For some time, owing to an undue importance being ascribed to the want of a particular branchial cavity, Mysis, Leucifer, and Phyllosoma were referred to the Stomapoda, which are now again limited, as originally by Latreille, to the Mantis-shrimps (Squilla), the Glass-shrimps (Erichthus) and their nearest allies. Of the developmental history of these we have hitherto been acquainted with only isolated fragments. The tracing of the development in the egg is rendered difficult by the circumstance, that the Mantis-shrimps do not, like the Decapoda, carry their spawn about with them, but deposit it in the subterranean passages inhabited by them in the form of thin, round, yellow plates. The spawn is consequently exceedingly difficult to procure, and unfortunately it becomes spoilt in a day when it is removed from its natural hatching place, whilst on the contrary the progress of development may be followed for weeks together in the eggs of a single Crab kept in confinement. The eggs of Squilla, like those removed from the body of the Crab, die because they are deprived of the rapid stream of fresh water which the mother drives through her hole for the purpose of her own respiration.

The accompanying representation of the embryo of Squilla shows that it possesses a long, segmented abdomen without appendages, a bilobate tail, six pairs of limbs, and a short heart; the latter only pulsates weakly and slowly. If it acquires more limbs before exclusion, the youngest larva must stand on the same level as the youngest larva of Euphausia observed by Claus.

(FIGURE 34. Embryo of a Squilla, magnified 45 diam. a. heart.

FIGURE 35. Older larva (Zoea) of a Stomapod, magnified 15 diam.)

Of the two larval forms at present known which are with certainty to be ascribed, if not to Squilla, at least to a Stomapod, I pass over the younger one* (* 'Archiv fur Naturgeschichte' 1863 Taf 1.) as its limbs cannot be positively interpreted, and will only mention that in it the last three abdominal segments are still destitute of appendages. The older larva (Figure 35), which resembles the mature Squilla especially in the structure of the great raptorial feet and of the preceding pair, still wants the six pairs of feet following the raptorial feet. The corresponding body-segments are already well developed, an unpaired eye is still present, the anterior antennae are already biramose, whilst the flagellum is wanting in the posterior, and the mandibles are destitute of palpi; the four anterior abdominal segments bear biramose natatory feet, without branchiae; the fifth abdominal segment has no appendages, and this is also the case with the tail, which still appears as a simple lamina, fringed on the hinder margin with numerous short teeth. It is evident that the larva stands essentially in the grade of Zoea.

CHAPTER 8. DEVELOPMENTAL HISTORY OF EDRIOPHTHALMA.

Less varied than that of the Stalk-eyed Crustacea is the mode of development of the Isopoda and Amphipoda, which Leach united in the section Edriophthalma, or Crustacea with sessile eyes.

(FIGURE 36. Embryo of Ligia in the egg, magnified 15 diam. D. yelk; L. liver.)

The Rock-Slaters (Ligia) may serve as an example of the development of the Isopoda. In these, as in Mysis, the caudal portion of the embryo is bent not downwards, but upwards; as in Mysis also, a larval membrane is first of all formed, within which the Slater is developed. In Mysis this first larval skin may be compared to a Nauplius; in Ligia it appears like a maggot quite destitute of appendages, but produced into a long simple tail (Figure 37). The egg-membrane is retained longer than in Mysis; it bursts only when the limbs of the young Slater are already partially developed in their full number. The dorsal surface of the Slater is united to the larval skin a little behind the head. At this point, when the union has been dissolved a little before the change of skin, there is a foliaceous appendage, which exists only for a short time, and disappears before the young Slater quits the brood-pouch of the mother.

(FIGURE 37. Maggot-like larva of Ligia, magnified 15 diam. R remains of the egg-membrane. We see on the lower surface, from before backwards:—the anterior and posterior antennae, the mandibles, the anterior and posterior maxillae, maxillipedes, six ambulatory feet, the last segment of the middle-body destitute of appendages, five abdominal feet, and the caudal feet.)

The young animal, when it begins to take care of itself, resembles the old ones in almost all parts, except one important difference; it possesses only six, instead of seven pairs of ambulatory feet; and the last segment of the middle-body is but slightly developed and destitute of appendages. It need hardly be mentioned that the sexual peculiarities are not yet developed, and that in the males the hand-like enlargements of the anterior ambulatory feet and the copulatory appendages are still deficient.

(FIGURE 38. Embryo of a Philoscia in the egg, magnified 25 diam.)

To the question, how far the development of Ligia is repeated in the other Isopoda, I can only give an unsatisfactory answer. The curvature of the embryo upwards instead of downwards was met with by me as well as by Rathke in Idothea, and likewise in Cassidina, Philoscia, Tanais, and the Bopyridae,—indeed, I failed to find it in none of the Isopoda examined for this purpose. In Cassidina also the first larval skin without appendages is easily detected; it is destitute of the long tail, but is strongly bent in the egg, as in Ligia, and consequently cannot be mistaken for an "inner egg-membrane." This, however, might happen in Philoscia, in which the larval skin is closely applied to the egg-membrane (Figure 38), and is only to be explained as the larval skin by a reference to Ligia and Cassidina. The foliaceous appendage on the back has long been known in the young of the common Water Slater (Asellus).* (* Leydig has compared this foliaceous appendage of the Water Slaters with the "green gland" or "shell-gland" of other crustacea, assuming that the green gland has no efferent duct and appealing to the fact that the two organs occur "in the same place." This interpretation is by no means a happy one. In the first place we may easily ascertain in Leucifer, as was also found to be the case by Claus, that the "green gland" really opens at the end of the process described by Milne-Edwards as a "tubercule auditif" and by Spence Bate as an "olfactory denticle." And, secondly, the position is about as different as it can well be. In the one case a paired gland, opening at the base of the posterior antennae, and therefore on the lower surface of the SECOND segment; in the other an unpaired structure rising in the median line of the back BEHIND THE SEVENTH SEGMENT, ("behind the boundary line of the first thoracic segment," Leydig).) That the last pair of feet of the thorax is wanting in the young of the Wood-lice (Porcellionides, M.-Edw.) and Fish-lice (Cymothoadiens, M.-Edw.) has already been noticed by Milne-Edwards. This applies also to the Box-Slaters (Idothea), to the viviparous Globe-Slaters (Sphaeroma) and Shield-Slaters (Cassidina), to the Bopyridae (Bopyrus, Entoniscus, Cryptoniscus, n.g.), and to the Cheliferous Slaters (Tanais), and therefore probably to the great majority of the Isopoda. All the other limbs are usually well developed in the young Isopoda. In Tanais alone, all the abdominal feet are wanting (but not those of the tail); they are developed simultaneously with the last pair of feet of the thorax.

(FIGURE 39. Embryo of Cryptoniscus planarioides, magnified 90 diam.

FIGURE 40. Last foot of the middle-body of the larva of Entoniscus Porcellanae, magnified 180 diam.)

The last pair of feet on the middle-body of the larva, consequently the penultimate pair in the adult animal, is almost always similar in structure to the preceding pair. A remarkable exception is, however, presented in this respect by Cryptoniscus and Entoniscus,—remarkable as a confirmation of Darwin's proposition that "parts developed in an unusual manner are very variable," for in the peculiarly-formed pair of feet there exists the greatest possible difference between the three species hitherto observed. In Cryptoniscus (Figure 39) this last foot is thin and rod-like; in Entoniscus Cancrorum remarkably long and furnished with a strongly thickened hand and a peculiarly constructed chela; in Entoniscus Porcellanae very short, imperfectly jointed, and with a large ovate terminal joint (Figure 40).

Some Isopods undergo a considerable change immediately before the attainment of sexual maturity. This is the case with the males of Tanais which have already been noticed, and, according to Hesse, with the Pranizae, in which both sexes are said to pass into the form known as Anceus. But Spence Bate, a careful observer, states that he has seen females of the form of Praniza laden with eggs far advanced in their development.

(FIGURE 41. Entoniscus Cancrorum, female, magnified 3 times.

FIGURE 42. Cryptoniscus planarioides, female, magnified 3 times.

FIGURE 43. Embryo of a Corophium, magnified 90 diam.)

