174 Scientific Proceedings, Royal Dublin Society. 
that they owe their origin to natural agencies. ‘These variations, 
if small in actual amount, are re/atively large, and correspond with 
fluctuations of at least 10 per cent. of the total quantity, and accord- 
ing to some observers to a much higher figure. 
- Thus the oscillations in the amount of atmospheric carbonic 
anhydride must be admitted, even by those who take the lowest 
estimate, to be comparable in extent with the variations which 
occur in atmospheric pressure, and they may have a significance 
quite as important as these latter. 
But in attempting to unravel the separate effects of those 
natural agencies which influence the proportion of atmospheric 
carbonic anhydride, very great difficulty is experienced, and indeed 
the task appears to be almost impossible in some cases with the 
material at present available for discussion. 
This is due in great measure to the fact that many of the natural 
agencies are antagonistic in their effect, and that several of them 
may be acting at the same time, thus giving a mixed result. Then, 
too, a difficulty arises as to the trustworthiness of the work of some 
of the observers—the fact that they have in many cases employed 
different methods of determination, and that very few of these have 
been tested as regards their degree of absolute accuracy—and also 
thatthe observers have collected their air samples at different heights 
above the ground. 
In our opinion, the subject is an important one, and is worthy 
of a systematic re-investigation by a number of skilled observers, 
working in different localities and employing the same method of 
determination, which shall have been proved to give results which 
do not vary from the true amount by more than two or three parts 
per million of air. 
In the following pages we propose merely to collect and review 
the evidence which has been brought forward regarding the extent 
and the causes of variation of the amount of atmospheric carbonic 
anhydride. 
Locality and Local Effects. 
The influence of locality is, as might be expected, well marked 
in the air of towns compared with that of their suburbs, or of 
country and sea-side places, and abundant evidence is forthcoming 
as to an excess of carbonic anhydride in the former. ‘he main 
