GEOGRAPHICAL DISTRIBUTION 173 



Marchantiaceae. This, of course, means a long period 

 of evolutionary development from similar forms to the 

 more complex, preceding the geological age of the rocks 

 containing the fossil record, and one such bit of positive 

 evidence fully substantiated, is of itself sufficient to 

 establish the antiquity of the liverworts. Moreover, 

 when such testimony is in agreement with the evidence 

 derived from other sources, such as comparative mor- 

 phology and geographical distribution, the fact of anti- 

 quity would seem to be reasonably well established. Now, 

 in addition to the evidence of comparative morphology, 

 there are, as Campbell points out, certain facts of dis- 

 tribution that can only be satisfactorily interpreted on 

 the basis of the comparative antiquity of the liverworts. 1 

 The liverworts are a widely distributed group ; some of the 

 genera are cosmopolitan, i.e., they are found practically 

 everywhere, in all continents, climates, and habitats, 

 and widely on oceanic islands. Riccia and Marchanlia 

 are cosmopolitan genera of continuous distribution. 

 Other genera are of wide, but discontinuous distribution, 

 such, for example, as Targionia, a genus containing only 

 two species, which are found in Southern and Western 

 Europe, Africa, Java, Australia, and Western America, 

 but are absent from Eastern America and from most of 

 Asia. The familiar Lunularia cruciata of our greenhouses 

 has a distribution similar to Targionia in the eastern 

 hemisphere, but is unknown in the western hemisphere 

 except where introduced. 



1 Throughout the discussion of liverworts I have drawn freely on Camp- 

 hell's article, cited above, and have, to a certain extent, adopted his 

 wording, asking the reader and the author quoted to accept this statement 

 in lieu of frequent quotes. 



