GENERATION OF ANIMALS, III. v. 



clearly be seen oozing out from males of both groups 

 at the proper season. Also, the females have a 

 uterus ; but if the whole tribe of fishes really were 

 female, some of them being unproductive of young," 

 then not only those fishes which lay eggs but all the 

 others as well ought to have a uterus, though no 

 doubt different in form from that of the ones which 

 lay the eggs [hke female mules in the class of bushy- 

 tailed * animals].'' In fact, however, while some fish 

 have a uterus, others have seminal parts, and this 

 distinction is found in all species except two, the 

 erythrinus and the channa'^ : some have seminal 

 parts, others have an uterus. The puzzle which 

 makes people put forward this theorj- is easily solved 

 when we hear what the facts are. These people 

 allege — and here they are quite correct — that none 

 of the animals which copulate produces many young, 

 for of all the animals which generate out of themselves 

 either perfect animals or perfect eggs, none is so 

 proUfic as the oviparous fishes, the number of their 

 eggs of course being something enormous. But this 

 point they have overlooked : eggs of fishes do not 

 behave in precisely the same way as those of birds. 

 Birds, o\iparous quadrupeds, and any o^^parous 

 Selachians there may be,* produce a perfect egg, and 

 once it has left the parent it grows no further ; fish 

 on the other hand produce imperfect eggs, which do 

 grow after they have left the parent. Furthermore, 

 the same occurs in the case of the Cephalopods and 

 Crustacea ; and these creatiu-es can actually be seen 



the word arcKvoi, but they are meaningless and irrelevant 

 anywhere in the text. 



■* For erythrinus see note, 741 a 36 ; the channa is another 

 species of Serranus, probably S. scriba. 



' The fishing-frog ; but see 754 a 26, n. 



309 



