Art. XXI.—On the Structure of the Head in Palinurus, with especial Reference to the Classification of the Genus.
[Read before the Otago Institute, 11th September, 1883.]
On my voyage out from England I was much interested to find that the marine crayfish (Palinurus lalandii), brought on board at the Cape of Good Hope, differed in a remarkable way from the common English species P. vulgaris, in that it had no trace of the stridulating organ, one of the most specialized characteristics of this very specialized genus.
The species of Palinurus are arranged by systematists in two groups or sub-genera, originally founded by Milne-Edwards,* who defines them as follows:—
“Sous-genre das Langoustes ordinaires.
“Les Langoustes ordinaires présentent sur le milien du front une petite dent rostriforme plus ou moins saillante; l'anneau antennulaire est très-étroit, de façon que les antennes externes se touchent presqu' à leur base, et recouvrent les antennes internes; enfin celles-ci se terminent par deux tigelles multi-articulées très-courtes.
“Sous-genre des Langoustes longicornes.
“Dans cette division naturelle il n'existe sur le bord anterieur de la carapace aucun vestige de rostre médian; l'anneau antennulaire est très-large et presque carré, de manière à écarter beaucoup les antennes externes et à laisser à découvert les antennes internes; enfin ces derniers organes se terminent par deux tigelles multi-articulées très-longues.”
The two species mentioned above (P. vulgaris and P. lalandii) as presenting such remarkable differences, occur in Milne-Edwards's work as the first and second species of Langoustes ordinaires; they are also placed together in Heller's table† as the third and fourth species of the same sub-genus, no hint being given in either work, or in any other at my disposal, of any important morphological differences between them.
In the present paper I hope to show that a more natural classification of the species of Palinurus may be obtained by taking into consideration certain points in the anatomy of the head, which have hitherto been largely ignored by systematists, although, as it seems to me, of fundamental importance. The chief of these have to do with the sound-producing orstridulating organ, first mentioned, I believe, by Leach,‡ in P. vulgaris, and described at length, in the same species, by Möbius,¶ and subsequently by myself.§
[Footnote] * Hist. Nat. des Crustacés, ii., p. 289.
[Footnote] † Reise der Novara, Zool. ii., Crustacea, p. 94.
[Footnote] ‡ Malacostraca Podophthalmata Britanniæ.
[Footnote] ¶ Archiv für Naturgeschichte, 1867.
[Footnote] § Proc. Zool. Soc., 1878, pp. 292 and 442.
An examination of several species of Palinurus has shown that while the stridulating organ is present in all the Langoustes longicornes associated with long antennulary flagella and obsolete rostrum, there are some Langoustes ordinaires in which it is well developed, others in which it is wholly absent; and further that in the latter group some species have a rostrum quite comparable with that of the Astacidæ, others having one so small as to merit special description only from its position.
I propose to describe, somewhat in detail, the structure of the head in P. edwardsii, P. vulgaris and P. interruptus, in order to bring out the points of likeness and of difference in three typical forms.
In Palinurus edwardsii, one of the common New Zealand species, the carapace is produced in front into an upturned rostrum (figs. 1 and 4, r) which is confluent laterally with the large supra-orbital spines (s. or. sp). The ventral faces of these latter pass insensibly into the “epimeral plates” (epm. pl.)* which form the posterior walls of the orbits, and are fused externally with the anterior border of the carapace and internally with the antennulary sternum (ant. st.). This latter is a vertical bar, just wide enough below to furnish the articular facets for the antennules (ant. 1′), and narrowing to its dorsal end where it becomes bent backwards at a right angle (fig. 1) to join with the epimeral plates. The latter, which are considered by Huxley as representing antennulary epimera,† bound a transversely oval area, consisting of uncalcified chitin, and bearing the small movable ophthalmic segment (oph.) or pseudo-segment, the middle portion of which is calcified, forming the so-called ophthalmic sternum. The epimeral plates are thus united with one another above and below the eye-bearing space forming supra-ophthalmic and infra-ophthalmic bars.
