NOVEMBEB 29, 1907] 



SCIENCE 



751 



disprove that should be the condition of all of 

 the earth's elements toward the center. The 

 study of seismic vibrations will probably 

 settle this question. A. 0. Lane 



Modern Chemistry, Theoretical and System- 

 atic. By Sir William Eamsay. 12mo, pp. 

 327. New York, The Macmillan Co. 1907. 

 Price, 70 cents. 



Sir William Ramsay's epitome of modern 

 chemistry, issued originally by Dent as two 

 volumes of his dainty series of Temple En- 

 cyclopedic Primers, can not fail to find many 

 new readers in the present one-volume form. 

 Chemists should not need to be told of its 

 merits, but if there be any who have over- 

 looked the book they can only be envied for 

 the treat which its perusal has in store for 

 them. Students of other sciences will find 

 in the book that for which many of them have 

 been looking. They will find an account of 

 the science, in which the chief results of 

 modern physicochemical work are not. only 

 described, but are so incorporated into the 

 chemistry of our school days that the nature 

 of the debt of the latter to physical chemistry 

 is plainly visible. The book fulfills its pur- 

 pose singularly well, for it is brief, yet ad- 

 mirably clear and readable. To bring the 

 present edition up to date a few minor changes 

 only were required. The chief of these seemis 

 to have been the addition of a paragraph on 

 the radium emanation. 



Alexander Smith 

 The University of Chicago 



Selection and Cross-breeding in Relation to 

 the Inheritance of Coat-pigments and Coat- 

 patterns in Rats and Guinea Pigs. By H. 

 MaoCuedy and W. E. Castle. Carnegie 

 Institution of Washington, Publication No. 

 70, May, 1907. 



The authors publish some important data 

 concerning the heredity of the spotted coat in 

 rats and in guinea pigs. They confirm the 

 conclusion of previous observers that the uni- 

 form or self color is dominant to the spotted 

 coat and the latter to the albino. Two types 

 of spotted rats were used, the Irish, with 

 white on the ventral side, only; and the 



hooded, with black head and rump and a 

 dorsal black stripe — the remainder of the body 

 being white. Both of these types the authors 

 call partial albinos, although the Spotted con- 

 dition appears to be a different "unit-char- 

 acter" from that of the albino. In other 

 words, the albino is the allelomorph of the 

 spotted coat and not a graded condition of 

 the latter as the term partial albino might 

 seem to indicate, although the authors recog- 

 nize the distinction just given. It has not 

 been found possible to produce an albino by 

 selection or otherwise from the spotted coat. 



Within the range of the spotted coat the 

 authors find that it is possible by selection to 

 produce races that breed approximately true to 

 any special degree of spottedness. They argue 

 from this that selection of a continuous or 

 fluctuating variation may produce fixed types 

 within the limits of the variation. They eon- 

 tend, therefore, that their results are opposed 

 to the conclusion of de Vries that fixation of 

 fiuctuating variations by means of selection 

 can not take place. But do the results really 

 establish this point. May not there be several 

 or even many semi-stable states of the spotted 

 coat that ordinarily overlap, i. e., may there 

 not be within the limits of apparent fluctua- 

 tion certain individuals that reproduce the 

 parent type? A comparison with Lang's re- 

 sults on snails and of Tower's on the potato 

 beetle would have been a welcome addition to 

 the paper in this connection. However this 

 may turn out in the end — and there is clearly 

 something peculiar in the inheritance of spot- 

 ted types that does not conform to the idea of 

 unit characters — the authors' data are a valu- 

 able contribution to the subject. 



The attempt of the authors to fix certain 

 color patterns in guinea-pigs — nose spots or 

 head spots, or Dutch marked individuals, 

 gave a negative result — a fact already familiar 

 to practical breeders. The experiments led, 

 however, also to a positive conclusion of no 

 little interest. It was found from the study 

 of 1,048 guinea pigs " that one can, by selec- 

 tion, either increase or decrease the extent of 

 the pigmented areas, but it is impossible by 

 selection to fix this pigmentation in a par- 

 ticular pattern, retaining pigment areas on 



