m

l

5?

O

m

o

THE

RAY SOCIETY.

INSTITUTED MDCCCXLIV.

This volume is issued to the Subscribers to the BAY SOCIETY for the Tear 1879.

LONDON:

MDCCCLXXX.

BfJ

A MONOGBAPH

FREE AND SEMI-PARASITIC

COPEPODA OF THE BRITISH ISLANDS.

BY

G. STEWABJDSON BEADY, M.D., F.L.S.,

PROFESSOR OF NATURAL HISTORY IN THE UNIVERSITY OF DURHAM COLLEGE

OF PHYSICAL SCIENCE, NEWCASTLE-ON-TYNE ; CORRESPONDING

MEMBER OF THE ZOOLOGICAL SOCIETY OF LONDON,

OP THE ACADEMY OF NATURAL SCIENCES

OF PHILADELPHIA, ETC.

VOL. III.

LONDON: FEINTED FOR THE RAY SOCIETY.

MDCCCLXXX.

PBINTED BY J. E. ADLARD, BABTHOLOMEW CLOSE.

A MONOGRAPH

BRITISH FREE AND SEMI-PARASITIC COPEPODA.

VOL. III.

GENERAL ANATOMY AND DEVELOPMENT OF THE COPEPODA.

IN preparing this monograph it was not part of my plan to enter at all into the consideration of the physiology or internal anatomy of the Copepoda, but, in compliance with the wishes of some of my friends, expressed while the first volume was passing through the press, I have put together, in the form of a preface to this third volume, a condensed account of some of the more important observations which have hitherto been made on this part of the subject. It will be at once apparent that what I have attempted is nothing more than a general outline, in the production of which I have been greatly indebted to the works of Glaus, Gegenbaur, and Huxley. Had I included the truly parasitic species, in whose anatomy, physiology, and general habits of life so many points of the highest interest occur, I should have been travelling beyond

VOL. III. A

45279

BRITISH COPEPODA.

the professed limits of my work, and have extended this preface beyond reasonable bounds, — on which account those species are noticed in the most cursory manner, and only in elucidation of the proper subject of the memoir.

The non-parasitic COPEPODA may be described as ENTOMOSTRACA, having an elongated body of variable form, but generally cylindrical, without a bivalved shell, and showing more or less completely a division of the body into numerous rings, or somites. There are two pairs of antenna, three pairs of prehensile and masticatory, or suctional, mouth organs, and five pairs of feet, adapted chiefly for swimming. The females are mostly fertilized by means of spermato- phores ; the ova are usually carried in external ovisacs ; when first hatched the larvae have only three pairs of limbs, and go through several metamorphoses before attaining the mature form.

The parasitic species, at one end of the series approach very nearly in structure and general appear- ance to the non-parasitic ; at the other end they are extremely different, exhibiting, especially in the males, many most remarkable examples of retrograde development, so that without the study of their metamorphoses it would be quite impossible to re- cognise them as Copepoda, or even as Crustacea of any kind. Yet even in these degraded forms — at any rate in the females — natatory limbs in a very much atrophied condition are almost constantly found.

GENERAL FORM. — The animal is usually somewhat pear-shaped, rounded in front and tapering towards the

GENEEAL ANATOMY. 3

hinder extremity, convex on the dorsal and flattened on the ventral surface. The degree of dorsal convexity, however, is variable, the most common form of the body being sub-cylindrical, but in many cases — notably amongst the Poecilostoma, Siphonostoma, and in some genera of the Harpacticidse (Zaus, Peltidium, For- cellidium, Idya, Scutellidmni) — the animal is markedly flattened, constituting a type of structure almost as widely distinct from the normal Copepoda as are the Isopoda from the Amphipoda, amongst the sessile-eyed Crustacea. But in the genera of Harpacticidae here referred to, the flattened form does not coincide with any deep-seated differences of structure, for, indeed, in some normal genera (Thalestris, Laophonte) we find certain species assuming this form, but retaining all the other characteristics of the genus to which they belong. The directly opposite type of structure, in which the animal is compressed laterally, occurs only in the curious genus Amymone, a group of rare occur- rence so far as our present knowledge extends, not having been noticed outside of the European area.

The front of the body is covered by a mem- branous shield or carapace, which falls rather loosely over the bases of the limbs below, but behind is continuous with the coverings of the thoracic somites ; in front it is usually projected from the forehead in the middle line, forming a rostrum, short or entirely absent in some few cases, but often moderately long and curved. In the Calanidse it is usually long and slender, much curved, and furcate nearly to its base. Behind the rostrum is commonly

BRITISH COPEPODA.

placed the median eye, often single to all appearance, but really composed of two closely approximated lateral eyes, embedded in a mass of black or crimson

pigment.

