THE SURVEY 479 



ripened judgment. It is one of the most difficult forms of 

 survey-work, if it is to have real value. It must be much more 

 than a car-window exercise. When properly undertaken, it is a 

 new and useful application of geography. There is a great dan- 

 ger that the overhead reconnaisance will be little more than prac- 

 tice in aviation. 



If a survey of any region or phase is to be a record of fact, 

 then it must be strictly scientific in spirit, as I have already in- 

 dicated. It must discover and set down every fact of signifi- 

 cance, wholly apart from any prejudice or bias in the mind of 

 the observer: the fact is its own justification. The work can- 

 not be as precise as that in the mathematical and physical 

 sciences; but in its purpose it must be as scientific as any work 

 in any subject. 



If the work is scientific, then it will not be undertaken for 

 the purpose of exploiting a movement, recruiting an associa- 

 tion, spreading a propaganda, advertising a region, sustaining a 

 political organization, or promoting the personal ambition of any 

 man. There is indication that survey work will soon become 

 popular ; there is danger that it will be taken up by institutions 

 that desire to keep themselves before the public and by locali- 

 ties and states that desire to display their advantages. It will 

 be easy to marshal statements and arrange figures, and par- 

 ticularly to omit facts, in such a way as to make a most attrac- 

 tive showing. Even some honest investigators will be likely to 

 arrange the material in such a way as to prove a point rather 

 than to state the facts, unless they are very much on their guard. 

 If country-life surveys have possibilities of great good, they have 

 equal possibilities of great danger. I am glad that the move- 

 ment is going slowly at first. 



The intention of survey work in agriculture is to make a rec- 

 ord of the entire situation and to tell the whole truth. Frag- 

 mentary surveys and piece-work, however good they may be in 

 themselves, do not represent the best effort in surveys. Prac- 

 tically all our surveys have thus far been fragmentary or unre- 

 lated, but this is the work of a beginning epoch. We shall al- 

 most necessarily be obliged to do- still further fractional and 

 detached work ; but it is time that we begin to train the imagina- 

 tion on completer and sounder programs. The whole basis and 



