STHUCTUEE OF LEPIDOSTEOBUS. 367 



(Wild's Cone, C.N. 1776 a, 1776 b, 1776 c) have been compared with Williamson's type 

 slides and with other sections cut from the same specimen which are now preserved in 

 the Binney Collection at Cambridge. So far as the corresponding structures cjiu be 

 compared in the absence of a transverse section of the newer specimens, the latter appear 

 to be identical with those described by Williamson, and they will therefore be described 

 as Lepidostrobus oldhamius ((3). The general morphology is quite similar to that of 

 the L. oldhamius (a) already described. 



Structure of the Axis. — The axis is traversed by a central xylem cylinder, but unfor- 

 tunately the specimen from which the longitudinal sections have been cut was in a 

 somewhat flattened condition and the axial tissues disarranged. Notwithstanding tliis, 

 the xylem elements are beautifully preserved. PI. XXXVIII. fig. 24 shows some of the 

 larger xylem elements in longitudinal section. As can be seen, they exhibit the scalariform 

 type of thickening in a very perfect manner, and show also the very fine vertical striae 

 connecting the transverse bars so commonly seen in these old Lepidodendroid plants. 

 This latter character, long thought by Williamson to be a diagnostic one of Lepidoden- 

 dron mundmn. Will., but afterwards shown to be equally characteristic of well-preserved 

 tracheae of other forms (including Lepidodendron selaginoides, L. Marcourtii, and 

 L. Wunschianum), appears so constantly in the best-preserved sections of all these forms 

 that it can hardly be referred to one of the effects of mineralization, but must rather be 

 regarded as a structural peculiarity. These larger elements appear to be tracheides with 

 very long oblique septa, and they have a maximum diameter of about -O^-'OS millim. 



Some of the smaller, peripheral, xylem elements are shown on PI. XXXVIII. fig. 25. 

 They do not clearly show septa, and the thickening is often of a very loose nature 

 (spiral ?). Doubtless the development of the xylem was centripetal, and these elements 

 constitute the protoxylem. 



Unfortunately, in none of the slides of this form is there any trace of the tissues 

 betwx'cn the periphery of the xylem and the inner cortical zone. 



Bower * has described the structure of the axis of Lepidostrobm Brownii, Schimp., in 

 which the axial tissues are completely preserved as far outward as the middle cortex. 

 He has also recognized what he considers to be the eudodermis, a band corresponding in 

 position with the innermost layer of the cortex, distinguished by Hovelacque as the 

 " game " f . If this determination of the endodermis be correct, then the tissue between 

 it and the xylem cylinder in L. Brownii, and representing presumably both phloem and 

 pericycle, must have been exceedingly scanty. Opposite the points of the tracheal 

 cylinder this tissue is only about 2 cells in radial thickness. Prof. Bovver points out that 

 if a true phloem was present it can have existed only in comparatively small quantity, 

 and he proceeds to make a comparison with the Psilotaceae, in which there is very little 

 tissue referable to phloem. 



Returning to L. oldhamius (a), PL XXXVI. tigs. 1, 2, 3, & 5, it will be seen tiiat the 

 arrangement of parts is quite different. Here the space between the xylem and inner 



* Ann. Bot. vol. vii. 



t " Recherchcs sur le Lepidodendron selaginoides. Stern," Memoires de la Societe Linncenne de Normandie, Caen, 

 1892, p. 52. 



3k2 



