J. Anim. Sci. 2006. 84:469-480
© 2006 American Society of Animal Science
Effects of season and breed on the feeding behavior of multiparous lactating sows in a tropical humid climate1,2
J. L. Gourdine*,
J. P. Bidanel
,
J. Noblet
and
D. Renaudeau*,3
* Station de Recherches Zootechniques, Institut National de la Recherche Agronomique (INRA), 97170 Petit Bourg, Guadeloupe, F.W.I, France;
and
Station de Génétique Quantitative et Appliquée, 78352 Jouy-en-Josas, France; and
and
UMR Systèmes d’Elevage, Nutrition Animale et Humaine, 35590 Saint-Gilles, France
 |
Abstract
|
|---|
The effects of breed and season on performance and feeding behavior were studied during 76 lactations in multiparous Large White (LW; n = 17) and Creole (CR; n = 23) sows reared in a humid tropical climate. The experiment was conducted in Guadeloupe (French West Indies, latitude 16°N, longitude 61°W) between May 2002 and July 2004. Average daily ambient temperature was greater during the hot season than during the warm season (26.0 vs. 23.8°C), but relative humidity was similar in both seasons (85% on average). The daily fluctuations of ambient temperature and relative humidity were similar for both seasons. At farrowing, BW was lower (187 vs. 265 kg) and backfat thickness was greater (40 vs. 22 mm) in CR than in LW sows (P < 0.01). Sows were offered feed ad libitum between the fifth and the 26th day of lactation. There was a breed x season interaction (P < 0.05) for ADFI. During the hot season the reduction of ADFI was more pronounced in LW than in CR sows (–1,100 vs. –300 g/d). Irrespective of breed and season, the daily number of meals was 9.0. The meal size and the rate of feed intake were greater in LW than in CR sows, respectively (555 g and 153 g/min vs. 390 g and 83 g/min; P < 0.01). The diurnal proportion of ADFI was greater in CR than in LW sows (0.60 vs. 0.41; P < 0.01). The reduction in ADFI in LW sows during the hot season was mainly related to a reduction in feed intake during the day rather than during the night (–1.3 vs. –0.2 kg; P < 0.01). Duration of standing was not affected by breed or season, and it averaged 120 min/day. This study confirms the negative effect of the hot season on feeding behavior of lactating sows. It also suggests a better acclimation to daily high temperatures and a greater heat tolerance in CR compared with LW sows, at least for eating behavior.
Key Words: breed • feeding behavior • feed intake • sow • tropical climate
|
INTRODUCTION
|
|---|
To understand the control and regulation of feed intake, and to establish an appropriate feeding strategy, it is important to study factors affecting feeding behavior. Voluntary feed intake (VFI) is influenced by numerous factors, many of which are environmental factors. Numerous studies have shown that ambient temperature plays a critical role in VFI. Indeed, VFI is reduced when lactating sows are exposed to ambient temperatures above the evaporative critical temperature (estimated to be below 22°C by Quiniou and Noblet, 1999
). In tropical conditions, because of opened or semiopened buildings, animals are more directly exposed to daily variation of the outside climatic conditions (Renaudeau et al., 2003).
The appetite is also affected by sow characteristics such as breed. Little information is available on the effect of breed on VFI of lactating sows (Sinclair et al., 1999). The Creole pig (CR) is the most important local Caribbean pig breed. It is characterized by a low reproductive and growth performance and an apparently good adaptation to harsh climatic conditions encountered in tropical humid areas (Canope, 1982). For this reason, the CR breed was introduced in our experimental unit to study genetics of heat tolerance in pigs. Moreover, considering its high adiposity, the CR pig is an interesting model to study the relationships between body composition and feeding behavior. Furthermore, little information is available on the components of feeding behavior of lactating sows raised in tropical conditions (Renaudeau et al., 2003) and their variations with season or breed.
The aim of the current study was to evaluate the effect of season on the performance and feeding behavior of Large White (LW) and CR lactating sows. Lactation performance and thermoregulatory aspects were previously studied during 179 lactations (Gourdine et al., 2006). The present paper concerns the feeding behavior aspects obtained on a subgroup of 76 lactations.
|
MATERIALS AND METHODS
|
|---|
Experimental Design
A total of 179 lactations on 71 multiparous sows (30 CR and 41 LW) divided in 24 successive groups of 9 to 11 LW and CR sows were used in a trial conducted at the INRA experimental facilities in Guadeloupe, French West Indies (latitude 16°N, longitude 61°W). Feeding behavior was recorded only on 17 groups of sows and on 3 to 6 multiparous sows per group; a total of 76 lactations was measured on 23 CR and 17 LW sows. The data covered the period between May 2002 and July 2004. Two seasons were determined a posteriori from climatic measurements: a warm season between November and April (23.8 ± 0.8°C) and a hot season between May and October (26.0 ± 0.5°C). Relative humidity was comparable for both seasons (85%).
