206 Scientific Proceedings, Royal Dublin Society. 
and Roland came to the opposite conclusion, namely, that there is 
a higher proportion in winter than in summer. 
The experimental evidence available for the discussion of the 
effects of the seasons is also of a very conflicting nature, as will 
be seen from a comparison of the results obtained by the same 
observer in different years and of those obtained by different 
observers in the same year. 
Thus, Marié-Davy found in the year 1877 the maximum 
amount in December, while two years later he recorded the 
minimum quantity for the year in the same month. In this latter 
year he noticed a very marked decrease in October, and especially 
on the 24th of the month.! 
These differences he sought to account for by the oscillations of 
the great equatorial atmospheric currents, and was of opinion that 
they might be of service in predicting meteorological phenomena, 
such as seasons of rain and of drought, long beforehand. Without 
discussing the value of this hypothesis, Reiset pointed out that 
Marié-Davy’s results were entirely different from those obtained 
by himself in the same year (1879). ‘Thus, he did not find any 
abnormal diminution in the proportion of atmospheric carbonic 
anhydride between the 2nd of October and the 14th of November. 
“During this period,’ he says, “thirty analyses have given a 
mean of 3:01 for 10,000 of air: this number is slightly in excess of 
the general mean.” On Plate X VI. we have plotted graphically all 
the results we have been able to collect of importance, in our judg- 
ment, which bear on the subject of seasonal variations. ‘The 
amounts are the mean monthly quantities found by each observer, 
with the exception of those of Reiset which we have ourselves 
calculated from the observations made by him on certain days 
each month, but not daily. ‘Thus, the quantity given as the 
monthly average for October, 1879, isthe mean of thirteen analyses 
made by him in that month. The results, then, which we give 
as his are not strictly comparable with those of the other observers, 
but we did not like to omit them, as his work was so thoroughly 
trustworthy, and is of considerable importance. 
It does not seem possible to draw any very definite conclusions 
from the chart; but, treating the curves statistically as regards 
1 The figures were really obtained by Lévy and Allaire at the Montsouris Obser- 
vatory, of which Marié-Davy was the Director. 
