REFRANGIBILITY of LIGHT. 31 
On this account I attempted to introduce a third medium, by 
filling this vacancy with a fluid of the leaft difperfive kind, and 
of lefs mean refractive denfity than the difperfive fluid. For 
this purpofe I employed fometimes rectified fpirit of wine, and 
fometimes vitriolic zther ; and by giving to the lenfes the pro- 
per degree of curvature, in which great variety may be intro- 
duced, I fucceeded in forming object-glafles, in which both 
aberrations are removed, and hardly any more light loft than 
in a fimple object-glafs. 
Havine gained this point, I now determined to try how 
far the aperture of the object-glafs might be increafed, without 
increafing its focal length, expecting, at leaft, to equal reflectors 
in this refpe&t. But the firft trials to execute object-glafles on 
this principle, though they left no reafon to complain of want 
of fuccefs, when compared with fuch inftruments as are now in 
ufe, furprifed me with new phenomena, and new obftacles to 
the perfection of the theory of telefcopes, more unaccountable 
and perplexing than any I had before encountered. Thefe I 
fhall now proceed to give an account of. 
Of the imperfet corretion of prifmatic colour which is obtained by a 
combination of mediums of different difperfive powers. 
I TooK a compound object-glafs of the conftruction laft 
mentioned, compofed of three lenfes, two of them plano-con- 
vex and the other a menifcus. The radius of convexity of one 
of the plano-convex lenfes is about four inches, and the con- 
vex fide is turned towards the objet. The radii to which both 
fides of the menifcus are ground, are about five inches, one 
fide being convex and the other fide concave. The concave fide 
is made to refpect the plane fide of the above mentioned plano- 
convex, and the vacancy between them is filled with vitriolic 
ether. The third plano-convex lens is ground to a radius of fix 
imches. Its convex fide is turned towards the. convex fide of 
brioy the: 
