Hill. — Problems of Arctic Exploration. 391 



it was through the channels of partially fresh water that he 

 pushed his way along the Asiatic coast. The question is 

 whether these warm areas from the rivers tend to form cur- 

 rents in the direction of the pole, and whether they supply 

 some of the varieties of ice which are mentioned by Norden- 

 skjold as occurring along the margin of the ice-pack iu the 

 Arctic Ocean. Fresh water freezes at a higher temperature 

 than salt, and it may be that, as the fresh and warmer water 

 of the rivers pushes northward, an open water-way may be 

 produced in summer in the direction of the pole ; but it is im- 

 possible to suppose that the movements can be regular and 

 continuous, seeing that the physical conditions are undergoing 

 seasonal modification both with regard to ocean- and river- 

 waters. 



Now as to aerial conditions : These, it appears to me, are 

 even less favourable to a successful voyage to the pole than 

 the oceanic. During the time that the " Vega" was wintered 

 in latitude 67° N. the temperature fell as low as 45-7° below 

 the zero of centigrade, and for months it appears that the 

 cold was more severe than anything experienced in more 

 southerly latitudes. But this low range of temperature, we 

 may suppose, would be greatly exceeded by ascending in a 

 balloon over the areas where tests were taken. Within the 

 tropics an ascent of less than 20,000 ft. brings us to the condi- 

 tions of an arctic climate, but it would be difficult to say what 

 temperature would be probable at an elevation of 2,000 ft. 

 above a spot where the thermometer showed 46° of frost, 

 centigrade. 



Andree and his fellow-passengers held that they could regu- 

 late their balloon so as to keep within 400 ft. or 500 ft. of the 

 surface. But the cold even at this moderate elevation would 

 be likely to increase in intensity as the pole is approached, 

 and it is doubtful whether life could be sustained within the 

 polar area at an elevation of some hundreds of feet above the 

 surface. It has been pointed out how the ocean temperatures 

 at great depths tend to coincide or approximate each other in 

 all zones, but the same thing takes place in the atmosphere,, 

 with this difference : that we know absolutely nothing as to 

 temperature for heights beyond 40,000 ft. Suppose, how- 

 ever, that an imaginary line is drawn from an elevation of 

 20,000 ft. at the equator to the sea-level at 80° of north 

 latitude, it would represent roughly the descending line of 

 corresponding temperature in the atmosphere, or, say, the 

 freezing-point on a centigrade thermometer. Now, it will be 

 manifest that, if the freezing-point at the equator, or within 

 the tropics for that matter, is constant in the atmosphere at 

 an elevation of 20,000 ft., and the freezing-point in the shade 

 is constant at 80° N. latitude at the sea-level, just as in the 



