The performance and health of weaned piglets hinges on their ability to consume feed quickly after weaning (McCracken et al., 1995; Pluske et al., 1996), yet most piglets still consume an insufficient amount of feed and experience BW loss at this time (Berkeveld et al., 2009). To reduce these weaning-related problems, highly digestible creep diets are often provided before weaning. However, the usefulness of this practice appears equivocal because most piglets do not eat significant quantities of feed until after 19 d of age (Pajor et al., 1991; Fraser et al., 1994). Interestingly, piglets given a liquid or gruel diet vs. a dry diet at weaning have greater DMI and BW gain (Partridge et al., 1992; Kim et al., 2001). These diets allow piglets to meet their water and nutrient requirements simultaneously, and they also require very little chewing activity.
Recently, we have shown that the majority of deciduous premolars erupt between the first and fifth week of the life of a piglet (Tucker and Widowski, 2009). Studies on miniature breeds have described in detail the changes in muscle, teeth, and bone that are required for suckling piglets to adapt to feeding and drinking (Herring, 1977, 1985; Herring and Wineski, 1986). Throughout the weaning process, the mode of intake changes and piglets must shift from using the jaw-opening muscles to generate suction power (i.e., to obtain milk) to using the jaw-closing muscles for the mechanical breakdown of solid food (Herring, 1985; Langenbach and van Eijden, 2001). Both motor learning and the eruption of teeth are important developmental precursors to the natural weaning process (Herring, 1985; Herring and Wineski, 1986), which begs the question: How does dental development relate to the onset of feeding in commercial piglets?
Therefore, the aims of this study were to determine 1) the effects of dentition on feed-oriented behavior and feed consumption before weaning, and 2) whether premolar eruption or occlusion at the time of weaning influences growth or behavior after weaning at 28 d. The preweaning predictions were that piglets would spend more time at the creep feeder and consume more feed if their premolars were erupted and occluded. Similarly, those piglets with a more advanced dentition at weaning would experience better postweaning growth and would perform more feeding behavior and less belly nosing.
MATERIALS AND METHODS
The University of Guelph Animal Care Committee approved all procedures for this study in accordance with the Canadian Council on Animal Care guidelines (Canadian Council on Animal Care, 1993).
A total of 233 Yorkshire piglets (3 trials of 8 litters each) were obtained from the University of Guelph Arkell Swine Research Station. Piglets were processed (tail-docked, injected with iron, ear-notched, boars castrated) before 3 d of age and were individually marked with nonammonia hair dye for identification on camera. To ensure natural dental development, teeth were not clipped. Litters remained with their sows in standard-sized farrowing crates (creep area: 152 × 208 cm) until weaning at 28 d.
Housing and Feeding.
The farrowing room had natural lighting but was also illuminated with fluorescent light from 0530 to 1730 h to ensure clarity of video recordings. Room temperature was maintained between 23 and 25°C with floor heating being provided in the creep area. On d 5, a commercial starter creep feed (Starter Advance Crumble, Floradale Feed Mill Limited, Floradale, Ontario, Canada; Table 1) containing 1.0% chromic oxide (Sigma-Aldrich Canada Ltd., Oakville, Ontario, Canada) was added to a single 4-hole corner creep feeder (41 × 29 × 29 cm) that was placed at the anterior of each farrowing crate. Water was provided ad libitum to piglets via nipple drinkers. All feed added to creep feeders was weighed and replaced every other day to ensure freshness. Once feed was removed, it was dried in a Fisher Isotemp Drying Oven (model 255g, 200 Series, Fisher, Pittsburgh, PA) at 100°C for a minimum of 2 d and weighed back to calculate the amount eaten per pen. Feed spillage was collected and recorded for the first 10 d of each trial but was found to be negligible so was not accounted for.
All animals were weighed on d 2, 6, 13, 20, 27, 31, and 35 (Pennsylvania M6400 Bench Platform Scale, Pennsylvania Scale Company, Lancaster, PA, with a Cardinal 738 Digital Indicator, Cardinal Scale Manufacturing Company, Webb City, MO). From BW data, preweaning ADG and ADG over the first 3 and 7 d of weaning were calculated.
