.no 



Horns of a 

 paraboloid form 

 difufed. 



Cylindrical 

 acouftic tubes 

 tranfmlt found 

 without re- 

 fleSion to a 

 ^re^c diilance* 



According to 

 the theory of 

 reflexion the 

 enlarged part 

 (pavillion) is of 

 no }»fe. 



SPEAKING TRUMPETS, 



would become larger than a riglit angle, and the ray would re« 

 turn upon itfelf. Thus the greateft number of the rays would 

 infue out of the moulh, after being feveral times refle61ed, and 

 the quantity of the rays which thus reach the ear, and which 

 are proportionalby lefs numerous as the cone is longer, cannot 

 fenfibiy augment the intenfity of the found. Neverthelefs, 

 acouftic horns do ftrengthen the found conliderably; this aug- 

 mentation muft thereiore depend on another caufe. 



It has been propofed to make horns of a paraboloidal form, as 

 being beft calculated to tranfmit found by refiedion; but whe- 

 ther they did not produce the effe<5l which was expe6ted from 

 them, or tiiat they were too difficult to conftrud, thefe horns 

 have been abandoned, and none but conic horns are made 

 ufe of. 



The acouftic tubes employed to tranfmit the found to great 

 diftances, fo that it cannot be heard by perfons placed between 

 the two extreme points, are ufually formed of cylinders, which 

 ferve to conduft the found. Lambert had already ftated that 

 refle6tion did not increafe the intenfity of found in cylinders, 

 becaufe whatever might be the direction of the incident rays, 

 being conftantly refleded between two parallels, the angle 

 which they form with the axis of the cylinder muft be the fame 

 at entering and quitting it; that thus the found muft experi- 

 ence a diminution at its iflue, which will be fo much the greater 

 as the number of refledions is more confiderable. Never- 

 thelefs the found is tranfmitted to a great diftance by means of 

 thefe tubes, and I have afcertained that the beating of my 

 watch, which ceafes to be audible at the diftance of 1"*. 10, 

 the medium of feven experiments (A), is heard at the dif- 

 tance of 2™ 23, the mean diftance of feven experiments (J) 

 when I place it in the mouth of a pafteboard tube, 0"". 038 in 

 diameter and 0™. 6 in length ; whence it follows that this tranf- 

 xniflion muft proceed from another caufe. 



All thefe confi derations have therefore induced me to ex- 

 amine the effedls of the fpeaking trumpet with the greateft 

 care. It follows from the theory of the refle£lion of found, 

 applied by Lambert to the fpeaking trumpet, that the opening 

 by which it is ufually terminated, ftiould be at leaft ufelefs, if 

 not injurious, for it has no influence on the concentration of 

 the reflefted rays; and therefore this gentleman propofed to 

 lay it a^gle entirely. The ^njilogy of form betv^eeu the trum- 

 pet 