Animal Management
An authorization to perform an experiment on living animals was given by the French Ministry of Agriculture. A detailed description of diet composition and management of animals was given in a previous paper (Gourdine et al., 2006). The lactation diet contained 14 MJ of DE per kg, 17.1% CP, and 0.92% crude lysine and was formulated using corn, wheat middlings, and soybean meal in order to meet or exceed amino acids requirements (NRC, 1998). During lactation, feed was distributed once per day and at the same time for both breeds (i.e., between 0700 and 0800) to minimize confounding of eating behavior. Sows had free access to water provided by a low-pressure nipple drinker. Feed allowance was progressively increased by 1 kg each day until d 5 (from 1 kg at d 1 to 5 kg at d 5). Sows were offered feed ad libitum from d 5 to 26. The day before weaning (d 27), sows were allocated 3 kg of feed, which allowed the assumption that they were weighed at weaning with an empty digestive tract.
Measurements
The sows were weighed within 24 h after farrowing and at weaning. Backfat thickness was measured ultrasonically (Agroscan; E.C.M., Angoulême, France) at 65 mm from the midline of the back beside the shoulder and the last rib on each flank, after farrowing and at weaning. Piglets were individually weighed at birth and every week until weaning (d 28). Electronic measurements of feed intake were made using a trough connected to a load cell and a computer. In addition to the electronic measurement of feed intake, every morning refusals were manually collected and weighed at the same time for both breeds, between 0600 and 0800; and the daily feed intake was determined as the difference between feed allowance and the refusals collected the next morning.
Standing and sitting duration (later named standing) was recorded using an infrared barrier located in the middle of the crate. The equipment did not allow standing and sitting to be distinguished. Individual feeding behavior and standing duration were recorded continuously during the ad libitum feeding period (i.e., from d 5 to 26). Each pen was equipped with a trough weighed continuously with a load cell. A visit corresponded to an unsteady period of the trough detected by the load cell. Signals from the infrared barrier and the load cell were continuously recorded.
Calculations and Statistical Analyses
Feed consumption per visit was calculated as the difference between the amounts recorded just before and just after the visit. For each visit, feed consumption lower than 20 g was considered an artifact due to the movements of the sows on the slatted floor, and it was not taken into account for further calculations. Because of electronic problems on the load cells and some power failures, 3 and 2% of daily recordings from CR and LW sows, respectively, were excluded from our study. Ingestion time of feed per visit corresponded to the difference between the time at the end and at the beginning of the visit. Sows exhibit short pauses during a meal, and these short intervals between visits (IV) must be differentiated from the longer ones between 2 different meals. For this purpose, a meal criterion (MC) defined as the maximum length of within-meal intervals between 2 successive visits was estimated. When 2 successive visits were separated by an interval shorter than MC, visits were merged into the same meal. The MC was estimated for each sow in each ad libitum days of lactation, using the log survivorship curve technique described by Bigelow and Houpt (1988), according to a nonlinear regression model (NLIN Procedure, SAS/STAT, Version 8.1, SAS Inst., Inc., Cary, NC). MC is defined as the breakpoint of the following model:
 |
in which y = log (1 – cumulative frequency of IV).
According to the nonlinear regression, 92 and 95% of CR and LW individual MC, respectively, were less than or equal to 5 min (Figure 1). Hence, from the calculated value of MC (i.e., 5 min), the following daily parameters of feeding behavior were calculated for each sow: number of meals per day, feed intake per day (g), total ingestion time per day (i.e., total duration of all the visits, min), total consumption time of feed (i.e., sum of the ingestion time and within-meal interval, min), rate of feed intake (i.e., total feed intake/total ingestion time, g/min) and feed intake per meal (g).
View larger version (12K):
[in this window]
[in a new window]
|
Figure 1. Cumulative distribution function of meal criterion of lactating Creole (dashed line) and Large White (thick line) sows; more than 90% of individual meal criteria were less than 5 min in both breeds.
