263 PROF. W. C. WILLIAMSON ON THE 



medullary rays. As the vascular wedges grew radially, 

 they also enlarged tangentially, and as they did so they 

 encroached laterally upon the peripheral prolongations of 

 the primary medullary rays {g,g), which, as we have seen, 

 ran parallel to, and on either side of, each wedge, through- 

 out the length of the internode. In this way the primitive 

 medullo-cortical origin of each such ray was lost sight of, 

 its peripheral extension becoming, both in its cambial 

 development and in its aspect, like an ordinary secondary 

 ray. It results that, when we examine the exterior of a 

 young decorticated Calamite, such as is represented in 

 the lower part of fig. 2, we find the longitudinally 

 extended vascular wedges, e', separated throughout their 

 entire length by tangential sections, g', of the parallel 

 primary medullary rays. In stems with a more developed 

 vascular growth, these alternations of tissue disappear, 

 as shown in fig. 3 g. 



The alternations of these vertical lines of cellular 

 and vascular tissue in contiguous internodes are brought 

 about in exactly the same way in living Equisetums 

 and in fossil Calamites. As each end of a vascular wedge 

 approaches the node above and below the internode 

 to which it belongs, it splits into two short diverging 

 branches (fig. 2, e) . Each one of these meets a similar 

 branch, derived from the contiguous vascular wedge of 

 the same internode, and the two halves thus derived 

 from two distinct wedges form a third one, which con- 

 tinues its upward or downward course through the next 

 internode, but in a line midway between those from which 

 it sprang, as in the living Equisetums; the internodal 

 canals, d, branch and recombine at the nodes of some 

 of the fossil Calamites in exactly the same way as the 

 vascular wedges do. 



Fig. 3 represents a restoration of a Calamite like fig. 2, 



