230 Clayton — Six and Seven Day Weather Periods. 



as points which can be easily determined, take first the most 

 southern point reached by the storms in their sudden deflec- 

 tion southward in the Missouri River valley, next the time of 

 their sudden bend northward in the eastern lake region, and 

 thirdly the most northern point reached before turning south- 

 ward near the mouth of the St. Lawrence River. The ob- 

 served times the storms were at these points and the times cal- 

 culated from the assumption of exactly equal intervals of 7d. 6*5 

 hrs. were as follows : 



II. 



III. 



Observed. 



Calculated. 





May 8, 



8 P. M. 



May 8, 



6 p. 



M. 



" 14, 



8 P. M. 



" 15, 



12 p. 



M. 



" 19, 



8 P. M. 









" 23, 



8 P. M. 



" 23, 



7 A. 



M. 



" 30, 



8 A. M. 



" 30, 



1 p. 



11. 



May 10, 



8 A. M. 



" 10, 



8 A. 



M. 



" 15, 



8 P. M. 



" 17, 



2 p. 



M. 



" 21, 



8 P. M. 









" 24, 



8 P. M. 



" 24, 



9 p. 



JI. 



June 1, 



8 A. M. 



June 1, 



3 A. 



M. 



May 1 0, 



8 P M. 



May 1 0, 



8 p. 



M. 



" IT, 



8 P. M. 



" 17, 



12 p. 



M. 



" 22, 



8 A. M. 



" 25, 



7 A. 



M. 



. " 26, 



S A. M. 









June 2, 



8 A. M. 



June 1, 



1 p. 



M. 



In the southern group of storms, the times of beginnings 

 and of the most southern positions reached may be taken as 

 points for comparing the times of occurrence of the storms 

 with the times computed on the assumption of exact regularity 

 in the intervals between them. The results are as follows : 



Observed. Calculated. Observed. Calculated. 



May 4, 8 p. m. May 4, 2 p. m. May 6, 8 a. m. May 6, 8 a. m. 



" 11, 8 p. M. " 11, 8 p. m. " 13, 8 p. M. " 13, 2 p. M. 



" .. ... " 19, 3 a.m. " .. ... " 20, 9p. M, 



" 26, 8 P. M. " 26, 10 A. M. " 28, 8 P. M. " 28, 3 A. M. 



The storm tracks arrange themselves into new groups at the 

 end of intervals averaging slightly less than 27 days, and pre- 

 sumably corresponding with a solar rotation. The preceding 

 chart gives only one of many striking instances which might 

 be given showing the arrangement of the storms in groups, 

 and the approximate regularity of the intervals between the 

 storms. 



Surely no one can believe all the data here presented acci- 

 dental, and no unfavorable data have been suppressed. There 

 is, however, one source of difficulty in investigating and 

 understanding these periods, and that lies in the fact that, 

 besides these frequent divisions of the six and seven day- 

 periods into periods of half these lengths, they are clearly 

 associated with others of greater lengths. The shortest of 