|
|
When a lactation occurred over 2 successive seasons, the lactation of a sow was attributed to the season in which the sow spent the largest number of days in lactation. Two parity groups were constructed, namely 2 to 4 (n = 38) and 
4 (n = 38). The effects of breed, season of lactation and first order interaction, the effect of group within the effect of season and the effect of parity number, and the random sow effect were tested on the mean lactation performance of sows and their litter, through an analysis of variance using the MIXED Procedure of SAS/STAT (Version 8.1, SAS Inst., Inc., Cary, NC). During the ad libitum period (i.e., between d 5 and 26), a total of 1,592 daily measurements of feeding behavior parameters were measured on 40 sows. These data pooled by sows on a daily basis were analyzed according a linear mixed model variance using the MIXED Procedure of SAS/STAT, including the fixed effects of breed, season of lactation and their interaction, the effect of group within season, the effect of parity number, and a random sow effect. Residual values were computed from the preceding models (without the random sow effect), and residual correlations between lactating performance and mean feeding behavior components were calculated using the CORR Procedure of SAS/STAT. The mean feeding behavior components per sow over the ad libitum lactation period were also calculated according to photoperiod (day vs. night) and were analyzed according to a linear mixed model including the fixed effects of breed, season, photoperiod, and their interactions, the effect of group within season and the random effect of sow. Finally, a mixed model was used to examine the fixed effects of breed, season, lactation stage, and their interactions on the average hourly sow feed intake during lactation. To take into account the correlations between repeated measurements on the same animal, after testing several time-series covariance structures, a heterogeneous first-order autoregressive structure was selected for both breeds based on Akaike’s Information Criterion (Littel et al., 1996). According to the hourly feed intake and hourly ambient temperature patterns, 4 periods were considered: 0600 to 1100 (first peak), 1100 to 1500 (heat stress), 1500 to 2000 (second peak), and the remaining hours. Average feed intake during each period was considered with respect to the total daily feed consumption and was analyzed according to a linear mixed model including the fixed effects of breed, season of lactation and their interaction, the effect of group within season, and the random effect of sow.
|
RESULTS
|
|---|
The daily variations of ambient temperature and relative humidity (RH) are presented in Figure 2. The daily fluctuation of ambient temperature was similar for the warm and the hot seasons with minimum and maximum values reached at 0600 (21.7 and 24.2°C, respectively) and at 1300 (28.2 and 31.9°C, respectively). Average daily RH and daily variation of RH were similar for both seasons. Contrary to ambient temperature, RH was greatest at about 0600 and lowest at about 1300 (i.e., 97 and 70%, respectively). According to data from the French national meteorological institute (Météo France), the length of the diurnal (here defined as daytime) period was slightly greater during the hot than during the warm season (12.33 vs. 11.67 h).
View larger version (21K):
[in this window]
[in a new window]
|
Figure 2. Daily fluctuations of ambient temperature (dashed line with  in warm season and dashed line with x in hot season) and relative humidity (solid line for warm season and dashed line for hot season). The values correspond to the mean hourly ambient temperature and relative humidity between May 2002 and July 2004. Warm season: November 2002 to April 2003 and November 2003 to April 2004; hot season: May 2002 to October 2002, May 2003 to October 2003, and May 2004 to July 2004. Diurnal period: 0620 to 1800 and 0550 to 1810, for the warm and hot seasons, respectively (Météo France data).
|
|
Whatever the breed, ADFI was depressed (P < 0.01) in the hot season during the whole lactation period and over the ad libitum intake period [i.e., from d 5 to 26 (3.46 vs. 4.15 kg and 3.90 vs. 4.67 kg, respectively; Table 1)]. Lactation BW loss was greater (P < 0.05), and average piglet BW at weaning tended to be lower (P = 0.09) during the hot than during the warm season (19 vs. 12 kg and 6.7 vs. 7.2 kg, respectively). The breed effect on ADFI varied whether expressed per kg/d or per g·kg0.75·d–1; ADFI was lower in CR than in LW when it was expressed in kg (3.1 vs. 4.5 kg; P < 0.01), but it was similar when it was expressed in g/kg0.75 (i.e., 69 on average). After farrowing, average BW was lower and average backfat thickness was greater in CR than in LW sows (187 vs. 265 kg, and 40 vs. 22 mm; P < 0.01). However, BW loss during lactation was not affected by breed (15 kg on average; P > 0.20), whereas the backfat thickness loss was greater in CR sows (4.4 vs. 2.8 mm; P > 0.05). Litter size and average piglet BW at weaning were lower in CR than in LW sows (7.9 vs. 9.2 and 6.3 vs. 7.6 kg, respectively; P < 0.01). In agreement with these results, milk production was greater in LW sows than in CR sows (6.04 vs. 4.30 kg/d; P < 0.01). However, this difference was no longer significant when milk production was considered with respect to metabolic BW (P > 0.30). The interaction between breed and season was significant only for ADFI and tended to be significant for average piglet BW at weaning: during the hot season, the reduction of ADFI (–0.30 vs. –1.1 kg/d; P < 0.05) and piglet BW at weaning (–100 vs. –900 g; P = 0.10) were attenuated in CR as compared with LW sows.
