ARY IS1TY OF :ORNiA MURICIN^E. PLATE 1. MANUAL OF GEOLOGY; STRUCTURAL AND SYSTEMATIC. WITH ILLUSTRATIONS OF THE SPECIES. BY GEORGE W./TRYON) JR. / CONSERVATOR OF TUK CONCHOLOGICAL SECTION OK TUK ACADEMY OF NATURAL SCIENCES OF PHILADELPHIA. VOL. II. MURICIN^ PURPURINJE, ^4y / 4*4 f PHILADELPHIA: Published by the Author, ACADEMY OF NATURAL SCIENCES, COR. ISTH & RACE STS. 1880. SCIENCE! ySjj LIBRARY WM. P. KILDARE, PilNTER, 734 & 736 SANSOM ST , PHILA. IN THE preparation of the introductory portion of the present volume free use has been made of Keferstein's admirable work, forming part of" Bronn's Klassen und Ordnungen d. Thierreichs:'' — the most comprehensive treatise on the anatomy of the mollusca which has }^et appeared. Of course many additional facts and some corrections have been published since the issue of that work, and abstracts of the most important of these have been herein incorporated. I have endeavored to avoid the prolixity of Keferstein without sacrificing clearness of language, but occa- sionally quote his very words. The few statements for which I alone am responsible, are made in the first person. The body of literature consulted is too extensive for quotation in a work like this, where much must necessarily be sacrificed for the sake of brevity. Whilst 1 have freely used the admirable illustrations of Keiner, Reeve, Sowerby, Kiister, and other monographs. I have added to these hundreds of figures, names and items of infor- mation gleaned from fugitive papers and scientific periodicals ; so that I am enabled to present a ca refully arranged, concise account of all that is known concerning the Murices and Purpurte. I may be thought 03- some, to have treated the species in a too conservative spirit; my object in the investigations of which this work is the result, has been solely to ascertain truth, not to establish, maintain or destroy scientific reputations. January. 1880. (T. W. T.. JR. If there be nothing new, but that which is, Hath been before, how are our brains beguil'd, Which, laboring for invention, bear amiss, The second burden of a former child ! SHAKSPERE. LIBRARY UNIVERSITY OF] PROSOBRANCHIATA. ANATOMY. External Features. The body in prosobranchiates presents the following, more or less, differentiated portions : 1. Head. This is sometimes partially separated from the body by an intervening- constriction or neck, but is very 'frequently only indicated by the possession of a mouth on its ventral and of a pair of tentacles on its dorsal surface. From the latter sometimes proceed pedicels bearing the eyes. 2. Mantle. This covers the posterior portion of the dorsal surface of the body. 3. Foot. Arising from the ventral anterior part of the body. The mouth may be either simple, in the plane of the head, or it may be produced at the end of a contractile rostrum, or of a retractile proboscis. Behind the head and on the under side of what ma}^ be called the neck is attached the foot, and behind the neck begins the posterior portion of the body covered by the mantle : this part of the animal is Arery long, usually, and as the shell is moulded upon it, it assumes a similar form — general!}" spiral. However, in Patella and similar limpet-like shells, the body is short and straight. The mantle encircles the body like a collar below, but above it is much produced over the back of the animal, and the respiratory cavity forms a sinus of its margin. The foot consists of the attachment to the neck, which is usually narrow, and an expanded portion or sole, or creeping disk. Such is its typical form, but in Patella the foot embraces the entire ventral surface of the animal. It is a muscular pro- cess of the body, and the only locomotive organ possessed by prosobranchiates. ('» EXTERNAL FEATURES. Viewed externall}7, the prosobranchiates, as well as the mol- lusca in general, must be regarded as bilaterally symmetrical ; a view which is strengthened by the position of the nervous system ; and the developmental history of the intestines, though its actual position, as well as that of the sexual organs, does not correspond with this symmetry. Thus we find that at first the anus is at the posterior end of the body, but gradually it ap- proaches nearer and nearer to the anterior end until finally it opens on the back or right side near the mouth. In its early stages the shell and mantle only occupy the extreme posterior end of the body, but in their growth they cover more and more of the latter, gradually pushing the anus forward. An external shell, usually sufficiently large to contain the entire animal, is common to all the prosobranchiatee. It is a secretion of the mantle, and conforms to its shape; and the hard- ness which it assumes by the addition of carbonate of lime, renders it an efficient protection to the animal, whilst the faith- fulness with which it reproduces the external features of the latter renders it extremely useful in classifying the niollusca. The spiral growth of shells is as nearly of true mathematical regularity as is possible in an organic body, forming the loga- rithmic spirals of Moseley. or conchospiral of Naumann. Cor- responding to the shell, which is produced by the mantle, and borne by the posterior portion of the body, the posterior dorsal part of the foot bears an operculum, secreted by an expansion of its skin called the opercular mantle. Typically, the operculum is a spiral also, but in the same plane ; yet in many cases its growth is annular. Usually horny, it is sometimes nearly, entirely or partially calcareous, and on the retirement of the animal within the shell it is brought into the aperture of the latter, which it more or less complete!}7 closes. Like the shell itself, it may be considered a protective organ, and when in apposition with the former suggests the two enclosing valves of the lamellibranchiate or bivalve mollusks. Notwithstanding the large portion of the animal which is always within the shell, even when the head and foot are ex- truded, the latter is only attached to the former at one point, on the columella, and by means of a columellar muscle, which, pass- ing through the foot, is attached at its other end to the oper- THE SKIN. 7 culum (when the latter is developed, which is not always the case . Although there is only this single actual bond of con- nection between shell and animal, the contact of the body serves to maintain the vitality of the shell, which soon bleaches, and finally decays when separated from its architect and inhabitant. We will now describe, in a necessarily cursory manner, successively, the skin, the foot, the muscles, the shell and operculum, the digestive, nervous, vascular, respiratory, secre- tive and sexual organs of the prosobranchiates. The Skin. The body is completely enveloped by MH external skin. Its epithelial layer is formed of quadrangular or prismatic cells, which have a distinct nucleus, and occasionally, when long- cylindrical in form, they have a tail-like end beneath, penetrating the cutis.* Externally the epithelium sometimes supports cilia upon the exposed portions of the body. The cutis consists essentially of tine interlacing muscular fibres with interposed cells; often attaining a perceptible thickness. The subcutaneous muscular layers of the body are immediately continuous with those of the cutis: the fibres of which they are formed may be clearly distinguished as an outer longitudinal and inner circular layer. These fibres are i as in molliisks gene- rally) compressed and band-like, with pointed ends and central oval nucleus. The skin forms a fold above and surrounding the foot, and this portion is technically termed the mantle; it surrounds the body, behind the cephalic portion like a collar, and thence spreads dorsally over the posterior part of the animal. In the terminal, thickened border of the mantle, the cutis becomes of greater thickness ; its upper stratum containing very numerous glands, furnishing the mucus and colors which are here mingled with the secreted shell-material. Similar glands, furnishing a copious supply of mucus are found also in the cephalic portion of the body as well as in the foot, and especially in the sole ; but no cells are found in that part of the animal permanently covered * It is possible that these last are epithelial terminations of nerves— hence sense organs. 8 THE SKIN. by the shell. Frequently the external skin is colored by a gran- ular pigment, which either is contained in cells, lying between or enveloping the glands, or else sometimes appears to lie free under the epithelium. The mantle border is the principal agent in the secretion of the shell : it is thrown out by the epithelial layer as a sort of cuticular development. With the organic basis of this secretion is mingled carbonate of lime, originating in the epithelial cells, where it may be separated from the blood : in hardening, the exuded material becomes half crystalline or laminated. Usually the external layer of the shell is a transparent or translucent skin, the epidermis ; having no lime in its composition. It is often colored by pigments lying in the outermost border of the mantle. Whilst the growth of the shell is thus provided for by addi- tions of the aperture margin from the mantle border, the whole mantle is equally capable of producing shelly substance ; and not only are shells thus thickened from within by the mantle surface, but breaks are repaired with new material by a similar provision : only such repaired and interior portions are devoid of epidermis and of color, the pigments being found only in the free border of the mantle. Mollusks are even able to secrete shelly matter to provide against threatening- dangers from the boring of other animals into their shell.* In certain genera, as Haliotis, Turbo, etc., the inner shelly layer is nacreous — that is, it forms mother- of-pearl. The mantle border by means of its sphincter muscles embraces the body closely, thus closing the mantle cavity except at one point, where a small opening allows the ingress and egress of water for respiration. This respiratory opening is a semi-circular notch, formed by muscles, and is sometimes prolonged on its dorsal wall into a half-closed tube or respiratory siphon; which, when present, assists by the phases of its development in the classification of the mollusca. This siphon usually forms an an- terior notch in the shell near the margin of the columella and the * A curious example of shell secretion by the visceral mantle occurs in a cone belonging to the cabinet of the late Dr. Gray. A section of this shell has been made, showing holes bored into the spire end by lithodomi and the repeated walls erected by the animal across the ends of the whorls to protect itself against the ravages of its insidious enemies, (plate 2, fig. 1.) THE SKIN. 0 existence of the latter thus predicates that of the former. The siphonal tube is sometimes greatly prolonged, and is then fre- quently covered for most or all its length by a prolongation of the aperture, which is technically known as the canal of the shell. The canal in Mu rex and Pusus is extremely long, at least in the typical species. Mollusks of which the shells are furnished with a canal or anterior notch are called siphonostomata, the first great division of the prosobranchiates. The siphon is principally confined to predatory or carnivorous mollusks. The second great division, termed holostoniata, have rounded apertures ; consequently no siphon but simply an opening for respiration. They are vegetable feeders usually (Natica is a remarkable ex- ception), and close the aperture of the shell completely by their operculum. At the posterior left border of the mantle, behind the branchht, is sometimes an opening from which a small siphon extends back- wards, and when it is present, it forms a notch in the posterior part of the shell, as in Cypraea and Conus, or a canal as in Ovula, or frequently it only forms a callosity on the upper part of the columella. close to its junction with the posterior part of the aperture margin. Probably this siphonal opening is for the exit of the water that has entered lay the branchial opening. In many of the siphonostomata it is not present. The mantle border can be freely withdrawn within the whorl, as it is not united to the shell at any point. It is frequently- prolonged into digitations, or exhibits prominences or invagina- tions, all of which develop similar features on the shell ; thus giving rise to the fingers of Pteroceras, the spines of Murex, etc. Occasionally, howrever, processes of the mantle do not secrete shelly coverings : Cerithium and the oriental Melanians, for in- stance, have delicately digitated mantle margins, these digita- tions forming no secretion, and sometimes thrown back over the shell. The mantle is occasionally largely developed into side lobes, which in Marginella and Cypraea are so extended as to be thrown up over the external surface of the shell, nearly or completely covering it. In such shells an epidermis is not present. The mantle lobes of Cypr*ea are beset with numerous papillae, which seem to partake the function of tentacles as tactile organs. In 2 10 FOOT. other genera, as in Oliva, the mantle is prolonged into filiform processes before and behind. The female Vermetus has the mantle cleft in the middle, ac- cording to Lacaze-Duthiers, although there is no corresponding cleft in the shell, and in Haliotis a similar mantle cleft impresses a groove in the shell, in which are situated the row of holes characteristic of the genus. The shell of Pleurotoma also has a sinus corresponding to a cleft mantle. The cause of the sutural sinus of the shell of the American fresh-water genus Schizostoma