARRELL, L. (1983). British red data books: 1. Vascular plants, 2nd ed., p. 81. Lincoln.

SIEBER, V. K. & Murray, B. G. (1979). The cytology of the genus Alopecurus (Gramineae). Bot. J. Linn. Soc.
79: 343-355.

SIEBER, V. K. & Murray, B. G. (1980). Spontaneous polyploids in marginal populations of Alopecurus bulbosus
Gouan (Poaceae). Bot. J. Linn. Soc. 81: 293-300.

Trist, P. J. O. (1981). The survival of Alopecurus bulbosus Gouan in former sea-flooded marshes in East
Suffolk. Watsonia 13: 313-316.

(Accepted May 1988)

308 P. J. O. TRIST AND M. J. WILKINSON

APPENDIX: SCORING SCHEME FOR HYBRID INDEX

A. bulbosus Intermediate A. geniculatus
Score 0 1 2
Vegetative characters
No. of nodes per culm 3 3-4 4
No. of branches per culm 0 N/A? 1
Rooting nodes absent N/A absent
Basal node swelling (mm) (1-)2-4-5(-6-0) (0O—)0-5—2-2 (—0-S—)0-—1-0(-1-5)
‘Twin’ basal swollen internodes present N/A absent
Basal leaves infalded mostly infolded N/A mostly flat
Stem erect fairly erect or spreading spreading, geniculate
Flag-leaf width (mm) 1-2 2-01-3 >3
Lower culm-leaf width (mm) 1-2 2-01-3 >3
Reproductive characters
Panicle length (cm) 2-0-2:-5 2:51-4-4 >4-4
Panicle width (cm) <3-5 3-5-4-5 >4-5
Glume sharpness acute acute to obtuse
obtuse
Awn length (mm) >5 4-5 <4
Hairs at top of lemma present sparse absent
Distance between glume tips >0-4 0-25—0-4 <0-25
& lemma tip (mm)
Pollen diameter (um) <29-4 29-4—33-0 >33-0
Abaxial epidermis
% short-cells being 0-20 21-37 >37
hooks/prickles
Short-cell frequency 0 1-2 >2
(per 0-11 mm”)
Mean long-cell length (um) >200 180-200 <180
Presence of rows of short-cells absent N/A present
Distance between stomata in >90 70—90 <70
continuous rows (um)
Stomatal length (um) <38 38-41 >41
Adaxial epidermis
Mean long-cell length (um) >210 190-210 <190
Short-cell frequency <3 3-4 >4
(per 0-11 mm?)
Prickles above nerves present N/A absent

* N/A: not applicable

Watsonia, 17, 309-317 (1988) 309

Potamogeton X lanceolatus Sm. in the British Isles

C, D..PRESTON

Institute of Terrestrial Ecology, Monks Wood Experimental Station, Abbots Ripton,
: Huntingdon, PEI7 2LS

ABSTRACT

Potamogeton X lanceolatus Sm., the sterile hybrid between P. coloratus Hornem. and P. berchtoldii Fieb., is
known only from the British Isles. The names P. lanceolatus Sm. and its synonyms P. X lanceolatus var.
hibernicus A. Benn. and P. X perpygmaeus Hagstr. ex Druce are lectotypified. The distribution of P. x
lanceolatus is outlined and the history of its taxonomic treatment reviewed.

INTRODUCTION

Potamogeton X lanceolatus Sm. (P. coloratus Hornem. X P. berchtoldii Fieb.) is one of the few
hybrids between the broad-leaved Potamogeton species in Sect. Potamogeton and the species with
narrow, linear leaves in Sect. Graminifolii. It is a sterile plant, only known from the Bnitish Isles.
Since its discovery in 1806 three names have been given to it: P. lanceolatus Sm., P. X lanceolatus
var. hibernicus A. Benn. and P. X perpygmaeus Hagstr. ex Druce. The purpose of this paper is to
outline the circumstances in which these names were published, to lectotypify them where necessary
and to describe the taxonomic history and distribution of this rare hybrid in the British Isles.

POTAMOGETON X LANCEOLATUS SM.

Potamogeton x lanceolatus Sm. in Sowerby, Engl. bot. 28: t. 1985 (1809) pro sp. (‘lanceolatum’).
Lectotype: North Wales, August 1808, H. Davies, herb. J. Sowerby, BM! Isolectotype in BM!.

The description of P. x lanceolatus in English Botany was written by J. E. Smith and illustrated by a
plate drawn by J. Sowerby. Smith (Sowerby 1809) stated that the plants he described were sent to
him by the Rev. H. Davies ‘“‘from the lakes of North Wales”. There is no reason to suppose that
Smith saw the earliest specimen of P. x lanceolatus, collected by Davies in October 1806 (BM).
There is, however, a specimen in Smith’s herbarium (LINN, Savage manuscript catalogue 246.17)
labelled in his own hand “Potamogeton lanceolatum Engl. Bot. t.1985. North Wales. Rev. H.
Davies 1807’. There are two specimens in J. Sowerby’s herbarium (BM) labelled “‘Potamogeton
lanceolatum nova spec. Received H. Davies. N. Wales. August 1808. E.B. t.1985”’. All these
specimens were determined as P. x lanceolatus by J. E. Dandy and G. Taylor, and I agree that they
all represent this taxon. All conform to Smith’s description, and none is a better match than the
others.

Both the specimen in Smith’s herbarium (LINN) and those in Sowerby’s (BM) must be regarded
as syntypes. Those in Smith’s herbarium were presumably used by him to draw up the description in
English Botany. Sowerby’s specimens were probably sent to him directly by Rev. H. Davies. The
original drawings for English Botany are preserved in the British Museum (Natural History) and
notes on them show that some, at least, were based on specimens sent directly to Sowerby (e.g.
Schoenus monoicus). There is no such evidence on the original drawing of. Potamogeton xX
lanceolatus, but as Sowerby notes that it was drawn from fresh specimens it seems likely that they
came directly to him. The labelling of the specimens also suggests this. There is therefore no

310 CrD. PRESTON

evidence that Smith saw them. Close comparison of the two sheets in Sowerby’s herbarium with the
original drawing show that one plant is so similar to the drawing that the conclusion that it is
the plant actually drawn is inescapable. It is less easy to identify the plant drawn by comparing ©
the specimen to the published plate as there are some differences between the original drawing and
the published version, notably a young leaf at the apex of the stem on the drawing which appears on
the plate as an inflorescence in bud.

Thus there are available for selection as lectotype two specimens closely related to the
protologue, that studied by Smith who drew up the description and that drawn by Sowerby who
illustrated it. J. E. Dandy and G. Taylor have labelled the specimen at LINN as the “‘type”’ and the
sheet at BM which includes the plant illustrated in English Botany as the “‘lectotype”’ of P. x
lanceolatus. This introduces a further complication. If one considers that the labelling of the BM
specimen constitutes effective lectotypification then it must be accepted as the lectotype, as it could
only be rejected if it seriously conflicts with the protologue. However if one considers that a
lectotypification is only effective when published then there is a choice of lectotype, as Dandy and
Taylor did not publish their selection. If I choose the specimen in Smith’s herbarium as lectotype the
actual lectotype of P. X lanceolatus will be left in doubt, as it would depend on one’s view of what
constitutes lectotypification. This would be undesirable, so I have decided to uphold Dandy and
Taylor’s choice and cite it above as lectotype. It is reproduced as Fig. 1.

It is unfortunate that the specimens available to Smith and Sowerby lacked broad floating leaves,
one of the characteristic features of P. x lanceolatus. This was probably the reason for a number of
erroneous nineteenth century records of P. X lanceolatus.

P. LANCEOLATUS VAR. HIBERNICUS A. BENN

Potamogeton lanceolatus var. hibernicus A. Benn. apud Praeger in /rish Nat. 5: 243 (1896).
Lectotype: Clonbrock River, Co. Galway, June 1896, R. L. Praeger, DBN!
P. lanceolatus forma hibernicus A. Benn. apud Praeger in Irish Nat. 5: 243 (1896), nom.. provis.

R. L. Praeger collected P. x lanceolatus (as well as P. coloratus) from the Clonbrock River not far
from the House on 21st June 1896, probably on a pre-breakfast ramble (McWeeney & Praeger
1896). Material from Clonbrock was submitted to Arthur Bennett who, after describing differences
between it and plants from Anglesey and Cambridgeshire, stated “it might be called var. hibernicus
(or f. hibernicus), characterized by its longer, and broader upper leaves, longer lower leaves,
slightly longer flower-spikes and the structure of the leaves” (Praeger 1896). I follow J. E. Dandy
and G. Taylor (in their unpublished monograph British species of Potamogeton L.) in regarding var.
hibernicus as a validly published name, with f. hibernicus as a provisional name, not validly
published at that rank. There is no material of Praeger’s 1896 collection in Bennett’s herbarium
(BM). There are, however, two specimens in Praeger’s herbarium (DBN). One is labelled
‘“‘Potamogeton lanceolatus Smith var. hibernicus, A. Benn. in litt.”’ and I have designated this as the
lectotype. It is reproduced as Fig. 2.

There is some variation in the leaf shape of Potamogeton X lanceolatus (Figs. 2, 3). This appears
to be correlated with the age of the plant and with environmental factors. The lower stem leaves are
more or less linear and superficially resemble the leaves of P. berchtoldii, although they are broader
and differ in venation. Sometimes such leaves occur all along the stem, even on a flowering plant.
Other plants have broader upper leaves which are elliptical in shape and are more obviously
intermediate between the putative parents. W. Wilson noticed that in Anglesey the broader leaves
‘‘are always found when the current is slow’’ (Hooker 1830). They are well developed in plants from
Burwell which were collected from a ditch or cultivated in a small pond, but are not always found on
plants collected in the swifter water of the Afon Lligwy or the Caher River. The type material of var.
hibernicus only has more or less linear leaves. However I can see no difference between the leaf size
and shape of these plants and that of some collections from the type locality of P. x lanceolatus, the
Afon Lligwy (Table 1). I cannot understand why Bennett thought that the leaves of var. hibernicus
were longer and broader than those of the type: there is no reason to regard them as broader, either
in absolute terms or in relation to their length. The lacunae along the midrib, said by Bennett to be
much less conspicuous in var. hibernicus, are similar to those of plants from Anglesey. I can

POTAMOGETON xX LANCEOLATUS IN THE BRITISH ISLES 311

ee z-
~ :

pen
Set. 3. E. Beads WR S Faster

K Potamugetan ixmewolates Sen.
ts AE ISM KAS IRR Re
MOAVERSYS BERSARIOM. B

Ficure 1. The lectotype of Potamogeton lanceolatus Sm. (BM). The plant on the lower right-hand side of the
sheet is the one illustrated in English Botany, t. 1985 (Sowerby 1809).

312 C.D«PRESTON

Lecte tyre ERE
LPB 0 Kish Ka Pod prphcstaat PE Set,
Deweninarit ts 8 ak 13H 7
Potameyctan % Lamoecletu
ve A
x Lt Aye
&
‘ y
‘ eae
iy
Ps f 3
é 4 }.
hk Z LS a £34 f
74
y j
$ 14

FiGurE 2. The lectotype of Potamogeton lanceolatus var. hibernicus A. Benn. (DBN). The specimen is a good
example of the form of P. x lanceolatus without broad upper leaves.

“

POTAMOGETON x LANCEOLATUS IN THE BRITISH ISLES 313

,

488 PL coromaren x POsILocs Po prerro wares Hagetr JONI ; September 1991, ] took Mra Weegwood to this place vaiicy hoping
on Aft. ined =P tonecodatus Sui, var. hibers tews Ar. Thos The fe “3 fa\ it mecht te found in fruit. The Cahir etree Was ther Yeoant
plant wae discover) by Mr P. O'Kelly in the Cahir Rive, Ce ¢ vo} 4} fully soticcid and we shortly found the above Pondwee! i pr
Ulare. In June 1907 Ht as chown me by how It crew in « rapid of2 7 3/ 6m and in tery much better vondition than that obia::
oman hea irea. but | sould not find any P oraminery 7 r first visit A careful search also resulted in finding /
growing » but the water wie turbad from foods #0 UY search sorne UBB NY —aariall AP linhews only, so rarady fristitiy-—-thins uf
wes ed erhaustive As I tail in Pep GRE 378 y9080 hid roborating the ahrewd augey 2 of Dre Hagstrian Supqlies of bert
Dot strike ~ As idetitical with the Anglewey plant. Dred 9 Hae Plants were sent him and alth wigh near his cod he wrote this
arte etamined ™Y specimen and (/n frtt.. 1919 suggested that ‘ hybrid which 1, ad ‘Glerim, name Po perpyginacus in the test ir
while OBe parent Was pousiilis be believed that the seeard s: * Ws4 teresting Ireland owns I will write 2 he F rf om book
tlleratus The latter SPOOIGE Wee Not on roars eorut the Potar MEO P tpg er espe t Pf. laneen

fetus which was disporerad by Mr Arthur i meat wy Barwi For
. Cambridgeshire, in 1880, PF ohopacs BOM ite hybrid, PB a pors
slag cecur in that place. TH ioak sstagt
J “

SOL 10 perpygmaens which those 4% *
Breen of the Anzviesey /

recorded from the Continens

FicuRE 3. The lectotype of Potamogeton x perpygmaeus Hagstr. ex Druce (OXF). The specimen shows the
broad upper leaves often produced by the hybrid P. coloratus x berchtoldii.

314 C. D. PRESTON

TABLE 1. COMPARISON OF POTAMOGETON x LANCEOLATUS SPECIMENS FOR
CHARACTERS ON WHICH VAR. HIBERNICUS WAS BASED
Based on dried specimens in BM & CGE (Afon Lligwy & Caher River) and DBN (Clonbrock, the type material
of var. hibernicus).

Afon Lligwy Clonbrock Caher River
Lower leaves (mm) 23-75 x 2-5-5 (22—)40-72 x (2-)3-5 25-65 x 2-5
Involucral leaves (mm) 17-65 x 3-5-10 40-65 x 3-6 21-48 x 3-7
Peduncles (mm) 9-50 39-42 10-51
Width of midrib and adjacent lacunae (mm) 0-3-0-7 0-3-0-7 0-3-0-7

therefore see no reason to regard the Clonbrock plants as distinct from those from Anglesey, and
support the reduction of var. hibernicus to synonymy.

POTAMOGETON X PERPYGMAEUS HAGSTR. EX DRUCE

Potamogeton X perpygmaeus Hagstr. ex Druce in Rep. botl Soc. Exch. Club Br. Isl. 6: 630 (1923).
Lectotype: Cahir River, Co. Clare, September 1921, G. C. Druce, OXF!, photograph in BM!
Isolectotypes in BM!, CGE!, E!, K!, OXF! and probably elsewhere.

P. X perpygmaeus Hagstr. ex Druce in Rep. botl Soc. Exch. Club Br. Isl. 6: 580 (1922), nom. nud.

P. X lanceolatus was collected by P. B. O’Kelly in the Caher River in 1891, the first Irish record
(Levinge 1891). In June 1909 G. C. Druce was taken by O'Kelly to see it, and noted that it did not
strike him as identical to the Anglesey plant (Druce 1910). Later Druce submitted specimens to J.
O. Hagstrom, who in 1919 suggested that they might be P. coloratus x P. berchtoldii, despite the
fact that the former was not then known from the area (Hagstrém thought that P. x lanceolatus
from Anglesey was P. alpinus x P. berchtoldii). In September 1921 Druce, making “‘a somewhat
adventurous journey to Ireland just after the truce’, returned to the Caher River with Mrs M. L.
Wedgwood. He found P. x Janceolatus “in profusion and in very much better condition than that
obtained on my first visit’, together with P. coloratus in some quantity (Druce 1923a, b). On
examining the 1921 material Hagstrom suggested the name P. X perpygmaeus for plants from
Ireland and Cambridgeshire, and promised to write a note on it in a book he was writing. At first
Druce (1922) published P. x perpygmaeus as a nomen nudum, but he wrote a longer note about it
after Hagstrom’s death in 1922 (Druce 1923a). Although the second note does not contain a formal
morphological description, Druce validated the name by his remark that “the reddish colour of
coloratus is also present in perpygmaeus which thus differs from the beautiful grass-green of the
Anglesey lanceolatus”’. (The fact that one would expect the Anglesey plant to have a reddish colour
if it actually was P. alpinus x P. berchtoldii is not commented upon).

P. X perpygmaeus should be typified by a specimen from Druce’s 1921 gathering, of which there
are three sheets in his herbarium (OXF). I have selected one of these as the lectotype (Fig. 3). The
extract from the B.E.C. Report in which Druce validated P. X perpygmaeus is attached to this
sheet, as is a note from Hagstr6m beginning “This hybrid which I ad int. [erim] names P.
perpygmaeus is undoubtedly the most interesting plant Ireland owns ...”. This note from
Hagstrém is dated 22nd July 1921, presumably an error as the material was not collected until
September of that year.

Judging by herbarium specimens, the colour of P. x lanceolatus in Anglesey is no different from
that of the populations in Ireland. Some Anglesey collections have clear green leaves but others — as
Bennett (1924) pointed out — have a reddish brown tinge. Ironically this coloration is well developed
in material collected by Druce himself in June 1900 (CGE). It is unfortunate that Hagstr6m was

* A truce between the Irish Republican Army and the forces of the British Crown had existed since 11th July
1921. The I.R.A. had retained their arms, and their units would have been in evidence at the time of Druce’s
visit.

-“

POTAMOGETON xX LANCEOLATUS IN THE BRITISH ISLES 315

unable to publish his reasons for separating the Irish and Cambridgeshire plants as P. x
perpygmaeus. I can see no differences which would justify such a distinction, and therefore agree
with the view that they represent a single taxon.

TAXONOMIC HISTORY OF P. X LANCEOLATUS

Most 19th century authors maintained P. lanceolatus as a species (W. J. Hooker 1830, Babington
1843, Sowerby 1869, J. D. Hooker 1870). Of the authors of the standard Floras only Bentham
(1858), whose species concept in Potamogeton was extremely broad, included P. lanceolatus as a
synonym of P. lucens. However, some of the specimens attributed to P. lanceolatus were
erroneously identified and the concept of this taxon, not all that clearly defined by Smith, became
even more blurred. Thus the fifth edition of Babington’s Manual of British Botany (1862), the most
critical of the more readily available Victorian Floras and one written by an author who had a special
interest in Potamogeton, contained erroneous records of P. lanceolatus from Buttermere, Killarney
and Antrim. Babington added a description of the fruit of P. lanceolatus to this edition of the
Manual, but this was based on material which he later recognized as P. polygonifolius. J. T. Syme
concluded that all except the Anglesey specimens had been erroneously identified, a conclusion
which Babington accepted (Sowerby 1869, Babington 1872).

Babington was able to visit Anglesey in 1880, where he was shown P. lanceolatus by J. E. Griffith,
author of The Flora of Anglesey and Carnarvonshire. In his account of his visit Babington (1881)
stressed the fact that the fruit of P. Janceolatus was still unknown, despite the fact that he had visited
the site at the end of August and Gniffith returned in late October. “If it had not been for Mr.
Bennett’s discovery”, Babington commented, “we might have thought that the plant of Anglesea
was a barren form of some species, and that it propagated itself by offsets . . . but the discovery of
exactly the same plant in the Fens renders this improbable’. Bennett’s discovery of P. x lanceolatus
in Cambridgeshire stimulated interest in the plant, and with the publication of his paper in the
Journal of Botany (Bennett 1881) an excellent drawing of P. x lanceolatus, accompanied by an
accurate and detailed description, finally became available to British botanists.

Alfred Fryer was the first to suggest that P. x lanceolatus was a hybrid. In an early discussion of
the possibility of hybridization in the genus (Fryer 1888) he called it “‘one of the best and most
distinct species of Potamogeton’’, citing it to demonstrate that “barrenness is no proof of hybridity”’.
He later came to realize that it was a hybrid between a broad-leaved and a narrow-leaved species,
suggesting the parentage P. gramineus (as P. heterophyllus) x P. berchtoldii (as P. pusillus {auct.])
for the Anglesey plant and P. gramineus x P. friesii for Cambridgeshire material (Fryer 1894).
Hagstrom (1916) considered that the Anglesey plant was P. alpinus x P. berchtoidii, and later
suggested the parentage P. coloratus x P. berchtoldii for plants from Cambridgeshire and Ireland
(see above). Bennett (1919, 1925) refused to accept that either P. alpinus or P. coloratus was the
broad-leaved parent of P. x lanceolatus, still favouring P. gramineus. However P. coloratus X P.
berchtoldii is now accepted as the parentage of P. x /anceolatus at all its localities (Dandy 1958,
1975).

DISTRIBUTION OF P. X LANCEOLATUS IN THE BRITISH ISLES

CAMBRIDGESHIRE (V.c. 29)
Bennett (1880) discovered P. x lanceolatus at Burwell Fen on 4th August 1880. The exact locality
was a ditch on the right-hand side of Burwell Drove, going from Wicken to Burwell (CGE). In his
first notice of the plant Bennett (1880) stated that it grew with P. coloratus; in a later paper he said
that it grew with P. gramineus in a ditch lined with Eleocharis acicularis (Bennett 1925). It was in the
later note that he argued for P. gramineus rather than P. coloratus as one of the parents of P. x
lanceolatus! Specimens of both P. coloratus and P. gramineus collected by Bennett at Burwell on the
day he discovered P. X lanceolatus were seen by J. E. Dandy and G. Taylor.

Alfred Fryer searched repeatedly for P. x lanceolatus at Bennett’s locality, without success. He
reported (1894) that the ditch in which it grew “thas now closed up, and its aquatic vegetation is
extinct’. He was unable to discover it in nearby water-courses. As a plant relying entirely on

316 2 C.D. PRESTON
vegetative reproduction for its survival, P. x lanceolatus is presumably ill-adapted to survive a
period when its habitat becomes overgrown.

Bennett cultivated the Burwell plant in a small pond in his Croydon garden — it “flourished
remarkably” and he was able to distribute 200 specimens of cultivated material (Bennett 1925). The
plate in Fryer & Bennett (1915) is based upon Cambridgeshire material.

ANGLESEY (v.c. 52)

Despite the fact that Smith stated that the plants he described as P. x lanceolatus were sent to him
by the Rev. H. Davies “from the lakes of North Wales” (Sowerby 1809), there is no doubt that they
came from the Afon Lligwy, a stream in Anglesey. Davies (1813) himself cites the locality as “the
rivulet between Bodafon and Lligwy” and the Afon Lligwy remains the only known Welsh locality
for the hybrid. It was collected there repeatedly after its discovery by Davies. In 1875 C. Bailey
found that it grew “in longish patches to the exclusion of anything else near it. In a few places where
the stream is slower, and more water in, it simply fills it up, from the bed to the surface, so that ducks
cannot comfortably paddle in it; in the swifter places it is in the middle of the brook, forming patches
of two to three yards long by six to twelve inches broad” (Bennett 1881). Five years later Babington
(1881) “found plenty of the plant growing in the quieter parts of the rather rapid stream, both above
and below the bridge of Penrhos-lligwy”’. Many botanists saw it subsequently, most recently R. H.
Roberts who recorded it in c. 1960 at grid reference 23/483.854 and in 1968 near the bridge where the
A5025 road crosses the Afon Lligwy, at 23/488.862. On ist August 1987 Ms Vv. Morgan and I
attempted to refind the plant in the Afon Lligwy, but without success.

CO. CLARE (v.c. H9)

Asin Anglesey, P. x lanceolatus has been regularly collected in the Caher River since its discovery
by P. B. O'Kelly in 1891. Webb & Scannell (1983) describe it as “plentiful in the Caher R. over most
of its course’’. On 26th June 1984 K.G. & J.O. Mountford recorded Apium nodifiorum, Eleocharis
palustris, Juncus articulatus, Mentha aquatica and Ranunculus flammula as associates of P. X
lanceolatus in the Caher River at grid reference M160.080.

N.E. GALWAY (v.c. H17)

Two localities for P. x lanceolatus are known in Co. Galway. It was discovered in the Clonbrock
River by R. L. Praeger in 1896 and refound in the same river in 1952 at Clonbrock (N. D. Simpson,
BM) and by Island Bridge near Ahascragh (R. B. Drummond, K). The other site is near Tuam,
where it was discovered by R. L. Praeger in 1899. The most detailed specimen was collected on 13th
July 1899 (BM) and is labelled “rapid shallow pebbly stream, Grange River, Barbersfort near
Tuam”.

OTHER EUROPEAN HYBRIDS BETWEEN SPECIES IN SECT. POTAMOGETON AND SECT. GRAMINIFOLII

Three more hybrids between Sect. Potamogeton and Sect. Graminifolii are known in Europe. Two
are similar to P. X lanceolatus: P. x rivularis Gillot, P. polygonifolius x P. berchtoldii, which is
known from France, and P. X miguelensis Dandy, P. polygonifolius xX P. pusillus, described from
the Azores (Dandy 1970, 1975). The third hybrid is P. X vartifolius Thore, P. natans x P.
berchtoldii, recorded in France and Ireland (Dandy & Taylor 1967). The identity of P. x heslop-
harrisonii Clark, described from the Outer Hebrides, has yet to be elucidated. Clark originally
reported this as P. gramineus X P. berchtoldii, later amending this to P. gramineus X P. perfoliatus
x P. berchtoldii (Clark 1942, 1943). Heslop Harrison (1949) commented that the parentage P.
alpinus x P. berchtoldii could not be excluded. According to Clark (1943) the type material was
deposited in the herbarium of the Department of Botany, King’s College, Newcastle upon Tyne
(NCE), but I cannot find it there now, nor is it at HAMU, in Clark’s personal herbarium (BM) nor at
K.

ACKNOWLEDGMENTS

I am grateful to J. R. Akeroyd for contributing the footnote describing the situation in Ireland in
1921, to K. G. & J. O. Mountford and R. H. Roberts for their records of P. x lanceolatus, to Miss

POTAMOGETON xX LANCEOLATUS IN THE BRITISH ISLES 317

M. Scannell for the loan of specimens from DBN, to Sir George Taylor for the loan of the manuscript
‘British species of Potamogeton L.’, to A. O. Chater, J. R. Edmondson and C. E. Jarvis for
discussion of the typification of P. x lanceolatus and to P. G. Ainsworth, A. O. Chater and Miss S.
K. Marner for obtaining photographs of herbarium specimens. Ms V. Morgan kindly joined me in
an unsuccessful search for P. x lanceolatus in Anglesey and P. S. Davis, T. C. Dunn, J. Heslop-
Harrison, A. J. Richards and D. A. Simpson assisted in an equally fruitless hunt for the type of P. x
heslop-harrisonii. J. R. Akeroyd, R. H. Roberts and P. D. Sell commented on a draft of the paper.

* REFERENCES

BaBINGTON, C. C. (1843). Manual of British Botany, p. 323. London.

BasinGTON, C. C. (1862). Manual of British Botany, 5th ed., p. 352. London.

BaBINGTON, C. C. (1872). Notes on Potamogetons. J. Bot., Lond. 10: 228-229.

BaBINGTON, C. C. (1881). On Potamogeton lanceolatus of Smith. J. Bot., Lond. 19: 9-11.

BENNETT, A. (1880). Potamogeton lanceolatus, Smith, in Cambridgeshire. J. Bot., Lond. 18: 276.

BENNETT, A. (1881). On Potamogeton lanceolatus of Smith. J. Bot., Lond. 19: 65-67, t. 217.

BENNETT, A. (1919). Notes on British Potamogetons. J. Bot., Lond. 57: 10-20.

BENNETT, A. (1924). Notes on Potamogeton. Trans. Proc. bot. Soc. Edinb. 29: 45-53.

BENNETT, A. (1925). Potamogeton lanceolatus Smith. Rep. botl Soc. Exch. Club Br. Isl. 7: 658-661.

BENTHAM, G. (1858). Handbook of the British Flora, p. 493. London.

CiarK, W. A. (1942). A hybrid Potamogeton new to British Isles. Vasculum 27: 20.

CLarRK, W. A. (1943). Pondweeds from North Uist (V.-C 110), with a special consideration of Potamogeton
rutilus Wulfg. and a new hybrid. Proc. Univ. Durham phil. Soc. 10: 368-373.

Danpy, J. E. (1958). List of British vascular plants. London.

Danpy, J. E. (1970). Potamogeton and Ruppia in the Azores. Bol. Soc. brot. 44: 1-7.

Danpy, J. E. (1975). Potamogeton L., in Stace, C. A., ed. Hybridization and the flora of the British Isles, pp.
444-459. London.

Danpy, J. E. & TayLor, G. (1967). Potamogeton berchtoldii x natans = P. X variifolius Thore, in SELL, P. D.
ed. Taxonomic and nomenclatural notes on the British Flora> Watsonia 6: 314-315.

Davies, H. (1813). Welsh Botanology, p. 18. London.

Druce, G. C. (1910). Potamogeton lanceolatus, Sm., var. hibernicus, Ar. Benn. Rep. botl Soc. Exch. Club Br.
Isl. 2: 476.

Druce, G. C. (1922). Potamogeton perpygmaeus Hagst., ined. Rep. botl Soc. Exch. Club Br. Isl. 6: 580.

Druce, G. C. (1923a). 2488. P. coloratus X pusillus = P. perpygmaeus Hagstrom, in. litt., ined. Rep. botl Soc.
Exch. Club Br. Isl. 6: 630.

Druce, G. C. (1923b). 2493. P. gramineus x polygonifolius = P. Seemenii Asch. & Graebn. Rep. botl Soc.
Exch. Club Br. Isl. 6: 631-632.

Fryer, A. (1888). Notes on pondweeds. Potamogeton fluitans Roth. J. Bot., Lond. 26: 273-278.

Fryer, A. (1894). Potamogeton rivularis Gillot. J. Bot., Lond. 32: 337-340.

Fryer, A. & BENNETT, A. (1915). The Potamogetons (Pond Weeds) of the British Isles. London.

HaGstr6M, J. O. (1916). Critical Researches on the Potamogetons. K. svenska VetenskAkad. Handl. nov. ser.,
55 (5): 1-281.

Harrison, J. W. Hestop (1949). Potamogetons in the Scottish Western Isles, with some remarks on the general
natural history of the species. Trans. Proc. bot. Soc. Edinb. 35: 1-25.

Hooker, J. D. (1870). The Student’s Flora of the British Islands, p. 371. London.

Hooker, W. J. (1830). The British flora, p. 75. London.

LeEviInGE, H. (1891). Potamogeton lanceolatus in Ireland. J. Bot., Lond. 29: 344.

McWEENEY, E. J. & PRAEGER, R. L. (1896). Notes on the fauna and ffora of Clonbrock, Co. Galway. Prefatory
Note. Ir. Nat. 5: 217-221.

PRAEGER, R. L. (1896). Notes on the fauna and flora of Clonbrock, Co. Galway. Flowering plants and vascular
cryptograms. /r. Nat. 5: 239-244.

Sowerby, J. (1809). English Botany 28: t. 1985.

Sowerby, J. (1869). English Botany, 3rd ed., 8: 34-35.

Wess, D. A. & SCANNELL, M. J. P. (1983). Flora of Connemara and the Burren, p. 222. Cambridge.

(Accepted February 1988)

See ae
‘sibs € i ay
a enetecchn sabi yom!
ng oat} 30h tecutl zeedniunt vil an

Stee

oe aT

<2 a
eS rs i Revs qo batnomaye beagee
; Sagpite tne Cf “fats pa Stated Watts spline

& a
ita) ninth. A om
— i 4 eile "yet -

3 al UT —~ aS 37 ‘

he * hah ike the ai ake a
rsa as x Wiss oA

cicact com lita biripe ts hakiaestit Aratinagiecaadiogant eae
att 9 20 (here itirdetne: eres an ioopbarn imag sabi orheiene ? (Horatnssi

‘ .
+ .
‘ ae » WE .s Te Pa
— ath tes | Setar hh tne OFS: Pane st ett iw splencasth
> - : . : : rue ¢, >? oF elem, Ves oo 47 a2
; = C Ayvles Wea ee a 1 Shale y IIL DE sgt ‘
— bs 5 a =. os
? a. mye SS 4 5 ca mas oh
abcapen wt Se Pt We ot ae bo PS et i
a4: , i t 2 Al oo 4 = en oho 71%
io a Sat ey “ ;
i ‘a es 2Ee SRR Sie } mas) x oo hg tury Oe ay Apt wo yh nermel
‘ 3 * *. ; tad bar | /*)
— = @*: t.. 7 i : ,
2 ay feist awe > 0
é —_ 7 7 7 J 3 .
‘ : ‘ so ad rn t> 1
: bide a A S bread
a A eu PVeR At ee ae versa
% iS Tega Ube A ee Ate
« = e wv ete Ae = . .
4 . D te 4.4 ¢
= 4 i =~ )*
Y . . 42400}
=) > - Y pe
he '« = =< Sit: f Age ANT y ’
= a . ~~, © 4». Deon etee Ae st ese
a * < a ¥ P ‘
‘ : ; ‘ q ‘ } a>. it . / pies .
ae ey as a 7 f
= ’ ® Pd #4 4 7 -
as — - 4 : er i ay es, § ™ hee tah
i ee e338 Arts é Pr ak aE 1 he - ~~ ap tases ee VS PASSAT ay ’ 4 ,
. 7 \

; 1S RESTE 8 evn evoby denise sate a pasion tt utetniscie

Y, 5 4
ar rity # 5% # Lyte oloitai
pa ; ws > z r ~* P ‘4 . » at ? seu » if ‘ ~,
2-54, 1 re AS Poi H $f Pad . ae a, aes way
£7 Sree Sa ee PVET ie O57
“ei —. ~~ 4 : aA rein . ~~ an eon A . :
, : 15 2 tee ; ae) = & ame ES ¢. x irine Haicaly Silk
hit. Se ah poe , BoP caver So ee supa ekytopd d eee
Ps,
igs iain ae Stile. xrinaie of ten ll
.. 23 ; , hl : i > Vise Fae Se Tipe
4 ~~ pi ES 1— J a r ee \ = ‘
Tree “4 sflanie? a ou, — x 4 = AVUGY gayi & eo CE ey ha xt aS.
. : - 4 ? ”
ae Seas 8. €
+
a ee She 7 » oe : t mm Ab “a <% oh .
f 2 Pa “s > ; i ' sh 4 Di ; ws ate t < bese bnog Bi 10° ‘aor
a - _e f “+ ee 2) :
a ’
' tT . . . *) Clssoit Bus)
_ * —. |
P s *24t B43 A? ’ Mik: f
t t \ ~~ -— = eu af
‘ . = d ‘ f q riv he \
: Dis RP eA ai Dy SIH iMigl mieseoF Msi yOoe Sat or 2tolaes!
7 m4 mrt AA MAS 43 yO AT
o . ; : - . i » >. “Se
¥ , is. tees : j wll iy ~ Sn? ta sh be f :
: a1 3 hate OTs ye ‘a c rales mh
eA « ~ + ‘

_ hor conll C% e643 agnky Wiciaaes ne wet Phy aioe i
opygiete th], gan ab oo .deotleoag wk teil ee ne eitul gett) 1b. 5

Ph

- ry's : i 7) 7’) iL ogee
: ee ' a ; nelane tt tars + eee ee
en Dose @) 1 eee Le | wig “ (A 8) Ms icy PMES bfiatt i it ¥ +) 4 * Z 73)
. 2 “ :
‘ A j ‘ i} "4 3% i} ‘
j _—— "7 PRU. . he
‘ ‘v4 GAP 4 Vital iad
7 r ' ye +3 apt? mel Ee ae
uO, TS , eS oe

— m! q Wao s4 i Sraseramnrgy.} io gol4 ARR)

(wees 1 re 4 7 —~
+ r - 4 6 Vitoes./
Fi
. 4 —
a pte
i > rs ig 3 Ka gy
'
- =
oe » rag &
q Li 7
a
*
¥) -

Watsonia, 17, 319-349 (1988) 319

A reappraisal of the British and Irish dactylorchids, 3.
The spotted-orchids

R. M. BATEMAN
Paleobiology Department, Smithsonian Institution, Washington D.C. 20560, U.S.A.
and
I. DENHOLM

Rothamsted Experimental Station, Harpenden, Herts., AL5 2JQ

ABSTRACT

52 morphological characters were recorded for ten individuals from each of 20 populations of Dactylorhiza
fuchsii (Druce) So6 and 23 populations of D. maculata (L.) So6. Multivariate analyses of data for both species
demonstrate largely continuous morphological variation polarized into two independent trends representing
vegetative vigour and degree of pigmentation. Both trends reach greater extremes in D. fuchsii. Possible causes
of this variation, including environmental modification, are discussed. Univariate, bivariate and multivariate
analyses demonstrate considerable overlap between existing and potential infraspecific taxa of both species.
Revised diagnostic descriptions are provided for five British and Irish varieties of D. fuchsii: var. fuchsii (Druce)
Bateman & Denholm, comb. nov., var. hebridensis (Wilmott) Bateman & Denholm, comb. nov., var.
cornubiensis (Pugsley) So6, var. alpina (Landwehr) Bateman & Denholm, comb. et stat. nov., and var. okellyi
(Druce) Bateman & Denholm, comb. nov. The morphological distinction of some taxa previously regarded as
subspecies has been exaggerated due to insufficient or unrepresentative data and unsubstantiated assumptions
regarding their population biology. Their distributions have been confused by a priori identifications of
populations based on geographical locations rather than morphological criteria. No justifiable infraspecific taxa
can be detected within D. maculata. Several long-standing taxonomic and nomenclatural controversies are
resolved by detailed study of both the new data and past literature. Solutions to other problems, including the
possible conspecificity of D. fuchsii and D. maculata, require comparable data from Continental spotted-orchid
populations.