The head and the first thoracic segment are usu- ally fused together, a fact which may be recognised by the position of the first pair of swimming feet (the first thoracic appendages), these limbs being generally fixed to the hinder part of the first body- segment, which is thus seen to be composed of all the cephalic, and the first of the thoracic, somites. In some cases a transverse indentation maybe noticed, probably a trace of the " cervical suture " which is so conspicuous a feature of the carapace in Crayfishes and Lobsters. Instances are frequent, however, in which the head is quite distinct from the thorax.

Theoretically, the Copepoda, like other Crustacea, are composed of twenty or twenty-one somites (twenty-one according to most authors, but twenty if we follow Huxley, who does not look upon the telson or last abdominal segment as a true somite)! the entire series not being developed, how- ever, in any one animal. Sometimes one or many of the somites are suppressed, at other times several are united into one segment, the real nature of which is rendered evident by the attachment to it of several appendages, each pair indicating the position of an anchylosed somite. Probably in no case amongst the Copepoda can more than sixteen or seventeen somites be recognised by enumeration either of appendages or segments. The somitic appen-

GENERAL ANATOMY. 5

dages or limbs, though very various in form and function, may all be reduced to three component parts — a basal portion or protopodite, which gives support to two branches, termed respectively endo- podite and exopodite. These parts are most clearly developed in the swimming-feet, which are distinctly made up of a peduncle and two branches, but in the mouth organs the same structure may be traced, though often modified to such an extent as to be obscure and difficult of recognition.

The cephalon in the Copepoda is composed of six somites, its appendages being one pair of eyes, two pairs of antennge, one pair of mandibles, one pair of maxillas, and two pairs of foot-jaws ; the thorax consists of five somites, and has five pairs of appen- dages in the form of swimming-feet ; the abdomen has no appendages, but consists of five somites and terminates in a forked tail, which ought probably to be considered as a sixth somite ; in the female the first two abdominal somites are generally united, forming one large genital segment with a pair of vulvar apertures. In parasitic species the abdo- men is often very much, reduced in size, both as respects the number and bulk of its somites, and this is the case also in some genera which are only partially parasitic, as Corycaus, Acontiophorus, Arto- trogus and others. As regards the cephalic appendages, it must be noticed that the so-called two pairs of foot- jaws are in reality portions of one and the same somitic appendage, but having the appearance of perfectly distinct organs they have come to be con-

6 BRITISH COPEPODA.

sidered as such, and their characters so taken note of by all systematists. The process of development, however, has been traced by Glaus, so as to leave no doubt as to the true nature of the organs.

APPENDAGES OF THE HEAD. — The eyes in their simplest form — in Cyclops, for instance— appear as a red or black spot in the middle of the frontal region, directly over the brain, with which they are connected by a large nerve, the spot when closely examined being found to consist of two lateral eyes, closely approximated and embedded in a mass of pigment ; the visual part of the apparatus is composed of two refracting bodies, or crystalline cones, and when more highly developed may possess numerous lenses, so as to form something like a facetted cornea. In some cases the eyes are widely separated, and have between them, in the median line, a simple, globular, pedunculated eye (PontelUnce) ; in other cases, as in some Corycceidce, the median eye is very small, while the lateral eyes are large, destitute of pigment, and consist of simple highly refracting lenses. In some parasitic species only are the eyes entirely wanting. In Pleuromma there is a supplementary eye, consisting of lenses with black pigment matter, on one of the thoracic segments.

The anterior antennce are usually large and conspicu- ous organs, rising from hollows in the front of the head on each side of the rostrum. They act in many species (Calanidce, Cyclopidce) as very powerful swim- ming organs, it being by their agency chiefly that these animals propel themselves through the water; the

GENERAL ANATOMY. 7

motion is thus a succession of rapidly recurring jerks corresponding with the separate sweeps of the antennas. These organs vary greatly in length and in the cha- racter of their setose armature. In the Harpactinidce, as well as in most parasitic and semi-parasitic species, the length of the anterior antenna falls short of, or at any rate does not usually much exceed, that of the first division of the body (carapace), while in the Cyclopidce it often equals, or even exceeds, that of the cephalothorax. In many Galanidce the length of the antennae is still greater, not unfrequently exceeding — sometimes very much exceeding — that of the entire animal. There is, however, no instance of this kind among British species, nor, so far as I know, among any but distinctly pelagic forms. And it is remarkable that, with this extreme length of the antennae, there is usually asssociated a greatly in- creased development of the apical lash of swimming- set^ with which the organ is armed. Sometimes also, as, notably, in the pelagic genus Euchceta, the mar- ginal hairs, though few, become wonderfully long. The antennas have, however, other important functions besides those of locomotion ; some of the variously formed setae with which they are in most cases largely provided act, no doubt, as organs of special sensation— the more simple hairs, perhaps, as tactile, the flattened and club-shaped setae as olfactory organs. The flat- tened, ensiform kind of appendage is seen on the antennae of most of the Harpacticidce (e. g. Idyafurcata, Plate LXVII, fig. 2 a), and the club-shaped form is seen in great force in Isias clavipes (Plate VII, figs. 4, 5) as

