IN CERTAIN MINERAL WATERS OF SOUTH BRITAIN. 
227 
seen the liquor, either at the time or shortly after the operation of the re-agent, 
assume a pinkish hue, owing, as I suppose, to the reddish tinge of the bromine 
given out mixing with the blue colour of the iodide of starch. In stating there- 
fore, as I have done in the Table, the proportions of iodine in several of the 
w r aters, I am far from pretending to offer more than an approximation to the 
relative quantity in which it occurs, and am fully aware of the necessity of 
more precise experiments, conducted on a different principle, before the points 
in question can be considered as satisfactorily determined. 
The starch test I find will readily indicate a quantity of iodine not exceeding 
one grain to 7 gallons of water, or one 450,000dth part; but as in no case that 
has occurred to me the proportion exceeded one grain to 10 or 12 gallons, and 
in many appeared scarcely to amount to one- 10th of that quantity, I despaired 
of arriving at more accurate results, by adopting any other method that aspired 
to greater precision than the one already stated. 
In every case in which I have noted that no iodine could be detected, the 
water had been concentrated at least as far as to one-30th of its original 
quantity without effect ; so that the proportion of this principle, supposing 
after all any of it to exist, could not well amount to a grain in 200 gallons. 
In some cases, indeed, where the spring was one of weak impregnation, I 
have carried the concentration much further, as may be seen in the Llandrin- 
dod waters, where no traces of iodine appeared, until they had been reduced 
to nearly one-50th of their original volume. 
In my trials for bromine, I have in great measure conformed to the directions 
of Balard ; first boiling down the water to about a fourth of its original quan- 
tity with a portion of quicklime to prevent the bromine from being dissipated 
by the heat ; and then, after filtering the residuum, introducing chlorine as long 
as any sensible yellowness was caused by its addition. The water was then 
strongly agitated with ether, which collects on the surface, carrying with it 
the bromine with which it had combined, and was then poured off into a separate 
vessel. [The bromine, immediately upon being thus removed from the water, 
was treated with a quantity of a concentrated solution of pure soda sufficient 
to render the ether containing it colourless ; the latter alkali being employed 
for this purpose in preference to the vegetable one, as I found that bromine 
formed with sodium a salt more soluble in alcohol than it did with potassium.] 
