22 On the Causes of Change in the 



the configuration of land above-mentioned, is to determine how 

 far it might afford this littoral communication with a temperature 

 of the ocean sufficiently high to admit of the dissemination along 

 it of the species alluded to. 



The next case is that in which the basin of the Atlantic should 

 be converted into dry land, so as to unite the old and new conti- 

 nents. This would give to our own region the extreme continental 

 climate of Northern and Central Asia. According to my estimate, 

 we should then have for Snowdon, 



Temperature of January . . 7° F. 1 ^. w - Q0 _ 

 July .... 66°-5 ) Dlfk ' 16 b ' 

 Mean annual temperature . 29°*75 



The summer temperature would be increased 5°-5 F., but the 

 winter temperature would be reduced 45°, and the mean annual 

 temperature 20°. 



In discussing the fourth case, in which the Gulf Stream is not 

 supposed to exist on our own shores, and a great part of Europe 

 is assumed to be submerged beneath the ocean, I have shewn that 

 the mean annual temperature would be very nearly the same in 

 western Europe and in the latitude of Snowdon, as in the case 

 above considered of simply the absence of the Gulf Stream. The 

 conditions under which the Welsh and Irish mountains would be 

 placed, supposing them extant above the sea while the neighbouring 

 region was submerged, would be very similar to the existing 

 conditions of the Falkland Islands, and the island of S. Georgia; 

 and a comparison with these islands leads me to conclude that the 

 estimate above given of the mean annual temperature of Snowdon 

 (42° F.) is two or three degrees too high. I have considered 39° 

 or 40° F. to be a nearer estimate. In fact, a great part of the 

 misconception which has existed respecting the possible past tem- 

 perature of this region has arisen from our regarding its present 

 temperature as the normal temperature for our own latitude, and 

 that of places like the island of S. Georgia, in corresponding south 

 latitudes, as the abnormal temperature ; whereas the exact reverse 

 of this is the actual case. 



Having determined the positions of the isothermal lines for any 

 particular hypothetical case, we can determine, for that case, the 

 mean annual temperature at any assigned place. The object 

 which I have next preposed to myself in this paper is more espe- 

 cially to determine the conditions under which glaciers would exist 

 in those parts of western Europe where traces of their former 

 existence have been observed. The principle on which I have 

 proceeded is easily explained. The snow-line is that line on the 

 side of a mountain which forms the highest limit to which the 

 boundary of the snow ascends during the year. It bears an im- 

 portant relation to all glaciers, being that limit below which the 



