BREEDING-HABITS OF SOME WEST-AFEICAN FISHES. 119 



of the basal lobe is in this young larva much longer than the dorsal or preaxial border ; 

 while the fin-rays become successively longer in passing from the preaxial to the post- 

 axial border. It follows that the shape of the fin is triangular, the apex being at the 

 extremity of the postaxial border (PL XI. fig. 1). 



Though the spot where this larva was caught was carefully searched, I did not 

 succeed in capturing another. 



Later on the same day, the 19th of August, I had another female P. lavradei, which 

 must have finished laying its eggs some weeks before. 



On the 21st of August, in my own fish-trap at the mouth of the small creek which 

 led from the river to the swamp, I found a female P. lapradei vihich had finished laying 

 its eggs, and it looked as though, in this case, it had spawned in the river or else at 

 the mouth of the creek. I am inclined to believe, however, that it had temporarily 

 returned to the river side of the trap after depositing its eggs in the swamp. 



On this same day I had a Polyjjterus senegalus, still crammed with eggs, but not one 

 free egg in the body-cavity. 



During the last week in August and the first in September, I killed fifteen of my 

 captive females ; but in no case could I attempt artificial fertilization, as the ova would 

 not come away from the ovaries, and in more than one case there were signs of 

 degeneration. 



On the 5th of September I left for England, leaving five pairs of P. senegalus in 

 charge of a native, who was to preserve eggs for me if any should be laid. 



Though I have little success in this direction to report, I have thought it well to 

 put on record the difficulties which I encountered in the search for the eggs of 

 Polijiderus in order that any future investigator who may attempt to obtain develop- 

 mental material of this fish may in being forewarned be also forearmed. 



The main difficulties in obtaining the eggs seem to lie in the fact that 

 PolypteruH probably makes no nest, and certainly lays but few eggs at a time, these 

 being scattered, probably broadcast, throughout the thick vegetation of the flooded 

 grass-lands. The eggs are minute, and therefore the chances of finding them in a 

 state of nature are small in the extreme. 



III.— The Habits and Life-history of Protopterus. 

 a. Nesting -hahits. 

 Although the development of Polyfterus had been the chief aim and object of my 

 second journey to the Gambia, I was also very anxious to obtain a series of the eggs and 

 embryos of Protopterus. When I was on the Gambia the previous year, I had brought 

 me a number of eggs of Protopterus, but I suspected that the way in which the native 

 told me that they had been laid was quite abnormal or altogether untrue. 



I had expected, in wading about the swamps, to come across deep holes in the ground 

 similar to the nests of Leindosiren, which I had become familiar with when in the 



