of Edinburgh, Session 1885-86. 
967 
In Group VII. we have tastes produced which are in the main 
salt. In the case of both the sodium and potassium compounds 
there is a tendency to the production of the saline and hitter among 
the higher members of the group, and this is sooner seen in the case 
of potassium — itself a higher member — than in the case of sodium, 
a lower member of Group I. 
From a study of these groups we may learn many important facts. 
It will be seen that much depends upon the electro-negative group 
with which the element is combined. Thus sodium combined to 
form a chloride has a different taste from the sulphate of the same 
element. 
With the same electro-negative group similar elements give 
similar tastes, but with a curious and very uniform change, as we 
pass from an element of low to one of high atomic weight. As we 
shall soon see, a change in physical property may, in like manner, be 
seen as we pass from a lower to a higher member of a group. 
So far then we have reduced taste to a function of elements and 
then compounds, and we see that it obeys laws which are the same 
for so-called physical properties of these elements; just as from a 
knowledge of the components of a compound we can account for 
its physical properties, so we have to take to pieces the sapid sub- 
stance before we get the clue to relationship between its nature, 
as indicated by its chemical and physical properties and the sensation 
it produces. 
This is much already to learn, but can we go a step further, and 
ask “ in what essentials are elements alike that produce similar 
tastes.” We turn naturally to the question with which we started, 
and ask, Do these elements vibrate in any way that is similar ? 
Eapid advances are being made in the more exact and extended 
inquiry, How do the ultimate particles of matter move ? In a few 
years, no doubt, the investigations into the ultra-red and ultra-violet 
spectrum will shed a flood of light on this question, and will enable 
us to come to more definite conclusions than at present. Still 
much is known, and it may not be premature even now to make 
use of this incomplete knowledge. 
No one would expect to find very closely allied spectra when 
comparing elements even of the same group, and for this reason. 
Suppose potassium and lithium to have each a fundamental tone and 
