47 
Fundamental Units of Vegetation. 
a floristic unit,.it is far from being a geographical unit of 
any sort. It would seem better to put many conifers with the 
lithophytes and psammophytes, whilst others are certainly oxylo- 
phytes, and others still pronounced mesophytes, the most mesophytic 
forests of the United States being dominated by conifers.” Several 
closely allied formations are, in my judgment, placed apart by the 
first divisions of Warming and Vahl’s new system. It must be 
remembered too that many plants, e.g., “ bog xerophytes,” possess 
some organs which, considered by themselves, would result in these 
plants being termed hydrophytes, whilst other organs would result 
in their being termed xerophytes; and certain plants, such as 
deciduous trees, are extreme xerophytes during winter, yet possess 
leaves which are “ mesophytic” during summer. Hence it is to be 
feared that any attempt to place every plant in a verbal class with 
the termination -phyte is either impossible or premature. 
Graebner’s (1901, 1909) classification of plant formations into 
larger groups has much to recommend it. Graebner (1909) makes 
three main divisions:— 
(a) Formations where the soil-water is rich in nutritive salts. 
(b) Formations where the soil-water is poor in nutritive salts. 
( c ) Formations where the soil-water is saline. 
These main divisions are further subdivided chiefly on the 
basis of the water-content of the soil, and then still further sub¬ 
divided by other physical factors. Thus the formations themselves 
are ultimately reached. This classification, although its author, of 
course, makes no claim that it is final, is, on the whole, and within 
an area of fairly constant climatic conditions, a natural one because 
it takes into account the whole of the factors comprising the habitat, 
and because no conflicting factor, such as plant form, is introduced. 
Doubtless future research will suggest some emendations ; but it 
may be regarded as the most successful of the existing systems, 
and as planned on a general and scientific basis which will lead to 
ultimate success. 
Gradmann (1909 : 93) has maintained that, in the matter of 
classifying formations, plant geographers should be given full 
liberty. It is sufficient to point out that plant geographers, have, 
in the past, exercised such liberty ; and the result is chaos. 