According to the analysis of variance, the number of meals was not affected (P > 0.60) by breed or season and averaged 9.0 ± 1.4 meals/d (Table 2). As a consequence and in connection with the seasonal effect on ADFI, the meal size was reduced during the hot season (428 vs. 517 g/meal; P < 0.01). The rate of feed intake was not affected by season (P = 0.85), and the total ingestion time was reduced (P < 0.05) during the hot season by about 7 min/d. Meal size was lower in CR than in LW sows (390 vs. 555 g; P < 0.01), and ADFI during the ad libitum period was then lower in CR than in LW sows (3.0 vs. 4.6 kg/d; P < 0.01). However, no significant between-breed differences were found when meal size was considered with respect to metabolic body weight (BW0.75; P > 0.40). The rate of feed intake expressed as g/min or as g·kg–0.75·min–1 was lower (83 vs. 155 g/min or 1.6 vs. 2.4 g·kg–0.75·min–1; P < 0.05), and the daily ingestion time was greater (39 vs. 32 min/day; P < 0.10) in CR than in LW sow. Irrespective of breed and season, standing activity averaged 120 min/d.
View this table:
[in this window]
[in a new window]
|
Table 2. Effect of season1 and breed on feeding behavior and duration of standing of lactating sows between d 5 and 26 postpartum (least squares means)
|
|
Meal size and rate of feed intake were lower during the night (430 vs. 530 g/meal and 110 vs. 130 g/min; P < .05; Table 3). The duration of standing activity was also lower during the nocturnal period (50 vs. 70 min; P < 0.01). Most of the other feeding behavior parameters were affected by a breed x photoperiod interaction. In LW sows, the number of meals and total feed ingestion time were greater during the nocturnal period than the diurnal period (5.2 vs. 3.7 meals/d and 20 vs. 12 min/d; P < 0.01). In contrast, the number of meals and feed ingestion time in CR sows were greater during the diurnal period (5.4 vs. 3.8 meals/d and 24 vs. 16 min/d; P < 0.01). Consequently, the diurnal proportion of ADFI was greater in CR than in LW sows (60 vs. 41%; P < 0.01). During the hot season, the reduction in diurnal feed intake was more pronounced in LW than in CR sows (–1,260 vs. –220 g; P < 0.05).
View this table:
[in this window]
[in a new window]
|
Table 3. Effect of season,1 breed, and light pattern on feeding behavior and duration of standing of lactating sows between d 5 and 26 postpartum (least squares means)
|
|
From the comparison of the hourly feed intakes or variations from hour to hour (between 2400 and 0200), the nycthemeral pattern of feed intake in LW sows (Figure 3) or in CR sows (Figure 4) peaked twice a day in both seasons. The first and the second peaks were observed between 0200 and 0900 and between 1500 and 2100, respectively. On average, about 54 and 32% of the meals in LW sows occurred during the first and the second peaks, respectively. The corresponding values in CR sows were 40 and 35%. According to Figure 5, the hourly feed intake was also affected by breed during the warm season. The early morning peak of feed intake took place at about the same time for both breeds, but its size was smaller in CR than in LW sows. Indeed, the proportion of feed intake between 0600 and 1100 was greater in LW than in CR sows (32.8 vs. 22.4%; P < 0.05; Table 4). In addition, the second peak took place earlier in CR than in LW sows (between 1200 and 1900 and between 1500 and 2100, respectively). Between 1100 and 1500, the proportion of feed intake was greater in CR than in LW sows (18.3 vs. 8.7%; P < 0.05), but no difference was found between 1500 and 2000 (29%; P > 0.40; Table 4). During the hot season, the nycthemeral pattern of feed intake was also affected by breed (Figure 6); the proportion of feed intake between 0600 and 1100 did not differ between breeds (34% on average; P > 0.40; Table 4). The second peak took place earlier in CR sows (Figure 6); the proportion of feed intake between 1500 and 2000 did not differ between breeds (26% on average; P > 0.50). During the hottest period of the day (i.e., 31.9°C between 1100 and 1500), CR sows consumed more feed (630 vs. 246 g; P < 0.05) than LW sows. The amount of feed represented 12.8 and 3.0% of the total daily feed intake, in CR and LW sows, respectively (Table 4).
View larger version (24K):
[in this window]
[in a new window]
|
Figure 3. Effect of season ( = warm season; • = hot season) on the hourly feed intake in lactating Large White sows; each point is the least squares mean of 12 sows in the warm season and 14 sows in the hot season; daily ambient temperature (dashed line = warm season; solid line = hot season); * = times when hourly feed consumption was affected (P < 0.05) by season.
