November 14, 1913] 



SCIENCE 



707 



ment will be apparent to all geologists and 

 mining engineers, and it is to be hoped that 

 similar cooperation on work relating to other 

 state problems will be effective in the near 

 future. F. W. DeWolf, 



Secretary 



SPECIAL ARTICLES 



ON THE ACOUSTIC EFFICIENCY OF A SOUNDING 

 BOARD 



The experiments described below appeared 

 to yield such a variety of information, of so 

 definite a character, that it seemed worth 

 while to record them, in spite of their simplic- 

 ity. 



The chapel of Adelbert College, built in 

 1910, had proved unsatisfactory in its acoustic 

 properties. The architect prescribed a sound- 

 ing board, as likely to remedy the defect, and 

 sent a sketch embodying his suggestion. It 

 was thought worth while to make a prelimi- 

 nary test before erecting a permanent sound- 

 ing board, and the writer was asked to take 

 charge of the matter. 



The chapel is a building of late English 

 Gothic type. The nave is 104 feet long, with 

 narrow and low side aisles, barely 6 feet wide, 

 including the massive piers. The width of 

 the nave, not including the aisles, is 30 feet. 

 The chancel is 34 feet long and 30 feet wide, 

 without aisles. The chancel floor is raised 

 about 16 inches above that of the nave. Thus 

 the general shape of the building is a long and 

 narrow rectangle, 140 feet by 30, with no im- 

 portant recesses or irregularities. The ceiling 

 is arched, about 48 feet high to the top of the 

 arch. Its curvature is such that any focal line 

 which might be formed by reflection would be 

 not near the floor, but high up in the audi- 

 torium. 



Experiments gave little evidence of local 

 echo or interference. The acoustic difficulties 

 arise chiefly from general reverberation. The 

 problem was then to determine by direct com- 

 parison the value of a sounding board as a 

 corrective of general reverberation. 



It is evident that the experiments must be 

 of such a kind as would appeal not merely to a 

 physicist, but to any intelligent person. 



This means that they must be comparable 

 with the ordinary use of the chapel, and must 

 involve the hearing of ordinary speech. Tet 

 it was of course desirable that they should 

 have some quantitative character, and that the 

 individual and personal characteristics of the 

 hearers should be so far as possible eliminated 

 or averaged. 



Several members of the college faculty and 

 two or three advanced students gave their cor- 

 dial assistance. To their patience and careful- 

 ness is due whatever of value these experi- 

 ments may have. 



Three speakers took part, differing greatly 

 in characteristics and in quality of voice, but 

 all accustomed to public speaking. 



It is a commonplace that ordinary speech is 

 understood largely by context and association 

 throughout a whole sentence rather than by 

 actual hearing of the individual words. To 

 eliminate this factor, lists of unconnected 

 words were read from a spelling book, at a 

 rate and with intonation similar to that used 

 in a connected passage. One who has not tried 

 this can hardly realize how much we rely on 

 association in listening to an address. In 

 order that this association-factor might not 

 be left entirely out of account, a passage from 

 some oration (always the same oration in any 

 one set of experiments) was read in addition 

 to the spelling-book list. 



Three rows of seats on the floor, and the 

 front row of the gallery at the back of the 

 house, were selected as representative of the 

 whole auditorium. The seats on the floor 

 were the seventh, fourteenth and twenty-first 

 from the front, and were called in the tests 

 G, N and U, respectively. The position of the 

 listener in any one row of seats, whether in 

 the middle or on either side of the chapel made 

 no apparent difference in the ease of hearing. 

 The speaker was equally well heard from any 

 part of the row, whether he stood in the pul- 

 pit, or in the middle of the front edge of the 

 chancel floor. These facts were established 

 by experiment before the sounding board was 

 put in place. 



The sounding board, made after the design 

 of the architect, was of the horizontal type 



