504 BREAD 



46 oxaminations of ashes furnished him by Dr. Gilbert, and treating them by the 

 above process, ho (Dr. Odling) obtained, to use his own words, 'in 21 instances, the 

 celebrated white precipitate said to be indicative of alumina and alum, so that had 

 these samples been in a manufactured instead of the natural state had the wheat, 

 for example, been made into flour I should have been justified, according to the 

 authority quoted, in pronouncing it to be adulterated with alum. But a subsequent 

 examination of the precipitates I obtained, showed that in reality they were not duo 

 'to alumina at all. M. Kuhlmann's process, as above described, is possessed of rare 

 merits: it will never fail in detecting alumina when present, and will often succeed 

 in detecting it when absent also. The idea of weighing this olla podrida of a preci- 

 pitate, and from its weight calculating the amount of alum present, as is gravely re- 

 commended by great anti-adulteration adepts, is too preposterous to require a moment's 

 refutation.' 



In order, however, to render the process for the detection of alum in bread free from 

 objections, the following method is recommended. It requires only ordinary care, and 

 it is perfectly accurate : 



Cut the loaf in half; take a thick slice of crumb from the middle, carefully 

 trimming the edges so as to remove the crust, or hardened outside, and weigh off 

 1,500 or 3,000 grains of it; crumble it to powder, or cut it into slices, and expose 

 them, on a sheet of platinum foil turned up at the edges, to a low red heat, until 

 fumes are no longer evolved, and the whole is reduced to charcoal, which will re- 

 quire from twenty to forty minutes, according to the quantity ; transfer the charcoal 

 to a mortar, and reduce it to fine powder; put now this finely-pulverised charcoal 

 back again on the platinum foil tray, and leave it exposed thereon to a dark cherry- 

 red heat until reduced to grey ashes, for which purpose gas-furnace lamps will bo 

 found very convenient. Only a cherry-red heat should be applied, because at a 

 higher temperature the ashes might fuse, and the incineration be thus retarded. 

 Remove the source of heat, drench the grey ashes with a concentrated solution of 

 nitrate of ammonia, and carefully reapply the heat ; the last portions of charcoal will 

 thereby be burnt, and the ashes will then have a white or drab colour. Drench them 

 on the tray with moderately strong and pure hydrochloric acid, and after one or two 

 minutes' standing, wash the contents of the platinum foil tray, with distilled water, 

 into a porcelain dish ; evaporate to perfect dryness, in order to render the silica 

 insoluble ; drench the perfectly dry residue with strong and pure muriatic acid, and, 

 after standing for five or six minutes, dilute the whole with water, and boil ; while 

 boiling, add carefully as much carbonate of soda as is necessary nearly, but not quite, 

 to saturate the acid, so that the liquor may still be acid ; add as much pure alcohol- 

 potash as is necessary to render it strongly alkaline ; boil the whole for about three 

 or four minutes, and filter. If now, after slightly supersaturating the strongly alka- 

 line filtrate with pure muriatic acid, the further addition of a solution of carbonate of 

 ammonia produces, either at once or after heating it for a few minutes, a light, white, 

 flocculent precipitate, it is a sign of the presence of alumina, the identity of which is 

 confirmed by collecting it on a filter, putting a small portion of it on a platinum hook, 

 or on charcoal, heating it thereon, moistening the little mass with nitrate of cobalt, 

 and again strongly heating it before the blowpipe ; when if, without fusing, it assumes 

 a beautiful blue colour, the presence of alumina is corroborated. If the operator pos- 

 sesses a silver capsule, he will do well to use it instead of a porcelain one for boiling 

 the mass with pure caustic alcohol-potash, in order to avoid all chance of any silica 

 (from the glaze) becoming dissolved by the potash, and afterwards simulating the 

 presence of alumina, though, if the boiling be not protracted, a porcelain capsule is 

 quite available. It is, however, absolutely necessary that he should use potasse a 

 I'alcohol, for ordinary caustic potash always contains some, and occasionally consider- 

 able quantities of alumina, and is totally unsuited for such an investigation. Even 

 potasse a I 'alcohol retains traces of silica, either alone or combined with alumina ; so 

 that for this, and other reasons which will be explained presently, an extravagant 

 quantity of it should not be used. 



Lastly, carbonate of ammonia is preferable to caustic ammonia for precipitating the 

 alumina, since that earth is far from being insoluble in caustic ammonia. 



The liquor from which the alumina has boon separated should now be acidified with 

 hydrochloric acid, and tested with chloride of barium, which should then yield a 

 copious precipitate of sulphate of barytes. 



The only precipitate which can, under the circumstances of the experiment, simulate 

 alumina, is the phosphate of that earth, which behaves with all reagents as pure alu- 

 mina. Such a precipitate, therefore, if taken account of as pure alumina, would alto- 

 gether vitiate a quantitative analysis if the amount of alum were calculated from it ; 

 but the proof that a certain quantity of alum had been used in the bread from which 

 it had been obtained would remain unshaken ; since alumina, whether in that stato 



