664 GEOGRAPHY 



by British Government Departments in illustration of their reports 

 are rarely more than diagrams delimiting the areas dealt with, but 

 not depicting the distributions. This is the more regrettable because 

 the accuracy and completeness of the statistics in the reports are 

 inferior to none, and superior to most work of a similar character 

 in other countries. As frequently happens, private enterprise has 

 stepped in where official action is wanting, and it is a pleasure to the 

 geographer to turn to the recent maps of Mr. J. G. Bartholomew, 

 especially the volume of his great Physical Atlas, the Atlas of Scot- 

 land published some years ago, and the Atlas of England and Wales, 

 which has just left the press. Both of the latter works contain gen- 

 eral maps based on statistics that have not been subjected to carto- 

 graphic treatment before, and attention may be drawn in particular 

 to the singularly effective and suggestive mapping of density of 

 population. Another work similar in scope, and no less creditable 

 to its compilers, is the Atlas of Finland, prepared by the active 

 and enlightened Geographical Society of Helsingfors. In Germany, 

 France, and Russia, also, examples may be found of good work of 

 this kind, sufficient to whet the desire for the complete and system- 

 atic treatment of each country on the same lines. 



It seems' to me that the most useful application of youthful 

 enthusiasm in geography, such as breaks forth in the doctorial theses 

 of German universities, and is solicited in the programme of the 

 Research Department of the Royal Geographical Society, would be 

 towards the detailed comparison of the distribution of the various 

 conditions dealt with statistically in Government Reports with the 

 topographical map of selected areas. The work would, of course, 

 not stop with the maps, for these, when completed, should be tested 

 and revised as fully as possible on the ground, since geography, be 

 the scale large or small, is not advanced by maps alone. 



Such small portions of the coordination of existing surveys are, at 

 the best, no more than fragments of a complete scheme, but they 

 show what can be done with existing surveys and actual statistics, 

 and indicate where these may be appropriately reinforced by new 

 work. I have treated a special case of this kind pretty fully, in papers 

 to which it is only necessary to refer. 1 One section of the scheme 

 outlined and exemplified in these papers is the distribution of rain- 

 fall viewed in relation to the configuration of the land; and with 

 the active assistance of nearly four thousand observers in the British 

 Isles, I feel that there is some prospect, though it may lie far in the 

 future, of ultimate results from that study. 



The system of botanical surveys now being carried on with signal 

 success in many countries is in some ways even more interesting. It 

 includes the mapping of plant associations and the discussion of their 



1 Geographical Journal, vii (1896), 345-364; xv (1900), 205-226, 353-377. 



