1. ‘Resistant’ and ‘susceptible’ species of animals appear to be more or less equally susceptible to lung infection with equal doses of tubercle bacilli. Therefore it is pertinent to ask whether recognized natural differences in species resistant are in fact significant. For example, in terms of death rates (i.e. of overall resistance) the C57 mouse is at least as susceptible to tuberculosis as the guinea-pig and much more so than man.

2. Resistant, moderately susceptible and susceptible strains of mice as determined by death rates when exposed to large infecting doses, were equally susceptible to intranasal infection with small numbers of tubercle bacilli.

3. A state of tolerance of the parasite by the host lasting for about 3 weeks was observed in all mouse strains, regardless of ultimate strain resistance.

4. Pre-allergic deaths commenced in all groups when the tuberculous processes left insufficient normal lung to support life, but the deaths stopped first in the resistant strain and last in the susceptible strain, coinciding approximately in each strain with the onset of allergy.

5. Acquired immunity, once established, appeared not to vary in quality from one mouse strain to another, at least during 3 months' observation.

6. Racial or strain variations in the resistance of mice to tuberculosis are therefore natural, only in the sense that speed of onset of acquired immunity is probably genetically determined for each strain.

7. It is suggested that both species and racial variations in natural resistance to pulmonary infection are insignificant. Differences in the subsequent course of the disease appear to be explainable by the rapidity, efficiency and duration of the acquired immune response.

Trials were conducted in the Sierra Nevada Mountains of California to test the effectiveness of 10% DDT dust in insecticide bait-box stations against the fleas of native diurnal rodents. The principle host species involved were the chipmunks Eutamias amoenus and E. speciosus, and the golden-mantled ground squirrel, Citellus lateralis. The important flea species were Monopsyllus eumolpi and M. ciliatus on chipmunks, and Diamanus montanus and Oropsylla idahoensis on ground squirrels. When baited daily for 12 days and furnished with 50 g. of 10% DDT dust, the bait-box stations reduced fleas strikingly within 24 hr., but very little residual control was obtained. A second application in which boxes were baited only twice weekly but for 28 days, also resulted in rapid knockdown and a residual control still effective 42 days after removal of the bait-boxes. The effect of the DDT was apparent at a distance of 400 ft. from the nearest bait-box due to the relatively extensive home ranges of the hosts. Although, in these trials, bait-box stations were spaced 100 ft. apart, 200 ft. spacing probably would be as effective with these hosts.

Threads were infected with the spores of four species of bacteria and put inside rubber gloves which were sealed into plastic bags and irradiated with electrons or gamma rays. A dose of 1·5 Mrad. killed approximately 99·99% of the spores of each species and a dose of 2·5 Mrad. appeared to give an adequate margin of safety for sterilization. Spores were similarly killed inside plastic tubing and within the lumen of hypodermic needles.

The tensile strength of the gloves decreased with increasing doses of radiation so that the rough, solution-dipped gloves tested were significantly weaker after 8·0 Mrad., but the smooth gloves tested still complied with the British Standard after 30 Mrad. When vacuum-packed before irradiation, rough gloves were still satisfactory after 16 Mrad., and smooth gloves were apparently unaffected by 30 Mrad.; they could, therefore, be sterilized six and twelve times, respectively.

Radiation gives dependable sterilization of rubber gloves, and the use of a sealed plastic package obviates subsequent contamination. The possibility of using disposable radiation-sterilized gloves is discussed, the use of a plastic radiation indicator suggested and the practical applications of radiation sterilization in the hospital considered.

An outbreak of Salmonella typhimurium infection in a large general hospital is described. The outbreak was spread over some 20 weeks and gave rise to 102 cases of enteritis and there were at least 150 symptomless excreters of the organism.

Evidence is presented to support the view that the method of spread was by cross-infection, not by contamination of food in bulk. Control measures, which ultimately brought the outbreak to an end, were instituted on this basis.

We wish to thank Dr A. J. H. Tomlinson and his staff at the Public Health Laboratory, County HallS.E. 1 for the generous assistance they gave us in examining all the specimens from the hospital staff during the outbreak. The resources of the Department of Bacteriology of the Postgraduate Medical School were at that time largely devoted to teaching and adequate investigations could not have been made without this help. Dr Tomlinson also gave us much advice both during and after the outbreak for which we are most grateful.

Dr E. S. Anderson, Director of the Central Enteric Reference Laboratory and Bureau, Colindale, London, N.W. 9, very kindly undertook the phage-typing of many of the strains of Salmonella typhimurium isolated and we wish to express our thanks to him for this and other help he gave us in the investigation of the outbreak.

Our thanks are also due to Dr Joan Taylor, Director of the Salmonella Reference Laboratory, Colindale, for her help and interest.

