Replacing rice bran and duckweed with ensiled Taro leaf-stem foliage (Colocasia esculenta) in diets of growing ducks

Local ducklings (n=160) were used in a production function design with 6 levels (0, 15, 30, 45, 60 and 75% as DM) of Taro leaf-stem silage replacing the basal diet of a mixture of rice bran and duckweed. There were three replicates of each level. Rice bran and duckweed were mixed and fed first at the theoretical level assuming the total DM intake for growing ducks was 70 g DM/kg LW. After the rice bran/duckweed was consumed the Taro silage was offered also at the theoretical level. In both cases the quantities were adjusted so as to minimize residues and to maintain the planned ratios between the Taro silage and the rice bran-duckweed.

Replacing the mixture of rice bran and duckweed by Taro silage resulted in a curvilinear decrease in DM intake. By contrast, there were improvements in live weight gain and especially in DM feed conversion as the rice bran and duckweed were replaced by Taro leaf and stem silage. The implication is that the silage made from the combined leaf and stem of the Taro plant has a higher nutritive value than the combination of rice bran and duckweed. The Taro plant is easily cultivated in most tropical latitudes, has yields of high biomass and is easy to ensile. These characteristics facilitate its use as the basis of complete diets for growing ducks, thus reducing the need for purchase of expensive feed supplements from off the farm.

Introduction

Cambodia is the poorest country in Southeast Asia and the majority of the population depends on agriculture for their food and income. The research and development strategy promoted by CELAGRID is to develop appropriate ways to improve the income of farmers, by using alternative feed resources which are not subject to human consumption. In order to develop duck production systems for poor farmers, it is important that the feed resources to be used can be available or grown in the local area as poor farmers do not have the cash to buy "commercial" animal feed. Rice bran is the cheapest and most widely available energy resource in Cambodia but it is low in protein. Local feed resources which are high in protein and potentially available for feeding ducks are duckweed, water spinach and other aquatic plants. The Taro plant (Colocacia esculenta), which grows naturally in ponds and uncultivated areas in SE Asian countries, is a recent addition to the list of potential protein sources for non-ruminant animals (Preston 2006).

Duckweed grows well in waste water and can double its weight in 24h (Leng et al 1995). Duckweed leaves have a low fiber content and the protein content can be as high as 35% (Leng et al 1995; Bui Huy Nhu Phuc et al 2000). In terms of protein production, grown under ideal conditions it can produce 10 tonnes of protein per hectare per year. This compares with soybean which produces less than 1 tonne/year. Under experimental conditions the annual production reached 183 tonnes/ha of DM, however, under practical conditions a yield of up to 30 tonnes of DM/ha is more feasible (Leng et al 1995). With appropriate fertilization of the pond water the protein content can be as high as 35-40% in the dry matter (Le Ha Chau 1998). It has been shown that fresh duckweed can replace completely the soybean meal in diets of growing ducks with no reduction in growth rate (Bui Xuna Men et al 1995).

The leaves of the Taro species (Colocia esculenta, Alocasia macrorrhiza, Xanthosoma sagittifolium) have attracted attention because they are rich in protein, and good sources of vitamins and minerals. Taro species are found growing in the wild state in mountain areas, in forests and in ponds and other water surfaces (Ngo Huu Toan and Preston 2007; Pheng Buntha et al 2008). The taro leaf has a high nutritional value (Du Thanh Hang and Preston 2009); however, it has an anti-nutritional substance, calcium oxalate, which is found in all parts of the plant, causing irritation in the throat and mouth epithelium and indirectly affecting the feed intake (Wikipedia No date). The influence of calcium oxalate can be reduced by ensiling with molasses (Malavanh Chittavong et al 2008), or by ensiling the leaves and stems together without any further additive (Rodríguez and Preston 2009; Du Thanh Hang and Preston 2010; Nguyen Tuyet Giang and Preston2011).

Hypothesis

It was hypothesized that taro silage can replace the duckweed and part of the rice bran in the diet of growing ducks with no reduction in growth rate.

Materials and methods

Location and duration

The experiment was carried out at the Center for Livestock and Agriculture Development located in Preah Theat village, Rolous commune, Kandal Stung district, Kandal province, approximately 25 km from Phnom Penh city. The trial started on 6th July 2010 and finished 3rd September 2010.

Experimental design and treatments

A production function design was used with 6 levels (0, 15, 30, 45, 60 and 75%) of Taro leaf-stem silage replacing the basal diet of a mixture of rice bran and duckweed. There were three replicates of each level.

Ducks, housing and management

A total of 162 local male ducks at one day old were purchased from a local hatchery. They were given commercial feed for the first 30 days after which they were adapted gradually to the experimental diets around one week before starting to collect the data. They were vaccinated against Duck Plague and Chlorella and de-wormed before starting the experiment. They were housed in pens (1.5m x 2m) in a simple house constructed from bamboo and wire net. There were 9 ducks in each pen. The concrete floor of the pens was covered with rice husks.