INTRODUCTION

The taxonomy and nomenclature of the spotted-orchids have evolved progressively but spasmodi-
cally since Linnaeus described Orchis maculata L. in 1753. Dactylorhiza maculata (L.) So6 (Heath
Spotted-orchid) is now more clearly and narrowly delimited, and has been joined in the Subsection
Eumaculatae Vermeulen by D. fuchsii (Druce) So6 (Common Spotted-orchid; Druce 1915) and the
exclusively Continental D. saccifera (Brongniart) Soé (Brongniart 1832). Senghas (1968) and
Landwehr (1975) disagreed with Vermeulen’s (1947) surprising inclusion of the Madeiran endemic
D. foliosa (Solander) So6 in this subsection. Some workers have advocated the amalgamation of D.
fuchsii with D. maculata (e.g. Nelson 1976) and the splitting of D. saccifera (e.g. Baumann &
Kiinkele 1982). However, this degree of taxonomic instability at the species level is greatly exceeded
by a general lack of consensus over numerous infraspecific taxa, which reflects the difficulty of
partitioning largely continuous ranges of morphological variation into satisfactory taxa (Prentice
1986). This problem requires detailed morphological studies capable of identifying and quantifying
trends in variation that can be assessed for potential taxonomic value and interpreted in an
evolutionary context.

This paper continues our morphometric studies on dactylorchids (Bateman & Denholm 1983,

320 R. M. BATEMAN AND I. DENHOLM

1985), and concludes our survey of infraspecific morphological variation in British and Irish
populations.

MATERIALS AND METHODS

20 populations of D. fuchsii and 23 of D. maculata were sampled between 1982 and 1986, including
at least three populations of each of the subspecies listed by Heslop-Harrison (1954) and Clapham
(1962). Details of these populations and the sample localities are presented in Tables 1 and 2.

52 morphological characters were recorded for each of ten flowering plants per population
(twelve plants of D. maculata were measured at Kilmory). Details of these characters (and of ten
indices (a—j) derived from them) largely follow Bateman & Denholm (1985), though an additional
category (S = + solid blotch) was introduced for character 11 (type of labellum markings) and mean
peripheral bract cell lengths (character 26) represent 30-100 cells, not 10-30 as was erroneously
stated previously.

Tables listing mean values and standard deviations for all characters and indices of each study
population are available from the authors on request.

Data were analyzed by univariate and multivariate methods using the Genstat computer program
(Payne et al. 1987). Characters 46-51 were excluded from the multivariate analyses to avoid bias
caused by a series of zero values reflecting the absence of a single feature (i.e. leaf markings), and
indices were excluded to avoid duplication of their component characters. D. fuchsii and D.
maculata were analyzed separately. After standardization to unit variance, each species yielded two
similarity matrices (Gower 1971), the first using population means and the second using data for
individual plants. Each matrix was used to link either populations or individuals by their maximum
similarities to yield a minimum spanning tree (Gower & Ross 1969) and then compute principal
coordinates (Gower 1966), compound vectors incorporating correlated characters of potential
diagnostic value. The first two principal coordinates (PC1, PC2) from each analysis were plotted
together to assess the degree of morphological separation of potential taxa in these dimensions (see
Bateman & Denholm (1983, 1985) for further details of the application of these techniques to
dactylorchids).

MAJOR TRENDS IN MORPHOLOGICAL VARIATION

The multivariate analyses of both populations and individuals demonstrated strikingly similar
overall patterns of infraspecific variation in D. fuchsii and D. maculata. In each of the four analyses
(Figs. 1-4), the first two principal coordinates largely represented variation in vegetative vigour and
intensity of pigmentation; lower order coordinates consistently accounted for much smaller
proportions of the variance than the second and were of little biological or taxonomic significance.
For simplicity, we have used our revised classification of D. fuchsii into five varieties (shown in the
left-hand column of Table 1) throughout this paper.

ANALYSES OF POPULATIONS
19 of the 46 characters used for multivariate analysis of D. fuchsii populations contributed
appreciably to the first two principal coordinates, which together accounted for 50% of the total
variance (Table 3). The second coordinate (PC2 on Fig. 1) largely represented characters
determined by vegetative and floral anthocyanin pigments, and clearly separated four anthocyanin-
low populations from the Burren (var. okellyi; PC2 values <—0-2) from eight anthocyanin-high
populations (vars hebridensis, cornubiensis and alpina; PC2 values >0-1). The eight remaining
populations (var. fuchsii) had intermediate values for PC2 but were widely dispersed along PC1,
which represented characters such as stem, leaf, bract and inflorescence dimensions that determine
vegetative vigour. Thus, Oldshore and Polin were the least vigorous and Oaklands and Parham
were the most vigorous of the populations studied. Five populations of var. fuchsii showed greater
overall vigour than all populations of the other varieties.

The superimposed minimum spanning tree resolved variation along these axes into five branches,
radiating from the area occupied by less vigorous populations of var. fuchsii in the centre of the plot

BRITISH AND IRISH SPOTTED-ORCHIDS 321

0.4
Baleshare
AX
PS be
: bat
0.3 ; a
Lelant =
‘ Va “DO Robach
/
/
/
0.2 / ;
/
Tintagel w react abe
/ ram shape lie > Oldshore
/ 7 ! =O a
- Rd a : Ashes Gill ON Polin
a 7
PC2 ana I oer
, Oddy Hill ie
0 ecSteual T-
SS Se a og
Torrin
Harpenden Pitstone \
-0.1 \
\
\
Oaklands @ .
-0.2 ;
Khyber Pass
Black Head
“0.3 S Corrofin
Gelain
-0.4 :
{0S -0.4 -0.3 -0.2 -0.1 0 Ost Oo2 0.3 0.4

PG1

Ficure 1. Principal coordinates plot of PC1:PC2 with superimposed minimum spanning tree. D. fuchsii,
population means.

@ var. fuchsii (® individuals lacking floral anthocyanins; Fig. 3)

QO var. hebridensis

@ var. cornubiensis

O var. alpina (© coastal habitats)

A var. okellyi (A individuals lacking floral anthocyanins; Fig. 3)

Minimum spanning tree links: double line = similarity of populations >95%, single line = 92-6-95%, dashed
line = 90-92:5%, dotted line = <90% Inset: Main trends of variation represented on the plot.

V = vegetative vigour

F = floral dimensions

A = anthocyanin pigments.

(Fig. 1). The eight anthocyanin-high populations formed three branches: (i) var. hebridensis
(Baleshare, Robach) and var. cornubiensis (Tintagel, Lelant), (ii) upland populations of var. alpina
(Keltneyburn, Ashes Gill), and (iii) coastal populations of var. alpina (Oldshore, Polin). The
strongest links of the tree reflected a combination of geographical and ecological proximity; they
occurred between the four populations of var. okellyi from the limestone pavements of the Burren,
two adjacent populations of var. alpina from machair in Sutherland (Oldshore, Polin) and two
populations of var. fuchsii from chalk soils near the Hertfordshire/Buckinghamshire border (Oddy
Hill, Pitstone).

The first two coordinates for D. maculata populations incorporated 44% of the total variance,
largely encompassed by 21 characters (Table 4).-They resembled closely the corresponding

R. M. BATEMAN AND I. DENHOLM

322

6H OA

(dW Xd ‘(4) dW XoW 9UO}]SOUN] ‘QIR[D “OD ‘uyol0D
‘(1)4AXOW snosayiuoqied ‘N ‘NIVITE9
‘(O)AW *(2)9W. oc/L 001 — MEL ce Lr6 €LE/I1 y8no7 ‘juowaaeg

6H ‘9A ‘91RD
‘0D ‘NIAWOUUNOD

JU0}SOWIT] ‘A ‘oped
(DOW ‘(IDI ‘(0)9W ¢/9 OS pO°L SNOJO}UOGIe’) cl 076 PCE/TT ysnoy “MA ‘oinsed
(Ga)OWXd “GAS xd | 7 OS
‘(IAVOW XdI 9U0}SOUWIT] ‘QIRID ‘OD ‘Yysoo1mN
‘(A).LW ‘(DOW snoJaytuoqies) ‘d ‘SSVd YAdAHy
‘(a)dI ‘G@A)IT “GW UL 002 r0°9 peod uyL = =— 08 880°0ST/ZI aug Aste Ig)
QUO}SOUIT] 6H ‘9°A ‘aleID ‘09D
snosajuoqies) ‘ysooumny ‘GVaH
(OW ¢/9 0OT ee ATLL O¢ OTT OST/CI MOV TA ‘Wwouwsaed 14]]9YO “IRA
pOl ‘9° ‘sapnqgy
9uO}SOUNT] 'N ‘oA¥S “NIMUMOL
dUON VL CZ 9-1 snoJojIUOgies) Ov 107 8ZS/81 ‘mopeow dweqd
9U0]SOUIT] 9ZH *9°A ‘okey
snosojuoqies ‘0D ‘UVAATLSVO
QUON WL 0S LL AVL Or L68°S9T/T1 ‘mopeow Aysieyy
07
‘OA ““S}IOH “UOWWOD
yeyo NHCNHdauvVH
ouON GUL OOT 9°¢ /SYUNTJ-YIIM-ALTD) al Tel 6E1/CS ‘pur[poom Aqqnisg

ST “9A “HPOJING
‘preysire WVHUVd

osuON sb-£/9 009 vee WEL ct 609° €¢¢/79 “11B9 MOTTE

07 “O:A “sey

| ‘yqeay Aauyog ‘syd

(0)IdW Xt *(14)4d N s€-7/9 00001 O-L WL 08 Z80'S8I/7S PARAS SNWTAVO

vO 9A

‘“syong ‘YJIOMsSIey

ouON st—-C/9 000S ee AIBYOD Oc! CHI 8€6/CP ‘ud ANOLSLId

OZ ‘O°A “SHO

‘UOTISSIM “TTIH

(14) 1dW XA st7c/9 OOT ae 424 OLI 601 'PL6/Cr AddO puejuMog

SI ‘ovA “Quay

J ‘9U0}SH[O4 “TTIH
suON v/9 007 vL AIBYO Scl O8E E1C/19 ATLSVO ‘puejumMod nsyonf “es
SptyosopAjoep zpotod sjue|d jo (O°H ut) sje9jeu (aow) doudl19J01 AjITeoo] pue 1ejIQeH {UOxe |

rayjo yo (Aouonbaay SULIOMOY ye9qg ‘ou ‘xolddy yd ]IoSs jusied [10S opnintyv pug

pure) s0udso1g

HSHONA ‘d ‘SNOILVTNdOd AGNLS GNV SHILITVOOT ATdNVS JO STIVLAG ‘T ATAaVL

323

BRITISH AND IRISH SPOTTED-ORCHIDS

dUON

(J)IdW XA ‘(O)IdW

(O)dWxA
‘dW *()OI

(4) dW XA
‘dW ‘()O1

dUON

QUON

dUON

(4)dW x4
‘(O)dW ‘G)W

t-c9

£/9

Cale

C/L

e/L

e/L

(Aol 8

U/L-v/9

‘arei AIOA = .JA, ‘olel = J, ‘JeuoIses00 = (0, ‘juonboyy = J, ‘UOWUWIOD = 9

‘uosvas 91R] Ajaenonsed v ‘cg6] SuLNp painseayp e
‘uoseas Ajieo Ajiepnonied & ‘7g6, SuLINp poinseay .
‘ojdwies ul 10}}eW STURSIO Jo s}JUNOW 981e] AIDA 0} ONp aJeUTXOIdde anyjeA ,

actly}

saplomauiajsunva] “dsqns syvlpu “gq = LW

pssiudajapvad ‘dsqns syvlow ‘gq = IdWw
pjjaandind ‘dsqns syvlpu ‘gq = dW
sypjuapioz0 ‘dsqns syvinu ‘gq = OW
pjyjayajnd ‘dsqns vinuspoul ‘gq = dj
pjuandd ‘dsqns pivudvoul ‘qd = IO]
paui1d909 ‘dsqns pivusvoul ‘q = DI
pypudvoul ‘dsqns pjvudvoul ‘q = I]
DIDINIDU ‘Gq = IW

nsyonf ‘q = A,

“YJUOUW Jey} JO Syoom oY} Joye sIsQUINU ‘YJUOW dy} soyvoIpUT ONbIygO sy} d1OJoq Joquinu sy] “onyea payewunsyY -
"QINJL[DUIUIOU PISIASI OJ ,LUONRIYISSe]D, 99S ,

OC

OOT

OO

Of

000S

OO0S

OSt

007

6°S

6:9

Te

Ld

129

/ SPAR]
/sjjny ueQueUIg

$9}P]S URITUOAIG

/pues umMolg

SSIous UPISIMI]
/pues uMOo|g

SSIoUs UPISIMIT]
/pues uMOIg

SSIOUS UPISIMIT]
/pues umMOolg

SsIoUs UPISIMI]
/pues uMOIg

dUO}SOUIT]
uelquied/{LL

ISIYOS BOTU/|TLL ¢

OL

Of

067

768 °S90/07

C8E CPS/OL

COL EL8/80

819° 6LL/80

L8S 961/67

88S 861/67

96b CLL/LT

v8L LLL/ve

Z OTA “T[EMUIOD

‘d “THOVLNIL
‘sjjtjo Aouissog
‘puvjsseis ysnoy

[ (O°A “][BMUIOD “AA
‘S9AT IS “LNW TH 71
‘sounp 3[qeI1S

OIL 9A

‘sapliqoH J01NO ‘1sIQ
"N ‘ALIOJUOIMIN
“‘HOVEON “eyor
OTT “9A

‘sapluiqoH J9IN© ‘I1sIQ
'N ‘AUVHSATVA
‘TeYyoRUeUed |,
‘Teyoryy

SOT *9°A “puerpioying
“MM “dag D10Y4SPIO
“NITOd “Aleyor

801

‘OA “purpoyINS “AM
‘10W AUYOHSATO
“Weyoeyp|

QQ OTA

‘ylod PIN ‘Tesun104
‘NUNGAANLTAM
‘oinjseg

9 ‘OA **syIOX “Mm
“PIA ‘uoj2[9U] “TTIO
SAHHSV ‘enIsed

sisua1qnudoo
“IVA

sisuapliqoy
“ICA

DUId]D “IRA

‘

ZS oA ‘Aasa[suy —
‘uowWoD NVIDAGNVTI

(DdN X9W “(dW 00Z a S}SIYOS PUOWL/TIELL 06 Lvl 9LS/€Z ‘purjioow dueqg

9UO}SOUIT] —9 et )ereN ““syIoX

SNOIOFIUOGIe) ‘ N-PIAW “QUO}SIUOZ “NOON

(2).LW ‘(DA <£-7/9 08 8-¢ /WB/uMianyye +1839 g COZ 199° P86/rE ALIHM ‘Yysieuw apisuresiys

9UO}SOUIT] v9 or sN

(DdN Xd ‘GA)dNxoW SNOIOJIUOGIeD ““SylOX “M-PIA ‘UO}9[8uUT

‘(1DAXOW ‘(O)dW ‘(3 ot/9 000S 0°S Manyeod uryy 087 S8LLLL/PE “T1ID SAHSV ‘FnIseg

3 OZ (9°A “'S}IaH ‘UOWWOD

(4A) 4 xoW ‘(4A) 4 <t/9 OOT 9-¢ WEL 08 800° SZI/ZS GOOMLANOMMN A ‘We2H

IZ °9°A ‘xaso[ppry

‘uowWOD FYOWNV.LS

= sUON t/9 Bl 9-9 spues 9jedArID S7I Ip6 6ST/IS ‘YSIBU OpIsueaNS

5 puesusai3 LI “OA ‘Ad1INS “UOUILUOD

= (9)dI b-€/9 Ns +0°9 TOMOT/APIG =» $C OTP’ P06/LP AF TISUNHL “80g

Z TT cA

A “sjueH ‘S ‘oIpjog ‘wolnl0g

ouON €/9 00S 8h s[oAeis neaie[g O€ L86 6¢¢/0F CGYOAMOONUD 'S “WIe2H

= IT ‘oA ‘‘syuReH

‘§ ‘nolnesg ‘WwoNog

< (J)dI p-€/9 O01 rSP spurs uoVeg/jeodg 02 190°09€/IP TIIHda LS ‘N “30g

7. II ‘o°A ‘*sjueH “S ‘aspo7y

< SUON p-€/9 0001 IS spurs uoVeg/eod 02 €LO°€€€/Ib Auusg ‘sog AA TLLVW

= fp ‘oA “UOADG

e “N *IOOWXA “UOUWIWIO)

< dUON ob/L O01 Cr ayqAyd ueoasg Olr Z9P'9b8/1Z MOOTUOd “‘Wwe2H

A [ ‘O°A ‘][BMUIOD ‘prezrq

Ss sUON €-7/9 00001 ZS ayunuodias/}eog OL Zrl 789/01 ‘2002 FAONWNAN “W1e2H

pe -SprysrojAjoep 19410 Jo zpotsod sjurjd Jo (OH ut) sjeliojew (do w) DOUDIOJOI Ajlyeoo] pue yeyQey
(Aouanbayy pure) s0U9so1g SULIOMOY YeIg ‘ou ‘xoiddy Hd jos juored [Ios opnin[~y Puy

‘suUOTRIADIQge JO UONeUR]dxd pue $d}0UJOO} IOJ [ BIGeI, 99S
VIWINOVW ‘d ‘SNOILVTNdOd AGNLS GNV SAILLITVOOT ATdNVS AO STIV.LAC ‘7 ATA VL

324

325

BRITISH AND IRISH SPOTTED-ORCHIDS

JUON

JUON

JUON

JUON

dUON

(JA) dW XOW “(4)d

QUON

dUON

(AA)OW X9W ‘(O)OW

(0)4 ‘GOW

QUON

(44)4 ‘GA)JOW

1/8

ere

VL

UL

VL

CHL

[/L-¥/9

T/L-v/9

Lie

cT/9

1/9

1/9

OOOT

09

US

00S

OS

Of

000¢

OS

OS

OOT

OOS

6°S

rT

1-9

gS

6°S

vv

0-S

29'S

6°S

V9

£9

o1qqe3/pues yoraq
posterjeod uly]

ouojspues
UPIUOPLIIO | /1eIg

2uojspues
ueIUOpLIIO [/}e9g

[TIS SH MOTOp/{TN 789d

ayueis/jeod uly]

pues [elAnyoroeys
/pues UMOTG/1B9q

STyos
/pues [eIAngoreys
jyead uy

IsIyos OTWM/TN/IeI¢q

sstous
uvipeseq/jeod uy],
JUO}SOUNT]
snoJojIuogiey)

jeod uy
dU0]SOUIT]
snoloyiuogiey

jvead uly,

JUO}SOUWT]
SnNoIOJIUOgIe)
/yeod uty

$99

0s"

OLT

SIT

09t

SI

Sc

Sc

SC

856 SEE/SI
O10'P9¢/81
100°69€/81
ZS19€S/81
POL PLI/8Z

889 96/87

C91 16/87
Ilb LL9/L7
cOS1$9/CO
COL crl/cl

810°L80/71

810°980/CI

pO *9°A ‘sopngg” ‘N “wnyy
‘ARG STUUWH “Heyory]

~ POT “9A

‘sopng” “N “winyYy “us[D
AUYOW II Joddn ‘dog
vOT “9°A

‘sopnggd "N “winyy “Us|
HOOTNIN Jeddn “sog

POL “9A

‘sopnqg ‘N ‘aAxS “TOOT
“AN “ysiew opisulesls
296 *O°A ‘SSoUIO}SeA
‘uO}sSHOP UID “GINVNATO
yoo] o10ys “q'N ‘WeoH
496

‘OA “UITEN “3S9104 NIGATNO

‘YooTy ukID *§ “Ysie

6

‘OA ‘UdIBPIOQGY ‘§ ‘UdIeH
jo 1eOg “$ “NYIVONATO
‘oinjsed ysnoy

88 “OA “Yd

“PI ‘WOWUSYy “A'S ‘UIng
SUAMV'I 2A0ge ‘Yysieyy
9TH

‘O°A ‘ABMTRD “MA ‘NAGHITO

“AA ‘oinqsed ysnoy

6H *9°A ‘21RID “OD “peo
yeild “HOOOUUNW
‘quowoaed sWWeW

6H ‘9A ‘oleID

‘Oc ‘BUIJBAUOOPSTT "M
‘NVAUATIVA Wsuloaeg
6H 9A

‘QIRID ‘OD ‘euIeAUOOpsSY]

"‘M ‘HOV TIVS1NOd
‘yuowdsaed owe

326 R. M. BATEMAN AND I. DENHOLM

TABLE 3. VARIATES CONTRIBUTING APPRECIABLY TO THE FIRST TWO PRINCIPAL
COORDINATES FOR D. FUCHSII POPULATION MEANS, LISTED IN ORDER OF

DECREASING IMPORTANCE
Principal coordinate PE] PC2
Percentage of variance
accounted for 27-1 23-1
Variate name, and Width of widest leaf — Labellum colour, ‘y’ ~
direction of increase in Stem diameter — Labellum colour, reflectivity (Y) =
value of variate in relation Inflorescence length — Presence of leaf markings :
to increase in value of Plant height — Stem anthocyanin a
vector (e.g. populations Length of basal bracts — Bract anthocyanin . se
with the largest mean Length of longest leaf — Labellum markings, contrast BL
values for leaf width and Position of longest leaf up stem — Labellum markings, distribution +
other vigour characters Number of non-sheathing leaves — Labellum markings, type =
tend to occur towards the Number of sheathing leaves = ,
left side of PC1 on Fig. 2). | Number of flowers =
Length of floral bracts =
TABLE 4. VARIATES CONTRIBUTING APPRECIABLY TO THE FIRST TWO PRINCIPAL
COORDINATES FOR D. MACULATA POPULATION MEANS, LISTED IN ORDER OF
DECREASING IMPORTANCE

Principal coordinate PCi og
Percentage of variance
accounted for 29-2 15-1
Variate name, and Stem diameter — Bract anthocyanin . >
direction of increase in Length of basal bracts — Stem anthocyanin ti
value of variate in relation Length of floral bracts — Presence of leaf markings Ar
to increase in value of Number of flowers — Labellum markings, type +
vector. Ovary length — Labellum markings, contrast +

Width of widest leaf — L.o.p.s', solid markings 5

Spur, length — Labellum markings, distribution +

Labellum, width =
Plant height =
Inflorescence length =
Length of longest leaf =
Labellum, length to lateral lobe =
Labellum, length to centrallobe —
Labellum, length to base of sinus —

' L.o.p.s = lateral outer perianth segments.

coordinates for D. fuchsii populations; PC1 again represented vegetative vigour (though supple-
mented with some floral dimension characters) and PC2 represented pigmentation. At the extremes
of PC1 were the stunted, late-flowering populations from Harris and Porlock, and the especially
vigorous populations from marshes at Thursley and Llandegfan (Fig. 2). Separation of populations
along PC2 according to degree of pigmentation was less clearly defined than for D. fuchsii. The five
anthocyanin-low populations from Carboniferous limestone habitats (PC2 values <—0-15) differed
considerably in vigour, and consequently formed three separate branches of the minimum spanning
tree: Ashes Gill (Yorkshire, hay-meadow), Poulsallagh, Murroogh, Ballyryan (Burren, limestone
pavements), and White Nook (Yorkshire, upland pasture). No populations of D. maculata were as
anthocyanin-rich as the extreme populations of D. fuchsii. The Burren populations, and those from
New Forest heathlands (Matley, Stephill, Crockford), formed two clusters, but the three
populations from the Isle of Rhum (Kilmory, Kinloch, Harris) did not associate closely.

BRITISH AND IRISH SPOTTED-ORCHIDS 327

ANALYSES OF INDIVIDUAL PLANTS :
Both principal coordinates plots for individual plants (Figs. 3 & 4) represented more characters but
much less of the total variance than the equivalent ordinations of population means (Tables 5 & 6).
Characters reflecting floral dimensions made more important contributions to plots for individuals
than to those for populations.

Pigmentation replaced vigour as the strongest coordinate (PC1) in the plot of D. fuchsii
individuals (Fig. 3), which revealed almost complete overlap of the three dark-flowered varieties
(vars hebridensis, cornubiensis and alpina) and considerable overlap between them and the
supposedly predominantly pale-flowered var. okellyi. Var. fuchsii again tended to have interme-
diate values for the pigmentation coordinate (PC1), and included numerous vigorous plants that
dominated the positive end of PC2. The apparent discontinuity between anthocyanin-deficient
. plants (PC1 values <—0-15) and the remainder was exaggerated by zero scores for three characters
representing a single feature (absence of labellum markings) and was much less evident when the
data were re-analyzed without these characters. The anthocyanin-deficient plants were mostly var.

Stephill
Matley Porlock @

Cluanie

@ibursley @ Culbin
2
af

~~
‘

SS Z Crockford

SS
Llandegfan®

Glencairn
Harris

Bricket
wood

!

|

|

!

|
Stanmore |
|

Ashes Gill 6

Poulsallagh

\
O White Nook

-0.4 =0.3 502 sO 0 0.1 0.2 0.3 0.4
PC1
Ficure 2. Principal coordinates plot of PC1:PC2 with superimposed minimum spanning tree. D. maculata,
population means.
© Burren
O Other Carboniferous limestone regions (® individuals lacking floral anthocyanins; Fig. 4)
O Rhum
© Porlock’
@ residuum
See Fig. 1 for strengths of minimum spanning tree links.
Inset: Main trends of variation represented on the plot (see Fig. 1).

328 R. M. BATEMAN AND I. DENHOLM

a)
oOo oO
© @ OAD O
‘oe, 0) o me 0 a)
on 0. OO gs O sa
OOO COO © OA ®
10} eS 0O Oo #0
0 0 @ee0s8e s
40 OA 680 @ O
© 00.6) «8©)~ «60: COO ®@ ®@
@QAOAA @ B@0 SA
A @ @n Cee
& @ e &
OO ese eee 8&8 @ @
fo) AB@eCe © ee @
44 GAS 80008 Bee C0
@A @h @ASC0@ e @e@ @ ®
& ® e @

-

03° -0.2 -0.1 0 0.1 0.2 0.3
PC2
FiGURE 3. Principal coordinates plot of PC1:PC2. D. fuchsii, all individuals.

Inset: Main trends of variation represented on the plot.
See Fig. 1 for explanation of symbols.

okellyi, but also included some individuals of var. fuchsii. Some plants of both varieties lacked all |
fioral pigments (Fig. 3, Table 8).

The ordination of D. maculata individuals was generally similar to that for D. fuchsii, though the —
vigour coordinate remained slightly stronger than the pigmentation coordinate and the subordinate
trend in flower size did not parallel the trend in vegetative vigour (Table 8, Fig. 4 (inset)). The |
morphological extremes on Fig. 4 were anthocyanin-less plants from Carboniferous limestone
habitats in the Burren and Yorkshire (bottom), small, late-flowering plants from Porlock and Harris |
(left) and large, marsh-dwelling plants from Thursley, Llandegfan, Stanmore and Culbin (top |
right). However, geographically- and/or ecologically-related populations (including those from
Rhum) were widely dispersed on the plot, which showed considerable morphological overlap of any |
jet infraspecitic taxa that could be delimited using these criteria.

BRITISH AND IRISH SPOTTED-ORCHIDS 329

consistently correlated characters but the trends themselves are uncorrelated, so that most
combinations of intensity of pigmentation and degree of vigour can be encountered within the
species as a whole. Extensive intra-population variation further obscures resolution of the
populations into clearly delimited intraSpecific taxa. D. fuchsii achieves greater vigour and more
intense pigmentation than D. maculata, thus presenting a broader morphological range that offers
greater scope for taxonomic subdivision.

INTERPRETATION OF VARIATION IN THE SPOTTED-ORCHIDS

The major trends of variation in vigour and pigmentation obscure some interesting subordinate
trends, and could represent several causal factors. In this and the following section we examine in
greater detail the nature and taxonomic implications of the patterns of variation shown by the
spotted-orchids, and speculate on possible underlying mechanisms.

@
D0O0@eS@C@Cee © 8 @
® @
@® oO

O
ee0ee 6@
© 8 cee0cVeeee @ 0e@
®@e@ 00 88 @0@ @

o0¢e000
@0e800880000
© 08@@00 ee O

O@@00ee00
ToL TTT jefol T fel yyy je

@ ee0008000
®@ @e0000e088 80

O
Oo ®
© @ ©COS8e@e@ 806
OO 8080
O

Oo
Ss
©
O
©
Oo
©

@
&

© Oo O00 O08
@

©

=0.3 =0.2 -0.1 0 0.1 0.2 O38
PC1
Ficure 4. Principal coordinates plot of PC1:PC2. D. maculata, all individuals.

Inset: Main trends of variation represented on the plot.
See Fig. 2 for explanation of symbols.

330 R. M. BATEMAN AND I. DENHOLM

TABLE 5. VARIATES CONTRIBUTING APPRECIABLY TO THE FIRST TWO PRINCIPAL
COORDINATES FOR ALL INDIVIDUALS OF D. FUCHSII, LISTED IN ORDER OF
DECREASING IMPORTANCE

Principal coordinate Pet PCL

Percentage of variance

accounted for 17-0 13-0

Variate name, and Bract anthocyanin + Width of widest leaf. “=

direction of increase in Stem anthocyanin + Stem diameter a

value of variate in relation Labellum colour, reflectivity (Y) — Plant height Ee

to increase in value of Labellum colour, ‘y’ — Length of basal bracts =i

vector. Labellum markings, contrast + Inflorescence length +3
Labellum markings, distribution + Length of longest leaf +
Presence of leaf markings + Number of flowers se
Labellum markings, type + Position of longest leaf up stem “a
tarps: solid markings + Labellum, length to lateral lobe +
Labellum colour, ‘x’ — Length of floral bracts 3z
Plant height — Number of sheathing leaves ae
Length of longest leaf — Spur, length ate
Labellum, lateral lobe indentations + Labellum, length tocentrallobe +
Inflorescence length — Spur, median width +
Number of flowers — Labellum, length to base of sinus +
Number of sheathing leaves — Labellum, width +

' L.o.p.s = lateral outer perianth segments.

TABLE 6. VARIATES CONTRIBUTING APPRECIABLY TO THE FIRST TWO PRINCIPAL
COORDINATES FOR ALL INDIVIDUALS OF D. MACULATA, LISTED IN ORDER OF
DECREASING IMPORTANCE

Principal coordinate PCI PC2

Percentage of variance

accounted for 14-1 10-7

Variate name, and Length of basal bracts + Presence of leaf markings =

direction of increase in Stem diameter + Labellum, length to base of sinus +

value of variate in relation Number of flowers + Labeilum markings, type

to increase in value of Width of widest leaf + Labellum, length to centrallobe +

vector. Bract anthocyanin — L.o.p.s', solid markings +
Length of floral bracts + Labellum markings, contrast +
Inflorescence length + Labellum, length to lateral lobe +
Labellum colour, reflectivity (Y) + Plant height sf
Plant height + Labellum, width +
Labellum colour, ‘y’ + Length of longest leaf +
Labellum, width + Bract anthocyanin +
Labellum, length to lateral lobe + Spur, length +
Stem anthocyanin _ Labellum markings, +
Ovary length + — distribution ne
Spur length + Stem diameter a
Labellum, length to centrallobe + Stem anthocyanin +

' L.o.p.s = lateral outer perianth segments.

“

VIGOUR .

Stephenson & Stephenson (1920), Summerhayes (1951) and Ettlinger (1976) noted that D. fuchsii
from clay woodlands tend to be unusually tall and ‘leafy’. Harpenden, our only population from a
clay wood, had tall (mean height 43 cm) but not especially broad stems (the inflorescence, which was
not especially long, was consequently a small proportion (mean value 0.13) of the total stem length)

BRITISH AND IRISH SPOTTED-ORCHIDS 331

TABLE 7. VARIATES CONTRIBUTING APPRECIABLY TO THE FIRST TWO PRINCIPAL
COORDINATES FOR ALL INDIVIDUALS FROM DARK-FLOWERED POPULATIONS OF D.
FUCHSI, LISTED IN ORDER OF DECREASING IMPORTANCE

Principal coordinate PCI PC2

Percentage of variance

accounted for 15-0 11-1

Variate name, and Labellum, width — Labellum, lateral lobe indentations +

direction of increase in Labellum, length to lateral lobe — Labellum markings, distribution +

value of variate in relation lLabellum, length to central lobe — Presence of basal leaf ~

to increase in value of Width of widest leaf — Length anthocyanin —

vector. Inflorescence length — Bract anthocyanin +
Labellum, length to base of sinus — _ Leaf colour +
Spur, median width — L.o.p.s', position —

Length of basal bracts =
Number of flowers =
Spur, width at entrance =
Stem diameter =
Number of sheathing leaves =
Spur, length =

' L.o.p.s = lateral outer perianth segments.

and many long but not particularly broad sheathing leavés that were widely spaced along the stem.
These characteristics of lengthening of structures without concomitant increases in width were
shared by the population growing in an overgrown meadow at Castlebar, and probably represent a
compensation response to low light intensities rather than exceptional vigour per se.

