/?6 7-(>f ^^^ OCCASIONAL PAPERS OF THE California Academy of Sciences No. 61, 44 pages, 11 figures. February 28, 1967. AUTOTOMY IN THE MOLLUSCA By Charles R. Stasek California Academy of Sciences IVijr.ne Bi^Iofc^:! Laboratory LIBRARY MAR 1 3 1957 WOODS HOLE, MASS. SAN FRANCISCO PUBLISHED BY THE ACADEMY 1967 \ CCASIONAL PAPERS OF TFIE CALIFORNIA ACADEMY OF SCIENCES No. 61, 44 pages, 11 figures. February 28, 1967. AUTOTOMY IN THE MOLLUSCA By Charles R. Stasek California Academy of Sciences Contents Introduction 2 Historical resume 2 Conditions of autotomy in the MoUusca 5 Instances of autotomy in the Mollusca 6 Gastropoda . . . . = . . . 6 Prosobranchia . . . o 6 Opisthobranchia 9 Pteropoda 9 Sacoglossa • • • ^ Nudibranchia ^^ Pulmonata 15 Scaphopoda 17 Bivalvia 18 Pectinacea 18 Tellinacea 20 Solenacea 20 Cephalopoda , . 24 Octopodacea 24 Argonautacea 25 Synopsis of autotomy in the Mollusca , 29 Discussion o 36 Acknowledgments 38 References cited 38 2 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers INTRODUCTION When under duress many animal species have the ability to cast off particular portions of the body. Probably everyone is aware that a lizard can easily shed its tail and that a crab can drop its legs. Less well known are analogous instances in the Mollusca, the group with which the present paper is concerned and in which the phenomenon has been reported in repre- sentatives of four of the six living classes. Unfortunately, there seems to be no complete study of this ability in any single moUuscan species, with the occurrence of abscission often obscurely buried among other data, given a variety of names, or not related to previous investigations. From a review of the literature, I have concluded that an underlying reason for the present paucity of knowledge in this area has been that there does not exist a clearly expressed framework of ideas within which compar- ative investigations can be carried out; hence, the aim of the present paper is not only to bring together and to illustrate instances of the process in the Mollusca, but also to set what is known about each instance against sev- eral conditions that one might expect to be associated with this ability wherever it occurs. In this way one can compare and evaluate existing and, hopefully, as yet unreported information on anatomical, physiological, be- havioral, and ecological factors that make the usually violent loss of living tissue adaptively significant to the organism. Except for bringing this material together with a definite purpose in mind and including some speculation, original contributions in this paper are limited to notes concerning the siphons of the bivalve Solen sicarius Gould, 1850. HISTORICAL resume' The act by which an animal casts off a portion of its own body was termed "autotomie" (="autotomy") by Fredericq (1883, who, in 1882 (a and b) demonstrated in a crab and in a lizard that such loss was the result of a muscular reflex action initiated by stimulation of sensory nerves in the part to be lost. In 1886, the same author summarized the occurrence of auto- tomy in reptiles and arthropods and characterized its significance as one of escape from enemies. "Le sacrifice d'une partie du corps assure, dans ce cas, le salut du tout" (p. 613). He also suggested that a comparable situa- tion existed in echinoderms and predicted that one might encounter auto- tomy in all animals of small size with hard and slender extremities. In the same year Varigny (1886) remarked that it would be interesting to know if activity equivalent to autotomy in crustaceans occurs in other animal groups. Again in the same year, Oehlert (1886) became the first per- son to associate autotomy with conditions previously known to exist in cer- tain Mollusca. Oehlert was unaware that a more complete summary of mol- No. 61) STASEK: AUTOTOMY IN MOLLUSCA 3 luscan examples had been made by Crosse (1860), whose review predates inception of the word "autotomy." In 1887, Fredericq tabulated the taxa he knew to be characterized by autotomy, reserving the term for those instances of active self-mutilation brought about by accidental external circumstances. Losses of portions of the body resulting from purely internal stimuli, such as proliferation of ma- ture proglottids in tapeworms and release of the hectocotylized arm in male cephalopods, were not held by Fredericq to be examples of autotomy. Par- allel with these phenomena Fredericq ranked the laying of eggs and mammal- ian parturition. Giard (1887) noted that the table given by Fredericq was incomplete, and classified instances of autotomy according to function ra- ther than according to taxonomic occurrence. His outline, as more clearly organized by Riggenbach (1903, p. 851), is as follows: I. Defensive autotomy A. Evasive autotomy, in which the result is escape from enemies. B. Economic autotomy, in which the result is reduction of the ani- mal's volume under unfavorable nutritive or respiratory conditions. While economic autotomy may occur in nature, it is usually observed only under imperfect laboratory situations. II. Reproductive autotomy, in which the result is reproduction. This cate- gory included hectocotylization of male cephalopods and, at the end of the series, release of proglottids by tapeworms. III. Parasitic autotomy, in which the animal rids itself of parasites as in Asterias richardi, which severs rays infected by myzostomid worms. [This category was added in a later paper. See Riggenbach (1903, pp. 805, 851)7] Here, to help complete the above classification, Giard could have placed an additional category: IV. Ontogenetic autotomy, in which the developing organism reaches a stage wherein a previously functional element is discarded, such as the early shell and operculum of nudibranchs and the early shell whorls of many gastropods (Pelseneer, 1935, p. 375). Unlike categories I-IIl, in no instance of ontogenetic autotomy would the animal be expected to regenerate the ele- ment. Giard pointed out that these divisions are not rigorous; for example, autotomy of an evasive type in a starfish is a kind of fission and may re- sult in reproduction. He also offered other classifications of autotomy such as "regenerative" versus "non-regenerative," and "general," where sep- aration occurs at any point on the body, versus "localized," where separa- tion always occurs at a precise point. Giard recognized a gradient of auto- 4 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers tomy into the cellular level, as in the release of nematocysts and in the ex- plusion of gametes. Giard's views were much broader than those of Fredericq and obscure the important distinction that Fredericq made between losses sustained as results of external or internal stimuli. A paper by Riggenbach (1903) remains the most extensive and most thoughtful review of autotomy throughout the animal kingdom, including the MoUusca. He realized that the visible act of autotomy could not be disso- ciated from a host of correlated factors, all of which comprise the condition of autotomy. Because of his significance to the study of autotomy, I have paraphrased some of his views below: Riggenbach defined autotomy as an activity of animals in which a part of the body, an organ, or, generally speaking, a specific mass of living substance is completely detached from the body (p. 858). He considered that autotomy occurred in three easily distinguishable categories. There are ani- mals in which the entire body undergoes violent motion when a part of it autotomizes, others remain quiet with only the piece to be autotomized be- ing excited, others do not move at all. In the first case it is a question of the activation of all or most of the muscles of the body; in the second, of the work of certain muscles lying in the zone of amputation; and in the third, of the action of one or a few specialized "Brechmuskels" that exclusively serve the autotomy. These categories are not always sharply distinguish- able (p. 858). Riggenbach stated that, as a rule, autotomy takes the following course: Certain stimuli excite the nervous system. This sets either all the muscles of the body, a single group, or only the specialized "Brechmuskels" in ac- tion. The activity of the muscles under the influence of the stimuli results of the stimuli results in the detachment. The resulting injury is rendered harmless through muscles or through other definite arrangements. The pro- visional closure is soon followed by healing of the wound and the regenera- tion of the lost piece of the body. With termination of regeneration the pro- cess of autotomy is concluded (p. 859). Autotomy was noted to occupy us- ually only a few seconds or minutes, more seldom hours or days, and it can be repeated at short time intervals (p. 861). Riggenbach did not regard as autotomy such processes as moulting, shedding of teeth and antlers, pro- cesses associated with metamorphosis, parturition, or release of genital products (p. 884). Instances of autotomy in the Mollusca were cursorily enumerated by Cooke (1895), Pelseneer (1906. 1935). and Allan (1941). It must be obvious to anyone who has reviewed the literature on auto- tomy that, from the historical approach, exactly what is and what is not to be considered an act of autotomy is difficult to say. Practically any event in which an anatomical loss is experienced could find categorical prece- dence in having been said to be autotomous. Diverse processes from re- No. 61) STASEK: AUTOTOMY IN MOLLUSCA 5 lease of nematocysts by coelenterates, to asexual division in some poly- chaete annelids (Berkeley and Berkeley, 1954, p. 329), to shedding of the larval shell of nudibranchs (Pelseneer, 1935, p. 375), and to loss of a liz- ard's tail have all come under the term. Riggenbach (p. 889) suggested that the term "Autoperose," from the Greek for "I mutilate myself," might be used to encompass all such pro- cesses. The term "autotomy" could then be employed more as Fredericq originally intended; namely, to include losses sustained as the result of external stimuli, serving the broad function of escape, and with the animal having some degree of preparedness to facilitate the loss. I favor this view, and would thus exclude ontogenetic losses and, particularly, losses of se- creted structures such as the larval shell of nudibranchs and the sheet of dorsal cuticle covering the nudibranch Corambe pacifica, as described by MacFarland and O'Donoghue (1929). I would restrict autotomy to the organ- ismic level and not apply it to losses at the cellular level, although cellu- lar processes play an important role in autotomous activity. CONDITIONS OF AUTOTOMY' IN THE MOLLUSCA From a theoretical standpoint, autotomy within the Mollusca might be expected to have the following characteristics: 1. There should be an ecological stimulus that triggers the severance. Disturbance by field collectors may induce autotomy. This reaction would suggest that disturbance by natural enemies also occurs. 