8 BRITISH COPEPODA.

well as in other Calanidce ; these last-named organs are always most largely developed in the males, and it is not unlikely that they are subsidiary sexual organs, pos- sibly endowed with a highly-developed sensuous faculty. Besides the forms of antennal appendages here noted other modifications exist, of which the special uses are at present unknown.

The anterior antennas serve yet another essential purpose, being adapted in the males as clasping organs ; they, together with the fifth pair of feet, are in very many species specially fashioned so as to insure a firm grasp of the female. The number of joints found in the anterior antennas varies from five or six in some semi- parasitic species (Lichomolgus) and eight or nine in the Harpacticidce, to twenty-four or twenty- five in the Calanidce. In the females the antennas are always alike on both sides of the body, but in the males of many of the Calanidce the right antenna is modified for the purpose of clasping ; while in all other non- parasitic species the male antennas of both sides are specialized for that purpose. In the semi-parasitic Corycce-idce, Saphirinidce, &c., there is little or no sexual difference in the anterior antennas, the clasping function devolving on the posterior pair. The par- ticular structural adaptations differ in different cases ; in the Calanidce, as before stated, the right antenna only is differentiated, the alteration consisting in one of the joints not far from the apex being so articulated as to form a hinge, by means of which the distal portion of the limb can be flexed upon the basal portion. Above and below the hinge the inner margins of the

GENERAL AXATOMY. 9

antennae are frequently armed with denticulated plates, giving a firmer grasp. A very remarkable instance of this structure exists in Gandace pectinata (Plate X, fig. 2), also in the genera Centropages, Pontella, Para- pontella, &c. ; other genera, such as Temora and Diaptomus (Plate VI, fig. 7), are provided with one or more strong spines in lieu of, or in addition to, the denticulated plates. These spines are situated at various points of the internal surface of the antenna above the hinge-joint, and the limb itself is more or less swollen in the same situation to give room for a powerful flexor muscle. In the Pontellince this enlarge- ment of the limb is excessive (Plate X A, fig. 2 ; Plate XI, fig. 2). In the males of many Calanidce, however, there is little or no difference of form between the antennae of the two sides ; the difference is very slight also in the Misophriidce. In most Cyclopidce there are found differences of a kind similar to those already described, but affecting equally both right and left antenna (Plate XVII, fig. 5 ; Plate XXII, fig. 16) ; there are here no denticulated plates and few spines, but the limb is distinctly geniculated near the base as well as near the apex, and the articulations of the terminal segments are likewise very mobile. Amongst the Notodelphyidce the structure of the male antennas is somewhat similar to that of the Cyclopidce, but by no means so well marked. In the Harpadicidce, where the anterior antenna are very much smaller in com- parison with the size of the animal, the hingement of the male organ is not quite so obvious, though still sometimes quite of Cyclopid type (Cantlwcamptus

10 BRITISH COPEPODA.

minutus, Plate XLIV, fig. 3 ; Eobertsonia tennis, Plate XLI, fig. 3) ; the apex is, however, often strongly clawed (Longipedia, Plate XXXIV, fig. 2 ; Euterpe, Plate XL, fig. 2 ; Tachidius, Plate XXXVII, fig. 3, &c.), and very frequently the two or three joints at the proximal side of the claw are coalescent and greatly enlarged, forming a pyriform or subglobose swelling for the reception of strong muscular bands ; examples of this structure are seen in the genera Tachidius (Plate XXXVII, fig. 3), Harpacticus (Plate LXIV, figs. 2, 13), Jonesiella (Plate XL VIII, fig. 3), and in many others. The spinous and setose armature of the antennae is, as a general rule, more profuse in the males of all families ; a good illustration of this cha- racter is seen in the case of Longipedia (Plate XXXIV, figs. 2, 3). In Pcecilostoma and Siphonostoma the sexual distinctions of the anterior antennas are not strongly marked, consisting chiefly of imperfect hinge- ments or contractions of the articulating surfaces between various joints of the male organ.

