34 RAE. — SALINE MATTER IN ICE. 



The following are the mean temperatures of the months in 

 which the ice was measured, corrected as nearly as practicable for 

 error of thermometer ascertained by freezing mercury : — 



Below zero. 

 Mean temperature of December 1853 - - 25'5 F. 



„ „ January 1854 - - 32*4 „ 



„ „ February „ - - 38' 4 „ 



„ „ March „ - - 20-5 „ 



r 5, April „ - - 1-7 „ 



Hints towards Observations in the Arctic Regions. 

 By Prof. J. Tyndall, LL.D., F.R.S. 



1 beg to recommend the following subjects for observation : — 



1. The formation of snow crystals ; their shapes, sizes, and 

 the atmospheric conditions under which they occur. In connexion 

 with this point it will be useful to consult Scoresby's Arctic 

 Regions, Vol. I. 



2. In water contained in vessels surrounded by cold brine, I 

 once observed the formation of small hexagonal, and stellar 

 crystals. They were formed at some depth below the surface, 

 and rose to the surface. Water in the Arctic regions could be 

 rapidly exposed so as to render a verification and extension 

 of this observation possible. 



3. By permitting a thermometer to be frozen in water, and 

 exposed to a varying temperature, a rough notion of the rapidity 

 of the conduction of heat through ice might be obtained. The 

 experiment would be more valuable if a prism of ice, at a 

 very low temperature, were warmed at one end, and the flux of 

 heat determined by the observation of thermometers, sunk in the 

 ice at different distances from that end. Only, however, when 

 there is plenty of time at the observer's disposal should this 

 observation be made. 



4. The rate at which the ends of some of the Arctic glaciers 

 advance into the sea ought to be determined by a theodolite. 



5. If possible, it would be desirable to compare this terminal 

 motion with the motion at some distance from the sea. 



6. The question whether the glaciers break off to form icebergs 

 through being lifted by the water underneath their snouts, or 

 through the gravity of their overhanging ends would be worthy 

 of decision. In the former case the surface of the glacier would 

 be in a state of longitudinal compression, and no crevasses would 

 be formed ; in the latter case the surface would be in a state of 

 longitudinal strain, and crevasses might be expected. 



7. The height of some of the tallest icebergs ought to be 

 accurately determined. 



8. The moraine matter carried down by the glaciers and trans- . 

 ported by icebergs, would be worthy of observation. 



