200 BOTANICAL GAZETTE [SEPTEMBER 
elevation, the reverse process must have taken place, and the 
many distinct areas must have been merged again into one. 
This may have taken place as often as the bay has been up and 
down, and certainly has happened as often as the bay has risen 
and fallen since Jscetes saccharata entered it. 
Just how or when Isoetes entered Chesapeake Bay is, of 
course, impossible to say, except that, according to this hypothesis 
of its dispersal, it must have been introduced before the last’ 
sinking of the coastal plain. 
From what has been said of the requirements of its habitat 
and the means of dispersal of Isoetes, it will be seen that the 
barriers between Chesapeake Bay and Delaware Bay and between 
both of these and other fresh tidal waters, are of such definite char- 
acter as to render these bays virtually islands of water in oceans 
of land. As we find Jsvetes saccharata nowhere else than in 
Chesapeake Bay and Jsoetes riparia nowhere else than in Delaware 
Bay, it seems fair to assume that neither of these species ever 
existed as such outside of the body of water to which it is now 
limited, and that we have here examples of initial endemism 
entirely comparable with that so common upon oceanic islands. 
These two species are closely related and probably stand to 
each other in relation of parent and offspring; but which is the 
parent and which the offspring may not be easy to decide. Or 
perhaps they were the offspring of a common parent different 
from both. The nearness of this relationship was emphasized in 
the recognition of the two varieties, /soetes saccharata Palmeri A. 
A. Eaton and J. saccharata reticulata A. A. Eaton, both possessing 
characters intermediate between Jsoetes saccharata Engelm. and 
fsoetes riparia Engelm. The significance of these forms will be 
increased rather than lessened if they should prove to be unten- 
able as varieties. For if these varieties are shown to be simply 
stages in the development of a polymorphic species, the greater 
range of variability which must then be admitted as a character of 
I. saccharata Engelm., coupled with the fact that those variations 
in several different features are in the direction of Jsoetes riparia 
Engelm., would make almost certain the inference that an 
extreme variation of /. saccharata had become somewhat fixed 
through its isolation in Delaware Bay. 
