362 



SCIENCE. 



[N. S. Vol. XXIII. No. 584. 



Henry Cxat Weeks : ' The Practical Side of 

 Mosquito Extermination.' (Presented by A. C. 

 Eustis.) 



James Caeeoll : ' Without Mosquitoes there 

 can be no Yellow Fever.' 



H. A. Veazie : ' .^stivo-autumnal Fever — Cause, 

 Diagnosis, Treatment and Destruction of Mos- 

 quitoes which spread the Disease.' 



Discussions of the various papers by 

 Messrs. James Carroll, G. N. Calkins, L. 0. 

 Howard, J. H. White, S. E. Chaille, A. C. 

 Abbott, A. L. Metz, H. B. Ward, Quitman 

 Kohnke, A. C. Eustis and others. 



Wm. J. GiEs, 

 Secretary. 



THE EXPERIMENTAL METHOD IN SANI- 

 TARY SCIENCE AND SANITARY 

 ADMINISTRATIONS 



The value of experimentation in all 

 branches of inquiry is now generally recog- 

 nized ; its philosophical significance and its 

 limitations are less often understood and 

 appreciated. In the present paper I pro- 

 pose to show that there is no hard-and-fast 

 line between the results of experiment and 

 those of experience, and that in the field of 

 sanitation, in which of necessity laboratory 

 experiment is difficult when not impossible, 

 the data of experience carefully studied 

 and rigidly verified are capable of yielding 

 results no less valuable than those derived 

 in some other subjects by experiment. 



To avoid confusion let us understand 

 clearly at the outset exactly what we mean 

 by the terms experiment, experience and 

 observation. Originally signifying much 

 the same thing as experience, namely, both 

 the processes and the results of trial or test, 

 the word experiment has in recent times 

 come to be used chiefly for a prearrange- 

 ment, and an artificial arrangement, of con- 

 ditions in such a way that specific questions 

 shall be answered; while the term experi- 

 ence has come to be used most often to 



'Address of the vice-president and chairman of 

 Section K, American Association for the Advance- 

 ment of Science, New Orleans, December, 1905. 



describe a result rather than a process— a 

 result, moreover, obtained without prear- 

 rangement, and under natural or unpre- 

 meditated, rather than artificial or pre- 

 meditated, conditions. We speak, for ex- 

 ample, of an experiment upon the strength 

 of materials and an experience with a 

 broken rail, or railway bridge. The term 

 oiservation requires no special comment. 

 Without this, both scientific experiments 

 and scientific experience are worthless. 

 Verification is, of course, the checking, con- 

 trolling or testing of the data and results 

 of experiments or experiences, in order to 

 determine their accuracy or degree of 

 truth; and we are still indebted to George 

 Henry Lewes for his striking exposition of 

 the fundamental and indispensable func- 

 tion of verification in all sound and scien- 

 tific inquiry. 



Now a little reflection will show that 

 experiment, in the somewhat narrow sense 

 here laid down, is necessarily comparative- 

 ly limited, and very likely to be confined 

 largely to the laboratory, for the reason 

 that any prearrangement of conditions 

 which shall be actual tests or trials of nat- 

 ural phenomena must oftenest be made 

 under cover, in limited space, with com- 

 paratively few objects, and at comparatively 

 small cost. Hence it has come to pass that 

 the experimental sciences have been chiefly 

 those which could be advantageously pur- 

 sued in laboratories and workshops, and 

 upon the small, rather than the large, scale. 

 There can be little doubt that this fact has 

 tended to magnify the importance of sub- 

 jects or parts of subjects lending themselves 

 readily to laboratory experimentation, and 

 has served to draw off attention from, and 

 thus to hinder, the development of other 

 sciences, or portions of sciences, quite as 

 important as those capable of advancement 

 by laboratory experimentation. In phys- 

 ics, for example, we may well suspect that 

 materials at hand, such as air or water. 



