June 17, 1910] 



SCIENCE 



935 



much to our regret that while the excur- 

 sion was in the district of the sub-Alpine 

 lakes, where the party had reached nearly 

 a dozen, no member could from conviction 

 present the arguments of the anti-glacial 

 erosionists. We did the best we could in 

 their absence, but found it impossible to 

 explain the over-steepened trough walls 

 and the numerous hanging lateral valleys 

 of most typical development without ac- 

 eei3ting a strong measure for glacial ero- 

 sion. After crossing into France, two pro- 

 fessors from the universities of Grenoble 

 and Fribourg presented their views against 

 wholesale glacial erosion during a visit to 

 the strongly glaciated valley of the Ro- 

 manche; but it seemed to most of us that 

 their discussion was incomplete and uncon- 

 vincing. 



What with the varietj^ of landscape that 

 we studied and with the variety of training 

 represented in our cosmopolitan party, it 

 will, I think, be agreed that our discussions 

 as to methods of describing land forms 

 must have been profitably extended by the 

 time the excursion closed in the volcanic 

 district of central France. Without giving 

 further account of our results, let me 

 next present certain underlying principles, 

 which appear to be of much importance 

 in this connection. 



DESCEIPTION IN TERMS OP TYPE POEMS 



Whenever an observer attempts to tell 

 what he has seen, so that a landscape or a 

 region may be conceived by his readers, he 

 must describe the observed forms in terms 

 of certain similar forms previously known 

 to him, and hopefully known also to those 

 for whom he writes. It must always be in 

 terms of something previously known that 

 a verbal description is phrased. Hence the 

 most accurate verbal description will be 

 made by that observer who is eqiiipped 

 with the largest variety of previously 



knoAvn type forms. It is important to 

 consider how a young geographer is to 

 obtain such an equipment. The ideally 

 perfect method would be for him to travel 

 about the world and see with his own eyes 

 a great variety of actual forms, from which 

 he might gradually develop a complete 

 series of type forms. Then all other forms 

 could afterwards be described in terms of 

 these types. But this method is manifestly 

 impossible to general application. Some 

 equipment of types may be secured by 

 observation of actual forms; and this be- 

 ginning may be significantly enlarged by 

 the study of dascriptions, pictures, models 

 and maps of actual forms, as prepared by 

 other observers. 



The geographer who follows the em- 

 pirical method stops here. The geographer 

 who follows the explanatory method goes 

 much farther. He extends and system- 

 atizes the equipment, thus far gained, by 

 deducing many related forms; and thus 

 fills his mind with a series of more or less 

 ideal forms. It will then be chiefly in 

 terms of the ideal types, largely developed 

 by deduction, familiarized by diagrams, 

 and confirmed or corrected by expei:ienee, 

 that his explanatory descriptions of actual 

 landscapes will be phrased. But whether 

 the geographer follow the empirical or the 

 rational method, it will be only in propor- 

 tion to the completeness with which his 

 series of ideal forms provides him with 

 counterparts of actual forms, that his de- 

 scriptions of actual landscapes can be true 

 to nature. Only in proportion to the com- 

 pactness of the terminology in which the 

 ideal forms are verbally expressed, can the 

 observer's descriptions be tersely stated. 

 Only in proportion to the correspondence 

 existing between the ideal forms as con- 

 ceived and named by the observer and by 

 his reader, will the reader be able to appre- 

 hend the observer's meaning. 



