462 MK. p. W. HIRMEE ON THE [Aug. IpOf, 



eastward in winter into Siberia, penetrating at times even to 

 long. 90° E., but as a rule they die out when they reach the Ural 

 or the Caucasus Mountains, and the anticyclonic regions of Asia. 



The course of cyclonic storms seems to depend upon the daily or 

 seasonal variations in the position of anticyclonic systems. Iti 

 April, for example, when the winter-anticyclone of North America 

 moves northward, and extends to the north of the Icelandic cyclone, 

 pressing the latter southward, the prevalent storm-tracks no longer 

 approach Iceland, but are situated farther south than in winter. 

 In July, moreover (fig. 19, p. 454), although storms originate in 

 the China Sea, as indeed they do all the year round, they do 

 not cross the .North Pacific, that region being then anticyclonic. 

 Other storms, however, arise in North America and in the North 

 Atlantic, and some of them invade Europe. A few penetrate into 

 Asia at that season, but as a rule they die out, as in winter, when 

 reaching the Urals, or the Caucasus. Eor a great part of the year 

 the prevalent storm-tracks of the North Atlantic nearly coincide 

 with the statistical isobars of 29'90 and 29'95 inches, as shown 

 in the chart for July (fig. 19, p. 454), that is, with the division 

 between the high and low-pressure systems. Although cyclonic 

 storms, when once started, traverse high latitudes, sometimes even 

 crossing to the north of the 70th parallel, they originate for the 

 most part in regions farther south. Occasionally, however, they 

 arise as far north as lat. 60^ N., as, for example, in Behring Strait.^ 



Applying these considerations to the meteorology of the Pleistocene 

 Epoch, it will not be impossible, I think, so far as my hypothetical 

 charts may represent correctly the general isobaric conditions 

 which then obtained, to arrive at a reasonable idea as to the regions 

 in which storms may have been prevalent at the diiferent stages of 

 that epoch. 



Referring first to fig. 18 (p. 452), the isobaric chart for winter 

 during the maximum glaciation of North America, storms may 

 have arisen then in the China Sea, as they do now, and have crossed 

 the Pacific towards America, but so long as they continued to 

 travel eastward, they must have passed to the south of the anti- 

 cyclone of the American ice-sheet, the stronger contrast between 

 the climatic zones of that period coinciding with atmospheric 

 disturbances of a strongly-pronounced character. Storms crossing 

 the Atlantic might then have often hurled themselves with much 

 violence against the western shores of Europe and the British Isles, 

 and preceded alwa5^s, as they must have been, by southerly and 

 south-westerly winds, would have tended to ameliorate the climate 

 of those regions.^ Possibly their eastward progress may have been 

 arrested then, as at present, by mountain -ranges. 



^ See Maps of storm -tracks, Bull. Int. Met. Washington. 



^ Cyclonic storms would, I think, have continued their prevalent north- 

 easterjy track from the Atlantic towards Lapland, as at present, so long as the 

 Icelando-British Channel remained open, and warm currents flowed northward 

 along the shores of Scandinavia and the British Isles. 