The infra-ophthalmic bar gives off on each side of the middle line a pedate process (cl. pr) which extends forwards in contact with, but quite free from, both its fellow of the opposite side and the dorsal or horizontal portion of the antennulary sternum. Each of these processes ends in a sharp spine and, from its proximal end, sends off an upwardly directed offshoot ending in two small spines, which is so closely applied to the lateral surface of the rostrum that the latter appears to be actually squeezed between the two clasping processes as they may be conveniently termed (cl. pr, figs. 1 and 4).
These clasping processes appear to have been very imperfectly described hitherto. Milne-Edwards merely says that there are two small spines below the base of the rostrum (in P. lalandii which agrees in all important
[Footnote] * Huxley, “The Crayfish,” p. 156, fig. 40.
[Footnote] † Is it not more likely that these plates represent the ventral region of the unsegmented præpstomium? If embryology should answer this question in the affirmative it will be convenient to speak of a single præstomial plate consisting of lateral, supraophthalmic, and infra-ophthalmic portions.
respects with P. edwardsii), and the same statement is made by Miers* and by Haswell.† The processes are perfectly distinct from the rostrum, either being removable without injury to the other, and clearly belong to the epimeral plates, so that if the latter are, as Huxley supposes, antennulary epimera, the clasping processes are to be looked upon as epimeral—or possibly as partly sternal and partly epimeral—outgrowths of the antennulary segment.
The true relations of the rostrum are very imperfectly seen in an external view: owing partly to the presence of the clasping processes which form an apparent proximal boundary to it, partly to the fact that on its dorsal surface it widens out suddenly and becomes confluent with the supra-orbital spines, it appears externally as a very small structure, hardly larger indeed than the clasping processes (fig. 4). But a longitudinal section (fig. 1) shows that it is really a structure of considerable size, being continued backwards some distance behind the clasping processes, and a short distance behind the ophthalmic segment where its ventral plate turns sharply forward and becomes continuous with the supra-ophthalmic bar of the epimeral plates.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
A short distance on each side of the middle line the supra-ophthalmic bar is produced into a small hollow procephalic process (fig. 1, prc. p.) which passes backwards, upwards, and outwards into the interior of the head quite like the homologous structure in Astacus and Homarus, from which it differs only in its small size: in a moderate sized specimen (8 or 9 inches) of P. edwardsii it is only about 1/10 inch in length.
The proximal segments (coxocerites) of the antennæ are, as in other species of the genus, fused with the carapace, so that the apparently proximal segment is really the second or basicerite (bc); of the coxocerites only the ventral portions are left, their lateral and dorsal regions being as it were squeezed out of existence by the immense development of the free portions of the antennæ, the articular cavities for which are thus bounded above by the epimeral plates, internally by the antennulary sternum, externally by the anterior borders of the carapace, and only below by the coxocerites. The fused coxocerites thus come to occupy the position of the epistoma of Astacus,‡ from which, however, they are at once distinguished by bearing the renal apertures. Posteriorly they are continued in the middle line into a projecting transverse bar (ep) which gives attachment to the labrum, furnishes the antero-internal articular facets for the mandibles, and apparently is the much reduced representative of the true epistoma.
[Footnote] * Cat. of the Stalk- and Sessile-eyed Crustacea of N.Z., 1876.
[Footnote] † Cat. of the Australian Stalk- and Sessile-eyed Crustacea, 1882.
[Footnote] ‡ Huxley, “The Crayfish,” p. 155.
The bonndary between the two coxocerites is marked externally by a short median longitudinal groove; the point of a fine scalpel pushed into this is found to penetrate easily a short distance into the interior of the head. When the soft parts are removed, and the head is examined from within, the instrument is found to have passed into an extremely narrow cleft between two delicate closely applied plates (i.cx. pl), which together form a median crest projecting upwards into the cavity of the head. These plates evidently represent a small portion of the applied inner walls of the coxocerites, and may be called the internal coxoceritic, or, for shortness, coxal plates. The posterior and outer edge of each coxocerite is similarly produced into a vertical obliquely-set plate (e.cx. pl), which may be called the external coxal plate. It is very possibly formed, in part, by the epistoma.
The basicerite (bc), or proximal segment of the movable part of the antenna, articulates with its socket (ant. 2′) in the usual way by two hinges; one of these is dorsal and internal (fig. 4, h), and is situated at the upper end of the antennulary sternum just where it turns sharply backwards; the other is ventral and external, and is situated at the junction of the coxocerite with the carapace. The articulation is thus an oblique one, elevation being accompanied by abduction, depression by adduction.
