OCTOBEK 1<,», 18S:?.] 



SCIENCE. 



5:31 



pinnule lias had its special name, whilst the more 

 developeil structures forming the lower part of a 

 frond have supplied two or three more species. 

 Then the distinct forms of the fertile fronds may 

 have furnished additional ones, whilst a further cause 

 of confusion is seen in the wide difference existing 

 between a young, half-developed seedling and the 

 same plant at an advanced stage of its growth. Any 

 one who has watched the development of a young 

 Polypodium aureuni can appreciate this difference. 

 Tet, in the early stages of paleontological research, 

 observers could scarcely have acted otherwise than 

 as they did, in assigning names to these fragments, 

 if only for temporary working purposes. Our error 

 lies in misunderstanding the true value of such 

 names. At present the study of fossil ferns is afford- 

 ing some promise of a newer and healthier condition. 

 We are slowly learning a little about the fructifica- 

 tion of some species, and the internal organization 

 of others. Facts of these kinds, cautiously inter- 

 preted, are surer guides than mere external contours. 

 Unfortunately, such facts are, as yet, but few in 

 number; and, when we have them, we are too often 

 imable to identify our detached sporangia, stems, and 

 petioles, with the fronds of the plants to which they 

 primarily belonged. 



That all the carboniferous plants included in the 

 genera Pecopteris, Neuropteris, and Sphenopteris, are 

 ferns, appears to be most probable; but what the true 

 affinities of the objects included in these ill-defined 

 genera may be, is very doubtful. Here and there we ob- 

 tain glimpses of a more definite kind. That the Devo- 

 nian Palaeopterishibernicais anhymenophyllousform 

 appears to be almost certain; and, on corresponding 

 grounds, we may conclude that the carboniferous forms, 

 Sphenopteris trichomanoides, .S. Humboldtii,' and 

 Hymenophyllum Weissii,- belong to the same group. 

 The fructification of the two latter leaves little room 

 for doubling their position, whilst the foliage of some 

 other species of Sphenopteris is suggestive of similar 

 conclusions; but, until their fructification is diseov- 

 ere<l, this cannot be determined. An elegant fomi of 

 Sphenopteris (S. tenella, Brong. ; S. lanceolata of Gut- 

 bier), recently described by Mr. Kidson of Stirling, 

 abundantly justifies caution in dealing with these 

 .Sphenopterides. This plant possesses a true sphe- 

 nopteroid foli.ige, but its fructification is that of a 

 marattiaceous danaid. The sporangia are elongated 

 vertically, and have the round terminal aperture of 

 both the recent and fossil Danaiae, — agroup of plants 

 far removed from the hymenophyllaceous type of 

 sphenopterid already referred to. 



Whether or not this Sphenopteris was really marat- 

 tiaceous in other features than in its fructification, is 

 uncertain; but I think that we have indisputably got 

 stems and petioles of ilarattiaceae from the carbo- 

 niferous strata. My friend M. Henault, and I, without 

 being aware of the fact, simultaneously studied the 

 ^ledullosa elegans of Colta. This plant was long 

 regarded as the stem of a true monocotyledon, — a 

 decision the accuracy of which was doubted first by 

 Brongniart, and afterwards by Binney. M. Renault's 

 ' Scblinpcr, vol. I. p. 408. = Ibid., p. 413. 



memoir, and my part vii., appeared almost simulta- 

 neously. We then found that we had alike deter- 

 mined the supposed monocotyledon to be not only a 

 fern, but to belong to the peculiarly aberrant group 

 of the Marattiaceae. As yet we know nothing of its 

 foliage and fructification. 



M. Grand-Eury has figured ' a remarkable series 

 of ferns from the coal-measures of the basin of the 

 Loire, the sporangia of which exhibit marked resem- 

 blances to those of the Marattiaceae. This is espe- 

 cially the case with his specimens of Asterotheca and 

 Scolecopteris,- as also with his Pecopteris Marattiae- 

 theca, P. Angiotheca, and P. Danaeaetheca; but there 

 is some doubt as to the dehiscence of the sporangia 

 of these plants: hence their uiaratti.aceous character 

 is not absolutely established. 



That the coal-measures contain the remains of ar- 

 borescent ferns has long been known, especially from 

 their abundance at Autun. In Lancashire I have 

 only met with the stems or petioles of one species 

 preserving their internal organization.^ The Rev. H. 

 H. Higgins obtained stems that appear to have been 

 tree-ferns from Kavenhead, in Lancashire; .and it is 

 probable that most of the plants included in the gen- 

 era Psaronius, Caulopteris, and Protopteris, are also 

 tree-ferns. 



There yet remains another remarkable group of 

 ferns, the sporangia of which are known to us through 

 the researches of il. Renault. In these the fertile 

 pinnules are more or less completely transmuted into 

 small clusters of oblong sporangia. In one case, M. 

 Renault believes th.at he has identified these organs 

 with a stem or petiole of a type not uncommon at 

 Oldham and Halifax, belonging to Corda's genus 

 Zygopteris. Renault has combined this with some 

 others to constitute his group of Botryopterid^es, an 

 altogether extinct and generalized tj'pe. This review 

 shows, that whilst forms identifiable with the Hyme- 

 nophyllaceae and Marattiaceae existed in the carbonif- 

 erous ejioch, and we find here and there traces of 

 aflinities with some other more recent types, most 

 of the carboniferous ferns are generalized primeval 

 forms, which only become differentiated into later 

 ones in the slow progress of time. 



Equisetaceae and Asteropliylliteae {Brongniart), 

 Calamariae {Endlicher), EquUetineae (Schimper). — 

 Confusion culminates in the history of this variously 

 named group: hence the subject is a most difficult 

 one to treat in a concise way. The confusion began 

 when Brongniart separated the plants contained in 

 the group into two divisions, one of which (Equis^- 

 tacds) he identified with the living cquisetums, and 

 the other (Ast^rophyllitdes) he regarded as being 

 gymnospermous dicotyledons. To Schimper belongs 

 the merit, as I believe it to be, of steadily resisting 

 this division; nevertheless, paleobotanists are still 



' Flore cnrbonifcrc du Department de la Loire et du centre de 

 In Frnnce. 



' Loc. clt, tab. vlil., figs. 1-5. 



-* PsnroniuH Kenanltii, ^fcn1alr vll., p. 10; and Memoir xiL, pi. 

 iv., fig. 16. These and other similar references are to my scries 

 of memoirs on the organization of the fossil plants of the coat- 

 measures, ptitjlished in tlie Thiiosopliical transactions. 



