146 



BIOLOGIC BASIS OF SEX 



many cases extremely close; moreover, some 

 of the apparent discrepancies have since 

 been found to be due not to fundamental 

 differences in the two types of hormone but 

 to differences in experimental conditions 

 as regards such factors as timing and dos- 

 age. The most important objections to ster- 

 oid hormones of adult type as controllers of 

 sex differentiation have been noted previ- 

 ously in various connections. However, a 

 brief recapitulation is in order: they are 

 (1) the frequent occurrence of paradoxical 

 effects; (2) the failure of male hormones to 

 inhibit effectively the Miillerian ducts of 

 mammalian embryos; and (3) their failure 

 in nearly all cases to have significant ef- 

 fects on the differentiation of mammalian 

 gonads. The first objection has been dealt 

 with (p. 141) and will not be disscussed fur- 

 ther. In the other cases the failure is not 

 absolute ; moreover, it is confined to a single 

 group, the mammals, and is not of general 

 application. To an extent, group or species 

 differences may be involved. In the hedge- 

 hog, for example, the funnel region of the 

 oviduct is inhibited by male hormone 

 (Mombaerts, 1944), and in female opos- 

 sums the vaginal portion is suppressed on 

 one or both sides in about 50 per cent of all 

 individuals (p. 114). Furthermore, in mouse 

 and rabbit embryos male hormone prevents 

 the union of the posterior ends of the Miil- 

 lerian ducts to form the vaginal canal (Ray- 

 naud, 1942) and corpus uteri (Jost, 1947a). 

 These partial effects in themselves require 

 explanation. Failure of steroid hormones to 

 reproduce more fully the effects of the em- 

 bryonic hormone may lie, at least in part, in 

 experimental conditions other than the type 

 of hormone. On the other hand it is possible, 

 as suggested by Jost (1953, 1955), that in 

 mammals a special substance is required for 

 the inhibition of the Miillerian ducts other 

 than the ordinary testis hormone. 



The failure of steroid hormones to modify 

 the gonads of placental mammals (even 

 when the accessory sex structures are pro- 

 foundly transformed) is in marked contrast 

 with the striking results obtained in many 

 lower vertebrates and in the opossum. It is 

 also at variance with the strong modifica- 

 tions usually found in freemartin gonads, 

 and this has often been cited as proof that 



different types of hormone are involved. 

 However, the freemartin is still almost 

 unique among mammals as an example of 

 gonad transformation induced by another 

 embryonic gonad, and may yet prove to be 

 a special case of a type peculiar to the bo- 

 vine family.^^ The possibility must be con- 

 sidered that in placental mammals gonad 

 differentiation (as opposed to the differen- 

 tiation of the accessory sex structures) has 

 come to be under direct genotypic control; 

 nevertheless, the demonstration, after many 

 earlier failures, of a thoroughgoing trans- 

 formation of the testis in the opossum sug- 

 gests that certain essential experimental 

 conditions have perhaps not been fully re- 

 alized. In any case, it may be a difficult 

 matter to determine whether the refractori- 

 ness of mammalian gonads to steroid hor- 

 mones is indeed due to a fundamental dif- 

 ference in the character of the hormones 

 themselves or to a change in the status of 

 the embryonic gonads affecting their reac- 

 tivity to hormones. 



In many other situations it appears that 

 embryonic and adult sex hormones are in- 

 terchangeable without observable differ- 

 ences in the results. Testosterone propionate 

 or methyl-testosterone, administered to cas- 

 trated rabbit embryos at the time of opera- 

 tion, prevent the usual castration changes in 

 all male structures, in this respect fully re- 

 placing the embryonic testis (Jost, 1947b, 

 1950, 1953), although they do not inhibit 

 the Miillerian ducts. A similar effect of tes- 



"'' There is, in fact, a notable scarcity of free- 

 martins in a strict sense in other groups in which, 

 on the grounds of placental fusion, the phenome- 

 non might be expected to occur at least occasion- 

 ally. For the literature on scattered cases inter- 

 preted as freemartins, see Willier (1939) and 

 Witsclii (1939); and for a case (the marmoset, 

 Wislocki) in which no freemartin effect was found 

 although the essential conditions seemed to be 

 present, see Witschi (1939). It is possible that the 

 piesence of the hormone is not the only factor to 

 be considered ; lack of reactivity on the part of the 

 gonad may be the principal factor and one which 

 may vary in different groups, correlated perhaps 

 with the presence of maternal or placental estro- 

 gens during pregnancy. There is still a surprising 

 lack of information for many groups; for exam- 

 ple, the freemartin condition in sheep (at least as 

 regards sterility) may occur more frecjuently than 

 lias been supi)oyed (sec Stormont, Weir and Lane, 

 1953). 



