A phase III 2 X 2 factorial trial of cisplatinum and bleomycin in 116 patients with recurrent or advanced squamous cell carcinoma of the head and neck is reported. Thirty percent of patients proved to be unfit for chemotherapy, and of those treated progression of tumour was the commonest "response". However, 25% of patients achieved a partial or complete response, with no significant difference in response rates between the treated arms. The median number of courses received was 1 (range 0-6) and the commonest causes for discontinuation of treatment were renal toxicity and death. Bleomycin reduced survival, but not significantly so, whereas cisplatinum prolonged median survival significantly by 10 weeks. Significant predictors of survival, in addition to treatment by cisplatinum, were age, performance status, N status, number of courses and response of the tumor.

No abstract

This paper is intended to convey the essence of Taguchi's design approach to chemists and others with an interest in chemometrics. Although most Taguchi‐style applications worldwide have been in electronics and in elaborately transformed manufactures, examples are increasingly found in chemical processes and in the food industry. Foremost among Taguchi's contributions is the concept of designing processes and products to be robust to the uncontrollable environmental influences which they experience during their operation or lifetime. This concept is explained with a worked example.

Summary Blood sugar, serum insulin and growth hormone levels in response to 100 g. of oral glucose were studied under carefully controlled conditions on days 9, 18 and 27 of the menstrual cycle in 20 healthy women aged 18 to 20 years. The results indicate that blood sugar and serum insulin levels in response to oral glucose loading do not alter significantly in the menstrual cycle. Seven of the women had repeat studies on the same days three cycles later. No significant change in blood sugar or serum insulin levels occurred between cycles. The fasting growth hormone level was highest at the time of impending menstruation in 11 subjects (55 per cent) but the difference in mean values for the group as a whole was not statistically significant. It is concluded that the changes in endogenous hormone levels in a menstrual cycle do not produce significant alterations in carbohydrate metabolism in normal women in this age group. The reported divergence of blood glucose and insulin levels after oral glucose in the menstrual cycle may reflect the interaction of other factors such as age and parity on glucose tolerance. It is suggested that investigations into the metabolic effects of oral contraceptive steroids on these parameters should be performed in nulliparous women in a similar age range.

Many two-variety competition experiments have plants of the two varieties systematically arranged on a triangular lattice of hill plots . Designs, balanced in a certain sense, are introduced and their advantages explained. An example of such a design is presented and compared with one of a more conventional nature.