408 OBSERVATIONS IN CHEMISTRY AND MINERALOGY 



the cobalt reaction. The forming of the bluish and greenish crystals has an in- 

 fluence upon the reaction only in so far as their mass lessens the concentration 

 of the cobaltic molecules and causes different types of praseosalt crystals to form 



A blue ammoniacal solution of a cuprisalt of high concentration throws 

 out deeply blue crystals of cupraminosalt ; the addition of persulf ate causes sul- 

 fate of prismatic monoclinic type, while the nitrate is cubo-octahedral. Neither 

 seems to interfere with the praseo reaction. 



(d) In an ammoniacal solution of a manganosalt, ammoniumpersulfate gives 

 black-brown peroxyhydroxyd. This can interfere by obscuring the field. 



f 



(e) Zinc and cadmium salts do not interfere. 



Execution of the Method. — Concentration of the test liquid is very desirable, 

 since the praseosalt is quite perceptibly soluble, but also because the green 

 crystals will form at once. Let the precipitate, obtained by ammoniumsulfid 

 in ammoniacal solution in presence of sal ammoniac, be the material for the test. 

 We remove some of the paste with a steel spatula and dry it on the spatula over 

 the flame in a few minutes. A small part of the dry material is then taken up with 

 a borax bead and is roasted in the oxidizing flame, just within the point, so that 

 the borax remains solid. One or two minutes will suffice for the oxidation, 

 whereupon substance and bead are melted together. The color of the cooling 

 bead will then tell whether nickel so predominates that the blue of the cobalt- 

 glass is not visible. I do not advise oxidation of substance with nitric acid, 

 making ammoniacal and then adding alcoholic solution of dimethylglycoxime, 

 which gives a beautiful red precipitate with nickel-salts, because this reaction 

 is too delicate. The action with the bead is in such a case both quicker and more 

 decisive whether the praseo reaction for cobalt should be made or not. Let 

 there be placed a quantity of dry precipitate not exceeding 5 milligrams in a 

 clean porcelain crucible of the smallest size, add one drop of concentrated nitric 

 acid and one drop of concentrated hydrochloric acid and evaporate to dryness, 

 at waterbath temperature. The best shape of crucible is that of a cone. Hence 

 I prefer to use a small conical cup, which I make from a large test-tube in the 

 way familiar to glass-blowers. To the evaporated substance add a small quan- 

 tity (1 or 2 mg.) of ammonium chlorid and then, with a fine pipette, one tenth 

 of a cubic centimeter, i. e., two drops, of strong aqua ammonia. If a precipitae 

 of ferrihydroxyd or manganihydroxyd forms the two drops of liquid must be 

 filtered without dilution. This can be done with a tiny glass funnel, whose 

 stem is capillary. Such a one can readily be made by drawing out a thin glass 

 tube and then widening one end. A tiny plug of filter-paper is placed m tnis 

 enlarged end, which is then immersed in the liquid. By applying suction a 

 the other end with the lips the entire liquid can be drawn into the capillary. 

 After removing the plug, the widened end of the tube is taken between the lips a^ 

 by forcing air into the tube a drop is easily transferred to the center of a clea ^ nch 

 slide. When the margin of the drop has been focused under a one-ha I ^ 

 objective, a fragment of a crystal of ammonium persulfate is place 



