472 



PHYSIOLOGY OF GONADS 



gen required to e^'okc one response often 

 is different from that required for a second 

 response. For example, the amount of es- 

 tradiol necessary to produce perineal en- 

 largement in the baboon is less than that 

 required to produce withdrawal bleeding. In 

 the human female, the order of sensitivity of 

 three tissues to estrogen is uterine cervix, 

 vaginal mucosa, endometrium (Zondek, 

 1954). Theoretically, therefore, if the re- 

 sponse of tissues is to be used in estimating 

 the amount of hormone produced, the in- 

 vestigator should follow the response having 

 the highest threshold value. Gillman also 

 called attention to another complication. 

 The minimal amount of estrogen necessary 

 to produce a response does not necessarily 

 approximate the amount being produced, 

 because, in the instance he cites, larger 

 amounts do not produce a larger perineum. 



Another comi^licating circumstance is the 

 occurrence of ''inherent" cycles (in some 

 cases the length of a reproductive cycle) of 

 responsiveness which have been demon- 

 strated in ovariectomized females receiving 

 constant amounts of estrogen from exoge- 

 nous sources. Many such animals have dis- 

 played cyclic vaginal changes (del Castillo 

 and Calatroni, 1930; Bourne and Zucker- 

 man, 1941a), uterine l)leeding (Zuckerman, 

 1940-41 ), and running activity (Young and 

 Fish, 1945). This unknown factor must be 

 taken into consideration in any attempt to 

 estimate the rate of estrogen production. 

 Existence of this factor gives emphasis to 

 the importance of direct determinations, 

 wiien they can be made, either from follicu- 

 lar fluid or from freshly drawn blood. Corre- 

 sponding considerations would hardly be 

 tliough of as applying to progesterone. Most, 

 if not all, of its actions are synergistic or 

 potentiating. Presumably, therefore, they 

 are directly dependent, not so much on any 

 changes in the inherent responsiveness of the 

 tissues as on the extent to which the tissues 

 have been conditioned or primed by the es- 

 trogen. 



A part of the picture which must be 

 brought into context with the problem of es- 

 trogen secretion comes from a review of the 

 temporal factors in a normal cyclic animal. 

 In animals with short estrous cycles (4 or 5 

 days in the rat, mouse, and hamster) . and in 



species in which the cycles are longer as in 

 the guinea pig it has generally been assumed 

 that estrogen is produced maximally at the 

 time of estrus. This is an assumption based 

 on the simultaneous occurrence of the cor- 

 nified vaginal smear and the display of es- 

 trous behavior. When one considers that 48 

 to 72 hours are necessary for vaginal epithe- 

 lium to proliferate and then degenerate into 

 cornified cells, it is obvious that the estrogen 

 which starts these changes must be elabo- 

 rated 2 or 3 days before estrus and there- 

 fore before the size of the follicles is maxi- 

 mal. Zondek (1940) demonstrated that if an 

 immature rat was injected with HCG the 

 ovaries could be removed 27 hours later and 

 the rat would exhibit vaginal cornification in 

 84 to 96 hours. This emphasizes that the es- 

 trogen which caused the cornified smear had 

 been elaborated 2 or 3 days before vaginal 

 estrus (Zondek and Sklow, 1942; Green, 

 1956). The dilation of the uterus with fluid 

 late in the proestrum (Astwood, 1939) is 

 also evidence that the efi^ective estrogen had 

 been elaborated 24 to 30 hours earlier. 



Problems of assay are involved in any 

 attempt to estimate secreted estrogen and 

 are discussed briefly by Emmens (1950a, b). 

 They are more serious in the case of the es- 

 trogens than in the case of progesterone. The 

 bioassay of estrogens is usually based on 

 vaginal cytology or change in uterine weight. 

 The vaginal cytology or smear method is es- 

 sentially the original method of Allen and 

 Doisy(kahnt and Doisy, 1928; Allen, 1932). 

 Ovariectomized rats or mice are given sub- 

 cutaneous injections of the substance being 

 tested for estrogenic i)otency. Smears are 

 made of the vaginal contents 24, 48, 60, and 

 72 hours later. If the smear reveals the pres- 

 ence of cornified cells 60 to 72 hours after 

 the first injection, the tested substance is 

 judged to be estrogenic. By using groups of 

 animals at each of several dose levels, the 

 minimal eft'ectiA-e dose can be judged. The 

 smallest amount of substance which will 

 jiroduce cornified smears in 50 to 70 ]ier cent 

 of the test grouj) is usually designated as the 

 rat or mouse unit. Intravaginal tests, intro- 

 duced by Berger (1935) and by Lyons and 

 Templeton (1936), and refined, especially 

 by Biggei's ( 1953) and by Biggers and Clar- 

 ingbold (1954, 1955), are 200 times more 



