4i o STUDIES IN FOSSIL BOTANY 



consists of the usual alternating longitudinal strands of 

 sclerenchyma and cellular tissue. The whole structure 

 is much like that of the petiole of Lyginodendron 

 oldhamium, but the spinous emergences characteristic 

 of the latter plant are absent. 



The petiole branched freely, and the smaller rami- 

 fications of the rachis are only about one-tenth as thick 

 as the main leaf-stalk. Thus the leaf, like that of 

 Lyginodendron, was a highly compound one ; remains 

 of the leaflets are often found, but it has not as yet 

 been possible to determine their minute structure. The 

 evidence available, from the petrified specimens, leads 

 to the conclusion that the foliage was of the same type, 

 as regards both anatomy and general Sphenopterid 

 form, as that of Lyginodendron oldhamium. 



3. The Roots. — The bases of adventitious roots are 

 often found in connection with the stem ; in a specimen 

 from the Coal-measures, belonging to H. Lomaxii, a 

 species closely allied to H. Grievii} three such roots 

 were found together, ranged in a vertical row on the 

 stem. They were clearly of endogenous origin, and are 

 placed opposite one of the primary bundles of the stem. 

 The steles of stem and root are connected by a system 

 of oblique vascular strands, apparently of secondary 

 origin. 



Miss M. Benson, F.L.S., has recently obtained 



evidence as to the structure of the roots of H. Grievii, 



1 In the memoir by Dr. Williamson and myself ( ' ' Further Observa- 

 tions," etc., Part iii. Lyginodendron and Heterangium) these two species 

 are united. It is, however, better to separate them, as there are some slight 

 distinctive characters, and their horizons are widely apart. The name 

 H. Lomaxii was suggested by Williamson in MS. for the Coal-measure 

 form, which Mr. Lomax first found at Dulesgate. 



