FERNS— SUMMARY 349 



a secondary modification, though one which, in certain 

 groups, took place at a very early period. 1 



From an anatomical point of view, two main types 

 of stem are sharply distinguished among the Palaeozoic 

 Ferns. On the one hand, we have the large, highly 

 complex, polystelic stems represented by Psaronius, 

 and belonging, no doubt, to the Marattiaceous Tree- 

 ferns. On the other, we have the herbaceous monostelic 

 forms, such as the Botryopterideae. 



The monostelic group is generally further character- 

 ised by having only a single vascular bundle in the 

 petiole, and it is remarkable that this peculiarity also 

 extended to many of the polystelic forms (species of 

 Psaronius), which, in the simplicity of their petiolar 

 structure, differed widely from Marattiaceae, while 

 agreeing with them in the anatomy of the stem. On 

 the evidence now available, it appears that monostelic 

 structure of the stem was more common among 

 Palaeozoic than among recent Ferns, while monodesmic 

 structure of the petiole was the rule, 2 though not, of 

 course, without exceptions. Summing up the results, 

 we may say that the Palaeozoic, or at least the 

 Carboniferous Ferns, group themselves anatomically 

 into two main cycles — the Palaeo-Marattiaceae on the 

 one hand, and on the other that great synthetic 

 group of Ferns (Primofilices) with a simple anatomical 



1 Stur came to a similar conclusion, " Culm- und Carbonfarne," p. 161. 

 See, however, F. O. Bower, " Studies in the Morphology of Spore-producing 

 Members," iii. Marattiaceae, Phil. Trans, Roy. Soc. 1897, B, p. 35, for 

 an able statement of the opposite view ; also his Origin of a Land-Flora, 

 p. 523, 1908. 



2 See L. A. Boodle, " On the Comp. Anatomy of Hymenophyllaceae, 

 Schizaeaceae, and Gleicheniaceae," Ann. of Bot. vol. xiv. 1900, vol. xv. 

 1901, vol. xvii. 1903. 



