1876.] 
Recent Chemical Researches . 
3 1 
increase of temperature, as those of the metallic elements. 
Sodium exhibits a line spedtrum at a dull red-heat ; at the 
same temperature the spedtrum of sulphur is characterised 
by a continual absorption in the blue : as the temperature 
increases the line spedtrum of sodium remains unchanged, 
but the spedtrum of sulphur gives place to a channelled 
space spedtrum, which remains until the temperature of the 
eledtric arc is reached, when a line spedtrum makes its 
appearance. There are therefore three stages of non-metallic 
spedtra corresponding to three stages of heat, but these 
three stages do not correspond with the stages of the spedtra 
of metallic elements : hence the inner structure of the 
non-metals probably differs from the inner structure of the 
metals. 
The spedtra of the substances which are present in the 
reversing layer of the sun are all line spedtra ; hence it is 
inferred that these substances are all in a similar state of 
aggregation. But we saw that at comparatively low tem- 
peratures non-metals are characterised by channelled space 
spedtra, which are believed to indicate a more complex 
inner structure than is shadowed forth by line spedtra. 
The inference to be drawn from these facts (admitting the 
connection between simple spedtra and simple molecular 
constitution) seems to be that the formation of those 
molecular aggregations which are presented to us by the 
non-metallic elementary bodies on the earth is not possible 
at the temperature of the sun. 
This hypothesis is put forward by Mr. Lockyer merely as 
a means for guiding future work : from the chemical view 
point it is exceedingly interesting. That the non-metallic 
molecules should be more ready to undergo change than the 
metallic molecules is in keeping with many known chemical 
fadts. The phenomenon of allotropy occurs to a considerable 
extent among non-metals, and it is almost certain that alio- 
tropic changes are always accompanied by variations in mole- 
cular aggregation. The change of ordinary into amorphous 
phosphorus, and vice versa, would seem to depend essentially 
upon the tension of the phosphorus vapour ; and what does 
this statement imply but that, under certain conditions, 
the molecules of phosphorus are so packed together as to 
exhibit to our senses the phenomena associated with what 
we call ordinary phosphorus, while under other circum- 
stances the state of molecular aggregation is such as that 
the body exhibits the phenomena associated with amorphous 
phosphorus ? 
Mr. Lockyer defines a metal, provisionally, as a substance 
