34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. "JO 



sequence of a warm summer must be principally that more ice than 

 common is melted, particularly in the ocean eastward of Green- 

 land, and that the quantities of ice which may be available to drift 

 southward towards Iceland are thereby diminished. The result 

 therefore should follow the opposite direction from that which 

 Hildebrandsson assumes. When he points for the proof of the 

 accuracy of his assumption to the agreement between the tempera- 

 ture variations in northern Norway in summer and the tempera- 

 ture variations of Iceland in the fall and winter, it might be re- 

 marked we should expect such an agreement if, as Wojeikofif has 

 indicated, alternate variations of yearly temperature take place in 

 the odd and even years. 



It may be seen that the principal cause which Hildebrandsson 

 assumes for the variation of the ocean temperature eastward of Ice- 

 land is quite different from that which Hann has given, namely 

 the variations in the northeast trade wind. 



We shall not pursue further the details of Hildebrandsson's highly 

 interesting investigation on the action centers, because we shall 

 return to it in a later chapter when we speak of the great variations 

 in the climate of the earth in general. We may, however, remark 

 that Hildebrandsson suggests that climatic variations (especially 

 variations of temperature) of a higher order occur which tend to 

 overshadow these variations associated with the different action 

 centers. Since these variations of the higher order are noted over 

 the whole earth, they arc regarded as having cosmic causes and 

 one is apt from the first to think of them as dependent upon the 

 amount of radiation xvhieh the sun sends forth. 



H. N. Dickson (1901) has studied a great number of surface 

 temperature observations collected by the common trade ships and 

 dealing with the distribution of temperature and salt contents in 

 the surface of the North Atlantic Ocean in each month of the year 

 from the beginning of 1896 to the end of 1897. He believes him- 

 self to have shown thereby that great periodic seasonal changes 

 and also non-periodic fluctuations take place in the circulation in 

 the surface water of the Atlantic Ocean. These fluctuations appear 

 to him to be associated with the distribution of air pressure and 

 the circulation of the atmosphere, both as relates to the periodic 

 seasonal changes and to the long periodic variations. In agree- 

 ment with Pettersson and Meinardus, he thinks that the variations 

 of the surface tem})erature of the ocean influence the distribution 

 of the atmospheric pressure. 



