XVI FOSSIL ARCIIEGONIATES 583 



been described under the name Dancvopsis. The other hving 

 genera are not known as fossils, although certain fossil genera 

 seem to be related to them. Thus Asterothcca and Scolecop- 

 teris have been placed in the Angiopteridece, Ptyclwcarpus in 

 the Kaulfussieae. 



Besides the forms which are unquestionably to be referred 

 to the Marattiales, there are a good many types of Paheozoic 

 Ferns w^hich show apparent resemblances to the true Maratti- 

 acese in the structure of the sporangium, but which have the 

 individual sporangium entirely distinct, instead of more or less 

 united with its neighbours as in the typical synangium of most 

 Marattiacese. This free sporangium is structurally like that 

 of such forms as Angiopteris, in which the sporangia are nearly 

 separate, and not improbably represents a Marattiaceous type 

 in which this tendency is carried further than in any of the liv- 

 ing genera. In still other forms of supposed Marattiaceous 

 affinity, e. g., Urnatoptcris (Potonie (3), Fig. 68), the spo- 

 rangia are borne upon sporophylls, which are completely cov- 

 ered wath them, as in the fertile fronds of Osmuuda or Bo- 

 trychiiim. In all of the living Marattiaceae except Dancra, the 

 synangia are borne upon unmodified leaves. In Dancra, how- 

 ever, the segments of the sporophyll are much contracted, and 

 the large synangia almost completely cover the lower surface 

 of the pinnae, and in this respect it suggests an approach to 

 those Palaeozoic types in w^hich the lamina of the fertile leaves 

 is entirely wanting. 



It is not unlikely that some of the Carboniferous Maratti- 

 ales were more or less synthetic types, connecting the typical 

 Marattiaceae with the later developed Leptosporangiates. The 

 genus Senftenhergia (Potonie (3), Fig. 86), for example, 

 seems to resemble to a certain extent both Marattiaceae and 

 Schizaeaceae, while Renault ia {Sturiclla) has been compared 

 with both the Osmundaceae and Schizaeaceae. 



The Marattiaceae seem to have maintained their ascendency 

 well into the Mesozoic. Raciborski (see Scott (i), p. 303) 

 found in upper Triassic beds about 70 per cent, of the Ferns to 

 be Marattiaceae ; but in lower Jurassic beds there was a remark- 

 able falling off in their number, only about 4 per cent, being 

 referable to the Marattiaceae. At the present time their num- 

 ber is less than one per cent, of the living species of Ferns. 



While there is some evidence of the presence of leptospo- 



