536 



SCIENCE. 



[N. S. Vol. VII. No. 172. 



that system, the authors recognize the fact 

 that it is, as a rule, impracticable to pursue the 

 academical study of zoology on the broader 

 lines of natural history, and that students must 

 in general acquire their training through the 

 accurate and thorough study of a comparatively 

 small number of forms in the laboratory. Each 

 group (the class, as a rule) is accordingly intro- 

 duced by the thorough examination of a single 

 representative or ' example,' and the treatment 

 is such as to render the book as useful on the 

 laboratory table as in the study. From the 

 study of the ' example ' the student is led to a 

 brief account of the distinctive characters and 

 classification of the class, the systematic position 

 of the example, and finally to a more extended 

 comparative discussion of the general organiza- 

 tion, development and afiinities of the group as 

 a whole. We think the book would have been 

 much more useful had the authors followed the 

 example of Claus in giving some of the more 

 important families under the leading orders, 

 and by the use of smaller type this might have 

 been done without material increase in size. 

 The authors have shown good judgment in not 

 confining the descriptive part too closely to 

 anatomical detail, having added, wherever pos- 

 sible, accounts of embryological development 

 and larval metamorphoses, with something also 

 of habits, life-history and distribution. The 

 book is thus much increased in effectiveness and 

 is relieved in a measure of the dryness that has 

 often characterized zoological text-books. 



To criticise so bulky a work in detail would 

 hardly be possible within the limits of this re- 

 view.* The classification adopted will probably 

 he in its main outlines sufficiently acceptable to 

 most morphologists," though some of its details 

 are open to serious criticism. We are glad to see 

 the Porifera recognized as a phylum distinct 

 from the Ccelenterata, the Scyphoza separated 

 from both the Hydrozoa and the Actinozoa, and 

 the unspeakable ' Vermes ' consigned to a limbo 

 from which it is to be hoped they will never 

 emerge. On the other hand, the phylum Ar- 

 thropoda is retained with hardly an intimation 



* For a number of detailed criticisms pointing out 

 some important errors, see a review in Natural Science 

 for March, 1898, which has appeared since the present 

 review was written. 



of the opposing view, held by a considerable 

 number of morphologists, that this group falls 

 into at least two distinct phyla. The retention 

 of the group ' Gephrea ' in the old sense seems 

 little short of a blunder ; for it is generally ad- 

 mitted that the Sipunculacea are but remotely 

 connected with the Echiuroids, the latter being 

 degenerate annelids, while the former have 

 wholly distinct affinities. We think that many 

 morphologists will be disposed to question the 

 desirability of retaining Amphioxus among the 

 ' Vertebrata ;' for, although the discovery of the 

 nephridia and other recent investigations clearly 

 indicate its afiinities to the higher chordates, it 

 may well be doubted whether the gap between 

 Amphioxus and the tunicates is any greater than 

 that which still separates it from the lowest 

 craniate. 



The above are, however, minor criticisms. A 

 real and very obvious defect lies in the order 

 of treatment of the invertebrata. If the study 

 of morphology has shown anything regarding 

 the complicated relationships of the inverte- 

 brate phyla, it has shown that the ' Mollus- 

 coida ' and Echinodermata are but remotely 

 connected, either in structure or in develop- 

 ment, with the rotifers, annelids and mollusks, 

 while all three of these groups show well- 

 marked, if not direct, afiinities to the Plyathel- 

 minthes. The authors recognize the near rela- 

 tionship of the rotifers to the annelids and 

 mollusks by introducing the former group with 

 a description of the annelid trochophore (I. , p. 

 298); and this affinity is again recognized in a 

 genealogical tree nearly two hundred pages 

 farther on (p. 483). Yet the Trochelminthes 

 are separated from the Annulata by the Mollus- 

 coida and Echinodermata, producing a breach 

 of continuity which can be only misleading and 

 confusing to the student. It would seem from 

 every point of view preferable to place the lat- 

 ter two groups after the Annulata, Arthropoda 

 and Mollusca at the end of the first volume — 

 an arrangement which would allow a nearly 

 continuous treatment from the Platyhelminthes 

 up to these groups, and at the same time give 

 opportunity for more direct discussion of the 

 possible affinities of the echinoderms and some 

 of the ' moUuscoids ' to the lower Chordata. 

 Fortunately, such a transposition can readily be 



