We treated 183 patients with fractures of the odontoid process (109 type II, 74 type III) non-operatively. Union was achieved in 59 (54%) with type-II fractures. All type-III fractures united, but in 16 patients union was delayed. There was no correlation between union and the clinical or radiological outcome of the fractures. Selective vertebral angiography, carried out in 18 patients ten with acute fractures and eight with nonunion, showed that the blood supply to the odontoid process was not disrupted. Studies on ten adult axis vertebrae at post-mortem showed that the difference in the surface area between type-II and type-III fractures was statistically significant. Our findings show that an age of more than 40 years, anterior displacement of more than 4 mm, posterior displacement and late presentation contribute towards nonunion of type-II fractures.

We reviewed 67 consecutive patients with fractures of the coracoid process, classifying them by the relationship between the fracture site and the coracoclavicular ligament. The 53 type-I fractures were behind the attachment of this ligament, and the 11 type-II fractures were anterior to it. The relationship of three fractures was uncertain. Type-I fractures were associated with a wide variety of shoulder injuries and consequent dissociation between the scapula and the clavicle. Treatment was usually by open reduction and fixation for type-I fractures and conservative methods for type-II.

At follow-up of the 45 available patients, 87% had excellent results, with no significant differences between the operative and non-operative groups or between the type-I and type-II fractures. We consider that operative treatment should be reserved for patients with multiple shoulder injuries with severe disruption of the scapuloclavicular connection.

We have reviewed 12 fractures of the coracoid process. In two of these patients the fracture extended into the body of the scapula and resulted in displacement of the glenoid. In some cases, there were associated acromioclavicular and glenohumeral dislocations or fractures of the clavicle and the acromion. Two patients required internal fixation to restore congruence of the glenoid; the others were treated conservatively with success. We present a new classification of coracoid fractures which helps in their management.

Serial arterial blood-gas analyses showed a phase of primary hypoxaemia in thirty-two out of fifty fracture patients (64 per cent) without head, chest or abdominal injury. The incidence was greater in those with shaft fractures of the femur or tibia or both, than in those with fractured hips, and was related to the severity of injury and the nature of the accident. Most affected subjects were already hypoxaemic on admission to hospital: the arterial PO₂ commonly fell to between 60 and 70 millimetres of mercury, and the episode generally lasted a few days. The hypoxaemia was generally subclinical but four patients developed mild clinical fat embolism. Early hypoxaemia was not found in six patients admitted with only soft-tissue injuries. One or more subsequent attacks of subclinical hypoxaemia, each lasting a few days, occurred in half of those previously affected. Most episodes followed fracture operation or manipulation. Pulmonary thromboembolism seemed responsible in two patients, but it could be excluded in others given oral anticoagulant prophylaxis from soon after admission. Pulmonary fat embolism is the most likely explanation of the primary episodes and could account for most of the subsequent periods of hypoxaemia.

1. A method of treatment of posterior malleolar fracture is described which restores the proximal surface of the ankle joint to its normal position.

2. Reduction is achieved with a special clamp and the position held with a strong plate. Early ankle movements may therefore be encouraged without fear of redisplacing the fracture.

3. Perfect reduction is necessary to avoid the later onset of arthritis, and this was achieved in five of six patients reported.

1. A study of fifty femoral heads removed at operation for primary prosthetic replacement showed a remarkable constancy of the fracture line.

2. It is suggested that two sub-groups of this fracture-subcapital and transcervical-have been described as a result of radiological interpretation without consideration of the effects of varying degrees of rotation.

1. Intramedullary nailing in two-level tibial fractures provides the following advantages: it allows walking with full weight-bearing in an average time ofthree to four months; it decreases the rate of non-union ; it decreases the rate of malunion ; it should decrease the rate of infection in closed fractures when compared with other types of internal fixation, due to the technique of blind nailing without exposure of the fracture site.

2. Compound tibia! fractures treated by nailing are still often complicated by infection. Nevertheless, we have not been able to find studies in the literature based on series large enough to permit the conclusion that other methods could lower significantly the infection rate.

1. The literature on fractures of the postero-superior aspect of the calcaneus is reviewed.

2. The mechanical distinction between "beak" fractures and avulsion fractures is questioned, and the dangers of a purely radiological diagnosis are stressed.

3. When a complete avulsion is suspected on clinical grounds, open reduction should be done.

1. Five cases of avulsion fracture of the calcaneus are reported.

2. The "beak" fracture of the calcaneus is thought to be a variant of the avulsion fracture and not a separate entity.

3. The variable attachment of the calcaneal tendon to the calcaneus is described, and its relationship to the different forms of avulsion fractures discussed.

4. Operative reduction and fixation are appropriate for young and active patients in order to restore full heel-cord function.

5. Attention is drawn to the risk of pressure necrosis of skin overlying a displaced fragment. Early operative correction may be required to prevent skin damage.

1. Seventy patients with impacted fractures of the femoral neck treated from 1953 to 1965 have been reviewed. Forty-seven were treated conservatively and twenty-three by primary internal fixation.

2. The complications of both methods of treatment are recorded.

3. The prognosis following impacted femoral neck fractures is good. Seventy-nine per cent treated conservatively and 96 per cent treated by primary internal fixation had excellent or good results.

4. Primary internal fixation is the treatment of choice.