284 The Arnefican Geologist. November, i89s 
cupying extensive areas. The largest of these, having a cir- 
cumferance of something over a mile, has endured for the past 
forty or fifty years. The sand consists mainly of coarse, 
sharply angular quartz, but much feldspar is present, some 
mica and numerous fragments of schistose and gneissic rocks. 
Whenever, on the retreat of the tide, the beaches and the ex- 
posed bars are dried by the sun's heat, the returning water, if 
not too greatly disturbed by unfavorable winds, lifts as it 
creeps up the slope, the whole superficial film of sand, includ- 
ing large thin pebbles of schist, and flloats it gently away on 
its surface. The surface of the water, near the shores bearing 
the sand, commonly moves out toward the main river, even 
when the tide is rising, the incoming water flowing beneath. 
I have estimated that in the course of a year something 
like a thousand tons of sand, at a minimum, are lifted and 
borne away to new resting places by the floating power of sur- 
face tension at this locality alone. This is, of course, however, 
an insignificant factor in the general distribution of the sands, 
as compared with the waves and currents, which sometimes 
move acres of these materials to a depth of from twenty to 
seventy-five feet in a few hours. The phenomenon of the nat- 
ural floating of sand on water has been recorded by J. C. 
(iraham* and by F. W. Simonds.' 
The surface tension of water together with the mutual 
attraction which usually exists between its molecules and 
those of clay and rock minerals, is a factor profoundlx- affect- 
ing conditions on land surfaces. It accomplishes the shrink- 
age and consolidation of clays and rock dust. It facilitates 
the decomposition of rocks and retards the removal of the 
products of dicomposition to an enormous extent, although 
in exceptional cases and under certain peculiar conditions it 
acts locally to increase erosion. The existence of land veg- 
etation is directly dependent upon it. It is responsible for 
the occurrence of many efflorescent and dendritic minerals, 
and some ore deposits, and it constructs, of cla}', stalagmites, 
stalactites and coatings on other rocks. It hastens the sub- 
sidence of fine particles where streams and rivers enter the 
ocean, and further causes the removal of coarse sand and 
*Am. Jr. of Sci., III. Vol XL. P. 476, Dec, 1890. 
tAm. Geol. Vol. 17, No. I, P. 29, Jan., 1896. 
