Middle-ear effusions have been produced following eustachian tube obstruction in guinea pigs. This was achieved by plugging the pharyngeal orifice of the tube by an atraumatic lateral transpalatal incision. General anaesthesia was used throughout the entire procedure. The presence of fluid in the middle-ear cavity was determined by observing:

(a) The appearance of pallor of the drum.

(b) Retraction of the tympanic membrance.

(c) Cloudiness of the tympanic bullae on radiographs obtained at weekly intervals, confirmed by:

(d) Subsequent myringotomy.

Serial smears of the induced effusions showed an absence of chronic inflammatory cells except in the case of one animal which developed bacterial otitis media. The presence of secretory cells was demonstrated in the actual epithelial layer and lamina propria but no discrete glands were found. Squamous metaplasia of the lining of the middle-ear cleft was noted. The fluid from the tympanic cavity appeared watery, faintly turbid and non-mucoid, and resembled the serious type of effusion found in man.

The over-all distribution of the pre-operative bacteria of 806 ears and the post-operative bacteria of 109 post-operatively moist or discharging ears correlated very well, the only clear difference being the detection of diphtheroid bacilli more often post-operatively than pre-operatively.

When comparing the pre- and post-operative bacteriology of 109 post-operatively infected chronic ears which had undergone radical surgery and obliteration with Palva flaps (and which were drawn from a group of 806 ears originally operated on and followed yearly for 5–14 years) it was noticed that Pseudomonas aeruginosa and Proteus sp were cultured more often post-operatively than pre-operatively in the same ears. Statistically, other bacteria were not found to be significantly present in the same ears.

The differential diagnoses and the duration of symptoms are presented for a group of 288 patients encountered over a ten-year period with a mass in the region of the neck behind and below the angle of the mandible. While infections constituted the largest aetiological group (48·3 per cent), 109 cases (37·9 per cent) had some form of neoplasia, with malignancy being found in 48 (16·6 per cent). The duration of symptoms varied widely, only the acute infections having an acceptably short delay prior to hospital admission. The mean symptom duration for all the neoplastic cases was in excess of five months.

The patient with a lump in the neck is a frequently encountered problem in ear, nose and throat practice. An important precept, which has long been recognized in the literature, is that any persistent asymmetrical mass in the neck of an adult must be regarded as malignant until definite proof to the contrary is obtained (Martin and Romieu, 1952; Slaughter et al., 1956; Skolnik et al., 1965; Shaw, 1976). Differential diagnostic possibilities in all age groups are however numerous, and pre-operative conclusions must often be revised following histopathological examination. A simple and rational approach, which is frequently of considerable value in the clinical assessment of these patients, is a combination of topographical and temporal classifications.

In 1960, Skandalakis et al. proposed a rule-of-7, in which the average duration of symptoms for cervical masses caused by infections was 7 days; for neoplasms, 7 months; while an interval of 7 years was characteristic of developmental anomalies. The topographical approach involves subdividing the neck into anatomical regions. The most posterior area, covered by the trapezius muscle is, in this context, of minimal clinical interest, since the overwhelming majority of neck masses lies anterior to this muscle. The clinically important part of the neck is subdivided into the anterior and posterior triangles by the sternocleidomastoid muscle. The posterior triangle is much less frequently the site of a neck mass, although malignancy is relatively more common in this region (Moussatos and Baffes, 1963). The anterior triangle is further subdivided by the digastric and omohyoid muscles into four smaller triangles, readily recognizable in the living neck, since both muscles are attached to the hyoid bone which can be palpated in the vast majority of cases.

The thyroid and submandibular glands are the most frequent source of neck masses in the inferior carotid, or muscular, and the digastric triangles respectively, and involvement of these organs can usually be readily recognized in the clinic (Beahrs, 1955; Slaughter et al., 1956; Skandalakis et al., 1960). The small submental triangle contains few structures of importance, and the majority of swellings in this region are either thyroglossal cysts or enlarged lymph nodes.

The situation is radically different in the carotid triangle, the region of the neck posterior and inferior to the angle of the mandible, where a wide variety of anatomical structures is congregated, and the differential diagnostic possibilities are consequently greater. The present study is a review of swellings in the neck confined to this area.