2. The autotomized portion should be an extremity of the body exposed to the triggering stimulus. 3. The part stimulated should be the part lost; therefore, the body wall should be included in the process of severance. Hence, autotomy has been distinguished from evisceration, as occurs in sea cucumbers (Hyman, 1955, p. 205.) 4. The autotomized portion should be dispensible, not containing or- gans vital to continued existence of the organism as a whole. 5. There should be a restricted zone of weakness at which severance occurs. At least there should exist a predisposition for abscission, perhaps unlocalized, in the part to be autotomized. 6. Severance should be aided by specialized anatomical conditions, for example, sphincter muscles, which come into activity under the proper circumstances. 7. There should be structural or physiological adaptations preventing loss of body fluids and promoting healing of the resultant wound. 8. The body should have the capacity to regenerate a structure resem- bling that lost, with only a relatively brief period of time needed for comple- tion of regeneration. 6 CALIFORNIA ACADEMY OF SCIENCES fOcc. Papers 9. The severed portion, while unable to regenerate an entire organism, as occurs in some asteroids, might have limited capacity for independent, perhaps violent movement, the function being that of a decoy. 10. There might be additional behavioral specializations facilitating escape of the organism. 11. The loss should be functionally advantageous to the individual, as well as to the species as a whole. In any instance, autotomy should be interpretable as having arisen in evolution through the selective advantages conferred by ability to sustain the loss. INSTANCES OF AUTOTOMY IN THE MOLLUSCA ^- In the following account, and so far as present knowledge allows, each example of moUuscan autotomy has been presented with the above points in mind. A synoptic table of autotomy in the Melius ca may be found on page 29. Where knowledge is restricted to fact of occurrence alone, ref- erences are usually limited to the Synopsis. GASTROPODA Prosobranchia Passing statements that Harpa ventricosa (fig. 1) can autotomize the posterior quarter or third of the foot occur often in the literature, but the only information on the process seems to be that contained in the original reports by Quoy and Gaimard (1832, p. 617) and Reynaud (1834). According to the former authors, autotomy is easily induced in this species and is in- itiated by contractions of the foot, the separation resembling an ungluing rather than a tearing of tissues. On the intact organism they saw no line indicating the potential region of separation, but a large blood sinus traver- sing the foot suggested a zone of weakness. The muscle fibers anterior to the sinus are longitudinal, while behind it the tissue was said to be homo- geneous and like lard. The rejected mass was seen to contract indepen- dently for some moments, and they held that a new member is regenerated despite the large volume of tissue to be produced. Reynaud' s observations differ but slightly from those of Quoy and Gaimard. He noted in the single specimen available to him that a shallow transverse furrow marked the position at which autotomy occurred and that the animal of Harpa is too large to be entirely retracted into its shell. When the living specimen was placed in alcohol it contracted violently, a tear rapidly progressing from the left side of the foot for more than an inch along the transverse furrow. Death, however, prevented completion of autotomy. Reynaud speculated that if a natural enemy should attack a harpa, the snail No. 61) STASEK: AUTOTOMY IN MOLLUSCA ■■v,■•.^Sf:A -^»,.1 *• " t ■ "*? f » I 46% . •'^ Figure 1. Harpa ventricosa. Animal from the left side, and (below) ventral aspect of the foot showing autotomized posterior portion. (From Quoy and Gaimard, 1833, pi. 42.) 8 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers would withdraw into its shell with the posterior part of the foot protruding, since the shell is too small to accommodate that organ fully. The exposed part, being firm, would protectively close the aperture of the shell in much the same way as would an operculum, which harpa lacks. Should the aggres- sor prove redoubtable, it seemed probable to Reynaud that rupture of the foot would then occur at the transverse furrow, and the animal would be able to withdraw completely into the shell, thus effecting escape. That the end of the foot in Harpa may be cut off by pressure of the shell was suggested by Cooke (1895). Adams and Adams (1858, p. 436) reported that "Stomatia, like Harpa .... has the power of spontaneously throwing off the hind part of the foot when the animal is irritated, and Gena exhibits the same peculiarity; speci- mens in spirits have the foot usually truncated from this cause." Hedley (1902, p. 100) observed under more natural conditions that "As the conchol- ogist crooks his finger round a live StomateUa mariei, the creature falls asunder... Gena does the same in Sydney harbor..." The following informa- tion concerning these trochids comes from Mr. Robert Talmadge (personal communication) who received it from collectors in the Indo-Pacific region: Animals of the StomateUa group spontaneously cast off the posterior part of the foot when there is delay in placing collected specimens in water. The cast portion wriggles and, under natural conditions, is probably a de- coy for any predator, the identity of which is unknown. The animals regen- erate the cast portion, as Mr. Talmadge has observed small buds protruding from the healed posterior surface of the foot. Most recently, Fishelson and Qidron-Lazar (1966) reported pedal autotomy in Gena varia. A fine, trans- verse white zone that is histologically distinct from adjacent regions marks the site of autotomy. The severed part of the foot undergoes continued move- ments for 2 to 6 hours following autotomy, but the animal proper does not Figure 2. Herse columnella. Animal from which shell and mantle have been removed. Cervical organ (co). (From Tesch, 1946, fig. 12.) No. 61) STASEK: AUTOTOMY IN MOLLUSCA 9 immediately attempt escape. Instead, hidden under its shell, it remains firmly attached to a nearby stone. It should be noted that prior to autotomy all these shelled gastropods have bodies too large to be entirely contained within the shell, while after- wards the shell completely covers the soft parts. This also applies to the land snail Helicarion to be described below. Knowledge of autotomy of the posterior end of the foot of Haliotidea (^Calyptraea) is limited to mere mention by Pelseneer (1935, p. 375), who attributed the information to Adams, whose observations I have not been able to locate. Opisthobranchia Pteropoda The phallus of protandrous species of benthic gastropods is resorbed during sexual transition from the male to the female phase (Coe,1944, p. 88). In contrast, Bonnevie (1916), followed by Tesch (1946, p. 23), noted in the thecosomatous pteropod Herse columnella that the penis and its associated cervical organ (fig. 2, co) are thrown off from a definite pre-formed spot at the base of both organs. This species is protandrous with the loss of the copulatory apparatus occurring sometime after the female stage is entered and copulation accomplished. Since these organs are not regenerated, this condition might best be considered an ontogenetic loss. The question arises as to why the copulatory apparatus should be thrown off rather than resorbed as in benthic species. The highly specula- tive answer may be that the fertilized female experiences, and thus bene- fits from reduced specific gravity and increased swimming ability incurred tlirough immediate loss of the organs, even though these are small and dif- ferences must be slight. Sacoglossa Conor (1961, p. 390) observed that a portion of the foot may be cast when the Mediterranean species Lobiger serradifalci is roughly handled. "A clearly discernable dehiscence line can be seen at the base of the [^ara- podi£d] lobes and foot tail. This is the site where separation from the body takes place." Contraction at this site occurs in a sphincter-like manner. Caliphylla australis (fig. 3, A) was observed by Risbec (1928, p. 273) to have cerata that adhere strongly to the observer's finger while detaching from the body of the gastropod. Thus isolated, the cerata live several hours and move about in the water while writhing vigorously. The margin of each ceras is lined with defensive glands, each major gland being surrounded by accessory glands (fig. 3, B and C). Risbec postulated that if an animal should attack this species, Caliphylla would be able to defend itself by 10 CALIFORNIA ACADEMY OF SCIENCES (Occ- Papers D Figure 3. A, Caliphylla australis, dorsal aspect. B, C. australis, one of the cerata with marginal row of defensive glands. C, C. australis, enlarged aspect of a defensive gland; principal gland (p), accessory glands (a). D , Asteronotus boholien- sis, dorsal aspect showing irregular mantle margins resulting from autotomy, E, Doto coTonata, longitudinal section of a ceras showing defensive glands (g), sec- tions of digestive diverticula (d), blood spaces (I), muscles (m), and the zone pre- sumed to have a function associated with autotomy of the ceras (z). (A-D, from Ris- bec, 1928; E, from Hecht, 1895, pi. 3.) No. 61) STASEK: AUTOTOMY IN MOLLUSCA 11 gluing its enemy in its sticky, and "without doubt" venomous, spatula-like cerata. Trinchese (in Risbec, 1928, p. 21) reported similar autotomous behav- iour in Caliphylla borgnini. Cerata detached from individuals of this spe- cies thrash about and live for about 2 days. Such longevity was explained by Trinchese as having its source in the fact that each ceras contained portions of all of the organ systems of the body, with the exception of that of reproduction. Besides other details, the digestive diverticulum was said to be rich in "substances alimentaires," which presumably would be an energy source permitting both violent activity and relatively long life of the detached members. Risbec, however, believed it more reasonable that such muscular movements simply reflect continuation of the normal activity of the cerata, which, in