The posterior or second pair of antennce are generally much smaller than the anterior pair, and consist in most cases of two branches ; they seem to be used both as swimming and prehensile organs, and in para- sitic and semi-parasitic species are specialized for prehension much more decidedly than the anterior pair. The main branch consists usually of three or four joints, to the basal or second joint of which is attached a " secondary" or " inner" branch of smaller size, and composed of one or several joints. This branch is in some cases altogether absent ; in others

GENERAL ANATOMY. 11

it is equal in size to the primary branch, and may be numerously jointed (Longipedia, Calanidce, Misoph- riidce). In Cyclopida} the secondary branch is want- ing, and in most of the Harpacticidce it is reduced to small dimensions or, in some few cases, is quite absent. The limb is generally curvate, dilated at the apex, and provided with numerous curved or geniculated setse at the extremities and over the margins of both branches. In the PcBcilostoma and Siphonostoma it consists of only one branch, which in the male of the Corycceidce (Plate LXXXIII, figs. 13, 14) is powerfully clawed, and is used chiefly as a prehensile and clasping organ.

The mandible consists, in its fully developed form, of a masticating portion and a " palp ;" the first-named division is in the form of an elongated, more or less triangular plate, dilated at the distal extremity and cut up into a variable number of tooth-like processes, these being sometimes only slight serrations, some- times large and powerful. The palp is variously formed ; in the Calanidce, most of the Harpacticidce, and in many Cyclopidce, it is composed usually of a large basal joint, from which spring one or two small setif erous branches ; these branches are generally 1- or 2-jointed, but may be absent altogether. In the sub-families Porcellidiince and Idyince the setiferous portions of the organ are very largely developed, form- ing large laminated appendages, which are fringed with densely ciliated filaments (Plate LXVIII, fig. 4 ; Plate LXIX, fig. 11). In the genus Cyclops, while the biting part of the mandible is well developed,

12 BRITISH COPEPODA.

the palp is atrophied, and is represented only by two or three ciliated setee. In the Corycceidcv the mandible is small and weak, its palp obsolete or reduced to very small dimensions, while in some Siphonostoma* the mandible itself is converted into a long and slender piercing style, which is enclosed in a tube resembling considerably the antlia of Lepi- dopterous insects, and composed of prolongations of the upper and lower lip (Plate XCIII, fig. 3) ; in Cyclopicera, however (a genus here included amongst Siphonostoma), there is an intermediate condition of things ; the mandible is very much elon- gated, slender, and finely toothed at the apex, being, in fact, almost stylet-shaped, but not enclosed in any sheath; the palp, also, is quite rudimentary (Plate LXXXIX, fig. 4).

The maxillce are small appendages, composed of a chewing portion, which is divided at the apex into numerous rather long and slender curved teeth, and of a complex, lobed, and setiferous palp, which frequently has filamentous appendages, possibly of a branchial character. In Cyclops the maxillae are very small in all their parts ; in the Calanidce, on the contrary, they are largely developed and possess numerous plumose filaments (Plate I, fig. 5) ; in the Notodelphyidce, also, the filamentous appendages are large (e.g. Plate XXIX, fig. 5). The Harpacticidce

* The Pomologies of the mouth-organs in this group have been dis- cussed by Glaus, Thorell, Buchholz, and others. The subject is not without difficulty, and is treated under the description of the Siphono- stoma (p. 26) at greater length than would be suitable in this prefatory notice.

GENERAL ANATOMY. 13

have the maxillas very small, but numerously divided ; there is usually a chewing portion pretty strongly toothed, and a palp which is divided into two or three setiferous digits, and has likewise two lateral (? branchial) offshoots (Plate LYII, fig. 2, Plate LX, fig. 6, Plate LXII, fig. 5, &c.), but this arrangement is subject to endless variation as to the number and development of the various parts. In parasitic species the maxillae, though usually present, are often atro- phied, and in the semi-parasitic forms belonging to the groups Pcecilostcma and Siphonostoma, are ex- tremely minute, and sometimes attached to the base of the mandibles.

The anterior and posterior foot-jaws (which are in reality only the exo- and endo-podites of a single somitic appendage) do not present features requiring any lengthened description. They are in all cases adapted for prehension. In the Calanidce, Cyclopidce, and Notodelphyidce, the two pairs are not very dissimilar in structure, consisting generally of from four to six joints, which are in most cases marginally produced into digitiform or wart-like eminences on their inner side, and are more or less densely setiferous. In the Harpacticidce, while the first foot-jaw is like those of the preceding families, the second is usually in the form of a strongly-clawed hand, and in some sub- families (Tdyince, &c.) both pairs assume this cha- racter; such, also, is the case in the semi-parasitic Pcecilostoma and Siphonostoma. The subfamily Longi- pediince, unlike the rest of the Harpacticidce, has the posterior foot-jaw destitute of a clawed hand, those

14 BRITISH COPEPODA.

organs in Ectinosoma being excessively slender, and almost linear in form.

