4,0 Mr. Davy’s Lecture on some new analytical Researches 
The more approaches are made in chemical inquiries to- 
wards the refined analysis of bodies, the greater are the 
obstacles which present themselves, and the less perfect the 
results. 
All the difficulties which occur in analysing a body, are 
direct proofs of the energy of attraction of its constituent 
parts. In the play of affinities with respect to secondary com- 
pounds even, it rarely occurs that any perfectly pure or un- 
mixed substance is obtained ; and the principle applies still 
more strongly to primary combinations. 
The first methods of experimenting on new objects likewise 
are necessarily imperfect ; novel instruments are demanded, 
the use of which is only gradually acquired, and a number 
of experiments of the same kind must be made, before one 
is obtained from which correct data for conclusions can be 
drawn. 
2. Experiments on the Action of Potassium on Ammonia , and 
Observations on the Nature of these two Bodies. 
In the Bakerian lecture, which I had the honour of read- 
ing before the Society, November 19, 1807, I mentioned that 
in heating potassium strongly in ammonia, I found that there 
was a considerable increase of volume of the gas, that hy- 
drogene and nitrogene were produced, and that the potassium 
appeared to be oxidated ; but this experiment, as I had not 
been able to examine the residuum with accuracy, I did not 
publish. I stated it as an evidence, which I intended to pur- 
sue more fully, of the existence of oxygene in ammonia. 
In a paper read before the Royal Society last June, which 
they have done me the honour of printing, I have given an 