On d 6, 9, 13, 16, 20, 23, and 27, a fecal sample was obtained from each piglet via a fecal swab (cotton tip applicator). The color of each sample was immediately visually assessed for the presence of chromic oxide (i.e., green color). When detected, the sample was assigned a positive score, indicating prior ingestion of creep feed (Barnett et al., 1989; Kuller et al., 2007). Occasionally, because of diarrhea or the inability to retrieve fecal material, a sample could not be assessed and was removed from data analysis (depending on the day, this ranged from 0 to 26 samples from a total 233 samples).
Behavioral data were collected continuously via a digital recording system (Kodicom i31808WM, i3DVR International Inc., Scarborough, Ontario, Canada). Cameras (Panasonic WV-CP240, Panasonic Canada Inc., Mississauga, Ontario, Canada) were mounted from the ceiling directly over creep feeders, and piglets were considered to be at the feeder when their head and ears were directly over the feed trough. Video recordings were analyzed by 3 trained individuals, and continuous sampling of behavior occurred during two 3-h time periods, for a total of 6 h/d (0700 to 1000 h, 1300 to 1600 h) on d 7, 10, 14, 17, 21, and 24. These time periods were selected based on preliminary observations over a 24-h recording period, during which piglets were most active and observed most frequently at the creep feeder.
The total number of visits and total duration of time spent at the creep feeder was determined for each piglet. If a piglet was observed with its head over the feeder trough and then withdrawing its head, it was considered to have left the feeder (i.e., 1 visit had occurred). Interobserver reliability was determined using one 3-h time period, which was watched by all observers (kappa = 0.91; Lehner, 1996).
Procedures for dental exams are described by Tucker and Widowski (2009). All deciduous incisors, canines, and premolars are referred to by a lowercase i, c, and p, with a superscript (or subscript) number indicating its position within the maxilla (or mandible). For example, p3 is the third maxillary premolar, whereas i2 is the second mandibular incisor (Figure 1).
Full oral examinations were performed on all piglets on d 2, 6, 9, 13, 16, 20, 23, and 27, with every deciduous tooth within the oral cavity being recorded as erupted or not. Eruption was considered to have occurred when any portion of the tooth crown had penetrated the gingiva (Smith et al., 1994). Dental measures included both the presence (vs. absence) of individual premolars (p3, p3, p4, p4) and the occlusion status of premolars within the entire dental arch. Occlusion (i.e., the contact between maxillary and mandibular premolars) was present at 2 stages within the population. Initial occlusion occurred when some (but not all) of the cusps touched between p4 and p3, and full occlusion occurred when all cusps on p4 and p3 made contact in addition to at least 2 of the cusps of p4 (Figure 2).
Housing and Feeding.
After weaning, 7 piglets from each litter (approximately balancing for birth weight and sex) were weaned as litter groups to an on-site nursery. Animals were housed in 1.2 × 2.4 m raised-deck pens (n = 24), each containing one 4-hole stainless steel feeder and standard nipple drinker. Both feed and water were provided ad libitum for the duration of the study. During the first 7 d, the same preweaning creep diet was provided but was then replaced gradually with a pelleted starter weaning diet (University of Guelph, Arkell Feed Mill, Guelph, Ontario, Canada; Table 1). Temperature was maintained at 26°C, and fluorescent lighting was provided from 0530 to 1830 h.
To examine how preweaning feed intake affected behavior and growth after weaning, piglets were further classified as creep feed eaters, noneaters, and moderate eaters. Those individuals demonstrating prior ingestion of creep feed on d 23 and 27 were considered eaters, whereas those demonstrating ingestion on neither day were noneaters. Piglets having 1 positive and 1 negative sample on either day were moderate eaters.
The same digital recording equipment was used as in the preweaning period, with cameras being moved from the farrowing room to the nursery. Behavior was recorded continuously, and data were collected by a single individual using scan sampling every 5 min for three 2-h time periods (0600 to 0800 h, 1100 to 1300 h, and 1600 to 1800 h) on d 2, 4, 6, 8, 10, and 12 after weaning. At each scan, piglets were recorded as performing 1 of 8 mutually exclusive behavior patterns (Table 2) and daily time budgets were estimated on a per-pig basis by calculating the number of scans devoted to each behavior pattern as a proportion of the total scans for that animal on that day.