In this order we meet for the first time with an extensive retrograde metamorphosis as a consequence of a parasitic mode of life. Even in some Fish-lice (Cymothoa) the young are lively swimmers, and the adults stiff, stupid, heavy fellows, whose short clinging feet are capable of but little movement. In the Bopyridae (Bopyrus, Phryxus, Kepone, etc., which might have been conveniently left in a single genus), which are parasitic on Crabs, Lobsters, etc., taking up their abode chiefly in the branchial cavity, the adult females are usually quite destitute of eyes; the antennae are rudimentary; the broad body is frequently unsymmetrically developed in consequence of the confined space; its segments are more or less amalgamated with each other; the feet are stunted, and the appendages of the abdomen transformed from natatory feet with long setae into foliaceous or tongue-shaped and sometimes ramified branchiae. In the dwarfish males the eyes, antennae, and feet, are usually better preserved than in the females; but on the other hand all the appendages of the abdomen have not unfrequently disappeared, and sometimes every trace of segmentation. In the females of Entoniscus, which are found in the body-cavity of Crabs and Porcellanae, the eyes, antennae, and buccal organs, the segmentation of the vermiform body, and in one species (Figure 41) the whole of the limbs, disappear almost without leaving a trace; and Cryptoniscus planarioides would almost be regarded as a Flatworm rather than an Isopod, if its eggs and young did not betray its Crustacean nature. Among the males of these various Bopyridae, that of Entoniscus Porcellanae occupies the lowest place; it is confined all its life to six pairs of feet, which are reduced to shapeless rounded lumps.

The Amphipoda are distinguishable from the Isopoda at an early period in the egg by the different position of the embryo, the hinder extremity of which is bent downwards. In all the animals of this order which have been examined for it,* (* In the genera Orchestoidea, Orchestia, Allorchestes, Montagua, Batea n.g., Amphilochus, Atylus, Microdeutopus, Leucothoe, Melita, Gammarus (according to Meissner and La Valette), Amphithoe, Cerapus, Cyrtophium, Corophium, Dulichia, Protella and Caprella.) a peculiar structure makes its appearance very early on the anterior part of the back, by which the embryo is attached to the "inner egg-membrane," and which has been called the "micropylar apparatus," but improperly as it seems to me.* (* Little as a name may actually affect the facts, we ought certainly to confine the name "micropyle" to canals of the egg-membrane, which serve for the entrance of the semen. But the outer egg-membrane passes over the "micropylar apparatus" of the Amphipoda without any perforation, according to Meissner's and La Valette's own statements; it appears never to be present before fecundation, attains its greatest development at a subsequent period of the ovular life, and the delicate canals which penetrate it do not even seem to be always present, indeed it seems to belong to the embryo rather than to the egg-membrane. I have never been able to convince myself that the so-called "inner egg-membrane" is really of this nature, and not perhaps the earliest larva skin, not formed until after impregnation, as might be supposed with reference to Ligia, Cassidina and Philoscia.) It will remind us of the union of the young Isopoda with the larval membrane and of the unpaired "adherent organ" on the nape of the Cladocera, which is remarkably developed in Evadne and persists throughout life; but in Daphnia pulex, according to Leydig, although present in the young animals, disappears without leaving a trace in the adults.

The young animal, whilst still in the egg, acquires the full number of its segments and limbs. In cases where segments are amalgamated together, such as the last two segments of the thorax in Dulichia, the last abdominal segments and the tail in Gammarus ambulans and Corophium dentatum, n. sp., and the last abdominal segments and the tail in Brachyscelus,* or where one or more segments are deficient, as in Dulichia and the Caprellae, we find the same fusion and the same deficiencies in young animals taken out of the brood-pouch of their mother. (* According to Spence Bate, in Brachyscelus crusculum the fifth abdominal segment is not amalgamated with the sixth (the tail) but with the fourth, which I should be inclined to doubt, considering the close agreement which this species otherwise shows with the two species that I have investigated.) Even peculiarities in the structure of the limbs, so far as they are common to both sexes, are usually well-marked in the newly hatched young, so that the latter generally differ from their parents only by their stouter form, the smaller number of the antennal joints and olfactory filaments, and also of the setae and teeth with which the body or feet are armed, and perhaps by the comparatively larger size of the secondary flagellum. An exception to this rule is presented by the Hyperinae which usually live upon Acalephae. In these the young and adults often have a remarkably different appearance; but even in these there is no new formation of body-segments and limbs, but only a gradual transformation of these parts.*

(* In the young of Hyperia galba Spence Bate did not find any of the abdominal feet, or the last two pairs of thoracic feet, but this very remarkable statement required confirmation the more because he examined these minute animals only in the dried state. Subsequently I had the wished-for opportunity of tracing the development of a Hyperia which is not uncommon upon Ctenophora, especially Beroe gilva, Eschsch. The youngest larva from the brood-pouch of the mother already possess THE WHOLE of the thoracic feet; on the other hand, like Spence Bate, I cannot find those of the abdomen. At first simple enough, all these feet soon become converted, like the anterior feet, into richly denticulated prehensile feet, and indeed of three different forms, the anterior feet (Figure 44) the two following pairs (Figure 45) and finally the three last pairs (Figure 46) being similarly constructed and different from the rest. In this form the feet remain for a very long time, whilst the abdominal appendages grow into powerful natatory organs, and the eyes, which at first seemed to me to be wanting, into large hemispheres. In the transition to the form of the adult animal the last three pairs of feet (Figure 49) especially undergo a considerable change. The difference between the two sexes is considerable; the females are distinguished by a very broad thorax, and the males (Lestrigonus) by very long antennae, of which the anterior bear an unusual abundance of olfactory filaments.

Their youngest larvae of course cannot swim; they are helpless little animals which firmly cling especially to the swimming laminae of their host; the adult Hyperiae, which are not unfrequently met with free in the sea, are, as is well known, the most admirable swimmers in their order. ("Il nage avec une rapidite extreme," says Van Beneden of H. Latreillii M.-Edw.)

The transformation of the Hyperiae is evidently to be regarded as ACQUIRED and not INHERITED, that is to say the late appearance of the abdominal appendages and the peculiar structure of the feet in the young are not to be brought into unison with the historical development of the Amphipoda, but to be placed to the account of the parasitic mode of life of the young.

As in Brachyscelus, free locomotion has been continued to the adult and not to the young, contrary to the usual method among parasites. Still more remarkable is a similar circumstance in Caligus, among the parasitic Copepoda. The young animal, described by Burmeister as a peculiar genus, Chalimus, lies at anchor upon a fish by means of a cable springing from its forehead, and having its extremity firmly seated in the skin of the fish. When sexual maturity is attained, the cable is cut, and the adult Caligi, which are admirable swimmers, are not unfrequently captured swimming freely in the sea. (See 'Archiv. fur Naturgeschichte' 1852 1 page 91).)

(FIGURES 44 TO 46. Feet of a half-grown Hyperia Martinezii, n. sp. (Named after my valued friend the amiable Spanish zoologist, M. Francisco de Paula Martinez y Saes, at present on a voyage round the world.)

FIGURES 47 TO 49. Feet of a nearly adult male of the same species; 44 and 47 from the first pair of anterior feet (gnathopoda); 44 and 48 from the first, and 46 and 49 from the last pair of thoracic feet. Magnified 90 diam.)

Thus, in order to give a few examples, the powerful chelae of the antepenultimate pair of feet, of Phromina sedentaria, are produced, according to Pagenstecher, from simple feet of ordinary structure; and vice versa, the chelae on the penultimate pair of feet of the young Brachyscelus, become converted into simple feet. In the young of the last-mentioned genus the long head is drawn out into a conical point and bears remarkably small eyes; in course of growth, the latter, as in most of the Hyperinae, attain an enormous size, and almost entirely occupy the head, which then appears spherical, etc.

The difference of the sexes which, in the Gammarinae is usually expressed chiefly in the structure of the anterior feet (gnathopoda, Sp. Bate) and in the Hyperinae in the structure of the antennae, is often so great that males and females have been described as distinct species, and even repeatedly placed in different genera (Orchestia and Talitrus, Cerapus and Dercothoe, Lestrigonus and Hyperia) or even families (Hyperines anormales and Hyperines ordinaires). Nevertheless it is only developed when the animals are nearly full-grown. Up to this period the young resemble the females in a general way, even in some cases in which these differ more widely than the males from the "Type" of the order. Thus in the male Shore-hoppers (Orchestia) the second pair of the anterior feet is provided with a powerful hand, as in the majority of the Amphipoda, but very differently constructed in the females. The young, nevertheless, resemble the female. Thus also,—and this is an extremely rare case,* (* "I know of no case in which the inferior (antennae) are obsolete, when the superior are developed," Dana. (Darwin, 'Monograph on the Subclass Cirripedia, Lepadidae' page 15.)—the females of Brachyscelus are destitute of the posterior (or inferior) antennae; the male possesses them like other Amphipodae; in the young I, like Spence Bate, can find no trace of them.

