114 SCIENCE PROGRESS. 



features of the main petiole and its branches, the histology 

 of the leaf lamina has been described in some detail. The 

 veins in the lamina seem to be of the collateral type, and 

 the mesophyll is composed towards the upper surface of 

 closely arranged cells elongated at right-angles to the epi- 

 dermis, and towards the lower of more loosely disposed 

 and less regular cells. Stomata have also been recognised 

 in transverse sections of these sphenopteroid pinnules. 



Finally the adventitious roots of this interesting genus 

 have been thoroughly investigated. In certain regions of 

 the stem adventitious roots are given off on all sides, both 

 from nodes and internodes ; their origin is clearly endo- 

 genous, and their histological features such as characterise 

 true root structures. The roots, like the petioles, were 

 originally described as a distinct genus, and named by 

 Williamson Kaloxylon Hookeri. The stele of the root 

 consists of two to eight centripetally developed groups of 

 xylem with conjunctive parenchyma, alternating with 

 groups of smaller cells, consisting partly of short 

 parenchyma and in part of narrow elongated elements. 

 Surrounding the xylem and phloem is a one-layered peri- 

 cycle and a layer of stouter cells with the characteristic 

 radial markings of endodermal cells occasionally visible. 

 External to the innermost layer of the cortex, we have a 

 broad zone of lax parenchyma with conspicuous and abun- 

 dant secretory sacs, and externally a peculiar epidermal 

 layer made of two or three rows of thin-walled cells. Many 

 of the roots examined show a zone of secondary xylem and 

 phloem, formed in a manner essentially identical with the 

 method of development of secondary vascular tissues from 

 a typical root cambium. 



Having thus briefly referred to some of the more strik- 

 ing anatomical characteristics of Lyginodendron, it remains 

 to describe in a few words the habit of the whole plant and 

 to discuss its affinities with recent forms. So far we are 

 without any satisfactory evidence as to the nature of the 

 reproductive organs. To quote Williamson and Scott's 

 words : ^ " We must, therefore, picture Lyginodendron to 

 ^ Williamson and Scott, p. 743. 



