Evaluation of ensiling methods to preserve sweet potato roots and vines as pig feed

Hoang Huong Giang, Le Viet Ly and Brian Ogle*

National Institute of Animal Husbandry, Ha Noi, Viet Nam
*Department of Animal Nutrition and Management, Swedish University of Agricultural Sciences, Uppsala, Sweden
vokimthanh@hotmail.com

Abstract

A laboratory experiment on ensiling sweet potato (SP) included 5 different ratios of sweet potato roots (SPR) and vines (SPV): 70, 60, 50, 40 and 30% of SPR with 30, 40, 50, 60 and 70% of SPV on a dry matter basis, respectively, giving treatments SP7:3, SP6:4, SP5:5, SP4:6 and SP3:7. Samples of SP silage were analysed at 0, 7, 14, 21, 28, 42, 70 and 84 days after ensiling to determine chemical composition and fermentation and physical characteristics. When SPR levels decreased from treatment SP7:3 to treatment SP3:7 the colour changed from yellow to a deeper green because of the dark green colour of SPV. However, the colour did not change from 14 to 84 days of ensiling. The silage on all treatments had a good smell at all times up to 84 days. With increasing ensiling time, dry matter content increased and crude protein decreased in all treatments, but the changes were not significant. Other chemical components such as NDF, calcium, and phosphorus did not change during the 84 days of ensiling in all treatments. The pH value in all treatments decreased rapidly in the first week (from around 6.4 to around 3.8) (P<0.01) and continued to decrease up to day 14 (to around 3.6), then remained low until 84 days. Acetic acid and lactic acid increased quickly in the first 2 weeks (P<0.01) and then remained constant to 84 days (P>0.05). However, there was an effect of ratio of SPR:SPV on pH and the organic acids. As the SPR level of the SP silage increased pH values decreased (during the 84 days of ensiling) and lactic acid increased during the 2 first weeks of ensiling (P<0.05). The NH3-N content in all treatments fluctuated at around 2-3% of total nitrogen and the differences were non- significant during the whole period of ensiling (P>0.05) as well as between treatments (P>0.05). All of the treatments resulted in good quality products.

Key words: Organic acids, protein, roots, silage, sweet potato, vines.

Introduction

In Vietnam around 269,000 ha of land are used for growing sweet potato, producing 1,745,300 tonnes of sweet potato roots in 1998 (Statistical Year Book 1999). The root is a good energy source (15.6 MJ of metabolizable energy/kg dry matter) and the vine is a source of protein (17.7% CP in dry matter) (National Institute of Animal Husbandry 2001), and both are considered as good livestock feeds. One of the harvests of sweet potato coincides with the wet season, which means that large amounts of sweet potato need to be stored for use in the off-season. Sweet potato roots (SPR) have a delicate skin that is very easily broken, so the flesh also is easily bruised, broken or cut, which makes for good conditions for bacteria to attack, resulting in the sweet potato decaying. In addition, the stored roots often come under attack from weevils and rats, and farmers may lose as much as half of their stored feed. Sweet potato vines (SPV), especially after harvesting, are considered as waste, because animals cannot consume all of the huge amounts produced in the short time available before the vine decays, which occurs within 2 or 3 days. However, the vine is very expensive to purchase during the off-season.

Ensiling by-products is a simple and low-cost option, which can preserve feed for long periods (Lien et al 1994). Ensiling can also render some previously unpalatable products useful to livestock by changing the chemical nature of the feed (Kayouli and Lee 1998). Tinh et al (2000) concluded that SPV ensiled with chicken manure resulted in the highest quality feed, increasing diet crude protein content and dry matter conversion rates. Tinh et al (2001) reported that rice bran, cassava leaf meal and chicken manure are good additives for fermenting SPR, in combination with salt, and the fermented SPR can be stored at least for 90 days in the laboratory without any significant reduction of the quality, and for 4 to 5 months on farm.

This research then focuses on the effects of ensiling SPR and SPV together in different ratios, without any additives, and the aim was to preserve sweet potato for several months and evaluate it for later use as fattening pig feed in the off-season.

Materials and methods

Location

The experiment was conducted in the laboratory of the Animal Nutrition Department of the National Institute of Animal Husbandry, Hanoi, Vietnam (NIAH) from February to April 2002.

