748 



SCIENCE 



[N. S. Vol. XXXII. No. 830 



it can be divided. For these we have not 

 as yet even names. It may seem absurd 

 that there should be this want of terms in a 

 subject which is associated in the minds of 

 most people with a superfluity of names. 

 I have elsewhere suggested the use of the 

 terms major natural region, natural re- 

 gion, district and locality to represent dif- 

 ferent grades of geographical units, and 

 have also attempted to map the seventy or 

 eighty major natural regions into which 

 the earth's surface is divided, and to 

 classify them into about twenty types. 

 These tentative divisions will necessarily 

 become more accurate as research proceeds, 

 and the minor natural regions into which 

 each major natural region should be di- 

 vided will be definitely recognized, de- 

 scribed and classified. Before this can be 

 done, however, the study of geomorphology 

 and of plant formations must be carried 

 far beyond the present limits. 



At the opposite end of the scale, that is, 

 in the geographical study of localities, good 

 work is beginning to be done. Dr. H. R. 

 Mill, one of the pioneers of geography in 

 this country and one of my most distin- 

 guished predecessors in this chair, has 

 given us in his study of southwest Sussex 

 an admirable example of a geographical 

 monograph proper, which takes into ac- 

 count the whole of the geographical factors 

 involved. He has employed quantitative 

 methods as far as these could be applied, 

 and in doing so has made a great step in 

 advance. Quantitative determinations are 

 at least as essential in geographical re- 

 search as the consideration of the time 

 factor. 



The geomorphologist and the sociologist 

 have also busied themselves with particu- 

 lar aspects of selected localities. Professor 

 W. M. Davis, of Harvard, has published 

 geomorphological monographs which are 

 invaluable as models of what such work 



should be. In a number of cases he has 

 passed beyond mere morphology and has 

 called attention to the organic responses 

 associated with each land form. Some of 

 the monographs published under the super- 

 vision of the late Professor Ratzel, of Leip- 

 zig, bring out very clearly the relation be- 

 tween organic and inorganic distributions, 

 and some of the monographs of the Le Play 

 school incidentally do the same. 



At present there is a double need. Re- 

 search may take the form, in the first 

 place, of collecting new information, or, 

 in the second place, of working up the ma- 

 terial which is continually being accumu- 

 lated. 



The first task — that of collecting new in- 

 formation — is no small one. In many cases 

 it must be undertaken on a scale that can 

 be financed only by governments. The 

 ordnance and geological surveys of our 

 own and other countries are examples of 

 government departments carrying on this 

 work. We need more of them. We need 

 urgently a hydrographical department, 

 which would cooperate with Dr. Mill's 

 rainfall organization. It would be one of 

 the taslvs of this department to extend and 

 coordinate the observations on river and 

 lake discharge, which are so important 

 from an economic or health point of view 

 that various public bodies have had to 

 make such investigations for the drainage 

 areas which they control. Such research 

 work as that done by Dr. Strahan for the 

 Exe and Medway would be of the greatest 

 value to such a department, which ought 

 to prepare a map showing all existing 

 water rights, public and private. 



We shall see how serious the absence of 

 such a department is if we consider how 

 our water supply is limited, and how much 

 of it is not used to the best advantage. We 

 must know its average quantity and the ex- 

 treme variations of supply. We must also 



