Agroecology uses natural processes and local resources rather than chemical inputs to ensure production while limiting the environmental footprint of livestock and crop production systems. Selecting to achieve a maximization of target production criteria has long proved detrimental to fitness traits. However, since the 1990s, developments in animal breeding have also focussed on animal robustness by balancing production and functional traits within overall breeding goals. We discuss here how an agroecological perspective should further shift breeding goals towards functional traits rather than production traits. Breeding for robustness aims to promote individual adaptive capacities by considering diverse selection criteria which include reproduction, animal health and welfare, and adaptation to rough feed resources, a warm climate or fluctuating environmental conditions. It requires the consideration of genotype×environment interactions in the prediction of breeding values. Animal performance must be evaluated in low-input systems in order to select those animals that are adapted to limiting conditions, including feed and water availability, climate variations and diseases. Finally, we argue that there is no single agroecological animal type, but animals with a variety of profiles that can meet the expectations of agroecology. The standardization of both animals and breeding conditions indeed appears contradictory to the agroecological paradigm that calls for an adaptation of animals to local opportunities and constraints in weakly artificialized systems tied to their physical environment.

Agroecology uses ecological processes and local resources rather than chemical inputs to develop productive and resilient livestock and crop production systems. In this context, breeding innovations are necessary to obtain animals that are both productive and adapted to a broad range of local contexts and diversity of systems. Breeding strategies to promote agroecological systems are similar for different animal species. However, current practices differ regarding the breeding of ruminants, pigs and poultry. Ruminant breeding is still an open system where farmers continue to choose their own breeds and strategies. Conversely, pig and poultry breeding is more or less the exclusive domain of international breeding companies which supply farmers with hybrid animals. Innovations in breeding strategies must therefore be adapted to the different species. In developed countries, reorienting current breeding programmes seems to be more effective than developing programmes dedicated to agroecological systems that will struggle to be really effective because of the small size of the populations currently concerned by such systems. Particular attention needs to be paid to determining the respective usefulness of cross-breeding v. straight breeding strategies of well-adapted local breeds. While cross-breeding may offer some immediate benefits in terms of improving certain traits that enable the animals to adapt well to local environmental conditions, it may be difficult to sustain these benefits in the longer term and could also induce an important loss of genetic diversity if the initial pure-bred populations are no longer produced. As well as supporting the value of within-breed diversity, we must preserve between-breed diversity in order to maintain numerous options for adaptation to a variety of production environments and contexts. This may involve specific public policies to maintain and characterize local breeds (in terms of both phenotypes and genotypes), which could be used more effectively if they benefited from the scientific and technical resources currently available for more common breeds. Last but not least, public policies need to enable improved information concerning the genetic resources and breeding tools available for the agroecological management of livestock production systems, and facilitate its assimilation by farmers and farm technicians.

The objective of this work was to study the changes that, selecting for environmental variability of birth weight (BW), could bring to other interesting traits in livestock such as: survivability at weaning (SW), litter size (LS) and weaning weight (WW), their variability assessed from standard deviations of LS, standard deviation of WW (SDWW) and also the total litter weight at birth (TLBW) and total litter weight at weaning. Data were registered after eight generations of a divergent selection experiment for BW environmental variability in mice. Genetic parameters and phenotypic and genetic evolution were assessed using linear homoscedastic and heteroscedastic models in which the traits were attributed to the female, except BW and WW that were in some models also attributed to the pup. Genetic correlation between the trait and variability levels was −0.81 for LS and −0.33 for WW. Clear divergent phenotypic trends were observed between lines for LS, WW and SDWW. Although animals were heavier in the high line, TLBW and at weaning was greater in the low line. Despite the negative genetic correlation that was obtained, SDWW was also higher in the high line. Heritabilities were 0.21 and 0.06, respectively, for LS and SW. Both phenotypic and genetic trends showed clear superiority of the low line over the high line for these traits, but inferior for WW. Heteroscedastic model performed similar to the homoscedastic model when there was enough information. Considering LS and survival, the low line was preferred from a welfare point of view, but its superiority from the productivity perspective was not clear. Robustness seemed higher as shown by a low variation and having a benefit to the animal welfare, but this still remains unclear. It was concluded that low variation benefits the welfare of animals.

