372 O. N. Rood—Nature and Duration of 
t. These, from their rarity, were a 
little difficult to study, and I am not — 
sure that in all cases the act” was 
terminated by a spark. The violet 
streak was of course invisible in dise 
experiments. 
Striking distance 2 millimeters—The form was the same as 
fig. 5, but the dimensions were somewhat reduced. 
Total duration with WAP "seins ss 01830 sec. 
dis 01588 “ 
AVOTORG,. fs einiin d+ cw cds 
Average duration of are Pe ee 
The other alge (fig. 7) also sometimes occurred with a dura- 
tion of ‘027 s 
P wapieerd pen 3 millimeters.—The form was still the same; 
ro ag ion AB consisting pean’ of only three or four 
ee but often of eight or 
Total duration with A i. eee 0124 
* dise 
The form of fig. 7 sometimes occurred, tipo with a 
duration about twice as great as that just give 
Striking distance 4 millimeters.—The form was sometimes ope 
same as those just ota but more often consisted solely ° 
a compact series of from four to six seapesuincnr rt sparks. 
‘0 
Duration with — sparks, oe re 3 sec. disc. 
uk Oya pees 0082 “ 
Striking a: 5 millimeters.—Merely two, three oF four 
isolated sparks. 
Average total duration, ...-..37.:. <3 0085 sec. (disc). 
Striking distance 6 pees —Same form as es last, 
Average total duration,..............++ 0025 sec. (dise). 
Striking distance 7 hence —T wo, three, may four jiaig 
Average total duration,.......5..60..0% ‘0018 disc). 
Striking distance 8°75 millimeters.—Only two wales one quite 
faint. Interval between them 0009 sec. (disc). 
Striking distance 10 millimeters.—One spark; seldom two- 
10-75 was the maximum striking distance; at it, and half @ 
millimeter under it, only a single spark was produe 
Form and duration of the discharge of the small jar with platr- 
num points as electrodes. 
The length of the simple induction spark was 58 millimeters. 
