Page images
PDF
EPUB

The two Ganoid scales (K1/316 ; Pl. 18, fig. 2 & 3; text-fig. 13) are of small size and rhombic in outline. Except for their size,

TEXT-FIG. 13.—A Ganoid scale (K1/316) from the Inter-trappean bed at Paharsingha.

x 104.

they agree very closely with the scales of Lepidosteus indicus Woodw. (Pl. 18, fig. 4; a loose scale from specimen 9366 of the type series). The Inter-trappean (possibly Lameta) cycloid scale from Dongargaon (K1/317; Pl. 18, fig. 10) belongs to the genus Clupea, and is of the type in which the transverse radii are complete.

This scale corresponds with similar scales described above from the Inter-trappean beds at Deothan and Kheri. It may here be noted that in the collection of the Geological Survey of India there are two specimens of Clupeoid scales with complete transverse radii (G 152, Pl. 18, fig. 8 & 9; text-fig. 1 b) collected from the infra-trappean bed at Dhamni, east of Warora, by F. Fedden in September 1874. This record shows that Clupea was found during the infra-trappean period also. The second cycloid scale (K1/318, Pl. 18, fig. 7; textfig. 14) from Pahársingha appears to be a ctenoid scale of the Percoid type in which the apical teeth are not well preserved or are not exposed. In its minute structure, it bears a close resemblance to the scales of the family Nandidae (vide supra, p. 281), a family whose close allies are at the present-day found as far apart as south-eastern Asia, Africa and South America, thereby showing its great antiquity. The fourth specimen of a fish roe (K1/319, Pl. 18, fig. 1), though of considerable interest from the point of view of its excellent state of preservation, is of little value for the determination of the fish to which it may have belonged.

[ocr errors][merged small]

TEXT-FIG. 14.-A Nandid scale (No. K1/318) from the Inter-trappean bed at Paharsingha.

X 30.

As to the fish bone from Tákli with 21 small teeth, it may be safe to presume, on Hislop's testimony, that it probably belonged to the Ganoid Lepidosteus.

In view of what is stated above it may now be possible to infer the relative ages of the various beds; but it should be borne in mind that the material is not sufficient to come to any

definite conclusion.

Mr. Crookshank tells me that doubts have recently been expressed by Matley on Hislop's determination of the age of the Dongargaon beds, and has given me the following note on this point :

· Hislop states definitely that the Dongargaon fish-remains were found in yellow limestone beneath the Trap. Hughes, who subsequently mapped the area, confirms this, and it is clear from his map that the fossil fish in question were found in beds underlying the Trap, and overlying the Gondwanas.

Matley? re-visited the area in 1919. He admitted that the fish beds underlay Trap, but he could not say whether they rested on Trap, or on older rocks, as he had not seen the base of the beds. It is clear, therefore, that he saw nothing in the field inconsistent with the views of Hislop and Hughes. Nevertheless he concluded that the Dongargaon beds were not Lametas, but were of Inter-trappean “In making this change he did not rely on the obvious field evidence of the superposition of one set of beds on another. He relied on the lithological dissimilarity of the Dongargaon beds and the Lametas of Jubbulpore, and on the fossil evidence obtained from the Dongargaon beds. These contain fossil fishes said to be early Tertiary, Bullinus princepii and a Melania similar to those in the Deccan trap Inter-trappeans, and the caudal vertebrae of Titanosaurus blanfordi Lyd. deker, all of which he considers younger than the Lameta fossil remains of Jubbulpore.

age.

1 Mem. Geol. Surv. Ind., XIII, Pt. I, (1877), pp. 89-90.
? Rec. Geol. Surv. Ind., LIII, Pt. 2, (1921), pp. 159-162.

As the Dongargaon beds were clearly infra-trappean, and were at the same time unlike the true Lametas, and seemed to be younger than them, Matley explained the occurrence as a case of infra-trappean beds of Inter-trappean age. While Matley may be right it is legitimate to point out that no other geologist has found infra-trappean beds of Inter-trappean age, that Saurian bones have never been found in undoubted Inter-trappean beds, that Melania and Bullinus have been found elsewhere extending down into the Cretaceous, and that the lithological evidence is of very little importance. On the whole Hislop's and Hughes' views seem more credible than Matley's, the more so as there is probably a big gap between the end of the Gondwana and the beginning of the Deccan trap period, a gap which is quite wide enough to hold fossil faunas differing as widely as those of the Jubbulpore Lametas and the Dongargaon beds.

Assuming then that Hislop is right the fossil fish from Dongargaon are probably older than those at Deothan and Kheri. If on the other hand Matley is right they should all be of approximately the same age.”

It may be noted, however, that Mr. Crookshank's material was collected on the borders of Hoshangabad and Betul, 100 or more miles distant from Dongargaon, and that it is impossible to correlate the two areas exactly as they are separated by a wide area of unfossiliferous Deccan trap.

