the 'Trigonometrical Operation* 147 
attended to, which is, that fixty on the micrometer head 
fhould hand nearly vertical, fo as to he conveniently feen. A 
few feconds of inclination to one fide or the other are of no 
moment, becaufe the dart or index being brought to that por- 
tion, whatever it may be, muft at all times remain there with- 
out alteration, unlefs fome derangement that may have hap- 
pened to the inftrument, in tranfporting from one place to ano- 
ther, fhould have rendered a frefh adjuftment neceflary. But if, 
when the wires coincide with their refpedlive dots and the firft 
notch, fixty on the micrometer head fhould happen to be under- 
neath, or fo far over from the vertex on either fide as to be feen 
with difficulty, then the gold tongue muft be moved a little by 
means of the capftan- headed fcrews, which a£t againft each other 
on theoppofite extremities of its axis. Thus, by repeated trials, 
the wiffied-for object will at length be effedted, that is to fay, 
fixty, to which the dart is to be fet, will ftand in a place eafily 
feen. But it is not to be expe&ed, that each microfcope will 
give juft nine hundred feconds for the run of fifteen minutes. 
Without great lofs of time this cannot be done ; befides that 
two obfervers, of different fights, will adjuft the microfcopes 
differently. Accordingly, in 1787, after many trials of the 
runs in meafuring fifteen minutes on the different parts of the 
limb, microfcope A was found to give only while B 
gave at a medium 90 1". But in 1788, microfcope Agave 
goo // , while B gave no more than 894/h Thefe differences 
were of courfe registered and allowed for in the eftimation of 
the angles for computation, whereby any difference between 
them almoft wholly difappeared. 
The gold tongue, which is extremely thin, applies very 
clofely to the furface of the circle. In the plan it is fuppofed 
to be feen through a thin plate of brafs covering the whole 
U z pedeftal, 