Sweet potato vine and root preparation

After harvesting the sweet potato vines were chopped into very small pieces (1-2 cm) and the root was washed to remove soil, then ground (1-2 mm) by machine.

Mixing of sweet potato roots and vines

The prepared SPR and SPV were mixed together in 5 different ratios as 5 experimental treatments, namely 70, 60, 50, 40, 30 % of SPR with 30, 40, 50, 60, 70% of SPV, respectively, on a dry matter basis, without any additives, and designated SP7:3, SP6:4, SP5:5, SP4:6, SP3:7, respectively. Each SPR and SPV mixture had a total weight of 24 kg (fresh basis) after mixing, and was then divided into 24 equal parts (each 1 kg) and placed in 24 plastic bags that were sealed to avoid air contamination. Three bags from a treatment were put into a 10 litre covered jar to prevent external mechanical damage and each jar thus represented 3 replications of one sample analysed for chemical composition for every ensiling period (there were 8 different ensiling periods). All jars were stored at room temperature (25-30^o C).

Measurements

Each jar of 3 bags of sweet potato silage (SPS) on each treatment included 3 replications, with samples taken at: 0, 7, 14, 21, 28, 42, 70 and 84 days after ensiling for analysis of chemical composition, including dry matter (DM), crude protein (CP), NDF, calcium (Ca), and phosphorus (P) and fermentation characteristics such as organic acids, pH and NH₃. Physical characteristics of SPS such as colour and smell were also observed and recorded.

The analyses were done in the Department of Feed Analysis of NIAH. DM, CP, NDF, Ca and P were determined by using standard AOAC procedures (AOAC 1990). pH was measured in the liquid extracted from SPS samples.

Statistical analysis

The data were analysed using the General Linear Model procedure of ANOVA in MINITAB 12.21 program (1998). Tukey pairwise comparisons were used to determine the differences between treatments with confidence level 95.0%.

Results

Physical characteristics

The SP silage had a pale yellow colour from the SPR and a dark green colour from the SPV. In the first three treatments (SP7:3, SP6:4 and SP5:5) the yellow colour was dominant, while the dark green colour of SPV was dominant in treatments SP4:6 and SP3:7. The colour in all treatments did not change after 14 days of ensiling up to 84 days. The silage on all treatments had a good smell, that did not change up to 84 days.

Chemical composition

The analysed chemical composition of fresh sweet potato shows that the DM of SPR (19.1%) was higher than the DM of SPV (15.0%). Other parameters, namely CP, NDF, Ca and P in SPR were much lower than in SPV (dry matter basis) (Table 1), which resulted in DM decreasing while CP, NDF, Ca and P were increasing from treatment SP7:3 to SP3:7 at 0 days.

Table 1. Analyzed chemical composition of sweet potato vines (SPV) and sweet potato roots (SPR) (% of dry matter)
Parameter	SPV	SPR
Dry matter, %	15	19.1
Crude protein	16.2	4
NDF	29.8	13.9
Calcium	1.16	0.4
Phosphorus	0.42	0.23

With increasing ensiling time, DM contents increased and CP decreased in all treatments, but the changes were not significant (P>0.05). However, DM and CP contents were different between treatments at all sampling times, due to different SPR and SPV ratios. NDF, Ca and P in all treatments were unchanged during the 84 days of ensiling (P>0.05) (Table 2, 3 and 4).

