912 



SCIENCE. 



[N. S. Vol. XI. No. 284. 



they have worn in the rock ; but above the cas- 

 cades, one sees the wooded valleys of the 

 streams as mountain hollows enclosed by grad- 

 ual slopes that lead up to the heights north of 

 Mt. Webster. These features seem to corre- 

 spond to the hanging valleys of the Alps, al- 

 though their dimensions are comparatively in- 

 significant. East of Bit. Washington, the point 

 of the spur between Tuckerman's and Hunting- 

 ton's ravines has the appearance of having been 

 sheared off ; and over the verge of the steep 

 slope thus formed swings the white thread of 

 Raymond's cataract, recalling the gauzy veils of 

 the Yosemite cliffs. 



Another characteristic of the glaciated and 

 over-deepened Alpine valleys with their cliff 

 walls is repeated iu the rock falls from the 

 sides of the White mountain valleys. The 

 floor of Carter notch is heaped with rough rock 

 fragments, forming two little ponds. The ava- 

 lanche that killed the dwellers in the old Willey 

 house is famous ; the sides of the upper Saco 

 trough towards Crawford notch are scarred with 

 the paths of many other slides; and the Saco 

 flows beneath heaped granite blocks that have 

 fallen from the enclosing cliffs. Long talus slope's 

 descend from the foot of the shear cliifs of the 

 Franconia notch. These phenomena are not nor- 

 mally chai'acteristic of mountains so old as those 

 of New Hampshire, although they seem appropri- 

 ate enough to ice-cut notches in old mountains. 



The notch streams are not large enough to 

 have produced prominent waste fans ; yet 

 north of the gate of Crawford notch — evidently 

 the old divide — the streams which come down 

 from the east and west, in Gibb's falls and 

 Beecher's cascades respectively, have built a 

 double fan where the Crawford house now 

 stands, thus forming a new divide, on which 

 one stream turns south to form the Saco. 

 Probably the waste here was largely accumu- 

 lated as a delta when the district to the north 

 was a glacial lake that had its outlet southward 

 over the old divide into the notch. The atti- 

 tude of the headwaters of the north-flowing 

 Peabody river would indicate that they had 

 been diverted similarly from a southern course 

 into the south-flowing Glen Ellis river by a 

 waste filling on the floor of Pinkham notch near 

 the old divide. 



The general form and the recent changes in 

 the White mountain notches thus indicate that 

 they are the result of strong glacial erosion 

 along the course of north and south valleys. 

 Philip Emerson. 



Lynn", Mass. 



floating sand and stones. 

 Pesusal of Dr. Erland Nordenskiold's de- 

 scription {Nature, vol. 61, p. 278, January 18, 

 1900) of the floating stones which he observed 

 during his journej^ along the southwestern coast 

 of Patagonia and of Professor Simonds's discus- 

 sion of the topic in Science (Vol. XI, N. S., 

 pp. 510-512, March 30, 1900), prompts me to 

 add a locality to those already mentioned at 

 which sand and stones have been observed 

 floating on water. While camping on the 

 'Thumb,' or West Arm, of Yellowstone Lake, 

 Wyoming, in July, 1899, the writer and other 

 members of the party, among whom were sev- 

 eral geologists, saw dark patches from two to 

 six inches iu diameter on the surface of the lake. 

 These patches were numerous near the shore, 

 and occasional ones were noted as far out as 

 we could see. Examination showed that they 

 were composed of the coarse black and red 

 obsidian sand which forms much of the lake 

 shore. The sand consisted for the most part 

 of subangular particles 2 or 3 mm. in diameter, 

 but pebbles 5 or 6 mm. across were frequently 

 seen, and at least one fully 10 mm. long and 

 rudely ellipsoidal in shape was observed by the 

 writer among these floating aggregations. The 

 material was solid glass, not cellular, and prob- 

 ably had a specific gravity of 2.345 (see J. P. 

 Iddings, in 7th Ann. Rep. U. S. G. S., p. 291). 

 The sand was not very drj^, on the other hand, 

 it seemed to be rather damp, when it was 

 picked up from the shore by the gentle ripples 

 and carried out by the moderate current pro- 

 duced at this locality by the inflow from an 

 adjacent hot spring area. Ripples which did 

 not break the surface of the water did not de- 

 stroy the patches of floating sand, but crested 

 wavelets precipitated the particles at once and 

 stopped the formation of other patches along 

 the shore line. The conditions therefore at this 

 locality are a sand composed of material some- 

 what repellent to water and motion of the 



