i8o 



THE SCOTTISH BOTANICAL REVIEW 



He arrived at the following conclusions in so far as monocoty- 

 ledons are concerned : — " i. It is probable that monocotyledons and 

 dicotyledons appeared at about the same time, namely early in the 

 Mesozoic or late in the Palaeozoic.^ 2. The hypogynous monocoty- 

 ledons appear to have preceded the epigynous monocotyledons, and 

 similarly the petaloideous hypogynous species seem to have 

 somewhat preceded the spadiceous and glumaceous species." 



Orders. 



Apocar/^(e{A\isma.ce3i, Triuridese 

 Naiadacese) .... 



CoronariecE (Stemonacese, Li- 

 liacese, Pontederiacete, Fhily- 

 draceas, Xyridaceae, Mayacese, 

 Commelinaceai, Rapateacese). 



yV^«afy?(7;vz'(Pandanacea;, Cyclan- 

 thacese, Typhaceae, Aroidca, 

 Lemnacege) .... 



Calyciiut (Flagellariacese, Jun- 

 cacese, Palmaceas). 



Glumacece (Eriocaulaceas, Cen- 

 trolepidaceae, Restiacese, 



Cyperacese, Gramineae). 



ZTj/o'ra/^j- ( HydrDcharitaceae) 



Epigyniv ( IJioscoreacese, Tac- 

 caceae, Amaryllidacese, Iri- 

 daceae, Haemudoraceas, Bro- 

 meliacefe, Scitamine?e) . 



Microsperma: (Burmanniaceas, 

 Orchidacese). 



The occurrence of Alismacites, on the one hand, in the Wealden, 

 and Liriodendron, Magnolia, and other primitive dicotyledonous 

 genera in the Cretaceous of Europe and America, and the persistence 

 of Bennettiteas in the Mesozoic floras with a broadly homologous 

 floral structure is a fact which cannot be ignored in studying the 

 origin of Angiosperms as revealed by the records of fossil evidence. 



As a broad generalisation it is the only outstanding feature which 

 suggests any material for the construction of a phylogenetic scheme, 

 and must therefore be accepted as a working hypothesis. 



Thus palaeontology leads us so far. The remaining links in the 

 chain must be supplied from the study of the embryology, 

 morphology, ecology, and the systematic characters of the group. 



^ We prefer to say /ate in the Mesozoic. 



