46 M. E. Metsclmikoff on the 



would have said something about it, especially seeing that in 

 general he describes his subjects diffusely and circumstantially 

 (as, indeed, may be seen from the quotations just given). It is 

 evident that he has compared the free-swimming larva with a 

 young but on the whole fully-formed sponge, without con- 

 sidering that in this way he might very easily be led astray, 

 as in fact has happened. The most remarkable circumstance 

 is that, in several parts of his monograph, Hackel puts forward 

 his "directly inferred transformation " as an actually existing 

 fact, and not as a more or less probable hypothesis. Thus, 

 for example, he says at p. 160, " I give the name of syn- 

 cytium in the Calcispongia? to the whole mass of tissue which 

 is produced by the fusion of the cells of the exoderm of the 

 ciliated larva] 11 and at p. 216, "Each cell of the entoderm 

 stretches forth a long vibratile process &c." In these cases 

 he forgets entirely that he has never seen either the fusion or 

 the extension of the cilia*. Is this the philosophical " method 

 of scientific investigation " so celebrated by Hackel, and for 

 the non-employment of which the embryologists (ontogenists) 

 are so severely blamed by him?f (p. 472). 



I now pass to the question how far the developmental 

 history of the Calcispongia3 can be made available for the 

 comparison of the principal layers of these organisms with 

 those of other animals. In this respect Hackel has arrived 

 at a settled conclusion. He regards as one of the most im- 

 portant results of his work the statement that the two layers 

 of the sponge -body are homologous with the ectoderm and 

 entoderm of the Coelenterata. By the ectoderm (or exoderm) 

 he understands the so-called syncytium — that is, the skeleton- 

 forming outer layer of the sponge, whilst he characterizes the 

 flagellate epithelium as the entoderm. He gives this con- 

 clusion as the result of his investigations in developmental 

 history. Thus he says, for example : — " The relationship of 



* I must indicate the following - passage as exceedingly naif : — "The 

 structure of the flagellate cells of the exoderm in the Gastrula is exactly 

 similar to that of the flagellate cells of the entoderm in the fully developed 

 Calcispongia " (p. 335). And yet this striking agreement did not suffice 

 to raise any doubt in Hackel as to whether his a priori conception of the 

 germ-lamella represents the truth. 



t It is truly surprising to read how this method has been employed in 

 the representation of the Ascula, Protascus, Protosponyia, and other form- 

 stages invented by Hackel. Thus, for instance, it is said at p. 339 : — 

 "•Formerly I supposed that all Calcispongise in their earliest youth pass 

 through the characteristic form of the Protolynthus. But I must now 

 add as a limitation that in many cases the transition from the Ascula to 

 the Olynthus takes place not through the Protolynthus but through the 

 Protosponyia.' 1 '' All these conclusions are assumed without any single 

 fact ascertained by observation being cited in their support. 



