HELOBIALES 437 



extra gametophyte is formed — a structure sometimes present in higher 

 famihes. 



Pollen-grain form gives no support to the theory that Casumina is not 

 an angiosperm; the pollen closely resembles that of the Betulaceae and 

 Myricaceae and is wholly unlike the polyplicate pollen of Ephedra. (A 

 position close to the Betulaceae was once suggested for the Casuarina- 

 ceae. ) 



Chalazogamy, first known in Casuarina (Fig. 103), was considered 

 gymnospermlike, and the genus was, therefore, believed transitional to 

 the gvmnosperms. But chalazogamy was soon found in other families, 

 and its importance in phylogeny discounted. 



Little is known of the anatomy of the flower. The stamen is described 

 as having two traces, which unite in the filament, then separate in the 

 connective. The filament is described as "having a tendency to split," a 

 character considered, in the classification that removes Casuarina from 

 the angiosperms, evidence of dichotomy and support for the protangio- 

 sperm theorv. The two-trace stamen is indeed primitive but occurs in 

 many other families, as in the Proteaceae and Betulaceae — an example 

 of retention of the primitive nodal structure of angiosperms. The ovule 

 is described as "lateral," but the origin of its trace is apparently 

 unknown. 



The Casuarinaceae are unquestionably angiosperms, a highly spe- 

 cialized family derived from primitive stock. Only a complete disregard 

 of morphological characters, external and internal, makes possible the 

 placing of this family with the Gnetales in "Protangiosperms." Pollen 

 and flower morphology suggest the Betulaceae as the near relatives of 

 the Casuarinaceae; wood structure shows many similarities to that of the 

 Hamamelidaceae. The casuarinas remain an isolated family; the prob- 

 ability that they are nearest to some of the amentiferous families is of 

 little significance, since the Amentiferae have been shown in recent 

 years to be a highly artificial assemblage of diverse taxa, characterized 

 by reduction in inflorescence and flowers, in association with the adop- 

 tion of unisexuality and anemophily. 



HELOBIALES 



Among monocotyledons, the Helobiales have been generally ac- 

 cepted as the most primitive, because of the polymery of the flowers, 

 which have little or no fusion. 



The flowers of the Helobiales have numerous free stamens and carpels 

 with transitions from spiral to whorled. (Some taxa have slight basal 

 fusion in the gynoecium, and carpel number may be reduced to few 

 or one.) The stamens have fairly primitive form, with flattened fila- 



