139 



him two fish (on the chance of his retaining at least one) 

 and coax them gently down his gullet. Leaving the back- 

 fin untrimmed may also mechanically prevent the rejection 

 of the fish. 



Most of the ground colubrine snakes feed freely in 

 captivity ; Ptyas mucosus especially is very eager at his 

 food * and will bolt frogs of very large size. I feed these 

 kinds of snakes once a week ; I put two frogs per snake 

 into the boxes, and let them divide the total number 

 amongst themselves. Pythons need only be fed about once 

 a month ; big frogs, chickens, or bandicoots are the .best 

 food for them. 



CHAPTER III.— The Museum. 



The collector will rarely be obliged to kill a snake for the 

 purpose of examination. He will generally have quite 

 enough snakes brought to him dead, and most snakes can 

 with very little trouble be identified while alive ; he will 

 only have to kill snakes when they are quite new to him or 

 are rare specimens worthy of preservation. The best way 

 to kill a snake is to poison it or to asphyxiate it by a 

 narcotic vapour ; interesting experiments may be made by 

 causing it to be bitten by venomous snakes, and an easy 

 and painless way of killing a small snake is to put it in a 

 sufficiently large bottle and pour in a few drops of chloroform. 



* My attention was once drawn by the cries of four weaver-birds 

 fPloceus hayaj who were in a cage in my verandah. On going out I 

 found that a large Dhaman had lifted the lid of his box, carelessly 

 left unfastened, and had got out ; but instead of making instant 

 use of his liberty, he could not resist the temptation offered by this 

 cage of birds ; he insinuated himself between the bars, and was 

 bolting the second bird when I came to the rescue ; he caught hold 

 of the third before I could secure him again. 



