treatment, the upper fold still contains an abundance of granular conjunctiva, while the other parts of the lids are, for all practical purposes, restored to a healthy condition. This has an important bearing upon relapses, which originate in the majority of cases, from the upper fold. Again, the condition of the cornea is closely dependent on that of the superior retro-tarsal fold. Active pannus, for example, is almost always accompanied by a congested and thickened fornix; I may say, indeed, that, in my experience,

FIG. 2.-Excision of the upper cul-de-sac. The last step in the operation-viz., cutting through the posterior layer of the cul-de-sac.

corneal changes are seldom met with when the condition of the upper cul-de-sac is really satisfactory. Lastly, I may add my conviction that an excision limited to the superior fold is capable of favourably influencing the disease in other parts of the conjunctiva, and that local remedies, useless before operation, often act well after its performance.

At this point the simple method of removing the upper cul-de-sac adopted in all my cases may be briefly described.

The upper lid is everted and the fold brought well into view, which may be readily accomplished in most instances by directing the patient to look down, and at the same time pressing the eyeball upwards. In cases of special difficulty, the fornix may be seized and drawn out by means of toothed forceps. The next step consists in passing through each end of the fold a strong silk suture, the free ends of which are given to an assistant, with instructions to keep the cul-de-sac everted and "on the stretch" during the subsequent proceedings. As shown by Fig. 1, the upper lid is meanwhile kept in position by the operator's thumb. With bluntpointed scissors an incision is then made along the line of attachment of the fornix to the tarsal conjunctiva, but this incision should include nothing beyond the anterior layer of the fold. As soon as the latter is freed from its attachments, a dissection is made into the sub-conjunctival tissue, and continued as far backwards as the surgeon deems necessary. The dissection will be materially assisted by an intelligent handling of the threads attached to the fornix. During this stage the insertion of the levator muscle will lie well away from the scissors, and is not likely to be damaged if ordinary care be exercised. The operation is completed by cutting transversely through the posterior layer of the cul-de-sac, which thus comes away bodily, together with the attached sutures (Fig. 2). During removal of the fornix, bleeding is often free, and it may be necessary to twist one or two small vessels. In my own practice, I do not employ sutures to close the wound, although some other operators-Czermak,257 for example-recommend their use.

When the palpebral apertures are narrow, there need be no hesitation in dividing the tissues at the outer angle of the eye before removing the cul-de-sac. Under these circumstances, it is well to secure a permanent widening of the fissure by the insertion of three sutures applied in a manner to be described later (Canthoplasty). should be noted in passing that this plan was recommended by Richet in 1874, and that many surgeons have since adopted the suggestion.

It

When both eyes are operated upon at one sitting, the lids of

the side first finished should be kept covered during the performance of the second excision. If this precaution be neglected, troublesome bleeding is liable to occur, and clots of large size may collect in the conjunctival sac.

The question of anesthesia is a matter of practical importance. My own plan is to render the surface of the conjunctiva insensitive by a few drops of a 2 per cent. to 4 per cent. solution of hydrochlorate of cocaine, repeated twice at intervals of three or four minutes. Then an injection of three or four drops of the same liquid is made into the tissues of the cul-de-sac by pushing the needle in deeply, and pressing out the solution drop by drop during its withdrawal. Five minutes later the operation may be performed practically without pain. In a few instances only have I been compelled to administer a general anesthetic.

Antiseptic precautions must, of course, be adopted, both before and after excision of the cul-de-sac. Thus, before the cocaine is used, the conjunctiva should be cleansed with a tepid sublimate solution (1 to 5,000), and special attention should be devoted to the roots of the eyelashes, which are likely to harbour infective germs. The dressings after operation consist of a piece of old linen shaped to fit the eye; after being sterilised by boiling it is smeared with iodoform-vaseline (iodoform, 1; vaseline, 8), and covered with a disc of alembroth wool. A firmly applied roller bandage will complete the dressing.

