130 MR. J. S. 15UDGETT ON THE 



The ova of lleterotix are shed into the coelorn as in the Salmon. ITeterotis belongs 

 to the group Ostooglossidic, which has much the same distril)ution as the Dipnoi, 

 though it seems doubtful whether this points to an antiquity of the group equal to tliat 

 of the Dipnoi. Giinther, however, regards the Osteoglossidse as one of the earliest 

 types of Teleostean fishes. 



VI. — The Nestihg-habits of Sarcodaces ovok B1. 



In these same flooded grass-lands the eye is frequently caught by masses of whi t 

 foam floating on the surface of the water. On close inspection it is seen to be filled 

 with numerous transparent ova, about the same size as those of Ileterotis {2^ mm.). 

 Soon these eggs hatch, and on hatching make their way through the foam, in which 

 they are laid, down to the surface of the water, and there the young larvae hang 

 holding to the surface of the water by a large adhesive organ situated on the front of 

 the head (PL XI. fig. 8 & 9, c.o.). 



The natives assured me that these were the eggs of the Sannko, more scientifically 

 Sarcodaces odoe. On rearing some of these larvae, I was able to confirm this statement. 



Sarcodaces is one of the Characinidce, of which family examples occur in Africa and 

 South America. , 



VII. — The Nesting-habits of Hyfesopisus BEBEljace-p. 

 It was a curious fact that of the six species of Mormyridce which I obtained in 

 the Gambia, only one besides Gymnarchu^ was found breeding in the swamps. This 

 was Hyperopisns hehe Lacep. 



Test-fig. 22. 



Eggs from ucst, supposed to be that of Hi/jjcrajjisus bebe. 



Although I did not succeed in finding fertilized eggs of Hyperomsus this year, I 

 obtained a number of females full of ripe eggs. I am practically certain that these 

 ovarian eggs are identical with the eggs which I studied last year under the impression 

 that they were the eggs of Polypterus. 



