54 CAKNEGIE INSTITUTION 



should be remote from powerful eleclrical or mechanical plants, 

 and should be surrounded by sutiicient land under its control to 

 secure the institution against the encroachment of such sources of 

 disturbance for the future. 



It is hardly possible that the present rapid extension of electrical 

 tramways to place such a laboratarj- permanent!}^ beyond the reach 

 of all electrical influence without removing to a point so inacces- 

 sible as seriously to inconvenience the daily life of those engaged in 

 the work. Fortunately, however, the number of problems requiring 

 conditions of extreme freedom from electrical disturbance is rather 

 limited, and these have been left by common consent to small lab- 

 oratories far removed from the centers of population, and equipped 

 solely for this work. The extreme requirement in this direction 

 may, therefore, be regarded as superfluous in the present plan. 



The laborator}' may be safely located within one thousand feet of 

 a well insulated, underground trolly, or a double overhead system: 

 ordinary overhead feeds, with return through surface conductors or 

 through the rails, will cause a considerable magnetic disturbance at 

 a distance of a mile with the passage of e^'ery car. Cars eqnipped 

 with accumulators, electrical carriages, or systems of electrical 

 lighting using direct current, exert little influence at a distance. 



Nearly all the finer physical measurements are 'seriously disturbed 

 by the jar of passing traffic, and by temperature changes. It is, 

 therefore, most important that the laboratory building bt. reasonably 

 remote from paved thoroughfares, from foundries and plants where 

 heav}'- manufacturing is done, and that it be so constructed as to 

 protect it, as far as may be, from temperature variations 



This applies as well with respect to the sources of electrical and 

 mechanical power for the laboratory iiself, and to its machine shop, 

 dynamos, air compressers, refrigerating machinery, to rooms for 

 carrying on such special researches, themselves as require powerful 

 machinery, steam pressure, gas furnaces, or anything by which the 

 observatory building or other work in progress there, coiild be dis- 

 turbed. These would, therefore, be best located in the general 

 power house of the institution at some distance (200 feet or more) 

 from the main laboratory. 



There are two methods of attacking the problem of constant tem- 

 perature, both of Avhich are necessary to a successful result. The 

 one involves the circulation of artificially coded or heated air with 

 the help of suitable regulating devices, the other that the building 

 be constructed in such a way that outside temperature changes pro- 



