NATURE OF THE GENETIC EFFECTS 387 



breaks are produced in interphases. If, as is usually held to be the case, 

 there is also a greater time limitation during most interphases, this would 

 constitute a second circumstance working in the same direction. Work- 

 ing in the same direction also, as far as the formation of inviable iso- 

 chromosomes is concerned, is the third circumstance, that during most 

 of the interphase stage the chromosome threads behave, in regard to 

 breakage by radiation, as though they were single, and also manage to 

 have their broken ends unite before they become effectively double, 

 while for chromosomes irradiated during or shortly before entering a con- 

 densed stage one or both of these conditions usually fails to hold, and 

 dicentric and acentric isochromosomes can be formed in consequence. 

 It is not yet known whether these three circumstances are the only ones 

 which lie at the basis of the special resistance of most interphases, as com- 

 pared with premitotic, mitotic, and spermatozoan stages, to having struc- 

 tural changes produced in them, or whether in addition the chromosomes 

 are actually less breakable during interphase, but there is at present no 

 way of measuring their breakability, uncomplicated by phenomena 

 involving union of broken ends. 



The greater vulnerability of spermatozoan chromosomes to structural 

 change constitutes the chief reason that spermatozoa have usually been 

 chosen for irradiation in experiments in which the production of such 

 changes was desired. Conversely, this is also the reason that it is desir- 

 able, for the production of offspring as free of structural chromosome 

 changes as possible, to allow a sufficient interval (in mammals, of some 

 months) to elapse after irradiation of the male before reproduction is 

 allowed. For in this way the germ cells which were in immature sperma- 

 togonial interphase stages at the time of irradiation have been given time 

 to replace those which were irradiated while in the mature condition. 

 As for the female, it has been noted previously that the higher degree of 

 dispersion of the nonhomologous chromosomes of oocytes results, in 

 Drosophila, in fewer translocations being produced by irradiation of 

 oocytes than of spermatozoa. In mammals, however, some transloca- 

 tions have been induced by irradiation of oocytes, although their fre- 

 quency remains to be better determined. It is not known how much, if 

 any, the frecjuency of these would be reduced by increasing the interval 

 between irradiation of the female and reproduction ; since in mammals the 

 oocyte stage persists for a very long period, it may be that the stage 

 sensitive to the production of structural changes lasts much longer in 

 female than in male mammals. 



Finally, the above principles appear to provide the chief reason that 

 there is so much more damage caused by radiation to somatic tissues in 

 which cell divisions are abundant than in those in which they are rare or 

 absent. Additional reasons for the greater damage to tissues with more 



