90 CALAMOCEINUS DIOMED.E. 



When Semon says that we know nothing of the inner organization of 

 the Cystidea, he is perfectly right ; but to assume therefore that it is 

 illogical to compare them in any way with the different primordial types 

 from which both they and the other classes of Echinodernis may have de- 

 scended, is another question. Even comparisons between existing types 

 are far more logical than comparisons between imaginary primordial types 

 with stages such as he figures as the Pentactula stages, not one of which 

 is a true copy of what has been observed in nature. 



We may remind those who do not see the grounds upon which homol- 

 ogies based upon comparison of the calycinal system are sustained, that 

 a limited number of these hard parts are among the first structures to 

 appear in the pluteus, and that they make their appearance much earlier 

 than the structures upon which the opponents of these homologies base 

 their conclusions. It is perfectly true that the development of the hard 

 parts often becomes so far modified that it cannot be further followed. 

 Still, the study of the hard parts, modified as they may be by their devel- 

 opment in time and during growth, will always remain a fruitful field for 

 speculation, based upon a more solid basis than homologies which call for 

 ancestral forms of which we know nothing, or which represent later stages 

 of development th.an those from which the calycinal homologies are drawn. 

 While there are, and no doubt always will be, differences of opinion as 

 to the interpretation of disputed points, that is no argument against the 

 validity of the homologies as a whole. They at any rate have the great 

 advantage of being based on facts, while the homologies based upon the 

 assumption of a definite primitive ancestor representing one class are 

 interesting speculations. 



As yet no observations have been made carefully enough to show that 

 the water canal of Holothurians occupies the same position which it does 

 in other Echinoderm embrj'os. In Echini, Starfishes, Ophiurans, and Cri- 

 noids, it lies in the point of junction of the two extremities of the spiral 

 water system. But, according to Semon, that is not the case in Sjmapta, 

 and its position in the space between the third and fourth primary tenta- 

 cles would go far to show that the primary tentacles are radial, and the 

 secondary diverticula interradial ; but no such interpretation can be given 

 to Semon's figures of the position of the madreporic canal, as Bury and 

 Hamann have both shown. 



If, as Semon always insists, and with good reason, and as all echinolo- 



1 



