i P- i ru i ru LO ~D O a a m a A TREATISE ON ZOOLOGY A TREATISE ON ZOOLOGY Demy 8vo, Cloth, price 15s. net each; or in Paper Covers, price 12s. $d. VOLUMES READY Part I. (First Fascicle) INTRODUCTION AND PROTOZOA. By Sir RAY LANKESTKK, K.C.B., F. R.S. ; Prof. S. J. HICKSOX, M.A., F.R.S. ; F. W. GAMBLE, U.Sc., F.R.S. ; A. WILLEY, M.A., D.Sc., F.R.S. ; J. J. LISTER, F.R.S. ; H. M. WOODCOCK, D.Sc. ; and the late Prof. WELDOX. Part 1. (Second Fascicle) INTRODUCTION AND PROTOZOA. By J. B. FARMER, D.Sc., M.A., F.lt.S. ; J. J. LISTEI:, F.R.S.; E. A. MINCHIX, M.A. ; and S. J. HICKSOX, F.R.S. Part II. THE PORIFERA AND COELENTERA. By Sir RAY LANKESTER, K.C.B., F.R.S. ; E. A. MINCHIX, M.A. ; G. HERBERT FOWLER, B.A., Ph.D. ; and GILBERT C. BOURNE, M.A. Part III. THE ECHINODERMA. By F. A. BATHER, M.A., assisted by J. W. GREGORY, D.Sc., and E. S. GOODRICH, M.A. Part IV. THE PLATYHELMIA, THE MESOZOA, and THE NEMERTINI. By Prof. BENHAM, D.Sc. Part V. MOLLUSCA. By Dr. PAUL PELSEXKER. Part VII. APPEND1CULATA (Third Fascicle: CRUSTACEA). By W. T. CALMAN, D.Sc. Part IX. VERTEBRATA CRANIATA (First Fas- cicle: FISHES). By E. S. GOODRICH, M.A., F.It.s. CANADA INDIA . AGENTS THE MACMILLAN COMPANY ti4 &, eo FIFTH AVENUE, NEW YOBK THE OXFORD UNIVERSITY PRESS 205 FLINDERS LANE, MELBOURNE THE MACMILLAN COMPANY OF CANADA, LTD. 27 RICHMOND STREET WEST, TORONTO MACMILLAN & COMPANY, LTD. MACMILLAN BUILDING, BOMBAY 309 Bow BAZAAR STREET, CALCUTTA TREATISE ON ZOOLOGY EDITED BY SIR RAY LANKESTER K.C.B., M.A., LL.D., F.R.S. HONORARY FELLOW OF EXETER COLLEGE, OXFORD ; CORRESPONDENT OK THE INSTITUTE 01 FIIAXCE; LATE DIRECTOR OF THE NATURAL HISTORY DEPARTMENTS OF THE BRITISH MUSEUM PART VII APPENDICULATA THIRD FASCICLE CRUSTACEA BY W. T. CALM AN, D.Sc. (ST. AND.) ASSISTANT IX THE XOOLOCK AI. M:l'Al!TMEXT OF THE BRITISH MUSEUM LONDON ADAM AND CHARLES BLACK 1909 7-H 9/7 PKEFACE THE present volume is an instalment of that part of the treatise dealing with the great phylum Appendiculata — a phylum which includes the Arthropoda, Chsetopoda, and Eotifera. Dr. Caiman having finished the present description of the Crustacea, it has been considered advisable to publish it at once, without waiting either for the general intro- duction on the classification and characters of the phylum Appendiculata or for the completion of the fascicles devoted to the Peripatoids, Chilopods, Arachnids, Chilognaths, Hexapod Insects, Chsetopods, and Rotifers. E. RAY LANKESTER. January 1909. CONTENTS CHAPTER I PAGE THE CLASS CRUSTACEA 1 CHAPTER II THE BRANCHIOPODA ...... 29 CHAPTER III THE OSTRACODA . . . . . .56 CHAPTER IV THE COPEPODA . . . . . .71 CHAPTER V THE CIRRIPEDIA . . . . . .106 CHAPTER VI THE MALACOSTRACA ...... 143 CHAPTER VII THE LEPTOSTRACA . . . . . .151 CHAPTER VIII THE SYNC A RID A . . . . . .162 vii THE CRUSTACEA CHAPTER IX PAGE THE MYSIDACEA . 171 CHAPTER X THE CUMACEA . . . . .183 CHAPTER XI THE TANAIDACEA . . . . . .190 CHAPTER XII THE ISOPODA . . . .196 CHAPTER XIII THE AMPHIPODA CHAPTER XIV THE EUPHAUSIACEA . . . . . .244 CHAPTER XV THE DECAPODA CHAPTER XVI THE STOMATOPODA . . 319 INDEX 333 CHAPTER I CLASS " CRUSTACEA. SUB-CLASS I. BRANCHIOPODA. Order 1. Anostraca. „ 2. Notostraca. „ 3. Conchostraca. ,, 4. Cladocera. SUB-CLASS II. OSTRACODA. Order 1. Myodocopa. „ 2. Cladocopa. „ 3. Podocopa. „ 4. Platycopa. SUB-CLASS III. COPEPODA. Order 1. Eucopepoda. „ 2. Branchiura. SUB-CLASS IV. CIRRIPEDIA. Order 1. Thoracica. „ 2. Acrothoracica. ,, 3. Ascothoracica. ,, 4. Apoda. „ 5. Rhizocephala. SUB-CLASS V. MALACOSTRACA. SERIES A. LEPTOSTRACA. DIVISION PHYLLOCARIDA. Order Nebaliacea. SERIES B. EUMALACOSTRACA. DIVISION 1. SYNCARIDA. Order Anaspidacea. 1 THE CRUSTACEA DIVISION 2. PERACARIDA. Order 1. Mysidacea. „ 2. Cumacea. „ 3. Tanaidacea. ,, 4. Isopoda. „ 5. Amphipoda. DIVISION 3. EUCARIDA. Order 1. Euphausiacea. „ 2. Decapoda. DIVISION 4. HOPLOCARIDA. Order Stomatopoda. Introductory. — The Crustacea form one of the Classes com- posing the Sub -Phylum Arthropoda, and include, besides the forms popularly recognised as Crabs, Lobsters, Crayfish, Prawns, Shrimps, Sandhoppers, Woodlice, Barnacles, and Water-fleas, u multitude of related organisms which are nameless in common speech. The Class presents so wide a range of structural diversity that it is all but impossible to give, in a few words, a definition which shall apply to all its members. Of the great majority it may be said that they are Arthropoda of aquatic habits, breathing by gills or by the general surface of the body, having two pairs of antenniform preoral appendages, and having at least three pairs of postoral appendages acting as jaws, the three corresponding somites being coalesced with the head. But while these characters are found in the more primitive members, actual or hypothetical, of all the sub-classes and orders composing the Class, the more modified types furnish exceptions to every statement of the definition. Thus, the land-crabs and woodlice are not only completely terrestrial in their habits, but are provided with special organs for aerial respiration ; the preoral appendages may be modified for locomotor or prehensile functions, or may be quite wanting ; and some or all of the mouth- parts may be suppressed. The most extreme modifications are found in parasitic forms, and some of these, such as the Rhizocephala, have lost, in the adult state, almost every trace, not only of Crustacean, but even of Arthropodous structure. In these cases, however, the larval stages afford indications of affinity, while less specialised forms provide connecting links with the typical Crustacea and leave no doubt as to the natural character of the Class as a whole. Historical. — In the Systema Naturae (12th edition, 1767), Linnaeus placed most of the Crustacea then known in his Class Insecta (equivalent to the sub-phylum Arthropoda as now under- THE CRUSTACEA stood), where they formed, with the Arachnida, one of the divisions of the Order Aptera. Three genera were recognised, Cancer, Monoculus, and Oniscus. The Barnacles, however, forming the genus Lepas, were placed among the Vermes testacea, between Chiton and Pholas, and the genus Lernaea, comprising certain parasitic forms, was classed under Vermes mollusca. The adjective crustata or Crustacea had been applied, more or less loosely, by the oldei naturalists to animals possessing a hard exoskeleton, and Brisson, in 1756, had used it as the designation of a group. Pennant, in 1777, appears to be the first post-Linnaean author to suggest the separation of a distinct class under the name Crustacea, and this step was definitely taken in Cuvier's Lemons d 'Anatomic comparee (1800), Avhere, however, the Isopoda still remained among the Insects. Lamarck in 1801 included the Isopoda, and Latreille made mairy changes in the classification, the most important being the division of the class into Malacostraca and Entomostraca (Genera Crust, et Insect., 1806). This arrangement, with the further division of the Malacostraca into Edriophthalma and Podopkthalma, proposed by Leach in 1815, has been widely adopted down to the present day. The researches of J. C. Savigny on the structure of the mouth- parts in Insects and Crustacea (1816) laid the foundations of modern conceptions of Arthropod morphology. Among his immediate successors in this line of research, perhaps the most prominent names are those of V. Audouin and H. Milne-Edwards. The Histoire Naturelle des Crustaces of the last-named author (1834-1840) marks an epoch in the history of Carcinology and is still indispens- able to the student. It is curious that, even at this date, Milne- Edwards did not include the Cirripedia in his survey of the Crustacea, although J. Vaughan Thompson had already in 1830 described their larval stages and recognised their Crustacean affini- ties. Apart from this omission, the limits of the Class adopted by Milne-Edwards differ from those now generally accepted in includ- ing the Pycnogonida, the Xiphosura, and the Trilobita. There can be little doubt that the affinities of these groups are with the Arachnida, though it is possible that the very primitive Trilobites were also phyletically related to the Crustacea. It is impossible to summarise here the numerous changes in classification introduced since the date of Milne-Edwards' work, but it may be mentioned that the establishment of a phylogenetic basis for classification is largely due to the numerous and important works of C. Claus. Further notes on the historical development of the subject will be given in the chapters dealing with the separate sub-classes and orders. THE CRUSTACEA GENERAL MORPHOLOGY. Exoskekton. — In the Crustacea, as in other Arthropoda, the exo- skeleton of the body consists typically of a series of segments or somites which may be movably articulated or more or less coalesced together. In its simplest form the exoskeletal somite is a ring of chitin, connected with the adjacent rings by areas of thinner integument permitting movement in various directions, and having a pair of appendages attached to its ventral surface. This ring may be further subdivided into a dorsal tergite and a ventral sternite, and the tergite may overhang the attachment of the appendage on each side as a free plate called the pleuron1 (Fig. 1). At the posterior end of the body is a terminal segment known ~~P- as the telson, upon which the anus opens. This segment never bears typical limbs and its nature has been variously interpreted. It has FlG- l- been regarded as a true somite or A transverse section of one of the rpsiiltinir frnm thp abdominal somites of the Crayfish (Antaeus). as resulting I a, appendage ; p, pleuron ; s, sternite ; t, Of a number of somites, while Some have described it as a " median appendage " or as a fused pair of appendages. Its true nature, however, is clearly shown by embryology. In the larval develop- ment of the more primitive Crustacea the body increases in length by the successive addition of new somites between the last-formed somite and the terminal region which bears the anus. The " grow- ing-point " is, in fact, situated in front of this region, and when the full number of somites has been reached, the unsegmented part remaining forms the telson of the adult. In no Crustacean, however, do all the somites of the body remain distinct. Coalescence of somites or suppression of segmenta- tion (lipomerism) involves more or less extensive regions where the component somites are only indicated by the persistence of the corresponding appendages. This is constantly the case in the anterior part of the body, where a varying number of somites are united to form the head. This fusion of cephalic somites is associated with what Lankester has termed the " adaptational shift- ing of the oral aperture " backwards from its primitive position at the anterior end of the body. As a result of this shifting, at least two somites, corresponding to the antennules and antennae, come to lie 1 Sometimes called the epimeron, but this term has been used in different senses and it seems better to abandon it altogether. THE CRUSTACEA 5 in front of the mouth in all Crustacea (Fig. 2). Perhaps an anterior somite bearing the paired eyes should also be recognised, but some doubt remains as to the appendicular nature of the eye-stalks, and it is possible that the eyes should be referred to a primitively preoral region corresponding to the prostomium of annelid worms. In any case, a prostomial element may be assumed to share in the formation of the head, and to it may be assigned the more or less problematical "frontal sense- organs " found in various larvae and some adult Crustacea. It has been suggested by Bernard that the back- ward shifting of the mouth was accomplished by a bending round of the anterior somites and that the prostomium is represented by the loV>riim nv nt-moy lir> incr in frrmt nf labrum 01 upper lip just in iront oi the mOUth. There 1S, however, little definite evidence in. favour of this view. In all addition to least three postoral somites, corre- sponding to the mandibles, maxillulae, and maxillae, are included in the head-region. Even where a larger number of somites are involved there is generally a more or less marked change in the character of the appendages after the third postoral pair, and, since the integumental fold which forms the cara- pace seems to originate at this point, it is usual to take the third postoral (maxillary) somite as the limit of the cephalon throughout the Class. It is quite probable, however, that in the primitive ancestral type of the Crustacea, the head-region included a smaller number of somites. The three anterior pairs of limbs (antennules, antennae, and mandibles), which alone are present in the " nauplius " larva, show peculiarities of structure and development which seem to place them in a different category from the succeeding limbs, and there is some ground for regarding the three corresponding somites as belonging to a " primary head-region." For descriptive purposes, however, it is convenient to treat the two following somites also as cephalic. Mention must be made here of a remarkable feature found only in the aberrant group of the Stomatopoda, among the Malacostraca. This is the reappearance of segmentation in the anterior part of the head, where two movably articulated rings carry the eye-stalks and Fio. 2. Diagram of the Crustacean head- region. (Modified from Goodrich.) The nervous system is shown in black, a', antennule ; a", antenna ; D, deutero- cerebrum, the division of the brain corresponding to the antennules ; K, paired compound eye : e, unpaired " nauplius " eye ; /.;>, frontal papilla or sense-organ ; in, mouth ; ind, man- dible ; maf, maxillula ; mx", maxilla ; P, .protocerebrum, the division of the brain corresponding to the paired eyes ; T, tritocerebrum, the division of the brain corresponding to the antennae. Behind the preoral Somites, at the mouth is seen the transverse com- existing Crustacea, in missure of the tritocerebrum. antennal ganglia or THE CRUSTACEA antennules. Whether or not these rings correspond to the primitive somites, their distinctness in the highly specialised Stomatopoda is clearly a secondary condition since it is not found in their larvae nor in any of the more primitive Malacostraca. For the same reasons no great morphological significance can be attributed to the less distinctly marked skeletal areas described as representing the ophthalmic, antennular, and antennal sternites in the higher Decapoda. In nearly all cases the post-cephalic somites can be further grouped into regions or tagmata distinguished by the shape of the somites or the character of their appendages. In descriptive carcinology two such regions are commonly distinguished as thorax and abdomen, but it must be pointed out that there is no morpho- logical equivalence between the tagmata so named in different groups. Throughout the Malacostraca, the thorax of eight and the abdomen of six somites are sharply distinguished by the appendages. In the other sub-classes the same names are sometimes applied to the limb-bearing and limbless regions of the trunk, while in the Branchiopoda they may denote respectively the regions in front of and behind the genital apertures. The total number of post-cephalic somites varies within very wide limits. In the Ostracoda, where the body is not distinctly segmented, the number of trunk-limbs does not exceed two pairs. In some Branchiopoda the number of trunk-somites exceeds forty. A structure which, from its occurrence in the most diverse groups of Crustacea, is probably a primitive attribute of the Class, is the dorsal shield or carapace, originating as a fold of the integument from the posterior margin of the cephalic region. In its simplest form, as seen in Apus among the Branchiopoda, the carapace loosely envelops more or less of the trunk. In many Branchiopoda and in the Ostracoda it forms a bivalve shell completely enclosing the body and limbs. In the Cirripedia it forms a fleshy "mantle" usually strengthened by shelly plates. In many cases among the Malacostraca the carapace coalesces with the tergites of some or all of the thoracic somites, though it may project freely at the sides, overhanging, as in the Decapoda, the branchial chambers. It is possible that, in those cases where some of the post-cephalic somites seem to be simply fused with the head-region, a reduced shell-fold is also involved in the coalescence. This is most probably the case in the Isopoda and Amphipoda, where the fusion of the first thoracic somite with the head is clearly the last vestige of a shell-fold, traceable, with progressively diminishing extent, from the primitive Mysidacea through the Cumacea and Tanaidacea. In the Copepoda, on the other hand, in which one or two trunk-somites coalesce with the head, there is less evidence that the dorsal THE CRUSTACEA "carapace" so formed really represents the shell-fold, and its lateral extensions, which cover the bases of the legs, may be the pleura of the coalesced somites. Apart from the Copepoda, the only Crustacea in Avhich there is no trace of a shell-fold are the Anostracous Branchiopoda, and perhaps also the anomalous Mala- costracan Anaspides. Before proceeding to discuss the true limbs, mention must be made of certain appendages which have sometimes been regarded as homologous with the limbs, but which probably do not belong to that category. In most Malacostraca and in certain other forms, notably among the Copepoda, the posterior margin of the oral aperture is bounded by a fold forming a lower lip (metastoma or hypostoma), usually cleft into two lobes, known as the paragnatha, which may bear movable terminal lappets. Since there is never any corresponding pair of ganglia on the ventral nerve-chain, or other evidence of the existence of a corresponding somite, there is little ground for the view that the paragnatha are a vestigial pair of limbs. Glaus has shown that in Apus the so-called paragnatha are probably the proximal lobes of the maxillulae, and he has suggested a similar connection in the Malacostraca, where, however, an independent origin of the lower lip seems more probable. The upper lip or labrum, already mentioned, is plainly an unpaired outgrowth. The terminal segment or telson often bears a pair of processes or rami forming the "caudal furca." These may be multiarticulate filaments as in Apus and a few Cirripedes ; in the Anostracous Branchiopoda, Copepoda, and Leptostraca they are unsegmented rods articulated to the body ; in other cases they may be simple processes of the telson. There seems to be very little reason for supposing that the furcal rami represent limbs, more especially since the telson, as stated above, has not the value of a true somite. Limbs : General Morphology. — The limbs of Crustacea differ very widely in form and structure, but it is generally possible to trace in them the modifications of a fundamental type consisting of a peduncle, the protopodite (or sympodite}, bearing two rami, the exopodite and endopodite. This simple biramous form is seen, for instance, in the swimming feet of the Copepoda (Fig. 3, B), the cirri of the Cirripedia, and the abdominal appendages of the Malacostraca (Fig. 3, A), and in the second and third pairs of limbs in the earliest and most primitive type of larva, the nauplius (Fig. 3, C). As a rule, the protopodite is composed of two segments known as the coxopodite (or coxa) and basipodite l (or basis), but one of these may be reduced or suppressed ; while, on the other hand, Hansen has shown that in some cases a pre-coxal segment can be 1 By some writers the term basipodite is applied to the protopodite as a whole. THE CRUSTACEA distinguished. The two rami may become specialised for different functions, as in the case of the thoracic limbs of Malacostraca (Fig. 3, D, E), where the endopodite forms a walking-leg, while the exopodite becomes a swimming-branch or may disappear altogether. ---•p. FIG. 3. Various types of Crustacean limbs. A, abdominal limb (pleopod) of Crayfish. (After Hux- ley.) B, swimming-foot of Calanus (Copepoda). (After Sars.) C, limb of second pair (antenna) of nauplius larva of Apus (Branchiopoda). (After Claus.) D, first thoracic limb of Anaspides (Syncarida). E, second thoracic limb of Anaspides. 6s, basipodite ; ex, coxo- podite ; en, endopodite ; ep, epipodite ; ex, exopodite ; gn, gnathobase ; prot, protopodite. The coxopodite often bears on the outer side an appendage (rarely more than one), known as the epipodite, which may function as a gill. In the appendages near the mouth one or both of the segments of the protopodite may develop, even in the nauplius, inwardly-turned masticatory processes or gnathobases. The occurrence of epipodites and gnathobases suggests that the primitive Crustacean limb was THE CRUSTACEA 9 Fir,. 4. Leaf-like trunk-limb of Lepidurus (Bran- chiopoda). (After Sars.) en, endites ; ex, exites. more complex than the simple biramous type. Lankester has called attention to the lobed leaf-like appendages of the Branchiopoda (Fig. 4), as probably approximating to the ancestral form. As will be shown below, it is not altogether easy to recognise the homologies of the various lobes even within the limits of the group Branchio- poda, and their exact relation to the parts of the biramous limb is still open to doubt, but it is probable that the Branchiopod limb preserves characters belonging to an early phyletic stage before the biramous type had become fixed. It does not seem profitable to go beyond this and to attempt, as some have done, to compare the limbs of the Branchiopoda in- detail with the Polychaete parapodium. The general character of the modifications which the original type of limb undergoes is often, though by no means always, plainly correlated with the functions which the limbs discharge. In swimming-limbs the rami are often flattened and oar-like, and fringed with plumose hairs or flattened spines. For walking or creeping one of the rami, generally the inner, is stout and cylindrical, tipped with a claw, and having the segments connected by definite hinge-joints allowing movement only in one plane. The oral appendages have the gnathobasic lobes developed at the expense of the rest of the limb, the rami persisting, if at all, only as sensory " palps." A multiarticulate flagelliform modification of the rami is generally associated with a sensory (tactile or olfactory) function, as in the antennules and antennae. A pincer-like (chelate or sub- chelate) form is frequently assumed by limbs used for prehension, the terminal segment being flexed against the penultimate, or opposed to a thumb-like process of the latter. Special Morphology of Limbs — Ocular Peduncles. — In many Crustacea, notably in the Anostracous Branchiopoda and in the majority of the Malacostraca, the eyes are set upon peduncles which are movably articulated with the head, and which may be divided into two or three segments. The view that these peduncles are homologous with the limbs was first suggested by H. Milne- Edwards, and has been widely but not iiniversally accepted. In spite of much discussion, however, it cannot be said that the point has been finally decided. The fact that the eye-stalks are most fully developed and most distinctly articulated not in the more primitive forms, but among the highly specialised Decapoda, is THE CRUSTACEA against the appendicular theory, and the evidence of embryology does not support it. In the development of the Branchiopod Branchipm, Glaus has shown that the eyes are, at their first appear- ance, sessilej and only become pedunculated at a later stage, the lateral lobes of the head on which they are set becoming produced and separated from the rest of the head by a movable articulation (Fig. 5). The most important evidence in favour of the appendicular nature of the eye-stalks is that afforded by the phenomena of f. Development of ocular peduncles in 'Branchipus. (After Glaus.) A, head of young larva ; B, head Of older la,rva. I (in A), lateral lobe of the head bearing E, the compound eye. In B this lobe has elongated to form the ocular peduncle, not yet niovably articulated, although one of the muscles for moving it is developed (in), e, unpaired or nauplius eye ; /, frontal seusi-- organs. regeneration. If the eye-stalk be removed from a living prawn or lobster, it is found that, under certain conditions, a many-jointed appendage, like the flagellum of an aiitennule, may grow in its place. The bearing of such cases of " heteromorphic regeneration " on questions of homology is, however, by no means clear, and their discussion would involve a reconsideration of some of the most fundamental conceptions of current morphology. For the present it must suffice to point out that the appendicular nature of the ocular peduncles cannot be assumed as definitely proved. The antennules (or first antennae) 'are almost universally regarded THE CRUSTACEA 1 1 Flo 6 Early nauplius larva of Copepod (Ci/dops s]>.) from below. «', antennule; a", antenna; gn, piatho- basic processof antenna ; tor, labrum ; wd, mandibl.-. as true appendages, although they dift'er from all the other append- ages in the fact that they are always innervated from the brain (or supra -oesophageal ganglia), and that they are uniramous in the nauplius larva (Fig. 6, a) and in the adults of all the sub-classes except the Malaco- straca. As regards the in- nervation, an apparent excep- tion is found in the case of Apus, where the antennular nerves arise, behind the brain, from the oesophageal connectives. This is un- doubtedly a secondary position, however, and the nerve-fibres have been traced forward to centres in the brain. In the Malacostraca the antennules are often Viiramrmt: /Fio- 7\ Vint tVi^rn 5 VriS- '/I DUt tnere is considerable doubt as to whether the two flagella correspond to the endopodite and exopodite of the other limbs. In most cases the antennules are sensory in function, but they may also be natatory or prehensile, and in the Cirripedia they form organs of attachment. The antennae (or second antennae) are of • special interest on account of the clear evidence that, although preoral in position in all adult Crustacea, they were originally postoral appendages. In the nauplius larva (Fig. 6, a") their position is beside rather than in front of the mouth, and they may bear hook-like masticatory processes (gn) which assist the similar processes of the mandibles in seizing the food. In the Branchiopoda and less distinctly in some other groups, the nerves to the antennae arise not from the brain, but from the oesophageal connectives, and the trans- verse commissure oj: th'e corresponding ganglia can be traced behind the oesophagus, even in those forms in which the ganglia have moved forward into the brain (Fig. 2, p. 5). The functions- of the antennae are very varied. As already stated, they act as jaws in some nauplius larvae. In many cases they are important organs of locomotion, and they may serve as sexual " claspers^" or as organs of attach- ment in parasites. In the Malacostraca they are mainly sensory, the endopodite being a long flagellum, while the . exopodite may form a flattened "scale" probably used as a balancer in Swimming, or may disappear altogether. Fio. 7. Antennulo of Crayfish. (After Huxl.'y.) 12 THE CRUSTACEA The mandibles, like the antennae, are, in the nauplius, bira- mous swimming-limbs with a masticatory gnathobase arising from the basal segment of the protopodite. This form and function are retained with little alteration in some adult Copepoda (Fig. 8, A). In most cases, however, the exopodite is lost and the endopodite (with the distal part of the protopodite) forms the "palp" (Fig. 8, B) or may disappear altogether (Fig. 8, C), while the " body " of the mandible is formed by the coxopodite (or perhaps by the precoxa), and has a masticatory edge armed with tubercles, teeth, or spines. In parasitic forms with suctorial mouth -parts the mandibles may take the shape of piercing lancets enclosed in a tubular beak formed by apposition of the labrum and metastoma. A. Fio. 8. A, mandible of Copepod (Calanus) (after Sars) ; ex, coxopodite (or precoxa, according to Ilansen), forming the " body " of the mandible ; bs, basipodite ; en, endopodite ; ex, exopoditr. B. mandible of Crayfish (after Huxley) ; letters as above. In both cases the basipodite and the segments distal to it form the " palp." G, mandible of LepMnrus (after Sars). In Ostracoda the mandibular palp aids in locomotion, and in a few cases the masticatory part is greatly reduced. The maxillulae and maxillae (or, as they are often termed, the first and second maxillae) are nearly always foliaceous appendages having gnathobasic lobes or endites borne by the segments of the protopodite (Fig. 9). The endopodite is reduced to a "palp" or is absent. On the outer side, lobes representing the exopodite and epipodite may be present. These appendages undergo great modifications in the different groups and exact comparative in- vestigations on their morphology are still wanting. The post-cephalic or trunk appendages vary greatly in number. In some Branchiopoda there are more than 60 pairs, while in some Ostracoda it is uncertain whether there are any. They present great diversity of form and function in the various groups THE CRUSTACEA and often also in the same animal. They may be nearly all alike as in the Branchiopoda, where, at most, one or two of the anterior pairs may be specialised as sensory or as grasping organs. Com- monly, as in the Copepoda, one or two of the anterior pairs are modified to assist the oral appendages and are known as maxillipeds. It is very characteristic of the Malacostraca that the series of trunk- limbs is differentiated into two well-defined "tagmata" or groups of similarly modified appendages, corresponding to the ep. en. en. C. FIG. 9. A, maxillula of Copepod (Calamts). (After Sars.) B, maxillula of Crayfish ; C, maxilla of Crayfish. (After Huxley.) en, endopodite ; cp.'epipodite ; ex, exopodite ; gn, gnathobasic lobes. (The plate lettered cp in C is possibly the exopodite rather than the epipodite ; see p. 268.) thoracic and abdominal regions respectively. The thoracic limbs have the endopodites forming, as a rule, more or less efficient walking-legs, and the exopodites, when present, form swimming- branches, while the abdominal limbs are usually biramous, with the rami similar and, in the more primitive types, natatory in function. The general similarity between the appendages of each tagma is usually qualified by minor modifications, sometimes leading to the formation of subsidiary groupings. Thus, for example, in the Decapoda a group of three pairs of maxillipeds is differentiated from the thoracic tagma. 14 THE CRUSTACEA Branchiae. — In many of the smaller Crustacea there are no special branchiae, and respiration is carried on by the general sur- face of the body. When present, branchiae are usually formed by differentiation of parts of the appendages, often the epipodites, but the shell-fold has probably in many cases a respiratory function, and processes from its inner surface (Cirripedia) or from the surface of the body (some Ostracoda) may develop as branchiae. In the more primitive of the Malacostraca, the gills are formed by the epipodites of the thoracic limbs (podobranchiae), and this was probably also the original nature of those branchiae which, in the Decapoda, are attached to the articular membrane between the limb and the body (arthrobranchiae), or to the body -wall itself (ftleurobranchiae). In the Isopoda the respiratory function is assumed by the lamellar rami of the abdominal appendages. Many terrestrial Crustacea have no special adaptations for aerial respiration. In land-crabs of different families, however, the lining membrane of the branchial chamber is covered with vascular papillae and acts as a lung. Still more remarkable are the breathing organs of many of the terrestrial Isopoda or Woodlice. These are ramified tubular invaginations of the integument in the abdominal appendages, and are precisely analogous to the tracheae of other air-breathing Arthropoda. Alimentary System. — In the great majority of Crustacea the alimentary canal is nearly straight, except at its anterior end, where it curves downwards to the ventrally placed mouth. The only cases hitherto described in which it is actually coiled upon itself are in certain Cladocera and in a single genus of Cumacea. As in other Arthropoda, it consists of stomodaeum, mesenteron, and prododaeum, the first and last with a lining of chitin continuous at mouth and anus with the exoskeleton. The relative proportions of these three divisions vary greatly, and the extreme abbreviation of the mesenteron found in the common Crayfish is by no means typical of the Class. Even in the closely related Lobster this section of the gut may be several inches long. The whole length of the alimentary canal is provided, as a rule, with circular and longitudinal muscle -fibres running in its walls, and there are often also muscle-bands extending to adjacent portions of the body-wall. In the anterior part of the stomodaeum these muscles are more strongly developed to perform the move- ments of deglutition. In a few Branchiopoda and Ostracoda the chitinous lining of this region develops spines and hairs which help to triturate and strain the food, and in some Ostracoda the armature assumes a more complex form as a series of toothed plates moved by special muscles. It is among the Malacostraca, however, and especially in the Decapoda, that this apparatus, the so-called " gastric mill," reaches its greatest complexity. It con- THE CRUSTACEA sists of a framework of movably articulated ossicles developed as thickened and calcified portions of the lining cuticle of the " stomach " or dilated part of the stomodaeum. These ossicles bear teeth and are moved by a complex system of intrinsic and extrinsic muscles. In the posterior division of the stomach a series of pads and ridges beset with stiff hairs form a straining apparatus. The mesenteron, in most Crustacea, has its surface increased by pouch-like or tubular outgrowths, which not only serve as glands for the secretion of the digestive juices, but may also become filled by the more fluid portion of the partially digested food and facilitate its absorption. These outgrowths vary much in their arrangement in the different groups. Most commonly there is a single pair, which may be more or less ramified, and may form a massive digestive gland (" hcpato-pancreas" or "liver"). In a few parasites (Rhizocephala and the Monstrillidae among Copepoda) the alimentary canal is absent or vestigial throughout life. Circulatory System. — The heart of the Crustacea is of the usual Arthropodous type, lying in a more or less well-defined pericardial sinus, with which it communicates by valvular openings or ostia. In some of the Branchiopoda, such as Branchipus, the heart is of the primitive tubular form, extending the whole length of the body, and having a pair of ostia in each somite. Even within the group of Branchiopoda, however, a progressive abbreviation of the heart, with a diminution in the number of ostia, can be traced, leading to the condition found in the Cladocera, Avhere the heart is a sub-globular sac and the ostia are reduced to one pair. Among the Malacostraca, an elongated heart with numerous ostia is found only in the Leptostraca and Stomatopoda. In other cases the heart is generally abbreviated, and even where, as in the Amphi- poda, it is long and tubular, the ostia are restricted in number. From the heart, the blood passes into one or more arterial trunks, which may ramify into a more or less extensive system of arterial vessels, or may open at once into the general lacunar system of the body (haemocoel). Sometimes, as in the Branchiopoda, even the arterial trunks are absent, and the blood is discharged from the anterior end of the heart directly into the lacunae of the haemocoel. In many Crustacea, especially those of small size (many Copepoda and Ostracoda, Cirripedia), there is no heart and no definite system of vessels, and the blood is simply driven hither and thither by the movements of the body and of the alimentary canal. Certain genera of parasitic Copepoda (LernanthrojMS, etc.) are unique among Arthropoda in possessing a closed vascular system, containing a coloured fluid, and shut off from the general cavity of the body. The morphological relations of this system are still obscure. i6 THE CRUSTACEA Excretory System. — The most important organs of renal excre- tion in the Crustacea are two pairs of glands, lying at the base of the antennae and of the maxillae respectively, which are probably the survivors of a series of segmentally arranged coelomoducts present in the primitive Arthropoda. The two pairs are never fully developed at the same time in one individual, although one may replace the other in the course of development. Thus, in the Branchiopoda, the antennal gland develops early and is functional during a great part of the larval life, but it ultimately atrophies and the maxillary gland takes its place as the excretory organ of the adult. In the Decapoda, where the antennal gland alone is well developed in the adult, the maxillary gland sometimes pre- cedes it in the larva. The structure of both glands is essentially the same (Fig. 10). There is a more or less convoluted glandular tube (t), of mesoblastic origin, connected internally , — Q. with a closed "end-sac" (e.s), representing a vestigial portion of the coelom, and generally a thin-walled duct which opens to the exterior. In the Branchiopoda the maxillary gland is lodged in the thickness of the shell -fold (when this is present); and from this circumstance it often receives the somewhat mis- leading name of "shell- gland." In the Decapoda, the antennal gland is largely developed and often very complex, and is known as the " green gland." Other excretory organs have been described in various Crustacea, but although their excretory functions have been demonstrated by physiological methods, their morphological rela- tions are in most cases quite obscure. In some cases they consist of masses of mesodermal cells, within which the excretory products are stored up instead of being expelled from the body. In other cases an excretory function is attributed to certain cells of the mesenteron or to some of its diverticula. Nervous System. — The central nervous system is constructed on the same general plan as in the other Arthropoda, consisting of a supra-oesophageal ganglionic mass or "brain," united by circum- oesophageal connectives with a double ventral chain of segmentally FIG. 10. Antennal gland of a larva of Estheria (Branchio poda). (After Grobben.) con, connective-tissue fibres e.s, end-sac ; o, external opening ; t, glandular tubule. THE CRUSTACEA arranged ganglia. In the primitive Branchio- poda the ventral chain retains the ladder-like arrangement found in some Annelids and lower worms, the two halves being widely separated and the pairs of ganglia cpnnected together across the middle line by double transverse commissures (Fig. 11). In the other groups the two halves of the chain are approximated and more or less completely coalesced, and, in addition, a concentration of the ganglia in a longitudinal direction takes place, leading ultimately, in many cases, to the formation of an unsegmented ganglionic mass. This is seen, for example, in the Brachyura, among the Decapoda. The brain consists, in the Branchiopoda, mainly of two pairs of ganglionic centres, the protocerebrum and deuterocerebrum (Fig. 2, P, D), giving origin, respectively, to the optic and antennular nerves. The antennal nerves arise, in this group, from ganglionic swellings on the oesophageal connectives. In the higher groups, as already mentioned, the centres for the antennal nerves have moved forwards and are included in the brain, forming the trito- cerebrum (Fig. 2, T), and other additional centres are developed, so that in the highly organised Decapoda the brain assumes an extremely complicated structure. Ei/cs. — Two kinds of eyes are found in Crustacea, the unpaired median or nauplius eye, and the paired compound eyes. The former alone is present in the nauplius larva, and it forms the sole organ of vision in the adult Eucopepoda. It may coexist with the paired eyes as in the Branchiopoda and in some of the more primitive Malacostraca, although, in \J \^ the latter, it is generally vestigial. When FIO. n. fully developed (Fig. 12), it usually presents Nervous system of Brand three divisions, each consisting of a cup- VSft^£SS£Sff*!K shaped mass of dark pigment (P), the cavity ^^n^nS^n^^d of which is filled With Columnar retinal of transverse commissures in the pos- n riii i t ^1 n terior limbless part of the trunk, in cells. 1 he outer ends oi these cells are con- indicates the position of the mouth. ti minus with trip norvp fibres (n\ whilp at The existence of a transverse com- erve- S ()l), Will e at mjssure in front of the mouth, as shown their inner ends they contain rod-like bodies fey Pi;of • Si>rs . '," ,t,his 'i,ra"'i"~ is «• . . T ', . , . . . r tremely donbttul. Possibly the structure (/'). Ill SOme Cases the three divisions OI the observed may be ' a portion of the i_ 1-11 visceral nerve-riii-' encircling the gullet eye are each supplied by a separate nerve in tha region of tfceisbrom. i8 THE CRUSTACEA arising from the brain. In many cases there is no special refracting apparatus, but a refractive body, or lens (I), is sometimes formed on the outer side of the retinal cells, while in the Copepoda, where the median eye may undergo considerable modification, cuticular lenses and other accessory structures may be developed. The compound eyes show considerable agreement in the details of their structure with those of Insects (Fig. 13). They consist of a varying number of orrvmatidia or visual elements, covered by a transparent region , . . , , , . , . a „ of the cuticle, the cwnea, which is usually divided into lenticular facets. Typically eacn ommatidiura has the structure shown in . the accompanying figure. Immediately under the cuticle lie a pair of wrneagen cells (by), by Avhich the cuticular lens is secreted and renewed on ecdysis. Below these are a group, generally two or four, of cells forming a refractive crystalline body (cr), FIG. 12. Horizontal section through the median eye of Cypris. (After ciaus.) Only two of P, pigment; r, rod-like bodies contained in the retinal cells. B. m. FIG. 13. A, horizontal section of the eye and ocular peduncle of Bmnfhiptt*. B, four onimatidia of same further enlarged, b, basement membrane ; c, corneal cuticle, which in this case is not thickened to form lenses ; cr, crystalline body ; cr.c, cells of the crystalline body ; /, nerve- fibrils ; g, optic ganglia in the peduncle ; hy, hypodermis or corneagen cells ; m, muscle of tin; peduncle ; r, retinula cells surrounding the rhabdome, which is here concealed from view by thu black pigment. (After Glaus.) the lower end of which is embraced by the tips of the elongated retinula cells (r). These surround a rod-like body, the rhabdome, of cuticular nature but penetrated by nerve -fibrils, and usually THE CRUSTACEA 19 divisible into rhabdomcrcs corresponding in number to the retinula cells by which it is formed. At their bases the retinula cells pass into nerve-fibres (/) which penetrate the basement membrane (b) and , enter the optic ganglia. Each ommatidium is more or less com- pletely isolated by a sheath of black pigment contained partly in the retinular cells, partly in special cells lying between them. By movements in the protoplasm of these cells the position of the pigment changes according to -the amount of light falling upon the eye. A layer of reflecting pigment, the tapetum, may also be present. It is impossible here to summarise the details of histo- logical structure or of physiological interpretation. It may be stated, however, that the variations in structure found in different Crustacea appear to be determined not so much by the systematic affinities as by the habits of the organisms. In this connection the work of Chun and of Doflein on the structure of the eyes in pelagic and abyssal species may be referred to. In many Crustacea, as already stated, the paired eyes are set on movable peduncles, and it is probable that this condition is the primitive one. In the primitive Branchiopoda the eyes are either pedunculated or, if sessile, movable in such a way as to suggest derivation from the pedunculate condition. Other Sense-Organs. — The other sense-organs of the Crustacea^ are formed by modification of the hairs or setae on the surface of the body and limbs (Fig. 14). As in other Arthropoda, many of these setae are tactile. Each consists of a hollow conical outgrowth of the cuticle, movably articulated at the base and containing a prolongation of some of the cells of the hypodermis. One or more nerve-fibrils may be traced into the interior, and, in some cases, a ganglion-cell in connection therewith lies at the base of the seta. When feathered, or provided with secondary barbs, the setae will respond to movements or vibrations in the surrounding water, and to some setae of this type an auditory function has been attributed (Fig. 14, C). In certain Malacostraca more specialised organs are found which have been regarded as auditory. In most Decapoda there is an invagination of the integument in the basal segment of the antennule having plumose " auditory " setae on its inner surface. In some cases the sac remains open to the exterior, permitting the introduction of sand -grains which act as " otoliths." In other cases the sac is completely closed, and may then contain a single " otolith " secreted by its Avails. Otocysts are found in a few other Malacostraca in various positions ; for example, in the endopodites of the uropods in the Mysidae. Eecent investigations have rendered it doubtful, however, whether aquatic Crustacea can hear at all, in the proper sense of the term, and it has been shown that one function, at least, of the THE CRUSTACEA so-called " otocysts " is connected with the equilibration of the body. They are more properly, therefore, termed " statocysts." Another modification of sensory setae is believed to be associated with the sense of smell. In most Crustacea the anntenules and often also the antennae bear groups of hair-like filaments, sometimes slender, sometimes more or less swollen, in which jthe cuticle is extremely delicate. These are known as olfactory filaments or aesthetascs (Fig. 14, A, B). They are ofteii more strongly developed in the male sex, and are believed to guide the males in pursuit of the females. FIG. 14. A, antenna of Daphnia magna bearing a group of "olfactory filaments." g, nerve-ganglion sending fibrils to the olfactory filaments. B, one of the filaments further enlarged. (After Scourfield.) C, "auditory" seta from peduncle of antennule of Mysis ri'livta. (After l^ars.) Glands. — Apart from the digestive and excretory glands already mentioned, and from other glandular structures to be described in connection with the reproductive system, many Crustacea possess dermal glands scattered over the surface of the body and limbs or grouped at certain points for the discharge of special functions. These consist of single cells or groups, traversed by ramified intra- cellukr canals, and communicating with the exterior by fine ducts. Such glands occurring on the walls of the oesophagus or in the neighbourhood of the mouth have been regarded as salivary in function. Others on the surface of the body and limbs may be used, as in some Amphipoda, for the construction of tubular cases or nests in which the animals live, or, as in some Copepoda, may secrete a gelatinous envelope enabling the animals to survive dessi- THE CRUSTACEA cation. Others, again, are believed to have a poisonous function. The greatly developed cement-glands of the Cirripedia are possibly related to the same group of structures. Phosphorescent Organs. — In this connection may be mentioned the power of phosphorescence possessed by many Crustacea. In some cases this is due to a luminous secretion produced by certain of the dermal glands. In the Euphausiacea and certain Decapoda, however, complex and remarkable light-producing organs are present, which were formerly described as "accessory eyes." Reproductive System. — In the great majority of Crustacea, as in other Arthropoda, the sexes are separate. Apart from certain isolated instances, possibly abnormal and probably non-functional, among Branchiopoda and Amphipoda, the only exceptions are the sessile Cirripedia and some parasitic Isopoda (Cymothoidae and Cryptoniscina), where hermaphroditism is the rule.1 Partheno- genesis is frequent among the Branchiopoda and Ostracoda, often in more or less definite cyclical alternation with sexual reproduc- tion. Where the sexes are distinct a more or less marked sexual dimorphism often exists. The male is frequently provided with clasping-organs for holding the female, and these may be formed by the modification of almost any of the appendages, often the antennules or antennae, or some of the trunk-limbs, or even the mandibular palp (some Ostracoda). In addition, some of the appendages in the neighbourhood of the genital apertures may be modified for the purpose of transferring the sperms to the female. In the higher Decapoda the male is generally larger than the female and has stronger chelae. In the other groups the male is often the smaller, and in many parasitic Copepoda and Isopoda this disparity in size is carried to an extreme degree, and the minute male is attached, like a parasite, to the enormously larger female. The remarkable and complex sexual relationships of the Cirripedia will be discussed in the section dealing with that group. The gonads of the Crustacea, as of other Arthropoda, are hollow organs, the cavity communicating Avith the efferent ducts. They are primitively paired but often coalesce partially or completely with each other on the dorsal side of the alimentary canal. The ducts are present only as a single pair, except in certain parasitic Isopoda (Hemioniscidae, Liriopsidae), where two pairs of oviducts are found. Various accessory structures may be developed in connection with the efferent ducts in both sexes. The oviducts may have diverticula serving as receptacula seminis (in cases where internal impregnation takes place), and may be provided with glands secreting envelopes or shells around the eggs. Similarly the glandular walls of the vasa deferentia may secrete spermatophores 1 According to recent observations (as yet unpublished) by Mr. Alf Wollebaek, certain deep-sea Decapoda also are normally hermaphrodite. 22 THE CRUSTACEA or capsules in which the sperms are transferred to the female. The terminal portion of the male duct is sometimes protrusible, and acts as an intromittent organ, or this function may be discharged by some of the appendages. The position of the genital apertures varies greatly in the different groups. The most anterior position is found in the Cirripedia, where the oviducts open on the first trunk-somite, while in certain Branchiopoda (Polyartemia), on the other hand, the genital apertures lie behind the nineteenth trunk-somite. It is characteristic of the Malacostraca that the position of the genital apertures is different in the two sexes, the female openings being on the sixth and the male openings on the eighth trunk-somite. In all the other groups, with exception of the hermaphrodite Cirripedia, the position is the same in the two sexes. While very few Crustacea are viviparous in the sense that the eggs are retained within the body until hatching takes place (some Branchiopoda), the great majority carry the eggs in some way or other after their extrusion. The various devices by which this is accomplished will be described in dealing with the different groups ; but it may be mentioned here that a few cases are known (Cladocera, terrestrial Isopoda) in which the developing embryos are nourished by a special secretion while in the brood-chamber of the mother. EMBRYOLOGY. The majority of the Crustacea leave the egg in a form more or less different from that of the adult, and pass through a series of free-swimming larval stages, but there are many cases of direct development in which the newly hatched young resemble the parent in general structure. The relative size of the egg is greater in those forms which develop without metamorphosis, except where other means exist for the nourishment of the developing embrj'os. The details of the early stages of development differ consider- ably within the limits of the class. They are chiefly of interest, however, from the point of view of general embryology rather than from that of the special student of the Crustacea, and can only be very briefly referred to here. An admirable summary of the whole subject will be found in Korschelt and Heider's Text-book of the Embryology of Invertebrates. Segmentation is usually of the superficial or centrolecithal type, or some modification thereof. The hypoblast is formed either by a definite invagination or by the immigration of isolated cells which wander through the yolk as " vitellophags " and later become associated to form the mesenteron, or by some combination of the two methods. The blastopore generally occupies a position corre- sponding to the posterior end of the body, j The mesoblast of the THE CRUSTACEA 23 cephalic (naupliar) region probably arises in connection with the lips of the blastopore and consists of mesenchymatous cells. In the region of the trunk, in many cases, paired mesoblastic bands are formed, growing in length by the division of teloblastic cells at the posterior end and becoming segmented into somites. The occurrence of true coelom-sacs is doubtful. The rudiments of the first three pairs of appendages, antennules, antennae, and mandibles, commonly appear simultaneously and, even in forms with embryonic development, they often show differences in their mode of appear- ance from the succeeding limbs. When this stage, corresponding to the nauplius stage of larval development, is passed through within the egg, it is often marked by the formation of a cuticular membrane within which the further development proceeds. The complex and varied larval metamorphoses of the Crustacea have been the subject of much discussion in view of their bearing on the phylogenetic history of the group. In those Crustacea in which the series of larval stages is most complete the starting-point is the form already mentioned under the name of nauplius. The typical nauplius (Fig. 6, p. 11) has an oval unsegmented body and three pairs of limbs, corresponding to the antennules, antennae, and mandibles of the adult. The antennules are uniramous, the others biramous, and all three pairs are used in swimming. The antennae may have a spiniform or hooked masticatory process at the base, and share with the mandibles, which have a similar process, the function of seizing and masticating the food. The mouth is overhung by a large labrum or upper lip, and the integu- ment of the dorsal surface of the body forms a more or less definite dorsal shield. The paired eyes are, as yet, wanting, but the median eye is large and conspicuous. A pair of papillae or filaments, probably sensory, are commonly present at the anterior end. Nauplius larvae, differing only in details from the typical form just described, are found in very diverse groups of the Crustacea, such as the Branchiopoda, Copepoda, Cirripedia, and some Mala- costraca. In many forms which hatch from the egg at a more advanced stage there is, as already mentioned, more or less clear evidence of an embryonic nauplius stage. It seems certain, there- fore, that the possession of a nauplius larva must be regarded as a very primitive character of the Crustacean stock. As development proceeds, the body of the nauplius elongates and indications of segmentation begin to appear in its posterior part. At successive moults the somites increase in number by the addition of new somites, behind those already differentiated, from a formative zone in front of the telsonic region. In the most primitive cases, the appendages posterior to the mandibles make their appearance, like the somites which bear them, in regular 24 THE CRUSTACEA order from before backwards. The limb -buds early become bilobed and grow out into typically biramous appendages which gradually assume the characters found in the adult. With the elongation of the body, the dorsal shield of the nauplius begins to project posteriorly as a shell-fold, which becomes the carapace. The rudiments of the paired eyes appear under the integument of the head, but only become pedunculated at a comparatively late stage. The course of development outlined above, leading from the nauplius to the adult form by the successive addition of somites and appendages in regular order, agrees so well with the process observed in the development of the typical Annelida, that it must be regarded as the most primitive. It is most closely followed by such Branchiopoda as Apus and Sranchipus, and by some Copepoda. In the majority of Crustacea, however, this primitive scheme is more or less modified. The earlier stages may be passed through within the egg or in the maternal brood-chamber, so that the larva only begins to lead an independent existence at a stage more advanced than the nauplius. Further, the gradual appearance of the successive somites and appendages may be accelerated, so that great advances in structure take place at a single moult. For example, in the Cirripedia, the latest nauplius stage gives rise directly to the so-called (7#pm- larva, which possesses all the appendages of the adult. Another common modification of the primitive method of development consists in the accelerated appear- ance of certain somites or appendages, disturbing the regular order of development. This modification is especially found in the Malacostraca, in which, even among those which have most closely' adhered to the primitive order of development, the last pair of abdominal appendages usually make their appearance before those immediately in front of them. The same process, carried further, leads to the very peculiar larva known as the zoea, in the typical form of which, found in the Brachyura, the posterior five or six thoracic somites are greatly retarded in development, and are still repre- sented by a short unsegmented region of the body at a stage when the abdominal somites are fully formed and even carry appendages. A remarkable phenomenon observed in the development of a few Malacostraca is the temporary retrogression and even dis- appearance of certain appendages which redevelop in later stages. The best- known instances of this will be further alluded to in ' describing the development of the Penaeidea among the Decapoda. In addition to the nauplius and zoea there are many other types of Crustacean larvae distinguished by special names, though, as their occurrence is restricted within the limits of the smaller systematic divisions, the}r are of less general interest. We need only mention here the metanauplius, a vaguely defined stage following the nauplius, THE CRUSTACEA 25 possessing rudiments of some of the post-naupliar somites and appendages. Most of the larval forms are pelagic in habit, and many show special adaptations to this mode of life, especially in the develop- ment of spines and processes which are probably less important as defensive organs than as aids to flotation. PHYLOGENY. The study of fossil Crustacea has hitherto contributed com- paratively little towards a precise knowledge of the phylogenetic history of the class. Although their remains are abundant in nearly all formations, from the most ancient up to the most recent, in very many cases only the carapace or shell is preserved, the limbs being lost or represented only by indecipherable frag- ments. Many important forms must have escaped fossilisation altogether owing to their small size and delicate structure. Further, many important groups were already differentiated when the geo- logical record began. The Branchiopoda, Ostracoda, and Cirripedia are represented in Cambrian or Silurian rocks by forms which seem to resemble those now existing, so that palaeontology can have little light to throw on the mode of origin of these groups. In the case of the Malacostraca the material is a little more promising. It is not improbable that the Phyllocarida, Avhich are found from -the Cambrian onwards, may include the forerunners of the true Mala- costraca, but nothing is definitely known of their appendages. The recent discovery, in the Tasmanian Anaspides and the Australian Koonunga, of what are believed to be representatives of the Carboniferous and Permian Syncarida, has given a clue to the affinities of forms hitherto problematical, and may throw light on the relations of other Palaeozoic fossils hitherto vaguely referred to "Schizopoda" or Decapoda. Remains of undoubted Decapods are first met with in Mesozoic rocks. They are abundant in many deposits, and are sometimes sufficiently well preserved to render possible fairly accurate determination of their affinities. The Isopoda and Stomatopoda are known from Mesozoic deposits, but have hitherto yielded no results of phylogenetic importance. In view of the scarcity of trustworthy evidence as to the actual forerunners of existing Crustacea, phylogenetic conclusions based on the data of comparative anatomy and embryology remain largely speculative. They are none the less a necessary preliminary to the attempt to construct a natural system of classification. The earlier speculations on this subject started from the assump- tion that the " theory of recapitulation " could be applied to the larval history of the Crustacea. The various forms of larvae, more especially the nauplius and the zoe'a, were supposed to reproduce, 26 THE CRUSTACEA more or less closely, the actual structure of ancestral types. As regards the zoea, this assumption was soon shown to be erroneous, and the secondary nature of this type of larva is now generally admitted. The nauplius, however, by the constancy of its general characters in the most widely diverse groups, shows itself to be a very ancient type, and the view has been advocated that the Crustacea have arisen from an unsegmented nauplius - like ancestor. To this view there are considerable objections. Several structures which can hardly have been absent from the common stock of the Crustacea, such as the paired eyes and the shell-fold, are not found in the nauplius. Other characters common to certain Crustacea and Annelids, such as the mode of growth of the somites, the structure of the nervous system and of the heart, can hardly be supposed to have arisen independently in the two groups. The view now most generally adopted is that the Branchiopoda, and especially Apus, which resemble the Annelids in the characters just mentioned and also in the large number and uniformity of the trunk-somites and their appendages, approach most nearly to the primitive Crustacean type. On the other hand, in some respects, such as the reduced mouth-parts, the Branchiopoda are considerably specialised. In some Copepoda the cephalic appendages are much more primitive than in the Branchiopoda, and the first three pairs of appendages retain throughout life, with little modification, the shape and function which they have in the nauplius stage. It is possible, however, that in these characters the Copepoda are persistently larval rather than phylogenetically primitive, and in other respects, especially in the absence of paired eyes and of a shell-fold, they are certainly specialised. In order to reconstruct the hypothetical ancestral type, there- fore, it is necessary to combine the characters of several of the existing groups. It may be supposed to have approximated, in general form, to Apus, with an elongated body of numerous similar somites, terminating in a caudal furca ; with the postoral appendages all similar, and with a carapace originating as a shell-fold from the maxillary region. The eyes were probably stalked and movable, the antennules uniramous, and the antennae and mandibles biramous and natatory, and both armed with masticatory processes. The trunk -limbs were probably biramous but with additional endites and exites, and all provided with gnathobases. It is to be noted that the Trilobita, which, according to the classification adopted in this work, are dealt with under Arachnida, approximate to the structure of the primitive Crustacean here sketched except in the absence of a shell-fold and in having the eyes sessile. It is not necessary, on this view, to deny all phylogenetic significance to the nauplius. It may be regarded as an ancestral THE CRUSTACEA 27 larval form, corresponding perhaps (as Hatschek suggests) to the stages immediately succeeding the trochophore in the development of Annelids, but Avith some of the later -acquired Crustacean characters superposed upon it. The five sub-classes into which the Crustacea are divided in the classification here adopted appear to represent independent or nearly independent lines of descent from the primitive stock. Their relations to each other and the probable course of evolution within each group will be dealt with in subsequent chapters. It may be mentioned that the classification introduced by Latreille in 1806, in which the Malacostraca are opposed to all the. other Crustacea grouped together under the name of Entomostraca, is still frequently adopted. The Entomostraca, however, like the Invertebrata, constitute a very heterogeneous group, defined only by negative characters and having no claim to retention in a natural system of classification. LITERATURE. (Selection of important works dealing with the Class as a ichole.) In this list and in those appended to the following chapters no attempt has been made to give a complete bibliography of the subject. The references given in the works mentioned will, however, provide the necessary guidance for the student who desires a further acquaintance with the vast mass of carcinological literature. 1. Brunts, L. Contribution ;\ 1'e'tude de 1'excretion chez les Arthropodes. Arch. Biol. xx pp. 217-422, pis. vii.-ix., 1903. 2. Glaus, C. Untersuchungen zur Erforschung der genealogischen Grundlage des Crustaceensystems. 4to. Wien, 1876. 3. N"eue Beitrage zur Morphologic der Crustaceen. Arb. zool. Inst. Wien, vi. pp. 1-108, pis. i.-vii., 1886. 4. Das Medianauge der Crustaceen. Arb. zool. Inst. Wien, ix. pp. 225- 266, 4 pis., 1891. 5. Dana, J. D. The Crustacea of the United States Exploring Expedition, 1838-1842. 2 vols. 4to, and Atlas, fol. Philadelphia, 1852-1853. 6. Dohrn, A. Geschichte des Krebsstammes. Jenaische Zeitschrift, vi. pp. 96-156, 1871. 7. Gerstaecker, A., and Ortmann, A. E. Crustacea, in Bronn's Klassen und Ordnungen des Thierreichs, Bd. v. Iste Abth. 1320 pp. 50 pis., 1866- 1879, by Gerstaecker. ' 2te Abth. viii + 1319 pp. 128 pis., 1881-1901, by Gerstaecker to p. 1056, continued by Ortmann. 8. Grobben, G. Die Antennendriise der Crustaceen. Arb. zool. Inst. Wien, iii. pp. 93-110, pi. ix., 1881. 9. Zur Kenntniss des Stammbaumes und des Systems der Crustaceen. Sitz.-Ber. Akad. Wien, ci. Abth. i. pp. 237-274, 1892. Translated in Ann. Mag. Nat. Hist, (6) xi, pp. 440-473, 1893. 28 THE CRUSTACEA 10. De Ilaan, W. Crustacea, in P. F. v. Siebold's Fauna Japonica. Fol. Lugduni Batavorum, 1833-1849.' 11. Hansen, H. J. Zur Morpli. der Gliedmasscn und Mundtheile bei Crustaceen und Insecten. Zoo}. Anz. xvi. pp. 193-198 and 201-212, 1893. Trans- lated in Ann. Mag. Nat. Hist. (6) xii. pp. 417-434, 1893. 12. Huxley, T. H. The Crayfish, an Introduction to the Study of Zoology. Internal. Sci. Series, vol. xxviii. London, 1880. 13. Korschclt, E., and Heider, K. Lehrbuch der vergleichenden Entwicklungs- geschichte der wirbellosen Thiere. 8vo. Jena, 1890-1893. Translated by M. Bernard, edited by M. F. Woodward, as Text-book of the Embryology of Invertebrates. 4 vols. (Crustacea in vol. ii.) 8vo. London, 1899. 14. Milne-Edwards, H. Histoire Xaturelle des Crustaces (Roret's Suites a Buftbn). 3 vols. and atlas, 8vo. Paris, 1834-1840. 15. Le Regne animal . . . G. Cuvier. Illustrated edition. Crustaces, 1 vol. and atlas, Svo. Paris, 1836-1849. (Only the plates are of much importance.) 16. Miiller, F. Fur Darwin. Svo. Leipzig, 1864. Translated as Facts and Arguments for Darwin. London, 1869. 17. Parker, G. H. The Compound Eyes in Crustaceans. Bull. Mus. Comp. Zool. Harvard, xxi. pp. 45-140, 10 pis., 1891. 