54 Proceedings of the Royal Society 
waters in a mica-slate country are slightly — very slightly 
coloured. 
The water of Loch Katrine is a well-known and characteristic 
example. Some years before the proposal was first entertained to 
use it for supplying Glasgow, I found it to contain only a^40, 000th 
of solids. When compared with a fine spring water, however, it 
now presents in a 16-inch glass tube an appreciable, yet very faint, 
yellowness. In hardness it indicates only O’ 65 by the soap-test, 
or the equivalent of a 108,000th of carbonate of lime. In corre- 
spondence with this great purity it acts powerfully on lead. In 
three weeks, a lead plate one inch and a half square, lost 2*53 
grains in weight, which is exactly the loss sustained in distilled 
water in the same time ; and crystals of carbonate of lead were 
formed in profusion. 
The water of Loch Lomond is a less familiar instance of the 
same kind. 
Loch Lomond is twenty miles long, and at its southern or outlet 
end, rather more than four miles and a half wide. Its average 
elevation is only 22 feet above high-water mark. Eight miles 
north of its outlet it suddenly contracts at Ross Point to rather 
less than a mile across ; and the northern division of twelve miles 
in length varies in breadth between a mile and only a fourth so 
much. The lower wide division of the loch, at a short distance 
from the shore, varies in depth on the whole from 8 to 12 fathoms ; 
and these soundings continue till near Point Ross, where there is 
a rapid increase to 32 fathoms. This continues to be the average 
in the middle of the lake, till at the next contraction in its width, 
opposite Rowardennan Point, where it singularly shallows at once 
to 9, 8, and 7 fathoms. A mile further up, after another swell, it 
quickly deepens at a new contraction at Rhuda Mor (the Great 
Point) to 65 fathoms ; and for five miles further north the sound- 
ings first steadily deepen by degrees to 105 fathoms, and then 
shelve to 80 opposite Inversnaid ; above which point the lake 
becomes both much narrower and greatly less deep (Admiralty 
Map). My observations on its waters were made near Tarbet, 
which faces the middle of the very deep five-mile reach, where the 
soundings in mid-channel are never under 85, and at one place, 
opposite Culness farm-house, attain the extreme depth of 100 and 
