Increased stocking densities are frequently reported to depress chicken growth performance, but the mechanisms behind this are not fully understood. This study was conducted to investigate the effects of stocking density on growth performance and digestive microbiota, known to be sensitive to environmental factors. Chickens were reared at 2 stocking densities, 12 or 17 birds/m(2). Growth performance was recorded between d 1 and 39, and litter was scored for quality on d 25, 31, and 37. Digestive microbiota was analyzed along the digestive tract (crop, ileum, ceca) of 3- and 6-wk-old chickens by using 2 molecular approaches: a qualitative method (fingerprinting by temporal temperature gradient gel electrophoresis) and a quantitative method (real-time PCR). An increase in stocking density was found to negatively affect the feed conversion ratio (+3.1%) and depress the daily BW gain of broilers (-5.5%) during the period from d 32 to 39 (P ≤ 0.05). Litter quality was reduced with the high stocking density as early as d 25. At 3 wk of age, stocking density strongly affected the fingerprint profiles of the bacterial community, with the highest modifications observed in the crop and ceca (R analysis of similarity = 0.77 and 0.69, respectively, P ≤ 0.05). At 6 wk of age, significant differences in the fingerprint profiles between the stocking densities appeared in the crop and ceca (R analysis of similarity = 0.52 and 0.27, respectively, P ≤ 0.05). The abundance of bacterial groups targeted by real-time PCR was affected by stocking density, but only to a limited extent. Because digestive microbiota may have consequences on the physiology of the digestive tract, its modification by an increase in stocking density may be involved in the reduced growth performance of the bird.

1. Four groups of hens, each of a different line, were inoculated at peak of lay, per os in the crop with 1 ml of a suspension containing 10(9) cfu/ml Salmonella enteritidis PT4 (SE). The kinetics of SE contamination in the environment, egg shell and yolk were studied during the first 28 d post inoculation. On the day of slaughter, intestines, caeca, spleen, liver, ovary, oviduct and content were investigated for SE contamination. 2. The commercial egg-type line L2 was found to be the most susceptible to SE. It laid many SE-positive yolks (13.8%) and internal and faecal organs were frequently infected. 3. Certain lines are found to exhibit a degree of resistance to SE; the cause of which is unknown and might be attributed to major genes.

The heritability of resistance of poultry to Salmonella enteritidis (SE) was investigated. Three m easurem ents of resistance were m ade: survival after intram uscular inoculation of 419 day-ol d chicks, absence versus presence of Salm onella in spleens and caeca 4 weeks after oral inoculation of 304 hens at peak of laying, and antibody response of 228 hens following two inoculations of an aroA m utant of this serotype. In the ® rst two models of infection, resistance appeared to be heritable. The heritability was estim ated from the sire and dam com ponents, respectively, at 0.14 6 0.10 and 0.62 6 0.16 for chick m ortality, 0.47 6 0.21 and 0.13 6 0.26 for resistance to spleen contam ination, and 0.24 6 0.15 and 0.53 6 0.26 for resistance to caecal contam ination in laying hens. By contrast the estim ated heritability of antibody response was very low (0.03 6 0.08 and 0.10 6 0.08 when estim ated from the sire and dam com ponents, respectively). These results suggest that a selection for increased resistance to SE m ay be ef® cient.

Experiments were carried out to determine the origin of dialkyl disulphides found in the larval frass of Acrolepiopsis assectella Zell. which enable the specialist parasitoid Diadromus pulchellus Wsm. to find its host. These kairomones are not specific: they are also found in the frass of the generalist Cacoecimorpha pronubana Hb. fed a leek-based diet. They are emitted only after the frass emerges from the gut. Bacteria are present in the gut of larvae and in their frass. The addition of antibiotics to the larval diet prevents bacterial development and suppresses disulphide emission. When grown in the presence of sulphur amino acids, precursors of sulphur volatiles, some bacteria metabolize kairomones. The production of volatile kairomones, responsible for hostparasitoid relationships, depends on the presence of additional organisms, i.e. bacteria, and not on enzymes arising from the plant or the phytophage.

The GV strain (serotype 0103:H2:K-), originally isolated from a diarrheic rabbit, is an enteropathogenic Escherichia coli strain that produces diarrhea without synthesizing the classical enterotoxins and that is not invasive. This strain is characterized by a 117-kb plasmid (pREC-l). Histological study of the gut by scanning electron microscopy and transmission electron microscopy was performed on the GV strain, on a derivative strain cured of pREC-l, and on transconjugants obtained by transfer of pREC-l to nonpathogenic strains E. coli K-12 and 6100, not belonging to the 0103 serogroup. The GV strain adhered to the epithelial cells of the ileum and large intestine, whereas the cured GV strain did not. Transfer of plasmid pREC-1 to E. coli K-12 or 6100 allowed the bacteria to attach to the intestinal mucosa in the same manner as that of the wild-type GV strain. Thus, pREC-l seems to play an important role in attachment to and colonization of the intestinal tract of rabbits by E. coli serogroup 0103. Scanning electron microscopy showed numerous bacteria attached together and closely associated with intestinal villi. Transmission electron microscopy revealed effacing lesions characteristic of enteropathogenic E. coli strains: effacing of microvilli and cuplike projections (pedestal formations) associated with an acute inflammatory and hemorrhagic response. In contrast with the results reported for rabbit pathogenic 015 strains, it appeared that the Peyer's patches were not involved in the early stages of infection with the 0103 GV strain. This strain may represent a model for the study of the virulence and pathogenic effects of enteropathogenic and enterohemorrhagic E. coli strains.