266 On the Conduction of Water. 
t’ in.ot’ 
cong? and u=HM=QL+KE+HK=!+ (3 +u'. Substitute 
these values of (1), (uv), in the equation above, there results 
; 1H 
a’?62 — b’? (pd+ cd?) 
oa (20/*ésin.g —W*(p + 2ed)) 4+-6/t/ 
+ 2a’? bu’ eu! 
2a’? sin.ct’u’ oa Me 4fitu! 
COs.0 
. t/? 
+(a’?sin.g? ee? ae 4-m't’* 
+a’? u/? J +n/ul, 
in which h’, 6’, &c., represent the corresponding coeflicients in the 
left member. But as before, t/=cos. (a —g)y. Wherefore it only 
remains to supply the places of the letters A, 6, &c., in (3), by the 
corresponding accented ones h’, 6’, &c. This furnishes the equa- 
ae —e’+f’cos.(w- 9)x Gok ly 
tion y ag On! mi! oe ° Sas era 
2 
— [x + sin. (»—9)x]*5 
in which the coefficients h’, 6’, &c. are simpler than h, 6, &c., and 
in which (n’) being equal to (a’’), is never zero. This equation, 
to be general, supposes the possibility of drawing a tangent to a 
conic section parallel to a given line, a problem which is evidently 
impossible in the case of the hyperbola, when the line parallel to 
which the tangent is required to be drawn makes a less angle with 
~ the principal axis, than the asymptote does with the same axis. 
h! +-0/cos.( —o)x +m’cos. (a—9)?x y’) 
n! 
Arr. VIII.—On the Conduction of Water ; by Prof. C. DeweY- 
In Vol. RXviii, p. 151, of this Journal, are some details on this 
subject. In that paper, the inadequacy of Dr. Murray’s expeti- 
ments on this subject was shown. It is certain that when the vessel 
containing the water and thermometer js formed of ice, the power of 
water to conduct caloric downwards cannot be shown, as the heated 
water, when its temperature is below 40° Fabr., will become heav- 
ier, and thence sink to the bulb, and cause the temperature to be 
higher. Ifthe vessel is ‘not made of ice, and the water on the ther- 
mometer is cooled to near 32° Fahr., it will be equally impossible 
to show its conduction of caloric from particle to particle, for the 
