174 



GYMNOSPERMS 



history. A prominent paleobotanist once proved that the monocots 

 must be older than the dicots, because an ancestor must be older 

 than its progeny. Now that it is generally believed that dichot- 

 omous venation is the most ancient type, the monocots can take a 

 more natural place as progeny of the dicots. Such recent mistakes 

 should temper one's enthusiasm when he constructs phylogenies, 

 especially when dealing with extinct plants where only scattered 

 fragments are available. 



g\ _ — -p — 5^ 



mv 



Fig. 184. — Poroxylon edwardsii (Permo-Carboniferous) : transverse section of stem. 

 Both bundles belong to the same leaf trace; p, pith; gc, mucilage canals; p.v, protoxy- 

 lem; x, centripetal mcta.xylem; .v^ secondary xylem; mr, medullary rays; X66. — .\fter 

 Bertrand and Renaud, in Scott's, Studies in Fossil Botany.^^ 



The name '^Cordaites'^ was originally applied to the leaves, and 

 the Cordailes section is still subdivided on leaf characters. Cordaites, 

 the Eu-Cordailcs of Grand d'Eury, includes those in which the leaves 

 are broad, with rounded apex and with strong veins mixed with 

 weaker ones. They were most abundant in the middle part of the 

 Upper Carboniferous. The Poa-Cordaites forms had linear, grasslike 

 leaves, up to 40 cm. in length, with veins about equal. The Dory- 

 Cordaitcs section contains those with broad lanceolate leaves with 

 acute apex, and rather fine equal veins. They are most abundant in 

 the Upper Carboniferous and Lower Permian. 



