CLIMATIC CONDITIONS OF THE UNITED STATES. 159 



represents a variation of 10 days in the average length of the frostless 

 season. West of the one-hundred-and-third meridian the stations are 

 so few, so poorly located, and the periods through which observations 

 have been accumulated are in many cases so short, that it was not 

 deemed advisable to continue the chart farther west. Here, however, 

 the data themselves are placed upon the map. 



The chart just described is the first one of the length of the frost- 

 less season to be published for the United States, and its appearance 

 is to be considered as marking a very great step in advance in the 

 climatology of the country. It should be of great value to agricul- 

 turists and ecologists, and it is to be hoped that the future may wit- 

 ness more uniformity and completeness in the keeping of the records 

 from which a more perfect chart of this feature, not unaccompanied 

 by an adequate tabulation of the source data, may evenually be 

 obtained. 



Since our own studies have been so largely based uj on the average 

 length of the frostless season as the duration factor for many climatic 

 conditions, the bare chart (as presented by Day) has been of com- 

 paratively little value in our work. After the publication of the 

 Summary by Sections, Mrs. Livingston made a recalculation of most 

 of the great mass of derived data that had already been prepared, 

 using the new average lengths of the frostless season obtained from 

 the average dates of the summary. The result exhibited some con- 

 siderable modifications in our derived data. Since the information 

 of the summary is later and much more complete than that given by 

 Henry, we have adopted the data from the latter source as the basis 

 of much of our work, using those from Henry's Bulletin Q only for a 

 few stations for which the summary fails to give frost data and for 

 which these are furnished by the other source. 



In table 2 are presented all the frost data that have been used in 

 our studies. In the first column are given the names of the stations, 

 alphabetically arranged under each State, the States being also 

 alphabetically arranged. An n after the station name denotes that 

 the observations were made in the vicinity of the place marked. In 

 the second column are given the altitudes of the stations (in feet) 

 so far as these have been available.^ The third and fourth columns, 

 respectively, contain the average data of the last killing frost in spring 

 and of the first killing frost in autumn. These data are quoted directly 

 from the Sunmiary by Sections (the number of the section in which 

 the station occurs being given in parentheses directly after the station 

 name) excepting in relatively few cases (indicated by an H after the 

 station name in the first column), where the average dates have been 

 obtained from Henry's Chmatology of the United States. In a few 



'These altitudes have been obtained from several sources and we have been unable to verify 

 all of them. If there are errors in some cases they are probably but slight ones. 



