Beriberi, White Rice, and Vitamin B: A Disease, a Cause, and a Cure. By Kenneth J. Carpenter. Pp. 282. (University of California Press, Berkeley and Los Angeles, CA, 2000.) 24.95, ISBN 0-520-22053-6, hardback

‘It seems a common observation’, remarks the author of this interesting book, ‘that men on their own are not very good at feeding themselves’ (p. 195). To be sure, beriberi seems to have affected predominantly institutionalized men, although women in the final trimester of pregnancy and their infants have also been at risk. The issues raised by this point would take one far from the central theme of this essentially narrative account; but it is one of many merits that it should stimulate the reader with such apposite aperçus. The approach is largely chronological for the first half of the book. Carpenter starts with early observations of the mysterious illness in Japan denoted by the term kakké, derived from the Chinese chiao ch’i (leg disease), and brings the reader right through to recent trends in research on thiamine. Japanese science, influenced in part by Chinese medical texts dating back to the 7th century AD, described the seasonality of the disease, its effects on the limbs and trunk, the involvement of the heart, and the often fatal outcome; and its sources and cures were conceptualized in terms of classical humoral theory. Western observers from the second half of the 19th century attributed the disease to a range of causes depending on the succession of concepts held to be plausible at a given time: poisonous ‘miasma’ or gaseous agents emanating from damp soil; microbial agents, following the work of Pasteur and Koch; the general diet; toxins; antinutrients; and the surprisingly difficult concept of nutritional deficiencies. The principal theme of the book is indeed the absorbing interplay between competing hypotheses, experimental tests, and epidemiological observations which culminated in the award of a Nobel Prize in 1929 to Christiann Eijkman and Gowland Hopkins for the discovery of ‘vitamines’. Between 1888 and 1895 in Java, Eijkman undertook a series of elegant and seminal feeding experiments on chickens. As so often, however, serendipity was critical. Eijkman had been feeding cooked white rice, left over from a military hospital, to birds, which rapidly developed the disease (loosely termed a ‘polyneuritis’); but a cook newly appointed in 1889 had refused ‘to give military rice to civilian hens’, forcing Eijkman to revert to uncooked feed-grade (unpolished) rice, thereby demonstrating the protective effect of lighter milling. Carpenter’s fascinating table (pp. 40–41), showing the rationale and outcome of each experiment in Eijkman’s first sequence of studies, should be read by anyone interested in the realities of hypothetico-deductive science. Subsequent work was undertaken in the Far East and by all the major colonial powers in mainland and insular South East Asia. Carpenter summarizes the studies by Grijns in Java, and by Braddon, Wright, Fletcher and Fraser in Malaya and