480 Palaeontologie. 



have been assigned to Dicotyledons and referred to such genera 

 as Ficophylliim, Myrica, Proteaephylliim, and others. Some of 

 these may well be small fronds of ferns with venation characters 

 like those of the Elk's Hörn fern (Platyceriiim), while others, 

 though presenting a close resemblance to Dlcotyledonoiis leaves, 

 afford insufficient data for acciirate generic Identification. In 

 dealing with fossil leaves of the dicotyledonous type, we must 

 not forget that the recent genus Gnetiim — a gymnosperm of 

 the section Gnetales — possesses leaves that may be said to 

 be indistinguishable in form and venation from those of certain 

 Dicotyledons. Before the close of the Potomac period these 

 few fragmentary relics of possible Dicotyledons are replaced by 

 a comparative abundance of specimens which must be accepted 

 as undoubted Angiosperms. 



Another indication of the sudden increase in the number 

 of dicotyledons is furnished by the Dakota flora of the United 

 States — in age somewhat more recent than the older Poto- 

 mac beds. In these plant-beds it is stated that Angiosperms 

 constitute two-thirds of the Vegetation. 



One of our most pressing needs is a thoroughly critical 

 revision of the late Cretaceoiis and earlier Tertiary floras, v/ith 

 the object both of determining the systematic position of the 

 older Angiosperms and of mapping out with greater accuracy 

 the geographical distribution of the floras of the world in post- 

 Wealden periods. This is a task which is sometimes said to 

 be impossible or hardly worth the attempt; the available evi- 

 dence is indeed meagre, and much of it has been treated with 

 more respect than it deserves, but it is at least a praiseworthy 

 aim, not to say a duty, to take stock of our material and to 

 compile lists of plants that may bear the scrutiny of experien- 

 ced systematists. 



In conclusion, the author urges the importance of taking 

 stock of our accumulated facts, and of so recording our obser- 

 vations that they may be safely laid under contribution as aids 

 to broad generalisations. Detailed descriptions and the enu- 

 meration of small collections are a necessity, but there is danger 

 of the Student neglecting the application of his results to problems 

 of far-reaching import. Condensed by E. A. N. Arber. 



Ausgegeben: 10. November 1903. 



Commissions-V^erlag: E. J. BriH in Leiden (Holland). 

 Druck von Gebrüder Gott helft, Kgl. Hofbuchdrucker in Cassel. 



