BAKING POWDERS. 597 



into the aspirator. The increase in weight of the potash bulbs gave 

 the weight of water absorbed. It was found, however, that the sample 

 would cake into a hard mass, through which a channel would form 

 which would permit the passage of the current of dry air, without af- 

 fecting the greater mass of the powder, and no exact results could 

 be obtained. Some improvement was made by mixing the powder 

 with dry oxide of zinc, so as to prevent the formation of a channel, 

 but still the results were not at all satisfactory, and the attempt to 

 make a direct estimation was finally abandoned. Even if the determi- 

 nation could be made exact, it is doubtful if all the water of combina- 

 tion could be obtained at 100 C., especially in phosphate and alum 

 powders, and probably a temperature high enough to accomplish this 

 would effect a decomposition of the starch. 



EXPRESSION OF THE RESULTS OF ANALYSIS. 



The results of analysis are given, first, as acid and basic radicals in 

 percentage composition while in the second part of each table an attempt 

 has been made to combine these into salts showing the constitution of 

 the powder. The difficulties attending this calculation of the pro- 

 bable combination of the acids and bases were so great that I was 

 frequently tempted to give it up entirely and state only percentage com- 

 position. I finally concluded to insert the calculation with the proviso 

 that it should be considered at best merely an approximation to the 

 exact composition ot the ponder. The obstacles in the way of an exact 

 calculation may be stated as follows : In the first place the amount of 

 total carbonic acid found is always less than that required to form bi- 

 carbonate of soda with the amount of sodium oxide found. This is 

 undoubtedly due to a partial action of the acid constituent of the pow- 

 der upon the bicarbonate, by which carbonic acid has been lost. The 

 percentage of bicarbonate is therefore calculated from the per cent, 

 of carbonic acid found, and the excess of sodium oxide left is stated as 

 " residual" sodium oxide, without attempting to make further hypotheses 

 as to the results of its combination with the acid constituent. It is 

 possible that part of the bicarbonate may become converted into the 

 normal carbonate under the conditions of being mixed and in contact 

 with other chemicals, though this is not likely ; then the indefinite 

 composition of many of the commercial salts used in the powders ren- 

 ders it an extremely difficult matter to arrive at any satisfactory con- 

 clusion as to the make up of the powder in which they are used. The 

 acid phosphate of lime, for instance, is a very variable substance, and 

 even ammonia alum, which might reasonably be supposed to be constant 

 in its composition, is found to vary widely from the theoretical. Its 

 content of water varies according to the greater or less amount of dry- 

 ing it has undergone, and aside from this the ratios of the ammonia, sul- 

 phuric acid, and alumina to one another are at variance with the formula. 