is as yet unknown; it may be due to a similar cause or it may be sexual. As the genus is restricted to the Coosa River and its neighborhood, I am inclined to think that it is a local disturbance of growth, especially as most of the species could not be distinguished from corresponding forms of Gonio- basis except by the lip notch or slit. Foot. The foot is a fleshy, expanded mass, attached to the under side of the body, in front of the mantle by a peduncle. In the heteropods the foot is divisible into three portions, termed, respectively, propodium, mesopodium and metapodium ; but in the typical gasteropods these three areas are blended in the sole, although the metapodium is indicated by the fact of its support- ing on its dorsal side the operculum. In Strombus (pi. 3, fig. 14), a transverse furrow separates the mesopodium from the propodium, and the metapodium is covered downwards and in front by the operculum. The peduncle of the foot is usually short and depressed", and covers the under side of the body between the mantle collar and mouth, the foot being expanded forward, but more extensively backwards ; but in Strombus and its allies the operculum is long and narrow, whilst the foot is slim and cylindrical. Haliotis, Patella and Chiton have the foot, on the other hand, very much expanded. Rapidity of motion appears to be in inverse ratio to the size of the foot ; those genera in which this organ is enor- mously developed, especially in those just cited, where it occu- pies the entire ventral surface of the body, being slow in movement. FOOT. 11 Generally, the anterior border of the foot is variously lobed (plate 3), and these lobes are mostly of glandular structure, though Keferstein has found them in Buccinum to consist of interwoven meshes of muscular fibres, in which are lodged beau- tiful nucleated cells. These lobes or filiform processes, like those of the mantle, are tactile organs. When the foot is greatly expanded, as in Cymba, Harpa, Doliuin, etc., it is generally laterally recurved over the shell, somewhat like the mantle in Cyprsea and Marginella. In Oliva (pi. 3, fig. 16) and in Ancillaria, the propodium, repre- sented by triangular lobes, lies flat, while the distinctly separated mesopodium covers a portion of the sides of the shell. In Natica the propodium is greatl}' developed forwards, extending beyond the head and reflected backwards over the latter in such manner as to conceal it, with its tentacles, and the anterior part of the shell itself, from view. The mesopodium in Natica, being like- wise largely reflected over the shell, the respiratory cavity is covered, but a canal is developed between the reflected pro- podium and the mesopodium, which conveys water to the branchiae. There is a rounded glandular opening in the sole of the foot of many prosobranchiates (Pyrula, pi. 3, fig. 18), which is the external opening of the pedal aqueous vascular system ; to be more particularly noticed hereafter. Nearly all of the prosobranchiates whilst in the larval state support an operculum on the dorsal side of the metapodium ; with some, the operculum is fugatious, and is lost in the trans- formation of the animal, yet it continues present in most of the groups. Its secretion from the operculigerous mantle, and its morphological relation to the shell, will be explained further on. The opercular mantle sometimes extends beyond the borders of the operculum itself, and is divided into processes or filaments, which may, as in Ampullaria, be reflected over it. The foot is the organ of locomotion of univalve mollusks : the anterior portion is protruded, and then by means of wave- like muscular contractions of the sole, the whole mass of body and shell is brought forward — when the former motion is re- peated. In Phasianella, Quoy and Gaimard have observed that the foot is divided lengthwise by a furrow, and that when motion 1 '2 COLUMELLAR MUSCLE AND OPERCULUM. is rapid the muscles of the sole are alternately used on either side, so that the effect of the motion is that of a pair of feet. In attached shells, like Vermetus, the foot is only rudimentary and serves merely as a support to the operculum. In retiring within the aperture of the shell the foot is generally doubled upon itself across the middle, so that its dorsal posterior side, bearing the operculum, comes outermost ; but in Oliva and Voluta it folds longitudinally, whilst the quadrate foot of Conus is withdrawn obliquely, without folding, first the right and then the left side. The Columellar Muscle and Operculum. As already stated, there is but one attachment of the proso- branchiate to its shell ; namely, by means of the columellar muscle, by which the inner face of the columella is directly united with the posterior portion of the body of the animal. It passes underneath the mantle, greatly thickening the body wall, and terminates upon the inner face of the operculum, so that by its contractions the operculum and shell are approximated. The form of this muscle depends on that of the shell, and in the conical, non-spiral shells especially, varies greatly from its normal development. Thus, it is horseshoe-shaped in Capulus ; it is divided into two portions, one of which lies on either side of the anterior part of the animal, in Fissurella. In Haliotis the animal is coiled around it, and its insertion, instead of being on the columella, is on the middle of the inner wall of the shell itself. At the ending of the columellar muscle in the dorsum of the foot, its fibres are nearly vertical to the plane of the operculum, which usually appears to be immediately superimposed upon them : in Buccinum, however, Keferstein finds interposed a layer of long cylindrical epithelial cells, with mostly distinct nuclei, and long divided processes entering between the muscular fibres. The operculum, a cuticular development of these cells, is com- posed, as may be seen in the corneous opercula of Murex, Purpura, Triton, etc., of very thin superimposed layers. With the microscope one may perceive in a thin section, the cylindri- cal cells with their head attached to the lowermost layer ; or, on COLUMELLAR MUSCLE AND OPERCULUM. 13 the inner face, the small rounded pittings where they have been attached. Dr. J. E. Gray was the first investigator who announced that the operculum is homologous with the second valve of the lamelli- branchiates or bivalve mollusks. He has shown that the oper- culum is developed on the embryo long before it is hatched ; that it is placed on and covers a particular part of the body called the lobus operculigerus, and which bears to it the same relation which the mantle does to the shell, and that its growth occurs in the same manner ; that this growth is made by the ad- dition of new matter to the inner surface and especially near the margin ; that it is attached to the animal by means of one or more muscles, which, as in the bivalve shell, pass from the larger valve or shell to the smaller one or operculum ; that the operculum, as it increases in size, is gradually moved on the end of its muscle — the many-whorled operculum of the Trochi re- volves as many times on the end of the muscle as the many- whorled spiral shell turns on its imaginary axis ; that the opercu- lum is often lined internally with a shelly coat like a shell, and sometimes, like the Cowries, its outer surface is covered also with a shelly deposit by a special development of the opercular lobe. The principal difference between the operculum and the valve or shell of the Gasteropods consists :— 1. In the operculum having no cavity, its cone being de- pressed, flat or even concave, or very much compressed, form- ing only a spiral riband, as in the spiral operculum. But this absence of a cavity is a difference only of degree, for the valves of some Gasteropods, as Umbrella, Patella, etc., are much flat- tened ; the first resembling the annular operculum of Ampul- laria and Paludina : but the greatest resemblance is to be ob- served in the small, flat valves of Gryphaea, Exogyra, Chama, and other genera of bivalve shells which are attached by one of their valves. These valves are often quite as flat and destitute of any cavity as the operculum of any Gasteropod ; and it is to be remarked that these valves exactly resemble a spiral opercu- lum in shape, the remains of the ligament forming a spiral mark on the outer surface, showing how the valve has rotated on the body of the animal as the operculum rotates on the foot of the Gasteropods. 14 COLUMELLAR MUSCLE AND OPERCULUM 2. The operculum is generally horny and formed of a sub- stance similar to the epidermis of shells ; but then some shells, like the Bulla, Aplysia, certain Uniones, etc., are entirely or almost destitute of calcareous matter, and some of the helices, when inhabiting granitic regions, are equally of epidermal sub- stance : on the other hand many opercula are thickened inter- nally with a calcareous deposit. Dr. Gray proceeds to show that in bivalve shells like Chama, where one valve (the attached one) has a spiral apex, whilst the other valve, is a flattened spire, the position of the hinge with reference to the spire must rotate slowly with growth, as in the spiral operculum in its growth. The direction of the spire of a spiral operculum is opposite that of the shell, showing another analogy. The conclusion arrived at by Dr. Gray is that the normal or typical form of mollusks is that protected by two valves or shells : indeed, some nudibranchiate gasteropod mollusks which have no shell in their adult state (Doridse), have their newly hatched young covered with two shelly valves which afterwards fall off.* Dr. Gray has always maintained that the opercula are of great value in the distinction of genera, and he does not fail to condemn severely the practice of preserving shells in museums, or of figuring and describing them in conchological works, with- out opercula. I have already alluded to the supposed opercular bodies found with the fossil Ammonites : f it ma}' be added that they occur as well in some of the Heteropods and Fteropods. In 1847, Loven proposed to consider the operculum as analo- gous to the byssus ; but Dr. Gray has pointed out that some genera of Gasteropods provided with an operculum, secrete a byssus also (Rissoa, Cerithium, Littorina, etc.). However, Prof. Huxley, one of the latest and best authorities, thus endorses Loven's views : " On the haemal aspect of the posterior portion of the foot, a chitinous or shelly plate, termed the operculum, may be de- veloped. This operculum appears to be the analogue, if not the * Ann. Mag. Nat. Hist., 2 ser., v, 476, 1850. f MAN. OP CONCH., vol. i, Cephalopoda, 267. COLUMELLAR MUSCLE AND OPERCULUM. 15 homologue, of tlie byssus of the lainellibranchs ; and is certainly not homologous with either of the valves of the shell of the latter, which are pallial structures." * The following interesting note by Dr. Gray, will throw some light on this very interesting subject : " Oti the reproduction of the lost part of an operculum., and its probable restoration ivhen entirely destroyed. " It is to be expected that the operculum of a gasteropodous rnollusk may be sometimes broken or injured, but I have never hitherto been able to find any very distinct example of the kind, so as to study how the repair of the lost part would be effected. That such an occurrence would most probabty be rare, is easily explained from its situation, as the operculum is protected by the last whorl of the spire of the shell when the animal is expanded, and by the mouth when it is contracted into the cavity of the shell. " I have lately met with a very distinct example in a specimen of Fusus in the British Museum collection. In this specimen the apical half of the operculum has been broken off (pi. 2, fig. 12 a), and the lost part has been renewed by an irregular round- ish process, nearly of the size of the lost part, not quite as thick as the original portion, and formed of rather irregular horny plates ; the smaller or first-formed portion being in the centre of the broken line, so that the restored part bears some similarity to the annular operculum of a Paludina. This restoration is exactly like that which would have taken place in a shell under similar circumstances, and is a further proof of the truth of the theory which I have long advocated, that the operculum is a rudimentary valve, and is homologous to the second valve of the bivalve mollusks. " In examining two specimens of Pleurotoma babylonica, pre- served in spirits, with the opercula attached, I was much sur- prised to observe that the opercula of the two specimens were exceedingly different in structure and belonged to two distinct modifications of that valve, one (pi. 2, fig. 12 b) being sub- annular, with the nucleus apical, like the other species of the genus, and the other (pi. 2, fig. 12 c) annular, with the nucleus Huxley, Anat. of Invert., 487, 1877. 