550 Proceedings of the Royal Society 
tion of saline matter. In by far the greater number of streams and 
lakes in Scotland, whether Highland or Lowland,Hhe salts met 
with are the same, viz., carbonates and sulphates of the three 
bases, lime, magnesia, and soda, and the chloride of their metalloids, 
calcium, magnesium, and sodium. Of these the chlorides are 
usually most abundant, the sulphates least so; and of the bases, 
lime is commonly predominant, magnesia the contrary. But fre¬ 
quently in the Highland streams the proportion of all is so small 
that most of the ordinary liquid tests scarcely affect them. In 
the water now under consideration, for example, magnesia, among 
the bases, was not indicated by the alkaline phosphate of ammonia; 
nor was sulphuric acid, among the acids, by nitrate of baryta; 
even lime was doubtfully indicated by oxalate of ammonia; 
chlorine, too, was scarcely indicated by nitrate of silver in a small 
test-glass, and required a quantity amounting to six or seven 
ounces to yield an undoubted faint mist; and permanganate of 
potash did not denote organic matter except faintly. Acetate of 
lead, however, by acting on both combined carbonic acid and 
organic matter, showed a haze even in a small quantity of the 
water; and so did tincture of potash-soap, by virtue of the decom¬ 
posing influence on it of earthy carbonates and free carbonic acid 
together. 
After frequent trials I am inclined to think, that for practical 
purposes, when organic matter does not require to be taken into 
account, we seldom need any other test for ascertaining the relative 
purity and usefulness of these waters than the late Professor Clark’s 
soap-test. In the present instance this denoted in several trials 
only 1*04 degrees of hardness, which is equivalent to that much of 
carbonate of lime in an imperial gallon of 70,000 grains of water. 
From frequent observation of the effects of this and other liquid 
tests, I feel assured that the total solid contents could not have 
been more than a 25,000th of the water, and was probably nearer a 
30,000th. 
In the third place, this composition, viz., little saline and ex¬ 
tremely little organic matter, would lead to the expectation that the 
water will corrode lead. And so it does, but not powerfully. A 
thin plate of lead, with 4J square inches of surface, weighing 437 
rains, was suspended by a lead rod in this water. In twenty-eight 
