ON SMELL. 
287 
Thus we can cliaracterise the compounds of cMorine and its 
oxides as chlorous ; indeed, we may group the three elements, 
chlorine, bromine, and iodine, together, and name the character- 
istic odour of them and their oxides haloid smells. 
Similarly sulphur, selenium, and tellurium in their compounds 
with hydrogen have a generic smell, and likewise arsenic and 
antimony. The only oxide of nitrogen which is smelt is nitric 
peroxide, so that it is impossible to pronounce on a generic smell 
for this substance. 
It is again easier to classify carbon compounds. The smell 
of the paraffins is generic ; so is that of the alcohols, the acids, 
the nitriles, the amines, with their irritation like that of ammonia, 
the bases of the pyridine series, the hydro-carbons of the benzene 
group, the higher hydro-carbons — ^such as naphthalene, anthra- 
cene, and phenanthrene. Give any one of these to a chemist, 
familiar with the smell of any one of each series, and accustomed 
to use his sense of smell, and he will at once refer the body to 
its class. 
The tendency of a rise in the series is to make the smell 
“ heavier,” less etherial, and more characteristic. It also becomes 
more able to affect the olfactory nerves. 
The rate at which smell travels is doubtless that rate at 
which the vapour which gives rise to it diffuses. Still it is 
impossible to test this experimentally. For the ease with which 
a smell is perceived varies with the molecular weight of the 
substance. Thus, if a piece of cotton-wool is impregnated with 
ethyl alcohol, and placed in one end of a long tube, which is 
immediately corked, and a similar arrangement be adopted with 
amyl alcohol — the fifth of the series of which the former is the 
second ; although their specific gravities have the ratio of 23 to 
44, and the ethyl alcohol should diffuse If times as rapidly' as 
amyl alcohol ; yet the smell of the latter will be perceived first, 
because a much smaller quantity produces the sensation. 
P 
