169 
AMMONIA AND NITRIC ACID IN RAIN-WATER. 
obtained in four or five separate testings, in each of which the 
error of observation is 0*25 of a septen. of ammonia at 1 , and that u 
requires considerable practice, and some dexterity and accuracy o o s 
ration, to avoid a greater range of error. Upon tlic whole, then, consu en g 
'tatin operating upon only a single litre of rain-water, the minimum eiror o 
"bservation, if any, will even with the most perfect manipulation genera > 
•mount to 3 or 4 per cent, of the total ammonia contained in the water, and 
further, that in any but the most practised and skilful hands the error may 
1 * much more than that supposed—indeed more than multiplied by the 
lumber of testings— we were led to abandon this second method also, not¬ 
withstanding its very obvious advantages so far as convenience and rapidity 
lri ‘ concerned. These, objections will of course apply with still greater 
We when rains containing less than tlu- average amount of ammonia arc 
'['‘•rated upon, and especially in the ease, of the waters of springs and rivers, 
^average amouut of ammonia in which would seem, according to M. 
boussingault, to be very considerably less than iri any rain-waters. 
hi subsequent experiments. th«.n-l«»iv. wi- adopted th.- phm 1,1 "[’''rating 
qion several litres of water in the first instance, reducing the bulk by suc- 
?*!** distillations to one-half, until thus brought to a convenient, amount 
*ur final distillation, and subsequent testing of measured proportional amounts 
0 urn distillate in the manner described in reference to Methods. Ibis 
0 ification, we have since observed, was suggested by M. Boussingault 
fi?T \j 1 would seem he did not generally adopt it. In the next lablc 
ar(! given the results obtained by Ibis Method 3, in which 3, 6, or 
" • _°f the rain-water were first operated upon, these quantities, as the 
be, being reduced to 24 ounces by successive distillations as 
farr a * ,0vc ' This product was then finally distilled, and the distillate 
mnr?!r illt0 quarters, the first three of which were separately super- 
ilkah t to‘ th the test - acid at 1°» and subsequently neutralized by the test- 
», Tablk III. 
' e hod S —Distillations to one-half, until reduced to 24 ounces, redistilled, 
_ and successive quarters tested. 
y ear 1854. 
Experiments. 
lbs. of 
Scptem9 of ammonia at 1° 
in 
Ammonia 
per 
million 
rain-water. 
Months. 
rain-water 
taken. 
1 st 
quarter. 
2 nd 
quarter. 
3rd 
quarter. 
Total. 
January | 
February/ 
March... ^ 
April* | 
May ...J 
1 
2 
1 
2 
1 
1 
2 
1 
2 
6 
6 
3 
3 
3 
12 
12 
12 
12 
12- 5 
13- 5 
6-5 
6-7.5 
8-25 
34 50 
35-25 
16-5 
10-25 
125 
0’75 
1-5 
1-5 
0-75 
1- 75 
2- 25 
1-25 
1(H) 
1*5 
10 
1-25 
1-00 
0-25 
1*0 
10 
10 
075 
15-25 
15-25 
0-25 
0-25 
025 
37- 25 
38- 50 
18-75 
1 S00 
} 0-779 
| 0 945 
0-945 
j 0-967 
J- 0-469 
Water. 
Tenths. 
Total. 
per 
million. 
FT - 
1 st. 
2 nd. 
3rd. 
4th. 
5th. 
1 'iOozs. 
9-0 
1-5 
0-5 
0-5 
0 5 | 12 0 | 0-980 
* April checked by Method 2. 
