CHAPTER V

REPRODUCTION--DEPOSITION OF EGGS--DEVELOPMENT OF THE FERTILISED OVUM--DIFFERENCES OF SEX--DIOECIOUS AND HERMAPHRODITE MOLLUSCA--DEVELOPMENT OF FRESH-WATER BIVALVES

Reproduction in the Mollusca invariably takes place by means of eggs, which, after being developed in the ovary of the female, are fertilised by the spermatozoa of the male. As a rule, the eggs are ‘laid,’ and undergo their subsequent development apart from the parent. This rule, however, has its exceptions, both among univalve and bivalve Mollusca, a certain number of which hatch their young from the egg before expelling them. Such ovoviviparous genera are _Melania_, _Paludina_, _Balea_, and _Coeliaxis_ among land and fresh-water Mollusca, and _Cymba_ and many _Littorina_ amongst marine. The young of _Melania tuberculata_, in Algeria, have been noticed to return, as if for shelter, to the branchial cavity of the mother, some days after first quitting it. Isolated species among Pulmonata are known to be ovoviviparous, _e.g._ _Patula Cooperi_, _P. Hemphilli_, and _P. rupestris_, _Acanthinula harpa_, _Microphysa vortex_, _Pupa cylindracea_ and _muscorum_, _Clausilia ventricosa_, _Opeas dominicensis_, _Rhytida inaequalis_, etc. All fresh-water Pelecypoda yet examined, except _Dreissensia_, are ovoviviparous.

The number of eggs varies greatly, being highest in the Pelecypoda. In _Ostrea edulis_ it has been estimated at from 300,000 to 60,000,000; in _Anodonta_ from 14,000 to 20,000; in _Unio pictorum_ 200,000. The eggs of _Doris_ are reckoned at from 80,000 to 600,000, of _Loligo_ and _Sepia_ at about 30,000 to 40,000. Pulmonata lay comparatively few eggs. _Arion ater_ has been observed to lay 477 in forty-eight days (p. 42). Nests of _Helix aspersa_ have been noticed, in which the number of eggs varied from about 40 to 100. They are laid in little cup-shaped hollows at the roots of grass, with a little loose earth spread over them. The eggs of _Testacella_ are rather large, and very elastic; if dropped on a stone floor they will rebound sharply several inches. The _Cochlostyla_ of the Philippines lay their eggs at the tops of the great forest trees, folding a leaf together to serve as a protection.

The eggs of the great tropical _Bulimus_ and _Achatina_, together with those of the _Macroön_ group of _Helix_ (_Helicophanta_, _Acavus_, _Panda_) are exceedingly large, and the number laid must be decidedly less than in the smaller Pulmonata. _Bulimus oblongus_, for instance, from Barbados, lays an egg about the size of a sparrow’s (Fig. 38), _Achatina sinistrorsa_ as large as a pigeon’s. The Cingalese _Helix Waltoni_ when first hatched is about the size of a full-grown _H. hortensis_. There is, in the British Museum, a specimen of the egg of a _Bulimus_ from S. America (probably _maximus_ or _popelairanus_) which measures exactly 1¾ inch in length.

The _Limnaeidae_ deposit their eggs in irregular gelatinous masses on the under side of the leaves of water-plants, and on all kinds of débris.

[Illustration: FIG. 38.--Newly-hatched young and egg of _Bulimus oblongus_ Müll., Barbados. _Natural size._]

The _Rachiglossa_ or marine carnivorous families lay their eggs in tough leathery or bladdery capsules, which are frequently joined together in shapes which differ with the genus. Each capsule contains a varying number of ova. The cluster of egg-capsules of _Buccinum undatum_ is a familiar object on all our sandy coasts. The capsules of _Purpura lapillus_ are like delicate pink grains of rice, set on tiny stalks. They are not attached to one another, but are set closely together in groups in sheltered nooks of the rocks. A single _Purpura_ has been observed to produce 245 capsules! _Busycon_ lays disc-shaped capsules which are all attached at a point in the edge to a cartilaginous band nearly 3 feet in length, looking like a number of coins tied to a string at equal distances from one another. In _Murex erinaceus_ the egg-capsules are triangular, with a short stalk. They are deposited separately in clusters of from 15 to 150, there being about 20 ova in each capsule. It appears that all the species of the same genus have by no means the same method of depositing their eggs, nor do they always produce eggs of at all similar size or shape. Thus, of two British species of _Nassa_, _N. reticulata_ lays egg-capsules in shape like flattened pouches with a short stalk, and fastens them in rows to the leaves of _Zostera_; _M. incrassata_, on the other hand, deposits solitary capsules, which are shaped like rounded oil-flasks. _Neptunea antiqua_ lays its eggs in bunched capsules, like _Bucc. undatum_ (Fig. 40), but the capsules of _N. gracilis_ are solitary.

[Illustration: FIG. 39.--Various forms of spawn in _Prosobranchiata_: =A= and =D=, _Pyrula_ or _Busycon_; =B=, _Conus_; =C=, _Voluta musica_; =E=, _Ampullaria_ (from specimens in the British Museum); all × ⅔.]

In _Natica_ the eggs are deposited in what looks like a thick piece of sand-paper, curled in a spiral form (Fig. 41). The sand is agglutinated by copious mucus into a sort of sheet, and the eggs are let into this, sometimes (_N. heros_) in regular quincunx form. _Ianthina_ attaches its eggs to the under side of its float (Fig. 42). The Trochidae deposit their eggs on the under side of stones and sea-weeds, each ovum being contained in a separate capsule, and all the capsules glued together into an irregular mass of varying size. The female of _Galerus chinensis_ hatches her eggs by keeping them between her foot and the stone she adheres to. They are laid in from 6 to 10 capsules, connected by a pedicle and arranged like the petals of a rose, with 10 to 12 eggs in each capsule. Those _Littorina_ which are not ovoviviparous deposit their spawn on sea-weeds, rocks, and stones. The eggs are enveloped in a glairy mass which is just firm enough to retain its shape in the water; each egg has its own globule of jelly and is separated from the others by a very thin transparent membrane.[242]

[Illustration: FIG. 40.--Egg-capsules of, =A=, _Nassa reticulata_ L. × ⅔; =B=, _Buccinum undatum_ L. × ⅔; =C=, _Neptunea antiqua_ L. × ⅓.]

