WORK DONE IN SPEAKING 411 



If zvork (})h()nati()ii only) is taken as ('(jiial to the produet of 

 the pressure and vohnne of air expelled, i.e. W — VH, we may 

 make a rough assessment of the amount of work done. During 

 ordinary eonversation, a man with a tracheal cannula developed 

 an air pressure of 200 mm. (HoO) and expired 300 litres of air 

 per hour, i.e. 



300 X 200 

 VH = — — = 60 kilogram-metres = 0-14 Cals. per hour. 



Speaking in a large hall the same man expired 1,440 litres of air 

 per hour and developed a mean pressure of 700 inm. (H2O). 



W = 1,440 X 0-7 = 1,008 kilogram-metres = 2-36 Cals. per hour. 



One may arrive at an estimate of the work of speaking by 

 measuring the oxygen used during rest and during speech. 

 Delivering an oration at the rate of 150 syllables a minute caused 

 the consumption of 28-78 litres of oxygen per hour. Subtracting 

 from this the amount used during a similar period of rest, viz. 

 16-96 litres, we find that 11-82 litres were used by the orator. In 

 large calories this amounts to 11-82 X 4-9 = 57-9 Cals. per hour. 



The value just obtained is about 25 times as great as that of 

 mere phonation. We must remember that the orator (and the 

 subject of the experiment was a Frenchman too) uses many 

 additional muscles in gesticulation, etc. A speech of an hour's 

 duration may tax his entire powers. To add to his troubles 

 he had to make himself heard in a large reverberating hall. 



Experiments have been carried out to determine the most 

 economical way of using the voice so as to obtain distinct enuncia- 

 tion and carrying power. During ordinary conversation in a 

 quiet room the oscillatory pressure (R.M.S.) in the sound wave 

 at 1 foot from the mouth appears to be about 1 dyne per cm. 2. 

 This value may be decreased or increased by about ten times 

 without appreciably decreasing distinctness. At the lower energy 

 level some of the consonants are difficult to differentiate. We 

 have seen that the main energy of the voice is, speaking broadly, 

 of low^ pitch. If we filter off the 60 per cent, or so of vibration of 

 lower frequency, the distinctness and carrying power of the voice 

 is not impaired, but if we filter off the 5 per cent, of the higher 

 frequencies we reduce the audibility of articulation by about 

 25 per cent. This applies especially to the dental consonants — the 

 " hiss " sounds having components of very high frequency, even 

 above 4,000 d.v. per second. 



The rate of emission of sound energv for ordinarv conversational 

 speech is about 125 ergs per second ; for ])ublic speech about 