Accurate measures of effective population sizes (Ne) in livestock require good quality data and specialized skills for their computation and analysis. Ne can be estimated by Wright’s equation Ne=4MF/(M+F) (M, F being sires and dams, respectively), but this requires assumptions which are often not met. Total census sizes Nc of livestock breeds are collated globally. This paper investigates whether estimates of Ne can be made from Nc; this would facilitate conservation monitoring. Some Ne methodologies avoid the assumptions of Wright’s equation and permit measurement, rather than estimation, of Ne. Those considered here employ, respectively, linkage disequilibrium (LD) of single-nucleotide polymorphisms (yielding Ne(LD)), and genealogical analysis (rate of increase of inbreeding, DF), yielding Ne(DF). Considering breeds of cattle, sheep, horses, pigs and goats for which Nc and either Ne(LD) or Ne(DF) are known (totals of 203 breeds and 321 breeds, respectively), proportionality has been investigated between Nc and these measures of Ne. Ne(LD) was found to increase with Nc, significantly in sheep and horses, less so in cattle, but not at all in pigs. Ne(DF) was correlated with log10(Nc) in cattle, sheep and horses (53, 56, 43 breeds, respectively). Ne(LD) was correlated in cattle (73 breeds) and pigs (31 breeds) with the log10 transformation of Ne as calculated by Wright’s equation. Further verification and refinement are needed, particularly of census data, but credible predictions of Ne are obtainable by applying the following multipliers to log10(Nc): cattle 17.61, sheep 97.72, horse 70.78. For cattle and pigs, multiplying log10(Ne(Wright)) by, respectively, 40.69 and 60.09, also gives credible predictions. Such census-based estimates of Ne could in principle be generated by non-specialists and are likely to be suited to audits of conservation activity when financial resources or availability of data are limiting. The ratio Ne/Nc varied among species with an overall median value of 0.03, less than a tenth of that typically observed in wild mammals. Characteristics were also investigated of a distinct herdbook-based methodology, namely the development of Wright’s equation to take into account variances of progeny numbers to yield what has been termed here Ne (Hill). Comparison of these values with Ne (Wright) could help to identify breeds with breeding structures conducive or inimical to genetic conservation. However, Ne(Hill) requires breed-specific values for these variances, and this restricts its applicability.

A total of 16 pure-bred Iberian (IB) sows, all of them suckling six piglets, were used, eight of them in each of the two consecutive trials (1 and 2). Daily milk yield and composition were determined weekly over a 34-day lactation period. Within each litter, one piglet at birth and four piglets on day 35 of life were slaughtered. Milk intake per piglet tended to be greater in trial 2 (832 v. 893 g/day; P=0.066), but piglets grew at 168±3.3 g/day, irrespective of the trial. In the IB sow milk, the linoleic (LA) : linolenic (LNA) acid ratio averaged 14.6 and 15.2 in trial 1 and trial 2, respectively. A fivefold increase in piglet body fat content was observed over lactation (P<0.001). Most of this fat (81.4%) was present in the carcass. After 34 days of lactation, whole-body relative content of palmitic, palmitoleic, stearic and oleic acids were very close to those in the milk consumed, suggesting direct deposition. Daily deposition of LA derivatives and of LNA and its derivatives was found to be extremely low (<0.02 g, on average). Moreover, some of the arachidonic acid (ARA) in tissues of the IB piglet at birth disappeared throughout the lactating period. An overall fractional deposition for total fatty acids (FA) was 0.409. Fractional oxidation (disappearance) rates were 0.939 and 0.926 for n-6 and n-3 polyunsaturated FA. The overall rate of disappearance for the major non-essential FA (myristic, palmitic, palmitoleic, stearic and oleic acids), estimated as 1−the overall fractional deposition rate, was 0.546. It is concluded that the high degree of FA unsaturation, high oxidation rate of LA and LNA, and poor synthesis of ARA from LA and of docosahexaenoic acid from LNA found in the suckling piglet might increase the energy cost of whole-body fat accretion, a contributor to the observed low efficiency of use of milk energy for growth.

