Odor exploration behavior of the domestic pig (Sus scrofa) as indicator of enriching properties of odors.

Introduction and aim: Although the sense of smell in pigs is widely recognized as being highly developed, surprisingly little is known about their sensory ability. This study aimed to (a) identify which non-social odors pigs were able to detect and distinguish between, (b) investigate the types of behavior expressed when exploring odors and, (c) compare pigs' responses to the different odors to evaluate their interest in the odors. Methods: Growing pigs (N = 192) of crossbred commercial breeds were enrolled in the experiment (32-110 days of age, weighing 64.9 ± 10.1kg). Littermate pairs of opposite sex were tested in test pens with two odor insertion points in the pen wall, 55 cm apart. All pigs were habituated to the test pens and experimenters. Twelve odors were tested (eight essential oils and four synthetic perfumes) in groups of three odors, with each pig pair tested once with one set of three odors (all possible orders of the three odors were tested on 24 pairs in total), always against a non-odor control (demineralized water). In a test, each of the three odors were presented during three trials in a row (a total of 9 trials per test; trial duration: 1 min; inter-trial breaks: 2 min; total test duration: 25 min). Response variables included: duration of sniffing, feeding-related behavior (licking, biting and rooting), agonistic behavior (biting, displacement and pushing) and no approach of the odor or control, recorded throughout each 1-min odor presentation. Results: All pigs sniffed an odor less when repeatedly presented (LMM: all odors P < 0.05), and significantly longer at the subsequent presentation of a new odor [LMM (3rd vs. 1st presentations): P < 0.001]. Specific odor and odor type (essential oil vs. synthetic perfume) had no significant effect on sniffing duration. Overall, feeding-related behavior and agonistic behavior were expressed significantly more when pigs explored the odor compared with the control insertion point (Paired t-tests: P < 0.001), and specific odor only affected the expression of feeding-related behavior. Conclusion: Collectively, pigs express sniffing, agonistic, and feeding-related behavior when exploring odors, which suggests that pigs perceive odors of non-social origin as a resource. Odors may thus constitute relevant enrichment material for pigs.

Evaluation of Poultry Stunning with Low Atmospheric Pressure, Carbon Dioxide or Nitrogen Using a Single Aversion Testing Paradigm.

Low atmospheric pressure stunning (LAPS) has been suggested for use in poultry under 4 kg in the abattoir as a more humane alternative to carbon dioxide (CO2). However, there are currently no studies offering a direct comparison of the aversion between methods. Here, we trained adult female broiler breeders to relinquish a food reward by moving to another area of the gas chamber in response to aversive stimuli. They were then stunned and subsequently killed using single exposure to either CO2, N2, LAPS or medical air as a control. Birds exposed to CO2 relinquished the food reward the quickest and exhibited gasping and headshaking more than the other groups. LAPS resulted in the quickest time to loss of posture (LOP) and birds in the N2 group took the longest. Birds exposed to N2 displayed the longest duration of ataxia of any group; however, they did not show any wing-flapping prior to LOP, unlike the LAPS and CO2. Collectively these data demonstrate that both LAPS and N2 are less aversive to poultry than CO2 and may offer a significant welfare refinement for poultry killed for meat production.

Characteristics of Piglets Born by Two Highly Prolific Sow Hybrids.

High piglet mortality constitutes a welfare challenge in Danish organic pig production with almost one in three piglets dying before weaning. Piglet characteristics such as birth weight, rectal temperature and intrauterine growth restriction (IUGR) affect piglet survival. Due to differences in breeding goals, these characteristics may be expected to differ between sow hybrids. Thus, the aims of the present study were (1) to investigate piglet characteristics in two highly prolific sow hybrids and (2) to study to which extent the aforementioned characteristics affect piglet mortality. Forty-nine sows (22 DanBred and 27 Topigs Norsvin) were followed in their first two parities. Sows were housed outdoors and gave birth in huts. On day 1 postpartum ( pp ) piglets were individually marked, weighed, their rectal temperature was recorded and they were scored for IUGR. Weight and rectal temperature were recorded again 3 days pp. Principal component analyses were conducted to explore relationships among variables. Early piglet death grouped with IUGR, lower rectal temperature and weight on day 1 pp. Late mortality grouped with increasing litter size and DanBred hybrid. Whilst, Topigs Norsvin hybrid grouped with increasing rectal temperature day 3 pp, longer crown to rump length, higher weight and more teats on the sow. Results of the statistical analyses showed that Topigs Norsvin piglets were heavier 1 and 3 days pp (p < 0.001) compared to DanBred piglets. Furthermore, Topigs Norsvin piglets had a higher rectal temperature than DanBred on day 1 pp (p = 0.023). The risk of IUGR depended on an interaction between sow hybrid and parity (p = 0.023). DanBred sows gave birth to more piglets (18.2 ± 0.6) than Topigs Norsvin sows (15.7 ± 0.5, p = 0.003), however, DanBred sows had fewer teats than Topigs Norsvin sows. Weight on day 1 pp affected both the odds of stillbirth (p < 0.001) and live born death (p < 0.001). Lower rectal temperature day 1 pp (p < 0.001) increased the odds of live born death. In conclusion, the investigated hybrids differed in several piglet characteristics related to piglet mortality. Use of sows giving birth to heavier and fewer piglets in the litter may thus be a useful tool to reduce piglet mortality in pig production with outdoor farrowing.

Causes of preweaning mortality in organic outdoor sow herds.

The aims of the current study were to identify the causes of preweaning piglet mortality and investigate the effect of season and parity on the proportions of mortality causes. The study was conducted in nine herds over a one-year period, and dead piglets were collected for necropsy once during each of the four seasons. In total, 2672 piglets were necropsied of which 60% were live-born but died post-partum (pp), 32% were stillborn and 8% were euthanised. Intra-partum deaths accounted for 88% of stillborn piglets. Eighty-one percent of necropsied live-born piglets died within four days pp. Half of these had a body weight of <1 kg at the time of death. Crushed and starved piglets accounted for 83% and 9% of mortalities until day four pp, respectively. Thirty-three percent of the crushed piglets (age zero to four days) had not suckled prior to death and the proportion of these piglets was lower in summer compared to the remaining part of the year (OR = 0.6, 95% CI [0.5;0.9]) and was higher for multiparous sows compared to first parity sows (OR = 1.7, 95% CI [1.1;2.6]). In conclusion, sow parity and season affected the proportion of crushed piglets before four days of age. Furthermore, the group of crushed piglets was heterogeneous and consisted of both heavy well-fed piglets and small piglets with empty stomachs. Thus, management implementations to lower mortality may differ depending on sow parity, season and individual piglet characteristics.