Experimental feeds and feeding

Duckweed was harvested from natural ponds near CelAgrid and was wilted overnight to reduce the water content before feeding to the ducks in the next morning. Taro foliage (Colocacia esculenta) (leaves plus stems) was harvested from ponds in the region near CelAgrid and was chopped into small pieces around 2-3 cm length with a knife and wilted 3-4 hours under sunlight before being packed tightly in plastic bags (50 liters capacity) without any additive. Rice bran was purchased from the rice mill near the Centre.

Rice bran and duckweed were mixed and fed first at the theoretical level (Table 1) assuming the total DM intake for growing ducks was 70 g DM/kg LW. After the rice bran/duckweed was consumed the Taro silage was offered also at the theoretical level. In both cases the quantities were adjusted so as to minimize residues and to maintain the planned ratios between the Taro silage and the rice bran-duckweed. Feeds were given three times daily at 8:00, 12:00 and 16:00h. Water was offered ad libitum in plastic bowls.

Table 1: Planned composition of the diets, % DM basis (based on prior analysis of the ingredients)
	T0	T15	T30	T45	T60	T75
Taro leaf - stem silage	0	15	30	45	60	75
Duckweed	31	25.5	20	14.5	9	3.5
Rice bran	67	57.5	48	38.5	29	19.5
Premix	1	1	1	1	1	1
Oyster shell	1	1	1	1	1	1
Total	100	100	100	100	100	100
% DM	61.1	55.1	49.1	43.1	37.1	31.1
% CP in DM	16.0	16.0	16.0	16.0	16.0	16.0

Data collection

The feeds offered and refused were recorded every day and samples of both taken for analysis of DM by microwave radiation (Undersander et al 1993) and N (AOAC 1990). The ducks were weighed as a group every 10 days during the 60 days of the experiment.

Statistical analysis

Data were analyzed by the General Linear Model in the ANOVA program of the Minitab software (MTAB 2000). Sources of variation were treatments and error. Regression analysis waas done with the same software.

Results and Discussion

Feed intake

The recorded ratios of of DM in the dietary ingredients consumed (Table 2) were closely related to the planned quantities (Figure 1). In the control diet (TF-0), 60% of the dietary crude protein was supplied by the duckweed, which was gradually replaced by the protein from the Taro leaf and stem silage to the point where 80% of the protein was from this source in the TF-75 diet (Figure 2). Replacing the mixture of rice bran and duckweed by Taro silage resulted in a curvilinear decrease in DM intake (Figure 3). The crude protein content of the DM consumed tended to decrease with increasing replacmentof rice bran and duckweed by Taro silage, especially in diets with 60 and 75% of Taro silage. Similar reductions in intake of DM and crude protein were observed by Nguyen Tuyet Giang et al (2010) when they fed growing ducks with increasing proportions of Taro silage in a mixture with rice bran.

Table 2. Mean value for intake of DM and crude protein by one duck fed rice bran and duckweed replaced with ensiled taro foliage
	TF-0	TF-15	TF-30	TF-45	TF-60	TF-75	SEM	P
DM intake, g/d
Taro silage	0	11.7	21.8	33.4	33.9	31.8	0.63	0.001
Duckweed	27.5	21	13.7	9.03	4.08	0.67	0.32	0.001
Rice bran	56.4	48.2	36.9	29.6	23.7	10.7	0.91	0.001
Total	83.9	80.9	72.4	72	61.7	43.1	1.29	0.001
DM, kg LW/d	82.8	82.2	76	76.6	68	50.7	1.3	0.001
CP intake, g/d
Taro	0	1.94	3.62	5.56	5.65	5.28	0.123	0.001
Duckweed	7.68	5.83	3.78	2.48	1.13	0.16	0.09	0.001
Rice bran	5.59	4.77	3.65	2.92	2.31	1.05	0.08	0.001
Total	13.3	12.6	11.0	10.9	9.07	6.50	0.215	0.001
CP, % in DM	15.7	15.4	15.1	15.0	14.8	14.5	0.14	0.001

Live weight gain and feed conversion

There were improvements in live weight gain and especially in DM feed conversion as the rice bran and duckweed were replaced by Taro leaf and stem silage (Table 3 and Figures 4 and 5). Similar improvements in feed conversion were reported by Nguyen Tuyet Giang et al (2010) when rice bran was replaced by Taro leaf and stem silage. Growth rates and feed conversion rates were poorer in the presemt study compared with those recorded by Nguyen Tuyet Giang et al (2010) but this can be explained by the greater body size of the "improved" ducks in their experiment (mature weight about 2.5 kg), compared with the smaller "unimproved" local ducks (mature weight about 1.3 kg) used in our study. There were also major differences in the quality of the rice bran which was richer in protein (13%) and oil (17%) in the experiment of Nguyen Tuyet Giang et al (2010) compared with the rice bran used in our study.

