NATURAL HISTORY OF THE BANKS OF THE TAY. 
87 
As in the first table, the analyses give parts of the different sub¬ 
stances in every 100,000 parts of the Tay water. For comparison, the 
numbers on the top of this table have been made to correspond with 
the numbers on the top of the former table. 
1. We have here not only the disappearance of suspended im¬ 
purities till after we pass Perth, but a considerable increase m the total 
■amount of both volatile and saline matter. We wish you specially to 
note that the total solid matter in the samples taken at high water 
at Perth, and from the town supply, is about 15 per cent, greater than 
in the sample either above or immediately below Perth. 
2. In regard to the chlorine, ikbove Aberfeldy it is almost exactly 
the same as in the first set of analyses, but it gradually increases, and 
is considerably greater in analyses No. 6 and No. 7 than it is above 
Perth, or at Inchyra at low water. These facts, taken with others I 
shall note in a little, point clearly to the conclusion that in the 
summer time at least the tidal water passes above the filter beds and 
finds its way into our water supply, accounting in great measure at 
least for the high percentage of chlorine, which Perth water 
undoubtedly has. 
3. The ammonias in the first 7 samples were also carefully deter¬ 
mined by Wanklyn’s process, and these afford valuable criteria as to 
the purity of the water, organically considered. “The free ammonia 
is that which can be removed by simple distillation of the water 
after adding a little sodium carbonate, which must first be carefully 
tested and freed from any trace of ammonia. The albumenoid 
ammonia does not pre-exist as ammonia in water, but is produced 
from the nitrogenous albumenoid substances present, by boiling the 
water, after removing the free ammonia, with a strongly alkaline 
solution of potassium permanganate. The ammonia got in this way 
is called albumenoid ammonia, and is proportional to the quantity of 
nitrogenous or albumenoid substances present in the water. In the 
sample drawn above Aberfeldy the free and albumenoid ammonias are 
both small, but after getting down the length of Logierait, both are 
very much increased. The free is increased 60 per cent., and the 
albumenoid is just doubled, and this is without doubt, we think, due 
to the introduction of sewage at Aberfeldy. The albumenoid 
ammonia goes on increasing all the way down to Perth, but the free 
ammonia decreases slowly from Delvine downwards to the filter-beds. 
This decrease is undoubtedly due to the fermentation or oxidation of 
the free ammonia into nitrous and nitric acid. In No. 5 the nitrogen 
present as nitrates and nitrites was *005 per 100,000 parts of the vater, 
and though, owing to an accident to our nitrometer, we were unable 
to determine the nitrogen as nitrates and nitrites in the other samples, 
we have no doubt but that the explanation given is the true one. 