18. The Retina and Optic Ganglia in Decapods, especially in Astacus. Mitth. Zool. Stat. Neapel, xii. pp. 1-73, pis. 1-3, 1897. CHAPTER II THE BRANCHIOPODA SUB-CLASS BRANCHIOPODA, Latreille (1817). Order 1. Anostraca. „ 2. Notostraca. „ 3. Conchostraca. ,, 4. Cladocera. Sub-Order 1. Calyptomera. Tribe 1. Ctenopoda. „ 2. Anomopoda. Sub-Order 2. Gymnomera. Tribe 1. Onychopoda. „ 2. Haplopoda. Definition. — Crustacea in which the carapace may form a dorsal shield or a bivalve shell or may be entirely absent ; the number of trunk-somites varies greatly ; the posterior part of the trunk is without limbs and usually ends in a caudal f urea ; the antennules are generally reduced and unsegmented ; the mandibles have no palp or only a vestige of one ; the maxillae are reduced or absent ; the trunk-limbs, which vary greatly in number, are generally of uniform structure, rarely pediform, generally foliaceous and lobed ; the position of the genital apertures varies greatly ; the paired eyes are rarely absent ; development usually with metamorphosis ; young hatched in nauplius or metanauplius stage. Historical. — The earliest mention of any of the Branchiopoda is found in the works of Swammerdam, who, in 1669, described a species of Daphnia as " Pulex aquaticus arborescens," and the name of " water-fleas " has since been commonly applied to the group of Cladocera. Apus and Branchipus were described early in the eighteenth century, and, together with Daphnia, formed the sub- jects of a series of remarkable memoirs (1752-1756) by J. C. Schiiffer. The classical work of O. F. Miiller, Entomostraca .v // Insecta testacea (1785), described a large number of new types and 29 THE CRUSTACEA B laid the foundations of classification. The order Branchiopoda, as first defined by Latreille in 1817, included Ostracoda, Cope- poda, and Limulus ; and Milne- Edwards, in 18 40, while exclud- ing these, retains the later added Nebalia. Later attempts to extend the limits of the group to readmit the Ostracoda and Branchiura have not met with support, while Claus's demon- stration of the Malacostracan affinities of Nebalia and its allies is now generally accepted. Among the authors who, in the first half of the nineteenth century, contributed to a know- ledge of the group, Jurine, Fischer, and Baird may be mentioned. Zaddach's mono- graph on Apus is still the chief source of information on many points of anatomy. Leydig's work on the Cladocera is especi- ally important as regards internal anatomy and histology ; and Weismann's series of papers deal, among many other points, with the remarkable phenomena of their reproduction, which had attracted attention from the time of Schaffer. As is the case with most other groups of Crus- tacea, modern conceptions of the morphology of the Branchio- poda are largely indebted to the works of Glaus. Lankester's FIG. 15. Branchineda paludosa, one of the Anostraca. X 4. (After Sars.) A, female ; B, male. a', antennules ; a", antennae, enlarged in the male to form clasping organs ; p, paired penes of the male ; ut, ventral prolongation of the genital segment in the female, containing the " uterus " tilled) with eggs. On the front of the head, between the antennules, is the unpaired eye, and,. just behind it, the stalked compound eye. Dorsal to the pear-shaped mandible is seen the groove which appears to divide the head-region into two segments. Following this, the trunk shows eleven limb-bearing and eight limbless somites (besides the telson), the first and second of the latter partly coalesced. paper on the appendages and nervous system of Apus greatly influenced opinion in favour of the primitive position of the Branchiopoda. Among the more purely descriptive and faunistic works the numerous papers of G. 0. Sars hold an important place, and mention may also be made of the fine monograph on the Cladocera of Sweden, published THE BRANCHIOPODA in 1900 by Lilljeborg, who had made impoi'tant contributions to the subject so long ago as 1853. MORPHOLOGY. The Branchiopoda present considerable diversity of general ehape. In the Anostraca there is no shell-fold, and the elongated FIG. 16. l.i'i'tiliirns !i!m:ial!!i. one of the Notostraca. X 2i. (After Sars.) Dorsal view on the left, lateral view on the right. In the latter, one-half of the carapace has been cut away to show tli'- segmentation of the trunk and the appendages. On the dorsal surface are seen the paired e\vs with the "dorsal organ" behind and between them, the median dorsal ridge of the posterior part of the carapace, and thu two transverse grooves just behind the eyes. body, composed of many distinct somites, has an almost vermiform aspect (Fig. 15). The flattened natatory feet are more laterally placed than is usual in Crustacea, increasing the resemblance to the Chaetopod worms. In the Notostraca also the body is elongated, and composed of numerous somites, but its anterior portion is covered by a broad arched carapace (Fig. 16). In the THE CRUSTACEA Conchostraca the body tends to be laterally compressed, and is completely enclosed in a bivalved shell closely simulating that of some lamellibranchiate molluscs (Fig. 17). In the Cladocera the body is composed of few somites and its segmentation is more or less obscured, while the bivalve shell, in most cases, covers the body and limbs, leaving the head free (Fig. 18). ad. md l-'u:. 17. Estlieria obliqua (Conchostraca). A, shell of female, from the left side. B, male seen from the side, after removal of left valve of shell. (After Sars.) a', antennule ; a", antenna; ml, adductor muscle ; /, caudal furca ; md, mandible. On the first and second pairs of trunk- linibs are seen the " claspers " of the male. Except in the Cladocera, the head is defined from the first trunk- somite. In the Anostraca it is divided into two portions by a transverse groove which crosses the dorsal surface just above the mandibles. This groove is also found in the Conchostraca, and it is no doubt homologous with the anterior transverse groove on the dorsal surface of the carapace in the Notostraca. In the Conchostraca and Cladocera the front of the head is more or less produced downwards forming a rostral process, and in the Notostraca it is inflected, forming a sharp semicircular anterior edge continuous on either side with the lateral margins of the carapace. THE BRANCHIOPODA 33 a. c. — f.-- The carapace may be directly continuous in front with the dorsal integument of the head as in most Cladocera, or defined from it by a groove as in some Cladocera and the Notostraca. In the Conchostraca the con- nection between the animal and the shell is reduced to a comparatively narrow neck and the lateral lobes of the shell extend forwards on each side so as to enclose the whole head (Fig. 17). Gener- ally the shell-fold does not coalesce with any of the trunk-somites which it en- velops, with the exception of the one or two anterior somites which in the Clado- cera are fused with the head. In the aberrant Cladoceran Leptodora, however, it coal- esces with the dorsal surface of the leg-bearing somites, and its free portion, which here, as in some other Clado- cera, forms merely a brood- sac, appears to arise from tllC posterior margin of the ''-c> brood -cham her ; br, brain; c, margin of carapace; • , i , •• . /TT -i n\ ''•*'> caudal setae; e, compound eyes coalesced into SlXtn trunk-SOmite (rig. 19). one;/, furca; gl, maxillary gland; h, heart; hep, Thp rnnivipp rvnv HP hePatic diverticulum of gut ; n.e, nauplius eye ; ov, UllclpdCC may e ovary (After Claus and Grobben.) more or less corneous, but it is never strongly calcified. In a few Cladocera (Monospilus, etc.) and in the Conchostraca (Fig. 17) the integument of the outer surface of the shell is not cast off in ecdysis, but remains in position, giving rise to a series of "lines of growth" marking the increased size of the shell at each moult. Special modifica- tions of the carapace for protection of the eggs Avill be referred to below. The number of trunk-somites varies very much. It is greatest in the Xotostraca, where 42 somites are found in certain species of Apus. In the Conchostraca the number is from about 13 to 28, and in the Anostraca 19 to 23. In the Cladocera the segmentation of the body is generally more or less obscured ; at least the first two somites are always coalesced with the head. In 1 ''"/ih niu, according to Claus, these are followed by three limb- bearing somites, and the succeeding apodous region is divided in the young into three "abdominal" somites and a " postabdomen " c.s. Fia. 18. female, a', antennule ; a", antenna; 34 THE CRUSTACEA or telson. This segmentation of the apodous region is distinctly retained in the adult Leptodora (Fig. 19). Apart from the presence or absence of appendages, the trunk of the Branchiopoda is not differentiated into distinct regions. By various authors the terms "thorax" and "abdomen" have been ap- plied respectively, sometimes to the pre- and post-genital, sometimes to the limb-bearing and limbless regions of the trunk. As the limits between these regions do not coincide, even approximately, r-f. FIG. 19. teptodora Tdndtii, female, x 10. «', antennule ; «", antenna ; c, carapace, reduced to a brood- sac ; c.s, caudal setae (compare figure of Daphnia, p. 33) ; e, compound eye ; /, furca ; p, first trunk-limb. (After Lilljeborg.) except in the Anostraca, it seems better to avoid altogether the use of the terms " thoracic " and " abdominal " in dealing Avith this group. In the Notostraca a varying number (4-14) of the posterior somites are without appendages. In the post-genital region of the body the number of pairs of appendages greatly exceeds the number of somites, some of the posterior somites carrying as many as six pairs. In the Anostraca there are from four to nine limbless somites, but at least the two anterior are coalesced to form the genital segment. In the Conchostraca the short post-pedal region is unsegmented. THE BRANCHIOPODA The last segment of the body, or tclson, carries, except in Limnetis and Thamnocephalus, a pair of furcal rami, long antenni- form filaments in the Notostraca (Fig. 16), unsegmented styles or flattened plates in the Anostraca (Fig. 20, C). In the Conchostraca and Cladocera the posterior part of the body is flexed ventrally and the furca is represented by a pair of strong curved claws (Fig. 17, /). The anus opens either at the end of the body 9n B. C. 4 T. pi Fio. 20. A, head-region of Lepidurus glaeialls from below : «', antenntile : «" . antenna ; yn, gnatlio- base of tirst trunk-limb; L, labrum or tipper lip turned forwards (in the natural position it covers the opposed edges of the mandibles) ; I, lower lip (according to Clans, the inner lobes of the maxillulae) ; in, mandible; mx', maxillula ; HIS", maxilla. B, posterior end of body of I.i-liiilii,-iis 'inlis ; o, position of anal opening : /''. supra-anal plate : r, rami of caudal fmra : '/'. terminal segment or telson. C, posterior end of body of Brnnflihn'i-t" />"/»<'".-•» ; letters as above. (After Sars.) between the furcal rami or, in many Cladocera, some distance in front on the dorsal surface. In the genera Lepidurus (Notostraca) and Thamnocephalus (Anostraca) the telson is produced as a thin plate above the anal opening (Fig. 20, B). Appendages. — The antennules are, for the most part, purely sensory in function, not segmented or obscurely so, and carrying tufts of sensory filaments. In many Cladocera they are very small and attached to the posterior surface of the deflexed beak-like 36 THE CRUSTACEA process forming the front of the head (Fig. 18, «'). In the males of some Cladocera they form clasping organs for holding the female. The antennae differ very much in the various orders. In the Notostraca they are vestigial (Fig. 20, A, a") and may even be absent in certain species of Apus. In the males of the Anostraca they are converted into powerful claspers for seizing the female, and may assume extraordinary and complex forms. A pair of filaments, known as frontal appendages (Figs. 21 and 32, /.«), arise, in many Anostraca, from the base of the antennae, though sometimes they are inserted on the front of the head and seem quite unconnected with the antennae. They may coalesce at the base, and may be more or less ramified, adding to the complexity of the apparatus PIG 21 carried by the male. In the females the Head of chirocephaius diapiiamis antennae are much simpler in form, and are Probablv Purelv sensoiT in Unction. antenna ; /.«, frontal appendages, It WOUld Seem from Claus's observations which in this species are large and , , , ,, „ , . Al branched. on the development of JJranc/npus that the endopodite of the antenna atrophies, and that the whole apparatus of the adult develops from the protopodite and exopodite of the larval antenna. In the Conchostraca and Cladocera the antennae are large biramous swimming - organs. In the former group they have multiarticulate rami, and they are protruded ventrally between the valves of the shell (Fig. 17). In the latter order the rami have few segments, and, since the head is not enclosed between the valves of the shell, the antennae project freely (Fig. 18). In the genus Lato-na (Cladocera) a large process from the proximal segment of the exopodite causes the antenna to appear as if three- branched. The mandibles (Fig. 8, C, p. 12) are devoid of palps in all existing Branchiopoda with the exception of Polyartemia, in which a vestigial palp has recently been found by Ekman. As a rule, they have broad toothed triturating surfaces, but in some predatory Cladocera (Leptodora) they become long sickle-shaped blades. The maxillulae (Fig. 22, A) are of small size and simple form. As a rule, each consists of a single lobe armed with spines on the inner edge. In the Notostraca they seem to consist of two lobes, but it is possible, as Glaus suggested, that the inner and anterior lobes represent the lower lip, or paragnatha, otherwise wanting in the group. The maxillae (Fig. 22, B) are also greatly reduced and consist of a simple lobe. In the Cladocera they are entirely wanting in the adult, but a distinct rudiment is present in the embryo. In the Notostraca THE BRANCHIOPODA 37 an external membranous lobe is present and was formerly regarded as corresponding to the gill or brad of the succeeding appendages. For this reason the appendage was regarded as belonging to the series of the trunk-limbs, and was called a "foot-jaw" or "maxilliped." It has been shown, however, by Glaus that the so-called "bract" is really a tubular process bearing the external opening of the maxillary gland, and is in no way related to the " bract " of the following limbs. There is therefore no reason to doubt that the appendage is homologous with the maxilla of other Branchiopoda. It is of interest to notice, in connection with the great reduction of this appendage, that there is, in Apus, no separate ganglion in the ventral nerve-chain corresponding to it, but that the nerves supply- ing it take their origin from the longitudinal connectives between the ganglia corresponding to the maxillula and the first thoracic appendage respectively. A. jf -~^ ,5F" \T g o. Fio. 22. A, maxillulae, B, maxilla, of Apus cancrifonnis. (After Clans.) p, the so-called inner lobes of the maxillulae, representing, according to Claus, the pusgnatha ; o, opening of maxillary •ilsind on a tubular process connected with the maxilla. The trunk-limbs of the Branchiopoda are generally of very uniform structure, and are not grouped into definite " tagmata." On the other hand, these appendages present very great differences in the different groups composing the sub-class, and it is not easy in some cases to determine the exact homologies of the various parts. The most primitive form is probably that found in the Noto- straca (Apus, Fig. 23, A). Each appendage consists of a flattened corm or axis to which are attached eight, lateral processes or lobes. Six of these processes spring from the inner margin of the corm and are termed endites, while two on the outer margin are termed exiles. The proximal endite (1), placed near the attachment of the limb to the body, is armed with strong spines and, like some or all of the other endites, is provided with special muscles which permit of a limited amount of movement on the corm. These basal endites working against those of the opposite side function as jaws, seizing and passing forward towards the mouth particles of food Avhich are THE CRUSTACEA Fio. 23. Trunk-limbs of various Branchiopoda. A, seventh trunk-limb of -ly/»* n/,ir, •;/,»•,//;.< (Xotostraca). B, third trunk-limb of Esthcria obliqua t'Conchostraca). C, fifth trunk-limb of Branchinei-ta paUidosn (Anostraca). D, eleventh trunk -limb of Polyortemia forcipata (Anostraca). (A after Lankester ; B, C, and D after Sars.) br, branchia or bract ; fl, tlabellum ; /, tactile process of the fifth endite in Estheria ; x, proximal exites of the Anostraca ; 1-6, the six endites, of which the first is the gnathobase. Between the flabellum and the sixth endite in A is seen the " sub-apical lobe." Regarding the homology of the plate lettered fl in C and D, see the discussion in the text. THE BRANCHIOPODA 39 drawn in between them by the rhythmical movements of the appendages in swimming. They are therefore distinguished as gnathobases. The Branchiopoda are the only Crustacea in which gnathobases are found on limbs far removed from the mouth. Of the two exites, the distal, known as the flabellum (/?), is a thin plate Avith setose margins. It is moved by special muscles and probably serves chiefly as a swimming-plate. The proximal exite, known as the bract (br), is branchial in function, having a very thin cuticle, without setae, and is not provided with muscles. The end of the corm, external to the sixth endite, is produced into a rounded process, the "sub-apical lobe." While most of the postmaxillary appendages of the Notostraca have the general structure of that just described, differing only in details of shape and proportion of the various parts, certain of them are specially modified. In the first and second pairs, the corm is divided, in the former into four, and in the latter into two segments. The endites, with the exception of the gnathobase, become more elongated in passing forwards along the series of limbs, and in the first pair the second, third, fourth, and fifth endites, counting from the base, are produced into filiform multiarticulate rami (Fig. 20, A). The fifth, in some species, is nearly as long as the body of the animal. The sixth endite, however, is much reduced and of peculiar form in the limbs of the first pair. The trunk-limbs of the Conchostraca (Fig. 23, B) can without difficulty be compared with those of Apus. The six endites are distinct though reduced in size and less sharply marked off from the unsegmented corm. The gnathobase is unprovided with masticatory spines, but, like the following endites, is fringed with setae. The fifth endite is produced in some cases into a long tactile process (t). Only the sixth endite is marked off by a distinct articulation. The flabellum is very large and the branchia is reduced in size. In the Anostraca (Fig. 23, C, D) the structure of the parts is rather different and their exact homologies are not quite clear. The flattened unsegmented corm has its inner edge more or less distinctly divided into six lobes. At the distal end, towards the outer side, is a broad oval plate (/) defined by a well-marked articulation and fringed with setae. On the proximal side of this, on the outer edge, is the branchia, characterised as such by its thin cuticle and by the lack of setae. Still nearer the base the outer margin of the corm is produced into a rounded plate (x), very thin and not defined by an articulation. In Poli/artemia and in some species, at least, of Chirocephalus there are two such plates (Fig. 23, D). It appears most probable that the six lobes of the inner edge correspond to the six endites of the Notostracan limb, in which case the distal setose plate (/) will represent 40 THE CRUSTACEA the flabellum. This view, however, is not accepted by Sars, who regards the distal setose plate as the sixth endite and supposes that the flabellum is wanting in the Anostraca. Sars's interpretation gains some support from a comparison with the Conchostracan type of limb, where the sixth endite has much the same position as the setose plate of the Anostraca. It is further supported by the fact that, in the development of the limbs of Apus, the flabellum only appears after some of the endites have become marked off, while in Branchipus the first differentiation to take place in the limb-buds is a bifurcation defining the distal setose plate from the terminal lobe of the inner edge. In any case it seems certain that the external basal plate or plates of the Anostraca are new formations unrepresented in the Notostraca and Conchostraca, and not, as Lankester supposed, homologous with the branchia (or bract) of these groups. In the Cladocera the structure of the limbs is still more difficult to interpret. In the Ctenopoda (Fig. 24, A), where the six pairs of "thoracic" appendages are comparatively uniform, the inner margin forms a small gnathobasic lobe followed by a broad lobe carrying a comb- like series of long setae. A distal lobe may correspond to the flabellum of Apus and Estheria, or perhaps to the sixth endite. The branchia is present and has the usual characters. In the Anomopoda there is considerable differentiation among the members of the series. In Daphnia, for example, the first limb (Fig. 24, B) is obscurely segmented and the five lobes on the inner edge are slightly developed. The second limb (Fig. 24, C) has the gnathobase enlarged and has an outer branch regarded as the exopodite. Both of these limbs are adapted by the possession of long curved setae to aid in the prehension of food. The third (Fig. 24, D) and fourth pairs are characterised by the great development of the proximal endite with its comb -like row of setae. They serve to keep a current of water flowing between the valves of the shell for the purposes of respiration and feeding. The last pair in Daphnia are greatly reduced (Fig. 24, E). In the Gymnomera, and especially in Leptodora (Fig. 19), the trunk-limbs have lost the characteristic leaf-like shape and have become cylindrical, elongated, and divided into four well-marked segments, without any trace of endites or exites, serving only for seizing and holding prey. Modifications of certain of the trunk-limbs subservient to the processes of reproduction are found in all divisions of the Branchiopoda. In the males of most of the Conchostraca the first two pairs have the terminal portion modified into a cheliform clasper (Fig. 1 7, B), but in Limnetis and in the Cladocera only the first pair is so modified. Sexual modifications of trunk-limbs in the female are found only in the Notostraca and in some Conchostraca. THE BRANCHIOPODA In |the former the appendages of the genital somite (the eleventh trunk-somite) are modified to form brood-pouches, and are known as " oostegopods " (Fig. 25). The sixth endite is coalesced with A. FIG. 24. Trunk-limbs of Cladoeera. A, first trunk-limb of Sida crustallma (CtenojKxla). B, first, C, second. L), third, E, fifth trunk-limbs of Daphnm pulex (Anomopoda). br, brancliia ; gn, gnathoba.se. (After Lilljeborg.) the sub-apical lobe, and together with it forms a hemispherical cup, closed by a movable lid formed by the flabellum. In some species of Conchostraca two or three pairs of limbs near the genital apertures have the proximal division of the flabellum produced THE CRUSTACEA and thickened, and to this the egg-masses are attached between the valves of the shell. As regards the homologies of the parts of the Branchiopod-limb with those of the biramous type found in other Crustacea, two views have been suggested. According to the interpreta- tion adopted by Huxley, among others, the " flabel- lum " or distal exite of the Apns-\im\) corresponds to the exopodite, while the distal part of the corm represents the endopodite. Lankester, on the other hand, considers the endopodite and the exopodite to be represented by the fifth and sixth endites respectively, the corm being the protopodite Fio 25 and the flabellum the epi- Eleventh trunk-limb or oostegopod of Apus cancri- P°dite. The former V16W formis, female. (After Lankester .) 1-6, the six endites, is supported by a Comparison of which the sixth is fused with the " sub-apical lobe . * * ~ , „ , which forms the pouch, p, containing the eggs; fl, With the leaf -like thoracic flabellum forming the cover of the pouch; br, the T i r it. T vestigial bract. limbs of the Leptostraca, while Lankester's interpreta- tion is based chiefly on a comparison of the pre-oral with the post- oral appendages in the larval Apus, and of the trunk-limbs of Apus with the maxilla and maxillipeds of various Decapods. Neither view is quite free from difficulties, and the divergences in structure mentioned above as occurring in the Anostraca still further complicate the matter, which requires further investigation. Alimentary System, — The oesophagus is narrow and is provided with constrictor and dilator as well as longitudinal muscles. It usually projects a little way into the more capacious mid-gut, and in Polyartemiii the terminal part is armed with setae. The hind-gut is short and muscular. In the aberrant Cladoceran Leptodora oesophagus is of great length and mid -gut hardly extends in front of terminal segment of the body. In many Cladocera belonging to the families Lyncodaphnidae and Lynceidae the mid-gut is more or less coiled, forming a simple loop or a spiral the the the Fio. 26. Pleuroxus u ncinatus (Cladocera) (after Lilljeborg), showing the coiled intestine and its ventral diverticulum. THE BRANCHIOPODA 43 of one and a half turns (Fig. 26). As a rule, the anterior part of the mid-gut gives off a pair of diverticula. In the Notostr'aca and Conchostraca these are much ramified, and in the Anostraca they are saccular and lobulated. In the Cladocera they are often absent, i\nd when present are simple caeca. In some Cladocera of the family Lynceidae the posterior part of the alimentary canal gives off an unpaired diverticulum on the ventral side (Fig. 26). It is usually short, but sometimes it is of considerable length. It is stated to arise from the hind -gut, and its walls contain large gland-cells. Circulatory System. — Except in the Cladocera, the heart is elongated and tubular. In the Anostraca it traverses almost the whole length of the trunk, and has paired ostia in each somite except the first and last. In the Notostraca the heart extends through the first eleven trunk-somites and has eleven pairs of ostia. In the Conchostraca it is restricted to three or four somites, with a corresponding number of ostia.- In the Cladocera the heart is greatly abbreviated, forming a sub-globular sac with a single pair of ostia, lying in the region of the first trunk-somite (Fig. 18, A). There are no distinct vessels and the blood is discharged directly from the anterior end of the heart (through three openings in the Notostraca) into the sinuses of the head-region. In some genera (Apm, Branchipus, Artemia, and some Cladocera) the fluid of the blood is coloured red, owing, as Lankester has shown, to the presence of haemoglobin. Excretory System. — The maxillary gland (Fig. 18, gl) is well developed in all Branchiopoda. Except in the Anostraca and some aberrant Cladocera (Leptodom), it lies within the "thickness of the shell-fold. It shows the typical structure, consisting of end-sac, glandular coiled tube, and short terminal duct. In the Notostraca, as already mentioned, the external opening is at the end of a tubular process arising from the under side of the head, close to the maxilla. The antennal gland is well developed in the larval stages of Anostraca, Notostraca, and Conchostraca (Fig. 10). A vestige of it persists in the adult Artemia. Glands. — Belonging to the dermal series of glands are the segmentally arranged ventral and leg glands of some Anostraca, and the groups of unicellular glands in the labrum which are probably always present. Special interest attaches to the structure variously referred to as " neck gland," " dorsal organ," or " adhesive organ," found in many Branchiopoda. In its simplest form this consists of a thickened and glandular area of the ectoderm, on the dorsal surface near the posterior limit of the head-region. The adhesive secretion reaches the exterior through pores or slits in the cuticle, and is used 44 THE CRUSTACEA for the purpose of attaching the animal temporarily to plants or other objects. In the Notostraca, however, where the organ is apparently well developed, it is not used for this purpose. In the Anostraca it is absent in the adult, though very large in the larva. In some of the Conchostraca (Limnadia) it is prominent and pedunculated. In many Cladocera it is greatly reduced or absent, but in some it is large and functional. In the Sididae the organ is divided into three parts, a large median anterior and a pair of smaller ones posterior to it. In this case certain muscles attached to the integument in the neighbourhood of the organ are believed to produce a sucker- FIG. -2~ Nervous system of Sinini-i'iilinlini (Cladocera) from the dorBal side. (After Cnnnington.) rer, brain ; n.f. 4. Endoskeleial Structures. — In many Branchiopoda there is a considerable de- velopment of trabeculae and plates of tendinous connective tissue giving attach- ment to muscles. This system, which almost merits the name of endoskeleton, n is most fully developed in the Anostraca, where the external cuticle remains almost membranous, but it is also found in the Cladocera. In the Notostraca the chief part of this endoskeletal system is a tendinous plate, known as the entosternite, lying under the anterior part of the alimentary canal and giving attachment to the adductors of the mandibles and to other muscles. Muscular System. — In the Conchostraca and Cladocera the valves of the shell are drawn together by a transverse adductor muscle. In some, at least, of the Cladocera this muscle is double. n.f.5. THE BRANCHIOPODA 45 Nervous System. — As has been mentioned above, the nervous system of most Branchiopoda shows very primitive characters in the ladder-like form of the ventral nerve-chain (Fig. 11, p. 17, and Fig 27). In some Cladocera, however (Polyphemidae and Lepto- dora), the ventral ganglia are more or less coalesced into a single mass. In all cases, the nerves to the antennae arise not from the supra-oesophageal ganglia, but from the first post-oesophageal pair of ganglia. This pair of ganglia is quite distinct from the next succeeding pair, which innervate the muscles of the mandibles, and, like all the ventral pairs of ganglia (with the occasional exception of the mandibular), it has a double transverse com- missure. It is closely connected with the origin of a " visceral " nerve-ring which encircles the oesophagus and bears an unpaired ganglion in the region of the labrum (Figs. 27 and 28, n.lbr). So far as their nerve-supply is concerned the antennae are, in the Branchiopoda, unmistakably post-oral. In the Noto- straca the nerves of the antennules (Fig. 28, n. a) arise from the oesophageal connectives just in front of those of the antennae. This was supposed by Lankester to be a primitive condition indicating the post-oral origin of these appendages. Pelseneer showed, however, that the fibres of these nerves pass forward to a pair of nerve-centres in the supra-oesophageal ganglion. In the other Branchiopoda the corresponding nerves arise from the posterior part of the brain, and there can be little doubt that the condition in Apus is a secondary one due to the brain having been shifted forwards and upwards in company with the paired eyes which lie on the dorsal surface of the head, while the antennules are in- FIG. 28. Sorted On its lower Surface. In ApUS Anterior part of the central ,, 1. T ,-, nervous system of Arum, from below, there is no ganglion corresponding to the semi -diagrammatic. r,-r, brain; reduced maxillae and these limbs receive gnM"' arvtennal ganglion ; ,(.«', nerve ui.i.MiitU., ami tuebe eive to antennule ; n.a", nerve in a pair of nerves arising from the longi- antenna ; n.f, nerves to first trunk- 1 . -n i limb ; n.Ior, nerve-ring encircling tudinal connective between the maxillular oesophagus; «.•>«./. nerves to man- i £. ,1 T /T71- r,0 dible ; n,maf. nerves to maxillula ; and first thoracic ganglia (Fig. 2 8r1M»xrf nerve to mariUa;o«, portion •a. in. /'). In the Cladocera, also, where the aL^pt^'nmo (Aft°r Lankestt'r maxillae are quite rudimentary or absent, there is no ganglion corresponding to this somite. In Branchipus, however, there is a distinct pair of ganglia with double transverse commissures. THE CRUSTACEA Sense-Organs. — The compound eyes are present in all Branchio- poda. In the Anostraca they are elevated on movable peduncles, but in all other Brarichiopoda they are sessile, preserving, however, a certain degree of mobility owing to the fact that they are sunk below the surface and covered by an invagination of the outer cuticle, the cavity of the invagination generally remaining in communication with the exterior by a pore (Fig. 29). Except in c e the Anostraca, the two eyes are closely approximated ; and in some Conchostraca, and more completely in the Cladocera, they fuse together into a single eye. The cornea is not distinctly faceted. The crystalline cone is divided into four parts. The rhabdom is surrounded by five rhabdom -cells, but (in ranchipus) is not distinctly divided into rhabdomeres. Except in some of the Cladocera, the nauplius eye persists in the adult of all Branchiopoda (Figs. 18 and 29, n.e). It is of the usual Diagrammatic longitudinal section through the tripartite Structure, and may compound eye and associated structures of Apus. i ' j •.•> ,1 i (After Bernard.) br, brain; c, external cuticle of be connected With the brain cotpinmd 3WTJ5faWE1?£=BS by three separate nerves or into the water-sac covering the compound eye and by a sin°"le One. In the sending a blind diverticulum into the nauplius eye ; *-*•, j . . j , . s, the water-sac, formed by an invagination of the LladOCCra it IS reduced 111 integument. The pore, p, is in the median line, so _• J eimnlifiprl in Qrrnp that, as a matter of fact, a strictly longitudinal Slze ana Simplltied m StrUC- section passing through it would pass between the ture, and mav be entirely compound eyes. * J absent. Frontal Organs. — In Branchipus a pair of organs, presumably sensory, lie on the front surface of the head (Fig. 5, /). Each consists of a large hypodermis cell surrounded by a group of ganglion cells and connected with the brain. Below this, in Branchipus, is a group of club-shaped cells containing peculiarly shaped rod -like bodies and connected with nerves. In the Cladocera the first-mentioned frontal organs are vestigial, and appear to be represented by a group of cells supplied by a continuation of the nerve of the median eye. The club-shaped cells, however (the " Nackenorgan " of Leydig), are well developed, but are situated high up on the sides or on the dorsal surface of the head. The pair of nerves supplying them originate from the ventral surface of the brain. FIG. 29. 47 Reproductive System. — The gonads are generally paired, but in some Cladocera they may be united in the middle line. In the Notostraca they are much ramified, but in the other Branchiopoda they are of simple tubular form. Probably all Branchiopoda are of separate sexes. In the Notostraca Bernard has described testi- cular tissue in the ovary, but the occurrence of normal functional hermaphroditism is still unproved. In the Notostraca the genital ducts open on the llth trunk- somite, and this is probably also the case in all the Conchostraca. In the majority of the Anostraca the first two apodous somites, namely, the 12th and 13th of the trunk, are more or less com- pletely fused, forming a " genital segment " on which the genital ducts open. The development indicates that the openings prob- ably belong to the 12th somite. In Polyartemia, however, the leg-bearing somites are 19 in number, and the 20th and 21st somites form the genital segment in the male sex. In the female all the apodous segments are coalesced. In the Cladocera the female genital apertures are lateral or dorsal in position on the posterior apodous division of the body. The male apertures are lateral or ventral and often placed farther back, sometimes quite at the end of the body. The genital openings are generally paired, but the female opening is unpaired in the Anostraca, where the oviducts unite to form a uterine chamber (Fig. 15, A, ut) with groups of gland- cells on its walls. In some Cladocera the male opening is un- paired. The terminal part of the vasa deferentia forms a paired eversible intromittent organ in the Anostraca (Fig. 15, B, p) and in some Cladocera (Sididae). The spermatozoa are immobile and usually spherical. In the Cladocera, however, they present a remarkable variety of form, differing greatly sometimes even in the species of one genus. In the Anostraca the eggs are retained, sometimes till they hatch, in the uterine portion of the oviduct. In the other Branchio- poda they are carried after extrusion either in special receptacles formed by the llth pair of trunk-limbs (Notostraca) or enclosed within the valves of the shell (Conchostraca and Cladocera). In the Conchostraca the egg-masses are attached to certain pairs of specially modified trunk-limbs. In the Cladocera a special brood-chamber is formed between the dorsal surface of the body and the shell, and is closed behind by fleshy folds or prominences. Further, in some Cladocera, if not all, a modification of the hypo- dermis of the dorsal surface takes place for the secretion of a nutritive fluid by which the embryos are nourished within the brood-chamber. In the Cladocera the sexually produced " resting "- or " winter "-eggs are deposited within the cast-off shell, and in many a part of the shell becomes thickened and indurated, and 48 THE CRUSTACEA separates from the rest at ecdysis to form a protective case known as the ephippiuin^ containing the resting-eggs. DEVELOPMENT. Excepting the great majority of the Cladocera and a few Conchostraca (Cyclestheria) in which the development is embryonic, FIG. 30. Larval stages of Apus cancriformis. A, metanauplius, just hatched; B, "second" larval stage; C, " fourth " larval stage, l.antennule; 2, antenna ; 3, mandible ; 4, maxillula ; I-XIII, first thirteen trunk-somites ; f.s, frontal sense-organ ; L, hepatic diverticula ; .•>•, carapace. (After Glaus, from Korschelt and Heider's Embryology.) the Branchiopoda have a free-swimming nauplius or metanauplius stage. Some differences exist even in closely allied forms in the stage of development reached at hatching. In the Notostraca (Fig. 30) and Anostraca the larva is a typical metanauplius with an oval body, showing posteriorly the commencing division of several trunk- somites and sometimes rudiments of their appen- dages. The aritennules are well developed, but simple. The antennae have a movable masticatory process. The corresponding THE BRANCHIOPODA 49 process of the mandible is feebly developed. The paired frontal sense-organs in the form of papillae develop at an early stage, though not present on hatching. The maxillulae and maxillae are generally not marked off till after the succeeding limbs have been differentiated. The " dorsal organ " or nuchal gland is very large in the early stages even in those forms in which it is greatly reduced or absent in the adult. The trunk-somites and their appendages become differentiated in regular order from before backwards. In the Conchostraca the earliest larva has no trace of the trunk - somites. The antennules are greatly reduced and the labrum is very large. The nauplius of Limnetis is remarkable for the broad dorsal shield and for the peculiar cruciform shape of the front of the head. In the Conchostraca and in the Cladocera the shell develops from paired rudiments. The later larval stages of Fio. 31. Metanaupliiis of Li'jitodom, hate-lied fiom a " winter "-egg. c', antennule ; «", antenna ; md, mandibular palp; ol, labrum; p'-p", rudiments of the six pairs of trunk-limbs. (After Sars, from Korsehelt and Heider's Embryology.) the Conchostraca, as, for instance, Estheria, correspond very closely with the adult structure of the Cladocera. In the Cladocera the eggs are usually large and rich in yolk, or, when the reverse is the case, a special provision is made for the nourishment of the developing embryos within the brood-chamber. The egg-membrane, which in the parthenogenetic ("summer") eggs is very thin, is early cast off, and the developing embryos lie free within the brood-chamber. In the sexually produced ("winter" or " resting ") eggs the whole development is gone through within the egg-membrane. A distinct nauplius stage is passed through, and, at least in some cases, is marked by the formation of a cuticle which is cast off later. The rudiments of the maxillulae and maxillae do not appear until after some of the trunk-limbs have already appeared, and the maxillae afterwards become reduced and disappear in the adult. In the embryos of some Cladocera in which only five pairs of trunk-limbs are present in the adult, six pairs of rudiments are formed, but the last pair disappears later. 50 THE CRUSTACEA The young of the Conchostracan Cyclestheria have an embryonic development very similar to that of Cladocera within the shell of the parent. A remarkable exception to the rule of embryonic development among Cladocera is afforded by the aberrant genus Leptodom. The parthenogenetic " summer "-eggs develop in the usual way within the brood-pouch of the parent. The " winter "-eggs, however, hatch out as metanauplii (Fig. 31). The body is unsegmented, but the six pairs of thoracic limbs are already visible as rudiments. The antennules are very short. The antennae, on the other hand, are unusually large, as they are also in the adult. They are without any masticatory process. The mandibles have long unsegmented palps. The compound eyes are not yet developed, but the nauplius eye is present, and persists throughout life in the individuals hatched from " winter "-eggs, while it is absent in those hatched from the " summer "-eggs. In the reduction of the antennules this larva shows some resemblance to that of Estheria. REMARKS ON HABITS, ETC. The great majority of the Branchiopoda inhabit fresh water. A few species of Cladocera, belonging to three genera, occur in the sea, and the Anostracan Artemia is found in salt lakes and brine- pools. The Cladocera form an important part of the plankton of lakes and ponds, and the larger Anostraca, Notostraca, and Con- chostraca occur chiefly in small ponds and rain-water pools. The occurrence, throughout the group, of sexually produced, thick- shelled " resting "-eggs, which can survive desiccation, in addition to the thin -shelled eggs produced by parthenogenesis, probably indicates the very ancient adaptation of the Branchiopoda to a freshwater habitat. No parasitic Branchiopoda are known. The Cladocera are nearly all of microscopic size, and some species which do not exceed 0*25 mm. in length are among the smallest known Arthropoda. Of the other orders, the Notostraca comprise the largest forms, some species of Apus reaching 70 mm. in length. PALAEONTOLOGY. The Conchostraca are well known as fossils, and forms referred to the existing genus Estheria occur as early as the Devonian. The Notostraca are more doubtfully represented by Protocaris from the Lower Cambrian. From the delicacy. of their structure, the Anostraca are less likely to be preserved, and almost the only undoubted example is Branchipodites of the Oligocene. The Cladocera are not certainly known earlier than Post-tertiary deposits. THE BRANCHIOPODA 51 Many palaeozoic fossils formerly classed with the " Phyllopoda " are now referred to the Phyllocarida. AFFINITIES AND CLASSIFICATION. The alliance of the groups included in the Branchiopoda is justified especially by the lobed foliaceous form of the trunk-limbs which they have in common, but the divergences of structure in other respects are greater than is the case in other sub-classes of Crustacea. Thus, while the other sub -classes are more or less strictly nomomeristic, each of the orders of Branchiopoda, and even some of the families and genera, are markedly anomomeristic. This is in agreement with the view that the Branchiopoda are a primitive group which has not attained to the fixity of general structure found in the other sub-classes. Their primitive character is further shown, as has been pointed out, by the general uniformity of the trunk -somites and their appendages, by the presence of gnathobases on all the trunk-limbs, by the " ladder-like " form of the ventral nerve-chain and the post- oral position of the antennal ganglia, and by the tubular heart and its segmentally arranged ostia. The primitive character of the larval development has also been alluded to. It may be mentioned here that, as in other groups of Arthro- poda, the possession by many Branchiopoda of a large number of somites can hardly be regarded as proof of their primitive position. In the Notostraca, the fact that the posterior pairs of appendages exceed in number the somites which carry them, shows that secondary changes, whether by coalescence of somites or, more probably, by multiplication of appendages, have taken place. A further argument in favour of a possible increase in number of somites is afforded by a consideration of the aberrant Notostracan genus Polyartemia (Fig. 32). Apart from the Cladocera, which the abbreviation of the trunk excludes from the comparison, Polyartemia forms the only exception to the rule that the genital apertures of the Branchiopoda are situated, approximately, in the region of the twelfth trunk-somite. Now, the close resemblance in all other respects between Polyartemia and the other Anostraca strongly suggests that the nineteen somites interposed between the head and the genital somite in that genus correspond, as a whole, to the eleven somites which occupy the same position in the other Anostraca; and the agreement of the latter in this respect with most of the other Branchiopoda seems to indicate that the smaller number of somites is here the more primitive, the larger the more specialised condition. If this be so, a similar multiplication of somites in the post-genital region of Notostraca may well account for the exceptionally large number found in that order. THE CRUSTACEA The classification of the Branchiopoda commonly adopted differs from that given below in grouping together the Anostraca, Notostraca, and Conchostraca as a single order, Phyllopoda (Latreille, 1802),1 distinguished from the Cladocera chiefly by the greater number of somites and appendages and by the prevalence of metamorphosis in development. The groups of the Phyllopoda, FIG. 32. Polyartemia forcipatu, one of the Anostraca. x 5. (After Sars.) A, female, dorsal view ; B, male, lateral view, a', antennules ; a", antennae, very small in the female but greatly enlarged and three-branched in the male; f.a, "frontal appendage"; p, paired penes of the male. There are nineteen limb-bearing trunk-somites, followed, in the male, by six limbless somites (besides the telson), of which the first and second are partly coalesced to form the genital segment. In the female the limbless region of the trunk is unsegmented. however, differ among themselves in characters which are at least as important as those separating the Cladocera from the Conchostraca, and it seems desirable to recognise this by giving them the rank of orders. 1 Unfortunately some writers, following Glaus, have transposed the names Branchiopoda and Phyllopoda, applying the latter to the sub-class and the former to one of its divisions, but this use is not sanctioned either by priority or by universal custom. THE BRANCHIOPODA 53 SUB-CLASS BRANCHIOPODA. ORDER 1. Anostraca. Carapace absent ; paired eyes pedunculate ; antennae prehensile in male, reduced in female ; trunk-limbs, 11 or 19 pairs, none post-genital ; furcal rami unsegmented ; development with metamorphosis. Family POLYARTEMIIDAK. Polyartemia, Fischer (Fig. 32). Family BRAXCHIPODIDAE. Brancldpus, Schaffer ; Ghirocephalus, Prevost (Fig. 21); Artemia, Leach; Ifranchinecta, Verrill (Fig. 15). Family THAMNOCEPHALIDAE. Thamnocephalus, Packard. ORDER 2. Notostraca. Carapace forming a dorsal shield ; paired eyes sessile ; antennae vestigial ; trunk-limbs, 40 to 63 pairs, of which 29 to 52 are post-genital ; furcal rami multiarticulate ; development with metamorphosis. Family APODIDAE. Apus, Latreille ; Lepidurus, Leach (Fig. 16). ORDER 3. Conchostraca. Carapace bivalved, enclosing head and body ; paired eyes sessile, coalescent ; antennae biramous, natatory; trunk limbs, 10 to 27 pairs, of which 0 to 16 are post-genital ; furcal rami claw-like ; development usually with metamorphosis. Family LIMNADIIDAE. Limnadia, Brongniart ; Estheria, Ritppel (Fig. 17) ; Cyclestheria, G. O. Sars. Family LIMNETIDAE. Limnetis, Love'n. ORDER 4. Cladocera. Carapace bivalved, generally enclosing body but leaving head free, sometimes reduced and serving only as a brood-sac ; paired eyes sessile, coalesced ; antennae biramous (except in the female Holopedium), natatory ; trunk-limbs, 4 to 6 pairs, none of which are post-genital ; furcal rami claw-like ; development embryonic, rarely with metamorphosis. SUB-ORDER 1. CALYPTOMERA. Carapace completely enclosing body and limbs. Tribe 1. CTENOPODA. Six pairs of trunk -limbs all similar and foliaceous. Family SIDIDAE. Sida, Straus ; Latona, Straus ; Penilia, Dana. Family HOLOPEDIIDAE. Holopedium, Zaddach. Tribe 2. ANOMOPODA. Five or six pairs of trunk-limbs, first two pairs more or less prehensile. Family DAPHNIIDAE. Daphnia, O. F. Miiller (Fig. 18); Moina, Paird ; Simocephalus, Schodler (Simosa, Norman). Family BOSMINIDAE. ttosmina, Baird. Family LYNCODAPHNIIDAE. Ilyocryptus, G. O. Sars ; Macrothrix, Baird. Family LYXCEIDAE. Lynceus, 0. F. Mu'ller ; Chydorus, Leach ; Eurycemis, Baird ; Pleuroxus, Baird (Fig. 26) ; Monospilus, G. 0. Sars. 54 THE CRUSTACEA SUB-ORDER 2. GTMNOMERA. Carapace not enclosing body and limbs. Tribe 1. ONYCHOPODA. Four pairs of trunk -limbs, more or less compressed. Family POLYPHEMIDAE. Polyphemus, 0. F. Miiller ; Bythotrephes, Ley dig ; Evadne, Loven • Podon, Lilljeborg. Tribe 2. HAPLOPODA. Six pairs of trunk-limbs, completely pediform. Family LEPTODORIDAE. Leptodora, Lilljeborg (Fig. 19). LlTERATURK. 1. Baird, W. The Natural History of the British Entomostraca. Ray Society, London, 1850. 2. Bernard, H. M. The Apoclidae : A Morphological Study. London, 1892. 3. Claus, C. Zur Kenntniss des Baues und der Entwickelung von Branchipus stagnalis und Apus cancriformis. Abhandl. kgl. Ges. Wiss. Gottingen, xviii. ; Phys. Cl. pp. 93-140, pis. i.-viii., 1873. , 4. Zur Kenntniss der Organisation und des feineren Baues der Daphniden und venvandter Cladoceren. Zeitschr. \viss. Zool. xxvii. pp. 362-402, pis. xxv.-xxviii., 1876. 5. Untersuchungen iiber die Organisation und Entwickelung von Branchipus und Artemia. Arb. zool. Inst. Wien, vi. pp. 267-370, 12 pis., 1886. , 6. Cunnington, W. A. Studien an einer Daphnide, Simocephalus sima. Jena. Zeitschr. xxxvii. pp. 447-f>20, pis. xxiv.-xxvi., 1903. 7. Ekman, S. Beitriige zur Kenntnis der Phyllopodenfamilie Polyartemiidae. Bihang Svenska Vet.-Akad. Handl. xxviii., Afd. iv., No. 11, 38 pp. 4 pis., 1902. 8. Grobben, C. Die Entwickelungsgeschichte der Moina rectirostris. Arb. zool. Inst. Wien, ii. pp. 203-268, 7 pis., 1879. 9. Lankester, E. Ray. Observations and Reflections on the Appendages and on the Nervous System of Apus cancriformis. Quart. J. Micr. Sci. xxi. pp. 343-376, pi. xx., 1881. 10. Leydig, F. Naturgeschichte der Daphniden, pp. iv. + 252, 10 pis. 4to. Tubingen, 1860. 11. Lilljeborg, W. Cladocera Sueciae. Nova Acta reg. Soc. Sci. Upsala, (3) xix. vi + 701 pp. 87 pis., 1900. 12. Packard, A. S. A Monograph of North American Phyllopod Crustacea. Twelfth Ann. Rep. U.S. Geol. Survey, pp. 295-392, 39 pis. and map, 1883. 13. Pclsenecr, P. Observations on the Nervous System of Apus. Quart. J. Micr. Sci. xxv. pp. 433-444, pi. xxx., 1885. 14. Richard, J. Revision des Cladoceres. Ann. Sci. Nat. Zool. (7) xviii. pp. 279-389, pis. xv.-xvi., 1895 ; and op. cit. (8) ii. pp. 187-363, pis. xx.-xxv., 1896. 15. Sars, G. 0. Norges Ferskvandskrebsdyr. Cladocera Ctenopoda. 71 pp. 4 pis. 4to. Christiania, 1865. 16. Om en dimorph Udvikling samt Generationsvexel hos Leptodora. Forh. Vidensk. Selsk. Christiania, 1873, pp. 1-15, 1 pi. THE BRANCHIOPODA 55 17. Sars, G. 0. On Cyclesthcria hislopi (Baird). Forli. Vidensk. Selsk. Christiania, 1887, No. 1, pp. 1-65, 8 pis. 18. Fauna Norvegiae. Vol. I. Phyllocarida and Phyllopoda. 4to, pp. vi + 140, 20 pis. Christiania, 1896. 19. Simon, E. Etude sur les Crustaces du sous-ordre des Phyllopodes. Ann. Soc. ent. France, (6) vi. pp. 393-460, pis. v.-vii., 1886. 20. Weismann, A. Ueber Ban und Lebenserscheinungen von Leptodora hyalina. Zeitschr. wiss. Zool. xxiv. pp. 349-418, pis. xxxiii.-xxxviii. , 1874. 21. — • Beitrage zur Naturgeschichte der Daphniden. Zeitschr. \viss. Zool. xxvii.-xxxiii., 1876-1879. 22. Zaddach, E. G. De Apodis cancriformis, Schaeff., anatome et historia evolu- tionis. Dissertatio inaug. vi + 72 pp. 4 pis. 4to. Bonnae, 1841. THE OSTRACODA SUB-CLASS OSTRACODA, Latreille (1802). Order 1. Myodocopa. „ 2. Cladocopa. „ 3. Podocopa. „ 4. Platycopa. Definition. — Crustacea in which the carapace forms a bivalve shell ; the trunk is indistinctly segmented, its posterior part is without limbs and ends in a caudal furca; the antennules and antennae are large and used for locomotion ; the mandibles have a palp ; not more than four pairs of limbs are distinctly developed behind the mandibles and they vary much in form ; the genital apertures are behind the last pair of limbs ; the paired eyes are sometimes present ; development with metamorphosis, the young hatched in the form of a modified nauplius. Historical. — Although various species of Ostracoda were seen and figured by the early microscopists, the scientific study of the group may be said to begin with 0. F. Miiller's Entomostraca (1785). The bivalve shell caused these animals to be associated in most of the earlier systems of classification with the Cladocera, from which they were separated by Milne-Edwards (1840). Almost the first to give an account of the internal anatomy was Zenker. The knowledge of their development is based chiefly on the works of Glaus. The systematic and faunistic works of G. 0. Sars, Brady, and Norman are among the most important ; while in recent years G. W. Muller has made noteworthy advances towards a precise and detailed knowledge of morphology and classification. MORPHOLOGY. The bivalved shell which completely encloses the body and limbs (Fig. 33) is usually elliptical in outline as seen from the side, often (Podocopa) somewhat flattened ventrally, but it may be nearly 56 THE OSTRACODA 57 globular or, in some Halocypridae, greatly elongated. The two valves are probably always more or less unsymmetrical, especially along the ventral margin. On the dorsal side they are connected by a hinge-joint, which may be merely an uncalcified strip of the integument (Halocypridae), or may be strengthened by inter- locking ridges and teeth. The most complex hinges are found among the Cytheridae. The almost globular shell of Gigantocypris is exceptional in that the free edges of the valves occupy only about one-third of the circumference. The outer surface of the valves is seldom quite smooth. It may be beset with setae or pitted or sculptured (Fig. 33, C), and is sometimes produced into wing-like processes. When the shell is strongly calcified and opaque more transparent spots may mark the position of the eyes (Podocopa, Fig. 33, e). The attachment of the adductor muscle is usually visible externally as a group of FIG. 33. Lateral view of shell of A, Philomedesbrenda (Myodocopa), x 8 ; B, Cyprisfuscata (Podocopa), x 19 ; C, Cythereis orn.ata (Podocopa), x 33. a, attachment of adductor muscles ; e, median «ye ; n, antennal notch. (A after Brady and Norman ; B and C after G. W. M tiller.) spots, the arrangement of which affords characters of systematic importance (a). When the edges of the valves are brought together they usually fit closely, but in some cases openings are left. The most important of these is the " antennal notch " found in most Myodocopa, and permitting the protrusion of the antennae (Fig. 33, A, n). Unicellular glands opening by pores on the surface of the shell are frequent in Myodocopa, less so in Podocopa. The fold which marks off the shell from the body on each side does not extend very far towards the dorsal surface, and certain of the viscera may extend, as in some Cirripedia, into the cavity between the outer and inner integument of each valve. This is the case with the hepatic caeca in some Cyprididae (Pontocyprinae), and more commonly with the reproductive organs, especially the ovaries, which in the Cyprididae are completely and in the Cytheridae partly lodged in the cavity of the shell-fold. In the Cypridinidae a network of blood-channels traverses each valve, radiating outwards from the muscle-impression. The posterior part of the body, which is free within the shell, THE CRUSTACEA shows, at most, only indistinct traces of segmentation (Cytlierella). The terminal part is curved ventrally, and ends in a caudal furca. In the Myodocopa the furcal rami are flattened triangular plates, with short, stout spines or teeth on the hinder margins (Fig. 37, E). In the Podocopa they are slender or styliform, bearing setae, or may be much reduced, as in the Cytheridae. A. FIG. 34. A, antennuleof CyprMinameditermnea, $. B, antenna of Conchoccia magnet-, ?. C, antenna of Cytlierella sordida. D, antenna of Duririnula stevensoni. (After G. W. Muller.) en, endopo- dite, ex, exopodite, according to Miiller'.s interpretation. Appendages. — The antennules may function as sensory organs, or they may be used for swimming, creeping, or digging in sand, and their form and armature of setae or spines are correspondingly varied. Each is composed of eight segments in Cypridina (Fig. 34, A) and in Pontocypis, and this seems to be the typical number from which the others have been derived by reduction. In the Halocypridae the segments are reduced to two or one. Sometimes THE OSTRACODA 59 the antennules are modified in the male sex as organs for clasping the female. This appears to be the function of certain highly peculiar setae provided with sucker-like organs found in the male of Cypridina. With very few exceptions the antenna appears to be the chief organ of locomotion, whether swimming or creeping. In the Myodocopa it is biramous, and the outer branch is more strongly developed than the inner, with eight or nine segments carrying natatory setae. The inner branch has not more than three segments, and is often modified as a clasping organ in the male. It is characteristic of the Myodocopa that the single segment of the peduncle is greatly expanded and occupied by large muscles (Fig. 34, B). In Polycope (Cladocopa) the two rami do not differ greatly in length. In the Podocopa only one ramus of three or four segments is well developed, and, according to the view of G. W. Miiller, it is the inner ramus, the outer being represented by a small process tipped with setae or being altogether absent (Fig. 34, D). In Cytherella (Platycopa) the antenna differs from that of all other Ostracoda in having the peduncle divided into two segments and bent or geniculate between them (Fig. 34, C). The two rami are well developed — the inner, of three segments, resembling that of the Podocopa in the disposition of the sensory setae. The outer ramus has two segments. In the Cytheridae a large seta, bent at the tip, occupies the place of the vestigial outer ramus present in the other families of Podocopa. It serves as duct to a large unicellular gland. This was formerly regarded as a poison-apparatus, but, according to Miiller, it is a spinning organ, by means of which the Ostracod covers with a network of fine threads the surface on which it is creeping, in order to obtain a secure foothold. The labrum is usually well developed, and is especially large in Cypridinidae, where it contains a group of gland-cells. The lower lip is usually small, and only rarely shows traces of a division into paragnatha. The mandible is characterised by the large size of the palp, which is often biramous and is sometimes pediform or variously modified. The gnathobase may be reduced to a small setose lobe (Cypridina), or to a curved serrated process extending into the gullet (Asterope). In Sarsiella it seems to be entirely absent, and the strong curved spines with which the tip of the palp is armed probably serve for seizing food and passing it into the mouth (Fig. 35, C). In the other families the gnathobase is usually well developed, and armed with spines and teeth. The palp usually consists of four segments, but the number is often reduced. In the Cypridinidae it is long and pediform, and is used in creeping. In Sarsiella, as already noted, it serves for the prehension of food, 6o THE CRUSTACEA and in the male of that genus it is apparently a clasping organ. In the Halocypridae (Fig. 35, B) and Polycopidae (Fig. 35, A) the first segment of the palp sends inwards a process tipped with spines, which lies alongside the gnathobase and assists in mastication. In CythereUa and some other genera the inner surface of the first and second segments of the palp bears a comb-like series FIG. 35. A ..mandible of Pdycope frequens. B, mandible of Conchoecia magna. C, mandible of Sar- sldlnlevis, 9- D, maxillula of Philomedfs interpwncta. E, maxillula of Polycopsis serrata. F, maxillula of Maarocypris suednea. (After G. W. Miiller.) ex, exopodite. of long setae. The exopodite is represented in most Myodocopa by a small unsegmented appendage. In the Cyprididae and allied families it is a flattened plate with radiating fan-like setae or setiform processes, the so-called branchial appendage. In Para- doxostoma and some allied genera the mouth-parts are modified for piercing and sucking. The mandible is styliform, with a slender palp, and is enclosed in a conical beak formed by the labrum and hypostoma. THE OSTRACODA 61 The structure of the maxillula varies very much in the different genera. The most primitive form is probably that found in the Cladocopa, where it consists of a protopodite of two segments, a three-segmented endopodite, and a small exopodite of one or two segments (Fig. 35, E). The segments of the protopodite are each produced inwards into a slight masticatory lobe. In the Halocypridae and most Cypridinidae (Fig. 35, D) the structure is similar, save that the exopodite is wanting. In most Podocopa the masticatory lobes are greatly produced, and the distal one is FIG. 36. A, third post-oral appendage (BO-called " second maxilla ") of Gyp rid inn mcditerraneu (young). B, third post-oral appendage (so-called "first leg") of Cythereis convexa. C, third post-oral appendage (so-called " maxilliped ") of Macroeypris suixinea, male. U, the same, female. (After (i. \V. Mtiller.) IM, endopodite ; ex, exoixxlite. divided into two. The three segments of the endopodite may be distinct or may fuse into one. A large "branchial appendage," with radiating setae, may perhaps represent the exopodite (Fig. 35, F). The third post-oral limb is of very diverse form, and has received different names in the various genera. Thus in the Cypridinidae it has been called the " second maxilla," in the Halocypridae and Cyprididac the " maxilliped," and in the Cy theridae and other families the " first leg." In the Cypridinidae it is quite maxilli- fonn, and in the adult shows hardly any trace of segmentation. In young stages, however, Miiller finds that it consists of six more or less distinct segments, of which two are assigned to the protopodite and the remainder to the endopodite (Fig. 36, A). A "branchial appendage" is present and is probably to be regarded 62 THE CRUSTACEA as the epipodite. In the other families the appendage is more leg- like. The " branchial plate " is commonly present, and sometimes an unsegmented appendage which is regarded as the exopodite. In many Cyprididae the limb forms, in the male, a strong clasp- ing organ (Fig. 36, C), which may be unsymmetrically developed on the two sides. In Cytherella it has, in the male, much the same structure as in the Cyprididae, while in the female it is entirely absent. Finally, in the Bairdiidae and Cytheridae it is quite pediform (Fig. 36, B), with a strong terminal claw, and with or without a branchial plate. The completely pediform character of this appendage in many Ostracoda suggests a doubt as to its homology with the maxilla of other Crustacea. This doubt is further strengthened by Miiller's statement that the limb appears to belong to the thoracic rather than to the cephalic division of the body. More important still is the fact that in the course of development a pause in the successive appearance of the limbs occurs before this limb is added to the series. On these grounds there seems to be considerable proba- bility in Miiller's view that the maxilla has been entirely lost in the Ostracoda and 'that the appendage which occupies its place is to be regarded as homologous with the first thoracic appendage of other Crustacea. In most Myodocopa the fourth post-oral limb is a maxilliform lobed plate, distinctly segmented only in Cypridina (Fig. 37, A). In Sarsiella and Asterope it is laminar and not lobed. In the Polycopidae it is absent. In the other families it is more or less leg-like (Fig. 37, B), with or without a branchial plate. In some Cytherellidae it forms a clasping organ in the male, while in the female it is reduced to the branchial plate. In the Cytheridae, Bairdiidae, and Darwinulidae the fifth post-oral limb is pediform, with a strong terminal claw, and is used for creeping. In the Cyprididae (Fig. 37, D) it appears less adapted for locomotion, and is probably chiefly used as a " cleaning foot " for cleaning the other appendages and the inside of the shell. This is its only function in the Myodocopa (except Halocypridae), where it has a remarkable structure, being long and cirriform, divided into numerous segments moved by two muscles running along its whole length (Fig. 37, C). It is set high up on the side of the body, and the terminal part is armed with setae and chitinous teeth. In the Halocypridae the limb is greatly reduced and in the Cladocopa it is altogether absent. A peculiar brush-like appendage found on the side of the body in the males of some Ostracods has been regarded as a vestigial sixth post-oral limb. It is found in a few Cyprididae and in Cytherella behind the last leg, and its position near the first leg (third post-oral) in the Bairdiidae and Cytheridae is explained as the result of a THE OSTRACODA secondary shifting of position. A similar organ has been found in one species of Ci/pridina. It is not unlikely that at least one pair of limbs is involved in the composition of the penes of male Ostracods. Alimentary System. — The oesophagus is narrow and has muscular walls. Its upper or posterior end projects into the capacious mid- FIG. 37. A, fourth post-oral ai>i>t'ii'lni;i> of i.'itpridina meditcrranea. 13. tlif same of Mucroeypris sitccinca. C, fifth post-oral appendage of Cypridina nudttenvnea. D, the same of Macrocypria succinea. E, one ramus of tlu- caudal furca'of Ci/prUlitut xqinimnxn. seen from the side. (After U. W. M tiller.) gut, and in the Podocopa is armed internally with chitinous ridges and teeth. In the Bairdiidae this apparatus is most fully developed and forms an efficient "gastric mill," moved by extrinsic and intrinsic muscles, for the trituration of the food. The mid-gut in the Podocopa is divided into two parts by a con- striction. Hepatic caeca may be absent as in Cgpridina, numerous and small as in Halocypris, or large and reduced to a single pair which may extend into the shell-cavity, as in Cyprididae. 64 THE CRUSTACEA The short rectum opens on the ventral side of the furca in Myodocopa, but on the dorsal side in the Cyprididae and allied families. Circulatory System. — A heart is present only in the Myodocopa (Fig. 38, H). It is placed near the dorsal surface just above the mandibles, and has one pair of lateral ostia and an opening in front through which the blood is expelled. No definite vessels exist, but a network of blood-channels is found in the shell of the Cypridinidae. The blood is coloured red in Krithe. Respiratory System. — In the majority of Ostracoda the respiratory function is probably discharged by the general surface of the body and limbs and by the inner surface of the shell. The so-called " branchial appendages " found on the various limbs serve to keep up a current of water within the shell. Only in the genus Asterope, and less distinctly in one or two species of Gypridina, are definite branchiae found. These are lamellar appendages attached to the dorsal surface of the posterior part of the body. Seven pairs are present in Asterope. Excretory System. — Little is known regarding the excretory system of the Ostracoda. " Segmental " organs in the form of small sacs lying at the bases of the ambulatory legs are described in Paradoxostoma and Bairdia. It is doubtful whether they open to the exterior. In Cypris a gland lying in the labrum and opening on the first segment of the antenna and a second gland opening on the basal segment of the third post-oral appendage (maxilla ?) have been described. Glands. — As already stated, dermal glands are frequently found in the valves of the shell. In some Cypridinidae their secretion serves to agglutinate the sand in which these animals burrow, and in some cases to form definite tubular dwellings. In the Halo- cypridae the glands are most numerous near the edge of the anterior part of the shell, and it is stated that their mucus-like secretion serves to entangle food-particles and is then swallowed. Specially large glands opening near the margins of the valves anteriorly are found also in some other Ostracoda, and, from their position, suggest a comparison with the " f ronto-lateral glands" similarly situated in the " Cypris-larva " of Cirripedes. Glands are ^sometimes found in the labrum (Cypridinidae and Cyprididae), and are believed, in Pyrocypris, to produce a luminous secretion. The spinning-glands of Cytheridae have already been mentioned. Musculature. — The adductor muscle (Fig. 38, S.M) runs between the valves of the shell on the ventral side of the alimentary canal. In some Ostracoda, especially in Cyprididae, a tendinous plate is found below the anterior part of the alimentary canal, serving for the attachment of the mandibular and maxillary muscles. It is worthy of note that the adductor muscle does not share in its formation. THE OSTRACODA Nervous System. — The supraoesophageal ganglion is largest in the Cypridinidae in correlation with the presence of paired eyes in that family. The ventral nerve-chain is represented by a single p FIG. 38. Cypridina mni:ti i-rnnra. Female (above) and male (below). One valve of shell removed to show appendages and internal organs. A", antenna (the antennule is just above) ; F' (below), fourth post-oral limb ; F' (above), tilth post -oral limb ; Fii, caudal furca ; G, brain ; H, heart ; M, mid-gut: Mi If, mandibular palp; MX', maxillula ; MX", third post-oral limb (maxilla); 0, pairrd eye ; ()', median eye ; P, penis ; S.M, adductor muscle of shell ; ,Ste, frontal tentacle ; T, testis. (From Claus's 66 THE CRUSTACEA undivided mass in the Cyprididae, but in the Myodocopa it may be more or less completely divided into two or three (Halocypridae) ganglionic masses. Sense-Organs. — Paired compound eyes are present only in the Myodocopa (excluding Halocypridae), where they project slightly from the surface of the head within the shell (Fig. 38, 0). The number of ommatidia in each varies from four in Sarsiella to about fifty in some species of Asteropc. The crystalline body in each is bipartite, and the rhabdom is longitudinally ribbed. There are no facets or cuticular lenses on the overlying cuticle, nor "eye-spots" on the shell. In Philomedes the eyes, though well developed in the male, are rudimentary in the female. The nauplius eye is present in the majority of the Osti^acods (Fig. 38, 0'), and consists of the usual three pigment-cups contain- ing each from 3 to about 100 retinal rods. In some Podocopa the three parts are widely separated from each other. A uni- cellular lens, or "crystalline body," is found in many Podocopa (Fig. 12, p. 18), and transparent "eye-spots" on the shell are common. In the Myodocopa the nauplius-eye lies at the base of a rod-like median process from the front of the head, the " frontal tentacle " (Fig. 38, Stz). Sometimes (Halocypridae) the distal portion of this tentacle is segmented off and is known as the "capitulum." Olfactory filaments are present, commonly in greater numbers in the male than in the female, on antennules and antennae, and also on the "brush-like appendage." Reproductive System : Female. — The ovaries are paired, and are generally lodged in the posterior part of the body. In the Cyprididae, and partly in the Cytheridae, they lie in the cavities of the shell-valves. The oviducts usually open separately behind the last pair of legs. In the Halocypridae the two oviducts unite shortly after leaving the ovaries, and the unpaired opening is situated on the left side just in front of the caudal furca. A receptaculum seminis, paired except in the Halocypridae, is always present. As a rule, it opens to the exterior by a special copulatory pore, and communicates internally with the oviduct, but sometimes only the external opening appears to be present, while in the Cyprididae the opening into the oviduct serves both for entrance and exit. Male. — The testes are simple in the Myodocopa (Fig. 38, 7'), four-lobed in most of the other families. In the Cypridinidae the two vasa deferential unite to open by a single median pore lying between the paired penes, which in this family alone are not traversed by the ducts (Fig. 38, P). In the Halocypridae the penis is unpaired and lies to the right of the middle line. The vasa deferentia unite just before entering it. In the other families THE OSTRACODA 67 the vasa deferentia are probably always united by a transverse duct shortly after leaving the testes. They are often of great length and are coiled in a very complex fashion, differing in the different groups. The distal part forms a ductus ejaculatmius of complicated structure, with muscular walls. The paired penes are traversed by the vasa deferentia, the terminal portion forming a protrusible copulatory tube, and the penis is often armed with chitinous hooks moved by special muscles. The spermatozoa are sometimes spherical (Asterope), more commonly filiform. In the Cyprididae they are of a size which relatively, if not absolutely, is unique in the animal kingdom. In /'inifiii'i//!