30 Bews . — Some General Principles; of Plant Distribution as 
of endemic monotypes with no obvious close relationships the difficulty 
of reaching a satisfactory explanation is greater. 
Fortunately there are not many such, and several which were once 
thought to be of this type have since had relations discovered in the tropics. 
If all other explanations fail, they must be considered relicts, whose im- 
mediate ancestors have disappeared. Guppy would class most of them 
as specialized types, the oddities of plant forms which have been produced 
not by differentiation from a more generalized type but by increased 
specialization. The Rosaceous Pygeum africanum or Leucosidea sericea are 
as good examples as any, but Noltea africana , Hippobromus alata , Curtisea 
faginea , are also monotypes. 
Other isolated types have been referred to in the general survey of 
the families. It may be noted incidentally that monotypes and genera 
containing few species have generally more restricted distribution than 
others. (Cf. de Candolle, ‘Origin of Cultivated Plants’, 2nd edit., 1909, 
p. 395, in connexion with the origin of Zea Mays.) 
General Applications. 
My object in bringing forward these facts and arguments has been 
mainly to afford illustrations of what I have begun more and more to 
realize in connexion with field ecological work, namely, the necessity for 
new view-points. The ordinary ecological methods of investigation are 
extremely useful up to a certain point. We find that certain plants occupy 
certain habitats, each with a certain range of the measurable factors, which 
vary to a considerable extent independently. As Clements has pointed out, 
and as can be seen by any one who has studied plant-life in the field, very 
fine measurement of each factor may be largely wasted effort, since (with 
the possible exception of some soil factors) it requires a relatively large 
difference in any single factor to cause any morphological adjustment. 
Nevertheless, since the factors vary independently, there are a fair number 
of permutations permissible. Thus as regards the moisture factor, habitats 
may be divided into aquatic, semi-aquatic, moist, mesophytic, dry and very 
dry, and any of these may be combined with habitats having full sunlight, 
diffuse light, and dense shade, and then with clay soils, sandy soils, and 
so on. In spite of their seeming complexity the essential inorganic 
differences between habitats are not so very difficult to recognize. Bio- 
logical factors, however, are more complex, and ecologists are giving more 
and more attention to them. In its widest sense, this aspect of ecology 
includes the vegetation itself as a factor and embraces, therefore, the whole 
question of plant succession. 
We are by no means near the end to be reached by the use of these 
ecological methods, and how very useful they are has already been shown 
by the results, but if we do not take into account the historical or geo- 
