Grofi, Uber intcrmediiire und alternative Vererbung. (i()7 



prevents the proper elimination of the antagonistic characteristics. 

 In the second series, if we take the grey batch as the F 2 D batch 

 the results agree well with one another, and the grey form is 

 dominant to the normal. In this case, however, there is a third 

 characteristic accompanied by the dominant forms. If we consider 

 them as a single characteristic, we get a pure Mendelian result. 



Similarly in the third series, if we take the spindle-shaped 

 eggs as the recessive female F 1? their successive generations are in 

 good accordance with the formula demanded by the principle, 

 excepting in the absence of uniform F 4 D eggs. This might be due 

 to the small number of worms reared in this case, since flacherie 

 made great havoc among them; and thus most of the worms 

 during the fifth stage died, and we only obtained a small number 

 of moths. 



Brood characteristics ("Voltinism") of the silk-worm. 



The phenomena of inheritance of the brood of the silk-worm 

 also afford-another example of matroclinous inheritance. In my opinion, 

 they come in the same category of inheritance above referred to, but 

 the phenomena are more complicated than in the former cases 

 because the divoltine characteristic is easily disturbed by the influence 

 of temperature during the embryonic stage. 



As is well known among Japanese breeders, divoltine breeds 

 may be changed into uni-, di- or multivoltine, by the influence of 

 the temperature during the incubation of the egg. As far as we 

 have experimented, if we expose the eggs to a temperature of about 

 80 F. and upwards, all the worms which emerge from them will 

 give moths which lay univoltine eggs without exception, while the 

 embryo developedundera temperature of about 65 F. or less, gave 

 moths which laid all divoltine or multivoltine eggs. Hence the result 

 of experiments made on the subject of brood-character might give 

 rise to great mistakes unless we pay special attention to the temp- 

 erature during the development of the embryo. On the other hand, 

 the univoltine characteristic was not so easily influenced by the 

 temperature as the divoltine was. 



So we come at last to the conclusion that nearly all the charac- 

 teristics of the silk-worms studied by us since 1900, even those 

 which seem to be non-Mendelian, nevertheless really are Mendelian. 



Uber intermediare und alternative Vererbung. 

 Von Dr. J. Grofs (Neapel). 



1. Einleitiiiig. 



In zwei friiheren Arbeiten habe ich versucht, der heute von 

 fast alien Vererbungsforschern geteilten Auffassung, die in den 



