1920.]| N. ANNANDALE & S. L. Hora: The Fish of Seistan. 165 
The name Garra was used in a double sense by Bleeker,! as 
that of a genus, in which he included species of both the genera 
recognized by us, and also (sensu stricto) as that of a subgenus, from 
which he excluded the species accepted by us as the type-species 
of Discognathus. Buchanan was not acquainted with any form 
belonging to this latter group, which is not found in the territory 
explored by him. 
Taking all these facts into consideration, we accept Jordan 
and Evermann’s ® finding that Garra, Ham. Buch. is the correct 
generic name of the species assigned by Day to Discognathus, but 
much of the synonymy in the Fishes of India under the latter 
name is incorrect. 
Discognathus adiscus, Annandale. 
1919. Discognathus adiscus, Annandale, Rec. Ind. Mus., XVIII, p. 68, 
Dex ne 2) ple xi, fie at 
The formula of the pharyngeal dentition is capable of two 
interpretations. Omitting the minute free teeth (found not only 
inthis genus but also in Garra and Cirrhina) it may be read 
either 5°4°22°4°5 or 5°3°33°3°5. The scales* are subcircular, 
but slightly longer than broad, sinuate at the base and rounded 
distally. Some have a pair of lateral processes as shown in fig. I. 
They have nine or ten radii, which proceed obliquely forwards. 
About half of these radii arise near the nucleus, which is 
situated at about a sixth of the distance between the base and 
the distal margin ; the others are much shorter and arise nearer 
the distal margin ; long and short radii alternate, but not always. 
There are about Io to 12 transverse striae near the base in fully 
formed scales. Dentritic blotches and minute round dots of 
pigment are scattered on the distal part. The measurements of a 
large lateral scale are as follows:—length 3 mm., breadth 2°7 mm., 
distance of nucleus 0°45 mm. 
We give measurements, etc. of a series of specimens from 
Seistan. 
In many respects this is the most primitive species of the 
genus known and the most closely related to Cirrhina. It is 
interesting to observe that the young of Garra nasuta*, one of the 
most highly specialized member of its genus, passes through a 
stage at which the mental disk is very similar to that of D. adiscus. 
D. adiscus lives in still or sluggish water and feeds on algae 
ona muddy bottom. Itis markedly gregarious and may some- 
times be seen on the surface of water-channels in the evening in 
shoals. In the plain of Seistan D. adiscus and the young of 
Schizothorax zarudnyi are almost equally abundant in pools left in 
1 Bleeker, Atl. Jchth., III, p. 24 (1863). 
2 Jordan and Evermann, The Genera uf Fishes, p. 115 (1917). 
> This statement is not in verbal agreement with that of Cockerell, Bz. 
Bur, Fish. (Washington), XXXII: 1912); but the question is one of degree. 
* See Annandale, Rec. Jnd. Mus. XVI, p. 132, pl. ii, fig. 2. 
