28 R. THRELFALL. 
examining the properties of sulphur—a substance chosen for — 
reasons I do not need to go into here—but which ultimately — 
turned out to have the immense practical advantage of being f 
commercially obtainable in a state of extraordinary purity. Most : 
Leblanc alkali makers now recover their sulphur by the Chance 
process, which consists in liberating sulphuretted hydrogen from 
the waste and then half burning it, so as to oxidise the hydrogen 
and deposit the sulphur. This sulphur is generally pure to about | 
one part in ten thousand, the residue consisting of a trace of water — 
and dust. By melting and filtering, the dust may be removed, — 
and then a single distillation yields a substance so pure that no 
tests I have been able to devise will indicate any impurity. It 
is well known that sulphur exists in two main modifications ; one 
of them is the form in which it is deposited by the electrolysis of | 
sulphides, the other by the electrolysis of its chloride or bromide. 
This important observation we owe to Berthellot.’ I considered 
it of the first importance to examine the relations of these two 
forms of sulphur, and to make a very long story short I found 
that one modification certainly and the other probably, refuses 
to conduct electricity at all. On the other hand a mixture of 
these substances conducts, not well, but distinctly, and the 
evidence, so far as it goes, tends to show that the conduction is 
electrolytic in a sort of way—for it is accompanied by the 
phenomenon of electric absorption, which is most satisfactorily 
explained on the whole by asort of electrolysis. There are a large — 
number of subsidiary phenomena also tending to support this view: 
We are then in this position. The two sorts of sulphur insulate 
separately but conduct when mixed, that is they combine, for 
the mixture has different properties from either of its constituents. 
If this is granted, what is there to prevent us from assigning t0 
each of the two modifications something of an elemental character? 
Similar results have been attained by totally distinct means by 
chemists working on the metals of the rare earths and 0D 
Beryllium. Is it possible that the idea of an absolutely hard and 
fast element may have to give way, and that we shall have to 
