220 
ASTRONOMY. 
as of another: Secondly; that out of an infinite number of 
possible laws, those which were admissible for the purpose 
of supporting the heavenly motions, lay within certain nar¬ 
row limits: Thirdly; that of the admissible laws, or those 
which come within the limits prescribed, the law that actu 
ally prevails is the most beneficial. So far as these propo¬ 
sitions can be made out, we may be said, I think, to prove 
choice and regulation ; choice, out of boundless variety; 
and regulation, of that which, by its own nature, was, in 
respect of the property regulated, indifferent and inde¬ 
finite. 
First then, attraction, for anything we know about it,* 
was originally indifferent to all laws of variation depend¬ 
ing upon change of distance, i. e. just as susceptible of one 
law as of another. It might have been the same at all 
distances; it might have increased as the distance increas¬ 
ed: or it might have diminished with the increase of the 
distance, yet in ten thousand different proportions from 
the present; it might have followed no stated law at all. 
If attraction be what Cotes, with many other Newtonians, 
have thought it to be, a primordial property of matter, not 
dependent upon, or traceable to, any other material cause; 
then, by the very nature and definition of a primordial 
property, it stood indifferent to all laws. If it be the agen¬ 
cy of something immaterial, then also, for anything we 
know of it, it was indifferent to all laws. If the revolu¬ 
tion of bodies round a centre depend upon vortices, nei¬ 
ther are these limited to one law more than another. 
There is, I know, an account given of attraction, which 
should seem, in its very cause, to assign to it the law 
which we find it to observe; and which, therefore, makes 
that law, a law, not of choice, but of necessity: and it is 
the account, which ascribes attraction to an emanation 
from the attracting body. It is probable, that the influence 
of such an emanation will be proportioned to the spissitude 
of the rays of which it is composed; which spissitude, 
supposing the rays to issue in right lines on all sides from 
a point, will be reciprocally as the square of the distance.* 
* Let the light of a candle fall upon a square object like A B C D, Fig. 
4, Plate XXXIX, and if a screen be placed parallel to the object and at 
double the distance, the shadow EFG H, received upon it, will be four 
times the size of the object itself. For the rays passing in straight lines 
by the angles A, B, C, D, the sides E F, F G, G II, II E, must be each 
double of A B, B C, C D, D A: therefore the shadow may be divided 
into four squares each equal in size to the object. At three times the dis¬ 
tance from the candle, the sides of the shadow would each be three times 
as large as the sides of the object, and its area would, therefore, contain 
