Recognized HTML document


fee, already described, for admission. That just defrayed the working expenses.

I t is by no means 'easy to select suitable instruments for such a purpose. They must be strong, easily legible, and very simple, the stupidity and wrong-headedness of many men and women being so great as to be scarcely credible. I used at first the instrument commonly employed for testing the force of a blow. It was a stout deal rod running freely in a tube, with a buffer at one end to be hit with the fist and pressing against a spring at the other. An index was pushed by the rod as far as it entered the tube in opposition to the spring. I found no difficulty whatever in testing myself with it, but before long a man had punched it so much on one side, instead of hitting straight out, that he broke the stout deal rod. It was replaced by an oaken one, but this too was broken, and some wrists were sprained.

I afterwards contrived, and used in a subsequent Laboratory, a pretty arrangement that gave the swiftness, though not the force of the blow, with absolute safety, and which could be used for other limbs than the arm. The hand held a thread, the other end of which was tied to an elastic band, capable of pulling it back faster than any human hand could follow ; so the hand always retarded its movement. Its speed was shown by the height to which a bead, actuated by the string (it is needless to explain details), was tossed up in front of a scale. This never failed, and was perfectly easy to manipulate.

The observations made in this Laboratory were of great use to me later on. Four hundred complete