20 DR. R. ANGUS SMITH ON THE 



sions an amount as small as 0*07 ; but, so far as we know, 

 they do not estimate tlie loss of oxygen, but the corre- 

 sponding increase of impurities. In the yard at the back 

 of the laboratory the amount is less than before the labora- 

 tory, and is as follows : — ■ 



Oxygen. 

 2o*8o 



21"OI 



20'94 

 20*84 

 2 1 '09 



Average 20-936 



Thus we have the series — 



In very wet weather, in front 20'98 



At all times, an average of 32 experiments 20*947 



Behind, in medium weather 20*936 



In foggy frost 20-91 



Over ash-pits 20-706 



These results surprise me as I write. They come from 

 analyses made some months ago, and without the hope of 

 such a fine gradation of qualities. They seem also to show 

 that we are exposed to currents of good air in the worst, 

 and of bad air in the best atmospheres, in towns like Man- 

 chester. This is suggested also by that number, 2i*oi 

 or more, so curiously turning up in the analyses of most 

 persons. 



In Dwelling -rooms, ^c. 



If we go to dwelling-rooms, &c., we find the same diminu- 

 tion of oxygen where there is insufficient ventilation : — 



Before the door of a house in a suburb of Manchester, 



the air gave of oxygen 20*96 



In the sitting-room, not very close 20-89 



In a very small room, with a petroleum lamp burning, 



a good deal of draught 20-84 



After 6 hours 20-83 



Pit of theatre, February 13th, 1864, 11.30 p.m 2074 



Gallery, February 15th, 10.30 p.m. 20"63 



