J. Wilson

THE END-RESULTS OF THE TONSIL AND ADENOID OPERATION IN CHILDHOOD AND ADOLESCENCE.

50 per cent. of the total danger sources are thus under observation. It may be of interest to state that in the year 1930 a total of 81,000 animals were reported ill and suspect, 43,880 head of cattle were killed in Prussia as a result of this method, and over nine million marks were paid the owners of the slaughtered animals in the same year as compensation. The owners of herds are in ever greater degree learning to appreciate the economic advantage resulting from the elimination of all tuberculous cattle. This growing interest, coupled with propaganda stressing the value of sending the cattle to pasture as compared with the unhealthy process of keeping them stabled, is resulting in a steady increase in stock free from tuberculosis.

Contributions to the Experimental Study of Epidemiology: The Effect of Vaccination on Herd Mortality.

In several previous reports we have described the behaviour of communities of mice, submitted over long periods of time to the risks attendant on the epidemic prevalence of a bacterial infection. These communities have been recruited in ways varying both as regards the rate of immigration and the nature of the immigrants.

The effect of diet on epidemics of mouse-typhoid.

IN previous reports (Watson 1937a, b) one of us has described a series of experiments in which various diets were tested in regard to their effect on the growth, fertility, survival and resistance of mice. It was found that, when part of the oatmeal in a particular diet was replaced by dried skimmed milk, the does were more fertile, and the young mice gained weight more rapidly and showed a lower mortality during the first 8 weeks of life. Mice bred and reared on the latter diet were found to show a significantly increased resistance to per os infection with living cultures of Bact. typhi-murium, and to the intraperitoneal injection of a toxic fraction derived from that organism. The results obtained in a few tests suggested that these mice were also more resistant to the intraperitoneal injection of living Bact. typhi-murium; but the differences observed were smaller, and of doubtful significance. In the present paper we record the effect of these two diets on the epidemic spread of mouse-typhoid. The extensive, and confusing, literature on the effect of diet on resistance to experimental infection has already been reviewed at some length (Watson, 1937b), and it is not necessary to refer to it here, except to note that Webster (1930) has reported a significant decrease in mortality following a change in the diet given to an infected herd of mice. The possible relation between our own observations and those recorded by Webster & Pritchett (1924) and by Pritchett (1927) has been considered in the earlier paper referred to above.

The Epidemic Potency of Strains of Bact. aertrycke of varying virulence: A Report to the Medical Research Council.

ONE of the questions, to which the experimental study of epidemics has as yet given no decisive answer, is the possible significance of variations in bacterial virulence. The experiments here recorded relate to this problem, in the particular case of mouse-typhoid spreading within a closed community. Webster and his co-workers (Webster, 1923, a, b, c, d; 1924, a, b; Webster and Pritchett, 1924) would regard the virulence of any single strain of Bact. aertrycke, the causal organism concerned, as but little variable. While admitting (Webster and Burn, 1928, a, b, c) that variations in virulence may occur in bacteria belonging to this group under the influence of a bacteriophage, they consider that any single strain maintains a constant level of virulence for an indefinite period of time, under any of the usual methods of cultivation, and that variations in virulence are difficult, if not impossible, to induce by any of the methods usually adopted. In particular, they believe that such variations play no significant part in the epidemic spread of disease. Our own experiments, and those of our colleagues (Lockhart, 1926), many of which have not yet been recorded, have convinced us that Bact. aertrycke, when cultivated in the laboratory under various conditions, varies in virulence to a greater degree, and more frequently, than Webster and his colleagues suppose; although our own experience is in accord with theirs, in so far as we have found that a single strain, maintained by massive subculture in a solid medium, and at infrequent intervals, usually maintains its virulence unaltered over months or years. This aspect of the problem will, however, be dealt with more fully in subsequent reports. The present series of experiments is concerned with the epidemic behaviour of strains of appreciably different virulence. With regard to the origin of these strains, we would merely note that they are all derived from strains isolated during the course of the experimental epidemics which we have studied during the past 10 years, and are, so far as we can tell, descendants of the original strain with which we started. Their after-history, as regards the time during which they have been maintained in subculture, and the

The Effect of Withdrawing Mice from an Infected Herd at Varying Intervals.

In a recent report (Greenwood et al. 1936) we included a short discussion of the few scattered observations that we had made on the effects of the dispersal of an infected herd (pp. 189–92). Briefly, we had found that the division of a herd, in which an epidemic due to Bact. typhi-murium was under way, into small isolated groups was followed by a greatly decreased rate of mortality in those groups when the dispersal was carried out at the beginning of the beginning of the epidemic period. Reaggregation of the groups resulted in a fresh spread of the disease, but the final mortality was lower than in a similar herd which had not been dispersed during the earlier stages of cage life (Topley, 1922). In a subsequent experiment (Topley & Wilson, 1925) dispersal was carried out at a later stage of epidemic spread, and very different results were obtained. For the first three weeks or so after division into small groups there was no material difference between the mortality experienced by the dispersed and not-dispersed mice. But at about the 25th day the death-rate in each of the dispersed herds showed a definite decline, while that in the undispersed herds continued unabated for some further length of time.

The Mortality of a Herd of Mice under "Normal" Conditions.

For the purpose of assessing the effects of exposure to a special environment on a herd of mice, one naturally desires a control series, viz. a herd not exposed to a specific risk but otherwise in pari materia with our colonies under experiment.

On the Mechanisms by which protection against Infectious Disease is acquired in “Natural” Epidemics

The problem which we shall discuss in this communication is of great importance, is perhaps the most important of all epidemiological problems, and we are far indeed from supposing that we have solved it. We think, however, that the work we have done is certainly useful in enabling us, and others, to state the problem correctly, and possibly useful in suggesting a first approximation to its solution.