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Figure 4. Effect of season ( = warm season; • = hot season) on the hourly feed intake in lactating Creole sows; each point is the least squares mean of 12 sows in the warm season and 14 sows in the hot season; daily ambient temperature (dashed line = warm season; solid line = hot season); * = times when hourly feed consumption was affected (P < 0.05) by season.
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Effect of season1 and breed on feeding behavior of lactating sows between d 5 and 26 postpartum (least squares means)
|
|
Residual correlations (2 breeds together) among lactation performance of sows and their litter and daily feeding behavior components are given in Table 5. Residual correlations describe correlations between residual values of the performance adjusted by the fixed effects of breed, season of lactation and their interaction, the effect of group within season, and the effect of parity number. Litter growth rate was positively associated (P < 0.01) with the number of meals (r = 0.45), ADFI (g·kg–0.75; r = 0.52) and ingestion time (r = 0.54), whereas ADFI was logically associated with ingestion time (r = 0.84; P < 0.01). Average backfat thickness was negatively associated with meal size and ingestion time (r = –0.21 and r = –0.26, respectively). According to the regression analysis, the rate of feed intake increased linearly with metabolic BW (Figure 7); each kilogram increase in metabolic BW resulted in a 3.7 g/min increase in the rate of feed ingestion (R2 = 0.60).
View this table:
[in this window]
[in a new window]
|
Table 5. Residual correlation between the sow’s body condition after farrowing, litter growth rate, and components of feeding behavior (the 2 breeds together)1
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 7. Relationship between rate of feed intake (RFI) and metabolic BW [RFI = – 97.3 (±18.4) + 3.74 (±0.34) x BW0.75; R2 = 0.60; = Large White; • = Creole].
|
|
|
DISCUSSION
|
|---|
The 40 sows used in the current study (76 lactations) were a part of a larger group of 71 multiparous sows (179 lactations) for which lactation performance and thermoregulation aspects were given in a previous paper (Gourdine et al., 2006). For the whole lactation period, sows and their litter performance in the present trial were similar to the values obtained in the total study. For instance, in the current study, ADFI and piglet growth rate during the whole lactation were 4.5 kg/d and 220 g/d in LW sows, and 3.1 kg/d and 192 g/d in CR sows, respectively (vs. 4.3 and 216 g/d in LW sows, and 3.2 kg/d and 190 g/d in CR sows; Gourdine et al., 2006). According to these observations, we consider that the subgroup of 37 sows used to measure feeding behavior is representative of the group of 71 sows used in the experiment.
Effect of Season on Feeding Behavior of Lactating Sows
In agreement with studies performed in lactating sows of conventional breeds raised in temperature-controlled rooms (Dourmad, 1993; Quiniou et al., 2000a; Renaudeau et al., 2002) or in tropical climates (Renaudeau et al., 2003), we have also measured a bimodal feeding pattern with a first peak in the morning and a second one in the late afternoon. As observed in growing pigs (Feddes et al., 1989), these results confirm that feeding pattern in lactating sows is mainly driven by light change. In our conditions for the morning peak, however, this photoperiod effect was partly confounded with the presence of the staff for the collection of feed refusals and feed distribution. In addition and in agreement with Renaudeau et al. (2003) in LW lactating sows, the feeding pattern is also affected by season and by the daily variation of ambient temperature and RH. In fact, during the hot season, the hourly feed consumption tends to coincide with the minimum daily temperature during the morning and the evening periods (Figure 3
). In other words, the sows reduced their feed consumption during the hottest periods of the day (i.e., between 1000 and 1500), and this decrease was partly counterbalanced by a greater feed intake during the cooler periods of the day (i.e., between 1700 and 2400, and between 0100 and 0800). One consequence of this adaptation is a change in the nycthemeral pattern of feed intake; in our study, more than one-half of the ADFI of LW sows occurred during the night period. This result is in accordance with previous values obtained in LW sows and in our experimental station (Renaudeau et al., 2003), but in contradiction with results obtained also in LW sows by Quiniou et al. (2000a) and Renaudeau et al. (2002) with temperatures kept constant during the day. These authors reported an increase in the diurnal proportion of feed intake with the ambient temperature. These results suggest that the proportion of diurnal feed intake also depends on whether daily temperature is kept constant or fluctuates. Quiniou et al. (2000b) reported that feed intake of lactating LW sows was less affected by high temperatures when ambient temperature fluctuated between 25 and 33°C than when it was kept constant at 29°C (4.7 vs. 3.5 kg). In the current study, even if the lower feed intake during the hot periods of the day was counterbalanced by a greater feed intake during the cool periods of the day, daily feed intake is critically lower than the value reported by Quiniou et al. (2000b; i.e., 4.2 vs. 4.7 kg) between 25 and 33°C for uncontrolled relative humidity ranging between 50 and 60%. It can be suggested that, in tropical humid conditions, feeding nycthemeral pattern is affected by photoperiod, temperature, and humidity (alone or in interaction with ambient temperature). In summary, the feeding pattern of lactating sows reared in tropical conditions is driven by both light intensity and daily temperature and humidity changes.
