STRUCTURE ANi) DEVELOPMENT OF MOSSES AND FERNS. 91 modern conditions of existence, we come to the Vascular Crypto- gams, with their three classes — Filicinea, Equisetinece, LycopodinecB. The Filicinem are arranged in two groups — Euspora7igiat(B and Leptosporangiatm. The former group comprises the homosporous orders Ophioglossacea and Marattiacem, and also the heterosporous Isoetacea. The provisional transference hither of the latter order from the LycopodinecB Mr. Campbell thinks to be justified by the weight of our present evidence. The Eusporangiatce are regarded as the remnants of widely divergent branches of a common stock, which indicate the ancestry of the Leptosporangiatm on the one hand, and on the other possibly that of the Monocotyledons and Dicoty- ledons along the line of the Isoetacece (the Gymnosperms having arisen probably along a different line, viz., that of the LycopodinecB via the SehujinellacecB). An interesting observation will be found on p. 243, where it is shown that Botrychium alone among the Ferns exhibits in its rhizome a true secondary thickening of its vascular bundles. A thickening, but of a modified nature, also occurs in the stem of Isoetes. The germination of the macrospore and the embryology of the sporophyte of this latter genus are amply described and figured. The higher BotrychiecB and the Marattiacece are both taken to show lines of approach to the leptosporangiate Ferns, of which Ostnunda is the lowest step. The LeptosporangiatcB are, like the Eusporangiata, arranged in two groups^— the homosporous Filices and the heterosporous Hydro- pterides [Marsiliacem and Scdviniace^B). As to the position of the various orders, and the points of origin at which the Salviniacece and MarsiliacecB arose, Mr. Campbell has a good deal to say ; and his views will be found summed up in the diagram on p. 421. The morphology and life-history of Azolla are discussed at some length. Chap. XIII. is devoted to a consideration of the Equisetinece. The nuclear changes that accompany the development of the spermatozoids and of the spores are well figured and described: the courtesy of the publishers enables us to give specimens of these illustrations (see next page). Corinack's recent observation of a slight secondary thickening in the nodes of Equisetum maximum is interesting in connection with the thickening that occurred in the fossil CalamitecB. Another fossil of this group — Calammtachys — is of importance as being heterosporous, and confirming the opinion that heterospory arose independently in several widely separated groups of vascular cryptogams — an opinion of which Mr. Campbell makes full use in his scheme of classification. In Chap. XIV. we reach the final class, the LycopodinecB, with two homosporous orders, Lycopodiacece and PsilotacecB, and the heterosporous order Selaginellece. The author has closely investi- gated the female prothallium of Selaginella Kraussiana, and shows that the diaphragm in the macrospore is not a true cell- wall, but a secondary thickening of the lower side of the prothalhum which occurs during a temporary cessation of the growth of the latter. The LycopodinecB on the one hand retain several primitive characters, and on the other, through Selaginella, show an affinity with the Gymnosperms.
