4thly, Having arrived at the lower part of the knee, he pushes the lower fragment upwards, and makes below it two or three circular turns to secure it. He then gives the roller into the hands of an assistant, and directing him who holds the long compress roller, to draw it forcibly upwards, pushes the integuments of the knee in the same direction, lest, by becoming interposed between the fragments, they might prove an obstacle to their reunion. Passing then the fingers of his left hand through the holes in the compress-roller he places them behind the superior fragment and pushes it forcibly downwards.

5thly, When the reunion of the fragments is exact, without any space intervening, he resumes the roller, and passing it obliquely under the ham, and bringing it up again behind the superior fragment, withdraws his fingers which held this fragment down. In place of his fingers, he then applies two or three tight circular casts, covers the knee with several oblique casts in form of the figure of 8, so as to leave no opening between them, and, then, continues the bandage up along the thigh, securing by it the compress-roller extended along the fore part of the limb.

6thly, When he has arrived at the upper part of the limb, the assistant who holds the compress-roller, drawing it forcibly upwards, doubles down its end over the circular casts. The surgeon next fixing this end by several additional casts, descends again along the thigh, covers the knee by a few more oblique turns, and finishes with the roller on the leg.

33. This first part of the bandage evidently fulfils the third and fourth indications (18). The compression of the roller on the muscles weakening their action and impeding their motions prevents their tendency to draw the superior fragment upwards: while the circular casts passed behind this fragment, acting in opposition to the muscular contractions, prevents it from moving upwards in obedience to them. The long compress-roller, stretched on the fore part of the limb, being first secured below, and then drawn forcibly upwards, presses the casts of the roller against each other, and prevents those that correspond to the thigh from slipping upwards, and thus abandoning the superior fragment, and prevents also those on the leg from slipping down and withdrawing their support from the inferior fragment. As there remains no vacant space between the circular turns, their pressure is uniform throughout: no swelling can consequently supervene (20).

34. But the first and second indications remain still to be fulfilled (18): it is necessary to prevent the separation of the lower fragment, by the extension of the leg on the thigh, and to throw the muscles into a state of relaxation by extending the thigh on the pelvis, and to maintain permanently, by the apparatus, that double position, which the assistant maintains only during the operation.

35. To obtain the first effect different means have been employed; but none answers so well, to extend the limb and retain it immoveably in that state, as a long and strong splint, placed, as Desault did it, subsequently to the application of the first part of the bandage, along the posterior part of the limb. An assistant must hold the end of this splint, while the surgeon secures it in its place by the second roller (31): in this way the extension of the leg is effected.

36. To obtain the extension of the thigh, it is necessary to place on the top of each other, two or three bolsters or little bags filled with chaff, so disposed as to form an inclined plain, considerably elevated towards the heel above the level of the bed, but which, gradually descending to the same level towards the tuberosity of the ischium, forms a supporting basis on which the whole limb may rest in a uniform manner. By this twofold extension of the leg and of the thigh, the lower fragment is kept up immoveably, and the muscles are kept in a state of relaxation.

Hence it follows, that this bandage fulfils extremely well the conditions laid down (18), and that it ought to be preferred to all the others (19...25), which answered the indications only in part.

37. Whatever may be the advantages of this bandage over the others, it must still be acknowledged to have its inconveniences. The rollers become relaxed in a short time; their compression is less active; the muscles, being less confined, contract more readily; hence the necessity of frequently repeating the application of the apparatus, a circumstance which is very troublesome, on account of the roller which composes it, and covers the whole limb. The resistance of it even when it is recently applied, is not always equal to the power of the muscles, whence the most assiduous attention is necessary, to obtain such a consolidation as to leave no trace of the fracture behind. Few persons ever possessed, like Desault, the art of overlooking nothing that might in any way contribute to the success of his treatment: from this, no less than from the excellence of his processes, arose the number of his cures. Let us confirm, by a few examples selected from among a great many, the doctrine here laid down. The following cases were collected by Julian and Bezard.

Case I. Francis Leclert, of a sanguine temperament, fell on the 7th of October, 1790, on his right knee, and produced a transverse fracture of the rotula. He was not able to rise; he was carried home, where a surgeon, on discovering the nature of his disease, advised him to be taken to the Hotel-Dieu.