98 REPORT— 1873. 



the iirst 200 or 300 fatlioms. But wliile the temperature of the Suhi Sea, which 

 is 50° at 500 fathoms, uevcr falls below this, even at ICOO fathoms, that of the 

 China Sea, which is 51° at 200 fathoms, falls to 37° at the same depth. As the 

 isocheimal of tlie Sulu Sea can never be any thing like as low as 50°, it is clear that 

 the reduction of the temperature of its deeper portion to that standard must depend 

 npou the entrance of cold water from the China Sea outside ; and it may be pretty 

 safely concluded that the depth of the channels of communication must be from 

 200 to 300 fathoms, so as to admit water of 50°, whilst excluding the deeper and 

 colder stratum. 



It is obvious that the existence of those peculiarities must have a very marked 

 influence on the Biological conditions of Inland Seas — and that, as like peculiarities 

 must have presented themselves in former periods of the Earth's history, the Lmow- 

 ledge of them may afford important aid in tlie interpretation of Palccoutological 

 phenomena. 



On the Ti-ee Ferns of iJie Coal-measures, and their Affinities with existhu/ 

 Forms. By W. Cakruthees, F.li.S. 



Lindley and Ilutton describe two species of Tree Ferns from the Coal-measures, 

 both from the Bath coal-field. I have been able to add eight species hitherto un- 

 described, chiefly through the assistance of J. M'Murtrie, Esq., of Kadstock. These 

 belong to tliree groups, which are remarkably distinguished by peculiarities in the 

 structure of the stems. Two of the groups belong to living forms, while the third 

 is extinct, being confined to Palfeozoic formations. Cuulopteris and Ttthicaulis be- 

 long to the same type as the living ferns which possess stems, including under this 

 term the humble stems (falsely called rhizomes) of many of our British species, as 

 well as the arborescent ferns of warmer regions, and excluding the rhizomatous 

 forms like Pteris, Fob/podiitin, and IlymenoplnjUum. In all these stems we have a 

 central medulla, surrounded by a continuous vascular cylinder penetrated regularly 

 by meshes, from the margins of which the vascular bundle or bundles to the fronds 

 are given oil) and through which the parenchj'ma of the medulla is continuous with 

 that of the stipes. In most tree ferns the medullary axis is larger, and the bases 

 of the stipes decay do\^■n to the circumference of the stem ; but in Osmiinda the 

 persistent bases of the stipes permanejitly clothe the small vascular cylinder, which 

 encloses a slender pith. To this latter form belongs the stipe vvith a dumb-bell- 

 shaped vascular bundle, separate specimens of which I have obtained from tlie Coal- 

 measures. These have been described, both on the continent and in this countrv, 

 under the name of Zi/f/opkris ; but they belong to Cotta's genus TnhicauUs, ai'id 

 they are very closely allied to a group of fern-stems which I have already placed 

 together under the name of Chekpteris. The stcm-structui-e of the comiiion tree 

 fern is represented by the genua Cauhjiieris, of which I have six species of carboni- 

 ferous age. 



The third and extinct group is represented by Corda's genus Steimnatopteris, only 

 now- known to be British, and by Psaronius, which, however, is not a separate ge- 

 neric form, but is only founded on specimens showing the internal structure of the 

 stems of which Corda's genus is the external aspect. The chief characters of Ptniro- 

 nius have been drawn from the structure of the aerial roots which iu\-est the stem 

 from which, indeed, the generic designation was derived ; while the structure of tJio 

 stem itself has been overlooked. But this is really of the first importance as wiU 

 appear from the follo-sving description which I have been al^le to make from' a finely 

 preserved specimen of an uudescribed species in the British Museum, and from the 

 fagures of Cotta and Corda. The circumference of the stem was composed of a con- 

 tinuous envelope of indurated tissue ; within this there were perpendicular tracts of 

 vascular tissue never penetrated by any mesh. Between these tracts the leaves were 

 given off in perpendicular series, the large single leaf-bundles coming right out from 

 the central parenchyma, in which the_y existed as well-formed bundles', filTiuf up more 

 or less completely the medullary cavity. In one form {Zipped) the leaves are oppo- 

 site, and the great proportion of the circumference of the stem is made up of the per- 

 sistent and common vascular tissue ; in others (species of Pmronim) the permanent 

 elements of the stem consist of throe, fom-, six or more perpendicular tracts. 



