460 



NA TURE 



[September 17, 1896 



/ 



Elementary PrMtical Chciiiistry and Qualitative Analysis. 



By Frank Clowes, D.Sc. Lond., and J. Bernard 



Coleman, A.R.C.Sc. Pp. xvi + 224. (London : J. 



and A. Churchill, 1896.) 

 This book, .which is founded on Prof. Clowes' larycr 

 "Practical Chemistry and Qualitative Analysis" is intended 

 for the use of general students and of technical students 

 in schools and colleges who are desirous of acquiring a 

 general elementary knowledge of chemistry, and who 

 propose to acquire this knowledge in the only true way, 

 viz. by themselves performing experiments in a laboratory. 

 For such students the book furnishes an admirable guide. 

 The first eighty pages contain excellent instructions as 

 to the preparation and use of apparatus, the methods of 

 carrying out ordinary chemical operations, and the modes 

 of demonstrating the properties of common gases and 

 liquids. The remainder of the book is occupied with a 

 course of qualitative analysis, which treats first, at con- 

 siderable length, of the reactions for metals and for acid- 

 radicles, and then of the actual analysis of simple and 

 complex substances. There is, further, an appendix of 

 useful tables and a good index. 



The hand of the experienced and careful teacher is 

 manifest throughout. The importance attached to 

 cleanliness, neatness, and system in the rules given for 

 working ; the directions for the verification of the state- 

 ments made, and for the keeping of the student's note- 

 book ; the precautions indicated as necessary for success 

 in certain expermients, the careful attention to detail, 

 and the emphasis given just where it is needed, show 

 that the authors have knowledge not only of chemistry, 

 but also of the "general and technical student," who, if 

 he will observe the instructions, and. work fairly through 

 the book, cannot fail to acquire a real knowledge of his 

 subject. 



For boys and girls at school, we ourselves should 

 recommend a course on somewhat different lines, starting, 

 for instance, with air rather than with oxygen, following 

 generally a historical sequence, taking the various 

 chemical operations not en bloc, but as required in the 

 course, and relegating qualitative analysis to a com- 

 paratively subordinate place. But taking things as they 

 are, and accepting as a fact the existing requirements of 

 various public examining bodies, this little work should 

 prove widely useful as a carefully-arranged, clear, and 

 accurate text-book. 



Entomological Notes for the Young Collector. By 

 William .A.. Morley. Pp. viii -f 129. (London: Elliot 

 Stock, 1896.) 



This is a little book of the most popular kind, written 

 with the intention of rendering the collecting of butter- 

 flies and moths easy to the youngest of beginners. It is 

 illustrated by eight pages of figures representing ap- 

 paratus, setting, &c., and the text is divided into twelve 

 chapters, corresponding to the months of the year, each 

 including a lesson on apparatus, collecting, rearing, iS;c., 

 and a list of some of the principal Lepidoptcra which 

 appear in each month. The book may be useful to 

 those for whom it is intended ; and we congratulate the 

 author on his good judgment in advising his readers to 

 learn the Latin names, and to forget the English. Here 

 and there a little revision would be useful ; thus Entom- 

 ology is defined as " that branch of natural history 

 which bears special reference to four-winged insects known 

 as butterflies and moths " (no other insects being even 

 mentioned in the book) ; Cambridgeshire is the only 

 locality given for Papilio macliaon and Vanessa antiopa ; 

 Lycccna artaxcr.xcs is said to be "generally distributed 

 in England " ; of L. corydon, we read " On chalk cliffs, 

 common " ; and moths which come to sugar are said, 

 as a rule, not to come to light. W. F. K. 



NO. 1403, VOL. 54] 



LETTERS TO THE EDITOR. 



[ The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can he underta/:e 

 to return, or to correspond with the writers of, rejected 

 ntamtscripts intended for this or any other part of NATURE. 

 No notice is taken of anonymous coiitmuniiations.'\ 



The Utility of Specific Characters. 



