SUMMARY After the administration of [4-14C]lynestrenol (17α-ethynyl-19-nor-androst-4-en-17β-ol) to 7 human subjects 31·–57·6% of the dose, whether administered orally or i.v., was excreted in the urine within 5 days. The biological half-life of radioactivity was 26·5 hr. After acid and enzymatic hydrolysis, 58·7 and 45·6% respectively of the urinary radioactivity was extractable. About 10% of the urinary metabolites were excreted as sulphate conjugates. A mean value of 1·75% of the administered dose was converted to phenolic compounds. The metabolites in the free fraction and enzymehydrolysed extract of urine were almost entirely polar compounds, whereas 70% of the metabolites in the sulphate fraction were much less polar. The chromatographic evidence showed that hydroxylation of lynestrenol must have occurred at two points in the molecule. Plasma radioactivity decreased more rapidly than after administration of norethisterone.

Some observations by Jeffcoate (1938) (Reynolds, 1949).The work of Csapo (1956) shows that during pregnancy progesterone reduces myometrial contractility. Short and Moore (1959) have shown that at the onset of labour there is no fall in blood progesterone, nor according to Shearman (1959) does the output of urinary pregnanediol decline in the weeks before the onset of labour. It is possible that toward the end of pregnancy some endocrine factor which overcomes the progesterone block on myometrial contractility comes into play. During pregnancy large amounts of oestriol are excreted, and in a recent study it was shown that a sharp rise in oestriol excretion occurs toward the end of pregnancy (Klopper et al., 1961). It seemed possible, therefore, that this oestrogen in particular is concerned in bringing about an endocrine milieu in which the myometrium will respond to the mechanical stimuli which bring about labour.In all the literature on the effects of oestrogen in labour we have been unable to find records of a trial controlled in accordance with modern statistical criteria. We therefore set out to test the efficacy of oestriol in such a trial. When the figures of this " double-blind " trial were analysed it became of interest to determine whether the results were peculiar to oestriol or would occur with other oestrogens also. This was therefore followed by a second, larger, trial in which the effects of oestriol were compared with those of a synthetic oestrogen, stilboestrol. MethodsFor the first trial a small homogeneous group of subjects who would show minimal variation in their obstetric performance was used. Women who were to be subjected to artificial rupture of membranes for clinical reasons were selected, but those with twins, hydramnios, antepartum haemorrhage, intrauterine foetal death, or previous caesarean section were excluded. The trial was limited to primigravidae under 30 years old who were within 21 days of their expected date of delivery. Messrs. Organon kindly supplied us with tablets containing 10 mg. of oestriol and a set of identical placebo tablets. The tablets for each patient were dispensed in advance by our statis;tical section and none of the observers knew whether oestriol or placebo tablets had been given. Oral administration was started 24 hours before the rupture of membranes and continued until delivery. The dosage was two tablets (20 mg.) initially, followed by one tablet (10 mg.) six-hourly thereafter. A record of possible side-elfects such as vomiting or flushing was kept for each patient, but such occasional phenomena were equally distributed and it proved impossible to tell from the patient's clinical behaviour whether she was having oestrogen or not. The trial was discontinued and the code broken when 44 patients had been treated.For the second trial a larger variety of patients was included and a greater number were studied in order to exclude chance concentrations of aberrant cases. The subjects in this trial were divided into four groups; (I) primigravidae 29 years old ...

The excretion of pregnanediol, pregnanetriol and the 6-oxygenated metabolites of progesterone by fifteen women has been measured throughout pregnancy. During the first trimester the level of excretion of the 6-oxygenated metabolites varied from 0\m=.\4to 2\m=.\2 mg./day and at the end of pregnancy the range was from 3\m=.\5to 11\m=.\6 mg./day. During the last 6 weeks of pregnancy, pregnanediol excretion was relatively constant, whereas in seven of the patients the excretion of the 6-oxygenated metabolites continued to increase. Except at the end of pregnancy there was a close correlation between the 6-oxygenated metabolites and pregnanediol excretion. The results suggest that the 6-oxygenated metabolites and pregnanediol have the same precursor and that little, if any, of a 6-oxygenated progesterone is secreted during pregnancy.There appears to be little increase in pregnanetriol excretion during pregnancy; six of the subjects showed a significantly higher excretion in the third trimester compared with the first trimester of pregnancy, but the values were not outside the normal range for non-pregnant subjects. The specificity of the method of Fotherby & Love (1960) for the estimation of pregnanetriol was examined when applied to pregnancy urine. Between 1\m=.\1and 2\m=.\4 % of administered progesterone was excreted as the 6-oxygenated metabolites. In only one of five subjects was pregnanetriol excretion increased after progesterone administration.

