The researches of Eijkman, of Schaumann, and of Grijns led Fletcher (16), a physician in the Kuala Lumpur Lunatic Asylum in the Federated Malay States, to carry out a thorough experiment on human subjects, with the object of determining the relation of rice to beri-beri. Fletcher divided the hospital into two sections, each containing nearly equal numbers of patients. To one group he supplied milled rice, and to the other "cured" rice. "Cured" rice was prepared by soaking unhusked kernels, or paddy, for 12-24 hours in water, then heating the soaked grains over a slow fire until they burst. The rice was then spread to dry in the sun. Such rice is easily husked and is ground for use without polishing. The entire kernel, including the germ, was therefore supplied to half the patients. The diet of all the inmates of the asylum consisted of four ounces of fresh meat four times a week; five and one-third ounces of fresh fish twice a week; five and one-third ounces of salt fish once a week; eight ounces of vegetables daily; one and one-third ounces of curry stuffs daily; two-thirds ounce of cocoanut oil daily, and thirty-eight ounces of uncooked rice daily. Since rice absorbs about three times its weight of water on cooking, and becomes very bulky, it seems incredible that so large an amount of rice could be eaten by man.

The only difference in the diets of the two groups of patients was in the character of the rice.

The diets were continued unchanged throughout a year with the following results: The group eating the milled rice developed a large number of cases of beri-beri, whereas no cases developed in the group fed the steamed rice which contained all parts of the kernel. Fletcher rearranged the patients and mingled some of the beri-beri patients with the well, and continued the experiment. The result was a complete demonstration that the disease was due to the consumption of a diet consisting in great measure of meat, fish and milled rice. We now know that the same result will follow the too free use of the endosperm of any of the cereal grains as food.

It has been found by experiments on man and animal, that the germ of rice is especially rich in the protective substance, water-soluble B, a lack of which leads to the development of beri-beri. The diet of the asylum patients observed by Fletcher contained a sufficient amount of this substance when the entire grain was its chief component, but not when the milled rice was used. The diet of the institution cannot be regarded as satisfactory for the promotion of health over a long period of time even when the steamed rice was employed, but it was then sufficiently good to prevent the development of beri-beri. The eight ounces of vegetables, the nature of which were not specified, sufficed to furnish enough anti-scorbutic substance to prevent scurvy. This diet, principally derived from rice, meat, and fish would not, if we may judge from the vast amount of experimental data obtained with animals, maintain over a period of several years normal health in an adult.