Ir is dangerous for a mere mortal to take part in the strife of 

 the gods, or for " gyrating"' mathematicians to join battle with 

 biologists. But as the enthusiasm of Prof. Wekion for his 

 subject has so largely perturbed my normal gyrations, that I 

 have devoted many months to the statistical theor)' of evolution, 

 perhaps I may be permitted a word or two on the subject of the 

 present controversy. 



To demonstrate that natural selection, whether secular or 

 periodic, is actually taking place in any species, and to measure 

 its amount, is in the present state of our knowledge one of the 

 grandest pieces of work that could be done. It would achieve 

 for the Darwinian theory what Hertz achieved for the Maxwellian 

 theory of light. At present no one has gone further in the 

 direction of this demonstration than Prof Weldon, and I 

 am inclined to think with Mr. Thiselton-Dyer— and I take it in 

 opposition to Prof. Lankester — that such a demonstration 

 can only be achieved by the statistical method. If, however, we 

 are to obtain a really solid result of that method, then the 

 mathematical theory, and the logic used, must both be beyond 

 suspicion. Now in any demonstration of the existence of natural 

 selection two points must be borne in mind : 



(a) A selective death-rate must be actually demonstrated. 

 This is a problem for fine statistical theory. 



(h) The correlation between organ and death-rate must be 

 shown in itself to be not fortuitous. The character must have 

 been selected because it is useful. This I take to be Prof. 

 Lankester's point. 



I propose to say a few words as to both (a) and {b). 

 It appears to me that both Prof. Lankester and Mr. Thiselton- 

 Dyer allow that a selective death-rate has been established in 

 the report on Carcinas imenas of 1894-5. This view I take to 

 be entirely erroneous, and I so expressed myself to Prof. Weldon 

 and several members of the Committee before the Report was 

 published. What Prof. Weldon demonstrates is \.V\s,—that if 

 crabs chanced to g)-ow in a particular manner, then there would 

 be a relation between death-rate and the size of a certain frontal 

 ratio. What is quite certain is that at the time the Report was 

 published, nobody knew how crabs grew ; and I very much doubt 

 whether I'rof Weldon, after his laborious two years' study of 

 the growth of crabs, would now uphold the hypotheses he then 

 adopted, e.g. : 



(i.) That size could be taken as a safe measure of age. 

 (ii. ) That young crabs of the same frontal ratio do not 

 "scatter " as they grow older. 



(iii.) That the amount of growth of crabs of any given frontal 

 ratio is entirely independent of that ratio. 



Yet if these — to me very improbable — hypotheses be not 

 accepted, the supposed demonstration of a selective death-rate in 

 Carcinas nucnas falls completely to the ground. The very 

 hypothetical character of "the conclusions of the Report of 

 1894-5, appears by his letter of August 26 to be now very fully 

 recognised by Prof. Weldon himself. I am not, however, sure 

 that it has been generally recognised. When the law of growth 

 of crabs has been accurately ascertained, then I am convinced 

 that it will require much more complex analysis than that of the 

 Report to ascertain whether a selective death-rate does or dees 

 not exist. I should not have said so much on this first point 

 did I not believe that next to blindly rejecting natural selection, 

 the most dangerous course open to biologists is to accept a 

 proof of its existence which is sure one day to be demonstrated 

 as fallacious by one of the many opponents of Darwinism. 



On the second point, surely Prof Lankester is entirely in the 

 right ? It is not sufficient to show that there is a correlation 

 between a certain frontal ratio and de.-ith-rate in order to assert 

 that the frontal ratio is a cause of death-rate. Very probably it 

 may be, but the demonstration is not logically complete, or at any 

 rate a definition of cause has been adopted which does not appear 



1 The term is due to Prof. Lankester, who thus described us— I think it 

 was to Mr. Thiselton-Dyer— in the early days of the Teaching UniverMty 



