inflammatory, if it impairs the normal action of the surface ciliated. epithelium is usually associated with septic secretion, then the impossibility of keeping the passages in an aseptic condition adds greatly to the difficulties attending operative measures.

Wurtz and Lermoyez found that as a rule the nasal mucus was sterile, and Thomson and Hewlett also made numerous cultures from the vestibule, the vibrissæ of the vestibule, and no less than seventy-six cultures from the interior of the nose; of the latter sixty-four remained absolutely sterile (84%). But numerous colonies were generally obtained by cultures from the vestibule and vibrissæ. Thomson and Hewlett do not affirm that micro-organisms are completely absent from the Schneiderian membrane, but that under normal conditions they are quite exceptional. They have further shown that if a culture be deposited on the septum the organisms are rapidly removed by the action of the ciliated epithelium, and that in two hours none of the myriads of organisms could be detected by their bacteriological cultures. They further show that while nasal mucus is not germicidal, it exerts an inhibitory influence on the growth of micro-organisms. Park and Wright have re-investigated these questions, making cultures from the mucus in the interior of the nose of thirty-six normal individuals, with the result that only six were sterile and thirty non-sterile; but in most of the latter the colonies were not numerous, and though these results demonstrate that the nasal fossæ are not sterile, they are not so rich in microbes as was formerly supposed à priori. They considered this to be due to the action of gravity causing a down-flow of mucus from the upper passages, to the action of cilia, to the fact that the mucus is not a good culture medium, to the filter action of the vibrissæ, and to the fact that inspired air usually contains few pathogenic micro-organisms. It is remarkable that many children with chronic membranous rhinitis which yields pure cultures of fully virulent diphtheria bacilli suffer from no constitutional symptoms whatever, while tubercle bacilli and various pyogenic cocci can rarely be found in the nasal passages of healthy individuals.

THE ACCESSORY SINUSES OF THE

NOSE.

A fairly accurate knowledge of the anatomical conformation and relations of these accessory sinuses is essential in rhinology; and it is therefore necessary to refer somewhat fully to the numerous features of clinical importance which they severally present.

Note. The student should read this section in connection with those on Diseases of the Accessory Sinuses, Sections XI and XII.

THE MAXILLARY SINUS (ANTRUM OF HIGHMORE).

(See Stereoplates XX, XXI, XXVII, XXVIII, and XXIX.) The antrum is a three-sided pyramidal cavity hollowed out of the body of the superior maxillary bone; the base of the pyramid is the

Section showing the relations of the frontal sinus, ethmoidal cells, and sphenoidal sinus to the nasal passages. A window has been cut out of the middle turbinated body (10), to show the

structures, etc., concealed by it and lying in the middle meatus.

6

7

8

10

The

outer wall of the nasal fossa, the apex is the malar process. roof is formed by the floor of the orbit, the floor by the hard palate and the alveolar process, the anterior wall by the facial and the posterior wall by the zygomatic surface of the superior maxillary bone.

Reference has already been made to the aperture of communication with the nasal passage beneath the middle turbinated body. In a number of subjects, about 10 per cent according to Zückerkandl, there is an accessory ostium in the mucous membrane closing in the large bony aperture; it is below the uncinate process of the ethmoid, behind the ostium, and uncovered by the middle turbinated bone. The clinical importance of this accessory ostium lies in the fact that an antrum containing pus may discharge its contents through an accessory ostium far back in the nose, and the pus will pass into the nasopharynx instead of forwards and into the middle meatus. The floor of the antrum is uneven from the projection of the sockets of the first and second molars, and several bony laminæ partially subdivide the cavity. These laminæ often interfere with drainage of pus from the antrum, and the position of the natural aperture shows that a collection of pus will escape more freely from the nose on lowering the head.

The relations of the posterior and anterior dental nerves, and of the superior maxillary nerve, which are well displayed in Stereo plate XXXI, explain the cause of direct and reflex pains in antral disease. If tumours grow in the antral cavity, or from other causes the walls of the antrum yield to distending pressure, the floor of the orbit is pushed up and displaces the eye, or the inner wall may bulge into the nasal passage, or the outer wall protrude outwards on the cheeks.

DEVELOPMENT.-The maxillary sinus varies considerably in size, not only in different individuals, and at different ages, but even on the two sides in the same individual In some persons the antra are very small, in others they extend beneath the nasal fossæ above the hard palate. It is formed earlier than any of the other accessory sinuses of the nose, its development commencing about the fourth month of fœtal life. The antrum is small at birth, and does not alter much in size until the second dentition, and then from about the sixth to the eighth year develops rapidly, but only attains its full dimensions in adult life. According to Turner the vertical diameter at birth is 3 mm.

THE ETHMOID CELLS.

The anatomical arrangement and relative development of the cells of the ethmoidal labyrinth, and their relationship to the frontal and sphenoidal sinuses, vary widely and with such frequency that it becomes impossible to draw definite lines between what is normal and what abnormal, but we may describe first the most constant as the typical conformation, indicating the commoner variations from this type, and finally illustrate unusual and hence abnormal developments.

Fig. 15.-Frontal section showing right side of nose. Cross section through ethmoidal labyrinth (SHAMBAUGH).

I, Meatus nasi communis; 2, Bulla ethmoidalis; 3, Processus uncinatus; 4. Concha media.

The subject is greatly simplified if we remember that developmentally the frontal sinus is only an anterior ethmoidal cell extending upwards into the frontal bone, and that similarly the sphenoidal sinus is a posterior ethmoidal cell extending into the body of the sphenoid; thus it is that both the frontal and sphenoidal sinuses may be encroached upon or even largely replaced by other anterior or posterior