234 Proceedings of the Royal Society of Edinburgh. [Sess. 
evident from the curve for Red clay Pacific on Plate II, which may be 
compared to the curve on Plate III for the Blue mud (a terrigenous 
sediment deposited in much shallower water). 
As a further still more striking contrast I also give on Plate III the 
distribution-curve for a sample of Yoldia Clay (from a layer of only 1 mm. 
thickness) taken at Uppsala, and probably deposited some 10,000 years ago 
at a depth not exceeding 150 m. in the old Yoldia Sea, which at that time 
occupied the space to the south from the Fennoscandian continent. 
The curve is seen to run quite close to the axis at r = 0*7, but rises 
abruptly from that point, no less than 91 per cent, of the amount having 
r<l/jc. 
Possibly this striking fact can be explained as due to an agglomeration 
to larger units of the finest particles under the enormous pressure at these 
depths. Another explanation might be found in a subsequent solution of 
the very finest particles, also furthered by the exceedingly high pressure. 
An investigation of this last assumption, viz. the influence of the size of a 
particle on its solubility, is at present in progress. 
VI. Summary. 
A new method for the mechanical analysis of soils and deposits has 
been developed. It consists in weighing, from time to time, on a specially 
constructed balance, the amount of sediment accumulated on a circular 
disc suspended from the balance near the bottom of a vessel containing an 
aqueous suspension of the sample. From the “accumulation-curve” thus 
obtained one finds by a series of mathematical operations a “ distribution- 
curve,” i.e. a curve showing how the amount of particles of a certain size 
varies with the latter quantity. 
This method has been applied to the study of some deep-sea deposits 
from the Challenger Office. The distribution-curves thus obtained show 
marked differences for the different samples, and reveal a surprising lack of 
very fine particles in the deposits from the largest depths, besides suggest- 
ing interesting problems for new research. 
University of Uppsala, Chemical Institute, 
September 1915.* 
* The printing having been delayed by the fire at Messrs Neill & Co.’s printing works. 