Genuinely vigorous D. fuchsii had stems that were unusually broad as well as tall, their leaves
were broad as well as long, and they had many (4—6) non-sheathing as well as sheathing leaves. They
also had long (>7 cm), many-flowered inflorescences and uppermost sheathing leaves that were
broadest well above the base. These increases in overall sizes rather than just lengths of structures
were exhibited by the Oaklands and Parham populations from flooded gravel-pits, and those
individuals from Pitstone that occupied the marshy floor of the chalk quarry rather than its dry
grassy margins. Thus, true vigour in D. fuchsii appears to be strongly correlated with high soil
moisture content, again suggesting a phenotypic response to environmental (in this case edaphic)
factors. Intriguingly, at least some of these marsh-dwelling D. fuchsii possessed hollow stems, a
feature that is supposedly confined to marsh-orchids.

Two populations of D. maculata (Thursley, Llandegfan), collected from amongst tall vegetation
in wet habitats, exhibited many of the vigour characteristics shown by the Oaklands and Parham
populations of D. fuchsii. They had tall, broad stems, long, many-flowered inflorescences, long,
broad (though not particularly numerous) leaves widely distributed along the stem and long basal
bracts. Three other populations (Culbin, Stanmore, Bricketwood) from Juncus-rich marshes
showed extreme values for various combinations of these characters, and all five populations
possessed rather long spurs.

Populations of both species exhibiting the smallest values for vigour characters grew among short.
vegetation in exposed coastal (Oldshore and Polin populations of D. fuchsii, Harris population of
D. maculata) or upland (White Nook and Porlock populations of D. maculata) habitats, and were
probably environmentally dwarfed.

Thus, size variation in vegetative characters may primarily reflect an interaction between the
ontogeny of spotted-orchids and environmental factors such as light intensity, soil moisture and
exposure (particularly to wind). The resulting effects on the phenotype, which are more readily
discerned in D. fuchsii than D. maculata, obscure any independent geographical trends such as the
latitudinal cline in leaf number suggested by Heslop-Harrison (1951).

PIGMENTATION 7
The eight populations of D. fuchsii with mean labellum reflectivities below 40% (Fig. 6) also tended
to have the lowest standard deviations for this character (all but Tintagel <10%), suggesting that

332 R. M. BATEMAN AND I. DENHOLM

-

TABLE 8. GEOGRAPHICAL AND ECOLOGICAL VARIATION IN FREQUENCIES OF
ANTHOCYANIN DEFICIENCIES IN D. FUCHSII AND D. MACULATA

D. FUCHSII D. MACULATA

eee S g 8

f= ay = nS aS re a E at AS)

gpa eee 8 oe eee
Bio Boke Bees eee eee
See) Boe her dnaaeins, Boe cee
Percentage of ere 2 € 4 g . baie =P Re = i=.
plants lacking: q Os UO Soa z Oo € 5 e Zz i
Stem anthocyanins 5g AO Sage ORR A gaarestogal= 99 ae 56
ce ae Bract anthocyanins dGve, ~ 68 FP SG ig ing ohoweiG7) <<) 70 ieee
y Leaf markings 13 43 20 9 15 53 de) 8
Floral L.o.p.s', markings 40 68 40 a2 10 10 35 7
Sethiecnaias Labellum markings Hi 48 0 5 5 0 35 2
y Labellum anthocyanins 2 48 0 3 25 3 35 27
All floral anthocyanins 8 35 0 2 3 0 30 0
All floral and vegetative 6 25 0 1 1 0 10 0

anthocyanins (i.e. albinos)
(Total number of plants) (200) (40) (10) (150) (232) (30) (20) (182)

'L.o.p.s = lateral outer perianth segments.

they constitute a distinct and consistent anthocyanin-high facies of D. fuchsii. This was supported by
the frequent occurrence of bract and stem anthocyanins in these populations, though leaf and
labellum markings were not especially large or bold. However, this group was very heterogenous in
other respects, notably labellum size, and included both the largest- and smallest-flowered
populations of D. fuchsii (Fig. 6). Marked bimodality in labellum dimensions (particularly width)
distinguished the large-flowered vars. hebridensis and cornubiensis from the small-flowered var.
alpina. As labellum dimensions were only minor contributors to the first two principal coordinates
for all D. fuchsii individuals (Table 5), this distinction was not apparent in that plot (Fig. 3). When
the analysis was repeated using only individuals of dark-flowered populations, PC1 largely
separated var. alpina from vars hebridensis and cornubiensis (Fig. 5) using floral dimensions (Table
7). PC2 gave an equal degree of separation of vars. hebridensis and cornubiensis, but using a more
diverse set of unrelated characters that are less easily interpreted (Table 7). Hence, this dark-
flowered facies comprised three groups of populations showing approximately equal levels of
morphological and ecological differentiation.

Two populations of D. maculata (Harris, Rhum; Porlock, Exmoor) resembled the dark-flowered
facies of D. fuchsii in having mean labellum reflectivities of less than 40%. They were also
distinguished by their late flowering (August; Table 2), narrow stems and small number of narrow
leaves. However, they were not especially rich in vegetative anthocyanins and do not appear to form
a cohesive morphological unit (see ‘Classification’).

Neither species of spotted-orchid included a distinct pale-flowered facies that could be
distinguished by differences in population means for labellum reflectivity, but both included
populations with unusually high frequencies of individuals lacking floral and/or vegetative
anthocyanins. Anthocyanin-deficient plants of D. fuchsii were most frequent in the Carboniferous
limestone habitats of the Burren but, contrary to popular belief, comprised a minority of plants in all
four study populations (Table 8); these Burren populations even included plants that matched
typical individuals of vars hebridensis , cornubiensis and alpina in intensity of floral pigments. Ashes
Gill, Yorkshire, the only other study population of D. fuchsii from Carboniferous limestone, was
attributed to the dark-flowered var. alpina. In contrast, D. maculata populations with the highest
frequencies of plants deficient in vegetative anthocyanins were associated with Carboniferous
limestone habitats in both the Burren and Yorkshire (Table 8). Floral pigments were also unusually
infrequent in D. maculata populations from Yorkshire but, surprisingly, not in those from the
Burren; most Burren plants of D. maculata possessed diffuse anthocyanins in their labella, albeit at

BRITISH AND IRISH SPOTTED-ORCHIDS 333

=Ur3 -0.2 =e 0 0.1 0.2 0.3
PGI

Figure 5. Principal coordinates plot of PC1:PC2. D. fuchsii, individuals of dark-flowered populations.
See Fig. 1 for explanation of symbols.

very low concentrations (over half had reflectivities >70% ) and often in very pale hues of red-purple
rather than the purple more characteristic of this species. All four anthocyanin-less plants of D.
maculata that we measured were from Yorkshire (Ashes Gill and White Nook).

Overall, the two species showed similar frequencies of vegetative anthocyanins, but plants lacking
all anthocyanins (i.e. true albinos) were more frequent in D. fuchsii (Table 8). Labella of D. fuchsit
more often lacked markings but less often lacked diffuse (i.e. background) anthocyanins than those
of D. maculata. Approximately 2% of individuals of both species possessed annular leaf markings, a
character much more common in marsh-orchids (notably D. majalis (Reichenbach) P. F. Hunt &
Summerhayes subsp. occidentalis (Pugsley) P. D. Sell) but evidently not confined to them
(Stephenson & Stephenson 1920; Godfery 1933; Helsop-Harrison 1948; Wiefelspiitz 1976).

TAXONOMIC AND EVOLUTIONARY IMPLICATIONS

The delimitation of spotted-orchid taxa and the ranks awarded to them should reflect available
knowledge of the nature and causes of their variation: whether the variation is continuous or
discontinuous, whether it exhibits geographically- and/or ecologically-related patterns, and to what
extent the phenotypic patterns reflect genetic and environmental influences.

Our multivariate analyses demonstrated that eachspotted-orchid species constitutes a morpholo-

334 R. M. BATEMAN AND I. DENHOLM

B Lelant (23%)

Fintagel (32%) O
gS Baleshare (30%)

Pitstone (65%)

@ Oaklands (52%)

7
Aa Torrin (52%) (—D Robach (25%)

Keltneyburn (31%) Castlebar (55%)
, 2 Black Head (69%)

© arcenden (57%)

@
Castle Hill (52%)

Ashes Gill (31%) ®
Oddy Hill (62%)

Polin (29%)

6 ©
© Oldshore (25%) A Khyber Pass (62%)

LABELLUM LENGTH TO CENTRAL LOBE (mm)

Corrofin (72%)
A

S Gelain (76%)

9 10 11 12
LABELLUM WIDTH (mm)

Ficure 6. Bivariate plot of mean labellum dimensions for D. fuchsii populations. Percentages denote mean
reflectivities of labella, the best measure of densities of floral anthocyanin pigments.
‘See Fig. 1 for explanation of symbols.

gical continuum in which all potential infraspecific taxa show considerable overlap. Moreover, much
of the variation in both species is polarized into two trends (sets of inter-related characters)
representing vigour and pigmentation. This contrasts with the more complex associations of often
unrelated characters observed in marsh-orchids, whose variation is also continuous but has strong
geographical and/or ecological components and can be resolved into subspecies showing only
limited morphological overlap (Bateman & Denholm 1983, 1985). Thus, no group of spotted-orchid
populations is sufficiently differentiated to merit subspecific status, and those considered worthy of
any taxonomic separation have been treated as varieties. We regard the subspecies and the variety
as essentially similar concepts; both describe geographically- or ecologically-related groups of
populations, but varieties show less morphological differentiation and are separated by fewer,
generally less predictive characters. The D. fuchsii segregates assigned to this rank are relatively
homogeneous and possess distinguishing characters (see ‘Classification’) that are unlikely to show
substantial environmental modification. Var. fuchsii remains a comparatively heterogeneous taxon,
but further subdivision on the basis of differences in vigour, which may be largely environmentally
determined, seems unwise. We have not recognized any infraspecific taxa of D. maculata, whose
overall range of variation is narrower and even less readily partitioned than that of D. fuchsii.

Our treatment of D. fuchsii taxa as varieties rather than subspecies contradicts most classifica-
tions published during the last 40 years. However, we believe that the taxonomic importance of the
two most widely accepted ‘subspecies’ of D. fuchsii (hebridensis and okellyi) has been inflated by
exaggeration of their morphological distinctness, failure to identify optimal diagnostic character
states, tautological arguments concerning their identification and speculative assumptions regarding
their evolution.

D. fuchsii subsp. hebridensis was well described from the Hebrides, initially qualitatively by
Wilmott (1939) and later quantitatively by Heslop-Harrison (1948, 1952, 1954), but it soon became
entrenched as a geographically and ecologically restricted taxon with dark flowers that occupied
machair along the western seaboard of the British Isles (Heslop-Harrison 1952, 1954, 1968). Var.

BRITISH AND IRISH SPOTTED-ORCHIDS 335

TABLE 9. TREATMENT OF SELECTED INTRASPECIFIC TAXA OF D. MACULATA BY
(a) BRITISH AND (b) CONTINENTAL WORKERS
— = taxon not mentioned (0 = taxon not yet described); V = considered to occur in Britain; X = not
considered British; abbreviations = taxon considered synonymous with subsp. maculata (ma), elodes (el),
ericetorum (er); ? = authors’ intentions unclear.

-_—_-,
(a) an nq
N fo. —~
nN — a >
— — Vay oO
rol 5 = a
b~ Dan « —
S" g Fehon rs Sri
va)
* Va) = fon ao} —_
: sepia amiee aes Bile she og g
« New = ym,
ae — Fes] Le, a g a = < =
. N N v i fo fo .
pil oN Dv > = =). = _ 7
= = = = as a <= 3 aes S =
fon Ee ° ~ = ae ~”a oO
= 2 > = E o D v
wm = 5 o o a. oS aa) So e- S
9) oO = = = 2) = r=] re v
oO = SS — a. = = oD —
=. S a. eS =I % x = = = =
a) fea) 7) Oo 7) ae O a ea) =) Q
Taxon
maculata V _- V x = x x = Bs = 9
ericetorum ma V ?ma el AVA a4 V V V V V
elodes ma er x V er 2er 2er - er er ?
praecox ma = ma el = er = - = _ -
rhoumensis 0 0 0 0 _ - V V ad _ V
(b)
~~ —_—
— =
a
= =
= _~ a4
— us Cow =) Ss
E a ZS Z & Sees
~_—
a S S Cc ~ 2 = =
Oo =) —_ 2) = SN rat —_~ =) 3%
i) S wm = = = S 5] :
3g = No) wa ae 2. ee) = v
<2) D 2) SS < ace $ Dy m oD
2 E a = 2 x aS) = 3 &
= = = = = oO = No) r= =
) o ros © o Se o) le) = 2
O > A 7) r=: s ee) XA D A
Taxon
maculata V 2 V ? er ? We V V V
ericetorum el Avi V el V \/ 74 ma ma V
elodes V a x V er x V V ma x
praecox el el er = = er Gr = = =
rhoumensis 0 0 V V ~ - - = =

cornubiensis and coastal populations of var. alpina, which were also dark-flowered and conformed
to these distributional criteria, were included in hebridensis without supporting morphological data
(e.g. Perring & Sell 1968) while inland populations of var. alpina were excluded. Consequently, |
hebridensis became a morphologically heterogeneous taxon whose individuals could not be
identified without knowledge of their geographical location (Heslop-Harrison 1968, p. 23). Its
morphological distinction was exaggerated by insufficient knowledge of the broad morphological
range encompassed by var. fuchsii and failure to recognize the small- and dark-flowered taxon that
we have named var. alpina.

The tecorded (predominantly western) distribution of subsp. okellyi resembles that of subsp.

336 R. M. BATEMAN AND I. DENHOLM

hebridensis (Perring & Sell 1968) and also covers a restricted range of habitats, typically limestone
pavements. It therefore received the same distributional ‘positive feedback’ as subsp. hebridensis,
and progressed from “‘an indistinct recessive colour race’? (Stephenson & Stephenson 1921) to a
subspecies exhibiting a typical Atlantic distribution (Heslop-Harrison 1954; Perring & Sell 1968).
However, okellyi differs from hebridensis in having been poorly described in its original diagnosis
(Druce 1909), which greatly exaggerated its anthocyanin deficiency. Our data show that flowers
from populations of var. okellyi in the Burren (its type locality) are not especially pale (their mean
labellum reflectivities resemble those of chalk populations of var. fuchsii at Oddy Hill and Pitstone),
nor are they unusually deficient in bract and stem anthocyanins. Anthocyanin-deficient plants are a
subordinate component of most populations of D. fuchsii in the Burren, including all four that we
studied, and the Burren populations are even less readily distinguished by other supposedly
diagnostic characters such as smaller labella and spurs, and narrower stems and leaves. Further-
more, anthocyanin-less D. fuchsii from the Burren cannot be distinguished from similar plants
growing in other regions and habitats (Fig. 3). Hence, var. okellyi is much less differentiated from
var. fuchsii then has generally been assumed, and differs mainly in the higher frequency of
individuals deficient in floral and/or vegetative anthocyanins. It coexists in the Burren with an
analogous anthocyanin-low variant of D. maculata (see ‘Classification’).

A third British spotted-orchid subspecies, D. maculata subsp. rhoumensis, has not received such
general acceptance. This enigmatic Rhum endemic reputedly has pale purple flowers, bold labellum
markings, blooms in August and is diploid (Heslop-Harrison 1948). Our three study populations
from Rhum differed appreciably in morphology and phenology both from each other and from the
original diagnosis of rhoumensis (see ‘Classification’), but not from populations of D. maculata
elsewhere in the British Isles. We cannot therefore tdegaly any cohesive infraspecific taxon peculiar
to, or present on, Rhum.

The infraspecific taxonomy of the spotted-orchids has also been strongly influenced in the past by
speculative interpretations of their population biology. Heslop-Harrison (1952, 1954, 1968)
justified subspecific status for hebridensis and okellyi by arguing that in their “extreme” forms both
approach ecologically and morphologically well-defined races, probably of monophyletic origin,
that are largely reproductively isolated from the type due to their disyunct geographical ranges.
However, he also noted that vars. hebridensis and okellyi intergrade extensively with var. fuchsii,
both by extensive geographical clines and by ‘hybridization’ at sites where two taxa occur (Heslop-
Harrison 1948, 1949, 1951, 1952, 1954; Allen 1971; Wiefelspiitz 1976). Our observations suggest
that populations of D. fuchsii are not as differentiated as the above argument implies, and we do not
believe that taxonomic decisions should be based solely on geographical and morphological
extremes. It seems more appropriate to regard D. fuchsii as a variable species whose populations
differ locally in ‘mean morphology’ in response to prevailing edaphic and climatic factors, both
through adaptive shifts in gene frequencies and environmental modification of phenotypes. We
believe that morphological continuity in both species probably reflects potentially unrestricted gene
flow within and between the poorly-differentiated groups of populations; if so, terms such as
‘intergradation’ and ‘hybridization’ are inapplicable in this context. However, neither hypothesis
can be substantiated without non-morphological data and new insights into the control and adaptive
significance of particular morphological traits.

POSSIBLE CONSPECIFICITY OF THE SPOTTED-ORCHIDS

Most British workers have regarded D. fuchsii and D. maculata as separate species, distinguished by
consistent differences in morphology (principally labellum shape) and chromosome number. The
supposed cytological discontinuity dates from a few chromosome counts by Hagerup (1938) of
2n=40 (diploid) for D. fuchsii and 2n=80 (tetraploid) for D. maculata subsp. maculata and subsp.
elodes. Much emphasis has been placed on this apparent discontinuity by subsequent authors,
including ourselves. However, in the same year, Heusser (1938) demonstrated the presence in
Switzerland of diploid and tetraploid populations that were both assignable to D. fuchsii on
morphological criteria, and the potential taxonomic significance of this observation was quickly
realized by Pugsley (1939). Hagerup (1944) then described D. fuchsii embryos with 2n=20 and
2n=60, and a few D. maculata embryos with 2n=40. Heslop-Harrison (1948) reported 2n=40 for D.

BRITISH AND IRISH SPOTTED-ORCHIDS 337

maculata subsp. rhoumensis from Rhum. Populations of D. fuchsii in Austria may be 2n=40, 60 or
80 without showing significant morphological differences (Groll 1965, 1968; Voth 1978; Voth &
Greilhuber 1980), and Averyanov (1977) reported D. maculata near Leningrad with 2n=60. Thus,
both spotted-orchid species appear to be capable of maintaining cytologically stable populations
that are diploid, triploid or tetraploid, and previous assumptions of chromosome number based on
extrapolation from morphology may have been rash.

When Druce (1915) first separated D. fuchsii from D. maculata he speculated that the
morphological differences between the two species may simply reflect differences in the soils that
they inhabit. Stephenson & Stephenson (1921) and Godfery (1921) questioned this hypothesis,
noting that the two species remained morphologically distinct in mixed colonies. Heslop-Harrison
(1948) compared soil pH from several populations of D. fuchsii and D. maculata in the Hebrides,
and concluded that the former occupied soils of pH less than 5-5 and the latter soils of pH greater
than 5-0. He subsequently repeated these figures as generally characteristic of the habitats of the two
species (Heslop-Harrison 1951, 1954). However, 11 of our 23 study populations of D. maculata
occupied soils of pH greater than 5-5 (Table 2), and four populations grew in neutral soils (pH >6).
D. fuchsii is almost confined to neutral or alkaline soils, though our study population of D. fuchsii
that most closely resembled D. maculata (Harpenden) occupied a slightly acidic soil (pH 5-8; Table
1). Its labella had relatively shallow sinuses and central lobes that equalled or only slightly exceeded
the laterals, resulting in a small labellum shape index (1-23) similar to that of the Harris population
of D. maculata (1-24). Furthermore, the Harpenden D. fuchsii had unusually broken labellum
markings and a small value for mean leaf shape index (0-11) more typical of D. maculata. Four
populations of D. maculata had appreciably larger mean values for the labellum shape index than
the remainder; two were from isolated southern populations in only mildly acidic soils (Bricketwood
and Stanmore) and two were environmentally-stunted (Harris and White Nook). Plants from
Bricketwood and Harris also had central labellum lobes that considerably exceeded the laterals,
approaching D. fuchsii in this respect (e.g. Stephenson & Stephenson 1921). Thus, spotted-orchid
populations showing intermediate characteristics tend to occupy intermediate habitats, and an
edaphic influence on their morphology is possible.

Biometric studies on spotted-orchid populations from mainland Europe (Heslop-Harrison 1951;
Groll 1965, 1968; Voth 1978; Voth & Greilhuber 1980; Reinhard 1985) indicate that D. fuchsii and
D. maculata are less distinct on the Continent than in the British Isles (see also Summerhayes 1951,
p. 273). Thus, Continental workers (e.g. Vermeulen 1949; Landwehr 1977; Sod 1980) have
attempted to partition a more restricted range of variation, and have consequently created a large
number of poorly-defined and often overlapping infraspecific spotted-orchid taxa. This observation
also explains why D. fuchsii was first described as a species by a British worker (Druce) as late as
1915; during the previous 80 years, Continental workers, including several who were content to
treat apparently much more trivial entities as species, had recognized D. fuchsii under several
different names at varietal and formal rank (see ‘Classification’). If D. fuchsii and D. maculata prove
to be both morphologically and cytologically indistinct on the Continent, there is a strong argument
for following Nelson (1976) and Sundermann (1980), and relegating D. fuchsii to D. maculata (L.)
So6 subsp. fuchsii (Druce) Hylander.

CLASSIFICATION

The classification and diagnostic descriptions that follow are based on our principal coordinates
(Figs. 1-5, Tables 3—7), minimum spanning trees, analyses of population means, and ranges for
individual characters. Published data on British plants (Heslop-Harrison 1948, 1951, 1954) and on
Continental plants (Heslop-Harrison 1951; Groll 1965, 1968; Voth 1978; Voth & Greilhuber 1980;
Reinhard 1985) have also been considered.

The following terms are used to describe the frequencies of character states in taxa: rarely, less
than 20% of individuals; occasionally, 20-50%; often, 51-80%; usually, greater than 80%.
Frequencies of the best diagnostic characters (italicized) show most discontinuity between varieties.
Some terms used in the descriptions are qualitative but are derived from quantitative measure-
ments:

Density of inflorescence: dense, index ‘f value greater than 7. Depth of labellum sinuses: very deep,

338 R. M. BATEMAN AND I. DENHOLM

index ‘b’ value greater than 1.5 (the term ‘labellum shape index’ refers exclusively to index ‘b’).
Roundness of labellum: broad, index ‘a’ value less than 0.4. Position of lateral outer T penianth
segments: + horizontal, categories 1 and 2 on our scale.

Soil pH: acid, <6; neutral, 6—7; alkaline, >7.

Genus Dactylorhiza Necker ex Nevski, Acta hort. bot. Acad. sci. URSS 4: 332 (1937).
Sect. Maculatae (Parlatore) Vermeulen, Stud. Dactyl. 65 (1947).
Subsect. Eumaculatae Vermeulen, Stud. Dactyl. 68 (1947).

1. Dactylorhiza fuchsii (Druce) So6, Nom. nov. gen. Dactylorhiza 8 (1962).

Orchis fuchsii Druce in Rep. botl Soc. Exch. Club Br. Isl. 4: 105 (1915).

O. maculata L. var. trilobata Brébisson, Fl. Normandie 313 (1835).

O. maculata L. var. meyeri Reichenbach f., Icon. Fl. Germ. 67 (1851).

O. maculata L. subsp. meyeri (Reichenbach f.) Camus, Mon. Orch. Eur. 193 (1908).

O. maculata L. var. obscura Neuman in Bot. Notiser 1909: 153 (1909).

O. fuchsii Druce subvar. albiflora Druce in Rep. botl Soc. Exch. Club Br. Isl. 5: 167 (1918).

Dactylorchis fuchsii (Druce) Vermeulen, Stud. Dactyl. 147 (1947).

D. fuchsii (Druce) Vermeulen var. meyeri (Reichenbach f.) Vermeulen, Stud. Dactyl. 146 (1947).

Dactylorhiza fuchsii (Druce) S06 var. meyeri (Reichenbach f.) Soé, Nom. nov. gen. Dactylorhiza 8
(1962).

Stem 7—50(—70) cm, 1-5—7-5(—11) mm in diameter, often lacking anthocyanins. Basal 1f or sheath
0-1, broadest at or above middle; sheathing lvs (2—)3—6(—7), usually strongly crowded towards
base of stem, usually recurved, narrowly to broadly lanceolate, usually broadest well above base,
longest 1f usually above widest, 4-5—16-5 cm long, widest 1f 0-5—4(—5-5) cm wide, width/length ratio
of lvs decreasing up stem, usually bright to dark green, tips rarely distinctly hooded; non-sheathing
lvs (1—)2-—6(-—9), narrow, broadest at base; lvs usually sparsely to densely marked on upper surface
only, markings usually solid (very rarely annular), usually + evenly distributed and transversely
elongated, usually 1-5—6 mm in mean diameter. Inflorescence 1-5—10(—22) cm, 8-30% of stem
length, fis 8—60(—95), fairly lax to dense (3-5—10(—13) fis/em). Basal bracts 7-5—23(—38) mm, 1-5-
2-5 times the length of the ovaries, floral bracts 5-12 mm, approximately equalling the ovaries,
often suffused with anthocyanins; peripheral bract cells (40—)55—135(—170) um long, barrel-shaped
to triangular. Labellum width exceeding length (usually considerably), 4-5—8-5(-9-5) x 7-12(-
13-5) mm, widest + at middle or above (obtriangular), base colour varying densities (reflectivity 10—
89%) of purple or less frequently white; markings pale to bold dashes and/or loops, occasionally
dots and dashes, rarely absent or a solid blotch, often covering about two-thirds of the labellum,
occasionally more or less; sinuses present (labellum three-lobed), usually deep, central lobe often
exceeding lateral lobes by >1 mm, labellum shape index usually >1.25; lateral lobes usually
indented, slightly deflexed to moderately reflexed; lateral outer perianth segments usually nearer
horizontal than vertical, often with solid (rarely annular) markings; median outer perianth segment
and inner perianth segments connivent; spur straight to moderately decurved, 3-5—8 x 1-2-5 mm at
entrance, 0-8—2-3 mm halfway along (2—3.7 mm at entrance, 1-6—3-2 mm halfway along in var.
cornubiensis), usually slightly tapering, half as long to as long as the ovary. 2n=?40. Flowering mid-
June to late July. Frequent and locally common throughout most of the British Isles, local in
northeast Scotland and the southwest. Neutral to alkaline (rarely slightly acid) soils.

Most early classifications recognized only a single species of spotted-orchid, Orchis maculata
(Linnaeus 1753). However, towards the end of the 19th Century, some workers began to realize that
two distinct taxa occurred in the British Isles. One had oval, planar lower leaves, deeply and
subequally three-lobed labella with the central lobe projecting beyond the laterals, and fairly
narrow, tapering spurs (D. fuchsii). It inhabited calcareous and neutral soils. The second had
lanceolate, keeled lower leaves, shallowly three-lobed labella with the lateral lobes equalling the
central lobe in length and considerably exceeding it in area, and very narrow, parallel-sided spurs
(D. maculata). It inhabited acid soils. Early attempts to recognize this taxonomic distinction were
made at infraspecific levels within D. maculata. D. fuchsii was regarded as O. maculata sensu stricto
and therefore treated as the type, and what we now call D. maculata was the segregate. Webster’s

“

BRITISH AND IRISH SPOTTED-ORCHIDS 339

(1886) O. maculata var. praecox was a restricted and unusual taxon, but Linton (1900) provided a
much more generally applicable description of British D. maculata as O. maculata subsp. ericetorum
Linton, a name that is still in general use in Britain (Table 9; see also discussion of D. maculata).

Few short papers can have initiated more taxonomic and nomenclatural controversy than that of
Druce (1915), which segregated the calcicolous spotted-orchids as a new species, Orchis fuchsii, and
treated the calcifuge as Linnaeus’ O. maculata. He was repeatedly obliged to justify this decision
(Druce 1915, 1918, 1924a, b, 1925) in barbed exchanges with Godfery (1921, 1923, 1924, 1933), who
believed that O. maculata was the calcicolous species. The debate was eventually settled in Druce’s
favour by Vermeulen (1947, 1968) and Heslop-Harrison (1951), who demonstrated that Linnaeus’
type specimen of O. maculata was the calcifuge taxon.

The evolution of spotted-orchid nomenclature on the Continent resembles that in Britain. Early
workers included a broad morphological range in O. maculata, but this was progressively
subdivided at subspecific or, more often, varietal level. D. fuchsii was repeatedly segregated from
D. maculata, most notably as O. maculata var. trilobata Brébisson (1836), var. meyeri Reichenbach
f. (1851) and var. obscura Neuman (1909) (Heslop-Harrison 1951). Most Continental workers now
include these taxa in D. fuchsii, but few are willing to treat D. fuchsii as a full species.

One notable exception is Holub (1983), who argued that D. fuchsii is a junior synonym of D.
longebracteata (F. W. Schmid) Holub, based on Orchis longebracteata F. W. Schmid (1791).
However, Jagietto (1986) convincingly demonstrated that the original description of longebracteata,
although ambiguous, is clearly discordant with D. fuchsit.

Extremes of pigmentation
About 20% of the D. fuchsii individuals at Oddy Hill occurred as an anthocyanin-rich form with
labellum markings more-or-less fused into a solid blotch and the upper surface of the sheathing
leaves with at least 50% areal coverage of large (mean diameter >6 mm) markings or entirely
covered with diffuse anthocyanins. Such plants also frequently possess spotted bracts and dashes on
their stems and ovaries, resembling some plants of var. hebridensis. This variant has also been
recorded in other chalk grassland populations at Beacon Hill, N. Hants. (D. M. T. Ettlinger pers.
comm. 1985; A. R. G. Mundell pers. comm. 1987), Mickleham, Surrey and Barton-le-Clay, Beds.
(R. M. Bateman), and “near Winchester” (Godfery 1933, p. 206 and Plate 50/22, 25), from
limestone grassland near Buxton, Derbyshire (D. M. T. Ettlinger pers. comm. 1987) and from
mildly acid coastal grassland at Armadale, Skye (R. M. Bateman). Nelson (1976, p. 100 and Plate
70c, d) illustrated similar plants from Scandinavia. Analogous variants occur in D. maculata (q.v.),
D. majalis subsp. purpurella (T. & T. A. Stephenson) D. Moresby Moore & So6 (near Hartlepool)
and D. ?incarnata (L.) So6 (near Southampton) (D. M. T. Ettlinger pers. comm. 1986).
Anthocyanin-less individuals have white or very pale cream-coloured flowers and have been
called subvar. albiflora Druce (1918). They form a small proportion of many populations of var.
fuchsii (e.g. Pitstone) and a much larger proportion of populations of var. okellyi.

a. var. fuchsii (Druce) Bateman & Denholm, comb. nov.

Orchis fuchsii Druce in Rep. botl Soc. Exch. Club Br. Isl. 4: 105 (1915).

Dactylorchis fuchsii (Druce) Vermeulen subsp. typica Vermeulen, Stud. Dactyl. 148 (1947).
Dactylorhiza fuchsii (Druce) S06 subsp. fuchsii (Druce) S06, Nom. nov. gen. Dactylorhiza 8 (1962).

Stem usually >20 cm, often >3-5 mm in diameter, occasionally suffused with anthocyanins (rarely
densely). Sheathing 1vs usually 4 or more, longest 1f often >9 cm long, widest 1f often >2 cm wide;
non-sheathing lvs often 4 or more; markings usually present, often covering >10% of the upper
surface. Inflorescence often >4 cm, occasionally >20% of stem length, occasionally dense, fis often
more than 25. Floral bracts often >8 mm, occasionally suffused with anthocyanins (rarely densely);
peripheral bract cells occasionally angular. Labellum often >6-7 mm long to apex of central lobe,
often >5-5 mm to apex of lateral lobe, occasionally >4 mm to base of sinus, often >9-5 mm wide,
occasionally broad, sinuses occasionally very deep; reflectivity of base colour often >40% (rarely
>80%); markings usually present; lateral outer perianth segments often + horizontal, markings
often present (rarely annular); spur occasionally >2 mm wide at entrance, >1-8 mm halfway along.
Flowering June/early July. Throughout the range of the species.

340 R. M. BATEMAN AND I. DENHOLM

Var. fuchsii is a very heterogeneous taxon, encompassing considerable variation in vigour and
pigmentation (see ‘Major trends in morphological variation’).

b. var. hebridensis (Wilmott) Bateman & Denholm, comb. nov.

Orchis hebridensis Wilmott in J. Bot., Lond. 77: 192 (1939).

O. fuchsii Druce var. hebridensis (Wilmott) Heslop-Harrison p. in Vasculum 25: 111 (1940).

O. fuchsii Druce subsp. hebridensis (Wilmott) Clapham, Fl. Br. Isl. 1317 (1952).

Dactylorchis maculata (L.) Vermeulen subsp. hebridensis (Wilmott) Vermeulen, Stud. Dactyl. 141
(1947).

D. fuchsii (Druce) Vermeulen subsp. hebridensis (Wilmott) Heslop-Harrison f. in Ber. geobot.
Forsch. Inst. Riibel 1953: 55 (1954).

Dactylorhiza fushsii (Druce) So6 subsp. hebridensis (Wilmott) Nelson, Mon. Ikon. Orchid. Gatt.
Dactylorhiza 102 (1976).

D. maculata (L.) So6 subsp. hebridensis (Wilmott) Nelson, Mon. Ikon. Orchid. Gatt. Dactylorhiza
102 (1976).

D. hebridensis (Wilmott) Averyanov in Bot. Zhurn. 71: 92 (1986).

Stem rarely >20 cm, occasionally >3-5 mm in diameter, usually suffused with anthocyanins (often
densely). Sheathing 1vs often 4 or more, longest 1f rarely >9 cm long, widest 1f occasionally >2 cm
wide; non-sheathing 1vs rarely 4 or more; markings usually present, usually covering >10% of the
upper surface. Inflorescence rarely >4 cm, rarely >20% of stem length, usually dense, f1s
occasionally more than 25. Floral bracts rarely >8 mm, usually suffused with anthocyanins (often
densely); peripheral bract cells usually angular. Labellum usually >6:7 mm long to apex of central
lobe, usually >5-5 mm to apex of lateral lobe, occasionally >4 mm to base of sinus, usually >9-5 mm
wide, usually broad, sinuses occasionally very deep; reflectivity of base colour rarely >40%;
markings usually present; /ateral outer perianth segments usually + horizontal, markings often
present (occasionally annular); spur occasionally >2 mm wide at entrance, >1-:8 mm halfway along.
Flowering late June/July. Hebrides and western Ireland.

