A total of 32 BEC pigs (16 barrows and 16 females, representing the offspring of 8 boars) were used for the trial. Animals were purchased in twelve farms located in the Ecuadorian coast and transferred to the swine farm “El Paraiso” (Los Rios province, coast of Ecuador). The area has a tropical climate with an average temperature of 25° C, an annual rainfall of 2400 mm and relative humidity of 82%. On arrival, all the pigs were identified, weighed, dewormed with Ivomec-Merial (0.08% ivermectin), housed in groups and offered a conventional diet for 15 days (period of adaptation). Males were castrated (under surgical anaesthesia with standard post-surgery treatment performed by a veterinarian). Then, 8 barrows and 8 females were randomly allocated to one of two treatments: traditional farming system and conventional indoor rearing system. The distribution of piglets in both rearing systems occurred at a weight of 18-22 kg and at an age of 165 days, approximately. All animals were manually fed twice a day (morning and afternoon) and water was provided ad libitum, and they received the same commercial diet, consisting of products of the area (16% crude protein, 6.2% ether extract, 8% crude fiber, 2.3% ash, 3200 kcal digestible energy/kg) ad libitum and 80% for C and SC groups, respectively. In C system, pigs were housed in individual pens (1 × 1.5 m) with feed and water bowls. In the SC system, pigs were also housed in individual pens of 1 × 1.5 m over night with concrete floor, feed and water bowls, and they were kept individually in a paddock of about 625 m²/animal during the day where they consumed banana (in an amount of about 275 g/day/pig, increasing 40 g every 5 days), roots, leaves and tubers of papaya, cassava, oranges, and insects, worms, etc. Individual body weight and feed consumption were recorded at 0, 15, 30, 45, 60, 75 and 90 days of experiment and these data were used to calculate average daily gain and feed conversion ratio. The study was carried out in accordance with Ecuadorian national recommendations for animal handling including animal welfare.

At the end of the trial (90 days), pigs were weighed in the experimental farm and transported to a commercial slaughterhouse. At the abattoir, animals were resting for 12 h with full access to water but not to feed. Then, pigs were electrically stunned, exsanguinated, scalded, skinned and eviscerated according to standard commercial procedures. The slaughter was performed, in compliance with ethical guidelines for animal care, in two different days in which both sexes and rearing systems were adequately represented. The slaughter weight, empty body weight (was calculated by deducting the weight of digesta) and hot carcass weight was recorded. In addition, the pH was measured at 45 min (pH₄₅) and at 24 h (pH₂₄) postmortem, in the left Longissimus thoracis muscle (LT) at the caudal end over 12^th rib, at right angles to the sagittal plane surface, using pH-meter with temperature compensation (Lutron mod. 212; pH Electronica, Buenos Aires, Argentina) equipped with a glass electrode (previously calibrated with pH 4.01 and 7.00 standard buffers equilibrated at 25ºC) enclosed with an unbreakable stainless steel knife, to facilitate the measurements in a muscle, and thermometer.

Carcasses were chilled at 4°C for 24 h, and cold carcass weight was recorded. The hot and cold carcass weights were used to calculate carcass yields. Then, the head was removed and the carcasses were split along midline. Several morphometric variables were measured using a vernier caliper or a flexible tape on the hanging right half of the carcass at 24 h postmortem following the method described by Peinado et al. (2004) and Fisher et al. (2003). These measurements were: backfat thickness (at the level of the 12^th rib, taken 1/4, 2/4 and 3/4 the length ventrally over the LT and perpendicular to the skin), carcass length (from the middle of the cranial edge of the first rib to the ischiopubic symphysis), shoulder length (from the middle of the dorsal edge of the scapula to the internal halfway of the facies articularis carpea), hand length (from the end of the olecranon to the distal point of the trotters), ham length (from the isquiopubic symphysis to the internal halfway point of the calcaneus), hind foot length (from the middle of the dorsal edge of the tarsus to the distal extreme of the trotters), maximum perimeter of the shoulder (at its widest, in the area of maximum amplitude), maximum perimeter of the ham (in the area of maximum amplitude, near the base of the tail) and wrist perimeters (at the narrowest point, in the middle of the metacarpus and metatarsus regions).

