NONCHROMOSOMAL GENES 253 



ing conditions. For example, summer field-grown plants are much 

 greener, with fewer white sectors, than sibling plants grown in the winter 

 in the greenhouse. Random segregation of pre-existing normal and 

 mutant plastids cannot readily explain such an environmental influence 

 on the extent of white sectoring. 



3. When the proportion of white seedlings is compared in the first and 

 subsequent generations of crossing striped plants, it is found that the 

 frequency of white increases with the numbers of generations of female 

 striped plants, as if some toxic substance were increasing slowly in con- 

 centration, affecting more and more of the determinants. Moreover, 

 the per cent of entire white seedlings in each generation is much less 

 than the per cent of white areas in the striped seedlings, again indicating 

 a process more complex than mere random segregation of particles. 



These observations favor the view that the nonchromosomal determi- 

 nant in this system is not the plastid itself. 



Currently, the problem of chloroplast heredity is being re-examined 

 with chloroplast mutants of microorganisms. The finding that strepto- 

 mycin-treated seeds give rise to albino and striped seedlings led some 

 investigators to test the effects of streptomycin upon algae. Strains 

 of Euglena were found which responded dramatically to streptomycin 

 treatment by the irreversible loss of chlorophyll-forming ability. Sub- 

 sequently, the same result was obtained in these strains by a short heat 

 treatment at 34° C and also by UV-irradiation. In all three treatments, 

 the effects were exhibited by most of the treated cells, with very little 

 death. Thus, it seems that an induced change, different from spontane- 

 ous mutation in affecting an entire treated population, has resulted in 

 loss of a hereditary property. It is not established whether the chloro- 

 plast itself or some other cell constituent is the site of the altered 

 or lost genetic determinant; the eventual loss of ability to form chloro- 

 plasts may be a secondary effect of a primary change elsewhere. In the 

 absence of a means to study recombination in this asexual organism, no 

 method has been devised to locate the relevant hereditary site. 



In Chlamydonwnas, streptomycin treatment has given rise to yellow 

 mutants, which have lost their ability to carry out a particular step in 

 chlorophyll synthesis. These mutants are phenotypically the same as 

 previously studied yellow mutants of spontaneous origin. In crosses 

 between spontaneous and streptomycin-induced yellow mutants, no green 

 recombinants have been found, indicating that the same site has been 

 altered in both groups of mutants. In crosses of yellow mutants with 

 normal green strains, the progeny segregate 1:1, but the y factor shows 

 only first-division segregation. Since there is but one chloroplast per 



