22 • D. P. PENHALLOW ON 



and to have an average diameter of 0-088 mm. Comparing these with corresponding data 

 for specimens from other localities, we find no essential variation. In two specimens repre- 

 senting the " iine-grained " variety (plate I., figs. 2 and 3) the resiu-passages show in 

 each case an average of thirty-seven per square centimetre, while the average diameter 

 is 0'093 and 0'121 respectivel}'. The distribution is approximately in rows in the spring 

 wood. 



From these facts it is evident that the slight deviations to be found in the smaller 

 number and size of the resin-passages are such as may be met with under ordinary con- 

 ditions, and therefore indicate no unusual alterations. 



Larix occidentalis. 

 (Plate I., fig. 4.) 



Desiccation in the open air of a dry room for a period of three months developed a 

 number of very prominent star and cup shakes. With the introduction of furnace heat 

 these increased both in numljer and size, so that at the end of eight months the log was so 

 strongly shaken as to suggest a speedy separation of the parts, were the supporting iron 

 bands to be removed. An examination of these fractures in microscopical section disclosed 

 the fact that (1) they are practically independent of the medullary rays (see fig. 1) ; (2) 

 that they follow between radial rows of the thick-walled summer tracheids ; and (3) that 

 their tangential extension is determined in the summer wood in a similar manner, or that 

 in the spring wood it is carried directly across the tracheids, the thin walls of whieli are in 

 consequence variously ruptured. 



The section removed from the log at the end of three months and submitted to a 

 farther separate desiccation of one month in a warm, dry room, showed no change beyond 

 the conditions noted at time of section, thus making it again clear that, as in the case of 

 Pseudotsiiga, seasoning in the log is an essential condition to the development of excessive 

 shakes. 



Section a, submitted to maceration for one month and afterwards desiccated for two 

 and one-half months, showed no farther change, although the maceration was accompanied 

 by a copious fungoid growth. Section b, subjected to maceration for four and one-half 

 months and afterwards to desiccation, also showed no change, although similarly accom- 

 panied by a copious fungoid growth ; so that here, as in the Douglas fir, such conditions of 

 incipient decay produce little, it any, eftect towards the final breaking up of the structure. 



Under the hatchet or chisel the wood splits up with great facility, a line of fracture 

 arising with every blow of the mallet, but taking a direction entirely independent of the 

 instrument. Here the first tendency seemed to be for the fracture to follow the line of the 

 medullary rays, and, secondarily, to take an irregular radial or tangential course, the result 

 beina:, however, the rapid breaking up of the structure in all directions, into rather small 

 pieces. Exudation of resin was observed in the sap wood, but it was not excessive. 



The growth-rings are here sharply defined, though rather irregular, measuring I'O to 

 1-5 mm. in thickness. The sunmier wood, which is dark, resinous and very prominent, is 

 generally one-fourth to one-half the spring wood, or upwards of one-third the total thickness 

 of the growth-ring. (Plate I., fig. 4.) 



Microscopical sections show the resin-passages to be scattering and not localized in any 

 special portion of the growth-ring. They number about fifteen per square centimetre, and 