Mutants of Bacillus subtilis can be developed to overproduce Val in vitro. It was hypothesized that addition of Bacillus subtilis mutants to pig diets can be a strategy to supply the animal with Val. The objective was to investigate the effect of Bacillus subtilis mutants on growth performance and blood amino acid (AA) concentrations when fed to piglets. Experiment 1 included 18 pigs (15.0±1.1 kg) fed one of three diets containing either 0.63 or 0.69 standardized ileal digestible (SID) Val : Lys, or 0.63 SID Val : Lys supplemented with a Bacillus subtilis mutant (mutant 1). Blood samples were obtained 0.5 h before feeding and at 1, 2, 3, 4, 5 and 6 h after feeding and analyzed for AAs. In Experiment 2, 80 piglets (9.1±1.1 kg) were fed one of four diets containing 0.63 or 0.67 SID Val : Lys, or 0.63 SID Val : Lys supplemented with another Bacillus subtilis mutant (mutant 2) or its parent wild type. Average daily feed intake, daily weight gain and feed conversion ratio were measured on days 7, 14 and 21. On day 17, blood samples were taken and analyzed for AAs. On days 24 to 26, six pigs from each dietary treatment were fitted with a permanent jugular vein catheter, and blood samples were taken for AA analysis 0.5 h before feeding and at 1, 2, 3, 4, 5 and 6 h after feeding. In experiment 1, Bacillus subtilis mutant 1 tended (P<0.10) to increase the plasma levels of Val at 2 and 3 h post-feeding, but this was not confirmed in Experiment 2. In Experiment 2, Bacillus subtilis mutant 2 and the wild type did not result in a growth performance different from the negative and positive controls. In conclusion, results obtained with the mutant strains of Bacillus subtilis were not better than results obtained with the wild-type strain, and for both strains, the results were not different than the negative control.

Reduction of the CP content in the diets of piglets requires supplementation with crystalline essential amino acids (AA). Data on the leucine (Leu) and histidine (His) requirements of young pigs fed low-CP diets are limited and have primarily been obtained from nonlinear models. However, these models do not consider the possible decline in appetite and growth that can occur when pigs are fed excessive amounts of AA such as Leu. Therefore, two dose-response studies were conducted to estimate the standardised ileal digestible (SID) Leu : lysine (Lys) and His : Lys required to optimise the growth performance of young pigs. In both studies, the average daily gain (ADG), average daily feed intake (ADFI) and gain-to-feed ratio (G : F) were determined during a 6-week period. To ensure that the diets had sub-limiting Lys levels, a preliminary Lys dose-response study was conducted. In the Leu study, 60 35-day-old piglets of both sexes were randomly assigned to one of five treatments and fed a low-CP diet (15%) with SID Leu : Lys levels of 83%, 94%, 104%, 115% or 125%. The His study used 120 31-day-old piglets of both sexes, which were allotted to one of five treatments and fed a low-CP diet (14%) with SID His : Lys levels of 22%, 26%, 30%, 34% or 38%. Linear broken-line, curvilinear-plateau and quadratic-function models were used for estimations of SID Leu : Lys and SID His : Lys. The minimum SID Leu : Lys level needed to maximise ADG, ADFI and G : F was, on average, 101% based on the linear broken-line and curvilinear-plateau models. Using the quadratic-function model, the minimum SID Leu : Lys level needed to maximise ADG, ADFI and G : F was 108%. Data obtained from the quadratic-function analysis further showed that a ±10% deviation from the identified Leu requirement was accompanied by a small decline in the ADG (−3%). The minimum SID His : Lys level needed to maximise ADG, ADFI and G : F was 27% and 28% using the linear broken-line and curvilinear-plateau models, respectively, and 33% using the quadratic-function model. The preferred model to estimate the His requirement was the curvilinear-plateau model. However, a 10% reduction in the SID His : Lys level was associated with an 11% reduction in the ADG. In conclusion, the SID Leu : Lys level needed to maximise growth was 108% when using the quadratic-function model as the best-fitting model. The minimum SID His : Lys level required to optimise growth was 28% when using the curvilinear-plateau model as the best-fitting model.

