378 Palaeontologie. — Eumycetes. 



allied, though probably not closely, to the Mesozoic and recent 

 Cycadophyta. The seeds of the Pteridosperms resemble the seeds 

 of the resent Cycads more closely than \hey do those of the 

 Betiettiteae : but the analogy between the microsporangium-bearing 

 fronds of the Pteridospertns and the Compound leafly sporoph3dls of 

 the Benettiteae is striking. On the whole of the evidence it appears 

 likeh^ that at least some of the Cycadophyta were derived from 

 forms resembling the MeduUosae though probably monostelic. Though 

 the habit of the Cordaiteae is very different from that of the Pteri- 

 dosperms; their seeds are in many ways closely similar, bilateral 

 as well as radial seeds occurring in the latter; in the anatomy too 

 a very complete series (not of course a phylogenetic one) may be 

 constituted passing from the Lyginodendreae to the tj^pical Cordaiteae 

 through Calauiopitys, Poroxylon the Pityeae and the exceptional 

 Cordaiteae of the English Lower Carboniferous. Thus an affinitj- 

 between Cordaitales and Gymnosperms is established, though the 

 connection must lie very far back. The vexed question of the 

 Lj'copodial or Filicinean origin of the Araucariae is discussed; very 

 strong reasons are given in favour of their being derived from 

 Perus ; these reasons include the generally accepted existance of an 

 affinity between Conijers and Cordaiteae coupled -with the Pteri- 

 dosperraous affinity of the latter group; the possession of a relatively 

 large number of pollen-sacs on the stamens of the Araucariae. The 

 principal point in favour of an afiinity with the Lycopods, the 

 presence of a Single ovule on the Upper surface of the sporophyll, 

 loses much of its weight if the Cretaceous Protodainniara with three 

 ovules on each scale is rightly referred to the Araucariae. As there 

 seems no reason to split up the Conijerae into two groups having 

 a separate origin it appears probable that they were all derived 

 from the Cordaitales though not necessarily from the Cordaiteae. 

 The bisexual nature of the Benettitean cones and the relative 

 Position of their microsporophylls and megasporophylls brings these 

 cones very near to an Angiospermous flower, especially to that of 

 the MagnoUaceae and other '' PoJypetalae" . The Benettiteae also show 

 some approach to the formation of a closed ovary, while their 

 dicotj^ledonous embryo and exalbuminous seed are very striking. 

 Though there may be a difference of opinion as to the nearness of 

 the relation of the Benettiteae to the Angiosperms we can hardh' 

 fail to recognize the existence of a real relationship. We must there- 

 fore look for the ancestors of the Angiosperms among the plexus of 

 Mesozoic Cycadophyta ; this view involves the origin of the Monoco- 

 tyledons from the Dicotyledons no doubt at an earl)^ period, for so 

 far the palaeontological record shows that the two classes are of 

 almost equal antiquity. Evidence has recently been adduced by 

 Arber and Parkin that the Gnetales may have been derived by 

 reduction of the floral organs from forms allied to the Benettitales. 

 Thus it seems likely that both Angiosperms and Gymnosperms are 

 ultimately descended, through primitive seed-plants like the Pteri- 

 dosperms, from the same stock as the Ferns. 



Isabel Browne (University College London). 



Biffen, R. H., First record of two species of Labotdbeniaceae 

 for Britain. (Trans. Brit. Myc. Soc, Season 1908 [1909]. p. 83.) 



The species recorded are: Stigmatomyces piirpnrens Thax. , on 