Table 3: Mean value of live weight gain and DM feed conversion ratio of duck on replacing rice bran with ensiled taro foliage in diets per head
Live weight, g	TF-0	TF-15	TF-30	TF-45	TF-60	TF-75	SEM	P
Initial	711	685	637	659	652	619	26.9	0.25
Final	1218	1225	1225	1203	1216	1204	32.8	0.99
Daily gain	8.27	8.50	9.97	9.30	10.5	10.2	0.51	0.045
DM conversion	10.2	9.67	7.29	7.79	5.86	4.23	0.58	0.001

Conclusions

Acknowledgments

The authors would like to express the gratitude to the MEKARN project financed by the SIDA-SAREC Agency and to the Center for Livestock and Agriculture Development (CelAgrid), for providing resources for conducting this experiment.

References

AOAC 1990 Official Methods of Analysis. Association of Official Analytical Chemists. 15th edition (K Helrick editor). Arlington pp 1230

Bui Huy Nhu Phuc, Jan Erik Lindberg, Brian Ogle B and Sigvard Thomke 2000 Nutritive evaluation with rats of tropical biomass for monogastrics: 1. Comparison of eight green products evaluated at two level of inclusion in relation to a casein diet, Doctoral thesis, Swedish University of Agricultural Sciences Uppsala 2000

Bui Xuan Men, Ogle B and Preston T R 1995 Use of duckweed (Lemna spp) as replacement for soya bean meal in a basal diet of broken rice for fattening ducks. Livestock Research for Rural Development. Volume 7.(3) Art. #2. http://www.lrrd.org/lrrd7/3/2.htm

Du Thanh Hang and Preston T R 2009 Taro (Colocacia esculenta) leaves as a protein source for growing pigs in Central Viet Nam. Livestock Research for Rural Development. Volume 21, Article #164. http://www.lrrd.org/lrrd21/10/hang21164.htm

Du Thanh Hang and Preston T R 2010 Effect of processing Taro leaves on oxalate concentrations and using the ensiled leaves as a protein source in pig diets in central Vietnam, from http://www.lrrd.org/lrrd22/4/hang22068.htm

Leng R A, Stambolie J H and Bell R 1995 Duckweed. A potential high protein feed resource for domestic animal and fish. Proc. 7^th AAAP Animal Science Congress Improving Animal Production Systems Based on Local Feed Resources pp. 103-117

Le Ha Chau 1998 Biodigester effluent versus manure, from pigs or cattle, as fertilizer for production of cassava foliage (Cassava Esculenta). Livestock Research for Rural Development. Volume 10, number 3, Article #27. http://www.lrrd.org/lrrd10/3/chau2.htm

Malavanh Chittavong, Preston T R and Ogle B 2008b Ensiling leaf of Taro (Colocasia esculenta (L.) Shott) with sugar cane molasses. Livestock Research for Rural Development. Volume 20, supplement, from http://www.lrrd.org/lrrd20/supplement/mala1.htm

Minitab 2000 User's guide to statistics. Minitab Statistical Software Version 13.31, Minitab Inc., USA.

Men Bui Xuan, Ogle B and Preston T R 1995 Use of duckweed (Lemnaspp) as replacement for soybean meal in a basal diet of broken rice for fattening ducks. Livestock Research for Rural Development, 7 (3) http://www.cipav.org.co/lrrd/lrrd7/3/2.htm

Ngo Huu Toan and Preston T R 2007 Evaluation of uncultivated vegetables for pigs kept in upland households. Livestock Research for Rural Development. Volume 19, Article #150. http://www.lrrd.org/lrrd19/10/toan19150.htm

Nguyen Tuyet Giang, Preston T R and Ogle B 2010 Effect on the performance of common ducks of supplementing rice polishings with taro (Colocacia esculenta) foliage. Livestock Research for Rural Development. Volume 22, Article #194. http://www.lrrd.org/lrrd22/10/gian22194.htm

Nguyen Tuyet Giang and Preston T R 2011 Taro (Colocasia esculenta) silage and water spinach as supplements to rice bran for growing pigs, Livestock Research for Rural Development. Volume 23, Article #45. Retrieved , from http://www.lrrd.org/lrrd23/3/gian23045.htm

Pheng Buntha, Borin K, Preston T R and Ogle B 2008 Survey of taro varieties and their use in selected areas of Cambodia. Livestock Research for Rural Development. Volume 20, supplement. Retrieved June 29, 2009, from http://www.lrrd.org/lrrd20/supplement/bunt1.htm

Rodríguez L and Preston T R 2009 A note on ensiling the foliage of New Cocoyam (Xanthosoma sagittifolium). Livestock Research for Rural Development. Volume 21, Article #183. http://www.lrrd.org/lrrd21/11/rodr21183.htm

Undersander D, Mertens D R and Theix N 1993 Forage analysis procedures. National Forage Testing Association. Omaha pp 154