THE APPENDAGES OF THE THOEAX are, in their simplest form, five pairs of swimming-feet, each foot consisting of a 2-jointed base and two 3-jointed branches, but the number of joints, though never exceeding three, is within that limit, subject to a good deal of variation. The second, third, and fourth pairs are in almost all cases constructed as simple swimming-feet, without any other function, but the first pair, not unfrequently (as in most Harpacticidce) has one of its branches con- verted into a clawed prehensile limb, and the fifth pair shows very extensive modification in almost every case, often differing very considerably in the two sexes, and sometimes being very much reduced in size, or even altogether wanting. Though, as has been already said, the second, third, and fourth pairs of feet are constructed simply as swimming organs, and present no peculiarities of form, there are some exceptions to that statement. In Metridia armata the inner branch of the second foot in the male has a very remarkably excavated notch with spinous margin (PI. LVI, fig. 20 &), and some foreign species, belong- ing, probably, to two distinct genera, Metridia and Undina, exhibit a similar structure,* sometimes in both sexes. Again, in the genus Harpacticus the inner branch of the second foot of the male has the middle joint produced downwards into a strong spine, which varies in character according to species, but, in some

* These species will be described in the ' Report on the Copepoda ' taken during the voyage of H. M.S. " Challenger."

GENERAL ANATOMY. 15

shape or other, is always present (PI. LXIV, figs. 7, 16; PI. LXV, fig. 11). The third foot of the male in the same genus is sometimes (perhaps not always) con- verted into a clasping organ, the outer branch being bent across the inner and having its last joint armed with several strong spines (PI. LXIV, fig. 20). In the Calanidce and Misophriidce the fifth pair of the male is usually specially adapted as a clasping organ, the limb of one or both sides being reduced to a single branch, and provided with an armature of spines or hooks, which either entirely supersedes the swimming function, as in Temora, or is superadded, as in Centropages; but in other genera belonging to these families the sexual alteration of the limb is not very great (Calanus). Some pelagic genera, which are not represented in the British seas (Euchceta, Undina}, though possessing in the male a strongly developed prehensile fifth foot, have in the female only four pairs of simple swimming- feet. In Undina the male fifth foot is remarkably long and very fantastic in shape, reaching sometimes even beyond the extremity of the caudal segments. The organ may not unfrequently be seen with sperma- tophores adherent to its apex,* and is possibly used as the means of conveying these bodies to the vulva of the female. f It is remarkable, too, that in species so constituted (especially in Undina Darwinii, Lubbock) the spermatophores are very commonly found affixed in a futile manner to the back of the thoracic rings of

* A similar condition is figured by Dr. Claus in the case of Euchata prestandrece (' Die frei lebenden Copepoden,' pi. xxx, fig. 9).

f A similar function is performed by tbe maxillary palps (chela?) of male spiders and by the hectocotylised arms of some cuttle-fishes.

16 BRITISH COPEPODA.

the female. In the Cyclopidce and Notodelphyidce the fifth feet are usually rudimentary and alike in both sexes, and in the HarpacticidcR they take the form of small, marginally setose, foliated expansions, slightly different in the two sexes, but generally larger in the female, in which sex they serve sometimes as a covering and support for the external ovisacs. In the semi-parasitic groups these organs are generally small, 1- or 2-jointed, and alike in both sexes.

THE NEEVOUS SYSTEM of the Copepoda is described as consisting of a brain, which gives off various sensory nerves, a sub-oesophageal ganglion, and a ventral nerve cord, on which are situated ganglionic enlarge- ments ; the antennary nerves are also thickened, forming ganglionic rings.

From an investigation of the nervous system of Cyclops, Mr. Marcus M. Hartog, F.L.S., of Owen's College, Manchester, has recently made out some other points which he kindly allows me to insert here. The most important of these are, that " gangli- onic swellings are found near the terminations of all sensory nerve fibres ; that the ventral nerve cord gives off at the end of the third segment of the body a pair of superficial cutaneous nerves, and at the fourth segment two pairs, one to the rudimentary fifth legs and another to two ventral muscles which rise from the sternal portion of the fifth segment. In the first abdominal segment is the fork described by Glaus and Leydig, but this takes its origin from the superficial (ventral) aspect of the cord which is continued onwards under the colleterial gland. After

GENERAL ANATOMY. 17

running obliquely outwards each branch of the fork subdivides into two, an anterior and a posterior branch, both sensory. At the commencement of the third abdominal segment the ventral cord forks, its branches diverge slightly in this segment, but more in the next, rising to the sides of the intestine, and having the ventral muscles of this segment superficial to them. In the last segment they have left the intestine and run about the horizontal median plane straight into the axis of either branch of the f urea.

