32 REPEODUCTION AND DEVELOPMENT OF THE CONGEH. 



flies of a given species, altliougli tliey breed only once, exhibit con- 

 siderable variation in size. 



The strongest evidence, however, in support of my conclusion is, 

 I think, the loss of the teeth and the atrophy of the bones, which 

 occurs during the ripening of the sexual organs. A conger, after 

 it had shed its milt or ova, would in all probability be entirely in- 

 capable of feeding itself ; without teeth it would be unable to hold 

 its prey, and without food it could not recover its former condition. 

 Cases of animals dying after breeding once are, of course, not un- 

 common in the animal, any more than in the vegetable kingdom. 

 Among insects it is rather the rule than the exception. But con- 

 fining ourselves to the Vertebrata, to which the conger belongs, there 

 are in that class cases of the phenomenon which are well established. 

 The common eel is known to go down to the sea in order to breed, 

 and the young elvers ascend the rivers in spring in countless multi- 

 tudes, but no adults have ever been known to return. It is very 

 probable that it will ultimately be found that all the members of 

 the eel family (Muraenidas) produce eggs only at the cost of their 

 own lives. Among the Cyclostomata my own investigations have 

 shown conclusively that the hag-fish, Myxine, does not die after 

 breeding. It breeds again and again, for I have taken, both on 

 hooks and in baited traps, numbers of females with ovaries showing 

 the collapsed follicles, from which the eggs had been recently dis- 

 charged. In fact, in an old female Myxine, the corpora lutea, i. e. 

 the old empty follicles in different stages of atrophy, belonging to 

 successively discharged crops of eggs, can always be seen in the 

 ovary. On the other hand, the river lamprey, Petromyzon planeri, 

 has been shown to die after breeding once. In this last case there 

 is a true metamorphosis from a sexually immature larva, the 

 Ammocoetes, which feeds and grows, to the sexually mature adult, 

 which feeds little or not at all, breeds, and then dies. 



With regard to the season of the year at which the spawning 

 of the conger takes place, my observations tend to show that it 

 is not confined to a very short period, but extends over several 

 months. It is impossible to decide how long a period would 

 have elapsed before each of the ripening females I have mentioned 

 shed its ova, if it had lived to do so. If we suppose that another 

 month was required to bring the ova to perfect maturity, then the 

 ova would have been shed in April, May, September, and October. 

 Similarly the female which was observed at Southport would, per- 

 haps, have spawned in July ; whence it may be provisionally in- 

 ferred that the female conger spawns in summer and autumn from 

 about April to October. But, on the other hand, I have had ripe 

 males in my possession from December to the end of August. If 