It is, however, to be particularly remarked, that the development of the sexual peculiarities does not stand still on the attainment of sexual maturity.

(FIGURE 50. Foot of the second pair ("second pair of gnathopoda") of the male of Orchestia Tucurauna, magnified 15 diam.

FIGURE 51. Foot of the second pair ("second pair of gnathopoda") of the female of Orchestia Tucurauna, magnified 15 diam.)

For example, the younger sexually mature males of Orchestia Tucurauna, n. sp., have slender inferior antennae, with the joints of the flagellum not fused together, the clasping margin ("palm," Sp. Bate) of the hand in the second pair of feet is uniformly convex, the last pair of feet is slender and similar to the preceding. Subsequently the antennae become thickened, two, three, or four of the first joints of the flagellum are fused together, the palm of the hand acquires a deep emargination near its inferior angle, and the intermediate joints of the last pair of feet become swelled into a considerable incrassation. No museum-zoologist would hesitate about fabricating two distinct species, if the oldest and youngest sexually mature males were sent to him without the uniting intermediate forms. In the younger males of Orchestia Tucuratinga, although the microscopic examination of their testes showed that they were already sexually mature, the emargination of the clasping margin of the hand (represented in Figure 50) and the corresponding process of the finger, are still entirely wanting. The same may be observed in Cerapus and Caprella, and probably in all cases where hereditary sexual differences occur.

(FIGURE 52. Male of a Bodotria, magnified 10 diam. Note the long inferior antennae, which are closely applied to the body, and of which the apex is visible beneath the caudal appendages.)

Next to the extensive sections of the Stalk-eyed and Sessile-eyed Crustacea, but more nearly allied to the former than to the latter, comes the remarkable family of the Diastylidae or Cumacea. The young, which Kroyer took out of the brood-pouch of the female, and which attained one-fourth of the length of their mother, resembled the adult animals almost in all parts. Whether, as in Mysis and Ligia, a transformation occurs within the brood-pouch, which is constructed in the same way as in Mysis, is not known.* (* A trustworthy English Naturalist, Goodsir, described the brood-pouch and eggs of Cuma as early as 1843. Kroyer, whose painstaking care and conscientiousness is recognised with wonder by every one who has met him on a common field of work, confirmed Goodsir's statements in 1846, and, as above mentioned, took out of the brood-pouch embryos advanced in development and resembling their parents. By this the question whether the Diastylidae are full-grown animals or larvae, is completely and for ever set at rest, and only the famous names of Agassiz, Dana and Milne-Edwards, who would recently reduce them again to larvae (see Van Beneden, 'Rech. sur la Fauna littor. de Belgique' Crustacees pages 73 and 74), induce me, on the basis of numerous investigations of my own, to declare in Van Beneden's words; "Parmi toutes les formes embryonnaires de podophthalmes ou d'edriophthalmes que nous avons observees sur nos cotes, nous n'en avons pas vu une seule qui eut meme la moindre resemblance avec un Cuma quelconque." The ONLY THING that suits the larvae of Hippolyte, Palaemon and Alpheus, in the family character of the Cumacea as given by Kroyer which occupies three pages (Kroyer, 'Naturh. Tidsskrift, Ny Raekke,' Bd. 2 pages 203 to 206) is: "Duo antennarum paria." And this, as is well known, applies to nearly all Crustacea. How well warranted are we therefore in identifying the latter with the former. However, it is sufficient for any one to glance at the larva of Palaemon (Figure 27) and the Cumacean (Figure 52) in order to be convinced of their extraordinary similarity!) The caudal portion of the embryo in the Diastylidae, as I have recently observed, is curved upwards as in the Isopoda, and the last pair of feet of the thorax is wanting.

Equally scanty is our knowledge of the developmental history of the Ostracoda. We know scarcely anything except that their anterior limbs are developed before the posterior one (Zenker). The development of Cypris has recently been observed by Claus:—"The youngest stages are shell-bearing Nauplius-forms."

CHAPTER 9. DEVELOPMENTAL HISTORY OF ENTOMOSTRACA, CIRRIPEDES, AND RHIZOCEPHALA.

The section of the Branchiopoda includes two groups differing even in their development,—the Phyllopoda and the Cladocera. The latter minute animals, provided with six pairs of foliaceous feet, which chiefly belong to the fresh waters, and are diffused under similar forms over the whole world, quit the egg with their full number of limbs. The Phyllopoda, on the contrary, in which the number of feet varies between 10 and 60 pairs, and some of which certainly live in the saturated lie of salterns and natron-lakes, but of which only one rather divergent genus (Nebalia) is found in the sea,* have to undergo a metamorphosis. (* If the Phyllopoda may be regarded as the nearest allies of the Trilobites, they would furnish, with Lepidosteus and Polypterus, Lepidosiren and Protopterus, a further example of the preservation in fresh waters of forms long since extinguished in the sea. The occurrence of the Artemiae in supersaline water would at the same time show that they do not escape destruction by means of the fresh water, but in consequence of the less amount of competition in it.) Mecznikow has recently observed the development of Nebalia, and concludes from his observations "that Nebalia, during its embryonal life, passes through the Nauplius- and Zoea-stages, which in the Decapoda occur partly (in Peneus) in the free state." "Therefore," says he, "I regard Nebalia as a Phyllopodiform Decapod." The youngest larvae [of the Phyllopoda] are Nauplii, which we have already met with exceptionally in some Prawns, and which we shall now find reproduced almost without exception. The body-segments and feet, which are sometimes so numerous, are formed gradually from before backwards, without the indication of any sharply-discriminated regions of the body either by the time of their appearance or by their form. All the feet are essentially constructed in the same manner and resemble the maxillae of the higher Crustacea.* (* "The maxilla of the Decapod-larva (Krebslarve) is a sort of Phyllopodal foot" (Claus).) We might regard the Phyllopoda as Zoeae which have not arrived at the formation of a peculiarly endowed abdomen or thorax, and instead of these have repeatedly reproduced the appendages which first follow the Nauplius-limbs.

Of the Copepoda—some of which, living in a free state, people the fresh waters, and in far more multifarious forms the sea, whilst others, as parasites, infest animals of the most various classes and often become wonderfully deformed—the developmental history, like their entire natural history, was, until lately, in a very unsatisfactory state. It is true, that we long ago knew that the Cyclopes of our fresh waters were excluded in the Nauplius-form, and that we were acquainted with some others of their young states; we had learnt, through Nordmann, that the same earliest form belonged to several parasitic Crustacea, which had previously passed, almost universally, as worms; but the connecting intermediate forms which would have permitted us to refer the regions of the body and the limbs of the larvae to those of the adult animal, were wanting. The comprehensive and careful investigations of Claus have filled up this deficiency in our knowledge, and rendered the section of the Copepoda one of the best known in the whole class. The following statements are derived from the works of this able naturalist. From the abundance of valuable materials which they contain I select only those which are indispensable for the comprehension of the development of the Crustacea in general, because, in what relates to the Copepoda in particular, the facts have already been placed in the proper light by the representation of their most recent investigator, and must appear to any one whose eyes are open, as important evidence in favour of the Darwinian theory.* (* I am still unacquainted with Claus' latest and larger work, but no doubt the same may be said of it.)

(FIGURES 53 AND 54. Nauplii of Copepoda, the former magnified 90, the latter 180 diam.)

All the larvae of the free Copepoda investigated by Claus, have, at the earliest period, three pairs of limbs (the future antennae and mandibles), the anterior with a single, and the two following ones with a double series of joints, or branchiae. The unpaired eye, labrum, and mouth, already occupy their permanent positions. The posterior portion, which is usually short and destitute of limbs, bears two terminal setae, between which the anus is situated. The form in this Nauplius-brood is extremely various,—it is sometimes compressed laterally, sometimes flat,—sometimes elongated, sometimes oval, sometimes round or even broader than long, and so forth. The changes which the first larval stages undergo during the progress of growth, consist essentially in an extension of the body and the sprouting forth of new limbs. "The following stage already displays a fourth pair of extremities, the future maxillae." Then follow at once three new pairs of limbs (the maxillipedes and the two anterior pairs of natatory feet). The larva still continues like a Nauplius, as the three anterior pairs of limbs represent rowing feet; at the next moult it is converted into the youngest Cyclops-like state, when it resembles the adult animal in the structure of the antennae and buccal organs, although the number of limbs and body segments is still much less, for only the rudiments of the third and fourth pairs of natatory feet have made their appearance in the form of cushions fringed with setae, and the body consists of the oval cephalothorax, the second, third, and fourth thoracic segments, and an elongated terminal joint. In the Cyclopidae the posterior antennae have lost their secondary branch, and the mandibles have completely thrown off the previously existing natatory feet, whilst in the other families these appendages persist, more or less altered. "Beyond this stage of free development, many forms of the parasitic Copepoda, such as Lernanthropus and Chondracanthus, do not pass, as they do not acquire the third and fourth pairs of limbs, nor does a separation of the fifth thoracic segment from the abdomen take place; others (Achtheres) even fall to a lower grade by the subsequent loss of the two pairs of natatory feet. But all free Copepoda, and most of the parasitic Crustacea, pass through a longer or shorter series of stages of development, in which the limbs acquire a higher degree of division into joints in continuous sequence, the posterior pairs of feet are developed, and the last thoracic segment and the different abdominal segments are successively separated from the common terminal portion."

