448 BOTANICAL GAZETTE [DECEMBER 
ecological knowledge and methods. The study does make fairly 
clear, however, the directions in which research must now 
proceed, and upon this I desire to offer some comments. 
The idea of ecological plant geography, the broadest and 
most important phase of ecology, is to interpret the physiog- 
nomy of vegetation; to tell precisely why each plant is where it 
is, in the company it is, and of the form, size, color, texture, etc. 
itis. Each plant, as it stands in nature, is an adjustment or 
equilibrium between its physiological powers and properties on 
the one hand, and the properties of the environment, physical 
and biological, on the other. Now, for a full understanding of 
these matters four principal things are needful. 
First, a collection and description of the actual facts as to the 
kinds of plants which occur in a given region, as to their visible 
features, and as to the way in which they are grouped. Our 
present-day ecology, especially that which is being actively 
pursued in this country, is strong in this descriptive work, to 
which indeed it is well-nigh confined, and it is giving an excel- 
lent basis for future advance. In the present paper I think 
these facts about the marshland vegetation are fairly represented. 
Even from this point of view the study is defective in one 
respect, common to most of our ecological studies, namely, 
the descriptions apply to the summer only; if followed through 
the year (as it should, but for practical reasons could not, be), it 
might, and probably would, Iead to conclusions somewhat 
different in details, as well as to much additional knowledge. 
Second, an exact study and clear expression of the facts as to 
the physical features of the environment which can affect plant 
life. For the study of the physics of the atmosphere, the 
methods have been thoroughly organized by meteorologists, but 
we need some way of expressing meteorological results in a form 
for ecological use. It is very common in ecological papers, as 
in the present one, to give elaborate tables of temperature, Pre" 
cipitation, etc., and then to dismiss them with a few words of 
general comment. This shows either that meteorological data ate 
obtainable more copiously and exactly than needful for ecolog!- 
cal use, or else, as is more likely, they are not expressed in a 
