Page images

The brain-case is cracked in several places, but the supraoccipital, the parietal and the frontal can be made out.

Vertebral Column:-In considering the vertebral column of a Cyprinoid fish the chief interest lies in the modification of the anterior vertebrae.1 This fossil specimen is unique, so far as we are aware, in having a rib of the first vertebra distinct and well developed. The existence of a separate distinct first rib is a very primitive character, but even in living forms the first vertebra possesses a well developed transverse process and in Catla catla this is usually an elongated rib-like structure, reaching to about the middle of the transverse process of the second vertebra. The scaphium and a portion of the inter-ossicular ligament of the weberian apparatus are also seen in our specimen slightly above the origin of the rib of the first vertebra. All the trunk vertebrae are covered with skin and muscles and it is difficult to make out their exact structure. Those of the tail region are very clear and are exactly similar to the tail-vertebrae of such fishes as Labeo rohita, Barbus tor and Catla catla. The skeleton of the caudal fin is also similar to that of these species.

Integument.--We can find no trace of scales either detached or in situ, but the lateral line is quite clear in the anterior half of the body and can be traced along the caudal peduncle. It lies below the vertebral column and has a slight downward curvature anteriorly, while on the tail it seems to have been nearly straight and to have run parallel to and just below the vertebrae.

Affinities of Daunichthys:-om what has been said above it is abundantly clear that the new genus belongs to the family Cyprinidae and probably to the subfamily Cyprininae. The following combination of characters, however, distinguishes our new genus from all living and fossil genera of the family.

The anal fin is provided with 9 branched rays and does not extend to below the dorsal; the lateral line runs below the vertebral column and in the tail it was probably situated in the lower half of the body; the dorsal fin possesses 12 branched rays and 3 spines, the last spine is deeply grooved longitudinally on the right side and is strongly denticulated posteriorly and the body is entirely scaleless.

In its general facies Daunichthys gregorianus resembles certain species of the genus Barbus (s. l.) and in its up-turned mouth those

1 Hora, Journ. As. Soc. Bengal, (n. 8.) XVIII, pp. 1-4 (1922).

of Catla. Neither Barbus nor Catla, however, possesses more than six branched rays in the anal fin and both are provided with well developed scales.


A deeply grooved dorsal spine is characteristic of certain living scaleless North American genera of Cyprinoid fishes such as Meda and Plagopterus. In these this spine is. however, composed of two "the posterior received into a longitudinal groove of the anterior." In the only scaleless Cyprinid fish of the Oriental Region (Sawbwa resplendens 2 from the Southern Shan States) the dorsal spine is not grooved and is normal in every respect.

In the following table are given some of the chief characters in which the fossil Cyprinid genera of the Oriental region are distinguished from one another. Of the other fossil genera of the family,3 some are known from America and others from Europe. Most of these are either described from the remains of the pharyngeal bones and teeth or are characterized by the possession of a long dorsal fin without an osseus spine. In none of these in which the dorsal fins are preserved, are the rays grooved like those of Daunichthys.

The Geological Survey of India has recently received from a boring in the Tenasserim coalfield at Kawamapyin, Mergui, certain samples of clay very similar to that in which Daunichthys is preserved. They were obtained at a depth of 208 feet. They contain fish spines, which at first sight are very similar to the last bony ray of Daunichthys, but closer examination shows that they differ in not being grooved as well as minor characters. It is impossible to assign them to any genus or family with certainty, but they are probably from a dorsal fin of a Cyprinid.

1 Jordan and Evermann, Ball. U S. Nat. Mus., XLVII, Part I, p. 328 (1896).

2 Annandale, Rec. Ind. Mus., XIV. p. 48 (1918). For an up-to-date list of fossil italics in Jordan's Classification of California: 1923).

genera of the family Cyprinidae see names in Fishes, pp. 139-144 (Stanford University,

[blocks in formation]
[blocks in formation]

Ventrals commencing in advance of dorsal.

Anal with 5, 6 branched Anal with 5 branched


[blocks in formation]

Barbus (s. 1.).*

Pharyngeal teeth molariform with flats or concave crown; 3, 2, 1-1, 2, 3.

Dorsal with or without osseous ray, the last osseous ray may or may not be strongly denticulated; with 7-9 branched rays.

[blocks in formation]
[blocks in formation]

* We have followed Day's Fishes of India in our definitions of these genera, which still persist, but Barbus is broken up by some (not by all) recent ichthyologists into a number of smaller genera (see Weber and Beaufort's Fishes of the Indo-Australian Archipelago III, pp. 89-238 (1916). + Marck, Palæontographica, XXII, p. 411, pl. xxiii, fig. 2; pl. xxiv, fig. 2 (1876).

