Whilst the assistant is staunching the hæmorrhage, the operator can introduce the raspatories through the openings thus made, and working them _from without inwards_, separate the whole of the muco-periosteal tissue. To effect this, different shapes of instruments will be required in order to follow the curves of the palatal segments, and those devised by Mr. Durham will be found most useful (Fig. 60). In loosening the flaps anteriorly, the advantage of the double-curved raspatory (Fig. 68) will be obvious. As the point of the raspatory reaches the inner free margin of the palatal segment, the separation of the muco-periosteal flap should be completed by the protrusion of the instrument into the cleft at the junction of the buccal and nasal mucous membranes. This is more readily accomplished if the edges have been previously pared; but it is better to postpone this step until the flaps have been detached, as the raw edges are less liable to be bruised by the sponging, and with the flaps loosened the margin can be pared with greater accuracy. In cases where the vomer is attached to one free edge of the palate (Fig. 11) the junction of the nasal and buccal mucous membrane should be incised to prevent its being lacerated by the raspatory. The attachment of the soft structures to the hamular process and back of the hard palate must be freely and fully divided. This is a most important and delicate part of the operation, and as the structures are here extremely thin, great care must be exercised. Should this separation be incomplete, the lateral incision cannot be carried down into the soft palate, and the flaps will not come into proper apposition. It may be attained by the use of a sharp cutting raspatory kept close to the bone, and as regards the hamular process, by a narrow probe-pointed bistoury, or a pair of curved scissors. The introduction of the left forefinger into the incision is of great assistance in effecting this with precision and thoroughness. [Illustration: FIG. 68.—Double-curved raspatories for detaching the anterior portion of the muco-periosteal flaps in uranoplasty.] After detachment the muco-periosteal flaps will often appear blanched or of a bluish-white colour as a result of the interference with the circulation, a fact which has been commented on by M. Trélat. The circulation, however, is soon re-established, and the normal colour returns in a few hours. When this proceeding has been satisfactorily accomplished on both sides, a temporary delay generally occurs for the assistant to arrest the hæmorrhage, and for the anæsthetist to get the patient more fully under control, so that the second most important stage may be conducted without any struggling. STAGE II.—_Paring the Edges of the Cleft._