256 THE MOSAIC STAGE OF DIFFERENTIATION 



medulla. Low temperature causes a differential inhibition of 

 medullary development with consequent delay in males of the 

 degeneration of the cortex and of the immigration of the primordial 

 germ-cells from it to the medulla. As a result, the primordial germ- 

 cells, exposed to cortical influences, become oogonia, and 40 mm. 

 tadpoles are all somatically females. Later, however, in the genetic 

 males among them, the delayed medulla succeeds in reaching the 

 stage of development requisite to inhibit the cortex, upon which they 

 become transformed into somatic males. 



High temperature, on the other hand, has a deleterious effect 

 upon the cortex, but not upon the medulla. As a result the sexes 

 are early differentiated in the normal i : i ratio, but later the females 

 show inhibition of the cortex. No further oocytes are differentiated, 

 those already embarked on differentiation degenerate after a short 

 period of further growth, the ovarian sacs derived from the medulla 

 begin to proliferate and form cords, and any undifferentiated pri- 

 mordial germ-cells migrate inwards and join the medullary cords, 

 where they differentiate into spermatogonia; thus the genetic 

 females become transformed into somatic males. Essentially 

 similar results have been obtained in Urodeles. 



These experiments clearly demonstrate the existence of local 

 sex-inductive agencies in cortex and medulla respectively. The 

 nature and action of these factors is more fully revealed by a series 

 of beautiful experiments on parabiotic twins. ^ 



Amphibian embryos are united parabiotically either side by side 

 (parallel pairs) or in series with the anterior end of one joined to the 

 posterior end of the other (chains). This has been effected both 

 homoplastically, between partners of the same species ; or hetero- 

 plastically, between different species. Here we shall confine our- 

 selves to homoplastic parabiosis, and to two-sexed pairs, in which 

 the partners are genetically of opposite sex. 



The most interesting experiments concern frogs {Rana). In 

 these, no effect on sex-differentiation is exerted in chain pairs : sex- 

 differentiation is normal both in the male and the female partner. 

 In parallel pairs, however, the sex-differentiation of one member of 

 the pair is modified. The affected partner is normally the female, 

 and the modification consists in a certain degree of inhibition of 



^ Witschi, 1932. 



