FERNS— FRUCTIFICATIONS 285 



It must be confessed that in such cases the reference 

 to Marattiaceae has always rested on very slender 

 evidence. Where no synangium is formed, and the 

 sporangia are not even grouped in definite sori, we 

 really have very little to go by, in default of anatomical 

 evidence, for the mere absence or rudimentary develop- 

 ment of the ring is of course no proof of Marattiaceous 

 affinities. The position of genera such as Dactylotheca 

 and Renaultia is entirely doubtful, and it is quite 

 possible that their fructifications may really be of the 

 nature of microsporangia, the plants being in that case 

 Pteridosperms and not Ferns. 



Among the Palaeozoic plants with definite synangia, 

 the circular or radiate type of synangium, now limited 

 to the genus Kaulfussia, seems to have prevailed. In 

 the genus Danaeites, however, which includes some 

 Pecopteroid forms from the Coal-measures, the synangia 

 were linear in shape, and appear to have closely 

 resembled those of the recent genus Danaea. In this 

 case the sporangia were arranged in a double row 

 along each secondary nerve of the fertile pinnule ; they 

 were sunk in the tissue of the lamina, and fused 

 laterally with one another, the constituent sporangia 

 each dehiscing by an apical pore. The agreement 

 with the recent genus was thus remarkably exact. 



The forms hitherto described bore their sporangia 

 on the ordinary vegetative fronds, as is the case with 

 a majority of the Ferns at the present day. A con- 

 siderable number, however, of the Palaeozoic Fern-like 

 plants were dimorphic, the sporangia appearing on fertile 

 leaves, or parts of the leaf, quite different in form 

 from the sterile foliage. 



