BY THE SELECTION OF SOMATIC VARIATIONS. 7 



Later, Baur (1910), in studies of seed progeny of variegated types 

 that appeared spontaneously, assumes that green is the combined 

 result of three factors. If a certain one (Z) is absent the tissue is color- 

 less, if Z and Y only are present then a chlorina type is produced, and 

 if N is present with Z and heterozygous Y, the aurea coloration results. 

 In Antirrhinum ma jus alhomaculatum, Baur found 8 types of variega- 

 tion arising spontaneously in cultures of about 30,000 pure-green 

 plants. His crosses with these show that certain cases appear to be 

 inherited only from the seed parent. To explain the results, Baur 

 assumes that hereditary qualities are localized in different parts of the 

 cell. The nucleus, the cytoplasm, and the chromatophores all possess, 

 he considers, different but definite factors concerned with variegation. 



Such results and conclusions illustrate very well the difficulties and 

 uncertainties which arise from attempts to analyze variations in terms 

 of unit factors and suggest most forcibly the need of a more thorough 

 investigation of such variations in a progeny derived by vegetative 

 propagation. 



The variations among the branches of a single plant which Correns 

 (1909, a and h) reports in connection with variegated types of Mirabilis 

 jalapa are apparently quite similar, in degree at least, to those I shall 

 report for Coleus. It would seem to be highly important that the 

 inheritance of these variations be studied in vegetative propagation. 

 Correns, however, made a study of seed progenies only. In the case 

 of the albomaculata type these were composed of green, white, and 

 albomaculata plants in quite different ratios for different plants tested. 

 All these classes appeared when pollen from a pure-green plant was 

 used, but when pollen from the albomaculata was used on pure green 

 there was no transmission of the quality of variegation. This case of 

 matrocliny is due, he assumes, first, to the localization in the cytoplasm 

 of the factor for variegation, and second, to the condition that male 

 sex-cells in this case do not carry cytoplasm. 



ShuU's (1914) studies with variegated types of Melandrium show 

 much the same results as those of Baur and Correns. He distinguishes 

 between chlorina, pallida, and pure-green types of Melandrium on the 

 basis of presence or absence of three factors which in crosses behaved 

 quite like units. In types with green-white blotched and with chlori- 

 nomaculata variegation, however, the variegation seemed to be trans- 

 mitted only through the seed parent, but not uniformly, for crosses of 

 variegated branches with pure green gave in the Fi generation plants 

 ranging from pure green through types of chlorinomaculata to yellowish- 

 green plants. It is highly interesting that Shull found that the greater 

 the amount of chlorina coloration in the calyx the greater was the num- 

 ber of variegated seed progeny. The aurea types, which possessed as 

 a rule small round flecks, gave such varied results with appearance of 

 different types in Fi progeny that Shull concludes it must be an infec- 