The rule in Crustacea is for each segment of a limb to be moved by two muscles, inserted into its proximal edge at points intermediate between the hinges. There are two such muscles to the basicerite of Palinurus edwardsii: one inserted dorso-externally, arising from the inner surface of the carapace and acting as combined elevator and abductor; the other ventro-internal, arising bi-pinnately from the outer surface of the internal, and from the inner surface of the external coxal plate, and acting as combined depressor and abductor. But besides these, the proximal edge of the basicerite is produced, immediately below the insertion of the elevator, into a strong serrated ridge, into which is inserted a muscle arising from the outer face of the external coxal plate. This being inserted above and external to the ventro-external hinge, must act as an additional elevator and abductor.
Essentially the same structure of the anterior part of the head is seen in P. lalandii and in P. hügelii: in the latter species, as I am informed by Mr. T. W. Kirk, the clasping processes are smaller than in P. edwardsii, but there is no difference of importance.
In P. vulgaris, on the other hand, great and important differences are at once apparent. A vertical section (fig. 2) shows that the rostrum (r) is a mere insignificant spine, smaller than those on the lower edge of the supra-orbital spines (s. or. sp), and no larger than the spiniform tubercles with
which the surface of the carapace is covered. As a result of this, the ophthalmic sternum (oph) is quite bare and fully visible in a view from above (fig. 5), there being, further, no trace of “clasping processes” developed from the antennulary sternum.
Still more striking are the modifications of the antennulary sternum (fig. 2, 5, and 7, ant. st.): it “forms a projecting vertical keel, with a rounded anterior border, and slightly convex sides; the anterior border is grooved in the middle line, the groove being bounded on each side by a strong smooth ridge (r); each lateral surface is marked with a shallow groove (g), and between the groove and the border, to which it is parallel, the surface (s) is so smooth as to have the texture of polished ivory.”* This keel, which constitutes the fixed portion of the stridulating organ, is broad at its postero-dorsal extremity, and thus interposes a wide space between the dorsal ends of the basicerites, but ventrally it becomes much narrower, so that the basicerites, being applied closely to its lateral surfaces, are very closely approximated at its antero-ventral extremity (fig. 5). For this reason the bases of the antennules are concealed, in a view from above, by those of the antennæ, as in P. edwardsii.
The articulation of the basicerite with its socket presents the noteworthy peculiarity of the absence of the dorso-internal hinge. From the spot where this hinge occurs in P. edwardsii, a strip of uncalcified chitin extends distalwards, and is so modified as to form the movable portion of the stridulating organ. The upper edge of the strip of chitin is produced into a large outstanding flap (figs. 5 and 8, fl.), immediately beneath which “the chitin becomes thickened and takes on the form of an oval area or pad (p) marked by a number of fine parallel ridges. In relation with the inferior edge of this pad is a small calcified tubercle (t). When the antenna is in place the flap works over the ridge (r) on the corresponding side of the anterior border of the antennulary sternum, the tubercle fits into the groove (g) on its lateral surface and the ridged pad is closely applied to the smooth space (s) between the groove and the border. Under these circumstances, when the antenna is moved upwards, the friction of the pad against the smooth surface produces a loud and grating noise, the principle being similar to that of the sound produced by the friction of india-rubber against paper. The apparatus can at any time be thrown out of gear and the antenna move noiselessly by slightly abducting the latter.”†
In consequence of the absence of the upper hinge the movement of the antenna is not, as in P. edwardsii, confined to one plane. I am disposed to think that the additional abductor muscle, described above in that species, has, in P. vulgaris, the special function assigned to it of abducting the antenna when it is to be moved noiselessly.
[Footnote] * Proc. Zool. Soc., 1878, p. 292.
[Footnote] † Proc. Zool. Soc., 1878, p. 292.
The internal coxal plates (i. ex. pl.) have the same relations as in P. edwardsii, indicating imperfect fusion of the coxocerites: external coxal plates are also present. There is, however, no trace of procephalic processes.
