ANGIOSPERMiE. 1 539 



a leaf with a net-like venation described under the name of Dicho- 

 neuron, has been regarded as a Monocotyledon, but apparently on 

 quite insufficient grounds. The genus (Ethophallum, from the Trias 

 of the Vosges, which has linear leaves arranged in groups of three 

 on the branches of a woody stem, and long seed-bearing spikes, 

 has also been referred to this subclass. Certain fruit-spikes from 

 the same deposits described as Echinostachys may belong to kindred 

 types. Again, the long ribbon-like leaves found in the Rhaetic 

 and Jurassic and known as Yucciies, have been looked upon as in- 

 dicating plants allied to the existing Drac<z?ia ; but it is quite pos- 

 sible that remains of totally distinct types of vegetation have been 

 included under this name. In the Upper Trias and Lias of Swit- 

 zerland occurs the imperfectly known Bambusium, which Mr J. S. 

 Gardner regards as an undoubted Monocotyledon, although Dr 

 Schenk suggests affinity with the Equisetaceas. The former writer 

 also regards as Monocotyledonous a stem from the Yorkshire Oolites 

 described as a Calamite. Plants from the Jurassic which have been 

 named Naiadita, Bensonia, &c, are not Phanerogams at all; while 

 Aroides of the Great Oolite is based on a Crinoid. Stems with a 

 rush or grass-like form from the Purbeck appear, however, to be true 

 Monocotyledons. Finally, Rhizocaulon, from the Upper Eocene of 

 the Paris basin, is based on the evidence of a stem which may indi- 

 cate a plant allied to the Cyperacea. 



Leaving these doubtful types, we may proceed to the considera- 

 tion of Monocotyledons which can be systematically placed. 



Order i. Liliiflor^. — This order includes the Lilies, Irises, 

 Yuccas, and their allies, most of which have conspicuous perianthed 

 flowers, which vary considerably in structure, and may be of large 

 size. The plants may be perennial, but are more usually annual, 

 with rhizomes or bulbs. The Liliacetz (Lilies, Tulips, Aloes, Yuccas, 

 &c.) are but of little palaeontological importance. From the Eocene 

 and higher Tertiaries plants have been described under the names 

 of Agavites, Yucca, and Dracana, the two last being existing genera ; 

 but Dr Schenk considers that most of these belong to Draccena 

 (Dragon-tree), now found in the Canaries, Africa, and India. Plants 

 of the genus Smilax — the type of a subfamily of Liliacece — occur in 

 the European Tertiaries from the Eocene upwards ; although it 

 seems doubtful if forms described as Smilacina and Majanthemo- 

 phyllum really belong to this group. The Juncacecz, or Rush family, 

 are known by species of Juncus from the Upper Miocene of CEnin- 

 gen ; and, according to Dr Schenk, a species of Iris from the same 

 deposits is the only fossil representative of the Iridacecs, although 

 several other fossils have been described under that name. In the 

 Dioscoreacece (Yams), Dioscorites, from the European Miocene, is 

 probably allied to Dioscorea ; while in the Pine-apple family, or 