/'/* Hionstrosa the total length of the animal is *6 mm., while the spermatozoa are 5'0-7'0 mm. long. DEVELOPMENT. The eggs are carried within the valves of the shell dorsally to the body of the animal in the Cypridinidae, Cytherellidae, some Cytheridae, and in the freshwater genus Darwinula, and in a few cases the young are carried for some time after hatching. In other cases the eggs are deposited on water-plants or shed free in the water. The course of development is best known in the Cyprididae. On hatching, the larva is already enclosed Avithin a bivalve shell, but otherwise corresponds in structure to a nauplius (Fig. 39). Three pairs of appendages are present, antennules, antennae, arid mandibles, the last two, however, not distinctly biramous. In the Cytheridae the mandible at this stage is rudimentary. The ss??\***&>^3^- -^ \ •»• . . ° v , ij j fff. Swrt'C^^^-AL--' » remaining limbs are added at suc- cessive moults in regular order from before backwards. According to j S Miiller, it is not the case that, as stated by Glaus, the fourth post-oral limbs appear before the third. There 30 is, however, a marked pause in the H«wH»rt»ofCtoHii «', antenna ; development before the appearance »•", antenna; ad, adductor muscle ofcai ii- .,,., , . J l , , . pace ; e, nauplius eye ; mil, mandible. or the third post-oral pair, and this, (Attcr ciaus.) as already pointed out, supports the view that a pair of appendages corresponding to the maxillae of other Crustacea is missing. REMARKS ON HABITS, ETC. The Ostracoda are abundant both in fresh waters and in the sea, generally burrowing in mud or creeping among weeds. The marine Halocypridae belong to the plankton. None are definitely 68 THE CRUSTACEA known to be parasitic, but one species found in the gill-chambers of Crayfish in North America may be so. The majority are minute, very many not exceeding 0'5 mm. in length. The Myodocopa are usually larger than the Podocopa, many being 1 or 2 mm. long. The largest known Ostracod, a species of Gigantocypris, is 23 mm. in length. PALAEONTOLOGY. . Fossil Ostracoda are abundant in all geological formations from the oldest to the most recent, but, with hardly an exception, only the shell is preserved and the affinities of the numerous genera which have been established remain quite uncertain. Most of the Ostracoda from Cambrian rocks belong to the genera Primitia, Jones, and Leperditia, Roualt, while Beyrichia, M'Coy, is common in the Silurian. Species referred to the recent genus Cypridina occur in the Carboniferous, and Gythere and other recent genera are stated to make their appearance in the Permian. AFFINITIES AND CLASSIFICATION. In the bivalved form of the shell, the ventrally flexed posterior part of the body, the form of the caudal furca (in the Myodocopa), and the biramous natatory antennae (especially of the Cladocopa and Platycopa), the Ostracoda present important resemblances to the Conchostraca and Cladocera among the Branchiopoda. There might, indeed, be little difficulty in regarding them as derived from the Conchostraca, were it not for the presence of the mandibular palp, which is absent from the Branchiopoda. The possession of this appendage and its biramous form in many cases point to the origin of the Ostracoda as an independent branch of the primitive Crustacean stock. There appears to be no reason, however, to suggest, as Glaus has done, that the small number of appendages is a primitive character. In the possession of paired eyes and a heart, the Cypridinidae are more primitive than the other Ostracoda, although in the form of the antennae and in some other points they are probably more specialised than the Cladocopa and Platycopa. The classification given below is that of Sars, as modified by Brady and Norman. In the more recent arrangement of G. "NY. Miiller the Cladocopa are included in the Myodocopa and the Platycopa in the Podocopa, and the number of families is considerably reduced. SUB-CLASS OSTRACODA. ORDER 1. Myodocopa. Shell generally with antennal notch ; antennae with massive basal segment, generally biramous, exopodite generally with eight or nine THE OSTRACODA 69 segments, endopodite minute and generally prehensile in male ; five pairs of post-oral limbs ; caudal furca with lamellar rami armed with spines. Family HALOCYPRIDAE (CONCHOECIIDAE). Halocypris, Dana ; Conchoecia, Dana. Family ASTEROPIDAE. Asterope, Philippi ( = Gylindro- leberis, Brady). Family CYPRIDINIDAE. Cypridina, H. Milne-Edwards ; Philomedes, Lilljeborg (Fig. 33, A) ; Pyrocypris, G. W. Mtiller ; Giganto- cypris, G, W. Mu'ller. Family RUTIDERMATIDAE. Rutiderma, Brady and Norman. Family SAHSIELLIDAE. Sarsiella, Norman. ORDER 2. Cladocopa. Shell without antennal notch ; antennae biramous, both rami well developed and natatory ; only three pairs of post-oral limbs ; other characters as in Myodocopa. Family POLYCOPIDAE. Polycope, G. 0. Sars ; Polycopsis, G. W. Miiller. ORDER 3. Podocopa. Shell without antennal notch ; antennae with basal segment not enlarged, one ram us (exopodite, Miiller) vestigial or absent, the other of not more than four segments ; five pairs of post-oral limbs ; caudal furca with styliform or vestigial rami. Family PARADOXOSTOMATIDAE. Paradoxostoma, Fischer. Family CYTHERIDAE. Cythere, O. F. Miiller; Cythereis, G. O. Sars (Fig. 33, C) ; Krithe, Brady. Family DARWINULIDAE. Darwinula, Brady and Robertson. Family BAIRDIIDAE. Bairdia, M'Coy. Family CYPRIDIDAE. Cypris, 0. F. Miiller (Fig. 33, B) ; Pontocypris, G. 0. Sars. Family CYPRIDOPSIDAE. Cypridopsis, Brady. ORDER 4. Platycopa. Antennae biramous, protopodite of two segments, endopodite of three, and exopodite of two segments ; only four pairs of post-oral limbs ; other characters as in Podocopa. Family CYTHERELLIDAE. Cytherella, Bosquet. LITERATURE. 1. Brady, G. S. A Monograph of the Recent British Ostracoda. Trans. Linn. Soc. xxvi. pp. 353-495, pis. xxiii.-xli., 1868. •1. - Report on the Ostracoda. Rep. Voy. "Challenger," Zool. i. 184 pp. 44 pis., 1880. 3. Brculy, G. S. , and Norman, A. M. A Monograph of the Marine and Freshwater Ostracoda of the North Atlantic and of North-Western Europe. Trans. Roy. Dublin Soc. (2) iv. pp. 63-270, pis. viii.-xxiii., 1889; and v. pp. 621-746, pis. l.-lxviii., 1896. 4. Glaus, C. Beitr. zur Kennt. der Ostracoden. 1. Eiihyicklnngsgeschichte von Cypris. Schr. Ges. Naturvviss. Marburg, ix. pp. 151-166, 2 pis., [1868] 1872. yo THE CRUSTACEA 5. Clans, C. Schriften zoologisehen Inlialts. 1. Die Familie der Halocypriden, pp. 1-16, pis. i.-iii. 4to. Wien, 1874. 6. Die Halocypridea des atlantisclien Oceans und Mittelmeeres. iv + 81 pp. 26 pis. 4to. Wien, 1891. 7. - Beitr. zur Kennt. der Siisswasser Ostraeoden. Arl>. zool. Inst. Wien, / x. pp. 147-216, 12 pis., 1893. 8. ^fuller, G. W. Die Ostracoden. Fauna und Flora des Golfes von Xeapel, Monogr. xxi., 1894. 9. Deutschlands Siisswasser Ostracoden. Zoologica, xii. Heft 30, 112 pp. 21 pis., 1900. 10. San, G. 0. Oversigt af Norges marine Ostracoder. Forh. Vid.-Sel.sk. Christiania, 1865, pp. 1-130, [1865] 1866. 11. Zenker, W. Monographic der Ostracoden. Ai-ch. Naturgesch. xx. pp. 1-87, pis. i.-vi., 1854. CHAPTER IV THE COPEPODA SUB-CLASS COPEPODA, H. Milne-Edwards (1830). Order 1. Eucopepoda. Sub-Order 1. Gymnoplea. Tribe 1. Amphaskandria. „ 2. Heterarthrandria. Sub-Order 2. Podoplea. Tribe 1. Isokerandria. „ 2. Ampharthrandria. Order 2. Branchiura. Definition. — Crustacea in which there is no distinct shell-fold, though the dorsal shield of the cephalon may be enlarged by the coalescence of one or more of the trunk-somites ; there are typically nine free trunk-somites (besides the telson), the last four without appendages ; the telson bears a caudal furca ; the antennules and antennae are generally large, the latter sometimes biramous, and both may be used for locomotion or for prehension ; the mandibles may have a palp, sometimes biramous ; there are typically six pairs of trunk -limbs, the last five biramous and natatory, but the sixth pair are frequently reduced or absent ; the genital apertures are on the seventh trunk -somite (except in Branchiura) ; the paired eyes are absent (except in Branchiura) ; development with metamorphosis, the young generally hatched in nauplius stage. (Most of these characters are subject to modifica- tion in parasitic forms.) Historical. — The relatively large size attained by some of the parasitic Copepoda renders it probable that they were the first to attract attention, and it is possible, though not certain, that they are referred to in certain passages of Aristotle. The minute free- living forms remained unnoticed till some of the microscopists of the seventeenth century — notably Leeuwenhoek — described and figured the common freshwater Cyclops. It was long, however, 71 72 THE CRUSTACEA before the affinity of the two groups was recognised. Linnaeus included some free-living Copepods in his genus Monoculus, among the Crustacea, but some of the parasites were placed under " Vermes mollusca " in the genus Lernaea, and others among the " Zoophyta " in the genus Pennatula. O. F. M tiller's great work on the " Entomostraca " marks the beginning of a new epoch in the study of the group. The genera Cyclops, Caliyus, and Argulus were established by Mtiller, besides two others based on larval stages, one of which, the Nauplius, still bears the name given to it by Mtiller. Jurine's Histoire des Monocles (1820) is noteworthy, among much else, for the description of the larval metamorphosis of Cyclops, the earlier stages of which had been seen long before by Leeuwenhoek and de Geer. Although an affinity between the highly modified Lernaeidae and the less -specialised parasites like Caligus was suggested, more or less definitely, by Oken, Lamarck, and others, it was not until von Nordmann's researches (1832) threw light on their development that the Lernaeidae and their allies were definitely accepted as Crustacea. In most systems of classification, however, the parasites were more or less widely separated from the free-living forms, and by Milne-Edwards they were even placed alongside of the Pycnogonida in a different sub- class. Milne -Ed wards introduced the name Copepoda for the free-living forms alone. Zenker, in 1854, seems to have been the first to associate together free-living and parasitic forms under the name Entomostraca. Modern conceptions of the morphology and classification of the Copepoda are largely based upon the long series of highly important memoirs (1857-1895) by C. Glaus. Among faunistic workers, J. D. Dana, H. Kroyer, Steenstrup and Ltitken, van Beneden, and Brady, may be specially mentioned. In more recent years, W. Giesbrecht, in his splendid monograph of the pelagic Copepoda and other important works, has given a new aspect to the problem of classification, by showing that the parasitic habit has been acquired not once but many times in the evolution of the group, and that the line of division, maintained even in Claus's classification, between free -living and parasitic forms must be abandoned. Hansen's fine monograph of the Choniostomatidae has raised that family, previously obscure, to the position of one of the best known among the parasitic groups. Finally, G. 0. Sars's work on the Copepoda of Norway, now in course of publication, is especially important as regards the bottom- living marine forms, which in recent years have been less studied than the more easily collected pelagic species. The small group of the Branchiura, associated by the older authors with the parasitic Copepoda, were removed by Zenker to the Branchiopoda. Glaus restored them to the Copepoda, and his arrangement is followed here, but in view of the important differences which separate them THE COPEPODA 73 from the Eucopepoda, it will be more convenient to deal separately with the morphology of the two groups. MORPHOLOGY OF EUCOPEPODA. Apart from the parasitic forms, which exhibit an endless variety of modifications, the Eucopepoda present a considerable uniformity in general shape. The body is divided into two regions, of which the anterior is sub -cylindrical or flattened, and is more or less sharply marked off from the posterior region, which is usually much narrower and terminates in a well- marked caudal furca. There are typically ten free segments (in- cluding the telson) behind the head, and the limit between the two regions is marked by a very movable articulation which falls either between the fifth and sixth of these (Gymnoplea) or between the fourth and fifth (Podoplea) (Fig. 40). Giesbrecht has suggested that the limit between these regions is really in the same position in both cases, that the fifth thoracic somite of the Gymnoplea is unrepresented in the Podoplea, and that the so-called fifth thoracic somite of the last-named group is a " pre-genital " somite which is suppressed in the Gymnoplea. The evidence in favour of this view, however, is very slight. The head - region is usually stated to include, in addition to the primitive cephalic somites, that corresponding to the maxillip'eds or first post- maxillary appendages. It appears, however, that in some larvae the line of articulation FIG. 40. Segmentation of the body in Gymnoplea and Podoplea. A, outline of the body of female I'tilinius (Gymnoplea), dorsal view. B, abdomen defining the first free somite of male Cawwws. c, outline of female cy/w>x . ° (Podoplea). ii, iii, vi. second, third, and sixth rilllS in front OI the maxilllpeds, thoracic somites (commonly reckoned as first, and '+ ;" -— '1-1- ^-*- +^~ second, and fifth respectively); 1 + 2, first and It IS possible a ile second abdominal somites, coalesced in the female hpqrino" fVipcpirmPvirlio-pe sex '" both instances ; t, telson ; /, caudal furca. bearing these appendages , marks the positjon of the movabiP articulation should be regarded as coalesced between the anterior and posterior divisions of •.LI j.i 11 • i . the body. not with the head-region but with the following somite. The next five somites bear swimming- feet, and constitute what is commonly called the thoracic region. It seems advisable, however, to extend the meaning of the term thorax to include also the somite of the maxillipeds. The remaining 74 THE CRUSTACEA four somites with the telson constitute the abdomen, and are limb- less, with the possible exception of the first or genital somite, which bears, in both sexes, the external openings of the genital organs, and may have what are sometimes regarded as the vestiges of a pair of appendages. In the majority of cases the number of free somites is reduced by coalescence. Very often the second (commonly reckoned as the first) thoracic somite, bearing the first pair of swimming-feet, is fused with the head, and not infrequently (in the Gymnoplea) the Fio. 41. A, Calanusfinmarchicus, female, from the side, x 16; B, antenna : C. maxilla; D, maxilliped. ii and vi, second and sixth thoracic somites ; 1+2, first and second alxlominal somites coalesced ; a', antennule ; en, endopodite ; ex, exopodite ; /, caudal t'urca. (After Sins.) last two thoracic somites fuse together. The abdominal somites are usually all distinct in the male sex (Fig. 40, B), but in the female this is rarely the case, the first two being generally fused (Figs. 40 and 41, A), and the number may be still further reduced. While there is no distinct shell-fold, the head-region usually possesses more or less well-developed pleural folds, which may be directed downwards or even inwards so as partially to enclose the bases of the limbs. These pleural folds are repeated on the free thoracic somites, but are as a rule absent on the abdomen. An- teriorly the tergal plate of the head-region is often produced into a rostral process which may be a flattened plate, movably articu- THE COPEPODA 75 lated with the head in some Harpacticidae, and resembling the rostral plate of the Leptostraca. More usually the rostrum is bent down under the head and is often forked. A very remark- able modification of the rostrum found in some Pontellidae is mentioned below in connection with the eye. The two rami of the caudal furca are usually movably articulated with the telson or terminal segment. Their setae are very constant in number and position, and afford valuable systematic characters. In some pelagic forms these setae attain an exaggerated development (Fig. 42), while in some commensal or parasitic forms they become converted into hook-like organs of adhesion (Doropygus). Between the furcal rami and somewhat on' the dorsal side is the anal aperture, covered by a small supra-anal FIG. 42. I'liini'iiium'- jiuiv, x 10. (After GieAbrecht.) plate which may represent the post-anal region of the telson in the Branchiopod Le/>iPIK/ suctorial siphon; vi, the rudimentary sixtli StaS6 ' pair of thoracic limbs ; ! + ->, the coalesced tirst CudopS and Other forms, the !m<1 second abdominal somites ; 4 + :j, the fourth , . , abdominal somite coalesced with the telson. tWO appeared tO arise trom a (After Giesbrecht.) U, larva of Ithincalanus cino-lp rnrlimpnt H«m"•" ' '» last metanauplins stage. Tlie outline single ruun ,nt. nanben, now- of tto body •* seen from abova The first three ever, has discovered, and the i'!lirs "f appenda-es are omitted, but the rudi- ments ot the posterior appendages are seen by observation has been confirmed transparency through tiie bodv. mi", maxillae ; 1 n- \ i,i iu r/"> (irst and second pairs of swimming-feet.' llldO I (After Giesbrecht.) certain marine Gymnoplea (Eucalanus, Ekincalanus, Pontella, etc.) in which the body is more elongated than usual, the rudiments of the two appendages are not only quite distinct, but are separated from each other by the suture line which marks off from the head the so-called first thoracic somite (Fig. 44, B). The " outer (or anterior) maxillipeds " are therefore the maxillae, while the inner (or posterior) pair, for which the name maxillipeds may be retained, must be regarded as the first of the thoracic series, and the somite corresponding to them is, at least in some cases, coalesced with that which bears the first pair of swimming -feet. The maxilla in its most fully developed form consists of a flattened and shortened axis of, at most, eight segments, of which the first and second each bear two, and the third a single endite (Fig. 41, C). In the Gymnoplea this appendage is beset with plumose setae, which act as a net in collecting food- So THE CRUSTACEA particles. In the Podoplea the setae are generally much reduced, and stout spines are developed on the endites. In many cases, especially in parasites, the armature is reduced to a single terminal claw-like spine and the limb is exclusively a clinging organ. In the parasitic family Lernaeopodidae, a very remarkable apparatus of attachment is formed by two appendages which unite to form a sucker-like organ of complicated structure (Fig. 45, B). In some cases, when the appendages are short, they coalesce com- pletely to the base, but in other cases they have the form of B. mxp. FIG. 45. Achtheres percarum (Lernaeopodidae). A, larva in first Copepodid stage, dorsal view. B, adult female, ventral view, x 25. a', antennule ; ft", antenna ; c.gl, cement gland opening into oviduct ; e, median eye ; /, caudal furca ; f.gl, frontal cement gland ; mx", maxilla ; mxp, maxil- liped ; o, opening of oviduct; ovd, oviduct distended with eggs ; up, openings of spermathecae. (After Claus, slightly modified.) long arms, united only at the tip. These appendages have been regarded sometimes as the " first maxillipeds " (maxillae) and sometimes as the " second maxillipeds " (here called maxillipeds). The fact that a pair of glands, identified as the maxillary (excretory) glands, open at their base, seems to show that the former interpretation is the correct one. An apical coalescence of paired appendages is not known to occur in any other Arthropods. The maxilliped (first thoracic limb) is always uniramous, and is generally more elongated than the maxilla (Fig. 41, D). It consists THE COPE POD A Si of, at most, seven (perhaps eight) segments, but the number is often much reduced, and, as in the case of the maxilla, the limb is commonly modified into a clinging organ with a strong terminal claw, or, as in a section of the Harpacticidae, into a subchelate, prehensile " hand." The five following pairs of appendages (the thoracic limbs of the ordinary terminology, the second to the sixth of the system here adopted) are in some Gymnoplea all similar and in the form of biramous swimming-legs. This form is retained by some, at least, of these limbs in all Copepoda in the later larval if not in the adult stage, and constitutes one of the most general characters of the sub-class. Each consists typically of a broad and flattened protopodite of two segments, arid of an endopodite and exopodite, each with, at most, three segments, flattened, and bearing marginal natatory setae together with, on the outer edges, strong spines (Fig. 3, B, p. 8). The proximal segments of the protopodites of each pair are con- nected with each other across the middle line by a plate formed by a transverse fold of the sternal integument (the " Bauchwirbel " of Zenker), so that in the backward and forward movement in swimming the two appendages move as one. The last (sixth) pair of thoracic limbs are similar to the preceding pairs only in the females of some genera of Gymnoplea. In the male sex of that order they are always modified into copulatory organs, often very complex, by means of which the spermatophores are affixed to the copulatory aperture of the female. This modifica- tion is asymmetrical on the two sides in correlation with the asymmetrical development of the internal generative organs in the group (Fig. 43, B). In the females of many Gymnoplea the appendages exhibit every stage of reduction even to complete dis- appearance. In the Podoplea the appendages are always present (Fig. 44, A, vi.), except in the more degraded parasites ; always vestigial, consisting of one or two small segments ; and are not specially modified in the male. In some Harpacticidae and Ascidi- colidae, hoAvever, they become enlarged in the female sex into plate-like appendages serving to protect the egg-masses. It is noteworthy that these vestigial limbs of the last pair may persist even in cases where the preceding pair of limbs is suppressed. In the Podoplea (but not in Gymnoplea) the genital apertures of the female on the first abdominal somite are guarded by valvular plates moved by muscles. These valves have been supposed to represent a vestigial pair of appendages. In the parasitic forms, with the loss of the power of locomotion, the thoracic limbs become more or less reduced. In Lernaeocem, for example, they persist as microscopic though completely formed limbs set at long intervals along the length of the unsegmented 6 82 THE CRUSTACEA body. In the Chondracanthidae the anterior two pairs alone are developed, and these become enlarged in the adult into clumsy, unsegmented bifid lobes. In other cases the thoracic limbs are reduced to minute, unsegmented processes, or some or all of them may disappear. In a few cases the adult is entirely without appendages, as in the Herpyllobiidae. Alimentary System. — The alimentary canal is in many cases of simple form, not divided into sharply defined regions and without diverticula. The stomodaeum and proctodaeum are short. In many Gymnoplea there is a short median diverticulum anteriorly, and in some cases, immediately behind this, a pair of small lateral (hepatic) caeca which may be bifid (Eucalanus). In some Cory- caeidae and Asterocheridae these caeca are large and much branched (Fig. 46). Groups of gland-cells described as salivary occur in the region of the labrum and epistome. The extrinsic muscles of the alimentary tract are well developed in Eucopepoda. Certain muscles running from the anterior part of the gut to the dorsal and anterior region of the body-wall are of importance in producing a rhythmical displacement of the whole alimentary canal, and serving, in the absence of a heart, to cause a circu- latory movement of the blood. In some parasites the dilator muscles of the oesophageal region are greatly developed and act as a suctorial apparatus. The short rectum (proctodaeum) is usually pro- vided with dilator muscles running outwards to the body- wall in addition to the usual constrictors, and the rhythmical movements of dilatation and contraction produced by them have been regarded as subservient to a process of anal respiration. The alimentary canal is usually nearly straight (except for the sternal flexure of the oesophageal portion), but in a few Gymnoplea and in Cancerilla (Asterocheridae) its course is slightly sinuous. Circulatory System. — A heart is present in most Gymnoplea and in the genus Misophria among the Podoplea. In all other Cope- poda it appears to be wanting. When present it has an abbreviated saccular form, and is situated in the region of the first or second Artotrogus orbicular!.*. Outline of body, from above, showing ramified lateral diverticula of the alimentary canal. (After Giesbrecht.) THE COPEPODA 83 free somite. There are three ostia, one median on the posterior surface and a lateral pair. Anteriorly the heart gives oft' an aortic vessel, usually very short. Only in Eucalanus is the aorta described as extending into the frontal region of the head and dividing into two pairs of lateral vessels. In the majority of the free-living Podoplea where the heart is absent the blood is kept in motion mainly by the continuous rhythmical backward and forward movements of the alimentary canal, effected by the extrinsic muscles already mentioned. In certain parasitic forms (Caligus, young Achtheres) a heart, or an apparatus having an analogous function, is said to be present, but exact details as to its structure are wanting. In Lernanthropus and some other Dichelestiidae a closed system of vessels is present containing a yellowish or reddish fluid. There is no heart, and the relation of this system to the circulatory apparatus of other Copepoda is quite obscure. Excretory System. — The maxillary ("shell") gland is the func- tional excretory organ in the adult stage of most, if not all, Copepoda. It is much larger in the freshwater forms, where the duct is long and convoluted, than in the marine forms, in which it is often hard to find, and sometimes apparently absent. The end sac is small, and the tube terminates in a short chitin-lined duct opening on the posterior surface of the maxilla. In the freshwater Harpacticid Belisarius a curious vibratile organ is found connected with, or in close proximity to, the maxillary gland. It has been supposed to be of the nature of a " flame-cell," but it is more probably a muscular fibre or membrane aiding the circulation of the blood in the neighbour- hood of the gland. Glands. — Unicellular dermal glands are present on the body and limbs of most Eucopepoda. Certain pelagic forms belonging to various genera of the Centropagidae and Oncaeidae are known to be phosphorescent, and Giesbrecht has shown that this is due to certain of the dermal glands, the secretion of which becomes luminous on issuing from the apertures of the glands. In fresh- water Cyclopidae and Harpacticidae the secretion of the dermal glands envelops the body when the water dries up, and forms a protective case enabling the animal to survive prolonged desiccation. Nervous System. — The ventral nerve-cord is always short, not reaching beyond the fourth free thoracic somite. It is divided into distinct ganglia in the Gymnoplea, but in the Podoplea, so far as is known, the ganglia are all coalesced. In the Corycaeidae and in the parasitic families the whole system is still more concentrated, forming a thick perioesophageal ring. Even in the Gymnoplea the distinction between ganglia and commissures is not sharp, nerve- cells being present abundantly on the latter as on the former. I 84 THE CRUSTACEA In Cyclops it has been observed that the pair of nerves to the antennae originate from the oesophageal connectives, so that as regards their nerve-supply these appendages are parastomial. Sense-Organs. — The paired compound eyes of other Crustacea appear to be unrepresented in the Eucopepoda, although rudiments of them were observed by Grobben in the development of Calanus. The mtuplius-eye, on the other hand, is almost universally present in the free-living Eucopepoda, and even in many of the parasitic forms, and in some cases, especially among the Gymnoplea, it attains a complexity of structure not observed in any other class of Crustacea. In the simplest and typical form it consists of three ocelli, each supplied by a separate nerve from the brain. Two of the ocelli are dorsal and look upwards and forwards, while the third is ventral, looking downwards. Each consists of a cup-shaped mass of pigment, containing in its cavity a number (up to ten) of retinal cells con- tinuous at their distal ends with the nerve-fibres, and having (at least in some cases) a rhabdome near the proximal end. In the genus Anomalocera (Pontellidae) and in some Asterocheridae the number of ocelli is increased to five, those of the dorsal pair being doubled. In some Eucopepoda the eye is movable by means of special muscles. These are wanting, however, in many cases (Cyclops}. In some cases the visual apparatus is perfected by the addition of a pair of corneal lenses formed by thickening of the cuticle over the dorsal pair of ocelli. These may be inconspicuous as in Cyclops, or large and well-defined as in Miracia (Harpacticidae). It is, however, in the two widely separate families of Pontellidae and Corycaeidae that the structure of the visual apparatus reaches its highest degree of complication. In the former the dorsal ocelli are often provided with cuticular lenses, and in addition there may be developed a vesicular crystalline body interposed between the retinal cells and the cuticle. The ventral ocellus approaches the sternal surface and sometimes projects as a papilliform or peduncu- late prominence, while the deflected rostrum in front of it becomes thickened in such a manner as to form a biconvex lens, serving to concentrate the rays of light upon it. In the Corycaeidae (Fig. 47) the three ocelli are widely separate and the median element remains small, while the dorso- lateral pair attain a much greater — sometimes relatively enormous — development. Each is provided with a large biconvex cuticular lens (/), and the retinal apparatus is at a considerable distance from this, at the apex of a conical space the base of which is formed by the lens and the walls by a delicate membrane. The pigment-cup (p) is elongated into a tubular form and at its mouth is set a vesicular " crystalline body " (c). In Corycaeus the posterior THE COPE POD A 85 part of the ocular apparatus reaches back into the region of the anterior thoracic somites, and in Copilia the pigment-cups enclos- ing the retinal cells lie together in a conical protuberance on the sternal surface of the body. In this account it has been assumed that the " paired eyes " of Pontellidae and Cbrycaeidae are derived from the dorso- lateral elements of the nauplius-eye. It must be mentioned, hoAvever, that Glaus, while admitting this derivation in the case of the Corycaeidae, regarded the lateral eyes of the Pontellidae as homo- logous with the paired compound eyes of other Crustacea, If it be the case, however, that in the Pontellid eye, as in those of other Eucopepoda, the retinal cells are "inverted" (or are connected with nerve-fibrils at the end turned towards the light), this homology would seem to be impossible. A problematical organ to which a visual function has been attributed is found in the genus Pleuromamma. It lies on one side FIG. 47. «//iVi<.s male, from tin* sidi>, showing one of the largo paired eyes. I, leiis ; c, crystalline body ; p, tubular pigment-cup. (After Leuckart.) of the cephalic shield, in the region of the maxillipeds, and consists of a globular refractive body enclosed in a mass of pigment, the whole projecting from the surface of the body in a little papilliform elevation. It has also been suggested that this is an organ of phosphorescence, but according to Giesbrecht this is not the case. The " acsthetascs" or "olfactory filaments," of the antennules have already been mentioned. Their number and arrangement vary very much in different forms, and afford valuable systematic characters. The "frontal sense-organs" are certain sensory setae, generally a single pair, on the front surface of the head above the rostrum, which are supplied by a pair of nerves arising from the brain, and which have been supposed to be the seat of some special sense. The existence of " auditory " organs in the Eucopepoda is doubtful. A pair of statocysts have been described in the anterior part of the brain in Eucodid stage. B, pupal sta- tion Of the region of the genital »"t'-""»l'- : '.<"- antenna ; / J», Mwt and second O pairs ot swimming - feet ; '.', mass of cement Segment. Impregnation takes produced by frontal jilarnl ; oc, nauplius-eye. (At't'-r Clans, from Kottchelt and Holder's FIG. 51. place in this free -swimming stage and the male does not develop further. The female, however, seeks a second host, fish of the family Gadidae, and becoming attached to the gills, burrows into the flesh so that the whole anterior region of the THE CRUSTACEA body is embedded. Three ramified processes grow out from the head into the tissues of the host, and the region of the genital somite becomes enormously enlarged, forming the greater part of the vermiform body (Fig. 52, C, D). The swimming - feet are retained in a vestigial condition near the anterior end. In some other forms parasitic on fish (Lernaeopodidae, Chon- dracanthidae) the male is attached to the female in a dwarfed condition throughout life. The larva becomes parasitic in the THE COPEPODA 93 first Copepodid stage, and the later stages of the typical series are suppressed, the characteristic features of the parasite being assumed at the next moult. In the Choniostomatidae, which are parasitic on other Crustacea, the larva hatches in the first Copepodid stage, with two pairs of swimming-feet, and becomes attached to the host by an adhesive frontal plate, corresponding probably to the "frontal thread" of other forms. This larva may give rise at once to the adult form, or a pupal stage with reduced limbs may intervene. The male is similar to the female in essential structure, but of much smaller size. The most extreme stage of degeneration is reached in the Herpyllobiidae, which are parasitic on Polychaete worms and on Crustacea. In Bhizorhina, which is the most thoroughly known, the adult female is entirely without appendages, and is attached by a tubular process which ramifies within the body of the host in a fashion recalling the " roots " of the Rhizocephala. The adult male is also entirely limbless, but remains enclosed within the last larval skin (of the first Copepodid stage). Within it are formed a pair of relatively enormous spermatophores, which are not expelled from the body but discharge their contents through ducts which pass out close to the point of attachment, and in front of the position of the larval mouth. A number of males are attached to the body of the female near the genital apertures. The Monstrillidae have a very remarkable life-history, which has only recently been made known. The adults of both sexes are free-swimming, and are without mouth-parts or alimentary canal. The newly hatched young are also free-swimming, but the inter- mediate stages are endoparasitic within various Polychaete annelids. The life-history is most fully known in the case of Haemocera danae, investigated by Malaquin (Fig. 53). The adults (F) have antennules and four pairs of swimming-feet, but no trace of antennae or mouth-parts. The fifth pair of thoracic feet are vestigial in the female and are stated to be absent in the male. The alimentary system is represented by a blind stomodaeal invagination and a mass of undifferentiated endoderm cells. The female carries a single packet of eggs (ov) adhering to a pair of very long setae (g.s) which spring from the genital valves. The young are hatched as nauplii, without mouth or alimentary canal, and with strong hook-like mandibles (A). The nauplius burrows into the body of its host (the Polychaete Salmacina), casting its cuticle in the process and losing its limbs, so that when it reaches the body- cavity it consists merely of a mass of embryonic cells without a cuticle (B). From the body-cavity the parasite passes into the vascular system of the host, where it undergoes its further develop- ment. A thin cuticle is now secreted, and a pair of processes begin to grow out from one end of the ovoid body (C, pi'). These 94 THE CRUSTACEA increase in size, in some cases, to several times the length of the body, and as the cuticle covering them is very thin it is believed f. FIG. 53. Stages in the life-history of Haemocei-n datutc. A, free-swimming nauplius-larva, x 280. B, embryo after penetrating into the body-cavity of the host; no organs remain except the degenerating nauplius-eye. C, later stage, from the vascular system of the host ; the absorptive processes have begun to develop. D, still later stage ; the absorptive processes are fully formed and the posterior end of the embryo is provided with rows of recurved hooks. E, tlm adult female, just before emergence from the host ; the anterior part of the body is distended with eggs. F, free-swimming female, carrying the extruded eggs, x 28 ; the hypodermis in the anterior part of the body has separated from the cuticle after expulsion of the eggs arid forms a sheath around the nerve-cord. «', antennule ; br, brain ; e, nauplins-eye ; /, swimming-feet ; <7.s, genital setae ; m, position of month ; mil, mandible of nauplius ; n,' nerve-cord connecting brain with ventral nerve-chain ; or, mass of eggs carried on genital setae ; ovy, ovary ; jn; absorptive processes. (After Malaqiiin.) that they act as absorptive organs. In some cases, but not always, a second pair of these processes is found behind the first. From the position which they occupy relatively to the rudiments of THE COPEPODA 95 appendages afterwards appearing, it is believed that they represent the antennae and mandibles of the nauplius. The organs of the adult are gradually differentiated within the cuticular sac, which enlarges with the growth of the animal. The pointed posterior end of the sac is surrounded by rows of recurved hooks (D), and these appear to be used in boring a way out through the tissues of the host when development is complete (E). The reproductive organs are developed before the parasites escape from the host. After escaping, a single moult takes place and sexual maturity is reached. The life-history of the Monstrillidae may be compared with that of the male Enterognathus described above, in which the earlier larval stages and the adult are free-swimming, while the intervening stages are parasitic and degenerate, and in which also the adult is incapable of feeding. MORPHOLOGY OF BRANCHIURA. The body is much flattened, and is divided into three regions, an unsegmented, cephalothoracic region followed by three free thoracic somites and an unsegmented abdomen (Fig. 54). The cephalothoracic region is covered by a greatly developed head-shield or carapace, which, while not projecting beyond the articulation of the succeeding somite as a distinct " shell-fold," is expanded on each side in a great wing-like pleural fold. In many species the lateral folds are produced backwards and cover the thoracic somites and their appendages, and sometimes even the abdomen. The abdomen is notched or bilobed, and bears a pair of minute furcal rami (/). Appendages. — The antennules are small, consisting of four segments, of which the first is divided into two parts and is provided with a large hooked claw and some smaller spines, used for attachment to the host. The antennae are also short, unirumous in the adult, of four segments, the basal part provided with stout spines. The mouth-parts are suctorial, the upper and lower lips together forming a proboscis (Fig. 54, p) within which are enclosed the mandibles and maxillulae. The maxillulae are never included in the proboscis in suctorial Eucopepoda. The mandibles are without palps in the adult and have sickle- shaped, serrated tips. The maxillulae are stated to be wanting in the genus Dolops. In Argulus they are simple, lancet-like blades. The maxillae (" first maxillipeds ") are very remarkably modified. Except in the genus Dolops, where they end in stout hooked claws, they are represented chiefly by a pair of large adhesive suckers (Fig. 54, mx"), situated some distance in front of the proboscis. The sucker is developed from the basal portion of the appendage, the 96 THE CRUSTACEA distal segments which are present in the larva persisting as a vestige under the outer margin of the sucker or disappearing altogether. The sucker itself is provided with strong muscles and its margin is strengthened by radiating chitinous rods. The first thoracic appendage ("second maxilliped ") (mxp} consists of five segments, the basal one produced into a lobe armed with strong teeth and the end of the limb carrying two stout claws. The next four pairs of appendages are biramous natatory feet. The first pair is attached to the cephalothoracic region, the remain- ing three correspond to the three free thoracic somites. Each consists of a stout protopodite, which, in some species at least, has sp ArrjnluK amrrimnus, female, from below, x 5. a', antennule ; o", antenna; e, paired eye ; /, caudal furca ; inx", sucker formed by maxilla (" lirst maxilliped"); inxp, first thoracic appendage ('• second maxilliped ") ; j>, suctorial proboscis ; sp, poison spine. (After Wilson.) three segments, and an exopodite and endopodite of about equal length. The exopodite is unsegmented, and the endopodite consists in the first pair of three segments, in the second of one, and in the last two pairs of two segments. Both rami are furnished with two rows of long plumose setae, set along the dorsal and ventral margins respectively. The insertion of the broadened bases of these setae gives to the exopodite and endopodite the appearance of being divided into numerous short segments, and they have been so described, but the arrangement of the musculature shows that these do not represent true segments of the limb. On certain of the anterior legs in most of the species the protopodite carries at its distal end, besides the endopodite and exopodite, a slender appendage, known as the flagellum, which THE COPEPODA 97 originates to the outer (or dorsal) side of the exopodite, and is bent backwards upon the protopodite. It is provided with two rows of plumose setae, and its probable function is to cleanse the lower surface of the carapace. The homology of this appendage is doubtful. It cannot be an epipodite since it springs from the distal segment of the protopodite. Nothing equivalent to it is found in the Eucopepoda. In the male sex the peduncles of the last two, sometimes of the last three, pairs of legs are modified for purposes of copulation. The details differ in the different species, but in all the peduncle of the penultimate pair is excavated to form a seminal pouch which ovy. -f-b. Fio. 55. A. ii'Tvou-i systria of Argvlusarnericanitfs. e, nauplius-eye ; immediately below is theopening through which the oesophagus passes. B, female reproductive organs, ovy, ovary; od, open- ing of oviduct : .-71, spi-rmatheca : the papillae on which the spermathecal ducts open are seen just below the opening of oviduct ; t.p, "tactile papillae." C, male reproductive organs, gl, accessory gland ; o, external opening ; «.v, seminal vesicle ; t, testis. (After Wilson.) opens on the dorsal surface of the limb. On the anterior surface of the peduncle of the last pair is a peg-like process which seems to be used for opening the mouth of the seminal pouch so that the latter can be filled with sperm. In some species in the female sex u pair of finger-like processes (Fig. 55, B, t.p) are situated at the sides of the genital orifice between the last pair of legs. These may possibly represent a vestigial sixth pair of thoracic appendages. In the genus Argulus a very peculiar poison apparatus is situated in front of the mouth. It consists of a hollow spine (Fig. r>4, .s/>), which can be withdrawn into and protruded from a sheath by the action of special muscles. The duct which traverses it com- municates with three groups of large unicellular glands lying at its THE CRUSTACEA base. These glands appear to be a specialisation of some of the dermal glands which (as in many Eucopepoda) are abundantly dis- tributed over the surface of the body. Alimentary System. — The alimentary canal has a narrow oeso- phageal region, with strong circular and longitudinal muscles, which projects, funnel-like, into the wide stomach. The latter gives off a pair of diverticula, which are much ramified, and occupy the greater part of the lateral lobes of the cephalothorax. When feed- ing, these diverticula, as well as the stomach and intestine, become filled with the juices of the host. The stomach is followed by a wide intestinal region which is separated by a sphincter from the narrow rectum. The latter is without the dilator muscles found not only in Eucopepoda but in most Crustacea. Circulatory System. — The heart is situated at the junction of thorax and abdomen, and sends off an aorta which reaches as far forward as the brain. It is provided with one or two pairs of lateral and inferior afferent ostia and sometimes a median inferior efferent opening at the base of the aorta. There appears to be doubt as to the existence of a posterior median efferent opening. In the earliest larval stages the heart is absent, and the circulation is mainly carried on, as in the later stages it is assisted, by rhythmical contractions of dorso-ventral muscles in the abdomen. Excretory System.- — An excretory gland of the usual type is present, and has been identified as the maxillary gland (shell- gland). It appears from the investigations of Glaus and of Nettovich, however, that it differs from the maxillary gland of the Eucopepoda and other Crustacea in opening not on the maxilla but on the first thoracic limb (the so-called " second maxilliped "), o'r on the sternal surface of the body close to the base of that appendage. Nervous System. — The ventral nerve-chain (Fig. 55, A) is shortened, but six ganglia can be distinguished. The statements as to the origin of the nerves to the two pairs of maxillipeds are con- flicting, but the last four ganglia supply the somites of the four- pairs of swimming-feet, the last also sending nerves to the abdomen. Sense-Organs. — A median (nauplius) eye of the usual structure is set upon the dorsal surface of the brain (Fig. 55, A, e). In addition, there are a pair of large compound eyes (Fig. 54, e) visible through the transparent integument of the dorsal surface. Each is supplied by a stout optic nerve which swells into a ganglion before entering the eye. The eye is movable, but it differs from the similarly movable, non-pedunculated eyes of the Branchiopoda in the fact that it moves not in a "corneal pouch" but in a blood-space which intervenes between the outer ends of the quadripartite crystalline bodies and the integument. Seen from above this blood-sinus is bounded on the anterior, external, THE COPEPODA 99 and posterior sides by a chitinous wall. It seems not unlikely that further research will show the wall to be formed by an invagination of the cuticle, in which case it may represent the vestige of a corneal pouch like that of the Branchiopoda. Reproductive System. — The ovary (Fig. 55, B, ovy) is unpaired and, at its first appearance in the larva, asymmetrically placed, afterwards assuming a median position. Rudiments of two ovi- ducts are found in the larva, but that situated on the same side of the body as the ovary atrophies, while the other develops further and ultimately opens (od) in the middle line between the bases of the last pair of legs. A pair of spermathecae (Fig. 55, B, sp) are found in the abdomen. They are not connected with the oviducal opening, but each has a short duct which gives off a blind diverticulum and terminates on a papilla with a retractile spiniform tip close to the anterior margin of the abdomen. It is believed that the sharp points of these papillae pierce the envelopes of the eggs when the latter are laid, so as to permit the entrance of the spermatozoa. The eggs are laid attached to stones or other objects. The testes (Fig. 55, C, /) are paired and lie in the abdomen. Their ducts unite to form an unpaired seminal vesicle (s.v), and after receiving the ducts of a pair of accessory glands (gl) lying in the thoracic region again unite to open by a median pore (0) between the bases of the last pair of legs. The seminal pouches on the penultimate legs of the male and the structures connected therewith have already been mentioned. DEVELOPMENT OF BRANCHIURA. In some species of Argulus the newly hatched larva has all the appendages similar to those of the adult with the exception of the first maxillipeds, which are not modified into suckers but are stout clasping limbs, each consisting of four segments and ending in a double claw. In other species, however, among which is the common European A. foliaceus, the newly hatched larva differs still more from the adult (Fig. 56). The antennae are biramous, having a large unsegmented exopodite which is lost in the adult, and the mandibles have a large uniramous palp of two segments. The antennal exopodite and the mandibular palp are tipped with plumose setae and serve as swimming-organs. The heart is not found in the earliest larvae but develops after the first moult. REMARKS ON HABITS, ETC. The majority of the Eucopepoda are marine, but numerous species, belonging chiefly to the families Centropagidae, Harpac- IOO THE CRUSTACEA ticidae, and Cyclopidae, are very abundant in fresh water. Most of the Gymnoplea are pelagic, forming a very important part of the plankton of the sea and of lakes. The non- parasitic Podoplea, on the other hand, with the exception of a few pelagic groups like the Corycaeidae, belong to the bottom-fauna. A large Newly hatched larva of Argulti* fuliweus. a', antennnle ; <<", antenna ; md, mandible ; milt, mandibular palp; mf, maxilla (first maxilliped); mf", (second) maxilliped; j*1, jv*, first and fourth pairs of swimming-feet. (After Clans, from Korschelt and Heider's Embryology.) fc proportion of the parasitic forms of various families attack fish, and some of these, such as Lernaeocera and Achthcres, occur in fresh water. Pennella is sometimes found on whales. Other parasitic and semi- parasitic forms are found on various groups of marine inverte- brates. The Branchiura are temporary parasites on fish and occur both in the sea and in fresh water. Most of the free-living Eucopepoda are minute, but some of the THE COPEPODA 101 parasitic species attain a greater size, the largest (Pennella) being more than a foot in length. No fossil remains of Copepoda are known. AFFINITIES AND CLASSIFICATION. On the hypothesis that the Nauplius represents the ancestral type of the Crustacea, the Eucopepoda would be regarded as the most primitive existing members of the class, retaining, as they do, naupliar characters in the form of the first three pairs of appendages and in the absence of paired eyes and of a shell-fold. As already indicated, however, it is much more probable that they are to be regarded as a specialised and in some respects degenerate group which, while retaining, in some cases, a very primitive structure of the cephalic appendages, has diverged from the ancestral stock in the reduction of the number of somites, the loss of the paired eyes and shell-fold, and the simplified form of the trunk-limbs. The prevalence of parasitism and the great structural changes associated therewith render the classification of the Euco- pepoda a matter of peculiar difficulty, and none of the schemes hitherto proposed is altogether satisfactory. In Claus's system (1880) the distinction drawn by most of the older authors between free-living and parasitic forms is still maintained in the two chief divisions of Gnathostomata and Siphonostomata, the latter includ- ing forms in which the mouth -parts are more or less distinctly suctorial. Giesbrecht has shown, however, that this arrangement, and also that proposed by Canu, based on the copulatory pores of the female, are quite unnatural, since forms with biting and suc- torial mouth-parts and with paired and unpaired copulatory pores may occur within the limits of the same family. Giesbrecht's own classification of the Eucopepoda, which is given below, marks a distinct advance, especially as regards the separation of the more primitive pelagic families to form the group Gymnoplea. His arrangement of the remaining families, which he groups together as Podoplea, is, however, less convincing, and he does not attempt to define the position of many of the parasitic forms. The system is therefore incomplete and can only be adopted as a temporary expedient pending further investigation. Some modern writers follow Zenker and Thorell in referring the Branchiura to the Branchiopoda, although the only character which can now be referred to in support of this arrangement is the presence of paired compound eyes. On the other hand, the com- parison instituted by Glaus between the appendages of Branchiura and Eucopepoda shows a general similarity of structure which can- not be disregarded. The only serious difficulty in the way of this comparison is the difference in the position of the maxillary gland, 102 THE CRUSTACEA which in the Eucopepoda, as in other Crustacea, opens on the maxilla, but in the Branchiura on the first thoracic appendage. It seems more probable, however, that a shifting of the aperture has taken place than that the appendages do not correspond serially in the two groups. SUB-CLASS COPEPODA, H. Milne-Edwards (1830). ORDER 1. Eucopepoda, Clans (1875). Paired compound eyes absent; genital apertures on the seventh trunk - somite ; thoracic limbs without flagellum ; fertilisation by epermatophores. SUB-ORDER 1. Gymnoplea, Giesbrecht. Last thoracic somite firmly connected with the preceding somite and movably articulated with the first abdominal somite ; last pair of thoracic appendages in female similar to preceding pair, or reduced or absent, in male always present and modified as copulatory organs ; one or neither of the antennules geniculate in male ; eggs deposited singly or carried in a single packet ; heart generally present ; vas deferens unpaired, its opening unsymmetrically placed. Free-living forms, mostly pelagic. » TRIBE 1. AMPHASKANDRIA, Giesbrecht. Antennules of male not geniculate, with more numerous aesthetasca than in female. Family CALANIDAE. This extensive group is divided by Sars into twelve families. Most of these correspond to sub-families recognised by Giesbrecht. Calamts, Leach (Fig. 41) ; Eucalanus, Dana ; Rhincalanus, Dana ; Calocalanus, Giesbrecht (Fig. 42) ; Paracalamis, Boeck ; Pseudo- calanus, Boeck ; Aetideus, Brady ; Euchaeta, Philippi ; 'PJiaenna, Glaus ; Scolecithrix, Brady ; Diaixis, G. 0. Sars ; Stephos, Scott ; Tharybis, G. O. Sars; Pseudocyclopia, Scott. (The families represented by the last eight genera are separated from the Amphaskandria by Sars to form a division Isokerandria.) TRIBE 2. HETERARTHRANDRIA, Giesbrecht. One of the antennules geniculate in male. Family CENTROPAGIDAE. Divided into eight families by Sars : — Centropages, Kroyer ; Diaptomus, Westwood ; Pseudodiaptomus, Herrick ; Heterocope, G. O. Sars ; Lucicutia, Giesbrecht ; Temora, Baird ; Metridia, Boeck ; Heterorhabdus, Giesbrecht ; Pleuromamma, Giesbrecht ; Arietellus, Giesbrecht. Family PSEUDOCYCLOPIDAE. Pseudocyclops, Brady. Family CANDACIIDAE. Candacia, Dana. Family PONTELLIDAE. Divided into four families by Sars : — Pontella, Dana; Parapontella, Brady; Acartia, Dana ; Anomalocera, Templeton ; Tortanus, Giesbrecht. THE COPEPODA 103 SUB-ORDER 2. Podoplea, Giesbrecht. (Most of the following characters are subject to modification in parasitic forms.) Last thoracic somite movably articulated with the preceding, and firmly united with the first abdominal somite, which it resembles in size and form. Last pair of thoracic feet vestigial in both sexes ; not modified as copulatory organs in male. Both or neither of the antennules geniculate in the male. Eggs generally carried in paired or unpaired masses. Heart absent (except in Misophria}. Male reproductive system usually paired. Free-living (rarely pelagic) or parasitic. TRIBE 1. ISOKERANDRIA, Giesbrecht. Swimming forms with antennules not geniculate in male, generally similarly segmented in both sexes, and parasitic forms allied to these. Family CLAUSIDIIDAE. Clausidium, Kossmann. Family CORYCAEIDAE. Corycaeus, Dana (Fig. 47); Copilia, v Dana ; Sapphirina, J. V. Thompson. Family. OXCAEIDAE. Oncaea, Philippi. Family LICHOMOLGIDAE. Licho- molgus, Thorell. Family ERGASILIDAE. Ergasilus, Nordmann. Family BOMOLOCHIDAE. Bomolochus, Nordmann. Family CLAUSIIDAE. Clausia, Claparede. Family NEREICOLIDAE. Nereicola, Keferstein. TRIBE 2. AMPHARTHRANDRIA, Giesbrecht. Swimming forms, with both antennules geniculate in male, generally differently segmented in the two sexes, and parasitic forms allied to these. Family MISOPHRIIDAE. Misophria, Boeck. Family HARPACTICIDAE. This extensive group corresponds rather to an assemblage of families (cf. Sars) : — Harpacticus, Milne-Edwards ; Longipedia, Glaus ; Cervinia, Norman ; Ectinosoma, Boeck ; Peltidium, Philippi ; Tegastes, Norman ; Porcellidium, Glaus ; Idya, Philippi ; Thalestris, Glaus ; Diosaccus, Boeck ; Canthocamptus, AJJestwood ; Belisarius, Maupas ; Laophonte, Philippi ; Kctelki, Dana ; Miracia, Dana ; Aegisthus, Giesbrecht ; Clytem- n extra, Dana, Family CYCLOPIDAE. Cyclops, Muller ; Oithona, Baird. Family MONSTRILLIDAE. Monstrilla, Dana ; Haemocera, Malaquin (Fig. 53). Family ASCIDICOLIDAE. Ascidicola, Thorell ; Notodelphys, Allman , Doropygus, Thorell ; Notopterophorus, Costa ; Enterognathus, Giesbrecht (Fig. 50). Family ASTEROCHERIDAE. Asterocheres, Boeck ; Acontio- phorns, Brady (Fig. 44) ; Cancerilla, Dalyell ; Scottocheres, Giesbrecht ; Artrotroyus, Boeck (Fig. 46). Family NICOTHOIDAE. Nicothoe, Audouin. Family DICHELESTIIDAE. Dichelestium, Hermann ; Lernanthropus, Nordmann. The position of the remaining families (consisting wholly of parasitic forms) with respect to this system of classification is not yet determined. The groups most usually accepted are : — Family CALIGIDAE. Caligus, Miiller. Family LERNAEIDAE. Lernaea, Linn. (Figs. 51, 52); Pennella, Oken ; Lernaeocera, Blainville. Family LERNAEOPODIDAE. Lernaeopoda, Kroyer ; Achtheres, Nordmann (Fig. 45). Family CHONDRACANTHIDAE. Chondracanthus, La Roche. Family io4 THE CRUSTACEA CHONIOSTOMATIDAE. Choniostoma, Hansen. Family HERPYLLOBIIDAE. Herpyttobiut, Steenstrup and Lutken ; Rhizorhina, Hansen. ORDER 2. BrancMura, ThoreJl (1864). Paired compound eyes present ; genital apertures on fifth trunk- somite ; thoracic limbs sometimes with flagellum ; no spermatophores. Family ARGULIDAE. Argulus, O. F. Miiller (Fig. 54) ; Dolops, Audouin ; Chonopeltis, Thiele. LITERATURE. 1. Brady, G. S. A Monograph of the Free and Semiparasitic Copepoda of the British Islands. Ray Society, 3 vols. 8vo, 1878-1880. 2. Report on the Copepoda. Rep. Voy. "Challenger." Zool. viii. pp. 142, 55 pis., 1883. 3. Canu, E. Les Copepodes du Boulonnais. Trav. du Lab. de "\Viniereux, vi. pp. 292, 30 pis., 1892. 4. Glaus, C. Ueber den Bau und die Entvvicklungsgeschichte parasitischer Crustaceen, pp. 34, 2 pis. 4to. Cassel, 1858. 5. Ueber den Bau und die Entwicklung von Achtkeres pcrcarum. Zeitschr. wiss. Zool. xi. pp. 287-308, pis. xxiii. and xxiv., 1861. •/ 6. . Die freilebenden Copepoden. pp. x + 230, 37 pis. 4to. Leipzig, 1863. 7. Beobachtungeu iiber Lcrnaeocera, Peniculus, und Lcrnaea. Schrift. Ges. Naturw. Marburg, ii. pp. vi + 32, 4 pis., 1868. *r 8. Ueber die Entwicklung, Organisation und systeniatisclie Stellung der Argnliden. Zeitschr. wiss. Zool. xxv. pp. 217-284, pis. xiv.-xviii., 1875. 9. Ueber Lernaeascus iiematoxys, Cls., und die Familie der Philichthyden. Arb. zool. Irist. Wien, vii. pp. 281-315, 4 pis., 1887. 10. Ueber die Maxillarfiisse der Copepoden und die morphologischen Deutung der Cirripedien-Gliedmassen. Arb. zool. Inst. Wien, xi. pp. 49- 64, 1 pi., 1895. 11. Giesbrccht, W. Pelagische Copepoden. Fauna und Flora d. Golfes von Neapel. Monogr. xix., 1892. 12. Asterocheriden. Op. cit. Monogr. xxv., 1899. 13. — • Mittheilungen iiber Copepoden. (14 papers.) Mitth. xool. Stat. Neapel, xi.-xiv., 1893-1900. v 14. and Schmeil, 0. Copepoda Gynmoplea. Das Tierreich, Lief. 6, 1898. 15. Grobben, C. Die Entwickelungsgeschichte der Cetochilus scptenlrionalis. Arb. zool. Inst. Wien. iii. pp. 243-282, 4 pis., 1881. 16. Haiisen, H. J. The Choniostomatidae. 205 pp. 13 pis. 4to. Copenhagen, 1897. 17. Hartog, M. The Morphology of Cyclops and the Relations of the Copepoda. Trans. Linn. Soc. Zool. v. pp. 1-46, pis. i.-iv., 1888. 18. Malaquin, A. Le parasitisme evolutif des Monstrillides (Crustacus cope- podes). Arch. Zool. exp. (3) ix. pp. 81-232, pis. i.-vii., 1901. 19. Nettovich, L. v. Neue Beitrage zur Kenntniss der Arguliden. Arb. zool. Inst. Wien, xiii. pp. 1-32, 2 pis., 1900. 20. Nordmann, A. v. Mikrographische Beitrage zur Naturgeschichte der wirbellosen Thiere, Heft 2, 4to, Berlin, 1832. THE COPEPODA 105 21. Sars, G. 0. An Account of the Crustacea of Norway. IV. Copepoda Calanoida, 1901-1903. V. Copepoda Harpacticoida, 1903 — (in progress). 22. Schmeil, 0. Deutschlands freilebende Siisswasser-Copepoden. Bibliotheca Zoologica, iv. Heft 11, 1892; v. Heft 15, 1894; vii. Heft 21, 1896- 1898. 23. Stcenstrup, J. J. S., and Ltitken, C. F. Bidrag til Kundskab om det aabiie Havs Snyltekrebs . . . Kgl. Danske Vidensk. Selsk. Skrifter, (5) v. pp. 341-432, pis. i.-xv., 1861. 24. Thiele, J. Beitriige zur Morphologic der Arguliden. Mitt. Mns. Naturkunde, Berlin, ii. (4), pp. 1-51, pis. vi.-ix., 1904. 25. Wilson, G. B. North American Parasitic Copepods of the Family Argulidae, with a bibliography of the group and a systematic review of all known species. Proc. U.S. Nat. Mus. xxv. pp. 635-742, pis. viii.-xxvii., 1902. A New Species of Argulus. . . . Op. cit. xxvii. pp. 627-655, 1904. 26. North American Parasitic Copepods belonging to the Family Caligidae. Pt. I. The Caliginae. Proc. U.S. Nat. Mus. xxviii. pp. 479-672, pis. v.-xxix., 1905. Pt. II. The Trebinae and Euryphorinae. Op. cit. xxxi. pp. 669-720, pis. xv.-xx., 1907. Pts. III. and IV. A Revision of the Pandarinae and the Cecropinae. Op. cit. xxxiii. pp. 323-490, pis. xvii.-xliii., 1907. CHAPTER V THE CIRRIPEDIA SUB-CLASS CIRRIPEDIA, Burmeister (1834). Order 1. Thoracica. Sub-Order 1. Pedunculata. „ 2. Operculata. Tribe 1. Asymmetrica. „ 2. Symmetrica. „ 2. Acrothoracica. „ 3. Ascothoracica. „ 4. Apoda. „ 5. Rhizocephala. Definition. — Crustacea which are sessile in the adult condition; the carapace (very rarely absent) forms a mantle completely enclosing the body and limbs, usually strengthened by shelly plates ; the posterior limbless part of the trunk is vestigial and usually ends in a caudal furca ; the antennules are organs of attachment, becoming vestigial in the adult, and the antennae generally disappear ; man- dibles without palp ; typically six pairs of biramous cirriform trunk- limbs ; usually hermaphrodite, female genital apertures on first trunk-somite, male apertures behind last pair of limbs ; paired eyes absent in adult ; development with metamorphosis ; young generally hatched in nauplius stage and passing through a so-called " cypris stage " with bivalved shell. Historical. — Some of the Cirripedia are sufficiently common and conspicuous to have attracted attention from remote times. They are probably referred to by Aristotle, and they formed the subject of a curiously persistent mediaeval myth, current in literature from the twelfth to the beginning of the eighteenth century, regarding the origin of the Barnacle Goose. While the earlier systematists not unnaturally classed the barnacles and acorn-shells as Mollusca, it seems strange to find this view of their affinities surviving the 106 THE CIRRIPEDIA 107 anatomical investigations of Cuvier. Lamarck, who was the first to give them the name " Cirrhipedes " (later corrected by Burmeister to Cirripedia), referred more or less vaguely to their affinities with Crustacea, as did also Oken and others, without definitely removing them from the Mollusca. It was not until J. Vaughan Thompson, in 1830, described the development of Balanus from the " Cypris " larva that the Crustacean nature of the group was placed beyond dispute. A little later Burmeister (1834) and Thompson himself (1835) completed the out- line of the life-history by discovering the earlier nauplius stage, confirming the long -forgotten observations of Slabber, who had figured the nauplius of Lepas as early as 1767. Although notable contributions to the anatomy of the group were made by Martin Saint- Ange (1835) and others, its taxonomy remained in the hands of conchologists, and new genera and species were established on the superficial characters of the shell alone. It is curious to note, as a survival of this period, that so recently as 1906 it was thought necessary to include a chapter on Cirripedes in a conchological work. Darwin's Monograph of the Cirripedia (1851-54) placed the study of the group upon a new basis, and indeed still remains the chief work of reference on the subject. The discovery of the " complemental males " and of the remarkable genera Cryptophialus and Proteolepas (the latter not since re-observed) are due to Darwin, while his systematic treatment of the normal Cirripedia (Thoracica) has scarcely been modified by subsequent workers, except as regards the addition of new species, for the most part from the deep sea. Some anatomical errors in Darwin's work were soon after corrected by Krohn (1859). J. V. Thompson had already in 1836 pointed out the resemblance of the nauplius larva of Sacculina to that of the Cirripedia, arid Lilljeborg (1859-60) established for that genus and its allies the group of " Cirripedia Suctoria," and showed that they passed through a "Cypris" stage. Fritz Miiller (1862-63) gave a more detailed account of the anatomy and larval stages of this group, to which he gave the name Rhizocephala. Delage, in a remarkable memoir (1884), made known, for the first time, the complete life- history of Sacculina, and his results, though received with scepticism by some, have recently been confirmed by G. Smith (1906). The group of Ascothoracica was established by Lacaze-Duthiers (1883) for the very remarkable genus Laura, and other genera have been added by Norman, Fowler (1889), Knipowitsch, and others. Among the more important contributions to the study of the normal Cirri- pedes may be mentioned the works of Hoek on the " Challenger " collections (1883-84), Aurivillius (1892-95), and Gravel (1904), and on the larval stages those of Glaus (1869), Groom (1895), and Hansen (1899) io8 THE CRUSTACEA MORPHOLOGY OF THORACICA AND ACROTHORACIGA. In general form and in many details of structure the Cirripedia as a whole depart more widely from the common type than do any of the other sub-classes of the Crustacea. This is correlated with the sessile habit of life which is universal within the group. In those forms which have become purely parasitic, as in the orders Ascothoracica, Apoda, and Rhizocephala, the modifications are still more profound, leading, in the last-named order, to the disappearance of every trace of Arthropod organisation. Neverthe- less, the life-history and even the minute characters of the larvae are so constant throughout the group that it is impossible to question the close relationship of the various forms. B. r+r.l.