substance must be held lower than the other end, and must be moved a little over the spirit-lamp before a draught of air is produced through the tube.  It is a good plan to have a number of these tubes on hand.  After having used a tube we cut off that end of it which contained the substance, with a file, and clean it from the sublimate, either by heating it over the spirit-lamp, or with a piece of paper wound around a wire.  It sometimes happens that the substance falls out of the tube before it becomes sufficiently melted to adhere to the glass.  To obviate this, we bend the tube not far from the end, at an obtuse angle, and place the substance in the angle, whereby the tube may be lowered as much as necessary.  Fig. 9 will give the student a comprehension of the processes described, and of the manner of bending the tubes.

[Illustration:  Fig. 9.]

(i.) Glass Tubes closed at one End.—­If we wish to expose volatile substances to heat, with the exclusion of air as much as possible, or to ascertain the contents of water, or other volatile fluids, or for the purpose of heating substances which will decrepitate, we use glass tubes closed at one end.  These tubes must be about one-eighth of an inch wide, and from two to three inches in length.  They should be made of white glass, difficult of fusion, and free from lead.  They should be closed at one end, as figured in the margin, Fig. 10.

[Illustration:  Fig. 10.]

When a substance is to be examined for the purpose of ascertaining whether it contains combustible matter, as sulphur or arsenic, and where we wish to avoid oxidation, we use these tubes without extending the closed end, in order that there may be as little air admitted as possible, as is represented in tube B. But when a substance to be examined is to be tested for water, or other incombustible volatile matters, we employ tubes with little bulbs blown at one end, such as represented at tube A. Here there is room for a circulation of air at the bottom of the tube, by which the volatile matter rises more easily.  In some cases, it is necessary to draw the closed end out to a fine point, as in the tubes C and D. Either one or the other of these tubes is employed, depending upon the nature of the substance used.  The sublimates condense at the upper part of the tube a, and can be there examined and recognized.  These tubes, before being used, must be thoroughly dried and cleaned.  In experimenting with them, they should not be exposed at once to the hottest part of the flame, but should be submitted to the heat gradually.  If the substance is of such a nature that it will sublime at a low heat, the tube should be held more horizontal, while a higher heat is attained by bringing the tube to a more vertical position.

VARIOUS APPARATUS NECESSARY.