June 17, 1880] 



NATURE 



'51 



vations as actually obtained without being smoothed. 

 The figure is a reduction to jij of the original ; the ab- 

 scissas are as the cube-roots of the various pressures in 

 millionths of an atmosphere, and show relatively the 

 number of molecules in a given linear space ; the ordi- 

 nates are as the number of cells. 



The observations were again plotted down as in Diagram 

 No. II., making the abscissK in the inverse ratio of the 

 cube-roots of the various pressures in millionths, so as to 

 represent relatively the mean distance of the molecules at 

 the various pressures in millionths of an atmosphere ; 

 this has the effect of extending the sc.-ile for decreasing 

 pressures beyond the minimum resist.ince of the tube, 

 and of compressing it on the opposite side for increasing 

 pressures. 



The following tables show the number of cells necessnry 

 to produce a discharge for various pressures in millionths 

 of an atmosphere : — 



11,400 



9,6co 



420 



358 



270 



'95 

 180 

 165 



90 

 100 

 200 

 300 



403 

 500 

 600 

 700 

 Soo 

 930 

 1,000 



S.280 

 5.145 

 4,200 

 3.600 

 3,120 

 2,670 

 2,280 

 1.830 

 1,320 



1,000 



135 

 C4"j 

 60 

 48 

 45 

 39 

 45 

 51 



An experiment was made in order to ascertain whether 

 there was either any condensation or dilatation of the gas 

 in contiguity with the terminals before the actual passage 

 of the discharge. In order to do this an apparatus was 

 constructed, as shown in Fig. I. 



It consists of a glass cylinder, 4'35 inside diameter, the 

 depth of which is accurately the same in every part, r6inch, 

 so as to insure the parallelism of two glass disks which 

 close its ends. Its cubical contents exclusive of the ter- 

 minals was found to be 3S5 cub. centims. 



These are held in contact with the ends of the cyhndcr 

 by means of screw-clamps m.iJe of ebonite, and the 

 whole apparatus is supported on a tripod ebonite stand, 

 which is fastened to a square wooden foot. Attached 

 parallel to the top and bottom glass disks, by means of 

 fl;mged-screw rods, are two brass disks with rounded 

 edges, 3' I inche; in diameter ; these are maintained at a 

 di=tan;e of o'i3 inch, 3 3 mm. at which the discharge of 

 11,000 cells would only just take place. 



The ends which project through the glass disks are 

 furnished with binding-screws for attaching wires from 

 the battery. 



On the side of the cylinder is a tubulure in which is 

 fitted a gauge containing strong sulphuric acid, so as to 

 dry the inside of the apparatus, and to indicate whether 

 any condensation or dilatation of the gas contained in 

 the cylinder occurs on connecting the metaUic disks with 

 the battery by means of the contact-key. The edges of 

 the cylinder were rubbed with grease, and care was taken 

 to prove that the apparatus was perfectly tight by causing 

 the fluid in the limb of the gauge to stand for some time 

 higher than that in the bulb. When connection was made 

 with a battery of 9,800 cells, there was not the slightest 

 indication of any alteration of volume of the contained 

 air, so that there was neither condensation about the 

 disks which woull have caused a contraction, nor repul- 

 sion from the disks which would have caused an expan- 

 sion of volume. The fluid in ihc stem was observed with 

 a lens, but not the slightest motion of it took place. The 

 same result was noti -eJ even when water was substituted 

 for sulphuric acid. So far, then, as this apparatus would 

 indicate it, the result is entirely negative. 

 Potential necessary to produce a discliarge between d/sAs 

 I '5 tnc/i diameter at a constant distance and at various 

 pressures 



The experiments were made by placing the micro- 