- sc- -c. — c. Kio. :>-. A, Lrjms anatifera; B, Balunvs licnnrri. (After Darwin.) c, carina ; <:!. o;iriiio-!ut<>ral ; !, lateral ; p, peduncle ; r+r.l, rostrum coalesced with rostro-lati-ial : se, scutum ; /, tergum. Leaving aside for the present the parasitic orders named above, the remaining Cirripedia, forming the orders Thoracica and Acro- thoracica (Abdominalia of Darwin), show considerable uniformity of structure, though differing widely in general appearance. In all cases, the animal is attached to some foreign object by the anterior portion of the body in the region of the antennules. The shell- fold is greatly developed, forming a " mantle " enclosing the body and limbs, and strengthened, as a rule, by shelly plates on its outer surface. Owing to a strongly marked dorsal flexure of the preoral region, the greater part of the body within the mantle comes to lie nearly at right angles to the anterior part. The anterior part of the cephalic region may be elongated into a flexible and muscular peduncle (Fig. 57, p), as in the Peclunculata, or represented only by THE CIRR1PED1A 109 R.L. FIG. 58. Turrileiius H'rii/htii. A, view of the whole fossil. B, a portion further enlarged. (', carinal scales ; C.L, cariuo-lateral ; L, lateral ; R.L, rostro-lateral ; the Opposite j;, rostral. (From Gravel's Monographic.) a flattened disc of attachment, as in the Operculata and Acrothoracica. When a peduncle is present, the rest of the body enclosed by the mantle is distinguished as the capitulum. On account of the confusion which would arise from applying the usual terms of orienta- tion to animals of such com- plex form, it is customary, in describing Cirripedia, to employ an arbitrary termin- ology in which the animal is supposed to be placed vertically with the capitulum above, and the peduncle or base of attachment below. What is, morphologically, the sternal aspect of the peduncle and the anterior side of the capitulum is called the "rostral, being the " carinal " side. The number and arrangement of the calcareous plates on the outer surface of the body afford valuable systematic characters, throwing light on the phylogenetic history of the group. It seems probable that in the most primitive Cirripedia there was no distinction of capit- ulum and peduncle, the whole body being en- veloped by a mantle, probably bivalved, and strengthened with shelly plates. Such a form is perhaps represented by the fossil Turrilepas (Fig. 58) from Silurian and Devonian rocks, in which the whole animal appears to be covered with im- C, carinal scales ; C.L, carino- bricating Scales arranged lateral ; L, lateral ; 7,'. /,, rostro - lateral ; S, rostral. . B Th-' i-apitular plates are not correctly shown. (From m transverse TOWS of five Gruvel's Moiinn-raiihie.) i -j rr>i_ on each side. Ihe genus Loriculu (Fig. 59), appearing in the Lower Cretaceous, has been supposed to represent the next stage of evolution, showing the beginning of the differentiation between peduncular and capitular plates. In the peduncular region the arrangement of the plates is the same as in Turrilepas. At one end of the animal, several C.L. L. FIO. 59. THE CRUSTACEA plates of the terminal row on each side are much enlarged and represent the capitular plates. Much older than Loricula, however, and probably much more primitive, is the still-existing genus Pollicipes, Avhich dates back to the Silurian, possibly to the Ordovician epoch. In Pollicipes the peduncle is covered with small scale-like plates which increase in size towards the capitular end, and in some cases (P. serins (Fig. 60)) show a complete gradation of size and arrangement up to the capitular plates themselves. The latter consist of unpaired rostrum and carina, with paired terga and scuta, with a varying number of lateral plates, forming a transition to the scales of the peduncle. From the type of Pollicipes two lines of modification may be FIG. 60. • Pollicipes serttis, showing transition from peduncular scales to capitular plates. (From Gravel's Monographic.) FIG. 61. Scalpcllumstearnsii, x §. (After Stebbing, from ]•:>>• cycl. Brit.) traced. On the one hand, in the group Pedunculata, we find the scales of the peduncle becoming horny and disappearing, while the capitular plates diminish in number as we pass from Scalpellum (Fig. 61), through forms like Lepas (Fig. 57, A) and Dichelaspis, to Alepas and Anelasma, where the mantle is entirely membranous. On the other hand, the Operculata may be supposed to have originated from a form resembling Pollicipes, or intermediate between it and Loricula. The peduncle has disappeared, but the whorl of plates immediately below the capitular valves have persisted, and, together with the rostrum and carina, have become united to form the outer " wall " of tubular or conical form, within the opening of which the scuta and terga are articulated to form the movable operculum. The most primitive genus among the THE CIRRI P EDI A HI Operculata is apparently Catophragmus (Figs. 02, 63), where the " wall " consists of eight pieces (or " compartments," as Darwin termed them), the unpaired rostrum and carina, and the paired lateral, rostro- lateral, and carino- lateral plates, and is further surrounded by several whorls of imbricating scales diminishing in size towards the periphery, and representing the armature of the vanished peduncle. In the other Operculata these scales are wanting, and a series can be traced in which the compartments diminish successively in number by coalescence, from Octomeris, which has eight plates, through Bcdanus (Fig. 57, B) with six, and Elminius with four, to Pyr- (jonui, where all the plates have coalesced and the "wall" is undivided. Each compartment presents three FIG. 6-2. Catophragmus polymerus. The upper figure represents the entire shell viewi-d from above. .S, scutum ; T, terguin, separated and further enlarged. (From Gruvel's .Mnnmjritiihie.) Diagram showing the constitu- tion of the " wall " in Catophragmus. The eight principal compartments are surrounded by several whorls of scales. (From Gruvel's Mono- divisions, a central paries flanked by two lateral portions known as radii or alae according as they overlap or are overlapped by the adjacent compartments. The exact manner in which the over- lapping takes place varies in the different genera and affords a basis for systematic divisions. Thus Darwin divided his Balanidae into two sub-families: (1) the Chthamalinae, in which the rostrum has alae on both sides, or, in other words, is overlapped by the adjacent compartments; and (2) the Balaninae, in which the apparent rostrum is really formed by the fusion of the rostrum with the rostro-lateral compartments (Fig. 57, B, r + rl), and con- sequently has radii overlapping its neighbours on both sides. By some recent writers the arrangement of the plates is interpreted somewhat differently, and, though its phylogenetic importance is recognised, the classification adopted is made to rest upon the number of compartments distinct in the adult. 112 THE CRUSTACEA Generally included with the Operculata, though possibly deserving to rank as a separate sub-order, is the curious little group of Asym- metrica, comprising the single genus Verruca. These are sessile, like the true Operculata, and have the shell composed of a wall and closed by a movable operculum. But the operculum consists of the scutum and tergum of one side only, those of the other side being fused to form one-half of the wall, which is completed, on the side of the movable opercular plates, by the greatly developed and displaced rostrum and carina. FIG. 64. Diagrammatic vertical section of Halanm. The cirri are cut short. A, amis ; ant, antennule; S, mouth; C, carina; c, cerebral ganglion; cap, lamellae of the "wall"; <•/, cirri; c.p, parietal canal; c.pal, mantle - cavity ; c.r, radial canal of the base; c.t, testicular caeca; E, stomach; gl.ce, cement-glands; i, intestine; in/, " infundibulum" (con- taining a prolongation of the mantle) ; l.ext, external lamina of the wall ; m.a.s, adductor scutorum muscle; m.d.s, depressor scuti muscle; rn.d.t, depressor tergi muscle; «, ventral nerve-mass ; o, opening of mantle-cavity ; ov, ovary ; p, penis ; R, rostrum ; K, scutum ; s.oe, egg-mass in mantle-cavity ; s.ro, rostral blood-sinus ; T, tergum; v.s, seminal vesicle. (From Gravel's Monographic.) In the Pedunculata the shell is formed of simple calcified plates, increasing in thickness by the application of successive layers on the inner surface, while the uncalcified cuticle between them scales off periodically to admit of growth, with the formation of a new cuticle underneath. In the simpler Operculata (Chthamalus), the compartments forming the wall are of this nature, but in most of the genera composing this group they have a more complex structure, being traversed by canals running parallel with the surface and containing prolongations of the hypodermis (Fig. 64, c.p and MI/). The complexity of the shell reaches its maximum in the genus Caronula, the species of which attach themselves to the skin of whales. In these, the folding of the wall gives rise to THE CIRRIPEDIA chambers on the inner and outer surfaces, receiving respectively branches of the ovaries of the Cirripede and prolongations of the epidermal tissue of the host. The base of the shell in the Operculata may be simple and membranous or it may become calcified, and in the latter case it may be traversed by radial canals (Fig. 64, c.r) carrying prolonga- tions of the hypodermis. In Xenobalanus (Figs. 65, 66) among the Operculata the wall is reduced to a small vestige and the opercular plates are absent. The mantle-sac is elongated and appears as if divided into capitular and peduncular portions, giving the I— jb animal an extraordinarily close re- semblance to certain Pedunculata such as AlejMi*. It may be mentioned here that in the Pedunculate Anelasma, which is parasitic on sharks, the peduncle becomes imbedded in the muscles of FIG. 65. Xenobalamis B, the reduced " wall " ; p, penis. (From Gravel's Mono- graphic.) , Fio. 66. Diagram of the reduced " wall " of Xeno- bahimtn. C, carina ; C.L+L, carino-lateral fused with lateral ; R.L, rostro-lateral ; R, rostrum. (From Gruvel's Monographic.) the host and gives off minute ramifying filaments which no doubt serve for the absorption of nutriment. The body, enclosed within the mantle, consists of a cephalic region ("prosoma" of Darwin), followed by a "thoracic" region corresponding to the somites bearing the six pairs of cirri. These somites are usually indistinctly defined in the membranous integu- ment and the first is always coalesced with the head. There is no distinct "abdomen" in the adult except in so far as it is repre- sented by the caudal furca which is present in most Pedunculata and a few Operculata. The furcal rami are usually small, unjointed or with few segments. Exceptionally they may be long and multi- articulate. Appendages. — The antennules, which serve for attachment in the larva, usually persist in a functionless condition imbedded in the cement which fastens the end of the peduncle or the base of the THE CRUSTACEA shell to the substratum (Fig. 64, ant, Fig. 67, A'}. They appear, however, to be absent in the Acrothoracica. The antennae disappear in the adult in all normal Cirripedia. Their possible persistence in the Ascothoracica will be referred to later. The mouth-parts consist of simply formed mandibles, maxillulae, and maxillae, the last united at the base to form a lower lip. The upper lip is large, often bullate, and at the sides, between it and the mandibles, are a pair of setose lobes which have been sometimes in- terpreted as lateral lobes of the labrum, but which seem undoubtedly to be the mandibular palps; The appendages of the " thoracic " region, of which there are normally six pairs, form the characteristic "cirri" from which the name of the order is ' derived (Fig. G 7, Cf). Each consists of a two -segmented protopodite bearing two long multiarticulate rami, the seg- ments of which are fringed with long hairs forming, when the cirri are pro- truded from the orifice of the shell, a " casting-net " for the capture of prey. The cirri increase in length and in the number of segments in the rami from before backwards, and the number of segments also increases with the age of the animal. The first pair are com- Flo 67 monly, at least in the Pedunculata, Dissection of Lepas from the side, separated by a little space from the following pairs, and more closely associated with the mouth - parts. Further, the first two or the first three pairs are distinguished from the posterior pairs by being shorter and by having the segments beset with stiff spines which prob- ably aid in the prehension of food. In the parasitic Anelasma the cirri are short, obscurely segmented, and quite devoid of setae (Fig. 68). In the Acrothoracica the cirri are reduced in number by the disappearance of the second and sometimes also of the third pair, and the first is separated by a wide space from the remaining pairs, which are crowded together at the posterior end of the body. The first pair are reduced to small papillae in Cr/ipfophialus, but in the remaining genera they are closely approximated to the mouth-parts, A', antennule ; C, carina ; Cd, cement gland and duct ; Cf, cirri (thoracic appendages) ; L, hepatic caeca ; M, adductor muscle ; Od, oviduct ; Ov, ovary ; P, penis ; Sc, scutum ; T, testis ; Te, tergum ; Yd, vas deferens. (From Claus's Textbook). THE CIRRIPEDIA and have two unsegmented rami. In Alcippe the last three pairs of cirri are uniramous. In many Pedunculata a membranous process (" filamentary appendage," Darwin) grows out from the side of the body just below the origin of the first cirrus. In Conchodernm similar appendages are also attached to the peduncles of some of the cirri, and in that genus and in Pollidpes others spring from the sides of the prosoma. These processes may be occupied internally by diverticula of the testes. In many Balanidae a setose plate is produced from the outer side of the peduncle of the third pair of cirri, and projects half-way across the dorsal surface of the body. In Cri/ptophicdus two pairs of filamentary processes spring from the dorsal surface of the body. It seems possible that some of those appendages which are attached to the peduncles of the cirri or in the neigh- bourhood of their articulation to the body may be of the nature of epipodites. In many Cirri- pedia a fold of the integument projecting into the mantle-cavity originates on each side of the cephalic region at the point where the mantle joins the body. In the Pedunculata these folds, the " ovigerous frena " of Darwin, usually serve for the attachment of the egg-masses, and are equipped for that purpose with knobbed spines. In the Verrucidae they are absent, but in the other Operculata they are represented by large plicated membranes no doubt branchial in function. Alimentary System. — The stomodaeum appears, as a rule, to form only the narrow oesophagus, the large stomach being without a cuticular lining. 2££ldMto* Thow'6 the In some Acrothoracica (Cryptophialus). however, body and the reduced ... . • j I j r i.v cirri- dwarf male of 7W« i//-i'• shell; sc, scutum; t, tergum ; x, dorsal fold; y, ventral fold. (From Korschelt and Holder's Embryology.) MORPHOLOGY OF ASCOTHORACICA. In the Ascothoracica, all of which are parasitic in Zoantharia or Echinoderma, the mantle may have a bivalved form (Synagoga and Pe.trarca), or it may form a capacious sac (Laura, Fig. 76) much larger than the body, with which it is connected by a narrow neck, and having only a small opening to the exterior. In Dendrogaster (Fig. 78) the mantle is still more developed, and is produced into branched lobes. In Laura the mantle is covered with stellate papillae penetrating the tissues of the host, and pre- sumably absorptive. In all cases the mantle contains ramifications of the enteric diverticula and portions of the gonads. The body 126 THE CRUSTACEA in Laurci (Fig. 77) is distinctly segmented into six "thoracic" and three limbless " abdominal " somites, and ends in a caudal furca. In Petrarca and Dendrogaster the body is unsegmented. In the three genera just named a pair of preoral appendages are present (Fig. 77, ant) and, except in Laura, are armed with hooked spines suggesting that they are organs of fixation. They differ from the adhering antennules of the Thoracica and Acrothoracica in being inserted, at least in Laura, at the sides of the buccal region and more or less enveloped by the mantle. It is possible that they are in reality the antennae, but there seems to be no definite evidence FIG. 76. Laura gerardiae. A, external view of the animal attached to a branch of the coral Gerardia (soft tissues of coral removed). B, the mantle-sac split open showing the body in the middle. The ramified hepatic diverticula, which are accompanied by branches of the ovary, are seen in the mantle on each side. (After Lacaze-Duthiers, from Encycl. Brit.) on the point. The cement-glands, so characteristic of other Cirri- pedia, appear to be absent. The mouth-parts are more or less reduced, but appear to be adapted for piercing. The thoracic appendages are biramous and articulated only in Synagoga. In Laura they are uniramous and indistinctly segmented, and the first pair differs from the others, being long and slender. A similar difference is observed in Petrarca, where, however, they are still further reduced, and in Dendrogaster they are represented only by some indistinct papillae. In all three genera the gut ends blindly, and the hepatic diverticula (Fig. 76, B; Fig. 77, F), which are large, extend into the mantle. A digastric "adductor" muscle is present on the THE CIRR1PEDIA 127 ventral side of the alimentary canal. The nervous system is reduced. An eye is stated to be present in Synagoga. In Laura the oviducts open at the base of the first pair of cirri. In the other genera they have not been traced. Petrarca is hermaphrodite, and the vasa deferentia open on a large penis which terminates the body. In Laura the testes are described as lodged in the bases of the second, third, fourth, and fifth cirri, opening to the exterior by numerous fine ducts, but this account is open to doubt. Dwarf males have recently been described in Dendrogaster. FIG. 77. Lnura gcmrdiae. A, body exposed by removal of one-half of the mantle, ant, antenna ; F, hepatic diverticula ; /, caudal furca ; g.s, cerebral ganglion ; i, intestine ; o, opening of mantle ; on, ovary ; o. 9 , opening of oviduct ; o. £ , opening of male ducts ; sac, wall of mantle. B, one of the papillae of the mantle, fi, vascular filaments ; v, blood-vessel. (After Lacaze- Duthiers, from Gravel's Monographic.) A nauplius larva peculiar among the Cirripedia in lacking the fronto-lateral horns of the carapace has been observed in Laura. In Dendrogaster the nauplius stage is suppressed and the larva hatches as a peculiar Cypris larva (Fig. 78, C), with only five pairs of biramous thoracic limbs, a long abdomen of five somites, stout antennules with hooks, and a very large olfactory filament (aesthetasc). There are no eyes, and the gut already sends a diverticulum into each valve of the shell. 128 THE CRUSTACEA MORPHOLOGY or APODA. This order was established by Darwin for the reception of a single species, Proteolepas bivincta, of which he found a solitary specimen parasitic within the mantle-cavity of a pedunculate Cirripede, Alepas cornuta, from the West Indies. No further specimens have been seen by later investigators. Proteolepas (Fig. 79) is referred to the Cirripedia mainly because it possesses adhering antennules which agree minutely with those of A. FIG. 78. Dendrogaster astericola. A, young specimen ; B, older specimen ; C, Cypris larva. a'r antennule ; ae, aesthetasc ; abd, abdomen ; br, supra-oesophageal ganglion ; /, caudal furca ; m.c, mouth-cone ; p, rudiment of penis ; st, stomach, sending a diverticulum into the shell- valve ; v.n, ventral nerve-mass. (After Knipowitsch.) normal Cirripedes. It differs most conspicuously in the absence of any trace of a mantle and of thoracic limbs. The body is elongated and maggot-like. It is divided into eleven segments, but as one of the segments is in front of that bearing the antennules, it seems clear that this segmentation does- not express the number of true somites present. The mouth-parts, borne on the first " segment," seem to be adapted for piercing and sucking. The labrum partly ensheathes the gnathites, of which there appear to be two pairs, turned outwards, and serrated on the outer margin. From the dorsal surface of the second segment THE CIRRIPEDIA 129 proceed two ribbon-like filaments (Fig. 79, A, ant), bearing at the tip the larval antennules, which agree in structure, as already noticed, with the type usual among normal Cirripedes. The six following "segments" were regarded by Darwin as thoracic and the three remaining as abdominal. All of them are devoid of any trace of appendages. The alimentary canal is greatly reduced. According to Darwin, only the oesophagus is present, and there is no trace of stomach, rectum, or anus. The ovaries lie at the sides of the anterior part of the body and the testes posteriorly. The vasa deferentia unite to open at the tip of the abdomen. There is no penis. 1!. enf I-'IO. 7'.'. 1'i-i'tii'li'iiiix lii ri iii-t', buccal coue : opening of genital atrium ; the outline of the colleteric gland is seen surrounding it. The outline of the testis is seen above, a little to the left of the peduncle. B, vertical section of Saceulina at right angles to the plane of the mesentery (semi-diagrammatic) ; at, genital atrium ; g, nerve-ganglion ; gl, colleteric gland opening into genital atrium ; ovy, ovary ; r, absorptive roots ; t, testis ; other letters as in A. (After Delage.) traces of segmentation and of appendages, and (excluding for the present the doubtful Sphaerothylacus) by the absence at all stages of life of an alimentary canal. The body (Fig. 80, A) has the form of a. simple sac attached by a short peduncle, from which root-like processes ramify through- out the body of the host. These absorptive roots appear to be absent in the aberrant genus Duplorlis. The visceral mass, or body proper, is completely enveloped by the mantle, which has a narrow aperture (o) capable of being closed by a sphincter muscle. In Sylon the opening is double, and in Clistosaccus and Duplorbis the mantle-cavity is completely closed. The mantle is attached to the THE CIRR1PEDIA visceral mass by a narrow mesentery (mes\ near to which on each side are the paired (more rarely unpaired) openings of the male and female generative organs. In the different genera the external form varies considerably, and with it the position of the mesentery and of the genital apertures. In Peltogaster (Fig. 81, B), which may be regarded as the most primitive form, the body has an elongated sausage-shape, with the mantle-opening at one end, and is attached by the peduncle about the middle of its length. The mesentery is longitudinal on the proximal side (next the peduncle). The genital apertures are placed on each side close to the mesentery, the female openings being nearer the end where the opening of the mantle is situated. Comparison with a normal Cirripede (Fig. 81, A), especially as regards the position of the genital apertures, suggests that the mesentery is on the dorsal side, and the mantle- FIG. 81. Diagram to illustrate comparison of Khizocephala with normal Cirripede. A, Le/iax : 15, I'llto'ii'xti'i-; (',>''«•,•/ iiin/i. »i, mantle ; »ws, mesentery ; n, tterve-gftnglion ; o, opening of mantle- cavity. <5, male generative aperture ; 9 1 female pMicrativc aperture. opening at the anterior (or rostral) end. In Sacculina (Fig. 81, C), which is parasitic on Brachyura, the whole body is flattened in the plane of the mesentery, and has assumed a secondary and superficial bilateral symmetry about a plane at right angles to this and coinciding with the median plane of the host. The mantle-aperture is in the middle of the distal edge, and the mesentery has suffered a corresponding displacement, extending from the mantle-opening to the point of attachment of the peduncle •on the side which, in the natural position, is turned towards the right side of the host. The genital openings, except that they are more widely separated from each other, occupy the same relative positions as in Peltogaster. In other genera, such as Lernaeodismx and Triangulus, the symmetry becomes still more complicated, and in Clistosaccus and Si/Ion the genital organs are unpaired. The peduncle perforates the integument of the host and gives THE CRUSTACEA off on the inside the absorptive roots which, in the case of Sacculina, penetrate into all the organs of the host, with exception of the gills and the heart, and extend to the terminal segments of the legs and into the antennules and eye-stalks. The roots are covered by a very delicate cuticle, beneath which is a layer of hypodermic cells, and the interior is occupied by reticular connective tissue, only the larger trunks having a central cavity. At the tips of the rootlets in Sacculina the outer layers are invaginated to form a cavity known as the "lagena." In the genus Duplarbis, where the root-system appears to be absent, the peduncle is hollow, its cavity communicat- ing with the closed mantle-cavity and opening at the other end into the body-cavity (haemocoel) of the host. Internal Anatomy. — Apart from a single nervous ganglion (Fig. 80, B, g) which lies close to the mesentery near the female genital openings, the only organs present are those of the generative system. The ovary (ovy) is divided into two much-lobed masses united by a median portion and giving off on each side a short oviduct, Avhich widens into a "genital atrium" (at) before opening into the mantle-cavity. The walls of this atrium, as in the normal Cirripedes, are glandular, often growing out into branched tubular glands (gl) ("colleteric" glands) supplying the material of the envelopes within which the eggs are packed in the mantle-cavity. In Sacculina the egg-masses are retained in position by barbed spines (" retinacula ") grouped on papillae on the lining membrane of the mantle. The testes (t) have each the form of an elongated sac narrowing towards a short vas deferens. The spermatozoa are filiform and actively motile. F. Miiller was the first to suggest that the Cypiis larvae, which, as Lilljeborg had previously observed, are often found attached near the mantle-opening of young specimens of Peltogaster, might be complemental males. Delage accepted this interpretation for the larvae which he observed in the same position in Sacciilina, but he was unable to obtain them alive or to observe any trace of male organs. At this stage the mantle-opening is still closed by a plug of chitin, and it is difficult to see how fertilisation could be effected. G. Smith, who has recently investigated the subject, finds that the larvae die very soon after they become attached, without developing farther. In only one instance did he find evidence of what seemed to be an abortive attempt of the cellular contents of the larva to pass into the tissues of the mantle in the way in which the contents of the larvae pass into the host in the ordinary course of develop- ment. G. Smith believes the attachment of these larvae to be an atavistic phenomenon ; that the larvae are, in fact, vestigial male individuals. The same observer has found in the mantle-cavity of THE CIRR1PEDIA 133 DuplorUs what appear to be extremely degraded, but still functional male individuals. In this genus the usual testes are absent, as they ;ire also in Sylon, but in the latter no male individuals have been found, and parthenogenesis may perhaps occur. DEVELOPMENT OF KHIZOCEPHALA. The development has been most fully worked out in the case of Sacculina by Delage, whose results have been confirmed and extended by G. Smith. The larva is hatched in the form of a nauplius (Fig. 82, A) showing a general resemblance to that of the normal Cirripedes, but differing in having no alimentary canal. The fronto-lateral horns are well developed and each contains a pair of gland-cells (gl). A process on the ventral side, called by Delage the rostrum, appears to represent the labrum, but there is no mouth. Posteriorly the body terminates in a caudal furca. The three pairs of limbs have the usual form, but the second and third pairs are without the masticatory hooks found in the normal Cirripedes. Frontal filaments (fs) are present, as is also the un- paired eye (ua) resting on a cerebral ganglion. In the later stages the thoracic somites and their appendages become differentiated within the posterior part of the body, not forming a postero- ventral process as in the normal Cirripedes. The Cifpris-stnge (Fig. 82, B) agrees in all essentials Avith that of the normal Cirripedes except" for the absence of a mouth and alimentary canal, and the vestigial condition of the mouth-parts. The fronto-lateral glands (gl) open in the usual position near the margin of the valves of the shell. The antennules consist of three segments only, and are without any adhesive disc ; the terminal segment bears two appendages which are probably sensory and a backwardly curved filamentous process which is stated to be the organ of attachment. The six pairs of thoracic limbs have the protopodite not distinctly segmented and the exopodite and en- dopodite each composed of two segments. The abdomen (ab) is unsegmented and terminates in a pair of furcal appendages. The unpaired eye persists, but beyond the muscles for moving the body and appendages no other internal organs are differentiated. It is worthy of note that there appear to be no cement-glands in connection with the antennules. After a free-swimming life of three or four days, the Cypris larva becomes attached to the host. In the case of Peltogaster, which infests hermit-crabs, it is probable that the larva settles at or near the spot where the .adult afterwards appears on the host's abdomen. In Sacculina, which attacks Brachyurous crabs, however, the place of attachment of the larva has no relation to the place of emergence of the adult parasite, the latter being always found under 134 THE CRUSTACEA the crab's abdomen in a position inaccessible to the larva. As a rule, the crab is attacked when the integument is still soft after ecdysis, and the larvae may attach themselves at any point on the dorsal surface of the carapace or on the limbs, at the base of one of the large setae where the articular rins; of uncalcified cuticle allows FIG. 82. Larval stages of Sacculina carcini. A, nauplius aftiT the first moult. B, free-swimming Cypris-stage. C, Cypris-stage after attachment to a seta (bV) of the host. D, formation of the Kentrogon larva. B, Kentrogon-Kt&ge, after the Ci/pris shell has been cast off and the "dart" lias been formed. F, penetration of the dart through the cuticle of the host. 1, 2, 3, the three pairs of nauplius-limbs. I -VI, thoracic limbs, alt, abdomen; bb, seta of the host; /," fat- globules ; f.s, frontal sense-organ ; gl, glands of the antero-Iateral horns ; ov, mass of mesoderm cells regarded by Delage as the rudiment of the ovary ; pf, " dart" ; ?<«, nauplius-eye. (After Delage, from Korschelt and Heider's Embryology.) more easy penetration (Fig. 82, C). Only one of the antennules is used for attachment, the filamentous process of the terminal segment clasping the base of the hair. The whole of the thoracic region of the larva, with its appendages and muscles, now becomes deta'ched and is thrown off, and later the bivalve shell is also shed, while the contents of the anterior region of the body become THE CIRRI P EDI A '35 retracted' and enclosed in a new cuticle, remaining connected only with the antennule which is fixed to the host (Fig. 82, D). Within the sac thus formed, a tubular chitinous organ known as the " dart " becomes differentiated. At first the dart is invaginated into itself, and in connection Avith it a second cuticular sac becomes separated within the first. The larva at this stage was designated by Delage the "Kentrogon" (Fig. 82, E). The point of the dart lies within Sections through successive stages in development of the "nucleus" of the fran-iiHim interan. -, from Korschelt and Heider's Embryology.) According ti> G. Smith, the mantle-cavity is already established when the inva.yination to form the peri- .somatic cavity takes place as in A. the fixed antennule, and when fully formed it becomes evaginated and forces its way through the cuticle of the host (Fig. 82, F). Through it the contents of the sac, consisting of a mass of undifferentiated cells surrounded by an ectodermal layer, pass into the body-cavity (haemocoel) of the crab, and become what Delage called the Sacculina internet. Probably the mass of cells is at first carried passively by the blood-currents of the host, but it ultimately becomes attached to the lower surface of the intestine 136 THE CRUSTACEA immediately behind the stomach. It now begins to send out processes which branch to form the absorptive roots, while the main part of the embryo grows backwards along the intestine towards the point at which emergence of the adult S'acculina takes place. A thickening near the growing border forms FIG. S4. Later stages in the development of Siunilinn. A, Sn,;-nruiii internu fully formed. B, Sacculiiu! externa just after emergence from the body of the host, a, genital atrium ; am, outer layer of mantle ; K, basal plate ; b, mantle-cavity ; C, central tumour ; d, opening of mantle-cavity ; D, intestinal wall of host ; ilr, colleteric gland ; /, opening of perisomatic cavity ; g, nerve-ganglion ; im, inner layer of mantle ; L, integument of host ; ov, ovary ; p, perisomatic cavity ; pe, ectoderm of visceral mass ; R, root-processes ; t, testis. (After Deluge, from Korschelt and Heider'.s Embryology.) the first rudiment of the organs of the adult, and is known as the "nucleus" (Fig. 83). Within this the mantle-cavity arises by delamination, and an outer cavity concentric with it, the " peri- somatic " cavity (p), is formed (according to G. Smith) by an in- vagination of ectoderm. When the young Sacculina is completely formed (Fig. 84, A) it has been carried" by its backward growth into the abdomen of the host, where it lies very close to the ventral THE CIRRIPEDIA 137 integument. Its presence causes a degeneration of the muscular and hypodermic tissues between it and the external cuticle, and at the next moult of the crab a hole is left through which the body of the parasite, protruding from the perisomatic sac, emerges on the surface (Fig. 84, B). The mantle-opening is at first closed by a plug of chitin, but in other respects the young Sacculina e.'-tcnni has already the essential structure of the adult. The development of Peltogaster seems to follow much the same course as that of Sacculina, except that no perisomatic cavity is formed. In this genus also the parasite penetrates the cuticle and reaches the exterior without any moult of the host. Very little is known of the development in other genera of Ithizocephala. In the genus Thompsonia the larvae are stated to reach the < '///•- stage while still within the mantle-cavity of the parent. APPENDIX TO KHIZOCEPHALA. The genus Sphaerothylacus, described by SI niter, is parasitic on a simple Ascidian (Polycarpd), living attached by ramifying roots to the inner wall of the branchial sac. The globular body is enclosed in a mantle which has a small opening. There are no appendages, but there is a complete alimentary canal with mouth and anus, the latter near the mantle-opening. The two ovaries each consist of two long caeca which unite to open, with those of the opposite side, into a common atrium near the mouth, surrounded by a mass of glandular tissue, no doubt representing the colleteric glands. The testes are paired simple tubes, opening close to the anus. The nauplius larvae differ from those of the true Rhizocephala in the absence of fronto-lateral horns. The genus Sarcotaces, Olsson, comprises two species which live embedded in the muscles of fish. Their structure is very imper- fectly known, but an alimentary canal is said to be present and there are no " roots." The nauplius larva is without fronto-lateral horns. The systematic position of both genera is quite obscure, and only further investigation can determine what relation, if any, they bear to the true Rhizocephala. HKMARKS ON HABITS, ETC., OF CIRPJPKDIA. The Cirripedia are exclusively marine, only a very few species penetrating into brackish water. In correlation with their sedentary habits the non - parasitic Cirripedia have developed a peculiar method of feeding by sweeping the water for floating particles of nutriment, with a net formed by the tendril -like branches of the thoracic limbs. Most species 138 THE CRUSTACEA are attached to rocks, seaweeds, etc., at the bottom, but the species of Lcpas and other genera are found on floating timber and ships' bottoms, while in L. fascicularis, the colonies of which gather round small floating objects such as dead Velellae or Spirula shells, additional buoyancy is gained by a vesicular mass of secretion from the cement-glands. Many Operculata are found attached to or embedded in corals, while the pedunculate Litholrya and the Acrothoracica burrow into corals and the shells of Mollusca. Various species of Pedunculate and Operculata are constantly found attached to large marine animals such as whales, turtles, and sea- snakes, or to the limbs and gills of large Decapod Crustacea. In the case of the operculate 1'ubicinella, found on whales, the shell becomes deeply embedded in the epidermis of the host. The line dividing commensalism from parasitism is definitely crossed by the pedunculate Anelqsma, in which the mouth-parts and limbs are reduced, and the peduncle, embedded in the flesh of a shark, absorbs nourishment by ramified " roots." From this it is but a step to the parasitic and degenerate Ithizocephala, of which the habits have already been indicated. The smallest Cirripedia are found among the Acrothoracica, some species of which are only two or three millimetres in length. Most of the Thoracica are much larger, the bulkiest being Balanus psittaeus, of which the shell is stated to reach nine inches in height by two or three inches in diameter. The peduncle of Lepas aimtifera may grow to 16 or 18 inches long. PALAEONTOLOGY. The characters and phylogenetic importance of some of the Palaeozoic Cirripedia have already been alluded to. It may be added that, like Pollicipes, the still existing genus Scalpellum dates back to the Silurian, and that both are well represented in the later Secondary rocks. Of the extinct genera, the Palaeozoic Turrilepas and the Cretaceous Loricula, already mentioned, are the most important. The earliest undoubted Operculate is Verruca (Asym- metrica) from the Upper Cretaceous. Many of the existing genera of Pedunculata and Operculata are found fossil in Tertiary deposits. AFFINITIES AND CLASSIFICATION. The great structural differences separating the Cirripedia from the other Crustacea show that they must have diverged very early from the main line of Crustacean descent. The simple biramous form of the trunk-limbs and their number have been regarded as indicating an affinity with the Copepoda, but there is little else to THE CIRRIPEDIA 139 support this vie\v. The resemblance of the bivalved shell of the Cypris larva to that of an Ostracod extends to such characters as the asymmetry of the valves and the presence of fronto-lateral glands near the margins of the shell ; but the rest of the organisa- tion differs widely from that of the Ostracoda. Among the characters which are probably significant of the isolated position of the Cirripedia are the difference in position of the genital apertures in the two sexes and the fact that the female openings are placed farther forward than in any other Crustacea. The nature of the larval development, especially the sudden transition between the sharply contrasted nauplius and Cypris stages, indicates a high degree of specialisation, and the structure of the adult animals is in many respects so clearly correlated with the sessile habit and the mode of obtaining food as to afford little help in deciphering their phylogeny. It may be noted here that, like other groups of sedentary organisms, the Cirripedia show a tendency to the assumption of a superficial radial symmetry, which becomes very marked in the Operculata. The Cirripedia were divided by Darwin into three orders, Thoracica, AbdonrinaMa, and Apoda. The order Abdominalia contained the single genus Cryptophiahis, which Darwin, misled by a superficial segmentation of the body, supposed to possess ap- pendages on the abdominal region. It has been shown, however, that the appendages in question are really thoracic, and that Cryptophiahis does not differ in this respect from Alcippe, which Darwin (although recognising the resemblance between the two genera) placed among the Thoracica. In the classification of Gruvel, adopted here, these two genera and some allied forms are grouped together as Acrothoracica, but it is to be noted that the differences separating this order from the Thoracica are probably less important than those distinguishing the other orders, and that the recently- described Koleolepas of Stebbing helps to unite the two. The genus Proteolepas, the sole representative of the order Apoda, is still known only from Darwin's description of a single specimen and its affinities are obscure. In the absence of a mantle it differs widely from all other Cirripedia. The Rhizocephala are plainly characterised as Cirripedia by their larval stages, but much remains to be done in elucidating the relationships of some of the genera at present referred provisionally to this order. The Ascothoracica form another order, established since the date of Darwin's work, but many points in their morpho- logy and development are still too imperfectly known to allow of their systematic relations being precisely defined. THE CRUSTACEA SUB-CLASS CIRRIPEDIA. ORDER 1. Thoracica. A mantle present. Six pairs of cirriform trunk-appendages. SUB-ORDER 1. Pedunculata. Peduncle and capitulum distinct. Outer plates of shell, when present, not forming a "wall." Family LEPADIDAE. Lepas, Linn. (Fig. 57, A) ; Diclielaspis, Darwin ; Pollicipes, Leach (Fig. 60); Scalpellum, Leach (Fig. 61); Ibla, Leach (Fig. 71); Lithotrya, Sowerby ; Conchoderma, Olfers ; A lepas, Rang ; An elasma, Darwin (Fig. 68) ; Koleolepas, Stebbing. SOB-ORDER 2. Operculata. Xo peduncle. Scuta and terga forming a movable operculum. Outer plates of shell coalesced to form a " wall." TRIBE 1. ASYMMETRICA. Scutum and tergmn of one side movable, without depressor muscles. Family VERRUCIDAK. Verruca, Schum. TRIBE 2. SYMMETRICA. Scuta and terga of both sides movable, with depressor muscles. Family BALANIDAE. Balanus, Lister (Fig. 57, B) ; Coronula, Lamarck ; Tubicinella, Lamarck ; Xcnobalam^ Steenstrup (Fig. 65) ; Elminius, Leach ; Pyrogoma, Leach. Family CHTHAMALIDAE. Chtha- malus, Ranzaiii ; Catophragmus, Sowerby (Fig. 62) ; Octomeris, Sowerby. ORDER 2. Acrothoracica. A mantle present. Trunk -appendages reduced in number, the posterior pairs widely separated from the first pair. Family ALCIPPIDAE. Alrippe, Hancock (Trypetesa, Norman) (Fig. 72). Family KOCHLORINIDAE. Kochlorine, Noll. Family CRYPTOPHIALIDAE. Cryptophialus, Darwin. ORDER 3. Ascothoracica. Mantle containing diverticula of alimentary canal. Trunk-append- ages more or less reduced. Family LAURIDAE. Laura, Lacaze-Duthiers (Fig. 76 . Family SYNAGOGIDAE. Synayoya, Norman. Family PETRARCIDAE. Petrarca, Fowler. Family DENDROGASTRIDAE. Dendrocjaster, Knipowitsch (Fig. 78). ORDER 4. Apoda. Mantle absent. No trunk-appendages. Family PROTEOLEPADIDAE. Proteolepas, Darwin (Fig. 79). THE CIRRI I1 ED I A 141 ORDER 5. Rhizocephala. Mantle present. No appendages nor alimentary canal. A system of absorptive roots nearly always present. Families not denned. Probably several should be recognised. Pelto- f/aster, Rathke ; Sacculina, Thompson (Fig. 80) ; Sylon, Krcyer ; Clistwic- cus, Lilljeborg ; LartMeodiscut, F. Milller ; Triangulus, Smith ; Duplorbis, Smith ; Thompsonia, Kossmann. LITERATURE. 1. AuricUlius, C. Jr. S. Ueher einige ober-silurische Cirripetlien aus Got- land. Bih. .Svenska Vet. Akad. Handl. xviii. Afd. iv. No. 3, 24 pp. 1 pi., 1892. 2. - - Studicn iiber Cirripedien. K. Svenska Vet. Akad. Hundl. (u.f.) xxvii. (7), 107 pp. 9 pis., 1895. 3. Benu.lt, W. Zur Biologic und Anatomic von Alcijtpe lumpus. Zeitschr. wiss. Zool. Ixxiv. pp. 396-457, pis. xix.-xxii., 1903. 4. Studien an bohrendcn Cirripedien (Ordnnng Acrothoracica, Gruvel, Abdominalia, Darwin). Arch. f. Biontologie, i. (2) pp. 165-210, pis. xiv.- xvii., 1906. 5. Clems, C. Die Cypris-iihnliche Larve (Puppe) der Cirripedien. . . . Sehr. Ges. Naturw. Marburg, Suppl. Heft v. 17 pp. 2 pis., 1869. 6. Dancin, 0. A Monograph of the Sub-Class Cirripedia, 2 vols. 8vo. Ray Soc. London, 1851-1854. 7. - - A Monograph of the Fossil Lepadidae . . . 1851. A Monograph of the Fossil Balanidae ami Vemicidae, 1854. Palaeontographical Soc. 8. DcliKjc, Y. Evolution do la Sacculine. Arch. Zool. ex pur. (2), ii. pp. 417- 736, pis. xxii.-xxx., 1884. 9. Fowler, G. H. A Remarkable Crustacean Parasite [1'elrarca]. Quart. J. Mi<-r. Sci. (n.s.) xxx. pp. 107-120, pi. viii., 1889. 10. Groom. T.T. On.the Early Development of Cirripedia. Phil. Trans, clxxxv. pp. 119-232, pis. xiv.-xxviii., 1895. 11. - - On the Mouth-Parts of the Cypris-Stage of Balanus. Quart. J. Micr. Sci. (n.s.) xxxvii. pp. 269-276, pi. xxix., 1895. 12. (f r itvel., A. Monographic des Cirrhipudes on Thecostraces. 8vo. Paris, 1905 [1904]. 13. Hanscn, H. J. Die Cladocereu und Cirripedien der Plankton-Expedition. Krgeb. Plankton-Exp. ii. 58 pp. 4 pis., 1899. 14. Hock, P. P. C. Report on the Cirripedia. Rep. Voy. "Challenger," Zool. viii. 169 pp. 13 pis., 1883 ; and x. 47 pp. 6 pis., 1884. 1"). Ki-ohii, An;/. Jjeob. ii. d. Cementapparat u. d. weiblichen Zeagangsorgane einiger Cirripedien. Arch. f. Naturgesch. xxv. pp. 355-364, 1859. 16. L«.-it:i--lnitli!i'i-s, 11. ilc. Histoire de la Laura Gerardiac. Mem. Acad. Sci. France, xlii. (2), 160 pp. 8 pis., 1883. 17. Lilljeborg, W. Liriopc et Pcltoyaster, H. Rathke. Nova Acta Soc. Sri. I'lisalensis, (3), iii. pp. 1-35, pis. i.-iii., and pp. 73-102, pis. vi.-ix., 1859- 1860. 18. Mi/ 1 1 IT, Fr. Die Rhizocephalen, eine neue Gruppe schmarotzender Krebse. \\-\\. Naturgesch. xxviii. pp. 1-9, pi. i., 1862. 19. Die zweite Entvvickelungsstufe der Wurxelkrebse. Airh. Xaturgesch. xxix. pp. 24-33, pi. iii., 1863. 142 THE CRUSTACEA '20. le Roi, Otto. Dcndrogastcr arborescent und Dcndrogaster ludwiyi, zwei * entoparasitisclie Ascothoraciden. Zeitsclir. wiss. Zool. Ixxxvi. pp. 100- 133, pis. vii.-viii., 1907. , 21. Smith, G. Rhizocephala. Fauna und Flora des Golfes von Neapel. Monogr. xxix. 123 pp. 8 pis., 1906. 22. Thompson, J. Vaughan. Zoological Researches, vol. i. Svo, Cork, 1830. Memoir IV. On the Cirripedes or Barnacles, pp. 69-88, pis. 9 and 10. 23. Discovery of the Metamorphosis in the Second Type of the Cirripedes, viz. the Lepades. . . . Phil. Trans. 1835, Pt. ii. pp. 355-358, pi. vi. 24. Natural History and Metamorphosis of ... Sacculina carcini. Entom. Mag. iii. pp. 452-456, 6 tigs., 1836. 25. Weltner, W. Verzeichniss der bisher beschriebenen recenten Cirripedien- arten. Arch. Naturgesch. Lxiii. Bd. i. pp. 227-280, 1898. 26. Willemoes-Suhm, E. v. On the Development of Lepas fascicularis and the "Archizoea" of Cirripedia. Phil. Trans, clxvi. pp. 131-154, pis. x.-xv., 1876. CHAPTER VI THE MALACOSTRACA SUB-CLASS MALACOSTRACA, Latreille (1806). SERIES I. LEPTOSTRACA. DIVISION PHYLLOCARIDA. Order Nebaliacea. SERIES II. EUMALACOSTRACA. DIVISION 1. SYNCARIDA. Order Anaspidacea. DIVISION 2. PERACARIDA. Order 1. Mysidacea. „ 2. Cumacea. „ 3. Tanaidacea. „ 4. Isopoda. „ 5. Amphipoda. DIVISION 3. EUCARIDA. Order 1. Euphausiacea. „ 2. Decapoda. DIVISION 4. HOPLOCARIDA. Order Stomatopoda. Definition. — Crustacea in which the carapace is variously developed or may be vestigial ; there are typically fourteen (rarely • fifteen) trunk - somites, all of which (except the fifteenth) bear appendages ; the telson rarely has a caudal furca ; antennules often biramous ; the mandibles may/ have a palp ; the trunk-limbs are differentiated into two tagmata, a thoracic of eight and an abdominal of six pairs ; female genital apertures on the sixth, 143 144 THE CRUSTACEA male apertures on the eighth trunk-somite ; paired eyes usually present ; development usually with metamorphosis, young rarely hatched in nauplius stage. The sub-class Malacostraca includes such a diversity of forms that it will be necessary to deal more fully than in the case of the other sub-classes with the separate orders composing it. Before doing so a brief account must be given of the general type of organisation found throughout the sub-class. Apart from the fixed number of somites, to which the Leptostraca offer the only exception, the most characteristic feature of the Malacostraca is the separation of the trunk -limbs into sharply defined thoracic and abdominal tagmata. This, together with the constancy in position of the genital apertures, on different somites in the two sexes, is sufficient to demonstrate the unity of the sub-class. Leaving the Leptostraca aside for the present, the more primi- tive members of each of- the " divisions " in the scheme of classi- fication here adopted approximate to a common type of structure from which the more specialised members of each group diverge very widely. Thus, the possession of a carapace enveloping the thoracic region, movably stalked eyes, biramous antennules, a scale-like exopodite on the antenna, natatory exopodites on the thoracic limbs, an elongated and ventrally flexed abdomen, and a " tail-fan " formed by the lamellar rami of the last pair of append- ages spread out on either side of the telson, are characters common to the Mysidacea, Euphausiacea, and the lower Decapoda, and, with some modifications, to the Anaspidacea and Stomatopoda. - It seems reasonable to suppose that this combination of characters, making up what has been called the "caridoid facies," must be attributed to the hypothetical common stock of the Malacostraca, At all events, it is possible to represent, in diagrammatic fashion, a generalised Malacostracan which serves as a convenient summary of the morphology of the group (Fig. 85). Some of the characters of this type require to be considered in more detail. The antennule, as already stated, is biramous, having two flagella springing from a peduncle of three segments. Since the antennules in the other sub-classes are always uniramous, as they are in the nauplius, it seems probable that the two flagella do not represent the endopodite and exopodite. When only one flagellum is present in the Malacostraca it is the outer which persists, and it alone, as is shown by the sensory filaments Avhich it carries, corresponds to the single ramus of the other sub-classes. In certain Decapoda (Caridea) and in Stomatopoda there are three flagella, the outer flagellum being divided into two. The protopodite of the antenna is composed of two or of three THE MALACOSTRACA segments. According to Hansen the latter number is, here as elsewhere, the more primitive. The endopodite is flagelliform, and the expedite has the form of a plate, the so-called " scale " or " squama," probably of use in swimming. As a rule the first three segments of the endopodite are enlarged and are counted with the two (or three) segments of the protopodite as forming a five- (or six-) segmented peduncle. The mandible has a "palp" of three segments, never biramous. The oral edge of the mandible is more or less distinctly divided into a "molar process" and an "incisor process," and between them is armed with bristles or spines. An accessory blade, the lacinia mobilis, lying close to the cutting edge of the incisor process and apparently abd.som FIG. 85. I Marram of a .uenemlised type of Malacostraea showing the " caridoid fades." a', antennule ; a", antenna ; ciln!.*uiii, abdominal somites ; c, carapace ; e, eye ; ex, exopodite of antenna; md, mandible ; m.i:', maxillula ; inx", maxilla ; pip, pleopods ; r, rostrum ; t, telson ; th.app, thoracic appendages; HI-UJI, uropods ; <$ and 9 indicate the positions of the genital apertures in'the male and female sex respectively. formed by the enlargement of one of these spines, is found in some of the orders, and is perhaps also a primitive character. The maxillulae have two endites and a " palp " of several segments. According to Hansen, whose investigations on the skeletal framework of the mouth-parts will be often referred to, the two endites belong to the first and third segments of the appendage, a small sclerite which Hansen supposes to represent the second segment having no endite in connection with it. An exite, in the form of a rounded plate, may be present ; according to Hansen it belongs to the first segment. The maxillae are more complex in form and the primitive plan of their structure is not quite clear. Apparently there are two endites, each of which is bifid, corresponding to the second and third segments, and a palp of one or two segments. The thoracic appendages (Fig. 86) are all similar, none of them, in the primitive type, being differentiated as maxiUipeds. Each has a protopodite of two segments, the coxopodite and basipodite, with, 146 THE CRUSTACEA according to Hansen, a pre-coxal segment (pleuropodite of Coutiere) which is distinct only in the Leptostraca and Stomatopoda. The axis of the limb is continued by the endopodite which forms an ambulatory leg, while the flagelliform exopodite is used for swimming. There are live segments in the endopodite, termed by Milne - Edwards respectively the ischiopodite, meropodite; carpopodite, propodite, and dactylopodite (often abbreviated to ischium, merus, carpus, propodus, and dactylus). Hansen con- siders that the terminal claw which is sometimes distinct from the dactylopodite represents an additional segment, making, with the pre-coxa, nine segments in the axis of the limb instead of the seven usually recognised. It seems probable, however, that this claw (termed stylopodite by Coutiere) is simply an enlarged spine and not one of the segments of the limb. At the bases of the thoracic limbs on the outer side are a series of epipodial appendages (exites) probably originally branchial in function. It is not quite clear how many of these appendages must be attributed to each thoracic limb of the primitive type, but probably at least two are to be recognised, an Diagram of a Malacostracan thoracic W<>dite attached to the COXOpodite appendage. Its, basipodite ; carp, carpo- and a pwepipodite to the pre-COXal segment. The oostegites, or brood- plates, attached to the inner side of the coxopodite in the female sex in some of the orders, and forming a pouch for the protection of the eggs and young, may possibly be derived from some of these epi- podial structures, as Glaus suggests. The terminology applied to the thoracic limbs in systematic works differs greatly in the various orders of Malacostraca. From one to three of the anterior pairs may be called maxillipeds, the second and third pairs are in some cases known as gnathopods, and the last five pairs are often termed peraeopods. ! The abdominal appendages are all biramous and are used in swimming, but the sixth pair differ in form and function from the others. The first five pairs are known as pleopods. They have the protopodite composed of two segments (occasionally there are traces of a third), and the rami are fringed with long setae and assist the thoracic exopodites in the ordinary swimming move- ments of the animal. The appendages of each pair are coupled together by a group of hooked spines (retinacula) either on the inner edge of the protopodite or on a special process of the FIG. 86. podite; ex, coxopodite; daet, dactylopodite; en, endopodite ; ep, epipodites ; ex, exopo- dite ; isch, ischiopodite ; mer, meropodite ; prop, propodite ; prot, protopodite. THE MALACOSTRACA 147 endopodite known as the appendix internet. The sixth pair of abdominal appendages, known as the uropods, are larger than the others, with a short, unsegmented protopodite and broad lamellar rami which lie at the sides of the telson and form with it the " tail-fan " which is used in swimming, or rather springing, back- wards by sudden flexion of the whole abdomen. The Leptostraca alone among the more primitive orders of existing Malacostraca stand apart from the scheme outlined above, and seem to have diverged from the main line of Malacostracan descent before the assumption of the caridoid form. Their systematic relations will be discussed more fully later. CLASSIFICATION OF THE MALACOSTRACA. The group Malacostraca, established by Latreille in 1806, has been accepted as a natural division by nearly all subsequent writers. Almost the only divergences of opinion as to its limits have had reference to the Leptostraca, which many zoologists following Milne -Ed wards have referred to the Branchiopoda or have regarded as occupying an intermediate place between Malacostraca and "Entomostraca." Claus's investigations on the structure of Nebalia, however, have been generally accepted as demonstrating its Malacostracan affinities. In the arrangement here adopted (following Grobben) the order Nebaliacea is included within the sub-class Malacostraca, but its distinctness from the other orders is marked by placing it in a separate division (Leptostraca) opposed to the other orders grouped together as Eiimalacostraca. In the arrangement of the Eumalacostraca most carcinologists hitherto have followed the lines laid down by Leach, who, in 1815, divided the group into two legions — the Podophthalma and the Edriophthalma — according to the condition of the eyes, movably pedunculate in the one and sessile in the other. As originally defined, the two groups Avere also distinguished from each other by the presence in the Podophthalma of a carapace which was absent in the Edriopbthalma, this character giving occasion for the names Thoracostraca and Arthrostraca applied to the same groups by Burmeister in 1834. The progress of research, however, rendered it increasingly difficult to frame satisfactory definitions of the two divisions. Thus, the sessile-eyed Tanaidacea were found to possess a true, though reduced, carapace, while the Cumacea were still more plainly intermediate between the two groups. The recent discovery of the very remarkable genus Anaspides, which has stalked eyes but no carapace, and the closely allied Koonunga with sessile eyes, makes the retention of the old arrange- ment quite impossible. 148 THE CRUSTACEA An important departure from the line of classification generally followed was made in 1883 by Boas, who showed that the order " Schizopoda " as then understood comprised two very different groups, which he separated as distinct orders, the Euphausiacea and Mysidacea. Boas discarded the old divisions Podophthalma and Edriophthalma, and divided the Malacostraca into seven orders, Euphausiacea, Mysidacea, Cumacea, Isopoda, Amphipoda, Decapoda, and Squillacea (or Stomatopoda). Hansen, in 1893, carried the reform of Ijhe classification a step -further. Setting apart (as Huxley had previously done) the aberrant Stomatopoda, as well as the Leptostraca, he showed that the remaining Malacostraca fell into two well-defined groups, the line of division passing through the old order Schizopoda ; on the one side he placed the Euphausiacea with the Decapoda, and on the other the Mysidacea with the Cumacea and the Edriophthalmate orders Tanaidacea, Isopoda, and Amphipoda. The classification adopted here is essentially that of Hangeri, as modified and extended by the present writer in 1904. It will be convenient to give here definitions of the main groups into which the Malacostraca are divided. The orders will be considered in greater detail in the subsequent chapters. SUB-CLASS MALACOSTRACA. SERIES I. Leptostraca, Glaus (1880). Abdomen of seven somites, the last of which is without appendages, and a telson bearing a pair of movably articulated furcal rami ; an adductor muscle runs between the two valves of the carapace ; thoracic limbs all similar, more or less foliaceous, with protopodite of three segments. SERIES II. Eumalacostraca, Grobben (1892). Abdomen of six somites (the number may be reduced by coalescence), the last of which typically bears a pair of appendages, and a telson which never bears movable furcal rami ; no adductor muscle of the carapace ; tboracic limbs rarely all similar (Euphausiacea), typically pediform, protopodite of two segments except in Stomatopoda. DIVISION 1. SYNCARIDA, Packard (1886). wjDarapace absent -first thoracic somite fused with the head or defined therefrom by a groove^protopodite of antenna of two segments ; mandible without lacinia mobilis ; thoracic legs flexed between fifth and sixth segments ;vno oostegites ; no appendix interna on pleopods ; hepatic caeca numerous ; heart much elongated, tubular. • THE MALACOSTRACA 149 DIVISION 2. PERACARIDA, Caiman (1904). Carapace, when present, leaving at least four of the thoracic somites distinct; first thoracic somite always fused with the head ;* protopodite of antenna typically of three segments ; mandible with lacinia mobilis (except in parasitic and other modified forms) ; thoracic legs flexed between fifth and sixth segments ;• oostegites attached to some or all of the thoracic limbs in female, forming a brood-pouch ; no appendix in- terna on pleopods ; ' hepatic -caeca few and simple ; heart generally elongated, extending through the greater part of thoracic region, or displaced into abdomen ; spermatozoa generally filiform ; development taking place within the brood-pouch, young set free at a late stage. DIVISION 3. EUCARIDA,' Caiman (1904). Carapace coalescing dorsally with all the thoracic somites^ eyes pedunculate ; protopodite of antenna with, at most, two distinct seg- ments ; mandible without lacinia mobilis in adult ; thoracic legs flexed between fourth and fifth segments; no oostegites; an appendix interna sometimes present on pleopods ; hepatic caeca much ramified ; Jieart abbreviated, thoracic ; spermatozoa spherical or vesicular, often with radiating appendages ; development as a rule with metamorphosis, a free- swimming nauplius stage in the more primitive forms. DIVISION 4. HOPLOCARIDA, Caiman (1904). Carapace leaving at least four of the thoracic somites distinct ; two movable segments are separated from the anterior part of the head, bear- ing respectively the pedunculate eyes and the antennules-; protopodite of antenna of two segments ; mandible without lacinia mobilis ; posterior thoracic limbs with protopodite of three segments (the relation of the segments of the anterior thoracic limbs to those of the limbs in the other divisions is doubtful) ; an appendix interna on pleopods ; hepatic caeca much ramified ; heart much elongated, extending through thoracic and abdominal regions ; spermatozoa spherical ; development with meta- morphosis, a free-swimming nauplius stage is not certainly known. LITERATURE. (General Morphology and Classification of Malacostraca. ) 1. Boas, J. E. V. Studien iiber die Yenvandtscliaftsbeziehungen der Mala- kostrakeri. Morphol. Jahrb. viii. pp. 485-579, pis. xxi.-xxiv., 1883. 2. Caiman, W. T. On the Classification of the Crustacea Malacostraca. Ann. Mag. Nat. Hist. (7) xiii. pp. 144-158, 1904. 3. G'laus, C. Neue Beitrage zur Morphologic der Crustaceen. Arb. zool. Inst. Wien, vi. pp. 1-108, pis. i.-vii., 1885. 4. Hanscn, H. J. Oversigt over de paa Dijmphna-Togtet indsamlede Krebsdyr. Dijmphna-Togtets zool.-bot. Udbytte, pp. 185-286, pis. xx.-xxiv., and Resmne, pp. 508-511, 1887. ISO THE CRUSTACEA 5. Hansen, U. J. Zur Morphologic der Gliedmassen und Mundtheile bei Crustaceen und Insecten. Zool. Anz. xvi. pp. 193-198 and 201-212. Translated in Ann. Mag. Nat. Hist. (6) xii. pp. 417-434, 1893. 6. Sars, G. 0. Histoire naturelle des Crustaces d'eau douce de Norvege, 1 Livr. Les Malacostraces, pp. iii + 145, 10 pis. 4to. Christiania, 1867. 7. Stcbbing, T. E. R. A History of Crustacea. Recent Malacostraca. Internat. Sci. Ser. London, 1893. THE LEPTOSTRACA SERIES LEPTOSTRACA, Glaus (1880). DIVISION PHYLLOCARIDA, Packard (1879). Order Nebaliacea. Definition. — To the characters mentioned on p. 148 as distinctive of the series the following may be added : Carapace present ; all the thoracic somites distinct ; eyes pedunculate ; mandible with- out lacinia mobilis ; no oostegites ; first four pairs of abdominal appendages biramous, with appendix interna, last two pairs reduced ; hepatic caeca few ; heart elongated ; development embryonic, young set free at a late stage. Historical. — The first -known member of the Leptostraca was the Cancer bipes of O. Fabricius, described from Greenland. Leach, who in 1815 established the genus Nebalia, placed it among the Macrura, but H. Milne-Edwards, while admitting its affinities with Mi/sis, ranked it as a Phyllopod, and this view was long and widely held. Metschnikoff in 1865, from a study of its development, replaced the genus among the Malacostraca as a " phyllopodiform decapod." Glaus, in a series of memoirs ending with his exhaustive monograph of 1889, vindicated the title of Nebalia to rank as a Malacostracan, and placed it in a group Leptostraca, alongside the Arthrostraca and Thoracostraca. The resemblance of certain fossil Crustacea to Nebalia had long been recognised, and Packard in 1879 proposed the name Phyllocarida for the group, including the living and fossil genera. Sars has called attention to the similarity in general form between Nebalia and certain Copepoda, but Glaus showed this resemblance to be merely superficial. t MORPHOLOGY. The carapace (Fig. 87) is compressed laterally so as to form a bivalved shell (though without any definite hinge-line), loosely enveloping the thorax and more or less of the abdomen, and quite 151 152 THE CRUSTACEA concealing, in most cases, the thoracic limbs. The two halves of the shell can be approximated by the action of an adductor muscle traversing the body in the region of the maxillae. Anteriorly, the carapace is produced into a movably articulated rostral plate (?•), not provided with special muscles but elevated or depressed by the movements of the underlying parts. The carapace is not attached to the body behind the maxillary region, and the eight thoracic somites which, except in Nebaliopsis, are very short and crowded together, are all distinctly marked off' by grooves on the delicate -integument. Of the abdominal somites the fourth alone may have distinct pleural plates, or these may be altogether absent. The telson (t) bears the two styliforin or lamellar furcal rami (/) PIG. 87. Neballa bipes, ?, from the side, a', antennule ; a", antenna; ab*, ab$, first and sixtli abdominal appendages ; ad, adductor muscle of carapace ; / caudal furea ; p, palp of maxillula ; r, rostral plate ; (, telson ; 1, 7, first and seventh abdominal .somites. (After Glaus.) articulated with it and moved by special muscles ; the anus opens between the rami towards the ventral side. Appendages. — The antennules (Fig. 88, A) have a peduncle of four segments bearing a flagellum of varying length and, external to it, a movable scale. It seems probable that this scale represents the outer ramus of the antennule of other Malacostraca, and the occurrence of four (instead of three) segments in the peduncle is paralleled in certain species of Tanaidacea. The peduncle of the antenna (Fig. 87, a") is apparently com- posed of four segments, of which the last two are coalesced in Nebalia and Paranebalia. Hansen recognises, in addition, a short basal segment and another, very short, between the second and third of the larger segments. The exopodite is absent. The distal part of the endopodite forms a flagellum which, in the adult male, may be nearly as long as the body. The mandible (Fig. 88, B) has a strong molar process ; the THE LEPTOSTRACA 'S3 incisor process is small and simple in Nebalia and Paranebalia, large and toothed in Nebaliella, and absent in Nebaliopsis. The palp is large and has three segments. The maxillula (Fig. 88, C) has the usual two endites. The palp is vestigial in Nebaliopsis ; in the other genera it is attached to the distal margin but curves outwards and backwards so that the A. Fio. 88. Cephalic appendages of AVfcci/iVi. A, antennule ; B, mandible ; C, maxillula ; D, maxilla. (After Clans.) very long, slender, and indistinctly segmented flagellum in which it terminates is directed obliquely upwards along the side of the thorax underneath the carapace (Fig. 87, p). The maxillae (Fig. 88, D) approximate in general form to the thoracic limbs, except for the absence of an epipodite. There are generally four endites, but the distal one is much reduced in Nebalia and Paranebalia. The endopodite is divided into two segments in Nebalia and, like the exopodite, is generally elongated. 