LIV
3005

Xtvcrpool fIDartne Bioloo^ (Tonimtttce.
LMB£, MEMOIRS

ON Typical British Marine plants s- Animals

EDITED BY W. A. HERDMAN. D.Sc, F.R.S.

XIII.

ANURIDA

A. D. IMMS, B.Sc. (Lond.),

Christ's College, Cambridge.

(With 7 Plates)

Price Four Shillings

LONDON

Williams & Norgai k

October, 1906

HARVARD UNIVERSITY.

I^ I B R A R Y

MUSEUM OF COMPARATIVE ZOOLOGY.

\^\y^ -

L.M.B.C. MEMOIRS.

XIII.

ANURIDA.

NOTICE.

The Committee desire to intimate that uo copies of these
Memoirs will be presented or exchanged, as the prices
have been fixed on such a scale that most of the copies will
have to be sold to meet the cost of production.

'J.'he Memoirs may be obtained at the nett prices stated,
from Messrs. Williams and Xorgate, 14, Henrietta Street,
Covent Grarden, London.

Memoii- 1. Ascidia published in October, 18:)!), (iO pp.
and five plates, price 2s.

,, II. Cardium — published m December, 1899,
92 pp., six plates and a map, price 2s. 6d.

,, III. Echinus — published in February, 1900.

36 pp. and five plates, price 2s.
„ IV. Codium — published in April, 1900, 26 pp.

and three plates, price Is. 6d.

„ Y. Alcyouium — published in January, 1901,.
30 pp. and three plates, price Is. 6d.

„ VI. Lepeophtheirus and Lernsea — published in
March, 1901, 62 pp. and five plates,
price 2s.

,, VII. Lineus — published in April, 1901, 40 pp. and

four plates, price 2s.
,, VIII. Pleuronectes — published in December, 1901,

260 pp. and eleven plates, jDrice 7s.

,, IX. Chondrus — published in July, 1902, 50 pp.
and seven plates, price 2s. 6d.

„ X. Patella — published in May, 1903, 84 pp. and

four plates, price 2s. 6d.
,, XI. Arenicola — published in March, 1904, 126

pp. and eight plates, price 4s. 6d.

,, XIT. Gammarus — published in July, 1904, 55 pp.
and four plates, price 2s.

,, XIII. Anurida — published in October, 1906, 105 pp.
and seven plates, price 4s.

Xivcrpool nPartnc Biologv Connnittce.
L.M.B.C. MEMOIRS

OK Typical British Marine Plants &- Animals

EDITED BY W . A. HeRDMAN. D.Sc, F.R.S.

XIII.

ANURIDA

BY

A. D. IMMS, B.Sc. (Lond.),

Chriifs College, Caiiihridoe.

(With 7 Plates)

Price Four Shillings
>^ LOND ox

Williams & XorgatI':
October, 7906

EDITOR'S PREFAf'E.

The Liverpool Marine Biology Committee was constituted
in 1885, with the object of investigating the Fauna and
Flora of the Irish Sea.

The dredging, trawling, and other collecting expeditions
organised by the Committee have been carried on inter-
mittently since that time, and a considerable amount
of material, both published and unpublished, has been
accumulated. Nineteen Annual Eeports of the Committee
and five volumes dealing with the '' Fauna and Flora "
have been issued. At an early stage of the investigations
it became evident that a Biological Station or Laboratory
on the sea-shore nearer the usual collecting grounds than
Liverpool would be a material assistance in the work.
Consequently the Committee, in 1887, established the
Puffin Island Biological Station on the North Coast of
Anglesey, and later on, in 1892, moved to the more
commodious and accessible vStation at Port Erin in the
centre of the rich collecting grounds of the south end of
the Isle of Man. A new and larger Biological Station and
Fish Hatchery, on a more convenient site, has now been
erected, and was opened for work in July, 1902.

In these nineteen years' experience of a Biological
Station (five years at Puffin Island and foiirteen at Port
Erin), where College students and young amateurs form a
large proportion of the workers, the want has been fre-
quently felt of a series of detailed descriptions of the
structure of certain common typical animals and plants,
chosen as representatives of their groups, and dealt with by
specialists. The same want has probably been felt in other
similar institutions and in many College laboratories.

VI.

The objects of the Committee and of the workers at the
Biological Station were at first chiefly faunistic and
speciographic. The work must necessarily be so when
opening- up a new district. Some of the workers have
published papers on morphological points, or on embry-
ology and observations on life-histories and habits ; but
the majority of the papers in the volumes on the " Fauna
and Flora of Liverpool Bay " have been, as was intended
from the first, occupied with the names and characteristics
and distribution of the many diiferent kinds of marine
plants and animals in our district. And this faunistic
work will still go on. It is far from finished, and the
Committee hope in the future to add still further to the
records of the Fauna and Flora. But the papers in the
present series, started in 1899, are quite distinct from these
previous publications in name, in treatment, and in pur-
pose. They are called " L.M.B.C. Memoirs," each treats
of one type, and they are issued separately as they are
ready, and will be obtainable Memoir by Memoir as they
appear, or later bound up in convenient volumes. It is
hoped that such a series of special studies, written by
those who are thoroughly familiar Avith the forms of which
they treat, will be found of value by students of Biology
in laboratories and in Marine Stations, and will be
welcomed by many others working privately at Marine
Natural History.

The forms selected are, as far as possible, common
L.M.B.C. (Irish Sea) animals and plants of which no
adequate account already exists in the text-books.
Probably most of the specialists who have taken part in
the L.M.B.C. work in the past will prepare accounts of one
or more representatives of their groups. The following
list shows those who have either performed or promised.

Memoirs from I. to XIII. have now been published.

Vll.

^0. XIY., on Ligia, \>j Mr. C. Gr. HeAvitt, will be out in a
few weeks. It is hoped that Cycloporns, Cancer, Antedon,
and llie Oyster will follow soon.

Memoir I. Ascidia, W. A. Herdman, 60 pp., 5 Pis., 2s.
,, II. Cardium, .J. Johnstone, 92 pp., 7 Pis., 2s. 6d.
,, III. EcHixus, H. C. Chadwick, 36 pp., 5 Pis., 2s.
,, lY. CoDiUM, R. J. H. Gibson and Helen Anld,

26 pp., 3 Pis., Is. 6d.
„ V. Alcyonium, S. J. Hickson, 30 pp., 3 Pis., ls.6d,
,, VI. Lepeophtheirus and Lern^a, Andrew Scott,

62 pp., 5 Pis., 2s.
„ YII. LiAEUs, R. C. Punnett, 40 pp., 4 Pis., 2s.
,, A'lII. Plaice, F. J. Cole and J. Johnstone, 260 pp.,

11 Pis., Ts.
,, IX. Chondrus, 0. v. Darbishire, 50 pp., 7 Pis.,

2s. 6d.
,, X. Patella, J. E. A. Davis and 11. J. Fleure,

84 pp., 4 Pis., 2s. 6d.
„ XL Arenicola, J. H. Ashworth, 126 pp., 8 Pis.,

4s. 6d.
,, XII. Gtammarus, M. Cussaus, 55 pp., 4 Pis., 2s.
,, XIII. Anurida, a. D. Ininis, 107 pp., 8 Pis., 4s.
,, XIY. Ligia, C. Gr. Hewitt.

Cycloporus, F. F. Laidlaw.

Cancer, J. Pearson.

Oyster, \V. A. Herdman and J. T. Jenkins.

Antedox, H. C. Chadwick.

OsTRACOD (Cythere), Andrew Scott.

Arciiidoris, Sir Charles Eliot.

BucciNUM, W. B. Randies.

Bugula, Laura R. Thornely.

ZosTERA, R. J. Harvey Gibson.

Himanthalia, C. E. Jones.

Pectp:n, W. J. Dakin.
Diatoms, E. E. Weiss.
Eucijs, J. B. Earmer.
BoTEYLLOiDEs, W. A. Herdmau.
CuTTLE-EisH (Eledone), W. E. Hoyle.
Actinia, J. A. Clubb.
Hydeoid, E. T. Browne.
Calcareous Sponge, E,. Hauitseh.

lu addition to these, other Memoirs will be arranged
for, on snitable types, such as S<igitta, a Cestode (by
Mr. vShipley), Pac/uius, and a Pycnogonid.

As announced in the preface to Ascidia, a donation
from Mr. F. H. Gossage, of Woolton, met the expense
of preparing the plates in illustration of the first few
Memoirs, and so enabled the Committee to commence
tiie publication of the series sooner than would otherwise
have been possible. Other donations received since from
Mr. Gossage, from Mrs. Holt, aud from others, are regarded
by the Committee as a welcome encouragement, and have
been a great help in carrying on the work.

W. A. Heedman.

University of Liverpool,

September. 1906.

L.M.B.C. MEMOIRS.

No. XIII. ANUEIDA.

BY

A. D. IMMS, M.8c. (Birm.), B.Se. (Loud.),
Christ's College, Cambridge.

CONTENTS.

I. Introductory Remarks
II. Bionomics.

III. Geographical Distribution

IV. Morphology

1. External Features

2. Integument and Colour

3. Mouth Parts

4. Digestive System .

5. Muscular System .

6. Ventral Tube

7. Cephalic Glands, etc.

8. Nervous System

PAGE

3

5
11
12
12
15
17
21 I
27
30
34
37

]
9. Sensory Organs .

PAGE

41

10. Circulatory System .

44

1 1 . Excretory System

46

12. Reproductive System

50

V. Embryology

56

VI. General Remarks on

Collembola

62

VII. Affinities of Collembola

71

Vlll. Marine Insects

75

IX. Literature .

80

X. Appendix

87

XI. Explanation of Plates .

90

PEEFACE.

Although the Insecta have received the attention of so
hirge an army of workers in many lands, very few have
deemed it worth their while to search for these animals
below the level of the high-water mark of the sea. It is
very generally believed that insects are totally unfitted to
endure the salinity of the sea-water and, consequently,
they have been very little sought for in that habitat.
The number of marine insects which are at present known
is, therefore, comparatively small.

The subject of this Memoir is the C'ollembolan
Anurida maritima (Guer.) and it has, perhaps, more to
commend itself to the attention of the general student of
Zoology than is the case with other marine insects. It
has not been my intention to confine myself solely to a
description of the type with which the Memoir deals, but
rather to make the work of a wider interest. With this
aim in view, I have included chapters on the general
structure and affinities of the Collembola, together with a
fairly extensive bibliography of the order, and some
additional remarks relating to other marine insects.

This work was commenced at the Port Erin Biological
Station, where the insects were collected and their habits
studied. The greater part of the anatomical work was
carried out in the Zoological Laboratory of Birmingham
University, while the remaining portion has been
completed in the Morphological Laboratory of the
University of Cambridge. I am indebted to Prof. E. "VY.
Wace Carlier, M.D., for helpful assistance as regards the
details of histological technique which were carried out in
the Physiological Laboratory at Birmingham. Dr.
David Sliarp has also supplied me with several valuable
suggestions.

It has been necessary to consult a considerable
amount of literature while preparing- this Memoir, and I
wish to acknowledge my indebtedness more particularly
to the writings of Fernald, Folsom and Willem. They are
numbered 8, 10 and 27 respectively in the bibliography
to be found at the end.

I. INTEODITCTOHY EEMAEKS.

The general characters of the class Insecta are familiar
to every student of Zoology. Among this vast assemblage
of animal forms, there is an obscure group of great
phylogenetic importance possessing certain definite
characters which separate it from the other members of
that class. Tntil comparatively recently, it has been
customary to regard this group as constituting a single
order, viz., the Aptera, and the latter was divided into the
two sub-orders CoUembola and Thysanura. It was in the
year 1885 that IJrauer recognised that the Aptera
possessed characters which merited their being raised to
something more than ordinal rank. He, therefore,
proposed to call them Apterygogenea in contradistinction
from the rest of the Insecta, which he termed
Pterygogenea. The investigations of the last thirteen
years all tend to establish Brauer's contention. In
preference, however, to his terms, the shortened names of
Apterygota and Pterygota are in more general i\se.

The following characters, taken collectively,
separate the Apterygota from the rest of the Insecta or
Pterygota : —

(1) The absence of any true process of metamorphosis.

(2) The total absence of wings.

(3) The presence of an evident pair of mouth-parts,

the maxilhdse, intercalated between *he
mandibles and first maxillae.

4

The AjDterygota are the most ji^eneralized of all
Insecta, and in their external form they resemble the
larvie of many of the Pterygota, or transient stages
in the larval development of others. A very marked
characteristic of the group is the existence of abdominal
appendages, either in the form of terminal processes at
the end of the body, or as processes in relation with the
sterna of the abdominal segments. The mouth-parts are
mandibulate, except in the few instances where they have
become imperfectly suctorial.

It is true that wingless forms are by no means of
infrequent occurrence among other groups of Insecta, but
in them the apterous condition is a secondary acquisition,
and not an ancestral character, as is the case in the
Apterygota. Furthermore, maxillulee may also be present
in other insects, but in the few instances where such
are known to occur they are found only in a vestigeal
condition.

The Collembola and Thysanura each possess certain
very marked characters, and are best regarded as two
sej)arate orders of insects.

The Collembola are popularly known as " spring-
tails." In them the antennae have from four to six joints,
and the abdomen consists of six segments and is never
terminated by any appendages. The first abdominal
segment always bears a ventral tube, and the fourth
segment usually carries a springing organ. The mouth-
parts are always sunk within the head, the nervous
system consists of a brain and four ventral ganglia, and
there are no Malpighian tubes.

The Thysaxuea are commonly known as " bristle-
tails." They have many-jointed antennae, and the
abdomen consists of ten segments and is provided with a
variable number of pairs of short, unmodified limbs. The

last abiloniiiial segnient is teriiiiiiated either by caudate
or forcipate appendages. A ventral tube is never present.
The mouth-parts may, or may not, be sunk within the
head, the nervous system consists of a brain and ten
or eleven ventral ganglia, and Malpighian tubes are
generally present.

The relations of the Collembola to the rest of the
Insecta are shown in the table on page G.

The Collembolan described in the present Memoir is
a convenient insect to study, as it can be obtained in
plenty in various localities scattered around the coasts of
the British Isles. According to Plateau it was first
discovered in IJeninark by Strum, who named it Poduva
amhulatoiia cauda carens. It was afterwards found in
Xormandy by Guerin-Menneville, who, in 18-jG, re-named
it Achonites maritimus \ Nicolet, however, refers to it as
Anouvd nuu-itima. In 1804 Laboulbene erected for its
reception the genus Anurida, but in Lubbock's " Mono-
grajih of the Collembola and Thysanura " it is described
as Liinira nitiritima. At the present day its nomenclature
has become settled, and it is now universally known as
A iiurida via rit hn a .

Those who have not the opportunity^ of working at the
sea-side can use with advantage the common Podura
aquatica as a substitute for Anurida. Podura is a very
jjlentiful species in many parts of Britain, and is to be
sought for on the surface of the water of waj^side ponds
and ditches.

XL BIONOMICS.

Anurida niaiitima is a dark blue-black insect measur-
ing 2-'5mm. in length. The individuals seen in fig. 2 show
the general appearance of the species to the naked eye, and
will aid in recognising it as it crawls about on the sea-shore.

A. APTERYGOTA.

1. COLLEMBOLA.

Spring-tails.

2. Thysanura.

Bristle-tails.

o

o

H - S
so ?= l-h

>
td
O
f

B. PTERYGOTA.

3. Dermaptera.

Earwigs.

4. Orthoptera.

Cockroaches, Locusts,
Crickets, etc.

5. Platyptera.

Book-lice, Termites,
Stone-flies, etc.

6. Thysanoptera.

Thrips.

7. Plectoptera.

Mayflies.

8. Odonata.

Dragonflies.

9. Hemiptera.

Bugs, Cicadas,
Green-fl}', etc.

10. Neuroptera.

Alder-flies,
Lacewing-flies, etc.

11. Coleoptera.

Beetles.

12. Trichoptera.

Caddis-flies.

13. Lepidoptera.

Butterflies and Moths.

14. DiPTERA.

Flies, Fleas,
Sheep-ticks.

15. Hymenoptera.

Ants, Bees,Wasps, Saw-
flies, &c.

HOMOMORPHA.

Young animal re-
sembling adult in
general form but
distinguished by the
absence of wings
and external genital
armature.

Heteromorpha.

Larva? dift'ering
very greatly from
the adults and al-
ways passing into
a pupal stage.

>

^
p
^

The habits of Aiuwiihi have been observed by the
pieseut writer principally on the shores of Port Erin Bay
in the Isle of Man.* About the rock-pools in that locality,
when the tide is low and the weather calm and bright,
numbers of this insect are generally to be seen. It may
be found both on the surface of the water and on the
surrounding rocks and weed. Anurida, unlike the
majority of the Collembola, is unable to perform any
leaping movements on account of the atrophy of the
springing-orgaii, and it is only capable of a moderately
active crawling motion. Those individuals which are to
be seen floating on the rock-pools are only able to walk
over the surface-film while the water remains motionless.
AVhen, however, the slightest ripple arises they become
drifted about from one part to another in a helpless con-
dition until thev are enabled bv chance to cling" to the
sides of the pools and crawl out on to the surrounding
rocks and weeds. Frequently numbers get blown together
into little groups on the surface of the water, and the
individuals composing them maybe seen actively crawling
over one another and apparently endeavouring to reach
the shore. Sometimes these groups are of a considerable
size and may consist of a hundred, or more, individuals.
There is every reason to believe that their occurrence on
the surface of the water is not due to the insect voluntarily
resorting to that medium, but that it is merely the result
of accident. Numbers of the insect frequently fall nto
the pools in their attempts to climb steep rock surfaces,
and also very many get blown thither by gusts of wind.

When the weather is cold and stormy, or otherwise
unfavourable, very 'few Anurida are to be found, and
often they appear to be absent altogether. At these times

* On the Continent the habits of the insect have been studied to
some extent in France by Laboulbene (2) and Moniez (4).

8

tliey seek skelter under weeds, or eusconce tkemselves in
crevices among- tke rocks or in crannies and nickes
between tke encrusting barnacles, and specimens are tken
to be obtained onh^ after a careful searck.

Tke distribution of tkis insect about Port Erin Bay
is confined to a tolerably definite zone of tke sea-skore.
Tke insect does not wander above tke kigk- water mark of
ordinary tides, neitker does it appear to venture furtker
seaward tkan about kalf-way between tkat limit and tke
Laminarian zone.

It is plentiful during spring and summer, but
towards autumn it becomes more sluggisk in its kabits
and decreases in its numbers. In winter it is not often seen.

As tke tide rises Anuiida retreats far into tke nickes
in tke rocks, wkere it is able to witkstand submersion for
a considerable period twice eack day. Since tke average
tide in Port Erin Bay is about fifteen and a kalf feet, it
would be immersed to a deptk of about eigkt feet. Tke
rocks round tke skorc of tkat locality are very markedly
jointed, and it is in tke crevices of tkese joints tkat
Anurida seeks refuge from tke incoming tide. An
abundant supply of specimens can usually be obtained by
following down tke retreating tide and examining tke
rocks as tkey become uncovered. Since tke insects work
tkeir way into tke crevices often to a deptk of four or five
inckes, it is necessary to split open tke rock witk tke aid
of a kammer and a stout steel cliisel. In suitable ckinks
as many as several kundred Anurida will sometimes be
found congregated togetker and, strewii about tke
crevices among tkem, tkere are usually to be seen numbers
of tkeir exuviae, wkick appear like minute wkite fiakes.
Tke relatively large orange-yellow eggs of tke species,
arranged in little irregular clusters, are often to be met
witk at the same time.

9

If a specimen of the insect be examined with a hand
lens, it will be seen to be covered with a coatinf>' of whitish
hairs. When the animal is submerg-ed under water, a
sufficient supply of air is retained by this hairy coating to
enable respiration to go on freely for a considerable
period, and at the same time it renders the insect
incapable of being wetted. In order to test the truth of
this, it is only necessary to take a few specimens and place
them in a deep glass vessel together with some sea-water.
If the vessel be shaken sufficiently, the animals will
become submerged below the surface of the water, and
each individual will then be seen to be enveloped in a
glistening coat of air. When once they are submerged
they are unable to reach the surface again, and they
crawl aimlessly about the bottom of the vessel. The
animals thus treated remained active for four and a half
days whenever the vessel was slightly shaken. On the
hfth day they had used up all the air investing them, and
on the sixth day scarcely anj^ movement was noticeable
among them, and they appeared to be in an asphyxiated
condition. On the seventh day they seemed to all
purposes to be quite dead.

In sandy localities Anurida resorts to burrowing for
protection from the incoming tide. In the neighbourhood
of Rhyl, jN'orth Wales, it occurs plentifully at low water
about the sands and on the surface of the pools left by
the retreating tide among the depressions and inequalities
in the shore. Since there are no rocks or other means of
shelter at hand, the insect has to bury itself in the wet
sand as the tide rises and to remain there until the next
ebb. It also appears to live under somewhat similar
conditions at Treport in Xormandy, for in that locality it
is mentioned as frequenting the mud at the mouth of a
rivulet. About the sandy beach on the north side of

10

Long- Island, Mass., there are found, according to
Davenport, (Ij three species of Collenibola beloug-ing-
to as many genera. Anurida, however, is the least
common among them in that locality. Among other
interesting observations, this author has proved that the
Collembola burrow to escape from the flood tide ; he
remarks that, with a hand lens, he has observed them as
they issue from the sand during the fall of the tide.

