The suture has a special use in so-called artificial strabismus; that is, in those cases where convergent is converted into divergent squint through unskilful treatment, or where tenotomy of the abducens, performed on account of "insufficiency of the internal recti," is followed by convergent strabismus. I have not found confirmation of the fear expressed by Arlt, that the method proposed by me could be scarcely practicable if it is a case of the advancement of a muscle too far forward, and I have corrected a large number of such cases in other practices. It is seldom profitable to take up things in which others have been unsuccessful, but it bring its own reward in the case of artificial squint. Periodic convergent squint offers a less certain ground for the operation. The change between normal position and a very considerable squint gives rise to the fear that an operation which would be able to remove the convergence might finally induce divergent strabismus. This fear is certainly not groundless, but at the same time it must be remembered that, with the exception perhaps of a few cases of clearly accommodative deviation, elastic preponderance of the internal recti or insufficiency of the external recti is generally the cause of periodic squint also. I have frequently, in periodic squint, performed double tenotomy of the internal recti with the slightest possible loosening of the conjunctiva. I have also attempted to confine the operation to the shortening of the external rectus without loosening the internal recti and with success, but not frequently enough to be able to deliver a certain opinion upon it. In periodic squint, the first care must always be to determine the condition of refraction, if possible with atropine, and to neutralise or over-correct hypermetropia if present. If squint is absent during the use of convex glasses, which happens frequently under these circumstances, the operation offers no further advantages, as the constant use of convex glasses afterwards can hardly be avoided. If the periodic deviation continues to exist, the operation can be carried out according to the above rules and so as to cause a slight effect. The final result is usually attained after two to three weeks in convergent squint; it is better to allow a slight degree of convergence to exist, as divergence, however slight, existing at this time, brings with it the fear of a gradual increase. It happens occasionally, that after years, convergence asserts itself again; I have observed it in spontaneous (see Case 39) as well as in operative cure of squint; still, this is so unusual, that I should like to give an illustration of the latter observation on account of its rarity.