All data were analyzed using SAS software (SAS Inst. Inc., Cary, NC), with the individual piglet as the experimental unit. All data were formally examined using the UNIVARIATE procedure, with comprehensive residual analyses being conducted to assess the ANOVA assumptions and determine the most appropriate method for data transformation.
For duration of time spent at the feeder, a repeated measures ANOVA was performed using the MIXED procedure. Duration data were transformed by taking the square root after the addition of a bias correction factor of 0.25. For analyses of creep feeder visits as well as fecal score, the GLIMMIX procedure was used. The Poisson and binary distributions were most suitable for number of feeder visits and fecal score, respectively, with neither variable being transformed before analysis. Because the range of eruption times differed for each premolar (for example, p3 was erupted in all piglets by d 16, whereas p4 did not begin to erupt within the population until d 16), separate analyses were carried out for each premolar (and extent of occlusion).
Duration of time spent at the feeder, number of visits to the feeder, and fecal score were evaluated using the following model:where Yijkl is the response variable (duration, number of visits, fecal score, and l is the error term that represents the litter error), μ is the overall mean, Gi is the fixed effect of sex, Aj is the fixed effect of piglet age (in days), Dj is the fixed effect of dental condition (the presence or absence of a particular premolar, or the occlusion of premolars), BWj is BW, Bj is birth weight (on d 2), Dj × Aj is the interaction between dental condition and piglet age, Iij is other relevant interactions, Tk is the random effect of trial, Ll is the random effect of litter, and eijkl is the random error term. Pair-wise differences between treatment means were assessed using t-tests. To account statistically for the passage rate of feed through the gastrointestinal system, a piglet was considered to have consumed feed on the examination day before when a positive fecal score was actually observed.
Postweaning behavioral data were transformed by taking the arcsine of the square root after the addition of a bias correction factor of 0.15. To examine how dentition and preweaning creep feed ingestion affected piglet behavior (ingestive behavior: head in feeder, at drinker; oral-nasal behavior: belly nosing, pen-mate nosing, pen rooting), repeated measures analyses of variance were used with the MIXED procedure, with the following model:where Yijkl is the time spent performing a specific behavior, μ is the overall mean, Gi is the fixed effect of sex, Dj is the fixed effect of dental condition (the presence or absence of premolars, the occlusion of premolars), Ej is the fixed effect of preweaning creep “eater” classification, Aj is the fixed effect of day (after weaning), BWj is BW, Bj is birth weight (on d 2), ADGj is the ADG, Iij is the relevant 2-way interactions, Tk is the random effect of trial, Ll is the random effect of litter, and eijkl is the random error term. Differences in behavior means across days were analyzed by linear contrasts (Kuehl, 1994).
To examine how dentition and preweaning creep feed ingestion influenced growth in the first 3 and 7 d after weaning, ANOVA were used in the MIXED procedure with the following model:where Yijkl is the growth performance trait, μ is the overall mean, Gi is the fixed effect of sex, Dj is the fixed effect of dental condition (the presence or absence of premolars, the occlusion of premolars), Ej is the fixed effect of preweaning creep eater classification, BWj is BW at weaning, Bj is birth weight (on d 2), ADGj is the ADG before weaning, Iij are the relevant 2-way interactions, Tk is the random effect of trial, Ll is the random effect of litter, and eijkl is the random error term. Pair-wise differences between means were assessed using a t-test.
BW Gain and Feed Intake.
Mean BW, and weekly ADG and ADFI are listed in Table 3. Feed data could not be collected on d 21 to 26 for 1 trial, so total feed intake and ADFI for wk 3 are from 2 trials of 16 litters. Total creep feed intake over the 22-d period ranged from 1,026 to 3,040 g/litter, which gave an average total intake of 216 g/piglet (litter range: 131 to 334 g; SEM = 15.8 g). The majority of creep feed was consumed in the week before weaning (d 21 to 28), when intake averaged 122 g/piglet (litter range: 60 to 220 g; SEM = 11.8).
