I. INTRODUCTION
In Tanzania, a majority of the SM farmers use mash diets for feeding broiler chickens, and this has resulted in lower performance. The feed industry in Tanzania is less developed due to lack of analytical facilities, low quality of ingredients and lack of government policy enforcement on feed quality (Geerts 2014). Improving feed quality through physical processing and microbial enzyme supplementation can enhance broiler performance. Feeding pellets improves nutrient density, reduces wastage during feeding and reduces feed selection by the birds (Glover et al. 2015). Grinding and pelleting disrupt the physical structure of dietary ingredients, thus exposing the feeds to the action of digestive enzymes, enabling high performance of broilers (Ghobadi and Karimi 2012). The application of microbial enzymes in poultry diets is important because poultry lack endogenous enzymes that can break cell wall structures from vegetable sources in feed ingredients (Ghobadi and Karimi 2012). Adding microbial enzymes will improve nutrient availability and also reduce costs related to production by increasing the productivity of chickens in Tanzania.
II. MATERIALS AND METHODS
A total of 480 day-old unsexed Ross 308 broiler chickens was randomly assigned to a 2 × 4 factorial arrangement of treatments, with two feed forms (mash and pellet), without enzyme, or supplemented (at 100 mg/kg) with Axtra XB (xylanase and beta-glucanase composite), QB (phytase) or a combination of Axtra XB + QB. Eight dietary treatments were compounded mainly from maize, soybean, cottonseed and fish meals. Birds were allocated to six replicates per treatment (10 birds per replicate) and housed in a climate-controlled deep-litter system with feed and water supplied ad libitum from 0 to 35 days. Birds were housed at the Centre for Animal Research and Teaching, University of New England, Armidale, Australia and raised in three phases, starter (d1-10), grower (d11-24) and finisher (d25-35). Measurements of the feed intake (FI) and body weight gain (BWG) were taken on d10, d24 and d35 to calculate feed conversion ratio (FCR). On d35 two birds per replicate were also euthanised and dissected to obtain the dressing percentage as well as absolute and relative weights of breast, thighs and drumsticks. Data were analysed using the general linear model of Minitab version 17 (Minitab 2014). Least square means were computed using the Tukey pairwise comparisons test, and considered to be significant at P ≤ 0.05.
III. RESULTS
Broiler gross performance (FI, BWG and FCR) results are shown in Table 1. Feed intake to d24 and d35 was higher (P < 0.05) for birds raised on pelleted diets compared to mash diets. Microbial enzyme supplementation increased FI (P < 0.05) at d24 and d35, with higher FI observed when a combination of Axtra XB and QB was used. The BWG to d24 and d35 was increased (P < 0.001) when birds were fed pelleted diets and when diets were supplemented with enzymes (P < 0.001) at both d24 and d35. The FCR for all age groups of birds was significantly (P < 0.001) reduced when feeding pelleted diets. However, enzyme supplementation decreased (P < 0.007) FCR only on day 24, with the combination of Axtra XB and QB resulting in the lowest FCR. There was no interaction effect between the main factors (feed form and enzyme supplementation) for any response variables (P > 0.05). The dressing percentage and relative weights of breast, thighs and drumsticks were superior (P < 0.001) in birds fed pelleted diets and the diets supplemented with enzymes, although the relative weight of drumsticks was not affected (P > 0.05) by enzyme (Table 2).
IV. DISCUSSION
The improved gross performance of broiler chickens consuming pelleted diets with or without enzyme supplementation could be due to increased nutrient availability, improved digestibility, reduced feed wastage and energy spent by bird during eating (Serrano et al. 2012; Abdollahi et al. 2013). These results agree with the findings of Truong et al. (2015) who reported that pelleting diets increases FI, reduces feed selection and improves nutrient digestibility. The result is also in line with those obtained by Lv et al. (2015) who found that broiler chickens fed with pelleted diets had better performance for the entire period of study. The improved BWG in this study could be due to the increase in feed intake (Amerah et al., 2008). The poorer performance in mash-fed birds might be a reflection of reduced ability of birds to eat bulkier, and possibly less palatable diets, which would result in lower nutrient availability (Brickett et al. 2007). The FCR values observed in this study are to some extent similar to those reported by Mabelebele et al. (2018) who found that broiler chickens fed pelleted diets had better growth performance and FCR than mash-fed broiler chickens.
The increased dressing percentage and meat yield (breast, thighs and drumsticks) of birds could be directly related to increased body weight as a result of feeding pelleted diets and or enzyme supplementation. Furthermore, pelleted diets improve the ingestion and passage rate of feeds leading to increased growth of bird and body parts weight (Amerah et al. 2007a). The heavy breast and thighs in this study are related to enzyme addition, which is similar to those reported by Erdaw et al. (2015) and Abdallh et al. (2017) who observed that adding microbial enzymes to broiler diets increases weight gain and meat yield. It can be concluded that adopting and applying these technologies will be beneficial to the SM poultry farmers in Tanzania. It will be necessary to assess the cost-benefit analysis of the diets when fed in Tanzania.
ACKNOWLEDGEMENT: The authors are grateful to Australia Awards scholarship and University of New England for financial support.
Abstract presented at the 30th Annual Australian Poultry Science Symposium 2019. For information on the latest edition and future events, check out https://www.apss2021.com.au/. 