life, are very motile and serve in locomotion. NUDIBRANCHIA There are two major anatomical regions knov/n to be autotomous in the Nudibranchia. The first involves the mantle margins, irregular pieces of which may be thrown off by several species of the Dorididae (see, for example, Risbec, 1928; here fig. 3, D). Eliot (1899) made the remarkable observation that, when handled, Discodoris fragilis may leave behind it a complete ring of mantle more than a quarter of an inch wide, while the cent- ral part crawls away, apparently none the worse for the loss. The process of autotomy is not rapid, and would not protect the nudibranch against a fish or any quick-moving predator, but might enable it to escape the attacks of a carnivorous mollusk, according to Eliot. Similar observations were made by Haefelfinger (1961, p. 334) on Peltodoris atromaculata, in which the phen- omenon was only seldom seen and then only under extreme conditions, for example, in the presence of rotting food or when the oxygen or circulation systems of the aquarium were shut off for a long time. He noted that regen- eration occurred and that this power seemed to be rather great. Allan (1932a, p. 88) characterized the genus Discodoris Bergh, 1877, as having the ability to discard a portion or the whole of the mantle margin when the animals are handled. Pruvot-Fol (1934, p. 221) used this feature to support a strong presumption that D, maculosa (Cuvier, 1804) was in this and not in some other genus, and Allan (1959, p. 222) held that the entire subfamily Discodoridinae was so characterized. However, mantle autotomy has yet to be observed in the Calif ornian species Discodoris heathi Mac- Farland, 1905. From the Synopsis one can determine that this ability is not restricted to Discodoris alone, for Bergh (1891, p. 122) stated that with strong irritation Argus and Peltodoris, as well as Discodoris, will throw off pieces of the mantle margins. Mantle autotomy in Kentrodoris inframaculata was thought by Risbec (1928. p. 22) to be aided by liquifaction of the tis- 12 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers sues as well as by endogenous spicules which, arranged in groups perpen- dicular to the body surface, form zones of weakness. The cerata, which project from the dorsal body surface, comprise the second major anatomical structure frequently autotomized by nudibranchs (fig. 4). Such cerata have been said to be "deciduous" or "caducous." Those of Melibe were observed to be cast off when the animals were placed in formaldehyde (Heath, 1917, p. 139) or were slightly disturbed (Pease, 1860, in Ostergaard, 1955, p. 135). Propemelibe mirifica, a very large form, was observed by Allan (1932b) to cast off cerata that showed signs of life for several days, curling up the edges when touched and exhibiting general movement. Allan believed it probable that the vivid coloring and sickly- sweet smell of the autotomized cerata were protective qualities. New cer- ata were regenerated and grew very rapidly to some inches in a few days (Allan, 1959). Parona (1891) brought about autotomy in Tethys leporina (fig. 4, B) by striking the animal, which then shed six of the 16 cerata. In contrast with Propemelibe, it required 2 months for the cerata of Tethys to regenerate to about one-third their normal length. In the autotomous cerata of Doto coronata (fig. 4, E) a flat layer or zone of intensely staining connective tissue was noted by Hecht (1895, p. 609). This layer is located near the point of insertion of the ceras upon the body and is interrupted only by the passage of the digestive diverticulum. It extends, presumably as a septum, from one side of the ceras to the other, clearly stopping against the layer of longitudinal muscle fibers underlying the epithelium (fig. 3, E, m). Hecht believed that this zone performs some unknown role, perhaps forming a point of least resistance, preparing an easy release of the ceras; perhaps acting as a closing plug after the autotomy; perhaps comprising a layer of embryonic tissue ready to assure the regener- ation of the ceras; or perhaps performing more than one of these roles. Hecht observed (p. 606) that new cerata were regenerated very quickly in Doto, and he concluded that autotomy was a means of defense. Baba and Hamatani (1965, p. 105) mentioned that the cerata easily fall off the body of Sakuraeolis enosimensis. In their plate 8, figure 3, they illustrated a longitudinal section of a ceras, the figure caption indicating a sphincter muscle and a septum at the base of this structure. These fea- tures were not discussed in the text, but the septum may bear some resem- blance to the peculiarities described by Hecht for Doto coronata. Hecht (1895) introduced the term "adhesive autotomy" to describe the condition whereby released cerata of Zephyrina mucronifera (fig. 4, G) adhere by their bases to the object that touched them. He described (p. 606) the cerata of this species as having well developed longitudinal muscle fibers, which, upon contraction, pull back the zone of implantation of a cer- as and constrict the opening of the digestive diverticulum within it. At the No. 61) STASKK: Al'IOrOMY IN MOLLUSCA H Figure 4. Various nudibranchs representing genera reported to autotomize the cerata. A, Doio affinis; B, Tethys leporina; C, Janolus cristatus; D, Glaucus pacif- icus; E, Doto coronata; F, ^eo/idia papillosa; and G, Zephyrina mucronifera. (From Adams and Adams, 1858, pi. 65.) 14 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers same time, the epithelium in that region forms a sort of [suction?! cup, which, to Hecht, explained why the cerata so strongly adhere by their bases, even to very smooth objects. Risbec (1928, p. 20), hov/ever, concluded that the cerata secrete an adhesive and perhaps venomous fluid. (See also the note on Caliphylla australis, p. 9). In all likelihood Hecht overlooked the pres- ence of one or more sphincter muscles on the proximal side of the abscis- sion, for the conditions he described, and also those noted by Edmunds (1966, p. 76) in Stiliger vanellus, do not seem to be the results of contrac- tion of longitudinal muscles, which would shorten the ceras but not neces- sarily constrict it or pull it away from the body. In Dirona albolineata, for example, MacFarland (1912) observed such sphincter muscles about large blood vessels at the base of each ceras. He thought that by contraction of the sphincters the lumina of the vessels would be closed should the ceras be accidentally lost. Graham (1938) investigated the alimentary canals of the aeolids Aeo- lidia papulosa (fig. 4, F), Cratena glotensis, Eolidina (-Aeolidiella) alderi, and Facelina drummondi. Unfortunately, where he writes of the autotomous cerata (p. 303) he does not note which of these species formed the basis for his observations. Perhaps his generalizations apply to all four, but be- cause of the lapsus I have not included these aeolids in the synopsis. Graham observed that each ceras has at its base a sphincter muscle that not only allows the ceras to be easily slied, but that also closes the opening left by the autotomy. As also noted by Risbec (1928), the autotom- ized cerata leave no trace upon the body, and both Risbec and Graham re- marked on the independent movement or writhing of autotomized cerata, an activity thought to divert the attention of a presumed predator that initiated the autotomy. In Aeolidia poindimiei, Risbec (p. 246) observed that at the time of separation the closed end of the digestive diverticulum forms a swol- len vesicle within the released ceras, while at the zone of separation there is found a tissue comprised of cells regularly disposed in palisades. No sug- gestion as to function was put forth to account for these conditions. Haefelfinger and Stamm (1959, p. 422) reported that the cerata of Li- menandra nodosa easily fall off and that they are regenerated, while Macnae (1954, p. 42) observed that an almost rectangular scar was left following ceratal autotomy in the pelagic nudibranch Glaucus atlanticiis. In this scar could be seen the narrow passage for the digestive diverticulum. Perhaps the specimens used by Macnae were dead or preserved, thus constriction of the wound had not taken place. A related species, G. pacificus, is illustra- ted in figure 4, D. Forster very early reported that Glaucus readily detached the long tail-end of its foot, but Bergh (1884, p. 14), who referred to Forster's state- ment, only rarely saw an individual that was without that iiieniber. The state- ment by Adams and Adams (1858, p. 436) tliat some nudibranchs have the No. 61) STASKK: AlTOTONn' IN MOLLUSCA 15 power to spontaneously throw off the hind part of tlie foot wlien tlie animal is irritated may refer to Ghwciis. Graham dealt at length with the presence of nematocysts in the tips of aeolidean cerata. These nematocysts, derived from the coelenterate food of the nudibranchs and retained undischarged within special cnidosacs, form a defense mechanism closely associated with autotomy in that family. Similarly, secretions, coloration, behavior and other attributes of opistho- branchs are significant as defense mechanisms (Thompson, 1960; Edmunds, 1966). PULMONATA Cockerell (1890a), in his description of the land slug Prophysaon coer- uleum, was probably the first to observe the effects of autotomy in the gen- us. He stated that "In one specimen [The posterior portion of the foot was] eaten off at the end." Later in the same year he recorded in a footnote that Hemphill had actually observed the autotomy (Cockerell, 1890b). Refer- ring to the same footnote, Raymond (1890) noted that autotomy of the foot does not occur in young specimens. Binney(1892) recorded another observa- tion made by Hemphill who found that the severed piece in what he called Phenacarion possessed as much vitality as the animal proper. B FiGLRE 5. A, Prophysaon foliolatum. Animal from the right side showing con- striction (c) at which autotomy occurs. B, P. coeruleuiu, dorsal aspect of preserved specimen from which posterior portion of foot has autotomized. C, Three views of autotomized piece from specimen shown in B. D, Helicarionviridis. aninuil from the right side. (A-C, from Pilsbry, 1948; D, from Adams and Adams, 1858, after Quoy and Gaimard, 1833.) 