276 STUDIES IN FOSSIL BOTANY 



latter series. The distinction, though not so absolute 

 as was at first supposed, coincides well enough with 

 natural divisions, for the only two Eusporangiate families 

 — the Marattiaceae and the Ophioglosseae — are very- 

 distinct from any of the Leptosporangiate groups. 1 

 In the living flora the latter enormously outnumber the 

 former, as regards both genera and species. In dealing 

 with fossil specimens it is, of course, impossible to 

 study the development of the sporangium, and our 

 judgment as to affinities must necessarily be based 

 on the mature characters. The sporangia of the 

 Marattiaceae and Ophioglosseae are, on the whole, of 

 relatively large size, attached to the frond by a broad 

 base, with a wall more than one cell thick, and without 

 a definite annulus, though its place may be taken by 

 a small group of thickened cells. In the former of the 

 two families, we have also to consider the grouping of 

 the sporangia, which are often united together in each 

 sorus to form a synangium. Among the Palaeozoic 

 plants with a Fern-like fructification we find a sur- 

 prisingly large proportion which present these characters, 

 or some of them, and it was hence inferred, apparently 

 with good reason, that in those early days the 

 Eusporangiate section of Ferns, now so restricted, was 

 predominant. It is especially the Marattiaceae to 

 which this conclusion applied ; the Ophioglosseae have 

 not been determined, from the earlier strata, with any 

 degree of certainty. 



We will first consider a few representative fructifica- 

 tions of the Marattiaceous type, postponing the question 

 whether they actually belonged to Ferns of that family 



1 See, however, Professor Bower's arrangement, given below, p. 352. 