In regard to the nature of the food of Anuiida.
Folsoni (10) states that it feeds on the soft tissues of the
Mollusc Littorina littorea. Laboulbene (2j also stated
that its food consists of the tissues of Molluscs. He
remarked: — " Les Acliondes [/.f., Anurida^ mangent
evidemment des petits MoUusques, si abondants sur les
roehers submerges a la maree haute, car lorsque j'ecrasais
un de ces MoUusques pendant mes recherches, je voyais,
au bout de cinq ou six minutes, un bon nombre
d'Achorutes qui se reunissaient sur cette proie, et, par ce
moyen, je pouvais en prendre jusqu'a une douzaine a la
fois." In the experience of the present writer, its food is
dependent upon the nature of the locality which it
frequents, and, consequently, varies a good deal. In the
rocky locality of Port Erin it subsists chiefly upon the
dead bodies of various small marine animals, principally
Mollusca and Crustacea. At times numbers of Anurida
may be seen congregated around, and also inside, the
shells of dead barnacles, actively engaged in feeding upon
the remains of the latter. Vegetable matter appears also
to enter into its food occasionally, for remains of Desmids
and other green Xlgse may be found at times among the
contents of the mid-gut. In barren, sandy localities the
insect has to rely for its food upon whatever dead organic
matter that is obtainable.

11

III. GEOGRAPHICAL DISTRIBUTION.

Aniu'ida maiitima has been recorded from a number
of localities scattered all aroimd our British coasts. It
appears to be abundant in the neighbourhood of Penzance
and Land's End (Marquaud, 75), it also occurs at
Plymouth (Parfitt, 80), and the present writer has met
with it sparingly at Torquay. On the coasts of Wales it is
known from Gower, jN'evin and Llandudno,* and it is also
common on the broad stretches of sand in the neighbour-
hood of Rhyl. On the west coast of Scotland it is
described as being common among seaweed-covered rocks
at Peaton, Loch Long (Evans, 60) ; it is also found on the
shores of Ayrshire and Buteshire (Boyd, 52). On the east
coast Lubbock (Lord Avebury) mentions it from St.
Andrews, and it is plentiful about the shores of East
Lothian at Aberlady and Xorth Berwick (Carpenter and
Evans, 57). In the Isle of Man it is abundant among the
rocks on the north side of Port Erin Bay, and can also
be met with at Poolvaish near Castletown. In Ireland it
occurs on the east, south and west coasts (Carpenter and
Evans), and Lubbock mentions having received it from
Ivinsale.

On the continent of Europe Anui-'uJa is plentiful in
Erance at Treport, Havre, Boulogne and other places,
together with the Channel Isles. It has also been recorded
from Holland, Belgium, Denmark, Heligoland and
Scandinavia.

In Xorth Americat it is reported from the Atlantic
coast from Cape Ann, Annisquam, Long Island, iN'ew

* For these localities, the writer is indebted to Dr. D. Sharp, P.R.S.,
for Gower, and to ISIr. F. A. Potts for Nevin and Llandudno.

t Packard (79) states that the American examples of Anurida
have been examined by Lubbock and found to be identical with the
European form.

12

Haven, Xantueket and AVood's Holl, and it extends as far
south as the shores of Ehjrida.

From the above remarks it will be noted that Anurkhi
marithiKi is a form ranging into both the Pakearctic and
Xearctic regions. This wide distribution of an animal
with practically no means of self-dispersal is remarkable.
Folsoni (63) believes that marine currents have been the
principal agents for its distribution. They would
certainly account for its presence on the barren rocky
islands off the Scandinavian coast and at Helio-oland and
the Isle of Man.

An allied terrestrial S2)eeies, Anurida (Aphoroinina)
(/i-anaiid. (Xic), has likewise a very wide geographical
distribution. It has a very extensive range in Europe,
being known from Grreat Britain, France, Scandinavia,
the Tyrol and Bohemia, and it is also recorded in Arctic
regions from Spitzbergen, Franz-Tosef Land and North
Siberia. ^4. tuUhergi, Schott, w^hich frequents the surface
of ponds of fresh water, is found in northern Europe and
has also been recorded from the United States. Another
species, A. clavata, Scliaif., is found in Tierra-del-Fuego ;
A. amorita, Fols , is known from Alaska and Siberia, and
A. steineni, Schiitf, is a South Georgian species.

IY._MORPHOLOaY.

1. GrENEKAL FEATURES OF ExTEKNAL AnATOMY.

The general appearance of Anuridd when viewed
under the lower power of the microscope will be seen on
referring to fig. 1. As in all insects, the body consists of
three well defined regions, viz., the head, the thorax and
the abdomen. Its exo-skeleton is but feeblv chitinised,

13

and consequently the different regions do not exhibit the
complex differentiation into the vorions selerites which is
seen among- the Pterygota.

The head (fig. 4) is relatively large in size and has
its long axis disposed in the same plane as the rest of the
body. In form it is somewhat triangular, and it is con-
siderably flattened, as if it had been compressed in the
dorso-ventral direction. In the mid-dorsal region of the
head there is situated a verv prominent elevation, shaped
rather like an inverted isosceles triangle having each of its
angles truncated. From the antero-lateral corners of this
protuberance arise the antennre, and on either side of it
is a well-marked oval elevation, bearing on its surface a
group of glistening black dots. These dots are the eyes of
the animal (figs, i and 5) , they are five in number on each
side. A curious circular sense organ, known as the post-
antennal organ (j>.((.o. in figs 4 and 5) lies on each side
just in front of the protuberance just mentioned. The
labrum and clypeus (fig. 1')) jut forwards in front of the
head som.ewhat in the form of a beak; they form the roof
of the pharynx and somewhat overhang the mouth. The
antennae are relatively short, being scarcely as long as the
head; they each consist of four joints, the second being
the largest. The articulation between the third and fourth
joints is but imperfectly developed, and is only visible on
the ventral aspect (fig. -'5). The extremity of each
antenna is abundantly clothed with hairs, aiid situated
among the latter is a minute trilobed sense-organ (fig. 7) ;
a short distance below this structure are several patches of
modihed cuticle (s'.o'.) which may perhaps also be of a
sensory nature. The mouth is slightly ventral in position ;
it is bounded dorsally by the labrum, ventrally by the
labium, and its sides are formed by the oral folds which
are prolongations of the sides of the head. The ventral

14

region of the head appears to be formed almost entirely
by the second maxillse or labium.

The thorax consists of three segments, viz., the
prothorax, the mesothorax and the metathorax (figs 1
and 3), the first being the smallest division of the three.
The three pairs of legs ditfer in no important features
from one another. Each leg is six-jointed and terminates
in a single claw, which, in the majority of individuals, is
armed with a small tooth on its inner margin (fig. 9).
Tarsi are absent in all Collembola, and, consequently, the
joint with which the claw articulates corresponds with
the tibia of other insects.

The abdomen consists of six segments, the third
segment being greatly enlarged. On the ventral aspect
of the first segment is situated a prominent bilobed papilla
known as the ventral tube {v.t. in fig. •'!). This organ is
the most characteristic of Collembolan structures, and it
varies a great deal in the degree of development to which
it attains in various genera. There arises from a point in
the mid-ventral line of the head a narrow furrow, which
is termed the ventral groove or linea ventralis (fig. ■')) ;
this groove is traceable backwards, passing between the
bases of the legs to the anterior aspect of the ventral tube,
where it terminates (Plate TV., fig 36). As will be
explained in a later chapter, the secretion of two pairs of
glands situated in the head flows down the ventral groove
and discharges over the surface of the ventral tube, which
it serves to moisten. The genital aperture is situated on
the ventral surface of the fifth abdominal segment near to
its posterior margin, and on the sixth and last segment is
situated the anus (fig. 3). The latter is surrounded by
three papillae, one of which is median and dorsal, and the
other two ventro-lateral.

There is no sexual dimorphism among the Collembola,

15

and the only difference between the sexes in Anurida is
that the females are, as a rule, somewhat larp^er and more
bulky than the males and, moreover, the ripe ovaries
usually show throuf^h the lesser pigmented cuticle on the
ventral side as a pair of yellow patches, liy the latter
character the females can generally be recognised with a
hand lens.

2. Integument and Colouration.

The integument consists of (a) the chitinous cuticle,
(b) the hypodermis or chitogenous layer, and (c) a base-
ment membrane {^vule Plate II., figs. 16 and 17).

The cuticle varies slightly in thickness in various
parts of the body, being thinner on the appendages than
elsewhere. It is quite colourless, and its surface is thicdvly
studded with minute, somewhat conical tubercles, which
are very evenly spaced apart (fig. G). Over the apex of
the labium these tubercles are replaced by irregular
protuberances of the cuticle, but all transitions are
observable between the two (fig. 12). On the labrum the
tubercles are also modified, and they exhibit a tendency
to be arranged in horizontal rows (fig. 1-3). The
tubercles are absent from the inner aspect of the basal
half of each of the legs, and scattered over the abdomen
are numerous small areas from which they are likewise
wanting; these latter areas correspond with the positions
of insertion of the tergo-sternal and other muscles
(Plate YI., fig. 60). The cuticle, when viewed in very
thin sections, is seen to consist of two layers, the line of
separation between them passing just below the bases of
the tubercles. These layers are best differential ed in
preparations stained with Mann's methyl-blue-eosin,
when the inner stratum of the two stains a brilliant blue,
while the outer one has a stronger affinity for the

16

eosin and appears pinkish in colour (figs. 16 and 17).
Interspersed among the tubercles are numerous hairs, some
of which are small and pilose, while others are larger and
stouter and rather of the nature of setse. The latter kind
are specially prominent about the sides of the head and
on the lateral margins of the abdominal segments (figs.
1 and -j). Both kinds of hairs arise from a circular base
(fig. 6), which, when viewed in section, is seen to be in
reality a cup-like depression in the cuticle (fig. 17).

The hypodermis, or chitogenous layer, is everywhere
of the nature of a syncytium, no cell boundaries being
distinguishable. Its cellular nature is evident, however,
by the presence of numerous large oval nuclei. It contains
a great quantity of a very dark indigo-blue pigment
[l). in fig. 16), and it is to this material that the
colouration of the animal is due. The pigment is
deposited in the form of minute granules and, except in
very thin sections, it appears black, and frequently
greatly obscures the nuclei. On the ventral aspect of the
body, which is less exposed to the influence of light, the
pigment is not quite so abundant as in other parts. When
a number of freshly-caught examples of Anurichi are
placed in a small quantity of strong alcohol a little
of the pigment becomes soluble in the latter and imparts
tf) it a faint bluish-green tinge. The bulk of the
pigment, however, appears to be incapable of being
extracted by most of the ordinary reagents. It is not
dissolved out by water, neither does ether nor strong
hydrochloric acid aifect it, and Fernald (8) remarks that
when sections of the animal were treated with 45 per cent,
nitric acid for ten minutes the latter failed to remove the
pigment. Weak solutions of strong alkalies turn it a
bright reddish colour. Distributed at intervals in the
hypodermis are large, pear-shaped cells (//. cell, in fig. 17),

which are situated not far from the bases of the hairs.
They ore the trieliogenoiis or liair-formiiifif cells; their
positions are easily recognised by their very large
nuclei.

The basement membrane (J).m. in figs. 10 and IT) is
an extremely thin and apparently structureless layer ; it
is easily seen in preparations stained with methyl-blue-
eosin, and appears bright blue.

3. The Mouth Parts.

The structure and development of the mouth-parts of
Aniirida have formed the subject of an elaborate paper
by Folsom (10), and much of the following account is
based on his memoir.

They consist of (1) the labrum or upper lip, (2) the
mandibles, (•)) the maxillula>, (4) the lingua or tongue,
foj the first maxilla^, and ((ij the second maxillw or
labium.

The principal mouth-parts are remarkable in that they
are deeply insunk within the head-capsule and in that they
are capable of being partially protruded from it when
in use (Plate Y., fig. 52). This deep-seated position of
the mouth-parts is a secondary condition, and has been
brought about in the following manner. In the embryo
the sides of the facial region of the head develop from
two lateral evaginations of the germ band. These
evaginations eventually fuse with the developing
fundaments of both the labrum and labium, and in this
way form a kind of enclosing box which, by further
growth, comes to surround all the remaining mouth-
parts. By this means, the latter appear in the adult
insect as if they were deeply pushed back into the cavity
of the head. This condition of the mouth-parts is

c

18

eminently characteristic of the Collembola and, with the
exception of certain Thysannra, it is found nowhere else
among insects.

The labrum (fig. 13) forms the dorsal or anterior wall
of the mouth-cavity. It is seen to be somewhat
rhomboidal in form, and on its dorsal surface are
distributed some stifiish hairs, whicli are probably of a
sensory nature. Situated immediately behind the labrum
is the clypeus (clyj).), which is separated from it by means
of a flexible suture. This suture functions as a hinge,
and admits of a certain amount of movement for the
labrum in the vertical plane. Posteriorly, the clypeus is
separated from the epicranial region of the head by a
shallow groove, while laterally it merges into the sides
of the head {j>l.). The posterior or pharyngeal surface of
the labrum is lined with a soft membrane; the latter,
however, does not exhibit any special chitinizations,
either in the form of hairs or gustatory organs, etc.,
which would constitute an " epipharynx."

The mandibles (figs. 11 and 15) are elongate tubular
structures, which are flattened and somewhat expanded at
their apices and provided with five sharp teeth, of
which the last is the largest. At the base of each
mandible is situated a large opening (Plate II., fig. 15,
cav.) through which muscles pass to be inserted on the
walls of the mandibular cavity. With the exception of
its basal attachment, each mandible lies free within a
pocket-like evagination of the pharynx (Plate IV.,
fig. 35). The mandibles of Anurida differ from the usual
Collembolan type in the absence of a masticatory surface.
This feature is correlated with the soft and unresisting
nature of the food in this species, which necessitates
rather the use of a cutting edge than a crushing
surface.

19

The maxillulae* (fi^s. 11 and 14) are intercalated
between tlie mandibles and tbe first maxillae. They are
situated immediately dorsad of the lingua, to which they
are closely applied. In form the maxillulse are broad
and plate-like, but are feebly chitinised and relatively
delicate structures. Along their anterior borders they
are finely serrated, and they each bear a tooth-like
projection at their outer angles. Unlike the first
maxillge, at no stage in their development do the
maxillulee exhibit any indications of a differentiation
into a ffalea and lacinia, neither are there anv traces of
palpi.

The lingua or tongue is a median unpaired structure,

and is moderately strongly chitinised (figs. 11 and 14).

It has a paired appearance owing to being lobed

anteriorly and in being grooved dorsally down the middle

line. Posteriorly, the lingua rests on two strong

chitinous props, which are termed the lingual stalks

(ling. St.). The latter structures are not, sensu stricto, parts

of the tongue ; they are developed as longitudinal thicken-

* Most writers on the Collembola have referred to these organs
as paraglossse. The latter name, however, is undesirable since it
renders them liable to confusion with parts bearing the same name
in the second maxillae of the Pterygota. Hansen [Zool. Anz., 1893)
was the first to recognise that they represented a complete and separ-
ate pair of gnathites and he termed them maxillulcv. Seven years
later the embryological work of Folsom (10) definitely established the
truth of Hansen's conclusions. On account, however, of their
intimate relations with the lingua, Folsom named them superlingnce.
In the generalized CoUembolan Isotoma palusfris Carpenter has
recently shown (Proc. Roy. Irish Acad., 1903, vol. xxiv. ser. B. p. 325)
that the maxillula; are of a very primitive nature, each being sub-
divided into a galea and laciuia which have become lost in Anurida.
Among the Thysanura the maxilluke are likewise differentiated into
a galea and lacinia, and a palp is present in addition. In the Pterygota
these gnathites have, as yet, been very little studied and they appear
to be either greatly reduced and very intimately associated with the
lingua or tongue (Orthoptera and Plectoptera) or atrophied altogether.
In the earwig-like genus Hemhnenis the maxillulse appear to approxi-
mate to the apterygote condition very closely (Hansen, Ent. Tidshr.,
189J-, p. 65). Folsom (10) regards" the '"• hypopharynx " of the
Pterygota as being a composite structure formed by a fusion of the
maxillulse with the tongue.

20

ings of the maxillary pockets, wliicli acquire a secondary
connection and ultimate fusion witli the lingua.

The first maxillse (figs. 11 and 14) each consist of an
apical portion or " head," which articulates by means of a
hinge joint with a long stalk-like stipes (5/.). Each stipes
in its turn articulates with a shorter piece, which is
termed the cardo (car.), whose opposite extremity is
attached to the proximal end of the lingual stalk of its
side. The " head " of the maxilla consists of three parts.
The uppermost of them is a thinly chitinised mem-
branous lamella, which is finely serrated along its inner
margin and appears to correspond to a greatly modified
palp (pJj).). Underneath the palp is situated the galea
(gal.) ; it is the most massiye piece of the three, and is
armed with strong teeth. The third piece is the lacinia
(lac.), which very closely resembles the palp in form, but
is ratlier more deeply toothed. The existence of a palp in
the adult insect seems to have been overlooked by Folsom,
notwithstanding the care with which he studied the
mouth-parts and the fact that it has been figured by
Fernald (8, pi. xlviii., fig. 9). The latter author, however,
made no suggestion as to its homologies. Its presence in
the embryo Aniirida is described and figured by Folsom,
who adds that in the newly-hatched insect no trace of a
palp can be made out, and he concludes that it must have
been resorbed. The palp is freqiiently difficult to make
out as it usually lies closely apposed to the sides of the
galea, and in this position it is easily overlooked on
account of its extreme transparency. Like the mandibles,
each maxilla lies in a separate pharyngeal pocket, which
is situated somewhat ventrad of the mandibular pocket of
its side fPlate lY., fig. 35).

The second maxillae or hibium form practically the
whole of tlie ventral region of the head. Although they

21

develop from paired fuudanients in the embryo (Plate
VII., tig. To), almost all traces oi' such an origin become
obliterated in the completed structure. Its distal
extremity, however, is seen to consist of two halves, which
are in close contact with one another along the mid-
ventral line without being fused together (fig. 12, Ih.).
Each half is to be regarded as the equivalent of an
undifferentiated galea and lacinia. Labial palpi are
present in the embryo, but they subsequently become
resorbed. The remaining portion of the labium extends
from the suture to the anterior margin of the pro-
thorax, and would appear to be the equivalent of an
undifferentiated mcntum, sub-mentum, and gula.

While feeding is taking place, both the mandibles
and first maxillie are protruded from their pockets for
the purpose of seizing the food, and they can be seen with
the aid of a strong lens to be projecting a short distance
through the mouth. By means of the retraction of these
parts, aided by the movements of the labrum and labium,
the food is taken into the mouth. Mastication is chiefly
performed by the combined action of the mandibles and
first maxillse. The part played by the maxillulse appears
to be a more or less passive and subsidiary one, for, since
they are closely applied to the lingua, they possess but
little power of movement. When the food is finely
comminuted, it is sucked up into the cBsophagus by means
of the alternate contraction and expansion of the radial
and circular muscles of the pharynx.

1. The Digestive System.

The digestive system is divisible into three well-
defined regions, viz., the fore-gut, the mid-gut or stomach,
and the hind-gut. It passes in a perfectly straight course

22

from the mouth to the anus, and is remarkable on account
of its extremely simple structure. AVith the exception of
a pair of salivary glauds there are no accessory organs in
association with it.

The fore-gut consists of the pharynx and the
oesophagus. The pharynx is a capacious cavity and
occupies about one-sixth of the total length of the fore-
gut. Its dorsal or anterior wall is formed by the labrum
and its ventral wall or floor by the labium or lower lip
(Plate III., fig. 27). The pharynx is evaginated into four
backwardly directed pockets, two on either side for the
reception of the mandibles and first maxillse (Plate lY.,
fig. -35). The lingua occupies a median position in the
floor of the pharynx, and the maxillulpe are closely applied
to the upper surface of the former.

The oesophagus is a narrow tube of a very uniform
calibre throughout its length (Plate III., fig. 27). At the
point where it joins the mid-gut it projects for a short
distance into the cavity of the latter in the form of a
papilla (Plate III., fig. 28). The lumen of the oesophagus
is roughly H-shaped in transverse section, owing to four
inwardly directed bulgings of its epithelial walls (fig. 29).

The walls of the fore-gut consist of three laN^ers ; the
innermost is a chitinous intima, the middle one is an
epithelium, and the outer one is muscular. The chitinous
lining, which is secreted by the underlying epithelial
layer, is continuous with the cuticle of the integument.
It is very thin and, with the exception of a few scattered
tooth-like projections in the region of the pharnyx and
the anterior part of the oesophagus, its surface is perfectly
smooth. Whether the epithelial layer is of the nature of a
true epithelium or not is uncertain. It has not been
possible to detect any cell-boundaries even after a
prolonged treatment of sections with suitable staining

23

reagents ; it appears as a syncytium containing' oval nuclei
dispersed at frequent intervals, and is similar in its
structure to the liypodermis, with which it is directly con-
tinuous (figs. 27 and 28). As Fernald remarks, there are
slight depressions in the chitinous lining which alternate
in position with the nuclei and may perhaps indicate the
boundaries between adjacent cells. Externally, the
epithelial layer rests on a delicate basement membrane
[b. m. in fig. 29), and immediately outside the latter is a
strongly developed layer of circular muscle fibres (figs. 27,
28 and 29). The structure of these fibres is remarkable;
they are all cross striated, and each fibre is in the form of
a ring completely encircling the gut, with its two ends
meeting in the mid-dorsal line. The sarcoplasm of each
fibre lies wholly external to' the contractile substance,
and in it is situated the nucleus of the fibre (fig. 29). The
nuclei of the fibres are all situated in the dorsal line, and
they form a continuous row, stretching right along the
fore-gut. Each muscle fibre is inverted externally by an
extremely delicate sarcolemma {sbnma in fig. 29).

