CARBONIFEROUS LYCOPODS AND SPHENOPHYLLS. 79 



been made, and that the smaller vessels at (b) and (c) belong to 

 the same series, but are only of weaker development, and Prof. 

 Williamson has pointed out that the smaller vessels are occasionally 

 developed in the central axis 1 and from the mode of increase in 

 the rootlet bundles, when smaller vessels are developed in the 

 axis, they may be added to the rootlet bundle. 



In the axis of a specimen in my collection from Pettycar, Fife, 

 some patches of small vessels occur, followed by the development 

 of others of normal size, 2 and in one of the rootlet bundles of 

 this example, seen in transverse section, immediately after leaving 

 the central axis, similar small vessels 

 occur within the rootlet bundle. In •"•• 

 other rootlet bundles in the same fossil, 

 admixtures of small and large vessels are 

 seen, having a different arrangement from 

 that just referred to. In another Stig- 

 marian axis from Oldham, these small 

 vessels form rings very similar in appear- 

 ance to the rings of annual growth in 

 Dicotyledons. These groups or bands of Fig. 12.— Rootlet bundle 

 smaller vessels seem therefore to be of Stigmaria (after 

 , , -iii ii • Williamson). For de- 



dependent on vital changes in the axis, . .. 



1 ° . scription see text. 



in which the rootlet strands participate, 



and do not represent in the rootlet bundle initial strands or 

 primary points of growth. This I believe to be the true explana- 

 tion of the formation of the small vessels shown in fig. 12 at (6) 

 and (c), to which so much importance has been given by Renault, 

 who regards such bundles as triarch, in distinction from the more 

 normal monarch form. 



The bundles of the " appendicular organs " appear, therefore, 

 to be of only of one form or type, that in which only one initial 

 strand exists. It may, therefore, be justly concluded that 

 whether the "appendicular organs" are looked upon as rootlets 

 or leaves, orily one of these is present, and that no admixture of 

 roots and leaves occurs in Stigmaria. The views advocated here 

 in regard to the structure of the rootlet bundles are very similar 

 to those expressed by Solms-Laubach in his Fossil Botany, who, 



1 Williamson, Monog. Stigmaria, p. 17, PI. IV., fig. 20. 



2 Slide No. 545a. 