[Illustration: FIG. 41.--Spawn of a species of _Natica_ (from a specimen in the British Museum) × ½.]

[Illustration: FIG. 42.--_Ianthina fragilis_ Lam. =FL=, float; =O=, ova; =Pr=, proboscis; =Br=, branchiae; =F=, foot. (Quoy and Gaimard.)]

_Chiton marginatus_, when kept in captivity, has been noticed[243] to elevate the posterior part of the girdle, and to pour out a continuous stream of flaky white matter like a fleecy cloud, which proved to be of a glutinous nature. It then discharged ova, at the rate of one or two every second, for at least fifteen minutes, making a total of 1300 to 1500, each being about 1/100 inch diameter. The ova were shot into the glutinous cloud, which seemed to serve as a sort of nidus to entangle the ova and prevent them being carried away. The subsequent development was rapid, and in seven days the young _Chiton_ was hatched, being then about 1/20 inch long. Lovén has described the same species as laying its eggs, loosely united in clusters of seven to sixteen, upon small stones. There is probably some mistake about the identification, but the observation illustrates the varying methods of oviposition among allied forms.

[Illustration: FIG. 43.--Egg-capsules of =A=, _Sepia elegans_ Orb., and =B=, _Octopus vulgaris_ Lam.]

Not very much is known with regard to the ovipositing of the _Cephalopoda_, especially those which inhabit deep water. Masses of ova arranged in very various forms have occasionally been met with floating in the ocean, but it is next to impossible to determine to what species, or even genus, they belong.[244]

In _Loligo punctata_ the ova are contained in small cylindrical cases measuring 3 to 4 in. by ½ in., to the number of about 250 ova in each case. Hundreds of these cases are attached together like a bundle of sausages or young carrots, and the movements of the embryos within can be distinctly noted. _Sepia officinalis_ lays large black pear-shaped capsules, each of which is tied to some place of attachment by a kind of ribbon at the upper end of the capsule, the whole forming a large group like a bunch of grapes. _Octopus vulgaris_ deposits thousands of small berry-shaped ova, attached to a string which runs along the centre of the mass (Fig. 43).

The so-called shell of the female _Argonauta_ is nothing more than a form of protection for the ova, and is in no sense homologous to the ordinary molluscan shell. The ova consist of a large granulated mass, attached to a many branched stem; they are contained in the spire of the shell, in contact with the posterior part of the body of the mother, but sometimes project externally beyond the coil of the spire.

Certain species possess the curious property of laying their eggs on the outside of their own shells. _Buccinopsis Dalei_ is not unfrequently found decorated with its own egg-capsules. Possibly this species, which lives on oozy ground, finds this the only secure place of attachment for its progeny. _Neritina fluviatilis_ has a similar habit, and so have many other species of _Neritina_ and _Navicella_. It is not quite clear, in the latter cases, whether the eggs are laid by the specimens on whose shell they are found, or whether they are deposited by others. In either case, perhaps the shell is the safest place for them in the rapid streams which both genera frequent. Specimens of _Hydrobia ulvae_ taken on the wet sands at the mouth of the Dee, are found to have several little rounded excrescences scattered over the surface of the shell. These, on examination, are found to be little masses of small sand-grains, in the centre of which is a clear jelly containing segmenting ova or young embryos. Here again, in all probability, the shell is the only comparatively stable object, in the expanse of shifting sands, on which the eggs can be laid.[245]

The pulmonate genus _Libera_, which occurs on a few of the island groups in the Central Pacific, is remarkable for the habit of laying its eggs within its own cavernous umbilicus, which is narrowed at the lower part. The eggs number from four to six, or the same number of very young shells may be seen closely packed in the cavity, each being in shape exactly like a young _Planorbis_. This constriction of the umbilicus does not occur till the formation of the last two whorls, _i.e._ till the animal is sexually mature. Some species, but not all, provide for the safety of their eggs more completely by forming a very thin shelly plate, which nearly closes the umbilical region, and breaks away or is absorbed to facilitate the escape of the young shells.[246]

=Union of Limax.=--With regard to the act of union itself, the method in certain species of _Limax_ deserves special notice. _L. maximus_ has been observed at midnight to ascend a wall or some perpendicular surface. A pair then crawl round and round one another emitting a quantity of mucus which at length forms a patch, 2 to 2½ inches in diameter. When this acquires consistency the pair begin to twist round each other in corkscrew form, and detach themselves from the wall, hanging by a cord of the thickened mucus, about 8–15 inches long, and still twisting round each other. The external generative organs are then protruded and copulation takes place, after which the bodies untwist, separate, and crawl up the cord again to the wall.[247]

=Periodicity in Breeding.=--In the marine Mollusca, the winter months appear to be the usual time for the deposition of eggs. Careful observations have been made on the Mollusca occurring at Naples,[248] and the general result seems to be that for all Orders alike the six winter months from November to April, roughly speaking, are the breeding time. Scarcely any forms appear to breed habitually in August, September, or October. On our own coasts, Nudibranchiata come in shore to deposit their ova from January to April. _Purpura lapillus_ may be observed depositing ova all the year round, but is most active from January to April. _Buccinum undatum_ breeds from October to May; _Littorina_ all the year round.

The land Mollusca exhibit rather more periodicity than the marine. In temperate climates they breed exclusively in the summer months. In the tropics their periods are determined by the dry and rainy seasons, where such occur, otherwise they cohabit all the year round. According to Karl Semper, the snails of the warm Mediterranean region arrive at sexual maturity when they are six months old, _i.e._ before they are fully grown. After a rest of about three months during the heat of summer, a second period of ovipositing occurs.[249] _Helix hortensis_ and _H. nemoralis_ ascend trees, sometimes to a height of forty feet, when pairing.[250]

=Hybridism= as the result of union between different species of Mollusca is exceedingly rare. Lecoq once noticed[251] on a wall at Anduze (Gard) as many as twenty specimens of _Pupa cinerea_ united with _Clausilia papillaris_. No offspring seem to have resulted from what the professor calls ‘this innocent error,’ for the wall was carefully scrutinised for a long time, and no hybrid forms were ever detected.

