October 20, 1923] 



NA TURE 



585 



meaning of his master and to delight in the subtleties 

 of that astonishing world of close-packed thought, 

 microscopic .and yet universal. He will, like Prof. 

 Joachim, find this treatise " fascinating and masterly," 

 and he will give thanks unstinted to him for his superb 

 exposition of it. Only those who ha\e wrestled with 

 the prodigious difficulties of such a work for themselves 

 can appreciate the learning and mastery shown by 

 him on every page of his commentary. The text also 

 is very greatly improved : it is something of a shock 

 to learn how untrustworthy is that of Bekker which we 

 have been in the habit of accepting without demur. 



(2) This miscellaneous work discusses various pheno- 

 mena of the heavens (such as clouds, comets, the 

 rainbow), the nature of the sea, earthquakes, wind, 

 thunder, many properties of " composite bodies " such 

 as iron, wood, honey, and plenty of other things 

 besides. The admirer of* Aristotle's biological works 

 will be sorely disappointed by it ; here are none of 

 the flashes of insight and the grand generalisations 

 which astonish us in those works, but here are his 

 vices to be seen in abundance, especially the almost 

 total absence of experiment, and the failure to test 

 his hypotheses, the need for doing which he ■ might 

 ha\e learnt from Socrates. One soon becomes weary 

 of reading one facile explanation after another, almost 

 al\\';i\ s on wrong lines : for example, the Milky Way 

 is ■■ a fringe attaching to the greatest tin Ic and due 

 to the matter secreted." At the same time, it is of 

 some interest, as testifying to the universality of its 

 author's outlook on the world ; " the number of things 

 that man spied into," said Goethe of Aristotle, " is 

 beyond belief." Perhaps the most interesting observa- 

 tion is that " we have only met with two instances 

 of a moon rainbow in more than fifty years," which 

 shows how Aristotle kept his eyes open ; how many 

 of us ha\e seen two of them ? But it is not given to 

 any one man to be supreme alike in biology and physics. 



The translation is excellently done, and Webster's 

 earlv death — he was killed in battle in 1917 — is a sad 

 loss to scholarship. 



A Survey of Scientific Literature. 



Statist I ml Bibliography in Relation to the Growth of 

 Modern Civilization : Two Lectures delivered in tJic 

 Universitv of Cambridge in May ig22. \\\ \\. 

 \\'\iiilliaiii Ihilni'-. i'p. 44 + 5 Tables -1-4 charts. 

 (London : Gralton and Co., 1923.) ds. net. 



THIS book contains two nut of the four le( tures 

 delivered by Mr. \\'\n(il:ain I Inline a> S.unKirs 

 reader m h;lili<i:Ta])li\ at tlie ( inxiTsitN' of ( 'ainlu'idLje 

 in May 19^.^ ■'■■•^^ mhih ;i not.iMe eunlnbulion to the 

 science of 111 Mr. I Inlnie's thesis is the need 



NO. 2816, VOL. I I 2] 



of co-operative action in bibliography, and in these 

 lectures he urges as an example of this need the 

 importance of bibliographical data as an aid to the 

 illustration and interpretation of changes in the pro- 

 gress of modern civilisation. 



The growth of scientific literature as a measure of 

 man's activity has not been generally recognised — 

 though the records previous to the nineteenth century 

 are as a rule much more full and trustworthy than the 

 ordinary data of the statistician — and Mr. Hulme here 

 shows by means of graphs and tables how bibliographical 

 statistics may not only serve to confirm conclusions 

 already reached from other sources, but may also aid 

 us to define and explain more precisely important move- 

 ments of our social and industrial history. He takes 

 as an example the International Catalogue of Scien- 

 tific Literature as being fairly representative of the 

 world's scientific literar\' output, and has compiled 

 statistics for the years 1901 to 191 3 for each of the 17 

 sections into which that work is divided, and correlated 

 these with statistics of patents for invention, trade, 

 population, etc. The figures given from the Inter- 

 national Catalogue admittedly cannot be taken as final, 

 for they are not only themselves subject to many adjust- 

 ments, but they are also confined to the literature of 

 pure science, and any influence that may have been 

 exerted by advances in technology is obscured. More- 

 over, each branch of science is treated as a whole, and 

 the behaviour of the various subclasses within each 

 branch and their interrelation cannot therefore be 

 studied. Nevertheless, the figures show certain features 

 which would probably not be greatly modified by a 

 more detailed examination. There is, for example, an 

 undoubted indication of the rhythmic progress of a 

 science, which appears to proceed in alternate periods 

 of growth and stagnation and rises to a period of 

 maximum output which in some cases it may be possible 

 to predict. The year 1910 seems to have been a peak 

 year, for there is evidence of a general falling off in all 

 sections of the Catalogue and in patents after that 

 year. l)nt unfortunately the confusion arising from the 

 War ha^ so \ itiated all statistics for years later than 

 1 91 3 that it is impossible to check the extent and the 

 duration of this depression. 



Another surprising feature to which Mr. Hulme 

 (lir(( ts attention is shown in the geographical dis- 

 tribution of the journals indexed in the Catalogue 

 throughout the period 1 901-13. The figure for Ger- 

 niaiiN and .\u.stna i.-> <inl\ jnst K ss than those for 

 l'"ranec, Kuvsia. the L luled States, and Great Britain 

 combined, \^!iile these four countries follow in the 

 iinler L;i\en. with Knssia appreciably higher than the 

 United Slates or Great Uritain. 



In connexion with the l.n; li>h patent statistics which 



Q3 



