GENETIC AND CYTOLOGIC AL EFFECTS 261 



dependence were quite different from those found in the experiments of 

 Noethling and Stubbe (1936) with monochromatic irradiation of Antir- 

 rhinum pollen. The results indicated (1) that wave lengths 313 m^ and 

 longer were relatively ineffective, (2) that wave length 302 m^ was geneti- 

 cally effective but less effective than wave lengths 297 m/x and shorter, 

 and (3) that wave length 254 m^u was much more effective than 297 mju. 

 Subsequent studies with monochromatic radiation, reviewed in a later 

 section, confirmed these indications. The maize data are for the fre- 

 quency of endosperm deficiency, while the Antirrhinum data are for the 

 frequency of point mutations. However, tests of mutation rate in the F2 

 progenies indicated the same spectral relations for mutations as for 

 endosperm deficiency, as far as could be determined from filtered radia- 

 tions (Stadler, 1941a). 



The frequency of induced embryo abortion was much lower with the 

 longer ultraviolet wave lengths than with the shorter. Comparing doses 

 approximately equal in frequency of induced endosperm deficiencies, the 

 frequency of induced embryo abortion was about nine times as high for 

 the discharge-tube radiation (chiefly wave length 254 m^u) as for the 

 filtered radiation (chiefly wave lengths 297 mn and longer). 



The frequency of endosperm deficiencies induced by ultraviolet treat- 

 ment of the pollen is very much higher than the frequency of deficiencies 

 affecting the embryos of the same seeds (Stadler, 1941a). Both values 

 can only be estimated, but the discrepancy is too great to be accounted 

 vor by any possible error in the estimates. 



With maximal doses of the longer wave lengths, the frec^uency of endo- 

 sperm deficiencies marked by A, Pr, and Su sometimes exceeds 40 per 

 cent. Since these marker genes can detect only a part of the deficiencies 

 occurring in 3 of the 20 chromosome arms, there are presumably several 

 hundred endosperm deficiencies per hundred seeds. These represent the 

 deficiencies (and perhaps gene mutations) realized under the conditions 

 of endosperm development, from alterations induced in one of the sperm 

 nuclei of the treated pollen grains. The embryos of the same seeds may 

 be checked to provide a maximal estimate of the frec^uency of deficiencies 

 realized under the conditions of embryo development, from alterations 

 induced in the other sperm nuclei. By this check, every sperm nucleus 

 in the tested population of treated pollen grains may be accounted for. 

 Assuming that every aborted embryo, every plant which failed to yield a 

 pollen specimen, and every plant which showed segregation of defective 

 pollen represents an induced deficiency, the maximum estimate of defi- 

 ciencies in the gametes tested by embryo constitution is only about 30 

 per cent. 



A similar comparison may be made for endosperm deficiencies and 

 embryo deficiencies marked by specific genes affecting both endosperm 

 and plant characters. Endosperm deficiencies of .4 are very common, 