(FIGURE 55. Nauplius of Tetraclita porosa after the first moult, magnified 90 diam. The brain is seen surrounding the eye, and from it the olfactory filaments issue; behind it are some delicate muscles passing to the buccal hood.)

There is only one thing more to be indicated in the developmental history of the parasitic Crustacea, namely that some of them, such as Achtheres percarum, certainly quit the egg like the rest in a Nauplius-like form, inasmuch as the plump, oval, astomatous body bears two pairs of simple rowing feet, and behind these, as traces of the third pair, two inflations furnished each with a long seta, but that beneath this Nauplius-skin a very different larva lies ready prepared, which in a few hours bursts its clumsy envelope and then makes its appearance in a form "which agrees in the segmentation of the body and in the development of the extremities with the first Cyclops-stage" (Claus). The entire series of Nauplius-stages which are passed through by the free Copepoda, are in this case completely over-leapt.

A final and very peculiar section of the Crustacea is formed by the two orders of the Cirripedia and Rhizocephala.* (* The most various opinions prevail as to the position of the Cirripedia. Some ascribe to them a very subordinate position among the Copepoda; as Milne-Edwards (1852). In direct opposition to this notion of his father's, Alph. Milne-Edwards places them (as Basinotes) opposite to all the other Crustacea (Eleutheronotes). Darwin regards them as forming a peculiar sub-class equivalent to the Podophthalma, Edriophthalma, etc. This appears to me to be most convenient. I would not combine the Rhizocephala with the Cirripedia, as Liljeborg has done, but place them in opposition as equivalent, like the Amphipoda and Isopoda. The near relationship of the Cirripedia to the Ostracoda is also spoken of, but the similarity of the so-called "Cypris-like larvae," or Cirriped-pupae as Darwin calls them, to Cypris is so purely external, even as regards the shell, that the relationship appears to me to be scarcely greater than that of Peltogaster socialis (Figure 59) with the family of the sausages.)

In these also the brood bursts out in the Nauplius-form, and speedily strips off its earliest larva-skin which is distinguished by no peculiarities worth noticing. Here also we find again the same pyriform shape of the unsegmented body, the same number and structure of the feet, the same position of the median eye (which, however, is wanting in Sacculina purpurea, and according to Darwin in some species of Lepas), and the same position of the "buccal hood," as in the Nauplii of the Prawns and Copepoda. From the latter the Nauplii of the Cirripedia and Rhizocephala are distinguished by the possession of a dorsal shield or carapace, which sometimes (Sacculina purpurea) projects far beyond the body all round; and they are distinguished not only from other Nauplii, but as far as I know from all other Crustacea, by the circumstance that structures which are elsewhere combined with the two anterior limbs (antennae), here occur separated from them.

The anterior antennae of the Copepoda, Cladocera, Phyllopoda (Leydig, Claus), Ostracoda (at least the Cypridinae), Diastylidae, Edriophthalma, and Podophthalma, with few exceptions relating to terrestrial animals or parasites, bear peculiar filaments which I have already repeatedly mentioned as "olfactory filaments." A pair of similar filaments spring, in the larvae of the Cirripedia and Rhizocephala, directly from the brain.

(FIGURE 56. Nauplius of Sacculina purpurea, shortly before the second moult, magnified 180 diam. We may recognise in the first pair of feet the future adherent feet, and in the abdomen six pairs of natatory feet with long setae.)

At the base of the inferior antennae in the Decapoda the so-called "green-gland" has its opening; in the Macrura at the end of a conical process. A similar conical process with an efferent duct traversing it is very striking in most of the Amphipoda. In the Ostracoda, Zenker describes a gland situated in the base of the inferior antennae, and opening at the extremity of an extraordinarily long "spine." In the Nauplii of Cyclops and Cyclopsine, Claus finds pale "shell-glands," which commence in the intermediate pair of limbs (the posterior antennae). On the other hand in the Nauplii of the Cirripedia and Rhizocephala the "shell-glands" open at the ends of conical processes, sometimes of most remarkable length, which spring from the angles of the broad frontal margin, and have been interpreted sometimes as antennae (Burmeister, Darwin) and sometimes as mere "horns of the carapace" (Krohn). The connexion of the "shell-glands" with the frontal horns has been recognised unmistakably in the larvae of Lepas, and indeed the resemblance of the frontal horns with the conical processes on the inferior antennae of the Amphipoda, is complete throughout.* (* In connexion with this it may be mentioned that, in the females of Brachyscelus, in which the posterior antennae are deficient, the conical processes with the canal permeating them are nevertheless retained.)

(FIGURE 57. Pupa of a Balanide (Chthamalus ?), magnified 50 diam. The adherent feet are retracted within the rather opaque anterior part of the shell.

FIGURE 58. Pupa of Sacculina purpurea, magnified 180 diam. The filaments on the adherent feet may be the commencements of the future roots.)

Notwithstanding their agreement in this important peculiarity, the Nauplii of these two orders present material differences in many other particulars. The abdomen of the young Cirripede is produced beneath the anus into a long tail-like appendage which is furcate at the extremity, and over the anus there is a second long, spine-like process; the abdomen in the Rhizocephala terminates in two short points,—in a "moveable caudal fork, as in the Rotatoria," (O. Schmidt). The young Cirripedes have a mouth, stomach, intestine, and anus, and their two posterior pairs of limbs are beset with multifarious teeth, setae, and hooks, which certainly assist in the inception of nourishment. All this is wanting in the young Rhizocephala. The Nauplii of the Cirripedia have to undergo several moults whilst in that form; the Nauplii of the Rhizocephala, being astomatous, cannot of course live long as Nauplii, and in the course of only a few days they become transformed into equally astomatous "pupae," as Darwin calls them.

The carapace folds itself together, so that the little animal acquires the aspect of a bivalve shell, the foremost limbs become transformed into very peculiar adherent feet ("prehensile antennae," Darwin), and the two following pairs are cast off; like the frontal horns. On the abdomen six pairs of powerful biramose natatory feet with long setae have been formed beneath the Nauplius-skin, and behind these are two short, setigerous caudal appendages (Figure 58).

The pupae of the Cirripedia (Figure 57), which are likewise astomatous, agree completely in all these parts with those of the Rhizocephala, even to the minutest details of the segmentation and bristling of the natatory feet;* they are especially distinguished from them by the possession of a pair of composite eyes. (* Compare the figure given by Darwin (Balanidae Plate 30 Figure 5) of the first natatory foot of the pupa of Lepas australis, with that of Lernaeodiscus Porcellanae published in the 'Archiv fur Naturgeschichte' (1863 Taf 3 Figure 5). The sole distinction, that in the latter there are only 3 setae at the end of the outer branch, whilst in the Cirripedia there are 4 on the first and 5 on the following natatory feet, may be due to an error on my part.) Sometimes also traces of the frontal horns seem to persist.* (* Darwin describes as "acoustic orifices" small apertures in the shell of the pupae of the Cirripedia, which, frequently surrounded by a border, are situated, in Lepas pectinata, upon short, horn-like processes. I feel scarcely any hesitation in regarding the apertures as those of the "shell-glands," and the horn-like processes as remains of the frontal horns.)

As the Cirripedia and Rhizocephala now in general resemble each other far more than in their Nauplius-state, this is also the case with the individual members of each of the two orders.

