SUMMARYFaecal oral spread is claimed by many to be the mode of transmission of the gastric pathogenHelicobacter pylori. This idea is based not on experimental data but because the epidemiology ofH. pyloriinfection resembles that of other pathogens known to be spread by the faecal-oral route. This is in spite of the observation that no-one has been successful in culturingH. pylorifrom human stool. In this study, a series of transmission experiments are reported on animals infected with the gastric spirilla,Helicobacter felisand ‘Gastrospirillum hominis’. Germfree mice and rats infected withH. felisdid not transmit their infection to uninoculated mice despite prolonged contact in the same cage nor could the bacterium be isolated from their intestinal contents. This was confirmed in specific pathogen free mice where infected dams did not pass the helicobacter to their progeny. Similarly, mice infected with a human isolate of ‘Gastrospirillum hominis’ did not transmit the infection while in close contact with uninoculated mice. In contrast, in a limited series of experiments, bothH. pyloriandH. feliswere transmitted from infected gnotobiotic Beagle puppies to uninfected animals in the same enclosure. In addition, the gastric mucus from a cat with indigenous ‘Gastrospirillum’-like organisms was infectious for mice, whereas faecal content from the same animal was not. It is suggested that the difference between the murine and canine experiments is that the dogs are more likely to have oral-oral contact than rodents. Unlike dogs, mice and rats do not vomit and are coprophagous. It is concluded that the case for faecal-oral spread ofHelicobacterspecies is ‘not proven’ and that the inter-oral route is more likely.

Summary. Spiral-helical shaped bacteria other than Carnpylobacter pylori have been shown to infect the human stomach. The characteristic helical morphology of these bacteria appears to be similar to that of bacteria found in the stomachs of many other animal species. Early reports on gastric bacteria suggested that rodents may be useful for investigation and isolation of stomach-associated bacteria. Therefore, anaesthetised mice were given, through a stomach tube, a heavy suspension of a spiral-helical bacterium from a cat, scrapings of gastric mucus from primates, or a homogenised whole-antral biopsy from a human patient. At intervals after inoculation, gastric biopsies were examined by lightmicroscopy and electronmicroscopy for the presence of spiral-helical bacteria. Significant colonisation was observed in 40% of mice 1 week, and in 80% of mice 1 1 weeks, after inoculation with suspensions of the cat isolate. Mice were also successfully colonised by spiral bacteria present in homogenised human biopsy material and by other spiral bacteria from a monkey. These observations suggest that mice may prove to be useful animals for the study of gastric bacteria that are, as yet, non-cultivable and for analysis of some of the attributes commonly thought to be involved in colonisation.

Colonisation of the human gastric antrum with non Campylobacter pylori spiral organisms is described. A moderately severe histological gastritis was present despite the fact that these organisms, in contrast to C. pylori, were not closely associated with the epithelial cell surface. Although in vitro culture of the organisms was unsuccessful, in vivo culture was achieved in the mouse stomach. Further study of these spiral organisms should provide important clues to the role of bacteria in upper gastrointestinal disease.

RoUS AND SMITH (1945) drew attention to the fact that embryo skin of BALB/c strain mice if minced and inoculated into the thigh muscle of adults of the same strain would survive and differentiate, and that by including methylcholanthrene in the inoculum, squamous carcinomas could be induced in the implantation cysts. The same procedure failed to yield such clear-cut results using other strains of mice.Our procedure for short term testing of substances for carcinogenic potential derived from these observations, is described below. It is essentially simple and we feel it is worth mentioning certain manipulations which permit a rapid turnover of material. Our experiments also confirm that " c " mice are the most satisfactory to use. MATERIAL AND METHODSSource of implant material BALB/c strain female mice were segregated into individual boxes when obviously pregnant, and when within 48 hours of term were killed by cervical dislocation. Collection of embryo implant materialThe uterine horns were removed aseptically to a sterile petri dish, where the foetuses were removed from the amniotic sacs, separated from the placentae and transferred to a second dry sterile covered dish.The foetuses were pinned out one at a time on clean sheets of paper on a cork mat and dissected under low power magnification. The individual tissues were removed aseptically, placed in labelled petri dishes and kept moist with sterile saline. Implants were usually made within 1 hour of the foetal dissections. Types of embryo implant materialRoutinely skin, lung (usually right lower lobe, to avoid trachea and oesophagus), stomach and urinary bladder were easily removed and when implanted yielded 100 per cent of growing implants after 16 weeks. Implants of kidney, adrenal, thymus, spleen were sometimes successful, but liver and brain failed to grow. Host animalsEight-week-old BALB/c mice of either sex were used as hosts for the embryo implants. This age was chosen because the mice were well past weaning and almost of adult size, which simplified surgical procedures.