460 EMBRYOLOGY 



life, which to-day characterizes the most of them. The 

 shape of the larva seems to have been produced indepen- 

 dently of this course of development. 



As regards the question what may have been the nature 

 of tlie bilateral progenitors of that radial ancestral form, 

 we are altogether in the dark. Ontogeny gives no answer 

 to this question, since, on the one hand, the larvse are much 

 changed, owing probably to phenomena of adaptation, and 

 since, on the other hand, no true relationships to other larval 

 forms — e.g., those of the worms — can be recognized. One 

 would most naturally compare the larvae of the Echinoder- 

 mata with those of the Turhellaria and Nemerteans or with 

 the Trochophore, but the difference in the distribution of the 

 ciliation and the absence of the apical plate make this diffi- 

 cult. Such larvae as that of Antedon, the Holothurian pupa, 

 and the vermiform larva of the Asteroidea (JoH. MC'LLEu) 

 recall segmented forms; but they may quite as well repre- 

 sent secondarily acquired stages of development. This is 

 especially difficult to determine in the case of the larva of 

 Antedon, for it is not impossible that larvse of the typical 

 form of the Echinoderm larvae make their appearance in the 

 development of the Crinoids, which is still little known. 

 The larva of Antedon even is modified, as the disappearance 

 of the blastopore show^s. Nevertheless, the resemblance 

 to the Holothurian pupa is striking, the latter certainly 

 representing a secondary stage of development. 



The Echinoderm larvae, as far as regards their internal 

 organization, resemble most closely such forms as the Anne- 

 lida, owing to the occurrence of coelomic sacs. We are 

 inclined to refer the formation of such a body cavity as 

 takes place in the Annelida to a like origin with that of the 

 Echinodermata, and accordingly to regard the mesoderm of 

 the Echinodermata and that of the Annelida as homologous 

 structures. Indeed, indications which point to relationships 

 of the Echinodermata with segmented forms are not lack- 

 ing. 



An internal segmentation would find expression in the body, if the con- 

 dition described by Bükt (No. 8)— the development of two pairs of 



