THE STRUCTURE AND AFFINITIES OF DIPLOLABIS ROMERI (SOLMS). 719 



Histology of the Stem. 



The stem-xylem is of great interest, since it consists of tracheides without any 

 admixture of parenchyma. Further, the thickenings on the walls of all the elements 

 are porose or reticulate, a type which recent work has shown to be the most primitive of 

 all. These reticulate thickenings can be distinctly seen in PI. I. figs. 5 and 6. The 

 inner zone of the wood, which is composed of short elements, appears to have offered less 

 resistance to crushing strains than the outer zone, and, as a result, its tracheides have 

 generally suffered considerable distortion during petrifaction (PL I. fig. 7). These inner 

 xylem elements vary considerably in size ; some are two or three times as broad as they 

 are long, while others are much longer than broad; the diameter varies from 100 /w. to 

 120 m. Where vertical rows of these short elements occur, there is a suggestion that 

 they may possibly have arisen by the septation of long tracheides. No trace of xylem - 

 parenchyma can be detected intermixed with these tracheidal elements, and hence we must 

 conclude that the plant possessed a solid xylem axis, i.e. the axis ivas a protostele. The im- 

 portance of this will be seen later, when the systematic position of the species is discussed. 



Towards the outer boundary of the inner xylem narrow scalariform elements may 

 occasionally occur. These are shown in longitudinal section in PI. I. fig. 8, where the 

 thickenings are clearly scalariform. Such groups constitute the protoxylem, though 

 probably they are petiolar protoxylems which are decurrent into the stem for a consider- 

 able distance, and not truly cauline. 



The elements of the outer xylem zone are greater in diameter than those of the 

 inner zone, and they consist of elongated, reticulate tracheides with a mean diameter of 

 160 m. Towards the periphery the elements tend to become slightly smaller. Some- 

 times groups of small elements can be detected, as in PI. I. fig. 15, prx. ; these are the 

 protoxylems of a petiole-trace which is being differentiated from the stem, and which 

 will become detached higher up the stem. As mentioned above, these protoxylem 

 groups die out as they pass downwards into the stem, but only after they have reached 

 the boundary between the outer and inner xylem zones. Owing to the pseudo-cellular 

 structure previously referred to, the tissues outside the stem-xylem are almost entirely 

 disintegrated. There is no trace of phloem or of inner cortex that would warrant more 

 than a passing notice, but small patches of the outer cortex are undoubtedly preserved. 

 This outer cortex has no special features. It consists of a thick-walled parenchyma, but 

 it contains no sclerotic layers. 



From the stem there depart, at intervals, petioles and roots. The distance between 

 the points of emission of any two petiole-traces is considerable. Tn no case, however, 

 is the outer xylem-ring of the stem ruptured by such emissions, so that the inner xylem 

 takes no part in the formation of the leaf-trace, i.e. these organs leave the stem in a 

 protostelic manner. The roots are quite irregular in their distribution, as can be observed 

 in PI. I. figs. 9, 10, and 11, where photographs of three successive sections are shown. 

 These sections were cut rather thicker than usual, and they are about ^ in. apart. In the 



TRANS. ROY. SOC. EDIN., VOL. XLVII. PART IV. (NO. 24). 106 



