Qbfervalions in Ele&ricity, 399 
by the combuftion had changed their colour, were almoft 
all dropped to the bottom oi the phial. 
I fhall conclude this paper with the defcription of a 
pocket electrometer which I have lately conftruCted, 
and which, on feveral accounts, feems preferable to thofe 
of the moft fenlible fort now in ufe. The cafe and 
handle of the electrometer is formed by a glafs tube 
about three inches long, and three-tenths of an inch in 
diameter; half of which is covered with fealing-wax. 
From one extremity of this tube, that is, that without 
fealing-wax, a fmall loop of filk. proceeds, which ferves 
occalionally to hang the electrometer on a pin, 8cc. To 
the other extremity of the tube, a cork is adapted, which, 
being cut tapering on both ends, can fit the mouth of 
the tube with either extremity. From one extremity of ’ 
this cork two threads proceed, a little fhorter than the 
length of the tube, fufpending each a. little cone of pith 
of elder. When this electrometer is to be ufed, that end 
of the cork which is oppofite to the threads is pufhed 
into the mouth of the tube, then the tube forms the 
infulated handle of the pith electrometer, as appears in 
fig. i . When the electrometer is to be carried in the 
pocket, then the threads are put into the tube, and the 
cork flops it, as is reprefented in fig. 2 , The peculiar 
advantages; 