" The ventral cord in Cyclops is not differentiated into distinct ganglia up to the second free (third) thoracic segment : beyond this is an enlargement con- taining ganglion cells at the posterior end of the fourth, and another (very small) in the last thoracic segment."

THE ORGANS OF SENSE, so far as they exist in the tactile and olfactory rods of the antenna?, have already received brief notice. Mr. Hartog has recently de- scribed certain vesicles in the frontal region of Cyclops, and others attached to the bases of the fifth pair of feet and seated on a ganglionic enlargement of the nerve supplying the feet, which vesicles he believes to be auditory organs. In the male they contain one or more highly refracting bodies floating freely in the interior. Claus found a pair of these vesicles in the brain of Calanella, and has figured them in * Die frei lebenden Copepoden ' (plate vii, fig. 9).

THE DIGESTIVE CANAL has a short, straight gullet, a large stomach, often with two ca3cal tubes, and an intestine opening on the dorsal aspect of the last (or

VOL. in. B

18 BE1TTSH COI'EPODA.

last but one ?) abdominal segment. The hinder por- tion of the alimentary canal is perhaps also uriniferous, but there exist near the bases of the foot-jaws other glandular organs, which may have a renal function. In the males of parasitic Copepoda the digestive canal disappears entirely.

RESPIRATION. — If we are right in assigning to certain appendages of the mouth-apparatus (to which refer- ence has been already made; see pp. 10, 11 *) a branchial function, then we cannot altogether assent to the commonly - received belief that respiration in the Copepoda is carried on entirely by the dermic and intestinal surfaces of the body, without the interven- tion of any specialized respiratory apparatus. A sub- rhythmic contraction of the hinder extremity of the gut has, however, been noticed by Mr. Hartog in several Copepoda (Cyclops, Diaptomus, Canthocamp- tus), and by several observers in other Crustacea (e. g. Astacus, Limnadia, Daphma) : this is, no doubt, a respiratory movement.

CIRCULATION. — In many Copepoda no special circu- latory organs have been found ; but in some there is a tubular heart, situated in the last thoracic segment, which drives forward the blood by a short vessel to the brain and anterior parts of the body, the blood

* There can be no doubt that these setiferous plates are homologous with the structures called in the Ostracoda branchial lamince by G. O. Sars. But they do not appear to have any special internal circulation of the vital fluids, and if their function be branchial, they must act only by propelling waves of aerated water over the neighbouring surfaces. It is, pei'haps, on the whole, most probable that the currents produced by these ciliated appendages are subsidiary to nutrition rather than to respiration.

GENERAL ANATOMY. 19

then passing through lacunae scattered throughout the tissues of the animal, and finding its way back to the heart, which it enters by slits in the walls of that organ.

REPRODUCTION. — The sexes in the Copepoda are always separate, sexual differences showing themselves even externally in the form and structure of the body ; in some, especially in parasitic species, the dimorphism is most remarkable, the male becoming little more than a motionless sperm-sac attached to the body of the female ; but in the species which come within the scope of this memoir the males are usually smaller, more active, and less numerous than the females, the chief external distinctions being found in the almost constant conversion of the anterior antennae — less constantly of the fifth pair of feet, and occasionally also of the posterior antennas and foot-jaws — into clasping organs. The ovaries and testes are placed in the middle or in the sides of the cephalothorax, com- municate with accessory glands, and have efferent canals, which open by distinct apertures on the sides of the first (or conjoined first and second) abdominal somites. The efferent canal of the ovary may be simple, or may give off laterally a number of pouches, which hold the eggs (Corycceidce), while in some para- sitic species it forms several terminal coils, in which the eggs are detained; in the Notodelphyidce the duct is converted into a large dorsal pouch or pseudo-uterus, covered only by the integument, in which organs the ova undergo partial development. In the free Cope- poda, however, the ova pass at once into two (often coa-

20 BRITISH COPEPODA.

lescent) external ovisacs attached to the first abdominal segment the coating of the ovisac being formed by the secretion of a special gland, situated near the termina- tion of the efferent duct, an enlargement of which forms in many cases a " receptaculum seminis." In the males the free-living forms have a simple testis ; many of the parasitic and semi-parasitic (Corycceidce, Sapphirinidce) a double testis, with two distinct vasa deferentia, the right duct being sometimes atrophied. In the coiled portion of the duct are formed the spermatophores — masses of spermatozoids enclosed in a capsule of hardened mucus, and usually fusi- form or club-shaped. During copulation the male affixes one or more of these bodies near the vulvar aperture of the female, the contents passing into the receptaculum seminis, and fertilizing the ova either in the interior of the body or during their passage into the ovisacs. In some cases the seminal fluid appears to be inserted directly into the vulva without the intervention of a spermatophore.

