1 6 Air, Smeaton’s Ohfervations on the 
as if they had been (like the reft) laid down from the bifettion 
from 85° 2o\ and therefore being the firft made, whatever error 
was in them, would be communicated to all connected with 
them, or taking their departure from them. Every heteroge- 
neous mixture fhould be avoided where equal divifions are re- 
quired. It is not the fame thing (as every good artift will fee) 
whether you twice take a meafure from a fcale as nearly the fame 
as you can, and lay them off feparately ; or lay off two openings 
of the compafles, in fucceftion, unaltered ; for though the fame 
opening, carefully taken oft from the fame fcale a fecond time, 
will doubtlefs fall into the points made by the firft, without 
fenfibleerror; yet as the floping ftdes of the conical cavities made 
by the firft point will condutt the points themfelves to the cen- 
ter, there may be an error which, though infenfible to the fight, 
would have been avoided by the more Jimple procefs of laying off 
the opening twice, without ever altering the compafles. 
The 96 arc was, I have no doubt, invented by Mr. Gra- 
ham, from having perceived, in common with all preceding 
artifts, how very much more eafy a given line was to bifett, than 
to trifett, or quinquefett ; and therefore the 96 arc which pro- 
ceeded by bifettions only (or by laying off the fame identical 
openings, which, as already fhewn, is ftill more fimpJe and 
unexceptionable) was wholly intended by him by way of 
checking the divifion of the arc of 90, which required trifec- 
tions and quinquefettions. But experience foon fhewed the 
fuperior advantage of it fo ftrongly, that the ufe of the 90 arc 
is now wholly fet afide, where accuracy is required ; whereas 
the ingenuity of Mr. Bird having fhewn a way to produce the 
90 arc by bifettion, when this is really purfued quite through 
the piece, by rejetting all divifions derived from any other 
origin, the 90 arc will have nothing in it to prevent its being 
equally 