A new method of assay was applied to the determination of pregnanediol in urines from patients in whom the presence of a placenta or a corpus luteum could be excluded. The normal variation of pregnanediol content in such urines was examined. It was shown that operative stress or the intravenous administration of corticotrophin caused an increased output of pregnanediol. Quantitatively significant amounts of urinary pregnanediol could not be detected after adrenalectomy nor did adrenalectomized patients respond to corticotrophin by an increased pregnanediol output. It is considered that the adrenal contributes to the pregnanediol found in urine and that the determination of urinary pregnanediol may be of value in the study of adrenal function.Pregnane-3a:20a-diol is a neutral steroid occurring in urine as a conjugate of glucuronic acid. It is a metabolite of progesterone. When progesterone is administered to human subjects, some 10-20% of the hormone can be recovered as urinary preg¬ nanediol. It is generally supposed that the excretion of urinary pregnanediol can be used as a measure of the progesterone produced by the corpus luteum and the placenta. Pregnanediol was first isolated from pregnancy urine by Marrian [1929]. Since then many methods have been proposed for its quantitative deter¬ mination in urine. As these methods have become more reliable, the changes in excretion during the menstrual cycle and pregnancy have been more clearly under¬ stood. It has also become evident that not all the pregnanediol in urine can be accounted for by the activity of either the corpus luteum or the placenta. The work of Engel, Thorn & Lewis [1941] showed that small amounts of authentic 5/?-pregnane3a:20a-diol could be isolated from a large volume of pooled normal male urine. These findings were supported by the work of Heard & Hoffman [1941] and of Westphal [ 1944]. The latter estimated that male urine contained 0-7 mg pregnanediol/1.Although a third source of urinary pregnanediol is now established, its origin has not been conclusively demonstrated. Some of the urinary pregnanediol is thought to originate from the adrenal. Many workers have found an increased pregnanediol output in patients suffering from adrenal disorders.

1. Pregnant rats were fed either low (<1 p.p.m.) Zn or control (40p.p.m. Zn) diets from day 10 of gestation. They were killed at intervals during the last 96h preceding the normal time for onset of parturition, and differences in plasma progesterone, oestradiol-17, and ovarian 20a-hydroxysteroid dehydrogenase were assessed. 2. Gestation was prolonged in Zn-deficient rats. 3. Although the preparturient decline in plasma progesterone began at the same time in all groups, at term, plasma progesterone concentration in Zn-deficient rats remained significantly higher than in normal females. 4. Induction of ovarian 20a-hydroxysteroid dehydrogenase activity was delayed by about 8h by Zn deficiency. This delay was not observed if prostaglandin F2. was injected previously. 5. The results suggest a Zn-dependent step(s) in uterine synthesis and/or release of prostanoids.Zinc was first reported to be required for normal parturition in the rat by Apgar (1968a). Female rats fed a diet containing 1 p.p.m. or less of Zn beginning on the first day after mating displayed loss of appetite and lethargy. Harderian-gland pigment often accumulated on the face and paws. Delivery was difficult and prolonged and accompanied by excessive bleeding, and most of the neonates were abandoned and placentae were left unconsumed. Further studies (Apgar, 1968b(Apgar, , 1970(Apgar, , 1972(Apgar, , 1973(Apgar, , 1975(Apgar, , 1976(Apgar, , 1977aGombe et al., 1973) showed that such effects were relatively specific to Zn deficiency and were not induced by restriction either of total food intake (pair-feeding) or

In this paper the problems of establishing the identity of a compound isolated from biological material with a steroid are outlined and discussed. The nature of evidence required for identification and some techniques for obtaining this are briefly considered. The importance of establishing the specificity of a method for the quantitative measurement of a steroid in a biological extract is emphasized. THE problemThe meaning of the term identification used here is the complete description of the chemical constitution (or structure) of the compound isolated, or the demonstra¬ tion that its chemical and physical properties are indistinguishable from those of a reference sample of a compound of established constitution.The identification of steroids in biological extracts is made difficult by the large number and low concentrations of closely related compounds which may be present.Even when consideration is limited to steroid hormones, their precursors and meta¬ bolites in biological extracts, the number of steroids likely to be present is formidable. The use of synthetic steroids in investigations and therapy adds to the problem. It may sometimes be possible to isolate and purify enough material to enable evidence of identity to be obtained by observation of infrared and mass spectra and, if possible, melting points, optical rotation and molar extinction coefficients. Usually very much smaller amounts of purified substance will be available for examination. The advent, in recent years, of an increasing variety of sensitive physical and chemical procedures for investigating steroids is thus fortunate. The use made of these pro¬ cedures and the criteria for the identification of steroids should be reviewed from time to time. This is the purpose of the present paper.Consideration might be given to three aspects of the problem : (a) the identification of unlabelled steroid-like compounds isolated from biological materials;(b) the identification of isotopically labelled steroids, when such steroids are formed in experiments with labelled precursors ;(c) the specificity of methods intended for quantitative estimation of steroids in biological extracts.

No abstract