The care and use of the pigs was performed following the Certificate of Authorization to experiment on Living Animals from UKZN Animal Ethics Committee (Reference Number: 096/11/Animal).

The experiment was conducted at Ukulinga Research farm, Pietermaritzburg, South Africa. The farm is located at 29°40′ S, 30°24′ E with the latitude of about 775 m above sea level. Daily temperatures average 29°C, with variation ranging from 28.2°C to 43°C. Mean annual rainfall is 735 mm, mostly received in summer, with light to moderate frost occurring occasionally in winter [10].

Twenty eight health clinically weaner (male, Pig Improvement Company) pigs weighing an average of 18±2.3 kg were used over a 4-week experimental period including 10 day of adaptation period. They were housed in a room with artificial lighting, heating and ventilation systems. The pigs were confined individually in pens measuring 1.5×1.0 m², containing plastic self-feeders trough and a low-pressure nipple drinker (Big Dutchman Lean Machine, Vechta, Lower Saxony, Germany). The ambient temperature and relative humidity were recorded throughout the experiment at 15 minute intervals using a HOBO TE MPERATURE, RH, 1996 ONSET logger (Onset Computer Corporation, Bourne, MA, USA). The house conditions were kept at a temperature of 21.9°C±2.24°C, 45.2% ±6.85% relative humidity and a 12 h dark-12 h artificial light cycle. The feed and water were provided ad libitum. The three fiber sources were chosen based on the availability in the tropics, and to dilute the basal diet. The LH, MC, and SH were ground to pass through a 2 mm screen. The fiber sources were then included at different inclusion levels of 0, 80, and 160 g/kg dry matter (DM) in diet of pigs. It was assumed that the incorporation of fiber would influence the physico-chemical characteristics excreta of pigs.

The conventional diet (Express Weaner) was purchased from a local feed manufacturing company (Meadow Feeds Limited, Pietermaritzburg, South Africa) with a low level of dietary fiber of 50 g/kg DM of total dietary fiber. Ingredients used in the formulation of basal diet were; yellow maize (426 g/kg), soya bean (176 g/kg), soybean oil cake (83 g/kg), wheat bran (100 g/kg), whole wheat (100 g/kg), Oil-sunflower (75 g/kg), cape fish (20 g/kg), and additives (20 g/kg). Table 1 shows chemical analysis of diet containing different fiber sources used in the study. The diet was designed to meet all the nutrient requirements of the pigs. Each pig, representing an experimental unit, was used in completely randomized design. All the four randomly selected pigs per treatment were fed into each of the seven diets including a control.

An amount of 250 g of feces was collected from each pig immediately after defecation. Urine collection was done through placing plastic trays underneath each pen to allow collection of urine that would have oozed through a 1 mm. To reduce the volatilization of nitrogenous compounds in the urine collected, 2 mL of 25% sulphuric acid was added to each tray within 5 minutes of collection. Collected feces and urine samples were stored at 4°C, pending analyses.

The excreta from each pig receiving the same treatment was mixed in 3 L buckets to make slurry for all the samples collected. Slurry was made by mixing feces and urine in a ratio of 1:2.3 (w/w basis of feces and urine) according to Canh et al [11]. Amounts of 200 g of feces and 460 g of urine per pig were sampled on day 1 and 16 of the incubation period of slurry. The slurry was left to stand and ferment at room temperature for a 16 d period in their respective buckets. Sampling for chemical analyses was done on day 1 and 16 of the incubation period of the slurry.