16 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers What little is known about autotomy in these slugs maybe found in the volume byPilsbry (1948, pp. 677, 680). He described Prop/iysaon as a genus in which the tail shows an oblique constriction at the posterior third to sixth of the animal's length (fig. 5, A-C), the constriction marking the site where self-amputation occurs, even in one's hand. Significantly, he observed that the body cavity does not extend beyond the constriction, and he pre- dicted that the tail would be found to regenerate. However, specimens of P. andersoni that were pinched, shaken, dropped, pricked, and put in pre- servative all failed to autotomize, so that Ingram (1948) concluded that not all specimens had the ability. This species was illustrated by Lovett and Black (1920), their painting showing the constriction described as charac- teristic for the genus. In 1943, Gregg recorded autotomy of the posterior third of the foot in Limax marginatus from southern California. The same author (Gregg, 1944) coined the term "auto-urotomy" for this ability and observed complete sev- erence of the last quarter of the foot after his specimens had been placed in a drowning jar. In relating his observations on snails of the genus Helicarion (fig. 5, D), Semper (1881, p. 375) stated that "Every species that I personally ex- amined possessed the singular property.. .of shedding their tail when they are seized somewhat roughly, at a little way behind the shell. This they do by whisking the tail up and down with extraordinary rapidity, almost convul- sively, till it drops off ..." Semper hypothesized that this behavior func- tions as protection against birds and reptiles, and he referred to similar ob- servations made by Guilding for snails of the genus Stenopus. Helicarion is characterized by a shell too small to fully accommodate the body, and the tail-whisking behavior may function by attracting attention away from the snail proper. It may aid the breaking process should the foot be seized, and, perhaps, it may supply sufficient force to throw the body of the captured snail some distance away from the predator, v;hich would be left with the tail only. Pfeiffer and Gundlach (1861) recorded that Pleurodonte (Polydontes) imperator, P. apoUo, and P. crassilabris are similarly able to throw off the hind part of the foot. Gundlach, who made the observations in Cuba, noted that the "tail" was very swollen in some individuals, absent in others, and that seizure always brought about autotomy at the same place. Four auto- tomized "tails" were found to live 54 hours following their release. Snails of the genus Nanina (now Xesta) are included in the same fam- ily as Stenopus. Benson (1835) observed that in Nanina "There is an ori- fice under the posterior caudiform appendage in the form of an isosceles triangle with the apex downwards, whence a thick greenish juice exudes when the animal is handled or irritated, the caudal appendage being turned up and protruded towards the exciting object." While autotomy does not oc- No. 61) STASKK: Al'TOIONn l.\ M()LLIJSC:A i: cur, the behavior of these snails aids in establishing that land gastropods do experience attacks from predators as yet unknown and that the posterior end is an advantageous location for glandular or autotomous protective de- vices.That snails may be eaten by reptiles is indicated by Wolcott's (1924) analyses of stomach contents of Puerto Rican lizards, of which six species yielded snails unfortunately not identified. Whether autotomous species are attacked remains speculative. SCAPHOPODA The only original record that I have seen to autotomy in the Scapho- poda is that of Lacaze-Duthiers (1856, p. 381). He found that when the aquarium water in which the animals were kept was changed, the mollusks would reject the captacular bundles (fig. 6). The filamentous captacula, which are extensile and ciliated feeding structures, are easily regenerated, according to Pelseneer (1906, p. 199). These filaments extend from two leaf-shaped shields of tissue, the medial captacula on each shield being short and immovable, while lateral ones grade into longer mobile filaments. Riggenbach (1903, p. 817) suggested that as new buds appear, older fila- ments are pushed laterally and, eventually, are entirely released from the body. Under normal conditions, probably only the outer filaments would be lost, newer ones appearing medially to replace them. Thus, as more fully described in the discussion, this would be a serial replacement rather than a true regeneration. Figure 6. Dentalium vulgare. A single captacular filament much enlarged, and the animal from the left side showing shell and extended captacuhi. (From Lacaze- Duthiers, 1856.) 18 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers In view of how little is known of the conditions under which the cap- tacula are norinally lost, inclusion of their release as an autotomous act is speculative. Figure 7. Lima dehiscens. Animal from the right side showing extended man- tle tentacles. (Photo by A. G. Smith.) BIVALVIA Pectinacea Gilchrist (1895, in Gilmour, 1963) noted the stickiness of the mantle tentacles of Lima (fig. 7) and that these were capable of autotomy, probably as a defense mechanism. Structural complexities of the tentacles of Lima hians were well described by Gilmour (1963). Each tentacle has along its external surface a large number of mucous gland cells arranged in rings, which give the tentacles of this genus their characteristic knobby appear- ance. Secretions from these cells are released upon stimulation and cause the tentacles to be tenaciously adherent. While the mucus is non-acid, Gil- mour did not exclude the possible presence of an irritant substance. Within the tentacle there is a longitudinal hemocoelic canal divided at irregular intervals by muscular septa, each of which is pierced by a sphincter-bound pore. A stout, pierced septum is also located at the base of the tentacle. There are no circular muscles within the tentacle, and when autotomy takes place it is brought about by contraction of radial muscles in a septum, at which separation occurs. Closure of the sphincter muscle accompanies autotomy, thus sealing the broken end of the tentacle. Crabs, to which limas had been presented as food by Gilmour, ap- peared to be irritated by the sticky, autotomized tentacles and made vigor- ous attempts to free themselves from them. Gilmour concluded that secretion of mucus in conjunction with autotomy may serve a defensive function. No. 01 stasi:k: adtotomv in m()lli;sc:a 19 \ m:.- WJ- Figure 8. Solecurtus strigillatus. Aspect of extended siphons, the animal auto- tomizing the ends of the siphons, and side view of one valve. (From Poli, 1791, pi. 12.) 20 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers Tellinacea Regarding Solecurtus coarctatus, Forbes and Hanley (1853, p. 262) reported that they had observed the annulated siphons to voluntarily break into fragments in the manner of the Mediterranean S. strigillatus (fig. 8). Rawitz (1892, p. 109 and following) observed the siphons of the latter spe- cies to be longitudinally striped. Of the stripes lacking pigment there are major and minor kinds. Mechanical stimulation of the major stripes resulted in siphonal contraction, while irritation of other parts of the siphons did not elicit a response unless the stimulus was strong enough to cause injury. With contraction of the siphons, characteristic ring-shaped constrictions appear along their length, as shown in figure 8. With excessive agitation, one or more ring-shaped segments of the siphons can be autotomized, a sim- ilar result being obtained when the animal is placed in preservative. Sim- ilarly, Atkins (1937, p. 392) observed that portions of the distal, unfused portions of the siphons of Solecurtus scopula were thrown off by sudden violent muscular contractions when the animals were rapidly removed from the water or when roughly handled. SOLENACEA Oehlert(1886) and Cooke (1895) mentioned that rouglily handled solens may, by violent muscular contraction, throw off portions of the foot. Since Oehlert did not refer to the species to which his observations pertained, it is not known whether this peculiar activity is widespread or applies only to S. siliqua, the species given by Cooke. Because of possible rupture of blood sinuses, such autotomy probably would inhibit rapid subsurface es- cape movements. Needless to say, the process seems a priori disadvanta- geous. Several species of Soien bear autotomous siphons, which, unlike those of Solecurtus, are mutually fused for their entire length. The siphons are characteristically annulated, as illustrated by Poli (1791), here shown in figure 9, A. Fischer (1887, p. 101) stated that he often found loose annulations on the surface of the substratum near the burrows of Solen marginatus. In this instance disturbance created by the collector as he walked near the burrow may have been the stimulus for the autotomy. While Ricketts and FiGiRE 9. A, Solen siliqua, showing annulated siphons. B, S. fonesi, showing extended foot and annulated siphons. C, S. fonesi, showing rings ol' tentacles on the siphons. D, S. fonesi, a cast-off section of the siphons, end view. E, S. sicarius, normal siphonal tips, end view; exhalant aperture (ea), flaring membrane (fm), in- halant aperture (ia). F, S. sicarius, end view of siphons immediately after removal of distal-most section; blood channel in siphonal septum (be), severed nerve bundle indicated by dotted circle (nb), buds of nunpient siphonal tentacles (tb). (A, from Poli. 1791; B-D, from Annandale, 1916.) No. 61) STASEK: AUTOTOMY IN MOLLUSC A 21 TT- >. m^mi- ^5S3!