236 J. P. Hall— Short Cycle in Weather. 



clear case, and most of the principal rises of temperature ex- 

 hibited in all these traces were similarly associated with more 

 than usually intense storms, whose centers kept north of New 

 York while passing eastward to the Atlantic. The warm wave 

 A, in the sixth curve, Series I, culminating on Jan. 2, 1890, 

 however, was not related to a deep depression but to a shallow 

 one. But as the pressure over Georgia at that time was 

 exceptionally high, the necessary gradient existed for an exten- 

 sive movement of air from the warm southwest. This instance 

 is typical ; and several other apparent exceptions may be ex- 

 plained by similar situations. After all, it is not the actual 

 barometric readings, but the contrasts, which explain wind 

 force. These cases of high pressure over the South Atlantic or 

 East Gulf States, moreover, suggest the possibility that not only 

 are rapidly moving high and low areas intensified at times, but 

 that the sub-permanent high pressure area which in an almost 

 continuous belt extends around the globe along the 30th par- 

 allel of latitude, is also subjected to an occasional intensifica- 

 tion. The warm wave B, in the first curve, Series II, cul- 

 minating in New York on January 14, 1892, was as conspicu- 

 ous as any exhibited in the whole set of traces; yet it had no 

 precedent at St. Paul or St. Louis. This was because the storm 

 to which it was related, instead of coming across the continent 

 near the northern frontier, entered from the Gulf of Mexico 

 and pushed up to the St. Lawrence Valley by a route which 

 brought New York under the influence of its southerly winds ; 

 while places west of the Alleghanies were not so affected. On 

 May 18 and 19, 1892, a storm of exceptional intensity prevailed 

 over Lake Michigan, and came eastward on the two following 

 days. New York, however, had a cold wave, not a warm 

 one, in consequence. This was because a high pressure area 

 over New England and the British Provinces obstructed the 

 advance of the storm along its proper route — the St. Lawrence 

 Valley — and forced it to reach the ocean by dipping down 

 toward the Virginia coast. The metropolis, therefore, ex- 

 perienced strong northeasterly winds instead of southerly 

 ones. Thus, the failure of the temperature-curve to rise at 

 the expected time is explained ; and the recurrence of storm 

 activity when the 27-day period required it, really happened. 

 Most severe cold waves, in like manner, may be shown to be 

 associated with the strong northwesterly winds prevailing be- 

 tween a very deep depression which has just passed and a mod- 

 erately high pressure area behind, as on Sept. 22, 1889. (E, 

 second curve, (Series I) or between a moderate low and a big 

 high, as on Oct. 23, 1889 (C, third curve, Series I), or between 

 a low and a high that are both very intense, as on Feb. 21, 

 1890 (F, eighth curve, Series I). But sometimes, as on Dec. 