Many authors (Summerhayes 1951; Clapham 1952; Akeroyd & Curtis 1980; Lang 1980; Davies et al.
1983; Webb & Scannell 1983) have argued that var. hebridensis is merely an environmentally
dwarfed ecotype of D. fuchsii with unusually dark flowers. However, Wilmott’s (1939) original
description of Orchis hebridensis stipulates unusually broad labella, and its large labellum and spur
were demonstrated biometrically by Heslop-Harrison (1954); see also Clapham (1962) and
Ettlinger (1976). Our data indicate strong bimodality in flower size that distinguishes dark-flowered
populations along the coasts of Cornwall, western Ireland and the Outer Hebrides from those
further east with similarly coloured but much smaller flowers (usually <6-7 X 9-5 mm; Fig. 6). The
larger-flowered populations also tend to have less prominent central labellum lobes (and therefore
smaller values (<1-5) for the labellum shape index), broader spurs (often >2-0 mm at the mouth),
and greater proportions of plants with annular lateral outer perianth segment markings (though
there are exceptions to all these criteria). We have therefore segregated the smaller-flowered plants
as var. alpina (q.v.), which includes some populations along the west coast of the Inner Hebrides
and Scottish mainland that have traditionally been regarded as var. hebridensis (e.g. Oldshore).

Populations of D. fuchsii with dark flowers and large labella occur in three areas (Heslop-
Harrison 1954; Perring & Sell 1968): the Outer Hebrides (Wilmott 1939; Heslop-Harrison 1948,
1951), western Ireland (Brenan & Simpson 1949; Heslop-Harrison 1949, 1950, 1951, 1952; Akeroyd
& Curtis 1980; Webb & Scannell 1983) and Cornwall (Pugsley 1940; Heslop-Harrison 1951, 1954;
Margetts & David 1981). Most of the populations occupy coastal machair (or machair-like) habitats.
The Cornish populations have been separated as var. cornubiensis Pugsley (q.v.).

Other characters that reputedly distinguish var. hebridensis from var. fuchsii are as follows:
shorter (c. 10 cm) stems (Lang 1980; Davies et al. 1983) bearing fewer, narrower leaves
(Summerhayes 1951; Lang 1980; Davies et al. 1983) that are less heavily spotted (Lang 1980; Davies
et al. 1983) or more heavily spotted (Summerhayes 1951), a denser inflorescence (Akeroyd & Curtis
1980; Davies et al. 1983), labella with more broken markings (Wiefelspiitz 1976; Lang 1980) and a
more prominent central lobe (Ettlinger 1976), and spurs more than 7 mm long (Heslop-Harrison
1954, p. 71) but narrow (Wiefelspiitz 1976), c. 1 mm wide (Clapham 1952). However, the Baleshare

BRITISH AND IRISH SPOTTED-ORCHIDS 34]

and Robach populations resembled many populations of var. fuchsii in all these respects.
Vermeulen’s (1947) suggestions that var. hebridensis is later flowering than var. fuchsii and
tetraploid are incorrect (Heslop-Harrison 1952). A population of var. hebridensis in Co. Galway
had taller stems and longer spurs than any of our study populations (Heslop-Harrison 1954), and
white-flowered plants may occur rarely in the Outer Hebrides (Wiefelspiitz 1976). Heslop-Harrison
(1948, 1954) thoroughly described var. hebridensis, but exaggerated its distinctness from var. fuchsii
as his data on the latter encompassed an unrealistically narrow range of morphological variation.

c. var. cornubiensis (Pugsley) Sod, Nom. nov. gen. Dactylorhiza 8 (1962).
Orchis maculata var. cornubiensis Pugsley in J. Bot., Lond. 78: 180 (1940).
O. fuchsii Druce var. cornubiensis (Pugsley) Clapham, Fl. Br. Isl. 1317 (1952).

Stem rarely >20 cm, rarely >3-5 mm in diameter, often suffused with anthocyanins (occasionally
_ densely). Sheathing 1vs often 4 or more, longest 1f occasionally >9 cm long, widest 1f rarely >2 cm
wide; non-sheathing lvs occasionally 4 or more; markings usually present, usually covering >10%
of the upper surface. Inflorescence occasionally >4 cm, often >20% of stem length, rarely dense, fis
rarely more than 25. Floral bracts often >8 mm, usually suffused with anthocyanins (occasionally
densely); peripheral bract cells rarely angular. Labellum usually >6-7 mm long to apex of central
lobe, often >5-5 mm to apex of lateral lobe, usually >4 mm to base of sinus, usually >9-5 mm wide ,
rarely broad, sinuses rarely very deep; reflectivity of base colour rarely >40%; markings usually
present; /ateral outer perianth segments rarely + horizontal, markings often present (occasionally
annular); spur usually >2 mm wide at entrance, >1-8 mm halfway along. Flowering June. Cornwall.

The Cornish populations were first described by Pugsley (1940) as Orchis maculata var. cornubien-
sis, and incorporated into Dactylorchis fuchsii subsp. hebridensis by Heslop-Harrison (1954). The
long obscure type locality, ‘a cliff-top near St Ives’’, may have been refound recently (M. Jenkinson
pers. comm. 1987). Otherwise, the only known populations of var. cornubiensis are those we
studied on a cliff-top at Tintagel and stabilized dunes at Lelant (Perring & Sell 1968; Margetts &
David 1981). However, Pugsley’s (1940) description of the St Ives plants is consistent with these
other Cornish populations, which tend to have wider spurs (median width >2 mm), more lax
inflorescences (<6 flis/em) and longer leaves (longest 1f >8 cm) than those in the Hebrides.
Wiefelspiitz (1976) over-simplified when describing var. cornubiensis as intermediate to var.
hebridensis and var. fuchsii.

Many populations of var. hebridensis and var. cornubiensis have acquired unusual characteristics,
perhaps as a consequence of geographical isolation. The large mean labellum width of the Baleshare
population (12-3 mm) exceeds those of all populations of D. majalis described by Bateman &
Denholm (1983). Similarly, the spur widths of the Lelant population (mean value for median width
2:5 mm) match those of the smaller-flowered populations of D. majalis (Bateman & Denholm
1983), D. incarnata (Bateman & Denholm 1985) and D. saccifera (e.g. Nelson 1976), and the stems
of the Tintagel plants are as broad (mean diameter 4-6 mm) as those of marsh-orchid populations of
similar stature, suggesting that they may be hollow. As in the type population of var. hebridensis on
Barra (Wilmott 1939), a small proportion of the Robach plants have spots on the undersides of the
leaves and bracts, and dashes on the stems and ovaries, features otherwise confined in British
orchids to the most anthocyanin-rich individuals of D. incarnata subsp. cruenta (O. F. Miller) P. D.
Sell and Neotinea maculata (Desfontaines) Stearn. In contrast, many of the plants in a population of
var. hebridensis from Co. Donegal lacked leaf markings (Heslop-Harrison 1949).

d. var. alpina (Landwehr) Bateman & Denholm, comb. et stat. nov.
Dactylorhiza fuchsii (Druce) So6 f. alpina Landwehr in Orchideeén 37: 78 (1975).

Stem occasionally >20 cm, rarely >3-5 mm in diameter, usually suffused with anthocyanins
(occasionally densely). Sheathing 1vs occasionally 4 or more, longest 1f rarely >9 cm long, widest If
rarely >2 cm wide; non-sheathing 1lvs occasionally 4 or more; markings usually present, often
covering >10% of the upper surface. Inflorescence rarely >4 cm, occasionally >20% of stem
length, rarely dense, fis rarely more than 25. Floral bracts rarely >8 mm, usually suffused with
anthocyanins (often densely); peripheral bract cells rarely angular. Labellum occasionally >6-7 mm

342 R.M. BATEMAN ANDI. DENHOLM .~

long to apex of central lobe, rarely >5-5 mm to apex of lateral lobe, rarely >4 mm to base of sinus,
rarely >9-5 mm wide, occasionally broad, sinuses often very deep; reflectivity of base colour rarely
>40%; markings usually present; lateral outer perianth segments occasionally + horizontal,
markings often present (rarely annular); spur rarely >2 mm wide at entrance, >1-8 mm halfway
along. Flowering late June/July. Scotland and northern England, possibly also Wales and Ireland.

Landwehr (1977) described and illustrated slender plants with small, dark flowers from the Austrian
Alps as D. fuchsii f. alpina. His original diagnosis fits well the populations of small, dark-flowered
plants that we have measured, though the maximum stem height of 25 cm would have to be
increased to allow the inclusion of the taller plants from Keltneyburn and Ashes Gill. Figure 40.8 of
Landwehr (1977) shows unusually large leaf markings on a plant of f. alpina taken into cultivation,
but they are not mentioned in his diagnosis. We have therefore adopted this name for fairly
homogeneous British populations of D. fuchsii with smali, dark flowers. As these populations are as
distinct from var. fuchsii as are vars. hebridensis and cornubiensis, we have raised alpina to varietal
status. The populations from inland meadows at Keltneyburn and Ashes Gill attain a stature (mean
height 23 and 30 cm respectively typical of var. fuchsii in such habitats, but the Oldshore and Polin
populations (mean height c. 10 cm) are from exposed coastal habitats and consequently environ-
mentally-dwarted. This taxon was recognized (though not named) by Clapham (1952, p. 1317):

“northern forms [of D. fuchsii] are commonly smaller with narrower basal lvs, stem purplish above,
and smaller fls, often pale reddish-violet in colour’.

When first describing D. fuchsii, Druce (1915) cited the French taxon O. maculata var. trilobata
Brébisson (1836, p. 313) as its earliest possible synonym, but regarded the brief original description
as ambiguous: “epi gréle; fl. petites; labelle 4 trois lobes profonds presque égaux” (inflorescence
lax; flowers small; labellum with three deep, subequal lobes). He later described var. trilobata as
‘‘very close to [the white-flowered] o’kellyi, but it has spotted leaves and tinted fiowers”’, a statement
followed by several British records for this taxon (Druce 1918, p. 167). Since flower colour is not
mentioned in the original diagnosis of trilobata, Druce’s logic eludes us, but it is a recurring theme in
subsequent treatments of trilobata. Godfery (1933, p. 206) applied the name to typical populations
of D. fuchsii occupying limestone grassland, which he regarded as having paler labella with shorter
central lobes than woodland populations. A similar morphological distinction between these
“habitat races’’ was described by Summerhayes (1951), but it is not supported by our data.

A second school regards trilobata as a small, dark-flowered form that is probably our D. fuchsii
var. alpina; indeed, records of trilobata from “‘Kenmore, Mid-Perth” (Druce 1916) and “near
Aberfeldy, Perthshire” (Wiefelspiitz 1976) may both refer to our study population of var. alpina at
Keltneyburn. Similarly dark-flowered plants, with exceptionally reduced lateral labellum lobes,
were illustrated as D. fuchsii f. trilobata (Brébisson) Landwehr (a combination that does not appear
to have been validly published) by Landwehr (1977, Plate 43.5, 6). As flower colour was omitted
from the original diagnosis of var. trilobata, and the stipulation for small flowers was probably
intended to distinguish var. trilobata from D. maculata rather than other infraspecific taxa of D.
fuchsii, we reject use of this epithet for both pale-flowered calcicolous populations and dark-
flowered populations occupying neutral or slightly acid soils.

e. var. okellyi (Druce) Bateman & Denholm, comb. nov.

Orchis maculata L. var. okellyi Druce in Irish Nat. 9: 211 (1909).

O. okellyi (Druce) Druce in Rep. botl Soc. Exch. Club Br. Isl. 4: 108 (1915).

O. maculata L. subsp. okellyi (Druce) Druce, in Hayward, Bot. Pocket-book, 15th ed. 277 (1917).

Dactylorchis fuchsii (Druce) Vermeulen subsp. okellyi (Druce) Vermeulen, Stud. Dactyl. 149
(1947).

Dactylorhiza fuchsii (Druce) So6 subsp. okellyi (Druce) S06, Nom. nov. gen. Dactylorhiza 8
(1962).

Stem rarely >20 cm, rarely >3-5 mm in diameter, occasionally suffused with anthocyanins (rarely
densely). Sheathing lvs often 4 or more, longest If rarely >9 cm long, widest If rarely >2 cm wide;
non-sheathing lvs occasionally 4 or more; markings often present, occasionally covering >10% of
the upper surface. Inflorescence rarely >4 cm, occasionally >20% of stem length, occasionally
dense, fls o¢casionally more than 25. Floral bracts rarely >8 mm, occasionally suffused with

|
|

BRITISH AND IRISH SPOTTED-ORCHIDS 343

anthocyanins (rarely densely); peripheral bract cells occasionally angular. Labellum rarely >6-7
mm long to apex of central lobe, rarely >5-5 mm to apex of lateral lobe, rarely >4 mm to base of
sinus, occasionally >9-5 mm wide, often broad, sinuses occasionally very deep; reflectivity of base
colour usually >40% (occasionally >80%); markings often present; lateral outer perianth segments
occasionally + horizontal, markings occasionally present (rarely annular); spur rarely >2 mm wide
at entrance, >1-8 mm halfway along. Flowering June/early July. Western Ireland, western
Scotland, Isle of Man.

Although discussion of var. okellyi usually centres on its type locality in the Burren, it was probably
first reported from Sutherland by E. S. Marshall in 1908 (as Orchis scotica). Several other
populations have since been located in western Scotland (Druce 1915, 1918; Heslop-Harrison 1954;
Perring & Sell 1968; Summerhayes 1968; Lang 1980) and the Isle of Man (Perring & Sell 1968; Allen
1963, 1971, 1986), leading to its description as an ‘‘Irish-Hebridean race”’ (Allen 1971) with “‘a wide
’ distribution exhibiting . . . a neat Atlantic pattern” (D. E. Allen pers. comm. 1987).

The origin of the epithet is confused. Druce was shown non-flowering plants of Orchis immaculata
in the Burren by P. B. O’Kelly, who subsequently supplied the type material (also in a non-
flowering state according to Praeger (1934)) described by Druce (1909). It proved to be a white-
flowered variant of D. fuchsii, though Praeger (1934) claimed that the plants were selected by
O’Kelly in error for the white-flowered form of D. maculata, which is also unusually frequent in the
Burren. Wiefelspiitz’s (1976) statement that okellyi was originally described as a variant of Orchis
mascula L. is undoubtedly incorrect.

Previous authors have disagreed over the range of pigmentation encompassed by var. okellyi. The
most stringent do not allow any floral or vegetative anthocyanins (McKechnie 1918; Stephenson &
Stephenson 1920, 1924; Godfery et al. 1924; Camus & Camus 1929; Godfery 1933; Clapham 1952;
Keane 1980), others permit labellum markings only (Druce 1909, 1915, 1918; Stelfox 1924;
Vermeulen 1947; Ettlinger 1976; Webb & Scannell 1983), and the most liberal accept a minority of
individuals with both labellum and leaf markings (Summerhayes 1951; Heslop-Harrison 1954;
Clapham 1962; Allen 1971; Wiefelspititz 1976; Landwehr 1977; Lang 1980; Davies et al. 1983).
Druce caused much of the confusion by (1) precluding all pigmentation from O. fuchsii var. okellyi
in his original diagnosis (Druce 1909) and in its redescription as a full species, O. okellyi (Druce
1915), but adding in the latter paper that some plants have labellum markings, and (2) including a
misprint in the original diagnosis, which should have read “‘leaves . . . unspotted”’ rather than
“leaves . . . spotted’’. Summerhayes (1951) provided the most accurate description: “‘Flowers are
very frequently almost white . . . with very faint pink or lilac markings on the lip, but in some
populations there may be quite a high proportion of plants with more heavily and brightly marked
lips”. In fact, pigmented plants predominated in all the populations that we measured. Taxonomic
recognition of okellyi at the population level therefore requires diagnostic characters other than
those dependent on pigmentation.

D. fuchsii var. okellyi supposedly has smaller flowers (Druce 1909, 1915, 1918; Camus & Camus
1929; Summerhayes 1951; Allen 1971; Wiefelspiitz 1976; Lang 1980; Davies et al. 1983) whose
labella have “more rounded” lateral lobes (Allen 1971; Ettlinger 1976) with crenate margins
(Stelfox 1924), and less prominent central lobes (Druce 1909; Allen 1971; Wiefelspiitz 1976; Lang
1980), resulting in a smaller labellum shape index (Heslop-Harrison 1954). Thus, the labella and
spurs should be intermediate in shape between those of D. fuchsii var. fuchsii and D. maculata
subsp. ericetorum. Our data (which are consistent with the single data set for Burren okellyi
published by Heslop-Harrison (1954)) reveal considerable variation between populations of var.
okellyi in flower size and shape. Black Head plants have similar labellum dimensions (means 6:7 X
9-5 mm) to var. fuchsii (Fig. 6). Khyber Pass differs from all other populations in having wide but
short labella, resulting in very small mean values for labellum roundness (0.37), labellum shape.
index (1.32) and central lobe prominence (0.6 mm). Corrofin and Gelain have small labella (mean
values c. 5-7 X 8-8 mm) of similar size to D. fuchsii var. alpina. Only one population (Gelain) had
unusually short (<5 mm), narrow (median width <1-5 mm) spurs.

Var. okellyi is also supposed to have short (Allen 1971), slender (Druce 1915, 1918; Godfery et al.
1924; Summerhayes 1951; Allen 1971) stems with few (Ettlinger 1976; Wiefelspiitz 1976), slender
leaves (Druce 1909; Camus & Camus 1929; Godfery 1933; Summerhayes 1951; Allen 1971;
Ettlinger 1976; Wiefelspiitz 1976; Lang 1980) and a narrow (Vermeulen 1947), dense (Druce 1909),

344 R.M. BATEMAN ANDI. DENHOLM_ .-

cylindrical/flat-topped inflorescence (Druce 1909, 1915; Camus & Camus 1929; Godfrey 1933; Lang
1980; Davies et al. 1983). However, our data show that its stem height and diameter are not
unusually small, nor are its leaves unusually short. Its leaves are fairly narrow (means 13-17 mm)
but are matched by populations of D. fuchsii var. hebridensis, var. cornubiensis and var. alpina.
Inflorescence shape is not a useful taxonomic character; it reflects the length of the inflorescence and
the proportion of flowers open, and therefore changes during anthesis. We have also been unable to
detect the strong changes during anthesis. We have also been unable to detect the strong fragrance
attributed to okellyi by Stelfox (1924), Godfery (1933). Clapham (1962), Landwehr (1977), Lang
(1980) and Davies et al. (1983), and there is little evidence to support the claim by Wiefelspiitz
(1976) that it can flower as early as April. Thus, populations of D. fuchsii var. okellyi are much less
distinct from var. fuchsii than has been suggested.

It is therefore tempting to regard var. okellyi as anthocyanin-less individuals rather than
populations, but it is equally difficult to distinguish Burren anthocyanin-less individuals from those
elsewhere in the British Isles. Druce (1915, 1918), subsequently supported by McKechnie (1918, p.
185), Stelfox (1924), Summerhayes (1951, p. 271) and Lang (1980), argued that this was possible,
but at the same time he described white-flowered plants from several sites in Yorkshire as var.
okellyi (Druce 1916). Similarly, Godfery (1933) listed numerous records for var. okellyi from
throughout the British Isles. Stelfox’s (1924) argument that var. okellyi always has yellow pollinia
while other white-flowered individuals of D. fuchsii have pink pollinia is void, as both pollinia colour
morphs occur throughout the range of D. fuchsii. Heslop-Harrison (1949, p. 293) stated that plants
of D. fuchsii in Co. Donegal with white flowers and unmarked leaves “‘are simply aberrant forms of
the natural population; the white [D. fuchsii] races of north Clare [i.e. var. okellyi on the Burren]
are more worthy of critical examination’’. However, three years later he wrote more cautiously “‘In
O. fuchsii populations elsewhere [i.e. outside the Burren] albino individuals occur which agree with
the type diagnosis of ssp. okellyi, but this does not remove the necessity for recognising as a distinct
taxonomic entity the race which occurs in Clare and Sutherlandshire”’ (Heslop-Harrison 1952, p.
105). We agree; anthocyanin-less individuals of D. fuchsii from the Burren cannot be reliably
distinguished from those elsewhere in the British Isles, and the name okellyi is best retained at
varietal level to describe heterogeneous, anthocyanin-low populations of D. fuchsii along the
western coasts of the British Isles.

D.E. Allen (pers. comm. 1986) argued that anthocyanin-low populations of D. fuchsii on the Isle
of Man that have been attributed to var. okellyi (Allen 1963, 1971, 1986; Perring & Sell 1968) are
more distinct from var. fuchsii than those in the Burren, having very small labella with three
subequal lobes and showing different geographical and ecological distributions from the type. If
these observations can be supported by biometric data, the Manx populations lie outside the range
of variation encompassed by var. okellyi and represent a new infraspecific taxon.

We cannot detect any characters in descriptions of the diploid Hungarian ‘endemic’ D. fuchsii
subsp. sooiana (Borsos) Borsos (Borsos 1959, 1961; Landwehr 1977; Sod 1980) that justify its
separation from plants of var. okellyi with pigmented but unmarked labella.

2. Dactylorhiza maculata (L.) Sod, Nom. nov. gen. Dactylorhiza 7 (1962).

Orchis maculata L., Sp. Pl. 942 (1753).

Dactylorchis maculata (L.) Vermeulen, Stud. Dactyl. 130 (1947).

Orchis candidissima Krocker, Fl. Silesiaca 3: 16 & tab. 2 (1814).

. elodes Grisebach, Uber. Bild. Torfs Emsm. 25 (1846).

O. maculata L. var. helodes (Grisebach) Reichenbach f., Icon Fl. Germ. Helv. 67 (1851).

. maculata L. var. praecox Webster, Brit. Orchid. 54 (1886).

. maculata L. f. candidissima (Krocker) Schulze, Orchid. Deutsch., Deutsch. Schweiz (1894).

. maculata L. subsp. ericetorum Linton, Fl. Bournemouth 208 (1900).

. maculata L. subsp. elodes (Grisebach) Camus, Mon. Orch. Eur. 192 (1908).

. ericetorum (Linton) Marshall in Rep. botl Soc. Exch. Club Br. Isl. 3: 127 (1912).

. maculata subvar. leucantha Druce in Rep. botl Soc. Exch. Club Br. Isl. 8: 213 (1916).

. maculata L. f. ericetorum (Linton) Hagerup in Dansk. Bot. Ark. 11: 3 (1944).

Dactylorchis maculata (L.) Vermeulen subsp. ericetorum (Linton) Vermeulen, Stud. Dactyl. 69
(1947). -

fe)

BDOOO000

BRITISH AND IRISH SPOTTED-ORCHIDS 345

D. maculata (L.) Vermeulen subsp. typica Vermeulen, Stud. Dactyl. 131 (1947).

D. elodes (Grisebach) Vermeulen, Stud. Dactyl. 137 (1947).

Orchis fuchsii Druce subsp. rhoumensis Heslop-Harrison f. in Trans. Proc. bot. Soc. Edinb. 35: 53

1948).

. maculata (L.) Vermeulen subsp. elodes (Grisebach) Vermeulen in Ned. Kruidk.
Archf. 56: 235 (1949).

D. maculata (L.) Vermeulen var. candidissima (Krocker) Vermeulen in Ned. Kruidk. Archf. 56:
235 (1949).

D. maculata (L.) Vermeulen subsp. rhoumensis (Heslop-Harrison f.) Heslop-Harrison f. in
Watsonia 4: 48 (1957). 4

D. maculata (L.) Vermeulen var. ericetorum (Linton) Vermeulen, in Fl. Neerlandica (1958).

D. fuchsii (Druce) Vermeulen subsp. rhoumensis (Heslop-Harrison f.) Clapham, Fl. Br. Isl., 2nd
ed. 1046 (1962).

Dactylorhiza maculata (L.) So6 subsp. rhoumensis (Heslop-Harrison f.) Sod, Nom. nov. gen.
Dactylorhiza 7 (1962).

D. maculata (L.) So6 subsp. elodes (Grisebach) S06, Nom. nov. gen. Dactylorhiza 7 (1962).

D. maculata (L.) So6 var. praecox (Webster) Sod, Nom. nov. gen. Dactylorhiza 7 (1962).

D. maculata (L.) So6 subsp. ericetorum (Linton) P. F. Hunt & Summerhayes in Watsonia 6: 132
(1965).

D. maculata (L.) So6 f. candidissima (Krocker) Landwehr in Orchideeén 37: 79 (1975).

Stem 4—40(—50) cm, 1-3—5-5(—8-5) mm in diameter, often lacking anthocyanins. Basal 1f or sheath
0-1, broadest at or above middle; sheathing 1vs (1—)2—4(-5), usually strongly crowded towards
base of stem, usually recurved, narrowly lanceolate, usually broadest well above base, longest 1f
often also widest, 4-16(—19) cm long, widest 1f 0-5—2(—2-5) cm wide, width/length ratio of 1vs
decreasing up stem, usually bright to dark green, tips occasionally distinctly hooded; non-sheathing
lvs (1-)2—5(—7), narrow, broadest at base; 1vs usually sparsely to densely marked on upper surface
only, markings usually solid (very rarely annular), + evenly distributed or concentrated towards
tips, + round to tranversely elongated, usually 1-3 mm in mean diameter. Inflorescence 1—8(-—10)
cm, 8—35% of stem length, fls 5—50(—60), lax to dense (2—10(—12) fls/cm). Basal bracts 6—20(—25)
mm, 1-5—2 times the length of the ovaries, floral bracts 4-12(—15) mm, approximately equalling the
ovaries, often suffused with anthocyanins; peripheral bract cells 55—120(—150) um long, barrel-
shaped to triangular. Labellum width usually exceeding length (often considerably), (4—)5—9-5(-
11) x (5-5—)6-5—13(—15) mm, widest + at middle or less frequently above (obtriangular), base
colour varying densities (reflectivity (10—)25—89%) of purple, less frequently red-purple or white;
markings pale to bold, ranging from dots to dashes and loops (rarely absent or only loops), often
covering about two-thirds of the labellum, occasionally more or less; sinuses present (labellum
three-lobed), usually shallow, central lobe rarely exceeding lateral lobes by >1 mm, labellum shape
index rarely >1-25; lateral lobes often indented, slightly deflexed to moderately reflexed; lateral
outer perianth segments usually nearer horizontal than vertical, usually with solid (rarely annular)
markings; median outer perianth segment and inner perianth segments connivent; spur straight to
moderately decurved, (2-5—)3-5—8(-9-5) x 0-6—2:2 mm at entrance, 0-5—2 mm halfway along,
cylindrical or slightly tapering, half as long to as long as the ovary. 2n=?80. Flowering mid-June to
late July (rarely August in the west). Frequent and locally common in the north and west, local in
central and eastern England and Ireland. Neutral to moderately acid soils.

Five noteworthy intraspecific taxa of D. maculata have been recognized in the British Isles (Table
9). Subsp. rhoumensis, a relatively recent segregate confined to Rhum, is discussed later. No author
has chosen to recognize Orchis maculata var. praecox since its original description (Webster 1886;
Druce 1912), but there is considerable disagreement concerning which of the other three subspecies
encompass(es) British and Irish populations of D. maculata. Most British authors (e.g. Bennett
1921; Summerhayes 1951; Heslop-Harrison 1951, 1954; Ettlinger 1976; Lang 1980) have only
recognized one subspecies, subsp. ericetorum, and have regarded Grisebach’s (1846) Orchis elodes
as a synonym of ericetorum and subsp. maculata as exclusively Continental. Godfery (1921, 1923,
1933) dissented on a nomenclatural point, arguing that subsp. elodes is synonymous with, and has
priority over, subsp. ericetorum. In contrast, most Centinental workers believe that subsp. maculata

346 R. M. BATEMAN AND I. DENHOLM

occurs in Britain, together with a second subspecies. This may be subsp. ericetorum, with elodes
treated as a separate, exclusively Continental subspecies (So6 1960; Wiefelspiitz 1976; Delforge &
Tyteca 1984), or it may be subsp. elodes, with ericetorum treated as a synonym of elodes (Camus &
Camus 1929; Senghas 1968) or of subsp. maculata (So6 1980). Alternatively, all three subspecies
may occur in the British Isles (Vermeulen 1947; Landwehr 1977). The complex reasons for this lack
of consensus are a microcosm of taxonomic problems in general, and will be discussed in detail
elsewhere.

D. MACULATA SUBSP. RHOUMENSIS

D. maculata subsp. rhoumensis was originally described as a subspecies of D. fuchsii endemic to
Rhum by Heslop-Harrison (1948). He initially attributed it to D. fuchsii due to its apparently diploid
karyotype, but later transferred it to D. maculata, which it more closely resembled in morphology
(Heslop-Harrison 1957). Clapham (1952, 1962) and Perring & Sell (1968) considered it morphologi-
cally indistinguishable from D. maculata subsp. ericetorum, though Ettlinger (1976) and Davies et
al. (1983) stated that it has unusually bold, looped labellum markings. The exceptionally late
flowering period (August) attributed to rhoumensis by Heslop-Harrison has been overlooked by
most subsequent authors, whilst most Continental workers have overlooked this taxon completely
(Table 9b).

Our three study populations from Rhum spanned a wide flowering period from early July
(Kinloch) to early August (Harris). Each population deviated considerably from the original
diagnosis of rhoumensis. Kinloch was the closest in overall morphology, but it flowered much too
early and had leaf markings that were much too dense (mean areal coverage 25%). Kilmory also
flowered too early, and had pale flowers (mean reflectivity 72-6%) with broken labellum markings
of only low to moderate contrast. Harris flowered at the prescribed time for subsp. rhoumensis and
had the stipulated dark flower colour (mean reflectivity 40%), but the plants were too short (mean
height 6-9 cm), had a large mean labellum shape index (1-24) intermediate to typical values for D.
fuchsii and D. maculata, and was tetraploid (2n = c. 80; A. Karp pers. comm. 1985). Furthermore,
Heslop-Harrison (1948, p. 53) specifically excluded from subsp. rhoumensis the populations of D.
maculata occupying the Harris machair.

Thus, none of our three study populations fulfilled the morphological criteria in the original
diagnosis of subsp. rhoumensis. Since Rhum populations of D. maculata are very variable, and
overlap almost completely populations elsewhere in the British Isles, we conclude that rhoumensis
is not a convincing taxon.

OTHER INFRASPECIFIC TAXA

Nonetheless, the August-flowering population from Harris provided an interesting comparison with
Porlock, an August-flowering population from Exmoor (Table 1). These two populations were
much the darkest-flowered of those studied (mean reflectivities <40%) and had narrow (c. 2 mm)
stems bearing few (c. 3), narrow (c. 7 mm) sheathing leaves. However, they differed from each
other considerably in several other characters. The shorter stem, shorter, more basally concentrated
leaves and proportionately longer inflorescences of the Harris plants may merely reflect environ-
mental dwarfing, but this is unlikely to explain their more sparse leaf markings, much shorter (c. 4
mm) spurs and more prominent (c. | mm) central labellum lobes. It would therefore be unwise to
unite these populations as an infraspecific taxon.

None of the plants measured approached the 18 mm labellum width of Druce’s (1920, p. 579) D.
maculata var. macroglossa, which was probably a hybrid, but several exceeded the 12-5 mm
labellum width required for D. maculata f. grandiflora (Vermeulen) So6 (Landwehr 1977).
However, we see no merit in perpetuating a taxon diagnosed by a single size character.

The tall, relatively slender Llandegfan plants resembled D. maculata var. traunsteinerifolia
(Harz) So6 (see Landwehr (1977, p. 37)). Other morphologically distinct populations were
Stanmore, with an unusually large number of non-sheathing leaves, and Thursley, with unusually
reflexed lateral labellum lobes more characteristic of D. incarnata (Bateman & Denholm 1985).
Some British and Irish populations of D. maculata, particularly those growing in exposed habitats,
resemble Scandinavian endemics described by Landwehr (1975, 1977): D. maculata subspp. deflexa
Landwehr, montellii (Vermeulen) Landwehr and elodes var. darnalensis Landwehr. The latter

BRITISH AND IRISH SPOTTED-ORCHIDS 347

closely resembles the late-flowering Porlock population. Comparative studies of British and
Scandinavian populations are desirable to investigate these relationships.

EXTREMES OF PIGMENTATION

Stephenson & Stephenson (1921, p. 123) described a plant of D. maculata subsp. ericetorum with
“the whole centre of the lip taken up by a patch of bright magenta’”’. This anthocyanin-rich form of
D. maculata, analogous to that already described for D. fuchsii, has also been reported from the
Continent. It was named Dactylorchis maculata var. concolor Vermeulen (1949, p. 232), later
demoted to Dactylorhiza maculata f. concolor (Vermeulen) Landwehr (1975, p. 79).

The well-known anthocyanin-low mode of D. fuchsii in the Burren (var. okellyi) occurs with a
similar but less distinct mode of D. maculata which includes a few anthocyanin-less individuals
(Druce 1915, 1918) that may have been confused with albino D. fuchsii (Praeger 1934; Heslop-
Harrison 1952, 1954). Burren D. maculata also have labella that are on average unusually broad
. relative to their length (index ‘a’ means 0-40—0-42) and wide spurs (means 1-4—1-7 mm for median
width) resembling those of D. fuchsii, though we could not detect the fragrance attributed to them
by Clapham (1952, 1962). Anthocyanin-low populations of D. maculata occur in other limestone
districts (Heslop-Harrison 1954); all four anthocyanin-less plants that we measured were found at
Ashes Gill and White Nook, populations from the Carboniferous limestone of Yorkshire. However,
even here the frequency of albinos has been exaggerated due to lack of close scrutiny; Marshall
(1912) described white-flowered D. maculata as “‘not uncommon” in Yorkshire, but qualified this by
stating that they were “‘seldom quite pure white”’!

Interestingly, the epithet candidissima (currently D. maculata f. candidissima (Krocker) Land-
wehr (1975, p. 79)) is usually applied solely to albinos (e.g. Heslop-Harrison 1954). The original
description of Orchis candidissima (Krocker 1814, p. 16 & Plate 2) is sufficiently detailed to show
that it is indeed a variant of D. maculata (Druce 1925; Vermeulen 1947) but also that it is not a
suitable basionym for a taxon composed of anthocyanin-less individuals; Krocker specified lightly
spotted leaves, a pigmented spur and purple pollinia for candidissima. As Druce was aware of this
(Druce 1925, p. 139), it is surprising that he suggested synonymy of candidissima with his Orchis
maculata subvar. leucantha Druce (1916) (see also Godfery (1933)).

ACKNOWLEDGMENTS

We thank the Botanical Research Fund for financial support of fieldwork, Belinda Denholm and
Orpah Farrington for assistance in the field, R. Bush, S. Brown and R. H. Roberts for providing
material from some study populations, and D. M. T. Ettlinger and J. T. Temple for critically
reading the manuscript.

REFERENCES

AKErROYD, J. R. & Curtis, T. G. F. (1980). Some observations on the occurrence of machair in western Ireland.
Bull. Ir. biogeog. Soc. 4: 1-12.

ALLEN, D. E. (1963). O’Kelly’s orchid in the Isle of Man. Peregrine 3: 128-129.

ALLEN, D. E. (1971). Dactylorhiza fuchsii subsp. okellyi (Druce) SOO behaviour and characteristics in the Isle
of Man. Watsonia 8: 401-402.

ALLEN, D. E. (1986). Flora of the Isle of Man. Douglas.

