On the Surface Temperature of North Sea and North Atlantic. 21 



Irish, and still more noticeably, the Norwegian coasts; and this 

 strong inflection of the isotherms is more than can be accounted, for 

 by the convection of heat due to direct current-action. On the other 

 hand, the isotherms fjirllier to the south, as they approach the coast 

 of France, are but slightly detlected. All this is part of a phenomenon 

 which was diligently studied about the middle of last century by 

 various meteorologists, and especially by Professor Dove of lierlin, 

 and Professor James Forbes of Edinburgh.* 



As these investigators showed, the effect of land in modifying 

 clinuitic temperature conditions is twofold. In the first place, the 

 effect of a neighbouring land-mass or continent is to exaggerate the 

 Variation of Temperature which is due to the seasons: the temper- 

 ature of the ocean is always steadier, or more nearly constant, than 

 that of the land or that of the inshore waters. Secondly, the Mean 

 Annual Temperature is also affected by the presence or neighbourhood 

 of land ; and this inliuence is also in the way of exaggeration, a 

 northern latitude being all the colder, and a tropical latitude all the 

 hotter, on or in the neighbourhood of land. It follows from this 

 that there is an intermediate zone where the presence of land makes 

 no apparent difference, and this, in the northern hemisphere, would 

 seem to be somewhere about 45° N. lat. As Professor James Forbes 

 pointed out, the accumulation of land in tropical Africa produces a 

 temperature in excess of the mean temperature of the earth's surface 

 at a corresponding parallel of latitude, while in Siberia the effect of 

 the great concentration of land is precisely the reverse, the tempera- 

 ture there being considerably below the mean of the parallel. In 

 short, every place upon the earth's surface has a mean temperature 

 of its own, which we may look upon as compounded of two parts — 

 (1) that which depends upon latitude alone, and (2) that which de- 

 pends on the distribution, and the relative areas, of land and water. 

 This latter factor is an exceedingly complicated one, and it is not 

 necessary for our purpose to analyse it in detail. Its influence is in 

 part direct, depending on the very different " thermal capacities " of 

 land and of water, whereby a given quantity of solar heat produces 

 a much greater change of temperature on land than on water ; while 

 the indirect effects are both numerous and important, and are chiefly 

 seen in the wind-currents as they in turn are affected by the unequal 

 heating of the land and water areas, and in the sea currents, as these 

 are profoundly modified by the topographical features of the coast- 

 line. 



But since it is not our business to attempt to analyse the nature 

 of these and other similar phenomena, let us merely attempt to ex- 

 amine the general effect of the presence of land, and of its particular 

 topographical characters, upon the sea temperatures of our area ; and 

 this we may do by taking as a standard of comparison the tempera- 

 ture features along some line in the Atlantic, sufficiently far from 

 shore, and by then representing our own North Sea and other coastal 

 temperatures, not as they actually are, but in the form of differences 

 from the oceanic standard of comparison. This is the method (already 

 alluded to) which was introduced about the middle of last century 

 by Dove, and which he called the study of " thermal anomalies." 



* " Inquiries about Terrestrial Temperature," Trans. R.S.E., xxii. pp. 75-100(1859), 

 1861. 