Afterwards, the left sides of the carcasses were quartered, according to Ecuadorian methods (Santos Ricalde et al., 2011), into five joints: fore limb (shoulder with front foot: means that portion of the fore limb which is separated by a straight cut at the wrist (carpal) joint), top loins (1^th-6^th thoracic vertebrae), loins (7^th-11^th thoracic vertebrae), tenderloins (12^th- thoracic vertebrae to last lumbar vertebra) and legs (hams with hind foot: portion of the leg which is separated by a straight cut passing through the hock (tarsal) joint). The different joints were weighed using calibrated scales sensitive to 1 g (GRAM DSN-30; BD COM, Zafra, Badajoz, Spain). Proportions of these joints based on the cold carcass were also determined. All the joints were dissected into skin, fat, muscle and bone. The cross sectional area of the LT on the 7^th rib was measured with a transparent cellulose acetate grid divided into l-cm squares. Then, a portion of the LT including the 7^th to the 11^th ribs was taken for later determinations.

The analysis of mineral composition was evaluated according to Galián et al. (2007). Briefly, approximately 1 g of meat (LT) was used for the process of obtaining ashes in a furnace oven (Model HK-11; Herotec, Ahlen, Germany). The ashes were dissolved with 2 mL of 65% nitric acid and 5 mL of 37% hydrochloric acid fuming on a hot plate and the volume was made up to 50 mL with distilled deionized water (obtained from the device Milli-Q Gradient A10, Molsheim, France). After that, the samples were analyzed for the determination of minerals in a plasma absorption spectrometer (Model Optima 200DV; Perkin-Elmer, Waltham, MA, USA). The minerals analyzed were: calcium, magnesium, iron, copper, zinc, phosphorus, manganese, potassium and sodium. The results obtained for each mineral were expressed in mg/100 g of raw meat, after application of the corresponding dilution factors. In this study, two replicates for each meat sample were performed.

Water holding capacity was determined only 1 day postmortem. Drip loss was measured following the method of Honikel (1998) on cubic pieces of 1.5 cm thick, free of external fat and connective tissue. The samples were weighed, hung by a nylon cord and put on top, inside a container which was closed after filling to avoid evaporation, ensuring the meat had no contact with the container walls. This container was placed in a chamber at 4 °C during 24 h and then reweighed. Drip loss was calculated as the difference in weight of the meat sample (after superficially wiping dry), before and after the hanging × 100, divided by initial sample weight yields. In order to assess water-holding capacity from cooked meat and to carry out the sensory analysis, a portion of the LT of the left half of each carcass was extracted and vacuum packed. The pieces were stored at 4ºC during 24 h. Then, two 1.5-cm thick steaks were obtained from each portion. Cooking loss was expressed as percentage weight loss of the initial weight. The cooking procedure was as follow: all the vacuum packed steaks were thawed previously to each session. They were reserved in a cooler at 4ºC during 24 h and, 2 h before the analysis the steaks were taken out the cooler until they achieve an internal temperature of 15-17ºC. Next, the meat samples, packed in nylon/polyethylene vacuum bags, were heated by immersion at 80ºC until they reached an internal temperature of 72ºC in a water bath (JP Selecta, Barcelona, Spain) with controlled temperature, followed by cooling in cold tap water for 40 min. Internal temperature of the cooked muscle was monitored by an iron/constantan thermocouple wire connected to a thermometer (Hanna Instruments, USA) and inserted into the geometric center of the sample. Before slice they were weighted to determine cooking losses. Subsequently, the pork samples were sliced (cubes of approximately 1.3 × 1.3 × 1.3 cm) taking care to avoid an excess content of fat or connective tissue. The cubes were placed in covered glass containers in a circulating water bath (approximately 60ºC) to equilibrate their temperature and immediately after were served on glass plates at room temperature (20–23ºC).

