56 W. G. SMITH : explained that at the time his attention had been distracted, it was arranged to make a series of experiments in which distraction was caused by reading an interesting book. These tests, carried out with one of the subjects whose antagonistic reaction was extremely constant, did not show any resultant appreciable effect on the form of reaction. A large number of observations were made with the object of determining whether the reaction had relation to any special set of muscles. Varying position of hand and arm and the use of first or second finger appeared to make no noticeable difference. Three persons, whose customary re- action was antagonistic, found that they could depress the finger without any preliminary lifting. It is, however, to be noted that one of these subjects was less constant in his form of reaction, while another showed in the curve of depression the rounded form characteristic of very slow muscular response. After attempting to secure the relatively isolated action of the finger muscles without the use of special apparatus, a series of determinations was made on two subjects, very constant in reacting antagonistically, in which the forefinger was firmly clamped between the second and third joints, the tip of the finger resting as before on the button of the sphygmograph. The mode of reaction was by no means so constant as before, but there was evidence of persistence of the antagonistic form, the evidence being less decisive in one subject. Similar experiments were made with the same subjects in which the back of the finger just behind the nail was laid on the button of the sphygmograph ; the reaction movement in this group was carried out by flexor muscles, not as in the former experiments by extensor muscles. The result was similar to that in the previous group of experiments : the antagonistic reaction was present, but in a less clear and decisive form. It must however be noted that the effort to innervate the finger muscles alone, the upper part of the finger being held firmly, is an exceed- ingly difficult and unsatisfactory performance. It was only occasionally, according to the record of the reagent's observa- tions, that it was possible to make a satisfactorily rapid and isolated contraction of the muscles in question : the effort usually brought with it at the same time involuntary and sometimes energetic contractions of the muscles of the arm, neck and trunk, or of all these together. In these circum- stances I do not feel inclined to lay any great stress on the experimental results. A much more satisfactory set of observations was obtained in determining the behaviour of the arm muscles in reaction