Behavioral data could be collected only from 2 trials. The duration of time spent at the creep feeder and the number of visits to the feeder over the course of the experiment are presented in Table 4. Duration of time spent at the feeder on d 7, 10, and 14 was variable but did not differ until d 17 (P > 0.05), when it decreased (P < 0.05). Increases were then observed on each successive observation day (P < 0.0001). The frequency of visits to the creep feeder decreased from d 7 to 10 (P < 0.05) and also from d 14 to 17 (P = 0.0002), and then increased successively thereafter (P < 0.0001).
Compared with d 10, piglets visited the feeder more frequently on d 7 (P < 0.05), although the duration of time did not differ (P > 0.05). As described by previous authors, this may reflect greater investigatory behavior toward a novel structure (or its edible contents, or both) that was introduced to the farrowing crates on d 5 (de Passillé et al., 1989; Delumeau and Meunier-Salaün, 1995).
Dentition in Relation to Feeding Behavior.
The percentage of piglets with various premolars erupted and occluded over the course of the experiment is shown in Figure 3. Eruption times of all teeth, as well as factors that influence that timing, have been reported (Tucker and Widowski, 2009). Sex was not found to be a factor for either feeding behavior or feed ingestion before weaning (P > 0.05).
The first premolar to erupt in the population was the maxillary p3. This tooth has 3 triangular elements (i.e., is trigonal), 2 major distal cusps (1 each on the buccal and lingual side), and 1 major mesial cusp on the buccal side (Figure 1). One piglet had this premolar erupted at 2 d of age, and all individuals had at least partial eruption by 16 d of age. From Tables 5 and 6, it is clear that on d 7, piglets with p3 erupted spent less time (P = 0.005) at the creep feeder and visited it less frequently (P < 0.0001 for number of visits on d 7 and P = 0.001 for number of visits on d 10) than those piglets without it.
The second premolar to erupt was the mandibular p4. This is the only tricuspid tooth in the deciduous dentition, having 3 cusps on both the buccal and lingual sides that are fused into transverse ridges (Figure 1; Herring, 1976). Again, 1 piglet (the same individual as mentioned above) had this tooth erupted by d 2, and all individuals had it erupted by d 20 of age. As with the p3, those piglets having p4 erupted made fewer visits to (P < 0.0001), and spent less time at (P < 0.0001 for duration on d 7 and P = 0.003 for duration on d 10), the creep feeder on d 7 and d 10 compared with those piglets with premolars at a lesser developmental stage. However, by age 17 d, piglets having this premolar erupted began to visit the feeder more frequently (P = 0.024).
The third premolar to erupt was the mandibular p3. Unlike the others seen to erupt in the time course of this experiment, this premolar is not molariform in shape but sectorial, with an elongated shearing cusp (Figure 1). No individual had this tooth present until 13 d of age, and even by 27 d, it was still not erupted in 6 animals. On d 14, piglets with this tooth erupted visited the feeder less (P = 0.037) than piglets without it, but by d 21 and 24, they visited the feeder more (P = 0.0002 on d 21 and P < 0.0001 on d 24). Likewise, piglets having this tooth by d 24 spent more time at the feeder (P = 0.025).
The last premolar to erupt during the course of this experiment was the maxillary p4. This tooth has 4 main bunodont cusps and hypoconulid lobes, with subsidiary cusps and crenulations surrounding them (Figure 1; Herring and Scapino, 1973). Within the current population, this tooth erupted in piglets as young as 16 d, although by 27 d, 53 individuals (27.5% of the population) still did not have it erupted. Piglets that had their p4 at 17 d of age spent less (P = 0.012) time at the feeder and visited it less frequently (P = 0.024), but at age 24 d, they spent more time (P = 0.0005) at the feeder and visited it more frequently (P = 0.0002).
The percentage of piglets displaying initial occlusion between p3 and p4 is shown in Figure 3. The same individual having both p3 and p4 erupted at 2 d of age also had initial occlusion between these teeth. Only 1 individual had not yet obtained initial occlusion by d 27. On d 7 and d 10, negative associations were found between measures of feed-oriented behavior (duration, P < 0.0001 for d 7 and P = 0.0004 for d 10; feeder visits, P < 0.0001 for d 7 and P = 0.003 for d 10) and premolar occlusion. By d 17, piglets with occlusion visited the feeder more (P = 0.010), and on d 21 and 24, both the number of visits (P < 0.0001 for d 21 and 24) and the duration of time (P = 0.001 for d 21 and P = 0.0001 for d 24) were positively associated with occlusion.
