346 



SCIENCE 



[N. S. Vol. XLVII. No. 1214 



phila supply in themselves all the evidence 

 that even an extreme selectionist should ask. 

 To explain why large changes are observed as 

 well as small ones it is suggested that we may 

 be "witnessing the disintegration of highly 

 developed apparatus instead of its build- 

 ing up." 



Dr. Castle's contribution also deals with the 

 controversy over the nature of evolutionary 

 change, whether continuous or discontinuous. 

 He, however, is less inclined than Dr. Jen- 

 nings to resolve the differences between the 

 two schools into differences in the use of 

 terms. After allowing for the effects of the 

 confusion in terminology, he sees " two con- 

 trasted sets of ideas," which he arranges under 

 the headings Darwin and DeVries. 



Students of paleontology, geographical dis- 

 tribution and classification are shown, in gen- 

 eral, to favor the belief in gradual evolution 

 and the efficacy of selection. The opposite 

 view, that of discontinuous variation and sta- 

 bility of new- forms, is held by a majority of 

 the students of experimental breeding. Sup- 

 port of the former view by one who has done 

 such thorough work as an experimental 

 breeder must have great weight. 



From the point of view taken by Jennings, 

 it would seem that in contrasting the mode of 

 origin and the stability of new types Castle 

 is himself open to the criticism of using the 

 term " type " ambiguously. If by a new type 

 is meant anything new, most geneticists would 

 range themselves with DeVries, but if a new 

 type is something more comprehensive, De- 

 Vries and his followers must stand alone. 



The crux of the difference between Castle's 

 views and those of practically all classes of 

 mutationists would seem to be in Castle's 

 holding that selection determines in some 

 measure the range of variation in subsequent 

 generations. Confining the question to in- 

 herited variations, does the selection of ex- 

 treme variations form new centers of distri- 

 bution? To do so, it would seem that small 

 variations must be more numerous than large 

 ones, an assumption which would be ques- 



tioned by most mutationists, including Mor- 

 gan. 



Although cases were encountered in Dr. 

 Castle's own work in which selection gave no 

 tangible results, in many characters progress 

 was rapid and continuous, with no indication 

 of other than physical limits, and it is held 

 that in the smaller mammals, which Dr. 

 Castle has studied intensively for many years, 

 there are few characters which can safely be 

 referred to the agency of perfectly stable 

 genes. 



That selection does, from a practical stand- 

 point at least, produce results is abundantly 

 shown by Dr. Castle's work. If by the use of 

 refined reasoning his critics are able to show 

 that change in a single Mendelian character is 

 not the only possible explanation of the re- 

 sults, these critics may then be referred to 

 Jennings's results with Difflugia. 



Dr. Riddle has here brought together in a 

 concise and readable form the results of his 

 extensive experiments on the nature of sex 

 in pigeons and has coordinated these results 

 with the work of other investigators. 



Sexual differentiation is interpreted as the 

 expression of quantitative differences in the 

 rate of protoplasmic activity; the more active 

 metabolism resulting in males, the less active 

 in females. Many lines of evidence are pre- 

 sented, all of which are consistent with the 

 view that preponderance of one or the other 

 sex is conditioned by the rate or level of metab- 

 olism. These various lines of evidence show 

 the following characteristics to be associated 

 with the female sex, all of them being associ- 

 ated also with a low level of metabolism: 

 Large size of yolk, low per cent, of water in 

 the yolk, high per cent, of stored material, 

 high total stored energy, exhausted physical 

 condition of the mother, age of the mother. 



In crosses the percentage of males increases 

 with the width of the cross to the point of in- 

 fertility. Since males are characterized by a 

 more active metabolism, there is an agree- 

 ment between these results and the commonly 

 observed increased vigor of hybrids. 



Assertive mating of gametes and differen- 

 tial death rates are fully considered and neither 



