May, 1908.] Annual Wood-Increment of Acer rubrum L. 



345 



These frequency distributions are shown graphically in 

 Fig. 1. The abscissa give width in millimeters, the ordinates 

 frequencies of rings. The variation constants deduced from. 

 them are indicated on page 347. 



For the benefit of those unfamiliar with the biometric method 

 of study employed here, a brief discussion of the more sahent 

 points is appended. For a more complete statement the reader 

 is referred to Davenport's "Statistical Methods" (4) or the 

 more popular work of Pearson (6). 



JS 



«. 



.S /. U Z. S.S 3. J.r ¥. Ms- S. Ss 6 6.f y. j.s a. S.f f f.f /(?. 



Fig. 1 — Frequency curves showing variation in width of the annual wood- 

 increment in Acer rubrum. Continuous lines — bog habitat; Dotted 

 lines — woodlot habi;at. 



It will be noted that in the trees of the bog habitat, there 

 are more rings three mm. long than of any other length, 

 while in the second type of habitat the greater number of rings 

 has shown two mm. This highest frequency or most common 

 length is known as the mode. It shows clearly the prevailing 

 type of wood-accretion. The distribution decreases in both 

 directions from the mode, but least so in the woodlot habitat. 

 The practical importance of the information afforded by this 

 value is apparent. We have here the average prevailing state 

 or place-habit of a similar lot of individuals from two distinct 

 places. It is a characteristic which has been determined by 

 influences covering a period of time (the age of the trees) long 

 enough to eliminate the effect of incidental fluctuations in the 