Dentition in Relation to Feed Consumption.
The percentage of piglets having positive fecal scores over the course of the study is shown in Table 4. No piglet was seen to have consumed feed by 6 d of age, and only 44.9% of piglets demonstrated (by fecal analysis) consumption of feed at 24 d of age, and only 60.6% demonstrated feed consumption by 27 d of age. Linear increases in consumption were seen for every day that fecal scores were examined (P < 0.0001); however, no association was found between creep feed consumption and the eruption or occlusion of premolars (P > 0.05).
Postweaning Relationships Between Dentition, Growth, and Behavior
Because of equipment failure, postweaning data were collected from only 2 trials of 16 litters (112 piglets in total). Over the first week, gilts tended to gain more BW than barrows (P = 0.065). The percentages of noneaters, moderate eaters, and eaters during the preweaning period were 57.1, 30.4, and 12.5%, respectively. All 112 piglets had their p3 and p4 erupted and had attained initial occlusion before weaning. Full occlusion was achieved by 48.2% of the population by d 27, whereas eruption of both p3 and p4 had occurred in 96.4 and 81.2% of the population, respectively. None of the dental measures influenced postweaning growth rates (P > 0.05) or the performance of any behavior in the postweaning period (P > 0.05). Similarly, classifying piglets as noneaters, moderate eaters, or eaters before weaning did not influence either growth or the time spent performing any behavior after weaning (P > 0.05).
We predicted that piglets with erupted and occluded premolars would spend more time at the creep feeder, visit the feeder more frequently, and show earlier consumption of solid feed. Our hypotheses were correct regarding feed-oriented behavior; however, the effects were age dependent. Two general trends were seen: first, animals 17 d of age and younger spent less time at the feeder and visited it less often if they had premolars erupted and occluded, and second, piglets 21 d of age and older spent more time at the feeder and visited it more frequently if they had premolars erupted and occluded.
These findings indicate that younger piglets may be inhibited from feeding as their first premolars begin erupting and occluding. In the examination of the gums during oral exams, it was often noted that bleeding and inflammation were present around the erupting premolars, especially at the early stages. On a cellular level, eruption is a localized and programmed event that is a direct result of alveolar bone resorption. In addition to the influx of bone-mediating cells to the gingiva, there is a host of signaling proteins that accompany them (i.e., eicosanoids, cytokines, growth factors; Sandy, 1992). These compounds are common mediators of inflammation, which, after induction, are known to cause localized discomfort, pain, and protective behaviors (Funk, 2001; Shapira et al., 2003; Sommer and Kress, 2004).
In cattle, and to a lesser extent in sheep and goats, this localized reaction is termed an “eruptive collar” (Andrews, 1985; Cocquyt et al., 2005) and is thought to cause a loss of appetite and reduced body condition in what would otherwise be healthy individuals (Andrews, 1985). The symptoms attributable to the initial cutting or eruption of teeth (i.e., teething) have been best documented in human infants and include increased biting, drooling, gum and ear rubbing, sucking, irritability, wakefulness, decreased appetite for solid foods, and mildly raised body temperature (Macknin et al., 2000; Wake et al., 2000). Interestingly, many of these behavior patterns are common in the young pig and might therefore be related to dental discomfort. Although many animal practitioners assume that teething afflicts other mammals (Hale, 2000), whether a true loss of appetite is occurring or animals choose to avoid solid food remains to be determined.
The reversal of the trend between premolar eruption and feed aversion that was observed when piglets reached 21 d of age and older may indicate that a point of desensitization had been reached or a method of coping had developed with regard to oral discomfort. For example, if piglets became accustomed to their painful gums or found that chewing on items (such as creep feed) offered relief, they may have increased feed-oriented behaviors as further premolars continued to erupt. Alternatively, the discomfort of eruption may have diminished or ceased altogether by this age. However, because piglets with erupted premolars spent more than double the amount of time at the feeder compared with piglets without premolars, it seems probable that feeding motivation became significantly enhanced as this part of their oral physiology developed.