^i^«B^*3^i5J5;'>r: B 3S5?5555!iS«»'" 22 CALIFORNIA ACADEMY' OF SCIENCES (Occ. Papers Calvin (1952) somewhat facetiously suggested that the ability of siphonal autotoniy may be an adaptation to predation by birds, Carlisle and his co- workers (1960, p. 49) made the observation that in southern Californian wa- ters tiie barred surfperch eats Solen rosaceous. Orcutt (1950) found many siphonal tips among the stomach contents of the starry flounder. These si- phons were identified as those of Siliqua patula and of other species not possessing ;iutotomous siphons, but since thistlounder also frequents areas where Solen is found, that these clams are likewise preyed upon is probable (Orcutt, personal communication). In Great Britain, Smith (in Atkins, 1937, p. 393) directly observed autotom.y in Solen marginatus when a medium-sized plaice nibbled at the siphonal tips. That predatory fish comprised an eco- logical factor leading to the evolution of au.totomous siphons in the Solen- idae seems fairly well established, in my opinion. Under laboratory condi- tions, Faussek (1897. in Riggenbach, 1903, p. 820) initiated aiitotomy in S. marginatus by electrical stimulation of the visceral ganglia. A specimen of Solen fonesi was reported by Bloomer (1907) to have part of the siphon "broken off," and Annandale (1916, p. 355) also noted autotomy of the segmented siphons in this species. He observed that there is a ring of minute tentacles around the distal end of each segment of the siphons and thought that these tentacles probably have a sensory function, "...the new tip as well equipped as the old." Annandale' s figure 5 (here fig. d; B-D) shows these siphons, the rings of tentacles about the sections being on the external surfaces. Autotomy in this species is the result of contraction of radial muscles interconnecting the inner and outer epithelia of the siphons and found only at the junctures between sections (Ghosh, 1916, p. 370). Solen rosaceus will autotomize its siphons at annular constrictions when the animal is roughly handled or if the aquarium water becomes stale (Ricketts and Calvin, 1952, p. 241; Pohlo, 1963, p. 100). Polilo observed that "The siphons are annulated for their entire length, and each annula- tion has a fully developed set of siphonal tentacles. When the siphons are autotomized,...a new set of tentacles is already formed and apparently ready to fimction." Strikingly, the sets of tentacles are on the inside walls of the siphonal tubes rather than on the outside as in S. fonesi. Ricketts and Calvin (p. 252) noted siphonal autotomy in Solen sicar- ius, a species studied by the present writer at the Friday Harbor Laborator- ies, Washington. Solen sicarius, like other autotomous members of the genus, has annulated siphons. Friday Harbor specimens 55 mm., 87 mm., and 90m.m. in greatest dimension bore six, seven, and eight siphonal sections respect- ively, thus suggesting a direct relationship of size and number of annulae present. The normal tips of the siphons are illustrated in figure 9E. About the inhalant aperture (ia) is a series of, usually, six large lobed and ser- rated tentacles that bend inwardly over the aperture. These are here termed No. 61) STASEK: Al 'lOlOMV IN MULLL SCA 23 the primary tentacles. Lying between the primary tentacles are much smaller secondary tentacles, whi(;h bend outwardly. An Incomplete ring of simple tentacles lies proximal to the primary and secondary tentacles. Tlie exhalant aperture (ea) is bordered by a thin tlaring membrane {fml raised above an incomplete ring of simple tentacles. Irregularly placed ten- tacles may exist between the flaring membrane and the ring of tentacles. When the distal section was pulled with forceps, it was easily sep- arated along a clearly marked constriction, the separation resembling an ungluing rather than a rupture of tissues (see note on Harpa, p. 6). Longi- tudinal nerve cords traversing the zone of weakness seemed only slightly more tenacious than the rest of the tissue. The distal end of the former penultimate siphonal section is shown in figure 9 F. Upon separation at the annular constriction, the inner and outer surfaces about the inhalant and exhalant canals contracted. This contraction was not great enough to com- pletely cover the raw tissue lying between the surfaces, especially the tis- sue of the siphonal septum, which contains two conspicuous, although con- stricted, blood channels (be). On the distal margin of the more heavily pig- mented inner surface of the inhalant siphon one could observe a ring of swellings or papillae (tb), six of which lay adjacent to the ends of the six severed nerve bundles (nb>. Similar, although smaller raised areas were present on the inner surface of the exhalant siphon. Based on eight speci- mens from which one to three siphonal sections had been artificially re- moved, the gross process of regeneration proceeded as follows: On day 2, that is, on the day following that on which the separation was made, the inner and outer surfaces of the siphons had fused over the exposed tissue. The line of fusion was clearly indicated by differences in pigmentation, the inner surface being darker. The tissue covering the siphon- al septum was heavily pigmented, and was apparently derived from the inner surface of the inhalant siphon. Also on this day the small swellings of the inhalant aperture consisted not only of inner layer, but of outer layer also, as indicated by regional differences in pigmentation. On day 3, the distal end of the siphons had grown by about 0.4 mm., as suggested by a difference in pigmentation. Tlie largest swellings of the inhalant aperture were about 0.2 to 0.3 mm. from base to tip in the speci- mens from which 3 sections had been removed, about 0.4 mm. in that lacking 2 sections, and about 0.5 mm. in the two missing 1 section. Small swellings were visible just proximal to the edge of the exhalant aperture. On day 4, swellings suggesting incipient secondary tentacles were visible on the outer surface of the inhalant aperture of one specimen. The largest swellings, now recognizable as prospective primary tentacles, ranged from 0.3 to 0.75 mm. long, the specimens lacking only one section still hav- ing the larger tentacles. 24 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers On day 5, the flaring membrane about the exhalant opening was dis- tinctly present in all but one specimen from which 3 sections had been re- moved. Tentacle buds about the flaring membrane were distinct, those of two specimens having grown to 0.4 mm. long. On day 6, the primary tentacles of the new siphonal tip had become 0.5 mm. to 1.2 mm. long; the secondary tentacles measured up to 0.3 mm. long, and those about the exhalant aperture were 0.2 mm. to 0.5 mm. long. While the siphonal apertures closed upon stimulation, the secondary tentacles tended to bend outwardly, and the larger, primary tentacles bent inwardly over the closed aperture. By day 7, some of the longest primary tentacles had small serrations along their sides and were 0.7 mm. to 1.5 mm. long; secondary tentacles were only up to 0.4 mm. long. Tentacles about the exhalant aperture ranged from 0.25 mm. to 0.6 mm. long. Although these specimens were not followed beyond 7 days, it was apparent that the small pigmented buds present on the border of every sec- tion of the siphons represented prospective tentacles whose development was arrested until more distal sections had been removed, thus curiously resembling the condition in plants in which lateral stem buds may remain inactive until the terminal bud is removed. The condition in Solen sicarius is quite different from that in S. rosaceus in which proximal tentacular de- velopment does not appear to be inhibited, fully grown sets being present on subterminal sections. CEPHALOPODA All reported instances that I have seen of autotomy in the Cephalo- poda are limited to the order Octopoda, and in this order to the Octopodacea (Octopodidae) and the Argonautacea (Tremoctopodidae, Ocythoidae, and Argonautidae). Octopodacea While Lane (1960) reported that some octopuses will voluntarily break off an arm when it is held, a most remarkable instance of octopdian auto- tomy was observed by Straughan (1957). In trying to transfer an unnamed species of octopus from a jar, Straughan proded the specimen with a rod. The octopus "....struck out viciously at the rod, somehow detaching one of his arms in the process, and before I realized that had happened, one of the arms came crawling out of the jug and slithered up my arm almost to the shoulder....** Four arms were eventually so detached. If this independent brachial activity occurs in nature, its defensive qualities can hardly be doubted. Lange (1920), in studying regeneration of the arms of cephalopods, noted in the octopus that there was an absence of bleeding from the cut No. 61) STASKK: AUTOTOMY IN MOLLUSCA 25 surfaces of artificially amputated yrrns. She stated that such reteution of blood is probably associated with the ability to autoroinize distal portions of the arms. She noted no specialized region for the point of autotomy, which often, but not always occurred four- fifths of the distance from. the arm tip. Immediately after removal of the arm tip, the external rim of the wound con- tracted spasmodically, the central musculature and the axial nerve remain- ing unprotected. Delayed bleeding, however, occurred 5 or 6 hours following the operation, the clotted blood forming a preliminary covering for the wound prior to commencement of regeneration. It seems to the present writer that delay of bleeding might serve in aiding escape from predators that employ olfactory stimuli to track dov/n prey. That such an adaptation might prove useful is suggested by the statement by Robson (1929, p. 21) that octopods are often found with portions of the arms missing, even to the extent that all of them are reduced to stumps. Attacks by enemies, autotomy, and auto- phagy were given as possible explanations. Argonautacea Among the pelagic Argonautacea there are two kinds of autotomy. The first involves the pair of dorsal arms of Tremoctopus, as well described by Portmann (1952). In the females of Tremoctopus violaceus these arms are greatly elongated and flattened (fig. 10). Each is characterized by a longi- tudinal row of brilliant red spots or ocelli, each ocellus being bordered by dazzling white. The dorsal surface of the arm is gray to reddish-brown, and the ventral surface passes from brilliant orange to cherry red and to purple tlirough violet or deep brown. In addition there are flashes of irridescent green and metallic gold. Midway between any two ocelli there is a transverse line that marks a zone of weakness (fig. 10, la). These zones of weakness divide the arms into segments that the animal is capable of throwing off. Upon release from the arm, one segment had increased its span from 78 mm. to 198 mm., thus confirming earlier observations that segm.ents are held under partial contraction as long as they remain on the body. New segments seem to be formed at the base of the arms, as indicated by the incipient sections there. Arranged transversely within the tissues of these arms are numerous cords, each of which is composed of a muscle bundle spirally wrapped with a dense envelope of connective tissue. The zones at which autotomy occurs are located where two transverse bundles are adjacent to one another. Rup- ture always occurs along the thin part between two such bundles. Portmann referred to the observations of an earlier investigator who noted that when a female oi Tremoctopus was swimming peacefully, the dor- sal arms were rolled up by the ventral side. When irritated, the animal un- furled its arms and swam rapidly backwards, thus trailing the brightly col- 26 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers au No. 6n STASKK: AirOTONn' IN M()I.LIISC:A