The same observer noticed, in the Luxembourg garden at Paris, and M. Gassies has noticed[252] at various occasions, union between _Helix aspersa_ and _nemoralis_, _H. aspersa_ and _vermiculata_, between _Stenogyra decollata_ and a _Helix_ (sp. not mentioned), _H. variabilis_ and _pisana_, _H. nemoralis_ and _hortensis_. In the two latter cases a hybrid progeny was the result. It has been noticed that these unions generally took place when the air was in a very electric condition, and rain had fallen, or was about to fall, abundantly.

Of marine species _Littorina rudis_ has been noticed[253] in union both with _L. obtusata_ and with _L. littorea_, but no definite facts are known as to the result of such unions.

=Self-impregnation= (see p. 44).

=Development of the Fertilised Ovum.=--The first stages in the development of the Mollusca are identical with those which occur in other classes of animals. The fertilised ovum consists of a vitellus or yolk, which is surrounded with albumen, and is either contained in a separate capsule, or else several, sometimes many, ova are found in the same capsule, only a small proportion of which ultimately develop. The germinal vesicle, which is situated at one side of the vitellus, undergoes unequal segmentation, the result of which is usually the formation of a layer of small ectoderm cells overlying a few much larger cells which contain nearly the whole of the yolk. The large cells are then invaginated, or are simply covered by the growth of the ectoderm cells. The result in either case is the formation of an area, the blastopore, where the inner cells are not covered by the ectoderm. The blastopore gradually narrows to a circular opening, which, in the great majority of cases, eventually becomes the mouth. The usual differentiation of germinal layers takes place, the epiblast eventually giving rise to the epidermis, nervous system, and special sense organs, the hypoblast to the liver and to the middle region of the alimentary tract, the mesoblast to the muscles, the body cavity, the vascular, the excretory and reproductive systems. The next, or _trochosphere_ (_trochophora_) stage, involves the formation of a circlet of praeoral cilia, dividing the still nearly spherical embryo into two unequal portions, the smaller of which consists simply of the prostomium, or

## part in front of the mouth, the larger bearing the mouth and anus.

So far the series of changes undergone by the embryo are not peculiar to the Mollusca; we now come to those which are definitely characteristic of that group. The stage next succeeding the development of the trochosphere is the definitive formation of the _velum_, a process especially characteristic of the Gasteropoda and Pelecypoda, but apparently not occurring in the great majority of land Pulmonata.

[Illustration: FIG. 44.--Veligers of _Dentalium entalis_ L.: =A=, longitudinal section of a larva 14 hours old, × 285; =B=, larva of 37 hours, × 165; =C=, longitudinal section of larva of 34 hours, × 165; _m_, mouth; _v_, _v_, velum. (After Kowalewsky.)]

The circlet of cilia becomes pushed more and more towards the anterior portion of the embryo, the cilia themselves become longer, while the portion of the body from which they spring becomes elevated into a ridge or ring, which, as a rule, develops on each side a more or less pronounced lobe. The name _velum_ is applied to this entire process of ciliated ring and lobes, and to the area which they enclose.

[Illustration: FIG.45.--Veliger of _Patella vulgata_ L., 130 hours old: _f_, rudimentary foot; _op_, operculum; _sh_, shell; _v_, _v_, velum. (After Patten, highly magnified.)]

[Illustration: FIG. 46.--Developed larva of _Cyclas cornea_ L.: _br_, rudimentary branchiae; _by_, byssus; _f_, foot; _m.e_, mantle edge; _sh_, shell. (After Ziegler, highly magnified.)]

[Illustration: FIG. 47.--=A=, Advanced veliger of _Dreissensia_: _f_, foot; _m_, mouth; _sh_, shell; _v_, _v_, velum. (After Korschelt and Heider, much enlarged.) =B=, Veliger of a Pteropod (_Tiedemannia_): _op_, operculum; _sh_, shell; _v_, velum. (After Krohn, much enlarged.)]

In this so-called _veliger_ stage, the velum serves, in the first place, to cause rotation of the larva within the egg-capsules, and, after hatching, as an organ of locomotion. As a rule, the velum disappears entirely in the adult mollusc after the free-swimming stage is over, but in the common _Limnaea stagnalis_ it persists, losing its cilia, as the very prominent circum-oral lobes. Simultaneously with the development of the velum, and in some cases earlier, appear the rudiments of the shell-gland and of the foot, the latter being situated on the ventral side, between the mouth and anus, the former on the dorsal side, behind the velum, and above the surface of the eventual visceral sac. Thus the prime characteristics of the veliger stage, subsequent to the appearance of the velum itself, are the development of the visceral sac and shell-gland on the upper, and of the foot on the under side. According to Lankester the primitive shell-gland does not, as a rule, directly give rise to the shell of the adult mollusc, but becomes filled up by a horny substance, and eventually disappears; the permanent shell then forms over the surface of the visceral hump from the original centre of the shell-gland. It is only in _Chiton_, and possibly in _Limax_, that the primitive shell-sac is retained and developed into the final shell-forming area, which is much wider, and extends to the edges of the mantle. Within the velar area first appear the rudiments of the tentacles and eyes; when these become developed the velum atrophies and disappears.

Several of these veligers when captured in the open sea have been mistaken for perfect forms, and have been described as such. Thus the larva of _Dolium_ has been described as _Macgillivrayia_, that of a _Purpura_ as _Chelotropis_ and _Sinusigera_, that of _Aporrhais pes pelecani_ as _Chiropteron_, that of _Marsenia conspicua_ as _Brownia_, _Echinospira_, and _Calcarella_.

_Cephalopoda._--The embryonic development of the Cephalopoda is entirely distinct from that of all other Mollusca. The segmentation of the vitellus is partial, and the embryo is furnished with a vitelline sac, which is very large in the majority of cases (Fig. 48). There is no free-swimming stage, but the embryo emerges from the egg fully developed.

=Differences of Sex.=--In the Mollusca there are two main types of sexual difference: (i) sexes separate (_dioecious_ type), (ii) sexes united in the same individual (_hermaphrodite_ type).

[Illustration: FIG. 48.--Two stages in the development of _Loligo vulgaris_ Lam.: _a_{1}_, _a_{1}_, first, and _a_{2}_, _a_{2}_, second pairs of arms; _br_, branchiae, seen through _m_, mantle; _e_, _e_, eyes; _fi_, fins; _fu_, funnel; _v.s_, vitelline sac. (After Kowalewsky.)]

