Experiments and Inductions, 83 



ball. The pitted space was bounded by a perfectly circular 

 ridge, as if there had been a current of air or augmented atmo- 

 spheric pressure on the part. 



In the dark the usual appearance of brush was visible issuing 

 from the needle when it was positive ; when negative the star 

 appeared, but the wax coating of the knob was pitted exactly the 

 same as when positive. 



3. Other experiments of the 'same kind were made by present- 

 ing the point of a penknife to the outer surface of a disk of shell 

 lac upon which there was a thin coating of soft wax in a semi- 

 liquid state. The inner surface of the disk touched the wax- 

 coated knob of the charged jar. The effect was very peculiar. 

 When the point was within one-eighth of an inch from the coat- 

 ing of the disk, a sharply-defined perfectly straight line of inden- 

 tation suddenly appeared, forming a small fosse about one-fourth 

 of an inch long and one-fiftieth of an inch broad. This was quickly 

 followed by another of the same kind, radiating from the extre- 

 mity of the first at an angle of about 60°, and sometimes at right 

 angles ; others followed, and sometimes two together, forming 

 the letter T. 



The general effect, after the discharging point had been held 

 opposite different parts of the semiliquid wax surface, resembled 

 the marks left by crows' feet on a soft clay surface. 



When the rounded extremity of a brass wire was substituted 

 in place of the penknife, a similar effect took place, but with 

 much greater ease, lines being thrown off in rapid succession 

 forming stars. A small brass ball throws off stars. 



4. It is only after such lines appear that the disk acquires a 

 charge. Wheu the wax surface has cooled and become solid, no 

 indentation or mark is observed when the wire or knife is pre- 

 sented under otherwise the same circumstances ; but charge being 

 still given as before, it seems fair to infer that the particles or 

 molecules of metals are still projected upon the wax surface in 

 lines. 



The indented line is not simultaneously impressed. — The end 

 nearest the point of the conductor appears first. This was most 

 distinctly observed when the conductor presented to the molten 

 surface was a needle bent at the point ; also a small brass ball 

 half an inch in diameter. 



5. With highly-charged jar and knob the following experiments 

 were made with an insulated circular plate of tinned iron 4 inches 

 in diameter. 



a. Held with its centre facing the coated knob at a distance 

 of a quarter of an inch ; after a few seconds, removed it, and 

 found no charge. 



b. The same process repeated after a very small hole had been 



G2 