ANONYMOUS (1966). Royal Horticultural Society colour chart. London.

AVERYANOV, L. V. (1977). Chromosome numbers of some species of the family Orchidaceae for the Leningrad
and Volgograd oblasts. Bot. Zh. (Leningr.) 62: 547-553.

BATEMAN, R. M. & DENHOLM, I. (1983). A reappraisal of the British and Irish dactylorchids, 1. The tetraploid -
marsh-orchids. Watsonia 14: 347-376.

BATEMAN, R. M. & DENHOLM, I. (1985). A reappraisal of the British and Irish dactylorchids, 2. The diploid
marsh-orchids. Watsonia 15: 321-355.

BAUMANN, H. & KUNKELE, S. (1982). Die wildwachsenden orchideen Europas. Stuttgart.

BENNETT, A. (1921). Orchis elodes Griseb. and Orchis ericetorum Linton. J. Bot., Lond. 59: 354-355.

Borsos, O. (1959). Dactylorchis fuchsii Druce et son affinité dans le flores Hongroise et Carpathique. Acta Bot.
Acad. Scient. Hung. 5: 321-326. :

348 R.M. BATEMAN ANDI. DENHOLM .

Borsos, O. (1961). Geobotanische Monographie der Orchideen der Pannonischen und Karpatischen Flora, 5.
Dactylorhiza Il. Ann. Univ. Scient. Budapest Sect. Biol. 4: 51-82.

BrEBISSON, L. A. DE (1835). Flore de la Normandie. Paris.

BRENAN, J. M. P. & Simpson, N. D. (1949). The results of two botanical journeys in Ireland in 1938-39. Proc.
Roy. Irish Acad. 52: 57-84.

BRONGNIART, A. (1832). Expédition scientifique de Morée, 3.

Camus, E. G. & Camus, A. (1929). Sub-gen. Dactylorchis, in Iconographie des Orchidées d’ Europe et du Bassin
Méditerranéen. pp. 208-243. Paris.

CLAPHAM, A. R. (1952). Orchis L., in CLAPHAM, A. R., Tutin, T. G. & WarBuRG, E. F. Flora of the British Isles,
pp. 1310-1323. Cambridge.

CLAPHAM, A. R. (1962). Dactylorchis (Klinge) Vermeul., in CLAPHAM, A. R., Tutin, T. G. & WARBURG, E. F.
Flora of the British Isles, 2nd ed., pp. 1042-1049. Cambridge.

Davies, P., Davies, J. & Hux.ey, A. (1983). Wild orchids of Britain and Europe. London.

DELForGE, P. & Tyteca, D. (1984). Guide des orchidées d’Europe dans leur milieu naturel. Paris.

Druce, G. C. (1909). Notes on Irish plants. Jrish Nat. 9: 209-213.

Druce, G. C. (1912). Orchis maculata L. var. praecox Webster. Rep. botl Soc. Exch. Club Br. Isl. 3: 127-128.

Druce, G. C. (1915). Orchis maculata L. and O. fuchsii. Rep. botl Soc. Exch. Club Br. Isl. 4: 99-108.

Druce, G. C. (1916). Plant notes for 1915: Notes on Orchis. Rep. botl Soc. Exch. Club Br. Isl. 4: 211-214. -

Druce, G. C. (1918). Notes on the British orchids, chiefly the palmate section. Rep. botl Soc. Exch. Club Br. Isl.
(supplement) 5: 149-180.

Druce, G. C. (1920). Plant notes for 1919: British marsh-orchids. Rep. botl Soc Exch Club Br. Isl. 5: 576—580.

Druce, G. C. (1924a). Orchis fuchsii Druce. J. Bot., Lond. 62: 198-201.

Druce, G. C. (1924b). Orchis maculata L. and Orchis fuchsii Druce. Rep. botl Soc. Exch. Club Br. Isl. 7: 322-
328.

Druce, G. C. (1925). What is Orchis candidissima Krocker? J. Bot., Lond. 63: 138-139.

Ett LinGeER, D. M. T. (1976). British and Irish orchids — a field guide. London.

GoprFery, M. J. (1921). Orchis elodes Grisebach. J. Bot., Lond. 59: 305-308.

Goprery, M. J. (1923). Orchis fuchsii Druce. J. Bot., Lond. 61: 306-309.

Gop ery, M. J. (1924). Orchis fuchsii Druce. J. Bot., Lond. 62: 202.

GopreryY, M. J. (1933). Monograph and iconograph of native British Orchidaceae. Cambridge.

GoprFery, M. J., STEPHENSON, T. & STEPHENSON, T. A. (1924). The British dactylorchids — a comparative table.
J. Bot., Lond. 62: 175-178.

Gower, J. C. (1966). Some distance properties of latent root and vector methods used in multivariate analysis.
Biometrika 52: 325-338.

Gower, J. C. (1971). A general coefficient of similarity and some of its properties. Biometrics 27: 857-872.

Gower, J. C. & Ross, G. J. S. (1969). Minimum spanning trees and single linkage cluster analysis. J. R. statist.
Soc. C. 18: 54-64. _

GrISEBACH, A. (1846). Uber die Bildung des Torfs in den Emsmooren.

Gro_L, M. (1965). Fruchtansatz, Bestaubung und Merkmalsanalyse bei diploiden und polyploiden Sippen von
Dactylorchis (Orchis) maculata und Gymnadenia conopsea. Osterr. Bot. Z. 112: 657-700.

Gro t, M. (1968).Fruchtansatz, Bestaubung und Merkmalsanalyse bei diploiden und polyploiden Sippen von
Dactylorhiza (Dactylorchis) maculata s. lat. Jber. naturw. Ver. Wuppertal 21-22: 98-101.

Hacervp, O. (1938). Studies on the significance of polyploidy. II. Orchis. Hereditas 24: 258-264.

HaGervpP, O. (1944). On fertilisation, polyploidy and haploidy in Orchis maculatus L. sens. lat. Dansk. Bot. Ark.
11: 1-25.

Hes.op-Harrison, J. (1948). Field studies in Orchis L., I. The structure of dactylorchid populations on certain
islands in the Inner and Outer Hebrides. Trans. Proc. bot. Soc. Edinb. 35: 26-66.

Hes.op-Harrison, J. (1949). Notes on the distribution of the Irish dactylorchids. Verdff. geobot. Inst. Riibel,
Ziirich 25: 100-113.

Hes.op-Harrison, J. (1950). Notes on some Irish dactylorchids. /r. Nat. J. 10: 82-83.

Hes.op-Harrison, J. (1951). A comparison of some Swedish and British forms of Orchis maculata L. sens. lat.
Svensk Bot. Tidskr. 45: 608-635.

HesLop-Harrison, J. (1952). Notes on the distribution of the Irish Dactylorchids. Veréff. geobot. Inst. Riibel,
Zurich 25: 100-113.

Hes_op-Harrison, J. (1954). A synopsis of the dactylorchids of the British Isles. Ber. geobot. Forsch. Inst. Riibel
1953: 53-82.

Hes.op-Harrison, J. (1957). Dactylorchis maculata subsp. rhoumensis (H.-Harrison f.) H. Harrison f., comb.
nov. Watsonia 4: 48.

Hesvop-Harrison, J. (1968). Genetic system and ecological habit as factors in dactylorchid variation. Jber.
naturw. Ver. Wuppertal 21-22: 20-27.

Heusser, C. (1938). Chromosomenverhiltnisse bei schweizerischen basitonen Orchideen. Ber. Schweiz. Bot.
Ges. 48: 562-605.

BRITISH AND IRISH SPOTTED-ORCHIDS 349

Ho tus, J. (1983). Reclassification and new names for some European phanerogams. Folia Geobot. Phytotax.,
Praha 18: 203-206.

Hunt, P. F. & SUMMERHAYES, V. S. (1965). Dactylorhiza Nevski, the correct generic name of the dactylorchids.
Watsonia 6: 128-133.

Jacietto, M. (1986). Nowa nazwa gatunkowa dla Dactylorhiza fuchsii (Druce) So6 (Orchidaceae)? Fragm. Fl.
Geobot. 30: 213-216.

KEANE, M. (1980). The Burren. Dublin.

Krocker, A. J. (1814). Flora Silesiaca 3. Bratislava.

LANDWEHR, J. (1975). Het geslacht Dactylorhiza. Orchideeén 37: 76-80.

LANDWEHR, J. (1977). Wilde orchideeén van Europa. Amsterdam.

Lana, D. (1980). Orchids of Britain. Oxford.

LINNAEUS, C. (1753). Species plantarum. Stockholm.

Linton, E. F. (1900). Flora of Bournemouth. Bournemouth.

McKecunlie, H. (1918). Notes on the genus Orchis. Rep. botl Soc. Exch. Club Br. Isl. (Supplement) 5: 183-189.

Marcetrts, L. J. & Davin, R. W. (1981). A review of the Cornish flora, 1980. Redruth.

. MarsHALL, E. S. (1912). Orchis ericetorum Linton. Rep. botl Soc. Exch. Club Br. Isl. 3: 127.

NELSON, E. (1976). Monographie und Ikongraphie der Orchidaceen-Gattung, III. Dactylorhiza. Zurich.

NeEuMAN, L. M. (1909). Anteckninger rérande nordiska Orkis-former. Bot. Notiser 1909: 151-159, 229-246.

Payne, A. W. et al. (1987). Genstat 5 reference manual. Oxford.

PERRING, F. H. & SELL, P. D., eds. (1968). Critical supplement to the Atlas of the British flora, pp. 141-143.
London.

PRAEGER, R. L. (1934). The botanist in Ireland. Dublin.

PRENTICE, H. C. (1986). Continuous variation and classification, in StyLes, B. T., ed., Infraspecific variation of
wild and cultivated plants, pp. 21-32. Oxford.

Pucs.ey, H. W. (1939). Recent work on dactylorchids. J. Bot., Lond. 77: 50-56.

Pusey, H. W. (1940). Further notes on British dactylorchids. J. Bot., Lond. 78: 177-181.

REICHENBACH, H. G. (1851). Orchideae, in Jcones Florae Germanicae et Helveticae 13-14. Leipzig.

REINHARD, H. R. (1985). Skandinavische und alpine Dactylorhiza-Arten (Orchidaceae). Mitt. Bl. Arbeitskr.-
Heim. Orch., Baden-Wiirtt. 17: 321-416.

ScHMID, F. W. (1791). Die in BOhmen wildwachsenden Pflanzen aus dem Geschlecht der Orchis. Sammi. Phys.
Aufsdatze 15: 217-254.

SENGHAS, K. (1968). Taxonomische Ubersicht der Gattung Dactylorhiza Necker ex Nevski. Jber. naturw. Ver.
Wuppertal 21-22: 32-67.

So6, R. DE (1960). Synopsis generis Dactylorhiza (Dactylorchis). Annls Univ. Scient. bpest Rolando Eétvos
(sect. biol.) 3: 335-357.

So6, R. DE (1980). Dactylorhiza Necker ex Nevski, in TuTin, T. G. et al., eds. Flora Europaea 5: 333-337.
Cambridge.

STELFOx, A. W. (1924). The white form of Orchis fuchsii and Orchis o’kellyi. Irish Nat. 33: 143-144.

STEPHENSON, T. & STEPHENSON, T. A. (1920). The British palmate orchids. J. Bot., Lond. 58: 257-262.

STEPHENSON, T. & STEPHENSON, T. A. (1921). The forms of Orchis maculata. J. Bot., Lond. 59: 121-128.

STEPHENSON, T. & STEPHENSON, T. A. (1923). Orchis elodes Grisebach. J. Bot., Lond. 60: 337.

SUMMERHAYES, V. S. (1951). Wild orchids of Britain. London.

SUMMERHAYES, V. S. (1968). Wild orchids of Britain, 2nd ed. London.

SUNDERMANN, H. (1980). Europdische und mediterrane Orchideen — Eine Bestimmungsflora, 3rd ed. Hildes-
heim.

VERMEULEN, P. (1947). Studies on dactylorchids. Utrecht.

VERMEULEN, P. (1949). Varieties and forms of Dutch orchids. Ned. kruidk. Archf. 56: 204-242.

VERMEULEN, P. (1968). Dactylorchis maculata und ihre formen. Jber. naturw. Ver. Wuppertal 21-22: 68-76.

Votu, W. (1978). Biometrische Untersuchungen an Dactylorhiza maculata s. 1.-Sippen in Niederdsterreich
(Orchidaceae). Linzer Biol. Beitr. 10: 179-215.

VotrH, W. & GREILHUBER, J. (1980). Zur karyosystematik von . Dactylorhiza maculata s. 1. Und ihrer
Verbreitung, insbesondere in Niederdésterreich. Linzer Biol. Beitr. 12: 415-468.

Wess, D. A. & SCANNELL, M. J. P. (1983). Flora of Connemara and the Burren. Cambridge.

Wesster, A. D. (1886). British orchids. Bangor.

WIEFELSPUTZ, W. (1976). Uber einige Dactylorhiza-Sippen in Gréssbritannien und Irland. Jber. naturw. Ver.
Wuppertal 29: 41-51.

WivtmorTr, A. J. (1939). Notes on a short visit to Barra. J. Bot., Lond. 77: 189-195.

(Accepted August 1988)

2 hee ‘5 ’

- ae “mL Gee
E weiss a4 ;
A Do) is PH
Pac hy =

PELOMSRIART, Fe:
fe . ‘ oe > gu oa
Ratads fay = GE SA ia A, (8 OD Sadm ec

be * . “ _ ae
Riediterransin. oo. To. Peewee.

. = we = -*) * cere 7
as 5 ous nia SMO, Cekisk at eee x

ae ¢ ¢ ‘ s a Fe é cme ihe

onal ‘. mote ARTA
yi ; ’ oe or -- - é
ce 4 | cae By P
“LAPHAL R. teh. blecseylorehee lt A » Vetonca
: bh ye f ots =) ri «= Cte. Ec Teg ie
‘A. . ater F ty rey seu on ¥ +155 fiup mat
Lay c rh ‘ i :
r a. . ’ r “a. | Ps _—. -
—- a. Osi eoaL AY. oh xs) 3 es oe ee cas *; ri Hy ts) iy rye
Rei ies f oe vest
: E's ye. 4 -. y J tt aw it 4 a | ays
ryt, Sitka Aes opt Azinga ais ass
aa = “Tt a aes, i? 6 red ~ J “<* L@s.' 2 Af 7
a's t-te A “yi: os oS ty +. A “<tr
. (A *' o — nA) o> ta wh
‘ » *. oh 7 43 - eve bel
: AGT MEAN AO) Crys a) -S ae eI a Ba
ia e 5 Se *3
1" ae ? 4 >? ¢ pe: i ave * a aan ar r. fyi 3 duqtyy ofb
r. 2 a7. f a { te’ ‘ * oat |
alt 4 a : NS hs 4
bw i « iw 3%? 5
- a ‘
FS rsa wen ; Ai 4 4 nie aot "i
> z t S a
= s * ra ¥ ‘ ,. > 7
“= f —— Pa BYE > x
Pa sf 1 F m . "te 2 & ,
. oY 4 _ ' ’ ., : . oe », oe ee rel ws = Wes | wy a cite abahy Ay
% > ‘ 7 . -=P- per ‘ yu ~ ) Se
~ xt
ES is += a ’ ¢ ‘ - s }
AeA? oT at: 4 -~ : Viskbh da 3 ms
: ,
yy - «4 el ] x : Re Sees
‘ 2 -% pS RA * rae a! --
i ; Si “7 2 ] =. yy S '
“4 4 ‘
~< ie 47) *. : 2). ¢ t ca 4 7 . : ani) Smeke (eoe
~ ifs : a . st ae vA S45 ho
ry : ; _ - ee rs hes {S631 ai Reo k
ANY : } tear) os AA isha sedoehier ond ofr pi atte if
- 4 7 aa
~ a ? # § -
. Va ‘ i _ q Pee ? Bed Fi
‘ a : 2 . - c i all
4S paraattiell risa. ; oi] BSS tts 2b insieTae obiaeo
j , > =
a! 3 ° ri > a8
, s : SUV Tow) BISNIS. SF
‘
> * : MS Ee Mei
- : — nites . ie , poe +i,
Al + % z Ae'237) Ky. Lae DIA
_ ba ' a * - ' “ 5 > . _>'
c31 rf i }. A ai} iit) rt tS
* we 2 T 7 eri 4 ai \ <q 7 webign
. y awe Fii Ad » a- . hs
. * PS ~ } “= 4 . Ct 4 ire l »,
z. = * . S r
. : - gal ti :
. <% aed ae 2 : ate! obvihs.
; a
(Tpit arte yy ai ) UAel sep te a
: } ee
. | } | ; a4 7 nan
Ge) .sF ate >) 7
2 4 ‘ 7 ' . et, | ‘ uf “ao
: . > a TTT 7 ae iy
: ary F i ; : z ary + MAESTRO W DEL SRY Ty
_ .
. ‘ ie baw, *, " EE Man AD,
4 - > | —-:
= ; x ; . '

} ITO ice 4 3 sabes 2: :
* pa : / ys 5 Tz P 4 Rise 6 ek : i
F ; a2 5 ces GOs, P29 F ay bay rete gee Wat
re, o>) AO TRPSVEAAO) GS aay = me aio nbsieoent {35e
i gt ae
: - y : . . ._ ‘ ‘= ie age. = ' go < ria” wih kt

; “Fe wnt Jol wate Asis reietiohsit rey
2 mht ber sprenc lee egy

vit i seit ne acy. te if
; oF
*
' De ’ :
| 45
4 : of eel 4 » Pos
? 4 < é ™ m AG ache
« (See,
we "ls op Peatiomen 7
“ .
i pe
- ~~,
- _
- 7 .
. 5
e -

Watsonia, 17, 351-365 (1988) 351

Short Notes

OBSERVATIONS ON INLAND POPULATIONS OF VIOLA CANINA L. IN SOUTH-
EASTERN SCOTLAND AND NORTH-WESTERN ENGLAND

Viola canina is a rare inland speciesin southern Scotland and northern England. It is virtually
restricted to coastal habitats. Recently I discovered several small inland colonies in the Tweed valley
of south-eastern Scotland where it was previously unknown, and rediscovered several inland sites,
mainly in Cumbria in north-western England.

In 1983, colonies of V. canina were found by the River Yarrow in Selkirks., v.c. 79, and their
- identification confirmed by Prof. D. H. Valentine. Since then further colonies have been found by
the Rivers Yarrow, Ettrick and Tima Water in Selkirks., and by the River Tweed in Roxburghs.,
v.c. 80. Grassland sites near Ancrum in Roxburghs., and Peel Hospital in Selkirks. also support
small colonies. The altitudes of the localities vary from 45—252 m. In 1985 V. canina was refound on
the shores of Ullswater in both Westmorland, v.c. 69, and Cumberland, v.c. 70, where it was first
discovered at the end of the last century (Hodgson 1898; Wilson 1936). However it appears to have
gone from the rocky islands of Ullswater (G. Halliday pers. comm.). Small populations still occur on
the western bank of the River Lune west of Sedburgh in Westmorland and on the eastern bank in N.
W. Yorks., v.c. 65. I found an additional riverside colony by the River Eden in the Baronswood
gorge in Cumberland. The two grassland sites known to me are on Wan Fell near Penrith.

The riverside habitats consist of rock outcrops, which extend into the beds of the rivers and which
are inundated only by the most severe floods. The Selkirks. rock habitats and those by the River
Lune are composed of hard, fine grained and relatively basic Silurian strata. Those by the River
Tweed are Carboniferous basalts, and the Ullswater rocks are of the Borrowdale volcanic series
which are basic in places. Although the Permian sandstones by the River Eden are acid, the
riverside rocks support a calcicolous flora from flushing with lime-laden water and silt deposition. V.
canina grows on niches and ledges on these open rock outcrops, which support a flora that includes
Antennaria dioica, Campanula rotundifolia, Galium boreale, Leontodon autumnalis, L. hispidus,
Scabiosa columbaria, Solidaga virgaurea and Thymus praecox.

The grassland habitat in Roxburghs. is on the shallow soil overlying an outcrop of basalt, which
also supports Dianthus deltoides. It is of interest, and possibly relevant, that this outcrop also
supports the rare maritime lichen Ramalina polymorpha (Corner 1981). The Selkirks. grassland site
consists of a relatively basic, stony glacial till with Leontodon hispidus, Polygala vulgaris,
Ranunculus bulbosus, Rhinanthus minor and, more locally, Ophioglossum vulgare. Vigorous
colonies of the hybrid V. canina X riviniana occur at the former site and further colonies of this
hybrid have been found in Roxburghs. in dry shallow grassland from which V. canina is absent.
There are few documented accounts of this hybrid in Scotland. The two Cumberland grassland sites
are on light sandy soil overlying the Permian sandstone. Although Calluna is extensive, there are
more basic areas which can be heavily grazed by rabbits. Here V. canina is associated with
Centaurium erythraea, Conopodium majus, Echium vulgare, Galium verum, Gentianella campes-
tris, Leontodon taraxacoides, Lotus corniculatus, Myosotis ramosissima and Viola riviniana. The
hybrid V. canina X riviniana is common at one site and distributed on roadside verges in the
surrounding district, where it is known from five sites.

The British habitat of Viola canina is given as ‘“‘heathy places preferring sandy soils’”’ (Druce
1932), and “‘heaths, dry grassland, dunes and fens’”’ (Moore 1987). However in Scotland and Ireland
it is not infrequently found on lakeshores. Buchanan White (1898) gives its habitat as “‘shingly-
margins of rivers and lakes . . .”. Hadley (1985) mentions “‘stony loch shores, etc.”” and A. McG.
Stirling (pers. comm.) states that in Dunbarton, v.c. 99, it is confined to the stony or rocky shores of
the mainland or islands of the southern part of Loch Lomond. In Ireland there are references to the
lakeshore habitat by Scully (1916), Praeger (1934), Stewart & Corry (1938) and Webb & Scannell
(1983). ,

It is well known that V. canina is intolerant of shade. It seems likely that these specialized open

352 SHORT NOTES

waterside habits, and possibly the grassland ones also, have provided refugia for small relict
populations. With the abundance of open habitats and unleached soil available in the early post
glacial period, V. canina was probably a widespread species in Britain and Ireland. Indeed in
Iceland I have seen it associated with Alchemilla alpina, Galium boreale, Salix herbacea and
Vaccinium uliginosum. Godwin (1975) records it from glacial deposits in the Isle of Man dated at
10,000 years old, with tentative earlier records from the Lea Valley arctic plant beds in Essex as well
as from Bronze and Iron age sites. With the expansion of the forest in the Boreal period, V. canina
would have suffered widespread elimination. In the more northern and western parts of Britain and
in Ireland, the growth of peat and leaching of soils during the wetter Atlantic period would have
reduced the available habitats still further. It is now only able to exist as a relict in these open
relatively basic inland sites which probably never carried closed woodland even during the post
glacial forest maximum (Pigott & Walters 1954). Unfortunately these small populations are
vulnerable to habitat change and their future is uncertain.

ACKNOWLEDGMENT

I would like to thank Dr S. M. Walters for helpful discussion.

REFERENCES

BUCHANAN WHITE, F. (1898). The flora of Perthshire. Edinburgh.

Corner, R. W. M. (1981). A contribution to the lichen flora of south-east Scotland. Trans. Proc. bot. Soc.
Edinb. 43: 307-315.

Druce, G. C. (1932). The comital flora of the British Isles. Arbroath.

Gopwin, H. (1975). The history of the British flora. Cambridge.

Hap ey, G., ed. (1985). A map flora of mainland Inverness-shire. Edinburgh.

Hopcson, W. (1898). Flora of Cumberland. Carlisle.

Moore, D. M. (1987). Violaceae, in CLAPHAM, A. R., Tuttn, T. G. & Moore, D. M. Flora of the British Isles,
3rd ed., pp. 107-112. Cambridge.

Picott, C. D. & Watters, S. M. (1954). On the interpretation of the discontinuous distribution shown by certain
British species of open habitats. J. Ecol. 42: 95-116. .

PRAEGER, R. LL. (1934). The botanist in Ireland. Dublin.

ScuLLy, R. W. (1916). Flora of County Kerry. Dublin.

STEWART, S. A. & Corry, T. H. (1938). A Flora of North-east Ireland. Belfast.

Wess, D. A. & SCANNELL, M. J. P. (1983). Flora of Connemara and the Burren. Cambridge.

Witson, A. (1936). The flora of Westmorland. Arbroath.

R. W. M. CorNER
Hawthorn Hill, 36 Wordsworth St, Penrith, Cumbria, CAll 7QZ

SOME EARLY NOTICES OF ANTHOCYANIN-RICH VARIANTS OF COMMON BRITISH
PLANTS

The Lacnunga (British Museum MS. Harley 585; edited by Cockayne (1866) and Grattan & Singer
(1952)) is an Old English (Anglo-Saxon) magico-medical commonplace book dated at around 1000
A.D. but-compiled from earlier sources. Much of the Lacnunga is adapted from Latin herbals
(Cameron 1983a), although Christian as well as Saxon and Celtic pagan elements are represented.
The book takes the form of herbal recipes supplemented by charms: 194 entries in all. A class of
professional medical men (leeches) practised medicine at this level and had the task of trying to
alleviate the common ailments of the period: particularly “spring fever’ (endemic benign tertian
malaria), dysentery (infection with Entamoeba histolytica) and abscesses caused by sheep liver fluke
infestation (Cameron 1983b). As befits learned men, some able to read Latin, many Latin plant
names are used. Nevertheless, roughly 50% of the Lacnunga plant names are Saxon in origin and an
interesting source of information about our Saxon folk taxonomy. This will be discussed elsewhere,
but it is notable that colours are rarely used to distinguish or denote plants.

=“

SHORT NOTES 353

Of some 200 plant names in the Lacnunga only eleven are associated with colour, such as:
yelodwyrt (Blackstonia perfoliata (L.) Huds. , yellow-wort) from the bright yellow flowers; brunwyrt
(Scrophularia aquatica L., brownwort) from the colour of the stem; hwit cudu (white cud), and
Hwitmering (white ?mastic) both probably referring to the white gum resin of Pistachia lentiscus L..,
imported into Britain from the Mediterranean region during Saxon times. There is also hwewen
hnydele (hued nettle), said by Grattan & Singer (1952) to refer to Lamium purpureum L. This
seems to be the only case in Saxon folk-taxonomy of a plant species being distinguished from
another by a colour qualifier. However, no less than six of these eleven plants are associated with the
colour red (read).

1) reade yearuwe (red yarrow — I have avoided Anglo-Saxon letters in transliterations and use
appropriate modern equivalents where possible). Achillea millefolium L. (yarrow) is an important
ingredient in Saxon leechcraft and wortcunning, being called for in seven recipes of the Lacnunga.
One of these recipes, however (63 in Grattan & Singer’s enumeration), called for reade yearuwe
(red yarrow) specifically (i.e. Achillea millefolium forma rosea Desf. (Hegi 1928, p. 570)), a variant
mentioned in Gerarde (1597), ‘“‘ccommon Yarrowe .. . the flowers whereof are either white or
purple, which being rubbed do yeelde a strong smell, but unpleasant . . .”, Lyte’s (1578) translation
of Dodoens, “‘the floures most commonly all white, sometimes also in this countrey of a purplish
colour” and Bock’s (1552) De Stirpium. Simples in Lacnunga recipe 63 are in alliterative pairs, and
reade yearuwe has the happy property of alliterating with redic (Raphanus sativus L., radish) and
ribbe (Plantago lanceolata L., ribwort). Gerarde (1597), using Tabernaemontanus’ plates, also
illustrates a European species of Achillea of uncertain identity “the second kinde of Milfoile or
Yarrow, hath stalkes, leaves, and roots like unto the former [Achillea millefolium], saving that his
spokie tufts are of an excellent faire red or crimson colour, and being a little rubbed in the hand, is of
a reasonable good savour . . . The first groweth every where in drie pastures and medowes: red
Milfoile groweth in a field by Sutton in Kent, called Holly Deane [Holy-Deane in 2nd edition], from
whence I brought those plants that do grow in my garden; but it is not common every where as the
other is.” This may be a Continental red yarrow introduced for medicinal purposes.

2) reade wuducerfille (red woodchervil). The Anglo-Saxon “‘cerfille”’, from the Latin cerefolium,
refers to Anthriscus cerefolium (L.) Hoffm. (chervil), and it is not surprising that a Latin rather than
Germanic name is used for a plant not native in Britain or northern Europe; rather more surprising,
however, is that wuducerfille [wudufille] should be used instead of the Saxon name (now lost to us)
for Anthriscus sylvestris (L.) Hoffm. Five entries in the Lacnunga call for wuducerfille and two of
these (90 & 127) call for reade wuducerfille specifically (although entry 90 has “‘reade fille’? which is
probably a short form of reade wuducerfille). This variant has purple pigmentation at the nodes, or
sometimes througout the whole stem and even the leaves. Extreme variants with purple-black
spring leaves (Hegi 1926, p. 1020), i.e. Anthriscus sylvestris var. nigra Murr, are sometimes found
growing amongst the typical morph (Gray 1863). The anthocyanin polymorphism was familiar to
Johnson (in Gerarde 1636): “It hath a whitish wooddy root, from which arise round red and hairy
stalkes . . . The leaves are . . . of a dark greene or else reddish colour.”

3) reade hofe (red hove). The Saxons called Glechoma hederacea L. hove or hofe, which survives
in the name alehoof. Two entries in the Lacnunga call for it, one (31) specifically for reade hofe
which conveniently alliterates with ribbe (Plantago lanceolata). Plants of hove with purple stems
and young leaves (G. hederacea forma purpurascens Otruba) are frequently found. Gerarde (1597)
describes Glechoma hederacea as having ‘“‘manie stalkes, of an uncertaine length, slender, and like
those of the vine, something cornered, and sometimes reddish”’.

4) reade netele (red nettle). Netele, in five entries of the Lacnunga, refers to Urtica dioica L.
However, three of these entries (49, 98 & 134) call for reade netele. Entry 134 further specifies “red
nettle that groweth through into a house” (‘‘seo reade netele the thurh ern inwyxth’’). A variant
version (Meaney 1984) of entry 98 (for smallpox) occurs in Bald’s Leechbook (entry 1.xxxix.2 in
Cockayne (1865)); both clearly derive from the same earlier source and both specify red nettle, as-
does the next entry in Bald’s Leechbook (1.xl, also for smallpox) which specifies: “‘the crop
[flowering top] of red nettle’. Many stinging-nettles have anthocyanin pigmentation on the ridges of
their stems and petioles, and some have the entire stems dark purple and the young leaves suffused
purple. An extreme variant with the leaves “suffused with violet-purple”’ has been described (U.
dioica forma purpurascens Druce (1920)). Tabernaemontanus (1590) illustrates both variants:
Urtica maior, sive sylvestris asperior (GroB Nettel/Brennend Nettel) and Urtica rubra (Rot Nettel).

354 SHORT NOTES

Gerarde (1597) using these plates also describes them, of the second saying: ‘“‘our common red
Nettle, is knowne better to some than desired, and therefore needeth no description’’. Johnson, in
his 1636 edition of Gerarde lumps the two, as does Lyte’s (1578) translation of Dodoens, where he
notes: “. . . our common great Nettell . . . The leaves . . . most commonly of a swarte greene
colour, & sometimes reddish”’.

5) reade seales (red sallows). Sallow bark is required in entry 31, but in entry 155 leaf of red sallow
(“‘reades seales leaf’) is required. Several Salices have anthocyanin pigmentation to varying
degrees, but notably Salix purpurea L., which has the anthers, and in some plants the twigs and
buds, tinged purple. Of this species Gilbert-Carter (1936) notes “The bark is very bitter from the
presence of salicin . . . which was much used in medicine before it was supplanted by . . . aspirin

. It is an important osier, and its twigs are often boiled to strip off the bark, and to impart a red
colour to the wood”’.

6) mugewyrt (mugwort). The importance of colour is continued in entry 178: “If one is to have
mugwort for a remedy, then let one take the red for the cure of a male and the green for the cure of a
female” (“‘yif man scyle mugcewyrt to lecedome habban, thonne nime man tha readen wepnedmen
& tha grenan wifmen to lececrefte’’ which Cockayne (1866) translates as: “If a man must have
mugwort for a leechdom, then let him take the red males and the green females for a leechcraft’’).
Mugwort (Artemisia vulgaris L.) is an important item in leechcraft, used in six entries. The purple-
stemmed variant of mugwort is a common plant, the most abundant in some areas according to
Salmon (1931), who calls it Artemisia vulgaris var. coarctata Forselle. Others distinguish this variety
on purely morphological grounds without regard to colour: contracted inflorescence and linear-
lanceolate leaf segments (Hegi 1928, p. 639).

The purple mugwort variant is a familiar one. Fuchs (1542) says: ‘There are three Artemisias in
Dioscorides and commonly found today. The first is platyphyllos which may be called in Latin
Artemisia latifolia. This one is now simply called Artemisia however. Of this there are again two
types contrasting in colour only: one with red stems and flowers is called in German Rotbucken or
RotbeyfiiB[ss]: the other with a white stem and true yellow flowers is called by the Germans
WeiBbucken.” Bock (1552) gives a similar account and Lyte’s (1578) translation of Dodoens has:
‘““Mugworte . . . Of this herbe there be twoo kindes moe [sic], differing onley in colour. . . The one
hath redde branches & floures and is called redde Mugworte. The other hath greenish branches,
changing towardes white, and is called white Mugworte, in all things els like one to an other”.
Gerarde (1636) has: ““The second kinde of Mugwort hath a great thicke and wooddy root, from
whence arise sundry branches of a reddish colour . . .”’, the editor, Johnson, adds: “I know not how
this differeth from the former, but only in the colour of the stalke and flours, which are red or
purplish; whereas the former is more whitish’’. Gerarde uses the woodcuts in Tabernaemontanus
(1590) in which two mugworts are illustrated, viz. Artemisia rubra, Roter Beyfu8 and Artemisia
alba, WeiBer BeyfuB.

Thus all the ‘red’ plants in the Lacnunga are unequivocally anthocyanin-rich variants, flushed
with purple in their stems, leaves or flowers. It is interesting that the word ‘red’ is used, as Anglo-
Saxons sometimes reserved this word for yellowish-red (Barley 1974). Laxity in colour description is
continued today in the use of the name “copper beech”’ for the anthocyanin-rich variant of Fagus
sylvatica L. The classically derived term ‘purple’ is based on the important colour reference point of
Tyrrhenian purple, the dye derived from Murex spp. (in fact crimson (Stearn 1973)), which may not
have been familiar in tenth century Britain.

As red variants are biochemically different from green ones, a pharmacological basis for the
preference in leechcraft cannot be ruled out. The origin is more likely to lie in Roman sympathetic
colour magic, however. The Lacnunga entries which call for red plants are almost all from the Latin
and Christian sources (a & c of Grattan & Singer (1952)), while none come from the Saxon pagan
source (b of Grattan & Singer (1952)). A passage of Marcellus Empiricus quoted by Grattan &
Singer illustrates the importance of colour: ‘“‘cut open the crop of a swallow. Little stones, both
white and black will be found. Placed in a golden locket they will permanently avert all eye pain and,
if wrapped in a yellow cloth or flaxen sac and hung around the neck, they avail against fevers’’.
Another case is provided in the Lacnunga itself (entry 7) in which there are instructions for an eye-
salve to be dripped into the eye through a flax-blue (“‘linhewenne’’) cloth. It is possible that the red
colour conveyed more power (mana of the anthropologists); indeed the symbolic significance of red
persists today. These instances underline the fact that although the study of anthocyanin

-

SHORT NOTES 355

polymorphism is neglected today, these red variants were observed and used by our remote
ancestors.

