DISTRIBUTION OF AMMONIA. 271 



the ammonia to settle upon it. For this purpose small 

 flasks were hung in various parts of the laboratory, and 

 examined daily. The flasks would hold about 6 ounces of 

 liquid ; but they were empty ; and the outer surface was 

 washed with pure water by means of a spray-bottle : it 

 was done rapidly, and not above 20 cub. centims. (two thirds 

 of an ounce) of water was used. This was tested for am- 

 monia at once with the Nessler solution. The second 

 washing, taken immediately, produced no appearance of 

 ammonia. Ammonia could be observed after an hour and 

 a half's exposure at any rate; but I do not know the shortest 

 period. The results of the washings were as follows 

 (they are the average of 34 experiments for some, and 17 

 for others — in all, 238 experiments) : — 



Height Height 



fi'om floor. Ammonia, from floor. Ammonia, 



ft. in. mgm. ft. in. mgm. 



Front laboratory 7 3 '013 4 2 "019 



Second landing 6 o "032 



Balance-room 5 i '015 o 8 "009 



First landing 4 lo -007 



Back laboratory 4 5 'oio o 6 'oio 



Entrance- lobby 6 5 '007 



Office 4 7 003 



Backyard 4 8 "036 o 7 '042 



Back closet 2 3 -105 



Midden -572 



The first three, belonging to the working-laboratory, are 

 not very regular, as we might suppose; but they never rise 

 very high, nor do they sink to the lowest. The rest keep 

 a remarkable similarity to each other from day to day; but 

 among themselves the diff'erences are very great. In the 

 second there is a disturbance caused by sweeping the floors. 

 On the other days it was requested that every thing should 

 be kept still. This of course brings in a practical difficulty 

 and limits the use of the test to cases where care can be 

 used and thoughtful observation, since there are many ways 