The objective of this study was to determine if a moderate or high reduction of dietary CP, supplemented with indispensable amino acids (IAA), would affect growth, intestinal morphology and immunological parameters of pigs. A total of 40 barrows (initial BW=13.50±0.50 kg, 45±2 day of age) were used in a completely randomized block design, and allocated to four dietary treatments containing CP levels at 20.00%, 17.16%, 15.30% and 13.90%, respectively. Industrial AA were added to meet the IAA requirements of pigs. After 4-week feeding, blood and tissue samples were obtained from pigs. The results showed that reducing dietary CP level decreased average daily gain, plasma urea nitrogen concentration and relative organ weights of liver and pancreas (P<0.01), and increased feed conversion ratio (P<0.01). Pigs fed the 13.90% CP diet had significantly lower growth performance than that of pigs fed higher CP at 20.00%, 17.16% or 15.30%. Moreover, reducing dietary CP level decreased villous height in duodenum (P<0.01) and crypt depth in duodenum, jejunum and ileum (P<0.01). The reduction in the dietary CP level increased plasma concentrations of methionine, alanine (P<0.01) and lysine (P<0.05), and decreased arginine (P<0.05). Intriguingly, reducing dietary CP level from 20.00% to 13.90% resulted in a significant decrease in plasma concentration of IgG (P<0.05), percentage of CD3+T cells of the peripheral blood (P<0.01), also down-regulated the mRNA abundance of innate immunity-related genes on toll-like receptor 4, myeloid differentiation factor 88 (P<0.01) and nuclear factor kappa B (P<0.05) in the ileum. These results indicate that reducing dietary CP level from 20.00% to 15.30%, supplemented with IAA, had no significant effect on growth performance and had a limited effect on immunological parameters. However, a further reduction of dietary CP level up to 13.90% would lead to poor growth performance and organ development, associated with the modifications of intestinal morphology and immune function.

Vegetable oils are used to increase energy density of dairy cow diets, although they can provoke changes in rumen bacteria populations and have repercussions on the biohydrogenation process. The aim of this study was to evaluate the effect of two sources of dietary lipids: soybean oil (SO, an unsaturated source) and hydrogenated palm oil (HPO, a saturated source) on bacterial populations and the fatty acid profile of ruminal digesta. Three non-lactating Holstein cows fitted with ruminal cannulae were used in a 3×3 Latin square design with three periods consisting of 21 days. Dietary treatments consisted of a basal diet (Control, no fat supplement) and the basal diet supplemented with SO (2.7% of dry matter (DM)) or HPO (2.7% of DM). Ruminal digesta pH, NH3–N and volatile fatty acids were not affected by dietary treatments. Compared with control and HPO, total bacteria measured as copies of 16S ribosomal DNA/ml by quantitative PCR was decreased (P<0.05) by SO. Fibrobacter succinogenes, Butyrivibrio proteoclasticus and Anaerovibrio lipolytica loads were not affected by dietary treatments. In contrast, compared with control, load of Prevotella bryantii was increased (P<0.05) with HPO diet. Compared with control and SO, HPO decreased (P<0.05) C18:2 cis n-6 in ruminal digesta. Contents of C15:0 iso, C18:11 trans-11 and C18:2 cis-9, trans-11 were increased (P<0.05) in ruminal digesta by SO compared with control and HPO. In conclusion, supplementation of SO or HPO do not affect ruminal fermentation parameters, whereas HPO can increase load of ruminal P. bryantii. Also, results observed in our targeted bacteria may have depended on the saturation degree of dietary oils.

Metabolic adaptation includes an array of concerted metabolic and endocrine events that enable dairy cows bridging the period of energy deficit at the onset of lactation. The present study evaluated metabolic, endocrine and reticuloruminal pH changes in 30 (25 Holstein and five Simmental) periparturient dairy cows experiencing variable lipolysis early postpartum. Cows were fed the same close-up and fresh lactation diets and kept in the same management conditions. Blood samples were collected at day 14, and day 4, relative to expected parturition, and at day 2, and day 21 postpartum, and serum metabolites and hormones related to glucose and lipid metabolism, as well as concentrations of several liver enzymes and acute phase proteins were determined. Additionally, reticuloruminal pH was monitored every 10 min over the last 3 days of the observation period. BW and milk yields were recorded and balances of energy and protein were assessed. Based on serum concentration of non-esterified fatty acids (NEFA) postpartum, cows were retrospectively classified into low (n=8), medium (n=11), and high (n=11) lipolysis groups, with NEFA levels of <0.4 mmol/l, between 0.4 and 0.7 mmol/l, and >0.7 mmol/l, respectively. Overall, elevated NEFA concentrations in the High group went along with a higher ratio of NEFA to cholesterol and reduced insulin sensitivity. While serum glucose, energy deficit and BW loss did not differ, cows of the High group exhibited increased lactate concentrations in the serum, compared with the Medium group. No differences in liver enzymes and acute phase proteins were evidenced among fat mobilization groups, whereas concentration of serum billirubin was lowest in the Low group after parturition. Data of milk yield and milk energy output showed no differences among groups, despite divergences in calculated energy balance and BW change postpartum. Cows of the Low group tended to increase dry matter intake but also showed longer time duration of pH below 6.0 in the reticulorumen (on average 299 min/day compared with 99 and 91 min/day for Medium and High groups, respectively). Differences in metabolic, endocrine and reticuloruminal pH responses indicate diverse metabolic adaptation strategies of early-lactation cows to cope with energy deficit postpartum.