In some cases--_e.g._ certain Pelecypoda--what is practically a third type occurs. The animal is hermaphrodite, but the male and female elements are not developed simultaneously, _i.e._ the same individual is at one time female, at another male.

1. The sexes are separate in

All _Cephalopoda_.

_Gasteropoda Amphineura_ (except _Neomeniidae_).

_Gasteropoda Prosobranchiata_ (except _Valvata_ and some species of _Marsenia_).

_Scaphopoda_.

Many _Pelecypoda_.

2. The sexes are united in

_Gasteropoda Opisthobranchiata_.

_Gasteropoda Pulmonata_.

Certain _Pelecypoda_.[254]

In the dioecious Mollusca, sexual union is the rule, but is by no means universal. In some instances,--_e.g._ _Vermetus_, _Magilus_, _Patella_, _Haliotis_, _Crepidula_, _Chiton_, the _Scaphopoda_--the form and habits of the animal do not admit of it; in others (many _Trochus_) a male copulative organ is wanting. When this is the case, the male scatters the spermatozoa freely; the majority must perish, but some will be carried by currents in the direction of the female.

When the sexes are separate, the female is frequently larger than the male. This is markedly the case in _Littorina_, _Buccinum_, and all the _Cephalopoda_; in _Argonauta_ the difference is extreme, the male not being more than ¼ the size of the female.

Those hermaphrodite Mollusca which are capable of sexual union (Gasteropoda, Pulmonata, and Opisthobranchiata) are conveniently divided into two sections, according as (1) there are separate orifices for the male and female organs, or (2) one orifice serves for both. To the former section (_Digonopora_[2]) belong the Limnaeidae, Vaginulidae, and Onchidiidae, and many Opisthobranchiata, including all the Pteropoda; to the latter (_Monogonopora_[255]) nearly all the Nudibranchiate Opisthobranchiata, and all the rest of the Pulmonata. In the latter case during union, mutual impregnation takes place, and each of the two individuals concerned has been observed (compare p. 42) to deposit eggs. In the former, however, no such reciprocal act can take place, but the same individual can play the part of male to one and female to another, and we sometimes find a string of _Limnaea_ thus united, each being at once male and female to its two adjacent neighbours.

=The Reproductive System.=--Broadly speaking, the complicated arrangements which are found in Mollusca resolve themselves into modifications of three important factors:--

(_a_) The _gonads_ or _germ-glands_, in which are developed the ova and the spermatozoa. These glands are generally known as the _ovary_ in the female, the sperm-gland or _testis_ in the male.

(_b_) The channels which provide for the passage of the seminal products; namely, the _oviduct_ in the female, the _vas deferens_ or sperm-duct in the male.

(_c_) The external generative organs.

[Illustration: FIG. 49.--Generative and other organs of _Littorina obtusata_ L., _female_.

=A=, anus. =Br=, branchia. =Buc=, buccal mass. =H=, heart. =Hep=, hepatic duct. =I=, continuation of oesophagus. =Ki=, kidney. =Li=, liver. =M=, muscle of attachment. =O´=, female orifice. =Od=, oviduct. =Oes=, oesophagus. =Ov=, ovary. =Ra=, radula. =St=, stomach. =U=, uterus.

(After Souleyet.)]

[Illustration: FIG. 50.--Generative and other organs of _Littorina obtusata_ L., _male_.

=A=, anus. =Br=, branchia. =H=, heart. =I=, intestine. =Li=, liver. =M=, muscle of attachment. =Pe=, penis. =Te=, testis. =VD=, vas deferens.

(After Souleyet.)]

=Dioecious Mollusca.=--The common _Littorina obtusata_ will serve as a typical instance of a dioecious prosobranchiate, exhibiting the simplest form of organs. In the female the _ovary_, a lobe-shaped body, is embedded in the liver. An _oviduct_ with many convolutions conveys the ova into the _uterus_, an oblong chamber which consists simply of a dilatation of the oviduct. The ova descend into the uterus, which is sometimes furnished with a seminal pouch. In this seminal pouch, or above it, in the oviduct, the ova come into contact with the spermatozoa. The lower part of the uterus secretes a gelatinous medium (or capsule, as the case may be) in which the fertilised ova become enclosed previous to exclusion. In position the oviduct abuts on the kidney, while the uterus is in close proximity to the rectum, and the female external orifice is found close to the anus, within the branchial cavity.

The male organs of _Littorina_ are more simple. The _testis_ is lodged, like the ovary, in the liver; the _vas deferens_ is, like the oviduct, convoluted, and eventually traverses the right side of the neck, emerging near the right tentacle, and terminating in the _penis_ or external copulative organ (Fig. 50).

This system prevails, with but slight modifications in detail, throughout the prosobranchiate Gasteropoda. The most important modification is the passage of the seminal products in certain cases (many of the Diotocardia) through the right kidney, with which the oviduct and vas deferens always stand in close relation. The same arrangement occurs in the Scaphoda and some Pelecypoda.

The penis varies greatly in form and size. In the Strombidae (see Fig. 99) and Buccinidae (Fig. 62) it is very large and prominent; in _Littorina_ it is somewhat spinulose at one side; in _Paludina_ a portion of it is lodged in the right tentacle, which becomes atrophied and much more obtuse than the tentacle on the left side.

=Spermatozoa.=--The shape of the spermatozoa and of the ova in Mollusca is of the usual type. In _Paludina Ampullaria_, and certain species of _Murex_ two types of spermatozoa occur, one hair-like, the other worm-like, three times as long as the former, and not tapering at one end. The former type alone take part in fertilisation, and penetrate the ovum. It has been suggested that these worm-like spermatozoa are a kind of incipient ova, and indicate a possible stage in commencing hermaphroditism. And, since the nearest allies of the Prosobranchiata (in which these types occur) are hermaphrodite (_i.e._ the Opisthobranchiata and Pulmonata), it is not unreasonable to suppose that the Prosobranchiata should show some tendency towards hermaphroditism in their genital glands.[256]

=Cephalopoda.=--The special characteristic of the reproductive organs in female Cephalopoda is the development of various glands, some of considerable size, in connexion with the ovary and oviduct. _Sepia_, _Loligo_, and _Sepiola_ are furnished with two large _nidamental glands_, which open into the mantle cavity independently of the oviduct. Their purpose is to produce a viscid mucus, which envelops the ova at the moment of their emission and eventually hardens into the egg-capsules. A pair of accessory nidamental glands occur in _Sepia_, as well as a pair of smaller glands situated on the oviduct itself.