In relation with the j)harynx and a small tract of the
adjacent portion of the oesophagus a remarkable system
of radial muscles is developed and, furthermore, the
circular fibres are more strongly developed in this region
than elsewhere. These muscles fall into two groups,
viz., a dorsal or anterior series and a ventral or posterior
series (Plate III., fig. 27). The muscles forming these
two groups take their origin in the cuticle of the body
wall. Those of the dorsal series cross the cavity of the
head in an oblique direction and, passing between the
circular fibres, attach themselves to the chitinous intima
of the gut. In the case of the ventral series, they cross the
head cavity in the horizontal plane, and converge to
become inserted into the tentorium. A similar system of

24

radial muscles occurs in Tomocenis, Orchesdla and other
Collembola ; tliey are termed tlie dihdores liharyiigil, aud
act in coiijuuctiou with the circular iibres aud convert the
pharynx into a kind of pumping organ.

The mid-gut commences in the posterior end of the
prothorax, and extends backwards into the fourth
abdominal segment, where it is joined by the hind-gut
(Plate lY., fig. -35). The lumen of the mid-gut is
practically circular in section, but its diameter narrows
considerably in the posterior fourth of its course. In
regard to its structure, it is very uniform throughout its
length, and its Avails consist of a layer of large epithelial
cells resting externally on a relatively stout tunica
jjropria, and on the outside of the latter is a system of
circular and longitudinal muscle fibres, both of which are
markedly striated (Plate Y., fig. 49, and Plate YI., fig. 61).
The epithelial layer consists of a row of very regular
cubical cells, which have a tendency to become somewhat
columnar. The boundaries between adjacent cells are
best seen towards the posterior region of the gut ; over
the rest of the area they vary greatly in their degree of
definition, and for the most part are hard to distinguish,
but their distinctness depends a good deal upon the
particular phase of physiological activity the cells hapjDeu
to be in at the time of killing and fixation. Each cell
contains a large oval or rounded nucleus situated near to
its centre ; the chromatin granules of the nucleus are
congregated together into several large masses (figs. 30
and 49). Towards the base of each cell there is, as a
rule, a deeply-staining granule of chromatic material
(ch. b in fig. -iO) ; the significance of this body is
unknown. The free margins of the mid-gut cells have a
finely striated appearance, which forms a " striated hem "'
or '' Harehensaum " (_hs. in fig. oU).

Fernald has described a process of regeneration which
takes phice iu the mid-gut cells, but it has not been possible
to conhrm his observations with any degree of certainty.
He remarks that ''The nuclei of the epithelium divide, and
one of the two that are thus formed in each cell passes
towards the free face of the cell, while the other passes
towards its base. The cells' walls now become indistinct
and delaminatiou occurs, the outer half of each cell being
thrown off. These moulted cells collect in the lumen of
the gut, and remain there until the chitin of the
remainder of the body is thrown off, when this is also
removed from the body. The outer edges of the cells
remaining- form a new * Harchensaum/ and resume
their normal condition '' (8. p. 455). A somewhat similar
process of moulting or regeneration of the mid-gut cells
has been described b}' Sommer in Tomocerus i^M acrutouxa)
(23), and by Prowazek in Isotoma (21).

The musculature of the mid-gut consists of two layers,
an outer longitudinal and an inner circular set (Plate \ .,
fig. 49, and Plate YI., hg. (jl). The circular hbres each
consist of a single greatly elongated cell, the opposite ends
of which meet in the mid-dorsal line. Previous to uniting,
the two ends of each fibre break up into their component
fibrillse or sarcostjdes, and also receive contributions from
adjacent fibres (Plate YL, fig. 61). The longitudinal
fibres consist of greatly elongated cells placed end to end.
The nuclei of both the circular and longitudinal fibres are
hard to detect ; they are very minute, and those of the
circular muscles are principally dorsal in position. A
fine membrane of connective tissue closely invests the
outer surface of the muscle fibres, and also lines the
rectangular spaces enclosed between them (fig. 30).
Where the mid-gut joins the hind-gut its epithelium
becomes greatly thickened, and forms a ring-like bulging

2G

AA'Lieli reduces the lumen of the gut at that point to
less than one-third of its usual diameter (tig. 49).
Surrounding the adjacent portion of the hind-gut is a
strongly developed pyloric or intestinal valve, which is
formed by the circular muscles at that point becoming
enormously thickened {valve c. iti.).

The hind-gut extends from the posterior border of
the fourth abdominal segment to the anus. It is lined by
a perfectly smooth chitinous cuticle, directly continuous
with that of the integument (cuf. in fig. 49), and under-
lying it is a cellular layer (h. g. cpith.), which rests
externally on a basement membrane {t. ij.). In the
anterior third of the hind-gut its epithelial layer is
greatly thickened, the nuclei are larger and cell-
boundaries are discernible ; over the rest of its course
this layer is relatively thin and of the nature of a
syucj^tium. Circular muscle fibres of the striped variety
are well developed in relation with the anterior half of the
gut, but they dwindle away as they proceed further
backwards. The cuticular and epithelial lining of
the gut is thrown into numerous folds and bulgings,
but the circular muscles do not follow their course
and large spaces are consequently left. (Plate III.,
fig. o\). These spaces are in free communication
with the general body cavity between the muscles,
and usually contain blood corpuscles. In addition to
circular muscles, there is also a system of radial muscles
in relation with the hind-gut ; they fall into two groups,
the dilatores recti and the dilatores ani (fig. 49, dil. red. in.
and dil. an. m.). The dilatores recti chiefly arise from ■
the middle section of the gut, where it is considerably
enlarged and forms a kind of rectal chamber. They
cross the cavity of the fifth abdominal segment in dorsal,
ventral, and lateral directions, and are inserted into the

27

cuticle of the body-wall. The dilatores ani are a similar
series of muscles, situated immediately around the anus.

The anus is situated on the last segment of the body,
and is surrounded by three papilloe, the largest of which
is median and dorsal and the other two ventro-lateral in
position (Plate I., fig. 3, an.).

5. The Muscti.ar System.

In order to explain the muscular system at all
adequately a very lengthy description would be necessar}',
and such an account is beyond the scope of this Memoir ;
only its general features, therefore, will be referred to.

The musculature of the abdomen may be broadly
divided into three systems. Firstly, a series of longitu-
dinal tergal muscles, which are arranged in the form of an
outer and an inner set on either side of the dorsal vessel
in each segment (/. t. m. in hgs. 50 and 51). Secondly,
and corresponding with these on the ventral side, a system
of longitudinal sternal muscles (/. 5. ?«.) ; and thirdly, a
system of tergo-sternal muscles, some of which pass
vertically and others obliquely from the terguni to the
sternum in each segment {t. s. viusc).

In the thorax this general arrangement is adhered to,
but it has undergone modification in conformity with the
requirements of the mobility of the legs.

In the head the musculature is very complex ; it is
principally concerned with the movements of the various
mouth-parts. In addition to the usual muscles in connec-
tion with those organs, there are special muscles for
drawing the mandibles and first maxillae forwards, and
partially protruding them through the mouth, and there
is also an antagonistic set for retracting these organs back
again into the cavity of the head. In Anurida more than

28

twenty pairs of muscles can lie made out in connection
with the mandibles and the first and second maxillte ; the
jaw muscles have been described in detail by Folsom for
the CoUembolan OrcheseUa, and the reader is referred to
his paper (9).

Excepting the circular muscle bands of the walls of
the heart, the fibres of both the somatic and splanchnic
muscles of Anurida belong to the striated variety through-
out. The principal muscles of the body consist of a large
number of very small fibres, and the latter do not exhibit
any tendency to be grouj)ed together into bundles. When
view-ed in transverse sections (Plate lY., fig. 45) the central
portion of each muscle is seen to be composed of a large
number of fibres, which appear as deeply staining angular
fields separated from one another by interstitial proto-
plasm. The periphery of the muscle is formed of a thick
layer of protoplasm, in which are situated one or more
relatively large, oval nuclei. The whole muscle is
invested externally by a limiting membrane or
perimysium. In preparations treated with Heidenheim's
iron-alum-liaematoxylin the minute structure of the
muscles can be made out with great clearness, and in this
resjDect the Collembola are very favourite subjects for the
study of the histology of this tissue. The typical light
and dark transverse stripes are seen w^ith extreme
clearness ; the dark bands or " sarcous elements " are sub-
divided longitudinally into extremely fine lines, which are
the interspaces between the fibrils or sarcostyles of the
muscle (Plate YL, fig. 60). Each light stripe is seen to
be bisected by an irregular transverse line [h. m.) which
indicates the position of the transverse or Krause's
membrane ; the segment of a sarcostyle contained between
two transverse membranes constitutes a sarcomere or
" muscle segment." By careful focussing by means of a

29

high power a transverse line can also be made out
bisecting eaeli of the dark bands, and is known as
Hensen's line.

At the points where the muscles are attached to the
cuticle the hypodermis undergoes some modification, and,
moreover, the pigment is absent from those positions. In
many instances the points of attachment of the muscles
to the cuticle is indicated externally, for over such areas
the characteristic tubercles are frequently not developed
(fig. 60).

The muscles of the walls of the alimentary canal
consist of single fibres, and their structure is referred to
in the account of tlie digestive system.

In the head and thorax there are developed endo-
skeletal structures for the purpose of giving- a firm
attachment to certain of tiie muscles. The endoskeleton
of the head is known as the tentorium (Plate III., fig. 27),
and consists of a chitinous plate lying parallel with the
frontal plane, from which diverge two pairs of chitinous
arms extending respectively to the dorsal and ventral
integument of the head. Between the dorsal arms lie the
oesophagus and supra-oesophageal ganglion above it, and
the ventral arms embrace the infra-oesophageal ganglion
(Folsom 10). To the tentorium are attached a great
number of muscles, including the posterior or ventral
group of the dilatores pharyngii, and many of the muscles
moving the mouth-parts. In the thorax the endoskeleton
consists of three apodemes, one to each segment, and which
are termed respectively the antefurca, the medifurca and
the postfurca. These give attachment to many of the
muscles of the legs, as well as lending support to the
nervous system, and take the form of chitinous pillars,
which project upwards from the sterna and bifurcate into
a pair of forks or arms (Plate IV., fig. o7). The medi-

30

fiirca and pnst-fiirca are better develnped than tlie
corresponding- part in the prothorax.

6. The Ye>;tral Tube or Abdominal Vesicle.

The ventral tube is a large papilla-like organ
situated on the middle of the ventral aspect of the first
abdominal segment (fig. 3). It presents a considerable
range of variation, both in form and in degree of develop-
ment among various Collembola, but in Anurida it
is relatively simple in structure and remains in a
rudimentary condition.

As is demonstrated by its development, the ventral
tube is formed by the fiision together of the pair of
abdominal ajipendages belonging to its segment [vide
text-fig. 1). It consists of a short basal column,
which is divided distally by a median longitudinal cleft
into two lobes or vesicles, and these latter are the only
indication in the adult of the paired origin of this organ
(Plate IT., fig. 3G). It is invested by the general cuticle
of the body- wall, but where the latter is prolonged over
the two terminal vesicles it becomes extremely thin and
flexible, its tubercles are wanting and it is quite smooth
(Plate II., fig. 21). In the structure of its walls it closely
resembles the general integument of the body ; they
consist of a well-developed layer of hypodermis, but
contain very little pigment and there is no basement
membrane. Where the hypodermis lines the lobes or
vesicles of the organ, it alters very considerably in
character; certain of its cells being very large, elongated
in a vertical direction, and tapering into a point {h.e. in
fig. 21). Each of these cells contains a large nucleus with
abundant chromatin granules.

Arising from the terminal lobes of the organ are
several muscles; they pass down its central cavity and

31

converge towards one another to be inserted into the body-
wall on either side {ret. muse, in figs. 21, 2o and 36). By
means of the contraction of these muscles the distal lobes
are capable of being withdrawn into the basal column
when the organ is not in use.

The cavity of the ventral tube is in direct communica-
tion with the general body-cavity of the animal, and
contains numerous blood corpuscles [b.c. in fig. 21), and
it is by means of the pressure exerted by the blood that
the protrusion of the vesicles is effected. When the organ
is in a retracted state its cavity is shut off from that of the
body of the animal by the contraction of the longitudinal
sternal muscles, which are disposed in two bundles
situated slightly to the outside of the organ on either side
(/. s. m. in fig. 38). The effect of the contraction of the
muscles is to approximate its anterior and posterior walls.
When the full expansion of the organ is desired both its
retractor muscles and the longitudinal sternal muscles are
relaxed. The slackening of the latter set of muscles
results in a sudden flnw of blood into its cavity, and in
virtue of the turgidity thus acquired its terminal vesicles
are protruded to their fullest extent.

With regard to the function of the ventral tube,
there has been, and is still, a great diversity of opinion.
Almost all investigators who have given any attention to
the structure of the Collembola have made suggestions
concerning it, and some of the views which have been put
forward appear to rest on a very slender basis. By
Kolenati (67) and Latreille* the ventral tube was
believed to be the external generative organ, and Claypole
(31) suggests the possibility of it being the relic of a
former outlet of the reproductive organs. Burmeisterf

* Now. Aiw. cVHist. Nat., t. 1, 183-2.
t HandbucJi der Entomologie, 1838.

32

believed it to be an apj^aratus for affording^ additional
support to the body. By Boiirlet+ it was thought to
serve to moisteu the springing organ as well as
serving as an adhesive organ on which the creature
might alight after springing, thus lessening the effect
of the shock of that movement. Renter (22) looked
upon it as being a kind of hydrostatic organ.
He believed that the claws of the feet collected moisture
from the hairs of the general surface of the body
and conveyed it to the ventral tvibe, which served to
absorb it into the system. Sommer (23) believed it to be
an adhesive organ, and suggested that it also performed the
same service for the integument as does the preen gland
for the feathers in Birds. Scliott (93) maintained a
somewhat similar opinion.

The view which has received the widest support is
the one which regards the ventral tube as being an
adhesive organ enabling the insects to walk over smooth
or steep surfaces. This opinion is upheld by Bourlet (in
part), de Olfers (19) and Tullberg (96), who believed
that its power of adhesion was brought about by
means of suction. De Geer', Nicolett and Lubl)ock,
however, thought that the adhesive function was effected
by means of a special secretion of the tube itself. Lubbock,
who studied the action of the organ in Sminfhurus, where
it attains its greatest development, remarks that if one of
these animals be laid on its back, and a piece of glass be
brought within its reach, it will endeavour to seize it witli
the feet, but at the same time it will protrude one or both
of the vesicles, emitting as it does so a miniite drop of
fluid, which, no doubt, enables it to obtain a better hold.

I Mem. Soc. cV Agric. du di^partment du Nord, 1841.

* Abhandlungen zur (Jeschichte der hisekten, Bd. vii., 1783.

t Rechercli.es ixmr Sifroir d I'histoire natiirelle defi Fodurelles, 1841.

33

Tlie view that the ventral tube is an adhesive organ
which prodnces a glandular secretion is also upheld by
Haaset, Uzel (38) and Prowazek (21), but these writers
regard it as being only a secondary function of the organ,
and assert that its principal significance is as an organ of
respiration. The thinness of its investing cuticle, and the
fact that it always contains a large amount of blood,
adapt it for aerating the latter and, therefore, lend support
to this contention.

The two most recent investigators, viz., Willem (27)
and Hoifniann (12) both agree in their opinion that
though it may fulfil the secondary function of being
an organ of respiration, its main function is to
serve as an organ of adhesion to aid the animal in
climbing. These writers believe that the secretion which
moistens the surface of the ventral tube is supplied by a
pair of glands situated in the head. The ducts of these
glands were first shown by Fernald to open by means of a
small pore into the commeu cement of the ventral groove.
Willem and Hoffmann believe that the secretion flows
down the ventral groove and reaches the ventral tube and
bathes the surface of the latter. The observations of
these writers are confirmed in the case of Aniiridn in the
succeeding chapter in this Memoir. By keeping various
Collembola in a damp chamber, and observing them
climb steep and smooth surfaces, it can be made out that
the primary function, at any rate, of the ventral tube is
that of an organ of adhesion. The present writer believes
that it also plays an important part as a respiratory
organ. Both Willem and Hoffmann regard the modified
hypodermis cells lining the ventral tube {Ji.c. in fig. 21)
as being simply ordinary hypodermis cells which have
assumed an unusual form ; while on account of their large

; Die Abdoiiiinal-aiihange der In^ekten, Morph. Jahrb., 1889.
D

34

size and tlie prnrainence of their nuclei they have been
regarded by previous writers as being the glands ^hich
provide the secretion which moistens that organ.

7. The Cephalic Glands and Ventral Groove.

In Anurida three pairs of cephalic glands are present,
two of which discharge their secretion into the ventral
groove, and the third pair open into the buccal cavity and
are to be regarded as salivary glands. Three pairs of
cephalic glands have also been made out by Willem and
Sabbe in Sminthttrus (28), by Willem in Orchesella (26),
and by Hoffmann in Tomocerus (12).

The salivary glands of Anurida are lodged in the
hinder region of the head, one on either side (Plate VI.,
figs. 63 and 65). They are composed of groups of acini,
and each acinus consists of a group of eight or nine very
large cells, which are disposed in a radial fashion around
a central ductule {dtl. in fig. 63). The cells of these
glands have very large nuclei, rich in chromatin, and a
considerable portion of the inner half of each cell is filled
with a dense and compact mass of secretory granules
[s. g.). The ductules are of very small calibre and are
intra-cellular at their commencement ; a little lower down
they converge and unite with those from neighbouring
cells to form a fine canal, which becomes inter-cellular in
structure. The canals from adjacent acini in their turn
unite with one another to form the main duct of the gland.
This duct (fig. 67) is composed of flattened cells, con-
taining small elongate nuclei ; both it and all other parts
of the glands are invested internally and externally by a
delicate membrane. The course of the duct is difficult to
make out among the numerous other structures of the
head, but it appears to converge with its fellow, from the
opposite side, towards the median line and to approach

35f

the lingua ; it Has not been possible, however, to detect
the point where they open into the buccal cavity. AYillem,
who has specially studied these glands, states that in
SiJiiiithunis their ducts open on to the lingua on either
side. He remarks as follows: — " Le conduit en c^uestion
se rapproche de la ligne mediane de la tete en passant
contre le pivot de la mandibule, puis cotoie obliquement
la commissure perioesophagienne ; sa derniere portion, a
parol chitineuse plus epaisse, est logee dans une rigole de
la base de riiypopharynx ; elle aboutit dans la cavite
buccale sur le cote de cet organe, un pen au-dessus de la
region occupee par le bord superieur de la plaque molaire
de la mandibule "' (26, p. 655). According to him a
similar arrangement prevails in Orchesella.

The salivary glands are regarded by Willem as being
a pair of metameric glands pertaining to the first maxillae,
and he bases this conclusion on the embryological
researches of Lzel (38). He remarks that *' Morpho-
logiquement, on doit les considerer comme des glandes
metameriques appartenant an segment de la premiere
maxille : la partie de I'hypopharynx ou aboutissent leurs
conduits excreteurs se forme, en effet, an se depens d'une
protuberance mediane qui apparait chez I'embryon sur le
premier segment maxillaire." If this observation of
Uzel be confirmed, the important generalization that the
salivary glands of the Collembola are homologous with
the shell glands of the Crustacea would be evident. The
Crustacean shell glands open on the second maxillary
segment, and it will be seen on referring to the table
given on page 74 that the second maxillge of the
Crustacea are homodynamous with the first maxillae of
insects.

Of the two pairs of glands which open into the
ventral groove, the most important pair are known as the

36

tubular glands and the other as the globular or acinose
glands.

The tubular glands (Plate YI., figs. 63 and 65) are
situated very near to the salivary glands in the hinder
region of the head. Both the secretory and conducting
portions of these glands are tubular in structure, and are
inter-cellular throughout their course. The distinction
between the two portions is very slight ; in the glandular
part the cells and their nuclei are rather larger than those
of the duct. The ducts of opposite sides pass downwards
and forwards close under the hypodermis of the ventral
region of the head and, approximating towards one
another, eventually open into a common chamber (fig.
64). From the latter a short duct passes downwards
in an oblique direction, and opens on to the exterior on
the ventral surface of the head at the commencement of
the ventral groove. The walls of the common chamber
are invested with a layer of circular muscle fibres, which
evidently have the power of controlling the flow of the
secretion into the ventral groove. These fibres resemble
in their structure those of the fore-gut, and consist
(fig. 64) of a layer of sarcoplasm on the outside with the
muscle substance lying wholly internal to it.

