ASEXUAL REPRODUCTION IN SAGARTIA 223 



shown in my groups to that given by Davenport's series. The 

 large number of recently divided specimens in the latter 's series 

 would make this error possible in a large number of cases. The 

 cases figured by Davenport ('03, figs, 3 to 7) are evidently in too 

 early a stage of regeneration to permit of any difficulty on this 

 point; but it would be encountered in dealing with slightly later 

 stages. If we assume twelve to be the number of stripes typical 

 of specimens resulting from ontogenetic development (see p. 225 

 for discussion of this point), the large number of specimens with 

 twelve stripes as compared with those having higher numbers 

 suggests the likelihood that there were included in Davenport's 

 collection an unusual proportion of undivided specimens. In 

 this connection the length of time the species has inhabited the 

 region concerned may be significant. Possibly the entrance of 

 S. luciae into a region and its establishment there is accomplished 

 by migration of larval forms, while the propagation of the 

 species in a region already occupied is much more largely 

 brought about by the asexual method. That asexual repro- 

 duction has been in progress for a shorter time in the group 

 studied by Davenport than in my lots is indicated by the rela- 

 tively small number of specimens with more than twelve stripes, 

 since I have shown that repeated regenerations tend toward 

 the production of an average condition with about seventeen 

 stripes. 



FORM RESULTING FROM ONTOGENETIC DEVELOPMENT 



From what has been said it is evident that no form as regards 

 number of siphonoglyphs, mesenteries, or orange stripes cer- 

 tainly distinguishes regenerated individuals from those that have 

 not undergone fission. The development of transforming em- 

 bryos must actually be followed to get unquestionable evidence 

 as to the form resulting from that process. I have made a 

 rather careful search of certain restricted localities at all seasons 

 of the year for such specimens with Httle success. The dates of 

 my special searches at Woods Hole were January 21, 1905; 

 July 12 and 18 and September 22, 1909; November 28, 1910; 