16 COLUMELLAR MUSCLE AND OPERCULUM. sub-central, somewhat like the operculum of Paludina. Tlie examination of the restoration of the lost half of the operculum of the Fusus before referred to has solved the difficulty, and I have no doubt that one of these animals had by some accident lost its operculum, and that it had gradually restored it; com- mencing, as in the case of the restored part of the operculum of the Fusus, by a small nucleus in the centre of the opercular mantle, on the back of the foot, and gradually adding new Ia3^ers around the edge of it, until it formed an annular operculum nearly of the size of the original, but differing from it in shape, being less acute in front and nearly similar in form at the two ends. A more minute examination has strengthened this theory, for the operculum of this specimen is less regularly developed than is usual in the annular operculum of the kind, and is much thinner than the normal operculum of the genus, as is the case in both these particulars with the restored part of the operculum of the Fusus. u This change in the formation of the operculum when it is reproduced is just what might have been expected. The animal, when it has to form the operculum at its birth, begins its forma- tion at the tip, and increases its size, as the animal requires a larger operculum for its protection, by the addition of new layers to the outer edge of its larger and last-formed end ; but when it has to reproduce this organ, the opercular mantle having reached a certain size, it proceeds to cover its surface with a new pro- tection in the most easy and rapid manner, and, commencing from a more or less central spot on the surface, enlarges the surface covered by adding new matter to the entire circumfer- ence of the first-formed part ; it continues this process without waiting to making the operculum as thick and solid as the one which was lost, until it readies the size of the original, moulding itself on the opercular mantle, and adapting its form to the form of the throat of the aperture of the shell which it has to close. The change of form in the front of the restored and mended operculum is caused by the parts being moulded on the existing opercular mantle — consequently they have not the narrow front part which is found in the normal form caused by that part having been formed when the animal had this part of a small size; and as it increases in size the whole opercular mantle THE SHELL. 17 moves forward, leaving the small tip of the operculum free, and useless to the animal, and, therefore, not necessary to be repro- duced when the operculum is reformed in the adult age of the animal."* The Shell. All prosobranchiates (and nearly all mollusks) are provided with an external shell, a dwelling place and a citadel combined, the hardness and durability of which, as Keferstein remarks, " supplies us with the best means of knowing the animal ; in- deed, in many cases, it is the only part known, and was formally the only part valued and preserved in collections. Although the animal itself offers more weighty and striking characters for the separation of the higher groups, yet having learned the close relationship existing between shell and animal, we find therein ample justification for attaching especial importance to the shell in a systematic point of view." We have already shown how the shell is produced by the mantle. The form of the shell is throughout regular, and is normally a cone curved into a spiral, and descending in a screw-like manner from the apex or initial whorl to the aperture. Nothing can be more beautiful than the regular geometrical progression of the growth of a shell or the certainty with which each species and genus grows in its normal pattern, although these modes vary among themselves so widely : thus we have the simple depressed cone of the Patella, all aperture and no spire, and from it every gradation from the Haliotis, almost equally depressed and broad, the result, however, of a very rapidly enlarging spiral, to the long, many-whorled Turritella, or the Vermetus, which is a Turitella partially unrolled into a simple long tube : — the opposite of the Patella. The whorls of a spiral shell are, in most cases, closely wound around its axis, and, therefore, most part of their surface is in contact, each whorl partially covered and concealed by its successor ; and where the axis does not lengthen by the obliquity of the spiral, we find, as in the cone and Cypraea that the shell only shows externally its last whorl, with, perhaps, a * Dr. J. E. Gray, Proc. Zool Soc., 100, 1854. 3 18 THE SHELL. very small portion of its predecessor visible on the spire. On the other hand there are genera in which the whorls are not at all in contact, and where the axis becomes itself an imaginative cone, widest at the base. Besides the almost numberless modifi- cations of form resulting from the degree of obliquity and close- ness of the spiral, the direction of the latter may be mentioned as another factor in producing modification. In most spiral shells the spire normally curves to the right, that is to say, plac- ing the shell with its apex turned from the observer and its aperture in view, the latter will be found on the right hand. In others the volutions proceed in the opposite direction with such regularity as to be eminently characteristic of some species and genera. However, in certain genera, it is found that species normally dextral will exceptionally produce sinistrally coiled shells, and this abnormal growth probably is caused by disturb- ance of the relations of the embryo with its initial shell. Whilst the bulk and weight of shells are composed principally of carbonate of lime, yet they have always an organic basis, which is first developed, and then gradually impregnated with the lime. If the latter be removed by the use of acids the organic residuum (conchy olin) still retains the shape of the shell, forming a sort of membranous framework. It is this organic basis which maintains the life of the shell, for, the animal re- moved, as in beach-worn or fossil species, the conchyolin soon disappears and the shell becomes pure carbonate of lime, growing at the same time more and more brittle. Many of the long spiral shells, such as Bulimus decollatus, some Pupas, Truncatella, Melania, etc., withdraw the body from the earlier whorls in the course of growth, and partition off the unoccupied space with a shelly plate : in such cases the unoccupied whorls become brit- tle and are soon broken off. We must, therefore, believe that the shell is vitalized, or rather, that its vitality is maintained by simple contact of its organic basis with the living animal. There are two very distinct types of shell structure, the cel- lular or porcellanous and the membranous or nacreous. In bivalve mollusks the former is the outer and the latter the inner layer, in most cases, but in univalves the shell is usually porcel- lanous only ; although a few of the holostomata, such as Turbo, THE SHELL. 19 Troclms, Stomatia, etc., have both layers, the inner one pearly. Nacreous shell substance consists of very thin, superimposed membranes, lying parallel with the surface and impregnated with aragonite. The iridescence of this substance proceeds from the waved or wrinkled close edges of these membranes. The aragonite is of crystalline structure, showing beautiful five or six angled tables. (PI. 2, figs. 8, 9.) As porcellanous shell structure is essentially the same in almost all Gasteropods, and the only kind of shell structure in most of them, a few sections of Strombus gigas will give a clear idea of the disposition of the material. (PL 2, figs. 4-6.) There are, it will be perceived, three layers of deposition (of which the middle one is thickest in this case), each composed of a multitude of plates or prisms, but each differing in the direc- tion of arrangement of these. It will be readily perceived how much this diversity of arrangement adds to the toughness of the shell structure, as no line of fracture can penetrate the entire shell wail, except by the violent breaking across of part of these layers of prisms. As to the chemical composition of shells, the conchyolin or organic material is a small, varying percentage, carbonate of lime, existing in quantities varying, from Turritella, 88.70 per cent, to Strombus gigas, 99 per cent. There are traces of other constituents, of which carbonate of magnesia is perhaps one of the most important ; it varies from 0.12 per cent, in Teles- copiuni to 0.48 per cent, in Fusus antiquus. Silicic acid has also been detected. C. Schmidt has obtained almost 1 per cent, of phosphate of lime from the shell of Helix nemoralis. The calcareous operculum of Turbo (analyzed by Wicke) con- tains : Carbonate of lime, 98.72 per cent. ; organic material 1.28 per cent. Shells are perceptibly harder than, and will scratch calc-spar. Their specific gravity is somewhat higher than that of Carrara marble, being about 2.15 to 2.85 for the prosobranchiates, and varying but slightly in the pulmonates. With reference to the coloring of shells (terrestrial as well as marine species) Fischer points out that as a rule, brilliancy of coloring increases towards the equator. He suggests three 20 THE SHELL. zones of coloration corresponding to the thermal zones, which may be designated as : 1. Monochromic or frigid zone. 2. Oligochromic or temperate zone. 3. Polychroniic or tropical zone. Of course exceptions are numerous — modifications based upon habits, ocean currents, adaptive coloration, etc.. M. Fischer particularly cites the melanism which characterizes so many shells of the West Coast of North and South America, giving, among other examples, the numerous species of sad colored and lugubriously named Trochi, which pervade those regions.* On adaptive coloration. — A number of papers on mimicry or adaptive resemblance as a means of protection of animals against their enemies have been published, and the subject has awakened much interest and discussion. Coincidence of colora- tion of molluscous animals with their surroundings has been frequently remarked, but it is probably not so often for purposes of attack or security as the result of the food upon which the^y live or a dye obtained by contact with it. I have shown in my volume on the Cephalopoda that conscious adaptation of color has been noticed in the Octopus. Mr. Morse gives a number of instances of coincidence of color between American marine mol- lusks and their surroundings which he supposes to be evidences of protective adaptation.f " The thickness, the roughness, and the smoothness of the sur- faces of shells appear to depend, in a great measure, on the still- ness or agitated state of the water in which they reside. Shells which have branching or expanded varices. like the Murices, are also much influenced by circumstances, and hence many mere varieties, arising from local causes, have been considered as dis- tinct species. Thus Murex anguliferus is merely a Murex ramo- sus with simple varices ; and Murex erinaceus, M. torosus, M. subcarinatus, M. cinguliferus, M. tarentinus, and M. polygonus are all varieties of one species. Murex magellanicus, when found * Dr. P. Fischer, Jour, de Conch., xxiii, IQ.% 1875. f E. S. Morse, Proc. Bost. Soc. Nat. Hut., xiv, 141, 1871. SHELL. 21 in smooth water, is covered with large acute foliaceous expan- sions ; but the same shell living in rough seas is without any such expansions, and only cancellately ribbed. In such situa- tions it seldom grows to a large size ; but when it does so, it becomes very solid, and loses almost all appearance of cancella- tion. Triton maculosus is very widely spread over the ocean in different temperatures and different kind of seas ; it consequently offers a multitude of varieties both of size and surface, all gradu- ally passing into each other, and most probably produced by the operation of the foregoing causes. Indeed, a vast number of merely nominal species have been formed from the habit, too prevalent among conchologists, of describing from single speci- mens, or even from several individuals brought from the same locality, which would never have been considered as distinct had collectors kept in their cabinets a series of specimens found under different circumstances, and studied, on the coasts where they are found, the variations which shells undergo."* Shell. Rest periods in the growth of mollusks are sometimes, as in the Murices, marked by a thickening of the edge of the shell, caused by continuous depositions of shell material, forming a ridge or varix ; and the rate of growth may be thus traced readily, in numerous mollusks. Thus in Murex one group shows three varices upon each whorl, indicating that a period of three years, or at least three seasons of activity is required for the completion of a single whorl of growth. In another group of Murices the varices are more numerous, as many as four to ten being counted on a whorl. In Triton, the varices are two on each whorl, but nearly alternately situated, so that the varices of each whorl oc- cupy an intermediate position to those on the preceding whorl : in Ranella there are also two varices but they form a continuous fringe or wing-like expansion on each side of the spire, showing a very regular growth by periods of half-whorls. The accretion of surface during growth is not continuous but is made by minute layers, around the margin of the aperture, each extending a little beyond its predecessor, and the edges of Dr. J. E. Gray. PMlos. Trans., 771, 1833. 