130 THE FIELD OF EXPERIMENTAL KESEARCH. 



somewhat oUiptical and the ends of their axes in the line of their dis- 

 charge being cohered I'ed and purple respectively. This peculiar 

 appearance, not at any time dazzlingly bright, persisted for a few 

 seconds, while fading gradually. 



Again, our knowledge of the aurora is not as Aet much more 

 definite or precise than it is in regard to the obscure forms of light- 

 ning alluded to above. Whether these phenomena will ever be ))rought 

 within the field of research by experimental methods is an open 

 question. 



The endeavor in the foregoing rather disconnected statements has 

 been to indicate directions in which the field of experiment may be 

 extended and to enjphasize the fact that research nuist be carried on 

 by extension of limits, necessitating more lil)eral endowment of 

 research laboratories. 1 have tried to make it clear that the physicist 

 must avail himself of the powers and energies set in phiy in the larger 

 industrial enterprises, and finally that the field of possible ex})l()ration 

 in ph3'sics b}' experinu'ntal methods has its natural boundaries, out- 

 side of which our advances in knowledge must be derived from a 

 study of celestial bodi(>s. 



The riddle of gravitation is yet to be solved. This all-permeating 

 force must be connected with other forces and other properti(>s of 

 matter. It will l)e a delicate task, ind(>ed, for the total attraction 

 between very large masses closely adjacent, aside from the earth's 

 attraction, is very small. 



Scientific facts are of little valiu; in themselves. Their significance 

 is their bearing upon other facts, ena))ling us to generalize and so to 

 discover principles, just as the accurate measurement of the position 

 of a star ma}' be without value in itself, but in relation to other 

 similar measurements of other stars m'dy become the nunms of dis- 

 covering their proper motions. We refine our instruments; we ren- 

 der more trustworthy our means of observation; we extend our range 

 of experimental inquiry, and thus lay the foundation for the future 

 work, with the full knowledge that, although our researches can not 

 extend beyond certain limits, the field itself is, even within those 

 limits, inexhaustible. 



