GYMNOSPERMS— CORDAITALES— CORDAIANTHUS. 263 



merely the irregular transverse cracks or fissures in the carbon. The latter 

 may be only the result of shrinkage of an axis composed largely of cellular 

 tissue or they may bear some . elation to the chambering of the i)ith. The 

 specimens from Cannelton have the gemmules usually more crowded, the 

 scales being generally a little shorter. 



Very interesting, as furnishing the data for the correlation of this 

 species, is the tjq^e described on page 534 of the Coal Flora as the stem of 

 Cordaites communis Lx. In the specimen. No. 8946 of the Lacoe collection, 

 the impression of a segment of stem or branch 14 cm. long and 2.3 cm. in 

 diameter shows about thirty leaf scars. From the axils of every one of 

 these, so far as can be learned without injury to the specimen, in the upper 

 half of the segment, there radiate rather slender racemes of Cordaiantlms. 

 The pedicels and gemmules on the upper part of the slab are rather slender, 

 having about the proportions of the C. dicliotomus figured in the Coal Flora, ^ 

 but those nearer the base of the segment are unmistakable specimens of 

 Cordaiantlms ovahis, and indicate the specific identity of the latter with the 

 type stem and intermingled leaf fragments of Cordaites communis. 



The full length of the lineate bracts is rarely shown. In one rather 

 small specimen they are, however, seen as slender, slightly decurrent needles, 

 39 mm. in length, or over five times the length of the gemmules. 



The fragment of a very small raceme, shown in PL LXXII, Fig. 1, is 

 suggestive of the Cordaianthus gracilis of Grand 'Eury,^ or to^some extent 

 the C. Volkmanni (Ett.) Zeill.,^ though the resemblance to Ettingshausen's 

 Calamites Volkmanni* is more remote. 



Cordaiantlms ovatus appears to differ from C. ebracteatus Lx., to which it 

 seems closely related, by the absence of the bracts and the usually shorter 

 scales in the latter. 



The difference between it and C. dicliotomus consists perhaps in the 

 rather larger and longer scales and the possible dichotomy of the axis in 

 the type described under the latter name, though it appears somewhat 

 questionable whether the distinction between these two plants, found at the 

 same locality, is of even varietal rank. The characters in common will be 

 mentioned in the remarks on the latter s^secies. 



I PI. Ixxvi, tig. 6. 



= F1. carb. Loire, p. 230, pi. xxvi, fig. 7. 



"FI. foss. bassiu houill. Valenciennes, p. 637, pi. sciv, figs. 6, 6a. 



^ Steinkohlenli. Strndoiiitz, pi. v, figs. 1-3. 



