This study was carried out on Bauscat (B), New Zealand (N), Californian (C), N × C (NC) and B × NC (BNC) rabbits. Overall reproductive performance for does with premature parturition (pre-term) of the five breeds was determined and compared with that of does delivering at the normal time (full-term). Full-term does showed a reproductive performance superior to that of pre-term does in all five breeds with larger litter size at weaning, more pups weaned per doe and a lower proportion of stillbirths. Pre-term B, N and BNC does had a larger number of pups weaned than pre-term C and NC does (P < 0·01). There was a high incidence of pup mortality in premature litters, and the main causes of this high mortality were respiratory distress syndrome, crushing of the skull and starvation and chilling.

The mean serum oestradiol 17-β concentrations and degrees of cervical mucous ferning in pre-term B does were higher than in full-term does (P < 0·01) but the mean serum progesterone concentrations in pre-term B does were lower than in full-term does during the gestation period (P < 0·01). Treatment with either dexamethasone or progestin reduced pup mortality in pre-term does and the birth-to-remating interval was shorter in pre-term does treated with dexamethasone (P < 0·01) but longer when treated with progestin (P < 0·01).

The effect of the origin of dietary fibre on digestion and caecal fermentation in the rabbit was studied independently of variations in the level of the different cell wall constituents (cellulose, lignin, etc.). Relationships between microbial activity and the digestion of nutrients were also evaluated over a 4-week growth period. Three experimental diets were given ad libitum from, weaning (28 days) to three groups of nine New Zealand White rabbits, subjected to caecal cannulation at 35 days of age with in-vivo digesta sampling at 6, 7, 8, 9 weeks of age. The diets were similar in their content of neutral-detergent fibre (NDF) (320 g/kg dry matter) and in the respective proportions of cell wall constituents (hemicelluloses (NDF - acid-detergent fibre (ADF)) = 0·54, cellulose (ADF -acid-detergent lignin (ADL)) = 0·37, lignin (ADL) = 0·09), but they differed by their main fibre origin: lucerne meal = diet LM, sunflower meal = diet SM, wheat straw = diet WS.

As a consequence of the similar fibre level, the dietary intake was similar among the diets. The NDF digestibility ivas 0·03 units lower for the diet SM, and led to a lower quantity of NDF degraded (3·7 g NDF per day per kg live weight) compared with LM and WS groups (5·0 g NDF per day per kg live weight). Significant variations in caecal volatile fatty acid (VFA) levels were observed between LM and WS, although the intake of nutrients was rather similar, therefore suggesting a direct effect of the botanical origin of fibre. Compared with LM and SM diets, a significantly lower caecal VFA level was found for the WS diet and was associated with a higher incidence of digestive disorders. In the WS diet, fibres that come from only one botanical origin (wheat straw and bran) may be unfavourable to caecal fermentation and to the health status. The ammonia level, the caecal pH and the proportions of the different VFAs were not affected by the diets.

The total VFA level showed a significant quadratic trend between 6 and 9 weeks of age, whereas the butyrate molar proportion showed a significant linear increase from 0·08 to almost 0·10. In parallel, the daily quantity of hemicellulose digested was positively correlated to the butyrate molar proportion (r = 0·884, P < 0·001), and negatively correlated to the caecal pH (r = -0·705, P = 0·022).

Energy metabolism during flushing of mink (Mustela vison) in relation to a control group was studied by means of combining pooled data from balance and respiration experiments carried out in 1993 (experiment I) and 1994 (experiment II) and some blood metabolites from experiment II. The experiments, which were divided into six 1-week balance periods, started on 7 February and continued until 22 March. Flushing was performed by restricted feeding in periods 2 and 3 and refeeding in periods 4 and 5. Each balance period included a 24-h measurement of heat production by means of indirect calorimetry in open-air circuit respiration chambers. In experiment II blood samples were taken at weekly intervals and analysed for total triiodothyronine, total thyroxine, free thyroxine, glucose, insulin and fructosamine. Generally, intake of digestible nutrients and energy metabolism measurements within the flushed group were strongly influenced by period (energy supply), whereas they remained stable in the control group and differences between treatment groups over the total experiment were non-significant. The fluctuation in energy status induced by flushing did not involve major changes in body weight. Blood metabolites reflected the food intake by being fairly stable in the control group and decreasing during restriction and increasing during refeeding in the flushed group (P < 0·05). As an effect of flushing the number of corpora lutea increased significantly (P < 0·05).

