ТЪе tubes were kept on ice (for up to 15 min) until centrifugation, after which the plasma was drawn off, frozen in plastic tubes on dry ice, and stored at -70°C until assay. Pituitaries were excised, frozen, and stored in the same manner as the plasma. Testes were weighed except in the case of immature fish that had threadlike testes too small (GSI < 0.1%) to be weighed with accuracy. All sham-operated fish rematured except for one male, which was excluded from further study. The completeness of castration and the presence and integrity of the Silastic capsules were always checked by visual inspection. Testicular remnants were found only in one fish, which was excluded from further study. Also, at the end of the experiments, the capsules were full of androgens. Lengths, weights, and GSI for the various groups of fish are shown in Table 1.
Gonadotropins were measured by heterologous RIA for coho salmon gonadotropins as described previously. The FSH RIA has a cross-reactivity of 6.6% with coho salmon LH, whereas the LH RIA does not cross-react with FSH (< 0.1%); neither RIA reacts with growth hormone, somatolactin, prolactin, or thyroid-stimulating hormone. Each pituitary was sonicated in 0.3 ml 0.075 M sodium barbital, 10 mM EDTA, 1 mM PMSF (pH 8.6) and centrifuged at 10 000 X g; the supernatant was assayed for both FSH and LH. Serial dilutions of Atlantic salmon pituitary extracts showed displacement curves parallel to those obtained with purified coho salmon FSH and LH, respectively, in the RIAs (data not shown). In a previous study, parallelism was found both for pituitary and for plasma samples from Atlantic salmon with the same RIA. buy asthma inhalers