20 



Weather Bureau's cooperating station at Dartmouth College. 

 In each case the mean was obtained by averaging the maximum 

 and minimum temperatures. The normal daily mean was 

 obtained from the complete records of the years 191 0-1920 

 inclusive. 





Normal 

 DailyMean 



Deviation from normal daily 



mean temperature 



Week or Month 











Temp 



1917 1918 1919 



1920 



192 1 



January' 



173 



+ .6 



—9-35 



+ -55 



—4- 



+ 3-65 



February 



18.7 



—3-2 



—3 



5 



+3 



5 



— 



5 



+ 4-85 



March 



30.1 



— -4 



— 



9 



+3 



9 



+ 



9 



+ 8.5,5- 



April 1-7 



37-9 



+ -7 



+4 



7 



— I 



6 



— 2 



3 



+ 11. 



April 8-14 



39-4 



—6.7 



+ 



2 



+7 



7 



—4 



8 



+ 5-9 



April 15-21 



42.9 



+ -3 



— I 



4 



+ 1 



3 



+ 



7 



+ 6.4 



April 22-28 



47.7 



—4.1 



— I 



7 



—4 



3 



—4 



4 



+ 5-2 



April 29-May 5 



49-7 



—8.4 



+2 



9 



+2 





—6 



I 



+ 4-1 



May 6-12 



53-2 



—8.4 



+7 



2 



—4 



5 



—5 



I 



— 2.8 



May 13-19 



54-6 



—6.2 



+8 



3 



— 



8 



—3 



4 





May 20-26 



57-9 



—1-1 



+5 



I 



+2 



6 



— I 





+ 2.8 



May 27-June 2 



58.9 



-4.8 



+6 



6 



+3 



4 



+6 



4 



+ 4- 



June 3-9 



60.5 



+2.3 



+ 



•5 



+9 



7 



— 2 



I 



— .8 



June 10-16 



62.0 



+ .4 



—3 



8 



+4 



4 



+2 



9 



+ -4 



June 17-23 



63.2 



+2.2 



—9 



.6 



+3 



4 



—4 



5 



+ 2.4 



June 24-30 



65.8 



—3-2 



—3 





— 2 



4 



+ 



9 



+ 4- 



July 1-7 



68.6 



—2.9 



—3 



■7 



+4 



6 



—5 





+ 5-3 























The chronological chart of the 104 species recorded is ar- 

 ranged in the order of the earliest dates recorded for the species 

 for the years 1917-1921 inclusive. Practically it means that 

 the order is that in which they appeared in the year 1921, which 

 gave the earliest dates for all forms except those that appear 

 with a star after the date following the name of the species. 

 In a few of these cases it will be obsers^ed that no record appeared 

 for the given flower in 1921. 



The following interpretations as to the correlations with 

 temperatures seem to be plainly indicated. 



1917. In general the flowers were late in making their ap- 

 pearance and in the majority of cases were last of the five 

 seasons (out of 91 recorded for 191 7, 66 were last). Up through 

 No. 17 there seems to be a tendency for the season to be a little 



