20 D. P. PENIIALLOW ON 



that trees of that period must (1) have possessed unusually large resin-passages, and (2) 

 these must have been developed in enormous quantity in the individual trees, or else these 

 latter have been produced in enormous numbers, unless we admit the deposit to have taken 

 place over great periods of time. ISTo existing parallels could be found, and the hypothesis 

 therefore failed for want of sup[)ort in fact. 



At this stage of our inquiry the statement reached us that in the Kootanie valley of 

 British Columbia there are to be found certain species of tamarack and other coniferous 

 trees "which secrete plates of resin in such a manner that, when the wood is sawed up and 

 left ill tiic sun, it fulls to pieces from melting of the resin." Here, then, appeared to be a 

 possible solution of the question in hand, first, in the occurrence of resin in a very unusual 

 manner and in great quantity, and, second, in the occurrence of these trees in the same 

 region where, in past ages, the coal deposits had been formed. It was, therefore, of import- 

 ance that this evidence should be carefully examined. In this emergency Mr. D. A. 

 Stewart, an engineer in the employ of the Canadian Pacific Railway, wlio was engaged in 

 surveying a projected line of road througli the Kootanie valley, kindly consented to secure 

 ample specimens of such trees and forward them to us for examination. This he did, and 

 in June, 1892, we received from him four tine specimens of wood, representing complete 

 sections of trunks two feet in length. Upon examination they were found to reiiresent 

 Pseudotsuga douglasii, Larix occidentalis, Pinus ponderosa, Pinus albicaulis. 



As the detailed structural variations of these woods will be considered fully in another 

 connection, it is only necessary, at the present time, to draw attention to such of their 

 anatomical peculiarities as may serve to afford a solution of the cpie-stion now under con- 

 sideration. 



On September 16th, after drying in the air of a dry room for three months, a section 

 about two inches thick was cut from the end of each log. Each of these sections was then 

 cut in two. The first half was allowed to dry in the air of a very dry room, at an average 

 temperature of 68^ F., in order to supplement observations upon the log under similar con- 

 ditions. The second part of each section was again divided. Part (a) was macerated in 

 water at a temperature of 65° to 70 -■ F. for a period of one month, when it was taken out 

 and allowed to desiccate in the air of a diy room at the same temperature for a period of 

 two and one-half months. Part (b) was macerated at the same temperature for a period of 

 four and one-half months, and afterwards desiccated in dry air, at the same temperature, for 

 one month. The object in making these tests was to determine (1) what changes would 

 take place in the log under the influence of rapid desiccation ; (2) what alterations would 

 occur in a rather thin piece of the material under similar conditions ; (3) the effect of 

 maceration extending over different periods of time, particularly when supplemented by 

 subsequent desiccation. It was thus hoped to secure some direct knowledge of the manner 

 in which these woods break up in drying, and the relations which the resin bears to such 

 changes; also that in the decay of the structure through maceration we might establish 

 conditions parallel to those which may have led to the liberation of the amber-like rods 

 forming the Kootanie coal. The results obtained may be best stated liy considering each 



species in detail. 



Pseudotsuga douglasii. 



Plate I., figs. 1, 2, 3. 

 After drying in the log for a period of three months, star and cup shakes became 

 manifest, and were developed in about e(inal prc>|iortions, but not in excess of what any 