ACKOWLEDGMENTS

Information on various points was kindly supplied by Mr P. D. Sell, Prof. R. I. Page and Dr J. R.
Akeroyd; Mr Sell and Mr C. D. Preston commented on the text — my thanks to all.

REFERENCES

Bar ey, N. F. (1974). Old English colour classification: where do matters stand? Anglo-Saxon England 3: 15-
28

Bock, J. [Hieronymus Tragus] (1552). De Stirpium. Strasburg.

CaMERON, M. L. (1983a). The sources of medical knowledge in Anglo-Saxon England. Anglo-Saxon England
11: 135-155.

CAMERON, M. L. (1983b). Bald’s Leechbook: its sources and their use in its compilation. Anglo-Saxon England
12: 153-182.

CockayYnE, O., ed. (1865). Leechbook, in Leechdoms, wortcunning and starcraft of Early England 2: 1-360.
London.

CockaynE, O., ed. (1866). Lacnunga, in Leechdoms, wortcunning and starcraft of Early England 3: 1-80.
London.

Dopoens, R. [tr. H. Lyte] (1578). A niewe [nievve] herball. London.

Druce, G. C. (1920). Plant notes, etc., for 1919. Rep. botl Soc. Exch. Club. Br. Isl. 5: 575.

Fucus, L. (1542). De Historia Stirpium. Basle.

GERARDE, J. (1597). The Herball or Generall historie of plantes. London.

GERARDE, J. (1636). The Herball or Generall historie of plantes, 2nd ed., enlarged and emended by Thomas
Johnson. London.

GILBERT-CaRrTER, H. (1936). British trees and shrubs. Oxford. _

GraTTAN, J. H. G. & SINGER, C. (1952). Anglo-Saxon magic and medicine. London.

Gray, J. E. (1863). On two forms of plants growing under the same conditions. J. Bot., Lond. 1: 295-7.

Heal, G. (1926). Jilustrierte Flora von Mittel-Europa 5 (2). Munich.

Hea1, G. (1928). J/lustrierte Flora von Mittel-Europa 6 (2). Munich.

MEANEY, A. L. (1984). Variant versions of Old English medical remedies and the compilation of Bald’s
Leechbook. Anglo-Saxon England 13: 235-268.

SALMON, C. E. (1931). Flora of Surrey. London.

STEARN, W. T. (1973). Botanical Latin, 2nd ed. Newton Abbot.

TABERNAEMONTANUS, J. (1590). Eicones plantarum seu stirpium. Frankfurt am Main.

Q. C. B. CRonK
Corpus Christi College, Cambridge, CB2 1RH

DACTYLORHIZA TRAUNSTEINERI (SAUTER) SOO: VARIANTS IN NORTH-EAST
YORKSHIRE

Examples of albino D. traunsteineri (Sauter) So6 are apparently very rare in the British Isles but
records of single plants have been cited for two British localities (Roberts 1985). In June 1986 I
found three such plants growing within a large colony of typical D. traunsteineriin N. E. Yorks., v.c.
62. All the albino plants grew within 3 m of each other and had pure white flowers with unmarked
labella; the leaves were light green and also unmarked, whilst the upper stem and bracts lacked the
anthocyanin staining typical of this taxon. Other than as described above, they resembled normal D..
traunsteineri in habit, lax-flowered appearance and labellum shape and size, the plants being 10-14
cm tall, with 5-11 flowers, three or four sheathing leaves, one non-sheathing leaf, and with a
maximum leaf width of 8-5S—12 mm.

In the immediate area of these albino plants was a single plant of D. traunsteineri whose six
flowers were coloured an extremely pale, delicate pink, with markedly tri-lobed labella sparingly
spotted and loop-marked in bright red. This unusual and very attractive plant, whose flower colour

356 SHORT NOTES

lies well outside the normal range for D. traunsteineri, apart from having a greater maximum leaf
width, exhibited a marked resemblance to descriptions and illustrations of Orchis francis-drucei
recorded for W. Ross, v.c. 105, by Wilmott (1936) and Landwehr (1977), and recently interpreted as
D. traunsteineri (Lowe, Tennant & Kenneth 1986).

In a second colony of D. traunsteineri in N. E. Yorks., c. 4 km from the above, a single
pseudopeloric variant was found in June 1987. This had flowers whose labella were generally
completely undifferentiated, although in a few instances an extremely short (1-2 mm) vestigial spur
was present. All labella were narrow, sepal-like, and forward-pointing, of typical colour but
completely lacking any form of dot- or loop-markings. In each flower both lateral sepals were spread
horizontally (laterally), whilst the remainder were invariably arched forward in a closed configu-
ration. The plant was 30 cm tall and had 24 flowers, both these characters being much in excess of
those found for local D. traunsteineri. Pseudopeloria in British orchids was reviewed by Bateman
(1985) but appears to be previously unrecorded for D. traunsteineri.

Colour photographs of all the plants described are retained by myself.

REFERENCES

BATEMAN, R. M. (1985). Peloria and pseudopeloria in British orchids. Watsonia 15: 357-359.

Lowe, M. R., TENNANT, D. J. & KENNETH, A. G. (1986). The status of Orchis francis-drucei Wilmott. Watsonia
16: 178-80.

LANDWEHR, J. (1977). Wilde orchideeén van Europa. Amsterdam.

Roserts, R. H. (1985). Some unusual orchid variants from Anglesey. Watsonia 15: 275- TH:

Wivmotrt, A. J. (1936). New British marsh orchids. Proc. Linn. Soc. Lond. 148: 126-30.

Mey. oY: FOLEY
87 Ribchester Road, Clayton-le-Dale, Blackburn, BBI 9HT

THE BRITISH ROSE SURVEY OF 1952-54

The British Rose Survey was organized under the auspices of the Systematics Association by the late
Professor P. C. Sylvester-Bradley, a paleontologist at Sheffield University with a keen interest in
botany, in collaboration with Dr R. Melville of the Royal Botanic Gardens, Kew. Very little
information about the survey appears in print; it was the subject of an exhibit at the B.S.B.I’s
exhibition meeting of 1952 (Sylvester-Bradley 1953) in which the results of a trial run conducted that
summer and autumn were presented, and the strategy of the main survey for 1953 outlined. The
plan was to sample 30-50 bushes from randomly selected populations and to score each individual
for 38 characters relating to habit, foliage, armature, flowers and fruits (an example of a data-sheet
was illustrated in B.S.B.I. News (Gornall 1988)); the data were later to be encoded on punched
cards for statistical and taxonomic analysis. It was hoped that the results would “‘elucidate the nature
of [rose] ‘species’ and the degree of reticulation present” (Sylvester-Bradley 1953; Lousley 1953).
After the trial survey of 1952, extensive field work was carried out in 1953, involving 192 registered
observers, 64 of whom sent in returns that year relating to 2764 bushes in 96 localities. Work,
however, then rapidly tailed off over the following two seasons; the results were never analysed, and
the survey was more or less forgotten.

It came as some surprise, therefore, when a total cache of 941 pressed Rosa specimens were
exhumed from the basement of the herbarium at K and Mrs J. Sylvester-Bradley’s garage in
Leicester in the summer of 1987. These proved to represent an unknown proportion of the total
gathered during the dry run of 1952 and the main survey of 1953. There was also documentation and
a card index which connected the specimens with collectors and the localities visited.

A total of 213 individuals or organizations were registered as observers (many of them B.S.B.I.
members) and they apparently visited 140 localities altogether, although the surviving specimens
originate from only 54 of these; Fig. 1 shows the 10-km squares containing the survey sites.

Most of the specimens collected by Sylvester-Bradley had been identified but the rest (c. 800)
were without names. One of us (A.L.P.) has now made determinations where possible, although it
must be said that about half of the specimens were too poorly pressed, too poorly preserved, too

=“

SHORT NOTES 357

CHANNEL |SLANDS
O|WA
PLOTTED ON WV

UTM GRID

(@) geatiicile: fe 2 ie ae (@)
e 1 2 $} 4 5 6

Ficure 1. Distribution of 10-km squares containing one or more survey sites; ®@ with herbarium specimens, O
without herbarium specimens.

immature or otherwise inadequate to allow identification. A taxonomic summary is presented in
Table 1.

The Survey was, in our opinion, doomed to failure from the start partly because the organizers
had bitten off more than they could chew; their intention was to transfer the vast amount of data to
punched cards but this was never done. It is also clear from the surviving instructions and the printed
data-sheets that the field-work was time-consuming and uninteresting, and the B.S.B.I. distribution
maps scheme was beginning to compete successfully for the volunteer labour force.

The principal mistake which the organizers made was to confine the Survey to very small areas

358 SHORT NOTES

~

TABLE 1-TAXONOMIC SUMMARY OF THE DATA FROM THE BRITISH ROSE SURVEY

OF 1952-54
Species/Hybrid* No. specimens No. localities
R. afzeliana 36 5
R. afzeliana X canina 20 7
R. arvensis 101 12
R. arvensis X canina 1 1
R. caesia X canina a 2
R. canina 232 25
R. canina X obtusifolia 2 2
R. canina X stylosa 8 1
R. canina X tomentosa 2 1
R. dumetorum 10 4
R. micrantha 8 3
R. mollis 52 6
R. obtusifolia 5 5
R. rubiginosa 2 2
R. sherardii 22 3
R. stylosa 11 3
R. tomentosa 1 1
Indeterminate 418 43

* Parents of hybrids are given in alphabetical order.

containing a large stand of rose bushes. Such sites often contain only the commonest local taxon.
Thus the specimens from the island of Hoy in the Orkneys are nearly all R. mollis Sm.; collections
from the Midlands are often nothing else but R. canina L.; those from the south are often R. canina
or R. stylosa Desv. In all the hundreds of specimens there are only eight of R. micrantha Borrer ex
Sm.., five of R. obtusifolia Desv., two of R. rubiginosa L., one of R. tomentosa Sm. and none at all of
R. pimpinellifolia L., R. caesia Sm., R. agrestis Savi and R. elliptica Tausch (the occurrence of the
last of which in Britain is now considered to be doubtful) (Table 1). The supreme example of this
strategic mistake is the collection number 86/A. This was made by field-workers from the Juniper
Hall Field Centre. With classic rose sites in the vicinity, such as Box Hill or even the immediate
neighbourhood of Juniper Hall itself, the field-workers chose a ruderal site (the spoil-heap of a lime-
works) and collected 40 specimens of R. canina and nothing else. It would have been far more
informative from a taxonomic point of view if the field-workers had been instructed to examine a
few individuals in as many localities and habitats as possible.

Nevertheless, the data do confirm our knowledge of the geographical distributions of some Rosa
species. For example, all the specimens of R. mollis came from the north; all specimens of R. stylosa
came from the south; and those of R. arvensis decline in frequency the more northerly the locality.

We have examined all the specimens with care and discarded all those which, for various reasons,
are unidentifiable. The remainder have been divided into sets, with the top one housed at LTR; a
subsidiary set has been sent to NMW, together with the documentation, data-sheets and the
reference card index.

ACKNOWLEDGMENT
We should like to thank Mrs J. Sylvester-Bradley for bringing the specimens to our attention and for
assisting us in sorting through them and explaining the reference system.

REFERENCES

GornaLL, R. J. (1988). Specimens and data sheets from the British Rose Survey. B.S.B.]. News 48: 14-15.
Lous.ey, J. E.{1953). Study of the British flora during 1952. Nature 171: 335-337.

SHORT NOTES 359

SYLVESTER-BRADLEY, P. C. (1953). The British Rose Survey. Yb. Bot. Soc. Br. Isl. 1953: 66.

R. J. GORNALL & A. L. PRIMAVESI
Botany Department, The University, Leicester, LE] 7RH

A NEW VACCINIUM IN THE BRITISH FLORA?

A dwarf, prostrate, small-leaved Vaccinium uliginosum L. collected in 1968 by one of us (W.S.S.)
on Roga Field, Shetland (GR 411/3.8) has proved to have a chromosome number of 2n=39. Typical
V. uliginosum subsp. uliginosum from other British localities, e.g. Unity Bog, Cumbria (GR 35/
528.591) and Hermaness, Shetland (GR 412/606.176), has 2n=48 in agreement with other
European counts (Moore 1982). The more northerly subsp. microphyllum Lange from arctic and
sub-arctic Europe is reported as having 2n=24 (Moore 1982). It has never been recorded from the
British Isles. A hybrid between these two subspecies would be expected to have 2n=36.

Although the chromosomes are very small and difficult to count, consistent counts of 2n=39 have
been made on the Roga Field plant. There are three possible explanations. Firstly it might be a first
generation hybrid (F,) with accessory chromosomes (2n=36+3B), though no differences in size or
appearance could be detected in the chromosome set. Alternatively it might be a second (F>) or
third (F3) generation hybrid resulting from a backcross to the tetraploid parent (subsp. uliginosum).
Both diploids and tetraploids are probably self-incompatible (Vander Kloet 1977), so an F; is
unlikely to have resulted from a self-fertilization, and the chance of there being two F, hybrid plants
in the same area is probably low. Thirdly, as the Roga Field plant is similar in appearance to subsp.
microphyllum, it might be derived solely from the subspecies through non-reduction of some
chromosomes at meiosis in either pollen or ovule production.

We suggest, then, that the Roga Field area and other exposed summits and hillsides in the
Shetland [sles should be searched for small-leaved, prostrate, strongly rhizomatous plants of V.
uliginosum as they might prove to be the diploid subsp. microphyllum. The plant can easily be
propagated at any time of year from small pieces of twig (preferably young) with attached rhizome
wrapped in damp moss.

REFERENCES

Moore, D. M. (1982). Flora Europaea check-list and chromosome index, p. 155. Cambridge.
VANDER KLOET, S. P. (1977). Evidence against autogamy in Vaccinium uliginosum L. Proc. N. S. Inst. Sci. 28:
101-104.

H. A. MCALLISTER & W. S. STEWART
Liverpool University Botanic Garden, Ness, Neston, S. Wirral, L64 4AY

NARTHECIUM OSSIFRAGUM L. IN THE BURREN, COUNTY CLARE

Webb & Scannell (1983) provided distribution data for plants in the Burren, Co. Clare, v.c. H9, and
therein noted that Narthecium ossifragum L. occurred only in districts 1 (The Clare Shales) and 3
(The Burren Lowlands). District 2 (The Burren Hills) is not notable for extensive peat deposits yet,
enigmatically, it does not have an exclusively calcicolous flora; plants such as Calluna vulgaris L..,
Erica cinerea L. and Eriophorum spp. abound in the region, colonizing the small hummocks of peat
that rest on the limestone rocks, and the pockets of soil which are maintained in a perpetually damp
state solely by rain water. However, plants characteristic of the wettest parts of raised bogs — for
example Erica tetralix L. and Myrica gale L. — are absent from this region, or at least have not yet
been reported. Hitherto N. ossifragum was included among the absentees.

During field studies in The Burren Hills at the beginning of August 1987, I found N. ossifragum in
several places on the higher slopes of the western hills. The first sighting was near the summit of
Dobhach Bhrainin, at about 270 m, in a damp, peaty area facing north-west. About 100 flowering
spikes were visible in the colony which covered less than 100 m* (herbarium specimens collected 1

360 SHORT NOTES

August 1987 are in DBN). On the succeeding three days, other populations were noted on the north-
eastern slopes and summit plateau of Cappanawalla, and on the northern slopes of Gleninagh
Mountain; these colonies were substantial and much more extensive than the original one on
Dobhach Bhrainin.

What is the reason for the absence of earlier reports of this species in the Burren? It is not an
abundant plant, but it is certainly not rare — anyone walking on the higher slopes of the Burren hills
during the summer is likely to encounter it in blossom. Perhaps the silence of earlier botanists is
explained by the fact that the bright yellow flower-spikes of N. ossifragum can be mistaken, at a
distance, for those of Galium verum L. or Solidago virgaurea L.; even at close range there is a
confusing similarity between the inflorescences of N. ossifragum and the diminutive variant of S.
virgaurea that inhabits the Burren.

~

REFERENCE

Wess, D. A. & SCANNELL, M. J. P. (1983). Flora of Connemara and the Burren. Cambridge.

E. C. NELSON
National Botanic Gardens, Glasnevin, Dublin 9

NOTES ON THE ECOLOGY OF BLYSMUS RUFUS (HUDS.) LINK IN NORTH-EASTERN
FIFE

Blysmus rufus (Huds.) Link is one of the few species considered to be a northern element in the
British saltmarsh flora (Ratcliffe 1977; Adam 1978). In Scotland, B. rufus is found predominantly
on the western coast, having a more fragmented distribution in the east (Perring & Walters 1982). It
is therefore of interest that, during a coastal vegetation survey in Fife, v.c. 85 (Leach & Phillipson
1985), B. rufus was found to occur on 15 saltmarshes. The following observations were part of an
investigation of the factors controlling the distribution of B. rufus in Fife (Penford 1982).

In Fife, B. rufus occurs on grazed beach-head saltmarshes, usually on raised beaches with a
gravelly substrate. Stands are small (less than 10 m*), and are found along the strand line at the
transition zone between base-rich fen communities and Juncus gerardii-dominated saltmarsh. B.
rufus often occurs as mono-dominant stands, although associates may include Eleocharis quinque-
flora (occasionally co-dominant), Triglochin maritima, T. palustris, Agrostis stolonifera, Festuca
rubra, Glaux maritima and Plantago maritima. Its confinement to the top of saltmarshes suggests
that B. rufus is an obligate halophyte, but is unable to withstand prolonged exposure to strongly
saline conditions. Soil analyses of sodium along transects (Table 1) confirmed that salinity increased

TABLE 1. CALCIUM AND SODIUM CONCENTRATIONS IN SOIL SAMPLES FROM THREE POINTS
ALONG SALTMARSH TRANSECTS 8 M LONG
Figures are means + s.e.; n = sample size

Transect point

Fen zone Blysmus zone Saltmarsh zone
Om 2m 8m
[Na] ppm 10114284 1082+213 4703+1948
n=7 n=9 n=8
[Ca] log,oppm 4-03+0-14 4-16+0-15 3-69+0-21

SHORT NOTES 361

sharply immediately to the seaward side of B. rufus stands (t=1-85, p<0-10) and decreased less
sharply from the B. rufus zone to the glycophytic fen zone further up the marsh (t=0-20, n.s.).

The proximity of B. rufus stands to fen communities containing predominantly basiphilic species,
including Carex otrubae, C. hostiana, C. disticha, Glyceria maxima, Lychnis flos-cuculi, Caltha
palustris and Equisetum arvense, suggests that B. rufus is subject to base-rich flushing. Soil analyses
(Table 1) showed that mean levels of calcium increased slightly between the fen zone and the B.
rufus zone (t=0-61, n.s.) and decreased sharply (t= 1-83, p<0-05) where more markedly halophytic
vegetation became dominant. Thus the seaward boundary of B. rufus stands appeared to be
influenced by a pronounced increase in salinity and a decrease in calcium.

The influence of calcium on the distribution of B. rufus in Fife may be significant due to the ability
of calcium to ameliorate the effects of salinity. This occurs by affecting the selective ion transport
mechanism at the root cells (Jennings 1976) and also by reducing the availability of sodium in cation
exchange within soils (Waisel 1972). In this way, Ranwell (1972) suggests that high levels of calcium
- could exert a profound effect on the species composition of saltmarsh vegetation.

The similiarities in species composition between the beach-head saltmarshes of Fife that support
B. rufus and the marshes of the western Scottish sea lochs, where it is more widespread, has been
remarked on by Leach & Phillipson (1985). It is likely that, in Fife, B. rufus is confined to sites
where a combination of factors produce conditions similar to those where this species commonly
occurs. Low salinity is a feature of sea-loch marshes on the western coast and, whilst salinity is likely
to be higher on Fife saltmarshes, the influence of base-rich flushing, coupled with a gravelly
substrate to reduce the retention of sodium, may enable B. rufus to persist in localized habitats
within an otherwise unsuitable area.

ACKNOWLEDGMENTS

Thanks are due to Dr A. D. Q. Agnew and Dr Ian Francis for help with the analysis. I am
particularly grateful to Simon Leach for suggesting this study and for commenting on the
manuscript.

REFERENCES

Apa, P. (1978). Geographical variation in British saltmarsh vegetation. J. Ecol. 66: 339-366.

JENNINGS, D. H. (1976). The effects of sodium chloride on higher plants. Biol Rev. 51: 453-486.

Leacu, S. J. & PHILLIPSON, P. H. (1985). The saltmarsh and brackish swamp vegetation of the Fife peninsula.
Trans. Proc. bot. Soc. Edinb. 44: 357-373.

PENFORD, N. (1982). Factors affecting the status and distribution of Blysmus rufus (Huds.) Link in north-east Fife.
B.Sc thesis, University College of Wales, Aberystwyth.

PERRING, F. H. & WALTERS, S. M., eds (1982). Atlas of the British flora, 3rd ed. Wakefield.

RANWELL, D. S. (1972). Ecology of saltmarshes and sand dunes. London.

RATCLIFFE, D. (1977). A nature conservation review. London.

WAISEL, Y. (1972). Biology of halophytes. London.

N. PENFORD
Rural Surveys Research Unit, Department of Geography, University College of Wales, Aberystwyth,
Dyfed, SY23 3DB

TYPIFICATION OF POTAMOGETON SPARGANIFOLIUS LAEST. EX FR.
AND P. NATANS SUBSP. KIRKIT HOOKER FIL.

The hybrid between Potamogeton natans L. and P. gramineus L. is relatively frequent in northern
Europe and is widespread, but very local, in the British Isles (Dandy 1975; Perring & Sell 1968). It is
a variable plant, usually resembling P. natans but with laminar rather than phyllodal submersed
leaves. The correct name for the hybrid is believed to be P. x sparganifolius Laest. ex Fr.; a later

362 SHORT NOTES 2

synonym is P. X kirkii (Hooker fil.) Syme ex Hooker fil., based on P. natans subsp. kirkii Hooker
fil. The purpose of this note is to typify these names.

Potamogeton sparganifolius Laest. ex Fr., Novit. fl. suec. mant. prima 9 (1832). Holotype: Fluv.
Torn. Muonionissa, L. L. Laestadius (UPS).

P. sparganifolius was described by Fries (1832) on the basis of material collected by L. L. Laestadius
“in fluviis profundissimis ad Karesuando Lapponiae’’. The epithet sparganifolius had been applied
to the plant by Laestadius, but not published by him. There is in Fries’ herbarium (UPS) a specimen
labelled ““Potamogeton sparganifolius Laest. Fluv. Torn. Muonionissa L. L. Laestad.”’, i.e.
collected from the R. Tornealven at Muonio, 75 km south-east of Karesuando. There is no material
annotated by Fries at S, where Laestadius’ herbarium is deposited (F. Bjérkback, in litt. 1988). The
specimen at UPS therefore appears to be the holotype of P. sparganifolius, and I have labelled it as
such. If further type material is ever located, the specimen at UPS would almost certainly be the
most appropriate choice of lectotype. It conforms to Fries’ description of P. sparganifolius and with
the current concept of the hybrid P. natans X gramineus.

Potamogeton natans subsp. kirkii Hooker fil., Student Fl. Brit. Isl. 371 (1870). Lectotype:
Ballinabrack River, Lough Corrib, 7 September 1853, T. Kirk (herb. W. Borrer, K), designated
here.

P. kirkii Syme in Sowerby, Engl. bot., 3rd ed., 9: 31 (1869), nom. synon.

P. polygonifolius var. kirkii (Hooker fil Hi: Watson, Lond. cat., 7th ed., 21 (1874).

P. kirkii (Hooker fil.) Syme ex Hooker fil., Student Fl. Br. Isl., 3rd ed. 535 (1884). .

P. X sparganifolius forma kirkii (Hooker fil.) Hagstr. ex Pearsall in Rep. botl Soc. Exch. Club Br.
Isl. 11: 186 (1936).

In September 1853 Thomas Kirk collected an interesting Potamogeton from “deep water by the
bridge against Maam Hotel, Ballinabrack River, Lough Corrib, Galway’. He identified it as P.
longifolius Gay (a taxon now regarded as synonymous with P. lucens L.) and distributed numerous
specimens under that name. Amongst those who received these were W. Borrer and C. C.
Babington. Borrer (1854) published the record of P. longifolius but Babington realized that this was
a misidentification. He was able to compare Kirk’s plant with authentic material of P. sparganifolius
which he had been sent by Fries himself. By January 1854 he had concluded that Kirk’s collection
was referable to P. sparganifolius, as the two plants agreed “‘in every particular that the respective
specimens afford the opportunity of contrasting” (Babington to Borrer, letter in herb. Borrer, K).
He included it under this name in all subsequent editions of his Manual of British Botany, tentatively
in the 4th edition (Babington 1856) but emphatically from the 5th edition (Babington 1862)
onwards. Babington’s opinion has been supported by recent students of the genus, including
Hagstr6m (1916), Pearsall (1931) and Dandy (1975).

Babington’s contemporaries, however, did not accept that Kirk’s plant was P. sparganifolius.
J. T. Syme gave an illustrated account of it in his edition of English Botany (Sowerby 1869). In
discussing it he specifically mentioned the only flowering specimen he had seen — a specimen in
Borrer’s herbarium. He was uncertain of the identity of Kirk’s plant, concluding “I have seen too
little of the Irish plant to venture to affirm it to be specifically distinct from P. polygonifolius, so that,
though convinced it is not the P. sparganifolius of Lastadius, I keep it under this name for the
present,.though much tempted to designate it P. Kirkii, after its discoverer’. He also included the
name “‘P. Kirkii, mihi, MS” in this synonymy of the species. This name was taken up by J. D.
Hooker of Kew, who validly published it as P. natans subsp. kirkii in his Student’s Flora (1870). In
the third edition of this Flora, Hooker (1884) raised it to specific rank. It was described and
illustrated as P. kirkii in Fryer & Bennett’s (1915) monograph The Potamogetons (Pond Weeds) of
the British Isles.

As far as I am aware there has been no attempt to lectotypify the name P. natans subsp. kirkii
Hooker fil. There is no material in the general herbarium at Kew which could be selected as a
lectotype. However, the three sheets in the herbarium of W. Borrer of the plant distributed as P.
longifolius by Kirk would have been available to Hooker, as Borrer’s herbarium was bequeathed to
Kew in 1862. Although Hooker did not annotate the sheets, which had been labelled as P. kirkii by

SHORT NOTES 363

Borrer, there can be no doubt that he saw them. In the preface to the Student’s Flora Hooker (1870)
emphasized that his descriptions were derived from specimens rather than copied from books, and
we know he used Borrer’s herbarium in writing the account of Potamogeton as a specimen in it is
cited under P. obtusifolius. As the flowering plant of P. sparganifolius in Borrer’s collection was
mentioned twice by Syme it is reasonable to conclude that Hooker would have examined it. I have
therefore selected this sheet as the lectotype. The flowering plant drawn in English Botany (t. 1903)
matches the specimen and is clearly based upon it. The detached floating leaves illustrated must,
however, have come from another specimen.

P. X sparganifolius is a variable hybrid, variation occurring both within and between populations.
The extent to which this variation is genetic rather than phenotypic has not been tested by
cultivation experiments. I do not see any reason to recognize individual populations, such as that at
Maam, as distinct taxa, and therefore support the reduction of P. kirkii to a synonym of P. xX

sparganifolius.

ACKNOWLEDGMENTS

I am grateful to R. Moberg (UPS) and F. Bjérkback (S) for searching for type material of P. x
sparganifolius, to Lyn Nichol (CGE) for arranging the loan of the holotype, to D. A. Simpson for
help during visits to Kew and to P. D. Sell for comments on a draft of the manuscript.

REFERENCES

BaBINGTON, C. C. (1856). Manual of British botany, 4th ed., p. 351. London.

BaBINGTON, C. C. (1862). Manual of British botany, 5th ed., p. 352. London.

Borrer, W. (1854). Notes on the ‘Cybele Britannica’, Volume III. The Phytologist 5: 44-50.

Danpy, J. E. (1975). Potamogeton L., in Stace, C. A., ed. Hybridization and the flora of the British Isles, pp.
444-459. London.

Fries, E. M. (1832). Novitiarum florae suecicae mantissa prima. Lund.

Fryer, A. & BENNETT, A. (1915). The Potamogetons (Pond Weeds) of the British Isles, pp. 16-18, tt. 8-9.

Hacstr6M, J. O. (1916). Critical researches on the Potamogetons. K. svenska VetenskAkad. Handl., n.s., 55
(5): 1-281.

Hooker, J. D. (1870). The Student’s Flora of the British Islands, p. 371. London.

Hooker, J. D. (1884). The Student’s Flora of the British Islands, 3rd ed., p. 535. London.

PEARSALL, W. H. (1931). Notes on Potamogeton. II. The larger British species. Rep. botl Soc. Exch. Club Br. Isl.
9: 380-415.

PERRING, F. H. & SELL, P. D., eds (1968). Critical supplement to the Atlas of the British flora, p. 137. London.

Sowerby, J. (1869). English Botany, 3rd ed., 9: 31-32, t. 1903. London.

C. D. PRESTON
Institute of Terrestrial Ecology, Monks Wood Experimental Station, Abbots Ripton, Huntingdon,
PE17 2LS

EARLIEST RECORDS FOR TWO x FESTULPIA COMBINATIONS

While checking herbarium specimens in folders labelled Festuca rubra we have recently come across
two specimens of X Festulpia that in each case constitute the first known record anywhere of that
particular hybrid combination. Both are also new vice-county records. :
1. Festuca rubra L. agg. X Vulpia bromoides (L.) S. F. Gray. Ashley Hill, W. Gloucs., v.c. 34.,
May 1928, C. Alden, ex herb. Gibbons & Bell (LTR). The plant was labelled ‘‘F. rubra var.
grandiflora ?viviparous’’. Several of the spikelets are indeed proliferating, a feature we have often
encountered in X Festulpia specimens and in other hybrid Festuca plants.

The measurements of this plant fall partly outside the ranges given by Ainscough et al. (1986), as
shown in Table 1; in general the parts are somewhat larger, but the glume ratio is not exceptional.
2. Festuca rubra L. agg. X Vulpia myuros (L.) C. C. Gmelin. By railway line at Mitchell Troy, 2

364 SHORT NOTES ~

TABLE 1. DIAGNOSTIC MEASUREMENTS OF TWO x FESTULPIA SPECIMENS

F. rubra agg. X F. rubra agg. X
V. bromoides, v.c. 34 V. myuros, v.c. 35
Lower glume length (mm) 3-0—4-0 1-6—2-1
Upper glume length (mm) 5-5—7-2 3-7-4-7
Glume ratio 0-5—0-58 0-47-0-64
Lemma length (mm) 6-2-7-5 4-2-5-0
Awn length (mm) 1-7-4-0 3-7-5-9
Anther length (mm) 1-1—2-2 1-0-1-4

~

miles S.W. of Monmouth, Mons., v.c. 35., 16th June 1951, E. Nelmes 973 (K). The plant was
determined by C. E. Hubbard (undated) as “F. rubra ssp. commutata” (= F. nigrescens).

In this case the measurements are typical of those given by Ainscough et al. (1986) (Table 1).

Neither of the specimens shows any sign of rhizomes; the Festuca parent might be F. rubra subsp.
rubra or F. nigrescens Lam. in each case. In both cases virtually all the pollen is empty, and the
anthers indehiscent.

The earliest record of F. rubra Xx V. fasciculata known to us is the specimen collected by J.
Gosselin in Guernsey, probably between 1788 and 1791 (STPCM). However, there is also an
undated specimen in BM, collected as Festuca cambrica Hudson and cultivated in William Curtis’s
garden (Curtis died in 1799), that might pre-date it.

REFERENCE

ArnscouGH, M. M., BarKER, C. M. & Stace, C. A. (1986). Natural hybrids between Festuca and species of
Vulpia section Vulpia. Watsonia 16: 143-151.

C. A. Stace & A.-K.K.A. AL-BERMANI
Department of Botany, The University, Leicester, LE] 7RH

CRATAEGUS x MEDIA BECHST. IN MIDDLESEX HEDGEROWS

Studies of hedgerow Crataegus in Middlesex, v.c. 21, suggest that Crataegus X media Bechst. (C.
laevigata X monogyna) may be more common than is suggested in some local Floras. A survey of
surviving agricultural hedgerows in the London Borough of Brent (Kingsbury, Wembley and
Willesden) and a small adjacent part of the London Borough of Barnet was undertaken from 1983 to
1987, using a method for locating hedgerows discussed by Williams & McLauchlin (in press). The
study area was between approximately 7 and 18 km north-west of central London and covered an
area of about 46 km’, 45 of which were accounted for by Brent. Much of Brent was still farmland
until the suburban expansion of London in the 1920—30s. Only two enclaves of countryside remain,
Fryent Country Park and, to a lesser extent, the area around the Welsh Harp Reservoir. In the
south, much of Willesden could be described as inner-city.

For the main part of the study, Crataegus specimens were identified in the field, using the main
lateral vein curvature and leaf shape as the chief identification criteria (Byatt 1975, 1976; Williams
1986). (Slight differences in stone shape were noted between the species: C. monogyna had a stone
with a pointed apex and C. /aevigata had blunter stones, but this may simply be due to species
differences in style and stone numbers). The survey found 548 former hedgerows or remnants (520
in Brent) and Crataegus was present in 257 (47%) of these. C. monogyna occurred in 213 hedges, C.
x media in 134 and C. laevigata in 18 hedges. A few Crataegus in 13 of the hedges remained
unidentified due to access or other survey problems. At Fryent Country Park, a hedgerow planted in

SHORT NOTES 365

recent decades contained C. monogyna only, older hedges had both this species and C. X media,
while some hedges planted before c. 1547 and others originating from assarts (woodland clearance)
contained all three.

The survey area fell within 21 tetrads, though only four of these were fully within the study area.
Comparisons were made with the flora of London (Burton 1983) which included Crataegus from all
habitats and obviously from other Boroughs in the peripheral Brent tetrads (Table 1). Though the

TABLE 1. CRATAEGUS RECORDS FROM 21 TETRADS IN MIDDLESEX

¢

Brent hedgerow survey Burton (1983)
Crataegus monogyna 18 19
Crataegus X media 14 3
Crataegus laevigata 9 10

Brent study was only recording Crataegus from former agricultural hedges, C. monogyna and C.
laevigata were found in a similar number of tetrads to those in Burton (1983), but for C. laevigata the
two studies only agreed in five tetrads. C. X media, however, was found to be much more
widespread than Burton (1983) suggested. Forty-three of the 66 monads (1-km squares) partly or
wholly within the study area had hedgerow Crataegus. Forty-one of these had C. monogyna, 14 had
C. laevigata and 36 had C. X media. While some of the discrepancies between the two studies may
be due to the more comprehensive search of the hedgerow study, it is suspected that C. X media is
under-recorded in the London flora. C. X media may be unfamiliar to many recorders or easily
confused (Jermyn 1974) and is probably commoner than local Floras indicate. As explained by
Bradshaw (1975), C. X media has a range of characteristics intermediate between those of the two
parents, and such is the degree of hybridization that Byatt (1975) considered the concept of the two
species C. monogyna and C. laevigata to be largely irrelevant in south-eastern England. Main lateral
vein curvature, when used with other characters, does, however, appear to be a useful aid for the
identification of Crataegus in the field and may show C. X media to be more widespread in other
counties.

