1. A high meningococcus carrier-rate in a military unit denotes that dangerous overcrowding exists in this unit.

2. Severe overcrowding (i.e. when beds are less than 1 foot apart) is usually accompanied by a carrier-rate (serological) of at least 20 per cent.

(Twenty per cent. is the danger line indicated in the War Office Memorandum on Cerebro-spinal Fever, March 1917.)

3. A carrier-rate of this height will usually imply that the mobilization standard of 40 square feet has been infringed, and also that beds are less than 1 foot apart.

4. A carrier-rate of 20 per cent. (without awaiting the occurrence of any actual cases of the disease) should be regarded as a signal for prompt and effective action to abolish overcrowding, and to improve ventilation, and to increase the distance between the beds to at least 2½ feet.

5. The distance between beds is of paramount importance.

6. Carrier-rates of between 10 and 20 per cent. are unsatisfactory and imply a certain degree of overcrowding. They must be watched with suspicion, and the mobilization standard strictly enforced.

7. Under the same conditions of overcrowding “non-contact” carrier-rates are the same as “contact” carrier-rates.

8. Quite a moderate degree of “spacing out” of beds, combined with simple methods for improving ventilation are highly efficient agents in reducing high carrier-rates.

9. When, however, a unit shows a high carrier-rate, insistence on the restoration of the mobilization standard is not sufficient.

“Spacing out” must be carried further; and a distance of at least 2½ feet between beds insisted on.

The “peace” standard of 3 feet between beds and 60 square feet of floor space with 600 cubic feet of air space would, of course, be still more effective.

10. The mobilization standard, introduced for a grave emergency, is the lowest possible concession to military necessity, which can be allowed with safety.

It has been shown by numerous observers that well-marked serological differences are found between different strains of meningococci. And since the work of Dopter on the relation of “parameningococci” to meningococci and to cerebro-spinal fever, there has arisen a fairly general agreement that among meningococci capable of causing cerebrospinal meningitis two broadly separable groups can be defined by immunological methods. But beyond this point considerable divergences of opinion appear. On the one hand there are a number of recent workers who have failed to convince themselves that the two groups of meningococci are in reality clearly delimited, permanent and independent entities. On the other hand certain other workers claim to have still further subdivided these micro-organisms into four definite and independent types, each of which possesses in itself the value of a bacterial species [cp. Andrewes (1917)].

In a paper by Capt. Ainley Walker in this Journal1 reference is made to three experiments on three rabbits killed by the “saturation” method of Lieut.-Colonel Gordon which consists in injecting into the auricular vein slope cultures of organisms in massive does at frequent intervals. The organisms used in the present experiments were the Meningococcus, B. typhosus, and Sarcina lutea. Capt. Ainley Walker has handed over to me the organs of the rabbits used for histological examination as to any evidence that might exist of the cause of death. The organs had all been fixed in alcohol. Portions of each were cut out and embedded in the usual way in paraffin. The sections made were stained by haematoxylin and eosin and by such other methods as the appearances suggested. It will be best to set out in detail the features of each organ and then to summarise the whole at the end.

Two strains of B. dysenteriae isolated from cases of asylum dysentery, were found to be readily agglutinable with stock Flexner-Y serum but serum prepared from the one had only a feeble action on the other. Absorption experiments revealed a like specificity. They differed in certain particulars in their action on certain carbohydrate media, and the attempt was made to obtain, by various procedures, variants from each strain giving fermentation reactions as alike as possible.

In the course of this work, an interesting and probably hitherto undescribed type of variation was met with, viz. the capacity exhibited by one strain which fairly promptly fermented the substance employed (isodulcite), of throwing off variants having no action on the substance. These variants appeared as white papillae on the surface of the original red ancestor and the description and analysis of this “reversion” process form the main interest of this paper. Though a fairly close approximation in fermentation characters was realised by selection of variants from each of the original strains, little or no approximation was apparent in respect of the serological affinities of these final variants.

The Absorption or Saturation Test published by Castellani in 1902 is now becoming more extensively used by serological and bacteriological workers in European and American laboratories. Its use is also essential for the accurate study of certain tropical infections, so that it behoves all tropical workers to be familiar with this test. It is therefore believed that an account of the method together with a general review of the literature of the subject may not be out of place in this Journal.

With intense dusting all the dusts used produced much the same effects, both immediate and remote.

With moderate exposures some dusts are much more readily eliminated than others. While coal dust and shale dust enter the lung with great readiness they do not produce, under these conditions, permanent lesions; and the lung might pass for normal after a twelvemonth.