Figure 11. Argonauta argo. Male specimen, showing hectocotylus within its sac, and with the hectocotylus extended. (From Tryon, 1882, pi. 16.) ored structures behind it. Although Portmann cautiously avoided specula- tion on the function of this behavior, that it is a defensive mechanism seems more than likely (Young, 1959, p. 410). The sudden appearance of the bright colors may serve to startle a predator, and, if a segment of the arm is autotomized, its rapid increase in size may aid total escape l^y de- tracting the attention of the pursuing animal away from the octopod. In the families Tremoctopodidae, Ocythoidae, and Argonautidae the males, which may be about one-twentieth the length of the females, are pro- vided with an autotomous arm, elongated and otherwise modified as an or- gan of copulation. During its development this organ, known as tlie hecto- cotylus, is coiled up within a sac of tissue (fig. 11). At some unknown time prior to copulation the sac ruptures and allows the contained arm to extend. With this arm a packet of sperm is transferred to the mantle cavity of the female, the arm somehow detaching and remaining within that space. In the Tremoctopodidae and Ocythoidae, the third right arm is specialized in this way, while in the Argonautidae the hectocotylus is the third left arm. Hoyle (1908, p. 521) remarked that "...it is not known for certain whe- ther the fertilizing arm is deposited by the male in the mantle-cavity of the Figure 10. Tremoctopus violaceus. Female specimen m dorsal aspect; arms (l-IV), beginning of autotomy between two sections of one of the dorsal arms (au), lines indicating zones of weakness at which autotomy occurs (la), dorsal aquifer- ous pores (pa). (From Portmann, 1952.) 28 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers female (as I think is most probable), or whether (as stated by some writers) the arm breaks off when mature and finds its own way to its destination. This much is certain, that for some time after its detachment it possesses the power of independent movement." More recently. Young (1959, p. 409) stated that "The male [argonauT; is a minute creature, which develops little beyond the larval stage except for the production of an enormous sexual arm. This is introduced into the mantle cavity of the female and there brok- en off from the male, a process of autotomy... This takes place while the female is still young and the male arm, charged with spermatozoa, lives on within the female until she is mature and the eggs are ready for fertiliza- tion, though the details of this are not known." Considering the size differences between the sexes, one possible function of an autotomous hectocotylus might be not only that of ensuring deposition of the spermatophore, but also that of ensuring escape of the min- ute male individual from the female; that is, the loss could be evasive in character. That the male might be endangered is suggested by Wells (1962, p. 35). He noted in Octopus vulgaris that copulation is conducted at a dis- tance, and he indicated that the males possess certain anatomical features that may serve as signals inhibiting attack by the female, especially if the male is the smaller of the pair. Wells also suggested (p. 36) that "...it is possible tliat cephalopods, in general, spawn only once and then die." If small male cephalopods are in danger from the females and if spawned males may soon expect to die, evolution in the minute males of the Argonautacea of a complicated autoto- mous capability that, under other circumstances would tend to prolong the life of the male, seems unlikely: there would have been no way for natural selection to operate in favor of escaped males, that is, in favor of males with the most efficiently autotomous hectocotyli. In this regard it may be noted that although regeneration of the hectocotylus was assumed by Steen- strup (1857, p. 107) and by Coe (1944, p. 94), reasonable doubt was express- ed by Grimpe (1928). If this arm, once lost, is not regenerated, the implica- tion is that any male can reproduce but a single time. Loss of the hectoco- tylus might thus be regarded as an ontogenetic loss perhaps broadly com- parable to conditions reported in the pteropod Herse columnella (p. 9) , al- though protandry is out of the question in the Cephalopoda (Coe, 1944, p. 94). In view of the present paucity of knowledge, I feel free to suggest that male of Argonautacea might simply be eaten by the females, with only the sperm-charged hectocotylus remaining behind. Efficient autotomy of the hectocotylus would thus bo advantageous to the species, if not to tln^ male individuals, and provide some basis upon which natural selection could act. No. 61) STASI:K: AUTOrOMV IN MOLLUSC A 29 o 05 ^ o. L. < a; < u r<~, z iT) c\ ^" '^ ' — ' u^ J U a: < J o ■Ji ijj -^ — ^-' ^M UJ U. nr; JJ O UJ H ■„ u. Z < cc: r- J >^ >- < ^ o OS z H 2 d: U 0- H C u^ k— << h- X < Z 1- UJ >- u. ^ £0 o C ITj < :r, m q: n a. < Z o •a- i>^ , a o 0) a. o CO c. ^' CO 00 -a d s nil a a o X 00 in 00 en B T3 0 0) ■a o o 3 O 3 03 CO ca Cj CO CT3 c CO CO -a G ■f-H O -a a o G? 4-) ■4-J ■!-> ■*-3 «-3 ■•-^ O o o o O o n o o o o o n n C.-H ^4-4 =« =^•—1 (•-• -, , ^<-> it— 1 s« -, ^ ■i * o o o o o o c ? 4-1 «-3 4-> ■4-J <_1 ■4-^ ^ o E-t E-4 Ul a rf :tJ T1 :^ ^ :rt o Q. a- ^ ;-i Ui t-l t-< u< Ul (h O O O o o O E-H E-4 ^ E-i ^ u< Eh O) 4-3 a: f/3 c« CO U5 CO CO C/3 <■ O O o o o o o CX CIh cu Oh ex, CU Oh Oh ca en o ca S o CO CO Uh H "3 UJJ ca 0) ca l-H >^ ca O ca •a •^^ o ca a >-, ca a: 0) ca -a CO o CD 4-H > c c o oa CO c a bX) ca u z < o O o OS m n a: ca X r- UJ H C/) a. O ca »— 4 '•*4 0) e o CO {-c ^ CO CO , S ^ <^ CO ^^ T-H x: "-^ CD O tH 02 ci tu i3 OJ »H o 0) (D bij 72 bD — ' o ■Jr v O OS cu 4-> o Cfl .o be "^ CD Oj rC C ri t-t o ■-^s < =■ - 1/; a; -ri 3 I/, M ca o 4-3 ^*-» "^ o Oh 2^ o O < y. ca •4^ -a ca 0) to -J l«l c •»*4 )>i '"^ ca •»*4 -a o G :» >4 ca XI O O 30 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers w u u z Id a u b. U CC <■ a, U Z o: IX Q u N 1-4 o H O H < < to u u i < ■a a c o O < c/3 O -J o o u < 00 OJ 05 i-H ^-^ • Qj 0) O O O O c: o X5 O o 3 CT3 'a o o O c/^ 00 T3 o cd •a ■I— ( CO »3 3 CO l«l 3 •C3 U C nj ■-. _, J2 Co o < ^^ »^ z §£ O w^ 40 O CO a: EC 9 >^ Z on 00 D< a3 W CO 00 CO Oi ^w' ^•^v c X3 l-l ^^ t-l CO ^^ ■* d Hi 'f! CO ti ^^ •t-i ^^^ ^^ 05 O -a 00 o 00 ••-1 Ml a ■1— 1 00 05 o 00 Oi 1—1 u 1— 1 Ol 00 "o • *-> o; 02 00 1—1 Ol Oi o 0) J3 a 0) J2 05 X3 o > o a Cfi ;-• Ol en cn t-t ^ OJ ci o ^^ ^-^ , — 1 CJl Cci oa X Di q:; CQ Oh w < a •t-H ■3 o o Q E-I s CCCGCCCC fciD 5b bi 'faC 'bO 'fai 'bo bD t^ t-■ o ■a o o 03 c 3 [/i O a o CO CO 13 c -^ , -r wl 1^ 0 0 "w 1/5 K ui 0 C^ v: \r O c IJ „ c c >. c: c CC o c c >. a; 'tl ^ m C3 ■o rf) ■vD Oj ^J > •^ a. • f— 1 CM CO CO ^ Ctt C- D. c o o •S (J No. 61) STASEK: AUTOTOMY IN MOLLUSC A 31 10 w u z u cc u lb u OS < a. u z a: a O O s- < < a. t/: u ft. < z < -a o O X u z < CQ D 00 ^~. C\2 -—I ■•-I 00 o en Ml c s c c e rt en c: 0) o •CI o .o CO Oi . "O CO 00 O -a c O C3 > Q •a c 0) > o CO D Q c o 1-1 OS T3 C O 2 o — 00 00 c o o 13 c o 0) a. o -a a a> > o Oi Ml CO o ii C s 02 Oi 02 02 o 03 r-* 1^ CO 1—4 " 3 ■rH •a T-H 3 CO 00 oo f-H f-H f-H f-H f-H f-H rt f-H ^ ^ ^— ( N "*, ^^ ^^^ f-H 1-H T-H ^ — ' ^^■^ ^•^ ^^ — ^ — ' ^ — ' ^ ■ — ' [/: "^ Cj J-< 3 CJ 5 P £-1 ^-' ' ^^ o CJ o o o o c o 4>J 0) c3 CD Ki 3 J3 tn 4-3 ■t— 3 4-3 X3 XJ J2 CD X! XI iJ3 X3 5 O "3 iTi O o 7) ^ cd O CJ CJ c^ U3 in 03 m CO Cfi xf CO ■a ^ QJ -- :^ 7i si ►2 (Xi OJ ■^^ -H < S (5 2 2 c5 pa o s 3: Cd Qi c2 £ cS 5 5 5 Q u N S t- P 'C! ^ QJ < Z3 ^ cd rf cS 73 d oi 73 73 73 a rf cd cd 73 7t -^ rf 73 73 ca H a: < 4-3 4—3 4.J 4—3 ^-3 4-3 4-3 4-J 4-3 -t-d 4-:> ■*-J ■«— 3 *-^ ^j ■*-J 4-^ 4-) 4^ ••H ^ ri 73 a 73 73 73 a 73 cd cd c^ 73 ■ri 73 r3 73 73 ca !5 O CJ s S s 0) 5j Li CD s 55 s 5j CD Eh OJ tH s 5 0) 5j s 73 4-3 O 03 !/3 OJ T3 :3 O i3 CD S o 2 o CO O CO o o 73 00 "Si 73 03 o E 1: 1 •fH o 03 1 II 1 ca 4-3 ca 0) ■-H 3 • Q Q Q Q t:ij lij 00* : ^" '': • '5: • -i '^' ''f ^" ^* • [^' '3 o 0) iiU < S 03 ca ■::! "^ -5 zq 3S"^:o3o3 6 ^J^Q lijiiij^:^;-!: til ca C T3 T3 t/3 3 73 CO ■c: 3 "H CJ O o n -«o 0) ca 03 ca ^ ^ 3 3 fcS •i j CO CD CO »--4 00 00 00 oo c 00 rt ,— ^ t-H .r^ x: x: (-1 ^ N_^ o o o bJD ^^^ Clj cd CTJ o 00 ^—1 03 f— 4 00 OJ a, E 01 -a G 3 o XJ c a -a G 3 a -H 00 00 1— ( •<^ 3 5 ^" ^ T-H x_^ be o G c -a 0) 01 ^ a 03 C 3 OJ .-( C^J ct3 c:i rt ca Ti 7i ca ca Uh ■O o ex. a. a, Cm Q< c p '-^ Q. < 3 73 G s o O o o O 3 C cd ci cd ■^H tH -I-, ••-< 4^ 4_3 4^ Ol t-l ■t-J W ■t^ '- -s t-i 4-3 E-i 4^ E-, 4-3 L-, ■4J C ■4-> t-i ^ ■ i-H s s S. o o 03 O -^ O 03 ■4-J o o 03 O o 03 *-5 § .2 o o 03 o 4-3 r- < o d) QJ 0) Ol '3 < y) :*-l CO V-. 71 ^ M <:•-> M ;*-H CO :•— 1 CO CO CO ^ a. O O O O O < C PL, ■q o CL, o o o O 0. o O CL, <:•-> Q o cu cC "3 o . cu C < u o S .J 'S' z 3 _o < c^3 CU "^ ■-. l>3 OS « 75 o ••-< en C s o 3 1^ ca , — , ca >~> o •"■ O 7) CO 3 ca «o CO C/3 0. E 0) o • eg O XJ O • • Ol -.—1 o OJ Q. 3 03 o CJ 03 1 O, ca III 1 o 1 1 -^ ca ca S 03 O O) a • 03 S o o ex ca C/) tii li3 -i • • < dJ q:^ • < : C/2 qJ a; 03 •kJ X C3 O ^1 t3 0> "S >^ ■*o 3 a: "S O s l"l 0^ ■a < z O a o ca o ca O ■-• ca § ca a: o o o ca a: ^1 a a o 3 03 «^ 00 03 3 O ■a o >- 3 03 •—4 a, 03 ca o 03 ca ■a T3 4-3 >■ 0) T3 0) ca T3 -l-H 4-d G ca G O o _i -l-« u x: 2 c o ea s o 3 03 PL, 34 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers CD O 05 w 2 in w ;ii 00 < a w Cu • .,M On J3 ^ O .rH a u N w s ■— < "* Q = < w r^ =^ ? ^ < n I - < CQ cu H w 1^ Id CTJ U 6/) w =1 a ri. m Q < e ce =3 u z 03 w «o O c a> Q CD d X3 >- •t-t ^ ■cH ■^ oi 2 < CO CD 03 1-1 ^-^ /-^ 0 CO 1—1 ^ s /-^ 03 i> 0 C^ 1-H CO D-, 05 02 Oi 00 1-1 CO 00 »— ( d 0? lO ^ T3 G 03 Oi 00 (D a ■1— 1 1— t 1-\ 1-1 03 0:1 a a 0 < * > "3 > cc 0 -t-3 > CD 0 0 0 d X • 0 CO 0 0 t-^ ^^ s? i> ^-, 0 i-H 05 OJ ■a -a cc CO 00 00 1— ( 4-3 C (3i 00 CO •t-H i-H HI 00 1-H G d 4-3 c CC 4-j x: 0 Id tc ISJ i/3 Cfl Q^ c SI ■t-3 ■t-d C t-l 0) 4-> ^ •1— t -i4 QJ e J2 OJ 4-9 c» CD 0 cu X5 •r—t !^ 5- 0 0 0 CO I-H '? 0 G'3 s O ei -t^ 4J 0 (-H X3 05 rt CD (V &H cd < < s CQ a Pi D5 S s C 0 0 0 CO CO CO CO CO CO CO CO CO CO « C C c C < G c C C r— C«-t 0 0 0 0 0 U^ 0 0 0 0 0 0 X3 C X! ^ x: u ^ a J2 ^ j:3 a a. a a a. a < a. a, D. a, D. -rH •I-H H Ol u^ CO O) CW 0 CO CO C« CO CO C/D C3 C<5 3 D 3 — o i3 o -c: ^ o »5 1/3 o tyj CO CO &o to tio to 3 C3 ca •-H a tV) a 4-^ 0 0 CO 1/3 ■*o "^ \2 o 71 IZ) 3 1 «i C3 C i/j s mrgina S. vagi 3 CD 0 00 3 CJ 03 '-*4 0 tu 0 ••H •"-4 -i<; = ,, , t., CO n .L^ , . ^ CO 00 CO to CO CO CO a 40 ■-■ 3 «3 0 s 03 ••-1 0 "-J CO 0) cd 0) •T3 t3 ■a E CO a < 0) o CO 0) nJ c I— I o CQ No. 61) STASEK AUTOTOMV IN MOLLUSCA 35 in w u z u o: u b u as < U z OS Q U N i o H O H < is: < 0. 