ACKNOWLEDGMENTS

I am grateful to Prof. A. D. Bradshaw, R. M. Burton and A. P. Williams for helpful comments
during the course of this study.

REFERENCES

BrapDsHaw, A. D. (1975). Crataegus L., in Stace, C. A., ed. Hybridization and the flora of the British Isles.
London.

Burton, R. M. (1983). Flora of the London area. London.

Byatt, J. I. (1975). Hybridization between Crataegus monogyna Jacq. and C. laevigata (Poiret) DC. in south-
eastern England. Watsonia 10: 253-264. cs

Byatt, J. I. (1976). The structure of some Crataegus populations in north-eastern France and south-eastern
Belgium. Watsonia 11: 105-115.

JERMYN, S. T. (1974). Flora of Essex. Colchester.

WILuias, L. R. (1986). A Hawthorn identification guide for the local historian. Wembley Hist. Soc. J. 6: 38-46.

Wiiuiams, L. R. & McLAUCHLIN, J. (in press). A method for surveying old hedgerows in suburban areas using
standard trees. Local Historian.

L. R. WILLIAMS
Brent Leisure Services, London Borough of Brent, Brent House, High Road, Wembley, Middlesex,
HAY 6SX

= ee oa

2 ae .

“— S58 bate ad? a:
7 =!
ie

=~ ee

- ¥ “er ‘ 2 : Oe ~ oan po
s i, via ; _-_ : Tt 4 x Ma Flee Sav

& Set m4 \ oi on
28) Spd 1 (toed ete

)
hi ‘fig > tir o=arv 1 ie
sa oe 11, 4 aoe

~

wey 2 7 “> as 2
C1! - 4 : x
> . nF
! pit Ew 1Caibe b
pa Ae) i ~ ye
: anton
(PROT) ns es “ee
- i . - 7 -
: — hal —— ig : re os -
Soe =
a Rr 7 Py
d i 7 —
; 8 a . - a
= ' =
} = ; 7
= " -
7 4 + ¢ nie
———— = - — ~ - >_S a a ee me le
- 4 : 4 tox “ .? va :
* é oth AP Py ee, tt
. -~
al b= - °
> —n . o 4 ar é _ coe ak
Ji Ve + TEs oS mates OP TCR
: r; . ¢ o< t= > j
[ 5 ' .3 ne iif. a% PT ,
L a = a, 4 ' & ioe 7 a.
le wy sls é vie} ad ow CLiGS3 ‘ vit
> r = *y 7 bs = = es
“3s “—- “ , Paris VAT Gr
be ¥ 2 3 ah mS > ~ MAS f $ Ss oe Se izou ie te. s
~- —-_ - ~ Ss
ed ~~ J*et. ere ae =
' "~ ? oY » {* =) 6 visits eat iat
| 2 ew 4 airs a ui TF 4) oe!
pe “4 — a ‘ “ ¢
j « ~* ot * x ed Snes , ,, t 4
) 37u) Sow? 4 Pie iste 240 *) aubue Siw
— . Oe 4 —
F - iG a <
4 “ 4 + my ; os eae ho B
z 5) = } 2 Lod 2 ‘ io" Feel ans 2 Yn 4 ae
- r “ a : : oa
e i 4 on F an, e “ x
— - 4 4 + ¥ ~ ~ ® ~%2 we * - i ws r ’ Py
s af, F> ¥ as ‘3 7e) i “$4 ‘ ' » i 2 m, «i 2 te Gh Seek 5
: = = i oe & aa wade Shoe .
. 4 . “ . : s - it | ‘ es
; - , 24 ¢ 3 ' 5 l it ‘ HOGG. Jf .0y t “4
- ‘ 4 Ta oa =) . = 7 7 i “+
« v . *¢% ae d - . a ¥ ° ¢ — Coyne ¢ - <a + » * Le 2
-<5° £9 "3 ; -2uhensinoure te semss & eee aetna
- 2” 7 ' a A ¥
i coment CUTS teu sl 6 fentbinidvtad ase
_ +2 , ee oe es ; ; SEP eia a | Sa bn
a Ps, . os >= col ae
f a6 2% - OM. de ‘
. 2 } Co Bis
- 4 _ - ear ia
; ~ - +i fy
« -
w¥7 fad o> POR j
= ¢ ase bw
~
sstiicn sy 20
a x oe
4 ! 4 L a —_ -
= : i
- é
+ P 2 : % vhs
4 ; i .
e ~ \ 2
= ?
= .
- ;7 t >
: ' pz
- . 3 7
7 7 ‘ = . ol ae
a aa’s
is a 4 Lars
U ¢ Pe ~~ o's
a i> ah i‘ + ja Ws oe 4
- - ‘ - ad “J
= 4 ,

; rt : qi pant atl) vot abr Poste weids shen wong
té Ain? ge) ty) Sipe Wit Bago A. 1 emp I. oe eed
| . aller Eas PE OT AREA Missee

BAI - € ‘ 7, %
> y > ’ . c r
% 23 geet | ing ¥, LD ni vat j Oh.
iY fom ; at
‘ 4 “ hr. ae ~ale
i v ; od i

_*

ih ~ v oe i? sand ¥
= -
Lf -

Watsonia, 17, 367-378 (1988) 367

Book Reviews

Mordecai Cubbitt Cooke. Victorian naturalist, mycologist, teacher and eccentric. M. P. English. Pp.
xviii + 357, with numerous illustrations in black & white. Biopress Ltd., The Orchard, Clanage
Road, Bristol BS3 2JX. 1987. Price £18.00 (ISBN 0—948-—737-02-6).

It is a shame this book touches too marginally on the vascular plants of the British Isles to warrant
more than a brief review in these pages, for it must surely be one of the best biographies of a botanist
ever written. The product of an immense amount of research, taken to the point of tracing
descendants in many parts of the world (and even interviewing one of them in Kenya), it is at the
same time highly readable.

Socially unsure, tactlessly over-forthright, a crusty reactionary in a bushy red beard, ‘M.C.’ — as
he liked to be called (he abominated his first name of Mordecai) — was one of those archetypal
Victorian figures of almost superhuman energy who hardly seem to have stopped writing for a
minute. Author of innumerable popular books and articles, some so ephemeral that even the
indefatigable Dr English has been unable to run them to earth, he was driven all his life by a
missionary fervour to spread to one and all the gospel of natural history. He was employed initially
as an elementary schoolteacher and then, in a succession of humble posts sadly beneath his abilities,
in the exhibiting of plant products in museums. He early found that his real love was mycology, a
subject on which he also published prolifically, in which he was entirely self-taught and in which he
achieved a place second only in Britain to the revered and incomparable Berkeley. In addition, he
was the founder, in 1862, of the short-lived Society of Amateur Botanists, which met in Hardwicke’s
bookshop in Piccadilly and which included Trimen, Dyer, Newbould and Britten among its
members, with the trend-setting Flora of Middlesex as its indirect product. The still-flourishing
Quekett Microscopical Club is the offspring of that body. Hardwicke and Cooke subsequently
strengthened their partnership with that fascinating miscellany, Science Gossip, of which Cooke
acted as editor for its first seven years.

Not content with all that, Cooke also coped in secret with a family life of stunning bizarreness and
complexity. With some brilliant detective work Dr English has uncovered the extraordinary fact
that his children were not his wife’s, as the world was led to suppose — but, most unexpectedly, her
grandchildren. Cooke turns out to have had a series of illegitimate children by his stepdaughter and
for long stretches of his life dwelt in what seems to have been a perfectly amicable ménage a trois.
Truly, the more we learn of the Victorians, the more amazing they become.

D. E. ALLEN

Aroids. D. Bown. Pp. 266, with 24 colour plates and 44 line illustrations. Century Hutchinson Ltd.,
London. 1988. Price £25.00 (ISBN 0—7126-1822-8).

The Araceae are, without doubt, amongst the most commonly cultivated indoor plants, so it is
surprising that there has been until recently a lack of books written for the non-specialist. Deni
Bown’s book more than adequately fills the need for an introduction to the family.

The book is divided into chapters that lead the reader from the elementary aspects of aroid
identification through pollination, ecology, morphology, edible aroids and medicinal uses. To cover
so many aspects of the family may seem daunting, but the way in which the information is presented
is so readable that even most complex aspects are clear.

The Araceae are introduced in ‘Variations on a theme’. The distinguishing floral features,
vegetative architecture, distribution, position relative to other plant families and evolution are all

368 BOOK REVIEWS x,

covered in great detail. Here also habitats are briefly discussed as a precursor to the later chapters
devoted to them. With these basics covered the text moves on to look in detail at individual aspects
of the family.

A full account of the pollination biology of the Araceae could easily fill several volumes, and so
the chapter concerned with reproduction (‘Of tails and traps and the underworld’) is, of necessity,
simplified. However, this in no way implies that it is superficial; in fact, every pollination mechanism
employed by the family is covered in detail.

Most of the book is concerned with the varied habitats in which aroids are found. The chapter
titles give the setting (e.g. ‘Woodlanders’, ‘Aquatics and amphibians’, ‘A place in the sun’, ‘In the
shadows’ and “Towards the light’), and the text describes the aroid inhabitants by genus (e.g.
Arisaema, Cryptocoryne, Biarum, Aglaonema and Monstera respectively). Within these chapters
almost all the major genera are covered, together with many poorly known taxa.

The giant tuberous aroids are allotted their own chapter, appropriately entitled ‘The Titans’. The
greater part of the text is devoted to the Sumatran Amorphophallus titanum, a plant which has
fascinated both the public and the biologist alike since its discovery over a century ago. However,
this is by no means the only giant in the Araceae, and I am pleased to see that Dracontium ane
Anchomanes, which both contain very substantial species, are not overlooked.

The chapters concerned with the edible aroids (‘An acquired taste’) and chemistry and medicinal
uses (‘Acids and crystals’) contain a wealth of information, much of it drawn from scattered sources
and skillfully synthesized into a more readily accessible form. As much has been gathered from
obscure sources, this volume provides information that would otherwise be difficult to find.

Aroids is generously illustrated with superb photographs taken by the author as well as numerous
line drawings, although I feel that the latter could have been of higher quality. The book ends with a
very useful ‘Check list of Aroid genera’, selected bibliography, detailed glossary and an exhaustive
reference section. This book, with its wealth of information and beautiful illustrations, will be an
essential part of a plant-lover’s library.

P. Boyce

Wild flowers of Majorca, Minorca and Ibiza — with keys to the flora of the Balearic Islands. Elspeth
Becket. Pp. 221 with 8 colour plates. A. A. Balkema, Rotterdam and Brookfield. 1988. Price £28.50
(ISBN 906-—191-—634-8).

There can be little doubt that there is ample room for a concise guide to the flora of the Balearic
Islands that goes beyond basic picture books to provide a satisfying resource for the serious
naturalist. When I first visited Majorca some 35 years ago, friends responded with puzzled
comments like ‘‘Where on earth is that”? However, since then, with the immense development of
cheap package holidays, countless botanists must have cut their teeth on the Mediterranean flora
through visits to these islands. Does this book then meet the needs of the present-day botanical
tourist? It is essentially a compilation of keys based very closely on those in Flora Europaea, and, as
such, it should provide an effective means for identification, at least for those reasonably
experienced in the use of Floras. The keys are clearly set out, but contain redundant abbreviations,
e.g. Annl,.bnnl and usu, which are irritating and save no space at all. Distributions are indicated by
Ma, Mi and I for the three large islands, generally without any further information. Each family has
an introductory section describing its features, sometimes accompanied by supplementary matter of
general interest. Vernacular names are given in English, German, French, Spanish and Catalan.
Eight composite colour plates are provided which may help beginners, who will probably find the
keys hard going, but the large number of species depicted (71 on Plate 5) inevitably means that their
utility is rather limited. A general introduction explains the use of keys and very properly urges the
need for the conservation of this flora, which is under considerable pressure as a result of the tourist
trade. The section on classification betrays such a state of mental confusion about the basics of
nomenclature that we can only be heartily thankful that the text rests so firmly on the foundation of
Flora Europaea. What then is lacking? Some introductory information on habitats and biogeogra-

BOOK REVIEWS 369

phy would have been most welcome, especially with regard to the very interesting endemic element.
The space devoted to vernacular names would have been better allocated to more detailed
distributional data, as there are hardly any indications as to whether species are abundant on every
roadside or restricted to a few remote localities. The book is elegantly presented with high quality
printing, but a clearer typographical distinction between family and generic names would have been
an improvement.

I wish I could feel more enthusiastic about this book, which does seek to meet a real need. At
£28.50 it is very expensive for the information it contains, and it would have been much better if its
de luxe format had been bypassed in favour of a cheap, compact field guide that would have fitted
into the pocket and cost half the price or less.

J. F. M. CANNON

Key works to the fauna and flora of the British Isles and north-western Europe. Edited by R. W. Sims,
P. Freeman & D. L. Hawksworth. Fifth edition. Pp. xii + 312. The Systematics Association Special
Volume No. 33, Oxford. 1988. Price £35.00 (ISBN 0—19-—857706-0).

This work provides references to books and papers which are considered key works for identifying
living organisms in the British Isles and north-western Europe. Apart from a great deal of updating,
it differs conspicuously from the fourth edition (1978) in having a new and admirably clear layout,
each entry starting with the title on the left and ending with the author’s name in bold on the right. It
is arranged taxonomically, with no index, and under, for example, Spermatophyta, the entries are
arranged under countries in a Geographic section (‘Britain’ includes Ireland), under Trees and
Shrubs, Water Plants, Sedges and Grasses, and Orchids in a Monographs of Special Groups section,
and under Bibliography, History, Glossary, Anatomy and Morphology, and Pollen and Spores
Identification. It is this part, and the Pteridophyta, that should be of chief concern to Watsonia
readers, but it must be said that the flowering plants are poorly served in comparison with most of
the other groups for which I am familiar with the literature.

A number of entries are very out-of-date. For example, W. Keble Martin, The concise British
flora in colour (1969) is given rather than The new concise British flora (1982), J. Lid, Norsk og
Svensk flora ed. 4 (1963) is given rather than Norsk, Svensk, Finsk flora ed. 5 (1985), and W.
Rothmaler, Excursionsflora ed. 3 (1962-63) is given rather than ed. 12 (1984). No Floras of either
Spitsbergen or the Faeroes are included. It would have been more helpful to give A. Love, Flora of
Iceland (1983) in English rather than the Icelandic equivalent /slenzk — Ferdafléra ed. 2 (1977,
erroneously given as 1983).

The taxonomic coverage of flowering plants is very uneven and poor in comparison with that for,
say, bryophytes or many animal groups. Although P. F. Yeo’s 1978 revision of Euphrasia and A. J.
Richards’s Taraxacum Flora of 1972 are included, there is nothing at all on Hieracium, and C. A.
Stace & M. Ingrouille on Limonium are omitted. W. C. R. Watson, Handbook of the Rubi (1958) is
included, but not H. E. Weber’s works on the genus which are superior in almost all respects. No
choice of Floras and field guides will suit everyone, but it is difficult to see why J. D. Hooker, The
student’s flora of the British Isles ed. 3 (1937) and E. Step; Wayside and woodland ferns new ed.
(1949) are included; admirable though they once were, they have now been so totally superseded
that they are positively misleading. The inclusion of G. C. Druce, Hayward’s botanist’s pocket-book
ed. 19 (1948) is slightly more justifiable for its wealth of infraspecific taxa. It is a pity that works on
galls seem to have been excluded, except for incidental mentions under some of the causers, and
there are no works on seeds or seedlings although pollen and spores are well covered.

There are numerous errors in titles. The B.S.B.I. Handbook Sedges of the British Isles ed. 2
(1982) confusingly retains the title of ed. 1. Although in the headings, names such as Orchids are
given their Latin equivalents, the Docks and Knotweeds Handbook lacks them which is unhelpful,
at least to non-English users.

It is much to be hoped that in the next edition the section on flowering plants will be brought up to
the standard of the rest of the book. It is unlikely to be of much use to B.S.B.I. members for this

370 BOOK REVIEWS Z

section, but (and this is not, I think, just the reviewer’s perspective) it can be warmly recommended
to members wishing to extend their knowledge to other groups of plants and animals.

A. O. CHATER

Wildlife conservation in churchyards. Pp. 16. Norfolk Naturalists Trust and Diocese of Norwich,
Norwich. 1988. Price £3.00 incl. p. & p. Obtainable from Norfolk Naturalists Trust, 72 Cathedral
Close, Norwich, NR1 4DF.

The increasing awareness in recent years of the importance of churchyards in conservation,
stimulated to a large extent by the B.S.B.I. and R.S.N.C. churchyards survey, has led to the
production of a number of booklets on the management of churchyards. Most of these have been
done on a county basis, and the one under review, arising out of the Norfolk Naturalists Trust
Churchyard Conservation Scheme started in 1981, is one of the most practical, detailed and helpful
that I have seen. It explains the ecological and floristic value of churchyards, including the
remarkable fact that six species characteristic of old meadows (Conopodium majus, Pimpinella
saxifraga, Primula veris, Saxifraga granulata, Leucanthemum vulgare and Galium verum) now have
about 50% of their Norfolk populations in churchyards. It gives advice on grass cutting, grazing,
hedging, tree planting, clearance of overgrown churchyards, renovating and repointing stonework,
management for lichens, birds and bats, the importance of ivy and many other matters. A one-page
summary of the recommendations is an especially helpful feature, and it is attractively illustrated.

Although tailored to Norfolk, most of the booklet’s advice is of very widespread applicability. It is
strongly recommended to everyone interested in churchyards, and could profitably be taken as a
model by anyone intending to produce a similar booklet for their own area.

A. O. CHATER

Flowers of Cyprus — Plants of medicine, vols. 1 & 2. C. C. Georgiades. Vol. 1: pp. 103 + 89 colour
plates; vol. 2: pp. 103 + 91 colour plates. Cosmos Press Ltd, Nicosia. 1987. Price not stated (ISBN
9963-7540—1-5, vol. 1; 9963-—7540—2-3, vol. 2).

Mediterranean people have tended to take a pragmatic approach to plants and animals in the
country around them; indeed they can often ill afford the ‘luxury’ of sentimentality when competing
for the land. By concentrating in a wide brief on the medicinal plants of Cyprus, Christos
Georgiades has performed a service in showing the large number of species (180 illustrated) which
are ‘useful’ in this context.

The third largest island in the Mediterranean (area — 9144 sq. km), Cyprus can boast some 1500
species of flowering plants, 75 of which are endemics. Its unique position has not only made it of
strategic importance throughout its history but a botanical crossroads too. The isolation from
neighbouring land masses for a significant time-span has enabled endemics to evolve, and the varied
geology and wide range of habitat types has ensured floral diversity. Botanically the island can prove
confusing since popular floras on the Mediterranean area are only partially relevant.

Though not claiming to be exhaustive, these two volumes cover a wide range of plants not
illustrated elsewhere with a text that is clearly written and highly informative. On the whole, the
illustrations are good for a work of this sort; anyone who has selected slides for plates will know that
the pressure to publish necessitates one or two compromise shots having to be used. The
blockmakers have done a good job in combatting the shrinkage of paper that bedevils the printing of
colour in hot climes and only a few of the plates are ‘out of register’.

For the serious student of the island’s flora there is no recourse but to invest in light-weight
clothing and use the baggage allowance for R. D. Meikle’s incomparable two-volume Flora of
Cyprus. However, for an acquaintance with the likes of the endemic Crocus cyprius and C.
hartmannianus, unusual plants in the Troodos massif such as Smyrnium connatum (Perfoliate
Alexanders)and Phytolacca pruinosa (Red Ink plant) or the merits of herbal tea ‘Spazia’

BOOK REVIEWS 371

(pronounced Spacha) made from shade-grown leaves of Salvia fruticosa (syn. S. cypria) these
volumes are certainly a worthwhile holiday purchase.

P.H. & J. A. DAviEs

Supplement to the Wild Flowers of Guernsey. D. McClintock. Pp. 54. La Société Guernésiaise,
Candie Museum & Art Gallery, St Peter Port, Guernsey. 1987. Price £4.40 incl. p. & p.

The Flora to which this is a supplenient was published by Collins in 1975, and was reviewed in
Watsonia 11: 83 (1976). Subsequent field work by members of the Société, together with numerous
records submitted by visiting botanists, yielded much of the information in this tersely presented
booklet. The five-page: introduction updates the preliminary pages of the Flora with additional
historical information, but fails to remedy the principal weakness of the earlier work with its scanty
treatment of the vegetational history and ecology of the northern Channel Isles.

The information in the Supplement is cross-referenced to the Flora by page numbers in bold type.
Taxonomic novelties are highlighted in three footnotes: Hedera helix subsp. hibernica (based on
Kirchner’s var. hibernica), Elymus repens var. aristatus (transferred from Agropyron repens var.
aristatus),and X Agropyron [sic] robinsonii, which should read X Agropogon robinsonii as stated in
the text. This hybrid between Agrostis stolonifera and Polypogon viridis has nothing to do with
Agropyron.

Where taxonomic changes have been published elsewhere, such as Ingrouille’s revision of the
Limonium binervosum group, particulars are included. The author delights in alluding to obscure
observations, such as in Calluna where he first dismisses a monographer’s record of var. hirsuta from
the Channel Islands and then pops in a reference to an aphid recorded from Guernsey on heather.
Some of the information in this booklet verges on the whimsical (‘“‘four tons of Primroses were sent
to England in connection with the inauguration of the memorial to Lord Beaconsfield on 16 April
1893”’), but the underlying intention to stress the economic importance of cultivated plants to the
Channel Islands is commendable.

J. R. EDMONDSON

An introduction to the flowering plants and ferns of Lochbroom & Assynt. C. Scouller, Pp. 33 with
map. Lochbroom Field Club, 1988. Price £1.50 incl. p. & p. from P. Harrison, Leckmelm, Ullapool,
Ross-shire.

This booklet is a botanical guide to two of the largest parishes in the British Isles, covering nearly
1000 sq. km. The introduction deals mainly with the history of botanical exploration in the area and
the factors influencing plant distribution. Climate is given little coverage, but the geology is well
treated although some of the geological terms may be unfamiliar to botanists.

The largest part of the book is a description of the special plants of the region, divided into
ecological groupings of moorland, mountains, woodland, wetland and coastal plants, with a special
mention of orchids. Arctic-alpines are identified throughout the text by the letters ‘“‘AA”’ after their
name. ‘

The guide is intended to be used in conjunction with a field guide or flora with the key features of
local subspecies and varieties described. This is a very useful feature, although the description of
Empetrum nigrum subsp. hermaphroditum would lead you to think that it was monoecious.

A weak point of the booklet is the rather curious usage of some English plant names and terms,
referring to Cyperaceae such as cotton grasses as “‘sedges”’ or calling Ophioglossum vulgatum a
“fern ally’. A more serious reservation is the lack of a contents page or index.

Regional guides may seem to be part of an old-fashioned tradition which has been squeezed out
by general field guides on the one hand and local nature trail guides on the other, but as an
introduction to the area, or as a useful souvenir of a visit, it can be recommended.

A. S. GUNN

372 BOOK REVIEWS

Molecules and morphology in evolution: conflict or compromise? Edited by C. Patterson. Pp. x +
229. Cambridge University Press. 1987. Price £22.50 boards (ISBN 0-—521-32271-5), £9.50
paperback (ISBN 0—521-—33860-3).

Anyone who attended the systematic sessions at the recent XIV Botanical Congress in Berlin would
have been astounded at the phenomenal rise in the number of people now working on the molecular
aspects of evolution. A symposium entitled ‘““Molecules versus morphology” was held at the Third
International Congress of Systematic Biology, Sussex University, July 1985, to assess this
burgeoning mass of molecular data. It produced a wide-ranging debate on a variety of animal
groups. There were no plants considered at all, but any reader of Watsonia interested in the history
of life will find an enormous amount of information about general principles and broad patterns of
relationships achieved by modern methods.

This is one of the most interesting and digestible books to have been produced on a very complex
subject. Buying it for Patterson’s very lucid introduction alone provides an instant entrée for those
readers who, like me, tend to skip the fine print of molecular evolution.

C. J. HUMPHRIES

The Fauna and Flora of Exmoor National Park. A provisional Check-List. Compiled by members of
the Exmoor Natural History Society. Pp. 272, incl. 4 maps. Exmoor Natural History Society,
Minehead. 1988. Available from the Exmoor National Park Office, Exmoor House, Dulverton,
Somerset. Price £7.00, post free. (ISBN 0—9512893-0-6).

Exmoor National Park, established in 1954, embraces an area of 686 sq. km., comprising moorland,
woodland and sea coast, much of which is between 360 and 580 m above sea level. The biological
survey made was extensive and detailed, and the resultant check-list covers all branches of the plant
and animal kingdoms. The account of the flowering plants, ferns and fern allies, which number over
900 taxa has been compiled by our member Caroline Giddens and occupies pages 60 to 115. These
are arranged in the system of Holmes, Outline of Plant Classification (1986) and scientific names are
based on those given in Clapham, Tutin and Warburg, Excursion Flora of the British Isles, ed. 3
(1980). English names follow Dony et al., English Names of Wild Flowers, ed. 2 (1986). The list is
supplemented by frequency data, habitat preferences and in a few instances locations.

D. H. KENT

The Natural History of the Chew Valley. K. Allen etal. Pp. 80. 1987. Available from 143 High Street,
Pensford, Avon, BS18 4BH. Price £3.90, post free.

The Chew Valley lies some 11 km S. of Bristol in North Somerset (now Avon), and this account of
its natural history by Keith Allen and 24 other contributors, including our members Clive Lovatt
and Ron Payne, as well as that great authority on trees, Alan Mitchell, is unlike any local account
that I have read before. Aimed not only at naturalists, it features walks and cycle rides in the area,
chapters on local characters past and present, and information on orchards, geology and habitats.
Several chapters are devoted to vascular cryptogams and a detailed list of species recorded from the
valley is given.

This is a delightful and very readable pot-pourri of biological data presented in an unusual
manner, and its profits are designated to the Avon Wildlife Trust, Friends of the Earth and the
World Wildlife Fund.

D. H. KENT

BOOK REVIEWS 373

Natural Science Collections in Scotland (Botany, Geology, Zoology). H. E. Stace, C. W. A. Pettitt,
C. D. Waterston. Pp. xxvi-xxvii-xxx + 373 + 8 microfiches. National Museums of Scotland,
Edinburgh. 1988. Price £25.00 (ISBN 0—948636—08-4).

Data on the natural science collections held at 286 Scottish institutions are given, as well as
information on collections in the hands of 180 private individuals.

The list of institutions surveyed is arranged alphabetically by location, followed by a combined
arrangement of institutions and individuals with details of the collections. Some biographical data
are provided, but for botanical collectors it is patchy and very inadequate, and many years of birth
and death which could have been easily ascertained are omitted.

A short series of microfiches covers subject indices to botany, geology and zoology and literature
references. They are excellent for institutions with the necessary reading equipment, but are of little
use to private individuals like the reviewer who lack that facility.

The book offers an insight into Scottish natural history collections, but with more skilful editing it
could have been greatly improved.

D. H. KENT

The correspondence of Charles Darwin, Vol. 3, 1844-1846. Edited by F. Burkhardt and S. Smith.
Pp. xxix + 523. Cambridge University Press, Cambridge. 1987. Price £30.00 (ISBN 0—521-—25589-
9).

This volume commences with the 35-year-old Darwin married and ensconced at Down House. 381
pages provide us with a clear insight into his preoccupations during the two years the book
encompasses. An additional 142 pages include appendices which explain translations (13 pp.),
chronology (4 pp.), Darwin’s notes arising from conversations with Hooker (4 pp.) and his interest
in breeding captive animals (2 pp.), together with comments on the manuscripts (18 pp.), a
bibliography (25 pp.), a biographical register (36 pp.) and an index (35 pp.).

During the two years concerned, Darwin was evidently still digesting the results of his voyage
aboard Beagle. He published his Volcanic Islands (1844) and Geological Observations on South
America (1846), whilst also bringing forth the much-revised second edition of his Journal of
Researches (1845) and a number of papers relating to the Beagle collections.

The impression has not infrequently been given that, once he settled at Down, Darwin became a
hypochondriac recluse. The correspondence published here shows such a view to be far from the
truth. He was a member and, at times, an officer of the Geological, Royal Geographical and Royal
Societies of London, regularly travelling to the capital to attend their meetings. He maintained his
lively correspondence with Forbes, Henslow, Lyell, Owen and Waterhouse, seen in the earlier
volumes of this series, but the emergence of the young Hooker as confidant and ‘sparring partner’
begins to become clear here, with an exchange of no less than 107 letters.

In 1843 Joseph Dalton Hooker returned from his four-year voyage with James Clark Ross, who
took Erebus and Terror into and around the southern oceans. Hooker, a botanist, was exercised by
the relationship between the plants of the austral continents, and his perceptive Introductory Essay
to the Flora of Tasmania is still required reading for students of these affinities. Henslow, who was
originally charged with identifying Darwin’s Beagle plants, found that his East Anglian responsibili-
ties made this virtually impossible. Hooker took on this task, and the Beagle plants are dealt with in
his mammoth Botany of the Antarctic Voyage. In the volume under review the exploitation by
Darwin of Hooker’s botanical knowledge is seen to be gradually transformed into an increasingly
close friendship. Darwin was assembling his notes and mulling over his ideas on evolution, and,
increasingly, he used Hooker as his ‘sounding board’. It is interesting to read that he was also
considering his mortality. Should he die, he wrote to his wife, his preliminary essay should be
published. Possible editors of his essay were Lyell, Henslow, Forbes, Londsdale, Strickland or,
perhaps, Owen; various emendations to this list resulted in his final choice - Hooker. The sum to be
devoted to the publication was £400! Think on that!

As I have said before in relation to the previous two volumes of this Correspondence (Watsonia

374 BOOK REVIEWS z
16(2), 17(1)), this is a book to be read and savoured; a review is inadequate. The high scholarship
involved is maintained, the presentation is excellent, the price acceptable.

D. M. Moore

The flora and vegetation of County Durham. G. G. Graham. Pp. vi + 526. The Durham Flora
Committee and Durham County Conservation Trust. 1988. Price £30.00 (ISBN 0-—905362-02-0).

This book is divided broadly into three main parts. Firstly, an introduction (64 pp.) covers such
matters as climate, soils, geomorphology and topography, as well as lists of recorders, abbreviations
and herbaria consulted. Secondly, there is a list of all plants (263 pp.), from bryophytes and lichens
to angiosperms. For each species are given other names by which it has been known in literature
dealing with Durhanz,, its status (native, introduced etc.), the phytogeographical element to which it
pertains, its frequency and abundance, and up to eight records, including the first; most species are
mapped by their occurrence in 2 km X 2 km tetrads. Thirdly, an account of the vegetation (149 pp.)
in which, grouped under general structural/habitat headings, are listed the phytosociological
associations, together with the relevés on which they are based. The text is adorned by line drawings
of the major vegetation-types, prepared at the Sunderland Polytechnic. A bibliography (19 pp.), a
gazetteer (7 pp.) and an index to plant names (21 pp.) complete the book.

In the British Isles there is a long and distinguished tradition of producing local Floras; i in giving us
the volume under review, Gordon Graham and his associates have added a jewel to this record —
must be a model for all future ventures of this sort. The organizational flair required to co-ordinate
the activities of the numerous people and groups who contributed to the programme must be
remarkable, while acknowledged financial support for the project reflects personal, regional and
national commitment to, and affection for, the County Palatine.

Born and brought up in Barnard Castle, I spent the first 18 years of my life exploring the scenic,
geological and biological riches of Teesdale, not to mention the search for fat brown trout in the
river. During the next six years, as a student at Durham, I was introduced to the ‘lowlands’ of the
eastern part of the County. Whilst I am not exactly languishing in the Thames valley, Graham’s
Flora has afforded me innumerable opportunities to return to the land of my youth. The vegetation
described is recognizable, the plants listed recall memories of many happy days in the field, while
the maps and places names bring a lump to the throat. The colour paintings of Gentiana verna and
Epipactis atrorubens, by Derek Hall, are stunning.

As a botanist I like this Flora, and as a Durham patriot I am proud of it. I was intrigued to see that
Phalaris arundinacea, amongst others, grows on the banks of ‘rivulets’; are they only called becks
amongst we of the hill-tribes in the west of the County? David Bellamy’s Foreword is well worth a
read in its own right.

The book’s publication was delayed by problems with a vital component of the computer-printing
system. The wait was worthwhile; the Flora is well-produced and attractively packaged. The price is
very competitive and I commend this volume to anyone with an interest in, or love for, the plants
and landscape of part of the land once controlled by the Prince-Bishops of Durham.

D. M. Moore

The genus Cyclamen (A Kew Magazine monograph). C. Grey-Wilson. Pp. 147, 12 colour plates.
London; Royal Botanic Gardens, Kew, in association with Christopher Helm & Timber Press,
Portland, Oregon. 1988. Price £13.95. (ISBN 0—7470-—1221-0).

Cyclamen (Primulaceae) is not native in Britain and Ireland but it is a popular garden plant; several
species are naturalized in woodland gardens (C. hederifolium and C. coum, in particular), and C.
persicum in the guise of its various, over-blown cultivars is ubiquitous as a pot-plant. The natural
habitats of the 19 species recognized by Dr Grey-Wilson range from southern France eastwards
through southern Europe into Asia Minor and northern Iran, with outlying populations in the
Balearic islands, north Africa and, as has recently been discovered, the Horn of Africa.

BOOK REVIEWS 375

This Kew Magazine monograph, illustrated with full colour portraits of species (and some garden
variants) by Mary Grierson and line drawings by Judith Gauden, follows the pattern established in
the first one which treated Paphiopedilum (Orchidaceae). Species are the primary concern; each is
described in detail, its history is discussed, and natural variation patterns are noted. Cyclamen is a
notoriously difficult genus to understand, and many wild populations do not readily fit into a neat,
‘pigeon-hole’ taxonomic system. The long horticultural history of Cyclamen, the numerous garden
hybrids and minor intraspecific variants recognized and named by gardeners, add to the confusion.
Chris Grey-Wilson discusses these problems and defines a number of new taxa in his monograph.

As it is the second book in the Kew Magazine series, I compared it with the first one
(Paphiopedilum) and was not favourably impressed. While the text is printed clearly, the colour
plates are reproduced on such poor, thin paper that the image of the Cyclamen on the reverse of
each sheet, as well as the text opposite that, are visible through the page; this is most distracting. The
line drawings of Cyclamen are not as expertly executed nor as crisply printed as those of
Paphiopedilum. It is a pity that the quality of the production has deteriorated, and I hope that this
trend will be arrested promptly.

