414 EMBRYOLOGY 



description of the author, evidently must come from the mesenchyma. 

 Examples of both genera were previously studied by Balfour (Com- 

 parative Emhrynlogy) and Metschnikoff, and were fou id to belike other 

 Echinoderras in regard to the structures in question. 



The studies of Bury (No. 8) on the early development of the entero- 

 hydroccele are, on the contrary, important. If, in spite of this, we have 

 not given them a correspondingly prominent place in our account of these 

 conditions, it is owing chiefly to the fact that Bury's statements on this 

 subject are almost directly oi^posed to those of other authors, and that, 

 furthermore, they neither trace the fundaments of those organs back to 

 the earliest stages, nor devote sufficient attention to the later history 

 of them. For these reasons, Bury's investigations, which, after all, 

 include only a few stages from the middle, do not seem to us to be suffi- 

 ciently conclusive upon the development of this important system of 

 organs, deviating so fundamentally as they do from all other desci'iptions. 



Bury assumes that all Echinoderm larvfe have not two enteroccelic 

 sacs, as had previously been believed, but two pairs of them, either 

 actually present or to be recognized from their fundaments. Thus the 

 larval body would exhibit an internal segmentation. These conditions 

 can be clearly seen in the larvje of Oj)hiurans and Echinoids, in which 

 the larger, anterior enterocceles lie at the side of the oesophagus, and the 

 smaller, jjosterior ones next to the stomach. The anterior and posterior 

 pairs have arisen by division of the primary enterocceles. The left 

 anterior enterocoelic sac opens to the exterior by means of the water 

 pore. The union of the latter with the enteroccElic sac does not corre- 

 spond to the subsequent stone canal, for the hydroccele does not arise 

 until later, and then either from the anterior or posterior enterocoele, 

 from which it is constricted off ; it is only secondarily that it unites with 

 the anterior enterocoele. Originally, then, only the body cavity commu- 

 nicates (by means of the dorsal pore) with the outer world. The hydro- 

 ccele does not unite with the body cavity, and thus with the outer world, 

 until later. Such conditions were also found by Ludwig in later stages 

 of Asterina (comp. pp. 408 and 436), and are retained throughout life 

 in the Crinoidea (comp. p. 447 and Fig. 224, p. 453). 



Bury's observations seem to coincide with those of Metschnikoff, who 

 also observed a division of the right enterocoelic vesicle in Opitiurans and 

 Kr.hinnids. but referred it to the formation of a right hydrocoele, which 

 subsequently degenerates. Thus Metschnikoff argues for a primitively 

 paired fundament of the hydrocoele, whereas Bury, like other authors, 

 derives it as an unpaired structure from one of the two enterocceles of 

 the left side. 



In other Echinoderm larvae Bury finds the internal segmentation less 

 sharply expressed. In the Asteroidea an anterior and posterior entero- 

 ccfile can still be distinguished ; they are, however, no longer separate, 

 but coalesce with each other. The Holothurioidea are said to have, 

 in addition to the two posterior enterocceles, a left anterior one, which, 