The pupae in both orders attach themselves by means of the adherent feet; those of the Cirripedes to rocks, shells, turtles, drift-wood, ships, etc.,—those of the Rhizocephala to the abdomen of Crabs, Porcellanae, and Hermit Crabs. The carapace of the Cirripedes becomes converted, as is well-known, into a peculiar test, on account of which they were formerly placed among the Mollusca, and the natatory feet grow into long cirri, which whirl nourishment towards the mouth, which is now open. The Rhizocephala remain astomatous; they lose all their limbs completely, and appear as sausage-like, sack-shaped or discoidal excrescences of their host, filled with ova (Figures 59 and 60); from the point of attachment closed tubes, ramified like roots, sink into the interior of the host, twisting round its intestine, or becoming diffused among the sac-like tubes of its liver. The only manifestations of life which persist in these non plus ultras in the series of retrogressively metamorphosed Crustacea, are powerful contractions of the roots, and an alternate expansion and contraction of the body, in consequence of which water flows into the brood-cavity and is again expelled, through a wide orifice.* (* The roots of Sacculina purpurea (Figure 60) which is parasitic upon a small Hermit Crab, are made use of by two parasitic Isopods, namely a Bopyrus and the before mentioned Cryptoniscus planarioides (Figure 42). These take up their abode beneath the Sacculina and cause it to die away by intercepting the nourishment conveyed by the roots; the roots, however, continue to grow, even without the Sacculina, and frequently attain an extraordinary extension, especially when a Bopyrus obtains its nourishment from them.)

(FIGURE 59. Young of Peltogaster socialis on the abdomen of a small Hermit Crab; in one of them the fasciculately ramified roots in the liver of the Crab are shown. Animal and roots deep yellow.

FIGURE 60. Young Sacculina purpurea with its roots; the animal purple-red, the roots dark grass-green. Magnified 5 diam.)

Out of several Cirripedes, which are anomalous both in structure and development, Cryptophialus minutus must be mentioned here; Darwin found it in great quantities together in the shell of Concholepas peruviana on the Chonos Islands. The egg, which is at first elliptical, soon, according to Darwin, becomes broader at the anterior extremity, and acquires three club-shaped horns, one at each anterior angle and one behind; no internal parts can as yet be detected. Subsequently the posterior horn disappears, and the adherent feet may be recognised within the anterior ones. From this "egg-like larva"—(Darwin says of it, "I hardly know what to call it")—the pupa is directly produced. Its carapace is but slightly compressed laterally and hairy, as in Sacculina purpurea; the adherent feet are of considerable size, and the natatory feet are wanting, as, in the adult animal, are the corresponding cirri. As I learn from Mr. Spence Bate, the Nauplius-stage appears to be overleaped and the larvae to leave the egg in the pupa-form, in the case of a Rhizocephalon (Peltogaster ?) found by Dr. Powell in the Mauritius.

(FIGURES 61 TO 63. Eggs of Tetraclita porosa in segmentation, magnified 90 diam. The larger of the two first-formed spheres of segmentation is always turned towards the pointed end of the egg.

FIGURE 64. Egg of Lernaeodiscus Porcellanae, in segmentation, magnified 90 diam.)

I will conclude this general view with a few words upon the earliest processes in the development of the Crustacea. Until recently it was regarded as a general rule that, by the partial segmentation of the vitellus a germinal disc was formed, and in this, corresponding to the ventral surface of the embryo, a primitive band. We now know that in the Copepoda (Claus), in the Rhizocephala (Figure 64), and, as I can add, in the Cirripedia (Figures 61 to 63) the segmentation is complete, and the embryos are sketched out in their complete form without any preceding primitive band. Probably the latter will always be the case where the young are hatched as true Nauplii (and not merely with a Nauplius-skin, as in Achtheres). The two modes of development may occur in very closely allied animals, as is proved by Achtheres among the Copepoda.* (* I have not mentioned the Pycnogonidae, because I do not regard them as Crustacea; nor the Xiphosura and Trilobites, because, having never investigated them myself, I knew too little about them, and especially because I am unacquainted with the details of the explanations given by Barrande of the development of the latter. According to Mr. Spence Bate "the young of Trilobites are of the Nauplius-form.")

CHAPTER 10. ON THE PRINCIPLES OF CLASSIFICATION.

Perhaps some one else, more fortunate than myself, may be able, even without Darwin, to find the guiding clue through the confusion of developmental forms, now so totally different in the nearest allies, now so surprisingly similar in members of the most distant groups, which we have just cursorily reviewed. Perhaps a sharper eye may be able, with Agassiz, to make out "the plan established from the beginning by the Creator,"* (* "A plan fully matured in the beginning and undeviatingly pursued;" or "In the beginning His plan was formed and from it He has never swerved in any particular" (Agassiz and Gould, 'Principles of Zoology').) who may have written here, as a Portuguese proverb says "straight in crooked lines."* (* "Deos escrive direito em linhas tortas." To read this remarkable writing we need the spectacles of Faith, which seldom suit eyes accustomed to the Microscope.) I cannot but think that we can scarcely speak of a general plan, or typical mode of development of the Crustacea, differentiated according to the separate Sections, Orders, and Families, when, for example, among the Macrura, the River Crayfish leaves the egg in its permanent form; the Lobster with Schizopodal feet; Palaemon, like the Crabs, as a Zoea; and Peneus, like the Cirripedes, as a Nauplius,—and when, still, within this same sub-order Macrura, Palinurus, Mysis and Euphausia again present different young forms,—when new limbs sometimes sprout forth as free rudiments on the ventral surface, and are sometimes formed beneath the skin which passes smoothly over them, and both modes of development are found in different limbs of the same animal and in the same pair of limbs in different animals,—when in the Podophthalma the limbs of the thorax and abdomen make their appearance sometimes simultaneously, or sometimes the former and sometimes the latter first, and when further in each of the two groups the pairs sometimes all appear together, and sometimes one after the other,—when, among the Hyperina, a simple foot becomes a chela in Phronima and a chela a simple foot in Brachyscelus, etc.

And yet, according to the teaching of the school, it is precisely in youth, precisely in the course of development, that the "Type" is mostly openly displayed. But let us hear what the Old School has to tell us as to the significance of developmental history, and its relation to comparative anatomy and systematic zoology.

Let two of its most approved masters speak.

"Whilst comparative anatomy," said Johannes Muller, in 1844, in his lectures upon this science (and the opinions of my memorable teacher were for many years my own), "whilst comparative anatomy shows us the infinitely multifarious formation of the same organ in the Animal Kingdom, it furnishes us at the same time with the means, by the comparison of these various forms, of recognising the truly essential, the type of these organs, and separating therefrom everything unessential. In this, developmental history serves it as a check or test. Thus, as the idea of development is not that of mere increase of size, but that of progress from what is not yet distinguished, but which potentially contains the distinction in itself, to the actually distinct,—it is clear, that the less an organ is developed, so much the more does it approach the type, and that, during its development, it more and more acquires peculiarities. The types discovered by comparative anatomy and developmental history must therefore agree."

Then, after Johannes Muller has combated the idea of a graduated scale of animals, and of the passage through several animal grades during development, he continues:—"What is true in this idea is, that every embryo at first bears only the type of its section, from which the type of the Class, Order, etc., is only afterwards developed."

In 1856, in an elementary work,* (* 'Principles of Zoology' Part 1 Comparative Physiology. By Louis Agassiz and A.A. Gould Revised Edition Boston 1856.) in which it is usual to admit only what are regarded as the assured acquisitions of science, Agassiz expresses himself as follows:—

"The ovarian eggs of all animals are perfectly identical, small cells with a vitellus, germinal vesicle and germinal spot" (paragraph 278). "The organs of the body are formed in the sequence of their organic importance; the most essential always appear first. Thus the organs of vegetative life, the intestine, etc., appear later than those of animal life, the nervous system, skeleton, etc., and these in turn are preceded by the more general phenomena belonging to the animal as such" (paragraph 318). "Thus, in Fishes, the first changes consist in the segmentation of the vitellus and the formation of a germ, processes which are common to all classes of animals. Then the dorsal furrow, characteristic of the Vertebrate, appears—the brain, the organs of the senses; at a later period are formed the intestine, the limbs, and the permanent form of the respiratory organs, from which the class is recognised with certainty. It is only after exclusion that the peculiarities of the structure of the teeth and fins indicate the genus and species" (paragraph 319). "Hence the embryos of different animals resemble each other the more, the younger they are" (paragraph 320). "Consequently the high importance of developmental history is indubitable. For, if the formation of the organs takes place in the order corresponding to their importance, this sequence must of itself be a criterion of their comparative value in classification. The peculiarities which appear earlier should be considered of higher value than those which appear subsequently" (paragraph 321). "A system, in order to be true and natural, must agree with the sequence of the organs in the development of the embryo" (paragraph 322).

