184 ITS DEATH. 



poor little fish now could only float at the surface ; 

 and as that could not last long, I resolved to attempt 

 a more extensive puncturing. I accordingly took it 

 into my fingers, and pierced the bladders here and 

 there in various parts of the body, and then returned 

 it to the water. At first I was afraid I had killed it 

 by keeping it out of water, though only for so brief a 

 period as a few minutes (certainly not more than two 

 or three); for it floated belly uppermost, and appeared 

 much exhausted, but gradually recovered. Though 

 it did not appear immediately that the bladders com- 

 municated with each other, yet they certainly did, for 

 the next day they had greatly diminished, and in a 

 few days they had entirely disappeared ; the skin had 

 healed and become smooth and healthy, and the little 

 creature was able to enjoy itself again. 



July %th. — I found my pet dead, on my return after 

 a week's absence from home ; it had apparently been 

 dead about three or four days ; so that it has lived in 

 captivity rather more than four weeks. 



The difficulty of delineating with accuracy objects 

 that can be defined only with microscopic powers 

 would hardly be imagined by those who have never 

 attempted it. In the case of this little fish, every 

 glance at its form or colours, in order to transmit 

 them to the paper, was taken through a triple pocket- 

 lens, which had to be exchanged for the pencil at 

 each stroke. The focus of this glass was about half 

 an inch, but the fish was swimming freely in a large 

 glass vase five inches wide ; so that it was only when 

 it spontaneously approached close to the side of the 

 transparent vessel, that I could get a sight. It was, 