In many of the male Cephalopoda the vas deferens is long and dilated at its outer end into a glandular reservoir, within which are formed the spermatophores, or narrow cylindrical packets which contain a very large number of spermatozoa. When charged, the spermatophores pass into what is known as Needham’s sac, where they remain until required for use. These spermatophores are a very characteristic part of the reproductive arrangements in the Cephalopoda. The male of _Sepia_ has been noticed to deposit them, during union, upon the buccal membrane of the female. During the emission of the ova by the female, the spermatophores, apparently through the agency of a kind of spring contained at one end, burst, and scatter the spermatozoa over the ova.

=The Hectocotylus Arm.=--Perhaps the most remarkable feature in the sexual relations of all the Mollusca is the so-called _hectocotylus_ of the Cephalopoda. In the great majority of the male Cephalopoda, one of the ‘arms,’ which is modified for the purpose in various ways and to a greater or less extent, becomes charged with spermatophores, and sometimes, during union, becomes detached and remains within the mantle of the female, preserving for some considerable time its power of movement.

The hectocotylus is confined to the dibranchiate Cephalopoda, and its typical form, _i.e._ when part of the arm becomes disengaged and left with the female, occurs only in three genera of the Octopodidae, _viz._[_Argonauta_, _Ocythoe_ (_Philonexis_), and _Tremoctopus_. In all of these, the male is many sizes smaller than the female. In _Argonauta_ the third arm on the left side becomes hectocotylised. At first it is entirely enveloped in a kind of cyst, in such a way that only a small portion of the tip projects; subsequently the cyst parts asunder, and allows the arm to become expanded to its full length, which considerably exceeds that of the other arms. At a certain point the acetabula or suckers terminate, and the remainder of the arm consists of a very long, tapering, sometimes thread-like filament, which is pointed at the extreme tip. It is not yet known how the spermatophores find their way into the hectocotylus, or how the hectocotylus impregnates the ova of the female. The arm thus affected is not always the same. In _Tremoctopus_ it is the third of the right side, in the _Decapoda_ the modification usually affects the fourth of the left.

[Illustration: FIG. 51.--Male of _Ocythoe tuberculata_ Raf. (= _Philonexis catenulatus_ Fér.), Mediterranean, showing three stages, =A=, =B=, and =C=, in the development of the hectocotylus arm: _h.cy_, hectocotylus still in the cyst; _c´y´_, spoon-shaped cyst at the end of the arm when freed; _th_, thread-like organ freed by the rupture of _c´y´_. Natural size. From specimens in the British Museum.]

This singular property of the male Cephalopoda has only recently been satisfactorily explained. It is true that Aristotle, more than twenty-two centuries ago, distinctly stated that certain of the arms were modified for sexual purposes. Speaking of what he calls the polypus (which appears to represent the _Octopus vulgaris_ of the Mediterranean), he says: ‘It differs from the female in having what the fishermen call the white sexual organ _on its arm_;’ again, ‘Some say that the male has something of a sexual nature (αὶδοιῶδές τι) on one of its arms, that on which the largest suckers occur; that this is a kind of muscular appendage attached to the middle of the arm, and that it is entirely introduced within the funnel of the female’. Unfortunately the word translated by _introduced_ is corrupt, and can only be restored conjecturally. He again remarks, ‘The last of the arms, which tapers to a fine point and is the only whitish arm, it uses in sexual union.’[257]

The typical hectocotylus seems to have entirely escaped notice until early in the present century, when both Delle Chiaje and Cuvier described it, as detected within the female, as a _parasite_, the latter under the name of _Hectocotylus octopodis_. Kölliker, in 1845–49, regarded the Hectocotylus of _Tremoctopus_ as the entire male animal, and went so far as to discern in it an intestine, heart, and reproductive system. It was not until 1851 that the investigations of Vérany and Filippi confirmed a suggestion of Dujardin,[258] while H. Müller, in 1853, completed the discovery by describing the entire male of _Argonauta_.

In all genera of dibranchiate Cephalopoda except _Argonauta_, _Ocythoe_, and _Tremoctopus_, one of the arms is sexually modified in various ways, but never becomes so much prolonged, and is never detached and left with the female. In _Loligo Forbesii_ Stp. the fourth arm on the left has 23 pairs of regularly developed acetabula, which then lessen in size and disappear, being replaced by long pedunculated papillae, of which there are about 40 pairs. In _Loligo vulgaris_ Lam. and _L. Pleii_ Orb. 18 or 19 pairs of acetabula are regularly formed, and then occur 40 pairs of papillae, as in _Forbesii_. In other species of _Loligo_ (_gahi_ Orb., _brevis_ Bl., _brasiliensis_ Orb.) only the outer row of suckers becomes modified into papillae after about the 20th to the 22nd pair. In _Sepioteuthis sepioides_ the modification is the same as in the _Loligo_ last mentioned, but the corresponding arm on the right side is so covered with acetabula towards its extreme end, that it is thought that it in some way co-operates with the hectocotylised left arm.

In _Octopus_, the third arm on the right side is subject to modification. This arm is always shorter than the corresponding arm on the other side, and carries fewer suckers, but is furnished at the extreme tip with a peculiar kind of plate, which connects with the membrane at the base of the arm by a channel of skin, which probably conveys the spermatophores up to the tip.

In _Octopus vulgaris_, the species referred to by Aristotle, the hectocotylised arm is short, thin in its outer half and pointed at the extremity, while the fold of skin is very white, and gives the arm an appearance of being divided by a cleft at the side. At the same time, an unusual development of one or two suckers on the arm is not uncommon.[259]

[Illustration: FIG. 52.--_Octopus lentus_ Baird, N. Atlantic, showing the peculiar formation of the hectocotylus arm, _h.a_. (After Verrill, × ½.)]

It is believed that in the Tetrabranchiate Cephalopoda (_Nautilus_) a union of the four inner ventral arms may correspond functionally to the hectocotylising of the arm in the Dibranchiates.

=Hermaphrodite Mollusca.=--(_a_) _Monogonopora._--The reproductive system in the hermaphrodite Mollusca is far more complicated than in the dioecious, from the union of the male and female organs in the same individual. As a type of the Monogonopora, in which a single orifice serves for both male and female organs, may be taken the common garden snail (_Helix aspersa_), the accompanying figure of which is drawn from two specimens found in the act of union (Fig. 53).

