The study was conducted at the farrowing facilities of the Swine Sector of the Fernando Costa campus city hall, at the University of São Paulo, Fernando Costa Campus, in Pirassununga, São Paulo. The site is at an altitude of 340 meters, south latitude of 21°80'00" and west longitude of 47°25'42", a Cwa climate with average annual temperatures of at least 13 and at most 31°C, according to Koppen (2011). This study was approved by the Ethics Committee on the Use of Animals CEUA No. 3758260116 of FZEA/USP.
The system used was semi-confinement in individual stalls for females and their litter, measuring 1.80 m wide, 4.20 m long, without cages, cemented floor with bed of sugarcane bagasse, an anti-crushing grid measuring 3.20 m long, drinking fountains for females and piglets and trough-type feeder for piglets and concrete ones for females. The 26 stalls included pens with a lamp to heat the piglets, separated from the stall by a 1.65 m high wall that allows access only to the piglets from an opening of 0.5 m high and 0.3 wide. In the stalls there were two hanging chains used as environmental enrichment for both ages (females and piglets).
The installation had a ceiling height of 2.70 m, a roof consisting of Marseille tiles and wood and concrete floor. The farrowing unit was divided into fan + sprinklers and control areas. In half of the experimental stalls three-propeller fans of the brand Ventisol of 60 cm were installed, with power of 1/5CV − 147W; and a maximum 1200 rpm. A fan was used for each two females and their respective litters at a height of 1.80 m from the floor, fixed on the pillar and a sprinkler/pulse irrigator with a TRAPP - DY-1013 rod for irrigation of water on the roof, fixed on the wall so that the water covered the entire roof, with one sprinkler used for each three females (stalls). The equipment was turned on at 6 am and turned off at 5 pm, remaining on continuously throughout the day. The control area consisted of stalls with a bed of sugarcane bagasse, environmental enrichments and the same physical structure of the fan + sprinklers area, but without the presence of fans and water sprinklers on the roof. Both environments had pens where the lamps remained on for the piglets throughout the day experimental period.
The facilities were separated by plastic canvas as a way to separate the environments, with a distance of 1.5 meters between treatments, preventing the circulation of air from the fans and humidity for the control treatment. The internal temperature in the control environment at 7 am was 23.7°C ± 0.64 with humidity of 74% ±1.64 and 26.2°C ± 1.23 at 1 pm with humidity of 70.4%± 1.68. In the fan + sprinklers environment, at 7 am the ambient temperature found was 21.1°C ± 0.66 with humidity of 81.9%±1.66 and at 23.5°C ± 0.85 at 1 pm with humidity of 85.7%±1.34.
Twenty-six females, F1, Landrace x Large White, of second birth order and their respective litters were used, totaling 281 piglets, 154 females and 127 males, with weaning at twenty-one days.
Nutritional management was performed in the morning and afternoon for females and piglets, with 7 kg per day for females and with progressive increase for piglets, starting with 200 grams in the first week and 500 g in the week of weaning.
On the second day after birth, piglets conventionally undergo routine practices, such as castration of males, tooth grinding, docking, Australian massaging and iron application. At this moment, the piglets were numbered with Arabic numerals using a non-toxic pen, in order of sex and weight, always starting with the heaviest males, followed by the lightest males, heaviest females and finally the lightest females. The animals were distributed in the following classes according to the average weight of the piglets at birth: light (≤ 1.36kg); medium (1.37 to 2.13kg) and heavy (> 2.14kg). Piglets were weighed at birth and three times a week until weaning and females at the entry of the farrowing phase (7 days before delivery) and at 21 days postpartum.
The litters were organized with 14 piglets. For the analysis of the hierarchy of piglets in the mammary gland, the total number of teats of each sow was obtained first. The minimum number of pairs found were seven and a maximum of nine. Sequentially, a classification was established for the location of the teats in the gland, in the craniocaudal direction, always initiated by the left side followed by the right side. The teats were numbered from 1 to 9 on the left side and from 10 to 18 on the right side. If there were no teats to follow the numbering, the digits remained in the sequence, following the classification described for the side of the gland (Fig. 1).
The registration of the teats where the piglets breastfed and the dispute for hierarchy was carried out from the second day of birth to weaning, three times a week, with one observation in the morning and another in the afternoon, by direct focal observation, in a strategic position, seeking less interference in suckling, performed by two observers previously trained for this function. The indication of the location of the teat where the piglet breastfed and the presentation of the behaviors were evaluated according to the side that the female remained in lateral decubitus position for suckling. Therefore, if she was lying on the left side, the observation and marking of the teat began at 1, if she was on the left side, it began at 10, always following the classification.
The classification into dominant, intermediate and subordinate was carried out by sow, aggregating different nomenclatures for males and females in relation to the weight category. Male piglets were defined by Arabic numerals, starting with the heaviest to the lightest, and females by capital letters, in the same order of weight. The imposition of the hierarchy was attributed to the registration of the occurrences of agonistic behaviors among the piglets, by direct observation with a sampling route of the behavior (MARTIN and BATESON, 1993), in a specific ethogram for the registration. For each agonistic interaction, the sender, receiver, the type of behavior performed by the aggressor and the counter-reaction of the victim were identified.
The record of the behavior performed during the imposition and definition of the suckling hierarchy was obtained concomitantly with the observation of the location of the animal to the classified teat, using collection routes by the direct focal method and continuous or instantaneous recording, according to the observed behavior (MARTIN and BATESON, 1993).
The evaluation of the hierarchy of piglets in the mammary gland was based on the occurrences of agonistic interactions observed between the piglets during the first stage of the study. The binary matrix of dominance was constructed based on the occurrences of agonistic interactions identifying the individuals of the lines that dominate individuals of the columns (DE VRIES, 1995). Based on the number of circular triads (d), the Kendall consistency coefficient (K) was used to test the linearity of the dominance relationships, which should be significantly stronger than expected, in an attempt to organize individuals into a linear hierarchy (APPLEBY, 1983; DE VRIES, 1998). For the division into three levels of hierarchy, a methodology adapted from (ECCLES and SHACKLETON, 1986 adapted from SOARES, 2015) three levels of hierarchy (dominant, intermediate, subordinate) was used, using fourteen animals per litter. Data were analyzed by variance with fixed effects of treatment (fan + sprinklers or control), time of day (morning or afternoon), sex (male and female), weight category and their interactions. For the significant effects, the means were compared by t test at 5%.