422 GENETICS OF SOMATIC CELLS 



ment in which the position of the genes with respect to heterochromatic chromosomal 

 regions determines their behavior. The patches of mutant tissue appear in individuals 

 of the proper constitution when a locus is abnormally placed next to a heterochromatic 

 chromosomal region by means of the chromosomal rearrangement, and subsequently 

 becomes variable in its expression. The size and number of patches and the type of 

 mutant allele appearing depend on "the distance of the locus from the point of re- 

 arrangement, the kind of heterochromatic region, the quantitative relations of other 

 heterochromatic regions in the chromosomal complement, and a galaxy of modifiers 

 distributed through the chromosomes. Moreover, the influence of these factors, at 

 least in part, is exerted differentially at different stages of development: the effects 

 are maximal at the later stages, but are not detectable in tissues differentiating early, 

 such as the gonads." 1171 The extent of the variegation can be changed considerably 

 by a folic-acid analog such as amethopterin, and this can be reversed by thymidine. 1173 



Variegation in maize and Drosophila are phenomena of such complexity and subject 

 to such precise genetic control that they have served as models in trying to bridge the 

 gap between mutations and differentiation and between heredity and develop- 

 ment 133, 821, 1173 



Another field in which somatic mutations and selection of clones have been impli- 

 cated recently is that of antibody formation. Much interesting discussion has followed, 

 but the relationship between the underlying changes and genetic phenomena versus 

 differentiation is entirely a matter of conjecture at the present time. The interested 

 reader is referred to the thoughtful and provocative theoretical treatises of Talmage, 1308 

 Burnet, 149 Lederberg, 772 and Szilard. 1306 



NEOPLASTIC CELLS in vivo 



Compared to normal tissues, neoplastic cells offer one advantage as tools for the 

 study of somatic variation: their ability for unlimited multiplication under appropriate 

 circumstances facilitates transplantation experiments and progeny tests of various 

 kinds. It must be kept in mind, however, that the findings are not necessarily 

 applicable to normal cells. Furthermore, transplantation per se will impose a selection 

 pressure of its own on the cell population. When purposefully used, such selective 

 pressures can be helpful in concentrating and identifying rare variants that would have 

 escaped detection otherwise. On the other hand, continuous selection in the course 

 of serial transfer will remodel the population, and variation as found in transplanted 

 tumor cells must not be extrapolated, without further inquiry, to the original tumor. 



With tumor cells, as with normal somatic cells, the main approaches presently 

 available for the study of genetic variation can be registered as the use of phenotypic 

 marker characteristics and the detailed morphologic examination of the chromosomes. 

 In addition, the relationships between cellular phenotype and viral infection, particu- 

 larly as regards tumor viruses, have received increasing attention recently, and the 

 concept of infective heredity has entered the tumor field. 