154 THE CRUSTACEA In Nebaliopsis, however, the endopodite is a short lobe and the exopodite is hardly indicated. The eight pairs of thoracic limbs are all similar, except in Nebaliopsis, and they present considerable differences of structure in the four genera. In Nebalia (Fig. 89) the whole limb is much flattened. The broad coxopodite and basipodite are distinctly separated, and on the proximal side of the former Hansen has detected a small pre-coxal segment. To the outer margin of the coxopodite is attached the broad lamellar epipodite (ep), obscurely divided into FIG. 89. First thoracic limb of Ncbalid. en, endopodite ; ep, epipodite ; ex, exopodite. (After Clans.) a proximal and a distal lobe by a slight notch on the outer margin opposite the point of attachment. The basipodite bears externally the oval flattened exopodite (ex) and is continued without any distinct line of articulation into the narrower endopodite (en). From the distal end of the endopodite three, or, in the case of the eighth pair, four segments are marked off, so that, except for the absence of an articulation between tlie basipodite and ischiopodite, all the segments of the typical malacostracan leg can be distinguished. In NebalieUa (Fig. 90, B) the epipodite is absent, the exopodite has more numerous marginal setae, and the articula- tion between the basipodite and ischiopodite -is more distinctly THE LEPTOSTRACA 155 marked. In Paranebalia (Fig. 90, A) the endopodite is long arid slender, projecting beyond the edges of the carapace in the natural A. ex. en FIG. 00. Thoracic limbs of Leptostraca. A, 1'nmnebalmlongipcs, fifth limb. B, Ncbaliellaantarctiea, fourth limb. C, Nebalwpeis tupicn, lirst limb. D, the same, second limb, en, endopodite ; ep, epipodite ; ex, exopodite. (A after Sars ; B, C, D alter Thiele.) position. The exopodite is also long and narrow, and is provided, on its outer edge, with numerous long plumose setae which suggest a natatory function. The epipodite is very small. The ischio- 156 THE CRUSTACEA podite is defined from the basipodite, at least in the last pair, and Hansen recognises a minute terminal segment as well as a pre-coxal, making altogether nine segments in the axis of the limb. In Nebaliopsis the thoracic limbs are very different from those of the other genera. Except the first and last pairs, they are unseg- mented lanceolate lamellae (Fig. 90, D), having a slight lobe on the outer edge to represent the exopodite, and a more distinctly marked bilobed epipodite ; the inner edge is beset with setae along its whole length. In the last pair, which are almost without setae, a terminal segment is marked off, and in the first pair (Fig. 90, C) this part, although not distinctly segmented off, is produced into a finger-shaped distal lobe (en). In Nebalia the tip of each of the thoracic limbs carries, in the breeding female, a fan of long plumose setae turned inwards to form the floor of a basket-like brood-chamber which is closed behind by rows of long setae on the inner edges of the last pair. The first four pairs of abdominal appendages are biramous and are used in swimming ; the last two pairs are small and uniramous. The former (Fig. 91) have a stout protopodite of two segments and long indistinctly segmented rami fringed with spines and plumose setae. From the inner edge of the endopodite close to its base there springs a short appendix inter-no, (a.i) bearing a group of hooked spines at its tip. Alimentary Si/stem. — The masticatory stomach, in Nebalia, is of a comparatively simple type. It is divided into a " cardiac " and a "pyloric" portion, the former with masticatory ridges moved by muscles, and the latter with two pairs of lateral setose lobes and a dorsal groove which is continued as a delicate chitinous funnel, open below, some distance into the mid-gut. The mid-gut extends back to the penultimate segment of the body. Four pairs of hepatic caeca open near its anterior end and a pair of short caeca open separately on the ventral side in the same region. Near its junction with the proctodaeum the mid-gut gives off an unpaired dorsal caecum, bifid at the tip. Circulatory System. — The heart, in Nebalia, extends from the cephalic region to the fourth abdominal somite. It has seven pairs of ostia, and the last pair, which are larger than the others, are situated in the region of the sixth thoracic somite. The exopodites and epipodites of the thoracic limbs are traversed by a close net- work of blood-channels and no doubt serve as respiratory organs. The valves of the carapace probably also assist in respiration, and the blood circulating in them is returned to the pericardial sinus by a definite venous channel on each side. Excretory System. — Both the antennal and the maxillary glands are present in a vestigial condition, the former lying in the proximal segment of the antennal peduncle, while the latter is 157 placed close to the adductor muscle of the carapace at the base of the maxilla. The gland in each case consists of a minute saccule giving off a short duct, but the external openings have not been detected. Eight pairs of glands lying at the bases of the thoracic limbs are also believed to be excretory. Each consists of a thickening of — en. FIG. 91. Pleopod of second pair of Nebalia. a.i, appendix interna ; en, endopodite ; ex, exopodite. (After Claus.) the hypodermis partly surrounding the efferent blood-channel from the epipodite. These glands alone become coloured by intra vitam treatment with indigo-carmine, while the antennal and maxillary glands excrete particles of carmine when the animal has been fed with that substance. Muscular System. — The adductor muscle (Fig. 87, ad) of the carapace has two muscular heads attached to the valves and con- 158 THE CRUSTACEA nected by a median tendinous part which passes under the stomach in the region of the maxillae. The muscle is innervated from the maxillary ganglion. Nervous System. — The brain is complex, admitting of close comparison with that of other Malacostraca. The oesophageal connectives are short and are united by the commissure of the antennal ganglia behind the oesophagus. In the ventral chain the ganglia of the mandibular, maxillular, and maxillary somites are distinct, though, like the eight thoracic ganglia, they are closely crowded together. Six abdominal ganglia are present, the seventh somite, like the telson, having no ganglion in the adult. In the embryo, however, a transitory seventh abdominal ganglion has been found. Sense-Organs. — The eyes are stated to resemble, in their intimate structure, those of the Mysidacea. In Nebaliella and Nebaliopsis and in one species of Nebalia (N. typlilops) the eyes are vestigial, though the peduncles persist. On the upper and inner surfaces of the ocular peduncle, in Nebalia, there are two small tubercles, supposed to be sensory organs. Olfactory filaments of the usual type are present on the antennules and, in the male, also on the antennae. Reproductive System. — The most conspicuous external difference between the sexes consists in the much greater length of the antennal flagellum in the male. The gonads have the form of paired tubes extending, when mature, through nearly the whole length of the body. The short vasa deferentia open on papillae on the coxopodites of the last thoracic limbs. The oviducts are hard to detect, but they appear to open on the sixth thoracic somite. The spermatozoa are spherical and are aggregated into globular spermatophores. The eggs are carried, as already stated, between the thoracic feet of the female. Development. — The development of Nebalia takes place within the brood-chamber of the parent without any free-swimming larval stages, and presents many points of resemblance to that of the Mysidacea. The first three pairs of appendages appear simul- taneously, giving a well-marked nauplius stage. The remaining appendages develop in order from before backwards. The embryo becomes free from the egg-membrane at a stage when all the thoracic appendages, and sometimes also those of the abdomen, are marked off (Fig. 92). The carapace at this stage does not extend beyond the second thoracic somite. When the young leave the maternal brood-chamber they have attained, in all essential respects, the structure of the adult. THE LEPTOSTRACA 159 REMARKS ON HABITS, ETC. All the Leptostraca are marine. Most of the species occur in shallow water or at moderate depths, but Nebaliopsis belongs to the " ba thy pelagic " fauna at depths exceeding 1000 fathoms. Many have an extremely wide distribution, the common European Nebalia bipes, for instance, ranging from Greenland to Chile and Japan. The species named appears to be very resistant to unfavourable conditions, i • • . vi.' Embryo of Xrlialid., lust hatched. «', antennule ; thriving in Water Wmcn IS a»( antenna; /, caudal furra ; //,, palp; s, setose lobe. 1 66 THE CRUSTACEA Sometimes, but not always, it is absent from the last pair also in Anaspides. In the males of both genera the first two pairs have the endopodites modified as copulatory organs. In the first pair of Anaspides the endopodite is a thick lobe, curved inwards and having a group of retinacula on a short process (perhaps a vestigial appendix interna) near the distal end of its inner edge. The endo- podite of the second pair is composed of two segments, the first elongated, bearing some spines and a group of retinacula near the distal end, and the second curved and spoon-shaped. In the natural position these appendages are turned forwards, the endo- podites of the second pair lying within the trough formed by the apposition of those of the first pair, and between the latter and the sternal surface of the thorax. The uropods in Anaspides are large, with lamellar rami, fringed with spines and setae, and form, with the telson, a tail-fan of the usual type. The exopodite is crossed by an incomplete suture or line of articulation near the distal end. In Koonunga the protopo- dite is relatively longer and the rami are not so broad, so that the fan-like arrangement is not quite so typical. The exopodite is undivided. As regards the internal anatomy, our information is as yet very restricted, and refers only to Anaspides. Alimentary System. — The masticatory stomach appears to be of very simple type, its armature consisting of longitudinal chitinous ridges beset with setae. The extent of the mid-gut has not been ascertained. The hepatic caeca are numerous, very long slender tubes. There are two median dorsal caeca — one in the region of the first and the other in the fifth abdominal somite. Circulatory System. — The heart is a long tube extending through a great part of the length of the body. The number of the ostia has not been ascertained. There is stated to be an unpaired descending artery originating from the under-surface of the heart between the last two thoracic somites. Excretory System. — On each side of the head, posterior to the mandibles, is a glandular mass of considerable size, showing in sections a convoluted tubular structure. No duct has yet been traced from it, but its position suggests that it may be the maxillary gland. Sense-Organs. — The paired eyes of Anaspides are set on short movable peduncles ; those of Koonunga are very small and are sessile on the sides of the head. In both genera a saccular invagination of the integument, sup- posed to be an otocyst (or statocyst), is found in the basal segment of the antennular peduncle. It opens by a small slit on the dorsal surface of the segment. Internally, on the upper side, is a row of peculiarly modified setae. Each is divided into two segments, the THE SYNC ARID A 167 distal one swollen and pyriform. While resembling in its position the otocyst of Decapods, this organ differs strikingly from it in the nature of the setae. . Reproductive System. — The ovaries form an elongated lobed mass on each side, extending through the posterior part of the thorax and into the abdomen. The oviducts open on the inner face of the coxopodites of the sixth pair of thoracic limbs. Between the bases of the last pair of legs on the sternal surface of the thorax is a rounded prominence directed forwards. At its tip a slit -like aperture gives entrance to a blind sac, with thick and apparently muscular Avails. At the base of the sac on each side is a racemose gland, apparently opening by a short duct into its cavity. It seems probable that this structure (originally described as the opening of the oviducts) is a receptaculum seminis. A similar organ is present in Koonunga. The testes are a pair of very long slender tubes, convoluted anteriorly, lying above the alimentary canal. The vasa deferentia terminate in a pair of oblique slit-like apertures on the sternal surface of the last thoracic somite. The development, unfortunately, is still entirely unknown. EEMARKS ON HABITS, ETC. Anaspides occurs in rocky pools at an elevation of about 4000 feet in the mountains of Tasmania. It reaches a length of about 38 mm. Koonunga is found in freshwater pools near Melbourne, and does not exceed 9 mm. in length. 'PALAEONTOLOGY. A group of fossil Crustacea found in Carboniferous and Permian rocks in Europe and America, for which the name Syncarida was proposed by Packard, appear to be closely allied to the living Anaspides and Koonunga. The structure is best known in the case of Uronectes (Gampsonyx) (Fig. 97), described by Jordan and von Meyer from the Lower Permian of Saarbrucken. The exact number of free somites is doubtful, but there appear to be eight in the thoracic region, and there are indications that the sixth abdominal somite was divided in a manner recalling the condition found in certain Mysidacea. The eyes are pedunculated. The antennules are biramous, and the antennae have a rounded scale- like exopodite. One of the anterior pairs of thoracic limbs is enlarged and armed with stout spines. The presence of exopodites on the thoracic limbs is probable, although denied by Fritsch, but the structure of the appendages is very obscure. The uropods are lamellar, forming a tail-fan with the telson ; the exopodites are 1 68 THE CRUSTACEA divided by a suture. In Praeanaspides (Fig. 98), recently described by Dr. H. Woodward from the English Coal-measures, the seg- mentation of the body agrees with that of Anaspides, and exopodites are certainly present on some of the thoracic limbs. Other genera are Palaeocaris, Acanthotelson, and Gasocaris. FIG. 97. Uronectes [ = Gampsonyx]fimbriatus. a', antennule ; f, supposed eye ; ex, traces of exopodites on thoracic limbs ; I, enlarged thoracic limb, probably the second ; /, tirst thoracic somite. (After Jordan and von Meyer.) AFFINITIES AND CLASSIFICATION. The existing genera of Syncarida present characters which indicate for them a very isolated place among living Malacostraca, while suggesting more or less remote affinities with widely divergent FIG. OS. Praeanasjndes praeeursor, from the Coal-measures of Derbyshire. (From H. Woodward in Ceol. Mag.) groups. They have retained characters of the primitive caridoid type in the tail-fan, the biramous antennules, the scale-like exopodite of the antenna (in Anaspides), and the natatory thoracic exopodites. With the loss of the carapace, however, the segmentation of the body comes to resemble that of the Isopoda and Amphipoda, though the demarcation of the first thoracic somite from the head (if this THE SYNC ARID A 169 be indeed the significance of the " cervical groove ") is not so distinct in any other Eumalacostraca. The homologies of the segments of the endopodite in the thoracic limbs are not quite clear, but the fact that the main flexure of the limb is between the fifth and sixth segments in Anaspides is a point of agreement with the Peracarida, to which group some slight resemblance may be traced in the structure of the maxillae. The possession of a statocyst in the basal segment of the antennule is only paralleled among the Decapoda, and the presence of a receptaculum seminis on the last thoracic sternite of the female and the modification of the first two pairs of pleopods in the male may also point to an affinity with that group. On the other hand, the double series of epipodial lamellae on the thoracic appendages of both genera, and the double gnathobasic lobes on the coxopodite of the first pair in Anaspides, are important features not found in any other Malacostraca. The fossil genera mentioned above show that already in Palaeozoic times a group of Malacostraca existed which, while retaining caridoid features in tail-fan, antennules, antennae, and pedunculated eyes, had a completely segmented body and no carapace. Anaspides alone among living Crustacea agrees with them in this combination of characters, and there appears to be no reason to doubt that it and Koonunya, are really descendants of the Syncarida of Carboniferous and Permian times. Bcdhynella natans, to which allusion has been made, is a minute Crustacean (TO mm. in length), described in 1882 by Vejdovsky from two specimens found in a well in Prague. It has not since been rediscovered. It appears to possess eight free thoracic somites. There are no eyes. The antennules are uniramous, and the antennae have a small exopodite. The first seven pairs of thoracic limbs are biramous, and each has a single vesicular epipodite. The last pair are vestigial. Only the first and last abdominal somites bear appendages. If its structure has been correctly interpreted, Bathy- nella would seem to be a degenerate member of the Syncarida, but only the discovery of further specimens will enable its systematic position to be definitely fixed. ORDER Anaspidacea, Caiman (1904). Family ANASPIDIDAE. Anaspides, G. M. Thomson. Family KOOXUNGIDAE. Koonunga, Sayce. LITERATURE. 1. Caiman, IV. T. On the Genus Aiiaspides and its Affinities with certain Fossil Crustacea. Trans. Roy. Soc. Edinburgh, xxxviii. (4), pp. 787-802, 2 pis., 1896. 2. - - On the Characters of the Crustacean Genus Hathynella, Vejdovsky. Journ. Linn. Soc. Zool. xxvii. pp. 338-344, pi. xx., 1899. 170 THE CRUSTACEA 3. Fritsch, A. Fauna der Gaskohle und der Kalksteine der Permformation Bohmeus. Bd. iv. Heft 3, Crustacea, etc., 1901. 4. Jordan, H., and von Meyer, H. Ueber die Crustaceen der Steinkohlenfonna- tion von Saarbriicken. Palaeontographica, iv. pp. 1-16, pis. i.-ii., 1854. 5. Packard, A. S. On the Syncarida, etc. Mem. Nat. Acad. Sci. Washington, iii. Pt. 2, Mem. 15, pp. Ik3-139, 4 pis., 1886. . 6. Sayce, 0. A. Description of a new remarkable Crustacean with Primitive Malacostracan Characters. Victorian Naturalist, xxiv. No. 7, pp. 117-120, 1907 ; reprinted in Ann. Mag. Nat. Hist. (8), i. pp. 350-355, 1908. 7. Thomson, G. M. On a Freshwater Schizopod from Tasmania. Trans. Linn. Soc. (2), Zool. vi. (3), pp. 285-303, pis. xxiv.-xxvi., 1894. 8. Vejdovsky, F. Thierische Orgauismen der Brunnenwiisser von Prag. Prag, 1882. 9. Woodward H. Some Coal-measure Crustaceans with Modern Representatives [Praeanaspides]. Geol. Mag. (n.s.) Dec. 5, vol. v. pp. 385-396, 1908. ADDENDUM. Since this chapter was in type G. Smith has described a third living representative of the Syncarida, Paranaspides lacustris, from the Great Lake of Tasmania, and has given further details as to the habits and internal anatomy of Anaspides ("Preliminary Account of the Habits and Structure of the Anaspididae . . .," Proc. Royal Soc. (B) Ixxx. pp. 465-473, pi. xiii., 1908). CHAPTER IX THE MYSIDACEA Order Mysidacea, Boas (1883). Definition. — Peracarida which retain more or less completely the primitive caridoid facies ; the carapace extends over the greater part of the thoracic region, but does not coalesce dorsally with more than three of the thoracic somites ; the eyes, when present, are movably pedunculate ; the antennules are biramous ; the antennae have usually a large scale-like exopodite ; the thoracic limbs (except sometimes the first and second pairs) have natatory exopodites ; the first and sometimes also the second pair are modified as maxillipeds ; a lamellar epipodite is present on the first pair ; ramified branchiae may be attached to the body-wall close to the bases of the thoracic limbs ; the pleopods are often reduced ; the uropods are lamellar, forming a tail-fan ; the young leave the brood-pouch provided with all the appendages of the adult. Historical. — The group Schizopoda, established by Latreille in 1817, and long approximated to the Stomatopoda on the authority of H. Milne -Ed wards, finds a place in most modern systems of classification, as comprising, after exclusion of many larval Decapods formerly referred to it, the forms here treated of together with the Euphausiidae. Boas, in 1883, however, discarded this grouping, and established the two orders Mysidacea and Euphausiacea, point- ing out that they were by no means closely related, and this view has been also advocated by Hansen. Our present knowledge of the structure and classification of the Mysidacea is very largely due to the work of G. 0. Sars. » MORPHOLOGY. From five to seven of. the thoracic somites are distinct, and the last two or three of these may be left uncovered by the carapace on the dorsal side. The carapace may be produced in front as a short rostrum, and there is usually a transverse groove or " cervical sulcus" (Fig. 99, c.s) on the dorsal surface in the region of the 171 172 THE CRUSTACEA mandibles. The possible homology of this groove with the cervical groove of Anaspicles has been already mentioned. The abdominal somites have the pleural plates generally reduced or absent ; but in Gastrosaccus the pleura of the first somite are greatly enlarged in the female to help in forming the brood-pouch. The FIG. 00. Mysis relicta, female, c.s, cervical sulcus ; .in, brood-pouch. (After Sars.) sixth somite is generally longer than any of the others, and in Qnathophausia it is divided by a transverse groove (Fig. 100, gr) about the middle of its length. It is possible that we have here two somites in process of coalescence, and that seven somites are represented in the abdomen of the Mysidacea as in that of the FIG. 100. Giuithophausia w'dlemoesii, Sars. (According to Ortmann this' is the fully adult form of G. zoeo, .W. Suhm.) x £. gr, groove partially dividing the last somite. (After Sars.) Leptostraca. It is important, however, to note that the last somite in this case bears appendages (uropods) and that the penultimate does not, while in the Leptostraca the reverse is the case. The indications of a similar division of the sixth abdominal somite in certain fossil Syncarida have already been alluded to. Appendages. — The ocular peduncles are peculiarly modified in many deep-sea forms in which the eyes are imperfect or absent. In Dadylerythrops the peduncle is produced as a finger-like process THE MYSIDACEA 173 — a: beyond the eye ; in Boreomysis scyphops the distal end of the peduncle is expanded and excavated in a cup -like form and is without pigment or any trace of ocular structure, but in other species of the same genus the eyes are normally developed ; in some Petal- ophthalmidae the peduncles are leaf-like or spiniform ; while in Psewlomnia, AmUyops, and some allied genera, they are repre- sented by broad plates extended horizontally in front of the carapace. In the Mysidae the three-seg- mented peduncle of the antennules carries in the male sex, in addition to the two flagella, a conical process beset with numerous sensory fila- ments. The antennae have the protopodite distinctly composed of three segments (Fig. 101, 1, 2, 3). A lamellar ex- opodite or " scale " (sc) is always present except in Arachnomysis and allied genera, where it is represented by a spine. In many Mysidae it is divided into two segments by a trans- verse suture near the tip. The mandibles have generally a well -developed lacinia mobilis (Fig. 102, l.m), differing in form on the two sides, and a row of spines (s) in- terposed between the incisor and molar processes. The row of spines is absent in the Lophogastridae and th1.— --— o. FIG. 101. Cephalothoracic region ofAfysis relicta, young female, from below. Most of the thoracic appendages have been removed, a', external flagellum of antennule ; a", . , . ttagellum of antenna ; ex, exopodites of and SOme Mysidae, and thoracic appendages ;l, labrtim ; mil, palp of mandible ; o, oostfgites, not yet fully developed ; sc, scale or exopodite of an- ,i i • • VM- • in SOme Cases the laClllia mOblilS IS •wontinfr TliA mnlnr -nrnpooa is email tftnll«a ! th l-th 3, first, second, and third wanting. nan thoracic ai,penda,,PS : ^ L)J 3; tlie three segments of the protx>podite of the an- tenna. (After Sars.) ^ Or absent in a few Mysidae. A palp . , ,* *\,r is always present, and becomes greatly enlarged in the aberrant Petalophthalmus, where it appears to have a prehensile function. The maxillulae (Fig. 103, A) have two endites arising, according to Hansen, from the first and third segments, and a slightly developed laminar exite, which Hansen states belongs to the first segment. In the genus Gnathophausia (Fig. 104) a palp of two segments is present, directed backwards beneath the carapace like that of the Leptostraca. The maxillae (Fig. 103, B) have a complex structure. There are two endites corresponding to the second and third segments (Hansen), the first of which is incompletely and the second com- 174 THE CRUSTACEA A. B. Lm. -j pletely divided into two. A plate -like lobe on the outer side (/) is regarded as the exopodite and springs from the third segment (Hansen). The palp is composed of two segments. In Gnatho- phausia a pigmented papilla on the outer side close to the base bears the opening of a gland producing a luminous secretion. The thoracic appendages have the coxopodite very small, and have usually an exopodite con- sisting of a peduncle and a multi-articulate setose flagellum attached near the proximal end of the basipodite. The ex- opodites of the first pair may be reduced (Lophogastridae, Eucopiidae) or absent (some species of Gnathophamia, Petal- ophthalmidae), and those of the A, mandible of Mi/sis. B, oral edge of same. j . , , . n , , further enlarged, i, incisor process ; l.m, lacinia Second pair are absent in retai- mobilis ; m, molar process ; s, spine-row. (A after onJitJidlmUS oars. ) -t The first pair of thoracic limbs are always specialised as maxillipeds. In the Lophogastridae and Eucopiidae they are without distinct endites. In the Mysidae (Fig. 105, A) an endite is generally borne 'by the basipodite, and sometimes also by each of the two following segments. In Petal- ophthalmus the first and second thoracic limbs (in the other genera FIG. 102. B. — f. of Petalophthalmidae only the second) have a large lamellar endite developed from the meropodite. In the Lophogastridae (Fig. 106) the last seven pairs of thoracic limbs are all similar, and exhibit the usual number of seven segments, FIG. 103. the dactylus being large and ^gff^+w^fi^^vj, having generally a claw-like respective segments according to Hansen's earlier 3 , T i interpretation (in his later papers the endites here Spine at tne apex. in tne numbered 3 and 4 in the maxilla are regarded as Fnpnrmrlap thp tjprnnrl tn thp resilltinS from the division of a single endite cor- JlrUCOpliaae tne Sec I to tue responding to the third segment, and the segments fifth pairs are Subchelate, and here numbered 5 and 6 become 4 and 5 respectively) ; , . , /, flabellum or exopodite. (After Hansen.) the next three pairs, which are exceedingly long and slender, have also the dactylopodite flexed against the propodite to form a prehensile organ. In the Mysidae there is, as a rule, a distinction between the second and the follow- ing limbs. The former, sometimes called second maxillipeds or " gnathopoda," are bent inwards towards the mouth, with the normal number of segments and with a rounded dactylopodite THE MYSIDACEA 175 without a claw. The remaining pairs (Fig. 105, B) have the propodite (except in most Petalophthalmidae and some species of ^iri'll(i) divided into secondary segments from two (Boreomysis, Siriella) to eight or nine in number. The dactylopodite is usually small, and terminates in a claw-like spine. In Hetcromysis the third pair are enlarged and prehensile. In all Mysidacea the first pair of thoracic limbs have a simple lamellar epipodlte (Fig. 105, A, ep} directed backwards beneath the carapace. In the Lophogastridae and Eucopiidae a series of ramified gills (Fig. 106, br) are developed in connection with the last seven thoracic limbs. Each consists of three or four main branches, which are again bipinnately or tripinnately divided. The largest branch is bent round on the sternal surface of the thorax between the insertion of the limbs, a point of some . interest in connection with the position of the gills in the Amphipoda. Although the gills of the Lopho- Fio. 104. Maxillula of Gnatho})hausia p, palp. (After A. d. ex. Fio. 105. A, first thoracic appendage (maxilliped) of Mysis. B, third thoracic appendage of same The minute coxopodite is omitted in each case. 6.s, basipodite ; longispina. br, branchia ; ep, epipodial process ; LllCOplldae, and r etalophthal- cx, exopodite. (After Sars.) miflap and in the jo-pn Kor mysis among the Mysidae. In the other Mysidae the number does not exceed three pairs, and it is often reduced to two, corre- sponding to the last two thoracic somites (Fig. 101, o). The pleopods are well developed in both sexes in the Lopho- gastridae and Eucopiidae, where they have multiarticulate rami fringed with setae, and no special modifications are found in either sex. In the Petalophthalmidae and Mysidae the pleopods are vestigial in the female (except in the little-known Archaeomysis), but are often well developed in the male. When they are reduced in the latter sex some of the pairs, most commonly the fourth, are specially modified. In the males of some Mysidae the peduncle bears distally, in addition to the two rami, a lobe or process of varying form, to which a branchial function has been attributed. The uropods have the exopodite divided into two segments by a transverse suture in the Lophogastridae, Eucopiidae, and Petaloph- thalmidae, and, less distinctly, in certain Mysidae. A statocyst is present near the base of the endopodite in most Mysidae, but it is absent or vestigial in Boreomysis and in the other families. Alimentary System. — The stomach in Mysis (Fig. 107, st) is divided into a globular cardiac portion occupying the greater part of the cavity of the head in front of the cervical sulcus and a much smaller pyloric portion. The interior of both chambers has numerous ridges and prominences armed with spines and setae. In particular, a tongue-shaped process directed backwards on the THE MYSIDACEA 177 FIG. 107. Alimentary canal of Mysis. st, cardiac chamber of stomach. (After Sars.) floor of the pyloric division, and bearing on each side anteriorly a double comb-like row of iridescent setae, can be identified with a similar process found in Amphipoda and Isopoda. The extent of the mid-gut does not seem to have been determined. Five pairs of hepatic caeca (Fig. 107, hep) are found in Mysis, opening by a common duct on each side just behind the tongue-shaped process on the floor of the pyloric chamber. Two pairs are very short and directed forwards ; of the three pairs directed backwards, the upper and lower extend through the greater part of the thorax while the middle pair are much shorter. In Siriella there are only three pairs, one turned forwards and two longer pairs turned backwards. An unpaired dorsal diverticulum (d) is given off at the junction of stomach and intestine. Circulatory System. — The heart of the Mysidae is elongated, fusiform, extending the whole or the greater part of the length of the thoracic region. There are only two pairs of ostia, one dorsal to and slightly in advance of the other. Anteriorly and posteriorly the heart is continued into median aortic vessels each flanked at its origin by a pair of lateral vessels. From the under-side of the heart a number of median vessels are given off to the underlying viscera, and near the posterior end there originates an unpaired descending artery which passes on one side of the intestine. On approaching the sternal surface it divides in the median plane into three branches which pass between the con- nectives of the nerve-chain in the fifth, sixth, and seventh thoracic somites. The anterior branch is continued forwards as a subneural artery through the anterior thoracic somites ; the- middle branch supplies the sixth and the posterior the seventh and eighth somites and their limbs. The abdominal aorta gives off in each somite, besides paired vessels which terminate in the pleopods, a median branch which passes on one side of the intestine and runs for a little way alongside the ventral nerve-cord, sometimes anastomosing with its neighbours in front and behind. The interest of this dis- position of the arterial trunks lies in the fact that the descending artery given off from the posterior end of the heart, which is clearly homologous with the similar vessel found in the Decapoda, is here seen to be one of a series of median vessels originating from the under-side of the heart and of the abdominal aorta, and contribut- ing to the formation of a discontinuous sternal or subneural vessel on the ventral side of the body. THE CRUSTACEA Fine capillary networks surround and penetrate the optic ganglia and those of the thoracic region. Converging venous channels on the sides of the thorax convey the blood from the limbs to the pericardium. Small papilliform elevations of the integument on the course of these channels in Mysis have been credited with a branchial function. The chief seat of respiration in the Mysidae, however, appears to be the carapace, in which is a rich network of blood-channels receiving blood from the sinuses of the anterior part of the body and returning it to the pericardium. The epipodite of the maxilliped may also have a branchial function, and at all events serves, by its movements, to maintain a current of water under the wings of the carapace. The branchiae of the Lophogastridae and Eucopiidae have already been described. Excretory System. — The antennal gland is well developed in Mysis. The canal is much convoluted and expands into a small bladder before opening to the exterior on the second segment of the antennal peduncle. Groups of excretory cells are present also at the bases of the thoracic limbs. Nervous System. — The oesophageal connectives are elongated and a post-oesophageal (antennal) commissure appears to be present. In Boreomysis the full number of eleven pairs of ganglia can be distinguished in the cephalo thoracic part of the ventral chain, but in Mysis all are coalesced into a continuous mass within which only ten pairs of ganglia can be made out. In Gnathophausia the first three pairs are completely coalesced and the fourth is closely approximated to them, but the remaining seven pairs are distinct. In all cases six abdominal ganglia are present. Sense-Organs. — The eyes have the cornea slightly faceted ex- ternally. The crystalline cone is bipartite and the elongated rhabdome is quadripartite. In certain bathypelagic Mysidae the ommatidia are divided into two groups differing in structure. In Gnathophausia there is, on the upper surface of the ocular peduncle, a small prominence which probably corresponds to the sensory papillae found in the Leptostraca. The statocyst, which is found in the endopodite of the uropods in nearly all Mysidae (Fig. 108), consists of a spacious vesicle originating as an invagination of the integument and remaining in communication with the exterior (in some species at least) by a narrow fissure. It contains a single large discoidal statolith (st), consisting of an organic nucleus surrounded by a thick shell of calcium fluoride and resting on a group of setae springing from the floor of the cavity. The tips of the setae are imbedded in the substance of the statolith. Each statocyst is supplied by a large nerve (n) from the last pair of abdominal ganglia. Holt and Tattersall have observed in Hansenomysis a pit on the upper surface of the proximal segment of the antennule. Although THE MYSIDACEA 179 no sensory setae could be discovered, it seems possible that this pit may represent the antennular statocyst of Syncarida and Decapoda. System. — The two tubular ovaries are connected with each other by a narrow bridge. The oviducts proceed from their hinder ends and probably open in the usual position at the base of the sixth thoracic limbs. The testes are closely approximated in the middle line and consist each of a number of pyriform follicles opening into the vas deferens, which is dilated near the front of each testis to form a seminal vesicle. The external openings are situated on papilliform elevations at the bases of the last pair of thoracic limbs. The spermatozoa have the form of slender rods with a filiform tail attached at an acute angle at one end. St.-- Via. ins. A, bison and one uroi>ot near its l>a*e ; o, exopodite of uropod ; n, nerve (applying senJOry setae of statocyst ; at, statolith ; t, telson. (After Sars.) Development. — The whole course of development takes place within the brood-pouch. In the Mysidae segmentation is of the discoidal type. The embryo becomes freed from the egg-membrane after the appearance of the first three pairs of appendages, at which stage, corresponding to the nauplius, the first larval cuticle is formed (Fig. 109). The caudal region, which within the egg was flexed ventrally, becomes extended and the body acquires a slight dorsal curvature. An important feature is the presence of two immovable setose styles (/) terminating the abdomen and re- presenting the caudal fuira of the Leptostraca. Under the cuticle of this maggot-like nauplius stage the remaining cephalic and thoracic appendages (th) appear simultaneously, followed by the uropods and, after an interval, by the pleopods. The elongation of the body in the post-naupliar region is effected by successive divisions of a series of teloblastic cells as in Isopoda. A pair of lateral i8o THE CRUSTACEA thickenings of epiblast appear very early and, approaching each other on the dorsal side, fuse to form an invaginated " dorsal organ " (d). The young animal leaves the brood-pouch with all the appendages developed. Advanced embryos closely similar to those of the Mysidae have been observed in Lophogaster and Eucopia. REMARKS ON HABITS, ETC. The great majority of Mysidacea are marine, but a few Mysidae occur in fresh water either as apparently recent immigrants from the sea, or as "relict" forms like the My sis r dicta (Fig. 99) found in lakes in Northern Europe, Ireland, and North America. A few Mysidae are members of the surface plankton, and a number of peculiarly modified genera of that family, like all .. a. the members of the other families, are Fio 1Q9 bathypelagic at great depths. Embryo of Mysis \MatromysU). Most of the Mysidae are of small size, a', antennule; a", antenna; d, few approaching Boreomvsis SCUpllOVS. which dorsal organ ; /, caudal furca ; md, 3 . y . " \ ' mandible; th, thoracic limbs, reaches 85 mm. in length; on the other hand, Anchialus pusillus is only 3 mm. long. Among the Lophogastridae many species are of considerable dimensions, and Gnathophausia ingens, the largest member of the order, reaches a length of 157 mm. PALAEONTOLOGY. The genus Pygocephalus was established by Huxley in 1857 for a species occurring in the Coal-measures of Scotland which he compared with the existing genus Mysis. Dr. H. Woodward has recently made the highly important discovery that Pygocephalus possessed a brood-pouch formed by six or seven pairs of imbricating oostegites (Fig. 110), thus showing that it must be classed with the Peracarida. Pygocephalus has a broad and apparently flattened b6dy, with a carapace covering the thoracic region. The antennules are biramous and the antennae have a broad scale -like exopodite. Seven pairs of thoracic limbs are visible in the fossils (the first pair were probably folded inwards as maxillipeds and are therefore invisible), carrying each a multiarticulate exopodite. The uropods and telson form a broad tail-fan. This combination of caridoid and Peracaridan characters justifies us in assigning Pygocephalus a place among the Mysidacea, although it is impossible at present to define more precisely its relation to the existing families of the order. It is not known whether the tergal portions of any of the thoracic iSi somites were distinct beneath the carapace in Pygocephalus, but in the genus Crangopsis of Salter, from the Lower Carboniferous, Ortmann has shown that at least four somites are distinct. It is probable that other fossil genera, such as the Carboniferous Anthrapalaemon, may be found to possess Mysidacean characters. AFFINITIES AND CLASSIFICATION. The removal of the Mysidacea from their old association with the Euphausiidae in the group " Schizopoda " is justified by the fact that the two groups do not have in common any characters except those of the primitive caridoid facies which they share with the lower Decapoda and, in part, with the Stomatopoda. On the other hand, the con- nection of the Mysidacea, through the Cumacea and Tanaidacea, with the Isopoda and Amphi- poda is shown not only by the characters given in the definition of the division Peracarida, but also by many connecting characters which link together the individual orders. For example, the retroverted palp of the maxillula in Gnatho- phausia is repeated in the Cumacea and Tanaidacea, and th& branchial epipodite of the first thoracic limb in these two orders may be derived from the simpler appendage found in the Mysidacea. The families mentioned below are those accepted by G. 0. Sars, with the addition of the Petalophthalmidae, established by Holt and Tattersall (following Czerniavsky). The sub-families of the Mysidae as adopted here have been regarded by Norman as distinct families. Fio. 110. Pygocephalus Cooperi, from the Coal-measures. Female specimen, from below, showing the imbri- cated oostegites. (After Woodward.) ORDER Mysidacea, Boas (1883). Family LOPHOGASTRIDAE. Lophogaster, M. Sars ; Gnathophausia, Willemoes - Suhm (Fig. 100). Family EUCOPIIDAE. Eucopia, Dana. Family PETALOPHTHALMIDAE. Petalophthalmus, Willemoes-Suhm ; Han- senomijsis, Stebbing. Family MYSIDAE. Sub-Family ARCHAEOMYSINAE. Archaeomysis, Czerniavsky. Sub-Family LEPTOMYSINAE. Leptomysis, G. 0. Sars ; Erythrops, G. 0. Sars ; Dactylerythrops, Holt and Tattersall ; Euchaetomera, G. 0. Sars ; Amblyops, G. O. Sars ; Pseudomma, G. 0. Sars. Sub-Family ARACHXOMYSINAE. Arachnomysis, Chun. Sub-Family MYSIDETINAE. Mysidetes, Holt and Tattersall. Sub-Family MYSINAE. Mysis, Latreille (Fig. 99) ; Macromysis, White ( = Praunus, Leach). Sub- Family STILOMYSINAE. Stilomysis, Norman. Sub-Family HETEROMYSINAE. Heteromysis, S. I. Smith. Sub-Family MYSIDELLINAE. Mysidella, G. O. 182 THE CRUSTACEA Sars. Sub-Family SIRIELLIXAE. Siriella, Dana. Sub-Family GASTRO- SACCINAE. Gastrosaccus, Norman; Anchialus, Kroyer (=Anchialina, Norman and Scott). Sub-Family BOREOMYSINAE. Boreomysis, G. 0. Sars. LITERATURE. 1. Beneden, E. van. Recherches sur 1'embryogenie des Crustaces : II. Developpement des Mysis. Bull. Acad. Belgique (2), xxviii. pp. 232- 249, 1 pi, 1869. 2. Bergh, R. S. Beitriige zur Embryologie der Crustaceeu : I. Zur Bildungs- geschichte des Keimstreifens von My sis. Zool. Jahrb. Anat. vi. pp. 491-528, pis. xxvi.-xxix., 1893. 3. Chun, C. Atlantis. Biologische Studien liber pelagische Organismen : V. Ueber pelagische Tiefsee-Schizopoden. Bibliotlieca Zoologica, vii. Heft 19, pp. 137-190, pis. viii.-xv., 1896. 4. Claus, C. Zur Kenntniss der Kreislaufsorgane der Schizopoden und Decapoden. Arb. zool. Inst. Wien, v. pp. 271-318, pis. xxi.-xxix., 1884. 5. Czerniavsky (Chernyavskii), V. I. Monographia Mysidarum imprimis imperil Rossici. Trudui St. Peterburg. Obshchest. Estest. xii. (2), xiii. (1), and xviii., 1882-83. 6. Delage, Y. Circulation et Respiration chez les Crustaces Schizopodes (Mysis, Latr.). Arch. Zool. expe'r. (2), i. pp. 105-130, pi. x., 1883. 7. Holt, E. W. L., and Tatter sail, W. M. Schizopodous Crustacea from the North-East Atlantic Slope. Rep. Fisheries, Ireland, 1902-1903, pt. ii. App. iv. pp. 99-152, pis. xv.-xxv., 1905 ; and Supplement, Fisheries, Ireland, Sci. Invest., 1904, v. 50 pp. 5 pis., 1906. 8. Norman, A. M. British Schizopoda of the Families Lophogastridae and Euphausiidae. Ann. Mag. Nat. Hist. (6), ix. pp. 454-464, 1892. 9. On British Mysidae, a Family of Crustacea Schizopoda. Op, cit. (6), x. pp. 143-166 and 242-263, pis. ix. and x., 1892. 10. Nusbaum, J. L'embryologie de Mysis chameleo (Thompson). Arch. Zool. exper. (2), v. pp. 123-202, pis. v.-xii., 1887. 11. Sars, M. Beskrivelse over Lophogastcr typicus, en maerkvaerdig Form af de lavere tifoddede Krebsdyr. Pp. iv + 37, 3 pis. 4to. Christiania, 1862. 12. Sars, G. 0. Histoire naturelle des Crustaces d'eau douce de Norvege. 1 Livr. Les Malacostraces. Pp. iii + 145, 10 pis. 4to. Christiania, 1867. 13. Carcinologiske Bidrag til Norges Fauna : I. Monograph! over de ved Norges Kyster forekommende Mysider. 3 pts. 42 pis. 4to. Christiania, 1870-79. 14. Nye Bidrag til Kundskaben om Middelhavets Invertebratfauna : I. Middelhavets Mysider. Arch. Math. Naturvid. ii. pp. 10-119, 36 pis. , 1876. 15. Report on the Schizopoda. Rep. Voy. "Challenger," Zool. xiii. 228 pp. 38 pis., 1885. 16. Willemoes-Suhm, E. von. On some Atlantic Crustacea from the " Challenger " Expedition. Trans. Linn. Soc. (2), Zool. i. pp. 23-59, pis. vi-xiii, 1875. 17. Woodward. H. On Pygocephalus, a Primitive Schizopod Crustacean. Geol. Mag. (n.s.) Dec. 5, vol. iv. pp. 400-407, pi. xviii., 1907. CHAPTER X THE CUMACEA Order Cumacea, Kroyer (1846). Definition. — Peracarida in which the carapace coalesces dorsally with the first three or four thoracic somites, overhangs on each side to enclose a branchial cavity, and is produced in front into two plates Avhich usually meet each other above in front of the head to form a pseudorostrum ; the telson may be coalesced with the last somite ; the eyes are generally coalesced into a single organ set on an immovable process of the head ; the antennules may be biramous ; the antennae have no exopodite ; some of the thoracic limbs have natatory exopodites ; the first three pairs are modified as maxillipeds ; the first pair have an epipodite generally pro- vided with branchial lobules and an exopodite forming a respiratory siphon j the pleopods are absent in the female and often reduced in the male ; the uropods are styliform ; the young leave the brood- pouch before the appearance of the last pair of thoracic limbs. Historical. — The first described Cumacean .was the Oniscus scorpioides of Lepechin (1779), and other species were described by Montagu (1804), Say (1818), and H. Milne-Edwards (1828). The last-named author established the genus Cuma, from which the name of the order is derived, but he later regarded this as being a larval decapod, and he maintained his opinion of the larval nature of the group as late as 1858, although Kroyer had described o vigorous females in 1841 and his discovery had been confirmed by H. Goodsir and others. While Spence Bate, Norman, Lilljeborg, Hansen, Dohrn, and others have contributed descriptions of species and observations on structure and development, by far the greater part of our present knowledge of the group is based on the elaborate and beautiful memoirs of G. 0. Sars. MORPHOLOGY. The general shape of the Cumacea is usually very characteristic, owing to the sharp distinction between the inflated cephalothoracic 183 1 84 THE CRUSTACEA region and the slender and very mobile abdomen carrying at its tip the styliform uropods (Fig. 111). The extreme specialisation of the respiratory system, with the concomitant modification of the anterior part of the carapace, are the most striking features differentiating this from the neighbouring orders. As in the Mysidacea and Tanaidacea the first thoracic limb (maxilliped) carries a backwardly directed membranous epipodite lying in a cavity between the carapace and the side wall of the body ; but in the Cumacea this epipodite is of relatively great size, and is usually (though not always) furnished with respiratory processes or lamellae (Fig. 112, br\ which may be very numerous and are often better developed in the more active males than in the females. Anteriorly, the branchial cavity is continued as a narrow channel covered by a forward extension of the lateral plate of the carapace, " ,-m:m^ "^ v. FIG. 111. Dictstylis Goodsiri, 9i from the side, x 4. a', antennule ; I1, I5, first and fifth legs (fourth and eighth thoracic appendages) ; m, brood-pouch ; ps, pseudorostrum ; t, telson ; ur, uropod. (After Sars.) the two lateral plates generally meeting each other above in front of the head and forming a more or less prominent pseudorostrum (ps) divided by a Y-shaped fissure (Fig. 112, fr). Within this channel lies the exopodite of the same appendage in the form of a narrow stalk bearing distally a membranous expansion, which is rolled upon itself to form a tube, or unites with its fellow of the opposite side in a single tube (ex) capable of protrusion from the front of the head below the pseudorostrum. Sometimes the pseudorostral plates do not quite meet in front of the head, and in certain genera (Zygosiphon) these plates, and the respiratory channels which they cover, are placed wide apart at the sides of the broadly expanded frontal region, while the exopodal tubes project as long transparent siphons. Eyes are altogether deficient in many genera. When present they are usually coalesced to form a single median organ (Fig. 112, e) borne on the front of the head, which is produced into an oculiferous lobe lying between the two plates of the pseudorostrum. THE CUM ACE A 185 When the lateral plates are removed, the conformation of the head and the position of the eye show some similarity to the arrange- ment met with in the Oedicerotidae ex among the Amphipoda. The two eyes are distinct in the embryo, and in one genus (Xciiinaxtacus) also in the adult. .-l]>/>en>tages. — The antennules rarely have both flagella well developed, the inner being usually reduced or absent. The antennae differ remarkably in the two sexes. In the female (Fig. 112, a") they are vestigial, while in the male they consist of a stout peduncle of five seg- ments, of which the last two are enlarged and clothed with a brush of long setae, while the flagellum is filiform and may exceed the length of the body (Fig. 113). In life this long flagellum is usually carried folded close to the side of the body, protected by the lower edge of the carapace and by the pleural plates of the abdomen, or in a special groove which runs along the sides of the abdominal somites. In the genus Lamprops the antennal flagella of the male are short anrl '1 into. one; c?' ex°Pod,ite OI first maxilliped, forming the re- Ihe mandibles never carry a palp, but spiratory siphon; fr, lateral ,i r .1 branch of "frontal fissure" be- m Other respects COniOrm to the type tweeu the head and the lateral r>Jvivnptpri«rip r>f tViP Ppr-inarirla Tti tVio plate of pseudorostrum ; h, heart ; 1 tne hep, hepatic caeca; 1 1, first pair Leuconidae and in the genus Diastiiloides °f less ; ov, ovary ; ps, pseudo- ,, , , 0 ,, j-ii • T i rostrum. (After Sara.) the body or the mandible is short and triangular and the row of spines is reduced. In Campylaspis the molar process is styliform. Fio. 112. Diagram of anterior part of body of Diastylis, <} , from above, showing the internal organs, a', anteunule ; a", antenna ; br, branchial epipodite of first maxil- liped enclosed in branchial cavity Pio. 113. 1'tci-' ••.'»»!, (J, from the Caspian Sea. (After Sars, from Ency. Brit.) The niaxillulae (Fig. 114, A), except in Platyaspis and Para- lamprops, carry a retro verted palp as in Tanaidacea and Lopho- gastridae. The maxillae (Fig. 114, B) with their two terminal 1 86 THE CRUSTACEA -2. PIG. 114. endites present a close approximation to those of Tanaidacea and Isopoda. In Campylaspis the endites are suppressed. The first thoracic appendages with their respiratory epipodites (Fig. 115) have a close resem- blance to those of the Tanaidacea, in which group, however, there is no exopodite. The basipodite has an endite directed distally, and carries on its inner edge two or three hooked spines ("coupling-hooks") which inter- A, maxillula, B, maxilla, of Diastylls Goodsiri. lock with those of the Other side. 1-4, segments of the appendages; l^-l*, endites rrc<_ j j J.T.* 3 ii. of the respective segments ; /, flabellum (ex- 1 he SCCOnd and third thoracic opodite) ; p, palp. According to Hansen's later limfoc though less SDecialised interpretation.thechitinous piece here numbered l DS» l OU»U K .' 4 in the maxilla does not represent a distinct may also be reckoned as maxilli- segment, and the endites 1 3 and / •» result from r . , , - the division of a single endite belonging to the peds, SlllCC they are turned tor- wards and applied to the oral region with their basipodites flattened and meeting in the middle line, while the terminal segments are relatively weak and carried in a folded position. The second pair are without exopodites, but in the female a small scale bearing a fan-like fringe of setae is attached behind to the base of the limb, projecting backwards into the marsupial chamber and serving to keep in motion the eggs or embryos contained therein. This scale is doubtless homologous with the oostegites, which are well developed on the four succeed- ing pairs of limbs, where they are firmly attached to the small coxal segments. The third pair are only rarely devoid of exopodites. The fourth pair of thoracic limbs are long raptorial or prehensile legs (Fig. Ill, I1), though their broad basipodites, sometimes meeting in the middle line, are not dissimilar to those of the preceding pair. The fifth pair (2nd legs) often have a reduced number of joints and differ in details from the succeed- First maxiiiiped of Dia- ,1 . i • i n • M i stylis styqia. bs, basipodite ; ing three pairs, which are all similar and appear i>r, branchial lamellae on to be fossorial in function. The small terminal SS&T-JSBffi segment in these limbs is tipped with a com- forming the respiratory , . ' siphon. (After Sars.) paratively weak spine, which only m the Nannastacidae becomes a stout curved claw. Natatory exopodites are always present on the fourth pair (1st legs), and usually on one or more of the succeeding pairs in the female sex; in the FIG. 115. THE CUM ACE A 187 males (Fig. 113) they are generally present on all save the last pair of legs. The pleopods are always absent in the female and sometimes in the male, but more usually in the latter sex from one to five pairs are well developed and biramous. The uropods are always con- spicuous, their slender rami furnished with comb-like rows of spines apparently used in cleaning the anterior appendages, for which purpose the abdomen can be flexed ventrally and sometimes also dorsally. Internal Anatomy. — The masticatory stomach is stated to resemble closely that of the Tanaidacea. There are from one to four pairs of hepatic caeca (Fig. 112, hep). In one genus (Platycuma) the anterior part of the intestine is coiled, forming a spiral of two and a half turns within the carapace, but it has not been ascertained whether the coiled part belongs to the mesenteron or to the proctodaeum. In this genus also the hepatic caeca appear to be absent. The heart (h) is usually somewhat elongated, but in Platycuma it is subglobular. There are three pairs of ostia. Besides anterior and posterior median arteries, the heart gives off a pair of antero-lateral vessels and an unpaired descending artery. A well-developed maxillary gland is present, which, according to Glaus, resembles closely that of Apseudes (Tanaidacea). The ventral nerve -chain consists of ten thoracic and six abdominal ganglia. The simple tubular paired ovaries are connected, at least in the young, by a narrow transverse bridge. The openings of the oviducts have not been seen. The testes are separate, tubular, with four small caeca anteriorly. The short vasa deferentia open on the sternal surface of the last thoracic somite. DEVELOPMENT. The development appears to resemble, in its main features, that of the Isopoda. In the earlier stages the embryo is curved dorsally. As in the Tanaidacea and Isopoda, the young leave the brood-pouch with the last pair of legs still undeveloped. In certain species this deficiency persists very late, and possibly in some cases throughout life. REMARKS ON HABITS, ETC. The Cumacea are exclusively marine (including under this term the group of peculiar species inhabiting the Caspian Sea), and are generally found burrowing in sand or mud. No species appears to be truly pelagic, although the actively swijaming males of some species, and less commonly the females, are found in the plankton 1 88 THE CRUSTACEA of inshore waters. In Arctic seas the Cumacea are conspicuous by their abundance and relatively large size, but they are also common in suitable localities in tropical waters. Many species, often also of considerable size, occur in the deep sea. The thirteen species known from the Caspian Sea all belong to the family Pseudocumidae, and were originally referred to the single genus Pseudocuma, which comprises only three truly marine species. It is interesting to note that several of the Caspian species " mimic " in their general aspect widely different genera of other families, and have on this account received such specific names as diastyloides, eudorelloides, and campylaspoides. The largest known Cumacean is the Arctic Diasti/lis Goodsiri (Fig. Ill), which reaches a length of 35 mm. Some deep-sea species are not much inferior to this, but the average size is much less, and some species do not exceed 1'5 mm. in length. No fossil Cumacea are known. AFFINITIES AND CLASSIFICATION. The systematic position of the Cumacea has been somewhat obscured by the customary classification of the Malacostraca into Edriophthalma and Podophthalma, since they unite the sessile eyes of the former group with the carapace of the latter. As a matter of fact, apart from characters of specialisation which seem to have arisen within the order, they clearly stand midway between the Mysidacea and the Tanaidacea, just as they combine, to some extent, the swimming powers of the one with the burroAving habits of the other. ORDER Cumacea, Kroyer (1846). Family BODOTRIIDAE. Bodotria, Goodsir (= Cuma, H. Milne- Edwards) ; Cyclaspis, G. 0. Sars ; Zyyosiphon, Caiman. Family VAUX- TOMPSONIIDAE. Vauntompsonia, Spence Bate. Family LEUCONIDAE. Leucon, Kroyer ; Eudorella, Norman. Family NANNASTACIDAE. Nan- nastacus, Spence Bate ; Gampylaspis, G. 0. Sars ; Platycuma, Caiman. Family CERATOCUMIDAE. Ceratocuma, Caiman. Family PSEUDOCUMIDAE. Pseudocuma, G. O. Sars ; Pterocuma, G. O. Sars (Fig. 113). Family LAMPROPIDAE. Lamprops, G. 0. Sars ; Paralamprops, G. 0. Sars. Family PLATYASPIDAE. Platyaspis, G. 0. Sars. Family DIASTYLIDAE. Diastylis, Say (Fig. Ill); Diastyloides, G O. Sars. LlTERATUKE. 1. Caiman, W. T. On the Cumacea. Rep. Pearl Fisheries Ceylon. Roy. Soc. ii. pp. 159-180, 5 pis., 1904. 2. The Marine Fauna of the West Coast of Ireland : Part IV. Cumacea. Fisheries, Ireland, Sci. Invest., 1904, i. 52 pp. 5 pis., 1905. THE CUM ACE A 189 3. Caiman. W. T. The Cumacea of the "Siboga" Expedition. Uitkomsteii . . . H.M. "Siboga," Monogr. xxxvi. 23 pp. 2 pis., 1905. 4. On New or Rare Crustacea of the Order Cumacea from the Collection of the Copenhagen Museum : Part I. Trans. Zool. Soc. London, xviii. pp. 1-58, pis. i.-x., 1907. 5. Dohrn, A. Untersuchungen iiber Bau und Entvvickelung der Arthropoden : I. Ueber den Bau und die Entwickelung der Cumaceen. Jenaische Zeitschr. v. pp. 54-81, pis. ii. and iii., 1870. 6. Sars, G. 0. Om den aberrante Krebsdyrgruppe Cumacea og dens nordiske Arter. Forh. Vidensk. Selsk. Christiania, 1864, pp. 128-208. 7. - Beskrivelse af de paa Fregatten Josephines Expedition fundne Cumaceer. Kgl. Svenska Vet.-Akad. Handl. ix. No. 13, 57 pp. pis. i.-xx., 1871. 8. Nye Bidrag til Kundskaben om Middelhavets Invertebratfauna : II. Middelhavets Cumaceer. Arch. Math. Naturvid. iii. pp. 461-512, pis. i.-xxi., 1878; and op. cit. iv. pp. 1-144, pis. xxii.-lx., 1879. 9. Report on the Cumacea. Rep. Voy. "Challenger," Zool. xix. 73 pp. 11 pis., 1887. 10. Crustacea caspia : Part II. Cumacea. Bull. Acad. Sci. St. Peters- bourg, Nouv. Ser. iv. (xxxvi.), pp. 297-338, 12 pis., 1894. 11. An Account of the Crustacea of Norway : III. Cumacea, 115 pp. 72 pis., 1899-1900. CHAPTER XI THE TANAIDACEA Order Tanaidacea, Hansen (1895). Definition. — Peracarida in which the carapace coalesces dorsally with the first two thoracic somites, and overhangs on each side to enclose a branchial cavity ; the telson is not defined from the last somite ; the eyes, when present, are usually set on short, im- movable peduncles ; the antennules may be biramous ; the antennae may have a small exopodite ; vestigial exopodites are sometimes present on the second and third pairs of thoracic limbs ; the first pair of thoracic appendages are modified as maxillipeds ; they have an epipodite lying in the branchial cavity, but no exopodite ; the second pair are chelate ; the pleopods are usually present and biramous ; the uropods are slender ; the young leave the brood- pouch before the appearance of the last pair of thoracic limbs. Historical. — The first known member of this order was that described by Montagu in 1808 as Cancer gammarus falpa, now placed in the genus Apseudes. Leach, Latreille, and others ranked the species known to them among the Amphipoda, while H. Milne- Edwards placed them among the Isopoda, remarking, however, that they established a transition to the Amphipoda. Dana (1852) united them with certain parasitic and other Isopoda in a group Anisopoda interposed between Isopoda and Amphipoda, and Spence Bate in 1868 combined them with the Isopod Anthuridae and Anceidae (Gnathiidae) in a no less heterogeneous group of " Isopoda aberrantia." Van Beneden (1861) called attention to the im- portance of the carapace of Tanais as a systematic character, and approximated that genus to Curna and Mysis. F. Miiller (1864) also attached great weight to the same character. Gerstaecker (1886) once more included the Tanaidacea among the Amphipoda ; but this retrograde step has met with little support, and most modern writers follow Sars in ranking them as one of the tribes of Isopoda (Chelifera). Glaus, however, in 1888 placed them in an independent order, for which he adopted Dana's name Anisopoda, 190 THE TANAIDACEA 191 and which he placed between Isopoda and Cumacea. Hansen adopts the same view of their affinities, and suggests for the order the name which is here used. Claus's monograph on Apseudes is the most important source of information on the anatomy of the group, and the works of G. 0. Sars are no less important as regards the description and classification of genera and species. MORPHOLOGY. The general form of the body (Fig. 116) resembles that typical of the Isopoda, being depressed or sub-cylindrical, with a compara- tively short abdominal region and with the telson coalesced with the last somite. The homology of the reduced carapace with that of the Cumacea and Mysidacea is indicated by the small cavity which it overhangs on each side and within which lies the epipodite of the first thoracic appendage. Only the first and second thoracic oc sc- ar. FIG. 110. Apseudes gpuiosus, female, ex, vestigial exopodites of second and third thoracic limbs ; oc, ocular lobe or peduncle ; sc, scale (exopodite) of antenna ; it /•, flagelliform uropod. "(After Sars.) somites are coalesced with the carapace, but in Sphyrapus the third somite is firmly attached to the head-region, although, according to Sars, it is not completely fused with it. The eyes are often absent, but when present they are usually set on frbnto-lateral processes of the head (Fig. 116, oc}, which in many species are defined by grooves and, although not movable, appear to correspond to the ocular peduncles of the podophthalmate groups. Appendages. — The antennules are biramous in the Apseudidae. In some species the two flagella are stated to arise from a common basal segment, so that the peduncle consists of four segments. The antennae have, in the Apseudidae, a small exopodite (Fig. 116, sc), and the protopodite consists of two segments, not, as is usual in the Peracarida, of three. The mandible carries a palp in the Ap- seudidae, but not in the Tanaidae. - The maxillulae, in the former family (Fig. 117, B), have two endites and a palp of two segments, but in the Tanaidae the proximal endite is wanting and the palp is unsegmented. The maxillae of the Apseudidae (Fig. 117, C) 192 THE CRUSTACEA have a small palp. In the Tanaidae the whole appendage is vestigial. The lower lip of the Apseudidae (Fig. 117, A) is peculiar in having each of the lobes terminating in a movably articulated lappet. The first pair of thoracic appendages are muxillipeds (Fig. 117, D), agreeing with those of the Cumacea and Isopoda in having coupling- hooks on the endite of the basipodite. The epipodite (ep) projecting backwards into the branchial chamber is most fully developed in the Apseudidae, where it consists of a peduncle bearing a spoon-shaped membranous plate which terminates posteriorly in a filiform pro- cess and is produced anteriorly as a rounded lappet fringed with setae. The resemblance of this apparatus to the branchial epipodite of the Cumacea is unmistakable, and the filiform termination may be compared with the inflected apex generally found in that group. A small lobule found in some Tanaidae may perhaps represent the exopodite. In the Tanaidae the maxillipeds- B — en. Mouth-parts of Apseudes spinosus. A, lower lip with movable lobes. B, maxillula ; ->>, palp C, maxilla. D, inaxilliped ; en, terminal lobe or endite of basipodite ; ep, epipodite. " (After Sars.) are more or less united at the base, the coxopodites, and some- times also (as in the Amphipoda) the basipodites, being coalesced. The second pair of thoracic limbs are in nearly all cases com- pletely chelate and are usually much stronger than the succeeding pairs. The third pair in the Apseudidae are flattened and apparently fossorial in function. The minute exopodites (Fig. 116, ar) with which these two pairs are provided in many Apseudidae are placed close to the exhalent and inhalent openings of the branchial cavity respectively, and by their vibratory motion assist in producing the respiratory current. In all the thoracic limbs the coxopodite is very small, and in the posterior five or six pairs the limb ends in a curved claw. In the Apseudidae there are five pairs of oostegites (the first pair very small) attached to the thoracic limbs, from the second to the sixth pair. In some, perhaps all, Tanaidae only one pair of oostegites is present, on the sixth thoracic limbs. The pleopods may be fully developed in both sexes, but some- times they are reduced in number or altogether absent in the THE TANAIDACEA 193 female. The uropods are nearly terminal on the last segment of the body. In Apseudes the two rami are long, multiarticulate, and flagelliform. In other cases the rami may be much reduced and the exopodite is sometimes wanting. Internal Anatomy. — In Apseudes, the alimentary canal has three pairs of hepatic caeca. The heart is elongated, extending through the six free somites of the thorax, giving off a median aortic vessel anteriorly and a pair of diverging abdominal arteries (but no median vessel) posteriorly. In Leptochelia and in the young of Apseudes it has two pairs of ostia, but in the adult Apseudes the right anterior ostium disappears and those of the posterior pair become asymmetrically placed. In Tanais only one pair persists. The anterior aorta, after dividing to encircle the brain in the median plane, forms a circumoesophageal ring, but there is no subrieural sternal artery. The lateral folds of the carapace are traversed by a network of blood-channels supplied by branches from the anterior aorta, and no doubt form the chief organs of respiration, possibly assisted by the epipodites of the maxillipeds. Considerable importance has been attached to the thoracic position of the heart in the Tanaidacea as differentiating them from the Isopoda and indicating affinity with the Amphipoda. As a matter of fact, however, in certain Isopoda (Jaera) the anterior end of the heart extends as far forward as it does in Apseudes, and the suppression of the abdominal portion, leaving intact the paired abdominal arteries, would produce a disposition of parts essentially similar to that of the Tanaidacea. The maxillary gland is well developed in Apseudes, and a vestige of the antennal gland has been described. Dermal glands are commonly found on the body and limbs, and in some Tanaidae the secretion appears to be utilised in forming the tubes of mud in which the animals live. In Heterotanais groups of gland-cells are described situated on each side of the anterior thoracic somites and opening by long ducts on the terminal segments of the correspond- ing legs, an arrangement which recalls that found in certain Amphipoda. The nerve-chain in Apseudes has all the ganglia of the post-oral somites distinct. It was stated by F. Miiller that a species of Tanais possessed an open statocyst-cavity containing a statolith in the basal segment of the antennule, but the observation has not been confirmed. The reproductive organs of both sexes are of a simple type. The vasa deferentia, in Apseudes, open close together on a median process of the last thoracic sternum. A seminal vesicle is formed in Tanais and in Leptochelia by fusion of the two vasa deferentia, but here also the external opening is paired. Sexual differences are often strongly marked. The olfactory filaments of antennules and antennae are, as usual, more numerous 13 194 THE CRUSTACEA in the male ; the chelae are often much stronger and differently shaped ; and the pleopods are always well developed in that sex even when they are reduced or absent in the female. In several genera of Tanaidae the oral appendages, with the exception of the maxillipeds, are entirely lost by the sexually mature male. Fritz Miiller described a remarkable dimorphism of the males in a species of Leptochelia. One form of male was distinguished by the great development of the olfactory filaments on the antennules, and had chelae very similar to those of the female. In the other form the olfactozy filaments were less numerous, but the chelae were greatly elongated and slender. These observations have been doubted by subsequent writers, but they have recently received partial con- firmation from the work of G. Smith. Development. — The embryo shows at first a dorsal curvature as in the Isopoda. A paired " dorsal organ " is present. The larvae leave the brood-pouch with the last pair of thoracic limbs and the pleopods undeveloped. In Apseudes, according to Glaus, the lateral plates of the carapace are at first extended as wing-like processes, becoming afterwards bent downwards over the branchial epipodites of the maxillipeds and fixed in position by peg-like outgrowths of the sternum on each side. REMARKS ON HABITS, ETC. The Tanaidacea are exclusively marine and occur from between tide-marks to depths of over 2000 fathoms. Many burrow in mud, some inhabit tubes of mud agglutinated by the secretion of the dermal glands, and several species of Tanaidae are recorded as living in rock-crevices among a felt-like mass of filaments, pre- sumably also secreted by the animals. Most Tanaidacea are minute. Many species do not much exceed one millimetre in length, but some Apseudidae reach 13 mm. or more. No fossil Tanaidacea are known. AFFINITIES AND CLASSIFICATION. Mention has been made above of the very varied opinions which have been held regarding the systematic position of the members of this order. Among recent writers, however, there is general agreement that the Tanaidacea must either stand as a distinct order or be merged in the Isopoda as the most primitive sub-order of that group. "While their resemblance to the Isopoda in general form is considerable, their marked divergence in such characters as the possession of a distinct carapace, with its branchial chambers, and the form of the epipodites of the maxillipeds, fully justifies their separation. THE TANAIDACEA 195 The two families into which the order is divided are dis- tinguished by many important characters, and it may prove necessary in the future to rank them as sub-orders. ORDER Tanaidacea, Hansen (1895). Family APSEUDIDAE. Apseudes, Leach ; Sphyrapus, Norman and Stebbing. Family TANAIDAE. Tanais, Audouin and Milne-Edwards ; Heterotanais, G. 0. Sars ; Leptochelia, Dana. LITERATURE. 