‡ Günther, Geol. Magazine (n. s.) III, p. 439, pl. xvi, figs. 2, 3a, 3b, 3c (1876).

ON A FOSSIL AMPULLARIID FROM POONCH, KASHMIR. BY B. PRASHAD, D.SC., Officiating Superintendent, Zoological Survey of India. (With Plate 15.)

COME opercula of a Gastropod mollusc recently sent me by Mr. D. N. Wadia, Assistant Superintendent, Geological Survey of India, consist of beautifully preserved and cleaned specimens of an extinct species of apple-snails of the genus Pachylabra Swainson. The occurrence of a species of this genus so far north in India is of great interest.

The genus Pachylabra 1 is, at the present day, represented by a number of species all over Peninsular India with the exception of Hyderabad, in Assam, Burma and the greater part of the Gangetic Plain. In the Gangetic Plain the range of distribution is limited by a line uniting Lucknow to Aligarh and then running south-west through Bharatpur and Ajmer in Central India to midway between Bombay and Ahmedabad in the Bombay Presidency. In spite of careful collecting at different times, I have failed to find specimens anywhere round Delhi, above Lucknow in the United Provinces, or in the Punjab and Kashmir. The fossil opercula brought back by Mr. Wadia from the Kashmir territory point to a greatly extended range of distribution of the family in former times. This is substantiated by the subfossil specimens of the opercula of an Ampullariid discovered in the Salt Range of the Punjab some years ago by the late E. Vredenburg. Unfortunately none of these specimens is now available for comparison and description.

The fossil opercula from Poonch, Kashmir, represent an undescribed species for which I propose the name Pachylabra prisca. The operculum of P. polita (Deshayes)—a species from Tonkin, Cambodia and Indo-China-was described by Houssay in detail and the operculum of P. prisca resembles it in essentials, but differs in certain well-marked characters. It is also different from that of the common Indian species P. globosa (Swainson), which I figure here for comparison.

1 See Preston, Fauna Brit. India Freshw. Mollusca, pp. 96-103 (1915), and Kobelt, Martini and Chemn. Conch. Cab. (ed. Küster and Kobelt) Ampullariidae, pp. 71-105, (1912-13).

2 Arch, Zool. Exper. Gen. (II ser.) II, p. 232, pl. xi, figs. 4, 10, 11 (1884),

The operculum of P. prisca is a concentric, patelliform, calcareous plate of somewhat pyriform shape. The nucleus, which is only to be made out on the external surface, lies near the middle of the plate, quite close to the inner margin, and is surrounded by 7 to 8 rings of growth, which may correspond to the age of the individual. Internal to the nuclear area is the narrow nearly straight region, which, as has been shown by Houssay, is secreted by the posterior part of the foot and consists of vertical plates lying one over the other. Externally this region in P. prisca is much narrower than the corresponding one in P. globosa. The area external to the nuclear region is secreted by the anterior part of the foot of the animal and is laid in horizontal layers; it is this region which increases in size with the age of the animal and shows the regions of growth. Internally the nuclear region is not distinguishable and its place is taken up by an ovoidal area for the attachment of the opercular muscle. The muscle is mainly attached to a depressed narrow area along the margin and in the centre there is a smooth raised region. In P. prisca as compared with P. globosa the ring of attachment is more excavated and extensive, while the central smooth region is more raised and convex. The opercula of P. prisca, compared with those of P. globosa, as seen in photographs of the side views of both species reproduced in Plate 15, figs. 2b, 3b, are very much thicker and consist of many layers. This appears to indicate that they belong to a species which lived in areas liable either to desiccation and a long dry season or to a long cold winter, in which it was necessary for the preservation of the species to close the mouth of the shell tightly.

Type-specimens:-In the collections of the Geological Survey of India. According to Mr. Wadia, the thick calcareous opercula were found in situ in soft, grey, micaceous sandstone interbedded with bright brick-red clay-shales similar to those characteristic of the Lower Siwalik (Chinji) series, about 1 miles south-west of Palandri, in the bed of the stream below Phalian village and some 200 yards north of the junction of the Palandri stream. The exact horizon of these beds is in course of investigation by Dr. Pilgrim.

The Gastropod remains were associated with relics of a rather varied vertebrate fauna found in the same vicinity within a few yards, including :—

Cheironeryx silistoensis,--molar and pre-molar.

Rhinoceros mandibular ramus with 1 premolar and 2 molars. incisor fragments.

[ocr errors]
« PreviousContinue »