[Illustration: FIG. 53.--Genitalia of _Helix aspersa_ Müller, drawn from two individuals in the act of union, from a dissection by F. B. Stead.

=A.G=, albumen gland. =C=, coecum. =Cr=, crop. =D.S=, dart sac. =E=, eye (retracted). =Fl=, flagellum. =H.D=, hermaphrodite duct. =H.DF=, ditto, female portion. =H.DM=, ditto, male portion. =H.G=, hermaphrodite gland. =L=, liver. =M.G=, =M.G=, mucous glands. =Ov=, oviduct. =P.S=, penis sac. =R.M=, retractor muscle of penis. =Sp=, spermatheca. =V=, vagina. =V.D=, vas deferens.]

Beginning from the inside and proceeding outwards we have firstly the _hermaphrodite gland_ or _ovo-testis_ (H.G.), a yellowish white mass of irregular shape, embedded in the _liver_ (L.) and forming part of its spiral but not reaching quite to the apex. Within this gland are developed the ova and spermatozoa. The former are rather large round cells, produced within the outer wall of the gland, while the spermatozoa, which are produced in the more central part, are thread-like bodies, generally aggregated in small bundles. From the hermaphrodite gland the ova and spermatozoa pass through the upper part of the _hermaphrodite duct_ (H.D.), which is always more or less convoluted. Below the convoluted portion, the duct opens into the _albumen gland_ (A.G.), a large linguiform mass of tissue which becomes dilated at the time of pairing, and secretes a thick viscid fluid which probably serves to envelop the ova. Up to this point both the male and female elements follow the same course, but on their exit from the albumen gland they diverge. The hermaphrodite duct becomes greatly enlarged, and is partially divided by a kind of septum into a male and female portion. These run parallel to one another, the larger or female portion (H.DF.), through which the ova pass (and which is sometimes termed the _uterus_) being dilated into a number of puckered folds, while the smaller or male portion (H.DM.) is comparatively narrow, and not dilated. At their anterior end, the two portions of the duct separate completely from one another, the female portion being then termed the _oviduct_ (OV.) and the male portion the _vas deferens_ (V.D.).

Following first the oviduct, we find that it soon widens into the _vagina_ (V.), which is furnished with a pair of _mucous glands_ (M.G.), one on each side. These are much branched, and resemble little bunches of whitish seaweed. A little above the mucous glands a long tube diverges from the vagina, which is furnished with a produced _coecum_ (C.) and a pouch, the _spermatheca_ (SP.) at the extreme end. In this pouch, and in the duct leading to it, is stored the spermatophore received in union with another snail. Just below the mucous glands the vagina is joined by the _dart sac_ (D.S.), which is more fully described below. Finally, at its lower end the vagina unites with the penis sac at a point just posterior to the common orifice.

Returning now to the male organs, we find that the _vas deferens_ is the continuation of the male portion of the hermaphrodite duct, after its final separation from the female portion. It passes under the retractor muscle of the upper right tentacle, which has been cut away in the specimen figured, to dissect it out. Just before the vas deferens widens into the penis sac, it branches off into a long and tapering tube, the _flagellum_, in which the spermatozoa are stored and become massed together in the long packet known as the _spermatophore_. The _penis sac_ (P.S.) is the continuation of the vas deferens beyond the point at which the flagellum diverges. It joins the vagina at its extreme anterior end, uniting with it to form the common genital aperture, which cannot be exactly represented in the figure. The _penis_ itself lies in the interior of the penis sac, and is a rather long muscular tube which is protruded during union, but at other times remains retracted within the sac.

In the Helicidae generally, the form of the generative organs varies with each separate species, sometimes merely as regards the size of the different parts, at others in the direction of greater simplicity or complication. The mucous glands may be absent, and the flagellum greatly reduced in size, or absent altogether.

=The Dart Sac.=--A remarkable part of the reproductive system in many of the true Helicidae is the so-called _dart_, _Liebespfeil_, or _telum veneris_. It consists of ‘a straight, or curved, sometimes slightly twisted tubular shaft of carbonate of lime, tapering to a fine point above, and enlarging gradually, more often somewhat abruptly, to the base.’ The sides of the shaft are sometimes furnished with two or more blades; these are apparently not for cutting purposes, but simply to brace the stem. The dart is contained in a dart sac, which is attached as a sort of pocket to the vagina, at no great distance from its orifice. There are four different forms of sac. It may be single or double, and each of these divisions may be bilobed, each lobe containing one dart at a time. In _Helix aspersa_ the dart is about 5/16 in. in length, and ⅛ in. in breadth at its base (see Fig. 54).

It appears most probable that the dart is employed as an adjunct to the sexual act. Besides the fact of the position of the dart sac anatomically, we find that the darts are extruded and become embedded in the flesh just before or during the act of copulation. It may be regarded, then, as an organ whose punctures induce excitement preparatory to sexual union. It only occurs in well-grown specimens. When once it begins to form, it grows very rapidly, perhaps not more than a week being required for its entire formation.

[Illustration: FIG. 54.--Darts of British land snails: =A=, _Hyalinia excavata_ Bean; =B=, _Helix hortensis_ Müll.; =C=, _Helix aspersa_ Müll. (After Ashford.)]

The dart is almost confined to _Helicidae_, a certain number of exceptions being known which border on _Helix_. _Hyalinia nitida_ and _excavata_ are the only British species, not _Helices_, which are known to possess it. It has not been noticed to occur in the slugs, except in the N. American genus _Tebennophorus_. About one-third of the British Helices are destitute of the dart.[260] _H. rufescens_ possesses a double bilobed sac, but only two darts, which lie in the lower lobes. It does not use the darts, and could not do so, from the relative sizes of dart and sac; it has often been watched when uniting, but the use of the darts has never been observed. From this it has been inferred that the darts are degenerate weapons of defence, and that they were in fact at one time much stronger organs and more often used.[261] This theory, however, does not seem consistent with the whole circumstances of the occurrence, position, and present use of the darts.

