1851.] HOPKINS ON CHANGES OF CLIMATE. 85 



feet, upwards of 700 feet below the snow-line. On the south coast 

 of Iceland the mean annual temperature is about 40°, and the height 

 of the line of 32°, consequently, less than 3000 feet. The height 

 of Eyafialla Jokul is about 5500 feet, and the height of the snow- 

 line is probably much the same as at Snsefell. The glaciers there 

 are stated to descend nearly to the level of the sea. 



I have estimated the mean annual temperature of the island of S. 

 Georgia at 38°. Consequently, the line of 32° will be at about the 

 elevation of 2000 feet. It is very desirable that more accurate ob- 

 servations should be made on the height of the snow- line in that island 

 than, I believe, have hitherto been obtained. The vague assertion 

 that the snow-line there descends to the level of the ocean, has pro- 

 bably arisen from confounding that line with the level to which the 

 glaciers descend. All we seem to know is, that glaciers descend to 

 the margin of the sea ; but before we can reason conclusively on 

 this, as a case analogous to that of Snowdon or of other mountains in 

 our own islands, it is necessary to know more than I have at present 

 been able to ascertain, respecting the height and configuration of the 

 mountains from which the glaciers descend. Mr. Darwin in his 

 Journal quotes Cook's description of the island, but it contains no 

 accurate information on the points in question, although it would lead 

 to the inference that the snow-line must be considerably below the 

 line of 32° F. 



It would seem very possible then, that the snow-hne on Snowdon 

 in the present hypothetical case might not be higher than the line of 

 32° F., the height of which is above estimated at 2200 feet. Glaciers 

 might thus descend from a snow-line little more than 2000 feet high 

 to the level of the sea. 



27. If, in addition to the hypothesis of the absence of the Gulf- 

 stream, we adopt that of a cold current from the north, sweeping over 

 the submerged portions of northern and western Europe, we shall 

 have an additional cause which might probably lower the mean annual 

 temperature of Snowdon and the neighbouring region by 3° or 4° 

 below that above assumed. Such a current would also tend to equalize 

 the summer and winter temperatures, since its effect would there be 

 principally or entirely produced on the summer temperature, which 

 might thus possibly be lowered 6° or 8°. The snow-line would thus 

 be brought at least 1000 or 1200 feet lower than above supposed. 

 This would be sufficient to account for glaciers descending to the sea, 

 not only on Snowdon, but also on the lower mountains of the west of 

 Ireland. 



28. Conclusions, but somewhat vague, have been drawn respecting 

 the former possible existence of glaciers in western Europe, from the 

 actual existence of glaciers descending to nearly the sea-level in 

 South America, in comparatively low latitudes. But in this com- 

 parison the relative heights of the mountains in the two regions has 

 been frequently, I think, overlooked. In the case we have been dis- 

 cussing, the mean annual temperatures in corresponding latitudes in 

 the two hemispheres would be almost exactly the same, and probably 

 the quantity of moisture in the atmosphere and the quantity of snow 



