October 31, 1902.] 



SCIENCE. 



709 



finally imbedded in the terrace from which it 

 was recovered after the manner detailed by 

 Mr. Huston. The bones of the cranium are 

 exceptionally heavy and massive, indicating 

 that the skull pertained to an adult male. 

 The frontoparietal surface is gently but regu- 

 larly concave, the depth of the concavity being 

 18 mm. This surface is likewise very rugose. 

 The horn cores are directed abruptly down- 

 ward and a little inward distally, in so far as 

 they are preserved. The extremities of both 

 horn cores are wanting. The expanse of the 

 horn cores equals but does not exceed that of 

 the orbits. There are rather large frontal 

 sinuses. These, as well as the various foram- 

 ina, contain a considerable number of peb- 

 bles, nearly all of which are of local origin. 

 Among them are two rather large fragments 

 of coal. The character of the enclosed peb- 

 bles would seem to indicate that the specimen 

 had not come much in contact with glacial 

 detritus from the north. 



The characters of the skull are such that I 

 have no hesitancy in referring it to Ovihos 

 cavifrons Leidy, first described by Dekay in 

 1828 as Bos pallassi in the Annals of the Ly- 

 ceum of Natural History of New York. The 

 chief interest attached to the present specimen 

 comes from the additional evidence it affords 

 as to the faunal changes brought about over 

 this region during the glacial period. The 

 remains of this animal have now been authen- 

 tically reported from Fort Gibson, I. T. ; St. 

 Louis, New Madrid and Benton Co., Mo. ; 

 Trumbull Co., Ohio; Big Bone Lick, Ky.; 

 from two different localities in Pennsylvania; 

 and from Council Bluffs, Iowa and West Vir- 

 ginia. In every instance these remains have 

 been recovered either directly from glacial de- 

 posits or from deposits that have been corre- 

 lated with some stage of the glacial period. 

 Since there would seem no good reason for 

 assuming that the musk ox at that time pre- 

 ferred climatic conditions very different from 

 those with which they are at present sur- 

 rounded, the reasonable inference would seem 

 to be that with the advancing ice they moved 

 southward until their range reached an ex- 

 treme limit averaging a few degrees, perhaps 



three or four, beyond the southern limit of 

 the ice. J. B. Hatcher. 



Caenegie Museum. 



EXCEPTIONS TO Mendel's law. 

 De Vries, Correns, and some other writers 

 have called attention to a number of appar- 

 ently important exceptions to Mendel's law. 

 In order to show the relation of these excep- 

 tions to the law, the law itself may be illus- 

 trated as follows: A and B are two plants, 

 each of which is self fertile and which may be 

 hybridized. Regarding any single respect in 

 which these two plants differ, the resulting 

 hybrid is a mono-hybrid. We will assume 

 that the character B is recessive in the hybrid, 

 representing the character by a small letter 

 in cases where it is latent. The following dia- 

 gram shows the results of hybridization, as far 

 as the second generation. 



kinds of pollen and two kinds of ovules pro- 

 duced by the hybrid plant Ab we get four 

 fertilizations : A X A, which gives jDlants of 

 the type of the parent A; B X B, which gives 

 plants like the parent B ; A X B and B X A, 

 which give again the hybrid Ab. 



It should also be stated that since each pf 

 these four crosses will occur an equal number 

 of times according to the law of probabilities, 

 the type A will constitute one fourth of the 

 second generation, B one fourth, and Ab one 

 half. 



Mendel's law, as first stated independently 

 in this country (Bui. 115, Off. Ex. Sta., p. 

 93) and essentially as stated by himself, is as 

 follows : In the second and later generations 

 of a hybrid there occur all the possible com- 

 binations of the characters of the parents, 

 and in definite proportions. 



But hybrids have been found in which this 

 seems not to be the case. The explanation of 

 a number of these is here offered. Millardet, 

 De Vries, Correns and others report cases of 



