DEVELOPMENT AND GROWTH 335 



post larval. From December onwards there are, in Fig. 15, still slight indications of three size groups 

 and it is possible that a small part of the population does not spawn in the third spring but lives to 

 spawn in the spring of the fourth year. The males of over 30 mm. in December carry ripe spermato- 

 phores, and in the next two months an increasing number of undoubted i-year group males attain 

 sexual maturity until, by March, practically all those over 25 mm. have reached the mature condition. 

 On the other hand, during these months all females, including those over 30 mm. in length, have 

 immature ovaries. Thus it is not possible to be certain whether these larger size groups are made up 

 of animals that have grown more rapidly and the males have become sexually mature rather early, 

 or whether they represent a small part of the population having a three-year cycle. 



Considering only those males which clearly belong to either the o-year group or i-year group the 

 time taken to reach various degrees of maturity can be determined. The period of adolescence, that is 

 the period during which the reproductive system is still developing (Bargmann's stages 1-5), lasts 



SHIlHBtB OCIOlSIl NOVEMBEK DECEMBEB MNUA«Y |EO«U»«» 



Fie i^ Monthly percentage length frequencies of E. triacantha. The number of specimens measured in each month is 

 shown at the top. The pecked lines divide the population into year groups. These groups have been used m preparmg the 

 growth curve, Fig. 16. 



from April to about the following November. During this and the next month an increasing number 

 of males are found with imperfect spermatophores in the spermatophore sacs or ejaculatory ducts, 

 and using Bargmann's criteria, these can be considered as having entered the adult phase. Durmg 

 January and February a large proportion are found with mature spermatophores m both the spermato- 

 phore sacs and the ejaculatory ducts, and by March nearly all have reached this fully adu t condition 

 and are presumably ready for pairing. It seems probable, however, that pairing is generally delayed, 

 because, although from January onwards a few females are found carrying spermatophores, it is not 



until June that the majority do so. . , . . u u 



A complete analysis of egg sizes has not been made, but it is apparent from those that have been 

 measured that eggs of over 0-3 mm. in diameter begin to appear in the ovaries about May, and that 

 the number of females with such eggs and the number of eggs of this size present in each ovary 

 increases until, in August very nearly all the adult females have ovaries with at least some large eggs. 

 The two females, mentioned above, with the appearance of having spawned, were taken in late 



^'Thrgrowth rate during the two years of the life cycle can best be seen by separating the year groups 