To our knowledge, there is limited information on the meal criterion in lactating sows. The meal criterion in the current study (i.e., 5 min) was similar to the study of Renaudeau et al. (2003) in the same experimental facilities but greater than the 2 min value chosen by Quiniou et al. (2000a) and Renaudeau et al. (2002), using a similar electronic trough but in other facilities. According to Renaudeau et al. (2003), this discrepancy between the 2 sets of results could be partly explained by the fact that the trough was fixed directly on the slatted floor in the tropical conditions, which made it more sensitive to sow’s movements and more cancelled visits (<20 s). Moreover, it has been shown that an overestimation of the meal criterion has little effect on the feeding pattern and the number of meals, whereas an underestimation results in an important increase of the number of meals (Dourmad, 1993; Renaudeau et al., 2002). Subsequently, results in the current study can be compared with literature results usually obtained with shorter meal criteria.
Over the ad libitum feeding period, the daily number of meals (i.e., 9.0 meals/d) is consistent with the value obtained by Renaudeau et al. (2003; i.e., 8.4 meals/d). In accordance with these authors, the reduction in ADFI during the hot season was related to a reduction in meal size, whereas the number of meals remained constant. In fact, Quiniou et al. (2000a) using several levels of ambient temperatures (18 to 29°C) suggested that the decrease of feed intake is related first to a reduction of meal size, second to a reduction of both meal size and frequency as temperature increases. In the current study, the rate of feed intake in LW sows is in agreement with the results of Quiniou et al. (2000a) and Renaudeau et al. (2002) in controlled hot conditions, and Renaudeau et al. (2003) in tropical conditions. In agreement with these authors, no significant effect of season on rate of feed intake was observed in the current study. Consequently, the lower total ingestion time measured during the hot season was directly connected to the reduced ADFI. However, time spent in sitting or in standing positions was not affected by season. As described in a recent literature review (Renaudeau et al., 2005a), it can be estimated that about 50% of standing time in lactating sow is dedicated to feed consumption. In addition, Renaudeau et al. (2002) found an increase in water consumption in hot conditions. Subsequently, even if drinking behavior was not recorded in the current study, it can be suggested that the increase in drinking activity during the hot season may explain the lack of season effect on total standing activity.
Effect of Breed on Feeding Behavior of Lactating Sows
To our knowledge, little information is available on the effect of breed on feeding behavior in lactating sows. Comparing a conventional breed of sows (i.e., LW) to fat-type sows (i.e., CR), our results show a significant breed effect on feeding behavior. In particular, CR sows consumed significantly less feed than LW sows. According to the lack of breed effect on ADFI and on milk production when it was expressed per g·kg–0.75·d–1, this between-breed difference would mainly be explained by metabolic body size differences (Gourdine et al., 2006).
In the current study, the reduction in the appetite of CR sows was explained by a reduction in meal size, which was not compensated by an increase in the number of meals. Therefore, no breed difference was found when meal size was considered with respect to metabolic BW. Consequently, as shown for ADFI, the lower meal size of CR sows would be partly due to the between-breed difference in BW. In contrast, in a between-breed comparison involving 80 group-housed growing pigs (40 LW and 40 CR) within a similar BW range (i.e., from 45 to 90 kg), Renaudeau et al. (2005b) found a greater meal size in CR pigs associated with a reduced number of meals. Similar results were also found by Hyun et al. (2001) in a comparison involving Meishan and LW group-housed pigs. Such a breed difference could be related to the housing conditions driving social competition phenomena (Nielsen and Lawrence, 1993). However, in individually kept Meishan and LW growing pigs, Quiniou et al. (1999) also found a reduced number of meals counterbalanced by a greater meal size in fat Meishan than in lean LW pigs. This breed effect on feeding behavior may indicate differences in maturity (mouth and gut sizes) and body composition: Meishan or CR pigs are older and closer to mature BW than conventional breeds at the same BW. As reviewed by Woods et al. (1998), the regulations of meal number and meal size are influenced by many factors, and their impacts are directly or indirectly modulated by the amount of adiposity in the body. Moreover, the relationship between body composition and feeding behavior seems to be different between growing pigs and lactating sows.