THE DEVELOPMENT of the free Copepoda from the moment of rupture of the ovum to the attainment of matured form presents a complex series of metamor- phoses. The parasitic species present some of the best marked examples of " retrograde development" to be found in the whole animal kingdom, but these do not come within the limits of our present subject. The form of the young Copepod on its escape from the egg is that known as NaupUus, having been de- scribed by Muller under that name before its relation to the Copepoda was known. The larva in this

GENERAL ANATOMY. 21

stage is oval, has a single frontal eye, three pairs of limbs arranged round the mouth, and no frontal appendages ; the mouth-organs proper are entirely absent, and the posterior part of the body has no appendages except a couple of setae in the neighbour- hood of the anus. The anterior portion of the body is equivalent to the three anterior cephalic somites, its three pairs of limbs becoming eventually antennae and mandibles. At the first moult the body becomes elongated and new limbs appear in the following order : — a fourth and fifth pair representing respec- tively the maxillae and foot-jaws ; a sixth and seventh which become the two anterior pairs of swimming- feet. At this stage the larva still resembles a Nauplius, and does not take on a Cyclopoid appear- ance until after the next moult.* It then resembles more closely the adult Cyclops as to the antennae and mouth-organs, but the number of limbs and somites is smaller ; the body in this condition is composed of an oval cephalothorax, three thoracic and one long terminal segment, which in succeeding moults be- comes forked. In the Cyclopidce the posterior antennae and the mandibles lose their accessory branches, but in other families these parts are usually retained. All the free, and many of the parasitic species pass through a further series of moults, in the course of which the still-wanting limbs and body- segments appear, the limbs attaining, by successive

* In some Nauplii, if not in all, the terminal part of the intestine is subglobular, and contracts periodically like the " contractile vesicle " of a Rotifer.

22 BRITISH COPEPODA.

steps, their full number of joints and perfect develop- ment.* Those parasites which miss the Nauplius stage are hatched in Cyclops-form ; many of the retrogressive species become fixed to some animal, segmentation is lost, limbs and eyes disappear or become atrophied ; the males, often dwarfed, being permanently fixed near the sexual apertures of the female.

* For details of this process in the genus Cyclops, see vol. i, p. 100.

NOTE.

CHANGE OF GENERIC NAMES.

The generic name LopJwphorus (vol. i, p. 121), having been previously used to designate a genus of Phasianidse, must be withdrawn. I therefore propose to substitute the word Pterinopsyllus

Cylindrosoma (in Table of Genera, vol. i, p. 31) is for a like reason discarded for Cylindropsyllus, and Solenostoma (loc. cit.} for Acontiophorus.

BEITISH COPEPODA.

Section II. — PCECILOSTOMA, Thorell.*

THORELL'S division of the Copepoda into three groups, G-nathostoma, Poecilostoma, and Siphonostoma, the distinctions between which are found in the characters of the mouth-organs, is disapproved by Glaus and some other authors, chiefly, as I understand them, on the ground of the gradual lapse of one series into the other rendering it impossible to draw perfect lines of demarcation, but partly, also, on the ground of a difference of interpretation of the homologies of some of the appendages. While differing from M. Thorell as to the nature of some of these organs, I myself think that his proposed division is a very natural one, the three groups presenting characters which, though differing in degree in various species, do point, on the whole, to habits of life very remarkably different, and deserving of expression in any natural classification.

The three groups are denned by M. Thorell as follows :

Series 1. GNATHOSTOMA.

Os mandibulis duabus libens tribusque paribus maxillarum instructum, siphone nullo.

* See p. 31, vol. i.

26 BRITISH COFEPODA.

Series 2. PGEOTLOSTOMA.

Os mandibulis et siphone carens, maxillarum paribus 3 — 1 ( — 0) instructum.

Series 3. SIPHONOSTOMA.

Os in siphonem, mandibulas 2 plerumque inclu- dentem, productum, et maxillarum paribus 3 — 0 in- structum.