Without entering into the geological merits of the controversy concerning the age of the beds at Dongargaon, whether infra- or inter-trappean, it is possible to say from a study of the fish-remains enumerated above that the beds at Dongargaon, associated with a relatively primitive fauna of two Ganoids, Clupea and one very primitive Serranid, must be regarded of an earlier age than the Inter-trappean beds at Deothan and Kheri, whence Mr. Crookshank obtained such a great variety of the modern teleostean fish. It is true that Ganoid scales and a piece of bone of the Lepidosteustype have been found in the Inter-trappean beds at Takli and Katerul but the scales are of a relatively much smaller size than those of L. indicus, and would thus seem to be of a later period when

1 Mr. S. R. Narayan Rao discovered the remains of Lepidosteus scales from the Inter-trappean bed at Kateru, near Rajamundry, and very kindly allowed me to see and identify the specimen for him. He proposes to publish a separate note on the specimen.

sea.

the large deep rivers of the area had presumably been replaced, through the outpouring of lava, by shallow streams.

It may be noted that when large fishes of the type of Eoserranus, Lepidosteus indicus and Pycnodus flourished in the Peninsula (the age of the infra-trappean beds at Dongargaon) the main river of the area may have been like a large deep river, but at the time when the Deothan and Kheri beds were laid down marshy conditions, with lakes of varying sizes in the neighbourhood of the mouth of a large river, had been established. The conditions then were probably similar to those now prevailing in the lower reaches of the Ganges. The presence of a considerable number of Clupeoid scales of fairly large size clearly indicates the deltaic nature of such a river not very far from the beds investigated by Hislop, Fedden, and Crookshank. Clupeoids are mostly marine fishes, but some ascend into large rivers for long distances. The occurrence of the Nandidae, the Pristolepidae, the Polyacanthidae and the Serranidae in the Inter-trappean beds at Deothan and Kheri also indicates the maritime nature of the area, for fishes of these families, even at the present day, are commonly met with in freshwater areas not very far removed from the The typically freshwater Cyprinoids are represented by a single scale of a primitive type, but it will be premature to draw any conclusions from this solitary record, as extensive collections of fish remains, especially of the type dealt with here, have not yet been made.

With regard to the ecological observations made above Mr. Crookshank has very kindly given me the following note from a geological point of view :

Arguing from the habits of the living fish most closely related to the fossil remains it seems that Deothan and Kheri, where the fossils were found, must have lain on the course of some deep sluggish stream not very far from the sea. The geological evidence on this point is of some interest.

“ All geologists who have worked in connection with the Deccan trap have remarked on the extraordinarily low relief of the pre-trappean land surface. Any large river draining this would certainly have been deep and sluggish. The first flows of trap would probably have disturbed such a river but might not have divert. ed its course greatly. The sediments associated with the fish scales are fine calcareous muds such as one would expect to find in ox-bows or other deep pools associated with large rivers of the present day. The only unusual feature is the presence of volcanic lapilli. Doubtless these were introduced by a shower of volcanic asb immediately preceding the lava flow which buried the pool.

“ It is idle to pretend that the details of the drainage and the coastline of wes. tern India in pre-trappean times are known, but it is supposed that a great river whose valley is now hidden by the Deccan trap must have flowed westward towards

K

the Arabian sea somewhere south of the present Narbadda valley. The exact location of such a river is not known, but its upper reaches are probably represented by the Gondwana valleys of the Satpuras and of Wardha district. The lower reaches are entirely hidden by the Deccan trap.

Marine sedimentaries of late Cretaceous age extend up the Narbadda valley to within 100 miles of Deothan. Nummulitic limestones of Parisian age occur 150 miles further west along the margin of the trap in Broach and Surat. The Deothan fish must have lived at some time between the end of the Cretaceous and the Parisian. It is a reasonable guess, therefore, that there was open sea somewhere between 100 and 250 miles to the west of Deot han in early Deccan trap times.

“The very meagre geological evidence therefore tends to support the conclusions concerning the habitat of the Deothan fossil fishes reached on purely zoological evidence."

In connection with the ecology of the Deothan and Kheri beds attention may be directed to an observation made by Sahni (1934, p. 136) regarding the flora of the Inter-trappean bed at Chhindwara, a place not very far distant from where Crookshank collected his material. In the course of his arguments regarding the Eocene age of these beds he stated :

“ One more point seems important although Rode himself does not refer to it. Nepadites is not only a genus very characteristic of the eocene period but, unless these palms have changed their mode of life since then, its occurrence in the northern part of the Deccan indicates the existence an estuary, during the early part of the Inter-trappean period in the proximity of Chhindwara. In their valu. able memoir on the London Clay flora recently published Reid and Chandler have shown that nearly all the fossil records of Nipa-like palms “lje approximately along the margins of the ancient Nummulitic, or Tethys, sea and its extensions'. So, most probably a north-flowing river debouched into the great Tethys sea or into an arm of that sea, not far to the north of Mr. Rode's home !"

It is thus seen that geological, palaeontological and faunistic evidences all point to the same conclusion, that is, an estuary must have existed in the neighbourhood of the infra- and Inter-trappean beds of the Central Provinces.

[blocks in formation]

Observations on Fish Scales. Bull.

Bur. Fisheries, Washington, XXXII,

pp. 119-174, pls. xxxii-xl. Geology of the Northern Slopes

of the Satpuras between the Morand and the

and the Sher Rivers. Mem. Geol. Suro. India. LXVI,

« PreviousContinue »