In the current study, a significant effect of breed on the rate of feed intake was observed. As reviewed by Quiniou et al. (2000c), the rate of feed intake proportionally increased with BW when growing pigs and lactating sows were considered together (i.e., +5 g/min per extra 10 kg increase in BW). In the current study, a linear relationship was also found, corresponding to an increase of 9 g/min per extra 10 kg of BW (Figure 7). The increase in the rate of feed intake with BW could be explained by the increase of mouth size (ingestion capacity) with BW (between and within breeds). Moreover, in accordance with the results of Renaudeau et al. (2005b) in growing pigs, CR had a lower rate of feed intake even when it was expressed with respect to metabolic BW. These results suggest that apart from BW, other factors such as body composition and mouth and gut characteristics may also affect the rate of feed intake of lactating sows.
Breed to Season Effect on the Feeding Behavior of Lactating Sows
Our study shows that nycthemeral feeding pattern is influenced by breed (Figures 5 and 6). In CR sows, the second peak took place earlier than in LW sows, so that the proportion of ADFI consumed during the hottest period of the day was greater in CR than in LW sows. These results suggest that CR sows may have a better ability to tolerate high ambient temperature than LW sows. This was confirmed by the lower rectal temperature value in CR than in LW sows during lactation (38.8 vs. 39.1°C on average, Gourdine et al., 2006). The attenuated effect of hot season on ADFI in CR sows was mainly related to the between-breed difference in thermotolerance. The superior thermoregulatory ability of CR as compared with LW was also reported in growing pigs (Renaudeau, 2005). In 2 or 3 consecutive thermal challenge experiments involving 12 LW and 12 CR growing pigs, the author reported that in response to heat challenge (from 22 to 34°C between 0900 and 1500, and from 34 to 22°C between 1500 and 2100), the upper critical temperature, above which rectal temperature increases, was greater, and the increase of respiration rate and rectal temperature above 30°C were lower in CR than in LW pigs. The better ability to tolerate heat in CR breed may be the result of a reduced heat production, an increased capacity for loss of heat in the environment or some combination of both. Clearly, the low metabolic rate resulting from the reduced BW in CR sows can be considered as a major contributing factor to thermotolerance. Moreover, it is not excluded that CR sows may have greater ability to lose heat. Nonevaporative heat loss is partly dependent on cutaneous heat conductivity of the animal. When ambient temperature exceeds the lower critical temperature, pigs increase their heat conductivity by changing the blood flow to skin vessels. Berbigier (1975) showed better nonevaporative heat loss capacity in crossbred CR x LW pigs than in purebred LW pigs. However, the results were obtained on a low number of observations (4 piglets), and further investigations are needed to understand the biological mechanisms that make the CR breed better adapted to heat conditions.
|
IMPLICATIONS
|
|---|
The present results confirm the effect of high ambient temperature and humidity fluctuations on feed intake and feeding behavior of lactating sows reared in tropical climatic conditions; a larger fraction of the daily intake was consumed during the night and during the cooler periods of the day. This also means that to maximize daily feed intake, feed should be made available at these preferred periods of feed consumption. Moreover, our results indicate that feed intake and feeding behavior in lactating sows are also affected by breed, principally to body weight differences between breeds. Finally, in our experimental conditions a better acclimation to daily high temperature and a better heat tolerance were found in Creole sows, in the sense that unlike Large White sows, they managed to eat under heat stress. Further studies are then required to characterize the adaptation of lactating sows exposed to fluctuating temperature and the mechanisms of heat tolerance.
|
Footnotes
|
|---|
1 The financial support of the Guadeloupe Region and the European Union social funds is gratefully acknowledged. 
2 The authors also wish to thank C. Anaìs, K. Benony, B. Bocage, M. Giorgi, G. Gravillon, A. Racon, F. Silou, and J.-L. Weisbecker for their technical assistance.
3 Corresponding author: renaudeau{at}antilles.inra.fr
Received for publication May 24, 2005.
Accepted for publication October 21, 2005.
|
LITERATURE CITED
|
|---|
Berbigier, P. 1975. Echanges thermiques au niveau de la peau des porcelets élevés en climat tropical 1. Influence des conditions climatiques et de la race. Ann. Zootech. 24:423–432.
Bigelow, J. A., and T. R. Houpt. 1988. Feeding and drinking patterns in young pigs. Physiol. Behav. 43:99–109.[Medline]
Canope, I. 1982. Etude des interactions entre le type génétique et le régime alimentaire chez le porc en milieu tropical humide. Ph.D. Diss. Institut National Polytechnique de Toulouse, France.
Dourmad, J. Y. 1993. Standing and feeding behaviour of the lactating sow: Effect of feeding level during pregnancy. Appl. Anim. Behav. Sci. 37:311–319.
Feddes, J. J. R., B. A. Young, and J. A. DeShazer. 1989. Influence of temperature and light on feeding behavior of pigs. Appl. Anim. Behav. Sci. 23:215–222.