I do not know whether any one at the present day will be inclined to subscribe to this proposition in its whole extent.* (* Agassiz' own views have lately become essentially different, so far as can be made out from Rud. Wagner's notice of his 'Essay on Classification.' Agassiz himself does not attempt any criticism of the above cited older views, which, however, are still widely diffused. With his recent conception I am unfortunately acquainted only from R. Wagner's somewhat confused report, and have therefore thought it better not to attempt any critical remarks upon it.) It is certain, however, that views essentially similar are still to be met with everywhere in discussions on classification, and that even within the last few years, the very sparingly successful attempts to employ developmental history as the foundation of classification have been repeated.

But how do these propositions agree with our observations on the developmental history of the Crustacea? That these observations relate for the most part to their "free metamorphosis" after their quitting the egg, cannot prejudice their application to the propositions enunciated especially with regard to "embryonal development" in the egg; for Agassiz himself points out (paragraph 391) that both kinds of change are of the same nature and of equal importance and that no "radical distinction" is produced by the circumstance that the former take place before and the latter after birth.

"The ovarian eggs of all animals are identical, small cells with vitellus, germinal vesicle and germinal spot." Yes, somewhat as all Insects are identical, small animals with head, thorax, and abdomen; that is to say if, only noticing what is common to them, we leave out of consideration the difference of their development, the presence or absence and the multifarious structure of the vitelline membrane, the varying composition of the vitellus, the different number and formation of the germinal spots, etc. Numerous examples, which might easily be augmented, of such profound differences, are furnished by Leydig's 'Lehrbuch der Histologie.' In the Crustacea the ovarian eggs actually sometimes furnish excellent characters for the discrimination of species of the same genus; thus, for example, in one Porcellana of this country they are blackish-green, in a second deep blood-red, and in a third dark yellow; and within the limits of the same order they present considerable differences in size, which, as Van Beneden and Claus have already pointed out, stands in intimate connexion with the subsequent mode of development.

"The organs of the body are formed in the sequence of their organic importance; the most essential always appear first." This proposition might be characterised a priori as undemonstrable, since it is impossible either in general, or for any particular animal, to establish a sequence of importance amongst equally indispensable parts. Which is the more important, the lung or the heart—the liver or the kidney?—the artery or the vein? Instead of giving the preference, with Agassiz, to the organs of animal life, we might with equal justice give it to those of vegetative life, as the latter are conceivable without the former, but not the former without the latter. We might urge that, according to this proposition, provisional organs as the first produced must exceed the later-formed permanent organs in importance.

But let us stick to the Crustacea. In Polyphemus Leydig finds the first traces of the intestinal tube even during segmentation. In Mysis a provisional tail is first formed, and in Ligia a maggot-like larva-skin. The simple median eye appears earlier, and would therefore be more important than the compound paired eyes; the scale of the antennae in the Prawns would be more important than the flagellum; the maxillipedes of the Decapoda would be more important than the chelae and ambulatory feet, and the anterior six pairs of feet in the Isopoda, than the precisely similarly formed seventh pair; in the Amphipoda the most important of all organs would be the "micropylar apparatus," which disappears without leaving a trace soon after hatching; in Cyclops the setae of the tail would be more important than all the natatory feet; in the Cirripedia the posterior antennae, as to which we do not know what becomes of them, would be more important than the cirri, and so forth. The most unimportant of all organs would be the sexual organs, and the most essential peculiarity would consist in colour, which is to be referred back to the ovarian egg.

"The embryos, or young states of different animals, resemble each other the more, the younger they are," or, as Johannes Muller expresses it, "they approach the more closely to the common type." Different as may be the ideas connected with the word "type," no one will dispute that the typical form of the penultimate pair of feet in the Amphipoda is that of a simple ambulatory foot, and not that of a chela, for the latter occurs in no single adult Amphipod; we know it only in the young of the genus Brachyscelus, which therefore in this respect undoubtedly depart more widely than the adults from the type of their order. This applies also to the young males of the Shore-hoppers (Orchestia) with regard to the second pair of anterior feet (gnathopoda). In like manner no one will hesitate to accept the possession of seven pairs of feet as a "typical" peculiarity of the Edriophthalma, which Agassiz, on this account, names Tetradecapoda; the young Isopoda, which are Dodecapoda, are also in this respect further from the "type" than the adults.

It is certainly a rule, and this Darwin's theory would lead us to expect, that in the progress of development those forms which are at first similar gradually depart further from each other; but here, as in other classes, the exceptions, for which the Old School has no explanation, are numerous. Not unfrequently we might indeed directly reverse the proposition and assert that the difference becomes the greater, the further we go back in the development, and this not only in those cases in which one of two nearly allied species is directly developed, and the other passes through several larval stages, such as the common Crayfish and the Prawns which are produced from Nauplius-brood. The same may be said, for example, of the Isopoda and Amphipoda. In the adult animals the number of limbs is the same; at the first sight of a Cyrtophium or a Dulichia, and even after the careful examination of a Tanais, we may be in doubt whether we have an Isopod or an Amphipod before us; in the newly-hatched young the number of limbs is different, and if we go back to their existence in the egg, the most passing glance to see whether the curvature is upwards or downwards suffices to distinguish even the youngest embryos of the two orders.

In other instances, the courses which lead from a similar starting-point to a similar goal, separate widely in the middle of the development, as in the Prawns with Nauplius-brood already described.

Finally, so that even the last possibility may be exhausted, it sometimes happens that the greatest similarity occurs in the middle of the development. The most striking example of this is furnished by the Cirripedia and Rhizocephala, whether we compare the two orders or the members of each with one another; from a segmentation quite different in its course (see Figures 61 to 64) proceed different forms of Nauplius, these become converted into exceedingly similar pupae, and from the pupae again proceed sexually mature animals, differing from each other toto coelo.

"If the formation of the organs occurs in the order corresponding to their importance, this sequence must of itself be a criterion of their comparative value in classification." THAT IS TO SAY, SUPPOSING THE PHYSIOLOGICAL AND CLASSIFICATIONAL VALUE OF AN ORGAN TO COINCIDE! Just as in Christian countries there is a catechismal morality, which every one has upon his lips, but no one considers himself bound to follow, or expects to see followed by anybody else, so also has Zoology its dogmas, which are as universally acknowledged, as they are disregarded in practice. Such a dogma as this is the supposition tacitly made by Agassiz. Of a hundred who feel themselves compelled to give their systematic confession of faith as the introduction to a Manual or Monographic Memoir, ninety-nine will commence by saying that a natural system cannot be founded upon a single character, but that it has to take into account all characters, and the general structure of the animal, but that we must not simply sum up these characters like equivalent magnitudes, that we must not count but weigh them, and determine the importance to be ascribed to each of them according to its physiological significance. This is probably followed by a little jingle of words in general terms on the comparative importance of animal and vegetative organs, circulation, respiration, and the like. But when we come to the work itself, to the discrimination and arrangement of the species, genera, families, etc., in all probability not one of the ninety-nine will pay the least attention to these fine rules, or undertake the hopeless attempt to carry them out in detail. Agassiz, for example, like Cuvier, and in opposition to the majority of the German and English zoologists, regards the Radiata as one of the great primary divisions of the Animal Kingdom, although no one knows anything about the significance of the radiate arrangement in the life of these animals, and notwithstanding that the radiate Echinodermata are produced from bilateral larvae. The "true Fishes" are divided by him into Ctenoids and Cycloids, according as the posterior margin of their scales is denticulated or smooth, a circumstance the importance of which to the animal must be infinitely small, in comparison to the peculiarities of the dentition, formation of the fins, number of vertebrae, etc.

And, to return to our Class of the Crustacea, has any particular attention been paid in their classification to the distinctions prevailing in the "most essential organs"? For instance, to the nervous system? In the Corycaeidae, Claus found all the ventral ganglia fused together into a single broad mass, and in the Calanidae a long ventral chain of ganglia,—the former, therefore, in this respect resembling the Spider Crabs and the latter the Lobster; but no one would dream on this account of supposing that there was a relationship between the Corycaeidae and the Crabs, or the Calanidae and the Lobsters.—Or to the organs of circulation? We have among the Copepoda, the Cyclopidae and Corycaeidae without a heart, side by side with the Calanidae and Pontellidae with a heart. And in the same way among the Ostracoda, the Cypridinae, which I find possess a heart, place themselves side by side with Cypris and Cythere which have no such organ.—Or to the respiratory apparatus? Milne-Edwards did this when he separated Mysis and Leucifer from the Decapoda, but he himself afterwards saw that this was an error. In one Cypridina I find branchiae of considerable size, which are entirely wanting in another species, but this does not appear to me to be a reason for separating these species even generically.