Discussions with Dr R. M. Fry and Dr Joan Boissard in Cambridge in connexion with their investigations of a similar outbreak were most interesting and valuable and it gives us pleasure to thank them.

We also acknowledge gratefully the help and co-operation of our colleagues at Hammersmith Hospital, especially Dr C. E. Roberts, the Medical Superintendent, and of those at the Postgraduate Medical School.

1. It has been shown that cows infected with brucellosis of the udder can usually be detected by an application of a modified milk ring test, based on the finding that milks from infected cows, in addition to showing a high bulk MRT titre also show considerable quarter variations. The effects of adult vaccination with strain 19 vaccine on the MRT are reported; low bulk titres with uniformity between quarters, except at the beginning and end of lactation and immediately following vaccinations, were observed, thus making differentiation between vaccinated and infected cows possible.

2. In a survey of raw milk samples from three areas in southern Scotland, 3·5, 5·3 and 9·0% of samples were shown to be infected with Br. abortus.

A total of 849 samples of pig sera were tested for antibodies to L. canicola, L. icterohaemorrhagiae, L. pomona, L. ballum, L. sejroe, L. grippotyphosa, L. hyos (mitis Johnson), and L. bataviae. Positive results were obtained in 57 samples to all but the last three serotypes.

I wish to thank Mr J. Norval of Edinburgh and Mr W. R. P. Hildebrand of Belfast for supplying the sera and Dr A. Wilson Taylor for his interest and supervision.

I am greatly indebted to the late Dr J. C. Broom of the Wellcome Laboratories of Tropical Medicine for the supply of leptospira.

During the 8½ years, February 1951 to July 1959, 672 strains of myxoma virus were recovered from the field in Australia, all but thirteen, which were derived from mosquito pools, being extracted from the tissues of infected wild rabbits. These were tested for virulence by the intradermal inoculation of small groups of rabbits with a small dose of virus. Based primarily on survival times, strains have been allocated to five grades of virulence, ranging from very high to very low. In spite of inadequate sampling during the first 5 years there has clearly been a trend from a predominance of highly virulent strains in the initial epizootics to a mixture of strains of virulence varying from very high (rarely) to very low. For the last 6 years the proportion of strains allotted to the different virulence grades has remained almost constant, just over half the strains tested each year being classified as moderately virulent.

1. Many immunologically related components occur in bovine serum and colostrum, but certain components in each are absent in the other (specific components). Two components (or groups) occur in colostrum which do not appear in serum, and these can be associated with electrophoretically slow and fast colostrum components.

2. Pre-colostral and adult urines contain low concentrations (<0·2 g./ 100 ml.) of protein which, though immunologically related to serum proteins, does not closely resemble any major serum component in physical properties.

3. Urines collected during the period of proteinuria contain increased concentrations of protein (up to 2 g./ 100 ml.) which is immunologically related to both serum and colostrum. From the presence of material related to specific colostrum components (and the absence of specific serum components), the added protein evidently originates from colostrum rather than serum. The total added protein possesses components with electrophoretic mobilities covering a wide range, with appreciable protein faster than albumin at pH 8. From sedimentation studies, the added proteins resemble the slower sedimenting colostrum proteins (β-lactoglobulin and lactalbumin) and contain no observable material with the sedimentation behaviour of immune lactoglobulin.

4. Immunologically, the added protein is related only to the electrophoretically faster of the specific colostrum components, and this applies also to post-colostral lymph in spite of its very high content of immune lactoglobulin. It follows that the specific electrophoretically slower colostrum components represent only a small proportion of the total immune lactoglobulin fraction.

5. The presence in urine, collected during the period of proteinuria, of antibody activity (deriving originally from the calf's mother), and of protein with the approximate electrophoretic mobility of, but without the sedimentation behaviour of γ-globulin, is to be explained probably on the basis of molecular degradation at some stage between absorption from the intestine and excretion in the urine.

6. The low molecular weight colostrum components, though present in low concentration in the lymph, are absent or below the detectable threshold in post-colostral sera (containing much material derived from immune lactoglobulin), and this is thought to be partly due to selective action by the kidney, allowing the passage of, and causing concentration of low molecular weight protein in the urine, whilst retaining higher molecular weight protein in the serum.

7. The results show that the proteinuria of the new-born calf arose mainly from low molecular weight protein in the colostrum which was absorbed from the gut together with the immune lactoglobulin. Owing to the small molecular weight, this, the former protein, was subsequently cleared from the circulation probably by glomerular filtration, accounting for the difficulty of detecting them electrophoretically and ultracentrifugally. On the other hand, the immune lactoglobulin with its much higher molecular weight was not so cleared from the circulation and is readily detected there.

The authors wish to thank Sir Alan Drury, F.R.S., for his support during the course of this work, Mr N. Buttress and Mr D. Hardman for technical assistance and Mr J. Clark for assistance in the care and handling of the calves.