1887.] Mr A. Scott on some Vapour Densities. 
415 
with the exception of one with sodium, which was the first done 
in a new platinum vessel. It may be taken then as conclusively 
proved that the molecules of potassium and sodium are monatomic 
at high temperatures. 
Mercury, sulphur, and iodine were used to test the apparatus 
and the degree of accuracy to be expected, and gave results quite in 
accordance with those of other experimenters, as did also mercuric 
sulphide. Both chlorides of mercury seem to undergo a very large 
amount of dissociation into their elements, though not to the same 
extent as the sulphide. 
The results for the chlorides and iodides of caesium, rubidium, 
and potassium point to the ordinarily received formulae as the 
correct ones, as is the case with the chlorides of manganese, silver, 
and lead, which last seems to dissociate to a certain extent. 
The bromide and iodide of cadmium also seem to dissociate largely 
at the temperature employed. Ferric and chromic chlorides give 
results which seem to point conclusively to the formulae FeCl 3 and 
CrCl 3 as the true ones, the chromic chloride giving results very 
closely corresponding to this, and the ferric chloride considerably 
lower, which may be due to water absorbed during weighing (which 
was done as above described for potassium), and this giving a larger 
volume of hydrochloric acid than the ferric chloride from which it 
would be produced, gives too low a number for the molecular 
weight. 
Potassium fluoride was tried, but gave no vapour whatever, and 
phosphorus at once destroyed the vessel. 
One experiment with arsenic trioxide gave results pointing to 
As 4 0 6 as its formula, but further experiments are in progress with 
it and several other bodies, the results of which I hope to be able 
to communicate shortly. 
6. On the Determination of the Plane Curve which forms 
the Outer Limit of the Positions of a certain Point. 
By Dr G. Plarr, Communicated by Professor Tait. 
2 D 
VOL. XIV. 
28/1/88 
