[ *63 j 
XIIK Defcrift'ton of a Lamp-Micrometer , and the Method of 
ufmg it. By Mr. William Herfchel, F. R, S, 
Read January 3?, 1782. 
T HE great difficulty of meafuring very fmall angles, fuch. 
as hardly amount to a few leconds, is well known to 
aftronomers* Since I have been engaged in obfervations on 
double ftars, I have had lb much occahon for micrometers that 
would meafure exceeding fmall diftances exactly, that I have 
continually been endeavouring to improve theie inftruments. 
The natural imperfections of the parallel \Vire micrometer 
in taking the diltance of very clofe double ftars are the follow- 
ing. When two ftars are taken between the parallels, the dia- 
meters mulf be included* I have in vain attempted to find lines 
fufficiendy thin to extend them acrofs the centers of the ftars 
fb that their thicknefs might be negleCted. The fingle threads 
of the filk-worm, with fuch lenfes as I ufe, are fo much mag- 
nified that their diameter is more than that of many of the 
liars. Refides, if they were much lefs than they are, the 
power of deflection of light would make the attempt to mea- 
fure the diftanc-c of the centers this way fruitlel's : for I have 
always found the light of the ftars to play upon thole lines and 
feparate their apparent diameters into two parts. Now fince 
the fpurious diameters of the ftars thus included, to my certain 
knowledge, are continually changing according to the ftate of 
the air, and the length of time we look at them, we are, in 
1 a home 