Nevertheless, we did not find any associations between the eruption or occlusion of premolars and feed ingestion during the preweaning period. Often, it was noted that piglets appeared clumsy in their feed-handling abilities and dropped much of it from their mouths as they attempted to consume it. This has been reported by researchers studying feeding behavior in miniature breeds of swine (Herring and Wineski, 1986) and may explain why, during the first few days after weaning, commercial piglets appear to be eating (i.e., they have their heads in the feeder trough), yet are not actually consuming appreciable quantities of feed (Gardner et al., 2001; McGlone and Anderson, 2002; Torrey and Widowski, 2004).
These observations indicate that motor learning is an important aspect in the development of feeding behavior in young pigs. For example, the periodontal mechanoreceptors within the ligaments attaching tooth roots to alveolar bone are the main tactile sensors providing tooth load information to the brain. As opposing teeth erupt into initial contact, these receptors will undoubtedly require a period of readjustment. Likewise, handling and chewing of food items requires significant coordination of the tongue muscles in a fashion distinct from that needed for suckling (Gordon and Herring, 1987), and also varies significantly based on the properties of food (Kakizaki et al., 2002). Within miniature breeds, the masticatory muscles themselves have been well documented as requiring periods of motor learning in the development of effective and coordinated chewing behavior (Herring and Scapino, 1973; Herring and Wineski, 1986; Herring et al., 1991). Huang et al. (1994) went further to classify 3 separate stages of premolar occlusion and then related them, using electromyography, to masticatory ability and efficiency. Miniature piglets with 1 cheek tooth per quadrant in occlusion (first stage) chewed more slowly, required more chewing cycles to process a food bolus, and had less regular alternation in the chewing side compared with piglet in further stages. To our knowledge, no studies have yet examined this period of adjustment in commercial piglets.
The lack of association between feed intake and premolar eruption may also be due to the method by which we accounted for passage rate of feed through the gastrointestinal system. Piglets were considered to have consumed feed on the examination day before when a positive fecal score was actually observed, assuming a passage rate of between 72 and 96 h. Kuller et al. (2007), who orally administered chromic oxide-marked creep feed to nursing piglets, reported occasional intermittent excretion of marked feed within the same day and concluded that the best time to sample feces was 48 to 58 h after ingestion. However, they also noted that when piglets are allowed continuous access to marked feed, chances of accurate detection for creep-eating individuals will increase because those who ingest feed tend to continue to do so (Appleby et al., 1991; Pajor et al., 1991; Kuller et al., 2004). Because fecal samples were collected only once every few days in our study, it is possible that the optimal detection period for marked feces was missed. Furthermore, ingestion of small quantities of marked feed may be masked by large intakes of milk; hence, the absence of color may not reliably indicate that no feed was eaten (Barnett et al., 1989).
It should also be noted that only one 4-hole creep feeder was placed in each farrowing crate so that simultaneous feeding of the whole litter was not possible. Because litter size ranged from 7 to 13 piglets, competition may have varied between litters. Although not recorded, displacement was frequently seen whereby 1 piglet would push another away from the creep feeder; however, the dentition of these piglets was not determined.
Regarding the dentition itself, it should be noted that although occlusion could be determined in this study from visual assessment of contact between premolars, the quality of occlusion could not be assessed. The animals were not sedated, and even the slightest resistance or movement in jaw muscles would give a misrepresentation of how well premolars aligned between the maxilla and mandible; therefore, the quality of occlusion was undetermined. In addition, piglets in this study were categorized simply as having their premolars either erupted or not erupted. This means that piglets with only small portions of their premolar cusps erupted through the gingiva would be statistically grouped with those having larger proportions of their premolar cusps erupted.
The postweaning phase of this study was designed to investigate whether dentition influenced postweaning growth or behavior. However, no evidence was found for an influence of dentition on the transition at weaning when piglets were weaned at 28 d of age. One possible explanation for the lack of relationship between dental development and feeding behavior or growth after weaning was the small variation in many of these dental measures across the population by 27 d of age. Except for full occlusion, which was present in only 48.2% of the population at weaning, all other measures were present in at least 80% of piglets. Additionally, the variation in the number of teeth and premolars was skewed toward the maximum possible number by this age.
