Bigelow — Meteorological Elements of the United States. 415 



studies, and the record is made homogeneous for the years 

 since 1873. For the purposes of this paper there have been 

 employed 50 stations, separated into 10 groups of 5 stations each, 

 covering the United States as fairly as possible, as follows : 



1. Missouri Valley. St. Paul, St. Louis, Keokuk, Daven- 



port, La Crosse. 



2. Lake Region. Marquette, Milwaukee, Chicago, Alpena, 



Detroit. 



3. Middle States. Pittsburg, Lynchburg, Buffalo, Oswego, 



"Washington. 



4. North Atlantic. Philadelphia, New York, New Haven, 



Boston, Portland. 



5. South Atlantic. Montgomery, Knoxville, Jacksonville, 



Augusta, Wilmington. 



6. West Gidf. Shreveport, Galveston, Memphis, Vicksburg, 



New Orleans. 



7. North Plateau. Denver, North Platte, Bismarck, Yank- 



ton, Omaha. 



8. South Plateau. Yuma, El Paso, Santa Fe, San Antonio, 



Palestine. 



9. North Pacific. Portland, Roseburg, Red Bluff, Winne- 



mucca, Spokane. 

 10. South Pacific. San Francisco, Los Angeles, San Diego, 

 Yuma, Sacramento. 



For the purposes of exhibit, these 10 groups have been 

 united into three groups, I (9, L0) Pacific States, II (6, 7, 8) 

 Plateau and West Gulf States, III (1, 2, 3, 4, 5) Central and 

 Eastern States, and this is justified by comparing the curves 

 together, as well as by their geographical characteristics. It 

 should be remembered that each station is independent of 

 every other, so far as the making of the observations is con- 

 cerned, and therefore whatever harmony is found to exist 

 among the residuals for any of the 10 sections cannot be due 

 to local peculiarities. In order to exhibit the kind of agree- 

 ment that has been found, the residuals of temperature, vapor 

 pressure and barometric pressure for the group No. 1, Missouri 

 Valley, are collected in Table I. It shows that there is gen- 

 erallv a fair agreement in the signs of the residuals for the 

 same year in each of the three elements. In some years the 

 magnitude of the residuals is large, in others it is small, and 

 in still others the residuals are scattering, in which case there 

 was no pronounced external force determining the prevailing 

 circulation to move aside from the normal in a given direction. 

 The means of the groups of five stations each have been car- 

 ried forward, and combined, as stated, into three groups, and 