7] CI. c/3 < q: tu z u O < <; Q O D- O < cx w u < u < D O Q. O H (J c O CO Oi 0) 05 c on aj <: tn 3 D, O O o a a. o o O -a •I— ( X3 O D- O O O CD 00 oj m 2 c a, < u < b < z o < o CO ^-v Oi w 1-1 CO s*^ 05 CD Cvi '^"N W CO 05 05 ^ 05 lO T-t ,^^ 1— t (Ti CD V. c o i2 T-H bson 1960) D O c o -^ O B C 4-3 ex is tn G o Oi tH 0) nstrup ng (195 5S Q^ >a cu 3 o a, 0^ -t-3 00 o 2 ° CO O Q OJ en a o o G CX C<3 . 3 Vi Q, 3 ■a -3 o o. o o o S 0) -a ^ (U N) o CJ o 3 (D CO O ca (U --: x: ■2 o ■a N c o a o S K rf X cd 3 CO c O h IT( 1.1 OJ o ^ o J3 >^r) CO 3 o >.. E-i 4-3 w <1) 'O JG ■^ o o J5 CO 3 >5 3 o O 3 3 O 03 4-3 a c o fcJD 36 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers DISCUSSION Autotomy, as the term is used in this paper, does not encompass onto- gentic losses. In the pteropod Herse columnella, and possibly in the Argo- nautacea, abscission of the male copulatory apparatus can be regarded as a loss of this type. The former is included in this paper as -an instance of autotomy because of the lack of information on the circumstances under which actual separation takes place. The latter is included for the same reason and because it is not positively known whether regeneration does or does not occur. Both phenomena might fall under reproductive autotomy in Giard's classification. However, that evasive qualities are involved seems to be a possibility. I have seen no references to parasitic autotomy in the Mollusca. Fur- thermore, what might be interpreted as instances of economic autotomy in the phylum, as in Peltodoris atromaculata, Limax marginatus, and Solen ro- saceous, seem always to be unnatural reactions brought about by preserva- tion or by unfavorable laboratory conditions. Identical, but probably evasive responses are known to occur naturally in related forms. Some instances of economic autotomy, however, have not been observed outside the laboratory (for example, in scaphopods). Therefore, in my opinion, the great majority of instances of autotomy in the Mollusca have, or, where definitive information is lacking, can rea- sonably be predicted to have, evasive qualities. In most instances, the na- tural predator or the situation inciting autotomy is unknown and can only be surmised at this time. The part to be autotomized is directly exposed to the environment as a projecting extremity. In some instances, for example, the cerata of opis- thobranchs and many of the tentacles of Lima, the autotomous parts are up- wardly directed, more or less perpendicular to the direction taken during lo- comotion. This position might suggest a large or natant predator that normal- ly approaches from above. In other instances, the autotomous portion is the last, or most posteriorly projecting part of the body. Gastropods that shed the tail-end of the foot, bivalves that release fragments of the siphons, and cephalopods that autotomize parts of the arms, which are trailed behind dur- ing escape locomotion, are called to mind. In these examples autotomy takes place in a direction opposite that taken during locomotion, and the autoto- mous piece is that likely to be first seized by a pursuing organism. Other instances more difficult about which to conjecture include those nudibranchs that shed any portion or whole of the mantle edge, scaphopods that lose the No. 61) STASKK: AIITOIOMV IN MOLLUSCA 37 anterior captacular bundles, and Solen siliqiia, which has been said to pinch off the tip of the foot. Pilsbry (1948) reported that the body cavity, and hence the vital or- gans of Prophysaon do not extend into the autotomous end of the foot. Such may also be deduced from Quoy and Gaimard's description (1832) to be true of Harpa. Indeed, in every example, organs necessary for continued exist- ence of the individual would never be expected to be included in the part lost, although in most instances anatomical details are not known. Most of these mollusks have one or more restricted zones of weak- ness and localized arrangements to facilitate autotomy. Nudibranchs that lose the edge of the mantle and octopods that autotomize variously sized pieces of the arms are examples of mollusks having a more or less diffuse autotomous capability. There is a significant dichotomy in the ways in which the Mollusca replace lost parts. Most often, removal of a piece of the anatomy results in a distal proliferation of cells at the site of the wound. This is regeneration in the strict sense and normally results in the production of a single part, in appearance quite like that lost. The captacula of scaphopods, the siphons of Solecurtus and Solen, and the dorsal arms of Tremoctopus are autotomous organs in which the part lost is serially replaced rather than regenerated. The mantle tentacles of Lima are possibly also to be included here, if these grow from the base rather than at the tip. In serial replacement there is a proximal zone of pro- liferation that is presumably always in activity and that forces older por- tions outwardly, much as in strobilization of a cestode flatworm or of a coelenterate scyphistoma. Whether proximal proliferation is actually con- tinuous, or is stimulated by injury to more distal autotomous elements, pos- es a fascinating question. Serial replacement seems to have advantages over regeneration in that a new and functional element is immediately avail- able, while in regeneration some variable, and perhaps critical period of time is required before a new element is produced. Finally, most autotomous structures in the Mollusca play some other, more usual role in the physiology of the organism, perhaps in locomotion, respiration, or feeding. Rarely, as in females of Tremoctopus, does the part to be autotomized seem to have no other role than that of decoy. In this sense autotomy may be regarded as a specialized condition superimposed upon the normal functions of an organ. It is, however, a condition having its source in the primitive ability of all organisms to regenerate parts lost to hostile factors or accidents of the environment. 38 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers ACKNOWLEDGMENTS The above reported observations of Solen sicarius were made while the author was on a grant from the Office of Naval Research, Department of the Navy, and the University of Washington, NR 104-142, It is a pleasure to acknowledge Dr. R. L. Fernald, Director of the Friday Harbor Laborator- ies, and Dr. D. L, Ray for their aid and consideration during my tenure there. Mr. AUyn G. Smith took the photograph of Lima and gave permission to include it here. He, Dr.Dwight Taylor, Mr. Lawrence C.Andrews, and Mr. Paul Gregory directed me to references on autotomous mollusks or re- lated matters. Dr. A. Portmann, University of Basel, kindly permitted me to use his figure of Tremoctopus. Mr. Steve D. Nagy translated significant parts of the reference by Rawitz. Dr. Ralph 1. Smith, Dr. Ross Pohlo, and Mr. Dustin Chivers had cogent criticisms of the original manuscript. Miss Joan Steinberg and Dr. Melbourne R. Carriker helped, respectively, with the taxonomy of nudibranchs and with possible sources of ecological informa- tion. To all these persons I hereby express my sincere appreciation. This work has been supported by National Science Foundation Grant GB 1535. REFERENCES CITED Adams, H., and A. Adams 1853- The genera of Recent Mollusca. John Van Voorst, London, 3 vols. 1858. Allan J. K. 1932a. Australian nudibranchs. Australian Zoologist, vol.7, pp. 37-105. 1932b. A new genus and species of sea-slug, and two new species of sea- hares from Australia. Records of the Australian Museum, vol. 18, pp. 314-320. 1941. New parts for old in the shell-fish world. Australian Museum Maga- zine, Sydney, vol. 7, pp. 280-285. 1959. Australian shells. Georgian House, Melbourne, pp. 1-487, 1-XXl, pis. 1-44. Annandale, N. 1916. Fauna of the Chilka Lake. Mollusca Gastropoda and Lamellibranchi- ata. Memoirs of the Indian Museum, Calcutta, vol. 5, pp. 327-366. Atkins, D. 1937. On the ciliary mechanisms and interrelationships of lamellibranchs. Ft. III. Quarterly Journal of Microscopical Science, vol. 79, pp. 375-421. Baha, Kikutaro 1949. Opisthobranchia of Sagami Bay. Publ. by Iwanaini Shoten, Tokyo, pp. 1-194, pis. 1-50. No. 6n STASEK: AUTOTOMY IN MOLLUSCA 39 l<\H\, K., AM) I. HAMATAM 1965. The anatomy of Sakuraeolis enosimensis (Baba, 1930), N.G. (=//ervia ceylonica C) Eliot, 1913) (Nudibranchia - EolidoideaV Publica- tions of the Seto Marine Biological Laboratory, vol. 13, pp. 103-113. Benson, W . H. 1835. Conchological notices; chiefly relating to the land and freshwater shells of the Gangetic provinces of Hindoostan. Zoological Jour- nal, vol. 5. pp. 458-466. Bkrgm, H. 1884. Report on the Nudibranchiata dredged by H.M.S. Challenger. Report of the scientific results of the voyage of H.M.S. Challenger, pt. 26, zoology, vol. 10, pp. 1-154. 1891. Die cryptobranchiaten Dorididen. Zoologischeii Jahrbiichern, vol. 6, pp. 103-144. Berkeley, E., and C. Berkeley 1954. Notes on the life-history of the polychaete Dodecaceria fewkesi (noni. n.). Journal of the Fisheries Research Board of Canada, vol. 11, pp. 326-334. BiNNEY, W. G. 1892. A fourth supplement to the fifth volume of the terrestrial air-breath- ing mollusks of the United States and adjacent territories. Bulletin of the Museum of Comparative Zoology, Harvard University, vol.22, pp. 163-204. Bloomer, H. H. 1907. On the anatomy of Ensis macha, Solen fonesii, and S. viridis. Pro- ceedings of the Malacological Society of London, vol. 7, pp. 18-19. Bonnevie, K. 1916. Mitteilungen iiber Pteropoden. I. Beobachtungen iiber den Geschlechts- apparat von Cuvierina columnella Rang. Jenaische Zeitschrift fiir Naturwissenschaft, vol. 54, pp. 245-278. Carlisle, J. G., J. W. Schott, and N. J. Abramson 1960. The barred surfperch (Amphistichus argenteus Agassiz) in southern California. State of California, Division of Fish and Game, Bureau of Marine Fisheries, Fish Bulletin no. 109, pp. 1-79. Cockerell, T. D. a. 1890a. New western slugs. The Nautilus, vol. 3, pp. 111-113. 1890b. On the generic position oi Arion foliolatus, Gould. The Nautilus, vol. 3, pp. 126-128. Coe, W. R. 1944. Sexual differentiation in mollusks. II. Gastropods, amphineurans, scaphopods,andcephalopods. The Quarterly Review of Biology, vol. 19, pp. 85-97. Cooke, A. H. 1895. Molluscs in The Cambridge natural history. Macmillan and Co., New York, pp. 1-535. Crosse, H. 1860. Note sur les animaux de trois helices de Cuba. Joui'nal de Conchyli- ologie, ser. 2. vol. 4, pp. 225-233. 