The globular or acinose glands (figs. 65 and 66) are
situated in the anterior region of the head, and close to
the hypodermis on either side. They each consist of a
group of cells, which are similar in structure to those of
the salivary glands, and from them a fine duct passes
forward and opens into the common canal or chamber in
front of the ducts of the tubular glands.

The ventral groove (Plate I., fig. 3, v.g.) arises close
behind the suture {sut. in fig. 12) of the labium, and passes
backwards in the median ventral line until it reaches the
anterior aspect of the ventral tube, where it terminates

3T

(Plate IV., fig. 36). At its commeucement it is in the
form of a closed canal (Plate II., fig. 26), which appears
to have been developed by the union with one another
of a pair of parallel ridges of the cuticle ; traced
further backwards, in the prothorax, the canal is
incomplete, the ridges not having fused with one
another ; in the metathorax each ridge has been folded on
itself, and the two together form a j)air of tubes with a
groove or channel between them (figs. 24 and 26t'), and in
this condition it continues the remainder of its course on
to the ventral tube.

The secretions of the tubular and acinose glands fiow
down the groove to reach the surface of the ventral tube,
which they serve to moisten. The fact that the ventral
tube receives the secretion of cephalic glands was first
ascertained by Fernald, and, although doubted by other
investigators, it has been recently confirmed by both
Willem and Hoffmann. Fernald, however, appears to
have regarded what are here described as salivary glands
as being the glands which furnish the secretion, and what
he has regarded as their ducts opening into the ventral
groove, appear to be really the ducts of the tubular
glands.

8. The jSTervous System.

The nervous system consists of a chain of five ganglia
united together by means of a double series of nerve
cords or connectives (Plate IV., fig. 48). The first of these
is the supra-oesophageal ganglion or brain ; it is situated
in the head immediately above the oesophagus (fig. 35).
The remainder of the system forms the ventral nerve cord
and lies beneath the digestive canal. It comprises a large
sub- or infra-cjesophageal ganglion, which is t()lh)wed by
three ganglia, lodged respectively in the prothorax, the

38

mesotliorax and the metatliorax (%. -JS). The sub-
oesophag-eal ganglion is joined to the brain by means
of a pair of para-oesophageal connectives, which pass
around and encircle the gullet, one on either side [conn.).

The brain of Anurida arises in a similar manner to
that which has been found to obtain in other insects. It
is formed in the embryo by the fusion of the first three
pairs of primitive ganglia, which are termed the
protocerebrum, the deutocerebrum and the tritocerebrum
respectively (Plate VII., fig. 74). The completeil organ of
the adult Insect is divisible into three main regions, which
correspond with these primitive ganglia. They are as
follows : —

(«) The protocerebrum, which is the largest division
of the brain, comprises the optic and proto-
cerebral lobes (fig. 48). The optic lobe on either
side gives origin to the main optic nerve (oj?. ??.),
and the latter sub-divides into five branches,
each branch supplying one of the eyes. The
protocerebral lobes form the greater part of the
brain when viewed from above.

(h) The deutocerebrum, which is composed of the two
antennary lobes. From each lobe a stout nerve
(at. n.) passes to the antenna of its side, and on
nearing the apex of that appendage it divides
into a number of fine branches, supplying the
sense organs (Plate IL, fig. 22).

[c] The tritocerebrum, which consists of the two
oesophageal lobes {trc. in fig. 48). From the
latter arise the para-oeso^jhageal connectives and
the nerves which supply the labrum.

The sub-cesophageal ganglion or, more properly, the
sub-CBSophageal ganglionic mass, is formed by the fusion
of four pairs of primitive ganglia. From it arise paired
nerves, which supply respectively the mandibles, the

39

maxillulse and the first and second maxillse, together with
au unpaired nerve which passes to the lingua.

The prothoracic and niesothoracic ganglia each give
oii: a pair of principal nerves on either side, which supply
the various muscles of their respective segments. The
ganglion situated in the metathorax [th. g. 3) is in reality
a ganglionic complex formed by the fusion of the pair of
primitive metathoracic ganglia with those of the
abdomen. Several pairs of nerves arise from this ganglion,
and they appear to supply the musculature of the meta-
thoracic and first abdominal segments.

Directed backwards into the abdomen is a pair of
stout parallel nerve cords {nv.). These are, perhaps, to be
regarded as the remaining connectives of the abdominal
nerve chain, the nerve ganglia of whicli have migrated
forwards during embryonic development and become
fused with the metathoracic ganglion. Along their course
these nerves give off branches (/./'.) which can be traced to
the generative organs and to the musculature.

Histologically, the nervous system is constructed
uj)on an exceedingly simple plan. The brain is composed
on the outside of a layer of unipolar and bipolar ganglion
cells 23rovided with very small but deeply-staining nuclei.
Internally it consists of a mass of extremely fine nerve
fibres, which are arranged for the most part in transverse
and longitudinal directions. It presents none of the
complex arrangement of nerve cells and fibres which is
exhibited in the brain of most Pterygote Insects. The
succeeding ganglia also each consist of an outer layer of
nerve cells enclosing a central core of nerve fibres (figs. 46
and 47). Xerve cells are wanting from the thoracic
connectives, but a few are distributed along the course of
the abdominal cords. All the ganglia and connectives
are invested by a neurilemma [iieiir. in figs. 40 and 47) of

40

dense connective tissue ; it takes a bluish colouration in
preparations treated with. Mann's methyl-blue-eosin.

Lying immediately dorsad of the ventral nerve chain
there exists what ma}' be regarded as a median accessory
nerve cord [in.n. in figs. 46 and 48). It is confined to the
region of the thorax, and is present between and above
each pair of connectives joining the thoracic ganglia. It
does not appear as a continuous cord, but is seen to arise
from the sub-cesophageal ganglion, and from there to pass
backwards to the anterior border of the prothoracic
ganglion, where it appears to terminate. It then,
however, takes fresh origin from the hinder margin of
the latter ganglion, and is continued backwards to the
mesothoracic ganglion. After repeating itself again, it
finally terminates in the ganglion in the metathorax.
This system is very delicate, and is difficult to make out ;
it is probable that its fibres pass as a continuous strand
through each of the ganglia, and so connect together each
section of the cord. Three pairs of transverse nerves
belong to this accessory nerve cord [m.n. hr. in fig. 48),
and in one or two fortunate preparations they were seen
to pass to certain j^arts of the ventral muscvdature.

A similar accessory system to the above has been

described by Oudemans in the Thysanuran Machilis. It

is there found in relation with each pair of ventral

ganglia throughout the whole length of the body, and is

said to supply the tracheal trunks and the occlusor

muscles of the spiracles. There appear to be no grounds

for regarding this system, either in Anurida or Machilis

as being of a sympathetic nature.' In Anniida its

* In the CoUembola a "true sympathetic (stomatogastric) system
has been described and figured by Willem in Podura aquatica. It is,
however, only very feebly developed. No certain indications of such a
system have been met with in Anurida. Claypole states that in the
embryo of the latter a cord -like proliferation of ectoderm cells arises
from the hinder end of the stoniatodoeum, and regards it as a remnant
of a sympathetic system ; it disappears, however, before hatching.

41

fuiiftioii is quite obscure: in Machilis it might possibly
be analogous with the vagus system of Yertebrata. A
somewhat similar system of nerves occurs among various
Pterygota, notably in the caterpillars of Cossus lignijjerda
(Lyonnet) and Sn/iiihr liyustri (IS'ewpoa't), and in the larva
of Chironoinus (Miall and Hammond).

9. The Sexsory Organs.

The sensory organs of Anurida consist of the eyes,
the post-antennal oi'gan, the apical sense organ of the
antennse and sensory hairs.

The eyes form a group of hve glistening black dots
on either side of the head. Each group is situated on a
very marked oval protuberance a short distance behind
the base of the antenna of its side (Plate I., figs. 1 and 4).

Each eve, when viewed from above, is seen to be oval
in form and somewhat convex, it is invested by the general
cuticle of the body-wall, which forms a perfectly smootli
and transparent corneal lens. AVhen viewed in vertical
section each eye is seen to be shaped very much like an
inverted cone (Plate III., tig. 34). Situated immediately
beneath the lens are four small cells, which belong to the
hypodermis and are directly continuous with it ; they
contain no pigment and are full of clear protoplasm. Two
of these cells are visible in the figure, and as they secrete
the lens they may be termed the lentigen cells. Situated
internally to the latter are four very large, deeply pig-
mented cells, which together form the retina. The inner
ends of these cells pass into and are directly continuous
with a branch of the optic nerve. They are invested with
a mass of pigmented hypodermis which also fills the
interstices between adjacent eyes.

It will be noted that the eyes of Anurida are
structures formed on an extremely simple plan. The

42

researches of AVilleiii go to prove that in the g-reater
numbe]' of Collembola (i.e., in the SminthuricUe, the
Eutomobryidse and in some of the Aehorutidsej the eyes
are constrncted on the eucone principle, and he regards
them as being incipient compound e^'es. In several
genera of the Achorutidse (including Anurida) he points
out that there is no layer for the secretion of a crystalline
cone, and hence he looks upon the eyes of such forms as
being ocelli or stemmata. He remarks (27 p. 90): —
" La consideration que les ocelles s'observent parmi les
Podurides, chez des formes animals oii les orgranes visuels
sont manifestement regression, incite a admettre que ces
stemmates sont productions derivees de I'ommatidie
eucone par la disparition du systeme lentigene, con-
clusion qui accords avec des faits observes par Patten
dans le developpement des ocelles de Yespa et de larve
^VAcilms, et qui autorisent a admettre que les stemmates
a deux couches cellulaires de ces Insectes derivent d'un
oeil a trois conches (comme I'ommatidie euconej par
I'atrophie on le developpement incomplet de la strate
moyeune."

If the extreme simplicity of the structure of the
eves of Anurida is due to retrogression, and is not a
primitive character, it lends support to a belief
maintained by certain students of the Collembola, viz., that
ATniridd is, in some respects, a degenerate type.

The post-antennal organ (Plate I., figs. 4, 5 and 10,
Plate III., fig. .j4) is a curious cuticular structure, circular
in form and situated one on each side of the head
immediately in front of each protuberance which bears
the ocelli. It consists of a variable number of somewhat
triangular bodies, which are in close contact with one
another, with their apices directed towards its centre.
The number of these bodies varies in different individuals,

43

but there are most usually uine of them. Out of eight
individuals which were specially examined Avith reference
to this organ, four of them had nine triangular bodies,
two had ten, and the remaining two possessed seven and
eight respectively. Both Laboulbene and Feriuild have
observed specimens which had seven to the organ on one
side of the head, and eight to the organ on the other side.

The elements (triangular bodies) of which the organ
is composed appear to be nothing more than greatly
enlarged tubercles of the cuticle which have become
hollowed out and mounted on short pedicles and arranged
in a circular manner around a common centre (tig. '34).
The cavity of each of the elements is filled with
protoplasm containing a small amount of pigment, but
it has not been possible to detect any nuclei.

Each post-antennal organ is supplied by a small
nerve {p.o.n. in fig. 34).

It is clear that this structure is a sensory organ from
the fact of its having a special nerve supply, and the
extreme thinness of its cuticular investment shows that it
is adapted to receive impressions of external stimuli.
What function it is likely to perform is wholly a matter
of conjecture; possibly it is an olfactory organ, as is
suggested by Willeni.

The apical sense organ is situated somewhat towards
the inner side of the apex of each antenna (Plate I., fig. 7).
It is a hollow tri-lobed structure (fig. 8), and its lobes are
in free communication with one another at their bases.
It is innervated by a branch from the antennal nerve
(Plate II., fig. 22), which enters the organ through a small
oval foramen situated at its base (fig. 8).

The function of the organ is uncertain ; Fernald
remarks that " the three lobes are so placed relatively to
each other that they might easily give the animal some

44

idea of the form of any object, corresponding, as they do,
to the three dimensions in space." It seems more likely,
hoAvever, that it functions as an olfactory organ, if
it be safe to reason from analogy with what is known
concerning the antennal sense organs of the Pterygota.

The sensory hairs are found principally at the apices
of the antennae, and on the upper and lower lips, but
possibly some of the hairs scattered over the general
surface of the body may be of a sensor}^ nature al^o.

10. The Circulatory System.

The circulatory system of Anurida consists of the
heart and the general body-cavity.

The heart is a narrow tube situated just beneath the
integument in the median dorsal line of the body above
the digestive canal (Plate lY., fig. 35). It is divided by
means of segmeutally arranged constrictions iuto a series
of six consecutive chambers, which are in free communi-
cation with one another. Anteriorly, the heart extends
as far as the anterior border of the mesothoracic segment,
and from that point it is prolonged forwards as the
aorta [a.). Posteriorly, it increases in calibre, and
terminates rather abruptly in an enlarged and somewhat
bulb-like chamber in the fourth abdominal segment
[a. bib.). Situated on either side of the heart at each of
the constrictions is an oval oriface or ostium (ost. in
fig. 41), and in association with the heart at these same
points is a pair of alary muscles (///. m.).

The walls of the heart consist of an external invest-
ment of connective tissue {ic. in fig. 40), a middle coat of
obliquely disjDOsed muscle fibres {c.m.b. in fig. -59 and ;/(.»'.
in fig. 40), and an inner lining or endocardium. The
muscle layer consists of bands of non-striped muscle
fibres, and is the only part of the body where such are

45

fouud ; they are much more strongly developed in the walls
of the last two chambers than elsewhere. The endocardium
is an excessively thin membrane ; in certain places it can
be made out with the aid of an oil-immersion lens
(iVin.^, but only with considerable difficulty.

The alary muscles are attached to the dorsal integu-
ment between the two groups of longitudinal tergal
muscles on either side of the body. Each muscle is
composed of a small bundle of striated fibres [al. ??(. in
fig. 41), and these latter are prolonged with a tendonous
strand of connective tissue [al. t.). As each tendon
approaches the heart it broadens out and splits into two
fan-shaped sheets or layers, one of which spreads out over
the dorsal surface of the heart and the other over its
ventral aspect (fig. 43). In the space between these two
layers is situated an ostium or lateral perforatioii of the
wall of the heart, and altogether there are six pairs of
ostia and alary muscles.

The aorta («. in fig. 35) may be said to commence
immediately in front of the first ostium, and directly
above the junction of the oesophagus with the mid-gut.
At its commencement it is triangular in transverse
section (fig. 42j but, traced further forwards, it becomes
more flattened, and is closely applied to the dorsal wall of
the oesophagus (fig. 51). By examining a series of
consecutive transverse sections, it will be seen that the
aorta gradually comes to wrap itself round the fore-
gut, and eventually completely surrounds it (fig. 35).
Just beneath the supra-oesophageal ganglion the aorta is
seen to terminate in a slightly expanded funnel-shaped
mouth (fig. 27), which enables the blood to flow directly
from the heart into the hsemocoelic cavity of the head.
The walls of the aorta are entirely membranous, and
contain no muscle fibres.

4G

The body-cavity of Anurida, as in all tlie lusecta, is
the blood-containing space in which are situated the
various organs of the body. It extends throughout the
whole length of the animal, and sends prolongations into
the antennae, the legs and the ventral tube. It is much
intersected by the fat body which occupies a large
proportion of its cavity in many parts of the body. The
body-cavity is filled with the blood of the animal, and the
corpuscles of the latter are noticeable everywhere
throughout its extent ; they are specialh' abundant in
the head, where the vascular and haemocoelic systems are
in open communication with one another.

The blood consists of {a) the corpuscles and (h) the
plasma. The corpuscles are circular in outline, but at
times they become amoeboid and give out pseudopodia
(Plate III., fig. '-V^). They are practically colourless, and
each contains a sharply defined circular nucleus.
Frequently small granules are scattered through the
cytoplasm of the corpuscle ; they are probably of an
excretory nature. Claypole remarks that in aewly
hatched individuals the blood corpuscles contain a large
amount of food yolk.

11. Excretory System.

Unlike other insects none of the Collembola possess
Malpighian tubes.* The excretory function is performed
entirely by the fat-body, or " Exkretionsorgane," as it is
termed by Sommer.

In Anurida the fat-body takes the form of irregular
masses of tissue distributed in various parts of the body
and occupying a large portion of the cavity. In the region

* In the Thysanura IMalpighian tubes are absent in Japyx, and in
Campodea they are only represented by papillae (Oudemans). Among
the Pterygota they are also wanting in Aphides (Witlaczil).

47

nf tlie abdomen it is largely developed, and in places it
closely embraces the gut and also attaches itself at many
points to the hypodermis. In the thorax it is much
broken up by the various muscles ; two principal masses
occur in close contact with the fore-gut, one on either
side, and come in contact with each other in the mid-
dorsal line just above the heart. Prolongations of the
fat-body pass from the thorax into the head and partially
surround the cephalic glands (fig. i)'^^) ; they do not extend
forwards in front of the brain. The organ is also present
in many parts in the form of a thin layer of tissue situated
in close apposition witli the hypodermis and separating
the latter from the body-cavity. In this form it is
principally developed along the ventral side of the body,
especially in the thorax.

The fat-body is limited exteriorly by a layer of con-
nective tissue and, where it is in contact with the hypo-
dermis, this inveS'tment is directly continuous with tlie
basement membrane of the body-wall (fig. 16). It is
wanting, however, from that surface of the fat-body
which is in close contact with the hypodermis and other
parts of the body.

In a newly hatched Anurida, before any active
process of excretion has taken place, the fat-body is seen
to be composed of masses of large cells with very irregular
boundaries (fig. 18). These cells are composed of a some-
what granular protoplasm, and each contains a relatively
small nucleus. The first indication of excretion taking
place is seen in the fact that many of the cells undergo
liquefaction at their centres, and eventually become
vacuolated (figs. 18 and 19). These vacuoles frequently
become confluent with one another, and the remaining
protoplasm, together with the nuclei, becomes restricted to
the periphery of the cells. At a later stage, tliere appear

48

in the fluid contents of the A^acuoles, numerous fine
granules of excreton^ material (fig. 19) ; these granules
increase in size, some much more rapidly than
others, and they ultimately take the form of rounded or
oval concretions. After excretion has been going on for
some time, the fat-body takes the form of an irregular
protoplasmic meshwork, in which cell boundaries are not
discernable, and nuclei are distributed at various intervals,
and the spaces enclosed within the meshes are filled with
fluid which contains the excretoiy granules held in
suspension. The process of excretion is generally most
active in the central portions of the larger masses of fat-
body, and the cells of the latter ultimately break down
and completely liquefy, with the exception of those
situated around the periphery (fig. 20).

The excretory concretions are unaffected by the action
of water, alcohol, ether or toluol, but are readily soluble
in acids, and are consequently absent from preparations of
animals fixed by acid-containing reagents. They are best
seen in animals which have been killed and fixed in strong
alcohol. They are of a crystalloidal nature, faintly
yellowish in colour and very eosinophilous. Their
chemical composition is uncertain, and they do not exhibit
any indications of lamination or other structure.

Sommer (23) has given some attention to the con-
cretions found in the fatty body of Tomocerus phnnheus.
In this species they are often of large size and are easier
to study than those of Aiinrida. They are dirty white in
colour when viewed with reflected light, and pale green
and shinins' with transmitted light. In their structure
they are laminated like starch-grains, and sometimes
exhibit radial lines in addition. They are unaffected by
water or alcohol, but dissolve in acetic, hydrochloric or
nitric acids, accompanied by an evolution of gas. Each

49

coucretiou is made up of a covering coat investing a
central crystalline (?) mass. This is seen wlien nitric acid
is made to act very slowly, when the central contents
dissolve first and subsequently the coat vanishes also.
Sommer believes that they are composed of calcium
carbonate, together with an organic basis.

Willem has also studied these excretory concretions,
and he used for the purpose Sminthurus fuscus. He
remarks (27) that they have a concentric structure,
and that they are insoluble in water, alcohol, ether, or
chloroform, but are dissolved by acids. By means of
aj)propriate tests, he came to the conclusion that they are
formed of a neutral urate of sodium. It is probable that
those of Tomocerus are of a similar composition, rather
than of calcium carbonate, and Sommer mentions that he
did not apply any uric acid tests.

A remarkable feature regarding the excretory
function in Collembola is the absence of any ducts by
means of which the products can be eliminated. The
concretions consequently increase in size according to the
age of the animal, and render it impossible for the lives of
these insects to be prolonged for any lengthy period. An
analogous case is met with in the Ascidians among the
Chordata. The " renal vesicles " of these animals
similarly have no means of getting rid of the excretory
material stored up within them, and which increases
greatly in quantity as the individuals progress in age.

It cannot be decided, in the light of existing
knowledge, whether the fat-body of the Collembola is
homologous with that of other insects. In the case of
the latter, authorities are far from being in agreement as
to its mode of development, and nothing is known of the
development of the fat-body in the Collembola. Both
Sommer and Willem claim that among the latter

E

50

the fat-body, on account of its intimate relations with
the hyperdermis, is an ectodermal product. This
evidence, however, does not seem sufficient to justify that
conclusion.

