750 



SCIENCE 



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



mittee on which both the universities and 

 the ordnance survey were represented. 



The information which many other gov- 

 ernment departments are accumulating 

 would also become much more valuable if 

 it were discussed geographically. Much 

 excellent geographical work is done by the 

 admiralty and the war office. The meteor- 

 ological office collects statistics of the 

 weather conditions from a limited number 

 of stations; but its work is supplemented 

 by private societies which are not well 

 enough off to discuss the observations they 

 publish with the detail which these obser- 

 vations deserve. The board of agriculture 

 and fisheries has detailed statistical in- 

 formation as to crops and live stock for 

 the geographer to work up. From the 

 board of trade he would obtain industrial 

 and commercial data, and from the local 

 government board vital and other demo- 

 graphic statistics. At present most of the 

 information of these departments is only 

 published in statistical tables. 



Statistics are all very well, but they are 

 usually published in a tabular form, which 

 is the least intelligible of all. Statistics 

 should be mapped and not merely be set 

 out in columns of figures. Many dull blue 

 books would be more interesting and more 

 ■wddely used if their facts were properly 

 mapped. I say properly mapped because 

 most examples of so-called statistical maps 

 are merely crude diagrams and are often 

 actually misleading. It requires a knowl- 

 edge of geography in addition to an under- 

 standing of statistical methods to prepare 

 intelligible statistical maps. If Mr. Bosse 's 

 maps of population of England and Wales 

 in Bartholomew's survey atlas are com- 

 pared with ordinary ones the difference be- 

 tween a geographical map and a carto- 

 grapliie diagram will be easily appreciated. 



The coming census, and to a certain ex- 

 tent the census of production, and prob- 



ably the new land valuation, will give more 

 valuable raw material for geographical 

 treatment. If these are published merely 

 in tabular form they will not be studied 

 by any but a few experts. Give a geog- 

 rapher with a proper staff the task of 

 mapping them in a truly geographical 

 way and they wiU be eagerly ex- 

 amined even by the man in the street, 

 who can not fail to learn from them. The 

 presentation of the true state of the coun- 

 try in a clear, graphic and intelligible form 

 is a patriotic piece of work which the gov- 

 ernment should undertake. It would add 

 relatively little to the cost of the census 

 and it would infinitely increase its value. 



The double lack — the lacunse in the in- 

 formation and the absence of adequate 

 geographical treatment of such material as 

 there is — makes the task of studying the 

 huge natural divisions which we call con- 

 tinents a very difficult and unsatisfactory 

 one. For several years in Oxford we have 

 been trj'ing to gather together the material 

 available for the study of the continents 

 and to make as accurate maps as is possible 

 for geographical purposes. "We have 

 adopted uniform scales and methods, and 

 by using equal area projections we have 

 obtained comparative graphic representa- 

 tions of the facts. We hope before the end 

 of the year to issue maps of physical fea- 

 tures, vegetation and rainfall of each con- 

 tinent and other maps for the world. 

 These are being measured, and I hope will 

 yield more reliable quantitative informa- 

 tion about the world and its continents 

 than we possess at present. 



With such quantitative information and 

 with a fuller analysis of the major natural 

 regions it ought to be possible to go a 

 step further and to attempt to map the 

 economic value of different regions at the 

 present day. Such maps would necessarily 

 be only approximations at first. Out of 



