80 



Mr. J. Joly. 



found that both, glasses have shared in the marking, seemingly to 

 an equal extent. The marks, too, are apparently vis a vis. 



What has caused this structure ? The answer is, T think, not hard 

 to find. In the first place these marks are melted in the glass. Their 

 rounded contours and ridges leave no doubt of this. They are quite 

 distinct in appearance from the hair-like cracks to be observed 

 extending from the track at more or less regular intervals. If the 

 spark be regarded as a line of explosion from which atoms raised to a 

 high temperature are being repelled, the effects observed are quite 

 explicable. These atoms, in the free spark escaping all round, are here 

 compelled to make their escape by the narrow side ways only. The 

 side rays depict the lanes of escape cut by the heated, outrushing 

 molecules. Shift the slide under the microscope till the extremity of 

 track is approached. Here the " head-room " is increasing. Fig. 2, 

 PI. 3, is a photograph of such a part of the track. Here there is a 

 relief of pressure in the longitudinal direction. The outrush becomes 

 more axial along the spark. ISTot quite axial, because there is also a 

 possible sideway escape and the lines brush out in the intermediate 

 direction of easiest escape. Hence the tracks generally end up in 

 these beautiful feathery forms. 



Occasionally the spark reaches a point in its path where side 

 escape is difficult or impossible. It then assumes the appearance of 

 fig. 1, PI. 4. The central part is much pitted and lines of escape 

 appear diverging at each end. This is perhaps in some respects 

 remarkable. It might be thought that the unidirectional current 

 would polarize the motion of the atoms, but it would appear as if the 

 atoms, having received their energy from the ether, were left unin- 

 fluenced sensibly by any stress or motion in the medium to polarize 

 their movements, other than their tendency to move from the spark 

 axis in a radial direction. In fact the directions of longitudinal flow 

 are apparently obedient to conditions of pressure only, or seem at 

 least sufficiently explained by such. 



The lines on the concave side of a sharply curved track are shorter 

 by a little than those on the convex side. The reason of this is that 

 escape is not so free on the inner side. Passing round sharp obstacles 

 accidentally occurring in the path, the axis of the spark track is close 

 in near the obstacles, the lines chiefly radiating outwards. 



I made up a slide in which a couple of very fine lines had been 

 engraved with an etching diamond on the surface of the slip, extend- 

 ing across the field at right angles to the path of the spark. On 

 passing sparks it was found that for a little distance at either 

 side of the point where the track crossed these lines the side rays 

 were discontinued, evincing the relief of pressure afforded by the 

 channels. 



When a spark bifurcates the course of the outrushing atoms is 



