508 PROCEEDINGS OF PHILADELPHIA MEETING. 



The last paper of the session was entitled : 



NOTES ON GLACIERS 

 BY HARRY FIELDING REID 



[Abstmctl 



1. The request made last year for observations on the advance or retreat of 

 American glaciers has brought but few responses ; these, however, show that the 

 Carbon river and North PuyaUup glaciers on mount Rainier and the Illecellewaet 

 glacier in the Selkirk mountains are retreating. 



2. If we consider a glacier in equilibrium, the amount of ice flowing through any 

 section in a year must e(iual the ice added to the glacier above that section in the 

 same time. This, in connection with the flict that there is actual accumulation in 

 the reservoir and melting in the dissipator (the region below the ncn' line), shows 

 that the flow must be greatest in the neighborhood of the iit'ir line and must di- 

 minish as we ascend or descend the glacier frt)m that region. This law of flow is 

 exact and must replace the similar empirical rule of velocity first stated by Desor. 

 It is inferred that the greatest flow in the ice-sheet of the Glacial period must have 

 been near the neve line, and that this was much nearer to the outer edge of the 

 sheet than to the center of distribution. In general, the velocity will be greatest 

 where the flow is greatest. 



3. A consideration of the flow in a glacier of indefinite length resting on a bed 

 of uniform slope leads to the conclusion that with ordinary glaciers the parts near 

 the end must owe some of their motion to pressure from behind. This has been 

 generally believed, but not clearly reasoned out. 



4. As a result of this pressure there must be in tlie dissipator a*motion of the ice 

 away from the bed. In the reservoir, on the contrary, there is a motion toward 

 the bed. We are enabled to draw approximatel}' the lines of flow followed by the 

 ice from the time of its deposition as snow to the time of its melting, and also to 

 show the positions occupied by the successive strata. 



5. If we knew the distribution of velocity and of melting we could calculate the 

 form of the glacier's surface. The vertical or overhanging ends of some Greenland 

 glaciers described by Professor Chamberlin during the past year are due to the large 

 quantity of debris in their lower layers, causing more rapid melting there. 



G. Glaciers are rarely in com[)lete e(iuilibrium with their surroundings. The re- 

 lations establishing the form of the surface bring about a state of unstable equilib- 

 rium, and this would lead us to expect the great fluctuations in the extent of glaciers 

 which we actually observe. 



The paper was discussed by G. F. Wright, R. D. Salisbury, W. N. Rice 

 and the President. The paper will l)e published in full in the Journal 

 of Geology. 



The Societ}'' then adjourned. No evening session was held. 