The Cyclamen volume includes blank pages at the end — instead of wasting paper, could the series
editors not have included information on the sources of the plants depicted? In such a taxonomically
perplexing genus the provenance of illustrated materials is of considerable significance. Moreover
properly documented plates would greatly enhance the scientific and historical content of these
otherwise excellent books.

For the numerous gardeners with a passion for Cyclamen, this book will be more than welcome as
it contains substantial information about the cultivation of species and cultivars. For botanists it is an
excellent introduction to the genus.

E. C. NELSON

Plant Crib. Compiled by T. C. G. Rich & M. D. B. Rich with the editorial assistance of F. H.
Perring. Pp. (vii +) 141: numerous line drawings. B.S.B.I. Publications, 24 Glapthorn Road,
Oundle, Peterborough PE8 4JQ. 1988. Price £8.00, post paid (ISBN 0-—901158—17-8).

This latest spin-off from the Monitoring Scheme appeared in time for the second season of field
recording. The book is an aid to the identification of some of the more tricky British flowering
plants. It is largely based on Wigginton & Graham’s Guide to the identification of some of the more
difficult vascular plant species, published by the Nature Conservancy Council in 1981, which despite
its original terms of reference has proved so useful nationally. In addition to making corrections and
additions to ‘Wigginton & Graham’ (which remains indispensible, since the satisfactory accounts
are cross-referenced and indexed, but not reprinted), the chief virtue of the Plant Crib is in making
available in one place a vast range of previously scattered material from published and as yet
unpublished sources and from a wealth of anecdotal field-knowledge.

One of the commonest sources used has been the various publications of the B.S.B.I.; reprinting
of these has occasionally led to overlooking changes of mind on the part of the author (e.g. in
Gentianella). Some of the sources, however, are less easily available and the re-publication of
relevant details is a useful service (e.g. the interesting account of Erophila).

Summaries of much new information are given in Clapham, Tutin & Moore’s Flora of the British
Isles (3rd edition), but the Plant Crib makes this information more generally available. Accounts
vary in style and content, from the lavish 10 pages allotted to the Water Buttercups to the simple
drawings of leaf-shapes of the ‘London Pride’ Saxifrages. The many excellent thumb-nail sketches
of diagnostic characters make the Plant Crib especially useful.

The large, traditional ‘critical’ groups are generally omitted, though there are useful new keys to
Rosa and to the sections of Taraxacum. Four main types of plant group are concentrated on:
introduced species (e.g. Spiraea, Doronicum), hybrids (e.g. Reynoutria), infraspecific taxa and
what may be termed ‘old chestnuts’ (e.g. vegetative Littorella/Isoetes/Subularia/Lobelia).

My one reservation is the inclusion of so many infraspecific taxa (an extreme example of which

376 BOOK REVIEWS

may be seen in the key to Fumaria, which largely reverts to the excellent, but dated, work of Pugsley
of 1912). They are perhaps inappropriate in what is intended primarily as an aid for field recording.
Too many of the subspecies and varieties are not clear-cut enough to be recorded without collecting
vouchers on which any serious study will always rely.

The usefulness of the Crib will only emerge from extensive testing in the field, and I hope that
botanists will exploit it to the full. It is exactly the sort of floristic work that the B.S.B.I. should be
publishing — fiexible in format, practical and affordable. If your favourite nasty group (Callitriche
for instance) has not been adequately covered, then I am sure the editors would welcome additions
and amendments. Such a work should be frequently updated and reprinted, if it is to remain useful —
as implied in the Introduction.

« H. J. No.tie

The Flowering Plants and Ferns of the Shetland Islands. W. Scott & R. Palmer. Pp. 468 + ix, with 16
colour plates. The Shetland Times Ltd., Lerwick. Price £20.00 (ISBN 0-—900662-56-—5).

The authors of this book have been collaborating on its production for over thirty years and the
result is a well written, meticulously researched work of fine scholarship. In all aspects it meets the
authors’ hopes that it represents a considerable advance over previous Shetland Floras. The present
Flora begins with a ‘General Account of Shetland’ which describes the general environment. It is
followed by ‘An Outline of Shetland Plant Habitats’ which introduces the most important ecological
and floristic features of the fifteen main types of plant community recognized by the authors. The
third chapter is on “The Changing Vegetation of Shetland’ which emphasizes the recent pressures on
the flora. These are so severe that the authors sadly conclude ““The meagre flora of our islands has
been severely depleted, our rarities have now become even rarer; we cannot afford to lose much
more’’. There follows ‘A Botanical Itinerary through Shetland’ which will be of great use to the
visitor and can conveniently be used in conjunction with the 1:100,000 map of the Shetland Islands
which is bound in the book’s centre. The maps show the boundaries of the thirty-nine 10-km squares
which the authors used as the basis for their distribution records.

There follows an ‘Analysis of the Flora’ which discusses generally the status “ind geographical
affinities of the Shetland plants including some intriguing contrasts and similarities with the Orkney
and Faeroe Islands. There is a final introductory Chapter on the ‘History of Botanical Exploration in
Shetland’.

Most of the book concerns detailed accounts of the Shetland taxa and includes a full treatment of
critical general such as Hieracium and Taraxacum. It is good to read in depth about outstanding
Shetland rarities such as Arenaria norvegica subsp. norvegica and the.endemic Cerastium nigrescens
subsp. nigrescens. The authors give excellent taxonomic criticisms on many Shetland taxa. For
example they reject Druce’s Plantago edmondstonii but favour renewed investigations into the
taxonomic status of the Shetland sea-cliff Silene dioica. Many of the descriptions are entertaining: I
would like to know the reasons why (for Trientalis europaea) ‘“women seem particularly successful
at finding it.””!

I have little on the negative side to say about this book. I missed any attempt to deal with the non-
vascular cryptogamic flora of Sheltand. More details of the Quaternary history of the Shetland flora
would have been welcome along with some speculation on its source. I noticed only one factual
error: fertiliser addition and surface seeding began in 1966 on the Keen of Hamar — not 1967 as
implied on p. 30.

This book is essential for all naturalists who plan to visit Shetland. Its pleasing production and
high quality colour photographs, combined with its clear and informative text will make it a
desideratum for most botanical bookshelves. It is a pity that so much of the delight I found in the
book was tempered by the sombre messages of the third chapter which reveals the great and
continuing threats to the Shetland flora.

J. PROCTOR

BOOK REVIEWS are

The naturalist’s garden. John Feltwell. Pp. 160, with numerous black & white and colour plates.
Ebury Press, London. 1987. Price £12.95 (ISBN 0—85223-661-1).

There have been histories of gardening written from various viewpoints, e.g. landscape, cultivation,
plant introduction; but I know of no previous works that have considered the gardeners and garden
makers as naturalists. This new and revealing point of view has enabled the author to show, for
example, not only that the Moorish gardens of southern Spain reflect an interest in the ambient
plants and animals by their mediaeval designers, but that famous gardens of our own day, such as
Christopher Lloyd’s at Great Dixter, East Sussex, often betray an interest in local wild life. His ‘wild
lawn’ in famous, and he favours mixed borders rather than beds of roses, for instance, as having the
advantages of natural communities, attracting varied wild life and reducing attacks of diseases and
pests.

The book is attractively written and illustrated, giving a good general history of gardens and
gardeners as well as the abovementioned emphasis on their natural history interests. The historical
sequence is interrupted by chapters on relevant but tangential subjects, such as ‘Birds in the garden’,
‘North American explorers,’ “The parson-naturalists’ and ‘Encouraging wild flowers’, all of which
combine to make this book most useful as well as decorative. B.S.B.I. members, in particular those
who have hitherto confined their interests to truly wild plants, should read it in order to see how
many aspects of natural history (other than distribution, of course) can be studied in a garden.

N. K. B. RoBsSON

The heritage of Clonmacnoise. Edited by M. Tubridy. Pp. 136. Environmental Sciences Unit,
Trinity College [Dublin] in association with County Offaly Vocational Educational Committee.
1987. Price not stated (ISBN 0—9512627-0-X, hardback; 0-—9512627-1-8, softback).

Clonmacnoise, along the River Shannon, is right in the middle of Ireland in the Central Lowlands. It
is the site of a combination of conservation interests - Mongan Bog, one of the few remaining raised
bogs left in Western Europe; The Callow, a flat area adjacent to the river which regularly floods in
winter and occasionally in summer; eskers, formed in rivers 20,000 years ago, which now commonly
support species-rich grasslands and a number of interesting esker woods; a small lake, Fin Lough,
which is surrounded by a variety of vegetation types including fen and a distinct community
associated with lime-rich spring water; and areas of outcropping limestone. As well as this
considerable diversity of vegetation, the bird life is similarly rich and varied.

This book represents an attempt to bring together a number of experts, a team from Trinity
College, Dublin, to record their observations on the Clonmacnoise region. This has been combined
with fascinating accounts of the early history of the area including the very famous monastery and an
insight into farming in the area since the 1830s. The result is a most readable work, full of detail,
illustrations, photographs (some in colour) and maps which form an excellent guide to the history
and conservation interest of the region and which will become an indispensible guide for any visitor.
It should serve as a model for any inter-disciplinary approach to conservation areas and the authors
and sponsors are to be congratulated on a fine combined effort.

: B. S. RUSHTON

The difficult and critical plants of the Lizard District of Cornwall. L. J. Margetts, assisted by A. J.
Byfield, R. W. David & P. L. Smith. Pp. x + 77, with 16 b/w plates, 88 distribution maps &
transparent overlay. Grenfell Publications Ltd., Bristol. 1988. Price £7.95 (ISBN 0—948715-01-4). >

Much as the Lizard acts as a magnet, it is usually the rare ‘specials’ that attract botanists, with most
of Britain’s clovers present, masses of Erica vagans (Cornish Heath) and many other taxa of
extreme rarity in the British Isles. Some of the observations in this neatly presented and closely
written book are evocative: Polypody ‘‘makes a more or less continuous fringe at shoulder height”
(for the specialist, both Polypodium interjectum and P. vulgare are meant); the Prickly Sedge, Carex

378 BOOK REVIEWS
muricata subsp. lamprocarpa, grows ‘“‘where the cows are still driven for milking and lightly graze
the hedges as they pass’’, hedges in Cornwall being solid banks.

This approach means that other books will be needed for identification, despite the clear and
succinct descriptions. These include Rubus stanneus, with ‘“‘deeply toothed, overlapping leaflets”
very characteristic of this bramble, and a one-line description of the leaf, flower and odour of
Mentha X piperita. The 1-km square dot-maps show the amount of work achieved by the University
of Bristol’s Lizard Project, from which this book stems; they have researched the Lizard vascular
flora with great intensity.

The presentation of the book is stylish, the writing is excellent, and it should inspire readers to
take (or further) an interest in critical groups. The Lizard itself scarcely needs to be commended to
botanists.

3
-

K. L. SPURGIN

The flora of Lough Neagh. J. Harron and B. Rushton. Pp. iii + 270 with 1 colour plate, 3 figures and
334 plant distribution maps. Irish Naturalists Journal Committee, Belfast and University of Ulster,
Coleraine. 1986. Price £5.00 (ISBN 0—901229-—82-2).

Tansley in 1939 observed that L. Neagh fenlands were “‘probably the most extensive in the British
Isles still remaining comparatively unspoilt”. Unfortunately, even as Tansley was writing, Shephard
was ‘improving’ the discharge of the R. Lower Bann and regulating lough levels; subsequent
lowerings of lough levels by 1-5 m markedly reduced the fenlands and, by 30%, the open water area
of the small macrophyte-rich L. Beg.

Nearly 50 years after Tansley’s ‘probably’ and ss Spat te comes this excellent volume
actually recording what is there, a modern stock-taking of the wetland flora associated with L.
Neagh. Its Introduction concisely reports on the topography and geology of the area, reviews the
limited evidence of the pre- and post-glacial history of the vegetation and gives a synopsis of the
drainage activities. The characteristics of the present vegetation (comprising some 700 species) are
outlined in an Irish context; much of the flora is characteristic of much of lowland Ireland, although
several species are of particular local significance, such as Ranunculus fluitans which has its only
known Irish site in the Six Mile Water and Spiranthes romanzoffiana subsp. stricta which is also
beautifully illustrated in Piper’s coloured frontispiece, while for others L. Neagh appears to be the
centre of their Irish distribution. The regrettable disappearance of half-a-dozen species including
Carex buxbaumii and Thelypteris palustris is confirmed.

Some 200 pages of the book are devoted to the species records based on but authoritatively
updating three previous and separate floras of the North-East of Ireland, Tyrone, and Armagh
published in 1888, 1893 and 1942 respectively. There is a most helpful list of field workers, their
publications and active periods, a clear map of the basin, and a good index.

The flora is the botanical offspring of a happy marriage of John Harron’s amateur (in the loving
sense of that oft-misused term) skills and enthusiasm for plants and Brian Rushton’s equal but more
professional academic talents. Bridesmaids and groomsmen come from a circle of colleagues,
principally the Belfast Naturalists’ Field Club, the Ulster Museum and the Irish Naturalists Journal.

Merely to say that the Flora is welcome would be a gross understatement. It comes at a time of
increasing conservation activity in Ulster’s post-Balfour era and must help in that. At £5.00
(machine type-set) it is a bargain even if the contents will last much longer than the soft covers. This
— with the soon-to-appear revision of Stewart and Corry — will serve local botanists well.

R. B. Woop

Watsonia, 17, 379-381 (1988) 379

Obituaries

IRENE MANTON
(1904—1988)

Professor Irene Manton, born at the turn of the century and who died on 31 May 1988, was one of
the most remarkable botanists of our time. As a girl she was originally destined to be a musician but,
by her own account, public appearances as solo violinist at school concerts were traumatic
experiences. Although she abandoned the idea of music as a career, she remained an accomplished
performer throughout her life and often relaxed by playing in a quartet. In her latter years at school
she became a convert to science after reading Newton’s Laws of Motion, which struck her forcibly
by their elegant simplicity. Her interests were specifically channelled towards cytology by reading E.
B. Wilson’s book on The Cell (1902), but having won a scholarship to Cambridge she was
disappointed to find that her undergraduate years there (1923-26) did nothing towards realising her
ambition of studying chromosomes. However, the award by her College of a postgraduate
studentship of £150 per annum enabled her to travel to Stockholm to work in Prof. Rosenberg’s
laboratory (and incidentally to become fluent in Swedish in a very short time!). At this period the
study of chromosomes involved the lengthy and tedious process of embedding, sectioning and
staining root tips. This formed the basis of her doctoral thesis, involving a study of 250 species of
Cruciferae. Her external examiner pointed out a discrepancy between the text and drawings which
was to have an important effect later on, namely that watercress was described as having 2n = 32 but
figured with 2n = 48. Fresh material was hastily gathered and processed to determine how the
discrepancy arose and this proved to have 2n = 64! This was her introduction to a wild polyploid
series and their sterile hybrid. It was also a portent in that thereby she added a new species to the
flora by discovering its chromosomes first — a pattern that was to be repeated many times by herself
and her students in various parts of the world.

Irene Manton went to the University of Manchester in 1929 as an assistant lecturer on the staff of
Prof. W. H. Lang, co-author with R. Kidston of the famous series of papers on the Rhynie fossils.
At this time Lang was interested in the mechanism of aposporously produced plants of the Royal
Fern, Osmunda regalis, and asked her to look at them cytologically. From this work arose the classic
papers on chromosome structure, together with the autoploid series of diploid, triploid and
tetraploid plants which provided such excellent teaching material for future generations of
cytologists.

During a short visit to Egypt, Dr J. Philp at the Cotton Research Institute showed her some slides
of meiotic cells prepared by McClintock’s method. This consisted of staining in aceto-carmine and
placing a coverslip on top — the weight of the coverslip slightly flattening the soft cells. She took up
this new technique with enthusiasm and greatly improved it by applying heavy manual pressure to
the coverslip. This produced perfectly flat cells in which all the chromosomes could be seen in one
plane and made possible the accurate counting of the high numbers found in the ferns — something
that had defeated previous cytologists.

World War II brought restrictions on printing and Irene Manton had accumulated at Manchester
such a wealth of material on the entire British fern flora that to publish it piecemeal when
restrictions were lifted would have taken an unacceptable length of time and the decision was taken
to incorporate it all in book form. In the meantime she had been appointed to the Chair of Botany at
Leeds in 1946. Here, her incredible energy not only enabled her to run and re-equip the Department
(delegation was not in vogue in those days!), but also rapidly to build up a very flourishing research
school, and take on a full teaching load. Nevertheless at the end of each day she would settle down
to do up to nine hours research. Weekends were an added bonus only interrupted by the necessity of
a few hours sleep and snatched meals. One outcome of all this workload was the streamlining of the
mechanics of research by using photographic methods wherever possible in order to eliminate much
of the traditional drawing. This was of critical importance and set new standards, since she adopted
as her maxim ‘what cannot be photographed cannot be used in evidence’.

380 OBITUARIES

The book, Problems of cytology and evolution in the Pteridophyta, was published in 1950 and
proved to be a landmark of much wider importance than its title might suggest. Not only were many
problems in the British fern flora resolved, but it effectively disposed of the view held by many
European flowering plant cytologists that polyploidy was a direct outcome of the cold experienced
during the Ice Age. This was reinforced by the joint publication with W. A. Sledge in 1954 of the
cytological survey of the ferns of Ceylon which showed an even higher percentage of polyploidy in
this tropical island. The advent of the book stimulated research all over the world with the result that
the ferns are now one of the best known groups of plants cytologically, despite the late start.

The frontispiece of the book depicts a fern spermatozoid as seen under the ultra-violet
microscope at a magnification of x 3000. In this some of the cilia can be seen to have disintegrated
into their fibrillar components and the observation led to the next great phase of her research which
was to occupy Irene Manton for the rest of her life, namely the fine structure of cells and their
components. Following what was literally a flying visit to the Rockefeller Institute in New York to
work with an electron microscope, she soon established at Leeds the first electron microscope
laboratory in the world devoted to the fine structure of plants. Initially work was concentrated on
the structure of cilia in motile plant cells such as spermatozoids and zoospores. The structure of two
central strands surrounded by nine outer ones was demonstrated to be common to all groups of
plants with such motile cells. The quality of the photographic proof and the manner of presentation
were, as always, of the highest standard and I well remember the whole audience giving her a
standing ovation after her lecture at the International Botanical Congress in Paris in 1954. This work
continued with investigations into other organelles such as plastids, vacuoles, and Golgi apparatus
and much of what is now student textbook knowledge dates from this work. —

The algae had always been a great interest of hers and much of the early scanning electron
microscopy was based on these plants, some of it done in collaboration with Mary Parkes at the
Marine Laboratory, Plymouth. The remarkable scale structures seen in the Chrysophyceae linked
up with another of her passions, and the walls of Botany House were covered with SEM
micrographs alternating with prints of modern abstract art without one feeling any sense of
incongruity. During her time at Leeds, Irene Manton amassed a large and valuable collection of
Chinese and modern abstract art — both prints and originals — and she delighted in staging
exhibitions of these, the last one being at Lancaster University earlier this year.

Retirement in effect meant the shedding of administrative and teaching duties in order to devote
herself fulltime to research and her other interests. Much of this period was taken up with a study of
the structure and distribution of nanoplankton, taking her on collecting trips to Greenland, Hudson
Bay and Resolute Bay, Alaska, South Africa and the Galapagos Islands. Many papers resulted from
this, with one major paper a month being maintained over a number of years. Her style of writing
was instantly recognizable and her accounts were often racy, but always models of clarity. The
quality of her work was recognized by her election to the Royal Society and to the Presidency of the
Linnean Society of London, together with the award of honorary degrees from several countries and
other honours in the form of medals and citations.

From all this it will be gathered that Irene Manton was a person of enormous energy, a quality
much in demand in Botany House at Leeds, where her office was on the first floor, the SEM and
darkrooms in the basement and the optical bench on the first floor — to say nothing of the teaching
laboratories being located in a different part of the University and the experimental gardens about a
mile away! One of my abiding memories as a student is of trailing panting after her in an effort to
keep up.

Her relationship with her research students was warm, and they were collectively always
affectionately known to her as “The Young”. It was as “The Young” that we remained despite the
passage of the years and the acquisition of families of our own who were in their turn included in the
fellowship.

T. G. WALKER

OBITUARIES 381

CHARLES PLOWRIGHT PETCH
(1909—1987)

Charles Petch died suddenly at his home in Wolferton, Norfolk on 8 December 1987. Born and
educated in the county, he was recording the local flora whilst still at school at Gresham’s.

He was one of a party of six from the Universities of Oxford and Cambridge to visit the deserted
island of St Kilda in 1931 (the inhabitants had been evacuated in 1930), and his paper on “The
vegetation of St Kilda’ was published in J. Ecol. 21: 92-100 (1933). After obtaining first class
Honours in Natural Sciences from St John’s College, Cambridge, he taught at Stowe. He then
returned to his former college and frorh there went to St Thomas’ Hospital, qualifying as a physician
in 1939. During World War II he served as a Medical Officer in the R.A.F. In 1948 he was appointed
consultant physician to St Hellier Hospital, Carshalton, a position he held until his retirement in
1975.

Whilst still at St Thomas’, Charles married Margaret Stirling. She and their two sons would
frequently accompany him in the field. A most hospitable family, they were always ready to
entertain a grubby and noisy party of botanists to tea after a field meeting.

Charles joined the B.S.B.I. in 1952 and with the inception of the Mapping Scheme soon had his
family out ‘square bashing’ in Norfolk. He was a meticulous recorder. In 1962, West Norfolk plants
today, with E. L. Swann, was published as a supplement to Proc. bot. Soc. Br. Isl. 4(1962). After his
retirement Charles was appointed recorder for v.c. 28 and served on the Council of the B.S.B.I.

Although he lived and worked in Surrey, and Surrey botanists like to claim him as their own, his
heart was in Norfolk, and it was there that he carried outt the major part of his botanical and
ecological work, often in collaboration with Eric Swann. Every holiday and on many weekends,
Charles, his family and his ’cello could be seen setting off in his old silver-grey Bentley in the
direction of Norfolk, where he kept a caravan and a boat. He was a member of the Norfolk and
Norwich Naturalists’ Society, of which he was elected President for the year 1981-82. He
contributed many papers on West Norfolk for the Society’s Transactions, and to commemorate
their centenary they published the Flora of Norfolk by C. P. Petch and E. L. Swann in 1968.

More recently he was a major contributor to the Ecological Flora of Breckland, edited by P. J. O.
Trist, though he was critical of the random sampling method of recording, especially as it always
seemed just to miss the rarities. He certainly knew the precise location of the interesting species in
his neighbourhood. None of this “Well, it’s along here somewhere”’: he knew exactly where to stop
his car to point out to me the Berberis in the hedgerow or the rosettes of Verbascum pulverulentum.
This was just as well if the weather was too appalling to open the car doors!

In Surrey Charles was a founder member of the Surrey Flora Committee in 1957, and a
contributor to the Flora of Surrey (Lousley 1976). Following the sudden death of Ted Lousley in
1976, Charles Petch and Cecil Prime saw the Flora through to publication. Charles was elected to
the S.F.C. in 1971, serving on it until his death. He was a very active member of the Committee,
undertaking site recording and regularly attending field meetings. The area of Surrey that he chose
to work in detail was the Lower Greensand around Leith Hill — perhaps because this was
reminiscent of his native Norfolk, to which he later retired.

He was also a competent mycologist. This had been his father’s profession. He led fungal forays in
Surrey and Norfolk, and as his elder son writes, “‘the quality of our breakfasts was radically
improved by his knowledge.”

Charles was a reticent man, quick-tempered but with a pleasing sense of humour. ‘‘What has it to
do with Norfolk County Council?”’, he inquired, when I referred to the N.C.C. at a Surrey meeting.
I now realize that he must have had exceptional organizational ability. How else could he have
managed a hospital consultancy, a private practice, beagling, sailing, walking, playing the ’cello,
entertaining, and still contribute so much to the botanical and ecological knowledge of the counties
of Surrey and Norfolk?

J. E. SMITH

Watsonia, 17, 382-383 (1988)

Report

ANNUAL GENERAL MEETING, 7 May 1988

Following the Presidential Address, the Annual General Meeting of the Society was held at 12.15
hrs in the Jodrell Laboratory Lecture Theatre, Royal Botanic Gardens, Kew, by kind permission of
the Director. 120 members were present. Professor C. A. Stace, President, took the Chair,
apologies for absence were read, and the Minutes of the 1987 Annual General Meeting, as
published in Watsonia 17: 119-120 (1988), were approved and signed by the President.

REPORT OF COUNCIL

The total of 4,760 survey cards sent in by B.S.B.I. members by the end of the first year of the
Monitoring Scheme was added to the Report, and it was noted that the records were being entered
on to the computer and copies of the cards were being returned to participants for the second year of
field work. In the Conservation Committee report a sentence was amended to “letters had been sent
supporting the opposition to three major conservation threats . . .”. The adoption of the Report
was then proposed by Mr E. Milne-Redhead, seconded by Mr R. G. Ellis and passed unanimously.

TREASURER S REPORT AND ACCOUNTS

The Treasurer, proposing the adoption of his Report and Accounts, offered to clarify any points,
but there were no queries. The adoption of the Report was seconded by Mr G. LI. Lucas and
unanimously approved.

RE-ELECTION OF HONORARY GENERAL SECRETARY AND HONORARY TREASURER

The President proposed Mrs M. Briggs and Mr M. Walpole for re-election as Honorary General
Secretary and Honorary Treasurer, warmly thanking both for their major roles in the running of the
Society. Their re-election was seconded by Dr T. C. G. Rich and carried by the meeting with
applause.

ELECTION OF COUNCIL MEMBERS

In accordance with Rule 10, nominations had been received for Dr S. L. Jury, Dr P. S. Wyse
Jackson and Mr N. F. Stewart; these were elected unanimously. The President then thanked all the
Editors and Secretaries appointed by Council — their considerable work for the Society was very
much appreciated; in particular Mr R. Smith, Field Secretary, the Secretaries of all the Committees,
and the Editors of all the Society’s publications. The achievements of the Society through the year
were very largely due to the time and work given voluntarily by them to the Society.

ELECTION OF HONORARY MEMBER

Proposing Dr N. K. B. Robson from the Chair, the President referred to Dr Robson's 21 years of
service as an Editor of Watsonia. The Society was exceedingly grateful for this very long term of
service and Dr Robson would be one of our most deserving Honorary Members. In reply, Dr
Robson wished the whole Editorial team to be included in the thanks, saying that his work as Editor

REPORT 383

had brought him much interest and enjoyment, and many contacts. He added that all aspects of the
B.S.B.I. had been a pleasure to him since he first joined the Society.

RE-ELECTION OF HONORARY AUDITORS

The Chairman proposed the re-election of Grant Thornton, West Walk, Leicester, as Honorary
Auditors, and the Treasurer, asking for a letter of thanks to be sent, said that the Society was
fortunate to have a firm of international standing named below the Annual Accounts.

ANY OTHER BUSINESS

_ Mr E. Milne-Redhead asked if the Conservation Committee had considered his request for a
recommendation to the Ministry of Transport on sowing wild flower seeds on motorway verges.
Praising his County Council for their co-operation in sowing local wild flower seeds on minor roads,
Mr Milne-Redhead said that in Suffolk they would welcome a similar policy for motorways and
trunk roads. A lively discussion followed as Dr J. G. Dony reported a confusion of locally rare
plants recorded after wild flower seed sowing. In Bedfordshire and Hampshire there was a strong
feeling that no wild flower seeds should be sown, but that new verges should be left to natural
colonization. Opinions from the floor were divided and the subject was referred to the Conservation
Committee for discussion on the practicability of formulating a B.S.B.I. policy on this matter.
Mr R. M. Bateman asked whether both the date of receipt and of acceptance for publication
could be recorded for papers published in Watsonia. This point was taken and referred to the
Editors. Mr Bateman also asked if the general Editorial policy and aims for the journal could be
stated in more detail than in the ‘Instructions to Contributors’ currently printed in Watsonia. The
Editors replied that a longer statement on this was planned and will be published in B.S. B.J. News.
The meeting closed at 12.58 hrs. x

M. Briccs

i 2

*4.

a

eit

ag Wf &

*18.

Mi. 8

*20.

B.S.B.I. Conference Reports

BRITISH FLOWERING PLANTS AND MODERN SYSTEMATIC METHODS
Ed. A. J. Wilmott, 1948, 104 pages, 18 plates. £5.25.

. THE STUDY OF THE DISTRIBUTION OF BRITISH PLANTS.

Ed. J. E. Lousley, 1951. 128 pages, illustrations and maps.

. THE CHANGING FLORA OF BRITAIN

Ed. J. E. Lousley, 1953. 203 pages, 9 plates, 25 text figs.
SPECIES STUDIES IN THE BRITISH FLORA

Ed. J. E. Lousley, 1955. 189 pages, 2 plates and 23 text figs. £5.25.
PROGRESS IN THE STUDY OF THE BRITISH FLORA

Ed. J. E. Lousley, 1957. 128 pages, 4 plates and 9 text figs. £5.25.

. A DARWIN CENTENARY

Ed. P. J. Wanstall, 1961. 140 pages, 7 plates, 12 text figs.

. LOCAL FLORAS

Ed. P. J. Wanstall, 1963. 118 pages, 1 plate (map), 9 text figs.

. THE CONSERVATION OF THE BRITISH FLORA

Ed. E. Milne — Redhead, 1963. 90 pages.

. REPRODUCTIVE BIOLOGY AND TAXONOMY OF VASCULAR PLANTS

Ed. J. G. Hawkes, 1966. 182 pages, 1 plate, 9 text figs.

. MODERN METHODS IN PLANT TAXONOMY

Ed. V. H. Heywood, 1968. 312 pages, numerous text figs.
Held in association with the Linnean Society of London.

. THE FLORA OF A CHANGING BRITAIN

Ed. F. H. Perring, 1970. 158 pages, 21 text figs. £3.50 (1973 reprint).

. TAXONOMY, PHYTOGEOGRAPHY AND EVOLUTION

Ed. D. H. Valentine, 1972. 431 pages, numerous text figs and tables. Held in association with
The Linnean Society of London and the International Organisation of Plant Biosystematists.

. PLANTS WILD AND CULTIVATED

Ed. P. S. Green, 1973. 232 pages, 8 plates and 24 text figs. £3.20.

. THE OAK: ITS HISTORY AND NATURAL HISTORY

Ed. M. G. Morris & F. H. Perring, 1974. 376 pages, illustrations. £8.25

. EUROPEAN FLORISTIC AND TAXONOMIC STUDIES

Ed. S. M. Walters, with the assistance of C. J. King, 1975. 144 pages and 4 plates. Held in
association with the Linnean Society of London. £3.80.

THE POLLINATION OF FLOWERS BY INSECTS

Ed. A. J. Richards, 1978. 213 pages and 31 plates. Held in association with the Linnean
Society of London. £45.

THE BIOLOGICAL ASPECTS OF RARE PLANT CONSERVATION

Ed. H. Synge, 1981. 586 pages and numerous text figs. Held in association with the Linnean
Society of London. £49.95.

PLANT LORE STUDIES

Ed. R. Vickery, 1984. 260 pages. Held in association with the Folklore Society. £7.50.
ARCHAEOLOGY AND THE FLORA OF THE BRITISH ISLES

Ed. M. Jones, 1987. 128 pages and numerous text figs. Held in association with the Association
of Environmental Archaeologists. £15.

THE LONG TRADITION

Ed. H. J. Noltie, 1987. 192 pages, 25 black and white illustrations. Held in association with the
Botanical Society of Edinburgh and The Society for the History of Natural History. £21.50.

Items marked with an asterisk are in print and available from BSBI Publications, 24 Glapthorn
Road, Oundle, Peterborough PE8 4JQ, at the prices stated (postage included).

INSTRUCTIONS TO CONTRIBUTORS

Papers and Short Notes concerning the systematics and distribution of British and European
vascular plants as well as topics of a more general character are invited.

Manuscripts must be submitted in duplicate, typewritten on one side of the paper only, with wide
margins and double-spaced throughout. They should follow recent issues of Watsonia in all matters
of format, including abstracts, headings, tables, keys, figures, references and appendices. Note
particularly use of capitals and italics. Only underline where italics are required.

Tables, appendices and captions to figures should be typed on separate sheets and attached at the
end of the manuscript. Names of periodicals in the references should be abbreviated as in the World
list of scientific periodicals, and herbaria as in Kent & Allen’s British and Irish herbaria. Line
drawings should be in Indian ink on good quality white card, blue-lined graph paper or tracing
paper. They should be drawn at least twice the final size and they will normally occupy the full width
of the page. Lettering should be done in Letraset or by high-quality stencilling, though graph axes
and other more extensive labelling are best done in pencil and left to the printer. Photographs can be
accepted only in exceptional cases.

Contributors are strongly advised to consult the editors before submission in any cases of doubt.
25 offprints are given free to authors of papers and Short Notes. Further copies may be purchased in
multiples of 25 at the current price. The Society takes no responsibility for the views expressed by
authors of articles.

Papers and Short Notes should be sent to Dr R. J. Gornall, Botany Department, The University,
Leicester, LE1 7RH. Books for review should be sent to Dr J. R. Edmondson, Liverpool Museum,

William Brown Street, Liverpool, L3 8EN. Plant records should be sent to the appropriate vice-
county recorders. Reports of field meetings should be sent to Dr B. S. Rushton, Biology Department,
The University of Ulster, Coleraine, Co. Londonderry, N. Ireland, BT52 1SA.

& MANU
MAGGS BROS. LTD., LONDON

MAGGS BROS. LTD. Rare books

autographs, manuscripts and miniatures

SO PE RRE LEY SOUARE LONDON WI1X 6EL
Cable address: Bibliolite London W1 Telephone: 01-493 7160

MANY LOCAL FLORA AND GENERAL BOTANY BOOKS -
ARE INCLUDED IN OUR LATEST LIST OF
RARE AND SECOND HAND BOOKS ON
NATURAL HISTORY |

available free on request

hybrids i ift the British Isles © Bs

PARNELL, J. A. No & SIMPSON, D. A. oe id
mite Schrank, P. minus Huds - Farce

flora x? ieee ee ae aE
PRESTON, C. D. The spread of Elobin

wasters Fife

‘Preston, C.D. Types of Panangieen
and P. natans subsp. kirkii Hooker oe

Stace; C. A. & Al-Bermani, A
Festulpia combinations: <;:

Williams, L. R. Crataegus x. mene Bechst. in 0 Mi idl
Book REVIEWS °.. firs tne eal Rone oe
OBITUARIES es JR ean AG ag Kees

REPORT?
Annual General Meeting, 7 May 1988 pee

Published by the Botanical Society of the British Isles

UK ISSN 0043-1832

Filmset by

WILMASET, BIRKENHBAD, MERSEYSIDE
Printed in Great Britain by

EATON PRESS CIMPFTED: WESTRIFEL DROAD. WALLASEY. 1 Aig eet

“ee
whe

6G FUSE aes;

Ot

a6)
su tee

FSF OS.

Rtas Saeed

Deiy tae onlay ere

Chen Bee ee

ie Ao

ALT
el a a

Fh ehere tye

‘

ra