72 



Rendle states "The arrangement does not claim to be strictly 

 phylogenetic. Various attempts have been made to construct 

 a phylogenetic system of Angiosperms, but the results are not 

 convincing, bear no suggestion of permanence, and bristle with 

 difficulties for the student." "A key mostly emphasizes the 

 differences amongst plants," writes Hutchinson; "a truly 

 natural and phylogenetic classification, however, should rather 

 emphasize their resemblances, by which alone their true affin- 

 ities may be ascertained. This I have attempted to do in the 

 pages of this book. " 



Rendle writes: "While it is possible that some (of the Apet- 

 alae) may be reduced forms, it is on the other hand possible to 

 regard these as representing lines of development from earlier 

 extinct groups. It seems likely that the development of the 

 highly differentiated, insect-pollinated dichlamydeous flower 

 was preceded by numerous, so to speak, experimental stages, 

 arising from earlier, now long extinct, angiosperms, and it is 

 a tenable view that such stages are represented among the 

 Monochlamydeae." Hutchinson takes another view: "It 

 seems very probable that extreme reduction of the perianth 

 and consequent loss of attractiveness to insects would result 

 in the adoption of another mode of pollen transference, by the 

 wind, which in this case would not be a primitive condition, as 

 it undoubtedly is in the Gymnosperms. In my opinion, the 

 universally accepted theory of the foliar origin of the carpel is 

 fatal to the assumption that the " Amentiferae" are primitive. 

 Many of these have an ovary composed of the union of two or 

 more carpels, which must be the result of cohesion and reduction 

 from older groups which originally had free carpels." 



Possibly the views may not be entirely irreconcilable; ances- 

 tors of the Amentiferae may have had separate carpels, yet not 

 have resembled primitive insect pollinated groups like Magnoli- 

 aceae. 



The divergent interpretations in these works lend special 

 significance to their points of agreement, such as the following: 



Casuarinales are placed by both near Fagales (cf. Benson); 



Piperales are removed from the early position occupied in 

 Engler's system (cf. Johnson); 



Magnoliaceae precede Nymphaeaceae; 



Balsaminaceae come with Tropaeolaceae in the Geraniales; 



