September 22, 1910] 



NATURE 



385 



are modifying it ; and (3) tlie pliase in the cycle of forms 

 characteristic of such structure acted on by such forms. 

 We may say that any form is a function of structure, 

 process, and time. Tlie matter is even more complicated, 

 lor we have instances, e.g. in antecedent drainage systems, 

 of the conditions of a previous cycle affecting a subsequent 

 one — a kind of heredity of forms which cannot be 

 neglected. 



1 he georaorphologist is seeking for a genetic classifi- 

 cation of forms, and in the works of Betrand, Davis, de 

 Ja Noe and de Margerie, Penck, Richthofen, Suess, 

 and Supan and their pupils are being accumulated 

 the materials for a more complete and systematic classifi- 

 cation of forms. As you all know, the question of terms 

 /or the manifold land-forms is a difficult one, and apt to 

 engender much more controversy than the analysis of the 

 forms themselves. I believe that we shall find it advan- 

 tageous 10 adopt some notation analogous to that of the 

 cliemists. I have not yet had time to work such a nota- 

 tion out in detail, but it might take the form of using 

 different symbols for the three factors noted above — say, 

 letters for different kinds of structure, Arabic figures for 

 processes and Roman figures for the stage of a cycle the 

 form has reached. 



Take a very simple set of structures and indicate each 

 bv a letter ; — 



I homogeneous 

 I horizontal 

 .-, layered \ tilted... 

 [folded 

 [mixe " 



xed 



If pervious or impervious, a /) or an t could be added — 

 e.g. a tilted limestone with faults would be C'/>. 



Ne.\t, indicate the commoner erosion processes by Arabic 

 numerals : — 



j'movincr water ... ... ... I 



i'rocess 



I wind 

 [sea ... 



One process may have followed another, e.g. where a 

 long period of ice erosion has been followed by water 

 erosion we might write 2-1, w-here these alternate annually, 

 say 21. 



The phase of the cycle might be denoted by Roman 

 figures. A scale of V might be adopted, and 1, III, and 

 \' used for youthful, middle-aged, and old-aged, as this has 

 been called, or early, middle, and late phases, as 1 prefer 

 to term them. II and I\' would denote intermediate 

 phases. 



.iV scarped limestone ridge in a relatively mature phase 

 like the Cotswolds would be, if we put the process first, 

 I cm.; a highland like the Southern Uplands of Scot- 

 land would be denoted by the formula i. 2-1 E' III. 



This is the roughest suggestion, but it shows how we 

 could label our cases of notes and pigeon-hole our types 

 of forms — and prevent for the present undue quarrelling 

 over terms.' No doubt there would be many discussions, 

 for e.\ample, about the exact phase of the cycle, whether 

 ice, in addition to water, has been an agent in shaping 

 this or that form, and so on. But, after all, these discus- 

 sions would be more profitable than quarrels as to which 

 descriptive term, or place-name, or local usage should be 

 adopted to distinguish it. 



The use of such notations in geographical problems is not 

 unknown. They were employed by Koppen in his classifica- 

 tion of climate ; and now, in the case of climatology, there 

 is coming to be a general consensus of opinion as to what 

 are the chief natural divisions, and the use of figures and 

 letters to indicate them has been followed by several other 

 authors. This should also be attempted for oceanography. 



If any international agreement of symbols and colours 

 could be come to for such things it would be a great gain, 

 and I hope to bring this matter before the next Inter- 

 national Geographical Congress. 



• What I wish to make clear is that it is not necessary 10 invent a new 

 term for every new variety of land form as soon as it is recognised. It will 

 suffice at first to be able to label it. The notation will also stimulate the 

 search for and recognition of new varieties. 



NO. 2134, VOL. 84] 



Ihe Need for Selecting Natural Geographical Units. 



We have still to come to Geography proper, which con- 

 siders land, water, and air, not merely separately but as 

 associated together. What are the units smaller than the 

 whole Earth with which our science has to deal ? 



When we fix our attention on parts of the Earth and 

 ask what is a natural unit, we are hampered by preconcep- 

 tions. We recognise species, or genera, families, or races 

 as units — but they are abstract rather than concrete units. 

 The reason for considering them as units is that they 

 represent a historical continuity. They have not an actual 

 physical continuity such as the component parts of an 

 individual have. Concrete physical continuity in the present 

 is what differentiates the geographical unit. Speaking 

 for myself, 1 should say that every visible concrete natural 

 unit on the Earth's surface consisting of more than one 

 organic individual is a geographical unit. It is a common 

 dilliculty not to be able to see the wood for the trees ; it 

 is still more difficult to recognise that the wood consists 

 of more than trees, that it is a complex of trees and other 

 vegetation, fixed to a definite part of the solid earth 

 and bathed in air. We may speak of a town or State 

 as composed of people, but a complete conception of cither 

 must include the spacial connections which unite its parts. 

 .\ town is not merely an association of individuals, nor is 

 it simply a piece of land covered with streets and build- 

 ings ; it is a combination of both. 



It is true that in determining the greater geographical 

 units, man need not be taken into account. We are too 

 much influenced by the mobility of man, by his power to 

 pass from one region to another, and we are apt to forget 

 that his influence on his environment is negligible except 

 when we are dealing with relatively small units. The 

 geographer will not neglect man ; he will merely be careful 

 to prevent himself from being unduly influenced by the 

 human factor in selecting his major units. 



Some geographers and many geologists have suggested 

 that land forms alone need be taken into account in deter- 

 mining these larger geographical units. Every different 

 recognisable land form is undoubtedly a geographical unit. 

 .•\ vast lowland, such as that which lies to the east of the 

 Rocky Mountains, is undoubtedly a geographical unit of 

 great importance, but its geographical subdivisions are not 

 necessarily orographical. The shores of the Gulf of Mexico 

 could not be considered as geographically similar to those 

 of the Arctic Ocean, even if they were morphologically 

 homologous. The lowlands of the polar regions are verv 

 different from those at or near the tropics. The rhythm 

 of their life is different, and this difference is revealed in 

 the differences of vegetation. 



I wish to lay great stress on the significance of 

 vegetation to the geographer for the purposes of 

 regional classification. I do not wish to employ a bio- 

 logical terminology nor to raise false analogies between 

 the individual organism and the larger units of which it 

 is a part, but I think we should do well to consider what 

 may be called the life or movement going on in our units 

 as well as their form. We must consider the seasonal 

 changes of its atmospheric and of its water movements, 

 as well as the parts of the Earth's crust which they move 

 over and even slightly modify. For this purpose a study 

 of climatic regions is as necessary as a study of morpho- 

 logical regions, and the best guides to the climatic regions 

 are the vegetation ones. 



By vegetation I mean not the flora, the historically 

 related elements, but the vegetable coating, the space- 

 related elements. Vegetation in this sense is a geo- 

 graphical phenomenon of fundamental importance. It 

 indicates quality — quality of atmosphere and quality of 

 soil. It is a visible synthesis of the climatic and edaphic 

 elements. Hence the vast lowlands of relatively uniform 

 land features are properly divided into regions according 

 to vegetation — tundra, pine forest, deciduous forest, warm 

 evergreen forest, steppe, and scrub. Such differences of 

 vegetatien are full of significance even in mountainous 

 areas. 



The search after geographical unity — after general 

 features common to recognisable divisions of the Earth's 

 surface, the analysis of these, their classification into 

 types, the comparisons between different examples of the 

 types — seem to me among the first duties of a geographer. 



