472 BOTANY part ii 



were already present in the Carboniferous flora, wliile Isocfcs is only known witli 

 certainty from the lower Cretaceous strata. 



The discovery of seed-like structures borne by some of the palaeozoic 

 Lycopodinae {Lepidocarpon, Miadcsmin) is of special interest. In them the 

 macrosjiorangiura was surrounded by an integument leaving only a narrow slit- 

 like opening ; the sporophyll also took part in enclosing the sporangium. Only 

 one macrospore attained full development. As in Isoctes the prothallium re- 

 mained within the spore. The macrospores were produced on sporophylls 

 resembling those of Lepidostrobus. Probably pollination occurred while the 

 sporangia was still attached to the parent plant from which later the niacro- 

 sporophyll, together with its sporangium, separated as a whole. 



3. The small class of the Sphenophyllinae occupies an intermediate position 

 between Lycopodinae and Equisetinae. These three classes appear to have had a 

 common origin, and may be grouped together as the Lycopsidae in contrast to the 

 Pteropsidae which include the Filicinae and the Pteridospermae. 



The Sphenophyllinae were represented by two genera in palaeozoic times. 

 Gheirostrobus from tlie Lower Carboniferous had complex cones of similar structure 

 to those of the Calamarieae, but approached Lepidodendron. in anatomical structure. 

 The species of Sphcnophylhom which lived from the Devonian to the Permian 

 periods were herbaceous land-plants with elongated internodcs. The stems, which 

 underwent secondary growth in thickness, bore superposed whorls of wedge-shaped 

 or dichotomously divided leaves. The spike-like cones resembled somewhat those 

 of Equisetum ; each s]iorophyll bore one to four homosporous sporangia. 



Among existing Pteridophyta the Psilotaceae, M'hich were formeily placed in 

 the Lycopodinae, are apparently most nearly related to the Sphenophyllinae. They 

 include two genera Psi/o/iii/t with two tropical species, and Tmcsiptcris with one 

 species in Australasia. 



4. The class of Filicinae was richly represented in palaeozoic times, for instance 

 in the Carboniferous period, by the Marattiaceae and by the extinct family of the 

 Botryopterideae. 



The Hydropterideae are known with certainty from the Tertiary rocks, but 

 Salvinia and Marsilia can be traced back to the Chalk. 



So far as our knowledge goes the Equisetinae, Spheno'pliyllinae, and 

 Lycopodinae are branches of the Pteridophyte stock wliicli have undergone no 

 further development in the direction of the more highly organised plants. From 

 the Filicinae on the other hand the first seed-plants had arisen even in palaeozoic 

 times. A connecting group below the Ferns on the one hand, and the primitive 

 Gymnosperms (Cordaiteae, Cycadaceae) on the other, is provided by the extinct 

 class of the Pteridospermae (^^"') to which for example Lyyinodcndron and 

 Ncuropte/ris belong. These plants, which have only in recent times become 

 accurately known, formed a very important constituent of the Carboniferous flora. 

 In the general ai)pearance of their vegetative organs they resembled Ferns ; their 

 stems underwent secondary thickening. They were heterosporous ; the macro- 

 sporangia had a similar structure to those of the Cycadeae and must therefore be 

 termed seeds. Tliey were borne, as were the microsporangia upon fronds that 

 scarcely difl'ered from vegetative leaves. The sjiorophylls were thus not yet 

 arranged as in the Cycadeae in cone-like flowers. 



