30 
Eccles, owing, I presume to a less sensitive kind of test 
paper having been used, and probably also to a difference 
in the method of exposure of the papers. The annual 
curve shows a maximum in September, and another at the 
end of April and beginning 
of May, and a principal 
minimum in November. The 
means for the seasons are, 
Winter 
....... 0-91 
Spring 
....... 0-94 
Summer 
........ 0*94 
Autumn 
0-88 
The numbers for winter, spring, and summer, are much 
less than those obtained at Eccles, where I believe the 
observations were made with Moffat’s test papers; and 
very strikingly less than those obtained at Oxford, where 
Schonbein’s papers were used. 
From the results of observations which I have made 
during the last few months with the new papers prepared 
by Mr. Mackereth, it seems probable that the amount of 
ozone near the earth’s surface is dependent upon the height 
at which clouds are formed in the atmosphere. In order to 
test this view I have examined the results of Mr. Crosth- 
waite’s valuable series of observations of the heights of the 
clouds at Keswick, as given at page 40 of Dr. Dalton’s 
Meteorological Observations and Essays, 2nd edition, the 
only reliable series of the kind of which I have at present 
any knowledge, and I find that out of every 1 00 observations 
the number of times that the elevation of the clouds ex- 
ceeded 1,000 yards was for each month as follows : — 
January 
... 34-6 
July 
48*5 
February 
... 30-4 
August 
52-0 
March 
... 51-8 
September 
46-6 
April 
... 52-8 
October 
42*6 
May 
... 63*1 
November ...... 
38-3 
June 
... 58-5 
December 
31-1 
