260 Some Account of Electrical Experiments. 



in Mechanical Tables, and that he has consequently ascribed! 

 these effects to molecular attraction, rather than to the action of 

 an expansive force : but the tenacity of a metal as usually con- 

 sidered, nnist not be taken as a measure of its power to resist a 

 sudden impulse; and in my experiments it may be observed, 

 that the results appear to correspond sufficiently with the known 

 . mechanical properties of the metals. Possibly very different 

 results may have been obtained by M. De Neiis, because all his 

 first experiments were made with water, and mine have been 

 made almost exclusively with oil, which is more certain in its 

 action, not only as a more expansible fluid, but as it presents less 

 facility to the dissi])ation of the charge in a lateral direction. 



The power which bus been most fretiuently employed in these 

 experiments is very considerable; but the principal facts may be 

 exhibited with more facility by employing very thin cylinders 

 filled with oil, and having but a small bore, for these may be 

 distinctly affected by a small battery, or even a few jars. 



M. De Neiis has attempted to trace the course of the electric 

 fluid, and has instituted some curious experiments for that pur- 

 pose. A square pane of window glass, perfectly clean and dry, 

 is slightly powdered over on each side with lycopodium or with 

 powdered resin (the former adheres best). In this state it is 

 placed vertically between the points of Henly's discharger, which 

 are made to touch the opposite surfaces of the pane at its centre; 

 the charge of a small jar is then passed from one wire to the 

 other, and two radiated figures of different appearance are pro- 

 duced. Fig. 1, Plate IV, represents that on the positive side, 

 and fig. 2 that on the negative side. 



I find that the results of this experiment are not very regular, 

 and are much influenced by the height of the charge, and the 

 continuance of the- contact of the charged jar with the dis- 

 charger. Each figure is exhibited most distinctlv when the op- 

 posite surface of the glass is clean. I have therefore sometimes 

 employed two panes of glass placed together, and witii their 

 outer surfaces only powdered : the effect is usually best obtained by 

 keeping the jar in contact with the wires for a few seconds, and 

 even in that case the jar will be but partially discharged : the 

 effects are therefore aaalogoiis to those obtained in the Exjjeri- 

 ments of Lichtenberg and Bennett, and do not appear to me to 

 offer any satisfactory evidence of the course of the fluid in the 

 discharge. 



The same observation applies in some degree to another in- 

 genious experiment of M. De Neiis, with is interesting from other 

 considerations. The figure on the opposite page represents the ap- 

 paratus. 1 and 2 are glass tubes, nearly filled with water, and each 

 containing two thick silver wires A and B, which are disposed ia 



the 



