440 Vererbiijg,* Variation, Mutation. 



is no more variable than che parents, the variability of F^ being fouad slightly 

 greater than the average of the parents in six cases and less in five cases. This 

 result agrees with Johannsens Observation. Different variants in F^ give similar 

 resiilts in Fg, showing that the Variation in F^ is fluctuating Variation due to en- 

 vironment and is of no germinal value. The F^ generation is more variable than 

 the parents. W\/en sufficient numbers of variants were studied the Fg showed a 

 ränge of varia/ion equal to the combined ränge of the parents and Fy In the 

 two Grosses studied there was only a small negative correlation between average 

 leaf area and number of leaves per plant. This indicates that leaf number and 

 average leaf area are inherited independently ; therefore we can combine the de- 

 sirable leaf size sharacters of one variety with the number of leaves of another 

 form. The results show some Variation in the correlation between height of plant 

 and number of leaves. Thus, the correlation coefficients of the two Fg generations 

 of the cross between two of the types studies were -{- ■ 342 + • 058 and + • 408 

 + • 036, while in the F« of the cross between the two other types the correlation 

 coefficient was -(--814 +'016. There was foimd a large positive correlation 

 between length and breadth of leaf, which indicates that the inheritance of these 

 characters depends on the sarae cause or series of causes. — It is concluded that 

 these "results are entirely in accord with the Mendelian interpretation of quan- 

 titative characters, such as the size of various plant organs, by the hypothesis 

 that a multiplicity of factors exists, each independently inherited and capable of 

 adding to the character, the heterozygous condition being half the homozygous. 

 The difficulty of correctly determining the exact number of factors in any case 

 is greatly increased, however, by the presence of fluctuations which, although 

 of no germinal value, obscure the action of heritable factors. Moreover, some 

 characters seem independently inherited, others closely correlated in inheritance 

 and still others partially correlated. These facts make the analysis of pedigree 

 culture data yet more difficult." Pearl (Orono). 



1229) Waldrou, L. E., Hardiness in successive alfalfa generations. In: Amer. 

 Natural. 46, 463—469, 1912. 



In the species Medicago sativa there is wide diversity in hardiness auiong different 

 strains, depending chiefly upon their geographic origin. Thus the Mongolian alfalfas are 

 very hardy while those from Peru and Arabia are very tender. It is cousidered pro- 

 bable that each strain of alfalfa is composed of biotypes which differ in hardiness, but 

 since the peruvian alfalfa contains biotypes which are hardy, it is not necessary to as- 

 sume that these biotypic dift'erences have ai-isen by mutation very recently. 



Gates (London). 



1230) Sannders, Miss E. R. (Cambridge, Newnham College), Further Con- 

 tribution to the study of Inheritance of Hoariness in Stocks 

 {Matthiola). In: Proc. Roy. Soc. Vol. B. 85, Nr. 582, S. 540—545, 1912. 



The wirter concludes that Sap-colour in Stocks is due to the presence of 

 two factors (C and R) in the absence of either of which the sap is colourless. 

 Hoariness also depends ou the presence of two factors (H and K) between 

 which and the Sap-colour pair a certain interrelation may exist. This inter- 

 relation between the two pairs of factors is such that the hoary effect due to 

 H and K only appears when C and R are also present. Hence a non-sap-coloured 

 individual may contain both H and K and yet be glabrous, but in a sap-coloured 

 glabrous form H and K cannot both be present. Similarly a non-sap-coloured 

 glabrous form cannot contain both C and R. Doncaster (Cambridge! 



