350 THE AMERICAN NATURALIST. [Vol. XXXVIII. 



walled, and 

 degeneration 



m the hard pines this feature is emphasized by a 

 of the tissue to such an extent that it is readily 

 broken out in making sections, whence it 

 characteristically appears either much broken 

 up or entirely wanting. The principal feature 

 of such rays, and the one which determines 

 their form, is the presence of a resin canal 

 in each case. Such resin canals traverse the 

 ray continuously for its entire length. They 

 present the same details of structure as the 

 resin canals which lie within the xylem. In 

 Pseudotsuga, Larix and Picea, the central 

 canal is narrow, especially in the first two 

 genera, and the epithelium consists of a single 

 layer of thick-walled cells. In Pseudotsuga 

 and Larix (Fig. 28), thyloses are altogether 

 wanting, but in Picea they are of sporadic 

 occurrence. In Pinus (Figs. 29 and 30) on 

 the contrary the canals are always distin- 

 guished by their great breadth; the epithe- 

 lium is composed of one to several rows of 

 thin-walled cells which are often resinous and 

 often much disorganized, while thyloses are 

 an invariable feature of the central canal. 



A comparison of different genera and spe- 

 cies shows that there is a somewhat striking 



the 



rays 



(tangential) to a given area of section. Such 

 variations may arise within narrow limits in 

 ^ Fusifom^""" ^^Vi\^ species according to location and 



■ (arthe «sin Lfai conditious of growth, but apart from this 

 Snchymi'^'ceiZ'and ^^^^^ ^^'^ souiewhat constant variations be- 

 (c) the terminals com- twccn different species which may be ex- 

 h^TtracheiS.^ x"^28o! prcssed by the use of the relative terms few, 

 many. No attempt has been made to define 

 such variations and more exactly, but it is quite possible that a 

 determination of the average number to a square centimeter, or 

 other convenient unit, might disclose a somewhat greater differ- 



