72 Records of the Indian Museum. (Mor. 3< VDEr 
highly specialized structure consisting essentially of three parts—an 
anterior transverse band of soft tissue covered with minute tuber- 
cles, a central almost cartilaginous disk with asmooth surface, and 
a posterior and lateral free border of soft tuberculated integument. 
It is therefore a much more efficient organ of adhesion. In all the 
Indian forms with which I am acquainted the disk completely 
separates the antero-ventral margins of the opercula, but Gray 
and Hardwicke in their ‘‘ I/Justrations of Indian Zoology”’ figure these 
borders in D. gotyla as meeting behind the disk (Vol. I, pl. lxxxviii, 
fig. 3) and this also appears to be the case in certain African forms. 
In specimens I assign to D. gotyla the borders nearly meet. In the 
Indian forms the snout, either in the adult male or in both sexes, 
is tuberculate and often produced between or outside the nostrils. 
In the Indian species the number of rays in the dorsal fin and 
of scales in the lateral line as a rule affords little or no assistance 
in specific diagnosis. 
The fish of this group are mostly tropical, but a local race of 
D. lamta is found as far north as Palestine, while either D. jerdoni 
kangrae or a closely allied form inhabits mountain streams in the 
Aden hinterland. In Africa species are found in the Nile valley, 
in the great African lakes and in the eastern waters of Abyssinia. 
In Asia the range of the group extends from Palestine to Yunnan, 
Southern India and Borneo. It seems to have its headquarters in 
the hill country of Southern India and Assam, but the Assamese 
species or races have not been investigated since the time of 
McClelland. In streams at the base of the Nilgiris I found four 
distinct species. 
Discognathus lamta, Day'. 
1919. Discognathus lamta, Annandale, Rec. Ind. Mus. XVI p. 114, 
text-fig. 1 and p. 131, pl. i1, figs. 1, 1a. 
Dr. Chaudhuri has recently taken specimens of this species 
near Seringapatam in Mysore. ‘They differ slightly from North 
Indian specimens, but I have not sufficient material to establish 
their racial identity. 
Discognathus persicus (Berg.). 
‘““Garra persica, Berg, sp. n.? 
‘* Discognathus lamta (non Ham. Buch.) Nikolsky, Ann. Mus. 
Zool. St. Petersbourg, IV, 1899, p. 411 (No. 11706,11707). 
Do 7,, pase igies sear 
3314 
11707. River Bampur in Eastern Persia. N. Zarundy 1808, 
15-27. VII (6). 
to by W. T. Blanford in his ‘Zoology and Geology of Eastern Persia,” ir Per- 
sian Baluchistan some little distance inland from the Arabian Sea. 
“ Buchanan's Cyprinus (Garra) lamta was probably, from its habitat, identi- 
cal with McClelland’s Platycara nasuta (1838) rather than with the D. lamta of 
authors. It is, however, impossible to establish this with absolute certainty. 
2 Berg, Ann, Mus. Zool. St. Pétersbourg XVIII, p. Ixi (1913). é 
