Subsequent to these investigations was that of Hollingworth[312] which is at once the most comprehensive, carefully conducted, and scientifically accurate one yet performed. He employed an ample number of subjects in his experimentation; and both his subjects, and the assistants who recorded the observations, were in no wise cognizant of the character or quantity of the dose of caffein administered, the other experimental conditions being similarly rigorous and extensive. The purpose of his study was to determine both qualitatively and quantitatively the effect of caffein on a wide range of mental and motor processes, by studying the performance of a considerable number of individuals for a long period of time, under controlled conditions; to study the way in which this influence is modified by such factors as the age, sex, weight, idiosyncrasy, and previous caffein habits of the subjects, and the degree to which it depends on the amount of the dose and the time and conditions of its administration; and to investigate the influence of caffein on the general health, quality and amount of sleep, and food habits of the individual tested. To obtain this information the chief tests employed were the steadiness, tapping, coordination, typewriting, color-naming, calculations, opposites, cancellation, and discrimination tests, the familiar size-weight illusion, quality and amount of sleep, and general health and feeling of well-being. A brief review of the results of these tests is given in the tabular summary. From these Hollingworth concluded that caffein influenced all the tests in a given group in much the same way. The effect on motor processes comes quickly and is transient, while the effect on higher mental processes comes more slowly and is more persistent. Whether this result is due to quicker reaction on the part of motor-nerve centers, or whether it is due to a direct peripheral effect on the muscle tissue is uncertain, but the indications are that caffein has a direct action on the muscle tissue, and that this effect is fairly rapid in appearance. The two principal factors which seem to modify the degree of caffein influence are _body weight_ and _presence of food_ in the stomach at the time of ingestion of the caffein. In practically all of the tests the magnitude of the caffein influence varied inversely with the body weight, and was most marked when taken on an empty stomach or without food substance. This variance in action was also true for both the quality and amount of sleep, and seemed to be accentuated when taken on successive days; but it did not appear to depend on the age, sex, or previous caffein habits of the individual. Those who had given up the use of caffein-containing beverages during the experiment did not report any craving for the drinks as such, but several expressed a feeling of annoyance at not having some sort of a warm drink for breakfast.