It has been suggested that the maturity of the piglet is critically important for their ability to adapt at weaning (de Passillé et al., 1989; Pajor et al., 1991). Piglets generally have smaller distress responses (Weary et al., 2008) and better performance when weaned at older ages (Leibbrandt et al., 1975; Armstrong and Clawson, 1980; Main et al., 2004). In addition, the period of BW loss after weaning has been shown to be age dependent, with 21-d-weaned piglets taking twice as long to recover compared with piglets weaned at 28 d (Colson et al., 2006). Studies also indicate more feeding behavior among piglets weaned at 28 vs. 14 d of age (Metz and Gonyou, 1990) and at 28 vs. 21 d of age (Worobec, et al. 1999). Therefore, any effect of dentition on feeding and growth at weaning may be more likely when piglets are weaned at younger ages (e.g., between 17 and 21 d).
Any effect of dentition on ingestive behavior patterns in the postweaning period may have been masked by the husbandry conditions imposed. Piglets were given access to solid feed for an unusually long period (from the age of 5 d), and 54% of piglets were known to be ingesting feed on the day before weaning. Housing piglets in litter groups may also have allowed for increased learning opportunities for both feeding and drinking behaviors because social stresses would be reduced under these conditions (Olesen et al., 1996; Hessel et al., 2006). Morgan et al. (2001) found that food-naïve piglets housed with established eaters were more likely to discover and consume feed than when only food-naïve piglets were housed together. Retrospective analysis in the current study showed that at least 1 piglet in every weaning pen had demonstrated previous feed ingestion.
Interestingly, classifying piglets as creep eaters based on their preweaning feed ingestion did not affect either their time at the feeder or their postweaning growth. This is contrary to the results of Bruininx et al. (2002), who reported that creep eaters ingested more feed on their visits to the feeder and also had greater ADG compared with noneaters. Although we could not determine whether piglets were actually consuming feed while visiting the feeder, the fact that we found no difference in BW gain suggests preweaning feed consumption had little effect on postweaning feeding behavior. However, it should be noted that Bruininx et al. (2002) classified piglets as creep eaters by examining fecal scores on d 18, 22, and 27, and found that 17.4% of piglets were positive on all 3 d. If using the same methodology, only 3 individuals (2.7%) would have been considered creep eaters in the current study.
With regard to our predictions on the performance of belly nosing after weaning, it was thought that because this abnormal behavior is considered to be a juvenile form of ingestive behavior (i.e., redirected suckling; Torrey and Widowski, 2006), piglets developing it may also possess a less developed dentition. However, no association was revealed in this study to support this hypothesis. In fact, the occurrence of belly nosing was minimal throughout the duration of this study, with overall means remaining below 1% of the scans. This is comparable with values reported by Metz and Gonyou (1990) and Worobec et al. (1999), who also weaned piglets at 28 d of age. Although belly-nosing events were minimal, there was a characteristic increase during the second week after weaning (Metz and Gonyou, 1990; Gonyou et al., 1998; Worobec et al., 1999; Bruni et al., 2008). Unfortunately, no observations were made on the condition of the gingiva or the dentition during that period, when belly nosing was seen to spike within the population.
The results of this study showed that although the majority of piglets had many of their deciduous premolars erupted by 27 d of age, and almost all piglets had initial occlusion between their p4 and p3 by 23 d of age, only 45 and 61% of piglets were consuming creep feed by d 24 and 27, respectively. No associations were found between the eruption or occlusion of premolars and feed consumption; however, a more precise methodology is suggested for examining this relationship in the future. Feeding behavior was associated with dental eruption, with age-dependent responses being revealed. Younger piglets were inhibited from feeding when their premolars began erupting, possibly because it was their first experience with oral sensitivity. However, by 21 d of age, piglets with erupted premolars were more attracted to feed, indicating enhanced motivation for independent ingestion after the eruption of their teeth.
At the time of weaning at 28 d, neither the eruption nor the occlusion of premolars was found to influence postweaning growth or behavior. Very little variation in dental measures existed within this population by this age. Further studies using more commercially relevant conditions such as younger weaning ages, mixing of pen mates, varying genetics, and limited preweaning feed exposure would be beneficial in elucidating the effect of dentition on weaning transition.