480 STUDIES IN FOSSIL BOTANY 



The free Kalymma petioles contain many bundles and 

 are known to have branched; there is little doubt that 

 they represent the stalks of compound Fern-like leaves. 



Their structure is so different from that of the 

 petiole in Lyginodendreae that it is best to keep 

 the families distinct, especially as we know nothing of 

 the reproduction of Calamopityeae. I have recently 

 identified specimens of Calamopitys in material from the 

 Upper Devonian of Kentucky, sent me by Professor , 

 E. C. Jeffrey, and Kalymma has long been known from 

 similar horizons. The great antiquity of the genus is 

 thus well established. 



Two other species may be shortly described. In 

 C. fascicularis} from the Lower Carboniferous of Dum- 

 bartonshire and the North of England, only the pith 

 and wood are known, but their structure is well shown 

 (see Fig. 177). The small pith (2 to 3 mm. in diameter) 

 is surrounded by a circle of distinct primary xylem- 

 strands, eight or nine in number. These strands are 

 very unequal in size, attaining their maximum diameter 

 where they leave the pith to pass outwards through the 

 secondary wood (Fig. 177, B and A). Traced down- 

 wards from this point they diminish rapidly in size, 

 and each strand unites with a small " reparatory " strand 

 on the side corresponding to the downward (kathodic) 

 direction of the leaf-spiral (see Fig. 177 and the 

 description). The comparison of successive sections 

 shows that the phyllotaxis was § and that the course 

 of the xylem-strands was essentially the same as in 

 Lyginodendron. The strands can be followed out 



1 D. II. Scott, "On the Primary Structure of certain Palaeozoic Stems 

 with the Dadoxylon Type of Wood," Trans. Roy. Soc. Edinburgh, 

 vol. xl. Part ii. 1902. 



