534 



Dr. J. C. Willis. On the Lack of 



changes without natural selection. Lastly we shall consider in brief the 

 factors which seem to have had an influence in the evolution of these remark- 

 able families. 



The orders Tristichacese and Podostemaceae (for I have shown in a recent 

 paper* that the old family of Podostemaceae must be split into these two, 

 unless some intermediates can be found, in Africa), as is well known, 

 consist only of annual plants which live upon the surface of smooth water- 

 worn rocks in rapidly moving water, and are confined, with the exception of 

 one species found in Ohio, to the warmer regions of the globe. During the 

 wetter season of the year they carry out their vegetative life, and form their 

 flowers as the dry weather approaches. The flowers open above the water 

 level as the rivers sink, and the seeds are shed upon the rocks, where those 

 few that manage to retain their position germinate with the beginning of the 

 wetter season. 



Conditions of Life. — It has long been an axiom that the conditions of life 

 of the plants of still water are extraordinarily uniform, and that to this is due 

 the small number of these plants, and the small amount of evolution through 

 which they appear to have gone and to be going, together with their enor- 

 mously wide distribution. But it will soon be seen that the conditions of life 

 for the Podostemacese and Tristichacese are if anything even more uniform, 

 and yet in contrast to the plants of still water they show the most astounding- 

 morphological differences, and form about 30 genera with over 100 very 

 different species (and it is certain that very many species and probably genera 

 have yet to be discovered). The distribution of these species is more and 

 more localised the more specialised they are in structure, while the widely 

 distributed genera are those which are the most like the original ancestors 

 from which the Podosternaceas and Tristichaceee are descended. 



The conditions of life of these plants are, and must always have been, 

 as uniform as, or more uniform than, those of other water plants. In the 

 first place, as to substratum, they grow as a rule only upon rocks, but, as 

 they will also grow upon anything which may have become wedged between 

 the rocks so as to be immovable, the actual chemical or physical com- 

 position of the substratum cannot be of importance ; in all probability the 

 plants take nothing from it unless, perhaps, silica. They grow upon this 

 rock bottom by means of creeping roots, which give off secondary shoots. 

 They are thus absolutely compelled from the start to a plagiotropic 

 mode of life. In the second place, as to temperature conditions : the 

 orders are, with the exception of Podostemon ceratophyllum, found in Ohio, 



* Willis, "A New Natural Order of Flowering Plants — Tristichaceae," 'Linn. Soc. 

 Journ., Bot.' (in the press;. 