The weight of the experimental pigs was recorded weekly. Feed intake was determined weekly to compute the average daily feed intake (ADFI). The pigs were being weighed once a week to estimate average daily gain (ADG), by weighing feed supplied in and feed left each week. Then feed conversion ratio (FCR) was calculated by dividing feed intake by weight gain. The nitrogen content of the slurry was analyzed using the diffusion technique for soil nitrogen fractionation described by Mulvaney et al [12]. The pH of the slurry was measured on a pH meter, calibrated with certified pH 4 and 7 buffer solutions according to procedures by Lynch et al [13]. Concentrations of acetate, propionate, iso-butyrate, butyrate, and valerate were done following procedures described by Otto et al [14]. The chemical oxygen absorbencies were determined according to Quayle et al [15]. Odor offensiveness was determined using 18 panelists from day 1 and day 16 of incubation period at the Discipline of Animal and Poultry Science at the University of KwaZulu-Natal. Hedonic tone was used to evaluate the odor offensiveness, which was defined as the character of odor (unpleasantness or pleasantness) observed above the odor detection threshold, as described by Otto et al [14].

The effect of dietary fibre source and the fibre inclusion level on the chemical composition of the slurry for two incubation periods was analysed using PROC general linear model of SAS [16]. The PDIFF procedure of SAS [16] was used to separate and compare treatment means. Pair-wise comparisons of means were performed using the PDIFF option. Effects could be considered significant when probabilities were below 0.05. The model used was:

Where,

Data on odour offensiveness were square-root transformed before analyses. Pearson’s correlation test was run to cater for the relationship that could exist between the chemical composition of the slurry and the odour offensiveness numerical scales obtained by olfactory sensory evaluation.

Table 2 gives the impact of LH, MC, and SH incorporation on ADFI, ADG, and FCR of pigs. The ADFI decreased linearly with inclusion level of all fibre based diets (p<0.05). There was an effect of SH in ADG (p<0.05). There was a non-significant relationship between inclusion level of LH, MC, and SH. Increasing levels of all fibre diets illustrated a non-significant effect on FCR (p>0.05).

The levels of significance for the chemical composition of pig slurry fed different fibre sources at varying inclusion levels and incubation periods are portrayed in Table 3. Fibre source affected pH, propionate, iso-butyrate, butyrate (p<0.05), COD and nitrogen content (p<0.01). Their inclusion level, on the other side, had an impact on total short chain fatty acid (TSCFA) concentrations (p<0.05) particularly butyrate (p<0.05) and valerate (p<0.01). The pH, COD, TSCFA concentrations were influenced by period of incubation (p<0.01). An interaction between fibre source and fibre inclusion level on content of nitrogen and pH was observed (p<0.05). Interaction between fibre source and incubation period had effect on the pH, nitrogen content, the TSCFAs (p<0.05), including iso-butyrate and butyrate (p<0.01). Interaction among fibre sources, fibre inclusion level and the incubation period affected valerate (p< 0.05) and butyrate (p<0.01).

The pH of the slurry was different across all fibre sources during day 1 of incubation period (Table 4). The slurry from pigs fed MC and SH, was, however similar on day 16. The COD was also similar across all fibre sources on the first day of incubation (p>0.05). On day 16, the slurry of pigs fed into SH had a high concentration of COD as compared to the slurry of LH. Nitrogen concentration of slurry from LH tended to be significant (p>0.05). The concentration of TSCFAs were high for SH during day 1, but were highest and the same across all fibre sources during day 16 compared to day 1. The similarity of acetate, propionate and valerate concentration were observed across fibre sources (p>0.05). Isobutyrate and butyrate were the same on day 1, but was largest for MC on day 16 (p< 0.05).

Figure 1, illustrates the effects of dietary fibre inclusion on odour offensiveness. Odour offensiveness of the slurry varied amongst all fibre sources tested (p<0.01). Slurry from pigs fed on diets containing MC had a high odour offensiveness scores than pigs fed into SH. On day 1, the slurry from pigs fed MC was less offensive than the slurry on day 16. The offensiveness of the slurries from pigs fed diets containing LH and SH was similar for both days.

The correlation coefficients are shown in Table 5. There was a positive correlation amongst all chemical parameters tested, except for pH which was negatively correlated to all the chemical parameters tested (p<0.05).