SUMMARY A prolonged study relating plasma testosterone levels to coitus is reported in one male subject. Testosterone levels in samples taken during and immediately after sexual intercourse were significantly higher than those found under resting conditions. A significant difference was not apparent between testosterone measurements taken before and after orgasm, and coitus did not appear to affect plasma luteinizing hormone levels. A marked nyctohemeral rhythm was observed in testosterone levels which tended to be higher early in the day. Masturbation had no significant effect on plasma testosterone levels in seven subjects.

A method for the quantitative determination of gonadotrophins in the urine of non-pregnant human subjects is described.The reliability criteria for the method have been studied and the technique has been found to be satisfactory in terms of accuracy, precision, sensitivity and specificity.The importance of expressing results of urinary gonadotrophin assays in terms of a reference preparation such as HMG 20A is emphasized.The aim of the present paper is to describe a quantitative method for the estimation of gonadotrophins in urine of non-pregnant human subjects and to evaluate the method in terms of accuracy, precision, specificity and sensitivity. The method essentially consists of adsorption of the gonadotrophins on to kaolin, elution from kaolin, precipitation by acetone and, in the case of low-titre urines, treatment of the crude kaolin-acetone powders by tricalcium phosphate. The bioassay is performed using the mouse uterus test and results are expressed in terms of a standard prepared from the urine of menopausal and postmenopausal women and termed HMG 20A. The preliminary work on which the method is based has already been described [Loraine & Brown, 1954Loraine, 1957Loraine, , 1958a. MATERIALS AND METHODSThe kaolin was acid washed (Society of Leather Trades Chemists' specification) and was supplied by British Drug Houses, Ltd. The tricalcium phosphate suspension was prepared according to the method of Swingle & Tiselius [1951] and contained approx. 0-07 g Ca3(P04)2/ml. Hydrogen-ion concentrations were measured by means of a Cambridge pH meter using a glass electrode and were uncorrected for ionic concentration.The standard (HMG 20 A) was prepared by Messrs Organon (Newhouse, Scotland) from the urine of menopausal and postmenopausal women. The kaolin-acetone method was used. The crude powder thus obtained was treated with tricalcium phosphate, subjected to fractional acetone precipitation and then dried over phosphorus pentoxide. A 'unit' was defined as the gonadotrophic activity contained in 1 mg HMG 20 A [Report of the second meeting of the Gonadotrophin Club, 1955, unpublished].

Serial studies of endocrine function have been performed in four female patients and one male patient with anorexia nervosa during treatment by refeeding. None of the females menstruated or ovulated at any stage of the investigation. Hormone assay changes were most marked in severely ill patients. Refeeding produced an increase in human pituitary gonadotrophin (HPG) output in two patients in whom pretreatment levels were below or at the lower end of the normal range. In these two subjects a rise in HPG levels during refeeding was associated with an increase in the output of 17-hydroxycorticosteroids and 17-oxosteroids. In the remaining three patients the effect of this treatment on the excretion of these hormones was less marked. Excretion values for dehydroepiandrosterone increased during refeeding in three subjects. In the females the relative proportions of oestriol, oestrone and oestradiol in urine were altered by refeeding, proportionally more oestriol being excreted.