=Hermaphrodite Mollusca.=--(_b_) _Digonopora._--As an example of the _Digonopora_ or hermaphrodite Mollusca with separate generative apertures for the male and female organs, we may take the common _Limnaea stagnalis_ (Fig. 55). It will be seen from the figure that the relative positions of the hermaphrodite gland and duct, and of the albumen gland, are the same as in _Helix_. When the oviduct parts company from the vas deferens, it becomes furnished with several accessory glands, one of which (GL.E.) probably serves as a reservoir for the ova, and answers more or less to a uterus. The tube leading to the spermatheca is short, and there is no divergent caecum. The female orifice lies near to the external opening of the branchial cavity. The vas deferens, which is very long, is furnished with a large prostate gland. The penis sac is greatly dilated, and there is no flagellum. The male orifice is behind the right tentacle, slightly in advance of the female orifice (compare Fig. 102).

[Illustration: FIG. 55.--Genitalia of _Limnaea stagnalis_ L. (from a dissection by F. B. Stead), × 2.

=A.G=, albumen gland. =Ac.G=, accessory gland. =F.O=, female orifice. =Gl.E=, glandular enlargement. =H.D=, hermaphrodite duct. =H.G=, hermaphrodite gland. =Li=, liver. =M.O=, male orifice. =P=, penis sac. =Pr=, prostate. =R.M=, retractor muscle of penis. =Sp=, spermatheca. =V.D=, vas deferens.]

Most of the _Opisthobranchiata_, but not all, have separate sexual orifices. Numerous variations from the type just described will be found to occur, particularly in the direction of the development of accessory glands, which are sometimes very large, and whose precise purpose has in many cases not been satisfactorily determined.

_Pelecypoda._--In the dioecious Pelecypoda, which form the great majority, the reproductive system is simple, and closely parallel in both sexes. It consists of a pair of gonads, which are either ovaries or testes, and a pair of oviducts or sperm-ducts which lead to a genital aperture. The gonads are usually placed symmetrically at the sides or base of the visceral mass. The oviduct is short, and the genital aperture is usually within the branchial chamber, thus securing the fertilisation of the ova by the spermatozoa, which are carried into the branchial chamber with the water which passes through the afferent siphon.

Hermaphrodite Pelecypoda are rare, the sexes being usually separate. The following are assured instances: _Pecten glaber_, _P. jacobaeus_, _P. maximus_, _Ostrea edulis_, _Cardium norvegicum_, _Pisidium pusillum_, _Cyclas cornea_, _Pandora rostrata_, _Aspergillum dichotomum_, and perhaps _Clavagella_. The greater number of these have only a single genital gland (gonad) on each side, with a single efferent duct from each, but part of the gland is male and part female, _e.g._ in the Pectens above mentioned. _Pandora_ and _Aspergillum_ have two distinct glands, respectively male and female, on each side, each of the two glands possessing its separate duct, and the two ducts from each side eventually opening near one another. It appears probable that the _Septibranchiata_ (_Cuspidaria_, _Poromya_, _Lyonsiella_, etc.) must also be added to the number of hermaphrodite Pelecypoda which have separate male and female glands.

It is worthy of remark that all the hermaphrodite Pelecypoda belong to forms decidedly specialised, while forms distinctly primitive, such as _Nucula_, _Solenomya_, _Arca_, and _Trigonia_ are all dioecious. In Gasteropoda similarly, the least specialised forms (the _Amphineura_, with the exception of the _Neomeniidae_, and the _Rhipidoglossa_) are dioecious. It is possible therefore that in the ancestors of the Mollusca the separation of the sexes had already become the normal type of things, and that hermaphroditism in the group is, to a certain extent, a sign or accompaniment of specialisation.[262]

=Development of Fresh-water Bivalves.=--The vast majority of fresh-water bivalves either pass the larval stage entirely within the mother, and do not quit her except in a perfectly developed form (_Cyclas_, _Pisidium_), or assume a mode of development in which free larvae indeed occur, but are specially modified for adaptation to special circumstances (_Unio_). _Cyclas_ and _Pisidium_, and no doubt all the kindred genera, preserve their ova in a sort of brood-pouch within the gills, in which the ova pass the earlier stages of their development. But, even so, the larva of these genera retains some traces of its original free-swimming habits, for a rudimentary velum, which is quite useless for its present form of development, has been detected in _Cyclas_.

The larva of _Dreissensia_ (see Fig. 47, A), so far as is at present known, stands alone among fresh-water bivalves in being free-swimming, and to this property has been attributed, no doubt with perfect justice, the fact of the extraordinarily rapid spread of _Dreissensia_ over the continent of Europe (chap. xvi.). In expelling the ova, the parent slightly opens the shells and then quickly closes them, shooting out a small point of white slime, which is in fact a little ball of eggs. The general course of development is precisely parallel to that of marine _Pelecypoda_, greatly resembling, so far as form is concerned, certain stages in the growth of the larvae of _Modiolaria_ and _Cardium_, as figured by Lovén.[263]

In June and July the larvae appear in large numbers on the surface of the water, when in spite of their exceedingly small size, they can be captured with a fine hand-net. They pass about eight days on the surface, feeding apparently on minute floating algae. During this time, the principal change they undergo is in the formation of the foot, which first appears as a small prominence midway between the mouth and anus, and gradually increases in length and flexibility. When the larva sinks to the bottom, the velum soon disappears entirely, the foot becomes exceedingly long and narrow, while the shell is circular, strongly resembling a very young _Cyclas_.

=Larvae of Unionidae.=--The early stages of the development of _Unio_ and _Anodonta_ (so far as the species of North America, Europe, and Asia are concerned) is of extreme interest, from the remarkable fact that the young live for some time parasitically attached to certain species of fresh-water fishes. In order to secure this attachment, the larva, which is generally known as _Glochidium_, develops a long filament which perhaps renders it aware of the neighbourhood of a fish, and also a larval shell furnished with strong hooks by which it fastens itself to the body of its unconscious host (Fig. 56). According to some interesting observations made by Mr. O. H. Latter,[264] the ova pass into the external gill of the mother, in which is secreted a nutritive mucus on which they are sustained until they arrive at maturity and a suitable opportunity occurs for their ‘being born.’ If this opportunity is deferred, and the _Glochidia_ mature, their so-called ‘byssus’ becomes developed, and by being entangled in the gill filaments of the parent, prevents their escaping. It is interesting to notice that, when the nutritive mucus of the parent is used up, it becomes, as it were, the turn of the children to provide for themselves a secondary mode of attachment.