In order to evaluate the influence of sex and rearing system on the sensory characteristics, cooked slices of LT from 32 BEC pigs were assessed by a trained panel, using a quantitative–descriptive analysis method for 11 different attributes. Sensory quality of cooked meat samples was evaluated by 5 panelists (females, student and staff members from Quevedo University, Ecuador) previously trained for meat sensory evaluation. Training of panelists (Pillsbury, 1992) was performed with pork samples for 2 weeks. All sessions were done in a five booth sensory panel room at 22ºC equipped with white fluorescent lighting (220–230 V, 35 W). Rearing systems and sex were represented in all sessions in a balanced design with a randomized serving order within session. Panelists were asked to evaluate tenderness, color, brightness, lard flavor, pork odor, pork flavor, sweet taste, metallist taste, unctuousness, juiciness and persistence of flavor. The sensorial evaluations were performed according to Meilgaard et al. (2006), and each attribute was quantified using an unstructured 10 cm line, ranging from the lowest intensity of each trait (left side) to the highest (right side).

All pigs started the experiment and were slaughtered at a similar age. The live weight at the start (about 165 days of age) and end of the experimental period (90 days later) averaged 23.7 kg and 47.3 kg, respectively (Fig. 1). The total weight gain (kg) was 28.07, 27.78, 20.08 and 19.75 kg for barrows and females C, and barrows and females SC, respectively (data not shown). As expected, weight at slaughter was higher in C pigs than in SC pigs, whereas differences between barrows and females were not found (Table 1).

Average daily gain increased from 0.19 kg/day at start of trial (0-15 days on fattening) to 0.35 kg/day at end (75-90 days on fattening), and a significant effect (p<0.05) of rearing system and sex on growth was recorded. The C pigs and barrows grew faster than SC pigs and females, respectively. The daily feed intake in the groups SC and C during the study period increased gradually from 0.64 (0.47 and 0.275 kg of commercial diet and banana, respectively) and 0.6 kg/day in the first week to 2.8 (1.8 and 1.0 kg of commercial diet and banana, respectively) and 2.3 kg/day in the last week, respectively. Average feed conversion ratio during the experimental period was 4.99, and rearing system showed a clear influence (p<0.05) on this variable (4.54 vs 3.72 for C and SC, respectively). Sex had no significant effects (p>0.05) on this trait.

The effects of rearing system and sex on slaughter weight, hot and cold carcass weights, dressing-out percentages, backfat thickness, rib eye area and carcass measurements are shown in Table 1. As expected, the C pigs, with higher slaughter weights (p<0.001), produced heavier carcasses (p<0.001), but carcass yields were lower (p<0.001) to those found in SC pigs (73.5% vs 74.9%). No significant (p>0.05) effect of sex was observed for any of the carcass traits studied.

Average backfat thickness and rib eye area in the current work ranged between 1.99 and 2.23 cm, and 7.40 and 8.71 cm², respectively. Both variables were significantly affected by the rearing system (higher in C pigs than in SC pigs, p<0.01) and sex (higher in barrows than in females, p< 0.05).

The mean values of the measurements taken in the carcass ranged between 66.9 cm and 12.1 cm for carcass length and wrist perimeter fore foot, respectively. With regard to morphometric parameters, rearing system influenced (p<0.05) most of the measures of length and perimeter of the carcass. However, sex affected (p<0.05) only the leg length, ham length and maximum perimeter ham.

The carcass cut weights and percentages respect to cold carcass weight, and the effects of rearing system and sex are presented in Table 2. The sirloin was the heaviest carcass cut (3.9 kg), and the top loin (2.0 kg) the lightest. These cuts represented from 22.7% to 11.5%, respectively. The study of the meat cuts reveals differences because of the rearing system and of the sex. The C pigs produced heavier heads (p<0.001), top loin (p<0.05), sirloin (p<0.05) and hams (p<0.01). Head and sirloin percentages were higher in SC pigs, whereas the percentage of the loin was lower. Likewise, the sum of the most valuable primal cuts was similar in both systems. Sex had significant effect (p<0.05) on sirloin percentage, greater in barrows.

