xxv] THEORETICAL DEDUCTIONS 317 



remote in affinity from those hitherto considered, and is pro- 

 bably best interpreted as ultimately descended from the stock 

 from which the Liliiflorae were also derived. Solms-Laubach^ 

 regarded the genus Eichhornia as of older origin than Pontederia, 

 an opinion which accords well with the fact that Eichhornia has 

 a trilocular ovary with numerous ovules, while in Pontederia 

 the ovary is reduced to a single loculus containing one ovule. 

 Among the Farinosae we also find another entirely aquatic 

 family in the small group of the Mayacaceae. 



No other families among the Monocotyledons are exclusively 

 aquatic, but there remain certain cases of hydrophytic genera 

 and species, occurring among families which consist otherwise 

 of terrestrial or marsh plants. Examples from the British flora 

 are Scirpus lacustris and S.fluitans (Cyperaceae), Glyceria aquatica 

 and G. fluitans (Gramineae) and Sparganium nutans (Spargania- 

 ceae). The resemblance of Sparganium to the Pandanaceae is 

 so great that we may perhaps regard S. natans as representing 

 an aquatic off-shoot from the stock which also gave rise to the 

 Screw Pines. 



(2) THEORETICAL CONSIDERATIONS 2 

 From the foregoing section of this chapter certain general 

 conclusions may be deduced. The most obvious and striking 

 feature is the relative paucity of hydrophytes in comparison 

 with terrestrial plants. Contrasted with those that live on land, 

 the number of aquatic families is so small as to be almost 

 negligible, and even when all the individual hydrophytic genera 

 and species are added, the sum total is relatively insignificant. 

 This result is however hardly surprising when we consider that 

 the Phanerogams are essentially a terrestrial stock and are dis- 

 tinguished from the Cryptogams by their aerial mode of polli- 

 nation, which has won for them the freedom of the land. Under 

 these circumstances, the reversion to aquatic life could hardly 

 be expected to occur on any great scale. It must also be remem- 



1 Solms-Laubach, H. Graf zu (1883). 



2 This section of the present chapter is based on a recent paper by the 

 writer in the Journal of Botany. See Arber, A. (1919I). 