[Illustration: FIG. 56.--=A=, _Glochidium_ immediately after it is hatched: _ad_, adductor muscle; _by_, ‘byssus’ cord; _s_, sense organs; _sh_, shell. =B=, _Glochidium_ after it has been on the fish for some weeks: _a.ad_, _p.ad_, anterior and posterior adductors; _al_, alimentary canal; _au.v_, auditory vesicle; _br_, branchiae; _f_, foot; _mt_, mantle. (Balfour.)]

The mother _Anodonta_ does not always retain the _Glochidium_ until fish are in her neighbourhood. Gentle stirring of the water caused them to emit _Glochidium_ in large masses, if the movement was not so violent as to cause alarm. The long slimy masses of _Glochidium_ were observed to be drawn back again within the shell of the mother, even after they had been ejected to a distance of 2 or 3 inches.

It is a mistake to assert that the young _Glochidium_ can swim. When they finally quit the mother, they sink to the bottom, and there remain resting on their dorsal side, with the valves gaping upwards and the so-called byssus streaming up into the water above them. There they remain, until a convenient ‘host’ comes within reach, and if no ‘host’ comes within a certain time, they perish. They are evidently peculiarly sensitive to the presence of fish, but whether they perceive them by smell or some other sense is unknown. “The tail of a recently killed stickleback thrust into a watch-glass containing _Glochidium_ throws them all into the wildest agitation for a few seconds; the valves are violently closed and again opened with astonishing rapidity for 15–25 seconds, and then the animals appear exhausted and lie placid with widely gaping shells--unless they chance to have closed upon any object in the water (_e.g._ another _Glochidium_), in which case the valves remain firmly closed.”

In about four weeks after the _Glochidium_ has quitted its host, and the permanent shell has made its appearance _within_ the two valves of the _Glochidium_, the projecting teeth of the latter press upon the ventral edge of the permanent shell, at a point about half way in its lengthward measurement, retarding the growth of the shell at that

## particular point, and indenting its otherwise uninterrupted curve with

an irregular notch or dent. As growth proceeds, this dent becomes less and less perceptible on the ventral margin of the shell itself, but its effects may be detected, in well-preserved specimens, by the wavy turn in the lines of growth, especially near the umbones of the young shell.

Mr. Latter found that all species of fish with which he experimented had a strong dislike to _Glochidium_ as an article of food. Sometimes a fish would taste it “just to try,” but invariably spit it out again in a very decided manner. The cause of unpleasantness seemed not to be the irritation produced in the mouth of the fish by the attempt of the _Glochidium_ to attach itself, but was more probably due to what the fish considered a nasty taste or odour in the object of his attentions.

* * * * *

The following works will be found useful for further study of this portion of the subject:--

=F. M. Balfour=, Comparative Embryology, vol. i. pp. 186–241.

=F. Blochmann=, Ueber die Entwickelung von Neritina fluviatilis Müll.: Zeit. wiss. Zool. xxxvi. (1881), pp. 125–174.

=L. Boutan=, Recherches sur l’anatomie et le développement de la Fissurelle: Arch. Zool. exp. gén. (2) iii. suppl. (1885), 173 pp.

=W. K. Brooks=, The development of the Squid (Loligo Pealii Les.): Anniv. Mem. Bost. Soc. Nat. Hist. 1880.

„ „ The development of the oyster: Studies Biol. Lab. Johns Hopk. Univ. i. (1880), 80 pp.

=R. von Erlanger=, Zur Entwickelung von Paludina vivipara: Morph. Jahrb. xvii. (1891), pp. 337–379, 636–680.

„ „ Zur Entwickelung von Bythinia tentaculata: Mitth. Zool. Stat. Neap, x. (1892), pp. 376–406.

=H. Fol=, Sur le développement des Ptéropodes: Arch. Zool. exp. gén. iv. (1875), pp. 1–214.

„ Etudes sur le développement des Mollusques. Hétéropodes: ibid v. (1876), pp. 105–158.

„ Etudes sur le développement des Gastéropodes pulmonés: ibid. viii. (1880), pp. 103–232.

=H. Grenacher=, Zur Entwickelungsgeschichte der Cephalopoden: Zeit. wiss. Zool. xxiv. (1874), pp. 419–498.

=B. Hatschek=, Ueber Entwickelungsgeschichte von Teredo: Arb. Zool. Inst. Univ. Wien, iii. (1881), pp. 1–44.

=R. Horst=, On the development of the European oyster: Quart. Journ. Micr. Sc. xxii. (1882), pp. 339–346.

=E. Korschelt= and =K. Heider=, Lehrbuch der vergleichenden Entwickelungsgeschichte der wirbellosen Thiere, Heft iii. (1893), pp. 909–1177 (the work is in process of translation into English).

=A. Kowalewsky=, Embryogénie du Chiton polii avec quelques remarques sur le développement des autres Chitons: Ann. Mus. Hist. Nat. Mars. Zool. i. (1883), v.

=E. Ray Lankester=, Contributions to the developmental history of the Mollusca: Phil. Trans. Roy. Soc. vol. 165 (1875), pp. 1–31.

„ „ Observations on the development of the pond-snail (Lymnaeus stagnalis), and on the early stages of other Mollusca: Quart. Journ. Micr. Sc. xiv. (1874), pp. 365–391.

„ „ Observations on the development of the Cephalopoda: ibid. xv. (1875), pp. 37–47.

=W. Patten=, The embryology of Patella: Arb. Zool. Inst. Univ. Wien, vi. (1886), pp. 149–174.

=M. Salensky=, Études sur le développement du Vermet: Arch. Biol. vi. (1885), pp. 655–759.

=L. Vialleton=, Recherches sur les premières phases du développement de la Seiche (Sepia officinalis): Ann. Sc. Nat. Zool. (7) vi. (1888), pp. 165–280.

=S. Watase=, Observations on the development of Cephalopods: Stud. Biol. Lab. Johns Hopk. Univ. iv. (1888), pp. 163–183.

„ „ Studies on Cephalopods: Journ. Morph. iv. (1891), pp. 247–294.

=E. Ziegler=, Die Entwickelung von Cyclas cornea Lam.: Zeit. wiss. Zool. xli. (1885), pp. 525–569.

##