On the other hand, what do we know of the physiological significance of the number of segments, and all the other matters which we are accustomed to regard as typical peculiarities of the different organs, and to which we usually ascribe the highest systematic value?

"Those peculiarities which first appear, should be more highly estimated than those which appear subsequently. A system, in order to be true and natural, must agree with the sequence of the organs in the development of the embryo." If the earlier manifested peculiarities are to be estimated more highly than those which afterwards make their appearance, then in those cases in which the structure of the adult animal requires one position in the system, and that of the larva another, the latter and not the former must decide the point. As the Lernaeae and Cirripedes, on account of their Nauplius-brood, were separated from their previous connexions and referred to the Crustacea, we shall, for the same reason, have to separate Peneus from the Prawns and unite it with the Copepoda and Cirripedia. But the most zealous embryomaniac would probably shrink from this course.

A "true and natural system" of the Crustacea to be in accordance with the sequence of the phenomena would have to take into account in the first place the various modes of segmentation, then the position of the embryo, next the number of limbs produced within the egg and so forth, and might be represented somewhat as follows:—

CLASSIS CRUSTACEA.

Sub-class I. HOLOSCHISTA.—Segmentation complete. No primitive band. Nauplius-brood.

Ord. 1. Ceratometopa.—Nauplius with frontal horns. (Cirripedia, Rhizocephala.)

Ord. 2. LEIOMETOPA.—Nauplius without frontal horns. (Copepoda, without Achtheus, etc., Phyllopoda, Peneus.)

Sub-class II. HEMISCHISTA.—Segmentation not complete.

A. Nototropa.—Embryo bent upwards.

Ord. 3. Protura.—The tail is first formed. (Mysis.)

Ord. 3. Saccomorpha.—A maggot-like larva-skin is first formed. (Isopoda.)

B. Gasterotropa.—Embryo bent ventrally.

Ord. 5. Zoeogona.—Full number of limbs not produced in the egg. Zoea-brood. (The majority of the Podophthalmata.)

Ord. 6. Ametabola.—Full number of limbs produced in the egg. (Astacus, Gecarcinus, Amphipoda less Hyperia ?)

This sample may suffice. The farther we go into details in this direction, the more brilliantly, as may easily be imagined, does the naturalness of such an arrangement as this force itself upon us.

All things considered, we may apply the judgment which Agassiz pronounced upon Darwin's theory, with far greater justice to the propositions just examined:—"No theory," says he, "however plausible it may be, can be admitted in science, unless it is supported by facts."

CHAPTER 11. ON THE PROGRESS OF EVOLUTION.

From this scarcely unavoidable but unsatisfactory side-glance upon the old school, which looks down with so great an air of superiority upon Darwin's "intellectual dream" and the "giddy enthusiasm" of its friends, I turn to the more congenial task of considering the developmental history of the Crustacea from the point of view of the Darwinian theory.

Darwin himself, in the thirteenth chapter of his book, has already discussed the conclusions derived from his hypotheses in the domain of developmental history. For a more detailed application of them, however, it is necessary in the first place to trace these general conclusions a little further than he has there done.

The changes by which young animals depart from their parents, and the gradual accumulation of which causes the production of new species, genera, and families, may occur at an earlier or later period of life,—in the young state, or at the period of sexual maturity. For the latter is by no means always, as in the Insecta, a period of repose; most other animals even then continue to grow and to undergo changes. (See above, the remarks on the males of the Amphipoda.) Some variations, indeed, from their very nature, can only occur when the young animal has attained the adult stage of development. Thus the Sea Caterpillars (Polynoe) at first possess only a few body-segments, which, during development, gradually increase to a number which is different in different species, but constant in the same species; now before a young animal could exceed the number of segments of its parents, it must of course have attained that number. We may assume a similar supplementary progress wherever the deviation of the descendants consists in an addition of new segments and limbs.

Descendants therefore reach a new goal, either by deviating sooner or later whilst still on the way towards the form of their parents, or by passing along this course without deviation, but then, instead of standing still, advance still farther.

The former mode will have had a predominant action where the posterity of common ancestors constitutes a group of forms standing upon the same level in essential features, as the whole of the Amphipoda, Crabs, or Birds. On the other hand we are led to the assumption of the second mode of progress, when we seek to deduce from a common original form, animals some of which agree with young states of others.

In the former case the developmental history of the descendants can only agree with that of their ancestors up to a certain point at which their courses separate,—as to their structure in the adult state it will teach us nothing. In the second case the entire development of the progenitors is also passed through by the descendants, and, therefore, so far as the production of a species depends upon this second mode of progress, the historical development of the species will be mirrored in its developmental history. In the short period of a few weeks or months, the changing forms of the embryo and larvae will pass before us, a more or less complete and more or less true picture of the transformations through which the species, in the course of untold thousands of years, has struggled up to its present state.

(FIGURES 65 TO 67. Young Tubicolar worms, magnified with the simple lens about 6 diam.:

FIGURE 65.* Without operculum, Protula-stage. (* Figure 65 is drawn from memory, as the little animals, which I at first took for young Protulae, only attracted my attention when I remarked the appearance of the operculum, which induced me to draw them.)

FIGURE 66. With a barbate opercular peduncle, Filograna-stage;

FIGURE 67. With a naked opercular peduncle, Serpula-stage.)

One of the simplest examples is furnished by the development of the Tubicolar Annelids; but from its very simplicity it appears well adapted to open the eyes of many who, perhaps, would rather not see, and it may therefore find a place here. Three years ago I found on the walls of one of my glasses some small worm-tubes (Figure 65), the inhabitants of which bore three pairs of barbate branchial filaments, and had no operculum. According to this we should have been obliged to refer them to the genus Protula. A few days afterwards one of the branchial filaments had become thickened at the extremity into a clavate operculum (Figure 66), when the animals reminded me, by the barbate opercular peduncle, of the genus Filograna, only that the latter possesses two opercula. In three days more, during which a new pair of branchial filaments had sprouted forth, the opercular peduncle had lost its lateral filaments (Figure 67), and the worms had become Serpulae. Here the supposition at once presents itself that the primitive tubicolar worm was a Protula,—that some of its descendants, which had already become developed into perfect Protulae, subsequently improved themselves by the formation of an operculum which might protect their tubes from inimical intruders,—and that subsequent descendants of these latter finally lost the lateral filaments of the opercular peduncle, which they, like their ancestors, had developed.

What say the schools to this case? Whence and for what purpose, if the Serpulae were produced or created as ready-formed species, these lateral filaments of the opercular peduncle? To allow them to sprout forth merely for the sake of an invariable plan of structure, even when they must be immediately retracted again as superfluous, would certainly be an evidence rather of childish trifling or dictatorial pedantry, than of infinite wisdom. But no, I am mistaken; from the beginning of all things the Creator knew, that one day the inquisitive children of men would grope about after analogies and homologies, and that Christian naturalists would busy themselves with thinking out his Creative ideas; at any rate, in order to facilitate the discernment by the former that the opercular peduncle of the Serpulae is homologous with a branchial filament, He allowed it to make a detour in its development, and pass through the form of a barbate branchial filament.

The historical record preserved in developmental history is gradually EFFACED as the development strikes into a constantly straighter course from the egg to the perfect animal, and it is frequently SOPHISTICATED by the struggle for existence which the free-living larvae have to undergo.

Thus as the law of inheritance is by no means strict, as it gives room for individual variations with regard to the form of the parents, this is also the case with the succession in time of the developmental processes. Every father of a family who has taken notice of such matters, is well aware that even in children of the same parents, the teeth, for example, are not cut or changed, either at the same age, or in the same order. Now in general it will be useful to an animal to obtain as early as possible those advantages by which it sustains itself in the struggle for existence. A precocious appearance of peculiarities originally acquired at a later period will generally be advantageous, and their retarded appearance disadvantageous; the former, when it appears accidentally, will be preserved by natural selection. It is the same with every change which gives to the larval stages, rendered multifarious by crossed and oblique characters, a more straightforward direction, simplifies and abridges the process of development, and forces it back to an earlier period of life, and finally into the life of the egg.

