158 ENVIRONMENTAL CONDITIONS. 



determining the number of days intervening between the average dates 

 of the last kiUing frost in spring and the first in autumn. It was 

 reahzed that these average dates for many of the stations were not 

 based upon adequate observations, but it was thought desirable to 

 make what use was possible of the data at hand as being the best that 

 were available, and our studies progressed satisfactorily, using the 

 lengths of the frostless season thus derived. 



About 1910^ a series of frost data for the United States, much more 

 complete than that presented by Henry, became available through 

 the publication by the United States Weather Bureau of the 106 

 summaries by sections. In the following year appeared Day's ''Frost 

 Data of the United States" {loc. cit.,) which comprises several forms 

 of frost charts, including one (Chart V) of the average length of the 

 crop growing season, days. This bulletin contains no presentation of 

 the data from which the charts were prepared and no reference to any 

 publication wherein the student may find these,^ but, through the 

 kindness of Professor Day, we have been informed that most of the 

 data for the preparation of the charts were taken from the summary 

 above mentioned. 



Under the heading ''Source of Data," the following statements are 

 made in Bulletin V: 



To secure data that would show more nearly the actual conditions that prevail in the 

 fields, orchards, and gardens, the most extensive compilation of frost data yet undertaken 

 by the Weather Bureau has been accompHshed and the results have been spread upon the 

 accompanying charts. 



The data from approximately 1,000 of our cooperative stations having the longest records, 

 usually from 10 to 30 years, except in the most newly settled localities of the West, where 

 records for shorter periods only are available, have been summarized, and the local con- 

 ditions due to physical environments brought out in much greater detail than has heretofore 

 been attempted. 



These charts being based upon the results of observations made in the open country 

 and therefore not subject to the artificial conditions prevailing in the large cities where 

 the regular stations of the Bureau are mainly located, differ from any that have appeared 

 in the past in that areas having peculiar climatic features not heretofore shown on such 

 charts are now clearly set forth. 



Chart V of Bulletin V shows isoclimatic lines, for the country east of 

 the one hundred and third meridian, "the average length of the crop 

 growing season, days, being the number of days between the average 

 date of the last killing frost in spring and the average date of the first 

 killing frost in autumn. " The distance between any two adjacent lines 



* Summary of the climatological data for the United States, by Sections, U. S. Dept. Agric, 

 Weather Bur. This elaborate presentation of "all the available data as they stand" (reprint of 

 section 1, page 1, introductory remarks) comprises 106 separate pamphlets, all of them without 

 any date of publication. They include, for the most part, data for periods extending through 

 1908 or 1909.) 



^^ -On page 4 of Bulletin Vll we find the somewhat unsatisfactory bit of information which follows: 

 "Tlie chart showing the average length in days of the crop growing season was prepared from a 

 somewhat different list of stations than was used for the charts of average dates of frosts, hence 

 an actual determination of the length of the season from charts i and ii might differ a few days 

 from the date shown on chart v. " But the data are not given for any of the charts. 



