1898] SOME NEW BOOKS 275 



The class Gephyrea is another case in point. In spite of all the 

 evidence which has been brought forward to show that Sternasjns 

 is a highly modified Polychaete, probably nearly allied to the 

 Chlorhaeraidae, it is again placed with the Echiurids. Now the 

 chief, if not only, real differences between the Chaetopods and 

 Echiurids are the presence in the latter of a peculiar preoral pro- 

 boscis, and of posterior organs with ciliated funnels — these are both 

 absent in Sternaspis. The paired anterior processes described by 

 Sluiter do not contain the cerebral region of the perioesophageal 

 nerve ring, and appear to be comparable not to the proboscis of 

 EcJiiurus, but to the palps of Polychaetes. The brain of Sternaspis 

 is situated in a prostomium, in normal Polychaete fashion. Perhaps 

 more striking still, in such a fin de sihle zoology as this, is the 

 retention of the Echiurids and Sipunculids in one class. An attempt 

 to build up a ' type morphologique ' possessing the structure of both 

 these forms can only result in the production of a caricature of an 

 animal having the essential characters of neither. 



With regard to the new group Vernudea, we were at first inclined 

 to believe that the authors merely introduced the name as a convenient 

 but loose term, somewhat as we use the word Invertebrata. Such, 

 however, is not the case. Messrs Delage and Herouard claim that 

 the various animals they have brought together by a process of 

 epuration cles vers are really closely enough related to form a true 

 group. Discarding the Molluscoida and the Podaxonia, and indeed 

 paying little attention to the important work recently done on the 

 sul)division of the coelom in these forms, they make the artificial 

 assemblage of Gephyrea into a central group, having des ajinites 

 re'elles on the one hand with the Annelids, and on the other with 

 the Chordata. From the point of view of phylogeny, such a position 

 seems to us quite untenable. The fallacy of the method may be 

 illustrated by an exaggerated example. Suppose a systematist to be 

 classifying the mammalia ; what should we think of his system, if, 

 having first united the Cetacea with the Sirenia and Pinnipedia, he 

 then proceeded to argue that this new group has obvious and real 

 affinities on the one hand with the Carnivora, and on the other 

 possibly with the Ungulata ! Yet this, it seems to us, is just the 

 method pursued by the authors in the case of the Gephyrea and 

 Vermidea. The process of eimration, or sifting, may be very good for 

 the true Worms, which are not treated of in this volume ; but if we 

 pour all the rejected forms into one group, the Vermidea, the classifi- 

 cation is left in a worse condition than before, the confusion is worse 

 confounded. 



If we have insisted somewhat at length on what seem to us 

 defects in the taxonomy of this volume, it is because we believe no 

 greater mistake can be made than to consider classification as a 

 matter of secondary importance. If rightly understood, classification 

 is not only a valuable summary of the knowledge acquired, but also a 

 guide to farther progress. However clearly and accurately described, 

 facts lose half their value and interest if not presented in their proper 

 relations. When the Bryozoa were shown not to be Hydroids, when 

 the Brachiopods were shown not to be Molluscs, w^hen the Eotifers 

 were shown not to be Infusorian I'rotozoa, when the Sipunculids were 