A stochastic simulation was used to investigate the value of crossbreeding information in a two line crossbreeding system in poultry under different genetic scenarios. Populations consisted of 25 sires and 250 dams, which were mated to produce both purebred and crossbred progeny. The next generation parents were selected either based on purebred information (PLS) (sib means, and own performance for females) or additional crossbred sib means were included (CCPS). The trait under selection was controlled by 20 loci with varying degree of dominance. Pure lines differed in initial allele frequencies. If the trait was controlled by loci with partial dominance, little or no extra benefit was obtained from including crossbred information over the pure line information. Under complete dominance and overdominance CCPS outperformed PLS. As a practical rule, CCPS is better than PLS if the ratio between dominance variance and total genetic variance is around 0·3 or higher. In this case the most probable cause of the dominance variation is loci with full or overdominance.

This experiment was conducted to test two different protein sources as alternatives to the commonly used fish meal (FM) in the diet of the European eel (Anguilla anguilla). Six experimental diets were tested in three replicated lots of European eels. All diets contained the same protein and energy content (ca, 300 g crude protein per kg dry matter and 18·5 MJ/kg, respectively) but differed in the nature of the protein source: FM was the only protein source in the control diet and was fully or partially (0–5: 0–5) replaced by meat meal (MM) or sunflower meal (SFM) in four other diets; a sixth diet included SFM as the only protein source but was supplemented with several essential amino acids. Food intake, fish growth and several indices of diet and protein utilization were measured. MM clearly was the poorest protein source while SFM could replace, at least 0·5 of the FM with no significant reduction in performance. In addition, the European eel was able to utilize the supplement of essential amino acids. The full-SFM diet was improved significantly when supplemented and results were not statistically different from the control FM-based diet. Overall, a good correlation was found between the results of each diet and the respective essential amino acid index, calculated using as reference the essential amino acid requirements previously defined for another eel species, Anguilla japonica. This index could be used as a reliable measure for an a priori evaluation of alternative protein sources to be included in commercial foods for eels.

A selection index procedure which utilizes both, purebred and crossbred information was developed for the sire line of a three-path crossbreeding scheme in pigs, to predict response to best linear unbiased prediction (BLUP) selection with an animal model. Purebred and crossbred performance were treated as correlated traits. The breeding goal was crossbred performance but methods can be applied to other goals. A hierarchical mating structure was used. Sires were mated to purebred dams to generate replacements and to F^ from the dam line to generate fattening pigs. Generations were discrete, inbreeding was ignored. The selection index included purebred and crossbred phenotypic information of the current generation and estimated breeding values for purebred and crossbred performance of parents and mates of parents from the previous generation. Reduction of genetic variance due to linkage disequilibrium and reduction of selection intensity due to finite population size and due to correlated index values was accounted for. Selection was undertaken until asymptotic responses were reached. The index was used to optimize the number of selected parents per generation and the number of offspring tested per litter, and to make inferences on the value of crossbred information when the breeding goal was crossbred performance. It was optimal to test a maximum number of offspring per litter, mainly due to increased female selection intensities. Maximum response reductions due to linkage disequilibrium and correlated index values were 32% and 29% respectively. Correcting for correlated index values changed ranking of breeding schemes. Benefit of crossbred information was largest when the genetic correlation between purebred and crossbred performance was low. Due to high correlations between index values in that case, the optimum number of selected sires increased considerably when crossbred information was included.