Constant heating of platinum causes a slight crystallization of the surface which, if not removed, penetrates into the crucible.  Gentle polishing of the surface destroys the crystalline structure and prevents further damage.  If sea sand is used for this purpose, great care is necessary to keep it from the desk, since beakers are easily scratched by it, and subsequently crack on heating.

Platinum crucibles stained in use may often be cleaned by the fusion in them of potassium or sodium acid sulphate, or by heating with ammonium chloride.  If the former is used, care should be taken not to heat so strongly as to expel all of the sulphuric acid, since the normal sulphates sometimes expand so rapidly on cooling as to split the crucible.  The fused material should be poured out, while hot, on to a !dry! tile or iron surface.

IGNITION OF PRECIPITATES

Most precipitates may, if proper precautions are taken, be ignited without previous drying.  If, however, such precipitates can be dried without loss of time to the analyst (as, for example, over night), it is well to submit them to this process.  It should, nevertheless, be remembered that a partially dried precipitate often requires more care during ignition than a thoroughly moist one.

The details of the ignition of precipitates vary so much with the character of the precipitate, its moisture content, and temperature to which it is to be heated, that these details will be given under the various procedures which follow.

DETERMINATION OF CHLORINE IN SODIUM CHLORIDE

!Method A. With the Use of a Gooch Filter!

Procedure.—­Carefully clean a weighing-tube containing the sodium chloride, handling it as little as possible with the moist fingers, and weigh it accurately to 0.0001 gram, recording the weight at once in the notebook (see Appendix).  Hold the tube over the top of a beaker (200-300 cc.), and cautiously remove the stopper, noting carefully that no particles fall from it, or from the tube, elsewhere than into the beaker.  Pour out a small portion of the chloride, replace the stopper, and determine by approximate weighing how much has been removed.  Continue this procedure until 0.25-0.30 gram has been taken from the tube, then weigh accurately and record the weight beneath the first in the notebook.  The difference of the two weights represents the weight of the chloride taken for analysis.  Again weigh a second portion of 0.25-0.30 gram into a second beaker of the same size as the first.  The beakers should be plainly marked to correspond with the entries in the notebook.  Dissolve each portion of the chloride in 150 cc. of distilled water and add about ten drops of dilute nitric acid (sp. gr. 1.20) (Note 2).  Calculate the volume of silver nitrate solution required to effect complete precipitation in each case, and