HABITS, BREEDING, ETC. 199 



there are now hardly any kinds of fish to be found at 

 all in them. That this is not a natural condition 

 seems probable. At one place I remember a tank 

 about which the natives had a superstition, and in 

 this there were numerous fish, but they considered it 

 haunted, and never dared to destroy the fish, for 

 fear of offending the ghost or demon who lived 

 there. 



Thus river poisoning may in certain cases be 

 allowable, but only then under proper restrictions, 

 while as a general rule it ought to be strictly 

 forbidden. 



The taking of small fry in basket-work traps 

 should always be jealously guarded. It is a wasteful 

 and improvident system, and although it might 

 perhaps be permitted to a certain extent in certain 

 places, so long as provision was made for a fair 

 proportion of the little fish to make their way down 

 stream, yet in most cases it would seem that if the 

 practice could altogether be put a stop to there would 

 result a great improvement in the quantity and the 

 quality of the food supply. In deciding upon the 

 necessity of any restrictive measures in any par- 

 ticular district, the question would resolve itself into 

 this. Do the number of the fish destroyed in this 

 manner form an appreciable proportion of the total 

 supply of the district? If they do, measures are 

 decidedly necessary for their protection. 



In some parts of the country when the rains are 

 over, the streams are entirely diverted from their 



