Periodical Literature. 285 



and Gleditsia longitudinal rents often appear and separate bun- 

 dles of fibers. These splits do not occur along the rays nor even 

 along the large vessels which run through the mass of the wood, 

 but within the bundles of wood fibers' themselves. The rays in 

 the course of the rents are torn like the other elements, but those 

 imprisoned in the interior of the bundles are merely bent. Ex- 

 cept in the case of complete failure the lumen of the folded ele- 

 ments usually remains open, not only in the case of ligneous 

 fibers with thick walls, but also in the vessels. 



The author's studies of the natural failure of wood bears out 

 his conclusions from the test specimens. His main point is that 

 the distribution of the rays has little eflfect upon the mode of 

 failure. This is not in accord with the findings of M. Thil who 

 says: 



The sides of the medullary rays sometimes produce planes of 

 least resistance varying in size with the height of the rays. The 

 rays assume a direction more or less parallel to the lumen of the 

 cells on which they border, the latter curving to the right or left 

 to make room for the ray and then closing again beyond it. If 

 the stress acts parallel to the axis of growth, the tracheids are 

 more likely to be displaced if the marginal ray cells are weak- 

 walled. It is on that account that on a radial section of the test 

 •block the plane of rupture passes in a direction nearly that of a 

 ray, whereas on a tangential section the direction of the plane of 

 rupture is oblique, but with an obliquity varying with the species 

 and determined by the inclination of the spirals on which the 

 rays are distributed in the stem. — Constitution anatomique dii 

 hois, pp. 140-141. 



S. J. R. 



Etude anatomique de hois comprtmcs. Mitteilungen der Schweizerischen 

 Centralanstalt fiir Forstliche Versuchswesen. Band X., Heft I., Zurich, 

 1910, pp. 53-101. 



Kanehira of Formosa, Japan, after a 



Termites personal investigation, and after corres- 



Resisting pondence with seventeen different coun- 



Timbers. tries, gives a list of tree species which 



resist termites or white ants, and which 



may be termed termite-proof trees. He summarizes the 



reasons for this quality in the wood as follows: (a) presence in 



the wood of some substance which has a strong smell or taste 