40 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers Edmunds, M. 1966. Defensive adaptations of Stiliger vanellus Marcus, with a discussion on the evolution of 'nudibranch' molluscs. Proceedings of the Mal- acological Society of London, vol. 37, pp. 73-81. Eliot, C. 1899. Notes on tectibranchs and naked mollusks from Samoa. Proceedings of the Academy of Natural Sciences of Philadelphia, December, pp. 512-523. Eaussek, V. 1897. Biologische Beobachtungen ueber Lamellibranchiaten. II. Die Auto- tomie der Siphone bei Solen und Solecurtus. (In Russian.) Travaux de la Sociele' Impe'riale des Naturalistes de St. Petersbourg, vol. 28, pp. 249-270. Fischer, P. 1880- Manuel de conchyliologie. Paris, pp.I-XXIV, 1-1369, pis. 1-23. 1887. FiSHELSON, L., AND G. QiDRON-LaZAR 1966. Footautotomy in the gastropod Genavaria (Prosobranchia: Trochidae). The Veliger, vol. 9, p. 8. Forbes, E., and S. Hanlev 1853. A history of British Mollusca and their shells, vol.1, John Van Voorst, London, pp. I-LXXX, 1-846, pis. 1-64. Fredericq, L. 1882a. Amputation des pattes par mouvement reflexe chez le crabe. Arch- ives de Biologie, vol.3, pp. 235-240. 1882b. Sur la rupture de la queue chez I'orvet. Bulletins de I'Academie Roy- ale des Sciences, des Lettres et des Beaux-arts de Belgique, Bru- xelles, ser. 3, vol. 4, pp. 213-214. 1883. Sur I'autotomie ou mutilation par voie reflexe comme moyen de de- fense chez les animaux. Archives de Zoologie Experimentale et Ge''ne'rale, Paris, vol. 10, pp. 413-426. 1886. Les mutilations spontane'es ou I'autotomie. Revue Scientifique (revue rose), Paris, ser. 3, vol. 12, pp. 613-620. 1887. L'autotomie chez les etoiles de mer. Revue Scientifique (revue rose), Paris, ser. 3, vol. 13, pp. 589-592. Ghosh, E. 1916. The anatomy of the common solen of the Chilka Lake. -• ?A dwarfed form of S. fonesi. Memoirs of the Indian Museum, Calcutta, vol. 5, pp. 367-374. 1920. Taxonomic studies on the soft parts of the Solenidae. Records of the Indian Museum, vol. 19, pp. 47-78. GlAKD, A. 1887. L'autotomie dans la serie animale. Revue Scientifique (revue rose), ser. 3, vol. 13, pp. 629-630. GlLMOlR, T. H. L 1963. A note on the tentacles of Lima hians (Gmclin) (Bivalvia). Proceed- ings of the Malacological Society of London, vol.35, pp. 81-85. No. 61) STASFK: Al'TOTOMV IN MOLLUSCA 41 GONOR, J.J. 1961. Observations on the biology ot Lobiger sermdifalci, a shelled saco- glossan opisthobranch I'roni the Mediterranean. Vie et Milieu, vol. 12, pp. 381-403. CiKAllAM, A. 1938. The structure and function of the alimentary canal of aeolid molluscs, with a discussion on their nematocysts. Transactions of the Royal Society of Edinburgh, vol. 59, pp. 267-307. Gregg, W . O. 1943. Notes on land slugs of southern California. Minutes of the Concho- logical Club of Southern California, no. 25, pp. 3-8. 1944. Notes on land slugs of Los Angeles and Orange counties, California. The Nautilus, vol. 57, pp. 109-115. Grimpe, G. 1928. Uber zwei jugendliche Mannchen von Argonauta argo, L. Zoologisch- en Jahrbiichern, vol. 45, pp. 77-98. Haefelfinger, H.-R. 1961. Beitrage zur Kenntnis von Pellodoris atromaculata Bergh 1880 (Mol- lusca, Opisthobranchiata). Revue Suisse de Zoologie, vol. 68, pp. 331-343. Haefelfinger, H.-R., and R. A. Stamm 1959. Limenandra nodosa gen. et spec. nov. (Nudibranch., Aeolidiidae propr.), un opisthobranche nouveau de la Mediterranee. Vie et Mi- lieu, vol. 9, pp. 418-423. Heath, H. 1917. The anatomy of aneolid, C/iioraera dalli. Proceedings of the Academy of Natural Sciences of Philadelphia, vol. 69, pp. 137-148. Hecht, E. 1895. Contribution a I'etude des nudibranches. Memoires de la Socie'te Zoologique de France, vol. 8, pp. 539-712. Hedley, C. 1902. A day on the Great Barrier Reef. The Nautilus, vol. 15, pp. 97-100. Hoyle, W. E. 1908. Presidential address. Report of the 77th meeting of the British Asso- ciation for the Advancement of Science (Leicester), pp. 520-539. Hyman, L. H. 1955. The Invertebrates: Echinodermata. McGraw-Hill Book Company, Inc., New York, pp. 1-763. Ingram, U'. M. 1948. Terraria observations on Prophysaon andersoni (J. G. Cooper). The Nautilus, vol. 62, pp. 17-19. Jatta, G. 1896. Fauna und Flora des Golfes von Neapel, Monogr. 23, I Cefalopodi, pp. I-XI, 1-268, pis. 1-31, Lacaze-Duthiers, H. 1856. Histoire de I'organisation et du developpement du dentale. Annales des Sciences Naturelles (Zoologie), Paris, ser. 4, vol. 6, pp. 225- 281, 319-385. 42 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers Lance, J. R. 1962. A new Stiliger and a new Corambella (Mollusca: Opisthobranchia) from the Northwestern Pacific. The Veliger, vol. 5, pp. 33-38. Lane, F. W. 1960. Kingdom of the octopus. Sheridan House, New York, pp. I-XX, 1-300. Lange, M. M. 1920. On the regeneration and finer structure of the arms of the Cephalo- poda. The Journal of Experimental Zoology, Philadelphia, vol. 31, pp. 1-40. 1966. Studies of opisthobranchiate mollusks of the Pacific Coast of North America. Memoirs of the California Academy of Sciences, vol. 6, pp. I-XVI, 1-546. LOVETT, A. L., AND A. B.BlACK 1920. The gray garden slug. Oregon Agricultural College Experimental Sta- tion, Bulletin 170, pp. 1-43. MacFarland, F. M. 1912. The nudibranch family Dironidae. Zoologischen Jahrbiicher, Supple- ment 15, pp. 515-536. MacFarland, F. M., and C. H. O'Donoghlie 1929. A new species of Corambe from the Pacific Coast of North America. Proceedings of the California Academy of Sciences, ser.4, vol. 18, pp. 1-27. Macnae, W. 1954. On some eolidacean nudibranchiate molluscs from South Africa. An- nals of the Natal Museum, vol. 13, pp. 1-50. Morse, E. S. 1919. Observations on living lamellibranchs of New England. Proceedings of the Boston Society of Natural History, vol. 35, pp. 139-196. Oehlert, D. 1886. L'autotomie et les amputations spontanees. Revue Scientifique (revue rose), Paris, ser. 3, vol. 12, p. 701. Orclitt, H. G. 1950. The life history of the starry flounder Platichthys stellatus (Pallas). State of California, Division of Fish and Game, Bureau of Marine Fisheries, Fish Bulletin No. 78, pp. 1-64. Ostergaakd, J . \L 1955. Some opisthobranchiate Mollusca from Hawaii. Pacific Science, vol. 9, pp. 110-136. Parona, C. 1891. L'autotomia e la rigenerazione della appendice dorsalli (Phoenicur- us) nella Tetbys leporina. Zoologischer Anzeiger, vol. 14, pp. 293- 295. Pelseneer, p. 1906. A treatise on zoology, pt. 5, Mollusca. E. Ray Lankester, editor. Adam and Charles Black, London, pp. 1-355. 1935. Essai d'ethologie zoologique d'apres I'etude des moUusques. Acad- demie Royale de Belgique, Publications de la Foundation Agathon de Potter, no. 1, pp. 1-662. No. 6n STASHK: AUTOTOMY IN MOLLUSCA 43 PfEIFFER, I-., AND J. Gl'NDI.AC H 1861. Zur MoUuskeiitauua der Insel Cuba. Malakozoologische Blatter, vol.7, pp. 9-32. PiisnKV, H. A. 1948. LandMoUusca of North America (North of Mexico), vol.2, no. 2. Acad- emy of Natural Sciences of Philadeli)hia, Monograph no. 3, pp. I- XLVII, 521-1113. POHI.O, R. 1963. Morphology and mode of burrowing in Siliqua patula and Solen ro.sa- ceus (Mollusca: Bivalvia). The Veliger, vol. 6, pp. 98-104. POLI, J. X. 1791. Testacea utriusque Siciliae...(Parmae), vol. 1, Dedication, 4 pages (not paginated); Praefatio, pp. I-X; De Vita...Josephi Xaverii Polii, by Stephano delle Chiaie, pp. I-XXVI; Introductio ad Scientiam Testaceorum, pp. 1-90; Ordo Primus complectens Testacea Multi- valvia, pp. 1-50 [T]; Ordo Secundus complectens Testacea Bivalvia, pp. 1-74 [Tj ; four indexes to the whole of the preceding, and ex- planations of plates I-XVIII, pp. LXXIII. [Adapted from collation in the Catalogue of the British Museum (Natural History), vol. IV, P-SN, p. 1592, 1913T] PORTMANN, A. 1952. Les bras dorsaux de Tremoctopus violaceus Delle Chiaje. Revue Suisse de Zoologie, vol. 59, pp. 288-293. Prlvot- Fol, a. 1934. Les doridiens de Cuvier. Journal de Conchyliologie, vol.78, pp. 209- 261. Qloy, J.R.C, and J. P. Gaimard 1832. Voyage de decouvertes de I'Astrolabe. Paris. Zoologie, vol. 2, pp. 1-686. 1833. Voyage de la corvette I'Astrolabe. Pans. Atlas, Zoologie, vols. 1 and 2. 1834. Voyage de decouvertes de I'Astrolabe. Paris. Zoologie, vol. 3, pp. 1-954. Rawitz, B. 1892. Der Mantelrand der Acephalen. Jenaische Zeitschrift fiir Naturwis- scnschaft, vol. 27, pp. 1-232. Raymond, W. J. 1890. Why does Prophysaon shed its tail! The Nautilus, vol. 4, pp. 6-7. Reynaud, a. 1834. Observations sur I'animal de la harpe. Memoires de la Societe'' d'His- toire Naturelle de Paris, vol. 5, pp. 33-40. RlCKETTS, E. F., AND J. CaLVIN 1952. Between Pacific tides. Stanford University Press. Stanford, Cali- fornia, pp. 1-502. Thud edition, revised. Riggenbacu, E. 1903. Die Selbstverstlimmelung der Tiere. Ergebnisse derAnatomie undEnt- wicklungsgeschichte, vol. 12, pp. 782-903. 44 CALIFORNIA ACADEMY OF SCIENCES (Occ. Papers RiSBEC, J. 1928. Contribution a I'etude des nudibranches Neo-Caledoniens. Faune col- onies Francaises, vol. 2, pp. 1-328. ROBSON, G. C. 1926. On the hectocotylus of the Cephalopoda -- a reconsideration. Pro- ceedings of the Malacological Society of London, vol. 17, pp. 117- 122. 1929. A monograph of the Recent Cephalopoda, pt. I: Octopodinae. British Museum, London, pp. I-XI, 1-236, pis. 1-7. 1932. A monograph of the Recent Cephalopoda, pt. II: The Octopoda. British Museum, London, pp. I-XI, 1-359, pis. 1-6. Semper, K. 1881. Animal life. International Scientific Series. D. Appleton and Company, New York, pp. I-XVI, 1-471. Steenstrl'p, J. J. S. 1857. Hectocotylus-formation in Argonauta and Tremoctopus explained by observations on similar formations in the Cephalopoda in general. The Annals and Magazine of Natural History, ser. 2, vol. 20, pp. 81-114. Straughan, R. p. L. 1957. The octopus you didn't get. The Aquarium, vol. 26, pp. 135-138. Tesch, J.J. 1946. The thecosomatous pteropods. I. The Atlantic. Dana Report, no. 28, pp. 1-82. Thiele, J. 1931- Handbuch der Systematischen Weichtierkunde. Jena, 2 vols., pp. 1- 1935. 1154. Thompson, T. E. 1960. Defensive adaptations in opisthobranchs. Journal of the Marine Bio- logical Association of the United Kingdom, vol.39, pp. 123-134. Tryon, G. W. 1882. Structural and systematic conchology. Philadelphia, vol. 1, pp. I-VIII, 1-453, pis. 1-140. Varigny, H. dk 1886. L'amputation reflexe des pattes chez les crustaces. Revue Scientif- ique (revue rose), Paris, ser. 3, vol. 12, pp. 309-310. Wells, M. J. 1962. Brain and behaviour in cephalopods. Stanford University Press, Stan- ford, California, pp. 1-171. WOLCOTT, G. N. 1924. The food of Porto Rican lizards. Journal of the Department of Agri- culture of Porto Rico, vol. 7, pp. 1-37. YOUNC, J. /.. 1959. Squids, cuttlefishes and octopuses. Proceedings of the Royal Institu- tion of Great Britain, vol. 37, pt. 4, no. 168, pp. 394-411. »