30 The American Geologist. January, 1896 
region." Inasmuch as the rocks bordering the river are, for 
the most part, Archean granites, both red and gray, schists, 
etc., the sand resulting From their disintegration may with 
propriety be termed " granite sand," even though composed 
largely of vitreous and pink quartz with only subordinate 
quantities of feldspar, biotite, and accessory minerals. 
The morning after my arrival the river was found to be 
rising and, as I stood on the bank, at the point where we 
secured our water supply, I noticed a considerable froth and, 
what appeared to me at the time, scum passing down the 
stream. I spoke of the condition of the river to my compan- 
ion, Mr. Laurence I). Brooks, of Austin, who remarked that 
what seemed to be scum was really sand. I thereupon went 
down to the water's edge and, dipping up some of the floating 
material, was astonished to find that the patches were com- 
posed of sand, mainly of quartz. At this time, half-past nine 
or ten, the water supported a large number of patches, which 
varied in area from less than a square inch up to several 
square inches, all swept along by the current. 
This phenomenon was witnessed only while the river was 
rising. When the water began to recede no more floating 
sand was seen. The reason for this will appear later. 
Noting my surprise, Mr. Brooks stated that he had seen 
sand floating on some of the streams of Alabama, and that it 
had oftentimes been very troublesome in digging water- holes. 
To test this statement I at once scooped out some holes in 
the damp sand and, surely enough, as their sides caved in, a 
\\'\\ grains Moated away on the surface of the water which 
immediately filled the holes. Between the grains there 
seemed to exist a mutual attraction in consequence of which 
they ultimately grouped themselves into small patches similar 
to the smaller patches seen on the river. In the meantime, 
Mr. Brooks, by gently sifting dry sand from his hands onto 
the water, succeeded in forming much larger patches. 
A week later, when the river was well down and the sandy 
stretches of its bed had become quite dry on their surface, I 
gathered sand by the handfuls and sent it floating down the 
stream in such quantities that the sand-rafts actually cast 
shadows on the bottom as they passed. Then I dug several 
large holes, as I had previously done in the damp sand, but 
