5gS PHILOSOPHICAL TRANSACTIONS. [aNNO I789. 



or, if short-sighted, assisted by a proper concave lens; which may be thought a 

 sufficient approximation in an explication grounded on a dissipation of rays, to 

 which a precise limit cannot be assigned, on account of the continual increase 

 of density from the circumference to the centre. Certainly some such angle of 

 aberration is necessary to account for the stars appearing under any sensible angle, 

 to such an eye; and if we were, without reason, to suppose the images on the 

 retina to be perfect, we should be put to a much greater difficulty to account 

 for the fixed stars appearing otherwise than as points, than we have now been to 

 account for the actual distinctness of our sight. The less apparent diameter of 

 the smaller fixed stars agrees also with the theory ; for the less luminous the 

 circle of dissipation is, the nearer we must look towards its centre to find rays 

 sufficiently dense to move the sense. From Sir Isaac Newton's geometrical ac- 

 count of the relative density of the rays in the circle of dissipation, given in his 

 system of the world, it may be inferred, that the apparent diameters of the fixed 

 stars, as depending on this cause, are nearly as their whole quantity of light. 



In further elucidation of this subject Dr. M. adds his own experiment. When 

 he looked at the brighter fixed stars, at considerable elevations, through a con- 

 cave glass fitted, as he is short-sighted, to show them with most distinctness, 

 they appeared to him without scintillation, and as a small round circle of fire of a 

 sensible magnitude. When he looked at them without the concave glass, or 

 with one not suited to his eye, they appeared to cast out rays of a determinate 

 figure, not exactly the same in both eyes, somewhat like branches of trees, 

 which doubtless arise from something in the construction of the eye, and to 

 scintillate a little, if the air be not very clear. To see day objects with most dis- 

 tinctness, he requires a less concave lens by 1 degree, than for seeing the stars 

 best by night; the cause of which seems to be, that the bottom of the eye 

 being illuminated by the day objects, and so rendered a light ground, obscures 

 the fainter colours, blue, indigo, and violet in the circle of dissipation, and there- 

 fore the best image of the object will be found in the focus of the bright yellow 

 rays, and not in that of the mean refrangible ones, or the dark green, agreeable 

 to Newton's remark, and consequently nearer the retina of a short-sighted 

 person ; but the parts of the retina surrounding the circle of dissipation of a star 

 being in the dark, the fainter colours, blue, indigo, and violet, will have some 

 share in forming the image, and consequently the focus will be shorter. 



The apparent diameter of the stars here accounted for is different from that 

 explained by Dr. Jurin, in his Essay on distinct and indistinct vision, arising 

 from the natural constitution of the generality of eyes to see objects most 

 distinct at moderate distances, and few being capable of altering their conforma- 

 tion enough to see distant objects, and among them the celestial ones, with 

 equal distinctness. But the cause of error, which is here pointed out, will affect 



