,"> Fig. 0, page -0. Plan* in a porcelain or platinum crucible that, luu* been ignited to constant weight, incline the crucible and burn lit*' paper. When the precipitate is white plaee tin* crucible in an upright position, rover nnd heat with f IK* full flame of the Mast lamp or of the No. -t Meker burner until, after cooling for 15 minutes in the desiccator, the weight w eonstant. If the former burner w used tin* first weighing may be made after 15 minutes of heating. If the Me.kcr burner is used the precipitate .should lie heated for 30 minutes before the first weighing. From tin* weight of calcium oxide found e'alculiitc the per cent of cfdcium in the sample, Volumetric, by Titration with a Standard Solution of Potas- sium Permanganate.-- Instead of igniting calcium oxalate and weighing tin* oxide* t!u* purified oxalutr may 1«* n*(Iissolv<*(J in sulphuric arid and the* resultingoxalir acid t.iiratc»4 + 2KMn<>4 + :UI2S(>4.....^ K2S(>4 -f- 2MnS( )4 f XH2<) •( lOCOy. (2) Although this ronstitut.es a direct tit ration of oxalic arid it in indirect, so far as calcium is concerned, and the calculation of the latter curt be accurate only (1) if the precipitation has recovered all of the calcium, (2) if the calcium oxalafe has l«*en well purified and (Mj if the oxalic acid resulting front its deeom- position by sulpliuric acid has been recover***! ;« tem oflVi'H greater convenience, of course. If the M>lutiuu IM to IH« U8i»d also for the determination of iron and poHHibly of ot!n*r elementH, then the normal system offers udvantnges. In either cane it is necenHary to know the equivalent weights of ealrium and permanganate mid thene cunnot be cnlcniluted ILH simply HH in cases already coriHiderecL