no Referate. 



Study of Religions", of W. C. D. Whetham on "The Evolution of Matter", 

 and of Sir George Darwin on "The genesis of Double Stars". But perhaps 

 the most fascinating essay of all is that by Professor Judd on Darwin 

 and geology. Having intimabely known both Darwin and Lyell the author 

 has been able to catch and fi.x for us some of the personal charm of these 

 two great men. Fortunately Professor Judd's essay is one of the long- 

 est in the book. We need only add that all who come across this volume 

 cannot fail to be grateful for the admirable way in which it has been 

 turned out by the Cambridge University Press and for the judicious care 

 with which it has been edited by Professor Seward. 



R. C. Punnett. 



Spillman, W. J. The nature of "unit" characters. The American Naturalist 

 43 1909. p. 243—248. 



In this paper the author has laid emphasis upon a thought, which, 

 though simple in itself, has not before been brought to the attention of 

 Mendelists. It arose from Baur's statement that the chromosome could not 

 be the basis of the unit character, since he had demonstrated the possession 

 of fifteen allelomorphic pairs by Antirrhinum majus, which is more than 

 the number of chromosomes in the species. This argument is not valid 

 however, unless it can be shown that more independent dominant characters 

 can e.xist in a single individual of the species than there are chromosomes 

 after reduction division. 



To prove that Mendehan phenomena are beyond the functions of the 

 chromosomes, it should be shown beyond a doubt that in a species bearing 

 2n chromosomes, n + 1 dominant characters can exist in a single individual, 

 without gametic coupling. Such proof has thus far been lacking. In both 

 Pisum and Anthirrhinum there are more distinguishable characters than there 

 are chromosomes in their varieties, but since these characters are possessed 

 by different varieties it may very well be that they have arisen separately 

 by functional variation of the chromosomes, and could not exist together. 



The w riter further elaborates by supposing each chromosome to possess 

 different lunctions. The entire failure of one of these tunc > ions would 

 account for a normal simple Mendehan ratio. If two or three chromosomes 

 were concerned in producing a somatic character, the genetic analj'sis 

 would show it to be a compound character produced only when these two 

 "factors" interact. The case of polled (dominant) and horned (recessive) 

 cattle is used as an illustration. He says, "Let us assume that the pro- 

 duction of horny substance requires that the functions a, b, e and f (in 

 chromosome B) shall be normal. If in a given group of individuals the 

 function e fails, which function may represent the production of a given 

 chemical substance in the cell, then horns fail to develop." While there 

 is nothing illogical in this way of looking at the matter, it seems as if 

 a more fortunate illustration could be used, - — one which would not run 

 counter to the "presence and absence" hypothesis. It is certainly more 

 simple to say that the presence of something in the cells of the polled 

 animals has inhibited the production of horny tissue. 



The work of several cytologists is cited in support of the theory that 

 Mendelian is solely chromosomic. Dr. R. R. Gates has suggested before 

 that abnormal chromosome behavior may account for the mutations of 

 De Vries. If this is the case, normal evolutionary changes may be simply 

 normal changes in chromosome function of a chemical nature. 



E. M. East, Harvard University. 