1. Boas, J. E. V, Kleinere carcinologische Mittheilungen : 1, Eine neue Art der Gattung Apseudes. Zool. Jahrb. ii. pp. 109-116, 1886. 2. Glaus, C. Ueber Apseudes Latreillei, Edw., und die Tanaiden. Arb. zool. Inst. Wien, v. pp. 319-332, 2 pis., 1884 ; op. cit. vii. pp. 139-220, 7 pis., 1887. 3. Dohrn, A. Untersuclumgeii liber Bau und Entwickluug der Arthropoden : 7. Zur Kenntniss vom Bau und der Entwicklung von Tanais. Jenaische Zeitscbr. v. pp. 293-306, pis. xi. and xii., 1870. 4. Sars, G. 0. Revision af Gruppen Isopoda chelifera, med cbarakteristik of nye herhen hbrende Arter og Slaegter. Arch. Math. Naturvid. vii. pp. 1-54, 1882. 5. Nye Bidrag til Kundskaben om Middelhavets Invertebratfauna : III. Middelbavets Saxisopoder (Isopoda chelifera). Arch. Math. Naturvid. xi. pp. 263-368, 15 pis., 1886. 6. Hmith, G. High and Low Dimorphism ; with an account of certain • Tanaidae of the Bay of Naples. Mitth. zool. Stat. Neapel, xvii. pp. 312-340, pis. xx.-xxi., 1905. Many of the works quoted in the list at the end of the next chapter deal, in part, with the Tanaidacea. See especially those of Beddard, Delage, Hansen (Plankton Expedition), Norman and Stebbing, Richardson, Sars, and Spence Bate and Westwood. CHAPTER XII THE ISOPODA Order Isopoda, Latreille (1817). Sub-Order 1. Asellota. „ 2. Phreatoicidea. „ 3. Flabellifera. „ 4. Valvifera. „ 5. Oniscoidea. „ 6. Epicaridea. Definition. — Peracarida in which there is no distinct carapace, but the first thoracic somite (rarely also the second) is coalesced with the head ; the telson is rarely defined from the last somite ; eyes sessile or set on immovable processes of the head ; antennules uniramous (except in Bathynomus) ; antennae sometimes with a minute exopodite ; thoracic limbs without exopodites ; first pair modified as maxillipeds, the epipodite, when present, not enclosed .in a branchial cavity ; remaining pairs all similar or variously modified, coxopodite always short, often fused with the body and expanded laterally ; pleopods typically biramous, with lamellar, branchial rami, generally the second and sometimes also the first pair modified in the male ; heart lying wholly or partly in abdomen ; the young leave the brood-pouch before the appearance of the last pair of thoracic limbs. Historical. — The terrestrial habits of the more familiar members of this order, and the close resemblance which some of them bear to certain Diplopoda (Oniscomorpha), led to their being widely separated from the other Crustacea in many of the earlier systems of classifi- cation. Even Latreille, to whom the name Isopoda is due, placed them at first among the Insecta. Leach ranked them along with the Amphipoda in his group Edriophthalma, thus giving them the position which they occupy in most modern systems. Our know- ledge of the morphology and classification of the Isopoda is largely due to the work of Scandinavian naturalists, especially to the monographs of Schiodte, Meinert, G'. 0. Sars, Bovallius, Budde- 196 THE 1SOPODA 197 Lund, and Hansen. The remarkable group of the Epicaridea has attracted many investigators, but it is especially to the memoirs of Giard and Bonnier that the student must turn for a clear account of their structure and complicated life- histories. As regards the internal anatomy of the Isopoda, the literature is scattered and fragmentary, but Delage's study of the circulatory system must not be passed without mention. Bullar's discovery of hermaphroditism among the Cymothoidae is also noteworthy. MORPHOLOGY. In the majority of the free -living Isopoda the body (Fig. 118) is more or less flattened, with an oval outline as seen from above and with a short abdominal region. In the Arcturidae and Anthuridae, however, it is elongated and subcylindrical, and in some of the latter family it is almost vermiform. In Phrea- toicus (Fig. 119) the body is laterally compressed and the aspect is that of an Amphipod. In the parasitic Epicaridea the adult female FIG. 118. Porcdlio neater, $ , doi sal view, x 3. (After Sars.) FIG. 110. Pkrmtofau autmitta, ?, from the sido. x's. (After Chilton.) becomes more or less distorted and deformed and may even lose all trace of segmentation. In addition to the first, the second thoracic somite becomes united with the head in the Gnathiidae and Serolidae, in the genus Stenasellus among the Asellota, and in a few Valvifera. The last thoracic somite occasionally remains of small size and without appendages in the adult, as it nearly always is in the 198 THE CRUSTACEA young. Very often some or all of the abdominal somites are fused together, and the last somite is always coalesced with the telson except in certain Anthuridae. The abbreviation of the abdomen is least in the Phreatoicidae ; in the other families the anterior somites are short and crowded together, though the telsonic seg- ment is often of large size. The abdominal somites are usually expanded laterally into pleural plates, and the thoracic somites may be similarly produced over the bases of the legs, but in these the pleural plates are often supplemented or replaced by the expanded coxopodites which here form the coxal plates or so-called "epimera." Appendages. — The antennules are usually short, and in the Oniscoidea they are almost or quite vestigial. They are never biramous except in Bathynomus, where a minute vestige of the inner flagellum is present, and in the cryptoniscan larvae of some Epicaridea. The antennae vary much in length, being several times as long as the body in some Asellota, while in the Epicaridea they are hardly more than vestiges. The peduncle usually consists of five segments, but in the Asellota and in Bathynomus and Cirolana it has six. In the Asellota a minute movably articulated scale repre- senting the exopodite is sometimes present on the third segment. In all other Isopoda the exopodite is absent. In the Arcturidae the antennae are very large and strong and are used in seizing prey. The mouth-parts of the Isopoda show great diversity of struc- ture, many modifications being correlated with the parasitic or semi-parasitic habits of certain groups. In the more typical forms (e.g. Asellus] the mandible carries a palp of three segments, and has a serrated incisor process, with a lacinia mobilis, at least on the left side of the body, a row of spines, and a strong molar process. The molar process is movably articulated with the body of the mandible in the Cirolaninae (Fig. 120, A). The maxillula (Fig. 120, B) has two endites (corresponding, according to Hansen, to the first and third segments) directed distally and bearing strong setae or spines. There is no trace of a palp such as exists in the Tanaidacea. The maxilla (Fig. 120, C) has three endites, the proximal one directly continuous with the basal part and directed obliquely inwards ; the two others are directed distally, over- lapping each other and movably articulated with the basal part ; Hansen regards these as resulting from the division of a single endite corresponding to the third segment. The first thoracic limb is always specialised as a maxilliped (Fig. 120, D) and closely associated with the other mouth-parts. It has a short but distinct coxopodite bearing externally a lamellar epipodite (ep) generally more or less indurated, and not, apparently, branchial in function as it is in the Tanaidacea, but serving as an THE ISOPODA 199 operculum to cover the lateral parts of the other oral appendages. The basipodite is produced distally into a large endite (I2) which is sometimes movably articulated (Bathynomus, Cirolana, Chiridotea), and which bears one or more coupling-hooks (c) interlocking with those of the opposite side. The endopodite or palp consists, in the typical case, of five more or less lamellar segments, but the number is often reduced. In the ovigerous females of Asellus and of certain genera of Valvifera the coxopodite bears a small lappet, fringed with setae, projecting backwards into the marsupium and A. eP. '1. lam'A FIG. 1-20. Mouth-parts of variotts Isopoda. A, mandible of Cirolana boreaHs. B, maxillula-of Chiridotfa entomon. C, maxilla of same. D, maxilliped of same. E, maxilliped of adult female ofRocinela danmonietitiis. F, maxilliped of adult male of same, c, coupling-spine ; ep, epipodite ; i, incisor process ; /I, 7'-', 73, N, endites of successive segments of the appendages (according to Hanson's later interpretation the endites 73, 7* of the maxillula result from the division of a single endite) ; 7am1-2, laminar outgrowths of first and second segments of maxilliped ; m, molar process ; p, palp; s, lacinia mobilis and spine-row of mandible, which in this case are closely associated together; 1, 2, 3, successive segments of the appendages. (After Hausen.) resembling the coxal appendage of the second thoracic limb in the females of Cumacea. A similar plate, more largely developed, is found in the females of Cymothoidae (Fig. 120, E) and Epicaridea and of some genera of Sphaeromidae, where it is associated with an expansion of the basipodite and with the epipodite, giving a lamellar character to the whole appendage. In all these cases the vibratory motions of the maxilliped cause a current of water to pass through the marsupium for the aeration of the developing embryos. A similar expansion of the basipodite united with the epipodite occurs also in Serolis, where, however, the coxal lobe is not developed. The position of this coxal lobe and its occurrence 2OO THE CRUSTACEA only in the adult female suggest its possible homology with the oostegites of the following somites. As an example of extreme modification of the mouth-parts and of exceptionally marked changes taking place during development, E. . m. m. FIG. 121. Gnathia maxillaris. A, late larval form (Praniza-stage). B, head of same, from below, further enlarged. C, adult female. D, second thoracic appendage of same, further enlarged. E, adult male. F, maxilliped (first thoracic appendage) of same, further enlarged. G, second thoracic appendage of same, m, mandible of male ; mxp, maxilliped of larva; II, III, VII, second, third, and seventh thoracic appendages (the eighth is undeveloped in adults of this family). (After Sars ; B partly after G. Smith.) the oral apparatus of the very aberrant Gnathia may be here described. In the larval state (Fig. 121, A), Gnathia is parasitic on fish, and the suctorial mouth -parts form a short conical proboscis (Fig. 121, B). The mandibles are styliform, finely serrate on the inner edge, without palp, molar process, Jacinia mobilis, or spines. The maxillulae and maxillae are simple, slender THE ISOPODA 201 styles. The maxillipeds (ni.fp\ which close the oral cone below, have an elongated basipodite produced distally into a slender endite, with many coupling-hooks ; the next segment is also pro- duced distally as an acute, serrate point, and the remaining segments of the endopodite are coalesced into a short palp. The appendages of the second thoracic segment (II), which in this family is coalesced with the head, lie on either side of the oral cone and end in strong curved claws which no doubt assist in pene- trating the skin of the host. The adult animals are free-living and have the mouth-parts very differently developed in the two sexes. In the male (Fig. 121, E) a pair of relatively enormous mandibles (ra) are articulated at the outer corners of the front margin of the head and are apparently used only as defensive weapons. The small aperture of the mouth is placed much further back on the lower surface of the head. The maxillulae and maxillae are entirely lost. The maxillipeds (Fig. 121, F) are broad and flattened, composed of six segments tapering gradually to the apex ; the endite of the basipodite is small and the other segments are fringed externally with long plumose hairs. The appendages of the second segment (Fig. 121, G) have lost the pediform character which they have in the larva and in all other Isopoda, and form a pair of overlapping valves closing over the oral area. Each consists of a large oval plate showing traces of three segments and bearing distally one or two minute terminal segments. In feeding, these valves are opened out and the vibratory movements of the maxillipeds produce a current of water which is supposed to carry food-particles towards the mouth. In the female (Fig. 121, C) the mandibles have disappeared and the maxillipeds are similar to but smaller than those of the male. The second thoracic limbs (Fig. 121, D) are, however, very different, consisting of a small leg-like appendage of three segments, having at its base a large oval plate which probably represents an oostegite, although, in this genus, distinct oostegites are not found on any of the other limbs. In the Epicaridea the styliform mandibles are enclosed in a suctorial " oral cone " formed by the upper and lower lips. The maxillulae and maxillae are vestigial or absent, and the lamellar maxillipeds serve, as already mentioned, to keep a current of water flowing through the brood-pouch. In the o vigorous females of many Cymothoidae the mouth-parts are covered by the anterior oostegites, so that the animal is incapable of feeding, and in some genera of Sphaeromidae the mouth -parts of the adult females (with the exception of the maxillipeds) are reduced and functionless. The last seven pairs of thoracic appendages in the Isopoda are typically developed as walking-legs which may, as in the Oniscoidea, present a uniformity of size and shape justifying the name Isopoda, 2O2 THE CRUSTACEA but which in other cases undergo many modifications for prehensile, natatory, and other functions. The coxopodite has the form of a complete segment movably articulated with the body only in the Asellota (Fig. 1 '22, ex). In all other Isopoda it is more or less completely fused with the body and is often expanded into a coxal plate overhanging the attachment of the limb and replacing the pleural expansion of the somite to which it belongs. An interest- ing series illustrating this substitu- tion can be traced within the In Idotea hectica (Fig. 123, A), the pleural plates -pi. FIG. 122. Thoracic somite ofjaera marina, separated and seen from in front. 6s, basipodite of tho- racic leg ; ex, coxopodite ; pi, pleural plate. family Idoteidae. (pi) of the thoracic somites are well developed and completely cover the coxopodites (c.r), of which the outline can be traced on the B. C. Fro. 123. Margin of thoracic somites of, A, Idotea hectica; B, I. ochotensis; C, Chiridotea sobini. The upper figures show the under surface, the lower the upper surface, ex, coxal plate (dotted); pi, pleural plate ; o, rudiment of oostegite. under-side of the plates near the outer edge. In other species of the same genus (Fig. 123, B) the coxal plate projects beyond the outer margin of the pleuron for a part of its length so as to be visible on the upper surface. In yet other species of the genus and in the. allied Chiridotea (Fig. 123, C), the pleura are no longer to be distinguished, and their place is takefi by the greatly de- veloped coxal plates, which are marked off, on the dorsal surface, by distinct sutures, generally allowing a slight amount of move- THE ISOPODA 203 Brl ment. This condition is found in the Cymothoidae, Serolidae, some Sphaeromidae, and in the Tylidae among the Oniscoidea. When the suture line disappears, as in most Oniscoidea (Fig. 118), it is impossible to distinguish the coxal plate from a true pleuron. In all Isopoda, however, with the single exception of the genus Plakarthnum • (Sphaeromidae), the coxopodites of the second thoracic somite (the first free somite) are completely coalesced with the body. The ischiopodite of the thoracic legs is generally more or less elongated, not very short as it is in Tanaidacea and Amphi- poda. The dactylopodite gener- ally ends in a stout claw which may be completely coalesced with the segment or defined from it by a suture. In ./"////" and some other Asellota, how- ever, there are several stout claw-like spines. Very commonly one or more of the anterior pairs of thoracic legs assume a pre- hensile function and become more or less completely sub- chelate, through never forming a perfect chela such as is found among the Tanaidacea. The most perfect natatory legs are found in Munnopsis and related genera, where the last three pairs have the carpus and propodus expanded into oval paddles fringed with long setae. In Amesopons (Valvifera) the fourth and fifth pairs of thoracic appendages (first and second peraeopods) are unrepresented, except by the oostegites in the female, a condition curiously repeat- ing that found in the Caprellidea among Amphipoda. The oostegites (Fig. 124, Brl}, which, in the great majority of Isopoda as in other Peracarida, form the brood-chamber of the female, are clearly seen in the Asellota to be attached to the coxopodites of certain of the thoracic legs. In the other tribes the coxopodites are more or less completely coalesced with the corresponding somites and the oostegites appear to spring from the ventral surface of the body close to the bases of the legs. In certain Cymothoidae and Epicaridea a pair of oostegites is present FIG. 1.'4. Gyge brrnn-liinli.-i (Bopyridae). a, female, seen from ventral side ; Brl, oostegite ; K, abdominal appendages. t>, abdomen of same with adhering niale(cj), further enlarged. (After Cornalia and Panceri, from Clang's Textbook.) 204 THE CRUSTACEA on each of the seven free thoracic somites, the last two pairs, however, being very small. If, as has been suggested above, the coxal lobe of the maxillipeds is homologous with the oostegites, we have the possibility of these structures being developed on all the thoracic appendages. More usually, however, only the first five free somites bear oostegites, and the number may be still further reduced, certain genera of Arcturidae possessing only one pair attached to the fourth free somite. Though the development of the oostegites has been traced in only a few cases, it is known that important differences occur in this respect. In Asellus they appear as small buds from the coxopodites, increasing in size at successive moults until maturity is reached. In the Oniscoidea, on the other hand, no trace of oostegites is visible externally up to the fJwitlt which immediately precedes oviposition. Just before this moult they are developed underneath the cuticle, and when this i^ cast off they at once expand to their full size. Specif structures aiding or replacing the oostegites in containing the brood will be described later in connection with the reproductive system, but it may be noted here that in a few Isopoda (Cassidina and a few allied genera among the Sphaeromidae, and some Epicaridea) the oostegites appear to be entirely wanting. The pleopods are almost always biramous, with a short proto- podite in which three segments can be recognised in Bathynomus, and with lamellar rami generally overlapping each other with the exopodite in front. One or both of the rami may be crossed by a suture-line dividing it into two segments. In the simpler cases all the pleopods are similar (except for the sexual modifications to be described below), both rami serving as respiratory and in many cases also as natatory organs. The latter function is indicated by a marginal fringe of long setae and by the presence of a group of coupling-hooks on the inner side of the peduncle. Pleopoda of this type are found with comparatively slight modifications in the Phreatoicidae, Gnathiiclae, and Cymothoidae, but in some members of the last two families the natatory setae, present in the young, are lost in the adult. This is the case also in the Epicaridea, where the pleopoda of the adult may become much reduced or altogether suppressed, or may, on the other hand, develop into arborescent branchiae. In the aberrant Cymothoid Bathynomus the endo- podites bear tufts of ramified branchial filaments. Even in the Cymothoidae, however, the uniformity of the pleopods is slightly qualified by the fact that the endopodite of the fifth pair is always devoid of marginal setae. This leads to the specialisation of functions found in Sphaeromidae and Serolidae, where the anterior pairs (the first three in Serolidae and many Sphaeromidae) are exclusively natatory and the posterior pairs exclusively branchial. THE ISOPODA 205 plp.- FIG. 125. Under -side of abdomen of male In many Sphaeromidae one or both rami of the last two pairs are transversely plicated so as to increase the respiratory surface. In some members of these families the exopodite of the fifth pair is more or less indurated and forms a kind of operculum protecting the more delicate appendages behind it. Opercular structures having a similar function are formed in different ways in other families. In some Anthuridae the first pair of pleopods are enlarged and cover the remaining pairs. In the Asellidae and Stenetriidae the exopodite of the third pair performs the same function. In the Parasellidae an unpaired plate formed by the coalescence of the first pair forms the operculum in the female, the male having a tripartite operculum formed by the first and second pairs. In the Valvifera (Fig. 125), finally, the pleopods are covered in by the valve -like uropods. A very special line of modification has been followed in the case of the Oniscoidea. In the Ligiidae, which are in many ways the most primitive family, the pleopods are II . .| *il j.v J*i A u nut-i - siue ui uijMiMiu-n ui ilia:*1 all Similar, With the exopotlltes StOUter Idotea bitltica,, with one of the uropods than thp pnHnnnHitPS Vint shar-ino- in removed to show the pleopods. 1,2, than the enaopOCUtes DUt snaiing in 3; first three somites of abdomen ;«, the respiratory function. In many W«mlc segment comprising the tort . , . . .... , three abdominal somites coalesced genera or the remaining families, how- with the teison ; pip, pleopods ; ur, ,1 ]•, F ,1 £• j uropod ; p, penes, attached to sternum ever, the exopodites of the first and Of last thoracic somite. (After sars.) second, and sometimes of all five, pairs are specially adapted for aerial respiration by the development within them of small cavities opening to the exterior by slit-like apertures and giving rise internally to a system of ramifying tubules filled with air (Fig. 126). These tubules, which are lined with a delicate chitinous cuticle, are known as pseudo-tracheae (tr). In certain cases (Oniscus) in which pseudo-tracheae are absent, their place is taken by a system of air-filled spaces immediately under the cuticle of the exopodite. These spaces do not communicate with the exterior, and appear to become filled with air by diffusion through the cuticle. In the majority of Isopoda the second, and sometimes also the first, pair of pleopods present special modifications in the male sex, the only exceptions being the Epicaridea and the small and aberrant family of the Gnathiidae, among the Flabellifera, where no such modification is found. In the remaining families of the Flabellifera and in the Phreatoicidea and Valvifera, the lamellar endopodite of the second pair bears, in the male sex, a rod-like process (appendix masculina) (Fig. 127, m) articulated with its inner 206 THE CRUSTACEA edge and grooved for the reception of the bundles of spermatozoa which it is its function to transfer to the female. This rod appears tr. FIG. 126. A, exopodite of first pleopod of Porcellio scaber ; the tuft of pseudo-tracheae is seen through the transparent cuticle. B, vertical transverse section through same. C, part of section further enlarged, art, point of articulation of exopodite with peduncle ; c, cuticle ; gr, " grooved area " of cuticle; hy, hypodermis ; n, nucleus of hypodermis of pseudo-tracheal tube; o, external open- ing of pseudo-tracheae ; tr, pseudo-tracheae. (After Stoller.) to be the distal segment of the endopodite. In the groups just mentioned the pleopod s of the first pair are similar or present but slight differences in the two sexes, but in the majority of the Oniscoidea and in the Asellota the first pair are also modified in the male sex. In the males of the Oniscoidea the inner ramus of the second pair is styliform and com- m posed of two segments, of which the proximal corresponds to the main part of the endopodite in groups above mentioned. In the family Ligiidae this is the only modification of the pleopods in the male sex, but in all the other Oniscoidea the first pair have the endopodites also styliform, though Second pleopod of an oyigerous female Unsegmented. of Nerocila maculata, showing persistence Jn the Asellota the SBXUal Hiodi- of the appendix masculina (m). en, endo- podite ; ex, exopodite ; e, laminar expan- nCatlOnS OI the pleopods are more sion from outer edge of protopodite. i jj'/rr ii. rn complex and diner from those of all other Isopoda. The second pair are always absent in the female. In the females of Asellidae the first pair are small, uniramous, and separate ; in the Stenetriidae and Parasellidae they are coalesced, ex. FIG. 127. THE ISOPODA 207 forming, in the latter family, a large operculum which completely covers the following pairs. In the male sex the Asellidae have the first pair not greatly different from those of the female ; the second pair are small, with a short protopodite and two rami, each of two segments, the endopodite having a cavity in the distal segment. In the Stenetriidae the first pair are partly fused together and the second pair are to some extent intermediate in structure between those of the other two families. In the Parasellidae the first pair are fused together, their enlarged protopodites each with a small immovable terminal ramus forming the middle plate of a tripartite operculum, of which the lateral parts are the enlarged protopodites of the second pair. The rami of the latter pair are borne on the inner margin of the protopodite ; the exopodite of two segments is hook- shaped and serves to fasten the lateral to the middle plate of the operculum ; the endopodite forms a geniculate copulatory organ of two, sometimes of three segments, with a cavity in its distal part communicating with the exterior by a narrow canal. In most Isopoda the uropods differ widely in form and position from the pleopods. In the adult females of some Epicaridea, however, the uropods do not differ, except in size, from the appendages in front of them, and this is also the case in the Cymothoid genera Anuropus and Bran chur opus, where they resemble the pleopods in structure and position and appear to share their branchial function. Although the uropods are usually biramous, one or other of the rami may disappear in many Asellota, some Sphaeromidae, Valvifera, and Oniscoidea. The uropods are entirely wanting in the Sphaeromid Vireia and in some Epicaridea. The rami are never composed oi more than one segment except in the Asellotan Acanthocope, where the uniramous uropods present three or five segments, including the peduncle. In the Valvifera the uropods are curiously modified in form and position (Fig. 125). They are attached far forwards at the sides of the greatly enlarged telsonic segment, and are folded inwards so as to cover completely the branchial pleopods. Each consists of a large plate formed by the expanded protopodite with the small endopodite at its tip, while the exopodite l is vestigial or absent. In many Anthuridae the exopodite is attached near the base and the endopodite at the tip of the elongated peduncle (leading to the erroneous statement that the endopodite has two segments), and the exopodite is usually folded backwards over the dorsal surface' of the telson. Alimentary System. — The stomodaeum forms, in the majority of Isopoda, a masticatory stomach, which is comparatively simple in 1 What is here called the exopodite is usually regarded as the endopodite, and vice versa. The interpretation given above depends on the assuniptioii that the uropods have reached their present position by a movement of rotation, not of simple translation. 208 THE CRUSTACEA — 3. v.d.- vrr 4. Asellus and Chiridotea and more complicated in the Oniscoidea. The number of types in which its structure has been studied is, however, too small to admit of profitable comparisons between them. In parasitic forms with suctorial mouth-parts the fore-gut is probably always more or less modified, but little is known of the details except in some Epicaridea. In the Entoniscidae the short and narrow oesophagus leads into a spherical or bilobed chamber, the cephalogaster, lined with villi and rhythmically contractile. This is followed by a second chamber, of which the lumen is reduced to a crescentic section by a strong ridge or typhlosole on the dorsal side clothed with setae which constitute an effective strainer. A third chamber with muscular walls, .contracting alternately with the cepha- logaster, is known as Rathke's organ. In the Bopyridae the cephalogaster alone has been recognised. Three pairs of hepatic caeca are present in Idoteidae and Cymothoidae (Fig. 128, hep) ; two pairs in Asellus, Serolis, and the Oniscoidea (except Ligia, which has three) ; and only one pair in Gnathia, Paranthura, and the Epicaridea, in which group the caeca may give off diverticula. It Aega psora, thoracic limbs without exopodites ; first pair modified as maxil- lipeds, coalesced at the base, without epipodite ; remaining pairs variously modified, the second and third commonly prehensile, coxopodites always short, usually expanded as coxal plates, movably connected with the body ; branchial appendages on inner side of coxopodites of some of the thoracic limbs ; pleopods, when fully developed, divided into two sets, the first three pairs with rnultiarticulate rami, the last two pairs generally similar to the uropods, with unsegmented rami ; no sexual modification of pleo- pods ; the young leave the brood -pouch provided with all the appendages of the adult. Historical. — In establishing the order Amphipoda (1816) Latreille excluded from it the genus Cyamus, which he referred to the Isopoda. Later, he established a separate order, Laemodipoda, for Caprella and Cyamus, placing it between the Amphipoda and Isopoda. This arrangement was adopted by H. Milne-Edwards, who further divided the Amphipoda into tAvo families, Gammarina and Hyperina. Kroyer, in 1843, showed very clearly that the Laemodipoda present only an extreme modification of the Amphi- pod type. Dana, in 1852, subdivided the Order into three groups — Caprellidea, Gammaridea, and Hyperiidea, a classification which has held its own to the present time. There appear to be no sufficient grounds for establishing the additional divisions of 224 THE AMPHIPODA 225 Synopidea, Bovallius, and Subhyperini, Delia Valle, and still less reason for the inclusion of the Tanaidacea, advocated by Gerstaecker. On the other hand, the very remarkable genus Ingolfiella, Hansen, may conveniently, for the present, be kept apart in the sub-order Ingolfiellina which Hansen has established for it. Among the more important contributions to our knowledge of the group mentioned in the list of literature at the end of this chapter, attention may be called to Spence Bate's British Museum Catalogue, Spence Bate and Westwood's Monograph of the British species, and to the works of Boeck, Bovallius, Glaus, and Delage. As with so many other groups of Crustacea, the memoirs by G. 0. Sars are numerous and VUL. U 6. t.— Fio. 13-2. Gammartulomuta, £, from the sidp. x 4. a', Jintennule ; a", antenna ; ace, accessory (inner) flagellum of antennule ; br, branchia ; ex, coxal plate ; gn, gnathopods ; pip'", pleopod of third pair ; prp', prp", first and second peraeopods (fourth and tifth thoracic appendages) ; t, telson ; ur, uropod (sixth abdominal appendage); II, VIII, second and^ eighth thoracic somites ; 1, 6, first and sixth abdominal somites. (After G. O. Sars.) of the first importance. Delia Valle's Monograph is valuable for anatomical and biological details, but the systematic part of the work is to be used with caution. The bibliographical history of the Order has been given at length in the admirable Introduction to Stebbing's Report on the " Challenger " collection, and the same author has recently completed a masterly revision of the Gam- maridea for the Tierreich. P. Mayer's Monograph and his later memoirs on the Caprellidea will not soon be superseded as the chief sources of information on that sub-order. MORPHOLOGY. The body of a typical Amphipod, such as Gammarus (Fig. 132), is laterally compressed, with the abdomen of considerable size and 226 THE CRUSTACEA flexed ventrally between the third and fourth somites. The large coxal plates (ex) on the thoracic somites projecting downwards increase the depth of the body and add to the appearance of lateral compression. Even Avithin the sub-order Gammaridea, however, this typical form is sometimes departed from by attenuation of the body or by its dorso-ventral flattening, which, in Pereionotus and some other genera, is carried so far as to give the general appear- ance of an Isopod. In the sub-orders Hyperiidea and Caprellidea the range of modification is much greater. The Hyperiidea, while departing the less widely, on the whole, from the typical form, include such extreme types as the balloon-like Mimonedes and the almost linear Ehabdosoma (Fig. 133). The Caprellidea differ from both the other sub-orders in the vestigial condition of the abdomen FIG. 133. piratuin (Hyperiidea). (After Stebbing, from Eney. Brit.) and the coalescence of the second thoracic somite with the head ; they comprise two families of widely different facies, the filiform Caprellidae (Fig. 134) and the flattened, Isopod -like whale-lice, Cyamidae (Fig. 135). The eyes, when present, are sessile on the sides of the head. Sometimes they coalesce in the median line, and in some Oedi- cerotidae the fused eyes are borne on a projecting frontal lobe. In Ingolfiella, distinct eye-lobes are present (although the eyes are apparently deficient), defined by suture-lines from the antero- lateral margins of the head -region. It is possible that these lobes may represent the eye-lobes of Tanaidacea and the ocular peduncles of more primitive Malacostraca, though the great specialisation of Ingolfiella in other respects is rather against this view. Appendages. — The antennules (Fig. 132, a) consist typically of a peduncle of three segments carrying two flagella. The outer flagellum is usually long and multiarticulate, while the inner (ace) A. dbd FIG. 134. A. Phtisica marina, $, x 5. B, Caprella linearis, 9, x7 (Caprellidea). a', antennule ; a", antenna ; abd, vestigial abdomen, with small appendages in A ; lr, branchiae ; gn. gnathopods (second and third thortcic appendages) ; m, brood-pouch ; prp', prp", first and second peraeopods (fourth and fifth thoracic appendages), wanting in B; IV, V, fourth and lifth thoracic somites. (After G. O. Sars.) br. FIG. 130. ParaciiuuiH.-- bmijiis. A, male, dorsal view, x 4. B, maxillipeds. a', antennule ; a", antenna ; Htnl, vi'sti^ial abdomen ; br, branchiae ; gn, gnathopods (second and third thoracic appendages) ; IV, V, fourth and iifth thoracic somites. (After G. O. Sars.) 227 228 THE CRUSTACEA is frequently absent, and, when present, is generally inconspicuous and composed of few segments. The antennae (Fig. 132, a") when fully developed liave a peduncle of five segments and a more or less elongated nagellum. A scale or exopodite is never developed. A conspicuous conical or spiniform tubercle on the second joint of the peduncle bears the aperture of the antennal gland, indicating that the five segments of FIG. 136. Phronwna sedentaria. a, female ; b, male. A', antennule ; A", antenna ; D, intestine ; Dr, gland in chela of sixth thoracic appendage ; G, genital aperturei of male ; H, heart ; A', branchia ; i Kf, mandible; N, ventral nerve-chain; 0,|eye ; Ov, ovary. [.(Fr' the peduncle must be derived from the six-segmented condition by a coalescence of two segments distal to the gland-opening, probably the third and fourth. In many cases the antennules and antennae are more strongly developed in the male than in the female sex, and bear more numerous sensory setae. In some Hyperiidea (Fig. 136) the antenna, though well developed in the male, is represented in the female only by a rounded tubercle containing the antennal gland on the front of the head. In some cases the THE AMPH1PODA 229 antennae have almost a pediform character, the segments of the peduncle being long and stout and the flagellum reduced, as in the Amphithoidae and allied families and conspicuously in the Corophiidae. The mandibles have usually the typical Peracaridan structure with molar process, serrated incisor process, lacinia mobilis on the left mandible or on both, a row of spines, and a palp of three segments, but any of these parts may be modified or absent. The palp, in particular, may be present or absent in genera otherwise very closely related. The maxillulae (Fig. 137, A) are remarkable in that they commonly exceed the maxillae in size and complexity of structure. Two endites are present, springing, according to Hansen's inter- pretation, from the first and third segments, and the fourth and B. C. Flo. 137. A, maxillula, B, maxilla, C, maxillipeds, of Socarne* iiitiriitii'iiliitiis (Gammarklea). The distal segments of the left inaxilliped are omitted. 1-7, .segments of the appendages; l^-l3, endites of the respective scgnu-nts. (After Hansen.) fifth segments form a " palp " which is turned forwards and inwards, resembling in appearance, and no doubt also in function, a third endite. The maxillae (Fig. 137, B) are small in size and simple in form, consisting mainly of two plates which, according to Hansen, are the endites of the second and third segments. The lower lip may attain to greater complexity than in most other divisions of the Malacostraca and its modifications are of some systematic importance. The two main lobes (paragnatha) of which it else- where consists are in many cases supplemented by a pair of accessory lobes lying between them, while the main lobes them- selves may be produced at the outer corners or each divided by incision into two as in the Amphithoidae. The first thoracic appendages or maxillipeds (Fig. 137, C) are always coalesced at the base, the coxopodites fusing to form an unpaired plate. The basipodite is produced into an endite, usually referred to as the " inner plate " (/ 2), which may be armed with teeth, spines, or setae, but does not carry coupling-hooks. The 230 THE CRUSTACEA ischiopodite also bears an endite (the " outer plate," 1 3), and the remaining four segments form the " palp." The palp is not unfrequently abbreviated by the suppression of one or two of its segments, and the coalescence of the proximal region may involve the basipodites partially or completely. Both these modifications are carried to an extreme in the Hyperiidea (Fig. 138), where the maxillipeds are represented by an FIG. 138. unpaired plate carrying a pair of movable, un- Reduced maxim- jointed appendages, representing in all probability (After G^o^Sa'rs." the ischiopodites with their endites, and a median process corresponding to the coalesced inner plates. In the Cyamidae the maxillipeds are sometimes of normal structure, but they may be igreatly reduced (Fig. 135, B), and in Platycyamus they are represented only by an unpaired plate without any trace of articulations. In Cyamus nodosus the interesting observation has been made that the young animals taken from the brood-pouch have well-developed maxillipeds with the full number of segments, although in the adults of this species they are reduced to a pair of unjointed appendages attached to the common basal plate. Of the remaining seven pairs of thoracic appendages, the first two are commonly, though not invariably, modified for prehension, and are distinguished as gnathopods (Figs. 132, 134, 135, gn) from the remaining five pairs, the peraeopods, which are generally organs of locomotion. Each limb consists of the usual seven segments. The coxopodite 'is always short, but is usually expanded externally to form a coxal plate (Fig. 132, ex), sometimes of great size ; internally it bears the branchial plate (br) and oostegite when these are present. In the Caprellidea and Ingolfiellidea the coxopodite remains small. In some Hyperiidea it is entirely coalesced with the somite. The basipodite is usually more or less elongated ; the ischiopodite, on the other hand, generally short, contrary to the rule among the Isopoda. The terminal claw is usually coalesced with the dactylopodite. The lateral compression of the body in most Amphipoda has lead to a separation of the thoracic legs into an anterior group of four (the two gnathopods and the first two paraeopods) and a posterior of three, which are opposed to each other in the direction of the principal articulations. In the case of the anterior group the limb is bent forwards at the articulation between the ischio- podite and meropodite, and backwards at that between meropodite and carpopodite, and the dactylopodite points backwards ; in the posterior group these directions are reversed and the dactylopodite points forwards except in the case of certain Gammaridea, where the direction of the dactylopodite (but not of the other segments) THE AMPHIPODA 231 is reversed.1 The distinctness of these two groups is further expressed by the relations of the coxal plates. Where these are small or of moderate size they are of similar form throughout the series, but when, as in many Gammaridea (Fig. 132), enlargement takes place, it is mainly the first four that become expanded so as to cover from the outside the basipodites, or even the whole limb ; the last three coxal plates in this case generally remain small and their place in protecting the gills within is taken by the expanded basipodites. The two pairs of legs following the gnathopods (the fourth and fifth of the thoracic series) are, among the Gammaridea (Fig. 132, prp', prp"\ not unfrequently more or less different from the succeeding pairs. In the Capuellidea, these two pairs are vestigial or absent except in the genus Phtisica (Proto) (Fig. 134, A) and its immediate allies. In the gnathopods of the Gammaridea every gradation may be traced from the simple, non-prehensile leg to the well-formed, sub-chelate, or perfectly chelate type, and even to more complex forms, as in Leucothoe and Aora. The gnathopods of Ingolfiella have a very unusual structure, the propodite and dactylopodite together forming the movable, finger which is opposed to the expanded carpopodite. Some of the peraeopods may show modifi- cation for prehensile purposes in Gammaridea, and in Polycheria all of them are sub-chelate. In the Hyperiidea much greater variety occurs, and any of the peraeopods except the last pair may be transformed into a chela, sometimes of large size (Fig. 136). The gnathopods, in this group, are always small. In the Caprellidae (Fig. 134) the gnathopods are sub-chelate and the last three pairs of peraeopods are also fitted for grasping. A point of interest with regard to these peraeopods is the existence in the basipodite of a definite "fracture-plane" at which the limb breaks off in antotomy. A similar fracture-plane is found in the legs of many Decapoda, where the habit of autotomy is highly developed. The branchiae (Figs. 132, 134, 135, br ; Fig. 136, K] are attached to the inner surfaces of the coxopodites, near the posterior border. They are generally vesicular or lamellar in form, and in some Gammaridea the respiratory surface is in- creased by numerous transverse ridges or folds. In some species of Cyamidae (Of/amus physeteris) the branchiae are ramified. Accessory branchiae occur in certain Gammaridae and Talitridae on some of the thoracic lirnbs and also on the first abdominal somite. The greatest number of branchiae is six pairs, occurring on the last six thoracic limbs in many Gammaridea. The number 1 The correlation between the lateral compression of the body and this gronpi:ig of the legs is well illustrated by comparison with Phreatoicus, the only Isopod where the body is laterally compressed and where the legs are divided into two groups exactly as in Am phi pods. 232 THE CRUSTACEA in this sub-order, however, is not unfrequently reduced to five, four, or even three pairs. In Hyperiidea the last pair of thoracic legs never carry branchiae and the number may vary from five to two pairs. In Caprellidea as a rule only two pairs of branchiae are present, on the fourth and fifth thoracic somites, but in some genera of Caprellidae there is an additional pair on the third somite (that of the second gnathopods). The oostegites spring from the inner surface of the coxopodites on the proximal side of the branchial plates. Four pairs are commonly present, on the third to the sixth thoracic somites, but the number may be reduced. In the Caprellidea (Fig. 134, in} two pairs only exist, on the fourth and fifth somites, but in addition a pair of small valvular appendages covering the external apertures of the oviducts on the sixth somite are probably to be regarded as vestiges of a third pair. Vestigial oostegites are stated to occur normally in the male sex in a few cases (e.g. Cycimus globicipitis and Synurella polonica). It is very characteristic of the Amphipoda that the abdominal appendages are sharply divided into two groups differing in structure. In the majority of Gammaridea (Fig. 132) and Hyperiidea (Fig. 136) all the six pairs are present, the first three pairs are turned forwards and consist of a peduncle carrying two subequal rami, each of which is multiarticulate and fringed with long setae ; the inner side of the peduncle bears distally a number of retinacula, and the first joint of the endopodite has internally one or two peculiar "cleft spines" which no doubt serve the same purpose of coupling together the pair of appendages. These limbs are the chief natatory organs among Amphipoda, and they also serve when the animal is at rest to maintain a current of water over the gills. The last three pairs of abdominal appendages are directed backwards and are generally similar to each other, so that the name uropods is commonly used to include them all, though in not a few cases the last pair retain, in details of form and size, some mark of that differentiation from the preceding append- ages which they show in other orders of the Malacostraca. As a rule, the three pairs are biramous ; not unfrequently the exopodite of the last pair consists of two segments, but except for this the rami of all are unjointed. Sexual differences not unfrequently occur in the size and structure of the last pair, but in no Amphipoda are any of the pleopods modified as copulatory organs. The abdomen of most Caprellidea is unsegmented and may bear vestigial appendages to the number of three pairs (Fig. 134, aid). An interesting link with the normal Amphipoda is constituted by the genus Cercops, where five distinct somites and a terminal piece (perhaps the telson) are present. The first and second somites bear, in the male sex only, minute filiform appendages of two THE AMPHIPODA 233 segments which appear to correspond to the first and second pleopods of the normal type, while the fourth and fifth somites carry in both sexes stout two-segmented limbs answering to the first and second uropods. In most other Caprellidae the rudimentary appendages are more fully developed in the male than in the female sex, but in a few the abdomen is quite without appendages. In the male Cyamidae a median appendage is present which seems to result from the fusion of a pair of uropods. In Ingolfiella the first three pairs of abdominal limbs are represented by small triangular plates, sometimes with a minute basal segment, without any trace of rami. The last pair is vestigial. Alimentary System. — The alimentary canal of the Amphipoda appears to differ from that of most Isopoda in the much greater development of the mid-gut region, which forms the greater p'art of its length. The stomodaeum forms a masticatory or triturating stomach, the structure of which appears to be fairly uniform throughout the Gammaridea and Caprellidea, and to be more' or less simplified in the Hyperiidea. When fully developed it presents anteriorly two lateral ridges projecting into the cavity, armed with spines and stiff setae. These ridges are moved by powerful muscles passing outwards to the body-wall on either side, and appear to be the most important instruments of trituration. Posterior to these are two pairs of setose ridges running more or less transversely, while in the floor of the cavity is a strong ridge ending behind in a free tongue -like process and carrying anteriorly four comb -like rows of iridescent setae. The cbitinous lining of the stomodaeum projects backwards into the beginning of the mid-gut as a cuticular funnel. In Phronima, where the apparatus appears to be adapted for straining rather than for masticating the food, the Avhole stomach is telescoped for a little way into the capacious mid-gut. The hepatic caeca are generally four in number and of considerable length, but in a few genera of Gammaridea and in most Hyperiidea only one pair is present, while in the Caprellidea the ventral pair and in Phronima both pairs remain rudimentary. Just above the point where the hepatic caeca communicate with the gut by a common aperture on each side, a short dorsal, forwardly directed caecum, unpaired in Gammaridea, paired in Hyperiidea and Caprellidea, arises from the anterior end of the mid-gut. At the posterior end of the mid-gut, at its junction with the proctodaeum, a second pair of caeca of very varying size is commonly present in the Gammaridea. In a few cases only a single unpaired caecum is present (Melita), and in Synwella, though two tubes are present in the young, that on the right side atrophies and only that on the left persists. Similar paired caeca are only exceptionally present in Hyperiidea (I'ibilia) and Caprellidea (CupreUa). To these posterior caeca of the mid -gut an excretory function has been 234 THE CRUSTACEA attributed, mainly on the ground that they sometimes contain calcareous concretions ; as they are outgrowths of the mesenteron, it is impossible to regard them as homologous Avith the Malpighian vessels of insects. In the Hyperiidea the anterior part of the mid-gut may be dilated while the posterior part remains narrow. It is stated to be lined by a very delicate cuticle of non-chitinous nature. The proctodaeum is short, rarely reaching forward to the posterior limit of the thorax, or, in the Caprellidea, into the penultimate thoracic somite. An " anal gland " in the form of a small diverticulum is described in Synurella. Circulatory System. — The circulatory system of Amphipoda is remarkable for the imperfect development of the arterial vessels and the consequent lacunar character of the greater part of the circulation ; as compared with that of the Isopoda it differs also in the position of the tubular heart, which extends through the greater part of the thoracic somites but does not reach the abdomen. It lies in a spacious pericardium which may extend backwards into the abdomen. Three pairs of ostia are present in most Gammaridea, but Corophivm has only one pair ; the Caprellidea have three pairs ; in the Hyperiidea only two pairs are present as a rule, but in Phronima, although the male has only two, the female has three pairs. The heart is continued at either end into the anterior and posterior aortae, which are denned by valves. In addition, a pair of arteries supplying the eyes and adjacent parts may arise from its anterior end, but only in the Hyperiidea are there two or three pairs of lateral arteries arising from the heart in the thorax and comparable to the thoracic arteries of the Isopoda. The anterior aorta is stated to divide in the median plane to encircle the brain in a " pericerebral vascular ring " very characteristic of the order, with which is connected (except in Caprellidae) an oesophageal ring similar to that of the Isopoda. It seems probable, however, that these rings are rather lacunar spaces than well-defined vessels. There is no trace of the subneural artery commonly developed in Isopods, the oesophageal ring opening posteriorly into the great ventral sinus which extends through the whole length of the body, and into which the posterior aorta also empties its contents at or near the end of the abdomen. Except perhaps in those forms (Hyperiidea) which possess lateral thoracic arteries, the appendages of the thorax and abdomen receive their blood-supply from this ventral sinus. Within the appendages the afferent blood-streams are contained in well-marked vessels which send branches to the gills. The efferent vessels of the appendages unite to form in each somite (except sometimes in the abdomen) a pair of pericardial vessels by which the blood is returned to the heart. In comparing the circulatory system with that of the Isopods, it is clear that the course followed by the blood in the abdominal THE AMPHIPODA 235 region is similar in both cases, the ventral sinus sending blood into the appendages, from which it is returned to the pericardium by the centripetal sinuses ; the ventral sinus in Amphipoda receiving, however, arterial blood directly from the posterior aorta. In the thorax the course of the circulation is very different. In the Isopoda the appendages receive blood directly from the heart by means of the lateral thoracic arteries and return it to the ventral or lateral thoracic sinuses ; in the Amphipoda, on the other hand, the limbs with their branchial plates receive blood from the ventral sinus and return it directly to the pericardium by the peri- cardial vessels. The ventral sinus of the Amphipoda, however, does not wholly correspond to the ventral or lateral sinuses of the Isopoda, but, since it receives arterial blood both anteriorly and posteriorly from the aorta, represents in part also the sternal system of arteries in the last-named group, while the homologues of the thoracic afferent pericardial vessels of the Amphipoda are to be found in the blood-stream which enters the pericardium of the Isopoda through the minute apertures at its anterior end. Excretory System. — No trace of a maxillary gland has been recognised in any Amphipod. The antennal gland, on the other hand, is rarely, if ever, wanting. It has been most fully studied in certain freshwater Gammaridae, where it presents the usual divisions of end-sac, convoluted tube, and duct, the latter being some- times dilated into a vesicle. The gland is usually contained within the first segment of the antenna, and its external aperture is at the tip of a conical or spiniform process of the second segment. " Coxal glands" are described as present in all the thoracic and the first three abdominal appendages of Gammarus. They consist of small groups of cells, without duct or opening to the exterior, rendered visible in the living animal by feeding with carmine. Dermal Glands. — In many Amphipoda there are found groups of unicellular glands lying in the thoracic appendages and opening by fine ducts at or near the extremity of the limb. Among the Gammaridea such glands are Avell developed in the Photidae, Aoridae, Amphithoidae, Jassidae, Corophiidae, and some of the allied families, where they are confined to the fourth and fifth thoracic appendages, lying mainly in the basipodite and connected with a duct which opens on the tip of the claw. In the Ampelis- cidae similar glands occur but are more widely distributed on the other appendages of the body. In all the families named the glands secrete a cementing material which is used in building up with particles of mud or sand or fragments of Aveed the cases or tubes in which the animals live. In the Talitridae, which burrow in sand, glands of similar structure are scattered over the surface of the body and appear to secrete a mucinous substance which lines the burrows. Glands of very similar structure to those just 236 THE CRUSTACEA described, but apparently with different functions, occur in many Hyperiidea. In the Phronimidae, where they have been most carefully studied, they occur in all the thoracic legs, and the ducts open at the end of the limb or, in the case of the ante- penultimate pair, on teeth of the palmar edge of the chela. It is supposed that in this case they act as poison -glands. In many Caprellidae groups of gland-cells also supposed to secrete a poison are present in the hand of the second gnathopods, their ducts opening on a prominent tooth of the palmar edge. Finally, glands of similar 'structure may be found, as in many other Crustacea, in the oral appendages and on the oesophageal walls, and are supposed to be salivary in function. Nervous System. — The ventral nerve-chain presents, in the majority of the Gammaridea, twelve pairs of ganglia connected by double longitudinal commissures, the sub-oesophageal supplying the mouth-parts, followed by seven corresponding to the free thoracic somites and four abdominal ganglia ; but the number may be reduced, especially among the Hyperiidea, by the coalescence of one or two of the anterior thoracic with the sub-oesophageal ganglia, the fusion of the last two thoracic, and the restriction of the abdominal ganglia to three pairs. In the Caprellidea, though four pairs of abdominal ganglia may be distinct in the young, they become in the adult fused into a single mass approximated to the last thoracic ganglion and lying in the penultimate thoracic somite. In Phronima the nerves to the mouth-parts arise from the oeso- phageal commissures close to the under-side of the brain. A post-oral antennal commissure such as exists in the Isopoda is perhaps in- dicated by the presence (in Gammaridea and Caprellidae) of a median foramen piercing the sub-oesophageal ganglionic complex near its anterior margin and giving passage to an unpaired strand of muscle running between the lower surface of the stomach and the lower lip. Sense-Organs. — The paired eyes of the Gammaridea show great diversity of size and disposition. They are rarely so large as to occupy the greater part of the surface of the head (Trischizostoma), while on the other hand they are in not a few cases quite vestigial or apparently absent. In the Oedicerotidae, as already mentioned, they are coalesced to form a single organ which is advanced to the front of the rostral process, and in a few cases (Tiron, Synopia) a pair of small accessory eyes are placed below the main eyes. This leads to the very remarkable condition found in the Ampeliscidae, which possess two pairs of eyes, each of which is made up of numerous ommatidia differing only in details from those forming the eyes of other Amphipoda, underlying a single lenticular thickening of the cuticle. In the Caprellidea the eyes are always small, but in the THE AMPHIPODA majority of the Hyperiidea (Fig. 13G, 0) they are enormously enlarged, occupying nearly the whole surface of the head. In this case each eye is commonly divided into two parts, the dorsal differing in the great elongation of the crystalline cones and in other details from the lateral division. Apart from the Am- peliscidae, the eyes of the Amphipoda are characterised by the fact that the corneal covering is not faceted, or, in other words, that corneal lenses are not formed. In addition to " sensory filaments " of the usual type borne by the main flagellum of the antennule, and sometimes particularly numerous on the enlarged proximal segment of the flagellum (Lysianassidae, Hyperiidea), many Gammaridea have on the antennae, and sometimes on the antennules, peculiar organs known as calceoli. These are, in the simplest cases, little flattened vesicles attached by a narrow stalk, but in some the structure is more complex. They are often, but not always, confined to the male sex, and have been variously interpreted as olfactory or auditory organs, as adhesive suckers, and even as sexual ornaments. A pair of statocysts have been described in connection with the anterior part of the brain in Oxycephalidae. Rep-odudive System. — The paired ovaries and testes are of simple tubular form and lie in the thoracic region. The testes are continued posteriorly into short vasa deferentia which may be slightly dilated into seminal vesicles and open on short papillae on the sternum of the last thoracic somite. The oviduct leaves the ovary at about the middle of its length and opens on the inner surface of the fifth coxal plate (sixth thoracic somite). It is stated that an actual opening does not exist until the moment when the eggs are extruded. The spermatozoa consist of a slender, rod-like head, to which a filiform tail, stated to exhibit vermiform move- ments in some cases, is attached at an acute angle. Definite spermatophores are not formed, at least in the majority of Amphi- poda. The spermatozoa are deposited on the ventral surface of the body of the female immediately before the eggs are extruded, and fertilisation is external. The occurrence of ova within the testis has been observed in species of Orchestia, where it is perhaps universal in young individuals. DEVELOPMENT. Segmentation is at first total, later becoming superficial, with early differentiation of the blastoderm on the ventral side. The teloblastic mode of growth in the post-naupliar region of the embryo which occurs in many Isopoda and Mysidacea does not present itself. A " dorsal organ " is early developed as a median thickening 238 THE CRUSTACEA of the ectoderm, the cells of which become pyriform, projecting inwards and connected with the exterior by a narrow neck. The larval cuticle, which is formed soon after the differentiation of appendages has begun, remains adherent to the dorsal organ after separating from the rest of the ectoderm. Later, the cells lose their apparently glandular character and become invaginated as a thin-walled vesicle, which persists after the embryo is freed from the egg-membrane. The dorsal curvature of the blastoderm as it lies upon the spherical surface of the yolk is early exchanged for a ventral curvature as the abdominal region of the developing embryo becomes folded downwards and forwards. The embryo is not liberated from the egg-membrane until the body and appendages have assumed more or less their final form. The young, however, remain within the marsupium for some time longer, leaving it finally only at the ecdysis which precedes the next act of oviposi- tion. The accounts of some older writers, according to which the young after leaving the marsupium of the parent returned to it for shelter when alarmed, have not been confirmed by any modern observer. The post-embryonic development in most Amphipoda consists mainly in the gradual assumption of secondary sexual characters and other features of subordinate importance. Only in the Hyperiidea, and notably in the Platyscelidae, do the changes of form occurring after the young leave the brood-pouch deserve the name of metamorphosis. REMARKS ON HABITS, ETC. The Hyperiidea and Caprellidea are exclusively marine (the reported occurrence of a Caprellid in the Lake of Geneva rests on insufficient evidence), as are also the great majority of the Gammaridea. The Hyperiidea are pelagic in habitat, sometimes ranging from the surface to great depths, and having often an exceedingly wide horizontal distribution. The Caprellidea, though for the most part inhabiting shallow water and almost or quite without the power of swimming, include some species of almost cosmopolitan range. The Ingolfiellidea have also a very wide distribution, for the two species which at present compose the sub-order come from Davis Straits and from the Gulf of Siam, at depths of 1870 fathoms and of 1 fathom respectively. The marine Gammaridea are rarely pelagic ; they are abundant in the littoral region and penetrate to great depths. The vast abundance of individuals and species in Arctic and Antarctic seas is especially noteworthy. As regards the freshwater species, the predominance of the genus Gammarus and its nearest allies must THE AMPHIPODA 239 be noticed ; the occurrence of species from genera otherwise marine in the " relict " faunas of lakes in Northern Europe and Amei-ica and in the Caspian Sea ; the peculiar blind subterranean species (Niphargus, etc.) which come to the surface in wells and penetrate into the abyssal waters of deep lakes ; and the radiation of single genera into numerous species in a limited area in the Gammari of Lake Baikal and the Hyalcllae of Lake Titicaca. The only terres- trial Amphipoda occur among the Talitridae, which in northern latitudes live, for the most part, just above tide-marks, but in the warmer regions of the globe penetrate inland to great distances. Perhaps no Amphipoda except the whale-lice (Cyamidae) (Fig. 135) are truly parasitic, though some forms with suctorial mouth- parts seem to be semi-parasitic on fish (Trischizostoma). Many species, however, are commensal with sponges and other organisms. The pelagic Phronima (Fig. 136) lives in a transparent barrel-shaped case fashioned from the swimming-bell of a Siphonophore or from a test of a pelagic Tunicate. The Gammaridea probably include the smallest as well as the largest Amphipoda, for many species do not exceed two or three millimetres in length. The largest is Alicella gigantea, Chevreux, which reaches a length of 140 mm. PALAEONTOLOGY. Fossil remains of Amphipoda are exceedingly rare, and although various problematical fossils from Palaeozoic rocks have been referred to this group, it is only in the Tertiary and Post-Tertiary deposits that undoubted Amphipoda have been found. All belong to the Gammaridea, and though a genus Palaeogammarns has been established for a species found in Baltic amber, its generic and even specific distinctness from some of the living forms of Gammarus is uncertain. AFFINITIES AND CLASSIFICATION. Although the Amphipoda plainly belong to the Peracaridan division of the Malacostraca, their relation to the other Orders of the division is by no means so clear as in the case of the Isopoda. It seems very likely that their affinity to the Isopoda is not so close as has been supposed. Apart from the sessile eyes and the segmentation of the body, characters which, there is reason to suppose, have originated independently in at least one other case (Koonunga), almost the only point of agreement between the two Orders is found in the possession of coxal plates on the thoracic somites. But coxal plates are not developed in the most primitive Isopoda, the Asellota, and if their appearance in that Order was later than the acquisition of the typical Isopodan characters, the 240 THE CRUSTACEA coxal plates of the Amphipoda must have had an independent origin. The possession, by the typical Amphipoda, of a well- developed inner ramus on the antennule and of a palp on the maxillula are also among the characters which suggest that they diverged from the common stock before the origin of the Isopoda. It is quite probable, however, that their origin must be sought for still further back. They do not show the late appearance of the last thoracic limbs and the tendency to coalescence of the telson with the last somite — characters which are common to Cumacea, Tanaidacea, and Isopoda — while they possess a well -developed antennal gland, which is never more than vestigial in these Orders. Further, if the branchiae of the Amphipoda be, as Glaus suggests, epipodites which have become shifted from the outer to the inner surface of the thoracic coxopodites, it follows that the Amphipoda must be supposed to have diverged, not later than the origin of the Mysidacea, from the primitive caridoid stock of the Peracarida. A regards the four sub-orders recognised in the classification given below, it must be admitted that the differences separating the Gammaridea from the Caprellidea are not very profound. A Caprellid which should unite the abdomen of Cercops with the thoracic limbs of Phtisica would be very hard to exclude from the Gammaridea. The Hyperiidea hardly depart more widely from the Gammaridea, but the retention of the three divisions is at least convenient. The sub-order Ingolfiellidea, established by Hansen for the two species of the genus Ingolfiella, is not admitted by Stebbing, who places the genus among the Gammaridea, but there seems to be force in Hansen's contention that its inclusion in that sub- order would logically require the absorption of the Caprellidea also. The difficulties in the way of classifying the Amphipoda are of a different order from those met with among the Isopoda. We have to deal with a vast diversity of forms within a comparatively small range of morphological variation, with the consequence that genera and even families have to be established on trivial characters, and their limits are often hard to define. ORDER Amphipoda, Latreille (1816). SUB-ORDER 1. Gammaridea, Dana (1852). Head not coalesced with second thoracic somite ; palp of maxillipeds with two to four segments ; all the thoracic legs present, usually with well-developed coxal plates; abdominal somites generally distinct, with well-developed appendages ; eyes rarely very large. Family LYSIAXASSIDAE. Lysianassa, H. Milne-Edwards ; Alicella, THE AMPHIPODA 241 Chevreux ; Socarnes, Boeck ; Trischizostoma, Boeck. Family STEGO- CEPHALIDAE. Stegocephcilus, Kroyer. Family AMPELISCIDAE. Ampelisca, Kroyer. Family HAUSTORIIDAE. Haustorius, S. Mliller ; Urothoe, Dana. Family PHOXOCEPHALIDAE. PJwxocephalus, Stebbing. Family AMPHI- LOCHIDAE. Amphilochus, Spence Bate. Family LEUCOTHOIDAE. Leucothoe, Leach. Family ANAMIXIDAE. Anamixis, Stebbing. Family METOPIDAE. Metopa, Boeck. Family CRESSIDAE. Cressa, Boeck. Family STENO- THOIDAE. Stenothoe, Dana. Family PHLIANTIDAE. Phlias, Gue"rin ; Pereionotus, Spence Bate and Westwood. Family COLOMASTIGIDAE. Colomastix, Grube. Family LAFYSTIIDAE. Lafystius, Kroyer. Family LAPHYSTIOPSIDAE. Laphystiopsis, G. 0. Sars. Family ACANTHONOTO- ZOMATIDAE. Accmthonotozoma , Boeck. Family PARDALISCIDAE. Parda- lisca, Kroyer. Family LILLJEBORGIIDAE. Lilljeborgia, Spence Bate. Family OEDICEROTIDAE. Oediceros, Kroyer. Family SYNOPIIDAE. Synopia, Dana. Family TIROXIDAE. Tiron, Lilljeborg. Family CALLIOPIIDAE. Calliopius, Lilljeborg. Family PLEDSTIDAE. Pleustes, Spence Bate. Family PARAUPHITHOIDAE. Paramphithoe, Bruzelius. Family ATYLIDAE. Atylus, Leacli. Family MELPHIDIPPIDAE. Mel- phidippa, Boeck. Family EDSIRIDAE. Eusirus, Kroyer. Family BATEIDAE. Batea, F. Muller. Family PONTOGENEIIDAE. Pontogeneia, Boeck. Family GAMMARIDAE. Gammarus, Fabricius (Fig. 132) ; Nipkargus, Schiodte ; Synurella, Wrzesniowski ; Melita, Leach. Family DEXAMINIDAE. Dexamine, Leach ; Polycheria, Haswell. Family TALITRIDAE (ORCHESTIIDAE). Talitrus, Latreille ; Orchestia, Leach ; Hyalella, S. Smith. Family AORIDAE. Aora, Kroyer. Family PHOTIDAE. Photis, Kroyer. Family ISAEIDAE. Isaea, H. Milne-Edwards. Family AMPHITHOIDAE. Ainphitlwe, Leach. Family JASSIDAE (PODOCERIDAE auctt.). Jassa, Leach (Podocerus auctt.). Family COROPHIIDAE. Corophium, Latreille. Family CHELURIDAE. Chelura, Philippi. Family PODOCERIDAE. Podocerus, Leach ( = Platophium, Dana). Family HYPERIOPSIDAE. Hyperiopsis, G. 0. Sars. SUB-ORDER 2. Hyperiidea, Dana (1852). Head not coalesced with second thoracic somite ; palp of maxillipeds absent ; all the thoracic legs present, coxopodites small or coalesced with the body ; abdominal somites generally distinct, with well-developed appendages ; eyes generally very large. Family SCINIDAE. Scina, Prestandrea. Family VIBILIIDAE. Vibilia, H. Milne- Edwards. Family CYLLOPODIDAE. Cyllopus, Dana. Family LAXCEOLIDAE. Lanceola, Say. Family CYSTISOMATIDAE. Cystisoma, Guerin. Family PARAPHRONIMIDAE. Paraphronima, Glaus. Family PHRONIMIDAE. Phronima, Latreille (Fig 136). Family MIMOXECTIDAE. Mimonectes, Bovallius. Family HYPERIIDAE. Hyperia, Latreille ; Plironi- mopsis, Glaus ; Euthemisto, Bovallius. Family PHROSINIDAE. Phrosina, Risso. Family PHORCORRHAPHIDAE. Phorcorrhaphis, Stebbing. Family PLATYSCELIDAE. Platyscelus, Spence Bate. Family SCELIDAE. Thyropus, Dana. Family PRONOIDAE. Pronoe, Guerin. Family TRYPHANIDAE. Trypliana, Boeck. Family OXYCEPHALIDAE. Oxycephalus, H. Milne- Edwards ; Rhabdosoma, White (Fig. 133). 