[ 3*8 3 
negative and for fome time afterwards, are but 
fmall ; the effect produced before towards render- ' 
ing the excefs of DbtF above RA/Q always lefs 
than the excefs of RA/Q^ above DA/C, will be 
fuch as cannot be compenfated for afterwards. 
. A further proof of this will appear from confider- 
lng that even when R A t QJs increafed toRQW A 
it is but little fhort of the arithmetical mean between 
ADA and HD A. For from Art. n.and 20. it 
may be inferred that the ratio of the whole area 
R QJW A to this mean, or to is A x H, 
which is never far from the ratio of equality but 
when both p and q are of any confiderable magni- 
tude, it is very nearly the ratio of equality For 
example j when n — no, q = JO o, p — 10 ] t i s 
.9938. 
22. The ratio of DA/F to RA/Q/s lefs than that of 
1 "f* 2 ^ a b +— 1 ± to one. For by Art. 12. 
the ratio of DA^F to DA/C is lefs than that of 
1+2E//4. ihZjL to 1 ~ 2 E a l? — 2 E a * b 9 
n 
But by the laft Art. R h t QJs greater than the arith- 
metical mean between V) htV and OhfC, and i is 
exaaiy the arithmetical mean between i + 2 E ab ’ 
4 - ? E and i — 2 E atf — 2 E . From 
whence this Article is plain. 
23. The 
