144 SECTION A. GENERAL ZOOLOGY. 



the anterior portions are fused to form one cavity. In some 

 species of Asterias at any rate both the anterior portions open to 

 the exterior by pores although the right pore subsequently becomes 

 closed up. 



The hydrocoele is budded off from the left anterior cavity, and at 

 the same time a small closed sac appears in a similar position with 

 regard to the right anterior cavity. This little sac is the rudiment of 

 a rigJit hydrocoele, the fellow of the left one. This is shown (i) by 

 the fact that it is budded off from the right anterior cavity in a 

 corresponding position to that in which the left hydrocoele origi- 

 nates from the left anterior cavity and (2) above all by the fact that 

 it developes in some cases into a rosette shaped structure similar to 

 the left one. 



Bury has described the segmentation of the body-cavity vesicle 

 into anterior and posterior parts in Ophiurids and Echinids. This 

 I can confirm by sections through larvae of Echinus esculentus, as 

 also the assertion that the pore communicates with the left anterior 

 cavity, not into the hydrocoele directly. Strangly enough Bury 

 denies the existence of a right hydrocoele, though this has been seen 

 and figured in Ophiurid larvae by Johannes Miiller and Metschni- 

 koff. In a recent work on the development of Asterias pallida, 

 Seitaro Goto^ denies my interpretation of the sac which I have 

 above called the right hydrocoele. This work in my opinion calls 

 for severe comment, for Mr Goto has neither seen nor figured the 

 earlier stages of the organ in question. 



We have nearly reached the limits of discovery possible under 

 the old methods carried out with such splendid results by Johannes 

 .Miiller, Alex. Agassiz and many other distinguished zoologists. 

 If we are to penetrate any further into Echinoderm embryology, it 

 can only be by the use of the most refined modern methods, and 

 any attempt to jump to conclusions on the basis of insufficient 

 and indifferently preserved material can only hinder our progress. 



If the hydrocoele is really a paired structure, then the body 

 cavity will have three divisions on either side, and a fundamental 

 similarity to the Tornaria larva of Balanoglossus, as well as to 

 Actinotrocha, would exist. We should thus be induced to think 

 that three such widely divergent groups as the Enteropneusta (and 

 through them the Vertebrata), the Echinodermata and the Phoro- 

 nidea had remote pelagic ancestors of simple structure in common. 

 In this connexion it is interesting to notice that in Cephalodiscus 

 and in Actinotrocha the middle division of the body cavity grows 

 out into long arms comparable to the radial canals of Echinoderms. 

 Such a conclusion seems to me eminently reasonable. If the 

 various phyla of the animal kingdom really belong to the same 

 stock, it seems more probable that this stock was of a primitive 

 undifferentiated character, rather than that nature having elaborated 



^ " The Metamorphosis of Asterias pallida with special reference to the fate of the 

 Body Cavities," by S. Goto. Journ. Coll. Sci. Lnp. Univ. Tokyo, Vol. x. pt. iii. 



