Spawfiing-habits of the Pilot Fish. 117 



projection appeared, which in the normal condition wotild 

 penetrate the yolk-mass to a slight extent. It doubtless 

 has some function in the mechanism of fertilisation, though 

 no canal connecting it with the microphyle was detected 

 in the perivitelline substance. 



The mode of origin of the filament of the egg is different 

 from what is found in some other filamentous eggs of Teleosts. 

 Thus, in the egg of the South African species of Hemi- 

 rhamphus and Atherina, I have noticed that, in the immature 

 and even fairly small ovarian eggs, the filaments occur as 

 irregular streaks on the surface of the zona radiata, but, in 

 those of Nuucrates, the filament is already free, and serves 

 to attach the egg to the wall of the ovary. A number of 

 such filaments are inserted at one spot, so that the ovarian 

 eggs are often grouped in grape-like clusters, or the fila- 

 ments become twisted on each other to form a rope-like 

 structure, round which the eggs are grouped. 



The presence of filaments on fish-eggs, as a rule, has been 

 found to be associated with the fact that they are anchored 

 to each other or to foreign objects, floating or lying at the 

 bottom of the sea. Thus the eggs of Hemirhamphus, Belone, 

 and Exocoetus have been found attached to each other and 

 to sea- weed in this way, though Scombresox, another member 

 of the same family, is said to have pelagic eggs provided 

 with filaments. Another family, the Atherinidas, all the 

 members of which have eggs provided with filaments, so far 

 as is known, have demersal attached eggs. 



There is thus a reasonable presumption that the possession 

 of filaments indicates that the eggs are, ultimately at least, 

 attached to some object fixed or floating in the sea, and, if 

 we suppose that the filamentous eggs of the pilot fish are 

 attached to the shark, with which the fish is so intimately 

 associated, it may explain some peculiarities in its habits 

 which have received a variety of explanations. These are 

 not entirely convincing, partly on account of this variety, 

 but chiefly on account of conflicting facts or of lack of 

 confirmation. 



Thus the explanation that the pilot feeds on the fragments 

 of the food of the shark is not in accordance with the fact 

 that small fish have been found in its stomach. The same 

 objection applies to another conjecture that it feeds on the 

 excrements of the shark, and still another that it feeds on 

 the parasites on the skin of the shark. An explanation of a 

 different nature, that the pilot keeps close to the larger fish 

 for the purpose of protection from its enemies, is a more 

 plausible one, but is somewhat strained when its very close 



