3. That sutures should be inserted to keep the edges in apposition, yet not so tightly as to cause ulceration. They may be either of metal, silver being preferable, or of fine silk well waxed. The metallic sutures are now generally preferred. Some dexterity is required in their introduction, and various instruments have been devised; the best seems to be a needle with a short curve fixed on a long handle, which should be entered on the (patient's) left side of the fissure in front, and brought out on the right side. If silk sutures be used, the chief difficulty, that of passing the thread through the second side from behind forwards, can be avoided in the following manner.[123] A curved needle is passed through one side of the fissure, and then towards the middle line, till its point is seen through the cleft. One of the ends of the thread is then seized by a long pair of forceps, and drawn through the cleft; the needle is then withdrawn, leaving the thread through the palate, and both ends are brought outside at the angle of the mouth. Another needle is then passed through a corresponding point at the opposite side of the palate, till its point again appears at the cleft; this time a double loop of the thread is also brought out through the cleft by the forceps into the mouth. If then the single thread of the first ligature which is in the cleft be passed through the loop of the second one also in the cleft, it is easy, by withdrawing the loop through the palate, to finish the stitch (see Fig. XXIX.). All the stitches should be passed and their position approved before any one be tied, and it is most convenient to secure them from above downwards. To prevent confusion, each pair of threads after being inserted should be left very long, and brought up to a coronet fixed on the brow, which is fitted with several pairs of hooks numbered for easy reference. This will prevent twisting of the threads or any mistake in tying. FISSURE OF THE HARD PALATE.--This may vary in extent from a very slight cleft in the middle line behind, up to a complete separation of the two halves of the jaw, including even the alveolar process in front, and sometimes complicated with harelip. To close such fissures by operation is difficult, as the breadth of the cleft is so great as to prevent the apposition of the edges when prepared, without such extreme tension as quite prevents any hope of union. Through the researches of Avery, Warren, Langenbeck, and others, a method has been discovered of closing such fissures by operation, which, though certainly not easy, is, when properly performed, generally successful.