330 A CENTURY OF PROGRESS IN THE NATURAL SCIENCES 



(gamostele). The stele concept was universally accepted. Strasburger (1891) 

 did not think that astelic axes differed in principle from monostelic. More im- 

 portant was that van Tieghem's opinion on the origin of polystely proved to 

 be incorrect. 



Later Jeffrey (1899a-1899b) propounded a modified stelar theory, distin- 

 guishing two basic types of structures, viz., the single solid concentric strand 

 without any pith, the protostele, and the concentric fibrovascular tube perfo- 

 rated by gaps and including a central pith, the siphonostele. There were two 

 types of siphonosteles, viz., the amphiphloic, with external as well as internal 

 phloem, and the ectophloic, with only external phloem. When the gaps of the 

 stelar tube correspond to leaf traces, as in the gymnosperms, the siphonostely 

 is phyllosiphonic, and when there are no foliar gaps it is cladosiphonic. Jeffrey 

 derived the siphonostele from the protostele, and thought that the pith was, 

 phylogenetically, formed by invasion of cortical tissue into the stele. 



The study of gymnosperm vegetative anatomy progressed actively also in 

 other directions. Investigations concerned the pith of the stem, the phloem, 

 the wood rays, the tracheid walls with their bordered pits, the trabeculae in sec- 

 ondary w^ood, the behavior of the leaf traces, and the resin ducts in roots, stems 

 and leaves. Penhallow (1896) and Kraus (1886), studied the secondary conifer 

 wood for the purpose of defining genera and species and of establishing the re- 

 lations to fossil woods. 



The cj^cads attracted much attention from the phylogenetic point of view 

 (cf. Strasburger, 1891). Solms-Laubach (1890a) demonstrated the position of 

 the cones and the sympodial nature of the stem. The complicated course of the 

 leaf-trace bundles was found to change in the peduncle of the cone into a sim- 

 plified organization conforming to conditions in the bennettites. Scott (1897) 

 concluded from a comparison with certain Cycadofilices that the mesarch cyca- 

 dean type of vascular bundles represents a vestige of a primitive organization 

 that was once common to leaf and stem. AVorsdell (1898a, 1898b, 1901) found 

 two types of stem structure in cycads, viz., one in which there is a single stele 

 and another in which there is more than one cylinder. He derived the vascular 

 tissues of the cycad stems from those of the paleozoic Medullosaceae, while ac- 

 cording to Scott (1899) the primary ground plan of the stem structure of a 

 polystelic Medullosa is fundamentally different from that of monostelic cycads. 



In his comprehensive account of gymnosperm anatomy, Strasburger also 

 dealt with the leaves particularly as regards the vascular bundles and the origin 

 of leaf traces. Other botanists studied the stomatal apparatus in conifers, their 

 trichomes, and the general anatomy of juvenile and adult leaves. A good deal 

 of attention was given to the transfusion tissue. According to one view, this 

 tissue forms part of the conducting tissues to the vascular bundle. Others 

 thought that it belonged to the parenchymatous tissue of the leaf, while ac- 

 cording to van Tieghem (1891b) it is a part of the pericycle of the bundle. 

 Worsdell (1897) tried to prove that the transfusion tissue is, phylogenetically, 

 a direct derivative of the centripetal xylem, which normally occurred in primi- 

 tive extinct gymnosperms, and is still found fully developed in the adult cycad 

 leaves and in the cotyledons of Ginkgo. 



Bower (1885) treated the leaf as a potential branch system, and used the 

 term phyllopodium for its axis. This axis may develop in various ways with- 