Both white and brown adipose tissues are recognized to be differently involved in energy metabolism and are also able to secrete a variety of factors called adipokines that are involved in a wide range of physiological and metabolic functions. Brown adipose tissue is predominant around birth, except in pigs. Irrespective of species, white adipose tissue has a large capacity to expand postnatally and is able to adapt to a variety of factors. The aim of this review is to update the cellular and molecular mechanisms associated with pre- and postnatal adipose tissue development with a special focus on pigs and ruminants. In contrast to other tissues, the embryonic origin of adipose cells remains the subject of debate. Adipose cells arise from the recruitment of specific multipotent stem cells/progenitors named adipose tissue-derived stromal cells. Recent studies have highlighted the existence of a variety of those cells being able to differentiate into white, brown or brown-like/beige adipocytes. After commitment to the adipocyte lineage, progenitors undergo large changes in the expression of many genes involved in cell cycle arrest, lipid accumulation and secretory functions. Early nutrition can affect these processes during fetal and perinatal periods and can also influence or pre-determinate later growth of adipose tissue. How these changes may be related to adipose tissue functional maturity around birth and can influence newborn survival is discussed. Altogether, a better knowledge of fetal and postnatal adipose tissue development is important for various aspects of animal production, including neonatal survival, postnatal growth efficiency and health.

Feed efficiency is an important trait in the future sustainability of pig production, however, the mechanisms involved are not fully elucidated. The objective of this study was to examine nutrient digestibility, organ weights, select bacterial populations, volatile fatty acids (VFA’s), enzyme and intestinal nutrient transporter gene expression in a pig population divergent in feed efficiency. Male pigs (n=75; initial BW 22.4 kg SEM 2.03 kg) were fed a standard finishing diet for 43 days before slaughter to evaluate feed intake and growth for the purpose of calculating residual feed intake (RFI). Phenotypic RFI was calculated as the residuals from a regression model regressing average daily feed intake (ADFI) on average daily gain (ADG) and midtest BW0.60 (MBW). On day 115, 16 pigs (85 kg SEM 2.8 kg), designated as high RFI (HRFI) and low RFI (LRFI) were slaughtered and digesta was collected to calculate the coefficient of apparent ileal digestibility (CAID), total tract nutrient digestibility (CATTD), microbial populations and VFA’s. Intestinal tissue was collected to examine intestinal nutrient transporter and enzyme gene expression. The LRFI pigs had lower ADFI (P<0.001), improved feed conversion ratio (P<0.001) and an improved RFI value relative to HRFI pigs (0.19 v. −0.14 SEM 0.08; P<0.001). The LRFI pigs had an increased CAID of gross energy (GE), and an improved CATTD of GE, nitrogen and dry matter compared to HRFI pigs (P<0.05). The LRFI pigs had higher relative gene expression levels of fatty acid binding transporter 2 (FABP2) (P<0.01), the sodium/glucose co-transporter 1 (SGLT1) (P<0.05), the glucose transporter GLUT2 (P<0.10), and the enzyme sucrase–isomaltase (SI) (P<0.05) in the jejunum. The LRFI pigs had increased populations of lactobacillus spp. in the caecum compared with HRFI pigs. In colonic digesta HRFI pigs had increased acetic acid concentrations (P<0.05). Differences in nutrient digestibility, intestinal microbial populations and gene expression levels of intestinal nutrient transporters could contribute to the biological processes responsible for feed efficiency in pigs.

