PAPERS ON BOTANY 157 
the water somewhat below this stage, the play of waves, es- 
pecially of the strong scouring wash made by passing steamers, 
would make the germination of seeds impossible. 
It is therefore pretty certain that the dwarf plants in ques- 
tion could not have germinated until after July 28th, probably 
about the beginning of August. 
Now, notes concerning the general vegetation of the region 
indicate that back from shore, where conditions were normal, 
the germination of spring seeds was fairly well under way 
about April 27th. This would almost certainly include Am- 
brosva and Persicaria, which are among the earliest of seeds 
to germinate, and probably the rest of the species represented. 
From this date the general growing season extended until Oc- 
tober 26th, a period of 181 days, when it was terminated by a 
killing frost, the length of growing season being very unusual 
for this region. The growing season of the dwarf plants 
continued about a fortnight longer, or until November 14th. 
On account of their closeness, both to the ground and to the 
water’s edge, they escaped the first few killing frosts and per- 
sisted bright and green until a slight freeze. This makes their 
growing period (from about August lst to November 14th) 
about 107 days or 74 days shorter than that of the general 
vegetation. It was very probably considerably shorter than 
this; the germinating plants were not seen, and the plants 
were given the benefit of the doubt by assuming germination 
with remarkable promptness after the receding waters had left 
the ground bare where they were found. A part of the ground 
where they grew was indeed partly submerged by a 2 days rise 
September 15-17, and it is barely possible that the plants 
sprang up after that time, giving them a growing season ot 
only 39 days. Their exceedingly small amount of growth 
and their persisting cotyledons makes this short season seem 
not improbable. 
The plants collected were preserved in two forms (1) as 
dried specimens and (2) preserved in formalin, as it was 
feared that some of the more diminutive ones might be easily 
lost among the herbarium sheets. There is appended photo- 
graphs showing some of the dwarf plants natural size, and a 
chart showing river stages, rainfall, and the boundary-lines of 
the dwarf-plant area. 