12. The Reproductive System.

The Male Organs. — The testes consist of a pair
of organs disposed along- the ventro-lateral region of the
body, and lying for a considerable portion of their length
in close apposition to the mid-gut. Each testis is an
elongated chamber, which tapers anteriorly into a delicate
filamentous prolongation (Plate YL, fig. 54), and the
extremity of the latter is attached to the excretory tissue
lying in the mesothoracic segment. Posteriorly, in the
region of the second and third abdominal segments, the
calibre of the testis increases very considerably, and the
mature organ occupies a large portion of the body-cavity
in those segments. The hinder fourth of each testis lies
completely ventral to the alimentary canal, but as each
gland passes anteriorly it gradually comes to lie ventro-
laterally in relation to the latter, and finally in its
terminal fourth it is completely lateral. In the fifth
abdominal segment each testis gives off a short eiferent
duct which, uniting with its fellow to form a common
canal, opens to the exterior on a small papilla situated in
the median line near to the posterior margin of the
segment (fig. 54).

Each testis is in the form of a tube, and its walls
consist of a cellular layer containing scattered oval nuclei,
but without any cell-boundaries. Externally it is
invested by a coat of connective tissue (fig. 59). The vasa
deferentia are similar in their structure to the testis wall,
and appear to be tubular evaginations of the latter. The
median canal, or ductus ejaculatorius (Plate III., fig. 32),

51

is formed by the iiuion of the two vasa deferentia ; it is
roughly diamond-shaped when viewed in transverse
section, and is lined internally by a chitinous futicle
which rests on an epithelial layer containing numerous
nuclei. The ductus is covered externally by a coat of
connective tissue. Dilator muscles arise from its chitinous
lining" and, passing in an oblique direction outwards, are
inserted into the body wall imusc).

The lumen of the testes is filled with a dense mass of
nucleated tissue, which consists of different stages in the
development of the spermatozoa. At the apex of the
gland the testis-wall becomes drawn out into the terminal
filament (fig. 58). The filament is tubular in structure,
and numerous nuclei can be seen in its walls. Near to its
apex the cellular wall dwindles away, and the filament
becomes reduced to merely a thread-like prolongation of
the connective tissue coat of the testis. It is purely liga-
mentous in function, and serves to maintain the testis in
position in the body-cavity.

If the structure of the testes be examined from the
apex of the organ and be followed downwards near to
where the efferent duct is given off, the successive stages
in the development of the spermatozoa can be readily
traced. The germinal tissue is in the form of a mass or
ridge, situated at about the middle of the length of the
testis. It consists of a mass of cells (fig. 59a), which are
seen in places to be roughly polygonal in shape ; each cell
contains a large, very deeply staining nucleus. The
germinal nuclei divide, and the direct products of the
division fill almost the whole of the anterior third of each
testis {d. s. in figs. 58 and 59a). The nuclei are seen to
principally arrange themselves in groups, which are more
or less ovoid or circular in form, and contain an area of
clear protoplasm in their centres. Traced further buck-

52

wards [d'^, s^ in fig. 59) these nuclei divide and fragment
and form tiny masses of irregularly-shaped chromatic
bodies embedded in a dense matrix of protoplasm. In the
hinder third of the testis the cellular layer of its wall
gives off numerous branched intra-testicular prolongations
[i. y. in fig. 59). These prolongations divide up the
substance of the testis into a number of separate masses
by means of their branches, which freely anastomose with
one another. In this way the cavity of the testis becomes
sub-divided into a number of irregular chambers, each of
which is filled with a mass of developing spermatozoa.
The next stage in the development of the sex cells is seen
where the chromatin bodies retreat towards the periphery
of the masses of protoplasm in which they are distributed
{cl?, 5-), and eventually become congregated together over
a very limited area of the latter. Traced a little further
back, the chromatin bodies gradually elongate to form the
heads of the spermatozoa, and from them are seen to
extend delicate filaments, and each mass becomes easily
recognisable as being a bunch of spermatozoa {s. -p.).
During the last stage in development the intra-testicular
prolongations, which have served to nourish the developing
sperms, are seen to break down and ultimately liquefy,
and the posterior end of the testis becomes filled with
spermatozoa floating in a thickish fluid.

The process of spermatogenesis has also been studied
in Anurida by Fernald, but the above account differs
very considerably from the description detailed by that
author. According to his observations, it would seem
that the process is a much more lengthy one than that
just described. What are in the present account
interpreted as bundles of fully-developed spermatozoa,
correspond with his stage where there are a number of
elongate granules with fine threads passing from them

53

forming- bundles lying in various directions. According
to him, this is not the final condition, but is merely a
prelude to a further series of changes. He states that
this structure soon changes, and the entire substance
begins to undergo degeneration, and an entirely homo-
geneous mass results containing many deeply-staining
granules. Fat globules appear in small numbers, and a
columnar epithelium becomes formed near to the external
connective tissue of the testis wall. A similar epithelium
becomes develoj)ed in the granular mass nearer to the
testis. Traced further back the fat globules increase in
number, and among them are multitudes of spermatozoa
which have been formed from the epithelial cells.

These later changes described by Fernald are very
remarkable, and there can be little doubt that they do not
represent the normal state of aflairs, and most probably
they are of a pathological nature. The observations of
the present writer confirm in all essentials the brief
account of the spermatogenesis given by Lecaillon (16).
In many examples of Anurida the writer has found
Xematode worms,'' and their developmental stages in the
testes (Plate YIL, figs. 79 and 80). In several instances,
these parasites were found to be so numerous as to
destroy the whole internal structure of the testes, leaving
behind little more than the bare walls. In other cases
the worms were less numerous, and the structure of the
organs was only partially destroyed, while many oil
globules were present. It is the belief of the present
writer that the cycle of later changes described by
Fernald are not stages in the process of spermatogenesis,
but are of a pathological nature, probably due to the
presence of Nematode worms in the testes of the animals
which he studied. Unless the presence of these worms is
* These parasites were also occasionally met with in the ovaries.

54

suspected, sections of the testes coiitaiuing them present
a deceptive and puzzling appearance.

The Female Organs — The female organs (Plate VI.,
fig. 58) are extremely simple, and in their external form
they very closely resemble those of the male. They
consist of a pair of ovaries, which have similar relations
to the alimentary canal as the testes (Plate IV., fig. 35).
Each ovary is in the form of a simple tube, tapering
anteriorly into a delicate terminal filament [t. /.). This
filament, as in the male, functions as a suspensory
ligament which serves to retain the ovary in position, and
is attached to the fat-body near to the median dorsal line
in the mesothoracic segment. The ovaries increase in
calibre as they are followed posteriorly, and in the fourth
abdominal segment each gives off a short oviduct (fig. 35).
The two oviducts quickly unite with one another and form
a median canal, which may be termed the vagina {vag.).
The vagina opens to the exterior by a median aperture
situated near to the posterior margin of the fifth
abdominal segment. From the ventral wall of the vagina
there arises a forwardly-directed flattened diverticulum
{die). The homology of this structure is very doubtful,
and Fernald has suggested that it may be a receptaculum
seminis, but in no instance has it been possible to detect
any traces of spermatozoa in its cavity.

The wall of the ovary consists of a thin cellular layer
containing scattered flattened nuclei, but does not show
any cell boundaries (fig. 55). On the outside the ovarian
wall is limited by a coat of connective tissue {t. j).). At
the apex of the gland the wall becomes prolonged
forwards as the terminal filament, which becomes reduced
near its extremity to only a strand of connective tissue.

The oviducts are outward prolongations of the ovarian
frails, and have the same essential structure as the latter.

55

The walls of the vagina consist of a layer of columnar
ejiithelium, resting externally on a basement membrane
and lined internally by a layer of chitin (Plate III.,
fig. 31). The median diverticulum from the vagina has
a closely similar structure, except that its nuclei are larger
and no cell boundaries are distinguishable in its epithelial
layer. Its lumen, on account of the flattening of the
organ, is extremely narrow and is in the form of a
transverse slit.

A few muscle fibres are attached to the walls of the
vagina, and pass outwards in an oblique direction to be
inserted into the neighbouring body-wall. They probably
aid in the extrusion of the ova by bringing about the
alternate expansion and contraction of the cavity of the
vagina.

The germinal tissue is situated in each ovary as a
protruding ridge lying in the region of the third and
fourth abdominal segments. The ridge is dorsally placed
in each ovary, and is inclined somewhat towards the
outside of the body (Plate YL, fig. 57). It consists of a
mass of irregularly-arranged cells, each cell with an oval
nucleus with the chromosomes regularly disposed around
its periphery. By means of rapid mitotic divisions the
germinal ridge buds off a mass of cells, which lie free in
the ovarian cavity. At an early stage two kinds of cells
become distinguishable, viz., vitellogenous or nutritive
cells, which have rounded nuclei rich in chromatin
material, and the egg cells or ova, which are much less
numerous and are scantily supplied with chromatin,
which is arranged in eight small granules disposed around
the periphery of each nucleus (fig. 55).

There is no arrangement of the ova into follicles, but
a number of vitellogenous cells become grouped around
each ovum. As development goes on the ova accumulate

56

food yolk within them, and increase greatly in size (fig. 56).
When the eggs are fully developed the vitellogenovis cells,
which have served to nourish them, having nothing
further to do, rapidly degenerate and eventually
disappear. According to Lecaillon (14), in some Collembola
{e.g., Pajnrius) intra ovarian jDi'olongations of the wall of
the ovary are developed, and serve to assist the vitello-
genous cells in the elaboration of food m.aterial for
absorption by the eggs. In Anurida, according to that
writer, these prolongations are only very slightly
developed. For the details of the process of oogenesis the
student is referred to the works of Lecaillon and of
Claypole (31).

v.— EMBEYOLOGY.

The eggs of Anurida when freshly laid are pale
yellow in colour, with a smooth, glistening surface. As
development proceeds they darken, and eventually become
dark orange. They are spherical in form, and each
measures on an average about 30 mm. in diameter. They
are deposited by the female in little groups, which are
usually to be found in plent}^ about the shores of Port
Erin Bay. They occur in the furthermost recesses of the
fissures in the rocks, and may be obtained by splitting
open the latter in the manner already mentioned (p. 8).
These groups consist of from seven or eight up to twenty
or thirty eggs, which are laid together in an irregular
fashion (Plate VIL, fig. 69). The individual females
appear to take no pains to keep their own eggs separate
and distinct from those of their fellows, and through this
cause large confluent masses, often consisting upwards of
a hundred eggs, may be met with.

The relatively large size of the eggs of Anurida

57

renders them favourable objects for the study of
Apterygote embryology. AVhat is kuowu of the develop-
meut of this iusect is due to the researches of Eyder (37),
Wheeler,* Miss Claypole (31) and Folsom (32). I'he
following account is mainly derived from the observations
of the two latter writers.

The egg is invested by two clearly distinguishable
membranes. Firstly, an outer shell-like coat (ch'. in
fig. 71), which is very brittle and can be easily dissected
off with the aid of a pair of fine needles and, underlying
this, is a thin crenated membrane [memh.) These coats
may, perhaps, be homologovis with the chorion and
vitelline membrane respectively of other Insecta. During
the early stages of development, a third and very delicate
membrane closely envelopes the embryo, except where it
is interrupted by the dorsal or pre-cephalic organ. In
addition to these coats, Claypole distinguishes two other
Qgg membranes, and the authoress maintains that all five
membranes arise from the egg or the blastoderm.

The unsegmented ovum (fig. 70) is formed of a large
central mass of protoplasm, which gives off outwardly
directed strands. These strands ramify as they approach
the periphery of the egg, and enclose numerous yolk
bodies within their meshes. The germinal vesicle early
becomes invisible, and does not again appear until after
the extrusion of the polar bodies. The latter do not
completely separate from the egg, and are eventually
resorbed into it.

The segmentation is especially noteworthy, it being
holoblastic but slightly unequal, and results in a solid
morula. After the morula has been formed, a breaking
down of the cell-boundaries takes place, and many of i\\e
nuclei begin to migrate to the periphery of the egg,
* A contribution to Insect Embryology. Journ. MorpJi., viii, 1893.

58

leaving the yolk bodies in the central protoplasmic mass.
The nuclei undergo division as they pass to the exterior,
and they eventually arrange themselves to form a two-
layered blastoderm — the future ectoderm and mesoderm.
Of the cells which are left behind in the yolk, some are
scattered singly through its substance, and have large
and deeplj^-staining nuclei ; the rest have a tendency to
remain congregated together in little clusters. The
former are the yolk cells or vitellophags, Avliich serve to
transform the yolk for the nutrition of the embryo ; the
latter, at a subsequent stage in development, form the
rudiments of the future endoderm {i/.c. and e.n. in
fig. 76).

About the time of the completion of the blastoderm
a modification of its cells takes place at the upper pole of
the egg, forming the precephalic or dorsal organ. At
this point the cells become markedly columnar, and when
viewed in section they form a lens-shaped mass of cells
(fig. 7G, 2'X-'- 0.). This remarkable structure is apparently
a vestigeal organ ; it soon commences to degenerate, and
eventually disappears altogether. AVheeler* homologises
this organ with the " dorsal organ '' of certain Crustacea
with the iudusium of the Orthopteron X/phitliitm. There
are no traces at any stage in the development of structures
corresponding with the amnion and serosa of other
Insecta.

The ventral plate or germ band first aj^pears as a
narrow area of mesoderm. It is formed by the cells of
the latter migrating from their original position and
becoming restricted to a band-like zone. This zone, with
the exception of where it is interrupted by the dorsal
organ, forms a girdle completely encircling the egg. As
the result of this migration of the mesoderm, the greater

* Loc. cit. p. 55, Vid<j also Willey. Quart. Juurn. Mic. Sci., xii, 1899.

59

part of the egg is left only covered by the ectoderm
(fig. 78, mes.).

The various appendages appear for the most part in a
regular succession from before backwards (fig. 71). The
antennae are the first pair to develop, and they are followed
by the medium unpaired rudiment of the labrum and
clypeus. Behind the former there arise in rapid succession
the mandibles, the first and second maxillse, the three pairs
of thoracic feet, and a pair of appendages to each of the
first four abdominal segments. Between the antennae and
mandibles there are developed a pair of transient trito-
cerebral or intercalary appendages (fig. 7o). They were
first discovered in this insect by Wheeler, and both
Claypole and Folsom have since confirmed their existence.
A little later there arise between the mandibles and first
maxillae the rudiments of another pair of mouth-parts
iiii.vuJ. in fig. 75). These have been recently discovered by
Folsom, and were named by him the super-linguae, but,
as already stated (footnote p. 19), the term maxillulae is
preferable for them. About this time the rudiments of
the stomatodieal and proctodaeal invaginations become
visible, and the dorsal organ commences to degenerate.
A very marked flexure of the embryo now takes place, and
when it is complete the latter becomes more or less folded
on itself (fig. 72). The result is that whereas the anterior
end of the embryo retains its relations with the pre-
cephalic organ, the anal portion becomes retracted away
from the latter, and the embryo itself becomes restricted
to less than one-half of the whole circumference of the
egg, instead of completely encircling it, as happened in
the earlier stages. Furthermore, the mouth-parts become
crowded together to form a definite head. The last mouth-
part to appear is the lingua, which does not become
visible until after the flexure of the embrvo has

60

commenced. It arises as a median unpaired rudiment,
situated between the fundaments of the hrst maxillce {ling.
in figs. 7 and 75).

Mention has already been made of the origin of the
endoderm from certain of the cells which are left over in
the yolk during the formation of the blastoderm. At a
stage a little subsequent to the condition of the embryo
represented in fig. 72, these cells commence to separate.
They increase iu size, and, engulphing yolk, they
so arrange themselves as to form a definite mesenteron.
It is remarkable that none of the yolk is enclosed in the
mesenteron during the process of formation of the latter ;
much of it remains in the body-cavity, while the rest
becomes included in the reproductive organs. Whether
the yolk cells have anything to do with the endoderm is
not clear ; although they take no part in the formation of
the mesenteron, Claypole suggests that they may,
perhaps, be endoderm cells which have acquired their
digestive powers at a very early stage.

The nervous system arises as a proliferation of
ectoderm cells in localised positions in the various
segments. The masses of cells so formed give rise to the
primitive ganglia, and they subsequently join up with
one another to form the brain and ventral nerve chain.
The brain is formed from the first three primitive ganglia,
which are termed respectively the proto-, deuto-, and
trito-cerebrum (fig. 74). There follows then in succession
ganglia pertaining to the segments of the mandibles,
maxillulse, and first and second maxillse, together with
three thoracic ganglia and a ganglion for each of the six
abdominal segments. The subsequent changes result in
the union of the three cerebral ganglia to form the
completed brain, while the remaining cephalic ganglia
consolidate into a siug-le g-uioiionic mass to form the sub-

61

(psopliageal ganglion, aud ultimately th.e abdominal
ganglia become involved in the meta-thoracic ganglion.

The germ cells appear at a comparatively late period
in development ; they arise a short time previous to the
commencement of the formation of the mesenteron.
They take their origin from the splanchnic mesoderm in
the second and third abdominal segments, and according
to Claypole their method of development is as follows: —
Those germ cells which ultimately give rise to the male
gonads arise from the inner side of the mesoderm, and
come to lie free among the yolk (fig. 77). In the case of
the female they are budded off from the outer side of the
splanchnic mesoderm, and form a group of cells lying free
in the mesoblastic somites of those segments. In the
case of the male the ccelomic spaces are not formed. In
the female the splanchnic mesoderm forms one side of a
spacious cavity, which evidently results from a fusion of
the contiguous cavities of the mesoblastic somites. The
inner wall of this cavity eventually breaks through, and
the coelom becomes confluent with the general htemocoelic
body-cavity, and at the same time the germ cells come in
contact with the yolk. The further development of the
werm cells has not been followed in anv detail but, in
both sexes, as they develop they incorporate among
themselves a large amount of the food yolk distributed in
the body-cavity. In newly-hatched individuals much of
the yolk is still present in the gonads, and it conduces
greatly towards the rapid maturition of the generative
elements.

Of the abdominal appendages, the first pair fuse
together and persist in the adult as the ventral tube.
The remaining pairs become resorbed, and ultimately
completely disappear.

The eyes and post-antennal organs first become

62

clearly indicated about the stage represented in fig. 72.
At the time of its first appearance, the post-antennal
organ consists of an irregular group of cuticular
elevations. In the newly-hatched animal the latter have
decreased in number and arranged themselves in a
circular form, and before the permanent form is attained
the organ undergoes further changes and development
[vide Laboulbene (2)].

The recently-hatched insect is perfectly white, except
for the eyes, which are enveloped in a patch of dark
pigment on either side.* Except in size and colouration,
the young animal does not differ from the adult. The
pigmentation of the integument i§ acquired after exposure
to daylight, but it is some time before it is developed to its
fullest intensity. After several moults the full growth of
the animal is attained.

YI.— GENERAL REMARKS ON THE
COLLEMBOLA.

The general characteristics of the Collembola have
already been noticed (Introduction, p. 4). About four
hundred species have been described up to the present,
most of which are very small in size, and but few forms
attain a length greater than 5 mm. The Collembola are
divided into three families, and these are classified by
Borner (48) under two sub-orders, as follows: —

Sub-Order I. — Arthropleona.

Body linear and sub-cylindrical, the limits between
each of the segments of the abdomen distinctly indicated.
Vesicles of the ventral tube short and sac-like.

* Folsom remarks that if the eggs be exposed to sunlight the
embryos become blackish blue before hatching.

63

F a m . I . , E n t n ni o b r y i d ae . — Scales present
or absent. Springing organ well developed. Isotoma,
Entomohrya, OrcheseUa, Lepidoeyrtiis, Actaletes, Beckia,

F a m . II., A c li 0 r u t i d se . — Scales never present.
Springing organ greatly reduced or absent. Podvra,
Achorutes, Atuirida, XenijUa, Aphorura, Neanura, &c.

Sfb-Order II. — Symphypleona.

Body sub-globular, the abdominal segments more or
less fused together. Vesicles of the ventral tube long and
tubular or short and sac-like.

F a m . III., S m i n t h u r i d se . — ^With the
characters of the sub-order. Smiiithurus, Sminthnrides,
Pajnrius, Neelus (MegcdotJiorax).

These insects are found in a great variety of situa-
tions ; they occur among dead and decaying vegetable
matter of all kinds, they may be found among moss,
under the bark of trees, and can be obtained from road-
side herbage of almost every description. They are
plentiful on the borders of ponds and streams, and are
also to be met with on the surface of the water itself ; some
few species inhabit the sea-shore and others are to be
found on the surface of the snow and glaciers. The only
condition which seems essential for their welfare is that
considerable amount of moisture is present, for they seem
incapable of surviving in very dry situations.

The order is practically world-wide in its distribution,
and it is remarkable also on account of the extremely
wide distribution enjoyed by many of its individual
genera and species. The genus Isotoma, for instance,
seems almost cosmopolitan, it being known from both
Arctic and Antarctic regions, and is distributed through-
out Europe and many parts of Xorth America. It has

64

been recorded from Argentina, Sumatra, tlie Sandwioli
Isles, and the Bismarck Archipelago, and also from the
Azores. Among species, Aehorutes armahis is found
throughout the greater part of the Holarctic region, as
well as in South America, New Zealand and Sumatra ;
and Sminfhunis hortensis has a wide distribution in
northern and central Europe, and is also known from
North America, Tierra del Fuego, and Japan.

In colouration a wide range is exhibited among the
various species. Many forms are of a uniform dull blue-
black, as in Anurida; others are green or yellowish, with
irregular patches of a darker colouring ; a few species are
banded or ringed ; some are all white ; and metallic forms
are not infrequent.