We investigated the diagnostic reliability of pregnancy detection using changes in interferon stimulated gene (ISG) messenger RNA (mRNA) levels in circulating immune cells in ewes. Two different groups of ewes (an experimental group, experiment 1 and a farm group, experiment 2) were oestrus-synchronized and blood sampled on day 14 (D0=day of insemination in control animals, experiment 1) and day 15 (experiment 2). Real-time PCR were performed to evaluate the abundance of different ISG mRNAs. In the experimental group, peripheral blood mononuclear cells of 29 ewes born and bred in experimental facilities were isolated using a Percoll gradient method. Gene expression for Chemokine (C-X-C motif) ligand 10 (CXCL10), Myxovirus (influenza virus) resistance 1 (MX1) and Signal transducer and activator of transcription 1 (STAT1) mRNA were, respectively, 8.3-fold, 6.1-fold and 2.7-fold higher (P<0.001) in pregnant compared with non-pregnant ewes. The receiver operating characteristic (ROC) curves generated from the real-time PCR data demonstrated that a reliable cut-off could be established for CXCL10, MX1 and STAT1. In the farm group of animals, peripheral blood leucocytes of 37 cross-bred multiparous ewes bought from several herds were isolated using the PAXgene® procedure. This blood sampling procedure is achievable in farms, whereas the Percoll method is not. No significant differences (P>0.10) in CXCL10, STAT1, MX1, Myxovirus (influenza virus) resistance 2 (MX2) and ISG15 ubiquitin-like modifier (ISG15) mRNA expression were found between pregnant and non-pregnant ewes. The ROC curves and the hierarchical classification generated from the real-time PCR data failed to discriminate between pregnant and non-pregnant animals. In this group of animals, our results show a strong variability in ISG expression patterns: 17% of animals identified as non-pregnant by the five tests were in fact pregnant, only 52% of pregnant animals had at least two positive results (two genes above threshold), whereas up to five positive results (five genes above threshold) were needed to avoid misclassification. In conclusion, this study illustrates the high variability in ISG expression levels in immune circulating cells during early pregnancy and, therefore, highlights the limits of using ISG expression levels in blood samples, collected on PAXgene® tubes on farms, for early pregnancy detection in sheep.

The objectives of the present study were to compare the cortisol response caused by ear tagging piglets with the distress caused by other known painful husbandry procedures (e.g. castration and tail docking) and to evaluate the effectiveness of analgesia with meloxicam to reduce the cortisol response caused by these procedures. In total, 210 male piglets were randomised to equal numbers (n=30) into one of seven groups: a control group which was only handled (H), an ear tagged group that received no analgesia (ET), an ear tagged group with analgesia (ETM), a castration group with no analgesia (C), a castration group with analgesia (CM), a tail-docked group with no analgesia (TD) and a tail-docked group with analgesia (TDM). The procedures were carried out on day 3 or 4 after farrowing. Five blood samples were taken from each piglet: 30 min before the respective procedure (baseline value), and 30, 60 min, 4 and 7 h after processing, to assess cortisol concentrations. Means as well as the area under the curve (AUC) value were analysed and the effective sizes of the procedures were established. At 7 h after the experimental treatment, cortisol concentrations had returned to base values in all groups. ET evoked a greater cortisol response than H piglets at 30 min (P<0.001) and 60 min (P=0.001). The cortisol response to ET was lower than C at 30 min (P=0.001) but did not differ significantly at the other sample times. The mean cortisol response was similar between ET and TD piglets over all sample times. Taking both intensity and duration of the cortisol response into account (AUC), ET evoked a greater response than TD. Analgesia (ETM) resulted in significantly lower cortisol levels than ET at 30 and 60 min post-procedure. Castration (C) provoked the highest cortisol response of all procedures; a significant analgesic effect (CM) was shown only at 4 h post-procedure. TD resulted in significantly higher cortisol levels than H piglets only at 30 min; analgesia (TDM) significantly reduced the cortisol response at 30 min. We conclude that ear tagging causes a dramatic increase in cortisol levels compared with handling alone in piglets, which suggests that this procedure causes substantial distress. However, further research is needed to confirm these results.

Early weaning may be used in beef cattle production to improve reproduction rates in range conditions. However, weaning causes a stress response in cows, which may be especially strong in early weaning management, as the bond between the cow and the calf is still strong. We hypothesized that weaning calves in two steps, with the aid of anti-sucking devices (nose flaps) would reduce the behavioural stress response in the cows separated from their calves 2 months after parturition. We compared the behaviour frequency and weight change in cows that were weaned abruptly, by separation of the calf on day 0 of the study, or in two steps, consisting of the use of anti-sucking nose flaps for 5 days before permanent separation; a third group was not weaned to serve as control. Thirty-six crossbred multiparous Aberdeen Angus×Hereford cows and their calves (n=12/treatment) were managed in three paddocks with similar pasture availability, with four dyads from each treatment per paddock. Cows’ behaviour was observed by direct visual instantaneous sampling, at 10 min intervals from days −3 to 11. Weaning the calves in two steps clearly attenuated the behavioural stress response observed in abruptly weaned cows, which included reductions in grazing and lying, and increases in pacing, walking and vocalizing. Our results corroborate those previously shown for cows nursing older calves, and indicate that step weaning can reduce the behavioural stress response of cows at weaning, even when the calf is weaned shortly after birth, when the bond between the cow and calf is still very strong.

