268 ENVIRONMENTAL CONDITIONS. 



is thus possible to consider two different periods of plant growth in this 

 region, both of which he within the limits set by the average frostless 

 season, but neither of which is as long as that season. In some other 

 regions there is but one period of general plant growth, but this is not 

 as long as that of the average frostless season. When moisture condi- 

 tions have been more thoroughly studied it may become possible to 

 consider both the frostless season and that with moist soil, in deriving 

 the duration factor for plant growth in general, but we have not found 

 it expedient to attempt this at the present time; the relations encoun- 

 tered are too complicated and information is too meager. 



Nevertheless, we have been able to derive two duration factors for 

 precipitation, which may be superimposed upon the temperature 

 duration factor here generally employed. These two new factors 

 are the lengths of the longest rainy period and of the longest dry period 

 within the period of the average frostless season. The first of these 

 is considered here and the other will receive attention under the next- 

 following heading. 



In attempting to derive an index of the normal duration of moist 

 conditions, we have again begun our computations with the data of 

 normal daily precipitation given by Bigelow (Bulletin R). Our pro- 

 cedure has been quite arbitrary. In the case of each station the series 

 of daily normals given by Bigelow has been considered as separated 

 into a series of overlapping groups of 5 days each. Numbering the 

 days of the period of the average frostless season consecutively, days 

 1 to 5 constitute the first group, days 2 to 6 constitute the second 

 group, days 3 to 7 constitute the third, etc. The 5 daily normals for 

 each group are averaged to give the mean daily normal precipitation 

 for that group, and these means are set down to form a new series. 

 Beginning with the first (in the period of the avei;age frostless season) 

 of these new group means, the groups are marked as rainy or dry, 

 accordingly as the value of their means are or are not greater than 0.10 

 inch. Thus, a 5-day group is called rainy if its group-mean is over 0.10 

 inch, dry if this mean is 0.10 inch or less. If we designate the succes- 

 sive 5-day groups throughout the normal frostless season by the alpha- 

 bet letters, and if we follow each letter by an R or D, to denote ''rainy" 

 or "dry," as the case may be, we obtain a series more or less similar 

 to the following: AR, BR, CR, DR, ED, FD, GR, HD, ID, JR, KR, 

 LD, MR, NR, etc. In this example the first four groups (A to D) are 

 seen to be "rainy." Then follow two " dry " groups (E, F), after which 

 is a single "rainy" group (G), which in turn is followed by two "dry" 

 groups (H, I), etc. Now, the last day of each "rainy" group is con- 

 sidered as occurring in a normally "rainy" period, and the last day of 

 each "dry" group is considered as within a normally "dry" period, 

 and it thus becomes possible to determine the extents of the various 

 "rainy" and "dry" periods thus established. If, for example, group 



