l64 THE SCOTTISH BOTANICAL REVIEW 



The Past History of Monocotyledons, with Remarks 

 on their Origin. By A. R. Horwood. 



(With Two Plates.) 



I. Introduction. 



Much progress has been made during the last decade in our 

 knowledge of the affinities of the Palaeozoic fossil plants, in so 

 much that it has now become possible to sketch out, in a more or 

 less realistic manner, the history of the early vascular Cryptogams. 

 Side by side with this increase in our knowledge, which is largely due 

 to Williamson, whose work has been carried on by Scott, we have 

 become acquainted with two other groups of an entirely unique 

 character, which are also found in excellent preservation in the 

 Coal-measures. 



The Pteridospermeae are at once a synthetic race, combining with 

 a fern-like foliage and stem anatomy of cycads a fructification of the 

 Phanerogamic type. Here, in fact, we may look for the origin of 

 the seed-bearing plants. Most of the Palaeozoic ferns formerly 

 regarded as Marattiaceous ferns are now found to belong to a 

 distinct group having cycado-filicinean affinities. 



Another group of more decided gymnospermic type is that 

 of the Cordaitales. In this extinct race we also have synthetic 

 characters. 



The stem anatomy resembles that of conifers of the Araucarian 

 type and the leaves are reminiscent of cycads, whilst the fructifi- 

 cation is distinctly like that of Ginkgo, a monotypic group represented 

 now by the Maidenhair tree of Japan. In the Mesozoic epoch the 

 researches of Carruthers and Seward in England, Lignier in France, 

 and Wieland in America have made it clear that in Beiviettites we 

 have also another synthetic type, combining cycadean affinities — in 

 some respects resembling, but really distinct from, Williamsonia — 

 but possessing also a distinctly Angiospermic tendency. Thus the 

 receptacle bears numerous cylindrical slender pedicellate structures 

 which bear the seeds. Between these are the inter-seminal scales. 

 At the periphery the distal portions of the scales form a pericarp 

 entirely enclosing the seeds. (PI. I. fig. 2.) 



In Be?inettites gibsonianjis, a section through the micropylar end 

 exhibits the embryo with remains of the endosperm above, the 

 slightly bifid micropyle, and a three-layered testa. The connection 

 between the adjacent scales and pericarp can also be clearly seen in 

 longitudinal section. (PI. I. fig. i.) 



The embryo was dicotylar, the radicle pointed, and situated at 

 the micropylar end. Traces of the plumule have been detected. 

 The exterior of the pericarp is furrowed being penetrated by the 

 micropyles of the seeds, between which there is a network formed 



