87 
are functions of V only. Then (5) may be put under the 
form 
SF. 
dy 
-SG. 
(7) 
a dx dv dx h ^ JXJ ' b ' /y h dvdx h 
and when, as for the Boolians, we can so determine the 
functions F and G as to render the sinister of (7) a perfect 
differential coefficient with respect to v we have a result of 
the form 
dv 
dY 7 
= sg 6 w 6 o hTv + h 
(8) 
whence 
/hdv = Z-/s,G t \V l ,Q, ll a\ (9) 
Now v only enters O through Y, and the whole expression 
under the last sign of integration is a function of x and Y. 
Hence if W and £2 can each be developed in a series of 
ascending powers of Y we shall have for integration a series 
whereof each term will be integrable. And if the series of 
integrated terms have, between and for the limits of the 
integration, a value or sum finitely assignable the conver- 
sion will be effected. 
There is a defect of generality, similar to that adverted 
to at the commencement of this paper, in art. 12 pp. 36 — 37 
of my paper “ On a Certain Boolian Trinomial,” published 
in the Messenger of Mathematics , No. XYII. (vol. Y). The 
process above given will supply the defect. 
“Oakwal,” near Brisbane, 
Queensland, Australia, 
November 26, 1869, 
Mr. Spence repeated the experiment he had made at the 
Exeter meeting of the British Association, showing that the 
temperature of saturated saline solutions could be raised to 
their boiling points by merely passing through them ordi- 
nary steam at a temperature of 212°. Thus, a solution of 
chloride of sodium was raised to a temperature of 221 °, and 
