CYTOPLASM AND ACTIVATION OF THE GENE 273 



in question contains cither a lethal gene or is as a whole lethal, 

 the gene acting only in major cytoplasm in a homozygous state. 

 A case studied by Honing in Canna subspecies crosses is similar 

 but less extreme in plasmatic effect; others found by Skalinska 

 and Gairdner fall in line. In addition to this case, still more 

 cases were found that resemble our Lymantria case in that not 

 only Fi in reciprocal crosses is different but also the segregants 

 in F2 are carried in one direction by the cytoplasm. In our case, 

 this direction was the direction toward the maternal type from 

 which the cytoplasm sprang: light cytoplasm, for example, made 

 all gene combinations lighter. In Sirks' case, however, the minor 

 cytoplasm generally increased the action of certain growth genes 

 but decreased that of others; similarly with color genes. This 

 difference between the two cases cannot, however, be considered 

 as of primary importance. The subspecies of Lymantria, which 

 were used, are distinct natural forms, isolated from each 

 other and therefore presumably well balanced in themselves 

 through long selection. The subspecies of the beans, however, 

 are cultural varieties and therefore probably not so distinctively 

 balanced. It must be stated that, unfortunately, in Sirks's 

 cases, the analysis of further extracted generations is missing. 



We mentioned previously the fact that all the cases of plasmatic 

 differences thus far known and recognizable by genetic methods 

 were found in crosses involving at least subspecies. This does 

 not mean, however, that all such crosses must give this result. 

 Baur (1932), for example, reports that in his numerous crosses 

 between geographic so-called species of Antirrhinum, no recipro- 

 cal differences were found. But as this material has not yet 

 been worked up quantitatively, later work might change this 

 statement. There has come to my attention, however, only a 

 single case in which positive results have been found in an ordinary 

 cross with mutations. Kuehn (1927) reported the isolation of a 

 strain of the wasp Habrobracon differing in intensity of body 

 pigmentation from the stock. In crossing this form to the type, 

 he reports results that practically repeat our results in Lymantria 

 and are represented sufficiently by the diagram given above, with 

 the exception of multiple genes being involved. 



We have dwelt thus far mostly upon cases in which the con- 

 clusions regarding plasmatic influence upon the gene-controlled 

 phenotype could be tested at least in the F 2 generation. But 