The effect of rearing system and sex on tissue composition of commercial cuts are given in Table 3. The average content of skin, fat, lean and bone and was 6.2%, 38.9%, 46.1% and 8.8%, respectively; values that indicate its propensity to fatness, especially considering the low weight at slaughter. The proportion of fat in the carcass was lower (p<0.001), whereas the proportions of skin, lean and bone (p<0.01) and fat/lean and lean/bone ratios (p<0.001) were higher in the SC pigs than in the C pigs (data not shown), in accordance to the higher backfat percentage observed in their carcasses (Pugliese et al., 2003). Also, sex significantly affected all traits studied. Barrows had significantly lower skin and fat proportions (p<0.01), and higher lean (p<0.01) and bone (p<0.001) proportions when compared with females (data not shown). The skin content of cuts ranged from 4.96% (loin) to 8.57% (top loin), the fat content between 29.9% (ham) and 48.0% (loin), the muscle between 38.4% (loin) and 58.2% (ham), and bone between 6.8% (ham) and 14.5% (top loin). The leaner primal cut was the ham and the loin had the highest fat content. The rearing system significantly (p<0.05) affected the tissue composition of most of the pieces, while there was a slight effect of sex, except on loin and sirloin.

Animals from all groups showed similar values (data not shown) of pH₄₅ (6.23 to 6.24), pH_24h (5.34 to 5.45) and for water holding capacity (2.7 to 3.4% and 25.6 to 28.7% for drip and cooking loss, respectively).

Table 4 shows the mineral composition of the LT in BEC pigs. Sex had a significant influence on the ash content of meat. Females of the SC group had an ash content (1.69%) well above that in the other three groups (1.21, 1.30 and 1.04%, respectively). Average values of the mineral composition of the meat were 232.1, 340.6, 7.1, 58.6, 23.8, 0.23, 2.34, 1.58 and 0.033 mg/100 g for P, K, Ca, Na, Mg, Cu, Fe, Zn and Mn, respectively.

The rearing system had a little influence on the meat mineral levels. The sex had an effect on the levels of the minerals P, Na, Cu and Fe (p>0.001) and Mg (p>0.05). The barrows showed higher levels of P and Mg and lower of Na, Cu and Fe than the female pigs. Sex differences in the Fe content, essential ingredient of the myoglobin and hemoglobin, did not translate into differences in meat color (Table 5).

The average ratings for the sensory attributes of each rearing system and for sex are given Table 5. Based on these results, the meat of the animals studied can be defined as moderately tender, with little brightness and color, with high odor and pork flavor, little unctuous on the palate, juicy and moderate to high persistence of flavor. The evaluation by the sensory panel of the pork loins did not show large differences in tenderness, color homogeneity, lard flavor, pork flavor and sweet taste. However, significant differences (p<0.05) were obtained for the remaining attributes studied between rearing systems. Metallic taste, juiciness and persistence of flavor were higher (p<0.05) for meat from C pigs (rating 0.77, 5.6 and 6.5, respectively) than SC pigs (rating 0.5, 5.0 and 6.0, respectively), while meat color (4.1 vs 5.0), brightness (4.4 vs 4.8) and unctuousness (4.0 vs 4.8) were lower. In contrast to rearing system, few differences in sensory attributes were observed between sex: metallic taste (higher in barrows: 0.7 vs 0.5, p<0.01) and brightness (higher in females: 4.4 vs 5.1, p<0.01).

Average daily gain was lower in our study than that recorded by Fortina et al. (2005), Renaudeau & Mourot (2007) and Galián et al. (2009) in others autochthonous pig breeds, so that the Creole BEC, reared indoors or outdoors, needs longer to reach the same weights than these breeds. During the growing period, average daily gain tends to increase although these results were not in agreement with those provided by Latorre et al. (2008). As in the current experiment, a higher growth for barrows compared to females has been well documented in previous studies (Latorre et al., 2008; Serrano et al., 2008). Nevertheless, these results contrast with those of Renaudeau & Mourot (2007) in Creole pigs, who indicated that daily gain was not affected by sex. The C pigs grew significantly more than SC pigs, in accordance with informations of Daza et al. (2006). These differences have been related to the higher energy requirements for physical exercise and thermoregulation in SC pigs (Pugliese et al., 2003), among others.

