CONSTANCE: SYSTEMATICS OF THE ANGIOSPERMS 449 



original groups — Pandanales, Helobiae (Alismatales), and Glumiflorae — was in- 

 dicated by Engler and favored by Campbell; Ponzo accepted these three and 

 added Liliflorae (Liliales) as a fourth, all in turn to be regarded as offshoots of 

 dicotyledons. Finally, Suessenguth thought he recognized such interclass con- 

 necting links as Cabombaceae-Butoniaceae, Eanunculaceae-Alismataceae, Ber- 

 beridaceae-Liliaceae, and Taccaceae-Aristolochiaceae. 



Depending upon what kinds of plants were postulated as the source or sources 

 of monocotyledons and the mechanism of cotyledonary change, it has frequently 

 been suggested that monocots arose as an adaptation to aquatic or marshy habi- 

 tats (Henslow, Bews, Andreanszky), or to a geophilous mode of life (Sargant, 

 Hill, Ponzo), or to both (Arber and Parkin). Henslow felt that the aquatic 

 environment had a "degenerating effect" upon structural features and pointed 

 to the parallelism in loss of cambium and secondary growth, the scattering of 

 vascular bundles, and parallel leaf-venation in hygrophilous dicots and mono- 

 cots. Sinnott (1914) and Calestani emphasized that the development of sheath- 

 ing leaf-bases might have led to a multiplication of both foliar and cauline traces, 

 and hence to the production of multilacunar nodes and "endogenous" stem struc- 

 ture. Arber and Parkin suggested that in monocotyledons the evolutionary 

 sequence might have been from herbaceous to woody types, reversing that 

 assumed for dicotyledons. All of these explanations are, of course, highly 

 speculative. 



In her excellent monograph on monocotyledons, Arber expressed her well- 

 known antipathy to the intrusion of phylogenetic speculation into the study of 

 form pursued for its own sake. In her opinion (1925, p. 217), 



. . . the great groups of Monocotyledons have not achieved unlikeness by divergent modi- 

 fication, but they must have been of different types from the moment of tlieir appearance. 

 . . . We have no evidence from Palaeobotany for the former existence of synthetic types 

 uniting any of the Monocotyledonous cohorts, and I am inclined to suppose that these 

 great groups will ultimately be traced back to a very remote antiquity, without display- 

 ing a common origin. 



She agrees that it would be logical to suppose an ultimate derivation of such 

 stocks from a single, primeval "Urmonocotyledon," but believes that such an 

 assumption is by no means proved and that we must, for the present, give up all 

 hope of discovering either connecting links between the great groups of mono- 

 cots or between monocots and dicots. In striking contrast with these rather pes- 

 simistic views is the brash comment of Ankermann that "today we may accept 

 the system of monocotyles as perfectly worked out" (1927, p. 46) ! Is there any- 

 thing positive to be gained from this welter of hypothesis and counterhypothesis 

 as to the origin and evolution of monocotyledons and their affinities with dicoty- 

 ledons or other vascular plants? Or must we assume, on the basis of majority 

 vote, that there is some kind of connection between florally unspecialized Ranun- 

 culaceae, Nymphaeaceae, and Alismataceae, and leave the matter at that point? 

 The recent extensive investigations by Cheadle on the vascular tissues of 

 monocots, like those of Bailey on dicots, appear to offer us some hope, although 

 Cheadle has been exceedingly chary of drawing phylogenetic conclusions. His 

 work has revealed : (1) that in essential features the vessels in the primary xylem 

 of monocots parallel the unidirectional trends established for those in the sec- 

 ondary xylem of dicots; (2) that a progressive series can be established for sieve 



