284 



NATURE 



[November II, 19 15 



conductor is, in fact, constant, so long as its 

 physical properties arc unaltered," and this is 

 the only meaning- Ohm's law can have if it is a 

 law and not a statement without physical signi- 

 ficance. But this is a small point to criticise 

 where so much is excellent. 



It is as a historian and recorder that Prof. 

 Thompson has earned the gratitude of electrical 

 engineers and physicists as much as in any other 

 direction ; and from that aspect alone his " Ele- 

 mentary Lessons in Electricity and Magnetism " 

 is worth reading. 



Market Gardening. By Prof. F. L. Yeaw. 



Pp. vi + i02. (New York: John Wiley and 



Sons, Inc. ; London : Chapman and Hall, Ltd., 



191 5.) Price 3-?. 6d. net. 

 This little book is intended to serve not only as 

 an elementary text-book of market gardening-, 

 but as an introduction to vegetable-growing- in 

 school and home gardens as well. 



To attempt to deal with so complex a subject 

 within the space of 68 pages of bold type was 

 heroic, and only an author well acquainted with 

 his subject could have given so much trustworthy 

 information in so small a space. We could wish 

 that its admirable points had been still further 

 elaborated for the use of beginners, who are 

 assumed to know more of manual processes than 

 is likely to be the case. 



Viewing it as an account of American market 

 gardening, one is struck with the remarkable 

 omissions concerning things which we believe, on 

 this side of the Atlantic, to be of the first import- 

 ance. Intercropping is illustrated by four excel- 

 lent plates, but only once incidentally mentioned 

 in the text; rotational and successional cropping 

 are not mentioned at all. No help is given as to 

 the selection of profitable varieties. The spade 

 apparently finds no place even in the home garden, 

 and many labour-saving devices, methods, and 

 appliances well known here seem to be unknown 

 there. Sub-irrigation is not mentioned, and" only 

 twenty-three crops are dealt with, whereas our 

 market gardens accommodate nearly double that 

 number (not including fruits and flowers). More 

 notes upon the cost of raising crops, the returns 

 to be expected, and the like, would have been 

 valuable ; onlv here and there is this mentioned. 



F. J. C. 



The Student's Handbook to the University and 

 Colleges of Cambridge. Fourteenth edition. 

 Revised to 30 June, 1915. Pp. 701. (Cam- 

 bridge: At the University Press, 1915.) Price 

 35. net. 



Though the statements made in the handbook 

 are not official, every care has been taken by 

 reference to competent authorities to make them 

 accurate. The temporary emergency legislation 

 occasioned by the war is recorded in this new 

 edition. This legislation refers, among other 

 matters, to the allowance of terms and examina- 

 tions to students on active service, and the modi- 

 fication in their case of various fees ; and also to the 

 modification of certain examination regulations. 

 NO. 2402, VOL. 96] 



LETTERS TO THE EDITOR. 



[The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notire is 

 taken oj anonymous communications.] 



The Origin of New Adaptive Characters. 



The author of the article in Nature of July 19 

 entitled "The Material Basis 01 Evolution" has mis- 

 understood some of the chief bearings of my address 

 entitled " Origin of Single Characters as Observed in 

 Fossil and Living Animals," ^ and I would regret to 

 have certain passages in this well-intentioned review 

 quoted as expressing my opinions. My address was 

 based on researches carried on continuously since 1889 

 on the actual modes of origin of single specific and 

 generic characters as observed in several lines of 

 mammalian descent (horses, rhinoceroses, and 

 titanotheres) in which such characters can be traced 

 from their beginnings to their final development with- 

 out any break. The term numerical was used as a 

 convenient designation for all characters which may 

 be expressed in numbers and formulae, while the term 

 proportional was used for such characters as may be 

 expressed in indices and ratios. ^ There is a clear 

 genetic distinction between these two kinds of char- 

 acters, for while changes of proportion, such as the 

 reduction of the lateral digits of the feet of horses, 

 gradually lead into changes which may be expressed 

 numerically, such reduction-changes fall under laws 

 which govern quantitative changes in general rather 

 than laws which govern the origin of new characters. 



It is in the origins of new adaptive organs, such 

 as the horn rudiments of the skull of mammals, that 

 we observe another evolutionary principle in opera- 

 tion, whereb}' descendants of the same ancestors give 

 rise to simlar numerical characters, both in near 

 and remote lines of descent, while at the same time 

 they may give rise to entirely dissimilar proportional 

 characters. In other words, there is some principle 

 of ancestral hereditary predisposition or control in the 

 origin of many new numerical characters which we 

 cannot at present understand or explain. Nor can we 

 now explain the causes of the origin of proportional 

 characters, because both the Lamarckian and the Dar- 

 winian explanations fail in certain cases. The principle 

 of ancestral predisposition also operates in the propor- 

 tional evolution of nearly related lines of descent, 

 although not in the remote lines. 



The distinction between numerical and proportional 

 is further sustained by the fact that modern 

 mammalogists, such as Miller and Osgood, in their 

 systematic treatises on carnivores and rodents, employ 

 either proportional or numerical characters In all their 

 definitions ; In practically all systematic treatises on 

 the hard part of mammals this distinction Is used in 

 definition. Proportional characters are far more 

 common than numerical. The observations of Osgood, 

 moreover, which are on an unprecedented scale, em- 

 bracing the examination of more than 27,000 speci- 

 mens of Peromyscus, are in absolute accord with my 

 results obtained in palaeontology, namely, that there is 

 a complete continuity (rather than discontinuity or 

 saltation) in the origin of proportional characters and 

 in the origin of several, If not all, of the numerical 

 characters observed. 



As to the De Vrles mutation hypothesis, it is ex- 



1 Presidential address before the Palaeontoloeical Soc'ely of America 

 delivered in the Academy of Natural Srienoes of Philadelphia, December 31 

 1914. 



'^ Technically these terms are replaced by the designations " rectigrada- 

 tions " and "allometrons." 



