ESTIMATION OF ASH. 81 



of the ove)i. To prevent loss of heat the oven may be lagged 

 with felt, asbestos, or kieselguhr. 



The most efficient condenser has been found to be a spiral coil 

 of tubiiiK (preferably copper), which fits rather closely into a 

 tube. The cold water enters at the top, and passes by a straight 

 portion of the tubing to the lowest coil, whence it circulates 

 upwards, and finds an exit at the top. 



A diafrrammatif figure of the bath will assist compieheusion 

 of the details (Fig. 5). 



A ver\- convenient water-bath for milk analysis is that devised 

 by Vietli ; the chief advantage of this lies in the lid, which, 

 instead of merely having holes made in it in which the basin.'^ 

 fit, has a copper ring fastened into each. This enables the 

 basins to bo taken ofi without contact between the fiiiiifr.s and 

 the hot lid. This bath may be heated by a steam coil or a gas 

 fianie, and is conveniently supplied with wator by a constant 

 level apparatus. 



If di'y steam is available it is very (;(jnvcnient to use steam 

 coils to heat tlir bath, and to condense the steam after it has 

 passed the coils. The condensed steam can be used as distilled 

 water, and is usually pure enough for all purposes. It is liable, 

 however, to contain traces of copper, if this metal is em- 

 ployed in the construction of the coils, condenser, etc. : if 

 the boilers prime to any extent, it will also contain impurities 

 (salts. scali'-))rov('nting composition, I'tc.), derived from the 

 boiler. 



Estimation of Ash. — The residue of total solids serves excel- 

 lently for the determination of the ash. By igniting over a 

 small Bunsen flame, or an .\rgaud burner, or in a muffle, a white 

 ash can lie obtained. The temperature must not be allowed 

 to rise above a barely jierceptible red heat, or distinct volatilisa- 

 tion of alkaline chlorides may occur. If the asbestos method 

 of total solid estimation has been used, a somewhat higher tem- 

 perature may be employed than if simple evaporation in platinum 

 lias been resulted to 



A more c^xact determination is obtained by evaporating a 

 larger (|uantity of milk than is usually taken for total sohd esti- 

 mation— '25 to •")() grammes — and gently igniting till thoroughly 

 charred : the nuiss is extracted with hot water and filtered, the 

 insoluble portion and the filter being (after washing) ignited at 

 a red heat till white ; this will give the insoluble ash. 



By e\aporating the filtrate and cautiously igniting at a low 

 temperature, the soluble ash is obtained. The sum of the soluble 

 and insolul)U' ash gives the total ash ; the results obtained in 

 this way are usually slightly higher (about 0'02 per cent.) than 

 the ash olitained bv i'.niition of the total solid residue. 



(! 