In the after-treatment of the operation, the less done in the way of interference the better. Under ordinary circumstances, I do not, for my own part, inspect the wound for four or five days.

Certain immediate sequels may be expected after removal of the cul-de-sac. For example, ecchymosis of the lids, hæmorrhages in the conjunctiva, and a greyish film over the operation-site are constantly observed; while, more rarely, the lids become markedly swollen, or chemosis is seen. In none of my cases was there any rise in body temperature, or complaint of pain. The average length of time between operation and complete healing of the wound was nine to ten days.

Two complications call for particular notice-viz., ptosis and the growth of wound-granulations. With regard to the first, it is necessary to distinguish carefully between temporary and permanent ptosis resulting from excision of the upper cul-de-sac. The temporary and common form is, no doubt, due to general swelling of the parts. It possesses no particular significance, and soon disappears. Permanent ptosis, on the other hand, follows excision of the fornix only when the operator has damaged the tarsal insertion of the levator palpebræ muscle. As already pointed out, the chance of such a mishap will be small if the operation be carried out according to the above directions. In this connection the fact must not be lost sight of that almost every case of advanced trachoma, apart from any question of surgical interference, is accompanied by more or less drooping of the upper lid. Hence, unless careful measurements have been taken, it becomes a difficult thing to say that ptosis in a particular instance has become more marked after operation. From a series of observations, I am convinced that a properly performed excision does not permanently increase existing ptosis, nor does it give rise to that condition. The second complication-the growth of wound granulations is by no means uncommon after removal of the upper retro-tarsal fold. It was observed in about one-third of my own cases. The growths may attain a large size, and often bear a close resemblance to a cock's comb, both as regards their shape and colour. They are exceedingly vascular, and bleed at the slightest touch. It has been asserted that granulations appear only when antiseptic precautions have failed, but I cannot accept that view. As to treatment, it is useless to apply caustics to the growths. They should be snipped away, as often as may be necessary, with curved scissors.

Attention may now be drawn to some of the special advantages of the method of removing the cul-de-sac that has been described in the foregoing pages. First of all, the operation is as easy of performance as it is rapid in execution. So far from any special instruments being needed, a pair of scissors, a needle, and some silk, complete the surgical outfit. Then, if the fold be removed in

exact accordance with my directions, the levator tendon runs little, if any, risk of damage. In other words, the operation is not likely to be followed by permanent ptosis-a point of obvious practical importance. Lastly, there is the fact that my method permits one to remove just as much or as little conjunctiva as the nature of the case requires.

With regard to the last point, generally speaking, an oblong piece of the fornix, measuring, say, 28 mm. by 9 mm., is taken away, but there are cases in which a smaller or larger excision may be performed with benefit. The trachomatous cul-de-sac, at a certain stage in the disease, is of course larger than it would be in a state of health. It is therefore likely (as insisted on by Walther and other authors) that the piece removed from a diseased fornix is at least twice as large as would be the case under ordinary circumstances.

I have not infrequently observed obstinate lacrymation relieved, if not actually cured, by excision of the superior cul-de-sac. This is, doubtless, related to the fact that the palpebral gland is wholly or partly removed by the operation.*

If we next enquire what class of case is most likely to be benefited by the operation, we shall give the foremost place to those instances where the upper sinus of the conjunctiva is stuffed, as it were, with a large accumulation of granular material. Next in order comes trachoma, in which ordinary treatment by escharotics has proved unavailing. In both conditions, excision will often render yeoman's aid to the surgeon. So far as I know, there is only one state of the lids that contraindicates the operation, namely, when the conjunctiva is extensively scarred, and the fornices almost obliterated by the disease. It is quite conceivable that were excision attempted under these circumstances, the result might be disastrous.

They

The advocates of the operation claim excellent results. affirm that the time taken up in the treatment of trachoma is con

This body forms, of course, the palpebral portion of the lachrymal gland. It varies in size in different persons, but may be always recognised as a lobulated mass, lying at the outer extremity of the upper fornix. Complete exposure of the latter structure is, however, essential to the demonstration.