In the greater number of Collembola the body is
clothed with hairs, but in some genera, such as Tomncerux
and Lepidocijrt\is, it is scaled, although hairs are not
altogether wanting. The hairs (text-fig. 4) may be simple
and tapering, clavate, or flattened and partially resemb-
ling scales, OT compound as in the Entomobryidse. The
scales bear a close resemblance to those of the Lepidoptera
but, acording to Sommer (23), they differ from them in
that they have no connection with the underlying
hypodermis. They are familiar to microscopists on
account of their being favourable test objects for the
capabilities of high power objectives.

The head varies considerably in the three families.
In the Smintliurid?e it looks almost vertically downwards,
among the Entomobryidae it is also pointed downwards,
but in rather an oblique direction, while in the
Achorutidse it is directed straight forwards in a line with
the rest of the body. The antennae have from four to six
joints ; they may be short and stout, as in the
Achorutidtc, or long and slender, as in OrrJiese/Ia and

65

Tomocerus. lu tlio latter genus and some species of tlie
vSniintliuriclae the terminal joint is much longer than the
rest and is annulated. With the exception of Neelus,
the Sminthuridse usually have elbowed antennae.

The thoracic segments are usually sharply defined ;
the prothorax is generally the smallest division of the
three. In the Sminthuridse the thorax is greatly com-

Text-Pig. 1. — Abdomen of an embryo Isotoma, showing appendages
(after Prowazek, enlarged). v, ventral tube; hm,
hamula or ' catch ' ; />, furcula or ' spring.'

pressed between the relatively large head and abdomen
and, furthermore, the metathorax becomes partially fused
with the first abdominal segment. Neelus [Megalo-
thorax), however, is remarkable on account of the enormous
development of the thorax, which greatly exceeds the
abdomen in length and bulk.

The legs have no true tarsal segments, and they
terminate usually in two claws, an upper and a lower,

F

66

which are inserted on the end of the tibia (text-figs. 2, 3
and 4). In frequent instances the lower claw is greatly
reduced, or it may be wanting altogether, as is the case
in Anurida. There are frequently found in relation with
the claws several (usually three) stiff hairs or setae, which
are swollen at their extremities.* These structures are

oo

Text-Fig. 2. — I., Isotoma palustris : m, manubrium; d, dens;
t, mucro. II., Extremity of fore foot. III., Byes and
post-antennal organ (the arrow is directing forwards.)
IV., Mucro. v., ' Catch ' : cp, corpus or basal portion ;
r, ramus.

inserted into the distal end of the tibia, and aie known as

tenent hairs (text-fig. 4). They appear to be modified

glandular setee, and exude from their extremities a small

drop of a sticky semi-gelatinous fluid. These hairs are

adhesive in their function, and they aid tlie insect in

climbing smooth or steep surfaces.

* These hairs are not peculiar to the Collenibola, but are found in
many larva? as well as in those aduit insects in which the adhesive
pads or pulvilli are wanting.

67

The abdomen uever consists of more than six seg-
ments ; the first segment always bears the ventral tube,
which represents a pair of fused appendages. Appendages
may also be present on the third (the " catch ") and fourth
(the springing organ) segments respectively (text-fig. 1).
The ventral tube is the most constant of all Collembolan
structures. It attains its highest degree of development in
the Sminthuridse, where it can emit a pair of long, tubular
vesicles (text-fig. 3). These may exceed in length the

V.t.---

Test-Fig. 3. — Smi7ithurus viridis. I., Springing organ: m, manu-
brium ; d, dens ; t, mucro. II., Extremity of fore foot.
III., Animal viewed from the side: vt, ventral tube;
k, ' catch ' ; ?», manubrium : d, dens ; t, mucro.

whole animal, and they are frequently studded with small
papillae. Almost every transition can be found between
this greatly developed condition to where the ventral tube
exists only in the form of a bilobed tubercle and capable
of only a very limited amount of protrusion.

By far the majority of CoUembola possess a pair of
partially fused appendages in relation with the fourth

68

abdominal segment and which function as a springing
organ to enable the Insect to take sudden leaps when
alarmed or otherwise disturbed. The force necessaiy io
generate the leap is obtained partly by the action of
special muscles and partly in virtue of the elasticity of the
organ itself. In form the springing organ is fork-like
(text-fig. 3), and is composed of an unpaired basal piece

Text-Fig. 4. — I., Compound hair of an Entomohnjn. II., Clavate
hair of a Tomocerus. III., Scale of Tomocerus vulgaris.
TV., Extremity of middle foot of Isotoma sensibilis,
showing three tenent hairs.

or manubrium (m), carrying at its distal end a pair of
arms, which are termed the dentes (d). Each dens
terminates in a small claw-like process — the mucro (t.) ;
the mucro varies greatly in form among different species,
and affords a useful character for the recognition of the
latter. In the Achorutidse the springing organ is either
greatly reduced in size, as in Poclura, Achorutes, and

69

Neaniira, or is wanting altog-etlier, as in AmiriJn,
and Anuiophorus.

Many species also retain a pair of appendages on the
third segment of the abdomen. These resemble the
springing org\m in their general form, but are very much
smaller in size. Similar to the spring, they consist of
a basal piece, which carries two arms ; the former is
termed by Tullberg the corpus [cp. in text-iig. 2), and the
two latter the rami (r.). This organ is usually known as
the " catch," or hamula, and it is believed to serve to
retain the spring in position when the latter is stowed
away under the abdomeu while not in use. The manu-
brium of the springing organ is retained between the two
rami of the " catch."

A variable number of eyes are geueraly present on
either side of the head behind the antennae ; their number
may vary even among different species of the same genus.
In Temjjletonia there is but a single eye on each side ; in
Anuni muscoruni there are three; in XcnyJla and Anuvida
marithna there are five ; in I'omoceiua and OvclieseUa six ;
while in Achorutes they reach the maximum number, viz.,
eight. In some CoUembola they are absent, as in
Lipur<i, Bechla and Anuvida granaria.

The post-antennal organ assumes a great variety of
forms among various genera. In Isotuma and Anuro-
phorus it is simple and ring-like, but in Lipura and
Anura it attains considerable complexity of structure.
In some few forms (i.e., Lipura) there may be present
curious structures, known as pseud-ocelli or puncta
ocelliformia. They are situated in front of the post-
antennal organ, and between it and the antenna; they
appear to have nothing to do with eyes, and their
function is unknown. In certain species of Apliovura
they have a much wider distribution over the body.

70

Sensory organs of varied form are usually present on
the ultimate and penultimate segments of the antennee.
An account of these structures is given by Absolon (6)
and also by Borner (7).

The mouth-parts in all Collenibola appear deeply
insunk within the head capsule ; this condition is a
secondary one, having been brought about through their
becoming overgrown and partially surrounded by the side
folds of the head. Those of Anura are greatly specialised
and ditfer from the mouth-parts of other Collembola in
being purely suctorial. In this genus the labrum and
labium together form a conical tube, in which the rest
of the mouth-parts are contained, the latter being
modified into stylets and used for piercing purposes.

The digestive canal is a simple tube passing from the
mouth to the anus without presenting any convolutions,
and this condition appears to be constant in all the genera
yet studied. The mid-gut of Neelus is remarkable on
account of its being sub-divided into a series of four
successive chambers. With the exception of salivary
glands, no accessory structures are associated with the
alimentary canal.

The nervous system appears to exhibit very little

degree of variation throughout the order.

Special respiratory organs are wanting, except in
species of Sminthurus, Smintliurides, and in Actaletes
nephini Giard ; in these forms tracheae are present. Trachse
are well known to be found in Sminthurus, and it is in
that genus that they are best developed. According to
Willem, on either side of the body, between the head and
the prothorax, there is placed a simple stigmatic crypt,
from which bunches of tracheae arise and are distributed
to the head, the legs and the abdomen ; * no anastomoses

* The exact position of the spiracles appears to require further
study. Lubbock and Tullberg maintain they are situated on the
head, while Haase states that they are placed on the prothorax.

71

take place between the traclieie of opposite sides of the
body. In Sminthurides and Actaletes "Willeni states that
the tracheal branches are entirely confined to the head,
while in Papirius he has discoyered two invaginations
which correspond in their position with the stigmata of
Sminthurus, and regards them as being the last remnants
of a tracheal system.

The heart in the Arthropleona is similar to that of
Aniirida ; in the Symphypleona it is much contracted,
not extending further backwards than about one-third of
the lensrth of the abdomen. Willem states that in this
sub-order it is only furnished with two pairs of ostia.

The reproductive system presents very little range of
variation ; a median ventral diverticulum of the genital
canal mav occur in the male [Sminthurus) as well as in
the female [Anurida).

YII.— THE AFFINITIES OF THE COLLEMBOLA.

The most marked characteristic of the Collembola is
the fact that the abdomen consists of only six segments,
and at no stage in the development are there known to be
more than that number present. Among all other insects,
including the Thysanura, wherever the embryology has
been studied, the abdomen has been found to consist of
not less than ten segments, together with a telson. The
significance of this feature of the Collembola has not been
emphasised by niorphologists, although it is one of the
most fundamental characteristics of the order. It is
reasonable to believe that this character has been acquired
from the primitive ancestors of the Insecta, whose meta-
merizatioii had not yet become fixed and limited to a
definite number of segments. In this respect, therefore,
the Collembola are very much isolated from the rest of the
Insecta, and the question arises as to where their affinities

72

lie. The sum total of tlie characters of the order indicate
that they are more closely allied to the Thysanura than to
any other order of Insecta, but that they also possess certain
generalised features in common with the lower Arthropoda.

The CoUemhola are allied to the Thysanura in that
they are the only two orders of Insecta which possess an
evident pair of mouth-parts — the maxillulee, which are
intercalated between the mandibles and first maxillse.
They further resemble the Thysanura in the absence of
any traces of wings, and in the fact that they undergo no
metamorphoses. It is with the genus Campodea that
their affinities come closest, and they resemble that genus
in their mouth-parts being withdrawn into the head
capsule, in the ovaries consisting of a single tube on
either side, there being no ovarioles, and in the absence
of an amnion and serosa in the embryo.

The Thysanura have been divided by Glrassi into two
divisions, viz., the Entotrophi, including the genera
Campodea and Japy.r, and the Ectotrophi, which com-
prise Machilis and hephma. The Entotrophi, as their
name implies, have their mouth-parts retracted within
the head, and it is true that they are more closely related
to the Collembola than is the case with the Ectotrophi.
The relationship, however, is not sufficiently close to
warrant the Collembola being included as a sub-division
of the Entotrojihi, as has been suggested by Stummer-
Traunfels (24). The difference in the number of segments
to the abdomen in the two groups presents a serious
difficulty to any such system of classification. It is by
no means certain that the entognathous condition of the
Collembola and Entotrophi is anything more than a
parallelism in evolution and, if that be so, it would be of
little value as a character implying a close relationship
between those two groups.

73

The presence of maxillulse in tlic Collembola is a
primitive feature wliieli is shared with the Symphyhx and
Diplopoda ; these mouth-parts are the representatives of
the Crustacean first maxillae {^vide Table on p. 74).
Furthermore, the occurrence of embryonic tritocerebral
appendages — the homologues of the Crustacean second
antennae, is also a primitive character which they only
share with the Thysanura.

In the general structure of the ovary, and the absence
of follicles, the Collembola, as Claypole has pointed out,
resemble the Myriapoda more closely than the Thysanura,
and come nearer 'to Scolopendrella than any other type.
An additional Myriapod feature is perhaps seen in the
post-antennal organs, which are peculiar to the Collembola
among Insecta. They are regarded by Willem as being
homologous with Tomosovary's organs in Myriapods, and
the recent researches of Hennings* and Hallert support
that conclusion.

In addition to the above mentioned characters, there
are a number of generalised features which suggest that the
Collembola are primitive animals. The most important
are : — (1) The extremely simple condition of the spiracles
where they are present ; as AVillem has pointed out, they
are nothing more than mere stigmatic crypts, and are to
all appearances similar to those of Peripatus. (2) The
nature of the digestive system, which is a perfectly
straight canal passing from the mouth to the anus
without ^Jresenting any convolutions. (3) The absence of
any external genital armature. (4) The relatively simple
nature of the eyes, each of which may be compared to a
single ommatidium of a compound eye.

Side by side with these generalised characters,

* Zeitschr. f. iciss. ZuoL, 7(J, lyOi, p. -2.6.
t Arch. f. itiiJ;. Anat., C5, 1905, p. 181.

74

1

1

c3 3 c

§3 S

Sj

S)

bo

o

1

C3

3 ®

0)

o

0)

o

o

a)
o

c

cd

S p

„ d p

d

d

d

'd

<

.=;

<

g

5i

M

= "3

ft

ft

ft

ft

a

rH

a o

.^

<n

IS

_J3

o

<

tH

■&

■'J'

s

CC

M

(A

c»

>^

■ J^

i

J .i

be

lU

bc

bC
lU

bxi

§

3

i

<u

s-t.

J

J

J

1^

? "^

1

i

s

3 C «
X S 0)

S 3 !U

§3 a

o

o

o

o

3

d

cj c3 p

Z; d d

'3

d

'3

'3

X 1

^

§

§K d

M

c-W

ft

ft

ft

ft

o

I—I

CQ ""'

-♦J

-3

r^

xj

3

1

'^

c5

CO

■<*

g

3

c
o

-*3

8

•la

d 3

01

o

bJO
0)

o

tJ)
o

.2S
S ^

O d

C
<

3

60

>

S3

3

%4

'd

ft

3

'3
ft

-^ 3

"^

o3

£

8 1

ffi

--

■3 §
Ik'

p

■- S'3"^

r- r^

^

^ §

t4-<

e^

u-<

^•-st:

<
O

<

1

X
d

§

o

ft

O

'3

ft

c

O

'd

ft

CO

d^ S
^ o S.

O CB =

t1

hH

i

o

H 1

s

ft

d

»

\A

f<

ce*

^

t

t' <u n

III

111
3
-a
§

2 .
Sot

8

8

li
SI

be

o

'5

o

'3

1

"o

d

(V

3

B

d

<

K

3
IM

ft

ft

ft

3
>

S go

l-(

CO

H =3

^

3

^

^

^

.s

H
O

O

<

o

c;

H

d

>1

3

0)

3

t3 O

3

d

lU

1

CD

.2*

d

8
%

5

3

d !-
ftS

+3 J2

3

<:

■g 3

Si

s

■a

433

a

g

I ■■'■

1

^niCi.

-<«

-

IN

I

"*

ira

o

c-

»

m

o

-

75

indications of very considerable specialization are
exhibited in the concentration of the nervous system, in
the mouth-parts being insunk within the head capsule,
in the structure of the springing organ and the ventral
tube, and in the reduction of the number of abdominal
segments to six.

Summing up, it may be said that the Collembola
show certain features in which they resemble the
Thysanura on the one hand and the lower Arthropoda on
the other while, at the same time, they have undergone
very considerable specializaition along lines of their own.
There do not appear to be any grounds for regarding them
as beius' degenerate animals. If the results of future and
more extended investigations definitely establish that
none of the Collembola have more than six abdominal
segments present in the embryo, their relationships with
the Thysanura and the rest of the Insecta will probably
prove to be much more distant than is implied by the
above remarks. It would assuredly be a sufficiently
fundamental character to separate them as a group from
all other Insects.

YIII.— MAEINE INSECTS.*

Contrary to what is usually believed among
naturalists, a considerable variety of insects are known to
inhabit the sea-shore below high-water mark, and to
undergo daily submersion during one or more periods in
their life-history. Very little attention, however, has
been devoted to them at present, and most probably a

* The term "marine" is only strictly applicable to insects in a
very limited number of instances. Although many of the species
included in the above account undergo frequent submersion by the
tide during one or more periods in their life-histories, others do not,
and are more strictly speaking to be regarded as "littoral" insects.

76

much larger nuniber are awaiting discovery. With very
few exceptioiis, marine insects are small in size, seldom
exceeding" more than a few millimetres in length, and
perhaps the only one which attracts the attention of the
casual observer is the fly Coelopa frigichi. This insect at
certain seasons of the year may be seen iiying in immense
swarms over the surface of decaving- Fuel and other Alsrse
cast up on the shore by the tides.

Plateau has drawn up a list of 40 genera, and
embracing 80 species of Tracheate Arthropods, w-hich are
known to inhabit the sea-shore and to be subjected to
submersion by the water. In the greater number of these,
he remarks, the power of living under such conditions is
not due to their possessing anj- special mechanisms to
enable them to do so, but to the general property these
animals have of being able to resist asphyxiation for
prolonged periods. His list includes Insects, Myriapods,
Pseudoscorpions, and Acari.

Most of what is known concerning marine insects is
comprised in the following works : —

A. S. Packard. — On Insects inhabiting Salt Water.

Am. Journ. Sci., 1871 ; also Froc. Essex Inst.,

18G8, and Ann. Mag. Nat. Hist., ser. -4, vol. viii.
r. Plateau. — Les Myriapodes Marins. Journ. de

r Anat. et Phys., xxvi., 1890.
L. 0. Miall. — The Natural History of Aquatic Insects.

London, 8vo.
G. H. Carpenter. — The Insects of the Sea. Knowledge,

1901.

The following list includes most of the marine Insects
inhabiting the British coasts, and has been principally
compiled from the writings of Plateau and Carpenter.

77

COLLEMBOLA.

Amouff the Collenibola a marine habitat has been
independently acquired by species belonging to both the
Entomobryidfe and Achorutidse but, up to the present,
none of the Sminthuridpe are known to frequent the
confines of the sea-shore.

In addition to Auuiida maritima, the following
species of marine Collenibola occur in the British Isles.
Those species indicated with an asterisk are not
exclusively marine, but are also found far inland.

Isotoma UttomliH, Men. *Xi'nyUa hnmicola, Tullb.

*I. paliistris (Miill.) *X. maritima, Tullb.

/. Schofti, Dalla Torre. Actalcte-s it<-j)tt()ii, Giard.

/. Beselsii, Pack. { = Isotoma rras.sicaiida,

*I. maritima, Tullb. Tulllj.)

*/. sexocnlata, Tullb. *Achon(tes viaticKS, Tullb.

Thysanura.
Machilis maritima (Leach).

COLEOPTERA.

The greater number of marine representatives of this
order consist of wingless forms, and many of them have
their bodies very much flattened, which enables them to
ensconce themselves in small crannies and fissures among
the rocks during the time they are submerged by the sea.
Their bodies are frequently covered with a coating of
hairs, which prevents them from being wetted, and also
encloses a certain amount of air, which is utilised during
respiration under water. Marine beetles hide themselves
by burrowing in the sand, or lurking under stones or
decaying seaweed ; at low tide they may be often seen
running over the rocks or sands.

78

Cercyon Uttoralis, Gyll.
C. depressiis, Steph.
Heteroceriis inarginatus.

Bose.
Li/nuiaeiiiii n'Kjropiceuni,

Marsh.
Cillemis lateralis Sam.
Treehina lapidosus, Daws.
Bemhidinm convinnum,

Putz.
B. ephippiinn, Marsh.
Aepus marinns, Stroem.
A. Bohi)i}i, Lab.
Micralymma bn'ripeunf,

Gyll.

TrogopJilof'Ks anglicanus,

Sharp.
Mynnffopova (Xen Jisa)

urida, Er.
M. sulcata, Kies.
Cafms cicatricosus, Er.
C. xanthnloma, Grav.

C. sericcus, Holme.
C.fitcicola, Curtis.

A ctocharis Beadingil.

Sharp.
Phytosus spiuiffr Curtis.
P. halticKs, Kraatz,
P. nigrirentiis, Chev.
Diglossa mcrsa, Hal.

D. silmaticolUs, Rey

DiPTERA.

A marked feature among marine flies is tlie reduction
or total atrophy of the wings. The first stage in the
reduction is seen in Chersodromia arenaria, where the
wings are much shortened so as to be of very little service
in flight. This is carried a step further in a Southern
France species, P samathiomya pedinata, Derby, in which
the wings have become greatly narrowed and strap-like,
and the nervures atrophied ; a similar condition is seen
in a Californian marine fly, Eretmoptera Browni, Kellogg.
An extreme case occurs in the females of the genus
CJitnio, Avhere the wings have disappeared completely,
although in the males they are tolerably well developed.

This reduction of the wings has been explained as
being an advantageous condition, since it prevents the
insects from being blown out to sea. A similar explana-

79

tion lias been made to account for the frequent occurrence
of wingless insects on oceanic islands. Among the few
insects, for instance, known from Kerguelen Land there
are three species of flies whose wings are atrophied, in
one instance they have gone altogther, and in the other
two cases they are reduced to mere scale-like appendages.

Marine Diptera are principally met with flying over
the masses of Ftici cast up by the tide on the shore : some
few, however, may be seen skimming lightly over the
surface of the water in the tide-pools. Most of the larvse
of these flies feed on thrown-up Fucus and are covered at
each tide, but those of the Chironomidse are submerged
during the whole course of their life, and subsist on
CladopJiora and other green Algse growing in the pools
left among the rocks by the receding tide and, further-
more, they have also been dredged up from a depth of
over ten fathoms.

The following British Diptera are all marine : —

Fucellia fucorum (Fall). Aphrosyliis raptor, Halid.

Coelopa frigida, 'Meig. A.ferox, Halid.