22 SHELL. these layers as exposed on the external surface are called growth- lines. Those shells which have a simple, or sharp-lipped aper- ture and which do not develop varices, nevertheless distinctly show the rest-periods by the greater impression of the growth- lines. Many shells retain the sharp aperture for a variable period, which may be called their juvenescence, but finally acquire adult characters, consisting of a thickened, reflected, inflected or lipped aperture — which is sometimes more or less contracted by inflected calcareous projections called teeth. Growth, however, frequently continues after adult characters have been acquired, and then these are absorbed away when accretion recommences, leaving the mark of their former position in a more prominent growth- scar or line. The power of dissolving their shells is possessed by certainly a large portion of the mollusca and is habitually exercised by many of them : thus the cone, which we have seen partitioning off its whorls against an enemy and thus seriously incommoding itself for room, would under normal circumstances acquire for itself additional accommodation by absorbing away most of the thickness of the enclosed whorls or partitions, and Cypraea, Nerita and Auricula assist themselves in the same manner so as to become eventually an external shell only, with a single cavity. Tn species with lengthened spire, this method is not so practi- cable, because only a small portion of the whorls are enclosed within succeeding ones. The growth of many of these animals is such that they cease after awhile to occupy some of the earlier whorls? and they then partition them off as a regular habit, in the same manner that the cone has worked under the pressure of excep- tional necessity. These partitions made, the portion thus cut off' from contact with the animal loses vitality, becomes brittle and breaks off, forming the truncated shells which are characteristic of numerous groups — especially of land shells. From what has been said of the mode of formation of shells it will be readily seen that details of sculpture as striae, sulcations, ribs, nodes, spines, etc., result from similar ornamentation of the applied mantle : thus the spine of a Murex, if closety examined, will be found to have a longitudinal seam upon its front face, showing that it has formed by a corresponding digitation of the mantle. It is scarcely necessary to pursue this subject DIGESTIVE ORGANS. 23 further at this time : reference to the plates of this and succeed- ing volumes will demonstrate this relationship of shell and animal. Neither shall we investigate further the forms of shells or their opercula, monstrosities, etc. All these matters, where general in scope, may well be relegated to an introductory treat- ise on Conchology ; or we may at some future time include them with other particulars of more general than special application in a volume of appendix to this series.* Digestive Organs. The digestive organs in the Prosobranchiates are well devel- oped. The mouth, which is sometimes in the lower plane of the head, and sometimes at the end of a proboscis capable of protru- sion and retraction, is frequently encircled by an extensible lip ; within, it is often armed with a jaw on either side, and the tongue is usually armed on its upper surface with numerous transverse rows of teeth, constituting the lingual ribbon. The oesophagus is often beset with appendages and salivary glands, and leads to the stomach; whence the intestine turns forward, passing close to the kidney and heart and into the respiratory cavity, the right side of which it traverses and finally empties into the anus. The intestine and often a portion of the stomach is embraced by an * Such a volume might embrace the facts of geological and geographi- cal distribution, in addition to the outlines of molluscan structure and a history of classification. One can readily conceive that such a volume would be a natural outgrowth and completion of a monographic series, a collation of the information contained in its predecessors, and an appli- cation of the same to the various important generalities which occupy modern scientific thought. It is a reproach to natural science, and to 110 department thereof more than to conch ology, that most of its votaries consider the determination of species and genera its legitimate end ; that they are more actuated by the selfish ambition of acquiring reputation than by the love of knowledge. Thus it happens that in most treatises very few structural details are given, whilst the technical descriptions of external features occupy the bulk of the work— such descriptions being the necessary justification for the imposition of generic or specific names and the consequent glorification of the namer. I propose, as far as practi- cable, to reverse this procedure, to consider the necessarily arbitrary and artificial nomenclature simply preliminary ; as a facility towards the ac- quisition of knowledge of nature and its laws— not as the end of knowl- edge. As the builder finds it convenient to express the kinds of instru- ments used in his labor, by technical names, so do naturalists find necessary a succinct designation of the subjects of their studies ; and the 24 DIGESTIVE OEGANS. enormous liver, filling nearly the whole of the first whorl of the shell, and pouring its secretions into the former (and often into the latter also) by several openings. We will successively ex- amine these various organs more in detail. The proboscis is a production of the skin of the anterior or head portion of the body, bearing the mouth at its end. When it remains permanently protracted it receives the name of ros- trum, that of proboscis being more properly limited to this organ when provided with muscles by which it can be retracted within the body. The typical proboscis is quite characteristic of the siphonostomated Prosobranchiates, or those carnivorous mol- lusks of which the shell is canaliculate or notched at its lower extremity; whilst those animals provided with a rostrum or snout, or with a simple mouth are members of the usually phytophagous holostomata In Dolium, a remarkable exception, the exceed- ingly long proboscis accompanies phytophagous habits. The invagination of the proboscis is effected by means of pow- erful retractor muscles supplied along its entire length and espe- cially numerous at its base, where the retraction begins. The protrusion of the proboscis, on the contrary, is effected by press- ing forward the blood towards the head, an operation assisted by the contraction of the annular muscles of the fore-part of the body. In Natica, according to Troschel, the invagination of the naming and technical description of species, in the same manner dis- tinguishes for us the implements which we should use in our investigation of nature— implements by which the Great Builder has worked, in which He has expressed His thought. The proper acquisition of a language requires the preliminary knowledge of its grammar, the knowledge of letters precedes reading : even so, the simple name of a species, then of a genus, and its recognition when met with or referred to forms the mere alphabet of science, from which we proceed gradually to the consideration of individual properties, then to intimate and to wider relationships, until we are fortified with sufficient knowledge to generalize. In these latter days generalizations are numerous enough also, but unfortunately they are usually the product of minds not furnished with the requisite intimate knowledge of the factors with which they build their theories. The details of anatomy and physiology of the mollusca will be given for each group as it occurs in the course of this work, and I shall thus neces- sarily repeat myself frequently ; and I hope that by the time that my reader and I have finished the volumes containing the technical descrip- tions, we will both be better prepared than we are at present to under- stand the subject as a whole. This course has also the important personal advantage, that if at any time I shall be attacked for sins of commission or omission, I may make the obvious defenses—wait and see. DIGESTIVE ORGANS. 25 proboscis commences at its extremity (like that of the tentacles of snails) by means of two retractor mnscles attached to the oral mass. ' The same observer describes a muscular disk on the under side of the proboscis, behind the mouth, in Natica and Sigaretus. This disk possesses suctorial action and probably enables the mollusk to attach itself firmly to the shells of other species when drilling them for the purpose of devouring the soft parts. The oral mass is usually an oval body formed by invagination of the external skin through the mouth, from the upper posterior end of which proceeds the 4 3 00 Devon. ,0 08 0 £ £ o 1 g c« — S3 1-5 1 Q Tertiary. Bj 3 o H Recent species. Total Species,i fossil & recent. CHITON 18 1' 01 QAA PATELLA g 10 1 1 A 10 00 ol Oft 400 1 no 231 FISSURELLA EMARGINULA. . . . CAPULUS 1 2 10 -j 4 5 7 23 23 1 9 yo 30 34 1UU 84 26 198 114 60 PlLEOPSIS 2 9 A 93 24 Af\ 31 CREPIDULA CALTPTR^EA SlGARETUS Q 16 11 1 9 4U 16 11 40 50 (\0 4U 56 61 NATICA OH oq K({ ftP. 10 /CO 1 fin 41 NERITA .... Q Q 9 OA ^oU 1UU 330 NERITINA 1 •i OA O^ f\4 1/cU 1 Afl 172 \VELLANA . . 1 ^ o4 1 Q IvMJ NERIN^A Kft 4fi lo Q9 lo TURBONILLA LOXONEMA 2 9 20 1 1" 22 y« 32 oq 92 32 MACROCHEILUS. . SCALARIA 1 14 1 1 1 I 1Q ftO 17 mo -1 AA 60 17 9ftA. TURRITELLA PHASIANELLA .... 4 I 29 5 2 4 0 1 6 1 17 2 71 9 Q 107 11 1 ^ 296 29 qi 30 22 fin £00 326 51 01 TURBO 18 32 I 0 Q Kf\ 5R Kiy 9fil IJK »i qQO DELPHINULA. . . . EUOMPBALUS SOLARIUM ROTELLA 28 2 60 4 7 2 2 4 34 1 36 65 4 55 90 102 1 K 30 25 10 ooy 85 90 127 f)K PHORUS 14. 1 n ^0 TROCHUS K 01 i 1 (\C Ptl 1 78 q/»9 1 (\n ^4 MURCHISONIA SCHIZOSTOMA 18 4 30 13 1 48 IDU OSsJj 48 PLEUROTOMARIA . CIRRUS 28 128 2 1 5 1 41 7 63 ^ 2 310 2 1 4 312 1 /i CERITHIUM . \ 1 Q« O97 qf»7 QA 14 Vfrry ROSTELLARIA PTEROCERAS STROMBUS 4 14 11 60 17 5 16 Q1 94 27 Of* oU 6 10 7ft 40 < 100 37 1 ftA MUREX K 13 17Q 187 91 ft qQ7 Fusus \ I 7 53 9QO 0^7 1 ftft A.r\<7 PTRULA I 2 17 3fi Ka A(\ QA PLEUROTOMA FASCIOLARIA 1 6 2 344 32 Qfj 350 34 °,fi 370 15 qer 720 49 71 7 3 I 15 5 142 173 100 970 TEREBRA 9 on 07 1 1 ft 1 J.7 13 93 106 70 17fi 2 110 112 Q^O 4«9 32 32 190 1R9 CYPR.3SA 3 7Q 09 1 fifl 94.9 CONUS 3 89 92 270 362 Total . . 127 473 15 17 S3 367 701 2783 451 ft ^500 ftOlfi CLASSIFICATION. 57 Classification. The following rapid sketch of the history of the modern clas- sification of mollusks and exposition of the system, based prin- cipally on the lingual armature of the gasteropods, is translated and condensed from a paper by the late Prof. Morch, published in the Journal de Concliyliologie (xv, 232, 1867). Ancient authors classified shells according to external forms, from which they derived generic names. Linnaeus was the first to introduce characters independent of the form of the shell : as the teeth and ligament in bivalves, plications and sulcations in univalves. By these characters Voluta and Turbinella were separated from Murex, Buccinum, etc. Linnaeus classed the species of each genus, according to the height of the spire, in analogous sections, of which the most were adopted by Bruguiere as distinct genera. Thus the follow- ing genera were terminated by a section " turrita," Bulla by Achatina ; Buccinum by Terebra ; Strombus by Potamides and Pirena ; Murex by the spiny Cerithiae ; Trochus by Telesco- pium and Pyramidella ; Turbo by Turritella ; Helix by Melania and Lymnaea. Linnaeus was the first to take the form of the animal into con- sideration as a generic character ; but he indicated only five different types of animals, namely : Doris, Limax, Tethys. Sepia, and Ascidia. Thus the animal of Chiton is a Doris, that of Argonauta a Sepia : bivalves with simple mantle are Ascidiae (Solen, M}*a, Pholas) and those with fringed mantle Tethys. Nearly all the univalves are called Limax. Adanson must be regarded as the founder of Malacology, but the number of mollusca known in his time was too few to per- mit the elimination of the principal systematic divisions. He was also the first to take into consideration the operculum and the shell structure as characters, and to divide the bivalves ac- cording to the number of muscular impressions. The system of Cuvier, based on the respiratory organs, in- duced a great reform in Conchology. The shells of pulmonate mollusks, heretofore dispersed by all authors, with the excep- tion of Adanson, among the pectinibranchs, were assembled in one group, which still remains intact. Although it may be diffi- 8 58 CLASSIFICATION. cult to indicate by a description the difference which exists between the shell of a pulmonate and that of a branchiate mol- lusk, there are, nevertheless, few collectors who will not recog- nize it at sight. Ancient authors, like Lister, Miiller, Chemnitz and Schrb'ter, who have treated upon the terrestrial and fluviatile mollusks specially, have rarely mistaken these shells ; and a mistake of this nature is very rare among modern authors, although a few instances might be cited. One can say only that the shells of terrestrial pulmonates are inoperculate, with entire apertures (holostomate), never nacreous, rarely spirally striated ; but one cannot give a single character expressible by words, not- withstanding that all who have seen a certain number of species can distinguish them with facility. The fluviatile mollusks, nearly always unicolored, although they may resemble marine groups as to form, can also be readily separated at a glance. Ferrussac and several modern authors have thought that all mollusks inhabiting dry land respire by the aid of a pulmonary sack, but nothing is less certain. Among branchiferous genera, the Littorinas and many trophical Neritimv, live a long while out ovf water. Tbe larvae of Auricula swim in the sea, and con- sequently possess a branchial respiration during this period of their life. According to Semper, Ampullaria has an accessory pulmonary sack. If the inoperculate pulmonates are consid- ered, with so much reason, as an incontestible group, it is be- cause, apart from their pulmonary sack, they possess other collateral characters of equal importance, as for example, the position of the eyes, the organization of the mouth and of the sexual organs. The rest of the gasteropods. after the exclusion of the pulmo- nates, were divided by Cuvier into several very natural groups, according to the form and position of the branchiae (nudibranchs, heteropods, tectibranchs, scutibranchs, cyclobranchs). The magnificent work of Quoy and Gaimard is full of precious ma- terial for the amelioration of the great group of pectinibranclii- ates. The Trochidse are here shown to be inseparable by their characters, as much external as internal, from Haliotis and other scutibranchs, notwithstanding the presence of an operculum and an elevated spire in the former. At the same time the great systematic value of the nacre was proved. Stomatella, with an CLASSIFICATION. 