As this conversion of a development passing through different young states into a more direct one, is not the consequence of a mysterious inherent impulse, but dependent upon advances accidentally presenting themselves, it may take place in the most nearly allied animals in the most various ways, and require very different periods of time for its completion. There is one thing, however, that must not be overlooked here. The historical development of a species can hardly ever have taken place in a continuously uniform flow; periods of rest will have alternated with periods of rapid progress. But forms, which in periods of rapid progress were severed from others after a short duration, must have impressed themselves less deeply upon the developmental history of their descendants, than those which repeated themselves unchanged, through a long series of successive generations in periods of rest. These more fixed forms, less inclined to variation, will present a more tenacious resistance in the transition to direct development, and will maintain themselves in a more uniform manner and to the last, however different may be the course of this process in other respects.

In general, as already stated, it will be advantageous to the young to commence the struggle for existence in the form of their parents and furnished with all their advantages—in general, but not without exceptions. It is perfectly clear that a brood capable of locomotion is almost indispensable to attached animals, and that the larvae of sluggish Mollusca, or of worms burrowing in the ground, etc., by swarming briskly through the sea perform essential services by dispersing the species over wider spaces. In other cases a metamorphosis is rendered indispensable by the circumstance that a division of labour has been set up between the various periods of life; for example, that the larvae have exclusively taken upon themselves the business of nourishment. A further circumstance to be taken into consideration is the size of the eggs,—a simpler structure may be produced with less material than a more compound one,—the more imperfect the larva, the smaller may the egg be, and the larger is the number of these that the mother can furnish with the same expenditure of material. As a rule, I believe indeed, this advantage of a more numerous brood will not by any means outweigh that of a more perfect brood, but it will do so in those cases in which the chief difficulty of the young animals consists in finding a suitable place for their development, and in which, therefore, it is of importance to disperse the greatest possible number of germs, as in many parasites.

As the conversion of the original development with metamorphosis into direct development is here under discussion, this may be the proper place to say a word as to the already indicated absence of metamorphosis in fresh-water and terrestrial animals the marine allies of which still undergo a transformation. This circumstance seems to be explicable in two ways. Either species without a metamorphosis migrated especially into the fresh waters, or the metamorphosis was more rapidly got rid of in the emigrants than in their fellows remaining in the sea.

Animals without a metamorphosis would naturally transfer themselves more easily to a new residence, as they had only themselves and not at the same time multifarious young forms to adapt to the new conditions. But in the case of animals with a metamorphosis, the mortality among the larvae, always considerable, must have become still greater under new than under accustomed conditions, every step towards the simplification of the process of development must therefore have given them a still greater preponderance over their fellows, and the effacing of the metamorphosis must have gone on more rapidly. What has taken place in each individual case, whether the species has immigrated after it had lost the metamorphosis, or lost the metamorphosis after its immigration, will not always be easy to decide. When there are marine allies without, or with only a slight metamorphosis, like the Lobster as the cousin of the Cray-fish, we may take up the former supposition; when allies with a metamorphosis still live upon the land or in fresh water, as in the case of Gecarcinus, we may adopt the latter.

That besides this gradual extinction of the primitive history, a FALSIFICATION of the record preserved in the developmental history takes place by means of the struggle for existence which the free-living young states have to undergo, requires no further exposition. For it is perfectly evident that the struggle for existence and natural selection combined with this, must act in the same way, in change and development, upon larvae which have to provide for themselves, as upon adult animals. The changes of the larvae, independent of the progress of the adult animal, will become the more considerable, the longer the duration of the life of the larva in comparison to that of the adult animal, the greater the difference in their mode of life, and the more sharply marked the division of labour between the different stages of development. These processes have to a certain extent an action opposed to the gradual extinction of the primitive history; they increase the differences between the individual stages of development, and it will be easily seen how even a straightforward course of development may be again converted by them into a development with metamorphosis. By this means many, and it seems to me valid reasons may be brought up in favour of the opinion that the most ancient Insects approached more nearly to the existing Orthoptera, and perhaps to the wingless Blattidae, than to any other order, and that the "complete metamorphosis" of the Beetles, Lepidoptera, etc., is of later origin. There were, I believe, perfect Insects before larvae and pupae; but, on the contrary, Nauplii and Zoeae far earlier than perfect Prawns. In contradistinction to the INHERITED metamorphosis of the Prawns, we may call that of the Coleoptera, Lepidoptera, etc. an ACQUIRED metamorphosis.*

(* I will here briefly give my reasons for the opinion that the so-called "complete metamorphosis" of Insects, in which these animals quit the egg as grubs or caterpillars, and afterwards become quiescent pupae incapable of feeding, was not inherited from the primitive ancestor of all Insects, but acquired at a later period.

The order Orthoptera, including the Pseudoneuroptera (Ephemera, Libellula, etc.) appears to approach nearest to the primitive form of Insects. In favour of this view we have:—

1. The structure of their buccal organs, especially the formation of the labium, "which retains, either perfectly or approximately, the original form of a second pair of maxillae" (Gerstacker).

2. The segmentation of the abdomen; "like the labium, the abdomen also very generally retains its original segmentation, which is shown in the development of eleven segments" (Gerstacker). The Orthoptera with eleven segments in the abdomen, agree perfectly in the number of their body-segments with the Prawn-larva represented in Figure 33, or indeed, with the higher Crustacea (Podophthalma and Edriophthalma) in general, in which the historically youngest last thoracic segment (see page 123), which is sometimes late-developed, or destitute of appendages, or even deficient, is still wanting.

3. That, as in the Crustacea, the sexual orifice and anus are placed upon different segments; "whilst the former is situated in the ninth segment, the latter occurs in the eleventh" (Gerstacker).

4. Their palaeontological occurrence; "in a fossil state the Orthoptera make their appearance the earliest of all Insects, namely as early as the Carboniferous formation, in which they exceed all others in number" (Gerstacker).

5. The absence of uniformity of habit at the present day in an order so small when compared with the Coleoptera, Hymenoptera, etc. For this also is usually a phenomenon characteristic of very ancient groups of forms which have already overstepped the climax of their development, and is explicable by extinction in mass. A Beetle or a Butterfly is to be recognised as such at the first glance, but only a thorough investigation can demonstrate the mutual relationships of Termes, Blatta, Mantis, Forficula, Ephemera, Libellula, etc. I may refer to a corresponding remarkable example from the vegetable world: amongst Ferns the genera Aneimia, Schizaea and Lygodium, belonging to the group Schizaeaceae which is very poor in species, differ much more from each other than any two forms of the group Polypodiaceae which numbers its thousands of species.

If, from all this, it seems right to regard the Orthoptera as the order of Insects approaching most nearly to the common primitive form, we must also expect that their mode of development will agree better with that of the primitive form, than, for example, that of the Lepidoptera, in the same way that some of the Prawns (Peneus) approaching most closely the primitive form of the Decapoda, have most truly preserved their original mode of development. Now, the majority of the Orthoptera quit the egg in a form which is distinguished from that of the adult Insect almost solely by the want of wings; these larvae then soon acquire rudiments of wings, which appear more strongly developed after every moult. Even this perfectly gradual transition from the youngest larva to the sexually mature Insect, preserves in a far higher degree the picture of an original mode of development, than does the so-called complete metamorphosis of the Coleoptera, Lepidoptera, or Diptera, with its abruptly separated larva-, pupa- and imago-states.

The most ancient Insects would probably have most resembled these wingless larvae of the existing Orthoptera. The circumstance that there are still numerous wingless species among the Orthoptera, and that some of these (Blattidae) are so like certain Crustacea (Isopods) in habit that both are indicated by the same name ("Baratta") by the people in this country, can scarcely be regarded as of any importance.

The contrary supposition that the oldest Insects possessed a "complete metamorphosis," and that the "incomplete metamorphosis" of the Orthoptera and Hemiptera is only of later origin, is met by serious difficulties. If all the classes of Arthropoda (Crustacea, Insecta, Myriopoda and Arachnida) are indeed all branches of a common stem (and of this there can scarcely be a doubt), it is evident that the water-inhabiting and water-breathing Crustacea must be regarded as the original stem from which the other terrestrial classes, with their tracheal respiration, have branched off. But nowhere among the Crustacea is there a mode of development comparable to the "complete metamorphosis" of the Insecta, nowhere among the young or adult Crustacea are there forms which might resemble the maggots of the Diptera or Hymenoptera, the larvae of the Coleoptera, or the caterpillars of the Lepidoptera, still less any bearing even a distant resemblance to the quiescent pupae of these animals. The pupae, indeed, cannot at all be regarded as members of an original developmental series, the individual stages of which represent permanent ancestral states, for an animal like the mouthless and footless pupa of the Silkworm, enclosed by a thick cocoon, can never have formed the final, sexually mature state of an Arthropod.