The objective of this research was to evaluate the survival rate of primiparous heifers within a large sample of herds across the United Kingdom and specifically to assess the association between age at first calving (AFC) on their survival. Data from 437 herds were re-structured for analysis. Descriptive statistics were calculated, and a multilevel logistic regression model was used to explore factors associated with the risk of first lactation culling. Potential explanatory variables included AFC, herd size, culling rate within the whole herd, calving season, herd mean 305-day yield and herd mean calving interval. The mean within-herd culling rate for the primiparous heifers was 15.9%. The mean within-herd AFC was 29.6 months, with 35.9% of heifers having an AFC >30 months of age. Multivariable analysis revealed a negative association between survival rate of primiparous heifers and increasing AFC, and also associations with herd culling rate in older cows and calving season. This study highlights the importance of AFC for survival of primiparous heifers, as well as the need to address heifer wastage in herds with high culling rates.

The production of protein from animal sources is often criticized because of the low efficiency of converting plant protein from feeds into protein in the animal products. However, this critique does not consider the fact that large portions of the plant-based proteins fed to animals may be human-inedible and that the quality of animal proteins is usually superior as compared with plant proteins. The aim of the present study was therefore to assess changes in protein quality in the course of the transformation of potentially human-edible plant proteins into animal products via livestock production; data from 30 Austrian dairy farms were used as a case study. A second aim was to develop an approach for combining these changes with quantitative aspects (e.g. with the human-edible feed conversion efficiency (heFCE), defined as kilogram protein in the animal product divided by kilogram potentially human-edible protein in the feeds). Protein quality of potentially human-edible inputs and outputs was assessed using the protein digestibility-corrected amino acid score and the digestible indispensable amino acid score, two methods proposed by the Food and Agriculture Organization of the United Nations to describe the nutritional value of proteins for humans. Depending on the method used, protein scores were between 1.40 and 1.87 times higher for the animal products than for the potentially human-edible plant protein input on a barn-gate level (=protein quality ratio (PQR)). Combining the PQR of 1.87 with the heFCE for the same farms resulted in heFCE×PQR of 2.15. Thus, considering both quantity and quality, the value of the proteins in the animal products for human consumption (in this case in milk and beef) is 2.15 times higher than that of proteins in the potentially human-edible plant protein inputs. The results of this study emphasize the necessity of including protein quality changes resulting from the transformation of plant proteins to animal proteins when evaluating the net contribution of livestock to the human food supply. Furthermore, these differences in protein quality might also need to be considered when choosing a functional unit for the assessment of environmental impacts of the production of different proteins.

Residual feed intake (RFI) is the difference between actual and predicted dry matter intake (DMI) of individual animals. Recent studies with Holstein-Friesian calves have identified an ~20% difference in RFI during growth (calf RFI) and these groups remained divergent in RFI during lactation. The objective of the experiment described here was to determine if cows selected for divergent RFI as calves differed in milk production, reproduction or in the profiles of BW and body condition score (BCS) change during lactation, when grazing pasture. The cows used in the experiment (n=126) had an RFI of −0.88 and +0.75 kg DM intake/day for growth as calves (efficient and inefficient calf RFI groups, respectively) and were intensively grazed at four stocking rates (SR) of 2.2, 2.6, 3.1 and 3.6 cows/ha on self-contained farmlets, over 3 years. Each SR treatment had equal number of cows identified as low and high calf RFI, with 24, 28, 34 and 40/11 ha farmlet. The cows divergent for calf RFI were randomly allocated to each SR. Although SR affected production, calf RFI group (low or high) did not affect milk production, reproduction, BW, BCS or changes in these parameters throughout lactation. The most efficient animals (low calf RFI) lost similar BW and BCS as the least efficient (high calf RFI) immediately post-calving, and regained similar BW and BCS before their next calving. These results indicate that selection for RFI as calves to increase efficiency of feed utilisation did not negatively affect farm productivity variables (milk production, BCS, BW and reproduction) as adults when managed under an intensive pastoral grazing system.

