90 THE EVOLUTION OF PLANTS 



was not constant. In the genus Wielandiella of 

 Nathorst, derived from the Upper Trias of 

 Sweden, the habit is quite unlike anything which 

 we are accustomed to associate with Cycads 

 (fig. 11). The long, slender stem was repeatedly 

 forked, with rosettes of narrow, pinnate leaves 

 only about three inches long, at the angles. The 

 flowers were seated singly in the forks of the 

 stem between the two branches. As Prof. 

 Nathorst has lately shown, they had essentially 

 the same structure as the flowers of Bennettites. 

 Here, then, we have a member of the group 

 which departs widely from the usual habit, 

 and this instance, to which others are already 

 being added, is enough to show that external 

 characters do not constitute any impassable 

 barrier between the Cycadophytes and the 

 higher Flowering Plants. 



The complex, frond -like form of the stamens 

 and their immense output of pollen-sacs are 

 striking features, indicating an affinity to the 

 Ferns, and very unlike anything in Angiosperms. 

 It is true that branched stamens occur in the 

 latter, as in the Mallows and the Castor-oil plant, 

 but this appears to be only a secondary and not a 

 primitive character. In the great majority of 

 Flowering Plants the stamen has a very constant 

 structure, bearing four elongated pollen-sacs, 

 arranged two and two on either side. It has been 



