a Difficulty in the Theory of Vijion ... 261 
The like reafoning is applicable to the circle of diffipation 
on *th* retina of the human eye; and therefore we may ieffen 
the angular aberration, before computed at 15^ in the ratio of 
250 to 55, which will reduce it to f 18". 
This reduced angle of aberration may perhaps be double the 
apparent diameter of the brighteft fixed ftars to an eye difpofed 
for feeing mod diftinftly by parallel rays; or, if fhort-fighted, 
affifted by a proper concave lens ; which may be thought a fuf- 
ficient approximation in an explication grounded on a diffipation 
of rays, to which a precife limit cannot be affigned, on account 
of the continual increafeof denfity from the circumference to the 
center. Certainly fome fuch angle of aberration is neceflary to 
account for the ftars appearing under any fenfibie angle to fuch an 
eye ; and if we were, without reafon, to fuppofe the images on 
the retina to be perfedt, we fhould be put to a much greater 
difficulty to account for the fixed ftars appearing otherwife 
than as points, than we have now been to account for the 
a&ual diftinflnefs of our fight. 
The lefs apparent diameter of the fmaller fixed ftars agrees 
alfo with this theory ; for the lefs luminous the circle of difli- 
pation is, the nearer we mu ft look towards its center to find 
rays fufficiently denfe to move the fenfe. From Sir Isaac 
Newton’s geometrical account of the relative denfity of the 
rays in the circle of diffipation, given in his fyftem of the 
world, it may be inferred, that the apparent diameters of the 
fixed ftars, as depending on this caufe, are nearly as their 
whole quantity of light. 
In farther elucidation of this fubject let me add my own 
experiment. When I look at the brighter fixed ftars, at con- 
fiderable elevations, through a concave glafs fitted, as I am 
fhort-fighted, to fhew them with moft diftin&nefs, they appear 
Vol. LXXIX. Q q to 
