314 Scientific Proceedings, Royal Dublin Society. 
to the upper surface not remaining always at the level shown in 
the drawing. I have preferred to keep the water-surface nearly 
steady against a needle-point, and to measure the flow by a burette. 
In this case let the height of the block be / centimetres. ‘Then 
: Oe ONG 
one of our cubic units 1s G c.c.; and 
8 
d, 
ate ; 6) x 60 c.c. per minute 
the total flow = -92 x (4°30)? x 3 
= 3140K7' ce. per minute. 
By this method I measured the conductivity of some peat brought 
from the Lochar Moss, near Dumfries. It was a slightly 
muddy litter-peat. It was cut from a wet part of the bog, and 
preserved in a damp-tight box until required four months later. 
Three blocks were cut, of the form sliown in the diagram, and 
8 ems. high; that is, 4 = 8. They had better have been twice or 
thrice this height, but I was short of material. ‘They were placed 
on perforated zinc; and the flow measured with a burette and 
needle as described. 
The values found for K,;.; were : 
1:30 x 10° C.G.8. units, 
0:93 < 107 4 
Meh D SO: 3 Bs 
The first two resemble the value found by the glass tube 
experiment, namely ,.,;= 1:3 x 107 C.G.8. units. 
- The third was obviously more open in structure. 
§ 7. CapILLARITy In Puat. 
Boussinesq (1904e) assumes that the capillary attraction in a 
uniform soil has the effect of reducing the pressure at the water- 
air surface by an amount which is constant all over the surface. 
This is clearly the most reasonable assumption when the surface 
is everywhere sinking. But if the surface is rising into dry soil, 
the innumerable water-air surfaces will then be convex upwards, 
and the pressure will be increased instead of diminished. And if 
