86 ANATOMY OF THE GYMNOSPERMS 



In the distribution of the pits an important feature appears in 

 the numerical variation in different parts of the ray. For diag- 

 nostic purposes it is necessary to have reference to the number 

 of pits not upon the entire surface of an individual cell but 

 within the limits of a spring or summer tracheid, as the case 

 may be. They are invariably most numerous in the region of 

 the earliest spring tracheids, usually diminishing toward the 

 summer wood, where the change may sometimes take place ab- 

 ruptly, and in which they are most commonly reduced to one, 

 with occasional obliteration in the most highly modified tracheids 

 last formed. A similar law of distribution is applicable within 

 the vertical limits of the ray. When these structures are several 

 cells in height the number of pits is typical, and, within certain 

 narrow limits, constant for all except the marginal cells. Thus 

 if the normal number is one to two for the central cells, it may 

 sometimes rise to four, six, or eight in the marginal cells only, 

 and such exceptions must be noted in diagnosis. When the 

 ray is only one cell in height the number of pits agrees with 

 that for the marginal cells. Such numerical variations possess 

 but little value for generic purposes, but as a specific character 

 they may be held to constitute the principal differential feature 

 in the last analysis. These relations are expressed typically in 

 the genus Sequoia, the two species of which may be definitely 

 differentiated. S. gigantea is characterized by oval and com- 

 monly narrowly bordered pits, the broadly oblong orifice equal to 

 the outer limits of the pit and chiefly parallel with the cell axis, 

 one to two, more rarely three to four, per tracheid. In some- 

 what sharp and definite contrast to this, S. sempervirens has 

 large, oval, narrowly bordered pits, two to six per tracheid, the 

 round or broadly oblong orifice being either parallel with or diag- 

 onal to the cell axis. In Libocedrus the pits are small, narrowly 

 bordered, oval, with a lenticular, diagonal orifice, one to four 

 per tracheid. Or again, in Larix americana, the pits are "two to 

 six per tracheid, becoming distinctly smaller toward the summer 

 wood where they are abruptly reduced to two and finally one 

 per tracheid." In Cupressus pisifera the pits are " chiefly two 



