428 PHYSIOLOGICAL, REGULATIONS 



organism is integrated. The data of biology are for the use of all 

 travelers ; anyone may select materials, as I have, that belong on 

 certain quadrangles. New combinations of variables are made, 

 whereupon general features emerge. 



At first sight it might seem easy to use published data for a new 

 investigation. I have not found it so. Ideally, hard-won materials 

 might be rescued from the oblivion into which the hypotheses that 

 accompanied them have fallen. Rare indeed is the record of a 

 research that is complete enough to serve for other ends. Often, 

 too, it takes more time to set down the limitations of an unfamiliar 

 technique than to carry out new measurements. The data recorded 

 are already selected in view of one conclusion, and very commonly 

 the omission of one or two more correlatives defeats the new objec- 

 tive. So he who revels in details and relations can rarely be sup- 

 plied, to his exacting requirements, from sources other than his 

 own tests. 



The accomplishments of mapping are illustrated in another 

 science, meteorology. The prolonged study of many maps led to 

 the recognition of intricate combinations of quantities that now 

 together foretell tomorrow's weather. The first weather maps 

 foretold nothing, for it took years for Brandes to devise them, and 

 weeks for Galton to receive and record the information embodied in 

 them (Geddes, '30, pp. 7 and 8). The choice of variables was 

 arbitrary and depended on the recording instruments available. 

 Facility in constructing and in grasping the weather's characteris- 

 tics came with long practice and observation. Quantitative physi- 

 ology appears now to be clumsily discerning similarly elementary 

 combinations. 



The individual who knows how to divide and multiply numbers 

 can get along in many situations. But he who uses algebra can deal 

 with larger and more complicated relations, while he who uses cal- 

 culus is equipped to deal with multifarious and involved phe- 

 nomena. In the end the biologist chooses whether to ignore the 

 multiplicity of relations or to do whatever is now possible toward 

 understanding them. It has been suggested by others that descrip- 

 tion (correlation) becomes more important as the system studied 

 is more heterogeneous and complex ; I suppose nearly every vital 

 function is part of such a system. 



Already physiology has used quantitative description. The 

 multiple relationships in blood as modified during respiratory ex- 



