The short cross-lines about three-fourths up on the left sides of the undulations in Nos. 10, 11, and 12 designate the points at which the crank crosses the perpendicular at the top. There is quite a pressure, and it is a little odd that it should be found at this point; it can only be attributed to ankle-action back of the natural dead centre. In No. 6, and to some extent in all the others, observe the jagged appearance in the general advance of the curves, which must be due to vibration: these results were all obtained upon tolerably smooth roads, mostly in Druid Hill Park, Baltimore. No. 6 was taken upon a road perhaps a little rougher than the track around the lake, but still upon an unusually smooth surface, and it was a surprise, not to say an alarming discovery, that this vibration should occur under such circumstances. The lake track, upon which results 2 and 3 were found, was in perfect condition, smooth as a surface-plate, and without the customary sprinkling of pebbles so common when dry weather has loosed the settings of these tiny obstructions and suffered them to roll out upon the roadway; yet these figures show the saw-teeth, and I have been unable to find a road smooth enough, or jointed machine frames and springs good enough, to make unwavering symmetrical lines. These little deviations in the curves always seem to show themselves to the extent of several pounds in height in spite of all alleviating conditions, suggesting that we have much to strive for in the construction of the ideal wheel free from all concussion. In order to judge accurately of the total amount of power to turn the wheel, we have to consider the register of both pedals superposed, as in No. 8, but the curve made upon one generally answers all purposes. The possibilities in this study are unlimited, and, with a perfectly-accurate instrument, it strikes me, the results of much more definite bearing than those acquired in the silly practice of testing machines by the strength of men.