148 Morphology BOOK i 



that the Myeloxylon petioles were borne upon the supposed 

 Cycadean stems which Cotta had named Medullosa in 1832. 

 They did not, however, publish in full the reasons for their 

 opinion till 1896. Medullosa, first described by Cotta, was 

 studied by Goeppert and Stenzel in 1881, and by them 

 referred to the Cycadaceae. The curious structure of the 

 stem was interpreted as polystelic by Zeiller in 1890, and 

 his view was endorsed and emphasized by Solms-Laubach 

 in 1897. He compared the structure with that of Cla- 

 doxylon, a plant from the Culm, in which several steles 

 are found each showing secondary thickening, and indicated 

 an affinity between this so-called fern and Medullosa. 



Another type of somewhat similar character was de- 

 scribed as Protopitys Buchiana by Solms-Laubach in 1893. 

 There accumulated thus a considerable body of evidence 

 pointing to the existence in Palaeozoic times of a stock 

 showing such close affinities to both Ferns and Cycads as 

 to warrant the opinion that it might be regarded as the 

 ancestor of both. This group was named Cycadofilices in 

 1897 by Potonie. 



Williamson and Scott in their memoir on Lyginodendron 

 and Heterangium suggested that they were derivatives from 

 an ancient and non-specialized Fern-stock from which they 

 showed a marked divergence in the Cycadean direction. Of 

 the two, they held Heterangium to be the most ancient and 

 to stand nearest to the Filicinean ancestry, Lyginodendron 

 having advanced much further on the Cycadean lines. 



Further investigations at and immediately after the close 

 of the century led to the view that many of the fronds 

 originally held to be Ferns really belonged to this group, 

 and that the predominance of the Ferns in Palaeozoic 

 times had been overestimated. 



It is beyond the scope of the present work to do more 

 than allude to the important discovery, made in the first 

 instance by Oliver and Scott in 1903, that these plants 



