CLIMATE OF HISTORIC TIMES 54B 



tively constant regions and relatively shifting regions. Gregory, in the 

 article already cited, has gone even farther in this direction, and called 

 the attention of geographers to the fact that an impartial survey of the 

 evidence from the world as a whole shows that, so far as historic changes 

 of climate are concerned, there are three types of areas : those which have 

 grown drier, those which have grown moister, and those where no change 

 has occurred. He has wisely adopted the plan of determining the type of 

 change in each region separately and plotting it on a map of the world. 

 This map is so valuable that it is here reproduced as figure 15. Two 

 changes have been made in it. In the first place, I have added some plus 

 signs in Yucatan and Guatemala to indicate increased precipitation on 

 the basis of the ruins described above. In the second place, I have drawn 

 lines through the U^s, which indicate unchanged conditions in north 

 Africa, Syria, Makraii, and the Caspian region, and have added minus 

 signs in their places. This is because the U^s of the original map are 

 largely based on the evidence of the palm and other vegetation, while 

 almost no attention is paid to ruins, old strands, and other evidences 

 which seem to me conclusive. Moreover, as Gregory himself states in a 

 later publication,^^ he did not take the pulsatory hypothesis into considera- 

 tion, and hence when he found evidence both of drier and wetter condi- 

 tions he was forced to conclude that there had been no change. Other 

 portions of the map remain as originally drawn; I do not feel that my 

 knowledge is sufficient to justify any expression of opinion as to the por- 

 tions south of the equator. In the northern hemisphere there may per- 

 haps be some question as to China and the far northern portions of 

 America, but the map at least represents the best knowledge that is avail- 

 able thus far. 



Before proceeding to a consideration of the possible causes of changes 

 of climate during historic times, let us sum up our conclusions as to their 

 nature. The changes appear to be pulsatory in nature, but have no defi- 

 nite periodicity. The same phenomena recur in cycles of all magnitudes 

 from the little cycles now in progress to those that have a length of thou- 

 sands of years. In general the changes vary from region to region in 

 such a way as to suggest that they are due to an alternate poleward and 

 equatorward shifting of the great climatic belts. The matter is more 

 complex than this, however, for in the same latitude one side of a con- 

 tinent may differ from the other. So far as can be detected, historic 

 changes of climate do not seem to differ from those of the Glacial period 

 or from the little variations that we see from year to year except in degree. 



87 Geographical Journal, vol. 44, 1914, p. 210. 



