81 
Here it will be seen that five years elapsed between the cold 
winters of 1854-55, and 1859-60; but that six years elapsed 
between the cold winters of 1864-65, and 1870-71, and that 
eight years elapsed between the present cold winter and the 
previous one of 1870-71. It is now known that the 
sun spot period is irregular and not so nearly an interval 
of 10, 11, or 12 years as was imagined. The minimum of 
the sun spot period happened about two years ago, but it 
is still at a minimum, and very seldom have spots been seen 
on his disc during the past year. Whether these cold 
winters are traceable to this solar inactivity or not, the pre- 
sent coincidence is very striking. 
In the same diagram I have presented the ratios of the 
mean summer temperatures of the last 18 years. These 
ranges are included in the weekly mean temperatures of 
June, July, and August, and will be seen to be far less than 
the ranges of the winter mean temperatures. This perhaps 
may be accounted for by the relative difference of the 
amount of atmospheric vapour existing in the air in the two 
opposite seasons of the year. In the winter season the ratio 
of atmospheric vapour reaches 87 per cent, whilst in the 
summer season it reaches only about 75 per cent. But 
whether this is so or not, there is the fact. Here it will be 
seen that between the hot summer of 1861 and the still 
hotter one of 1868 there is an interval of seven years ; 
since then the coming summer will make an interval of 11 
years. In the interval of 7 years there was only one sum- 
mer with a temperature slightly above the average, 1865; 
all the other summers of this interval were mostly far below 
the average ; but in the 11 years’ interval all the summers, 
with only two exceptions, have either been above or equal 
to the average summer temperature. The ratios of the 
summer and winter temperatures have not the slightest 
relationship to each other. Hence it is impossible to form 
