on the Genus Drimys. 361 
group cannot with any degree of probability be regarded as a primitive 
feature. 
The material used in the present investigation was collected at the 
request of the senior author by travelling Fellows of Harvard University in 
New Zealand and in Java, and care was taken to secure seedlings as well as 
parts of mature plants. Wounded conditions were particularly sought for 
on account of the proved value of wound reactions in the case of both 
Conifers and Dicotyledons. Abundant material of Drimys colorata and 
a moderate amount of D. axillaris were available from this source. We 
are indebted to the Director of the Royal Gardens, Kew, England, for 
small twigs of D. Winteri and D. aromatica. Through his kindness we 
have likewise been enabled to study the organization of the interesting 
genera Trochodendron and Tetracentron> which are also devoid of typical 
vessels. 
The wood of Drimys , apart from the absence of vessels, strongly 
resembles in general appearance that of one of our northern oaks, for it is 
characterized by conspicuously large rays in striking contrast to others 
a single row of cells in breadth. Fig. 1, PI. VII, shows the general 
organization of the wood of D. colorata as seen in transverse section. To 
the right and left are indications of the broad rays of the oak type. The 
mass of wood in the centre is radially traversed by narrow and uniseriate 
rays. The annual rings are somewhat indistinct, a result of the mild 
climatic conditions under which the plant has flourished. They are most 
clearly marked by a terminal zone of parenchyma. Fig. 2, PI. VII, repro- 
duces the tangential aspect of the wood in D. colorata under a moderate 
degree of magnification. Laterally are the large rays as in the preceding 
figure, and in the centre lies the mass of wood penetrated by the linear or 
uniseriate rays. Wood parenchyma is not seen in the field of view. 
Fig. 3, PI. VII, shows a part of the wood of D. colorata , somewhat highly 
magnified. The narrow rays alone are evident, and the fact that they are 
uniseriate is now quite apparent. The mass of the wood is composed of 
tracheides which are thick-walled and angular in shape. Wood parenchyma 
appears tangentially as thin-walled elements which look black in the 
photograph. 
The characters of the longitudinal elements or tracheides are of 
greatest interest in the present connexion. Fig. 4, PI. VII, represents 
a rather high magnification of a radial view of the wood in D. colorata. To 
the right may be seen parenchymatous cells, the remainder of the field 
showing only tracheary elements. The pits in the latter are in a single 
row and do not usually become flattened by mutual contact, thus scarcely 
justifying the comparison with the wood of the araucarian Conifers which 
has often been made. Fig. 5, PI. VII, supplies a view of the wood of 
D. axillaris under a similar magnification. Parenchymatous elements are 
C c 2 
