ee 
1903] UTION OF ISOETES SACCHARATA 195 
that species of local range are of the greatest importance in 
solving the fundamental problems of geographic distribution. 
_ From this brief discussion it will be seen that there are two 
essentially different kinds of endemism. In one the species is a 
remnant, in the other it is a beginning. To the former of these 
the term ve/ict endemism has been applied ;*° to the latter I shall 
apply the term zmztial endemism."' Relict endemism is illustrated 
by such classic examples as the Sequoia of western North America 
and Ginkgo biloba L. of eastern Asia. Initial endemism is perhaps 
best exemplified by numerous endemic species of oceanic islands, 
though it does not follow that all endemic species of oceanic 
islands are initial. 
It is obvious that many difficulties will be encountered in 
determining whether any local species in question is a relict or an 
autochthon. It is also obvious that we may have initial species 
of a relict genus, though not the reverse. After examining the 
nature of the barriers which limit the distribution of Jsoetes sac- 
charata 1 shall suggest to which class of endemic species it 
probably belongs. 
It will have been noted in examining the map, or in looking 
over the descriptions of the several stations, that even within the 
narrow confines of Chesapeake Bay, this species is not generally 
distributed along the shores, but occurs only here and there in 
Closely limited areas, the largest of which is perhaps less than 
1oo™ long. This extreme localization within its range is due 
solely to the requirements of its habitat. The chief conditions 
necessary for its success are the following: 
a) It is limited to tidal beaches, which fact restricts it to a 
Narrow zone, never more than a few meters wide along the shore 
6) It requires fresh water, never occurring in water of more 
‘than slight salinity. South of Spesutie Island the rivers have a 
* DRubg, O., Handbuch der Pflanzengeographie 125. Stuttgart: J. Engelhorn. 
1890. 
** An initial species is called an autochthon and auéochthon endemism might be 
used in contradistinction to re/ict endemism. Drude (of. cit. p. 124) refers to such 
Species as “vicarious or corresponding forms,” from which initial endemism is some- 
times called vicarious endemism, but the significance of vicarious in this connection 
As too obscure to commend its adoption. 