In P. trigonus, another of the Langoustes ordinaires, I am informed by Mr. Haswell that there is a prominent squarish rostrum covering the ophthalmic segment, but that in other respects the resemblance to P. vulgaris is very close, there being a well developed stridulating organ, and no clasping processes.
In P. interruptus, as well as in all other Langoustes longicornes which I have examined or obtained descriptions of, there is a stridulating organ of essentially the same structure as in P. vulgaris. The antennulary sternum (figs. 3 and 6, ant. st.) is, however, truncated in front instead of pointed, so that no part of it overhangs the articular ends of the antennules, which are therefore visible in a view from above. Moreover, the antennulary sternum is nearly as broad in its antero-ventral as in its postero-dorsal region, so that the basicerites (bc) are kept widely separated, and do not, as in all the Langoustes ordinaires, partly hide the bases of the antennules when the animal is viewed from above.
As in P. vulgaris, there is no trace of procephalic processes. The rostrum also is completely aborted.
The fused coxocerites present no trace of the median groove mentioned in the preceding species, and in correspondence with this the internal coxal plates are absent (fig. 3), so that fusion of the coxocerites is complete. Instead of there being two distinct external coxal plates, one for each coxocerite, there is a single transverse sub-vertical posterior coxal plate (p.cx. pl) consisting of two laminæ, an anterior formed by the posterior walls of the coxocerites, and a posterior furnished by the epistoma (ep).
Coming now to the general bearing of these facts, it seems clear, to begin with, that the Langoustes ordinaires do not form a natural assemblage, since some of the species included in the sub-genus have a well-developed rostrum, simple unmodified antennulary sternum, etc., while others possess a perfectly formed stridulating organ,—always a mark of high specialization, rare under any form in Crustacea,* and in the present case of a quite unique type of construction.
The longicornes on the other hand form a strictly natural group, since they all possess the stridulating organ and long antennulary flagella, while none have any trace of a rostrum.
[Footnote] * See Wood-Mason, Proc. Entom. Soc. Lond., Nov. 1877, p. xxvii.
Moreover P. vulgaris and P. trigonus are seen to be intermediate between the species without a stridulating organ and the longicorns, approaching more nearly, however—at least so it seems to me—to the latter, in virtue of the possession of that mark of high specialization the stridulating organ.
Assuming, as I think one is bound to do, that the Palinuridæ are descended from some Astacine or Homarine ancestor, probably through some such intermediate form as Palinurellus,* one cannot but conclude that the species which have no stridulating organ, a well-developed rostrum, and imperfectly fused coxocerites come nearest to the primitive stock, and that those species in which the stridulating organ is present, the rostrum atrophoid, and the coxocerites completely fused, have undergone the widest divergence from that stock and present us with the extreme of modification of the Palinuroid type.
These views may be conveniently expressed in the form of a phylogenetic table, as follows:—
Assuming that this table is an accurate expression of the relationships of the species of Palinurus, I think there can be no doubt that, in classifying the species, the most important division must come along the line a b, which divides the comparatively generalized non-stridulating forms from
[Footnote] * I have unfortunately neither specimen nor description of this interesting genus, and take the fact of its being the most primitive of the Palinuridæ from a brief notice of a paper by Boas (Studier over Decapodernes Slaegtskabsforhold) in the “Zoological Record for 1880,” p. (Crust.) 32.
those which possess the stridulating organ. A second very natural subdivision, but of less importance than the first, is indicated by the line c d which divides the stridulating species into brevicorn and longicorn forms.
For systematic purposes it will probably be convenient to employ both these boundary lines, that is, to divide the species into three sub-genera instead of two as heretofore. The non-stridulating species or Silentes form the first sub-genus, which I propose to call Jasus:* the stridulating forms or Stridentes fall into two sub-genera, one of brevicorn species, to which the name Palinurus may be restricted, and one of longicorn species, equivalent to Milne-Edwards's Langoustes longicornes, Gray's Panulirus, and Pfeiffer's Senex.
Put in a tabular form the classification I propose is as follows:—
Genus Palinurus, Fabr.
A. Silentes. Stridulating organ absent; rostrum well developed, clasped by paired pedate processes of the epimeral plates; procephalic processes present; coxocerites imperfectly fused; antennulary flagella short.
Sub-genus Jasus, T.J.P.
