194 J. Starhie Gardner — Ifesozoic Angiosperms. 



be in more perfect preservation, there is no twist visible, 1 so that 

 it is probable that either the spiral arrangement was confined to an 

 inner part of the structure, or that the purpose they had to serve 

 could be equally fulfilled without the twist. They seem to have 

 been seated on a pedicel, and according to Schimper, a number were 

 joined to form a sort of umbel ; but in England at least they are 

 generally found detached. Examples were figured in the Trans- 

 actions of the Geological Society as long since as 1840, 2 under the 

 name of Carpolitlies helicteroides. No better idea of their enigmati- 

 cal nature can be given than the bare enumeration of some of the 

 guesses that have been made as to their affinities. They were 

 thought to be allied to Helicteres, a genus of Sterculiacea, by Prest- 

 wich and Morris : they are Palceobromelia of Ettingshausen, allied 

 to the Pine-apple family ; the Palceoxyris, of Brongniart, related to 

 Xyris, 3 a sedge-like Monocotyledon principally confined to the 

 Tropics ; the radicular appendages of Equisetum according to Les- 

 quereux ; they were named Sporlederia by Stickler, who regarded 

 them as Bromeliads ; and Spirangium by Schimper, who hazards 

 no opinion as to their position in the vegetable kingdom, beyond 

 such as is implied by placing them after the Monocotyledons. Finally, 

 they are claimed to be " pro-Angiospermous " fruits by Saporta 

 and Marion, but it is almost superfluous to say that their affinities 

 remain to this day completely unknown. They are found in all 

 stages, from the Carboniferous up to and including the Wealden, and 

 any discovery tending to shed further light upon them would be of 

 very great importance. 4 



Next in point of age comes a leaf with reticulated venation from 

 the Permian of Russia, and named Dichoneuron Hooheri, Sap. It is 

 stated to be of a firm consistence, with a long petiole, slightly dilated 

 at the base and detached naturally from the stem ; and an engraving 

 shows it to be bipartite, indented, and lobed and cut laterally into 

 two segments. Saporta and Marion 5 claim that the venation re- 

 sembles that of Pistia (a tropical Duckweed), and that the leaves of 

 the aroideous Amormophyllum are similarly bilobed, and suggest the 

 association with it of a spathe-like body from the same formation. 

 Sir Joseph Hooker, however, to whom the specimen was shown, 

 thought that it was probably a cryptogam allied to Ceratopteris, a 

 genus of tropical and aquatic Ferns. It is in any case quite obvious 

 that very little importance can be attached to a mere leaf with such 

 undecided characters, when we reflect that leaves greatly resembling 

 each other with not dissimilar venation are common alike to Crypto- 

 gams, Gymnosperms, and both divisions of Angiosperms. Another 



1 There is an admirable specimen showing this condition in the Museum at 

 Owens College. 



2 Prestwich, Geology of Coalbrook Dale, Geol. Trans, vol. v. part in. plate 38, 

 fig. 12, and explanation by Prof. Morris. 



3 A view recently upheld by Nathorst. 



4 The species described as Palceospathe by Unger from the Carboniferous and Per- 

 mian, may possibly also belong here. A list of them is given in Schimper, Pal. 

 vegetale, vol. ii. p. 505. 



5 Evolution des Phanerogams, vol. i. p. 231. 



