162 Records of the Indian Museum. [VoL. XVIII, 
isthmus. From Ciryhina and Crossochilus it is separated by the 
presence of an adhesive disk on the ventral surface of the head. 
The jaws are also less sharp and not so bony. ‘They have no trace 
of horny covering. Further, except in Cirrhina afghana from the 
Nushki desert, the characters of which are very divergent, the 
scales of the Indian species of Civrhina are always distinctly longer 
than broad. The teeth also are stouter than those of Dzscog- 
nathus. 
The relationship of Discognathus to Garra seems fairly clear. 
There can be no doubt that the former is the more primitive of 
the two, departing less from the normal Cyprinid type. This is 
borne out not only by the structure of the adult Discognathus 
TEXT-FIG. 3.—Pharyngeal teeth of Discognathus. 
a. D. adiscus. 
b. D. phryne. 
but also by the fact that the young Garra passes through a stage 
in which the structure of the head agrees with that of Dzscog- 
nathus. We figure a young specimen of G. nasutus 7°4 mm. long, 
illustrating this point, with one of about the same size of Psilorhyn- 
chus for comparision. It will be seen that its adhesive mental 
disk and also its branchial isthmus closely resemble those of 
D. adiscus (Rec. Ind. Mus., XVIII, pl. xi, fig. 1). We refrain 
from discussing this point further because Prof. D. R. Bhatta- 
charyya of Allahabad is at present engaged in a detailed study of 
the anatomy of the mouth-parts, etc., of these fish. 
The genetic relationship in the opposite direction between 
Discognathus on the one hand, and Crossochilus and Czirrhina on 
the other, though undoubtedly close, is not yet capable of full 
