AS COMPARED WITH THAT OF THE PRESENT DAY, 421 



* 



Judging from the great preponderance of Filices in the coal measures, 

 the probabilities are somewhat in favour of Sigillaria? belonging to that 

 class ; and this opinion has accordingly the most advocates. Lindley 

 and Hutton* indeed adopted it at first, but rejected it afterwards on 

 the examination of Caulopteris ; f a most decided tree-fern. 



The latest observations, however, on this subject, are those of 

 M. Brongniart, who after an examination of the S. elegans'l slightly modi- 

 fies his opinion, placing the genus between Lycopodiacece and Cycadece^ 

 but nearest to the latter. 



Assuming the S. elegans to be a true Sigillaria, it appears to afford 

 slender grounds for the adoption of the above view, as regards its 

 uniting such diverse and distinct orders as Cycadece and Lycopodiacece. 

 It is true that it departs signally from the ordinary structure of the 

 latter order, which however itself exhibits various modifications of the 

 arrangements of scalariform tissue ; but it requires stronger evidence 

 than the more perfect structure and regular arrangement of the bundles 

 of vascular tissue to ally it to Cycadece, with which, in general appear- 

 ance, habit, fluting, markings, stigmaroid roots, absence of accompany- 

 ing foliage, and many other points, it has nothing in common. 



The Cycadea, on the other hand, and especially the Zamia integrifolia, 

 which M. Brongniart instances as peculiarly favourable to his views, 

 possess broad vessels perforated with very large circular apertures § 

 (not seen in Sigillaria?), and they want the double zone of bundles of 

 vascular tissue and diverging fascicles of Sigillaria?. 



That the Sigillaria? were allied to Lycopodiacea? is evident, their 

 tissues and scarring being very like those of Lepidodendron, in which, 

 however, there is but one series or zone of vascular tissue. In both, the 

 great mass of the woody axis is formed of large tubular vessels iden- 

 tical in structure, and in both, fascicles are given off from the central 

 mass to the scars on the circumference, which fascicles consist of slen- 

 derer tubes than the axis does. In Lepidodendron there are no 

 medullary rays, the vascular zone being continuous and surrounded by 



* Fossil Flora, Introduction to, vol. i. 



t Fossil Flora, tab. 42, sub Caulopteris primava, et tab. 54, sub Sigillaria pachyderma. 



J Archives du Museum, t. i. p. 426. 



§ A very similar tissue is exceedingly common in the coal-beds : all the mineral char- 

 coals (commonly called mother-coals in S. Wales) present it abundantly, and appear 

 wholly formed of such vessels held together by fragments of medullary rays. In spe- 

 cimens of charcoals from the Virginian coal-mines given me by Mr. Lyell, the same tissue 

 occurs ; but it was only during the passage of this sheet through the press that I had the 

 good fortune to meet with a fragment of the fossil wood to which these charcoals belong. 

 In examining the collection of Mr. Darker, an acute observer and excellent lapidary, he 

 showed me relics of a small rolled nodule of fossil wood from the coal-seams of the Yard 

 Mine, Durham, in which I at once recognized this tissue, of which I had been long in search. 

 The nodule was little larger than a hasel-nut, and originally covered with quartz crystals 

 Mr. Darker had himself conjectured its probable alliance to Cycadeee. 



2 G 2 



