I ill \ Toyama, On certain characteristics of the Silk-worm etc. 



and the other from a cross between Theophila mandarina and Bombyx 

 mori (Tetravoltine white), both of them being established as a con- 

 stant form by ourselves, and the third, a breed extracted from 

 a Chinese univoltine white by Ishiwata. 



Of females from the first and the third albinos mated with 

 normal males, some gave characteristic white eggs, some a mixture 

 of both white and many other light coloured eggs, that is to say, 

 some intermediate coloured ones, rarely normal light coloured eggs ; 

 but they are never so dark as pure normal dark coloured eggs. 

 The reversed mating gave, on the contrary, all normal coloured 

 eggs, no influence of an albinotic character could be seen. 



Some albinotic forms derived from the second albino : namely 

 the albino derived from the cross, "Theophila X Bombyx" behaved 

 in a normal Mendelian way, the white being recessive towards the 

 normal coloured ones. In this case, reciprocal crosses gave the 

 same results, that is to say, only dominant characteristics make 

 their appearance in F r 



The results of the reciprocal matings above described, taught us 

 that there are certain egg-characteristics which behaved as matro- 

 clinous in inheritance, and even dominant ones when belonged to 

 males remained dormant in the offspring. As regards dominancy 

 and recessiveness, some are perfect, while others are imperfect, in 

 the latter some intermediate or mixed forms will be produced. We 

 found also that the segregation of parental characteristics in the 

 offspring takes place in the same way as Mendelian charact- 

 eristics do. 



As to the causes of the matroclinous inheritance above described, 

 there may be two. 



a) Where the origin of egg-characteristics is due to the shell 

 or yolk, which are the products of the female parent before fertiliza- 

 tion takes place; and, b) where their origin is due to the pigment 

 of the serosa, which is derived from the conjugation of paternal 

 and paternal nuclei. In the former, it is quite natural that the 

 egg-characteristics should be matroclinous in inheritance, and in the 

 latter, we attribute it in agreement with de Vries' theory, to the 

 migration of maternal pangens to the cytoplasm before the entry 

 of a spermatozoon into the egg, and thus paternal pangens have 

 no, or very little, influence upon the characteristics of the eggs. 



Now we return to Mendelian questions. Suppose there are 

 some characteristics which are predetermined by the maternal gens 

 before fertilization takes place. In such a case, even the dominant 

 characteristic, when belonging to males, will not act as a potent 

 characteristic in the offspring. If D represents a dominant factor 

 and R a recessive, their reciprocal crosses would give diametrically 

 opposite results as shown below: 