Flue dust and crystalline silica are not eliminated with such readiness.

Two methods are available for distinguishing the dusts—perhaps three.

1. Coal and shale are taken up by cells which are quickly shed and consequently do not set up processes which block lymphatics. These cells frequently break down, and masses of dust are to be seen on the surface of the alveolar walls.

Flue dust and the crystalline silica are taken up by cells which tend to remain in situ and form plaques, which appear early and persist. They are the only site of fibrosis made out in this investigation.

Dusts which form plaques are not readily eliminated.

2. Rate of elimination is more important than rate of invasion. Dusts which are not readily eliminated after moderate exposures are to be avoided.

3. There are two ways out of the lung: direct via the bronchi in the form of plugs of mucin, dust cells and dust; and indirectly via the lymphatics. These ultimately bring some of the dust to the large bronchi at the root of the lung, and the dust can be traced through the connective tissue, muscular, and mucous coats into the epithelium, whence it is excreted. Colourless dusts become pigmented, and to judge from the data as yet available, the more marked the pigmentation the readier the elimination. It is possible that pigment plays a part in this bronchial excretory process, on the analogy of the lipochromes and the deposit of particles in the skin. The different rates at which dusts reach the mucous coats of large bronchi illustrates their relation to lymphatic drainage.

It is of importance to make out whether dusts observe an “all or none rule” or whether they exhibit gradation.

Flue dust for instance in all these experiments produced permanent lesions; but a great deal entered the lung and a great deal was eliminated. Specimens taken from animals immediately after exposure to flue dust, Transvaal quartz, and flint, show more flue dust than crystalline silica; but in cases taken later, owing to flue dust elimination being more rapid, the dust remaining is nearly the same in quantity in all three.

Experiments were carried out (series 10) to determine whether these three dusts can be eliminated like coal and shale if the amount of dust in the air be slight and the invasion of the lung moderate. It is worth making great efforts to diminish the dust cloud if there are reasonable grounds for thinking that under these circumstances rate of elimination might approach rate of invasion. If, on the other hand, a considerable proportion of any dust that enters remains, then other methods must be sought.

The result of these experiments was not very encouraging although it was fairly clear that dosage was an important factor. While it is certainly not the case that concentration of dust in the air × duration of exposure is a constant; still with flue and silica the dose must be very small if accumulation is to be avoided.

More encouraging results were obtained by adding coal. The vigorous catarrhal reaction produced results in the relatively inert dust being carried out in the plugs along with the coal; and these experiments have certainly left the impression that under the conditions described the lungs would practically free themselves from flue or crystalline silica dust if these dusts enter in small quantities only and with coal. The fact that crystalline silica—even when present in considerable quantities—does not always tend to produce grave pulmonary disease, may perhaps be accounted for by assuming that when silica is relatively benevolent there is inhaled together with it some substance that provokes a reaction in the lung with shedding of epithelium and expectoration of débris. If flue dust has been used for stone dusting with impunity it may be due to the presence of coal, since, experimentally, it behaves like the silica group when used pure.

Under the conditions of these experiments:

Dusts that make mischief are dusts that accumulate.

Dusts that are eliminated are dusts that produce a marked initial reaction with much shedding of epithelium.

Dusts that accumulate do not produce so marked an initial reaction, there being much less shedding of epithelium.

Dusts that produce an initial reaction tend to carry out with them the more inert.

An outbreak of food poisoning which occurred at Brighton in November, 1917, showed several features of unusual interest which makes its publication of value. While the epidemiological inquires and pathological investigations were carried out separately a coherent narrative is best obtained by combined presentation. The pathological investigations were carried out for, and at the request of, the Local Government Board.

The outbreak occurred in the Royal Sussex County Hospital which at the time contained 369 residents, 227 being patients and 142 staff. Twenty-eight persons suffered from food poisoning, 24 of whom were patients and 4 members of the staff. Two deaths occurred but both were patients already suffering from severe diseases.

(1) Killed agglutinable broth cultures of the various dysentery bacilli may be prepared by the formalin and cold storage process and may be standardized by the method elaborated by Prof. Dreyer for the typhoid and paratyphoid bacilli.

(2) Since formalin (0·1 %) usually causes a gradual loss of sensitiveness to agglutination during the early life of the emulsion, a certain period (two months) must be allowed to elapse before the emulsion can be used as standard.

(3) After such a period the sensitiveness of the emulsions remains unaltered for at least ten months, and probably much longer.

(4) By none of the other methods of preparation investigated could satisfactory standard agglutinable cultures be made.