59 animal similar to the Trochidse, has a nacreous interior, whilst Sigaretus, with a non-nacreous shell, has an animal like Natica. The relations between the enamel of the shells of Cypraea, Oliva and Natica, and the structure of the animal were shown for the first time by the same work. As it has become evident to me that the presence of an operculum and the height of the spire, considered heretofore as prime characters, have, in reality, little value in distinguishing the families, I have sought to divine the natural affinities of mollusks according to the sculpture and structure of their shells. In 1847 Loven published four plates of lingual dentition, re- presenting 94 species of cephalous mollusks. The first glance at these plates suffices to show clearly that the lingual armature confirms the most of the ancient divisions. Thus, the Cephalo- pods, Pteropods, Heteropods, Scutibranchs (in the sense of Quoy and Gaimard. including the Trochidse), are also distinguished by the teeth. The conchological analogy between Pleurotoma and Conus had already been shown by Sowerby. There existed, nevertheless, certain anomalies until then inexplicable, on ac- count of the small number of observations made, as for example, the analogy between Philine and Scaphander and the Gymno- branchs. The want of teeth must not be considered as fatal to the systematic value of characters found in these organs. The teeth accepted as an exclusive character have, doubtless, incon- veniences, as in the whole animal kingdom, but it cannot be denied that all other organs taken as exclusive characters offer still greater inconveniences. Thus the shell may be wanting in very similar animals (for example, Notarchus, Aplysia ; Ptero- trachea, Cardiopoda ; Limax, Tebennophorus). The operculum is often wanting in the adult, although the young may have it. There are even operculated and inoperculated species in the same genus, as understood by many modern authors (for exam- ple, Pleurotoma and Bela, Oliva and Olivella, Yetus, Yoluta and Lyria, Spirialis and Limacina, Proserpina and Helicina). The organs of respiration and locomotion may be entirely want- ing in closely related species (Firoloidea and Phyllirhoe). Loven has characterized the families according to the teeth? and has given Latin diagnoses. In 1848 Troschel* mentions the * Handbook of Zoology, 60 CLASSIFICATION. teeth as characters of all his sub-orders, and introduces into the nomenclature, for the first time, several new names taken from the form of the teeth. Thus the section H. of Loven is called Rhipidoglossa, including the Scutibranchs, that is to say. Neritidae, Trochidse, Haliotidse and Fissurellidrc. For the sec- tion L. of Loven he proposed the term Toxoglossa. Tsenioglossa corresponds to the Ctenobranchous Gasteropods, having seven rows of teeth (3. 1. 3.), excepting the operculate pulmonates, although they may have the same form of teeth. Thus Cyclos- toma is placed in another sub-order from Valvata and Paludina. The Heteropods, which have the same general disposition of the teeth as Tsenioglossa, are regarded as an order, with the same value as the Gasteropods — an opinion still maintained by this author. In 1853, Dr. Gray, in adopting the names of Prof. Troschel, proposed several new groups, according to the form of the teeth.* 1. Hamiglossa. Three ranges of teeth (1. 1.1.), the lateral versatile. This last character appears to Prof. Morch to be con- sequent upon the rupture of the tissues ; it is observed above all when there is abundance of water under the compressor. 2. Odontoglossa. Including only Fasciolaria, Mitra and Tur- binella, which have the same form of teeth, but of which the laterals are not versatile. 3. Rachiglossa. A single row of teeth (0. 1. 0.) ; the laterals having disappeared. 4. Dactyloglossa. Only differing from Taenioglossa by their lateral teeth, which are wider, with very profound comb-like incisions. 5. Ptenoglossa. Teeth nearly subulate, in numerous longi- itudinal rows; Scalaria, Acteon. B. Gymnoglossa. No teeth : Arehiteetonidae, Acusidae, Can- cellariadae, Pyramidellidae. But teeth have been since discovered in the three first families. There are, doubtless, many genera indubitably deprived of teeeh. without, for that reason, forming separate groups. Gray has regarded these different groups as having a sys- tematic value inferior to that of the form of the proboscis. Thus Proc. Zool. Soc., 32, 1853. CLASSIFICATION. 61 he divides the Ctenobranchiates into two sub-orders : the Pro- boscidifera. which he believes zoophagous, furnished with an entirely retractile trunk, and the Rostrifera, having a contractile, but not retractile, trunk, and sometimes very long, as in Stru- thiolaria, which he supposes phytophagous. This author has thus placed the sections of Toxoglossa. Gym- noglossa, Ptenoglossa and Trenioglossa in these two sub-orders. The small value of the retractile trunk as an ordinal character is proved for example in the Bullidi? ( Bulla vexillum, possessing a very long retractile trunk). Odostomia has also a very long re- tractile trunk ; and the rather short trunk of Janthina is very often retracted into the head. In a more recent work,* Dr. Gray has reunited all the Toxo- glossre in a single division Toxifera, still retaining for the other divisions the separation into two widely removed sections. The name of Ctenoglossa is changed to Ptenoglossa ; the name Trapezodonta is proposed for the Coriocellae, the teeth of which do not appear to Prof. Morch to differ from the Tsenioglossae except in the want of the two internal teeth on each side (1. 0. 0. 1.0. 0. 1). Heteroglossa is proposed for the Cyclobranchiates. In 1854, Morch divided the cephalophora into five great groups, namely: I. Rhipidoglossata (including the Cyclo- branchs) ; II. Ptenoglossata ( Pulmonata, Tectibranchiata, Jan- thinidse) : III. Taenioglossata (including Pneumonopoma and Heteropoda) ; IV. Hamiglossata (Proboscidea of Troschel, Odontoglossa and Rachiglossa of Gray); Y. Toxoglossata (in- cluding Pleurotoma and Terebra). In 1857, Morch reduced the five principal divisions to three, namely: I. MUSIVOGLOSSATA (corresponding to the Ptenoglossata of 1856, but thus modified because this name has been used by M. Troschel to characterize the group of Janthimu and Scalariae), Pulmonata and Tectibranchia. II. ARTHIOGLOSSATA, including : 1. Tamioglossata, 2. Ancis- troglossata, 3. Toxoglossata. III. RHIPIDOGLOSSATA, with the section Orthodonta (Cyclo- branchiates). In 1861-2. circumstances having induced Prof. Morch to study the Planarians, he was struck with their great affinity Guide to Brit. Mus., 1857. 62 CLASSIFICATION. with the Pellibranchs, above all in the generative organs ; this caused him to make a comparative revision of the geni- tal organs of mollusks. He then ascertained that those be- longing to his first division were androgynous and furnished with a retractile male organ ; whilst those of the second section were dioecious, with a non-retractile male organ ; and those of the third section differed from the others by the want of a copu- lative organ. In other words, he had thus arrived, inde- pendently, at the three groups proposed in accordance with the sexual organs by Blainville and Labreille. In 1859 Morch perceived that Mollusks were divided into two great groups, according to the construction of the heart and that these groups accorded also with those furnished by the sexual organs. Thus the Phanerogama Latr, with a retractile or non- retractile copulative organ, have a heart with a single auricle (Monotocardia Morch), whilst the Agama Latr, which have no copulative organ, have a heart with two auricles (Diotocardia Morch). It appears, doubtless, rather strange that the acephala should form a group with a considerable portion of the gastero- pods (Rhipidoglossa and Heteroglossa). but there exists a simi- lar division among the vertebrates, namely : the cold-blooded vertebrates, where the fishes are united with reptiles, the latter provided with well-developed locomotive organs analogous to those of the mammalia. Stimpson proposed, a few years since,* to form a group Anan- dria, characterized by the want of a male copulative organ. This group includes the Melanians of North America, the Vermetidaj and Turritellidae and certain Cerithiai. M. Riippel. however, has figured a male organ in Yermetus inopertus, and M. Lacaze- Duthiers has found a single male individual which circumstances did not permit him to examine sufficiently. As to the Melanians, they may want an external conical male organ, but the sexual character is with them represented by a groove. In the Agama of Latreille there is not the least external sexual difference. Mr. Morch believes that naturalists of the most opposite schools could agree to a scheme of classification which he sub- mits, as follows : Am. Jour. Sci., 2 ser. 37, p. 47, 1864. CLASSIFICATION. 63 Sub-Kingdom 3,—MOLLUSCA. Supra-class 1. PHANEROGAM A Latr. (Monotocardia, Morch). Class 1. ANDROGYNA (Hermaphrodita, Latr.). Order 1. GEOPHILA, Fer. (Stylomraatophora, A. Schmidt). Phyllovora with jaw. Agnatha without jaw. Order 2. HYGROPHILA ; eyes at the interior base of the tentacles. Planorbis, Physa, Limnrea, Siphonaria, An- cylus, Auricula. Order 3. TECTIBRANCHIA (Pomatobranchia), Pyramidella (connecting with the preceding genus), Obeliscus, Odos- tomia, Chemnitzia, Actaeon, Bulla, Aplysia, Notarchus. Gasteropteron connecting with the Pteropods. Order 4. PTEROPODA. 1. Gymnosomata. Clione, Pneumodermon. 2. Ihecosomata. Clio, Hyalrea, Limacina, Heterofusus. Order 5. GYMNOBRANCHIA. 1. Pygobrancliia. Doris, etc. ; branchite near the anus. 2. Pleurognatha.* Pleurophyllidia. Dendronotus, Trito- nia, Bornella, ^Eolis, Glaucus, Phyllirhoe. Order 6. PELLIBRANCHIA. Tethys, Chioiwa, Herman, Ely- sia, Limapontia, Pelta. All these mollusks are placed at the head of the gasteropods by Cuvier. Messrs. Troschel and Gray arrange them between the Acephala and Patella, considering androgynism as probably a character of absolute inferiority ; but the Acephala have gen- erally separated sexes. Class II. DIOICA, Latr. (Exophallia, Morch). Order 1. T^NIOGLOSSATA, Troschel. This is the only division where the famity groups and their reciprocal relations do not appear to Morch to be perfectly clear. In any case, it appears incontestable that all mollusks having seven ranges of teeth form a characteristic group. Recently, Troschel has divided the Trenioglossata into three groups accord- ing to the trunk : * Dr. Mobius has recently shown that the Doridae have also lateral jaws, although very little developed. 64 CLASSIFICATION. 1. Trunk not retractile. 2. Trunk retractile only by the end. 3. Trunk retractile from the base. These differences appear to originate solely in the different length of the trunk. The old divisions Holostomata and Entos- tomata are not very faulty. Troschel's first section commences with the Pulmonates, but Morch considers very doubtful their having a true pulmonary sack closed by a contractile opening. Gray calls these respira- tory organs of the Cyclostomse " gills vascular, branched," and " gills indistinct in the form of series of vessels on the inner sur- face of the mantle." Sect. 1. Fam. 1. Aciculacea ; 2. Pomatiacea ; 8. Cyclotacea ; 4. Cyclostomacea. Sect. 2. Respiring by branchiae and lungs ; Ampullariacea. Sect. 3. Branchiferous Holostomata. Fam. 1. Valvatae ; 2. Hydrobiye (Lithoglyphus^ ; 3. Littorinidae ; 4. Rissoidse ; 5. Pa- ludinidae ; 6. Meianiidse ; 7. Potamidae ; 8. Cerithiidae (Planaxis) The Aporrhaidse form a passage between the Cerithiidae and Strombidee. Crepidula and Capulus belong with Hipponyx in a group, notwithstanding differences in the labial palpi. Onus- tidse connects the Crepidalidfe with the Heteropods. Ovulidae (including Pedicularia). generally placed close to Cy- praea, is strongly distinguished from the latter by its non-retrac- tile proboscis. Notwithstanding this character, Morch considers it intermediate between Cassis and Cyprsea. The 3d section of Troschel (trunk retractile from the base) contains the genera which Morch united in 1852 in the family Tritonidse, namely: Cassis, Dolium, Pyrula, Triton, Ranella. Onchidiopsis, Yelutina, Marsenia, Tylodina, form, probably, a very natural division, notwithstanding the want of the two lateral teeth in Marsenia. The family Naticidae stands next in relation- ship. It is in the Taenioglossata that the greatest incertitude relative to a natural grouping of the families exists ; in the orders which follow, these relationships are more positively defined. Order 2. RHACHIGLOSSATA, Troschel. Never more than three rows of teeth. All mollusks having coriaceous ovisacs, heretofore known, belong to this order. UNIVERSITY OF \ CALIFORNIA. 