In female cichlids both histochemistry (Livni 1971; Yaron 1971) and
ultrastructural morphology (Nicholls and Maple 1972) show that most
ovarian steroidogenesis is probably located in two kinds of follicular cells:
the granulosa cells (although the ultrastructural evidence is here somewhat
equivocal) and special theca cells, located mainly close to capillaries.
Estradiol-17-0 (E2) seems to be of generaloccurrencein fish. It is mainly
involved in the control of vitellogenin synthesis in the liver and of mobiliza-
tion of mineral and fat stores (Olivereau and Olivereau 1979; Mugiya 1978).
This seems to be true also in S. aureus where estradiol-170 was identified in
the ovary by Katz et al. (1971). Later a positive correlation was found
between plasma E2 concentration and ovarian weight, i.e., the stage of
vitellogenesis (Yaron et al. 1977; Terkatin-Shimony and Yaron 1978). In the
same species, some ovarian steroidogenic pathways were investigated by
Eckstein (1970) and Eckstein and Katz (1971), leading to the identification
of new steroid metabolites like 11-ketotestosterone and ll-hydroxytestos-
terone among others which are more classical (testosterone and progesterone).
Their exact function, although related to sexual activity (Katz and Eckstein
1974) is unknown. Deoxycorticosterone, a corticosteroid which can be
synthesized either in the interrenal tissue or in the teleost ovary (Colombo
et al. 1973) was found to increase 38-fold in the blood of S. aureus between
a sexual resting phase at 18 to 20°C and sexual activity at 30°C. As this
steroid is assumed to be one of those mediating gonadotropin action on final
maturation and ovulation in catfish, Heteropneustes fossilis (Sundararaj and
Goswami 1977), it is tempting to propose a similar model for final maturation
controi in Sarotherodon. However, the state of "sexual activity" (stage of
ovarian development) at which the blood of female Sarotherodon was
collected by Katz and Eckstein (1974) was not known precisely and further
work is needed to find out which steroids mediate pituitary gonadotropin
control over final oocyte maturation and their mechanism: for reviews on
this problem in other teleosts, see Jalabert (1976) and Sundararaj and
Goswami (1977).
Prolactin also has some control over ovarian steroidegenesis as can be
inferred from the observation of Bliim and Weber (1968) that ovine prolactin
greatly stimulates the activity of 30-steroid dehydrogenase in ovaries of the
cichlid Aequidens pulcher. This observation suggests that inhibition of
ovarian growth by prolactin is mediated through steroid action. An under-
standing of the precise mechanism for this action would be of great interest.
As in higher vertebrates, gonadotropin secretion by the pituitary appears
to be regulated by the nervous system through the hypothalamus which
secretes a releasing factor first indicated in fish (Cyprinus carpio) by Breton
et al. (1971) and Breton and Weil (1973). This factor is probably not very
different from mammalian luteinizing hormone-releasing hormone (LH-RH)
(Breton et al. 1972, 1975a) at least as far as the biologically active part of
the molecule is concerned. Recent studies show that LH-RH's from lower
vertebrates, including S. mossambicus, differ from those of higher verte-