THE tumours which appear in the intrasplenic ovarian graft in castrated rats, mice, guinea-pigs or rabbits apparently are due to the uncontrolled gonadotrophic activity of the prehypophysis, the ovarian steroid hormones which normally control hypophysial activity being inactivated during their passage through the liver before reaching the general circulation (Iglesias, Mardones and Lipschutz, 1953). The most striking proof in favour of this explanation is given by the administration of ovarian hormones. When oestrogen is administered the intrasplenic ovarian graft in the castrated guinea-pig offers the aspect of a normal ovary; hemorrhagic follicles and irregular luteal cords or nodules so characteristic of the intrasplenic graft in the guinea-pig fail to appear and seemingly normal corpora lutea are present (Lipschutz, Iglesias, Bruzzone, Humerez and Peinaranda.. 1948). On the other hand, corpora lutea, but also luteomata, are counteracted when progesterone is administered for several months (Mardones, 1948;Iglesias, Lipschutz and Mardones, 1950; Mardones, Bruzzone, Iglesias and Lipschutz, 1951). In mice the ovarian tumours arising in intrasplenic grafts can be prevented when oestrogen or androgen is given (Li and Gardner, 1949).It seemed, at first sight, rather contradictory when it was found that inactivation of oestrogen under these experimental conditions was in general incomplete, as evidenced by the increase of the weight of the uterus and vagina (Bernstorf, 1951), and that granulosa-cell tumours and luteomata may develop in the intrasplenic ovarian graft in castrated rats with persistent oestrus (4 cases described by Lacour. Oberling and Guerin, 1951).In the present paper the question is examined whether ovarian tumourigenesis in experiments of long duration in the guinea-pig presupposes suppression of oestrogenic action in the body. We shall be able to show (1) that this is not the case, and (2) that the hormonal imbalance from which ovarian tumourigenesis under the given experimental conditions derives can, and must, be expressed in quantitative terms.For the purpose of this study the secretory activity of 15 intrasplenic grafts, as listed in Table I of our previous paper (Iglesias, Mardones and Lipschutz, 1953) has been checked by the histological examination of the vaginal mucosa. the uterus and the mammary glands. The material included 8 grafts with luteomata and 7 without lutoomata. A series of 9 experiments is added in which after castration and intrasplenic grafting minute quantities of oestrogen were administered during 26 to 27 months.
15