164 



what behind the middle of a line, connecting end of snout 

 and base of caudal, separated by 14 15 scales from occiput. 

 Dorsal concave, 4^ spine strongly ossified and strongly denti- 

 culated, with its flexible portion somewhat shorter than head 

 and about i 1 ^ in height of body. Anal concave, its third spine 

 rather strong, the length of which equals the head without snout 

 or nearly so. Origin of ventrals separated by 4 to 4 ] / 2 scales from 

 9*h scale of lateral line. Length of ventrals about equal to that 

 of pectorals, equal to distance from hindborder of operculum 

 to nostril. Ventrals reaching anus, pectorals reaching on base 

 of ventrals. Caudal deeply forked, longer than head. Least 

 height of caudal peduncle less than twice in length of head, 

 somewhat less than 4 / 5 of length of caudal peduncle, surrounded 

 by 1 6 scales. Silvery, upper surface reddish brown, 7 or 8 rows 

 of small brown spots, corresponding to the rows of scales, the 

 upper and lower most incomplete. A large round blackish 

 blotch at the base of the caudal. Fins more or less darkish. 

 Length 230 mm. 



Nomen in dig. : Pojo (Samarinda); Puhing (Bo); Lawak, 

 Lalawak (Malay. Batavia); Wader (Javan.); Seren, Sakka 

 (Sundan.). 



Habitat: Sumatra (Lahat); ponds on Island Singkep !; Java 

 (Batavia, Perdana, Parongkalong, river Brantas near Surabaya !) ; 

 Borneo (river Kapuas : Sintang, river Mahakam : Samarinda ! 

 and river B6). - - Siam. 



We made this description after two specimens from the Island 

 Singkep and after one specimen from Samarinda, Borneo. We 

 are not sure that this is the Barbus armatus of C.V. and of 

 BLEEKER, as our specimens are higher. The differences in the 

 number of scales of the 1.1 is of no importance, as we would 

 find a higher number when counting the two small scales perfo- 

 rated by the 1.1, which runs upwards along the hindborder of 

 the operculum. Moreover BLEEKER gives for the 1.1 35 in his des- 

 cription of B. valenciennesi, which he himself afterwards con- 

 sidered to be the same as B. armatus. Principally on account 

 of the high and short caudal peduncle of our specimens, through 

 which characteristic they differ from B. repasson, we bring 

 them to B. armatus. Moreover there is a difference in the 

 relative length of eye and postorbital part of head. We find 

 only 2 barbels in the two specimens from Singkep, but we do not 