Gourdine, J. L., J. P. Bidanel, J. Noblet, and D. Renaudeau. 2006. Effects of breed and season on performance of lactating sows in a tropical humid climate. J. Anim. Sci. 84:360–369.[Abstract/Free Full Text]
Hyun, Y., B. F. Wolter, and M. Ellis. 2001. Feed intake pattern and growth performance of purebred and crossbred Meishan and Yorkshire pigs. Asian-Aus. J. Anim. Sci. 14:837–843.
Labroue, F., L. Maignel, P. Sellier, and J. Noblet. 1999. Consommation résiduelle chez le porc en croissance alimenté à volonté: Méthode de calcul et variabilité génétique. Journ. Rech. Porcine Fr. 31:167–174.
Littel, R. C., G. A. Milliken, W. W. Stroup, and R. D. Wolfinger. 1996. SAS System for Mixed Models. 1st ed. SAS Inst., Cary, NC.
Nielsen, B. L., and A. B. Lawrence. 1993. The effect of group size on the behaviour and performance of growing pigs using computerised single-space feeders. Pig News Inf. 14:127N–129N.
Noblet, J., and M. Etienne. 1989. Estimation of sow milk nutrient output. J. Anim. Sci. 67:3352–3359.[Abstract/Free Full Text]
NRC. 1998. Nutrient Requirement of Swine. 10th ed. Natl. Acad. Press, Washington, DC.
Quiniou, N., S. Dubois, Y. Le Colzer, J.-F. Bernier, and J. Noblet. 1999. Effect of growth potential (body weight and breed/castration combination) on the feeding behaviour of individually kept growing pigs. Livest. Prod. Sci. 61:13–22.
Quiniou, N., and J. Noblet. 1999. Influence of high ambient temperatures on performance of multiparous lactating sows. J. Anim. Sci. 77:2124–2134.[Abstract/Free Full Text]
Quiniou, N., D. Renaudeau, S. Dubois, and J. Noblet. 2000a. Influence of high ambient temperatures on feed intake and feeding behaviour of multiparous lactating sows. Anim. Sci. 70:471–479.
Quiniou, N., D. Renaudeau, S. Dubois, and J. Noblet. 2000b. Effect of diurnally fluctuating high ambient temperatures on performance and feeding behaviour of multiparous lactating sows. Anim. Sci. 71:571–575.
Quiniou, N., D. Renaudeau, A. Collin, and J. Noblet. 2000c. Effets de l’exposition au chaud sur les caractéristiques de la prise alimentaire du porc à différents stades physiologiques. INRA Prod. Anim. 13:233–245.
Renaudeau, D. 2005. Effects of short-term exposure to high ambient temperature and relative humidity on thermoregulatory responses of European (Large White) and Caribbean (Creole) restrictively fed growing pigs. Anim. Res. 54:81–93.
Renaudeau, D., M. Giorgi, F. Silou, and J. L. Weisbecker. 2005b. Influence du climat tropical et du type génétique sur les performances zootechniques et le comportement alimentaire du porc en croissance entre 45 et 90 kg. Journ. Rech. Porcine Fr. 37:259–266.
Renaudeau, D., J. L. Gourdine, N. Quiniou, and J. Noblet. 2005a. Feeding behaviour of lactating sows in hot conditions. Pig News Inf. 26:17N–22N.
Renaudeau, D., N. Quiniou, S. Dubois, and J. Noblet. 2002. Effect of high ambient temperature and dietary protein level on feeding behaviour of multiparous lactating sows. Anim. Res. 51:227–243.
Renaudeau, D., J. L. Weisbecker, and J. Noblet. 2003. Effect of season and dietary fibre on feeding behaviour of lactating sows in a tropical climate. Anim. Sci. 77:429–437.
Sinclair, A. G., J. M. Shaw, S. A. Edwards, S. Hoste, and A. McCartney. 1999. The effect of dietary protein level on milk yield and composition and piglet growth and composition of Meishan synthetic and European White breeds of sow. Anim. Sci. 68:701–708.
Woods, S. C., R. J. Seeley, J. Porte, and M. S. Schwartz. 1998. Signals that regulate food intake and energy homeostasis. Science 280:1378–1383.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
J. L. Gourdine, J. P. Bidanel, J. Noblet, and D. Renaudeau
Effects of breed and season on performance of lactating sows in a tropical humid climate
J Anim Sci,
February 1, 2006;
84(2):
360 - 369.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
INTRODUCTION
= Creole sow) on the hourly feed intake in lactation during the warm season; each point is the least squares mean of 12 Large White sows and 21 Creole sows; dashed line = daily ambient temperature; * = times when hourly feed consumption was affected (P < 0.05) by breed.