16 242 THE CRUSTACEA SUB-ORDER 3. Caprellidea, Dana (1852). Head coalesced with second thoracic somite ; palp of maxillipedd with from one to lour segments (rarely absent) ; rarely all the thoracic legs well developed, fourth and fifth pairs usually vestigial or absent, coxo- podites small or coalesced with the body ; abdominal somites generally all fused, appendages vestigial ; eyes small. Family CAPRELLIDAE. Cercops, Kroyer ; Phtisica, Slabber (Proto, Leach) (Fig. 134, A); Caprella, Lamarck (Fig. 134, B). Family CYAMIDAE. Cyamus, Latreille ; Platycyamus, Liitken ; Paracyamus, G. 0. Sars (Fig. 135). SOB-ORDER 4. Ingolfiellidea, Hansen (1903). Head not coalesced with second thoracic somite ; palp of maxillipeds with four segments ; all the thoracic legs present, coxopodites small ; abdominal somites distinct, first three and last pairs of abdominal appendages vestigial ; eyes wanting but articulated eye-lobes are present. Family INGOLFIELLIDAE. Ingolfiella, Hansen. LITERATURE. 1. Bate, 0. Spence, Catalogue of the . . . Amphipodous Crustacea in ... the British Museum, pp. iv + 399, 58 pis. 8vo, 1862. 2. Bate, C. Spence, and Westwood, J. 0. A History of the British Sessile-Eyed Crustacea, 2 vols., 1863 and 1868. Issued in parts, 1861-1868. 3. Boeck, A. De Skandinaviske og Arktiske Amphipoder. Pp. 711, 32 pis. 4to. Chvistiania, 1872-1876. 4. Bovallius, C. Mimonectes, a Remarkable Genus of Amphipoda Hyperiidea. Nova Acta R. Soc. Sci. Upsal. (3), xiii. No. 4, pp. 15, 3 pis., 1885. 5. Contributions to a Monograph of the Amphipoda Hyperiidea. Kg]. Svenska Vet.-Akad. Hand!., Ny Foljd. xxi. No. 5, pp. 72, 10 pis., 1887 ; and xxii. No. 7, pp. 434, 18 pis., 1889. 6. The Oxycephalids. Nova Acta R. Soc. Sci. Upsal. (3), xiv. No. 4, pp. 141, 7 pis., 1890. 7. Chevreux, E. Amphipodes provenant des Campagnes de 1'Hirondelle, 1885- 1888. Resultats des Campagnes scientifiques . . . Monaco, xvi., 1900. 8. Claus, C. Zur Naturgeschichte der Phronima sedentaria, Forsk. Zeitsohr. wiss. Zool. xxii. pp. 331-338, pis. xxvi.-xxvii., 1872. 9. Der Organismus der Phronimiden. Arb. zool. Inst. Wien, ii. pp. 59- 146, pis. iii.-x., 1879. 10. - • Die Platysceliden, pp. 77, 26 pis. 4to. Wien, 1887. 11. Delage, Y. Contribution a 1'etude de 1'appareil circulatoire des Crustaces edriophthalmes inarins. Arch. zool. exper. ix. pp. 1-173, pis. i.-xii., 1881. 12. Delia Valle, A. Ganimarini del Golfo di Napoli. Fauna und Flora des Golfes von Neapel, Monogr. xx. 1 vol. and atlas, 1893. 13. Dybowsky, B. N. Beitr. z. naheren Kenntniss der in dem Baikal-See vor- kommenden . . . Gammariden. HoraeSoc. Entom. Rossicae, x., Beiheft. pp. 190, 14 pis. 4to, 1874. 14. Hansen, H. J. The Ingolfiellidae, fam. n., a New Type of Amphipoda. Journ. Linn. Soc. London, Zool. xxix. pp. 117-133, pis. xiv.-xv., 1903. THE AMPH1PODA 243 15. Liitken, C. F. Bidrag til Kundskab om Arterne af Slaegten Cyamus, Latr., eller Hvallusene. K. Danske Aridensk. Selsk. Skr. (5), Naturvid. Afd. x. (3), pp. 231-284 + 3 pp. resume, pis. i.-iv., 1873 ; op. cit. (6), iv. (4), pp. 315-322, 1 pi., 1887 ; o/>. dt. (6), vii. (9), pp. 419-434, 1 pi., 1893. 16. Mayer, P. Caprelliden. Fauna und Flora des Golfes von Neapel, Moiiogr. vi., 1882 ; and Nachtrag zn den Caprelliden, Monogr. xvii., 1890. 17. Die Caprelliden der " Siboga"-ExpeJition. Uitkomsten . . . H.M. "Siboga," Monogr. xxxiv. pp. 160, 10 pis., 1903. 18. Nebeski, 0. Beitriige zur Kenntnissder Amphipoden der Adria. Arb. zool. Inst. \Vien, iii. pp. 111-162, 4 pis., 1880. 19. Sars, G. 0. Histoire naturelle des Crustaces d'eau douce de Norvege. 1 Livr. Les Malacostraces, pp. iii + 145, 10 pis. 4to. Christiania, 1867. 20. — - An Account of the Crustacea of Norway : Vol. I. Ampliipoda. Pp. viii + 711, 248 pis. Christiania and Copenhagen, 1890-1895. 21. Stebbing, T. R. R. Eeport on the Amphipoda. Rep. Voy. "Challenger," Zool. xxix. 2 vols. and atlas, 1888. 22. - - Amphipoda: I. Gammaridea. Das Tierreich, Lief. xxi. pp. xxxix+ 806, 127 text-figs., 1906. CHAPTER XIV THE EUPHAUSIACEA Order Euphausiacea, Boas (1883). Definition. — Eucarida which retain the primitive caridoid facies ; the exopodite of the maxilla is small ; none of the thoracic limbs are specialised as maxillipeds ; branchiae in a single series, attached to the coxopodites of the thoracic limbs (podobranchiae) ; young hatched in the nauplius-stage. Historical. — The genus Thysanopoda wag placed by H. Milne- Edwards in his tribe of " Mysiens," and most subsequent writers have retained the association of Euphausiacea and Mysidacea in the group " Schizopoda." As already mentioned, Boas, was the first to separate the two orders. Some of the later larval stages were described as distinct genera by Dana before their true nature was pointed out by Glaus. MetschnikofFs discovery of the nauplius- stage of Euphausia (1869) drew special attention to the larval his- tory of the group. The most complete account of the structure, development, and classification of the Order is that given by G. 0. Sars in his " Challenger " Report. The phosphorescence of certain Euphausians was first observed by J. V. Thompson ; the light-pro- ducing organs have been investigated by several zoologists, the most detailed account being that given by Chun. MORPHOLOGY. The general form of the body (Fig. 139) is completely caridoid. The carapace coalesces dorsally with all the thoracic somites, and is produced in front as a rostrum which is never of great length. The telson (Fig. 142, C) has a pair of large movably articulated spines (s) near the tip. Sars has shown that these are formed by enlargement of one out of several pairs of marginal setae which are present in the larvae, and it is therefore impossible to regard them as representing the caudal furca of the Leptostraca. Appendages. — The antennules are always biramous, and some- 244 THE E UP HA US I A CEA 245 times show sexual modifications in the male. The antennae have a protopodite of two segments, a large lamellar scale or exopodite, and an endopodite with three peduncular segments and a long FIG. 139. Mcganyctiphanes norvegiea, c5, x 2. iflagellum. The mandibles (Fig. 140, A) generally have a palp ; the incisor process is directly connected with the molar process, there being no lacinia mobilis or row of spines in the adult, A. Mouth-parts of tfeganyctiphanes norvegiea. A, mandible ; B, maxillula ; C, maxilla, e, exite of first segment of maxillula ; ex, exopodite of maxilla ; i, incisor process ; m, molar process ; p, palp. although some spines are present in the larval stages. The maxillulae (Fig. 140, B) have the usual two endites and a palp which is generally unsegmented, but has three segments in Bentheu- phausia and two in some larvae. An exite of the third segment (exopodite) is present in the larva, but is replaced in the adult by an exite (e] of the first segment, sometimes erroneously called 246 THE CRUSTACEA the exopodite. The maxillae (Fig. 140, C) generally consist of four segments ; the second and third have each two endites and the third a slightly developed flabelliform exite (exopodite) ; the fourth segment generally forms the palp, but in Bentheupliausia the palp consists of three segments. None of the thoracic limbs (Fig. 141) are specialised as maxilli- peds, the first pair being similar in general structure to the succeeding pairs. The coxopodite is distinct and of considerable size ; the main flexure of the limb is between the meropodite and carpopodite ; the terminal segment is blunt and is without a distinct claw, bearing, as a rule, only a group of setae and br.~ Fio. 141. Thoracic appendages of Meganyctiphanes norvegica : A, first ; B, second ; C, seventh ; D, eighth, br, branchia (epipodite) ; en, endopodite ; ex, exopodite ; p, luminous organ. spines ; the exopodite is flattened and fringed with long natatory setae. On the outer side of the coxopodite is attached an epipodite (br), which is small and simple in form on the first pair, but ramified in a complex fashion on the other pairs of thoracic limbs. These branchiae, which are apparently quite homologous with the podo- branchiae of Decapoda, are not covered by the carapace, but hang out freely at the sides of the body. The last pair or the last two pairs of thoracic appendages are commonly much reduced or quite vestigial, though the branchiae connected with them are of large size (Fig. 141, D). The coxopodite of the first pair is usually slightly produced inwards, but only in Bentheupliausia does this so-called "masticatory lobe" differ very perceptibly from that of the succeeding pairs. A more important exception to the uniformity of the thoracic limbs is found in the genera forming the sub-family Nematoscelinae. In these, one of the pairs, the second or the third, has the endopodite greatly enlarged and modified as a raptorial weapon. In Nematoscelis, the THE EUPHAUSIACEA second pair is modified and the terminal segments bear a group of long serrated " harpoon-like " spines. In Nematobrachion, the details of the raptorial limbs are very similar to those of Nematoscelis, but it is the third pair that is modified. In Stylocheiran, the enlarged third pair has a prehensile form, the terminal segment being, in most cases, opposed to a strong spine or a group of spines on the penultimate segment. In S. abbreviation, G. 0. Sars ( = S. chelifer, Chun), the spines are replaced by a process of the segment, form- ing a perfect chela. The pleopods (Fig. 142, A) are strongly developed in both sexes ; the rami are flattened and fringed with long setae ; the endopodite has an appendix interna (z) on its inner edge. The A. FIG. 142. Abdominal appendages of Meganyctiphanes norvegica (the marginal setae of the rami are omitted). A, pleopod of third pair. B, endopodite of first pair of male, with the copulatory apparatus unfolded. C, telson and uropod. en, endopodite ; ex, exopodite ; ?', appendix interna ; s, movable sub-terminal spines of telson ; t, telson. first and second pairs are modified as copulatory organs in the male sex, the endopodite having on its inner edge a secondary lobe, which in the first pair (Fig. 142, B) is very large and armed with a complex apparatus of hooks and spines. The uropods (Fig. 142, C) have the exopodite divided by a suture only in I>entheuphaitsia. Internal Anatomy. — The details of the internal structure of Euphausiacea are still very imperfectly known. The hepatic caeca appear to resemble those of Decapoda, being generally much rami- fied and forming a large mass lying on each side of the thoracic region. The heart is short, polygonal in outline, with three pairs of ostia. The disposition of the main arterial trunks appears to be similar to that in the lower Decapoda. An antennal gland is present, opening on the proximal segment of the antennal peduncle. In the thoracic portion of the ventral nerve-chain the full number 248 THE CRUSTACEA Pg- of eleven pairs of ganglia can be distinguished, all of them, however, united in a common sheath of connective tissue. The paired eyes are each divided into two parts in the sub-family Nematoscelinae, the fronto-dorsal division having elongated ommatidia of a structure supposed to be adapted for perception of very faint light. The nauplius-eye persists in the adult, lying between the paired eyes beneath the rostrum. No statocysts have been found in any Euphausiacea. A system of very remarkable organs, formerly regarded as " accessory eyes " but now known to be lumin- ous organs (photospheres), is found in all Euphausiacea except Benth- euphausia. As a rule, two pairs of these organs are situated in the coxopodites of the second and seventh pairs of thoracic append- ages respectively (Fig. 141, p), an unpaired series on the sternal sur- face of the abdomen between the bases of each of the first four pairs of pleopods, and a pair, differing in structure from the others, on the upper surfaces of the ocular peduncles. In Stylocheiron, only the posterior thoracic pair, the first fameiue ramranddiig the lens'; 'n, nTrve ; abdominal organ, and those on the r> reflector ; sfr' striated body' ocular peduncles are present. Each of the thoracic and abdominal photo- spheres (Fig. 143) is globular in form, lying beneath and detached from the hypodermis, and, in some cases at least, moved by special muscles. The centre of the organ is occupied by a "striated body " (sir) composed of radiating lamellae, which seems to be the actual source of light, and in which the fibrils of a special nerve appear to terminate. On the inner side of the photosphere is a concave reflector (r) composed of concentric lamellae, and on the outer side is a lens (/), the whole surrounded, except on the outer side, by a sheath of pigment (pg). The organs on the ocular peduncles differ in the absence of a lens, in their incomplete separa- tion from the hypodermis, and in other details of structure. The spermatozoa are simple round or oval nucleated cells. They are transferred to the female in lageniform spermatophores, which are formed within the widened terminal parts of the vasa deferentia. The ova are small in size and are sometimes carried, probably only for a short time, " loose among the thoracic legs, which, with their setae, form a sort of basket " (Holt and Tattersall, Euphausia}. Flo. 143. Section through one of the thoracic phosphorescent organs of Nematoscelis. THE EUPHAUSIACEA 249 In other cases the ova are agglutinated together in a paired or unpaired mass, which is carried attached to the sternal surface of the thorax, close to the bases of the posterior thoracic legs. These egg -packets are formed by some cementing material apparently extruded along with the eggs, and bear no sort of morphological resemblance to the brood-sac of the Mysidacea Avhich is formed by the oostegites. DEVELOPMENT. With the possible exception of the genus Stylocheiron, where the relatively large size of the eggs suggests an abbreviated meta- morphosis, the Euphausiacea leave the egg in the form of a typical nauplius, and reach the adult stage by a series of changes closely parallel to those occurring in the metamorphosis of the Penaeidea among the Decapoda. Glaus was the first to show that the forms described by Dana as distinct genera under the names Calyptopis, Furcilia, and Cyrtopia are successive stages in the development of Euphausiacea, and Sars has been able to fill in, with considerable detail for several species, the outline thus furnished. The newly hatched nauplius, as described by Metschnikoft" and by Sars, has an oval unsegmented body, without shell-fold, and with the three pairs of nauplius appendages in their typical form, the first uniramous, the second and third biramous ; the median eye is not developed till the next stage. The mouth is not at first open, and there is no masticatory process on the base of the antenna such as exists in Copepoda and Cirripedia. In the next stage the larva assumes the form of a metanauplius (Fig. 144, A), showing rudiments of three additional pairs of ap- pendages, maxillulae, maxillae, and first thoracic limbs ; the shell- fold is defined posteriorly and partly envelops the hinder end of the body. In later metanauplius-stages the third pair of appendages (mandibles) lose their natatory character and become reduced to the basal masticatory part, with only a very small rudiment to represent the palp ; the shell-fold now projects in front as well as behind and overhangs the head in a hood-like form. The metanauplius is succeeded by a series of Calyptopis- etages (Fig. 144, B, C) characterised by the elongation of the trunk- region and the differentiation of its somites in regular order from before backwards. The thoracic somites are very short and crowded together. The appendages already present become more fully developed, and, later, the rudiments of the last pair of abdominal appendages appear. The paired eyes develop but are covered at this stage by the frontal hood of the carapace. In the next succeeding Furcilia-stages the paired eyes become free and movable. The pleopods develop from before backwards and, later, the anterior thoracic limbs. The Cyrtopia -stage is 250 THE CRUSTACEA characterised chiefly by changes in the antennules and antennae. The former, which have become biramous already in the Calyptopis- stage, now elongate, while the latter lose their natatory function, and their rami, at first similar, become differentiated into " scale " and flagellum respectively. The Cyrtopia gradually assumes the form Larval stages of Euphausia. A, metanauplius. B, calyptopis. C, later calyptopis-stage. 1, antennule ; 2, antenna ; 3, mandible ; 4, maxillula ; 5, maxilla ; I, first thoracic appendage ; th, thoracic somites ; rili, abdomen ; (cijXfts), first five abdominal somites ; ««, uropod ; an, anus ; f.s, frontal sense-organs. (A after MetschnikofT; B and C after Claus ; from Korschelt and Heider's Embryology.) of the adult by the successive development of the remaining thoracic limbs. The development of the various appendages, which has been traced in detail by Sars, offers many features of interest. The palp of the mandible only begins to redevelop in the Cyrtopia- stage. The maxillula has at first a small lobe on its outer edge which appears to represent the exopod and is quite independent of the THE E L 'PHA US I A CEA 2 5 1 plate which in the adult has been regarded as the exopod, but which is in reality a process of the first segment of the limb. A very remarkable feature in the development of the paired eyes has been found by Sars in the case of Neinatoscelis. Here, a transitory larval eye consisting of seven ommatidia is formed at the apex of the eye- stalk and is later pushed to the outer side by the development of the permanent visual organ. Though this transitory organ is stated to disappear in the adult, it seems possible that it may really persist as the luminous organ of the eye-stalk, especially since in other genera (Nyctiphanes, Ewphautia) the luminous organ appears before the eye itself. The heart is visible already in the later metanauplius- stases. The liver at first consists of three caecal tubes on each side. KEMARKS ON HABITS, ETC. The Euphausiacea are characteristically pelagic animals, forming part of the surface-plankton of the ocean and descending to con- siderable depths. They are generally remarkably transparent, but Bentheuphausia, which appears to be a true deep-sea form (1000- 1800 fathoms), is said to be "quite opaque and of a similar vivid red colour to that of most other true deep-sea crustaceans." The adult size of most Euphausiacea lies between 10 and 40 mm.; a species of Thysanopoda reaches 55 mm. in length. No fossil Euphausiacea are known. AFFINITIES AND CLASSIFICATION. The differences in structure which justify the separation of the Euphausiacea from the Mysirlacea have already been insisted on. A certain degree of resemblance in general facies is sufficiently accounted for, on the one hand, by the approximation of the basal members of the Peracaridan and Eucaridan lines of descent to the common caridoid stock of the Eumalacostraca ; and on the other, by the similarity in habitat between the Euphausiacea and many of Mysidacea. The resemblances between the members of the present Order and some of the lower Decapods, especially the Penaeidea, are of much greater importance. The complex copulatory armature of the first pleopods has a general resemblance to that of the Penaeidea, the larval development of the two groups is closely parallel, and the presence in some Sergestidae of phosphorescent organs resembling, though differing in details from, those of the Euphausiacea may also be an indication of affinity. At present, the Euphausiacea are regarded as constituting only a single family. THE CRUSTACEA ORDER Euphausiacea, Boas (1883). Family EUPHAUSIIDAE. Sub-Family EUPHAUSIINAE. Euphausia, Dana ; Thysanopoda, Milne-Edwards ; Nyctiphanes, G. O. Sars ; Meganycti- plianes, Holt and Tattcrsall (Fig. 139). Sub-Family NEMATOSCELINAE. Nematoscelis, G. O. Sars ; Nematobrachion, Caiman ; Stylocheiron, G. O. Sars. Sub-Family BENTHEUPHAUSIINAE. Bentheupkctusia, G. O. Sars. LITERATURE. 1. Chun, C. Atlantis. Biologische Studien iiber pelagische Organismen. 5. Ueber pelagische Tiefsee-Schizopoden. Bibliotheca Zoologica, Bd. vii. (Heft 19), pp. 137-190, pis. viii.-xv. 6. Leuchtorgane und Facettenaugen, torn. cit. pp. 191-262, pis. xvi.-xx., 1896. 2. Clems, C. Uebev einige Schizopoden und niedere Malacostraken Messina's. Zeitschr. wiss. Zool. xiii. pp. 422-454, pis. xxv.-xxix., 1863. 3. Hansen, H. J. Preliminary Report on the Schizopoda. . . . Bull. Mus. Oceanogr. Monaco, N"o. 30, pp. 32 ; and Further Notes on the Schizopoda, op. cit. No. 42, pp. 32, 1905. 4. Holt, E. W. L., and Tattersall, W. M. Schizopodous Crustacea from the North-East Atlantic Slope. Rep. Fisheries, Ireland, 1902-1903, pt. ii. App. iv. pp. 99-152, pis. xv.-xxv., 1905 ; and Supplement, Fisheries, Ireland, Sci. Invest., 1904, v. 50 pp. 5 pis., 1906. 5. Metschnikoff, E. Ueber ein Larvenstadium von Eupliausia. Zeitschr. wiss. Zool. xix. pp. 479-482, pi. xxxvi., 1869. 6. - Ueber den Naupliuszustand von Euphausia. Op. cit. xxi. pp.-397-401, pi. xxxiv., 1871. 7. Sars, G. 0. Report on the Schizopoda. Rep. Voy. "Challenger," Zool. xiii. pp. 228, 38 pis., 1885. CHAPTER XV THE DECAPODA Order Decapoda, Latreille (1802). Sub-Order 1. Natantia. Tribe 1. PENAEIDEA. „ 2. CARIDEA. „ 3. STENOPIDEA. Sub-Order 2. Reptantia. Section 1. PALINURA. Tribe 1. ERYONIDEA. „ 2. SCYLLARIDEA. Section 2. ASTACURA. Tribe NEPHROPSIDEA. Section 3. ANOMURA. Tribe 1. GALATHEIDEA. „ 2. THALASSINIDEA. „ 3. PAGURIDEA. „ 4. HIPPIDEA. Section 4. BRACHYURA. Tribe 1. DROMIACEA. Sub-Tribe 1. DROMIIDEA. „ 2. HOMOLIDEA. Tribe 2. OXYSTOMATA. „ 3. BRACHYGNATHA. Sub-Tribe 1. BRACHYRHYNCHA. „ 2. OXYRHYNCHA. Definition. — Eucarida in which the caridoid facies may be retained or may be greatly modified ; the exopodite of the maxilla is very large (scaphognathite) ; the first three pairs of thoracic limbs are specialised as maxillipeds ; branchiae typically in several series, attached to the coxopodites of the thoracic limbs (podo- branchiae), to the articular membranes (arthrobranchiae), and to the 253 254 THE CRUSTACEA lateral walls of the thoracic somites (pleurobranchiae), very rarely absent ; young rarely hatched in nauplius-stage. Historical. — The great majority of the larger and more familiar Crustacea belong to the Decapoda, and this Order received far more attention from the older naturalists than any of the others. A considerable number of species are mentioned by Aristotle, who describes various points of their anatumy and habits with accuracy, and sometimes with surprising detail. A long series of purely descriptive writers who have added to the number of known forms without contributing much to a scientific knowledge of them begins with Belon and Rondelet in the sixteenth century, and perhaps does not altogether come to an end with Herbst's Naturgescliichte der Krabben und Krebse (1782-1804). Among the most noteworthy of early contributions to anatomy are Swammerdam's memoir on the Hermit-Crab (1737), and that of Roesel von Rosenhof on the Crayfish (1755). Reaumur's observations on the phenomena of ecdysis and the regeneration of lost parts in the Crayfish (1712- 1718) have become classical. The foundations of classification were laid by J. C. Fabricius (1793), who divided the Linnean genus Cancer into a large number of genera, the majority of which are still recognised. Latreille, to whom the name of the Order is due (1802), also began its subdivision into sub-orders and families. In this more than in any other group of Crustacea the works of H. Milne- Edwards, and especially his Histoire Naturelle des Crustaces, may be taken as marking the beginning of the modern period, and his classification of the Decapoda has been that most generally accepted until very recently. Almost contemporaneous with Milne-Edwards's great work, and often surpassing it in morphological detail and systematic insight, was de Haan's volume on the Crustacea of Japan (1833-1849). The first important departure from the general plan of classification laid down by these authors was made by Boas in 1880, and his system has been further elaborated by Ortmann and by Borradaile. J. Vaughan Thompson's discovery of the larval metamorphosis of Decapoda (1828-1831), confirming the earlier observations of Slabber and Cavolini in the eighteenth century, gave rise to a curious controversy in which Westwood and others denied the possibility of such a metamorphosis, basing their argu- ments chiefly on Rathke's memoir on the development of the Crayfish (1829). F. Miiller in 1863 made the highly important discovery that Penaeus is hatched from the egg in the form of a nauplius, and the clue thus given to the interpretation of the other larval stages was followed up especially by Glaus. The development of deep-sea exploration within the last thirty years has resulted in the discovery of a large number of important new types of Decapoda, which have been described by Spence Bate, Miers, Henderson, A. Milne-Edwards, Bouvier, Faxon, Alcock, and others. The numerous THE DEC APOD A 255 species of fossil Decapoda have been little studied from the point of view of phylogeny, but reference may be made to Bouvier's essay on the origin of the Brachyura as an example of the results which may be obtained in this department. Among other papers which have been fruitful in suggesting lines of research for later workers may be mentioned Huxley's memoir on the classification and distri- bution of the Crayfishes (1878); A. Milne-Edwards's note on the transformation of the ocular peduncle into an antenna-like organ in a Palinurid (1864), the forerunner of much recent work on regenera- tion and abnormalities ; Giard's papers on parasitic castration ; and Faxon's discovery of the alternating dimorphism in the males of Cambarus. MORPHOLOGY. Amid the great diversity of general shape exhibited by the Decapoda, two chief types may be distinguished. In the first or Macrurous type the general caridoid facies is retained, the body is elongated and subcylindrical, the abdomen is long and terminates in a tail-fan. In the Brachyurous type (which is not confined to the Brachyura, but recurs in several groups of Anomura) the cephalo- thorax is greatly expanded laterally and more or less depressed, while the abdomen is reduced and folded underneath the cephalo- thorax. A very peculiar modification is found in most Paguridea, where the abdomen is markedly asymmetrical and spirally coiled, in correlation with the habit of living in the empty shells of Gasteropod Molluscs. The carapace coalesces dorsally with all the thoracic somites and overhangs on each side as a branchiostegite, enclosing the branchial chamber within which the gills are concealed. Anteriorly it may be produced into a rostrum, which in a few genera of Caridea (Rhynchodnetes, etc.) is movably articulated. In most Brachyura the rostrum is reduced to a short but broad frontal plate, of which the relations to the adjacent parts will be described below. In some Macrura (Scyllaridae) and in many Anomura and Brachyura, where the cephalothorax is flattened from above downwards, the lateral portions of the carapace are abruptly bent inwards towards the bases of the legs. The lateral margin thus produced is commonly toothed or otherwise armed. The surface of the carapace is commonly marked by depressions and grooves corresponding in part to the insertions of various muscles, but in part independent of these. In this way several regions of the carapace are defined which, especially in the Brachyura, may be still further divided into sub-regions. For convenience of systematic description these various areas are denominated ac- cording to a scheme of terminology introduced for the most part by H. Milne-Edwards (1851). More recently the furrows of the 256 THE CRUSTACEA FIG. 145. Carapace of the Norway Lobster (Xephrops nor- vegicus) from the side. (After Boas.) The letters refer- ring to the grooves of the carapace are those used by Boas, e, e,', the "cervical groove " of Bouvier, "anterior cervical groove " of Borradaile ; c, " branchial groove " of Bouvier, " posterior cervical groove " of Borradaile. carapace have been studied by Boas and by Bouvier. Only a few points can be mentioned here. In the lobsters and crayfish a con- spicuous groove (Fig. 145, c) crosses the dorsal surface of the carapace transversely about the middle of its length and curves forwards on either side. This groove, named the " cervical groove " by Milne - Edwards (c in Boas's terminology, the " branchial groove " of Bou- vier), was supposed by him to indicate the line of division between the por- tions of the carapace aris- ing from the antennal and mandibular somites respectively. Other writers, for instance, Huxley, regarded it as marking the limits of the cephalic and thoracic regions. There appears to be no ground, however, for regarding this groove as of greater importance than some of the other grooves of the carapace. In some cases an equally con- spicuous transverse groove (e of Boas, " cervical groove " of Bouvier) (Fig. 145, e, e) crosses the carapace a little in front of the cervical, and as this is the only transverse groove, apparently, to be found in any of the lower Macrura (Stenopidea, Caridea), it seems at least as likely to e c I a afford an important morpho- Flo 146 logical landmark. In SOme Carapace of CalUanassa novaebntanniae (Tlia- r>a«pe nnrtirmo rvf tVia paranar-a lassinidea) from the left side. (After Borradaile. > cases portions tne carapace c, e> the gr'ooves so lettered by Boas (see Fig 144) . may be separated by a longi- I.a,Zlneaonom«ptca(perhap8jdsotlieZi««*wii*i- ,. , , . J3 . dim), the front part of which is the line & of tudinal groove Or Uncalcihed Boas; l.t, Unm thalassinwa (perhaps also the linp whirh mav PVPD form a z"ie« ^om"'ica). *he front part of which is the line line, wni may ever >rm a (? Of Boas ; r, rostrum. movable hinge. Of this nature are the linea thalassinica (Fig. 146, l.t) of the Thalassinidea, with which the I. homolica of the Homolidae may perhaps be identical, and the 1. anomurica (l.a) of many Anomura, identified with the I. dromiidica of Dromiidae and the unfortunately named " epimeral suture" of other Brachyura. The sternal surface of the cephalothorax is very narrow in many Macrura, but is often broad in those which have a depressed form. It is broad in many Anomura and in all Brachyura, with the exception of Raninidae. The thoracic sterna are usually clearly distinguishable, and, in the lower forms, seem to preserve a certain- THE DECAPODA 257 degree of mobility. In the higher forms they become firmly united, with the exception of the last thoracic sternum, which may be movable (Astacidae, Parastacidae, and Anomura). In front of the mouth, regions representing the sterna of the three preoral " somites " can be distinguished, but on account of the " cephalic flexure " the ophthalmic and antennular sterna are directed forwards, or even upwards. The antennal sternum is mainly represented by the epistome, a plate of varying shape lying between the labrum and the bases of the antennae. In the Natantia the epistome is comparatively narrow, and on each side is separated from the lateral portions of the carapace by the exhalent branchial channels. In most of the Reptantia the epistome (Fig. 147, A, ep) is broad and comes in contact with the carapace on each side, and in the Brachyura it becomes firmly united with it. In this way there is defined more or less distinctly a buccal frame within which lie the mouth-parts, and which in most Brachyura is closed by the operculiform third maxillipeds. The sides of this buccal frame are formed by the free antero -lateral margins of the carapace (Fig. 147, B, l.m), while in front it is more or less distinctly delimited by the epistome itself, or by a transverse ridge (Fig. 147, B, a.m) which divides the epistome into two parts, the epistome proper and the endostome or palate (end). In most Brachyura also (except the Dromiacea) the proximal segments of the antennae are fused with the epistome. In the Macrura the anterior margin of the carapace forms on each side of the base of the rostrum a more or less distinct " orbital notch," within which the eye rests when it is turned outwards. In the Brachyura this transverse direction of the eye-stalks is permanent, and the orbit' is usually (except in Dromiacae) completed by the downgrowth of a process (l.p) from the front, external to the antennules, which unites either directly or, more usually, by intervention of the second segment of the antenna, with the sub-orbital lobe (s.o) of the cara- pace. Further, in all .the Brachyura the rostrum or frontal plate sends downwards in the middle line a process (m.p) which unites .in front of the ophthalmic and antennular sterna with the epistome, and separates from one another the basal segments of the antennules. The greater part of the ophthalmic peduncle is in this way con- cealed in a kind of sheath, and only the terminal segment appears and is movable within the orbit. In the Dromiacea the second segment of the antennal peduncle is free and there is no corresponding process of the front, so that the orbits are incompletely or not at all defined. The arrange- ment is hardly more complete in certain Oxyrhyncha (Macrocheira) (Fig. 147, B), but in most Brachyura the antennal peduncle joins with the front to form a partition separating completely the orbits from the "antennular fossae," into which the antennules 17 258 THE CRUSTACEA may be withdrawn. There is, however, great diversity in the de- tails of structure of the " facial " region among the Brachyura, and these are of considerable value as systematic characters. In the Scyllaridae among the Macrurous groups the cephalic FIG. 147. Head and anterior part of body from below. A, Xephrops iwrvegicus. B, Macrocheira Kaempferi. C, Carpilius convextis. (Drawn by Miss G. M. Woodward.) «', antennnle ; a", antenna ; a.m, ridge forming anterior margin of the mouth-frame and dividing the epistomial area into epistome proper and endostome ; c, point where the lateral wing of the carapace touches, or, in B and C, fuses with, the epistome ; e, eye, in C retracted into, and partly concealed by, the orbit ; end, endostome ; ep, epistome ; ex, exopodite ; l.m, lateral margin of buccal frame ; l.p, lateral process of rostral plate, which in C comes in contact with the basal segment (2+3) of the antenna ; md, mandible ; mp, median process of the front (in B and C) uniting with anterior process of epistome ; mxp3, third maxilliped ; r, rostrum or (in C) frontal plate ; s.o, suborbital lobe forming floor of orbit ; t, in A, tubercle bearing opening of antennal gland, in B and C, operculum covering the opening and probably representing the reduced first segment of the antenna ; 1-5, the segments of the antennal peduncle. region is modified in a way that at first sight suggests the Brachy- urous type, the eyes being widely separated and lodged in complete orbits. In this case, however, the front unites in the middle line not with the epistome but with the greatly enlarged antennular somite. In the Alpheidae (Caridea) the anterior margin of the carapace THE DECAPOD A 259 is modified in a very peculiar manner, growing over and, in most cases, completely enclosing the eyes. The differences in the development of the abdominal region are no less conspicuous than in the case of the cephalothorax, and have been utilised as affording characters for the primary sub- divisions of the Order. In the Natantia the abdomen is large and, with its appendages, forms the chief organ of swimming. It is generally more or less compressed and its somites have well- developed pleura. It is dorsally humped or bent between the third and fourth somites in many Caridea (Eukyphotes, Boas) (Fig. 148), but the character is not so constant as to justify great Fio. 148. Hetei'ocarpus Alphonsi (Caridea, Pandalidae), showing the "humped" form of the abdomen and the multiarticulate meropodite and carpopodite of the second leg. (From Alcock, Xiititmlixt in Indian Seas.) systematic importance being attached to it. In the Palinura and Astacura the importance of the abdominal appendages as natatory organs is generally reduced, and the abdomen itself is not humped. Among the Anomura, the Thalassinidea (Fig. 149) have retained the extended abdomen of the Macrurous groups, but the pleura are more or less reduced; the Galatheidea (Fig. 150) have the abdomen more or less closely flexed under the cephalothorax but not greatly modified; the Paguridea, with the exception of some interesting transitional forms (Pylochelidae) (Fig. 151), have the abdomen and its appendages more or less unsymmetrically developed and its somites imperfectly indicated. In the hermit-crabs (Pagu- ridae and Coenobitidae) the abdomen is soft-skinned and spirally 26o THE CRUSTACEA FIG. 149. Icowaxiopsis andamanensis (Thalassinidea). (From Alcock, Naturalist in Indian Seas.) FIG. 150. Munida andamunica (Galatheidae), abdomen extended. (From Alcock, Naturalist in Indian Seas.) 26l coiled to fit into the Gasteropod shells inhabited by the animals. Only the sixth somite and the telson are fully calcified, the tergal portions of the other somites being merely indicated by widely Fro. 151. ' Pylocheles Mieruli (Paguridea). «, en. A, first, B, third maxilliped of Keptiinuspelagii-us (Brachyura). (After de Haan.) i . rndopodite ; ej>, epipodite (mastigobranchia) ; ex, exopodite. culum to the buccal frame and entirely lose their pediform char- acter. The ischiopodite and meropodite become broad plates and the terminal three segments are often hidden behind the meropodite. The peduncle of the exopodite may also be expanded and share in forming the operculum. Its terminal flagellum is either folded out of sight or may be entirely lost. The epipodite forms a long curved blade in most Brachyura (Fig. 159, B, ep). The remaining five pairs of thoracic appendages (peraeopods) are typically ambulatory legs, composed of the usual seven segments. Exopodites may be present on some or all of them in some Penaeidea and Caridea (Pasiphaeidae, Fig. 160, Hoplophoridae, some Atyidae, and Crangonidae), but elsewhere they are wanting. As a rule one or more pairs are chelate or sub-chelate, except in the Scyllaridea (where, however, the last pair are imperfectly chelate in the female sex) and in some Hippidea. The first three pairs are chelate in 270 THE CRUSTACEA most Penaeidea and in the Stenopidea and Astacura (Fig. 161), the first four or all five pairs in the Eryonidea (Fig. 162), the first two FIG. 160. Psathyrocaris fragilis (Pasiphaeidae), showing the greatly developed exopodites of the thoracic legs (a " Schizopod " character) and of the pleopods. (After Alcock. Fio. 161. Nephropsis Carpenteri (Nephropsidae), £ . A deep-sea species in which the eyes have almost disappeared but the vestigial eye -stalk can be seen below the rostrum. i(From Alcock, Naturalist in Itidian Seas.) in most Caridea, the first or the first two in Thalassinidea (Fig. 149, p. 260), and the first pair only in the other Anomura (Fig. 150, THE DECAPODA 271 p. 260) and the Brachyura. In most Anomura the last pair, and in a few Brachyura the last or the last two pairs, are subchelate. A very remarkable form of chela is found in the genus Psalidopus (Caridea) (Fig. 163), in which both fingers are movably articul- ated with the propodite, an arrangement resembling that found in the second maxilliped of Stylodadylus. In most of the Reptantia, where the first pair of legs are chelate and much larger than the others, they are commonly referred to as the chelipeds, and the following four pairs are distinguished as walking -legs. Frequently the chelipeds are asymmetrical in size and shape on the two sides, the larger chela having the fingers armed with blunt crushing -tubercles, while the smaller has sharp cutting -teeth. In many cases, as, for instance, in the lobster, the larger crushing- chela may be on the right or the left side indifferently, but in some Bra- chyura it is constantly on the same side of the body. A curious reversal of asymmetry sometimes occurs as a result of the loss of the larger chela ; at the next ecdysis the remaining chela assumes more or less com- pletely the characters of a large crushing-chela, while the re- generating limb has the form of a small cutting-chela. A modification of SOme Of . /'<•»'"'•>"<(« •"'"'-iooi- 'qOUBjqOJlUJV (N!MOS— II<1C^!MI^I^IM '" O3 u •qoiiBjqopoj g'Cog'g'ogooggoo£c££o=o o £goooo ^ •HDUBjqopoj gCoggggog"g'goo££££o=o lt:.•< /«.>//.-•!'/•«".*».•.• (Mmi-dlagnmnmtic). cm-/i, cardiac chamber ; d, lateral denticles ; I,, terminal lappets projecting into mid-gut ; l.r, lateral ridge ; m.r, median ventral ridge of cardiac chamber: m.t, im-dian tooth ; put, pyloric chamber; r, longitudinal ridge separating dorsal and ventral divisions of pyloric chamber ; »-, wedge- slwped ridge with straining apparatus. (After Bonnier.)Jj of which the former is a direct continuation of the cardiac chamber, while the latter is a diverticulum closed in front and communi- cating with the dorsal chamber only by a narrow slit between the setose margins of the longitudinal ridges. The ventral chamber has its cavity again divided into two portions by a strong median wedge-shaped ridge (w) which rises from its floor and is produced posteriorly into a tongue-like process overhanging the apertures of the hepatic ducts into the mid-gut. The lateral surfaces of this ridge are provided with a characteristic armature which seems to act as a straining apparatus ; it consists of numerous parallel plate- like ridges running longitudinally and standing at right angles to the surface, each bearing on its edge a comb-like fringe of setae lying parallel with the surface and covering in the groove lying between each ridge and the next. This wedge-shaped ridge and 282 THE CRUSTACEA straining apparatus appear to be very constant throughout the group, and are no doubt homologous with the very similar structures found in other orders of Malacostraca. At its posterior end the chitinous lining of the stomodaeum terminates in four tongue-like lappets (/) (much elongated in Penaeus) which project freely into the cavity of the mid-gut. In the other Penaeidea the stomodaeal armature is much more complex than that just described. A large number of sclerites, more or less calcined, are differentiated in the walls of both cardiac and pyloric chambers. A dorsal and a ventral series can be dis- tinguished, the dorsal and dorso-lateral pieces of the cardiac chamber being in relation to the strong median and lateral teeth. From this arrangement of the parts found in the Penaeidea, that charac- teristic of the Reptantia may be easily derived, the chief differences being due to the appearance of additional sclerites, and especially of a series of intermediate pieces on the lateral walls of the two chambers. The elements of the dorsal series are the more important, forming as they do a system of levers moving the dorsal and lateral teeth. The great majority of the Caridea diverge more widely from the Penaeid type owing to the disappearance of the whole of the dorsal series of sclerites and of the dorsal and lateral teeth associated with them, the roof of at least the cardiac chamber remaining quite membranous. Only the Atyidae and one or two others among the families of Caridea hitherto examined possess certain elements of the dorsal series well developed, but they are differently arranged from those of the other Decapods. Mention must be made here of the gastroliths or " crab's eyes," which are discoidal calcareous nodules in the lateral walls of the cardiac division of the stomach in Crayfish (Astacus) and Lobsters (Hornarus). They are periodically formed shortly before ecdysis takes place, and are shed into the cavity of the stomach, to be broken up and dissolved, apparently providing some of the material necessary for the calcification of the new integument. No similar structures are definitely known to occur in any other group of Decapods. The mid-gut varies very much in length in different Decapoda, but exact observations have been made only on a few types. In the Crayfish (Astacus) it is exceedingly short, so that the dorsal lappet which terminates the cuticular lining of the stomodaeum extends through it into the beginning of the proctodaeum. In most, if not all, Brachyura it is also very short. In the Lobster (Homarus), however, it occupies five-sixths of the post-gastric part of the alimentary canal. In species of AlpJieus (Caridea) the mid- gut extends as far as the last somite, and in Paguristes it is longer than the proctodaeum. From the upper surface of the mid-gut THE DEC APOD A 283 there arises anteriorly in Astacura and most Thalassinidea a short unpaired caecum. In Callianassa among the Thalassinidea and in most Paguridea a pair of longer or shorter caeca are present, and in most Brachyura they form two long and convoluted tubules. In the Caridea and Galathaeidea and in Pagunstes (Paguridea) no caeca are found. Dromia possesses a single short caecum, and so resembles the Astacura and differs from the other Brachyura. An unpaired caecal tube of considerable length springs from the dorsal surface of the intestine in the Brachyura, and a shorter caecum is present in the Lobster, in the Thalassinidea, and in some Paguridae. It is probable that in all these cases the caecum arises from the posterior end of the mid-gut. In Alpheus, according to Coutiere, the mid-gut is produced backwards beyond its junction with the narrower hind-gut into a number. of blind saccules. Groups of gland-cells on the walls of the oesophagus, on upper and lower lips, and on the maxillulae and maxillae, have been regarded as salivary glands. Quite similar glands, however, may occur through- out the whole length of the hind-gut also, and they are identical in structure with the dermal glands which occur in various situations on the surface of the body. With the single exception of Leucifer, which possesses only two pairs of hepatic caeca, the voluminous "liver" of the Decapods consists of a mass of minutely ramified tubules, lying mainly in the thorax. It communicates with the anterior part of the mid-gut by, as a rule, a single duct on each side, but in Alpheus (Coutiere) three ducts are present. In Paguridae the hepatic glands are displaced backwards, and lie for the most part in the abdominal region. Circulatory Si/stem. — The heart in all Decapods is short, polygonal in outline, and situated under the posterior part of the carapace. As a rule there are three pairs of venous ostia, of which one, or in the Brachyura two pairs are situated on the upper surface. Coutiere has demonstrated the existence of two additional pairs in certain Caridea, and possibly further research will show that these are present in other cases. Anteriorly the heart gives off a median ophthalmic artery which runs forward to supply the region of the eyes. On each side of this originates an antennal artery, which, besides supplying the antennae, sends branches also to the rostrum, eyes, and adjacent parts. In Astacus, Bouvier finds that terminal branches of the antennal arteries unite in front of the brain in a median vessel which runs backwards to anastomose on the Avails of the oesophagus with branches of the sternal (subneural) artery — an arrangement which recalls the circumoesophageal vascular ring of some Isopoda and Amphipoda. A second pair — the hepatic arteries — arise from the sides of the heart a little way behind the antennal arteries, and are distributed to the hepatic glands and adjacent viscera. 284 THE CRUSTACEA Posteriorly the heart sends off a median vessel, the superior abdominal artery, while the unpaired descending artery (some- times called the sternal artery) may arise separately from the heart (Brachyura) or may branch off from the superior abdominal artery just beyond the valves which mark its origin from the heart. The descending artery passes on one side (either to right or left) of the intestine and pierces the ventral nerve-chain in nearly all Decapods, passing between the connectives uniting the sixth and seventh thoracic ganglia. Only in some of the Brachyura, where the concentration of the nervous system reaches its highest point (Oxyrhyncha and some Brachyrhyncha), this perforation of the nerve- mass does not take place, the artery passing behind instead of through it. On arriving at the ventral surface the artery bifurcates in the median plane, a large branch, to which the name of sternal artery is commonly applied, running forwards to supply the ventral surface of the thorax and its appendages, while a smaller branch running backwards also beneath the nerve -chain is the inferior abdominal artery (absent only in Paguridea). These two arteries taken together form a median longitudinal trunk quite comparable to the subneural vessel of Isopods, and, like it, may communicate with the dorsal system of vessels by a circumoesophageal ring. A further communication is very often present at the posterior end of the abdomen, where a vascular ring encircling the intestine unites the superior and inferior abdominal arteries. A pair of posterior lateral arteries arising from the superior abdominal artery near its origin from the heart, and often unsymmetrically developed, are of importance since they irrigate the branchiostegal regions of the carapace which have a respiratory function. A venous sinus in the mid-ventral line receives the blood from the lacunar system of the body and appendages and distributes it to the gills, whence it is returned to the pericardial sinus by branchio-pericardial channels running in the inner wall of the branchial cavity. A minor circuit for the blood is afforded by the lacunar network of the branchiostegites, which, receiving blood partly from arteries and partly from adjacent venous sinuses, return it directly to the pericardium by special channels. In terrestrial Decapods various modifications of the respiratory and circulatory systems are met with. In those most completely adapted to a terrestrial life (Birgus, Cardisoma) the lining membrane of the branchial cavity is very vascular and covered with minute villi. The supply of venous blood to the sinuses of the branchio- stegal regions is more important and more definite than in aquatic Decapods, and the apparatus no doubt functions as a lung. In the terrestrial Hermit-crabs (Coenobita) a very peculiar respiratory organ is found. A rich vascular network is developed THE DEC APOD A 285 in the delicate skin of the abdomen, especially on the dorsal side anteriorly. Two pairs of venous trunks running along the sides of the abdomen return the blood to the pericardium, a pair of rhythmically contractile vesicles at the base of the abdomen serving to accelerate the flow. Excretory System. — In all Decapods the antennal gland is well developed, and generally presents a complexity of structure not found elsewhere among Crustacea. It has in most cases lost its original tubular form and assumed that of a compact gland. Three divisions are commonly distinct — the saccule, the labyrinth, and the bladder, with its efferent duct leading to the exterior. The saccule, which represents the end- sac of the typical antennal gland, may retain its simple saccular form, but more commonly it is complicated either by the development of partitions dividing up its cavity, or by numerous branches which ramify through the mass of the labyrinth. The labyrinth may be considered as derived from a sac which, by the rich development of partitions and trabeculae from its walls, has been converted into a spongy mass traversed by a complex system of canals. In the Crayfish (Astacus) the structure is still further complicated, mainly by the elongation of a portion of the labyrinth into a whitish cord of spongy substance which is convoluted upon itself, forming the "medullary" portion of the gland, the greenish "cortical" layer representing the proximal portion of the labyrinth which com- municates with the end-sac. The bladder may retain, as in the Crayfish, the form of a simple vesicle communicating with the exterior by a short duct. In many cases, however, it sends oft* prolongations which may extend through a great part of the body. In some Caridea this vesical system is very extensive, lobes from the two sides uniting with each other to form an unpaired vesicle above the stomach. In the Brachyura three main lobes are given off from the bladder, which are very constant throughout the group, such differences as do occur being correlated with the differences in shape of the carapace. In the Paguridae, however, the vesical system reaches its greatest complexity (Fig. 167). The bladder sends off prolongations which ramify between the organs and anastomose to form delicate networks and arborisations in the region of the thorax, and two long diverticula, which may unite with each other, pass backwards to traverse the whole length of the abdomen. In Palinurus an accessory gland not found in any other type opens into the duct of the bladder. The external aperture is in most cases placed on a papilliform elevation on the proximal segment of the antennal peduncle. In the Brachyura the aperture is covered by an operculum (Fig. 147, B and C, t), capable of being opened and closed by special muscles. It has been shown that this operculum in all probability represents the 286 THE CRUSTACEA reduced proximal segment of the antenna, and the muscles attached to it have been identified with those which move the proximal antennal segment in the lower Decapods. This structure was described by Audouin and Milne-Edwards as a kind of auditory ossicle. No trace of the maxillary gland is known to persist in any adult Decapod, though it is frequently well developed in the larval stages. Traces of glandular organs, presumed to be homologous with the antennal and maxillary glands, have been observed in embryonic stages in certain other somites of the trunk. In addition, certain other structures are found in adult Decapods, the excretory functions of which have been demonstrated physiologically, although their morphological significance remains obscure. The most important of these are the " branchial glands," which are masses of connective-tissue cells surround- ing the venous channels in the axis of the gills and are devoid of ducts. Other glands of the dermal type also occur in connection with the gills. Nervous System. — Great differences exist in the number and disposition of the ganglia composing the ventral chain. Arnnnty tVip Inwpr F)pr>nr>nrlo tVm oiV rrnnrrliu Among tne lower jjecapoas tne six ganglia corresponding to the six abdominal somites are distinct, but those of the cephalothorax may be more or less coalesced. The largest number of distinct ganglia appears to be found in the Astacura, where those of the five posterior thoracic somites are well separated, the remaining cephalotho- racic ganglia being more or less completely coalesced to form a large suboesophageal ganglion. In other cases coalescence has taken place to a greater extent, and in the Scyllaridea and some Caridea, at least, all the cephalothoracic sternal ganglia form a single mass. Among the Anomura the degree of coalescence varies, and sometimes the first abdominal ganglion is approximated to the thoracic mass. Among the Brachyura the concentration of the nervous system reaches its highest point ; the whole of the sternal ganglia are united into a rounded mass lodged in the thorax, from which the nerves radiate outwards. As a rule this mass is perforated in the centre for the passage of the descending artery. In the more primitive Dromiacea, however, the concentra- FIG. 167. Diagram of the excretory system of Eupagurus bernhtirdus. abd, unpaired vesicle lying in abdomen ; arb, arborisations of the vesical system in thorax ; I, labyrinth ; o, external opening ; s, saccule. (After Marchal.) THE DECAPODA 287 tion is somewhat less complete, the outlines of five pairs of ganglia can be distinguished in the central mass, while posteriorly is a shortened chain of five pairs of ganglia corresponding to the abdominal somites, although not extending beyond the posterior limits of the thorax. A system of visceral nerves is well developed in the Decapoda. A gastric plexus is formed by anastomosis of three nerves, a median one arising from the posterior surface of the cerebral ganglia and a pair from the oesophageal commissures. Special nerves to the rectum are given off by the last abdominal ganglion. Sense-Organs. — The paired eyes are well developed in the great majority of Decapods, although, as already mentioned, they may be reduced or entirely absent in deep-sea and cave-dwelling forms as well as in some parasitic and burrowing species. The cornea is generally distinctly faceted, the facets being square or hexagonal in outline. Sometimes they are square in the centre of the corneal area and hexagonal towards the margin. As a rule the crystalline cone is formed by four cells, and there are seven retinular cells enclosing a quadripartite rhabdome. The nauplius-eye has been found to persist in a vestigial condition in the adult in many of the lower Decapoda. A pair of statocysts lodged in the proximal segment of the anten- nules occur in the great majority of Decapods (Fig. 156, A, st, p. 265). Only in certain Caridea do these organs appear to be entirely wanting (Pandalus, Hippolyte). The statocyst develops as an invagination of the integument, and in most of the lower Decapods it remains in communication with the exterior, sometimes by a wide aperture (Crangori), more commonly by a narrow slit. Rarely among the Natantia the statocyst appears to be quite closed (Leucifer, Sergestes}, and this is the case also in the Galatheidea and Hippidea among the Anomura and in the whole of the Brachyura. In the Brachyura, after ecdysis, the statocyst is open to the exterior by a narrow slit, which, however, soon closes by coalescence of the newly formed cuticle covering its edges. In this group also the cavity assumes a complex form by the folding of its walls. In those cases where the statocyst remains open it contains a number of foreign particles, sand-grains, which act as statoliths, and are in some cases agglutinated together into a mass by an organic substance secreted by dermal glands on the inner surface of the sac. In this mass the tips of the sensory setae are embedded. When ecdysis takes place the chitinous cuticle lining the statocyst is thrown off and with it the contained sand- grains, and it has been shown that fresh grains are introduced by the animal either burying its head in the sand or placing the grains in position by means of its chelae. When the statocyst is without external opening it usually contains no solid particles. 288 THE CRUSTACEA This is the case in the Brachyura and the majority of the Anomura. In the few Natantia, however, which have closed statocysts, solitary statoliths, probably of organic composition, are present, which are no doubt formed in situ. As with the open statocysts, the lining membrane, and with it the statolith, is cast and renewed at each ecdysis. In all cases the inner surface of the statocyst bears plumose sensory setae arranged in one or more rows. In Leucifer the tips of the sensory hairs are embedded in the substance of the statolith. The development of the statocysts has been traced in the Lobster and the Shore-crab. In both cases the functional state is assumed rather suddenly ; at the fourth larval stage in the lobster and the first Megalopa-stage in the Crab. In the latter the statocyst is at first open to the exterior and sand-grains are found in it. Sensory filaments occur in most cases on the external flagellum of the antennules, commonly in larger numbers in the male than in the female sex. In many Caridea they are confined to a specially thickened portion of the flagellum, and when the flagellum bifurcates the filaments are borne by the secondary branch (Fig. 156, B, p. 265). In some terrestrial species (CoenoUta) the filaments are very short, forming a close fur. Phosphorescent organs are now known in a number of deep-sea Decapoda (Sergestidae, Penaeidae, Hoplophoridae, Pandalidae, Eryonidae), but the nature of the organs differs widely in the different groups. In Aristeus coruscans (Penaeidae) (Fig 168) and Heterocarpus alphonsi (Pandalidae) (Fig. 148, p. 259) Alcock observed a luminous fluid to be emitted from the base of the antennae, apparently from the orifices of the antennal glands. This case may be compared with that of Gnathophausia among the Mysidacea, where a luminous secretion is produced by a gland on the maxilla which may possibly be the excretory maxillary gland. Polycheles phosphorus (Eryonidae) was found by the same observer to be " luminous at two points between the last pair of thoracic legs, where there is a triangular glandular patch." Numerous phosphorescent organs have been found by Coutiere on the body and limbs of various Hoplophoridae, but their structure has not been examined. In Sergestes challengeri Hansen has found an extraordinary number of luminous organs (about 150) on the body and limbs, although they are not found in other species of the genus. In this case the structure recalls that found in the Euphausiacea. Each organ has, internally, a reflector, composed apparently of concentric lamellae, enclosing a mass of cells. There is nothing corresponding to the "striated body" of Euphausiacea, and the "lens" is double, the outer part being formed by a thickening of the cuticle which has no counterpart in the Euphausiacea. THE DECAPODA 289 Reproductive System. — The testes as a rule lie partly in the thoracic region and partly in the abdomen, and, except in some Paguridae, are connected with each other across the middle line. In the simplest cases, as in some Caridea, they are tubular in form, but as a rule they send off numerous caecal diverticula. In Leucifer the two testes unite with each other to form an unpaired mass lying below the intestine. In the Paguridae they are dis- placed backwards so as to lie wholly in the abdominal region, where they are unsymmetrically placed on the left side, either fused into a single mass or entirely separated from each other, the right testis lying in front of the left. The vas deferens presents typically three divisions (which, however, are not distinctly defined in many Caridea) : (1) a narrow Fid. 168. Aristeus coruscans (Penaeidae). (Prom Alcock, Naturalist in Indian Seas.) efferent duct leading from the testis ; (2) a glandular part, with wider lumen, often convoluted ; (3) a terminal ductus ejaculatorius with muscular walls. In Brachyura (except Dromiidae) the distal portion of the second or glandular division is provided with caecal diverticula which in some cases are very numerous, forming a large glandular mass. In Leucifer the structure of the vas deferens is very complex, and there are two distinct glandular regions. In nearly all cases the terminal portion of the vas deferens perforates the coxopodite of the last pair of legs, or emerges on the arthrodial membrane between the coxopodite and the body. In the majority of cases no penes are formed, but in some genera of Paguridae (Spiropagurus, etc.) a membranous tubular penis is present on one side only. In all Brachyura a pair of penes are present, the tips of which lie within the grooves of the first pair 'of abdominal appendages. In those families of Brachygnatha formerly 19 290 THE CRUSTACEA grouped together as Catometopa these penes either lie for a short distance from their base within grooves excavated in the last thoracic sternum, or else they perforate the sternum directly, the vasa deferentia in this case not entering the coxopodites of the legs at all, as they do in most other Decapods. In most Macrura and in the Dromiidae the spermatozoa when discharged are enclosed by a sheath of secretion which sets to a firm membrane, forming a continuous cord-like mass. In Scyllarus and in the Anomura this is broken up into separate spermatophores attached by one end in a row on a strip of membrane. In the Brachyura (except the Dromiidae) the spermatophores are quite separate. The spermatozoa are remarkably varied and complex in structure. As a rule they are provided with stiff radiating pro- cesses which serve to attach them to the surface of the egg, and, in some cases, an " explosive " apparatus is present which effects penetration of the egg-membrane. The ovaries generally resemble the testes in shape and position. In the Penaeidae they may extend through the whole length of thorax and abdomen, but in most cases they are of less extent. In Leucifer, and in most if not all Thalassinidea and Paguridea, they lie wholly in the abdomen. They are always united across the middle line, sometimes at more than one point. In the Crayfish the two ovaries (like the testes) are joined together posteriorly so that the organ has a trilobed form. Except in Leucifer and in the Brachyura, the oviducts are simple in form and open on the coxopodites of the sixth thoracic appendages (third legs). In Leucifer, which is peculiar in so many points of structure, the oviducts have receptacula seminis connected with them and unite to open by a median aperture on the sternal surface of the thorax. In the Brachyura, where intromittent organs are developed in the male, the terminal part of the oviduct is of considerable length and serves as a vagina, while a lateral pouch, sometimes double, with glandular walls, forms a receptaculum seminis. In the Dromiacea the receptaculum seems to be a temporary structure formed at the time of copulation. In the Brachyura, with exception of the primitive Dromiacea and of certain Oxystomata (Raninidae, some Dorippidae), the oviducal apertures are removed from the coxo- podites of the legs and open on the sternum of the corresponding somite. In addition to the internal receptacula seminis mentioned above, an external organ having apparently the same function is found in certain Decapods. It is best known in the Penaeidae, where it has been named the tlielycum and affords characters of systematic importance. It lies on the sternal surface of the thorax and appears to be formed by two or more outgrowths from the last THE DECAPODA 291 33. thoracic somite enclosing a ctavity within which may sometimes he found the large foliaceous spermatophores deposited by the male. In the Lobster (Homarus) (Fig. 169) a median pouch enclosed by three processes on the sterna „ of the last two thoracic „ somites has the same function, and in the Crayfishes of the genus Cambarus (but not in Astacus) a more complicated organ in the same position is known as the " annulus ventralis." These structures have not hitherto been studied from a comparative point of view, but it seems likely that an investigation of their mor- phology and their relation to 3P- LP f- Fio. 169. the structures occupying a Similar position in the Syn- , Sternal surface of posterior thoracic somites of " J female Lobster (Honutrus r/c iiimn /•».