In our study, feed conversion ratio was slightly higher than that recorded (4.2) by Fortina et al. (2005) in Casertana and Mora Romagnola breeds reared indoors, fed using diets formulated for commercial hybrid pigs and slaughtered at<60 kg of live weight. The differences between rearing systems (4.54 vs 3.72 for C and SC, respectively) could be attributed to differences in digestibility of diets and given that part of the diet of the SC pigs could not be controlled. In agreement with Peinado et al. (2012) in crossbred pigs (Landrace × Large White dams × Duroc sires), sex had no significant influence on feed conversion ratio. However, Renaudeau & Mourot (2007) and Serrano et al. (2008) noted in Creole and Iberian × Duroc pigs, respectively, that barrows had a higher feed conversion ratio than gilts.

This low body weight observed is not unusual or surprising if you consider studying Benitez & Sanchez (2001), where it is mentioned that the average slaughter weight of pigs reared in “traspatio” system in countries such as Ecuador was 45 kg. In common practice barrows and females of BEC pigs are usually slaughtered approximately at the same age, as was done in the present study. The poor dressing-out percentages are indicative of the low degree of muscularity from slightly fatness carcasses. The lowest carcass performance can be attributed to the weight/age at slaughter and the level of nutrition (Tartari et al., 1993) among other factors. The average carcass yields are lower than those recorded in other rustic pig breeds (77.6-83%) (Renaudeau & Mourot, 2007; Santos Ricalde et al., 2011; Franco et al., 2014). These results could be related with differences in slaughter weight since differences in the degree of fat are very light. In agreement with previous studies, in the current experiment the dressing percentage is similar in barrows than in females.

It should be noted that the carcass yields was significantly lower in the C pigs than those found in SC pigs (73.5% vs 74.9%), in agreement with the results obtained by Galián et al. (2009) who found higher carcass yield values in Chato Murciano pigs reared outdoors (74.2 vs 73.1% and 72.7 vs 71.7%, for hot and cold carcass yield values, respectively). However, Daza et al. (2006) in Iberian pigs found that the pigs fed in confinement had a carcass yield significantly higher than those fed extensively with acorns and grass. The higher crude fiber consumption in pigs fed extensively could be the cause of reduced carcass yield in these animals.

The values obtained for backfat thickness showed that the animals under this study were leaner because of their low weight and age at slaughter. However, if the levels of the fat thickness relate to slaughter weight, these values are considered to be high, and caused to a large extent by the rusticity of the breed. In the current experiment, the average values for backfat thickness were lower than those reported by Franci et al. (2005) and Poto et al. (2007) in several breeds (3.6-8.9 cm). These results are in disagreement with previous reports, which demonstrated that autochthonous pigs are characterised by stronger fatty deposition than commercial pigs (Wojtysiak & Poltowicz, 2014). However, these comparisons are hard because of differences in the experimental designs as the dorsal thickness has a strong relation to slaughter weight and diet composition. Thus, Aro & Akinjokun (2012) recorded backfat thickness values between 1.2 and 2.8 in Large White × Duroc pigs fed cassava peel and slaughtered at 68-70 kg liveweight. There were significant differences between the two systems for backfat, which is similar to found by Pugliese et al. (2003) and Daza et al. (2006) in Nero Siciliano and Iberian pigs, respectively. Backfat thickness was higher in females than barrows, in agreement with Renaudeau & Mourot (2007), Latorre et al. (2008) and Aro & Akinjokun (2012). The low weight and age of the animals at slaughter could be the main causes of the lack of differences between sexes. Nevertheless, our results disagree with Renaudeau et al. (2005) in Creole pigs, who indicated that the barrows deposit more fat than females mainly at lumbar level. The rib eye area was lower than that recorded in most of the literature consulted due to lower slaughter weight and even with respect to the results obtained by Wojtysiak & Połtowicz (2014) on Pulawska pigs slaughtered at 30 kg body weight. The absence of an improvement program and the poor quality of feed provided to the animals could be the main causes of poor muscle development, and consequently the low values of the rib eye area.

