1897-98.] Mr W. W. Taylor on Sodium Mdlitate. 
253 
II. — 0*5723 gm. substance gave 0*3142 gm. h7a 2 S0 4 = 1 7*80 per 
cent. Ea. 
Calculated for 
•N 0 6^'12^12 j 17H 2 0. 
Na 17*74 
H 2 0 39 24 
Found. 
I. II. 
17*61 17*80 
38*30 
The formula given in Beilstein’s Handbuch der Organisclien 
Chemie , on the authority of Erdmann and Marchand,* isNa 6 C 12 0 12 , 
18H 2 0. On reference to their paper, I found that the loss on 
heating to 160° C. is stated to he 38*88 per cent., which closely 
agrees with the formula given above. As at that time the formula 
assigned to mellitic acid was C 4 H 2 0 4 , the 38*88 per cent, loss was 
nearer 6H 2 0 than to 5II 2 0. 
Erdmann and Marchand gave no sodium estimation. 
The salt dissolves in water without any indication of hydro- 
lysis ; the most dilute solutions that were used were neutral to 
phenolphthalein. 
The arrangement of the apparatus was a modification of that 
used by Abegg.f 
The glass cylinder of 200 c.c. capacity was supported by cork 
wedges in a metal air-chamber, the depth of which was much 
greater than that of the glass. 
The metal cylinder was fixed in the wooden lid of a large 
porcelain cylinder, which, in turn, was surrounded by a large 
wooden box packed with waste. The freezing mixture of pounded 
ice, salt, and water was placed in the porcelain cylinder, and the 
temperature remained constant to within 0*1° C. during a series 
of determinations lasting several hours. The stirrer consisted of 
a platinum disc provided with two stout platinum wires fused 
into glass tubes, which moved vertically in brass slides. The 
length of stroke of the stirrer was easily adjusted to the depth of 
liquid by means of a movable eccentric on the driving pulley. 
The stirrer was worked by an electric motor, and the rate was 
kept constant at 32 strokes a minute throughout all the experi- 
ments. 
* Liebig’s Annalen , 68 (1848), 327. 
+ Zeit. fur phys. Chem. xx. (1896), p. 207. 