-), showing the Carida Would yield important receptacuium seminis. VI-VIII, bases of the last lf three pairs of legs; f '^-P. median process on last thoracic sternum; sp, 111 tne great majority r>r,, pleopods ; abg, THE DECAPODA 299 of which two races are known, the one found in Southern Europe being exclusively freshwater in habitat ; the other, found in Britain and Northern Europe, inhabiting brackish or salt water. The former hatches at a stage when all the limbs except the uropods are present, and the first two pairs of legs have exopodites. In B FIG. 17X Larval stages of Sergestes. A, Protozoea-, B, Sehizopod-stage(^ca?it7ioso)na). a', antennule ; o", antenna ; mf'-inf", the three pairs of maxillipecls ; MX', maxillula : mx", maxilla ; ol, labrum ; ml, maxillary gland. (After Glaus, from Korschelt and Heider's /•,'/«'/;•//'>/'<;///.) the northern race all the ambulatory legs are rudimentary on hatching, and there are no abdominal limbs. None of the Astacura are known to possess a Zoea-stage. The Lobster (Homarus) is hatched in the Schizopod-stage (Fig. 176), with natatory exopodites on all the thoracic limbs, but without any abdominal appendages. In the further course of development 300 THE CRUSTACEA the uropods are the last to appear. In Nephrops (Fig. 177) the course of development is very similar, but the larvae are distinguished by the long spines of the abdominal somites and telson. The freshwater Cray- fishes have a direct develop- ment, the young on hatching resembling the adult in most points, but lacking the first and the last pairs of abdominal appendages. The Scyllaridea have a very peculiar and character- istic series of larval forms, which were long described as adults under the generic name Phyllosoma (Fig. 178). These larvae are remarkable for the Larva of American Lobster (Homarus ameri- large size to which SOme attain, canus), in Schizopod-stage. «', antennule ; ,;<;* / ' . 1 . (Galatheidea). mf-mf", the three Space Will not permit Of aliUSlOn to pairs of maxillipeds ; p, rudiments ,i f ,1 of legs and inlls- (After Sars, from more than one or two of the more Korscheit and Heid.-i •'* salient points. 3°4 THE CRUSTACEA -d.S f: Truly freshwater Decapods (apart from estuarine or brackish- water species which may penetrate into fresh water) are found among the Natantia in the family Atyidae and in several genera of Palaemonidae ; of the Reptantia, the Crayfishes of the families Astacidae and Parastacidae, the monotypic Aegleidae among the Ano- mura, and the Potamonidae (Thelphusidae) and numerous species of Grapsidae among the Brachyura, are also dwel- lers in fresh water. Some of these are more or less am- phibious in their habits, like many Potamonidae, and some Crayfishes are found burrow- ing in the earth far from streams or ponds, their bur- rows reaching down to the ground -water. The same is reported of the marine or brackish-water Thalassina. Truly terrestrial species are found among Paguridae (Birgus and Coenobita) and Brachyura (Gecarcinidae), and it is interesting to note that these are derived not from freshwater but from marine types, and all (except, possibly, some Gecarcinidae) pass- their early stages in the sea. The Sergestidae belong to the plankton, occurring at the surface and descending to great depths. A few Brachyura (Planes and some other Grapsidae) lead a pelagic life, clinging to driftweed, floating timber, and the like. It is worthy of note that the deep-sea Decapods include the more primitive members of each of the chief subdivisions : the Aristeinae among the Penaeidea, the Hoplophoridae among the Caridea, the Eryonidea among the Palinura, the Pylochelidae among the Paguridea, and the Homolodromiidae among the Brachyura. Parasitism and commensalism in varying degrees are common. The Paguridae alone, which live in the empty shells of Gasteropod molluscs, present a whole series of cases of commensal association with Sponges, Coelentera (Fig. 183), and Polychaete worms; Fio. 181. First Zoea-stage (after the first moult) of Cnn-'ni \u maenas. a', antennule ; a", antenna ; abd, abdomen ; d.s, dorsal spine of carapace (the so-called "Zoea "- spine) ; /, furcate telson ; md, mandible ; inx', maxil- lula ; mx", maxilla ; r, rostrum ; th', th", first and second maxillipeds, biramousand natatory ; th^-tlv, rudiments of three following thoracic appendages. (After Faxon.) THE DEC APOD A 3°5 Spongicola, Typton, and Eiconaxius live in sponges, many Pontoniinae and Pinnotheridae (as Aristotle knew) in the mantle-cavity of bivalve molluscs; members of the first-named family inhabit the pharyngeal cavity of Tunicates, and some Pinnotheridae are found in the " respiratory trees " of Holothurians and the rectum of sea-urchins. Many Decapods are constantly found among living corals, and the Hapalocarcinidae live in "galls" on the branches of corals. A very peculiar habit is that of some crabs of the genus Melia, which carry in each cheliped a living Actinian and use it as a weapon. FIG. 18:?. Later stages of Curdim* > »<" •/'">. A, young Megalopa. B, older Megalopa. C, post-larval .stage. (/, dorsal spine of carapace ; r, rostrum (A after Spence- Bate ; B and C after Brook. From Kor.schHt and Heider's Embryology.) Special interest attaches to the stridulating organs, found in many Decapoda, since their possession is presumptive evidence that the animals do have some power of hearing. A few Penaeidae, some Palinuridae, and a considerable number of Brachyura are now known to have stridulating organs in various parts of the body. That of Ocypoda, shoAvn in Fig. 184, is one of the few of which the sound-producing function has been demonstrated by observation of the living animals. It consists of a file-like series of ridges (a) on the inner surface of the propodite of one of the chelipeds, which can be rubbed up and down upon a sharp ridge (b) on the ischiopodite of the same appendage, producing a hissing sound, which probably 306 THE CRUSTACEA serves to warn intruders from entering the burrows of these shore- living crabs. In the case of purely aquatic species, the function of these organs is less easy to understand. The range of size in Decapoda is greater than in any other group of Crustacea. Some Natantia do not exceed half an inch in length, one Pagurid is adult when 8 mm. long, a species of Porcellanid has a carapace measuring 3 mm. by 5 mm., and some Brachyura are no larger. The largest forms are found among the FIG. 183. Parapagwrus pilosimanus (Paguridae), lodged in a colony of zoantharian polypes. (From Alcock, Naturalist in Indian Seas.) Reptantia ; some Palinuridae and Astacura reach one or even two feet in length and are bulky in proportion. The largest living Arthropod is the Japanese crab Macrocheira (or Kaempferia) Kaempferi, of which the carapace may measure 15 inches in length, and the extended chelipeds of the male may span more than 10 feet. PALAEONTOLOGY. Fossil remains of Decapods are not known with certainty from any Palaeozoic deposits. Many genera from the Devonian upwards THE DECAPODA 307 have indeed been described as belonging to this group, but in no case is enough known of their characters to enable more to be said than that they agree with the " caridoid " groups of the Malacostraca in the possession of a carapace arid of a tail-fan. In the Mesozoic rocks many undoubted Decapods occur, includ- ing representatives of all the chief groups now living. Many genera of Penaeidea are found from the Jurassic, perhaps from the Triassic period onwards, some of the earliest even resembling closely the existing genus Penaeus, to which they have been referred. Aeger, from Triassic and Jurassic rocks, presents characters which suggest an affinity with the Stenopidea. True Caridea appear later, in the Upper Jurassic, some at least presenting primitive characters in the retention of exopodites on the ambulatory limbs. Fresh- water Caridea of doubtful affinities occur in the Miocene. The Eryonidea are especially interesting since the few existing deep- FKI. 1S4. Larger c.hi-liped of Oi'/^W" macrocera, from the inner side, showing the stridulating mechanism, c, tile-like scries of ridges on propodite ; /;, ridge or scraper on ischiopodite against which the ridges of the propodite can be rubbed when the limb is flexed. (From Alcoi-k, Suti'i-iiiixt . 260). Sub-Family MUNIDOPSINAE. Munidopsis, Whiteaves. Family PORCELLANIDAE. Porcellana, Leach ; Petrolisthe.^ Stimpson. TRIBE 2. THALASSINIDEA. Abdomen extended, symmetrical ; body compressed ; rostrum some- times well developed ; first legs chelate, rarely sub-chelate ; tail-fan well developed. Family AXIIDAE. Axiux, Leach (including Eiconaxius, Spence Bate, and Iconaxiopsis, Alcock (Fig. 149, p. 26*0), as subgenera) ; Calocaris, Bell. Family LAOMEDIIDAE. Laomedia, de Haan ; Jaxea, Nardo. Family CALLIANASSIDAE. Sub-Family CALLIANASSINAE. Callianassa, Leach ; Callianidea, H. Milne-Edwards. Sub-Family UPOGEBIINAE. Upogebia, Leach ( = Gebia, Leach). Family THALASSINIDAE. Thalassina, Latreille. TRIBE 3. PAGURIDEA. Abdomen nearly always asymmetrical, either soft and twisted or bent under thorax ; rostrum generally small or absent ; first legs chelate ; tail- fan not typical, uropods (when present) adapted for holding the body into hollow objects. Family PTLOCHELIDAE. Pylocheles, A. Milne-Edwards (Fig. 151, p. 261). Family PAGURIDAE. Sub-Family PAGURINAE. Pagurus, Fabricius ; Clibanarius, Dana. Sub-Family EUPAGURINAE. Eupagurus, Brandt; tipiropagurus, Stimpson; Parapagurus, S. I. Smith (Fig. 183, p. 306). Family COEXOBITIDAE. Coenobita, Latreille; Birgus, Leach (Fig. 152, p. 262). Family LITHODIDAE. Sub-Family LITHODINAE. Li tit odes, Latreille ; Neolithodes, Milne-Edwards and Bouvier (Fig. 153, p. 262). Sub-Family HAPALOGASTRINAE. Hapalogaster, Brandt TRIBE 4. HIPPIDEA. Abdomen bent under thorax, symmetrical ; rostrum small or absent ; first legs styliform or sub-chelate ; tail-fan not adapted for swimming. 314 THE CRUSTACEA Family ALBUNEIDAE. Albunea, Fabricius. Family HIPPIDAE. Hippa, Fabricius ; Remipes, Latreille. SECTION 4. BRACHYURA. Abdomen small, symmetrical, bent under thorax, tail-fan not developed ; carapace fused with epistome at sides and nearly always in the middle ; exopodites of maxillipeds with tfagella, when present, bent inwards; first legs always, third legs never, chelate ; no appendix interna on pleopods ; uropods rarely present, never biramous ; branchiae gener- ally few. TRIBE 1. DROMIACEA. Last pair of legs modified, dorsal in position ; female openings on coxopodites ; first pleopods present in female ; branchiae sometimes numerous ; month-frame quadrate. SUB-TRIBE 1. DBOMIIDEA. Sternum of female with longitudinal grooves ; vestiges of uropods usually present ; branchiae 14-20 on each side ; eyes completely sheltered by orbits ; no linea homolica on carapace. Family HOMOLOUROMIIDAK. Homolodromia, A. Milne -Ed wards. Family DROMIIDAE. Dromia, Fabricius. Family DYNOMEXIDAE. Dynomene, Latreille. SUB-TRIBE 2. HOMOLIDEA. Sternum of female without longitudinal grooves ; no uropods ; branchiae 8-14 on each side; eyes not completely sheltered by orbits ; linea homolica usually present on carapace. Family HOMOLIDAE. Homola, Leach. Family LATREILLIIDAE. Latreillia, Roux. TRIBE 2. OXYSTOMATA. Last pair of legs normal or modified ; female openings generally on sternum ; first pleopods wanting in female ; branchiae few ; mouth-frame triangular, produced forwards over epistome. Family DORIPPIDAE: Dorippe, Fabricius ; Ethusa, Roux ; Gyclodorippe, A. Milne-Edwards. Family RAXINIDAK. Ranina, Lamarck. Family CALAPPIDAE. Sub-Family CALAPPINAE. Calappa, Fabricius. Sub- Family ORITHYIXAE. Ortihya, Fabricius. Sub-Family MATDTIXAE. Matuta, Fabricius. Family LEDCOSIIDAE. Sub-Family LEUCOSIINAK. Leucosia, Fabricius ; Ebalia, Leach. Sub-Family ILIINAE. Ilia, Leach. TRIBE 3. BRACHYGNATHA. Last pair of legs normal, rarely reduced or dorsal in position ; female openings on sternum ; first pleopods wanting in female ; branchiae few ; mouth-frame quadrate. THE DECAPOD A 315 SUB-TRIBE 1. BRACHYRHYXCHA. Body not narrowed in front ; rostrum reduced or wanting ; orbits well formed. Family CORYSTIDAE. Corystes, Latreille. Family .PORTOXIDAE. Sub-Family CARCIXINAK. Carcinus, Leach (Carcinides, Ratnbun). Sub- Family PORTUMNIXAE. Portumnus, Leach. Sub-Family CATOPTRINAE. Catoptrus, A. Milne-Edwards. Sub-Family CARUPIXAE. Carupa, Dana. Sub-Family PORTUXIXAE. Portunus, Fabricius. Si\b-Family CAPHYRIXAE. Gaphyra, Guerin. Sub-Family THALAMITINAE. Thalamita, Latreille. Sub-Family PODOPHTHALMIXAE. Podophthalmus, Lamarck. Family POTAMOXIDAE. Sub-Family DECKENIINAE. Deckenia, Hilgendorf. Sub- Family PsEUDOTHKi.PHrsiXAi:. Pseudothelphusct, Saussure. Sub-Family POTAMOXIXAE. Potamon, Savigny (= Tlielphusa, Latreille). Sub-Family TRICHODACTYLIXAE. Trichoductylus, Latreille. ^Family ATELECYCODAK**- • Sub-Family ACANTHOCYCLINAE. Acanthocydus, Milne-Edwards and Lucas. Sub-Family THIIXAE. Thia, Leach. Sub-Family ATELECYCLIXAE. Atele- cyclus, H. Milne-Edwards. ^Family CAXCRIDAE. Sub-Family CANCRINAE. Cancer, Linnaeus. Sub-Family PIRIJIELIXAE. Pirimela, Leach. Family XAXTHIDAE. Sub-Family XAXTHIXAE. Xantho, Leach. Sub-Family CARPILIIXAE. Carpilius, Leach. Sub-Family ETISINAE. Etisus, H. Milne-Edwards. Sub-Family MEXIPPIXAE. Menippe, de Haan. Sub- Family OZIIXAE. Oziu*t H. Milne- Ed wards. Sub-Family ERIPHIINAE. Krijihia, Latreille. Sub-Family TRAPEZIIXAE. Trapezia, Latreille. »-Family GOXOPLACIDAE. Sub-Family RHIZOPINAE. Rhizopu, Ptimpson. Sub-Family PRIOXOPLACINAE. Prionoplax, H. Milne-Edwards. Sub- Family GONOPLACIXAE. Gonoplox, Leach. Sub-Family CARCINOPLACIXAK. Oardnoplax, H. Milne-Edwards. Sub-Family HEXAPODIXAE. Hexapus, de Haan. ( Family PIXNOTHERIDAE. Sub-Family PINXOTHERIXAE. Pinno- theres, Latreille. Sub-Family PIXXOTHERELIIXAE. Pinnotherelia, Milne- I'M wards and Lucas. Sub-Family XEXOPHTHALJIIXAE. Xenophthalmus, White. Sub-Family ASTHENOGXATHIXAE. Asthenoynathus, Stimpson. •^ Family PTENOPLACIDAK. Ptenoplax, Alcock and Anderson. Family * PALICIDAE. Palicus, Pliilippi ( = Cymopolia, Roux). -^Family GRAPSIDAE. Sub-Family PLAGUSIIXAE. Placjusia, Latreille. Sub-Family SESAR- MIXAE. Sesarma, Say. Sub-Family GRAPSIXAE. Grapsus, Lamarck ; I'lancs, Bowdich (= Nautilotjrapsus, H. Milne-Edwards). Sub-Family VARIXIXAE. Varuna, H. Milne-Edwards. A Family GECARCIXIDAE. Gecarcinus, Leach ; Cardisoma, Latreille. A Family OCYPODIDAE. Sub- *?Family MACROPHTHAI.MIXAK. Macrophthalmui, Latreille. Sub-Family OCYPODIXAE. Ocypoda, Fabricius (Fig. 155, p. 264); Gelasimus, Latreille. Sub-Family MYCTIRIXAE. Myctiris, Latreille.^ Family HAPALOCAR- CEXIDAE. Hapalocarcinus, Stimpson. SuB-TlUBE 2. O.XYRHYXCHA. Body narrowed in front ; rostrum usually distinct ; orbits generally incomplete. 316 THE CRUSTACEA Family HYMEXOSOMIDAE. Hymenosoma, Desmarest. Family MAIIDAE. Sub-Family IXACHINAE. Inachiis, Fabricius ; Macrocheira, de Haan ; Macropodia, Leach. Sub-Family ACAXTHONYCHINAE. Acanthonyx, Latreille. Sub-Family PISINAE. Pisa, Leach ; Hyas, Leach. Sub-Family MAIINAE. Maia, Lamarck ( = Mamaia, Stebbing) ; Pericera, Latreille ; Mithrax, Leach. Family PARTHENOPIDAE. Sub-Family PARTHENOPINAE. Parthenope, Fabricius ; Lambrus, Leach. Sub-Family EUMEDOXINAE. Eumedonus, H. Milne-Edwards. LITERATURE. 1. Alcock, A, Materials for a Carcinological Fauna of India. Nos. 1 to 6. (Brachyura.) Journ. Asiatic Soc. Bengal, Ixiv., Ixv., Ixvii., Ixviii., Ixix., 1895-1900. 2. An Account of the Deep -Sea Brachyura collected by the . . . "Investigator," pp. ii + 85, 4 pis. 4to. Calcutta, Indian Museum, 1899. 3. A Descriptive Catalogue of the Indian Deep-Sea . . . Macrura and Anomala . . ., pp. 286 + iv, 3 pis. 4to. Calcutta, Indian Museum, 1901. 4. A Catalogue of the Indian Decapod Crustacea in the . . . Indian Museum. Part I. Brachyura : Fasciculus 1, Introduction and Dromides or Dromiacea, pp. ix. +80, 8 pis., 1901 ; Part II. Anomura : Fasc. 1, Pagurides, pp. xi + 197, 16 pis., 1905; Part III. Macrura: Fasc. 1, The Prawns of the Pencus Group, pp. 55, 9 pis., 1906. 5. Bate, C. Spence. Report on the Crustacea Macrura. Rep. Voy. " Challenger," Zool. xxiv. pp. 90 + 942, 157 pis., 1888. 6. Bell, Thomas. A History of the British Stalk-eyed Crustacea. * Svo. London, 1853. Issued in parts, 1844-53. 7. Boas, J. E. V. Studier over Decapodernes Slaegtskabsforhold. Kgl. Danske Vidensk. Selsk. Skrifter (6), i. (2), pp. 25-210, 7 pis., 1880. 8. Kleinere carcinologische Mittheilungen. 2. Ueber den ungleichen Entwicklungsgang der Salzwasser- und der Siisswasser-Form von Palae- monetcs varians. Zool. Jahrb. Abth. Syst. iv. pp. 793-805, pi. xxiii., 1889. 9. Borradailc, L. A. Land Crustaceans ; Marine Crustaceans, pts. i.-vii., ix.-xi., xiii. Gardiner's Fauna . . . Maldive and Laccadive Archipelagoes. Cambridge, 1901-1904. 10. On the Classification of the Decapod Crustaceans. Ann. Mag. Nat. Hist. (7), xix. pp. 457-486, 1907. 11. Bouvier, E. L. Le systeme nerveux des Crustaces Decapodes et ses rapports avec 1'appareil circulatoire. Ann. Sci. Nat. Zool. (7), vii. pp. 73-106, pi. vii., 1889. 12. Recherches anatoraiques sur le systeme arteriel des Crustaces Deca- podes. Ann Sci. Nat. Zool. (7), xi. pp. 197-282, pis. viii.-xi., 1891. 13. Sur 1'origine homarienne des Crabes : Etude comparative des Dro- miaces vivants et fossiles. Bull. Soc. Philomath. Paris (8), viii. pp. 34- , 110, 1896. 14. Brooks, W. K. Lucifer : A Study in Morphology. Phil. Trans, clxxiii. pp. 57-137, 11 pis., 1882. 15. Bumpus, H. 0. The Embryology of the American Lobster. Journ. Morph. v. pp. 215-262, pis. xiv.-xix., 1891. THE DEC APOD A 317 16. Coutiere, If. Les " Alpheidae," Morphologic externe et interne, formes lurvaires, Bionomie. Ann. Sci. Nat. Zool. (8), ix. pp. 1-559, pis. i. -vi. , 1899. 17. Doflein, F. Brachyura. Wiss. Ergeb. deutschen Tiefsee-Expedition . . . " Valdivia," vi. pp. xii + 314, and Atlas, 58 pis., 1904. 18. Faxon, W. A Revision of the Astacidae. Mem. Mus. Comp. Zool. Harvard, x. No. 4, pp. vi+186, 10 pis., 1885. 19. Reports ..." Albatross." XV. The Stalk-Eyed Crustacea. Mem. Mus. Comp. Zool. Harvard, xviii. pp. 292, 67 pis. and map, 1895. 20. Henderson, J. 11. Report on the Anomura. Rep. Voy. "Challenger," Zool. xxvii. pp. 7 + 215, 21 pis., 1888. 21. Herrick, F. H. The American Lobster : A Study of its Habits and Develop- ment. Bull. U.S. Fish Commission for 1895, pp. 1-252, 64 pis., 1895. 22. Huxley, T. H. On the Classification and Distribution of the Crayfishes. Proc. Zool. Soc. London, 1878, pp. 752-788, 7 figs. 23. Marclial, P. Recherches anatomiques et physiologiques sur 1'appareil excreteur des Crustaces Decapodes. Arch. Zool. exper. (2), x. pp. 57-275, 9 pis., 1892. 24. Miers, E. J. On the Classification of the Maioid Crustacea or Oxyrhyncha, with a Synopsis of the Families, Sub-Families, and Genera. Journ. Linn. Soc. London, Zool. xiv. pp. 634-673, pis. xii., xiii., 1879. 25. Report on the Brachyura. Rep. Voy. "Challenger," Zool. xvii. pp. 50 + 362, 29 pis., 1886. 26. Milne-Edwards, H. Observations sur le squelette tegumentaire des Crus- taces Ducapodes et sur la morphologic de ces animaux. Ann. Sci. Nat. Zool. (3), xvi. pp. 221-291, pis. viii.-xi., 1851. 27. Milne- Edwards, A., and Bouvicr, E. L. Reports . . . "Blake." XXXIII. Description des . . . Paguriens . . ., Mem. Mus. Comp. Zool. Harvard, xiv. No. 3, pp. 172, 12 pis., 1893 ; XXV. Description des ... Gala- theides . . ., op. cit. xix. No. 2, pp. 141, 12 pis., 1897; XXXIX. Les Dromiaces et Oxystomes, op. cit. xxvii. No. 1, pp. 127, 25 pis., 1902. 28. Considerations generales sur la famille de Galatheides. Ann. Sci. Nat. Zool. (7), xvi. pp. 191-327, 1894. 29. Crustaces Decapodes . . . " Hiroudelle " . . . lre Ptie, Brachyures et Anomures. Resultats des Campagnes scientifiques . . . Monaco, Fasc. vii. pp. 112, 11 pis., 1894 ; also Crust. Decap. . . . " Hirondelle " (Supple"- ment) et . . . " Princesse Alice," op. cit. Fasc. xiii. pp. 106, 4 pis., 1899. 30. Exped. Scient. du " Travailleur " et du "Talisman." Crustaces Decapodes, lre Ptie, pp. 396, 32 pis., 1900. 31. Mocquard, M. F. Recherches anatomiques sur 1'estomac des Crustaces podophthalmaires. Ann. Sci. Nat. Zool. (6), xvi. pp. 1-311, 11 pis., 1883. 32. Miiller, F. Die Venvandlung der Garne<5len. Arch. Naturgesch. xxix. pp. 8-23, pi. ii., 1863. Translated in Ann. Mag. Nat. Hist. (3), xiv. pp. 104- 115, pi. iv., 1864. 33. Ortmann, A. Die Decapoden-Krebse des Strassburger Museums. 8 pts. Zool. Jahrb. Abth. Syst. v.-vii., 1890-1894. 34. - • Das System der Decapoden-Krebse. Op. cit. ix. pp. 409-453, 1896. 35. Prentiss, C. W. The Otocyst of Decapod Crustacea ; its Structure, Develop- ^/ ment, and Functions. Bull. Mus. Comp. Zool. Harvard, xxxvi. Xo. 7, pp. 166-251, pis. i.-ix., 1901. 36. Przibram, H. Experimentelle Studien iiber Regeneration. 4 papers. Arch. Entwicklungsmechanik, xi., xiii., xix., xxv., 1901-1907. 3i8 THE CRUSTACEA 37. Rathbun, Mary J. Les Crabes d'eau douce (Potamonidae). 3 parts. Xouv. Arch. Mus. d'Hist. Nat. Paris, vi., vii., viii., 1904-1906. 38. Reichcnbach, H. Studien zur Entwicklungsgeschichte des Flusskrebses. Abh. Senckenberg. naturforsch. Ges. xiv. (1), pp. 137, 19 pis., 1886. 39. Sars, G. 0. Bidrag til Kundskaben om Decapodernes Forvandlinger. 3 papers. Arch. Math. Naturvid. ix., xiii., xiv., 1884-1890. 40. Thompson, J. Vauyhan. Zoological Researches, vol. i. 8vo. Cork, 1828. CHAPTER XVI THE STOMATOPODA jJC DIVISION HOPLOCARIDA. Order Stomatopoda, Latreille (1817). FOR a definition of the Division Hoplocarida, see p. 149. Historical. — The common and conspicuous Squilla mantis of the Mediterranean can hardly have escaped notice in antiquity, and it is surprising that it cannot be identified with certainty among the Crustacea mentioned by Aristotle. It was described by Kondelet (1555) under the generic name which it still bears. The groTip Stomatopoda, as defined by Latreille in 1817, had practically the limits now assigned to it, though some larvae, were admitted to generic rank along with the adults. By H. Milne-Edwards the group was extended to include not only the " Schizopoda," but also some larval and adult Decapods (Pkyllosoma, Leucifer, etc.). Restricted by subsequent writers to the single family Squillidae, the Order has generally been ranked along with " Schizopoda " and Decapoda in the group Podophthalrna, though Huxley and, later, Grobben have pointed out the great differences separating the Stomatopoda from the other stalk-eyed groups. The first details as to the larval metamorphosis of the Order Avere given by F. Miiller (1862-64). Glaus, in a remarkable memoir (1871), traced out several developmental series. Later workers, especially Brooks (1886) and Hansen (1895), have succeeded in referring many larvae to the various genera and species of adults. It is to be noted, however, as Hansen has pointed out, that the number of specific forms among the larvae exceeds that of the known adult species. MORPHOLOGY. The general appearance of the Stomatopoda is highly character- istic and very constant throughout the group. Its most striking features are due to the great development of the abdominal region and its appendages, the small size of the carapace, and the large and peculiarly formed raptorial limbs. 319 320 THE CRUSTACEA The body (Fig. 186) is more or less flattened dorso-ventrally. The carapace is fused dorsally with at least two of the thoracic somites, two others are represented by indistinct vestiges over- lapped by its hinder edge, while the last four are free and com- pletely developed. The lateral wings of the carapace project more or less horizontally, roofing over on each side a widely open channel th. —HI*. FIG. 186. Squilla mantis, male, from the side, a', antennulf ; a", antenna ; p, penis ; sc, scale or exopodite of antenna ; tit1, UP, th$, first, second, and last thoracic appendages. within which lie the epipodites of the anterior thoracic appendages, and which corresponds to the branchial cavity of other forms. Anteriorly the carapace does not extend to the front of the head (Fig. 187), leaving uncovered two movably ar-ticulated segments, FIG. 1ST. Anterior part of body of Squilla mantis, from above, a', antennule ; a", antenna ; a.f, antennular segment of head ; c, carapace ; o.s, ophthalmic segment of head ; r, rostral plate ; sc, scale or exopodite of antenna. which carry respectively the eyes and the antennules, and which are commonly regarded as representing the ocular (o.s) and antennular (a.s) somites. A small rostral plate .(r), movably articulated with the front edge of the narapace, overlies the antennular segment. That part of the head lying between the point of attachment of the antennae and that of the mandibles is much elongated, forming THE S TO MA TO POD A 321 a narrow " neck," which, except for the lateral wings of the carapace projecting on either side, recalls the similarly formed "neck" of Leudfer and of the Trachelifer-lajrv& of Jaxea. The anterior thoracic somites are much abbreviated and crowded together. The first and second are apparently not distinct from the carapace in the adult. The third and fourth are at most represented dorsally by small sclerites overlapped by the hinder part of the carapace. The fifth and succeeding thoracic somites are complete, and movably articulated. The abdominal somites often increase in width posteriorly, and their horizontally extended pleural plates may become greatly expanded in certain species. The telson (Fig. 188, t) is very broad and its posterior margin is generally cut into sharp teeth ; it is firmly united to the preced- ing somite in certain species of Gonodactylus (Protosquilla, Brooks). Fn:. 188. Caudal fan of Squilla mantis, upper surface, en, endopodite ; ex, exopodite ; j), process from peduncle of uropod ; t, telson, The surface of the carapace and of the body-somites is often ornamented with longitudinal keels, and the telson is always more or less elaborately sculptured. Appendages. — The antennules (Fig. 187, a) have an elongated peduncle of three segments, which bears three comparatively short flagella. Of these, the two on the outer side spring from a common stalk which is unsegmented ; the inner flagellum is also unsegmented for a short distance from its base. The antennae (Fig. 187, a") have a protopodite of two segments, a large exopodite, and a comparatively feeble endopodite. The exopodite consists of a small basal segment and an oval mem- branous scale (sc) with setose margins ; the endopodite has two elongated proximal segments and a short flagellum. The mandibles (Fig. 189, A) carry a slender palp of three segments. The oral edge is crescentic and strongly serrate, its two cornua corresponding respectively to the incisor and molar processes of other Malacostraca. The proximal cornu projects upwards into the cavity of the mouth. The maxillulae (Fig. 189, B) have two endites, the distal 21 322 THE CRUSTACEA one terminating in a strong curved spine. A vestigial palp is present (p). The maxillae (Fig. 189, C) have a peculiar and characteristic form which cannot be closely compared with that of the corre- sponding appendage in other Malacostraca. They appear to consist of four segments, of which the first and second are indistinctly separated. The first five pairs of thoracic appendages are similar in structure and are commonly called maxillipeds, though, as they possess no endites or other adaptations for mastication, the name is hardly appropriate. Each consists of only six segments (there is no evidence to show how these are related to the seven segments commonly recognised in other Malacostraca) and terminates in a m. FIG. 189. Mouth-parts of Squilla mantis. A, mandible, seen from the inner, or oral, side ; B, maxillula ; C, maxilla, i, incisor process ; in, molar process ; o, papilla bearing opening (of maxillary gland ?) ; p, palp. prehensile "hand" or sub-chela; there are no exopodites, but epipodites (Fig. 190, A. ep) are present on all five pairs in the form of discoid membranous plates or vesicles attached to the basal segment by a narrow neck. The first pair of limbs (Fig. 190, A) are long and slender and the terminal segment is minute ; the second pair are very massive, forming powerful weapons (Fig. 186, th2) ; the third, fourth, and fifth pairs resemble each other and are less powerful. In each case the terminal segment is flexed upon the preceding one in such a way that its point is directed forwards, an arrangement which recalls the peculiar inverted chela of the Amphipod Trischizostoma. The last three pairs of thoracic limbs (Fig. 186, ths) are slender, birarnous, and without epipodites. The protopodite is very distinctly composed of three segments, of which the second is elongated. The inner (and anterior) of the two rami is the stouter and consists of two segments ; the outer is slender and unjointed. According to Glaus, the development of THE STOMATOPODA 323 the limb shows the outer branch to be the endopodite and the inner the exopodite, the relative positions of the two being reversed in the course of development. The pleopods (Fig. 190, B, C) are remarkable in carrying the branchial apparatus. The broad and flattened protopodite has articulated with it at some distance from each other the endopodite and exopodite, each of which is lamellar and membranous and is obscurely divided into two segments. From the inner edge of the endopodite springs a short appendix interna (i), bearing a group of coupling-hooks. The branchiae (br) consist of a main stem springing from the anterior face of the exopodite near its base, extending horizontally inwards, and carrying on its lower edge a ex. Fin. li>0. A. tiist thoracic appendage of >'4. Young larva of the Alima-type. a', antennule ; a", antenna ; ai-o5, pleopods ; I, II, first and second thoracic limbs ; 6-8, last three thoracic somites. (After Brooks, from Korschelt and Heider's Embryology.) and the Lebanon, in which latter larvae of the Erichthus-type have also been recognised. Species of Squilla also occur in the London Clay and other Tertiary deposits. AFFINITIES AND CLASSIFICATION. Perhaps the most aberrant character of the Stomatopoda, and one which separates them not only from the other Malacostraca but from all other Crustacea, is the presence of distinct and THE STOMATOPODA 331 movable ophthalmic and antennular " somites." "Whatever be the morphological value of these segments of the head, there can be no doubt that their separation in the Stomatopoda is a secondary and not a primitive character. The movable rostral plate is a character of some interest from its resemblance to that of the Leptostraca ; but it is to be noted that the spiniform rostrum of the larval Stomatopod is not articu- lated, while, on the other hand, the Decapod Rkynekocinetea shows the possibility of the ordinary rostrum becoming divided off by a movable joint from the carapace. The lamellar epipodites of the first five pairs of thoracic limbs recall those of the Syncarida ; the bifurcation of the outer flagellum of the antennules is only paralleled among the Caridean Decapods ; the modification of the first pair of pleopods in the male may be compared with that found in the Euphausiacea and the Penaeid Decapods ; the possession of an appendix interna on the pleopods is shared by the Leptostraca and the lower Eucarida. Other characters, such as the structure of the maxilla and the segmenta- tion of the thoracic limbs, cannot be closely compared with those of any other Malacostraca. It seems most probable that the Stomato- poda are a lateral offshoot from the main stem of the Malacostraca, of which, in the absence of connecting links, it is as yet impossible to determine the exact relations. The existing Stomatopoda form a very homogeneous group, within which only one family can be recognised, while many of the genera are separated by comparatively slight differences. ORDER Stomatopoda, Latreille (1817). Family SQUILLIDAE. Squilla, Fabricius (Fig. 186) ; Lysiosquilla, Dana ; Pseudosquilla, Dana ; Gonodactylus, Latreille ; Coronida, Brooks. LITERATI; UK. 1. Bigelow, R. P. Report upon the . . . Stomatopoda collected by the Steamer "Albatross" . . . Proc. U.S. Nat. Mus. xvii. pp. 489-550, 3 pis., 1895. 2. Brooks, Jl~. K. Report on the Stomatopoda. Rep. Voy. "Challenger," Zool. xvi. pp. 114, 16 pis., 1886. 3. Glaus, 0. Die Metamorphose der Squilliden. Abh. kgl. Ges. Wiss. Gottin- gen, xvi. pp. 111-163, 8 pis., 1871. 4. Die Kreislaufsorgane und Blutbewegung der Stomatopoden. Arb. zool. Inst. Wien, v. pp. 1-14, 3 pis., 1883. 5. Grobben, C. Die Geschlechtsorgane von Squilla mantis, Rond. S.B. Akad. Wiss. Wien, Ixxiv. Abth. i. pp. 389-406, 1 pi., 1876. 6. Hansen, H. J. Isopoden, Cumaceen und Stomatopodtn der Plankton- Expedition. Ergeb. Plankton-Exp. ii. pp. 105, 8 pis., 1895. 7. Jurich, B. Die Stomatopoden der deutschen Tiefsee-Expeditiou. Wiss. Ergeb. deutschen Tiefsee- Expedition . . . "Yaldivia," vii. pp. 361- 408, pis. xxv. -xxx., 1904. 332 THE CRUSTACEA 8. Lister, J. J. Note on a (? Stomatopod) Metanauplius Larva. Quart. Journ. Microsc. Sci. (n.s.) xli. pp. 433-437, 1898. 9. Miers, E. J. On the Squillidae. Ann. Mag. Nat. Hist. (5), v. pp. 1-30 and 108-127, pis. i.-iii., 1880. 10. Miiller, F. Bruchstiick zur Entwickelungsgeschichte der Maulfiisser. Arch. Naturgesch. xxviii. pp. 353-361, pi. xiii., 1862; Ein zweites Bruchstiick . . ., op. cit. xxix. pp. 1-7, pi. i., 1863. 11. Orlandi, S. Sulla struttura dell' intestine della Squilla mantis, Rond. Atti Soc. Ligustica, xii. pp. 112-133, pis. iii.-iv., 1901. INDEX Note. — Page-numbers in thick type refer to the sections on classification. Abdomen, 6 ; of Amphi- poda, 225 ; of Brancbio- poda, 34 ; of Cirripedia, 113 ; of Copepoda, 74 ; of Cumacea, 184 ; of Decapoda, 259 ; of Lso- poda, 198 ; of Lepto- straca, 152 ; of Malaco- straca, 144; of Mysidacea, 172 ; of Stomatopoda, 321 Abdominal appendages of Amphipoda, 232 ; of Leptostraca, 156 ; of Malacostraca, 146 Abdominalia, 108, 139 Acanthephyra, 311 Acantbepliyridae, 311 Acanthocope, 207, 219 Acanthocyclinae, 315 A can thocyclux, 315 Acanthonotozoma, 241 Acanthonotozomatidae, 241 Acantbouychinae, 316 Acanthonyx, 316 Acanthosoma, 297 Acanthotelson, 168 Acartia, 102 Acetes, 273 Achtheres, 80, 83, 100, 103 Acontiophorus, 78, 79, 103 Acrothoracica, 1, 106, 108; definition, 140 ; mor- phology, 108 Adductor muscle of Bran- cbiopoda, 44 ; of Cirri- pedia, 116 ; of Ostracoda, 57, 64 ; of Leptostraca, 157 Adhesive organ of Bran- chiopoda, 43 Aega, 208, 220 Aeger, 307 Aeginae, 220 Aegisthus, 103 Aeglea, 277, 313 Aegleidae, 304, 313 Aesthetascs, 20 ; of Cope- poda, 76, 85 Aetideus, 102 Affinities of Amphipoda, 239 ; of Branchiopoda, 51 ; of Cirripedia, 138 ; of Copepoda, 101 ; of Cumacea, 188 ; of Deca- poda, 309 ; of Euphausi- acea, 251 ; of Isopoda, 218; of Leptostraca, 160; of Mysidacea, 181 ; of Stomatopoda, 330 ; of Syncarida, 168 ; of Tan- aidacea, 194 Ala of Cirripedia, 111 Albunea, 265, 314 Albuneidae, 314 Akippe, 115, 117, 120, 139, 140 Alcippidae, 140 Alcirona, 219 Alcock, 254, 288 Alepas, 110, 128, 140 Alicella, 239, 240 A lima, 326, 328, 330 Alimentary system, 14 ; of Amphipoda, 233 ; of Brauchiopoda, 42 ; of Branchmra, 98 ; of Cirri- pedia, 115; of Copepoda, 82 ; of Cumacea, 187 ; of Decapoda, 279 ; of Euphausiacea, 247 ; of Isopoda, 207 ; of Lepto- straca, 156 ; of Mysi- dacea, 176; of Ostracoda, 63; of Stomatopoda, 323; of Syncarida, 166 ; of Tanai'dacea, 193 Alpheidae, 258, 311 Alpheus, 282, 283, 311 Amblyops, 173, 181 333 Amesopodidae, 220 Amesopous, 203, 220 Ampelisca, 241 Ampeliscidae, 235, 236, 241 Ampharthrandria, 71 ; de- finition, 103 Amphaskandria, 71, 77 ; definition, 102 Ampbilochidae, 241 Amphilochus, 241 Amphion, 311 Ampbioninae, 311 Amphipoda, 2, 143 ; affini- ties and classification, 239 ; definition, 224 ; de- velopment, 237 ; habitsf etc., 238; historical notes, 224 ; morphology, 225 ; palaeontology, 239 Amphithoe, 241 Amphithoidae, 229, 235, 241 Anal respiration in Cope- poda, 82 Anarcturus, 220 Anamixidae, 241 Anamixis, 241 Anaspidacea, 1, 143, 162, 169 Anaspides, 7, 8, 25, 147, 162-168, 169 Anaspididae, 169 Ancestral type of Crus- tacea, 26 ; of Malaco- straca, 144 Anceus, 219 Anchialina, 182 Anchialus, 180, 182 Ancinus, 220 Andasma, 110, 113, 114, 115, 138, 140 Anisopoda, 190 Annulus ventralis, 291 Anomalocera, 84, 102 334 THE CRUSTACEA Anomopoda, 29, 40 ; defi- Leptostraca, 156 ; of Asteropidae, 69 nition, 63 Stomatopoda, 323 Asthenognathiuae, 315 Anomostraca, 162 Appendix masculina, 274 Asthenognathus, 315 Anomura, 253, 309 ; defini- Apseudes, 187, 190, 191, Asymmetrica, 106, 112; tion, 313 193, 194, 195 definition, 140 Anostraca, 1, 29, 39, 48; Apseudidae, 191, 192, 193, Atelecyclidae, 315 definition, 53 194, 195 Atelecyclinae, 315 Antenna, 11 ; of Am phi - Apus, 6, 7, 8, 11, 24, '26, Atelecydus, 315 poda, 228 ; of Branchio- 29, 30, 36, 37-40, 42, Athanas, 265, 311 poda, 35 ; of Cirripedia, 43, 45, 46, 48, 50, 53 Atya, 311 114 ; of Copepoda, 77 ; Arachnomysinae, 181 Atyidae, 266, 268, 282, of Cumacea, 185 ; of Arachnomysis, 173, 181 304, 311 Decapoda, 265 ; of Archaeomysinae, 181 Atylidae, 241 Euphausiacea, 245 ; of Arckaeomysis, 176, 181 Atylus, 241 Isopoda, 198 ; of Lepto- Arcturidae, 197, 204, 212, Auditory setae, 19 straca, 152 ; of Malaco- 220 Audouin, 3, 286, 324 straca, 144 ; of- Mysi- Arcturus, 220 Aurivillius, 107 dacea, 173 ; of Ostra- Argathona, 220 Autotomy, in Amphipoda, coda, 59 ; of Stomato- Argathoninae, 220 231 ; in Decapoda, 273 poda, 321 ; ofSyncarida, Argulidae, 104 Axiidae, 313 164 ; of Tanaidacea, Argidus, 72, 95-97, 99, Axius, 313 191 100, 104 Antenna! gland, 16 ; of Arietellus, 102 Baird, 30 Amphipoda, 235 ; of Aristeiuae, 278, 304, 311 Bairdia, 64, 69 Decapoda, 285 ; of Eu- Aristeus, 288, 289, 311 Bairdiidae, 62, 63, 69 phausiacea, 247 ; of Aristotle, 71,106,254, 319 Balanidae, 115, 140 Leptostraca, 156 ; of Annadillidiiiae, 220 Balauiuae, 111 1 Mysidacea, 178 Armadillidium, 220 Balanus, 107, 108, 111, Antenuular fossa of Bra- Armadillo, 220 112, 122, 138,, 140 chyura, 257 Artemia, 43, 50, 53 Barnacle goose, 106 Antennule, 10 ; of Amphi- Arthrobranchia, 14, 275 Barybrotes, 220 poda, 226 ; of Apoda, Arthrostraca, 147 Barybrotinae, 220 129 ; of Branchiopoda, Artotrogus, 82, 103 Basipodite, 7 35 ; of Branchiura, 95; Ascidicolidae, 75, 78, 81, Basis, 7 of Cirripedia, 113 ; of 89, 103 Bate, C. Speuce, 183, 190, Copepoda, 75 ; of Cuni- Ascidicola, 103 225, 254 acea, 185 ; of Decapoda, Asconiscidae, 221 Batta, 241 264 ; of Euphausiacea, Asconiscus, 221 Bateidae, 241 244 ; of Isopoda, 198 ; Ascothoracica, 1, 106, 107 ; Bathynella, 162, 169 of Leptostraca, 152 ; definition, 140 ; develop- Bathynomus,198, 199, 204, of Malacostraca, 144 ; ment, 127 ; morphology, 211, 217, 218, 219 of Mysidacea, 173 ; of 125 " Bauchwirbel " of Cope- Ostracoda, 58 ; of Stoma- Asellidae, 205, 206, 216, poda, 81 topoda, 321 ; of Syn- 219 Belisarius, 83, 103 carida, 164 ; of Tanai- Asellota, 196 ; definition, Belon, 254 dacea, 191 219 Beueden, van, 72, 190 Anthura, 211, 219 Asellus, 198, 199, 204, 208, Benthesicymus, 311 Anthuridae, 197, 198, 210, 211, 213, 219 Bentheuphausia, 245, 246, 205, 216, 219 Astacidae, 257, 277, 304, 248, 251, 252 Anuropodinae, 219 312 Beutheuphausiiuae, 252 Anwropus, 207, 219 Astacilla, 220 Bernard, 5, 47 AOra, 231, 241 Astacillidae, 220 Beyrichia, 68 Aoridae, 235, 241 Astacura, 253 ; definition, Birgus, 261, 262, 284, 303, Apoda, 1, 106 ; definition, 312 304, 313 140 ; morphology, 128 Astacus, 264, 278, 282, Boas, 148, 171, 244, 254, Apodemes in Decapoda, 283, 285, 291, 312 256, 259, 268, 274, 309 263 Asterocheres, 103 Bodotria, 188 Apodidae, 53 Asterocheridae, 76, 78, 82, Bodotriidae, 188 Appendix interna, 147 ; of 84, 103 Boeck, 225 Decapoda, 273 ; of Eu- Asterope, 59, 62, 64, 66, Bomolochidae, 103 phausiacea, 247 ; of 67, 69 Bomolochus, 103 INDEX 335 Bonnier, J., 197, 214, 215, 279 Bopyridae, 208, 221 Bopyrina, 212 ; definition, 221 Bopyroid stage of Isopoda, 215 Bopyrus, 221 Boreomysinae, 182 Boreomysis, 173, 175, 176, 178, 180, 182 Bornulaile, 254, 277, 303, 309 livxniina, 53 Bosminidae, 53 Bouvier, 254, 255, 256, 276, 283, 307, 308 Bovallius, 196, 225 Brachygnatha, 253 ; defini- tion, 314 Brachyrhyncha, 253 ; de- finition, 315 Brachyura, 253 ; definition, 314 Bract of Branchiopoda, 37, 39 Brady, 56, 68, 72 Brain, 16 Branchiae, 14 ; of Amphi- poda, 231 ; of Cirripedia, 115 ; of Cumacea, 184, 186 ; of Decapoda, 275 ; of Euphausiacea, 246 ; of Isopoda, 204 ; of Mysi- dacea, 175 ; of Ostra- coda, 61, 64; of Sto- inatopoda, 323 ; of Syn- carida, 165 ; of Tanai- dacea, 191 Branchial formulae, 280 Branchial glands, 286 Branchinecta, 17, 30, 38, 53 Branchiopoda, 1 ; affinities and classification, 51 ; definition, 29 ; develop- ment, 48 ; habits, etc., 50 ; historical notes, 29 ; morphology, 31 ; palae- ontology, 50 Branchiostegite of Deca- poda, 255 Branchipodidae, 53 Branchipodites, 50 Branchipus, 10, 15, 18, 24, 29, 36, 40, 43, 45, 46, 53 Branchiura, 1, 71, 100, 101 ; definition, 104 ; development, 99 ; mor- . phology, 95 Branchuropus, 207, 219 Bresilia, 311 Bresiliidae, 311 Bri.sson, 3 Brooks, 292, 295, 319 Bruntz, 116 Brush-like appendage of Ostracoda, 62 Buccal frame of Brachyura, 257 Budde-Lund, 196 Bullar, J.F., 197 Bunneister, 91, 107, 147 Bythotrephes, 54 (.'iihii-Hjiti, 221 Cabiropsidae, 221 Calanidae, 102 Calanus, 8, 12, 13, 73, 74, 88, 89, 102 Calappa, 314 Calappidae, 314 Calappinae, 314 Calceoli of Amphipoda, 237 Caligidae, 76, 91, 103 Calif/us, 72, 83, 103 Callianassa, 256, 283, 301, 307, 313 Callianassidae, 313 Callianidea, 274, 313 Calliaxis, 301 Calliopiidae, 241 Calliojnus, 241 Calocalanus, 75, 102 Calocaris, 301, 313 Calyptomera, 29 ; defini- tion, 53 Calyptopis-etagft of Eu- phausiacea, 249 Cambarus, 291, 312 Campecopea, 220 Campylaspis, 185, 186, 188 Cancer, 3, 254, 315 Cancerilla, 82, 103 Cancridae, 315 Cancrinae, 315 Cancrion, 215, 221 Candacia, 102 Candaciidae, 102 Canthocamptus, 103 Canu, 86, 101 Caphyra, 315 Caphyrinae, 315 Capitulum of Cirripedia, 109 Caprella, 224, 227, 233, 242 Caprellidae, 226, 231, 232, 233, 236, 242 ^Caprellidea, 224 ; defini- tion, -242 Carapace, 6 ; of Branchio- poda, 31 ; of Branchiura, 95 ; of Cumacea, 184 ; of Decapoda, 255 ; of Euphausiacea, 242 ; of Leptostraca, 151 ; of Malacostraca, 144 ; of Mysidacea, 171 ; of Os- tracoda, 56 ; of Sto- matopoda, 320 ; of Tanai- dacea, 191 Carcinides, 315 Carciniuae, 315 Carcinoplacinae, 315 Carcinoplax, 315 Carcinus, 303, 305, 315 Cardiso-ma, 284, 315 Caridea, 253 ; definition, 311 Caridina. 266, 277, 311 Caridoid facies of Malaco- straca, 144 Carina of Cirripedia, 110 Carino - lateral plates of Cirripedia, 111 Carpenter, 218 Carpiliinae, 315 Carpilius, 258, 315 Carpopodite, 146 Carpus, 146 Carupa, 315 Carupinae, 315 Cassidina, 204, 220 Castration, parasitic, 292 Catometopa, 290, 310 Catophragmus, 111, 140 Catoptrinae, 315 Catoptrus, 315 Caudal furca, 7, 35 (see Furca) Caullery, M., 214 Ciivolini, 254 Cement glands, 20 ; of Cirripedia, 116 Centre/pages, 77, 102 Centropagidae, 83, 99, 102 Cephalic flexure of Deca- poda, 257 Cephalogaster, 208 Cerataspinae, 278, 310 Cerataspis, 279, 280, 310 Ceratiocaris, 159, 160 Oeratocuma, 188 Ceratocumidae, 188 Ceratothoa, 220 Cercops, 232, 240, 242 Cervical groove of Deca- poda, 256 ; of Syncand;i, 163 ; of Mysidacea, 171 Cervinia, 103 Oetochilus, 88 336 THE CRUSTACEA Chaetilia, 220 of Ostracoda, 68 ; of Corycaeus, 85, 103 Chaetiliidae, 220 Stomatopoda, 330 ; of Corystes, 315 Chalimits, 91 Syncarida, 168 ; of Tan- Corystidae, 279, 315 Chelipeds, 271 aidacea, 194 Coutiere, 146, 267, 277, OMura, 241 Claus, 3, 7, 10, 30, 36, 37, 283, 288, 291, 309 Cheluridae, 241 52, 56, 67, 72, 76, 77, Coxa, 7 Cfiiridotea, 199, 202, 208, 79, 85, 88, 98, 101, 107, Coxal glands of Amphi- 210, 220 146, 147, 151, 190, 191, poda, 235 Chirocephalus, 36, 39, 53 194, 213, 225, 240, 244, Coxal plates of Isopoda, Chirostylus, 313 249, 254, 275, 292, 319, 198, 202 ; of Amphi- Chondracanthidae, 82, 92, 322, 324 poda, 226, 230 103 Clausia, 103 Coxopodite, 7 Chondr acanthus, 86, 103 Clausidiidae, 103 Crangon, 276, 287, 311 Choniostoma, 103 Clausidium, 103 Crangonidae, 266, 268, 273, Choniostomatidae, 86, 93, Clausiidae, 103 311 103 Clibanarius, 313 Crangopsis, 181 Chonopeltis, 104 Clistosaccus, 130, 131, 141 Cressa, 241 Chthamalidae, 140 Clypeoniscus, 221 Cressidae, 241 Chthamalinae, 111 Clytemnestra, 103 Crinoniscidae, 221 Chthamalus, 112, 140 Coelom, 16 Crinoniscus, 221 Chun, 19, 244 Coenobita, 284, 288, 303, Crista dentata, 269 Chydorus, 53 304, 313 Cruregens, 219 Circulatory system, 15 ; of Coeuobitidae, 259, 261, 313 Crustacea, definition, 2 Amphipoda, 234 ; of Colomastigidae, 241 Cryptoniscan stage of Iso- Brauchiopoda, 43 ; of Colomastix, 241 poda, 214 Branchiura, 98 ; of Cirri- Conmiensalism of Decapoda, Cryptoniscina, 212 ; defini- pedia, 115; of Copepoda, 304 tion, 221 82 ; of Cumacea, 187 ; Compartments of Cirri- Cryptophialidae, 140 of Decapoda, 283 ; of pedia, 111 Cryptophialus, 107, 114, Euphausiacea, 247 ; of Complemental males, 107, 115, 139, 140 Isopoda, 208 ; of Lepto- 118, 132 Crystalline body, 18 straca, 156 ; of Mysi- Conchodenna, 115, 116, Ctenopoda, 29, 40 ; defini- dacea, 177 ; of Ostracoda, 117, 140 tion, 53 64 ; of Stomatopoda, 32 ; Conchoecia, 58, 60, 69 Cubaris, 220 of Syncarida, 166 ; of Conchoeciidae, 69 Cuma, 183, 188 Tanaidacea, 193 Conchostraca, 1, 29, 39, Cumacea, 2 ; affinities and Cirolana, 198, 199, 219 49 ; definition, 53 classification, 188 ; defi- Cirolaninae, 198, 216, 219 Conilera, 210, 219 nition, 183 ; develop- Cirrhipedes, 107 Copepoda, 1 ; affinities and ment, 187 ; habits, etc., Cirri of Cirripedia, 114 classification, 101 ; de- 187 ; historical notes, Cirripedia, 1 ; affinities and finition, 71 ; habits, etc., 183 ; morphology, 183, classification, 138; defini- 99 ; historical, 71 187 tion, 106 ; development, Copepodid stages of Eu- Cuvier, 3, 107 121 ; habits, etc., 137 ; copepoda, 88 Cyamidae, 226, 230, 231, historical notes, 106 ; Copilia, 103 233, 242 palaeontology, 138 Copulatory appendages of Cyamus, 224, 230, 231, Cladocera, 1, 29, 40, 49 ; Branchiura, 97 ; of Cope- 232, 242 definition, 53 poda, 81 ; of Decapoda, Cyathura, 211, 219 Cladocopa, 1, 56 ; defini- 274 ; of Euphausiacea, Cyclaspis, 188 tion, 69 247 Cyclestheria, 48, 50, 53 Claspers, 11 Corallana, 219 . Cydodorippe, 314 Classification of Amphi- Corallaninae, 219 Cyclometopa, 310 poda, 239 ; of Branchio- Cornea, 18 Cyclopidae, 76, 79, 83, poda, 52 ; of Cirripedia, Corneagen cells, 18 100, 103 138 ; of Copepoda, 101 ; Ooronida, 326, 331 Cyclops, 11,71, 72, 73, 76, of Cumacea, 188; of De- Coroniderichtkus, 326 78, 79, 84, 86, 103 capoda, 309 ; of Euphau- Coronula, 112, 140 Cyclops-stages of Eucope- siacea, 251 ; of Isopoda, Corophiidae, 229, 235, 241 poda, 88 218 ; of Leptostraca, Corophium, 234, 241 Cyclosphaeroma, 218 160 ; of Malacostraca, Corycaeidae, 82, 83, 84, Cyiindroleberis, 69 147 ; of Mysidacea, 181 ; 100, 103 Cyllopodidae, 241 INDEX 337 s, 241 Cymodoce, 220 f'l/HKif,,,/,',!, 315 I 'l/, /in//,, HI. 220 Cymothoidae, I',".'. 208, 204, 208, 209. 216, 219 < 'yinothoinae. 212. 213, '220 < 'yprididae. 57. 00, 62, ' (>:i. 04, 07, 69 Oypridina, 58, 59, til. 02. 03. 05. OS, 69 Cypridiuidae, 57, 59, 61, '(14, 05, 66, 67. 69 ( typridopsidae, 69 <'ii)>i-iil<>iixix, 69 , 18, 57, 67, 69 rv^/v's-stagf, of Ascotho- racica, 127 ; of Cirri- p.-dos, 24, 107, 121 ; of Rliizocephala,133 < lyproniscidae, 221 Cyproniscus, 221 Cyr-'reUa, 58, 59, 60, 62, 69 Cytherellidae, 07, 69 Cytheridae, 57, 58, 61 62, 66, 67, 69 Czeruiav.sky, 181 Dactylerythrops, 172. 181 Dartylopodite, 146 Daetylus, 146 Dajidae, 221 D'lfn.i, 221 Dana, 72. 190. 224, 244, 249 Itiniiilin, 221 I >«/>/> via, 20, 29, 33, 40, 53 Duphniidae, 53 Dart of Rhizocephala, 135 Darwin, 107, 108, 111, 115, 116, 117, 118, 123, 128, 129, 139 Darwinvla, 58, 67, 69 Danvinulidae, 02, 69 Decapoda, 2, 143 ; affinities and classification, 309 ; definition, 253 ; develop- ment. 292 : habits, etc., 303 ; historical notes, 254 ; morphology, 255 ; palaeontology, 300 l)ei-kenin, 315 Deckeniinae, 315 De Geer, 72 Delage, 107, 132, 133, 135, 197. 22-, Delia Valle, A., 225 Dendrobraoohiate, 277 Ill-nil ,-nffaster, 125, 126, 127/128, 140 1 tendrogastridat), 140 Dermal glands, 20 (see Glands) Desmosoi/ia, 219 Deuterocerebrum, 17 Development of Amphi- poda, 237 ; of Ascotho- racica, 127 ; of Brauchio- poda, 48 ; of Branchiura, (.".i : of Cirripedia, 121 ; of Cumacea, 187 ; of Decapoda, 292 ; of Encopepoda, 87 ; of Eupliausiacea, 249 ; of Isopoda, 213 ; of Lepto- straca, 158 ; of Ostra- coda, 67 ; of Mysidacea, 179 ; of Rhizocephala, 133 ; of Stomatopoda, 326 ; of Tanaidacea, 194 Dexamine, 241 Dexaminidae, 241 /,/// / :• t\~ ~i no l mi t.i i \ xu^ DiaptomMt, 102 Diastylidae, 188 Diastylis, 184, 185, 186, 188 Diastyloides, 185, 188 Dichelaspis, 110, 140 Dichelestiidae, 83, 103 Diclielestn/m, 103 Digestive gland. 15 Dimorphism, sexual, 21 ; in Decapoda, 291 ; in Tanaidacea, 194 /liiixtici.'nn, 103 Diporodelphya, 86 Doflein, 19 Dohrn, 123, 183 Dolops, 95, 104 Dnii/,/:,', 314 Dorippidae, 290. 314 /^/••vw/"s i:>. 103 Dorsal organ of Amphipoda, 237 ; of Branchiopoda, 43 ; of Isopoda, 213 ; of Mysidacea, 180 ; of Syn- carida, 164 ; of Tanai- dacea, 194 Drama, 283, 314 Dromiacea, 253 ; definition, 314 Dromiidae, 256, 289, 290, 308, 314 Dromiidea, 253 ; definition, 314 Dromioprit, 308 DuplarUs, 130, 131, 133, 141 Duveruoy, 324 Dwarf males of Cirripedia. 117 Dynamene, 220 Dynomene, 314 Dynouienidae, 308, 314 Ebalia, 314 Kctinosoma, 103 Edriophthalma, 3, 147 Edwards (see Milne- Edwards) Eiconaxius, 277, 305, 313 Ekman, 36 Elaphocuris, 297 Elminius, 111, 140 Embryology, 22 Endites, 37 Endophragmal system, 263 Endopodite, 7 Endoskeleton of Branchio- poda, 44 Endostome of Brachyura, 257 End-sac, 16 Enterognathus, 90, 95, 103 Entomostraca, 3, 27 Entoniscidae, 208, 217, 221 Entoniscus, 221 Entosteruite of Branchio- poda, 44 ; of Decapoda, 263 Ephippium of Cladocera, 48 Epicarid stage of Isopoda, 214 Epicaridea, 196 ; defini- tion, 220 Epimeral suture, 256 Epimeron, 4 (footnote) Epipodite, 8 ; of Cumacea, 186 ; of Decapoda, 275 : of Eupliausiacea, 246 ; of Isopoda, 199 ; of Lepto- straca, 154 ; of Malaco- straca, 146 ; of Mysi- dacea, 175 ; of Sto- matopoda, 322 ; of Syu- carida, 165 ; of Tanai- dacea, 192 Epistome of Decapoda, 257 Ergasilidae, 103 Ei'H'isiliis, 103 Erichthina, 296 Erichthus, 326, 330 THE CRUSTACEA Eriphia, 315 of Decapoda, 285 ; of Furca, 7 ; of Branchio- Eriphiiuae, 315 Euphausiacea, 247 ; of poda, 35 ; of Cirripedia, Eryma, 307 Isopoda, 210 ; of Lepto- 113 ; of Copepoda, 75 ; Eryon, 307, 312 straca, 156 ; of Mysi- of Leptostraca, 152 ; Eryonidae, 288, 312 dacea, 178 ; of Ostracoda, of Ostracoda, 58 Eryonidea, 253, 304 ; de- 64 ; of Stomatopoda, Fi'i'cil id-stage of Euphau- finition, 312 325 ; of Syncarida, 166 ; siacea, 249 Erytkrops, 181 of Tanaidacea, 1 93 Exlheria, 16, 32, 38, 40, Exites, 37 Galathea, 313 49, 50, 53 Exopodite, 7 Galatheidae, 313 Ethusa, 314 Exoskeletou, 4 Galatheidea, 253 ; defini- Etisinae, 315 Eyes, 17 ; of Amphipoda, tion, 313 Etisus, 315 236 ; of Branchiopoda, Galatheinae, 313 Jtucalanvs, 79, 82, 83, 85, 46 ; of Branchiura, 98 ; Gammaridae, 231, 235, 241 102 of Cirripedia, 117 ; of Gammaridea, 224 ; defini- Eucarida, 2, 143; defini- Cumacea, 184 ; of Deca- tion, 240 tion, 149 poda, 287 ; of Eucope- Ganimarina, 224 Etichaeta, 102 poda, 84 ; of Euphausi- <.<'idacea, 174 Harpacticus, 103 ', 314 Gnathostomata, 101 Hatschek, 27 Homolidae, 256, 308, 314 Gonads, -21 Haustoriidae, 241 Homolidea, 253 ; defini- Gonerichthus, 326, 328 Haitstorius, 241 tion, 314 Gonodactylus, 321, 328, Head-region of Crustacea, 4 Homolodromia, 314 331 Heart, 15 (see Circulatory Homolodromiidae, 278, Gonoplacidae, 315 system ) 304, 307, 314 Gonoplacinae, 315 Heider, 22 Homolopsis, 308 Gonoplax, 315 Helleria, 220 Hoplocarida, 2, 143 ; de- Goodsir, H., 183 Hemioniscidae, 211, 221 finition, 149 (see Sto- Grapsidae, 304, 315 Hemioniscus, 208, 212, 221 matopoda) Grapsinae, 315 Henderson, 254 Hoploparia, 307 Grapsus, 315 Hepatic caeca (see Ali- Hoplophoridae, 268, 288, Green gland, 16 mentary system) 304, 311 Grobben, 84, 147, 162, Hepato-pancreas, 15 II n />l< Chorus, 311 319 Herbst, J. F. W., 254 Huxley, 42, 148, 180, Groom, 107 Hermaphroditism, 21 ; in 255, 256, 319 Gruvel, 107 Cirripedia, 117 ; in Deca- Hyalella, 239, 241 Gurnev, R., 211 poda, 292 ; in, Isopoda, Hyas, 316 Gyrje, 203, 221 212 Hymenocaris, 159 Gyniiioniera, 29, 40 ; de- Herpyllobiidae, 82, 93, 104 Iliinicnocera, 265, 311 finition, 54 Herpyllobius, 104 Hymenocerinae, 311 Gymnoplea, 71; definition, Heterarthrandria, 71, 76 ; ffymenosoma, 316 102 definition, 102 Hymenosomidae, 316 Heterocarjtus, 259, 288, Hyperia, 230, 241 Haau, W. cle, 254 311 Hyperiidae, 241 Habits, etc., of Amphipoda. Heterocope, 86, 102 Hyperiidea, 224 ; defini 238 ; of Brancbiopoda, Heteromysinae, 181 tion, 241 50 ; of Cirripedia, 137 ; Heteromysis, 175, 181 Hyperina, 224 of Copepoda, 99 ; of Heterorhabdus, 102 Hyperiopsidae, 241 Cumacea, 187 ; of Deca- Heterotunais, 193, 195 Hyperiopsis, 241 poda, 303 ; of Euphau- Hexapodiuae, 315 Hypol>olyte, 266, 287, 311 Iconaxiopsis, 260, 313 Tanaidacea, 194 Hippolytidae, 311 Idotea, 202, 205, 209, Haemocera, 93, 103 Historical notes, 2 ; on 220 Haemocoel, 15 Amphipoda, 224 ; on Idoteidae, 202, 208, 220 Halocvpridae, 57, 60, 61, Branchiopoda, 29 ; on Idya, 103 62, 66, 67, 69 Cirripedia, 106 ; on Ilia, 314 Halocypris, 63, 69 Copepoda, 71 ; on Cu- Iliinae, 314 Hansen, 7, 72, 78, 79, macea, 183 ; on Deca- flyocryptus, 53 107, 129, 130, 145, 146, poda, 254 ; on Euphau- Inachinae, 316 148, 152, 154, 156, 171, siacea, 244 ; on Isopoda, Inachus, 316 173, 174, 183, 191,197, 196 ; on Leptostraca, Ingolfidla, 225, 226, 231, 198, 218, 225, 229, 240, 151 ; on Mysidacea, 233, 240, 242 266, 276, 288, 319, 326, 171 ; on Ostracoda, 56 ; Ingolfiellidae, 242 328 on Stomatopoda.- 319 ; Ingolfiellidea, 224 ; defi- Hansenomysis, 178, 181 on Syncarida, 162 ; on nition, 242 Hapalocarcinidae, 305, 315 Tanaidacea, 190 Ingolfiellina, 225 Hapalocarcinus, 315 Hoek, 107, 116 Isaea, 241 Hapaloyaster, 313 Holognathidae, 220 Isaeidae, 241 Hapalogastrinae, 313 Holognathus, 220 Ischiopodite, 146 Haplophthalmus, 220 Holopediidau, 53 Ischium, 146 Haplopoda, 29 ; defini- Holopedimn, 53 Isokerandria, 71, 102 ; de- tion, 54 Holt, 178, 181, 248 finition, 103 22 1 340 THE CRUSTACEA Isopoda, 2, 143 ; aber- Latreille, 3, 27, 29, 30, 52, Lilljeborgiidae, 241 rantia, 190 ; affinities 147, 171, 190, 196, 224, Limbs, general morphology, and classification, 218 ; 254, 309, 319 7 definition, 196 ; develop- Latrei/lid, 314 Limnadia, 44, 53 ment, 213 : habits, etc., Latreilliidae, 314 Limnadiidae, 53 216 ; historical notes, Latreutes, 311 Limnetis, 35, 40, 49, 53 196 ; morphology, 198 ; Laura, 107, 125, 126, Limnocaridina, '277, 311 palaeontology, 218 127, 140 Limnoria, 220 Lauridae, 140 Limnoriinae, 220 Jaera, 193, 202, 213, 219 Janira, 203, 219 Leach, 3, 147, 151, 190, 196 Leander, 311 Linea anonuirica, 256 ; droiniidica, 256 : homo- lica, 256 ; thalassinica, Jassidae, 235, 241 Jaxea, 301, 313 Jordan H. 167 Leeuwenhoek, 71, 72 Lepadidae, 140 Lepas, 3, 107, 108, 110, 256 Linnaeus, 2, 72 •« Liiiitparus, 307, 312 •Jurine 30 72 114, 116, 117. 123, 124 Lipomerism, 4 125, 138, 140 Liriopsidae, 211, 217, 221 Lepechin, 183 Liriopsis, 221 Kaempferia, 306 Leperditia, 68 Lister, J. J., 328 Kentrogon-stage of Rhizo- Lepidurus, 9, 12, 31, 53 Lithodes, 313 cephala, 135 Leptochelia, 193, 194, 195 Lithodidae, 261, 273, 275, Kishinouye, 292 Leptodora, 33, 34. 36, 40, 313 Knipowitsch, 107 42, 43, 45, 49, 50, 54 Lithodinae, 313 Kochlorine, 140 Leptomysiuae, 181 Lithotrya, 138, 140 Kochlorinidae, 140 Leptomysis, 181 Liver, 15 (see Alimentary Koehler, 117 Leptostraca, 1, 143 ; affini- system) Koleolepas, 139, 140 ties and classification, Longipedia, 103 Koontinga, 25, 147, 162- ] 60 ; definition, 148, Lophogaster, 181 167, 169 151 ; development, 158 ; Lophogastridae, 173, 174, Koonungidae, 169 habits, etc., 159 ; his- 175, 176, 181 Korschelt, 22 torical notes, 151 ; Loriatla, 109, 110, 138 Kowalevsky, 325 morphology, 151 ; palae- Lower lip, 7 ; of Amphi- Krithe, 64, 69 ontology, 159 poda, 229 ; of Branchio- Krohn, 107 Lernaeu, * 3, 72, 91, 92, poda, 36 ; of Copepoda, Kroyer, H., 72, 183, 224 103 78 ; of Ostracoda, 59 ; Lernaeidae, 91, 103 of Tanaidacea, 191 Labrum, 7 ; of Cirripedia, Ler-naeocera, 81, 100, 103 Lncicutia, 102 114 ; of Copepoda, 78 ; Lernaeodiscus, 131, 141 Lucifer, 311 of Ostracoda, 59 Lernaeopoda, 103 Luminous organs, 21 ; of Lacaze-Duthiers, 107 Lernaeopodidae, 80, 90, 92, Copepoda, 83 ; of Deca- Laemodipoda. 224 103 poda, 288; of Euphau- Lafystiidae, 241 Lernanthrojms, 15, 83, 103 siacea, 248 ; of Mysi- Lafystius, 241 Leucifer, 273, 275, 287- dacea, 174; of Ostracoda, Lagena of Rhizocephala, 291, 295, 296, 298, 311 64 131 Leuciferinae, 311 Liitken, 72 Lamarck, 3, 72, 107 Leucon, 188 Lynceidae, 42, 44, 53 Lambrus, 316 Leuconidae, 185, 188 Lynceus, 53 Lampropidae, 188 Leucosia, 314 Lyncodaphniidae, 42, 53 Lamprops, 185, 188 Leucosiidae, 279, 314 Lysianassa, 240 Lanceola, 241 Leucosiinae, 314 Lysianassidae, 237, 240 Lanceolidae, 241 Leucothoe, 231, 241 Lysioerichthus, 326 Lankester, 4, 9, 30, 40, Leucothoidae, 241 Lyaiosquilla, 326, 331 42, 43, 45 Leydig, 30, 46 Laomedia, 313 Lichomolgidae, 103 Afacrocheira, 257, 258, 306, Laomediidae, 313 Lichomolgus, 103 316 Laophonte, 103 Liffia, 208, 209, 213, 220 Macrocypris, 60, 61, 63 Laphystiopsidae, 241 Ligidium, 210, 220 Macromysis, 181 Laphystiopsis, 241 Ligiidae, 205, 206, 212, Macrophthalminae, 315 Lateral plates of Cirripedia, 220 Macrophthalmus, 263, 264, 110 Lill.jeborg, 31, 107, 183 315 Latona, 36, 53 Lilljeborgia, 241 Macrqpodia, 316 INDEX 34i Macrothrix, 53 ('umucea, 186; of Deca- Mimonectes, 226, 241 Macrura, 309, 310 poda, 268 ; of Isopoda, Mimonectidae, 241 Maia, 316 198 ; of Malacostraca, Mirncia, 84, 103 Maiidae, 316 146 ; of Ostracoda, 61 ; Misophria, 82, 103 Maiinae, 316 of Tanaidacea, 192 Misophriidae, 103 Malacostraca, 1, 3 ; classi- Maxillula, 12 ; of Amphi- Mithrax, 316 fication, 147 ; definition, poda, 229 ; of Branchio- Moina, 53 143 ; morphology, 144 poda, 36 ; of Branchiura, Monoculus, 3, 72 Malaquin, 93 95 ; of Cirripedia, 114 ; MniinUstni, 220 Males of Cirri pedia, 118 of Copepoda, 78 : of Monoporodelphya, 86 Alamaia, 316 Cumacea, 185 ; of Deca- Monospilus, 33, 53 Mandible, 12 ; of Amphi- poda, 266; of Euphau- Monstrilla, 103 poda, 229 ; of Branchio- siacea, 245 ; of Isopoda, Moustrillidae, 93, 103 poda, 36 ; of Branchiura, 198 ; of Leptostraca, Montagu, 183, 190 95 ; of Cirripedia, 114 ; 153 ; of Malacostraca, Miiller, F., 107, 132, 190, of Copepoda, 78 ; of Cu- 145 ; of Mysidacea, 173 ; 193, 194, 254, 292, 319 niacea, 185 ; of Decapoda, of Ostracoda, 61 ; of Miiller, G. W., 56, 59, 61, 266 ; of Enphausiacea, Stomatopoda, 321 ; of 62, 67, 68 245 ; of Isopoda, 198 ; Syncarida, 164 : of Miiller, 0. P., 29, 56, 72 of Leptostraca, 152 ; of Tanaidacea, 191 Manilla, 260, 313 Malacostraca, 145 ; of Mayer, P., 225 Muuidopsinae, 313 Mysidacea, 173 ; of Ostra- Median eye, 17 Munidopsis, 313 coda, 59 ; of Stomato- Megalopa, 303 Munna, 211, 219 poda, 321 ; of Syncarida, Megan i/ctiphanes, 245, 246, Miniiiopsis, 203, 219 164 ; of Tanaidacea, 191 247, 252 Muscular system of Bran- Mantle, 6 ; of Cirripedia, Meinert, F., 196 chiopoda, 44 ; of Cirri- 108 ; of Rhizocephala, Mi'iiii'i-tioda, 329 ; of Syncarida, Stilomysis, 181 >V/,n/, 241 167; of Tauaidaeea, T.'i Stomach, 14 Sciuidae, 241 Slabber, 107, 254 Stomatopoda, 2, 143, 319 ; Seolecithrix, 102 Sluiter, 137 affinities and classifica- Scottocheres, 8(5, 103 Smith, G., 132, 107, 136, tion, 330 ; development, Scidda, 330 165, 170, 194, 292 326 ; historical notes, Scutum of Cirripedia, 110 Socarnes, 229, 241 319 ; habits, etc., 329 ; Scyllaridae, 255, 266, 312 Solenocera, 265, 311 morphology, 319 ; palae- Scyllaridca. 253 : defini- Somite, 4 ontology, 330 tion, 312 Spence Bate (see Bate) Stomodaeum, 14 Xcyllarux, 290, 300, 312 Spermatheca of Eueope- Stridulating organs, 305 Scypliacinae, 220 poda, 86 (see Recepta- Stylocerite, 265 Scyphax, 220 culuni seniinis) StylocheiroH, 247, 248, 249, Segmentation of egg, 22 Siiermatophores. 21 ; of 252 Sella turcica, 263 Eueopepoda, 86 ; of De- Stylodactylidae, 268, 311 Sense-organs, 91 ; of Am- capoda, -290 Stylodactylus, '271, 311 phipoda, 236 ; of Bran- Spcrmatoxoa of Cirripedia, Stylopodite, 146 chiopoda, 46 ; of Bran- 118; of Decapoda, 290; Sub-apical lobe of Branchio- chiura, 98 ; of Cirripedia, of Ostracoda, 67 poda, 39 117 ; of Decapoda, 2S7 ; X/ili,ie>-oma, 210, 213, 220 Subhyperiiii, 225 of Eucopepoda, 84 ; of Sphaerornidae, 199, 203, Suctoria, 107 Isopoda, 211 ; of Mysi- 204, 205, 207, 209, 212, Summer eggs of Branchio- daei-a, 178 ; of Ostra- 216, 220 poda, 49 coda, 66 ; of Stomato- Sphaeroininae, 220 Swain, 218 poda, 325 ; of Syncarida, Sphaerathylacus, 130, 137 Swammerdam, 29, 254 166 Sphyrapus, 191, 195 Sylon, 130, 131, 133, 141 Sergestes, 287, 288, 297, Spinning -organ of Ostra- Symmetrica, 106 ; di-tini- 299, 311 coda, 59 tion, 140 Serge.stidae, 268, 273, 288, Spirontocaris, 311 Sympodite, 7 295, 304, 311 Spiropagums, 289, 313 Synagoga, 125, 126, 127, Srr.rotinae, 311 Spongicola, 305, 311 140 Serolidae, 197, 203, 204, Squama (see Scale) Syuagogidae, 140 220 .sv/// tfta, 319-323, 328. 330, Sj-ncarida, 1, 143 ; affini- Serulis, 199, 208, 220 331 ties and classification, Sesarmu, 315 Squillacea, 148 168; definition, 148, Si^anninae, 315 Squillidae, 331 162 ; habits, etc., 167 ; Setae, 19 Statocyst, 20 ; of Amphi- historical notes, 162 ; Setdla, 103 poda, 237 ; of Dei-apoda, morphology, 162 Setobranchia, 277 287 ; of Eueopepoda, 85 ; Xi/inijiia, 236, 241 Sexual diniorphisni, 21 of Isopoda, 211 ; of Mysi- Syuopidea, 225 Shell-fold, 6 dacea, 178 ; of Syncarida, Synopiidae, 241 Shell-gland, 16 166 ; of Tanaidacea, 193 'Synurella, 232, 233, 234, Sieyonia, 311 Stebbing, 139, 225, 240 241 Sicyoninae, 311 Steenstrup, 72 Sida, 41, 53 Stegocephalidae, 241 Tachaea, 219 Sididae, 44, 47, 53 Steyocephalus, 241 Talitridae, 231, 235, 239, Si.mocephalus, 44, 53 Stenasellus, 197, 219 241 Simosa, 53 Stenetriidae, 205, 206, r«iifi-x*, 241 Siplionostoinata, 78, 101 219 Tanaidacea, 2, 143 ; affini- Siriella, 175, 177, 182 Stenetritim, 219 ties and classification, Siriellinae, 182 Stenopidae, 311 194; definition, 190; Size of Amphipoda, 239 ; Stenopidea, 253, 309 ; de- development, 194; of Branchiopoda, 50 ; of finition, 311 habits, etc., 194 ; his- Cirripedia, 138; of Cope- Stenopus, 291, 311 torical notes, 190 ; mor- poda, 100 ; of Cumacea, Stcnnthoe, 241 phology, 191 346 THE CRUSTACEA Tanaidae, 191, 192, 193, ' Thoracica.l, 106 ; morpho- Uroptychus, 276, 313 194, 195 logy, 108 ; definition, Urothoe, 241 Tanais, 190, 193, 195 140 Tapetum, 19 Thoracostraca, 147 Tattersall, 178, 181, 248 Thorax, 6 Valvifera, 196 ; definition, Tegastes, 103 Thorell, 101 220 Telson, 4 Thyropus, 241 Varuna, 315 Temora, 102 Thysanopoda, 251, 252 Varuninae, 315 Tergite, 4 Tiron, 236, 241 Vasa deferentia, 21 Tergum of Cirripedia, 110 Tironidae, 241 Vauntompsonia, 188 Terrestrial Decapoda, 284 ; Tortanus, 102 Vauntompsoniidae, 188 Isopoda, 205 Trachelifer, 301 Vejdovsky, F., 162, 169 Thalamita, 315 Trapezia, 315 Verruca, 112, 138, 140 Thalamitinae, 315 Trapeziiuae, 315 Verrucidae, 115, 140 Thalassina, 304, 313 Triangulus, 131, 141 Vibilia, 233, 241 Thalassinidae, 291, 313 Triehobranchiate, 277 Vibiliidae, 241 Thalassinidea, 253 ; defini- Trichodactylinae, 315 Virbius, 311 tion, 313 Trichodactylus, 315 Vireia, 207, 220 Thalassocarinae, 311 Trichoniscidae, 220 Vitellophags, 22 Thalassocaris, 311 Trichoniscus, 220 Thalestris, 103 Trilobita, 3 Thamnocephalidae, 53 Trischizostoma, 236, 239, Walking-legs of Decapoda, Thamnocephalus, 35, 53 241 271 Tharybis, 102 Tritocerebrum, 17 Wall of Cirripedia, 110 Thelphusa, 315 Trunk appendages, 12 Water-fleas, 29 Thelphusidae, 304 Trypetesa, 140 Weismann, 30 Thelycuru, 290 Tryphana, 241 Westwood, 225, 254 Thenus, 312 Tryphanidae, 241 Williamson, 275 Thia, 315 Tubicinella, 138, 140 Winter eggs of Branchio- Thienemann, 211 Turrilepas, 109, 138 poda, 47, 49 Thiinae, 315 Tylidae, 203, 212, 220 Wollebaek, 21, 292 Thompson, J. Vaughan, 3, Tylos, 220 Woodward, H., 168, 180 107, 244, 254 Typton, 305, 311 Thompsonia, 137, 141 Thomson, G. M., 162 Upogebia, 274, 301, 313 Xanthidae, 315 Thoracic appendages of Upogebiiuae, 313 Xauthinae, 315 Amphipoda, 229 ; of Urdu, 218 Xantho, 315 Branchiura, 96 ; of Cu- Uronectes, 167, 168 Xenobalanus, 113, 140 macea, 186 ; of Cirripedia, Uropods, 147 ; of Amphi- Xeuophthalminae, 315 114 ; of Copepoda, 79, poda, 232 ; of Cumacea, Xenophthalmus, 315 81 ; of Decapoda, 268 ; 187 ; of Decapoda, 274 ; Xiphosura, 3 of Euphausiacea, 246 ; of Euphausiacea, 247 ; of Isopoda, 198 ; of of Isopoda. 207 ; of My- Malacostraca, 145 ; of sidacea, 176 ; of Sto- Zaddach, 30 Mysidacea, 174 ; of matopoda, 323 ; of Syn- Zenker, 56, 72, 81, 101 Stomatopoda, 322 ; of carida, 166 ; of Tanai- Zoea, 24, 25, 293, 296. Syncarida, 164 ; of dacea, 193 297, 301, 302 Tanaidacea, 191 Uroptychidae, 313 Zygosiphon, 184, 188 THE EXD Printed by R. & R. CLARK, LIMITED, Edinburgh. THE SENSE OF TOUCH IN MAMMALS AND BIRDS WITH SPECIAL REFERENCE TO THE PAPILLARY RIDGES BY WALTER KIDD, M.D., F.Z.S., Author of : Use Inheritance,' 'Direction of Hair in Animals and .Man,' etc. Demy 8vo, Cloth, containing 174 Illustrations. Price 5s. net. (Post free, price 5s. 4d.) "Tliis is, tor the purposes of exact science, undoubtedly a valuable book. . . . As regards the highest mammal, man. it is well ixjinted out that the sense of touch lias been from the very first of extreme importance, and that 'such use of this sense in man must have contributed greatly to his better equipment for the struggle of his life, and thus in a broad way have been governed by a slow, remorseless process of selection.' The book is eminently one for specialists, but the excellence of the numerous illustrations makes it also interesting to the general reader."— Loniloti Qiiarterlii /,'• u •• Dr. Kidd's book is the most important contribution to the matter since Miss Whipple's paper was published."— The Spevtntor. BY THE SAME AUTHOR USE -INHERITANCE ILLUSTRATED BY THE DIRECTION OF HAIR ON THE BODIES OF ANIMALS Demy 8vo, Paper Covers. Price 2s. 6d. net. (Post free, price 2*-. 3d.) " Such a book as this is calculated, at least, to give pause to the Weismannians. Dr. Kidd has broken Around in a commanding position, and we are anxious to see how his attack can be met." — The Irish Naturalist. •• This is an interesting contribution to thedynamic or I amarckian principles of evolution. . . . The author seems to have made out a good case and to have been led by the legitimate use of the inductive method to what seem to be valid and natural conclusions." — Science. " It is urged that . . . the doctrine that acquired characters are never inherited does not hold good, and hence a Lamarckian explanation of the phenomena must be accepted. The case, as argued by Dr. Kidd, appears to be a strong one, and it will be curious to not* what the Weisman- nists will have to say in reply." — Kitou-/"1*" . Demy 8vo, Cloth, Illustrated. Price 5s. net. (Post free, jiric' ").-•. 4rf.) " Dr. Kidd shows much ingenuity in explaining the various causes which, in his opinion, have produced the different hair-slopes, and the illustrations with which the book is liberally furnished are of considerable assistance to the reader." — Westminster lleriew. " The description of the facts can be commended. There is evidence that considerable care has been taken in the collection of observations, and the illustrations are excellent." — Mawlin'tri- Guardian. "That the direction of the hair-slope in mammals varies much is a fact well known to zoologists ; but we are not acquainted with any general work, such as the present, in which the facts are put together within one cover. The author has, therefore, in any case, accomplished a piece of work which supplies a definite want in the literature of zoology. The usefulness of this general summary is furthermore enhanced by numerous clear figures which show at a glance the essential facts." — Lancet.. PUBLISHED BY ADAM AND CHARLES BLACK, 4, 5, & 6 SOHO SQUARE, LONDON THE SCIENCE AND PHILOSOPHY OF THE OBGANISM THE GIFFORD LECTURES DELIVERED BEFORE THE UNIVERSITY OF ABERDEEN IN THE YEARS 1907-8 BY HANS DRIESCH, PH.D. HEIDELBERG IN TWO VOLUMES. DEMY 8vo. CLOTH PRICE I O/6 NET EACH (Post free, price I I/- each] NOTE THIS work rests chiefly upon the results of experimental embryology, the most recent branch of biology, in which the author is personally concerned. The first volume deals with the science of the organism and discusses the chief results of analytical biology in two parts, and, in the author's view, results in two main proofs of the autonomy of life, or what may still be called, if in a modernised sense, " vitalism," as distinguished from " mechanism." The second volume, continuing the exposition of the science of the organism into a third part, contains a detailed theory of animal movements, reflexes, instinct, and, in particular, action, resulting in a third main proof of the autonomy of life. The foundation of a complete system of scientific biology having thus been laid, the remainder of the volume — forming its larger half — is devoted to the philosophy, as distinguished from the science, of the organism. The central part of this concluding discussion is devoted to a thorough analysis of the concept of " teleology " and its relation to the inorganic sciences, PUBLISHED BY ADAM AND CHARLES BLACK, 4, 5, & 6 SOHO SQUARE, LONDON