1 Sect. 1. Marginella, Voluta, Volutilithes, connecting with Cryptochorda and Harpa ; Oliva, Ancillaria, Bullia, Nassa, Mitra, Columbella. Sect. 2. Buccinum, Fusus, Fasciolaria, Turbinella, Purpura, Murex, Magilus. Order 3. TOXOGLOSSATA. Two rows of teeth. Stimpson has recently discovered a median tooth in Clionella sinuata, Born. Conus, (Borsonia?), Pleurotoma, Clionella, Terebra, Can- cellaria, Halia, Lachesis ? Supra Class II. AGAMA, Latr. (Diotocardia, Morch). Mollusks without copulative organ. Heart with two auricles, placed nearly always around the intestine. Class 1. EXOCEPHALA, Latr. (Psendophallia, Morch). Order 1. RHIPIDOGLOSSA, Troschel. Proserpina, Helicina, Hydrocena, with sessile eyes. Gray, in figuring the teeth of Proserpina (Ceres), formed for this mollusk a distinct suborder, alongside of the Neritinse, which he called Pseudobranchia. In the same year (1857) Morch placed Helicinae in the Rhipidoglossates, with Neritina, notwith- standing the want of an operculum in Proserpina, an example followed by Troschel with some hesitation. Eyes sessile. Shell not nacreous. Neritina, Nerita. Shell nacreous. Phasianella, Turbo, Trochus, Margarita, Stomatia, Haliotis. Eyes not petiolate. Shell not nacreous. ? Scissurella, Emar- ginula, Fissurella, the affinity of which with Haliotis, is incon- testable. Order 2. HETEROGLOSSATA, Gray. (Orthodonta, Morch, Docoglossa, Troschel.) Patella, Tectura, Pilidium, etc., Chiton. Class 2. ACEPHALA, Cuvier. (Endocephala, Latreille ; Dithyra, Anst.) Dimyaria (Plagymiona, Latr.). Heteromyaria (Mytilacea). Monomyaria (Mesomyona, Latr.). Notwithstanding that the systematic value given above to these different groups varies from that of other authors, their order of succession differs but little from Cuvier. The separa- 66 CLASSIFICATION. tion of the Rhipidoglossata (Pectinibranchiates) and Scutibran- chiates, is the principal change. The arrangements of Gray and Troschel differ still more, above all in the Androgyna, which are placed between the Heteroglossa and Acephala, probably because androgynism is considered as a character of inferiority ; not- withstanding that the Acephala, which are inferior, have the sexes separate, with some exceptions. The little division, Pteno- glossa, including only Janthina, Scalaria, Solarium, is placed, in the system of Troschel, between the Rhachiglossa and Rhipido- glossa. If we admit a special concordance between the teeth of these three groups, Morch still does not consider the difference sufficient to justify a separation from the Androgyna. Janthina appears to him more close to the Pteropods by its lateral wings, and Scalaria, notwithstanding the position of the eyes, approaches Chemnitzia. As in the entire animal kingdom, the greatest difference exists amongst authors relative to the value of the swimming organs. Latreille has united the Cephalopods and Pteropods in a single division Pterygia,to which he attributes the same A^alue as to his Apterygia, comprehending all other mollusks. Gegenbauer and Huxley have demonstrated that the Pteropods are veritable Gasteropods furnished with a pair of accessory swimming organs. Already the discovery of Gasteropteron has shown the little value of the Pteropods as a division equivalent to the Gasteropods. The Heteropods merit still less to be considered as a division of equal value. The late Prof. O. A. L. Morch, although attaching as much systematic value to the lingual dentition as any other concholo- gist, acknowledged that no single organ could be used in classL fication unless its differential characters accorded with differences of other portions of the animal and shell ; but he endeavored to show that conchologists have erred in estimating too highly for systematic purposes the form of the shell, whilst neglecting other external characters, such as sculpture, structure (nacreous, porcellanous, etc.) and color. " According to my views, one must consider shells, so to say, from a mineralogical point of view." Having thus chosen conchological characters heretofore neglected, in grouping the genera and families, the discrepancies CLASSIFICATION. 6t between a natural classification of the shells and one founded on dentition, according to Prof. Morch, will disappear. " I have united in the family Tritonidae, according to the sculpture, Ranella, Triton, Pyrnla (Ficus), Dolium, Cassidaria and Cassis, placing them near to Cypraea." This is in accord with the character of their dentition, which widely removes Triton and Ranella from the Muricida?, close to which they have heretofore been placed, upon conchological characters — principally the form of the shell, the presence of varices, the operculum, and also upon a decided resemblance of the animals. I have carefully re-examined these genera and their relation- ships with others, in the point of view taken by Prof. Morch, and the result of this examination is to convince me that he has in the first place selected in the sculpture a character that is of generic importance only in the single genus Dolium — that is to sajr, its species happen to possess revolving ribs, and that even in the sculpture the relationship of Triton and Ranella with Murex is exceedingly close, whilst they have little or no affinity with Dolium, Cassis, etc. In fact, it is precisely because Prof. Morch has regarded lingual dentition a priori as an "infallible cri- terium"that he has been enabled to detect supporting resem- blances in the shells. It is easy to show in many other instances, as in the group under discussion, how heterogeneous is the assemblage united by means of the " infallible criterium." There is, besides, a growing conviction, in which I share, that there are no sharply-defined groups in nature ; that a generic character, for example, cannot be made to cover all its species ; that upon its borders occur forms which partake of the characters of other so- called genera, and that families, orders, etc., similarly coalesce upon their confines. I anticipate a future period when our larger collections, together with our better knowledge of external influ- ences and of the power of adaptation to them of these creatures, shall reveal to us a series of recent and fossil forms having rela- tionships so intimate, that our present system of classification and resulting nomenclature shall become utterly valueless. In this point of view classification is essentially arbitrary and we can only help ourselves by choosing that which does least violence to natural affinities. The value of a classification founded on a single organ (the lingual ribbon), which does 68 CLASSIFICATION. violence to most of the apparent affinities, whilst at the same time it fails of signification even in one of the most important func- tions with which it is connected, in that it does not enable us to certainly separate the phytophagous from the zoophagous animals, may be seriously questioned. We have many most im- portant characters of the mollusks which impress themselves upon their shells, so that they are in accord and enable us to predicate reciprocally their relationships ; and such characters appear to me to be much more useful for classification. If it be proposed that a single arbitrary standard shall be used because it is arbitrary and hence will remove all doubt as to the position of a given species, then the standard selected should be the most universal and the most apparent — namely, the shell. But if a natural arrangement be attempted, still less can we make account of any character which is not in accord with the assemblage of characters. A natural sequence can only result from an accordance of most of the organs and functions. That dentition in the mollusks is a character worthy of study, that it will throw light on many doubtful points, that it will cor- rect many errors is not to be disputed ; but the claims made for it are preposterous ; — for whilst a few hundred species only have had their tongues examined, described and figured, many thousands have been arbitrarily placed and displaced in conse- quence. Stimpson has examined the tongue of Ranella caudata and finds it to be that of a Murex ; accordingly he separates from Ranella a few other species because their shells resemble the shell of Ea,nella caudata and unites them also with Muricidae and this is practically the course (and necessarily so) pursued \)y all these dentition systematists. If conchological characters may be used to support the fabric reared upon the knowledge (I had almost written the want of knowledge) of a single structural character, why may they not be equally used against it. Is it not imperti- nent to make use of a few hundred observations of an organ which only pervades a portion of the mollusca, to establish a classification which is frequently in violent contrast with natural affinities ascertained by long examination of all the species, recent and fossil ? If the exo-skeleton or shell carries the impress of its animal, its right appreciation will afford us the only possible classification. CLASSIFICATION. 69 It is not partial, but pervades nearly the whole mollusca — as well those which have no lingual ribbon ; its universality is the proof of its higher systematic importance ; its relationships are not single,' it is the epitome of the modifications of molluscan struc- ture. Supposing the dentition of all living forms to be examined (an impossibility), we are still confronted by the fossil shells, which absolutely refuse to be classified by any other than con. chological characters. What shall we do with them ? Shall we use for these 30,000 species obvious external, universal charac- ters, yet discard these in the recent mollusca for the modifications of a partial character, the very slight observation of which has sufficed to show that it may not be predicated with certainty from either the shell, operculum, external features, or anatomy of the animal ? Whilst I shall continue to find in the shell the usual characters for the discrimination of genera and families, I shall not refuse all the aid which I can obtain from the study of lingual dentition as well as from all other sources which may enable me to more rightly appreciate natural relationships, to correct error, to avoid it. For the present, I prefer to treat Triton and Ranella as transitional genera having many relationships with the Murices, but partaking in their dentition and in some other structural details in the characters of Cassis, Dolium, etc. Prof. Theodore Gill published in 1811, under the auspices of the Smithsonian Institution an " Arrangement of the families of Mollusks," largely founded on their lingual dentition. Whilst this classification presents many features as novel as praiseworthy, reflecting the highest credit on the philosophical views and critical acumen of its author, it is, I think, equally unacceptable with those classifications heretofore proposed by European authors in which this character has been used as an exclusive guide. I reproduce Prof. Grill's arrangement below, as far as it relates to the Rhachiglossate mollusca. It will be noticed, as particularly affecting the groups of mollusks included in the present and succeeding volumes of this Manual, that the families of Fasciolariidae (including Fusinse) Biiccinidae,and Muricidse are widely separated, being each placed under a group founded on a modification of dentition, and that other families of mollusks otherwise not nearly related to these, are nevertheless interposed 70 CLASSIFICATION. between them, as for instance, to take an extreme case, the Olividae. Of course Ranellidae and Tritonidae do not appear at all in this table of Rhachiglossates — their dentition places them in the Taenioglossates. GASTEROPODA. Sub-class DICKGA. Order III. PECTINIBRANCHIATA. (Sub order TOXOGLOSSA. } Sub-order RHACHIGLOSSA. (Typica.} Formula (H'O- (PI. 5, tig. 48). Family 50. CYSTISCIDJE = Cystiscidae, Stm. Family 51. MARGINELLID^E < Marginellacea, Troschel. Family 52. VOLUTION = Volutacea, Tr. a. Volutomitrince { yolutpmitrina, Gray. I Amoriana, Gray. b. Volutina> fVolutina Gray. ( i etina, Gray. (Odontoglossa.) (PI. 5, figs. 49, 51), Family 53- FASCIOLARIID^E = Fasciolariacea, Troschel. a. Fusince. b. Fasciolfiriinw. Family 54. MITRID^ = Mitracea, Tr. (Duplohamata. } Family 55. MELONGENID^G = Cassidulina. Tr. Family 56. BUCCINID^E < Fusacea, Tr. a. Photinw = Photina, Tr. b. Buccinince = Buccinina, Tr. c. Chrysodomina = Neptunina, Tr. Family 57. NASSID^E == Nassacea, Tr. a. Cyclonassina>, b. NassininoR. Family 58. CYNODONTID.E < Fusacea, Tr. a. Cynodontince = Vasina, Tr. b. Imbricarimce = Imbricariina, Tr. ? Family 59. TURBINELLID.E ^ Vasidae, H. & A. Ad. (Hamiglowa.) (PI. 5, figs. 52, etc.) Family 60. TURRID^ = Strigatellacea, Tr. Family 61. OLIVID^E = Olivacea, Tr. a. OUmnas = Dactyliiia, Tr. b. OlivellincB = Olivellina, Tr. c. Ancillince «. Ancillina, Tr. CLASSIFICATION. fl Family 62. HARPIDJE = Harpacea, Tr. Family 63. PTYCHATRACTID^E = Ptychatractidae, Stm. Family 64. MURICIDJE. a. MuricincB = Muricea, Tr. b. Purpurince = Purpuracea, Tr. (Atypoglossa.} (PI. 5, fig. 50.) Family 65. COLUMBELLID.E = Columbellacea, Tr. The relations of the typical Fusi with Fasciolaria are intimate and a species of the former examined by Stimpson* has the saw- like lateral teeth of the latter ; on the other hand, Macdonaldf found the dentition of another species of the typical Fusi to agree with that of Murex in having simple, uncinate pleural teeth. I propose to follow a classification which does not differ very essentially from that most approved before the advent of the lingual system, but modified nevertheless by certain ameliorations which the latter has enabled us to perceive for the first time. It is not denied that this arrangement is exceedingly artificial — all are so ; but it it believed to be the most obvious, therefore, the easiest. It may be remarked here, that one of the inherent diffi- culties of arrangement in a lineal line is caused by inter- relationships ; thus, I have chosen to proceed from Murex to I'urpura, then to Triton, to Fusus and to Buccinum ; yet the three latter groups are equally related to Murex and with one another, and the passage from one group to another is so gradual that the assignment of some of the out-lying forms to a genus is very difficult. Classification adopted in this Work". FAMILIES. 