THE 



GEOLOGICAL MAGAZINE. 



NEW SERIES. DECADE IV. VOL. X.'* 



No. IX. — SEPTEMBER, 1903. 



(d:rxg-xist j^Xj j^^e-ticlies. 



I. — On Homceomorphy among Fossil Plants. 



By E. A. Newell Arber, M.A., F.L.S., F.G.S., Trinity College, Cambridge; 

 University Demonstrator in Palaeobotany. 



STUDENTS of palseobotany, when concerned with casts and 

 impressions of fossil plants as distinguished from petrifactions, 

 have often to face difficulties in the course of their examination 

 of such remains, some of which are peculiar to this branch of 

 palaeontology. Even when a large number of specimens of any 

 particular type of foliage, or other organs, are available for 

 comparison, it is often difficult to decide how far one set of 

 casts and impressions can be regarded as distinct from another. 

 Authorities differ in their ideas of the aggregate of differences 

 necessary to constitute genera and species. This difficulty, 

 although common to the systematist in the study of recent plants, 

 is greatly intensified when dealing with fossils, on account of the 

 fragmentary nature of the evidence. 



To take an illustration. Among Upper Carboniferous fern -like 

 plants none are perhaps more frequent in British rocks than 

 Neuropteris and Alethopteris, types of foliage in all probability 

 belonging to members, not of the true ferns, but of the Cycadofilices. 

 In these genera the habit of the frond is extremely well marked 

 and characteristic, and they are regarded as good form - genera. 

 But there also occur, frequently in the Continental Coal-measures, 

 though more rarely in the Carboniferous rocks of this country, 

 two other genera, Linopteris (also known as Dictyopteris) and 

 Lonchopteris, which correspond exactly with Neuropteris and 

 Alethopteris, respectively, except in one remarkable detail. In 

 Linopteris and Lonchopteris the secondary nerves of the pinnules 

 anastomose among themselves, forming a regular network, whereas 

 in the genera first mentioned no such reticulations occur. This 

 parallelism extends also to species within the genera. As Zeiller'^ 

 has pointed out, Linopteris Brongniarti, Gutbier, corresponds closely 



1 Zeiller: " Elements de Paleobotanique," 1900, p. 108. 



DECADE lY. — VOL. X. — NO. IX. 25 



