144 



CHEMISTRY. 



gamboge, etc. Prussian bine fades in moist 

 air; much more rapidly in an atmosphere of 

 carbonic dioxide ; but is permanent in dry air. 

 Mixed with cadmium yellow, Prussian blue 

 gave a green which was very sensitive to light 

 if moisture was present, but was permanent in 

 dry air. Vermillion was shown to fade in dry 

 and moist air, also in an inert atmosphere like 

 carbon dioxide. With cadmium yellow an ox- 

 ide was formed which blackened in moist air 

 in a few hours, though in dry air light was 

 without action on it. The author condemns as 

 unsafe those pigments which fade in dry air, 

 and shows that the greater number of paints 

 are stable in sunlight, provided moisture is 

 absent. 



When petroleum is stored in lead -lined 

 tanks, the lead is rapidly corroded, with the 

 formation of a heavy, brownish-colored pow- 

 der. This powder has been found to consist 

 of a carbonate and hydrated oxide of lead and 

 a small quantity of valerate of lead ; the brown- 

 ish color is due to organic matter. The hy- 

 pothesis that the white lead, of which the pow- 

 der practically consists, and a paraffin, is formed 

 by the action of an oxidizing agent and a small 

 quantity of valeric acid present in the petroleum 

 on the lead, is supported by experiments made 

 by William Fox. 



H. Le Chatelier has found that hydrated 

 cements treated with a large excess of water 

 give up not only the lime present as hydrate, 

 but also, in time and after treatment with fresh 

 quantities of water, they surrender nearly all 

 the lime in combination. Slow-setting cements 

 contain much calcium hydrate ; quick-setting 

 cements, very little. 



Analytical Chemistry. In the analysis of milk 

 as recommended by the Association of Official 

 Agricultural Chemists, the butter is estimated 

 by drying on the water-bath for thirty minutes, 

 or by drying with powdered asbestos for two 

 hours at 100 0. For casein, the milk is 

 digested with HaSO-4, or the dried residue is 

 rubbed up and transferred to the soda-lime 

 combustion tube, or is transferred to a di- 

 gestion flask, and the casein estimated by the 

 method of Kjeldahl. For the estimation of 

 the fat, a strip of blotting or filtering paper is 

 saturated with a measured quantity of milk, 

 and dried, after which the fat is extracted from 

 it ; or the milk is dehydrated by means of an- 

 hydrous sulphate of copper; the fat is ex- 

 tracted by means of the low-boiling products 

 of petroleum ; the butter is saponified with 

 solution of potassium hydroxide in alcohol, 

 and the excess of the alkali is determined by 

 means of a solution of hydrochloric acid. In 

 Babcock's method for estimating water in fat, 

 the milk is placed in ignited asbestos, and sub- 

 jected, at 100 C., to a slow current of dry air 

 till the water is expelled. The tube containing 

 the solids from this operation is placed in an 

 extraction apparatus, and exhausted in the 

 usual way. In Prof. Macfarlane's method the 

 milk is absorbed in asbestos fiber in a tube, 



which is then exposed, to dry it, for ten or 

 twelve hours at a temperature of 90 C. The 

 tube is then exhausted with petroleum ether, 

 dried, cooled, and weighed. The loss repre- 

 sents the butter fat. For sugar, the milk, its 

 specific gravity having been determined, is 

 treated with mercuric nitrate or mercuric 

 iodide solution tor precipitation of albumen, 

 shaken, filtered, and subjected to polariscopic 

 examination. For the estimation of ash, the 

 milk, treated with nitric acid, is dried and 

 burned at a low red heat till the ash is free 

 from carbon. 



In the analysis of butter, a portion of the sam- 

 ple, taken from the inside of the mass, is placed 

 on a slide, treated with a drop of pure sweet- 

 oil, and examined with a microscope and with 

 polarized light and the seleuite plate. Pure 

 butter will show neither crystals nor a parti- 

 colored field with selenite, while other fats, 

 melted and cooled and mixed with butter, will 

 usually present crystals and variegated colors. 

 The specific gravity and the melting-point are 

 determined with apparatus prepared for that 

 purpose. Volatile and soluble acids are esti- 

 mated by processes requiring considerable 

 manipulation. The amount of water is ascer- 

 tained by heating at 105 C. for two hours in 

 a flat-bottomed platinum dish full of sand. 

 Salt is volumetrically ascertained by adding 

 hot water, waiting till the melted fat has all 

 collected on the top, and running the water, 

 without any of the fat, into an Erlenmayer 

 flask. The salt is also determined in the filtrate 

 by means of solution of silver nitrate. The 

 methods of estimating curd depend on the 

 principle of drying a weighed portion and 

 extracting the fat with ether or petroleum. 

 The residual mass is then weighed, and the 

 curd determined by loss or ignition. In Bab- 

 cock's method for the determination of casein, 

 dried butter is treated with light petroleum 

 till all fat is removed. The residue is then 

 ignited with soda-lime or treated by the Kjedahl 

 method. 



For the determination of traces of arsenic in 

 tissues, yarns, and paper-hangings, R. Fresenius 

 and E. Hintz digest the chopped tissue with 

 hydrochloric acid for one hour ; add solution of 

 ferrous chloride, and heat till the excess of hy- 

 drochloric acid has passed off, and then boil 

 till the distillation is stopped by frothing. More 

 than two thirds of the liquid in the retort could 

 generally be distilled over. A second distilla- 

 tion with hydrochloric acid is effected, and 

 the sulphureted hydrogen treatment is applied. 

 After elimination of organic matter, the pre- 

 cipitate is filtered, treated with bromo-bydro- 

 cbloric acid and ferrous chloride, and distilled. 

 Treatment with sulphureted hydrogen gives 

 arsenic trisulphide. 



The state of combination in which quicksil- 

 ver is dissolved in natural waters has been 

 studied by G. F. Becker in the course of his 

 investigations of the quicksilver deposits of the 

 Pacific slope. Pyrite or marcasite almost inva- 



