29 



•of more recent times. In a certain number of cases the seeds have 

 been found in connection with the fronds. Thus in a thoroughly 

 fern-like species, Pecopteris Pluckeneti, in which no one had suspected 

 any connection with seed-plants, Graud Eury found the little flat 

 seeds, attached by hundreds to the fronds. In the fine " fern," 

 Neuropteris heterophylla, somewhat resembling a huge Royal Fern 

 { Osmu?ida ) in its foliage, Mr. Kidston found seeds of the size of a 

 hazel-nut. The seeds were attached to stalks which, happily, still 

 bore the characteristic leaflets, enabling the plant to be identified. 

 A good many years earlier the same observer had discovered in the 

 same plant another form of fructification, which we must now 

 regard as representing the pollen-bearing organs. 



Other cases of seeds in actual connection with " fern-fronds " have 

 been found by Mr. David White in American species of Aneimites, 

 resembling the fronds of the Maidenhair. In a large number of 

 instances, extending to most of the genera of fern-like Coal plants, 

 the association between particular fronds and particular seeds is 

 constant. 



In order to gain some idea of the organisation of one of these 

 " Seed- bearing Ferns " of the Palaeozoic period, we will take a 

 special case and select one of the commonest fern-like plants of the 

 Lower and Middle Coal-measures. The frond ( Alethopteris ) has 

 long been known — the name " true fern " or " true bracken" shows 

 what complete confidence the discoverer had in its fern-nature. It is 

 a large and handsome, repeatedly divided frond, such as might have 

 belonged to one of the Tree-ferns. The anatomical structure of the 

 frond is known, and its petiole is found in connection with a 

 remarkable type of stem, Medullosa, also well known in the petrified 

 condition with its structure perfectly preserved. The stem in this 

 case had three distinct woody cylinders, each of which grew 

 indefinitely in thickness by its own cambium, a curious, and, as it 

 seems to us, inconvenient arrangement, for the .different masses of 

 wood, as growth went on, must, one would suppose, have got in 

 each other's way. The leaf-stalks, with their numerous vascular 

 strands and peripheral system of fibrous strengthening-bands are 

 much like those of a recent Cycad, while the leaflets had the 

 typical structure of assimilating organs of the present day. 



The associated seeds ( Trigonocarftus ) have in most respects the 

 structure of the seeds of Cycads ; in particular, the pollen-chamber 

 is a striking feature, and the seed-coat consisted of a fleshy outer 

 layer enclosing a stone, within which was the embryo-sac. \i we 

 may judge from analogy with the related Neuropteris, already 

 described, the seeds were borne on the fronds, probably but little 

 modified as compared with the ordinary foliage. 



Another example of a fern -like seed-plant of Palaeozoic age is 

 Lyginodendron oldhamium, of which all the parts are now well 

 known ; it has taken many years to completely reconstruct the 

 plant from its fragmentary, though beautifully preserved remains. 

 The foliage is very finely cut and of great elegance of form, 

 recalling some of the living Davallias. The presence of spines on 

 all parts of the frond recalls especially the West Indian Bramble 