1. MuRiciD.'E. Including Muriciniii and Purpurimt. *2. TRITONIDJS. Tritoniuse, Ranelluue. FUSID^E. Fusinae (typical) Fasciolariimw, Ptychatractina.1. BUCCINID^E. Xeptuninae, Melongenina?, Buccininae, Pisa- uiinae, Photimt. 5. NASSID^E. Nassinse, Cyclonassinte. (6. Turbinellidffi. 1. Volutida?. 8. Mitridte. 9. Colum- bellidte. 10. Marginellidae. 11. Olividse, including Ancillaricinse. 12. Harpidae, etc.) *Am. Jour. Conch, i. 54. fAnn. Mag. Nat. Hist., 4 ser., ii. 243. 72 CLASSIFICATION. SYSTEMATIC. CLASS GASTEROPODA. Head distinct, furnished with eyes and tentacles. Body usually protected by a spiral or conical univalve shell. Lower surface of body developing a thickened, expanded, creeping disk or foot. ORDER I. PROSOBRANCHIATA. Animal, creeping or swimming, protected by a shell usually large enough to contain it. Breathing organs (branchiae), plume- like, situated in advance of the heart. Sexes distinct. (Section A. SIPHONOSTOMATA.) Animal with its mantle margin prolonged into a siphon conveying the water into the branchial chamber. Carnivorous. Marine. The shell is spiral, the axis usually imperforate, the aperture prolonged into a canal, or simply notched below. Operculum lamellar, horny. (Section B. HOLOSTOMATA.) Respiratory siphon wanting, or represented by a mere lobe in the collar of the mantle. Shell spiral or limpet-shaped, generally globular or pyramidal, with the aperture entire below, and rounded. Marine, fluviatile or terrestrial. Phytophagous usually. (Natica is carnivorous.) Operculum spiral, horny or calcareous. SIPHONOSTOMATA. Family MURICID^E. Shell spiral, fusiform ; aperture more or less canaliculate, or simply notched in front. Sub-family Muricinae. Shell canaliculate ; whorls crossed by varices. Operculum ovate. Nucleus subapical. Sub-family Purpurinse. Shell with very short canal, or simply notched in front ; but frequently ribbed or nodulous ; colu- mella usually broad and flattened ; operculum oblong, nucleus elongate, forming the long outer edge. The above division into two sub-families holds good as to a majority of the species, but does not cover them all. Thus, CLASSIFICATION. 73 some of the Muricinse have the operculum and short canal of the Purpuringe, whilst some of the latter possess the varices of the former. In the following synopsis of genera and subgenera, the dis- criminative characters used separate widely groups, which really appear to be closely related : thus, Cerostoma and Pteronotus are intimately allied, notwithstanding the difference of the oper- culnm ; and Phyllonotus and Chicoreus have the same general facies, although they differ in the number of varices. Owing to their inter-relationships, no attempt to present the genera in suc- cession can be other than exceedingly artificial. Sub-Family MURICIN^E. Synopsis of Genera. MUKEX. Shell ovate or oblong ; spire prominent ; whorls convex, crossed by three or more continuous varices ; aperture ending below in a canal, which is generally partly closed. a. Operculum with sub-apical nucleus. * Varices three. Typical Murices. Shell spinous ; spire elevated ; canal very long, narrow, nearly straight. Subgenus PTERONOTUS. Shell triangular ; varices fin-like or foliated ; canal moderate, closed, somewhat curved. Subgenus CHICOREUS. Shell ovate-pyriform ; varices foliated and sometimes spinose ; canal short, curved, wide, nearly closed. * * Varices four to ten. Subgenus RHINOCANTHA. Has the short body whorl and long canal of the typical murices ; differs in having more numerous varices. Subgenus HOMALOCANTHA. Whorls rounded and sutures very deep ; varices foliated, and peculiarly produced into expanded digitations ; canal long. Subgenus PHYLLONOTUS. Like Chicereus, but varices numerous. b. Operculum purpuroid. * Varices three. Subgeuus CEROBTOMA. Varices wing-like ; aperture usually dentate within the outer lip, with a produced tooth near its base. Analogous with Pteronotus. 10 t4 CLASSIFICATION. * * Varices numerous. Subgenus VITULARIA. Shell oblong ; spire short ; body whorl long ; canal very short, wide ; outer lip thickened and dentate within. Varices nearly obsolete. Subgenus OCINEBRA. Spire elevated ; canal more or less closed ; varices foliated, sometimes spinose. UROSALPINX. Fusiform. No proper varices, which are replaced by longitudinal ribs. EUPLEURA. Ranelliform, with a pair of lateral varices, one on either side, and intermediate smaller varices ; aperture dentate within. TYPHIS. Ovate or oblong, with projecting hollow tubes between the three spinose varices ; aperture sub-orbicular, prolonged in front into a closed siphonal canal. TROPHON. Varices numerous, lamelliform or 1 acini ated ; spire promi- nent ; aperture ovate ; canal open, usually turned to the left ; shell white, often dark-colored within the aperture ; typically, Arctic and Antarctic. Sub-Family PURPURIN.E. As already stated, whilst the Muricida) naturally subdivide into two groups, one of them (Murices) distinguished by varices on the shell, oper- ctilum with terminal initial point, whilst the other (Purpurae) has nodules but no varices, patulous columella, short canal or mere basal notch, oper- culum with lateral nucleus ; yet on the confines of these two groups occur forms which partake of the characters of either, and the classification of which is entirely arbitrary. Ocinebra, species of Trophon, Urosalpinx and Eupleura, have undoubted relationships with Purpura, yet are classed with Murex— partly because the species have usually been con- sidered or were described as Murices ; on the other hand, Purpura crispata and its allies possess the variceal features of Murex. Kobelt has, on this account, included them in his catalogue of the genus Murex ; but on account of the extreme variability of the species (some specimens being without varices) and the number of connecting forms between the smoother varieties and typical PurpuraB, I prefer to retain them in the group to which they have usually been referred. If the difficulty of defining these two sub-families is great, still greater does it become when we descend to the genera and subgenera of either of them. Various authors have attempted it, from the "groups" of Kiener's monograph to the genera and subgenera of H. & A. Adams. I adopt the latter as a mere convenience, premising that nature presents her specific forms here (as frequently elsewhere) in such continuous series, that no CLASSIFICATION. 75 real line of demarcation can be traced ; the characters represent simply the high tide of an osculation, which at its ebb merges into the next incoming wave. Synopsis of Genera. PURPURA. Shell oblong-oval, last whorl large ; spire generally short ; aperture ovate large, terminating in a very short, oblique channel, or notched ; columella flattened ; outer lip simple. PURPUROIDEA. Shell turriculated, ventricose ; summit of the spire sharp ; whorls convex, with a line of spines or tubercles on the shoulder ; columella smooth, rounded, excavated in front; siphonal notch wide ; outer lip thin. Fossil. LYSIS Gabb, Stomatiform, very oblique ; spire moderate ; whorls costate ; aperture narrow, outer lip simple inner lip straight, concavely expanded over the wide umbilicus so as to completely cover it. Fossil. IOPAS. Shell ovate, rugose, last whorl large ; spire acuminate ; aperture moderate, emarginate and channelled in front ; columellar lip covered witli a thin enamel, and with a prominent plait-like callosity at the hind part ; outer lip sinuous, crenate within. VEXILLA. Shell purpuriform ; inner lip flattened and depressed, but outer lip, when adult, thickened, inflected and toothed ; aperture wide. RICINULA. Shell ovate, solid ; spire short, whorls tubercular or spinous ; aperture linear, narrow, contracted by callous projections, with a short, oblique, emarginate canal in front ; inner lip tubercularly wrinkled ; outer lip internally with plait-like teeth, often digitate. MONOCEROS. Shell ovate, last whorl large; spire rather elevated; aperture semilunar ; inner lip wide and flattened ; outer lip crenated, with a prominent tooth at the fore part. PSEUDOLIVA. Shell ovate, solid, subglobose ; spire very short, suture slightly channeled, whorls tumid round the upper part ; aperture oval, canal very short ; inner lip arcuated, with a callosity at the hind part ; outer lip thin, furnished at the fore part with a small tooth or callosity. CHORUS. Shell laminately varicose, spinose on the shoulder ; canal rather long ; outer lip with a spine as in Monoceros. PIKAXIA. Shell conical ; spire short, acute ; aperture oval-oblong, emarginate anteriorly ; inner lip flattened, with several transverse plaits in the middle ; outer lip acute, grooved internally. CONCHOLEPAS. Shell ovate, last whorl very large, expanded ; spire very short, obliquely inclined towards the left side ; aperture very wide, slightly channeled anteriorly ; inner lip flattened ; outer lip with two small teeth in front. 6 CLASSIFICATION. CUM A. Shell pyriform ; spire elevated, acute, whorls angular or spinose ; aperture oval- oblong ; columella convex, sometimes with a strong an- gular tubercle in the middle ; outer lip acute, grooved internally. RAPANA. Shell ventricose, axis perforated to the apex ; spire de- pressed ; aperture oval, narrowed anteriorly ; canal open, slightly recurved; inner lip reflected, free anteriorly; umbilicus Avide, corrugated. RHIZOCHILUS. Shell when young free, resembling Rapana ; when adult, sometimes with more or less irregular solid shelly extensions of the outer and inner lips, which clasp the axis of coral or the surface of neighboring shells, and at length close the month with the exception of the anterior siphonal canal which is converted into a shelly tube. No operculum.(?) • SEPARATISTA. Shell turbinate, subdiscoidal, the first whorls con- tiguous, the last more or less separated ; aperture expanded, slightly angulated, the margin everted ; umbilicus very wide, infundibuliform with the whorls visible to the apex. No operculum. MELAPIUM. Shell ovate-pyriform, ventricose, imperforate, porcella- nous ; spire very short, apex papillary ; aperture expanded, inner lip with a thick, smooth callus at the hind part columella twisted anter- iorly, with a prominent oblique plait ; canal wide, recurved, directed towards the left. Operculum unknown. WHITNEYA. A cretaceous fossil possibly synonymous with Melapium. RAPA. Shell thin, globosely pyriform ; axis perforate ; umbilicus partly concealed by the reflected inner lip ; spire obtuse ; aperture oblong, produced anteriorly into a wide, subrecurved canal. Operculum unknown. MAGILUS. Shell when young, spiral, thin ; when adult, white, solid, tubular, spiral for three or four whorls, the last prolonged into an irregular straight or flexuous tube, solid posteriorly, and with a siphonal keel on the left side. Operculum ovate, nucleus sublateral. MAGILINA. Young shell free, formed of a single whorl ; finally prolonged into a tube which is attached by one side to the surface of submarine bodies. NISEA. Shell composed of a discoidal portion and of two tubes ; the last whorl recurved upon itself in the same way as Anastoma, in two tubes of variable length and less sinuous than the single tube of Magilus. (Fossil.) Relationships very doubtful. I NIVEKSITY OF OAL1FOUMA. MUREX. 77 Genus MUREX, Linn. Mr. W. Kobelt has published a catalogue of the genus Murex (excluding the genera Trophon, Yitularia and Typhis) in the •' Jahrbiicher " of the German Malacozoological Society, 1877. Whilst it is the most complete catalogue of the genus heretofore published, I find that it does not include more than about half the number of specific names amassed during my researches. His introductory remarks upon classification are important, and fairly state the difficulties attendant upon any attempt to separate the species into natural groups. FiATe monographs, illustrated by colored figures of the species, have been published, viz., in Sowerby's " Conchological Illustra- tions," in Reeve's " Conchologia Iconica," in " Kiener's " Co- quilles Yivantes," in Krister's " Conchylien Cabinet," and in Sowerby's " Thesaurus Conchyliorum." The fossil species number more than one hundred and fifty, beginning with the Eocene. On the coasts of the Adriatic Murex brandaris and M. trun- culus are constantly seen in the markets, where, under the names of " bulo niaschio," ubulo femina," and " garusola," they are sold to the poorer classes for food. 1. Typical Murices, or Tribulus Group. M. SCOLOPAX, Dillw. PI. 9, figs. 106, 108 ; pi. 24, fig. 208. The comparative smoothness of the surface of this species will distinguish it from M. ternispina,vf\i\i which it is too closely allied. The revolving ribs are much darker in color than the general surface, so that Reeve describes the specimens as banded. Full grown specimens attain 8 to 9 inches. Red Sea, Indian 0., China. M. occa, Sowb. (fig. 108), is a depauperate young state, the spines not so wrell developed, onl}' three of them prominent on the body, the intermediate ones becoming more noticeable with increase of growth. M. Macgilliurayi, Dohrn. (fig. 208), from Lizard Isles, Australia, is similar to occa. M. TRIBULUS, Linn. PL 9, figs. 107, 109. The whorls are crossed by numerous, somewhat nodulous transverse ribs (4 to 6 between varices), decussated by close, 78 MUREX. alternately larger and smaller revolving lines. The typical form has also several revolving series of brownish spots, but these are not constant The spines of the canal are not so numerous or so long as those of M. scolopax. Attains 4-5 inches. Red Sea, China, Japan. M. nigrospinosus, Reeve (fig. 100), is simply a state of this species in which the spines become dark colored. 