Trowel1: Treatment of Kwashiorkor 723 continue for two or three years and often with considerable benefit. This was often possible in Africa under the conditions of marital life which were a part of polygamy, but conditions and customs are changing rapidly. If a second pregnancy occurred prematurely, the first child was weaned and became ill. The second preknancy was considered to kill the first child, and the disease often carried a stigma. This illness was called kwashiorkor among the Ga people of the Gold Coast, among whom Doctor Cicely Williams was working when she published her accounts of this disease (1933). Other tribes have other names and about fifty different terms have been employed by different doctors. These names often refer unfortunately to multiple deficiencies, possibly those of the vitamins, and these terms have crept into the literature, especially that of South America. Yet treating these cases mainly by multiplevitamin therapy often results in fatalities. Recently it has become clear that Czerny & Keller (1906) were describing an identical condition in Germany under the name of MehlnSihrschaden. Their description of the disease, however, was very incomplete. No changes were noted in the pigment of the hair and the pancreatic lesion was not described; so that subsequent workers, such as Vdghelyi (1948, 1950), who almost rediscovered the condition in Hungary and called it “nutritional edema,’’were unable to identify this disease with Mehlnahrschaden. The old and correct teaching of the German pediatricians was kept alive and developed in Italy, where Frontali (1952) has confirmed the presence of the pancreatic lesion and the hair changes in this disease (called “starchy food dystrophy” in that country). This disease is present in nearly 5 per cent of children in Southern Italy. The presence of the characteristic “enamel-paint dermatosis” and “flexural fissures” of kwashiorkor has however not been described in any case of Mehlnahrschaden, or “starchy food dystropy.” In the literature of the United States, a few cases of kwashiorkor probably have been described, such as that of McEnery (1933). It is dacult to evaluate the more numerous reports on the cases of “nutritional edema” seen in young children in the United States and in other countries. In cases of “nutritional edema” in infancy, there has not been described the dermatosis, the fissures, the dyspigmentation, thc pancreatic lesion, or the high mortality, which are Characteristic features of kwashiorkor.“h’utritional edema” was ascribed at one time to protein deficiency and it is probable that many cases of this disease, especially among young children, were cases of kwashiorkor, and were not cases of hunger edema due to calorie deficiency, or beri-bcri due to thiamine deficiency, all three of which are different varieties of “nutritional edema.)’The latter is not a suitable term for kwashiorkor, for some severe cases show no edema and mild cases of kwashiorkor never show any edema. It is proposed to confine the attention to clinical aspects of the treatment of cases of severe kwashiorkor in childhood, since little is known concerning that of the milder state. There is fairly general agreement that mild kwashiorkor is probably very common in many tropical countries. It is, however, difficult to define (FIGURE 1). Thus, Welbourn (1953) has found signs suggestive of mild kwashiorkor in 46 per cent of the children attending child welfare clinics near Kampala, Uganda. Wherever possible, treatment and prevention of