Includes P. lalandii, P. edwardsii, and P. hügelii.
B. Stridentes. Stridulating organ present; rostrum variable, but rarely (? never) as well developed as in A; pedate clasping processes absent; procephalic processes absent.
a. Antennulary sternum narrow below and bases of antennæ consequently approximated; bases of antennules hidden, in a view from above, partly by the antennulary sternum, partly by the antennæ; coxocerites imperfectly fused; antennulary flagella short.
Sub-genus Palinurus, Fabr., in pant.
α. Rostrum well developed, covering ophthalmic segment.
β. Rostrum reduced to a small spiniform tubercle; ophthalmic segment uncovered.
b. Antennulary sternum broad below, causing a wide separation of bases of antennæ; bases of antennules visible in a view from above; rostrum absent; ophthalmic segment uncovered; coxocerites perfectly fused; antennulary flagella long.
Sub-genus Panulirus, Grey, Senex, Pfeiffer, Includes P. interruptus, P. japonicus, P. penicillatus, P. ornatus, etc.
[Footnote] * Jasus, according to Lemprière, was the father of Palinurus.
There appear to be some interesting facts about the geographical distribution of the Palinuridæ, but my information is too scanty to allow of my doing more than glance at this part of the subject. I think there can be no doubt that all the longicorn species (Panulirus) agree in essential characters with P. interruptus, but I can unfortunately obtain no definite information as to two of the brevicorns, P. longimanus and P. frontalis, and am, therefore, unable to say whether they should come under Jasus or under Palinurus, as restricted by me.
Leaving these two species aside, it is worthy of remark that all the species of Jasus are confined to the Southern Hemisphere (Ethiopian and Australian Regions), and those of Palinurus, as restricted above, to the Northern Hemisphere (Palæarctic Region), while those of Panulirus occur in both hemispheres, and, as far as I can make out, in all the zoo-geographical regions.
In concluding this paper I wish to offer a few remarks on Mr. C. Chilton's views of the affinities of Palinurus, as set forth in his paper on Paranephrops in the last volume of these “Transactions.”*
Mr. Chilton says:—“The most important result arising from the examination of Paranephrops setosus is that its affinity to Palinurus now seems to be placed beyond doubt. Paranephrops and the Parastacidæ generally resemble the Palinuridæ in that they have no appendages upon the first abdominal segment; in this they differ from the crayfishes of the Northern Hemisphere, and from Homarus and Nephrops. The Palinuridæ and the Parastacidæ also agree in having hooked setæ, while in the Potamobiidæ and the lobsters the setæ are not hooked. Moreover the branchial formulæ of Palinurus and Paranephrops are almost identical. Taking the presence or absence of the first abdominal appendage as the basis of his classification, Professor Huxley placed the Palinuridæ and the Parastacidæ together as the Astyla, while the Potamobiidæ, Homaridæ, etc., together form the Stylophora. This classification is confirmed by the structure of the male reproductive organs in Paranephrops setosus, for these agree in every essential particular with those of Palinurus vulgaris as described by Brocchi.” After quoting Huxley's theory of the origin of Potamobiidæ and Parastacidæ from a common ancestor, distinguished as Protastacus, Chilton goes on to say “I have only to add that the Protastacus stock appears to have left Palinurus, which has lost the chelate limbs possessed by its ancestors, as its marine representative in the Southern Hemisphere.”
[Footnote] * On some Points of Difference between the English Crayfish (Astacus fluviatilis), and a New Zealand one (Paranephrops setosus), Trans. N.Z. Inst., vol. xv., p. 150.
On re-reading the paper of Professor Huxley's referred to in the foregoing quotation,* it appears to me that Mr. Chilton has been misled by having taken a useful and suggestive table of morphological characters for a natural classification. A very similar table is given in Huxley's paper on Ceratodus,† in which the Ichthyopsida are divided, according to the mode of attachment of the mandible, into Autostylica, Amphistylica, and Hyostylica, but it is certainly not intended to convey the impression that the Dipnoi are more nearly related to the Amphibia than to the Ganoids from the fact that the two former belong to the Autostylica, while Ganoids are hyostylic.