Orgijma luctosa, Meig. Thalassomi/ia Fraiuni-
Actora aestuum, Meig. fcldii, Schiner.

Glenanthe ripicola, Halid. Chironomns sp. ?

Chersodromia arcnaria, Clunio mariuus, Halid.

Halid. Clvmo hieolor, Kieff'.

Hemiptera.

AepojyhUxis Bonnairei, Sign.

The species of HaJohates and Halohatoides are
exclusively marine and inhabit the tropical waters of the
oceans. They have not been found further northward
than Spain.

80

IX.— LITERATURE.

In the following catalogue the more important papers relating
to the Collembola are enumerated, but it is not intended as a complete
bibliography of the order. The earlier literature of the subject has
been purposely omitted as references to practically all papers prior to
1870 will be found in Lubbock's " Monograph " and Tullberg's
" Sveriges Podurider."

The only general treatise is Lubbock's " Monograph of the
Collembola and Thysanura." published by the Ray Society in 1873.

A. — Bionomics.

L Davenport. C. B. The Collembola of Cold Spring Beach,
with special reference to the movements of the Poduridae ; Cold Spring
Harbor Moiiographs, No. II., Brooklyn, 1903.

2. Laboulbene, a. Recherches sur VAnurida maritima,
insecte Thysanoure de la famille des Podurides ; Ann. Soc. Eniom.
Fr., Ser. 4, T. IV., 1864, p. 705.

3. Lecaillon, a. Notes sin- I'habitat et les moeurs de quelques
Collemboles ; Bull. Soc. Philom. Paris, 9e Ser., T. III., 1900-01, p. 67.

4. MoNiEZ, R. Acai'iens et Insectes marins du Cotes du
Boulonnais ; Rev. Biol. Nord. Fr.. T. II.. 1889-90, p. 347.

5. WlLLEM, V. L'influence de la Lumiere sur la Pigmentation
de Isotoma tenebricola ; Ann. Soc. Entom. Belg., T. XLV., 1901,
p. 193.

B. — Morphology and Physiology.

6. Absolon, C. Uber Neanura tenebrarum nov. spec, aus den
Hohlen der mahrischen Karstes ; iiber die Gattung Tetrodontophora
Renter und einige Sinnesorgane der CoUembolen ; Zool. Anz.,
Bd. XXIV., 1901, p. 575.

7. BoENER, C. iiber das Antennalorgan III. der CoUembolen
und die systematische Stellung der Gattungen Tetracanthella Schott
und Actaletes Giard ; Zool. Anz. Bd. XXV., 1902, p. 92.

8. Fernald, H. T. The Relationships of Arthropods ; Studies
Biol. Lab. John's Hopkins Univ., Vol. IV., No. 7, 1890, p. 431.

9. FoLSOM, J. W. The Anatomy and Physiology of the Jlouth-
parts of the CoUembolan Orchesella cincin, L ; B^dl. Mus. Comp.
Zool. Harvard, Vol. XXX\\, No. 2, 1899, p. 7.

ftl

10. FoLSOM, J. W. The Development of the ]\Iouth-parts of
Anurida maritima Guer. ; Bidl. Miis. Comp. Zool. Harvard, Vol.
XXXVL, No. 5, 1900, p. 87.

11. Heymons, R. Uber die Bildung and den Bandes Darm-
canals bei niederen Insekten ; SB. Ges. Naturf. Berlin, 1897, p. HI.

12. Hoffmann, R. W. Uber den Ventraltubiis von Tomocerus
plumbeus L. vind seine Beziehungen zu den gropen iinteren Kopfdriisen.
Ein Beitrag zur Kenntnis der Collembolen : Zool. An-^. XXVIII.. Bd.
Nr. 3, 1904, p. 87.

13. Hoffmann, R. W. Uber die IMorphologie und die funktion
der kanwerkzeuge von Tomocerus plumbeus L. II., Beitrag zur
Kenntnis der Collembolen ; Zeits. f. iciss. Zool., Bd. LXXXII.. 1905,
p. 638.

14. Lecaillon, a. Recherches snr I'ovaire des Collemboles ;
Arch. d'Anat. micros., T. IV., 1901, p. 471.

15. LficAJXLON, A. Sur le testicule cVAnuroplwrus laricis ;
Bidl. Soc. Philom. Paris, 9e Ser. T. IV., 1902, p. 46.

16. Lecaillon, A. Sur le testicule d\-inurida maritima ; Bull.
Soc. Entom. Fr., 1902, p. 64.

17. Lecaillon, A. Sur la disposition, la structure, et le
fonctionnement de I'appareil reproducteur male des Collemboles ; Bidl.
Soc. Philom. Paris, 9e Ser., T. IV., 1902, p. 99.

18. Nassonow, N. The INIorphology of Insects of Primitive
Organization : Lepisma, Campodea and Pod^ira ; Stud. Lab. Zool.
Mus. Moskair, III., 1887, p. 15. (In Russian, but useful for figures).

19. Olfers, E. de. Annotationes ad Anatomiam Podurarum,
Diss, inaug. Berolini. 1862.

20. OuDEMANS, J. T. Beitrage zur Kenntnis der Thysanura
und CoUembola, Bijdr. tot Dierh, Afiev. 16, 1888, p. 147. (Is a
translation of " Bijdrage tot de Kennis der Thysaniu-a und CoUembola,"
Amsterdam, 1887, with a few brief additions).

21. Prowazek, S. Bau und Entwickelung der Collembolen ;
Arbeit. Zool. hist. Wien, Bd. XII., 1899-00, p. 335.

22. Reuter, 0. M. Sur la fonction du tube ventral des
Collemboles ; Entom. Tidslcr., 1 Jahrg, 1880, p. 162.

23. Sommer, A. Uber Macrotoma plumbea. Beitrage zur
Anatomic der Poduriden ; Zeits. /. tviss. Zool., Bd. 41, 1884-85, p. 683.

24. Stummer-Traunfels, R. R. von. Vergleichende Unter-
suchungen iiber die Mundwerkzeuge der Thysanuren und Collembolen ;
SB. Alcad. Berlin. Bd. 100, Abth. 1, Heft. 4, 1891, p. 216.

25. Willem, V. Les yeux et les organes postantennaires des
Collemboles ; Ann. Soc. Ent. Bdg., T. XLL, 1897, p. 225.

G

82

26. WiiXF.M, Y. Les glandes oephaUques des? Orcheselles ;
Arch, de Biol.. T. XVII.. 1900. p. 6.53.

27. Wii.le:m. V. Eecberches sur les Collemboles ct les
TbysanoTires ; 2Wm. conr. VAcad. Sci. Belg., T. LVIII., 1900.

28. WiLLEM, V. ET Sabbe, H. Lc tiibe ventral et les glandes
cephaliques des Sminthurus ; Ann. Soc. Entom. Belg., T. XLI., 1897,
p. 130.

C. — Development.

29. Gael, J. Snr un organe einbroj'onnaire ehez un C'ollembole;
Arch. Set. Phys. Nat. Geneve. Ser. 4. T. 15. 1903, Compte-rendu p. 601.

30. Claypole, a. I\I. The Embryology of the Apterygota ;
Zool. Bull.. Vol. II., No. 2, 1892, p. 69.

31. Claypole, a. 'M. Tlie Embryology and Oogenesis of
Anurida maritima (Gner.) ; Jovrn. Morph., Vol. 14, 1898, p. 219.

32. FoLSOM, J. W. Loc. cit. No. 10.

33. Lemoine, V. Eeeherches sur le developpement des Podu-
relles; A.ssoc. Fr. p. VAvanc. ScL, 1882, p. 483.

34. OuLGANiNE, W. N. (or Uljanin). Sur le developpement des
Podurelles ; Arch. Zool. Gen. et Exp.. T. IV., 1875, p. xxix., and
T. v., 1876, p. xvii.

35. Packard, A. S. Embryological Studies on Diplax,
Perithemis, and the Thysanurous Genus Isotoma ; Mem. Peabody
Acad. Sci., Vol. I., 1871.

36. Prowazek, S. Loc. cit. No. 21.

37. Ryder, J. A. The Development of Aniiridn maritimn Gucrin ;
Amer. Nat., Vol. 20, 1886. p. 299.

38. UzEL, H. " Studien iiber die Entwickelung dcr Apferygoten
Insecten." BerUn, 1895.

D. — Systematic and Faunistic.

39. Absolon, K. Vorlaufigo Mittheilung uber die Aphoruriden
aus den Hohlen des mahrischen Karstes ; Zool. Anz., Bd. XXIII.,

1900, p. 406.

40. Absolon, K. Uber einige theils neue CoUembolen aus den
Hohlen Frankreichs und des siidlichen Karstes ; Zool. A')iz., Bd. XXIV.,

1901, p. 82.

41. Absolon, K. Weitere Naluicht iiber europaische Hohlen
CoUembolen und iiber die Gattung Aphoriira, A. D. MacG. ; Zool.
Anz., Bd. XXIV., 1901, pp. 375, 385.

83

42. Absolon, K. Loc. elf. No. 6 (and otlier papers by the autlior
in tlie same journal).

o

43. Agren, H. Lapplandische Collenibola ; Arkiv. ZooL. 1904.
II., No. 1.

44. BoRNEB, C. Vorlaufige Mittheilung ziir Systeinatik der
Sminthuridse Tullb., insbesondere des genus Sminthurus Latr. ; Zool.
Anz., Bd. XXIII., 1900, p. 609.

45. BoBNEB, C. Vorlaufige Mittheilung (iber einige neue
Aphorurinen und zur Systematik der C'oUembola ; Zool Ariz., Bd.
XXIV., 1901, p. 1.

46. BoRNER, C. Uber ein neues Achorutidengenus Willemia,
sowie 4 weitere neue Collembolenformen derselben Familie ; Zool.
Anz., XXIV., 1901, p. 422.

47. BoRNER, C. Neue Collembolen formen und zur Nomenclatur
der Collenibola Lubb. ; Zool. Anz., XXIV., 1901, p. 696 (and other
papers by the author in the same journal).

48. BoRNER, C. Zur Kenntnis der Apterygoten-Fauna von
Bremen und der Nachbai'distrikte. Beitrag zu einer Apterygoten-
Faima Mitteleuropas ; Abh. Ver. Bremen, XVII., Heft 1. 1901, p. 1.

49. BoRNER, C. Loc. cit. No. 7.

50. Brook, G. Notes on some little known Collenibola, and on
the British species of the genus Tomocerus ; Journ. Linn. Soc. (Zool.),
Vol. XVII., 1884, p. 19.

51. Brook, G. A Revision of the genus Entonidbrya, Rond.
{Degecria, Nic.) ; Journ. Linn. Soc. (Zool.), Vol. XVII., 1884, p. 270.

52. Boyd, D. A. Collembola and Thysanura in the Nat. Hist,
of Glasgow and West of Scotland, published by the Local Committee
for the meeting of the Brit. Assoc, for the Advanc. Sei., Glasgow,
1901, p. 317.

53. Carpenter, G. H. Tiie Collembola of Mitchelstown Cave ;
Irish. Nat., Vol. VI., 1897, pp. 225, 257.

54. Carpenter, G. H., and Evans W. The Collembola and
Thysanura of the Edinburgh District ; Proc. Roy. Phys. Soc. Edin.,
Vol. 14, 1899, p. 259.

55. Carpenter, G. H. Collembola from Franz-Josef Land ;
Sci. Proc. Roy. Dublin Soc, vol. 9 (n.s.), 1900, p. 271.

56. Carpenter, G. H. Collembola ; in Fauna Hawaiiensis,
Vol. III., 1904, p. 299.

57. Carpenter, G. H., and Evans, W. Some Spring-Tails new
to the British Fauna, with description of a new species ; Proc. Roy.
Phys. Soc. Edin.. Vol. XV., 1904, p. 215.

84

58. Carl, J. Ueber Scliweizerisclie Collembola ; Bei\ Suisse
Zool, VI., 1899, p. 273. Do. IX., 1901, p. 243.

59. Dalla Torre, K. W. v. Die Gattnngen iind Arten der
Apterygogenea (Braner), Sep. 40; Prog. K.K. St. Gym.. Innsbruck,
1895.

f>0. Evans, W. 8ome Recordsof Collembola and Thysanura from
the "Clyde" area; Ann. Scot. Nat. Hist., No. 39, 1901, p. 154.

61. Evans, W. A Preliminary List of Collembola and
Thysanura; Trans. Perth Soc. Nat. Sci., III.. 1901, p. 150.

02. FoLSOM, J. W. Japanese Collembola. pt. I. ; Bull. Esse.v
Inst., XXIX., 1897, p. 51 ; pt. II.. Proc. Amer. Acad. Arts and Sci.,
XXXIV., 1899, p. 261.

03. FoLSOM, J. W. The Distribution of the Holarctic Collem-
bola ; Psyche, vol. 8, 1901, p. 159 ; (and other papers by the author in
the same journal.)

64. FoLSOM, J. W. Apterygota, Papers from the Harriman
Alaska Expedition, XXVII. ; Proc. Washington Acad. Sci., Vol. IV.,
1902, p. 87.

65. Giard, a. Sur un nouveau genre de CoUembole marin et
sur I'espece type de ce genre : Actaletes Neptuni ; Le Naturaliste,
II., Ser. II., No. 53.

66. Guthrie, J. E. " The Collembola of Minnesota " ; Geol. Nat.
Hist. Survey. Minn., Zool. Series, No. 4, 1903.

67. KoLENATi, F. A. Zwei neue osterreichische Poduriden ; SB.
Alcnd. Wien, Bd. 29, 1858.

08. Lie-Pettersen, 0. J. Norges Collembola ; Bergens 7nus.
aarh., 1890, No. 8.

09. Lubbock, Sir J. On some Spitzbergen Collembola ; Journ.
Linn. Soc. (Zool.), Vol. XXVI., 1898, p. 610.

70. Lubbock, Sir J. On some Australasian Collembola ; Journ.
Linn. Soc. (Zool), Vol. XXVII.. 1899, p. 334.

71. MacGillivray, A. D. A Catalogue of the Thysanoura of
North America; Canadian Entom., Vol. XXIII., 1891, p. 207.

72. MacGillivray, A. D. North American Thysanoura ;
Canadian Entom., Vol. XXV., 1893, pp. 127. 173. 218, 313 ; Vol. XXVI.,
1894, p. 105.

73. MacGillivray, A. D. The North American species of
Isotoma ; Canadian Entom., Vol. XXVIII., 1896, p. 47.

74. Moniez, R. Notes sur les Thysanoures ; Rev. Biol. Nord. Fr.
IT., 1889-90, pp. 24, 365, 429; III., 1890-91. pp. 64. 68 (and other
papers by the author in the same journal).

85

75. ^Iarquand. E. D. The Thysanura and Coleoptera of the
Land's End District ; Rep. and Trans. Penzance Naf. Hist, and Antiq.
Soc, 1880-81, p. 52.

76. OuDEMANS, J. T. iSystcmatische Beschrijvino- der in
Nederland voor Komende Thysanura ; Tidschr. v. Entom.. Nederl.
Entom. Vereen, 38 D., 4 Afl., 1891, p. 104.

77. OuDEMANS, J. T. Apterygota des Indischen Archipels ;
Weber's Ergebn. Reise Nederl. Ostindicn, 1890, Bd. 1, p. 73.

78. Parona, C. Collembola. Saggio di \n\ C'atalogo delle
Poduridi italiane : Atti. soc. ital. sc. naf., 21, p. 559. (For references
to other papers by this author inde Schott, No. 93.)

79. Pack.\rd, a. S. Synopsis of the Thysanura of Essex Count}',
Mass., with Descriptions of a few extra limital forms ; Fiffh rep.
Peabody Acad. ScL. 1873. p. 23.

80. Parfitt, E. Devon Collembola and Thysamu-a ; Trans.
Devon. A.ssoc. Advnnc. Sci.. Vol. 23, 1891, p. 322.

81. Reuter, 0. jM. Catalogus prsecursorius Poduridarum
Fennise; Mcdd. Soc. Faun. Flor. Fenn., Vol. I.. 1876, p. 78.

82. Reuter. 0. M. For Finland nya Collembola ; Medd. Soc.
Faun. Flor. Fenn., VI., 1878, p. 203.

83. Reuter, 0. M. Podurider fran nordvestra Sibirien, samlade
af J. R. Sahlberg ; Ofv. finsk. vet. soc. fork., Bd. 33, 1891, p. 226.

84. Reuter, 0. M. Collembola in caldariis viventia ; Medd. Soc.
Faun. Flor. Fenn., XVII., 1890.

85. Reuter, 0. M. Apterygogenea Fennica ; Acta. Soc. Faun.
Flor. Fenn., Bd. XL. 1895. p. 1.

86. Reuter, L. and 0. M. Collembola and Thysanura found in
Scotland in the summer of 1876 ; Scot. Nat., Vol. V., 1879-80, p. 204.

87. Schaffeb, C. Die Collembola der Umgebung von Hamburg
und benachbarter Gebiete ; Mitt. Naturh. Mus. Hamburg, Jhg. 13,
1896, p. 147.

88. Schaffer, C. Hanil)urglier Jlagalhaensische Sammeheise —
Apterygoten, 1897.

89. Schaffer, C. Die Collembola des Bismarck-Archipel nach
der Ausbeute von Prof. F. Dahl ; Arch. Naturg., Jhg. 64, Bd. 1,
Hft. 3, 1898, p. 393.

90. Schaffer, C. Die Arktischen und Subarktischcn Collem-
bola; Faima Arctica, Bd. 1, 1900, p. 257.

91. Schaffer, C. Ueber Wiirttembergische Collembola ; Jahresh.
Ver. WiirHemb., Bd. 56, 1900, p. 245.

86

92. ScHERBAKOF, A. M. Einige Beinerkimgen iiber Aptery-
gogenea, die bei Kiew, 1896-7, gefiinden wurden ; Zool. Anz., Bd. 21,
1898, p. 57.

93. ScHOTT, H. Znr Systematik und Verbreitung palaearctischer
Collembola ; Kongl. Svenska Vetens.-Akad. Handl., XXV.. 1893, No. 11.

94. ScHOTT, H. Etudes snr les CoUemboles du Nord : Bih.
Svenska Vetens.-Akad. HnndL, Bd. 28, Afd. 4. 1902. No. 2.

95. ScHOTT, H. North American Apterygogena ; Proc. Calif.
Acad. Sc, Ser. 2, Vol. 6. 1896, p. 169.

96. TuLLBERG, T. Sveriges Podurider ; Kongl. Svenska Vete7is.-
Akad. Handl, X., No. 10. 1872.

97. TxjLLBERG, T. Collembola Borealia; Ofr. Kongl. Svenska
Vetens.-Akad. ForJi., XXXIII., 1876, p. 23.

98. UzEL, J. Thysamira Bohemia ; Sitzher. K. bohm. Gesell.
Wiss., 1890, p. 3.

99. Wahlgreen, E. Collembola wahrend der Schwedischen
Gronlands-expedition, 1899, aiif Jan Mayen iind Ost-Gronland ein
geesam melt ; Of v. Ak. Forh., LVII., 1900, p. 353.

100. WiLLEM, V. Description de Actaletes neptuni Giard ; Bull.
Sci. Fr. ct Belg., XXXIV., 1901, p. 474.

101. WiLLEM, V. CoUemboles; Bcsult. Voyage Belgica, Zool.,
1902.

102. WiLLEM, V. Les rapports cV Actaletes avec 1©3 autres Col-
lemboles; Ann. Soc. Ent. Belg., T. XLVI.. 1902, p. 11.

87

X.— APPENDIX.

The Parasites of Collembola.

Mincliiu (Lankester's Treatise on Zoology, Part I.,
2nd Fascicule) enumerates two species of Sporozoa as
being parasitic in Collembola, viz.: — Gregarina pcdurte
(Leger) from the gut of Orcliesella villosa and Glugea
thysaimra' (L. Pfr.) from the gonads of Poduni (iquatica.
The occurrence of Sporozoa in Collembola has also been
referred to by Lubbock and Sommer.

The presence of Nematodes in the gonads of Anurida
is recorded in this Memoir, and Sommer states that he has
met with them in Tomocerus {]\Iacrotoma).

The present writer has recorded {Ann. Mag. Xat.
Hist., ser. 7, Vol. XV., 1905) the habits of a marine
Pseudoscorpion, Ohisium mai-ithnum. Leach, which
probably preys upon Anurida.

Note on Anurida crassicornis, Peut.

L. and 0. M. Peuter (86) describe this species from
the river Tay, near Perth, and state that it is very like
and allied to A. maritima, but differs in its much shorter
and thicker antennae, in the structure of the head (the
inter-antennal lobe being pentagonal, with almost straight
sides and well separated from the rest of the head) and in
the more robust body. The description is unaccompanied
by any figures, and, according to Carpenter and Evans
(54), the types are unfortunately lost.

The Economic Aspect of Collembola.
Springtails have not up to the present attracted much
attention from economic biologists, though occasionally
instances are known where they have done injury to

88

cultivated plants [vide Carpenter, Prac. Ass. Econ. Biol.,
Vol. I., Pt. I., July, 1905). Dr. Traquair (Beport to the
Plans and Worlcs ('oinin/ft/e of the Edinburgh T'own
Council, March, 100(i ; and The Scotsman, March 2ud,
190G) has recently dealt with the occurrence of Collembola
in hydrant boxes and water cisterns in Edinburgh.
Although the presence of these insects in such situations
is in itself harmless, the abundance in which they
were found is evidence that they find an ample food
supph' there, and herein lies the danger, for it points to
the water being contaminated.

