450 
Scott ,— The Structure of Mesoxylon multirame. 
cuts deeply into the phloem, enclosing disorganized tissue which had 
probably died off while the tree was living. There is, however, nothing to 
show that the wood was affected—it appears quite normal throughout. 
There are one or two black cracks, filled perhaps with an exudation of resin, 
but we have no reason to suppose that these lesions occurred during life. 
I have observed what appears to be wood-parenchyma in another 
specimen (section 2578), but the evidence is not so convincing as in the 
cases just described. The presence of parenchyma in Cordaitean wood is 
exceptional. I have recorded its occurrence in a remote member of the 
Cordaitales, Pitys antiqua (Scott, 1902 , p. 352). 
The cambium is preserved in places, but the cells are usually somewhat 
collapsed. The same applies, to a certain 
extent, to the inner layers of the phloem , but 
the bulk of this tissue is extremely well 
preserved. 
In transverse section the phloem (about 
2 mm. in thickness) is seen to consist of 
roughly tangential bands of larger and smaller 
elements (PI. XI, Figs. 7 and 8). The 
former are 50-70 p. in diameter, and are usually 
almost filled by a brown mass, either solid or 
with a small central lumen, and somewhat 
retracted from the cell-wall. In transverse 
section one might take this mass for thickened 
cell-wall in an altered condition, but in longi¬ 
tudinal section (PL XII, Figs. 10 and 11) it is 
evident that the brown substance is of the 
nature of cell-contents ; it is irregularly broken 
up and resembles the supposed resin-masses often met with in the tracheides 
or wood-parenchyma. Transverse walls are seldom seen ; they may some¬ 
times be hidden by the dark contents, or may have broken down as 
Bertrand and Renault ( 1886 , p. 292) found in the case of Poroxylon. 
The smaller elements usually appear clear (PI. XI, Fig. 8, and PI. XII, 
Fig. 11); sometimes they have quite light brown contents, and occasionally 
they may be infiltrated with the darker substance found in the long tubes. 
They are often flattened tangentially like cambial cells; the tangential diameter 
ranges from 50 to 24 /x, and the radial from 30 to 12 /x. They have often, as 
is obvious from the measurements, undergone additional radial divisions. 
In longitudinal section it is found that these smaller phloem elements 
are of two kinds ; many of them have very distinct and frequent transverse 
walls; others are long and tapering and not septate (PL XII, Fig. 11). 
The former must be regarded as phloem-parenchyma, the latter as sieve- 
tubes. It is very rare to find the sieve-plates preserved ; but in the case 
Text-fig. 1. From a radial 
section of the phloem, showing 
sieve-tubes and parenchyma. s./>., 
three sieve-plates on radial wall of 
a sieve-tube, x about 105. R. S. 
S. 2780. 
