342 BOTANICAL GAZETTE [way 
to habitat is, I have no doubt, not adaptation in the old sense at all, but is 
a sifting-in of forms whose characters have been elsewhere and otherwise 
determined, but which happen to match, either by coincidence or by tolera- 
tion, with the physical conditions prevailing in that habitat. So far do I 
consider this true that I think our best working hypothesis in this field is 
this: any plant stands where it does for the reason that the physical demands 
made by the structure and habit it happens to possess overlap in some 
degree the physical conditions prevailing in that place, and the better they 
match the more nearly does the plant find its optimum, and the worse they 
match the more slender is the hold of the plant upon that place. Now the 
study and measurement of the physical factors of environments is steadily 
progressing; it behooves us to study and measure the physical demands 
of the plants in order that we may be able to compare the one with the other. 
This is the crux of what I mean by the need for physiological life-histories 
of plants. 
I come now to the practical part of the subject. I assume that the plant 
is a bundle of physical needs, and that our aim is to determine what and 
how great these are. Then these things are needful: 
1. We must devise methods and instruments, for the most part auto- 
graphic, for measuring and recording each physiological process. This, 
indeed, we are doing, but must do more and better. 
2. We must develop a system of standard units for each physiological 
_ process, such as will permit their exact expression and their comparison. 
We already measure photosynthesis in grams per square meter of surface 
per hour, and are approximating to similar definite units for respiration 
and transpiration. We must develop similar systems for the other processes. 
Further, we must develop methods of graphic expression of such data in 
order that they may be readily matched against the graphic results supplied 
by study of the environments. In most cases these will doubtless work 
themselves out in the form of minimum-optimum-maximum curves, but 
it will not be enough to make these curves represent means or averages; 
they must express the range of frequency from the mean, which can be done 
by means of shaded or penumbrate graphs, in which the mean forms the 
dark center and the shading merges off thence to the extremes. 
3- We must not confine our measurements to those invisible processes 
in the protoplasm which we commonly associate with physiology, but 
must include structure which is simply an external and visible manifesta- 
tion of physiological operations, a tool which physiology forges as an aid 
in its processes. Does a plant produce a long tap-root ? That is because, 
for reason sufficient unto itself even though unknown to us, the protoplasm 
