5. _Absence of ocular changes._ After division of the sensory root ocular changes are absent, due in part to the fact that the ocular nerves are left intact, and also to the probability that ‘sympathetic fibres pass to the eye after entering the trigeminal nerve through the Gasserian ganglion’ (Spiller). In the consideration of these advantages it should be noted that they are mainly directed against procedures in which complete extirpation of the ganglion is attempted, and are of less weight when contrasted with the partial excision advocated by Hutchinson. =Frazier’s operation[76] on the sensory root.= ‘A horse-shoe shaped incision is made, beginning at the middle of the zygoma and terminating a little below the apex of the lobe of the ear (see Fig. 91). The musculo-cutaneous flap is made purposely a little larger than the opening in the skull. An opening is made in the skull with a diameter not exceeding 3 cm. and extending down as far as the infra-temporal crest. The dura is separated from the base of the skull with a blunt retractor as far as the foramen spinosum. The middle meningeal artery is ligatured and the dura mater incised directly over the third division at the margin of the foramen ovale and separated from the upper surface of the ganglion as far back as the sensory root. The latter is picked up with a blunt hook, grasped with forceps and divided or avulsed. It is necessary only to expose the posterior portion of the ganglion; the second and third divisions of the fifth, the structures on the inner side of the ganglion, and the cavernous sinus, are left undisturbed.’ =Choice of operation.= There are both advantages and disadvantages to each of the various operations enumerated. There are also other operations. From my own experience I have come to the conclusion that the two operations of choice are resection of the second and third divisions of the fifth nerve and division of the sensory root. The former of these two operations is carried out when the general condition of the patient is not altogether satisfactory, presuming, of course, that the neuralgia is limited to those divisions of the nerve. Under more favourable circumstances excellent results are obtained by exposure and division or resection of the sensory root, with, I think, the minimum risk. =The results obtained by operation.= The prevailing impression that these operations are associated with a high mortality is totally incorrect. From my own experience and from numerous recorded cases, the mortality may be estimated at not more than 5 per cent.