266 ANNUAL OF SCIENTIFIC DISCOVERY. 



necting the northern point of Scotland with Iceland, and that island 

 with the continent of Greenland. 2. The next case assumes the ele- 

 vation of the land now constituting western Europe to a sufficient 

 height to produce such glaciers as those the effects of which we recog- 

 nize in that region as having been produced during the glacial period. 

 3. The northern portion of the Atlantic is supposed to be converted 

 into dry land by the elevation of its bed. 4. In the last case, the 

 absence of the gulf stream with its influences upon the western coast 

 of Europe is assumed, together with the submergence beneath the sea 

 of a large portion of northern and western Europe. In this part of 

 his memoir the author restricts himself chiefly to the consideration of 

 these cases with the view of ascertaining how the cold of the glacial 

 epoch can be best accounted for, together with its consequent glaciers 

 of sufficient magnitude to produce the phenomena now so universally 

 attributed to them. Having constructed the isothermal lines in any 

 of the above cases for Januaiy and July, he deduces the mean annual 

 temperature at any proposed place. He can then calculate the height 

 at that place of an imaginary surface in the atmosphere, the tempera- 

 ture of which, at every point, is equal to 32 Fahr. This imaginary 

 surface must of course meet the surface of the earth along a line for 

 every point of which the mean annual temperature is that just men- 

 tioned ; and any line upon this imaginary surface (as that in which it 

 intersects the surface of a mountain,) is called a line of 32 Fahr. 

 In estimating the height of this line, the author adopts the results 

 given by Hmnboldt and others, as to the decrease of temperature for 

 an assigned increase of height in ascending from the earth's surface. 

 The next step is, to ascertain the position of the snow-line in refer- 

 ence to the line of 32. In tropical regions the former line is below 

 the latter ; in higher latitudes it is generally above it. Whenever the 

 difference between the summer and winter temperature is small, the 

 snow line has a comparatively low position with respect to the line of 

 32. By means of these and other inferences, drawn from existing 

 cases, we are able fo estimate approximately the relative positions of 

 these two lines in our hypothetical cases, and thus knowing by calcu- 

 lation the height of the line of 32 at any proposed place, we can 

 estimate that of the snow line at the same place. Now, it appears by 

 observation that nearly all the well known glaciers, of sufficient mag- 

 nitude to be considered of the first order, descend about 4,000 or 

 5,000 feet below the snow line, and that the smaller glaciers descend 

 only to smaller distances below that line. We are thus enabled in any 

 hypothetical case to form an approximative estimate of the distance 

 which a glacier would probably descend beneath its snow line ; or, 

 knowing the height of that line by the means above stated, we can 

 thus estimate the height above the sea level to which the lower extrem- 

 ity of the glacier would probably descend. The author then proceeds 

 to apply these principles to cases 2, 3, and 4 above mentioned, and to 

 determine the conditions under which glaciers, sufficiently large to 

 produce certain observed glacial phenomena, would exist in Western 

 Europe. In case 2 it would be necessary that that region should be 



