Dr. R. Konig on Manometric Flames. 17 



The bands of light in the mirror appeared under their influence 

 like an alternately darker and lighter ribbon with irregular small 

 teeth; and the whole was so cloudy and undefined that I could 

 even discover no difference between the different vowels. The 

 semivowels m and n gave such similar pictures that I could not 

 distinguish between them. I have sketched them for the notes 

 e y g, e, c (fig. 9, PL I.); deeper notes showed longer, but still 

 misty and undefined periods. Of course in these experiments I 

 was obliged to put my nose instead of my mouth to the instru- 

 ment. The quivering R, silently pronounced, shows a series of 

 flame-summits of different elevations pretty regularly forked or 

 toothed. In the small rotating mirror, with a plate 15 centims. 

 wide, of which I generally make use, these summits appeared to 

 me to follow each other irregularly. But when I employed a 

 larger one, 40 centims. wide, I perceived the regular periodicity 

 of the whole group, which was repeated four or five times in the 

 width of the mirror. The teeth, which are spread over the whole 

 of the flame-summits, are simply caused by the air- current. Of 

 this we can easily convince ourselves by placing the tongue at a 

 little distance from the gums instead of permitting it to vibrate 

 on them, and then expelling the air violently through the 

 narrow aperture. The band of flame then appears serrated, 

 without any individual flame-summits rising above it. But if 

 we intone the R, the picture of the note unites with that of the 

 letter, and there ensues such a confused series of single flames 

 and whole groups of dissimilar height and form, that in the eva- 

 nescence of the picture it is impossible to decipher them. I have 

 sought to give the character of the voiceless R in fig. 10, PL I. 



The different characteristics of the voiceless explosives P, T, and 

 K are easily recognized. At P the flame suddenly rises straight 

 up high above the average line, then shows two or three similar 

 elancements, which are followed by a few rapidly decreasing ones. 

 Both the high and low chief movements show as at R the inden- 

 tations caused by the air-current. 



The rise of the flame is less sudden at T, neither is it so high ; 

 and the deep incisions are wanting, which at P in the commence- 

 ment show two or three rapid elancements. At K, which is arti- 

 culated further back in the mouth, there is still less a sudden rise 

 of the flame; but the picture begins with a regular rising and 

 falling wave, followed by a few rapidly diminishing ones of the 

 same form. The indentation of the whole picture is the same 

 as in P and T. 



If we utter one of these consonants many times in succession 

 while continually turning the mirror, we rarely see the picture 

 well ; it is therefore better to place the mirror so that the flame 

 shows in one corner, and with a slight turn must pass over its 



Phil. Mag. S. 4. Vol. 45. No. 297. Jan. 1873. C 



