472 BENNETTITALES [OH. 



long, and 2 3 mm. broad near the base, and some showed a small 

 oval lamina at the apex with a midrib and dichotomously branched 

 lateral veins. These bracts with the terminal lamina are regarded 

 by Thomas as almost certainly reduced leaves of the Taeniopteris 

 vittata type: they are represented in the restoration (fig. 573) 

 as occurring at the base of the flower-peduncle. Stomata were 

 found on the bracts exactly like those on the microsporophylls, 

 and this affords a strong argument in support of the view that 

 Williamsoniella belongs to the plant which bore Taeniopteris 

 fronds. There is a close parallelism between the bracts accom- 

 panying Williamsoniella flowers and those described by Nathorst 

 in connexion with Wielandiella 1 . It seems reasonable to regard 

 the bracts as serving the purpose of bud-scales. 



An important point is the constant association with the 

 flowers of Taeniopteris vittata fronds, a species described in Vol. n. 

 and by most palaeobotanists regarded as a Fern frond. Taenio- 

 pteris leaves occasionally show a clean-cut base 2 and in specimens 

 described by Thomas there are two small humps on the surface 

 of the proximal end of the petiole which represent vascular 

 bundles. Humps of similar size occur on the leaf-scars of stems 

 which are believed to have borne both Taeniopteris leaves and 

 Williamsoniella flowers. Fragments of the stems were found in 

 association with flowers : they are 7 mm. 2 cm. in diameter and 

 frequently forked and there is evidence that the flowers were 

 borne at the forks, the shoot forming a dichasial system. The 

 habit of the plant is represented in fig. 573 reproduced from 

 Mr Thomas's paper. 



Williamsoniella Lignieri (Nathorst). 



This species described by Nathorst as WiUiamsonia? Lignieri 

 from Whitby is regarded by Thomas as a Williamsoniella. The 

 stomata on the microsporophylls agree with those of Taeniopteris 

 and the spores obtained from Nathorst's flower-buds are like 

 those of Williamsoniella coronata. Evidence is adduced by 

 Thomas in favour of regarding W. Lignieri as bisexual and not 

 unisexual as Nathorst supposed. 



1 See page 463. 



2 Seward (00) B. xvi. fig. 1; Thomas, H. H. (15 2 ) PL xiv. fig. 26. 



