[ 39 * ] 
3. Bat when looking at the two Balls (Fig. 6 .) with 
the naked Eye in an open Room, we conftder that C D 
is as far again from the Eye as AB, we judge it to be 
as big as AB, (as it really is) notwithftanding it fub- 
tends an Angle but of half the Bignefs. 
4. Now if, unknown to the Spedator, (or while 
"he turns his Back) the Ball CD be taken away, and 
another Ball op of half the Diameter be placed in 
the fame Line, but as near again, at the Side of AB, 
the Spedator thinking this laft Ball to be at the Place 
of CD, muft judge it to be as big as CD, becaufe it 
fubtends the very fame Angle as CD did before. 
It follows therefore That if a Ball be imagin'd 
to be as far again as it really is, we make fuch an Al- 
lowance for that imagin'd Diftance, that we judge it 
to be as big again as it is, notwithftanding that the 
Angle under which we'fee it, is no greater, than when 
we look at it, knowing its real Diftance. 
For this Reafon the Moon looks bigger in the Hori- 
zon, and near it, than at a confiderable Height, or at 
the Zenith : Becaufe it being a common Prejudice to 
imagine that Part of the Sky much nearer to us which 
is at the Zenith, than that Part towards the Horizon ; 
when we fee the Moon at the Horizon, we fuppofe it 
much farther; therefore as it fubtends the fame Angle 
(or nearly the fame Angle) as when at the Zenith, we 
imagine it fo much bigger as we fuppofe its Diftance 
greater. 
The Reafon why this Experiment is hard to make, 
is becaufe the Light from the Ball 0 p is too ftrongly 
refleded on account of its Nearnefs; but if we could 
give it fo little Light as to look no brighter than the 
Ball C D, it would deceive every body. I have made 
E e e the 
