Specimens of Fossil Wood. 247 



devoid of structure, as if a more delicate tissue like combi- 

 um had been removed. Other and complete bundles, on 

 the other hand, show no such open tracts, nor do they, as 

 appears in figure 2, show more than one kind of tissue, so 

 that we are left somewhat in doubt as to their precise com- 

 position. The cross section shows an apparent absence of 

 vessels, and with one exception the same may be said of the 

 longitudinal sections. In one case a single cell shows five 

 transverse bars, possibly the remains of a spiral, annular 

 or scalariform structure. In other respects the cell is the 

 same as the other members of the bundle. Each bundle is 

 surrounded by a layer of sclerenchymatous tissue composed 

 of rather thick walled cells of very unusual dimensions 

 and form. 



The specimen is apparently identical with Unger's Kal- 

 ymma grandis, * which he considers to be related to the 

 Equisetacese — a view correctly based upon the general 

 structure, though the presence of an outer zone of vascular 

 structure must be regarded as exceptional, and, so far as I 

 am aware, it has no parallel in existing types. Uncertainty 

 as to the exact structural characteristics of the vascular 

 bundles renders a more decided opinion as to the affinities 

 of this plant undesirable at the present time. 



Additional interest is given to this specimen from the 

 fact that it is the first of the kind from the formation and 

 locality from which it was obtained, and that as already 

 stated by Sir W. Dawson, it aids in connecting the middle 

 Devonian flora of America with that of Europe. 



Explanation of Plate I. 

 Kalymma grandis. Ung. 

 Fie;. 1. — Transverse section of a double vascular bundle from the 

 outer portion of the stem. W. — woodcells ; Pr. — parenchy- 

 ma. The large cells forming a line nearly across tbe 

 figure, show tbe parencbyma separating the two bundles. 

 x 40. 



1 Richer andUnger, Devonian of Tburingia, p. 71. 



