492 BOTANY 



florinae, Opuntinae, Columniferae, Gruinales, Terebinthinae, Sapindinae, 

 Frangulinae, Thyrnelaeinae, Tricoccae, Umbelliflorae, Saxifraginae, Rosiflorae, 

 Legumirwsae, Myrtiflorae, and the provisional group of the Hysterophyta. 

 The first three orders comprise plants with simply constructed apetalous 

 flowers, often typically unisexual, and probably representing less highly 

 developed types ; while the group of the Hysterophyta, as is evident 

 from the parasitism of most of its members, is of more recent origin. 

 The sequence in which the orders are here given does not constitute 

 an ascending series, and it has no reference to their position in the 

 scale of development. 



Order 1. Amentaeeae 



Flowers hypogynous or epigynous, UNISEXUAL, SMALL, NAKED, or 

 with CALYCOID PBRIGONE ; the male in catkins (amenta) ; the female 

 in inflorescences of a different character. Number of stamens variable, 

 rarely the same as that of the perigone leaves. Gyncecium TWO- TO Six- 

 MISROUS. Seeds without endosperm. 



The Amentaeeae are all woody plants with alternate leaves. The 

 male inflorescences are characteristic of this order ; they have the form 

 of catkins, bearing the small flowers in the axils of scale-like bracts. 

 The female flowers are sometimes aggregated into catkins, as in the 

 Willow ; in other cases into capitate or spike-like inflorescences. The 

 fruit is usually a one-seeded nut, rarely a capsule or drupe. 



The diclinous flowers, the absence or imperfect development of the 

 perianth, the variable number and often irregular arrangement of the 

 parts of the flower in the same or in allied species, the almost uniform 

 wind-pollination, and the small degree of modification exhibited by the 

 flowers adapted to pollination by insects, make it probable that of all 

 the Dicotyledons the Amentaeeae differ less widely from the primitive 

 form, and represent phylogenetically the lowest stage of development. 

 That the primitive character of the flowers is not a result of reduction, 

 but of a low degree of development, is apparent not only from the 

 fact that all indications of such a reduction are lacking, but also 

 because the male and female flowers have so evidently not arisen, like 

 the unisexual flowers of the more highly developed types, from others 

 that were originally hermaphrodite. 



The close relation of the Amentaeeae to older, now extinct, types may be 

 assumed with certainty, if it should prove that the Casuarinaceae also belong or 

 are allied to this order. The last-named family, as Treub has shown, possesses 

 peculiarities which distinguish it from the Angiosperms and place it nearer 

 the Gymnosperms, or even the Pteridophytes (e.g. a multiplicity of embryo-sacs 

 with egg-apparatus, the presence of a cell-wall investing the still unfertilised egg- 

 cells, beginning of the endosperm formation before fertilisation). The Casuarinaceae 

 exhibit the additional peculiarity that the pollen-tubes do not approach the 

 embryo-sac through the micropyle, but by penetrating the chalaza (Chalazogamy). 

 According to recent investigations some undoubted A mentaceae are also chalazogamic, 

 a fact which would favour the inclusion of the Casuarinaceae within this order. 