Note on Histological Techmque, &c.

Specimens of Anurida are best collected from off the
rocks and weed by means of a camel-hair brush, and
transferred into a phial containing TO per cent, spirit ;
if it be desired to obtain them in large numbers, they can
frequently be scooped up in quantities from off the surface
of the tide-pools, using for the purpose a metal spoon or
some similar instrument.

The mouth-parts and sense organs can be easily
studied by treating the head with a weak solution of
potassic hydrate, which, by dissolving the soft parts,
renders it gradually transparent. This should be
performed on a slide, under the lower power of the
microscope, and the action of the alkali checked, as soon
as the mouth-parts assume a pinkish-violet, by running in
alcohol under the cover-glass.

If it be necessary to preserve the animals for histo-
logical purposes, they are best brought away alive in a
vessel containing sea water, and then transferred with a
camel-hair brush into hot TO per cent, alcohol, and after-
wards passed into that of 90 per cent, strength. Hot
corrosive sublimate or picro-sulphuric acid can also be

89

used with advantao'e for killing aud fixation. Whichever
reagents are used, the specimens shouhd be afterwards
placed in 90 per cent, alcohol, to which have been added
about five drops of potassium livpochlorite (concentrated
solution j to each fiuid ounce of the alcohol. After
remaining in this for about half an hour, they sliould be
washed in fresh alcohol of similar strength and then
passed into absolute alcohol. The potassium hypochlorite
serves to remove the pigment, and also softens the
chitinous parts and aids in section cutting.

The process of embedding is best performed in watch-
glasses, using paraffin of a low-melting point, which is
afterwards replaced by hard paraffin for cutting. For
staining, the best results were obtained by using
Manns methyl-blue-eosin and Heidenhains iron-alum
hsematoxylin. All staining was done on the slide by
prolonged treatment, and afterwards washed out to the
required degree of colouration.

The eggs were killed and fixed in both hot water and
hot TO per cent, alcohol, and afterwards kept in 90 per
cent, alcohol for five weeks for hardening purposes. After
this treatment, the egg membranes can be dissected away
from the embryos with the aid of a pair of fine needles
under a low-power microscope. The embryos can then be
stained with borax carmine, and mounted whole
in glycerine for examination. For instructions as to
sectioning the eggs, vide Folsom (10).

90

XI.— EXPLANATION" OF PLATES.

liEFEiiEACE Lettering.

a, b = Direction of the section in

fig. 52.
ab. app. 1-^ = Abdominal api)en-

abd., !• ^■'' =^'"1 '' = Abdominal seg-
ments.
a. bib. = Last chamber of heart.

a. gl. = Acinose gland of ventral

tube.
al. m. — Alary muscle.
al. t. = Tendon of alary muscle.
an. = Antenna.
an' = Anus.

at. n. = Antennal nerve.
b.c. = Blood corpuscles.

b. cav. = Body cavity.

b.m. = Basement membrane.

c. = Inner layer of cuticle,
c' = Outer layer of cuticle.
car. = Car do.

cav. — Foramen leading into

cavity of mandible.
ch. = Cliromatin.
ch.' = Outer membrane of egg.
chb. = Common Chamber.
ch. h. = Chromatin body.
ch. e. = Torn edge of outer egg

membrane.
circ. m. = Circular muscle fibres

of oesophagus.
chii. = Clypeo-labral rudiment.
ch/p. — Ch'peus.
c. tn. — Circular muscle fibres of

mid gut.
c. m. b. = Circular muscle l)and.s.
c. m. j. = Union of circular

muscle fibres.

conn. = Para-oesophagea con-
nective.

conn. '■ ''■ '■' — Connectives of ven-
tral nerve cord.

c.p. — Central mass of protoplasm.

c. t. j = Connective tissue

c'. t'. ?«'. i membrane.

c. t. ?«. = Connective tissue coat

of mid gut.
cut. = Cuticle.

d. b. — Dark transverse band.
d.ep. — Cup-like depression.
den' cell. = Deutocerebron.

dil. an. ni. — Dilator muscles of

anus.
dd. reel. in. = Dilator muscles

of rectum.
div. = Median diverticulum of

vagina.
drc. = Deutocerebrum.
d. s. j = Successive stages in

d.' s.' development of

d.^ s.'^ ] spermatozoa.

d. t. — Duct of the acinose gland.
dil. = Ductule of salivary gland,
ec. = Ectoderm.
em. nem. = Nematode embryo.
en. = Future endoderm cells.
ep. = Epithelium.
epcr. = Epicranium.
epith. = Fore-gut epitheUum.
epith. th. = Epithehal thickening.
ex. g. = Excretory granules.
exc. or. = Fatty body.
gal. = Cialea.
g.c. = Ganglion cells,
g ' c' = Germ cells.

91

rjl. = Tubular gland.

gl. ap. = Aperture of tubular

glands.
gl. dt. I = Ducts of tubular
gl. ' dt. ' ' glands.

gon. = Gonad.

g. p. = Lateral pouch of mid-gut.
g. t. = Germinal tissue.
/;. = Hair.

h. c. = Modified hypodermis cell.
h. cell. = Trichogenous cell.
M. = Hypodermis.
A.' d.' = Ventral surface of head

(2nd maxillae).
h. g. = Hind "gut.
/(. g. cpiV/i.=Hind gut epithehum.
li. g. m. = Circular muscles of

hind gut.
hs. = Striated hem or " hiirchcn-

saum."
M. = Heart.
inf. oes. g. = Infra -oesophageal

ganglion.
inf. oes. m. = Posterior series of
dilatores pharyngii miiscles.
i. p. = Intra -testicular prolon-
gations.
/,■. //(. = Krause's membrane.
I. = Lumen.
lac. = Lacinia.
lb. = Labium (2nd maxillse).
I. c. = Lentigen cell.
ling. = Lingua, or tongue.
ling. St. = Lingual stalk.
I. m. = Longitudinal muscle fibres.
In. — Lens.
I. 11. = Lateral nerve.
Ir. — Labrum.
I. s. in. = Longitudinal sternal

muscles.
/. t. m. — Longitudinal tergal
muscles.

m.

= Muscles.

md. = Mandible.

nvi. g. = Mandibular ganglion.

md. p. = Pharyngeal pocket of
left mandible.

med. f. = Medifurca.

memb. = Inner egg membrane.

mes. = ilesoderm.

?«. /. = Muscle fibres.

m. g. = Mid-gut.

m. g. epith. = Mid-gut epithelium.

m. n. = Median nerve.

m.' n.' = Mandibular nerve.

)n. n. br. = Lateral branch of
median nerve.

m. 71UC. — Nucleus of muscle fibre,

ino. = Mouth.

;«. pr. = Male pro-nucleus.

m. s. = Line of apposition of dis-
tal portions of 2nd maxilLe.

muse. = Dilator muscles.

m. IV. = Muscular wall of heart.

»n.r. Jid. = Head of 1st maxiUa

m.i-.' g. = Ganglion of 1st maxilla.

>n.r.- g. = GangUon of 2nd maxiUa.

)n.v.' n. = Nerve to 1st maxilla.

»«.r.- n. = Nerve to 2nd maxilla.

mx. p. = PharjTigeal pocket of
left maxilla.

mxid. = Maxillula.

III. V id. g. = Ganglion of maxillula.

inxul. n. = Nerve to maxillula.

n.

: Nucleus.

nuc. '

nem. = Nematode worm.

nem.' = Transverse section of

Nematode worm.
nenr. = Neurilemma.
n. /. = Nerve fibres.
n. V. \ = Abdominal nerve
«.' V.' I (or nerves).

92

o. — Ovum.

o.' = Ovules.

oc. — Eyes.

od. = Oviduct.

oes. = Oesophagus.

o.n. = Bi'anches of optic nerve.

op. n. = Optic nerve.

ost. = Ostium.

Of. = Ovary.

o. IV. = Ovarian wall.

J). = Pigment.

f. a. 0. = Post-antennal organ.

-pc. o. = Precephalic organ.

per. oes. a. = Peri -oesophageal
aorta.

ph. = Pharynx.

2J.i. = Protoplasmic island in
which female pro-nucleus is
present.

pi. = Side fold of head.

p. I. = Periplieral layer of proto-
plasm.

plm. = Coagulated blood plasma.

pip. = Maxillary palp.

p. o. n. — Nerve to post-antennal
organ.

pre. — Protocerebrum.

pr' ceb. = Protocerebron.

ps. = Pseudopodium.

r. = Protoplasmic strands.

r. nuc. = Nucleus of fat body.

rep. ap. = Reproductive aper-
ture.

ret. = Retinal cell.

ret. I. = Layer of fat body border-
ing on hypodermis.

ret. miisc. — Retractor muscles of
ventral tube.

s — Blood containing space.

scrp. = Sarcoplasm.

sec. = Secretion of gland.

s.g. = Secretory granules

s. gl. = Salivary gland.

slmma. = Sarcolemma.

s. o. = Sense organ.

s.' o.' = Possible rudimentary
sense ory-ans.

sp. = Cavity formed by breaking

down of central cells of fat

body.
spr. = Bunches of spermatozoa.
St. = Stipes.
str. p. = Striated portion of

muscle fibre.
sup. oes. g. = Brain.
sup. oes. m. = Anterior series of

dilatores pharvncrii muscles.
snt. = Suture.
^1 :;. (. _ Abdominal sterna.
tent. = Tentorium.
/. /. = Terminal filament.
th. = Thorax.
//t.i--' = Pro- meso- and mcta-

thorax.
Ih. app.^^-'' = Thoracic appen-
dages.
th. gr.i~-^= Pro- meso- and meta-

thoracic ganglia.
t. p. — Tunica ]oropria.
tr. CI pp. = Tritocerebral ?,ppen-

dage.
trc. = Tritocerebrum.
tr' ccb. = Tritocerebron.
t. s. = Substance of testis.
t. s. muse. = Tergo-stornal

muscles.
t. 10. = Wall of testis.
vars. = Vacuoles.
vaj. = Vagina.
valve c. m. = Pyloric valve.
V. c. — Nutritive cells.
V. e. — Epithelial layer of vagina.
V. g. = Ventral groove.
V. m. = Traasverse raiTScle.
vs. = Terminal vesicles of ventral

tube.
V. t. = Ventral tube.
iv. — Connective tissue coat of

heart.
x = Organ of unknown function.
y. = Yolk.
y. c. = Yolk cells.

93

Pl.\te I.

Fig. 1. Anurida maritima, adult male viewed dorsally.

X 35.
Fig. 2. A group of Anuvida showing their general

appearance when alive. Nat. size.

Fig. -3. Ventral aspect of adult female (sluidiiig
omitted). The ventral groove is seen as a
slightly sinuous line arising from a point in
the middle line of the head and passing mid-
way between the bases of the legs to terminate
on the anterior border of the ventral tube.
X 35.

Fig. 4. Outline figure of the dorsal side of the head
showing the position of the sense organs.
X circa 50.

Fig. 5. llight lateral oval protuberance of the head,
together with the five eves of its side. The
post-antennal organ is seen immediately in
front. X 230.

Fig. 6. A piece of the chitinous cuticle from the
abdomen showing the tubercles and some of
the large seta-like hairs. x 200.

Fig. 7. Terminal joint of left antenna viewed from its
inner aspect. It shows the trilobed apical
sense organ together with patches of modified
cuticle, which are probably of a sensory
nature. x 180.

Fig. 8. The apical sense-organ of the left antenna,
x 600.

Fig. 9. Claw of left fore foot, x 420.

Fig. 10. Eight post-antennal organ. x 1,200.

94

Fig. 11. Diagram for the purpose of showing the
relative positions of the mouth-parts to one
another when viewed from the ventral surface.
X circa 100.

Fig. 12. Ventral aspect of the apex of the 2nd maxillae
(labium). This figure and the succeeding one
are from preparations which have been depig-
mented to show the sculpturings of the cuticle.
X 160.

Fig. 1-3. Dorsal aspect of labrum and clypeus. x 160.

Plate II.

Fig. 14. Dorsal aspect of the maxillulee, the lingua and
the left hrst maxilla — soft parts removed.
X 360.
Fig. 15. Dorsal aspect of right mandible. x 360.
Fig. 16. A^ertical section of the integument, together
with a parietal extension of the fat-body.
From the ventral region of the meso'-thorax.
x 800.
Fig. IT. Vertical section of the integument passing
through the base of a hair and a trichogenous
cell. X 800.
Fig. 18. Mass of fat-body from the thorax : excretion

scarcely yet commenced. x 400.
Fig. 19. Mass of fat-body from the abdomen : excretion
actively taking place. x 400.
Mass of fat-body from the abdomen : excretion
far advanced. x 320.

Longitudinal and vertical section of the ventral
tube. X 200.

Apex of antenna showing fine branches of
antennal nerve passing to sensory organ and
hairs. x 180.

Fig.

20,

Fig.

21

Fiff.

95

Fisr. 2'3. Horizontal section taken across the vesicles of

the ventral tube. x 200.
Fig. 24. Portion of the ventral groove : region of the

meta-thorax. x 200.
Fig. 25. Organ of unknown function situated on the

floor of the fourth abdominal segment, x 820.
Fig. 26. Three successive transverse sections through the

ventral groove ; (a) in the head region ;

(h) in the prothorax ; (c) in the meta-thorax.
X 800.

Plate III.
Fig. 27. Eeconstruction of sections taken through the

head close to the middle plane. x ;360.
Fig. 28. Section passing through the junction of the

fore-gut with the mid-gut. x 360.
Fig. 29. Transverse section across the oesophagus.

x 600.
Fig. 30. Section of wall of mid-gut. x 1,200.
Fig. 31. Transverse section taken across the rectum

together with the vagina and its niediau

diverticulum. x 550.
Fig. 32. Transverse section of the ductus ejaculatorius.

x 550.
Fig. 33. Blood corpuscles from a living animal, x 1,200.
Fig. 34. Tertical section through the post-antennal

organ and two of the eyes (partly after

Willem). X 1,000.

Plate IV.
Fig. 35. Diagrammatic figure of AnuviJa showing the

relations of the principal internal organs to

O'ue another. x circa 40.
Fig. 36. Figure showing the termination of the ventral

96

groove on the anterior aspect of the ventral
tube. X 45.

Fig. 37. Section of the mid-ventral region of the meso-
thorax showing the relations of the nervous
system, medifurca and the ventral groove.
X 200.

Fig. 'IS. Vertical and transverse section of tlie ventral
tube. X 180.

Fig. 39. Portion of the heart showing the non-striated
circular muscle bands. x 550.

Fig. 40. Vertical and longitudinal section of a chamber
of the heart. x 550.

Fig. 41. Portion of the heart showing alary muscles.
X 550.

Fig. 42. Transverse section of the heart in the posterior
region of the oesophagus, x 420.

Fig. 43. Transverse section of the heart, together with
the alary muscles ; third abdominal segment.
X 420.

Fig. 44. Transverse section of the heart ; fifth
abdominal segment. x 420.

Fig. 45. Transverse section across three of the longi-
tudinal tergal muscles. x 400.

Fig. 40. Longitudinal section through the first and
second thoracic ganglia, taken slightly to the
left side to show the median accessory nerve.
X 150.

Fig. 47. Longitudinal section through the meta-
thoracico — abdominal ganglion. x 320.

Fig. 48. Semi-diagrammatic reconstruction of the
nervous system of Anurida. x circa 100.

97

Plate Y.

Fig. 49. Longitudinal section through the posterior

half of the alimentary canal. x 550.
Fig. 50. Transverse section across the fourtli alidoniiiial

segment. x 200.
Fig. 51. Transverse section across the posterior region

of the prothorax. x 200.
Fig. 52. Transverse section across the head. x 200.

The section is taken along the line a, h in

fig. 27.

Plate YI.

Fig. 53. Diagrammatic figure showing the position of

the ovaries (mature). x circa 10.
Fig. 54. Diagrammatic figure showing the position of

the testes. x circa 15.
Fig. 55. Apex of ovary and base of terminal filament.

X 400.
Fig. 56. Longitudinal section through the ovary

showing developing eggs. x 180.
Fig. 57. Transverse section across the ovary through the

germinal tissue. x 050.
Fig. 58. Apex of testis and base of terminal filament.

x 400.
Fig. 59. Longitudinal section through the hinder third

of the testis. x 400.
Fig. 59(7. Transverse section across the testis, passing

through the germinal tissue. x 400.
Fiff. GO. Longitudinal section of a tergo-sternal muscle

at its point of insertion. x 400.
Fig. Gl. Portion of musculature of mid-gut, showing

junction of the circular fibres in the median

dorsal line. x 400.
Fig. G2. Ditto of hind-gut. x 550.

98

Fig. 63. Saliyary and tubular glands of AnurlJa.

X 450.
Fig. 64. f'ommon chamber receiving the ducts of the

tubular glands (partly diagrammatic), x 400.
Fig-. 65. Diagram of head showing- relations of the

cephalic glands. x circa 65.
Fig. 66. Acinose gland of ventral tube in longitudinal

section. x 450.
Fig. 67. Aperture of the glands of the ventral tube;

section taken parallel with the long axis of the

body, x 400.
Fig. 68. Portion of duct of salivary gland. x 500.

Plate TIL

Fig. 69. Group of eggs of Anuridn. x 10.

Fig. TO. Section through nnsegmented egg (after
Claypole, reduced).

Fig. 71. Egg of Anurida, showing precephalic organ
and developing appendages. The outer egg-
membrane {ch.^) has been dissected away from
the lower half of the egg. x ^i^).

Fig. 72. Embryo ^4/i«m/« ; flexure almost complete. The
precephalic organ is seen to be degenerating
(after Claypole, reduced).

Fig. 78. Embryo Anurida at the time of development
of the tritocerebral appendages (after
Wheeler).

Fig. 74. Paramedian section to show primitive cephalic
ganglia about the stage in fig. 72 (after
Folsom, reduced.)

Fig. 75. ^'entral aspect of cephalic region, about the
stage in fig. 72 (after Folsom, reduced).

Fig. 76. Blastoderm with precephalic organ (after
Claypole, reduced).

99

Fig. 77. Transverse section through, second abdominal
segment on one side of the body of the male,
at the stage prior to that seen in fig. 72 fafter
Claypole, reduced).

Fig. 78. Transverse section showing early mesoderm
formation (after Claypole, reduced).

Fig. 79. Diagrammatic figure showing Nematodes in
the testis. x circa 100.

Fig. 80. a, Embryo of Nematode; h, the same in
transverse section. x 200.

C. Tinlins & Co., Ltd., Printers, 53, Victoria Street, Liverpool.

L.M. B. C. Memoir XIII

Plate I.

Imq.jl.

a.bd.. 6

A.D. Imms. del

ANURIDA,

Fig. 11.

: 5.f:rslfine,Ljth CdmV

L.M. B. C. Memoir XIII.

Plate 11.

^■^V^y^;•Mv,V;;'•V/^;-^^;iUy

A- D. Imms. del .

ANURIDA.

MfFajla.ac it Erakmf. Lttb-Edin

L.M. B. C. Memoir XIII.

Plate III.

■.i>.oes.a.

ri^.34.

M'Farlajie S: Ersl<inc.I.it.h Eain''

ANURIDA.

L.M. B. C. Memoir XIII

Plate IV:

per. oes.a. \^ per.afs.a.\ f.^ I

t' \p^' ^^' M^^

l.s.

m

V •-- con

• -ttijcilI. n.
_ True ! 71 .

•- -..-';^^

Fig. 47.

c.^J.

Fig. 42

[m

7vf. oe's a

z^>

■-- th.a.
mnbr.

Fig. 40.

fi^"^- ll

Fig. 43

Fi^. 41. 3#

Fig. 46.

■*,*«?:_

aZ.i. at.

**#*"-'

AD Ini m s . del .

,'"'^.^-««='=' "'t^

ANURIDA

L.M. B. C. Memoir Xm.

Plate V.

A.D Imms, del.

M<FarU.iic 1 Erskine. Liii EduiT

ANURIDA.

L M. B. C. Memoir XIII.

Plate VI.

Fie. 63.

A. D Irams. del.

alait.

ANURIDA.

M-Fa,rla,ne i. Er.skine, I.ifch.Cdii

L.M. B. C. Memoir XIll

Fig.78^ /^^°- ^.

ch.

Plate VH.

vj '-4^

4 '.^^ *i.a^. ''1i>,%^\f^

M.;.*/ ■ _^- -^^ £^^

x/.;.

Pig. 70.

1 .l^t'^fy

Fi^.72:^^ ,■■.:;- t

Fig. 77.

pc.o.

^^^i^!^

/.- ' Fi^.73.

■A.j^

,<!A.,».'

/f^

' / ^^^Mit<^M^^S\> (

^'■y-

'.bSKH^x';^is:,-:o^

tr 'ceb

frr'ceh

Fig 75

, ' , Una.

th . a.pp. •-'

AD-Imms id.. Figs. 69. 71, 79, 60 ,

lewi . t.

Fig. 80.

M'fivUnoiiErsliuic Litli Y.i\S

ANURIDA.

3 2044 093 366 474