erythrocytes. It is partly explained by acute phase increases in plasma fibrinogen concentration and viscosity. In particular, however, we found that the filterability of both polymorphonuclear and mononuclear leucocytes is reduced in acute cerebral infarction. This may not only explain the reduced filterability of blood but also promote malperfusion of the microcirculation and infarction.Reduced filterability of leucocytes may be part of the acute phase response to injury, as suggested by the correlation with fibrinogen concentration in the patients with stroke and by similar findings in the controls with chest infection. Preceding infection is an important risk factor for cerebral infarction.5 The risk of infarction also increases with age, and we observed decreased filterability of leucocytes in controls aged over 40. We suggest that mechanisms by which increasing age and infections increase the risk of cerebral infarction include increased plasma fibrinogen concentration and viscosity and decreased filterability of leucocytes, each of which tends to reduce capillary flow. Women attending our breast clinic with cyclic breast pain of at least two months' duration were included in the study. Women not menstruating regularly, taking the contraceptive pill, or with breast abnormalities were excluded. The women were asked to record their pain on a linear analogue scale from one to 100 daily from the first day of menstruation.2 The nodularity of the breasts was assessed by palpation on one day of the luteal phase of the cycle each month.Patients who kept full records during a control month were randomised into the six month trial. They were asked to apply 5 g of cream each evening from the tenth day of one menstrual cycle to the first day of the next. Two different preparations in identical containers labelled first or second were used; one contained 1% natural progesterone and the other was without the active hormone. The active and placebo creams were applied for three months, each in random order, and only the statistician knew the order of the treatments. For each patient a mean pain score was calculated from the seven worst daily pain scores during the premenstrual phase each month. The seven worst days were usually consecutive and most often the days before the next period. The pain and nodularity scores for each patient were analysed by non-parametric tests, which compared differences within patients.Thirty two women were entered into the study; 17 were randomised to use the active drug first and 15 the placebo. Seven did not return for their monthly appointments (usually after the first month). Thus complete data were available for 25 of the women, 14 of whom used the active treatment first and 11 the placebo.The data were first analysed for trends in the pain and nodularity scores over time, but no evidence of any such trends was found. To compare the drugs a mean pain score and a total nodularity score were calculated for each patient's three month exposure to each drug. The results (table) showed a small...

No abstract

Plasma levels of LH, FSH, oestradiol 17 beta and testosterone were measured in twenty-two men with gynaecomastia. The group was divided by age into those under and those over 50-years-old and the mean hormone levels of the two groups were compared with two groups of age-matched normal men. In the young men with gynaecomastia LH and consequently the LH:FSH ratio was lower than in controls. Older patients with gynaecomastia had higher values of both LH and FSH than normal controls but the LH:FSH ratio was similar in the two groups. A pulsatile pattern of LH was present only in the young controls. Older controls had higher FSH levels than younger controls. In older men with gynaecomastia oestradiol levels and the oestradiol:testosterone ratio were higher than in those without.

There is an association of pernicious anaemia with iron deficiency anaemia (Faber and Gram, 1924). There is also a high incidence (85 per cent) of gastric parietal cell antibody in the sera of patients with pernicious anaemia (Taylor, Roitt, Doniach, Couchman and Shapland, 1962) and a less marked, but significantly increased incidence (33 per cent) in patients with iron deficiency anaemia who have a histamine fast achlorhydria (Dagg, Goldberg, Anderson, Beck and Gray, 1964). Pernicious anaemia has a familial incidence (Askey, 1940; Callender and Denborough, 1957) and there is also an increased frequency of achlorhydria in relatives of patients with pernicious anaemia (Askey, 1940; Neel, 1947). In addition te Velde, Abels, Anders, Arends, Hoedemaeker and Nieweg (1964) have found the gastric parietal cell antibody in 20 per cent of relatives of patients with pernicious anaemia, compared with 6 per cent in a control series. Because of the association of iron deficiency anaemia with pernicious anaemia and the increased incidence of histamine fast achlorhydria and parietal cell antibody in both diseases, the occurrence of the antibody has been assessed in the families of patients with iron deficiency anaemia, histamine fast achlorhydria and gastric parietal cell antibody. In a group of 22 patients with iron deficiency anaemia, histamine fast achlorhydria and gastric parietal cell antibody, Dagg, Goldberg, Gibbs and Anderson (1966) found that seven had a 58Co vitamin B12 absorption test (Schilling, 1953) characteristic of pernicious anaemia, and six of these had reduced serum vitamin B12 levels; that is these patients had latent pernicious anaemia. In 10 of the 22 patients the serum vitamin B12 levels and the Schilling test were normal, and in the remaining five patients the serum vitamin B12 was low but the Schilling test was normal. Since it appeared possible that the familial incidence of gastric parietal cell antibody would be higher in relatives of the patients with latent pernicious anaemia than in relatives of those without latent pernicious anaemia, 11 families from these 22 patients were investigated, six of which had propositi with latent pernicious anaemia and five of which had propositi without latent pernicious anaemia (Table I). The relatives of the remaining 11 patients were unwilling to co‐operate in these investigations.