On definite Proportions. 245 
The foregeing experiments and observations will, if I mistake 
not; place the electric column in a new point of view; for to its 
other singular properties maybe added that of an hygrometer: 
they will likewise account in an easy, natural, and satisfactory 
manner for the phenomenon observed by Mr. De Luc, and de- 
seribed by him in Nicholson’s Journal; without compelling us 
to have recourse to the hypothesis of ‘* the preduction of a new 
quantity of electric fluid in the column by the sunbeams*.”’ 
The same may be said concerning the phenomena noticed by 
Mr. Singer when he placed an electric column, and a bell-ringing 
apparatus, before a fire t. 
in the experiments alluded to, the heat dispersed from between 
the extremities of the columns the moisture which impaired the 
insulation ; and thus enabled the electromotive power of the in- 
struments to produce a greater accumulation and deficiency of 
electric fluid at their opposite extremities than could possibly exist 
while that moisture was interposed between them. 
; Iam, &c. 
Tuomas Howipy. 
_ To Messrs. Nicholson and Tilloch. 
LI. An Attempt to determine the definite and simple Propor- 
tions, in which the constituent Parts of unorganic Substances 
are united with each other, By Jacop Berz2E.ius, Pro- 
fessor of Medicine and Pharmacy, and M.R.A. Stockholm, 
[Concluded from p. 175.] 
IV. Generar Virw oF THE RESULTS OF THE EXPERIMENTS 
RELATED IN THESE DIFFERENT Essays. 
I 
W; HEN two substances, which we consider as simple, ure ca- 
pable of being united in more than One proportion, these pro- 
portions, for a given quantity of the [positively] electrical body, 
are multiples by 14, 2,4... 0f the least proportion in which 
the [negatively] electrical body is capable of combining with it. 
{We must not, however, understand this multiplication of pro- 
portions in the strict mathematical sense of the terms, which is 
equivalent to the involution of powers, but simply as relating to 
the multiplication of the quantities by the numbers in question, 
Gilbert.) 
There are however many phenomena which make it probable, 
that the multiple \+ ts only apparent ; and hence it follows, that 
the combination, from which it is deduced, does not exhibit the 
* Nicholson’s Journal, vol. xxxvi. p. $15, + Thid. p. 374. 
~. Q38 true 
