[ 



FRESH-WATER FISHES OF SIAM, OR THAILAND 303 



sandy or gravelly bottoms. As a person approaches such streams 

 quietly so as not to disturb the fishes, he may see none, and on a number 

 of occasions, when no loaches were visible but subsequent events proved 

 them to be present in abundance, it seemed that they might be largely 

 nocturnal and thereby protected from some of their natural enemies. 

 The mountain people, who are very fond of eating loaches and other 

 small fishes, spend considerable time and labor on tiny streams that 

 would be regarded by the uninitiated as hopeless fishing grounds. 

 The method of procedure is to place a dam of stones and earth, some- 

 times supplemented by leaves, twigs, or branches, across a streamlet, 

 diverting its course into an old bed, or into a new bed that may be pre- 

 pared by removing stones and building up little banks of gravel, sand, 

 or earth. In the bed thus left dry, the gravel and sand are scooped up 

 with the hands, coconut shells, or other implements, and the fish are 

 disclosed. A short stretch of exposed streamlet bed, about a meter 

 wide and 10 to 15 meters long, has yielded scores of loaches, mostly 

 Noemacheilus, 



In a scholarly revision of the fishes of the genus Noemacheilus 

 recorded from Burma, Hora (1929a, p. 311) made the following 

 remarks, which apply also to Thailand : 



Among the Indian freshwater fishes there are few genera whose taxonomy 

 is involved in such great confusion as that of Nemachilus. This is in part due 

 to the fact that many of the species exhibit considerable individual variability. 

 Moreover, most of the specific standards, such as number of fin-rays, scale counts, 

 body proportions, colouration, etc., which are used in distinguishing species of 

 most of the other Cyprinoid genera, are of very little significance in the case 

 of Nemachilus. Fortunately most of the Burmese species are very peculiar and, 

 therefore, little diflBculty has been experienced in characterizing them. In 

 distinguishing closely allied species I have relied on the character of the lateral 

 line, the position of the anal opening with regard to the ventrals and the position 

 of the commencement of the dorsal with regard to its distance from the tip of 

 the snout and the base of the caudal. 



The species of Noemacheilus from Thailand may be differentiated as 

 follows : 



la. Body marked by numerous transverse bands or saddles. 

 2a. Cross bands black or dark brown. 



3a. A narrow cross band at base of caudal fin. 

 4a. Dark cross bands on body 6 to 9. 

 5a. Lateral line complete. 



6«. Cross bands 6 or 7, their width about equal to interspaces, 2 pre- 

 dorsal, 2 subdorsal, 2 or 3 postdorsal ; a small black spot at 



anterior base of dorsal fin ; head plain fowlerianus 



6&. Cross bands 7, wider than interspaces, 2 predoi-sal, 2 wholly or 

 partly subdorsal ; 3 postdorsal ; 2 black spots at base of dorsal 

 fin; upper surface of head with dark spots, a dark preorbital 

 blotch thai 



