Inielligence and Miscellaneous Articles. 477 



that is to say, the relation established which exists between the 

 duration of the current, supposed constant, and the deflection it 

 produces upon the galvanometer-needle. For this purpose M.Pouillet 

 made use of a contact-breaker consisting of a glass disk carrying 

 on one of its faces a strip of metal arranged as a radius. In its ro- 

 tation the metal strip rubbed lightly against a metal spring. One of 

 the poles of the pile being connected with the strip, the other with 

 the spring, from the velocity of rotation of the disk and the distance 

 of the sj;ring from the axis the duration of the current could be de- 

 duced and the galvanometer-deflections corresponding to difl^erent 

 durations could be measured. M. -Schneebeli substituted for this 

 another arrangement, consisting of a metallic pendulum carrying 

 at its lower part a tri])le spring ; this rubbed against a hori- 

 zontal strip ot steel fixed in the same vertical plane as the axis of 

 rotation of the pendulum. That there might be no shock at the 

 moment ofcontact, a glass plate ap])lied horizontally to the steel on the 

 same side as the apjjroaching pendulum forced the spring to bend 

 gradually. The pendulum was connected with one of the poles of the 

 pile, and the stri]) with the other; and the Meyerstein galvanometer 

 (with mirror and rule) was introduced into the circuit, which was 

 traversed by the current from one or two very clean Bunsen pairs. 

 The duration of contact in this aj)paratus was inversely proportional 

 to the square root of the height of the fall H of the pendulum. 



Now, on constructing the curve of which the values of — — were the 



abscissse, and the galvanometer-deflections the ordinates, the author 

 obtained aright line passing very distinctly through the origin; from 

 which it follows that the deflection of the galvanometer-needle is pro- 

 portional to the duration of the current which produces it. 



The intervals of time during which the current was allowed to 

 act varied between 000015 and 0*00070 of a second. It being so, 

 M. Schneebeli studied by means of this method the conditions upon 

 which the more or less prolonged duration of the shock of two elastic 

 bodies depends. He operated upon one substance only, and com- 

 menced with a simple case, that of impact on a plane surface. This 

 was the upper base, quite flat and smooth, of a right cylinder of steel 

 2 metres in length and of 36 millims. diameter, firmly fixed, and 

 connected with the galvanometer. The impact was always direct 

 and central. The striking body, a ball or a cylinder, was attached 

 to a conducting wire which put it in communication with the pile 

 and, besides, with the galvanometer. The current passing during 

 the time the contact continued, gave, in the galvanometer-deflection, 

 a relative measure of the duration of the impact. The author had 

 previously satisfied himself that the form and relative dimensions of 

 the two surfaces exerted no sensible influence on the conductivity of 

 the circuit, and consequently on the deflection. 



1. Influence of the mass of the striking body on the duration of the 

 collision. — On letting fall on the plane surface of steel some cylinders 

 also of steel, of the same length but of different diameters, and all 

 having their lower extremity identically spherical, it was ascertained 



