February 26, 1903] 



NA TURE 



589 



the few pages that are dignified by the name are merely 

 a mockery to anyone who is not one of the authors. 



However, even with these easily remedied defects, the 

 book is a good and useful one which can be recom- 

 mended to the student as one to be added to his library. 



OUR BOOK SHELF. 



Studies in the Cartesian Philosophy. By Norman 

 Smith, M.A. Pp. xiv -t- 276. (London : Macmillan 

 and Co., Ltd ; New York : The Macmillan Com- 

 pany, 1902.) Price $s. net. 

 The title of Mr. Smith's book conveys an adequate idea 

 of its scope. The author indicates the lines of his treat- 

 ment thus : — "In Descartes' system, as we have tried to 

 show, there are three fundamental tenets, viz. the doctrine 

 of representative perception, a very peculiar form of 

 rationalism, and the conception of spirit as an active 

 creative agency" (p. 115). The main portion of the 

 book (pp. 1-115) is devoted to Descartes, with appendices 

 on " Arnauld's denial of the doctrine of representative 

 perception " and on Descartes' views of perception, 

 time and consciousness (pp. 11 5- 136). The rest of 

 the book deals with Cartesian principles in Spinoza, 

 Leibniz and Locke, with Hume's criticism and "the 

 transition to Kant." 



The author's treatment is lucid throughout ; the main 

 lines of criticism are stated clearly, and, on the whole, 

 adequately. This is especially true of the chapters on 

 Descartes and Locke, where the author has allowed him- 

 self to treat the subject at greater length. As to Descartes, 

 the author says, "his philosophy of nature I have 

 reserved for further consideration, and in this present 

 volume limit myself, as far as possible, to his meta- 

 physics " (preface, p. vi). His reason for thus dividing 

 the subject is that Descartes' "metaphysical teaching is 

 perverted by principles wholly at variance with his own 

 positive scientific views " (preface, pi.). This point is 

 specially brought out in contrasting Descartes' physical 

 and metaphysical views of motion (pp. 70-71). 



With regard to the salient features of Descartes' 

 teaching, Mr. Smith considers that the cogito ergo 

 sum, so far from being "the really ultimate element in 

 his system," is "simply one consequence of the doctrine 

 of representative perception which is itself a consequence 

 of his dualistic starting point " (p. 14). The importance 

 of " method " as " not merely an instrument for construct- 

 ing knowledge" (p. 23) and the relation of method to 

 Descartes' view of intelligence is well brought out ; the 

 same may be said of the author's treatment of the de- 

 ductive mathematical method and its fallacies. In "the 

 proof of the existence of God," Mr. Smith thinks 

 " Descartes' scholasticism came to a height." He rightly 

 treats some of the Cartesian arguments as purely 

 "official" (p. 64). But we cannot say that he is alto- 

 gether clear upon the relation of the system as a whole 

 to God ; it may be disputed whether Descartes ever 

 meant by God "the all-comprehensive absolute reality." 

 Certainly we can recognise the universality of the criterion 

 without identifying the system with its maker. The 

 author's argument is scattered and somewhat divided 

 between what Descartes really meant and what he really 

 said, 



The discussion on Locke is an excellent chapter ; the 

 treatment of "substance" and "the unknown" may be 

 specially mentioned (v. p. 195). The treatment of Spinoza 

 and Leibniz, though suggestive, is too brief. As the value 

 of this book lies not so much in its originality as in the 

 accurate exposition of certain lines of thought which 

 have dominated modern philosophy, the author should 

 not have allowed the recent works on these philosophers 

 to cramp his own treatment. The section on Berkeley 



NO. 1739, VOL. 67] 



suggests the same criticism. Yet the many good quali- 

 ties of the book should recommend it to all students of 

 philosophy. It is adequately furnished with references 

 and has a good index. G. S. B. 



Die progressive Reduktion der Variabilitdt und ihre 

 Beziehungen zum Ausstcrben und zur Entstehung der 

 Arten. By Daniel Rosa, Professor of Zoology in the 

 University of Modena. Authorised Translation from 

 the Italian by Dr. H. Bosshard. Pp. 106. (Jena : Gustav 

 Fischer, 1903, published 1902.) Price 2.50 marks. 



Prof. D. Rosa begins his interesting essay by say- 

 ing that cuttlefishes might envy the obscurity which 

 multitudinous evolutionist-pamphlets— likened to "ink- 

 squirts" — have given to the troubled waters surrounding 

 the rock of the theory of descent ; but this somewhat 

 cynical outlook has happily not hindered him from pub- 

 lishing (in 1899) the booklet before us or from having it 

 translated into German by Dr. H. Bosshard. We have 

 both versions of the essay, and, so far as we can judge, 

 the translation is exceedingly well done. 



In his first chapter, Rosa pictures organic evolution as 

 a long-drawn-out " substitution process," in the course 

 of which many groups-, having reached their acme, give 

 place to others springing from a lower level of the 

 phylogenetic stem and retaining a capacity for abundant 

 and far-reaching variation. As a group becomes more 

 perfect, it tends to nirvana ; its variations are reduced in 

 number, or, in any case, in range ; and the extinction of 

 "lost races" like Graptolites, Trilobites, Ammonoids, 

 Pterodactyls, &c, is causally associated with a progres- 

 sive reduction of variation. It has to be admitted, how- 

 ever, that we do not really know much about the scope 

 of variation in the last days of lost races. 



In the second chapter, Rosa inquires whether the pro- 

 gressive reduction of specific variations is wholly due to 

 natural selection or in part also to internal organismal 

 conditions. He emphasises two points: — (1) that an 

 organ which disappears in the course of evolution never 

 reappears along that line of descent, that an organ 

 which has become retrogressive never reacquires the 

 capacity of progress ; and (2) that in many cases, there 

 is a constancy or fixity in the numerical relations of 

 parts, eg. segments, limbs and digits, from which the 

 type seems quite unable to free itself. These two sets of 

 facts point to a progre*ive reduction of variability, 

 especially in types towards the ends of the phyletic 

 branches. This theory is corroborated by detailed re- 

 ference to the limitations which structural and functional 

 differentiation seems to impose upon the variability of 

 tissues and cells. Evolution is dominated by the "law of 

 progressively reduced variability." 



The third chapter is less of a unity, for the author has 

 been impelled to speak briefly " de omni re scibili et 

 de quibusdam aliis." Rosa attaches little importance to 

 individual fluctuations ; he relies upon general changes 

 or mutations of the idioplasm occurring throughout the 

 species. He has done useful service in indicating the 

 tendency to reduction of variability in highly evolved 

 types ; his essay is very interesting and suggestive, 

 pleasantly free from dogmatism or verbal polemics ; but 

 we must wait for more detailed data, and admit that 

 " Thatsachen, nicht Ansichten, entscheiden." &.<r<-:-*: 



J. A. T. 



Steel Ships : their Construction and Maintenanse. A 



Manual for Shipbuilders, Ship Superintendents, 



Students and Marine Engineers. By Thomas 



Watson. Pp. xiv + 290. (London : Charles 



Griffin and Co., Ltd., 1901.) Price iSs. net. 



The title of this work led us to hope that a long-felt 



want had at length been supplied ; but we regret to 



have to say that on reading it we were disappointed. 



Mr. Watson does not appear to have the scientific 



