3.x : 6 M\'m-ueq-e 'vn'tbe Confiruciion.. 
ground together, though ever io irregular at firft, will 
(if thbavorking be uniform and proper, <: onfifting, efpe- 
cially at lalt, of crofs ftroh' 3 at : ’ poflible direction- 
acrofs the diameter) be formed ...;o :: ions of true and 
equal fpheres; had it not been for this lucky neceffity, it- 
would .have been impoflible to have produced that cor- 
reCtnefs which is eflential in the fpeculum of a good re- 
flecting telefcope by any mechanic contrivance what- 
ever. For when it is confidered, that the errors in reflec- 
tion are four times as great as in refraction, and that the 
leaft defect in figure is magnified by the powers of the 
inflrument, any thing Ihort of perfection in the figure: 
of the fpeculum would be evidently perceived by a want 
of diftinCtnefs in the performance. 
I mull not, however, quit this article without obferv-- 
ing, that I all along fuppofe, both in forming the tools, 
and at laft figuring the metal (and indeed the fame mult, 
be obferved in the future procefs of polifhing) that no 
kind of preffure is ufed that may endanger the bending: 
or irregularly grinding them; they fhould therefore be 
held with a light hand, and loofely between the fingers,, 
and the motion given fhould be in a horizontal direction, 
with no more preffure than their own dead weight. 
Having now finifhed the metal on the hones, and ren- 
dered it both in point of figure and furface fit for the 
laft 
