As soon as the visible impurities are removed and the tube when washed looks bright and clean, it may be wiped on the outside and held perpendicularly so as to allow the water film to drain down. If the tube be greasy (and perhaps in other cases) it will be observed that as the film gets thinner the water begins to break away and leave dry spots. For accurate work this grease, or whatever it is, must be removed; and after trying many plans for many years, I have come back to the method I first employed, viz. boiling out with aqua regia. For this purpose, close one end of the tube by a cork (better than a rubber bung, because cheaper), and half fill the tube with aqua regia; then, having noted the greasy places, proceed to boil the liquid in contact with the glass at these points, and in the case of very obstinate dirt--such as lingers round a fused joint which has been made between undusted tubes--leave the whole affair for twelve hours. If the greasiness is only slight, then simply shaking with hot aqua regia will often remove it, and the aqua regia is conveniently heated in this case by the addition of a little strong sulphuric acid. The spent aqua regia may be put into a bottle. It is generally quite good enough for the purpose of washing glass vessels with sand, as above explained. However carefully a tube is cleaned before being subjected to blowpipe operations, it will be fouled wherever there is an opening during the process of heating, unless the extreme tip only of an oxidising flame be employed. Even this should not be trusted too implicitly unless an oxygas or hydrogen flame is employed. When a tube or piece of apparatus has been cleaned by acid, so that on clamping it vertically, dry spaces do not appear, it may be rinsed with platinum distilled water and left to drain, the dust being, of course, kept out by placing a bit of paper round the top. For accurate work water thus prepared is to be preferred to anything else. When the glass is very clean interference colours will be noticed as the water dries away. Carefully-purified alcohol may in some cases be employed where it is desired to dry the tube or apparatus quickly. In this case an alcohol wash bottle should be used, and a little alcohol squirted into the top of the tube all round the circumference. The water film drags the alcohol after it, and by waiting a few minutes and then adding a few more drops of alcohol, the water may be practically entirely removed, especially if a bit of filter paper be held against the lower end of the tube. It is customary in some laboratories to use ether for a final rinse, but unless the ether is freshly distilled and very pure, it leaves a distinct organic residue.