This appears to me to be certain. There are three definite conditions all grouped together under this name. There is partial rupture of the tendon of the supraspinatus, there is injury (probably partial rupture) of the subscapularis, and there is a true sub-deltoid bursitis.

The latter is probably encountered whenever symptoms follow a "toss" on to the shoulder. Treatment should be by radiant heat baths, followed by exercises as already indicated for training a weak deltoid. The old bone-setter's advice could not be bettered.

It should be possible to distinguish between this condition and those due to injury to the subscapularis and supraspinatus. Creaking will probably be felt on palpation during active or passive movements of the shoulder in all these cases; but the situation of the pain on active contraction of the injured muscles should sufficiently indicate the nature of the injury when these are affected. Treatment should follow similar lines, but the addition of graduated faradic contraction to the supraspinatus should render recovery complete.

Muscle training of the rotators should also prove a valuable auxiliary in treatment. The fact that callus may be thrown out at the insertion of a muscle should warn us not to attempt to "push" treatment, but to regard scrupulously the laws drawn up for the treatment of all recent injury.

One reflection is of interest. It would seem probable that the pain and disability which follow fracture of the greater tuberosity of the humerus are largely due to the intensity of the sub-deltoid bursitis, which is almost inevitably haemorrhagic. This would account for the otherwise inexplicable increase to be noted in both symptoms, when comparing a case of uncomplicated fracture of the surgical neck of the humerus with one in which similar injury is complicated by separation of the greater tuberosity.

The above expression of opinion was written on the boat during my return home. Since then I have taken every opportunity of testing my theories, with the following results.

A dressmaker and a gymnast both came to hospital complaining of the sudden onset of acute "lumbago," the first while working in a cramped position, the second while doing gymnastic work. Neither gave any history of previous trouble. Attempts to secure rotation of the ilium effected rapid and complete relief in both cases. I use the word "attempts," because I could not detect any movement of the ilium on the sacrum at the time of manipulation.

I have fitted a large number of belts of the Goldthwait pattern to patients complaining of "muscular rheumatism," of "lumbago," and of "sciatica" - all of long standing, and all of whom had received every variety of treatment without avail. Manipulation in these chronic cases has so far failed me. The result has been varied, as was to be expected from the heterogeneous type of material. In a vast majority of cases relief has been marked, and several patients have returned to work - two to laborious work, I believe - after months and even years of incapacity. In two cases at least no benefit has been noted, and others are as yet sub judice. I have seen enough, however, to assert confidently that sacro-iliac strain does exist and that relief can be afforded by the fitting of proper support. This should, as a rule, tend to draw the two anterior superior spines together and to press backwards towards the sacrum that on the painful side.

I have attempted manipulation of the cervical spine with almost dramatic success in two cases of "brachial neuritis," with marked success in two cases of "migraine," and with varied success in several other cases. Complete failure has also to be recorded; but so far, in the cases I have treated, amelioration has been the rule. Both "migraine" cases were of interest. One was an officer with apparently typical attacks at frequent intervals, each attack ending in intense headache for two days. When I last heard of him the attacks had ceased. The other patient was hit by a bullet in May, 1915, which broke his jaw, travelled round and came out over the root of the neck. From that time to December, 1919, the patient had been completely incapacitated owing to the fact that on any sudden movement or on stooping he suffered intense pain (apparently in his great occipital nerve) which "bowled him over," faint and nauseated. He is now (April, 1920) about to return to work as a French polisher.

I have attempted manipulation of the thoracic spine for patients with chronic cough in three cases. One was a "dead failure." The second, an elderly woman of sixty-seven, had suffered from intense spasms of coughing for "five or six years." I gave a very complete examination, testing thoroughly all movements in the spine, and perhaps this amounted to manipulation. To my intense surprise I received a letter some time later saying: "I have had no recurrence of my terrible paroxysms of coughing, for which I am most thankful and grateful." I am sorry to say the patient is apparently relapsing, but I have had no chance of attempting a second manipulation, as she lives too far away. This case tempted me to try to help a young girl who is training for operatic singing. Her mistress was "in despair," so the student said. After two manipulations she was "able to sing better than for months," and her mistress sent most grateful messages. I have not the least doubt that this percentage of success will not be maintained; but the evidence is probably enough to indicate that the claims made to relieve chronic spasmodic cough by manipulation are not to be put aside too lightly. I did all in my power to exclude all psychical effect, but it is impossible to say how far this has operated.

I have only seen one case which I think was typical of "subdeltoid bursitis." I built up a very wasted supraspinatus with graduated faradic contraction and then manipulated once under gas. Very great improvement is reported, and I am hoping to hear of complete cure.

On the whole, four months' experience has tended to confirm the accuracy of most of the conclusions expressed above. At least this much is certain: an attempt should be made to relieve by manipulation, "lumbago" and "sciatic pains" which start suddenly as the result of cramped position or strain. A typical sciatica and lumbago of long standing should be given treatment by support, and many cases will be relieved. Neck manipulation in obscure cases of pain in the head, neck and upper extremity is worthy of a trial.