The conformation of the carcass is an aspect of interest in the meat industry. Such confirmation may be assessed by subjective variables and through objective measurements. The mean values from carcass measurements were lower than those found by De Jesús-López (2008) in Spanish autochthonous breeds (Celta). However, the values were higher than those recorded by Ramos (2008) in Peruvian Creole pigs, which is in consonance on the differences in body weight between the two populations (51.4 kg vs 35.4+5.1 kg). The carcass measurements are related to the genetic biotype and slaughter weight, resulting in a low score of carcass conformation. With regard to morphometric parameters, rearing system effects were significant in most of the lengths and perimeters of the carcass, in agreement with Galián et al. (2009) in Chato Murciano breed. However, Pugliese et al. (2003) found no significant differences for all carcass measurements, except for body length, between indoors and outdoors pigs. These differences were mainly caused by the carcass weight of both groups. No significant differences were observed between sex, except for leg length, ham length and maximum perimeter ham, in agreement with Franco et al. (2014) in Celta pig crossbred with Duroc and Landrace genotypes.

The comparisons with the results obtained by other authors both local breeds and commercial breeds are hard because of differences in the carcass cutting between studies. Also, differences in regional quartering methods might explain these differences. The study of the most valuable meat cuts reveals some differences because of the rearing system. These results are consistent with the research of Daza et al. (2006) and Galián et al. (2009). In contrast, authors like Galián et al. (2008), in Chato Murciano pigs, reported higher weights on the carcass cuts from pigs reared outdoor. These discrepancies between studies could be attributed to the availability and quality of the feeding system in both indoor and outdoor system. Primal cut yield, except sirloin percentage, was not affected by sex confirming the results of Peinado et al. (2012) and Franco et al. (2014) in crossbreed pig (LD × LW dams × Duroc sires) and Celta pig crossbred, respectively. Ayuso et al. (2014) reported that the pieces obtained from Iberian pigs showed similar weights and yields in both sexes, with the exception of the forelegs (heavier in barrows). Conversely, Serrano et al. (2008) found that primal cut yield was lower for barrows than for females.

The proportion of fat in the carcass was lower, whereas the proportions of skin, lean and bone and fat/lean and lean/bone ratios were higher in the SC pigs than in the C pigs, in accordance to the higher backfat percentage observed in their carcasses (Pugliese et al., 2003). Carcass from gilts were leaner than barrows with lower percentage of intermuscular fat in the loin, sirloin and ham, while the fat contents in the shoulder and top loin were similar in both sexes which agree with Serrano et al. (2008) in Iberian pigs.

In several cuts of the carcass, the system × sex interaction was significant. Males reared in the SC system accumulated more fat than females, while in the C system differences between barrows and gilts were not found. Most foraging capacity by barrows might be the main cause of these differences. Something similar, though in the opposite direction was observed for lean content.

Average pH values and the pH fall in the first 24 hours postmortem (0.9) revealed an acceptable postmortem glycolysis and could be associated to an appropriate pre-slaughter handling. The differences in pH₄₅ (6.2) and pH₂₄ (5.3) values obtained in our study with those recorded by Ramos (2008) in Peruvian Creole pigs (6.9 and 5.7, respectively) could be attributed to the period of fasting, stress at slaughtering and the age of the animals. The no significant influence of sex on the pH was in line with the results recorded by other authors (Renaudeau & Mourot, 2007; Franco et al., 2014). However, Serrano et al. (2008) found that initial and ultimate pH were higher for barrows than for females. These discrepancies could be related to stress factors and differences in the glycolytic potential between sexes (Larzul et al., 1997).

The water holding capacity values obtained in our study for can be considered as normal, and were similar to those recorded by Tejerina et al. (2012) in Iberian pigs free-range or confinement rearing, Franco et al. (2014) in Celta pig crossbred, and Franci et al. (2005) in Cinta Sinese breed, although higher than those recorded by Galián et al. (2008) in Chato Murciano breed. In the literature on meat quality in different livestock species, is widely known the strong relationship between pH and water holding capacity. Our data did not show such differences in pH₂₄ in any of the two different rearing systems, which could explain the lack of differences in the variables drip and cooked losses. These results are in accordance to Tejerina et al. (2012), but contrary to Rosenvold & Andersen (2003), Pugliese et al. (2004), Franci et al. (2005) and Galian et al. (2008). There were no differences in water holding capacity between sexes, in accordance to Pugliese et al. (2004).

