10 PROCEEDINGS OF THE NEW YORK MEETING 



States Geological Survey, 1899. Upon the four sheets, plate xxxv, A, B, C, D, 

 thirteen ice barriers obstructino; tributary valleys and thereby producing high level 

 sands are represented upon the east side and fifteen upon 'the west side. Most of 

 them extend northerly and southerly, because the obstructing ice occupied the 

 main valley. Other ice-tongues seem to have similarly occupied the Deerfield 

 and Millers river valleys. 



It has been objected to the existence of local glaciers that the flow of the lower 

 ice was influenced locally by the topography. This fact is conceded for the time 

 of excessive ice accumulation, and then it will be the valley rocks which will be 

 transported downward ; but the esker is characterized by the presence of the val- 

 ley rocks, and they must have been transported even after the accumulation of 

 some modified drift. I refer to the fragments of white mountain porphyries which 

 are common in the esker, but have not yet been discovered in the neighboring till. 

 These fragments increase in number and rise as one ascends the valley of the 

 Ammonoosuc, and constitute there an upper till overlying the ground moraine. 



The presence of this lobe of ice may confirm the contention of Mr Upham, that 

 certain tributary deltas are higher than the normal principal terrace of the Con- 

 necticut. The side stream may occasionally discharge an abnormal amount of 

 water which would pile up an unusual amount of sediuient. 



Following the reading of Professor Hitchcock's paper the Society 

 adjourned, at 12.40 o'clock, for the noon recess. At 2.10 o'clock tlie 

 Society reconvened and listened to a second paper by the same author. 



VOLCANIC PHENOMENA ON HA WAII 

 BY CHARLES II. HITCHCOCK 



Remarks were made by W. H. Hobbs and by visitors. The paper is 

 printed in full in this volume. 



The next paper was 



A THEORY OF ORIGIN OF SYSTEMS OF NEARLY VERTICAL FAULTS 



BY WILLIAM H. HOBBS 



[ Abstract'] 



The point is first em})hasized that joints and faults probably differ in degree of 

 displacement chiefly, and that i)rismatic fault systems formed of two parallel and 

 intersecting series iiiay be ex])lained by simple compression of a section of crust in 

 the -same way that prismatic systems of joints have been accounted for by Becker 

 and others. The conditions of rupture under compression are discussed (a) for a 

 homogeneous crustal block without preexisting structure planes, and {h) for a crustal 

 block possessed of a network of vertical fault planes. Stress is laid upon the fact, 

 too often overlooked, that an isotropic block during compression is in the aniso- 

 tropic condition of a non-isometric crystal. 



The relative depression of a crustal block along vertical rupture j)lanes due to 

 inadequate support of its load receives an independent discussion. In every area 

 where relative depression occurs there is a closed line, which may be termed the 



