1851.] HOPKINS ON CHANGES OF CLIMATE. 75 



At Snowdon, from its greater proximity to the ocean, this change 

 in the summer and winter temperatures will he less. If we suppose 

 it to be 2°, we shall then have for Snowdon (see Table, art. 13, 



P- 68),- 



DifF. 



January temperature 25° 1 o^o 



July „ 59 / ^^ 



Mean 42 



This latter mean annual temperature is about 2° higher than that 

 of the south coast of Iceland (see Table, art. 13, p. 68). The differ- 

 ence of summer and winter temperatures, 34°, is greater than for 

 Iceland, where it may be stated generally to be about 18° or 20° 

 only. This actual case, however, presents a good standard of com- 

 parison for our hypothetical case. The Falkland Islands, situated 

 in lat. 50° S., and the island of S. Georgia, in lat. 54° S., furnish also 

 good standards of comparison. The conditions of the station above- 

 mentioned in western Europe (long. 10° E.), and Snowdon, of which 

 the latitude is 53° N., would, in the case we are discussing (that of 

 the absence of the Gulf-stream, and the submersion of Europe), far 

 more resemble the actual conditions of all the above places than they 

 do at the present time. We have for the Falkland Islands, — 



DifF. 



January temperature 52° "I , c,o 



July „ 34 /'^ 



Mean 43 



This is less by 2° than the estimated mean temperature for the cor- 

 responding place (long. 10° E.) in N. lat. ; the difference between 

 the winter and summer temperatures is here less by 16° than in the 

 other case. Again, we have for the island of S. Georgia (lat. 54°), — 



Diff. 



January temperature 45° 1 , .© 



*July „ 31 / ^'^ 



Mean 38 



Here the mean temperature is 4° less than the estimated mean tem- 

 perature of Snowdon ; and the approximation to uniformity of tem- 

 perature throughout the year much greater. 



The table given above (art. 17, p. 72) shows the tendency, as I 

 have already remarked, of the predominance of sea to equalize the 

 summer and winter temperatures ; and this is in accordance with the 

 greater equality of those temperatures in the Falkland Islands and 

 S. Georgia, than in the corresponding northern latitudes. It also ap- 

 pears from the table just referred to, that for the latitudes there given 

 above 40°, the mean temperature for a whole parallel of latitude is 

 higher in the south than for the corresponding parallel in north lati- 

 tude. Now if this be due to the predominance of sea in the southern, 



* This temperature is not very certain, even should the map be exact, since 

 there is no July isothermal given in the southern hemisphere quite so far south as 

 South Georgia. 