Average ash content of meat samples (1.3%) was slightly higher than reported by Mienkowska-Stepniewska et al. (2007) in five pig breeds (1.1%), although well below that recorded by Aro & Akinjokun (2012) in Large White × Duroc pigs (3-4%). When we compare the values of mineral composition of meat with the reference values provided by other authors (Muñoz & Ledesma, 2002) referred to the mineral composition of the pig LT, note that the levels of P, K, Mg, Fe and Cu are high whereas Ca, Na and Zn are within the normal values, which agrees with Galián et al. (2007) and Poto et al. (2007) in Chato Murciano breed. The Cu content could be related to the rearing system (“traspatio” involving more physical exercise that in in-door system) and muscle oxidative metabolism of this breed, in line with that indicated by Poto et al. (2007) in Chato Murciano pig and its crosses. The mean values obtained in elements like K, Cu, Fe and Zn were similar to those obtained by other authors, who found values of 330-349 mg/100 g for K (Poto et al., 2007), 0.2 mg/100 g for Cu or 2.3 mg/100 g for Fe and 1.4 to 2.5-2.7 mg/100 g for Zn (Guang-Zhi et al., 2008). The values for P, Ca and Mg appeared at a higher quantity than in results obtained by Muñoz & Ledesma (2002), Galián et al. (2007) and Poto et al. (2007). The average values of Mn were 0.033 mg/100 g, level similar than those found by McCance & Widdowson’s (1991), although lower than those recorded by Ramos (2008) in Peruvian Creole pigs (0.008 mg/100 g). Galián et al. (2007) and Poto et al. (2007) in Chato Murciano breed only found traces of manganese. Due to the limited information about Mn values in pork meat, we cannot discuss this item further. However, comparisons must be made with caution given the large variations found among breeds (Guang-Zhi et al., 2008) and feed regimes (Mienkowska-Stepniewska et al., 2007). Guang-Zhi et al. (2008) found significant differences between mineral elements in LT in five pig breeds (three local and two imported breeds). On the whole, local breeds had higher Na, Ca, Fe, Cu and Zn levels and lower K level than that in the imported breeds; while, no significant differences were found among the five breeds for Mn levels.

The rearing system had a slight effect on the levels of the mineral content of the meat. The pigs reared outdoors (SC system) showed higher levels of Ca. In previous studies (Galián et al., 2007, 2008, 2009; Poto et al., 2007), the pigs reared indoors showed higher levels of Fe, Cu and Zn, and lower levels of Mg, P and K than the pigs reared outdoors. The differences between the results of our study and the aforementioned works can be partially explained by factors such as weight at slaughter, genotype and physical exercise, among other.

The differences in the sensory analysis could be due to the characteristics of the rearing system, exercise and/or physical variables, but not be ascribed to a pH difference as no significant differences in pH were found between the different systems and sex. It is widely accepted that the tenderness is the most important factor affecting overall meat acceptability. The tenderness was positively correlated with cooking loss (0.475), not reaching statistical significance other correlations with postmortem meat instrumental traits. Nam et al. (2009) reported that tenderness score was correlated with L* and drip loss, and negatively correlated with pH₄₅. Also, cooking loss was positively correlated with brightness (0.482), while showed negative correlations with taste attributes as sweetness (-0.408) and metallicness (-0.383), which agrees with Nam et al. (2009). As the above authors reported, these results suggest that panelists do not prefer pork in terms of taste which has high cooking losses and is perceived as tough meat. So, it may be natural that taste acceptance decreases as cooking losses increase.

In summary, Creole is a breed of slow growth and with a clear tendency to fatness. The rearing system significantly influences the growth and quality of the carcass and meat, while the influence of sex was light.