126 



SCIENCE 



[N. S. Vol. XXXII. No. 812 



Stratigraphioally, the Shawangunk grit may be 

 called early Salina in age. The red beds into 

 which it grades upward indicate increasing 

 aridity of climate, compatible with the suggested 

 desert origin of the salt and gypsum deposits of 

 this age in western New York and Michigan. 



Microseisms : Otto Kxotz, Ottawa, Canada. 



It is found that disturbances are registered by 

 the seismograph, which are not due to earth- 

 quakes. They have been noted in all parts of the 

 world, and are more frequent in winter than in 

 summer. They last sometimes for days with con- 

 siderable regularity, presenting a " sawtooth " 

 appearance, and have a period of about five sec- 

 onds. They have been examined by the writer for 

 the past three years, and he has found that they 

 are essentially due to barometric pressure, more 

 specifically to areas of low barometer with steep 

 gradients. Furthermore, these areas of low ba- 

 rometer are most eflfective when they rest or pass 

 over water, i. e., the ocean. A corresponding area 

 of low barometer even with steep gradients over 

 laud, approaching the earthquake station has 

 little effect in producing microseisms. Wind 

 affects the seismograph, but in a different manner. 

 Wind effect as shown by microbarogram has its 

 counterpart on the seismogram, but not as micro- 

 seisms. 



In Europe one should find, in conformity with 

 the above deductions, microseisms recorded before 

 the arrival of the low, as the storms travel from 

 west to east; in eastern America passing from 

 land to water, and in Europe from water to land. 



Bigh Terraces and Abandoned Valleys of Western 

 Pennsylvania: E. Wesley Shaw, Washington, 

 D. C. (Read by title.) 



The features indicated by the above title are 

 the well-known gravel-covered rock shelves found 

 along the large streams of the region about 200 

 feet above present stream channels. They have 

 been ascribed by different men to submergence 

 and marine erosion; to a large ice dam at Cin- 

 cinnati or Beaver; to normal stream work; and 

 to huge local dams of ice. The data gathered by 

 the present writer seem to indicate that the high 

 terraces and abandoned channels developed as a 

 unit through the overloading of the Allegheny in 

 early glacial time and a later redissection. The 

 aggradation of that stream caused every tributary 

 to aggrade, and the coarseness, slope and other 

 characters of the deposit indicate that the tribu- 

 tary streams built up as rapidly as the overloaded 

 master stream. As the stream beds rose they 



reached the heights of one after another of the 

 lowest places in divides between small tributaries, 

 and at such times and places the currents of the 

 rivers were divided and the cols occupied. When 

 final redissection began the rivers chose the chan- 

 nels momentarily most desirable, and thus many 

 parts of valleys were abandoned. 



Glacio-lacustrine and Post-glacial Features of the 



Connecticut Valley near Hanover, N. B.: J. W. 



GoLDTHWAiT, Hanover, N. H. 



The stratified drift deposits of the Connecticut 

 Valley near Hanover, N. H., and the surface fea- 

 tures developed on them by the sculpturing of 

 tributaries to the Connecticut River furnish ma- 

 terial which is of more than merely local interest. 



An extensive clay plain, formed of thin-bedded 

 silts of very uniform composition, is believed to 

 be the heavily aggraded floor of an extinct lake 

 of the Champlain stage, not a river deposit. 



A delta of coarse gravel and sand, built by a 

 small tributary stream at its debouchere into 

 this lake, fixes its water level at 560 feet above 

 the sea, or 30 to 60 feet above the clay plain. ■ 



There are features of post-glacial river erosion 

 within the plain, including terraces, which are 

 persistent up and down the valley, and are not 

 due to local protection of ledges. There are in- 

 trenched tributary streams of various sizes, whose 

 abandoned valley floors point clearly to temporary 

 baselevels between the level of the Champlain 

 Lake, 560 feet, and the Connecticut River, 375 

 feet. The chief of these are at 450 feet and 420 

 feet. Three working hypotheses for these stages 

 are suggested: (a) rock barriers farther down 

 the Connecticut Valley, which the master river 

 has removed or swung off from; (6) tongues of 

 ice or ice blocks which impounded the waters for 

 a time, during the Champlain stage, and (c) 

 postglacial regional upwarping which was not 

 single and continuous, but consisted in a series of 

 uplifts separated by pauses. 



Large-scale contour maps made by students of 

 the Thayer School of Civil Engineering furnish 

 precise data for comparing the heights of terraces 

 on different tributaries, up and down the valley. 



The Shorelines of the Glacial Lakes in the Ver- 

 million Quadrangle, Ohio: Frank Cabney,. 

 Granville, Ohio. (Read by title.) No abstract, 

 received. 



A Quantitative Measure of Maximum Arid Defla- 

 tion: Charles R. Keyes, Des Moines, Iowa. 

 (Read by title.) 

 In a normally moist land the volume of erosion 