Eco-efficiency is a useful guide to dairy farm sustainability analysis aimed at increasing output (physical or value added) and minimizing environmental impacts (EIs). Widely used partial eco-efficiency ratios (EIs per some functional unit, e.g. kg milk) can be problematic because (i) substitution possibilities between EIs are ignored, (ii) multiple ratios can complicate decision making and (iii) EIs are not usually associated with just the functional unit in the ratio’s denominator. The objective of this study was to demonstrate a ‘global’ eco-efficiency modelling framework dealing with issues (i) to (iii) by combining Life Cycle Analysis (LCA) data and the multiple-input, multiple-output production efficiency method Data Envelopment Analysis (DEA). With DEA each dairy farm’s outputs and LCA-derived EIs are aggregated into a single, relative, bounded, dimensionless eco-efficiency score, thus overcoming issues (i) to (iii). A novelty of this study is that a model providing a number of additional desirable properties was employed, known as the Range Adjusted Measure (RAM) of inefficiency. These properties altogether make RAM advantageous over other DEA models and are as follows. First, RAM is able to simultaneously minimize EIs and maximize outputs. Second, it indicates which EIs and/or outputs contribute the most to a farm’s eco-inefficiency. Third it can be used to rank farms in terms of eco-efficiency scores. Thus, non-parametric rank tests can be employed to test for significant differences in terms of eco-efficiency score ranks between different farm groups. An additional DEA methodology was employed to ‘correct’ the farms’ eco-efficiency scores for inefficiencies attributed to managerial factors. By removing managerial inefficiencies it was possible to detect differences in eco-efficiency between farms solely attributed to uncontrollable factors such as region. Such analysis is lacking in previous dairy studies combining LCA with DEA. RAM and the ‘corrective’ methodology were demonstrated with LCA data from French specialized dairy farms grouped by region (West France, Continental France) and feeding strategy (regardless of region). Mean eco-efficiency score ranks were significantly higher for farms with <10% and 10% to 30% maize than farms with >30% maize in the total forage area before correcting for managerial inefficiencies. Mean eco-efficiency score ranks were higher for West than Continental farms, but significantly higher only after correcting for managerial inefficiencies. These results helped identify the eco-efficiency potential of each region and feeding strategy and could therefore aid advisors and policy makers at farm or region/sector level. The proposed framework helped better measure and understand (dairy) farm eco-efficiency, both within and between different farm groups.

Crop–livestock production is claimed more sustainable than specialized production systems. However, the presence of controversial studies suggests that there must be conditions of mixing crop and livestock productions to allow for higher sustainable performances. Whereas previous studies focused on the impact of crop–livestock interactions on performances, we posit here that crop–livestock organization is a key determinant of farming system sustainability. Crop–livestock organization refers to the percentage of the agricultural area that is dedicated to each production. Our objective is to investigate if crop–livestock organization has both a direct and an indirect impact on mixed crop–livestock (MC–L) sustainability. In that objective, we build a whole-farm model parametrized on representative French sheep and crop farming systems in plain areas (Vienne, France). This model permits simulating contrasted MC–L systems and their subsequent sustainability through the following indicators of performance: farm income, production, N balance, greenhouse gas (GHG) emissions (/kg product) and MJ consumption (/kg product). Two MC–L systems were simulated with contrasted crop–livestock organizations (MC20–L80: 20% of crops; MC80–L20: 80% of crops). A first scenario – constraining no crop–livestock interactions in both MC–L systems – permits highlighting that crop–livestock organization has a significant direct impact on performances that implies trade-offs between objectives of sustainability. Indeed, the MC80–L20 system is showing higher performances for farm income (+44%), livestock production (+18%) and crop GHG emissions (−14%) whereas the MC20–L80 system has a better N balance (−53%) and a lower livestock MJ consumption (−9%). A second scenario – allowing for crop–livestock interactions in both MC20–L80 and MC80–L20 systems – stated that crop–livestock organization has a significant indirect impact on performances. Indeed, even if crop–livestock interactions permit improving performances, crop–livestock organization influences the capacity of MC–L systems to benefit from crop–livestock interactions. As a consequence, we observed a decreasing performance trade-off between MC–L systems for farm income (−4%) and crop GHG emissions (−10%) whereas the gap increases for nitrogen balance (+23%), livestock production (+6%) – MJ consumption (+16%) – GHG emissions (+5%) and crop MJ consumption (+5%). However, the indirect impact of crop–livestock organization doesn’t reverse the trend of trade-offs between objectives of sustainability determined by the direct impact of crop–livestock organization. As a conclusion, crop–livestock organization is a key factor that has to be taken into account when studying the sustainability of mixed crop–livestock systems.