It is evident from even a casual examination, and is implied, although not stated explicitly, in the paragraph of Huxley's paper beginning:—“Let it be supposed,”‡ that the Potamobiidæ and the Parastacidæ are more nearly allied to one another than is either of them to any other family of Crustacea. To say that the Parastacidæ are more nearly allied to the Palinuridæ than to the Potamobiidæ, because they are astylic, is not more reasonable than to say that Anguis is more nearly allied to the Ophidia than to the Lacertidæ, because it is apodal. In the same way, to unite the Parastacidæ with Palinurus, and to separate them from the Potamobiidæ, on the ground that the two former possess a more primitive type of male reproductive apparatus than the latter, is equivalent to grouping Ganoids with Elasmobranchs rather than with Teleosts, because of the abnormal condition of the urino-genital organs in bony fishes.
What Professor Huxley shows clearly enough is that, while the Homarina are more closely allied to the Potamobiidæ than to the Parastacidæ, the reverse is the case with the Palinuridæ. He does not discuss the question of the origin of the latter family, nor say whether he considers its affinities to be, on the whole, Astacine or Homarine.¶ It is perhaps an open question whether true genetic affinities are more clearly indicated by the loss of the first abdominal appendage, and by the structure of the male organs and the setæ, or by the structure of the gills; but, at any rate, it must not be forgotten that Palinurus agrees with the Homarina in having the podobranchs completely divided into gill proper and epipodite and thus differs widely, in an important structural peculiarity, from the Parastacidæ.
However, this may be, it is certain that the interval separating Palinurus from the Parastacidæ is far wider not only than that which separates the latter from the Potamobiidæ, but than that existing between the Astacina and the Homarina. To hold otherwise is equivalent to denying the value of both anatomy and embryology as guides to genetic affinity.
[Footnote] * On the Classification and the Distribution of the Crayfishes, Proc. Zool. Soc., 1878, p. 752.
[Footnote] † Proc. Zool. Soc., 1876, p. 24.
[Footnote] ‡ Quoted by Chilton, loc. cit., p. 152.
[Footnote] ¶ Boas, in the paper referred to above, seems to derive Palinurellus from a Homarine ancestor.
I am greatly indebted to Mr. T. F. Cheeseman, F.L.S., of the Auckland Museum, for procuring me a spirit specimen of Palinurus vulgaris, and to Mr. W. A. Haswell, M.A., and Mr. T. W. Kirk, M.A., for much needed information respecting species of Palinurus in the Sydney and Wellington Museums of which I was unable to obtain either specimens or accurate descriptions.
I shall be very grateful to anyone who will furnish me with descriptions of the head in Palinurus longimanus, M.-E., P. frontalis, M.-E., Palinurellus, Boas, and Palinustus, Alph. M.-E., or, still better, who will procure me specimens of these forms.
Explanation of Plate XXV.
ant. 1, antennule.
ant. 1′, articular cavity for antennule.
ant. 2, antenna.
ant. 2′, articular cavity for antenna.
ant. st., antennulary sternum.
cl. pr., clasping processes.
e. cx. pl., external coxal plates.
epm. pl., epimeral plates.
fl., flap of movable part of stridulating organ.
g., groove on lateral surface of antennulary sternum, for tubercle (t.).
h., dorso-internal hinge of antenna.
i. cx. pl., internal coxal plates.
oph., ophthalmic segment.
p., ridged pad of movable part of stridulating organ.
p. cx. pl., posterior coxal plates.
prc. pr., procephalic processes.
r., ridge on antennulary sternum.
s., smooth surface on antennulary sternum forming fixed part of stridulating organ.
s. or. sp., supra-orbital spines.
t., guiding tubercle of movable portion of stridalating organ.
N.B.—All the figures are diagrammatic, unessential details (spines, tubercles, etc.) being omitted.
|Fig. 1.||Longitudinal vertical section of the head of Palinurus edwardsii.|
|2.||" " " " Palinurus vulgaris.|
|3.||" " " " Palinurus interruptus.|
|4.||Anterior part of head of Palinurus edwardsii seen from above.|
|5.||" " Palinurus vulgaris. "|
|6.||" " Palinurus interruptus. "|
|7.||Antennulary sternum of Palinurus vulgaris, from the left side.|
|8.||Proximal end of antenna of Palinurus vulgaris, from the inner side.|