44. M. TENXispiNA, Lam. PI. 10, fig. 113. The distinguishing character of this, the most beautiful of the Murices, is the large number of long, parallel, curved spines which adorn the canal and lower portion of the body, with alternating, recurved, smaller spines. In its sculpture it is like M. fribulus, and some of the succeeding species. Length, 6 to 8 inches. Indian Ocean, Japan. Torres Sts., N. Australia, in 20 to 3Q fathoms, sandy bottom. M. TERNISPINA, Lam. PL 9, fig. 110; pi. 10, figs. Ill, 114; pi. 11, figs. 117, 118. Usually smaller, narrower, especially the canal, and with fewer and shorter spines than M. tribulus ; yet it is by no means readily distinguished from that species. It has the same sculp- ture, and even sometimes the same revolving. spots of brown color. It is more graceful usually, in form, its spines are not so stout, and the upper and middle series on the body whorl are not much larger than the others, as in tribulus. Its claims to specific rank are allowed with considerable hesitation. Length, 6 inches, but usually not over 3 inches. Indian Ocean, Japan, China, Philippines. The animal is figured by Quoy and Gaimard (under the name of tenuispina (fig. 117) as greatly extended from its shell in order to regain its normal position when placed on its back. The extended foot is quite large, subcylindrical, yellowish, marbled on the sides with yellow, brown and red. Two very long, cylindrical, thin and pointed tentacles carry eyes on their middle. The mantle is grayish ; its margins undulated. M. Martinianus, Reeve (fig. 118), and M. aduncospinosus, Beck (fig. 1 14) , I cannot separate even as varieties ; they are simply .MUREX. 79 degrees of spinous development of this species. M. Troscheli, Lischke (fig. Ill) is founded on a very large, stout specimen, which, when perfect, must have been nearly 7 inches in length. I place it here ; yet it has affinities with M. tribulus, particularly in its solid, stout appearance. M. SOBRINUS, A. Ad. Suppl. PL, fig. 536. The seven whorls are convex, longitudinally nodosely plicate and transversely lirate, very spiny ; the long, straight canal is spiny ; the body whorl has two reddish-brown bands. Length, 36 mill. Japan, 29 to 55 fms. M. RARISPINA, Lam. PI. 10, fig. 115 ; PI. 11, fig. 119. The best character of this form is found in the great develop- ment of the superior spines on the body whorl. The lower part of the canal is smooth, and the upper part has only a few short spines. Length, 3*5 inches. Indian 0. M. Mindanensis, Sowb., (fig. 119) I believe to be a depauperate state of this species. Messrs. Kiener & Reeve consider M. formosus> Sowb., (fig. 115), a synonym, and as the former cites rarispina in the Lamarckian collection, I am disposed to agree with them. Mr. Sowerb^y figures an entirely different shell for rarixpiiia — a shell which appears to me to be a M. tribulus. M. BREVISPINA, Lam. PI. 11, fig.121 ; PL 24, fig. 209. The spines are very short, scarcely longer than the tubercles which, in double series, two in each, divide the space between the varices ; these tubercles define a someAvhat flattened periphery. Length, 3 inches. Red Sea, Ind. 0., So. Africa, N. Australia, 6-11 fms. 31. seniUs, Jousseaume, (fig. 209), can scarcely be regarded as more than a thin variety of this species, in which the spines are better developed. It is said to inhabit the Philippine seas. M. CONCINNUS, Reeve. PL 11, fig. 122. The colored, narrow revolving lines, and short, direct spines give this shell a very lively appearance. Only a single specimen is known (Metcalfe Museum). It can scarcely be regarded as a well-established species because these narrow bands are found in 80 MUREX. individual specimens of several of the following forms (and have in these cases caused them to be described as new species), with- out being at all characteristic. Locality unknown. M. CABRITII, Bernard!. PL 11, fig. 123. The short, thick-set spines, extending over the canal, remind one of M. plicatus, Sowb., with wrhich it also has other relation- ships. Only a single specimen is known, without locality. Length, 2'5 inches. M. PLIOATUS, Sowb. PI. 40, fig. 508. This species is well distinguished from others of the group ; it is thick and heavy, the spines are obtuse, short on the whorls and long on the canal. The color is purplish white, darker within the aperture. Length, 3 inches. W. Coast of Central Am. to Gulf of California. Two specimens in the Mus. Philad. Acad. have thread-like brown bands like M. concinnus, Reeve. M. BEAUI, Petit. PI. 11, fig. 116. Described from a single specimen which, except in its much greater size, much resembles varieties of M. recurmroxtris. The shell is not in good condition, and presents no satisfactory spe- cific characters. Length, 5 inches. Isle of Marie- Galante, W. 1. _ M. EXIMIUS, Brazier. Whorls seven, with two short, blunt spines on each varix of the body and no spines elsewhere ; spaces between the varices longi- tudinally five-ribbed, crossed by alternately larger and smaller striae. Cream color, violet-tinged in the aperture. Length, nearly 2 inches. Not figured. Darnley M. Torres Sts. N. Australia; 30 fms. sandy bottom. M. RECURVIROSTRIS, Brod. PI. 11, fig. 193 ; PI. 10, fig. 112 , PI. 12, figs. 124-128. This is a comparatively small species, rarely exceeding two inches in length. The varices are thick, plait-like and tubercu- lated by the crossing of elevated lines — which also cut the three inter-variceal ribs into tubercles. These tubercles are sometimes developed into short spines, one on the upper part of each varix, MT REX. 81 and there are also one or two on each varix below the aperture. The color varies from whitish to livid with two or three broad brown bands — which are most visible within the aperture. West Indies ; W. Coast of Central America to Mazatlan. Having a very extensive suite of this species, including specimens from well-authenticated localities on both sides of the American continent, I find no difference in them whatever as to the typical form described above, and which may be recognized at once by its general dark color. M. nigrescens, Sowb. (fig. 124), is simply a large specimen of the typical form and M. livi- r/w.s, Carp., cannot be distinguished as a variety. M. funiculatus, Reeve (fig. 112) is founded on an unusually light colored specimen ; similar ones are in the collection before me as well as examples of intermediate coloration. M. messorius, Sowb., (fig. 125), is similar to M. f-uniculatus and also has relationships with the form with straight canal which the same author has named M. rec- tirostris, (fig. 126). There is also a West Indian form, much lighter in color, and with the spines usually somewhat more developed whilst the canal is generally shorter. If I had fewer specimens I might be, able to distinguish this form as a variety, but my material sup- plies all intermediate gradations. Of these lighter-colored shells Sowerby has made his M. similis (fig. 130), whilst another specimen has been figured by him in mistake for M. motaciUa, Chemn. Two eager English conchologists discovered this error and Mr. Hinds has added to his scientific laurels by describing the wrongly identified shell as M. Antillarum, whilst Mr. Reeve has called it M. nodatus and refigured it. Of course neither of these investigators felt called upon to study the group to which Mr. Sowerby 's wrongly identified species belonged and equally of course neither of them had time to ascertain whether any other naturalist had made and published a similar discovery. Thus I place eight so-called species in the synonymy of M. recurvi- rostris in addition to a number of synonyms acknowledged by pre- ceding authors and upon which I therefore refrain from dwelling. Mr. Arthur Adams has issued a short Latin diagnosis (Zool. Proc. 1851), of M. pulcher from St. Croix, W. I. No figure has been published and the description mentions neither color, 11 82 MUREX. dimensions nor distinctive characters (nearly all the species pub- lished by him in this paper are similarly indefinite), so that it is impossible to ascertain what species he intended to describe ; the description itself, however, as far as it goes, will characterize very exactly our present species, and such being the case, I have a right to so identify it. M. MOTACILLA, Chemn. PL 12, figs. 129, 131 ; pi. 13, figs. 132-134. There are usually two prominent longitudinal tubercles be- tween the varices instead of three, as in the last species. The surface is closely covered with revolving lines which become tuberculous on the elevations, and here and there slightly spinous. The canal is very narrow, and turned to the right. There are two or three brown bands, usually. Length, 2*5 inches. Senegal, West Indies. This species is rather variable, and presents, among others, two forms which may retain the specific names given them, as varieties : VAR. CAILETI, Petit. Figs. 131, 132. Shell smoother, more angulated on the periphery, more spinous, the nodules smaller, forming two or three ribs or tubercles between the varices. The principal revolving lines are colored brown so that the surface appears variegated with revolving, narrow, thread- like bands. I figure a specimen from the original lot collected by M. Caillet, which is much more angulated than the type^fig. 131) and I have before me a fine series of intermediate forms. VAR. ELEGANS, Beck. Figs. 133, 134. Longitudinal tubercles two between the varices, prominent, spines not developed, periphery only obtusely angulated, the raised revolving, thread-like lines dark-colored. This is a very beautiful variety which may include as a synonym M. trilineatiis, Reeve, (fig. 134). M. CHRYSOSTOMA, Gray. PL 13, figs. 136, 135. In general appearance not unlike M. motaciUa, Yar. elegans, this species presents the following distinctive characters, which appear to be permanent : it is larger, stouter, the canal is pro- MUBEX. 83 portionally shorter and straight or but little turned to the right; there are one or two spur-like spines on the varices at the lower part of the mouth ; — generally on the left-hand varix and some- times on the back varix, but very seldom on the right hand or lip varix ;. the lips of the mouth are tinged with more or less brilliant orange color. I consider M. bella, Reeve, (fig. 135), a synonym. West Indies. M. AUSTRALIS, Quoy. Shell fusiform, a little ventricose, with long canal turned to the left ; with revolving striae, and short tubercles on the varices. Yellowish-white. The animal has short tentacles of a sooty color, white towards the points, near which are placed the very small eyes. The rest of the body is pale yellowish, mottled with white. Length, 33 mill., breadth, 14 mill. Port Western, Austr. This species has not been figured nor identified ; I think it belongs to the Tribulus group. M. LIGNARTUS, A. Adams. Ovately fusiform, sub-umbilicated ; spire acuminate, reddish brown ; whorls excavated above, with two medial, elevated, nodulous lines; transversely lirate, the lines unequal, elevated, rugulose ; longitudinally three-varicose, the varices with two elevated spines ; aperture round-ovate, white within ; canal as long as the aperture, subrecurved. West Africa. This shell has not been figured nor are dimensions given. I cannot, therefore, be sure that it belongs to this group. Sowerby (Thes. Conch.) refers Ugnarius to M. quadrifrons; I do not know whether it be this species, but the description of it does not agree at all. M. HAUSTELLUM, Linn. PL 13, fig. 137. This well-knowrn species will end the section of typical Murices. Like M. chrysostoma it is never spinous. Adult specimens vary from 2 to 6 inches in length. Red Sea, Ind. 0., China, Mauritius, Philippines. 84 PTERONOTl S. Sub-genus Pteronotus, Swainson. This group may be advantageously restricted to those species possessing a muricoid operculum and three varices. Its position is somewhat difficult to define, inasmuch as its relationships are varied — with Chicoreus rather than with Phyllonotus perhaps, and on the other hand with Cerostoma : — which latter may include the forms, likewise with three varices, with or without the lip- spine which Conrad makes the typical character of his group, but with purpuroid operculum, and in general aspect forcibly suggesting a connecting link with Purpura. Kobelt has indeed, included such species as Purpura plicata and lactuca in his group of Cerostoma, but I think the line between the Purpuroid Muri- ces and Purpura proper may be drawn just here, if anywhere — that Cerostoma is a normal Murex in its three continuous varices : whilst the numerous varices of P. pi tea la :nxl its allies insensibly degenerate into shells without varices = typical Purpurae. It may be remarked that Pteronotus, as thus restricted, consists of East Indian and African species, whilst the distribution of Cerostoma, is North Pacific, extending from the west const of North America to Japan. a. With two or three ribs or tubercles between the varices. M. TRIGONULUS, Lam. PL 11, fig. 120. Owing to the insufficiency of the original description and the want of reference to a published figure this species has been variously identified with several others since described. I prefer the decision of Reeve, because it relieves these other species from doubt. Kobelt gives "Antilles '• as locality, which shows that he has supposed it related to the motacil/a or >•>'<•// /-rirostris group — which it certainly does somewhat resemble. Tapparone in quoting it from the Red Sea, confounds it with tri