IIKKillT GROWTH. 31 



of lii;]it at tlic disposal of tlic cm-owu tlie rate of growth ditt'ers, and there is loiiiul, thcrefor(% in tlie 

 forest trees, though very nearly the same age, trees of diHeront heights, according to the success 

 of the struggle for liglit which they have had with their neighbors. At every stage of the devel- 

 opment of a I'orest growtii, after its Juvenile period, the trees can be classilied into dominant, the 

 tallest, which grow with their entire crown in full enjoyment of light and space, oveitopping their 

 neighbors; codominant, which, although of sanie height, have their crowns luirrowed in, but still 

 unimpeded at the top; while others (oppressed) are pressed in from sides and top, and linally are 

 entirely suppressed and die. This relationship of individuals changes from time to time, some of 

 the codominant gradually falling into the class of oppressed, and of these a large number become 

 suppressed. Occasionally a codominant becomes dominant, or an oppressed one, by liberation of 

 its oppressors, througli storms or accident, finds opportunity to push forward aiul nnike up for 

 lost time. Thus, a natural growth may start with a hundred thousand seedlings i)er acre; by the 

 twentieth year these will have been reduced by death to 6,000, and by the hundredth year hardly 

 300 may be left, the rest having succumbed under the shade of the survivors. 



It is owing to these changes that in analyzing tree growtli we find great, often unaccountable, 

 variation in the rate of growth of even the same individual, and hence, in order to recognize the 

 average, a very large number must be measured to even out the deviations from the law. 



For the same reason it is desirable to classify the trees as indicated above and ascertain the 

 rate of growth of trees grown under different light conditions. To be sure trees behave also 

 somewhat differently uuder varying conditions of soil, climate, and exposure; hence, a further 

 classilication is necessary if it is desired to establish more than the mere general law of progress 

 and also to ascertain the influence of these variable conditions. 



In a general way, we find, as iu the trees grown in the open, the slow seedling stage followed 

 by a very rapid increase in the annual rate of growth, beginning with the sixth year and reaching 

 a maximum of 16 inches with the tenth year iu dominant trees. With trees wliicdi have not 

 enjoyed access to light to the same extent the maximum occurs later; hence, in codominant trees 

 it is reached, with 13 inches, in the twentieth year, while the oppressed trees reach their maximum 

 current accretion still later, namely at forty years, with less than 12 inches for the year. As soon 

 as this highest rate is reached decline takes place gradually in all classes, much faster in the 

 dominant trees than in the less-favored ones, which decline iu the rate of annual height growth 

 much more slowly. 



By the one hundreth year the annual height growth is reduced to from (> to 7 iuclies, the 

 dominant trees showing the lower rate, which continues to decline until about the one hundred 

 and sixtieth to one hundred and seventieth year, when all tree classes have come to a rate of 

 about 2 inches, at which ttiey continue to grow, slowly but evenly, for another century. 



This persistence of the height growth, which makes old trees tower 40 to 50 feet above their 

 broad-leafed neighbors, influences also the shape of the crown, whic'i does not flatten, as is the 

 case with most pines. Very old trees, four hundred years and over, rarely exceed a height of IGO 

 feet, although exceptional individuals have been found of the unusual heiglit of 200 feet. 



It will thus appear that the principal height growth is made during the first century, the 

 second century noting a persistent but only slow jjrogress. 



If we take the average of all tlie yearly accretions at any one year of the life of llie tree 

 (the average aniuial accretion at that year), the influences which have been at work during the 

 whole lifetime are of course reflected; therefore, since the juvenile period shows a slow growth, 

 the average accretion attains its maxinuim much later. This culmination of the average annual 

 accretion takes i)lace much earlier in the more favored tree classes, namely, at about the twentieth 

 to fortieth year, after that declining, while in the opju'essed it does not occur uutil the seventieth 

 year, maintaining itself afterwards for a long period. 



This difference would also appear if we compared better and poorer sites. In other words, 

 when the auTuial rate of growth is slow it remains more persistent tluui when it is rapid. The 

 persistence noted in oppressed trees indicates also the shade endurance of the species. From 

 Table IV, which gives the accretions from decade to decade (periodic accretion), we see the 

 ca])acity of the species to thrive iu si)ite of the shade, even in later stages of its life. Even after 

 ninety years of op2)ressiou, when the tree is given opportunity by increase of light, it is still able 



