10 
exposures. This frame was long enough so that when in position the 
wire trav would be entirelv within the room containing the formalde- 
hyde, the outer end of the frame being pushed just far enough into 
the tube through which it was introduced to close the outer open- 
ing of the tube with a tight-fitting rubber stopper. Some of these zinc 
tubes were also utilized for drawing air from the room in making the 
chemical determinations. 
DETERMINATION OF THE AMOUNT OF FORMALDEHYDE IN THE 
ATMOSPHERE OF THE ROOM, 
The quantity of formaldehyde in the atmosphere of the room during 
any particular experiment was determined by the potassium cyanide 
method of Romijn. This method depends upon the fact that formal- 
dehyde combines with potassium cyanide in the sense of this equation: 
/H 
KCX + HCHO = HC— OK 
\CN 
and that the cyanide thus combined can not be precipitated by means 
of silver nitrate. On the other hand the excess of potassium cyanide, 
over and above that required by the aldehyde, reacts with silver nitrate 
in the sense of this equation : 
KCX + AgXOg = AgCX + KX0 3 , 
and hence can be determined quantitatively by means of decinormal 
solutions of silver nitrate and potassium sulphocyanate. 
TTie potassium cyanide solution . — This is made by dissolving 3.3 gm. 
(of purity of 96 per cent or over) in water to the volume of 500 c. c. 
The solution is standardized against a decinormal solution of silver 
nitrate in the f ollowing manner : 
Into a flask graduated to 100 c. c. an excess of the silver nitrate solu- 
tion is drawn from a burette, say 12 or 13 c. c. ; to this are added 4 or 5 
drops of strong nitric acid (about 50 per cent), and then 10 c. c. of the 
cvanide solution from a burette. The flask is then shaken and filled 
up to the mark with water and the contents thoroughly mixed. The 
liquid is then filtered into a dry burette and 50 c. c. of the filtrate 
drawn into a beaker. To this is added some ferric ammonium alum 
indicator and the excess silver is titrated with a standard solution of 
potassium sulphocyanate in the usual way. The sulphocyanate solu- 
tion must also be titrated against the silver solution in the presence of 
nitric acid and iron alum, to establish its equivalence to the silver 
nitrate. The number of c. c. of sulphocyanate required is multiplied 
by 2 and this is converted into its equivalent of decinormal silver 
nitrate solution. The latter is the excess of silver solution and the dif- 
ference between it and the volume of silver solution originally taken 
