36 D. P. PENHALLOW ON 



specimen 789 the total volumes of zones 2 and 3 are 2-85 cm. and 4-77 cm. respectively. In 

 specimen 428, from the same locality, these zones are 16*38 cm. and 0'70 cm. respectively. 

 It may very correctly he urged that these zones in the two trees were not developed syn- 

 chronously. Admitting this to he the case, we may, then, be permitted to assume that in 

 789 zone 2 was develojaed synchronously with zone 1 of 428. We would then have the 

 following relations : 



No. 789 17-07 ... 2-85 ... 4-77 ... 1-36 



ÏTo. 428 2-72 ... 16-38 ... 0-70 



We now find a practical agreement in volume in these two zones, but this agreement 

 fails completely for the subsequent zones, and the argument itself therefore fails. It is thus 

 clear that we must look to some other cause for an explanation of these variations. 



It has already appeared that (a) the growth-rings are broadest in zone 1, becoming con- 

 tinually narrower and more uniform with increasing diameter, and {b) that where the growth- 

 rings are broadest there is the widest variation in thickness. This may bo taken as the 

 expression of two important facts in the growth of the plant. 



It is a well-recognized fact that the growth-rings of trees mark, and arc primarily 

 dependent upon, alternating periods of rest and activity, which, being determined by sea. 

 sonal changes, are, in this latitude, chiefly annual, though, as is well known, they may be 

 semi-annual, and thus give rise to more than one ring.' In this sense, therefore, the origin 

 of such rings is physiological. It is, therefore, quite within bounds to assume that in the 

 gradual decline of a tree the energy of growth does not diminish at a uniform rate, but is 

 subject to a certain periodicity dependent in the first instance upon phj^siological conditions. 

 In this, therefore, we would find an explanation of the occurrence of distinct zones of growth. 



We must not lose sight, however, of the important influence of mechanical pressure in 

 inducing structural alterations. Each growth-ring is found to consist of two parts — the inner 

 or spring wood, which is distinguished by the tracheids being very large and thin-walled, 

 and often elongated radially ; and the outer or si;mmer wood, at once recognized by the 

 smaller and usually very thick-walled tracheids, Avhich are often strongly compressed 

 radially. It is the direct apposition of these widely difterent structures which serves to 

 define the growth of separate seasons. But these structural differences have their origin, as 

 Sachs pointed out several years since, and as de Vries has proved l)y direct experiment, not 

 in physiological conditions but in conditions of mechanical pressure established between the 

 investing cortex and the growing tissues of the vascular cylinder. It is in this, therefore, 

 that we must seek an explanation of those structural differences to be met with in the growth- 

 rings, which include variations in the density and general character of the summer wood, as 

 well as variations in the relative proportions of spring and summer wood. 



The growth-rings present variations in thickness which call for somewhat more extended 

 notice. If we compare the sections exhibited in plate I., figs. 1-3, and plate II., figs. 1-6, 

 these variations will be made clear. In fig. 1, plate I., the growth-rings are broad, and well 

 represent the character of the rings in zone 3. This specimen was from the Kootanie valley 

 of British Columlna. 



In plate II., fig. 5 is a section fn)m zone 3 of No. 35. By comparing this with fig. 6, 

 which was taken from zone 1 of the same tree, a fair conception may be gained of the varia- 



' " Canadian Record of Science," I., 162, 1885. 



