THE GENETICS OF TUMOR TRAXSPLAXTATIOX 303 



clear and persistent difference. The amount of temporary growth which 

 each exhibited was also different. The use of stocks of mice in which 

 temporary growth of transplanted neoplasms is followed by regression gives 

 a very delicate physiological test of the nature and activity of that tumor. 

 A series of tumors compared in this way often reveals more subtle and minute 

 differences than are detectable by any other known test. Cloudman has 

 made an intensive comparative study of the transplantation of mammary 

 tumors arising spontaneously in a single mouse and has shown that in the 

 case of three adenocarcinomas of the breast very different genetic constitu- 

 tions were involved. These tumors appeared at essentially the same time. 

 It is clear that the transplantation test provides a method of determining 

 whether these three tumors were independent primary growths or metastases 

 of the same primary neoplasm. 



It is also evident that by a comparison of the genetic factors in such 

 a series of tumors much information can be derived as to the factors which 

 all possess in common and those w^hich are specific to a single growth. It 

 is quite conceivable that if extensive studies of this type were made we 

 might, by plotting the relationships of the genetic factors, obtain a valuable 

 picture of the process of tumor formation as a whole from a biological point 

 of view. 



Similarly during the lifetime of an individual successive neoplasms occur- 

 ring at intervals as the age of the animal increases may be maintained 

 through transplantation and studied in comparison with one another to 

 find out whether older animals give rise to tumors which are character- 

 istically different from those produced by younger ones. All this type of 

 work in its various implications should contribute very definitely to our 

 knowledge of the process of disintegrating individuality in ageing animals. 



Transplantation experiments and somatic mutation. — The question 

 of somatic mutation is discussed further in Chapter 6. For the present it 

 will suffice to point out that the occurrence of mutations in transplanted 

 tumors which increase the percentage of takes of these tumors is a well- 

 established phenomenon supported by the work of Strong, Bittner and 

 Cloudman. Ordinarily in tumors involving a number of genes these changes 

 also aft"ect more than one gene, also there have been cases where apparently 

 a change in a single gene resulting in a change from a two factor to a one 

 factor ratio has been observed. In these tumors the mutational change is 

 clearly somatic since the tumor in which the changes occur is composed of 

 somatic and not germinal tissue. Tyzzer and many others subsequently 

 have suggested that the change from a normal to a tumor cell may be in 