As regards the debateable anatomical points, it may be useful if I quote, in the first place, some remarks of M. Thorell, taken from a letter which he was good enough to address to me some few years ago — before the publication of Dr. Claus's 'Neue Beitrage zur Kenntniss parasitischer Copepoden.' M. Thorell writes as follows : — " You know, of course, that I have pro- posed to divide the Copepoda into three parallel series, Grnathostoma, Poecilostoma, and Siphonostoma, and that I consider the Poecilostoma to be characterized by having the parts of the mouth free, and formed for stinging or licking, as also by the mandible being absent. This view of the oral apparatus of the Psecil- ostoma has been accepted by Claparede and a few others, but it is not admitted by, for instance, Glaus, who considers that the Poecilostoma have true man- dibles, and who rejects the subdivision proposed by me. The reasons which induced me to believe that the Paacilostoma were destitute of mandibles, and that what Glaus calls mandibles are the true maxillge, and that his maxillas are maxillar-palpi, were — first, that the " mandibles " are always placed far more back- wards than the true mandibles of the Gnathostoma;

P(ECILOSTOMA. 27

secondly, that I sometimes, as, for instance, in Licho- molgus albens, found a small longitudinal, semi-pipe- formed depression or groove exactly at the place where the sipho of, for instance, Dyspontius and Ascomyzon is inserted, and which I therefore considered to indicate the place where the sipho and mandibles ought to be found if any mandibles existed ; and thirdly and chiefly, that sometimes, as in the genus Lichomolgus, the so- called ' maxillae ' are fixed on the ' mandibles ' (quite as an ordinary palpus is fixed on a mandible or a maxilla), and directed from the oral aperture, a cir- cumstance with which I could find nothing analogous in the class Crustacea, supposing Glaus' ' mandibles ' really to be mandibles."

The greatest difficulty which besets the discussion of this question is the minuteness of the mouth-organs in these animals, and the liability to displacement or mutilation of the various parts in conducting a dissec- tion, so that the organs of one and the same species will often present very different appearances in differ- ent preparations of the animal. There can be no doubt, however, that the fact so strongly insisted on by M. Thorell, — that of the coalescence, in Lichomolgus, of the maxilla and mandible (or maxilla and palp) — does really exist : the question remains, What is this palp- like organ ? In appearance it is not unlike the poorly- developed mandible- or maxilla-palp of many Grnathos- toma, but it is also much like a single branch of such a maxilla as we find in the genus Cyclopicera or Arto- trogus, so that not much can be learned by comparison of structure only. The point next to be considered is

28 BRITISH COPEPODA.

the nature of the appendages, — by Thorell called the first pair of maxillae, by Glaus the mandibles, — to which the palp-like organ is attached.

In considering this question we shall do well to take a somewhat wider survey than merely of the order Copepoda. Among the nearly related order Ostracoda, for the most part consisting of true Gnathostomous Crustacea, we find a group, — including chiefly the genus Paradoxostoma, — in which the mouth is modified for suctorial purposes in a manner at once reminding one of the siphonostomous Copepoda. In Paradoxostoma the tubular mouth is formed by the coalescence of the labrum and labium, and the man- dible assumes the form of a stilet, having a very slender filiform palp, the almost exact counterpart of the same organs in A.contioplwrus , Cydopicera, &c. (see PI. LXXXIX, fig. 4, and PL XC, fig. 4). There can be no doubt, I think, that in the well-marked siphonostomous Copepoda, such as Acontiophorus, Dys- pontius, and Artotrogus (Ascomyzon), the tubular mouth is formed, as in Paradoxostoma, by the union of the upper and lower lips, and that the filiform organs lying immediately by the side of the siphon (see PI. XC, fig. 1 c, and PI. XCI, fig. 6 c c) are modified man- dibles and palps; in Artotrogus, indeed (PL XCIII, fig. 3 b &'), we find this stilet-shaped mandible dis- tinctly toothed at its apex. In the genus Cydopicera, of which I have fortunately collected and examined many specimens with great care, all the mouth- organs are largely developed; there is an unmistak- able mandible with a well-developed palp, a distinct

PCECILOSTOMA. 29

2-branched maxilla, and a- stout proboscidiform suc- torial mouth (PI. LXXXIX, figs. 4, 5, 6). And it can scarcely be doubted that the two-branched organ shown in Plate LXXXVII, fig. 10 (Acontiophoi*us armatus), is homologous with the mandible and palp of Cyclopicera nigripes represented in* Plate LXXXIX, fig. 4, and of G. gracilicauda (Plate LXXXIII, fig. 3). To recur to the genus Lichomolgus. If we examine again the disputed organ (PI. LXXXV, figs. 4, 12, PL LXXXVII, fig. 3, PL LXXXVIII, fig. 11) we shall find that it bears a very strong structural resemblance to those appendages of siphonostomous genera which have just occupied our attention, and the mandibular nature of which is, I think, pretty conclusively shown. I have therefore little doubt that this organ in Licho- molgus ought to pass for a mandible, and inasmuch as the mandibular palp amongst the Siphonostoma is sometimes nearly or quite suppressed, but the maxilla never, and, moreover, as when the mandible-palp does exist it assumes a form totally different from the palp- like appendage of the mandible of Lichomolgus; for these reasons I am. disposed to regard this appendage as a rudimentary maxilla. The opinion is confirmed by a