47 
By an oversight in our analysis of November, 1876, the 
potassium chloride is made 84 instead of 8‘4, which of course 
affects the proportions of the two alkalies, but does not 
alter the relations of the other salts. The potash in the 
present analysis was determined directly, but in the former 
by difference from the total alkali and chlorine. 
On comparison, the water of this spring in the different 
years alluded to will be found to contain a very constant 
amount of mineral matter, which does not vary in compo- 
sition to any great extent. The respective amounts of dif- 
ferent years are as follows : 
1868 — Thorpe 508 '5 grains per gal. 
1875 — J. B. and H. G 5 10 ’3 ,, 
1876— J. B. and H. G, 514-6 „ 
1878— C. G. and H. G 522-5 „ 
The flow of water on August 16th, 1878, was at the 
rate of half a gallon per minute. This is just half the 
rate found by Thorpe in 1868, who during the previous 
two years had found the flow very constant. A long period 
of small rainfall occurred just previous to the date of our 
sample. 
We may add, that in a paper read before this Society 
(Second Series, Yol. XII.) by Mr. Binney, F.B.S., will be 
found some additional information as to the geological 
strata of the locality of Humphrey Head, accompanied by a 
section of the Head and the neighbouring point, which 
agree with the remarks on this point in our previous paper. 