Table 2. Effect of sweet potato root and vine ratio on dry matter (DM, %) and crude protein (CP, % of DM) content in sweet potato silage
Parameter	Treatment^*	Time of ensiling, days								SE	P
Parameter	Treatment^*	0	7	14	21	28	42	70	84	SE	P
DM	SP7:3	18.0	18.3	18.3	18.7	18.3	18.5	18.8	19.0	0.329	0.426
	SP6:4	17.2	17.3	17.4	17.7	17.5	17.6	17.9	17.9	0.293	0.712
	SP5:5	16.8	16.9	17.0	16.7	17.0	17.2	17.3	17.6	0.248	0.258
	SP4:6	15.9	16.1	16.2	16.1	16.3	16.3	16.0	16.3	0.218	0.838
	SP3:7	15.1	15.2	15.3	15.3	15.3	15.4	15.2	15.2	0.138	0.831
CP	SP7:3	8.1	8.1	8.0	8.0	7.9	7.8	7.9	7.9	0.134	0.762
	SP6:4	9.4	9.2	9.3	9.1	9.2	9.1	8.9	9.0	0.199	0.784
	SP5:5	10.5	10.3	10.3	10.3	10.2	10.2	10.3	10.1	0.134	0.745
	SP4:6	11.7	11.4	11.5	11.2	11.3	11.2	11.3	11.2	0.285	0.877
	SP3:7	13.0	12.9	12.9	12.9	13.0	12.7	12.6	12.5	0.208	0.565
*SP7:3 consists of 70%SPR and 30%SPV, SP6:4 consists of 60%SPR and 40%SPV, SP5:5 consists of 50%SPR and 50%SPV, SP4:6 consists of 40%SPR and 60%SPV, SP3:7 consists of 30%SPR and 70%SPV SPR: Sweet potato roots SPV: Sweet potato vines

Table 3. Effect of sweet potato root and vine ratio on ammonia nitrogen (NH₃-N, % of total N) and NDF content in sweet potato silage
Parameter	Treatment^*	Time of ensiling, days								SE	P
Parameter	Treatment^*	0	7	14	21	28	42	70	84	SE	P
N-NH₃	SP7:3	2.08	2.20	2.15	2.45	2.35	2.19	2.49	2.56	0.155	0.310
	SP6:4	2.28	2.52	2.45	2.51	2.61	2.26	2.40	2.58	0.286	0.977
	SP5:5	2.19	2.43	2.76	2.76	2.58	2.67	2.56	2.64	0.227	0.676
	SP4:6	2.24	2.39	2.53	2.85	2.69	2.50	2.65	2.37	0.217	0.585
	SP3:7	2.21	2.67	2.75	2.86	2.75	2.31	2.88	3.09	0.251	0.286
	SE	0.321	0.493	0.158	0.108	0.114	0.144	0.212	0.174
	P	0.993	0.969	0.126	0.075	0.225	0.227	0.585	0.142
NDF	SP7:3	19.5	18.4	18.2	18.2	18.6	19.2	18.3	18.5	0.332	0.126
	SP6:4	22.7	22.5	21.3	20.7	21.6	20.2	20.8	21.6	0.792	0.379
	SP5:5	24.0	24.0	22.8	22.6	23.3	22.2	23.4	22.1	0.665	0.332
	SP4:6	26.5	25.3	25.3	25.6	26.3	27.1	25.2	25.1	0.562	0.174
	SP3:7	28.1	27.8	27.7	27.9	27.6	28.1	27.8	27.6	0.275	0.710
*See footnote in table 2

Table 4. Effect of sweet potato root and vine ratio on calcium (Ca, % of DM) and phosphorus (P, % of DM) content in sweet potato silage
Parameter	Treatment^*	Time of ensiling, days								SE	P
Parameter	Treatment^*	0	7	14	21	28	42	70	84	SE	P
Ca	SP7:3	0.58	0.55	0.56	0.62	0.61	0.59	0.61	0.59	0.020	0.172
	SP6:4	0.67	0.63	0.67	0.63	0.65	0.70	0.67	0.66	0.018	0.345
	SP5:5	0.80	0.79	0.77	0.79	0.83	0.81	0.82	0.81	0.014	0.198
	SP4:6	0.85	0.88	0.85	0.86	0.87	0.91	0.82	0.83	0.023	0.306
	SP3:7	0.92	0.93	0.95	0.97	0.94	1.00	0.97	0.93	0.024	0.286
P	SP7:3	0.26	0.27	0.24	0.26	0.28	0.25	0.25	0.24	0.016	0.790
	SP6:4	0.30	0.33	0.31	0.34	0.30	0.33	0.29	0.32	0.014	0.329
	SP5:5	0.32	0.31	0.30	0.29	0.33	0.30	0.31	0.31	0.012	0.457
	SP4:6	0.34	0.35	0.32	0.34	0.34	0.32	0.35	0.32	0.011	0.291
	SP3:7	0.37	0.35	0.35	0.33	0.36	0.37	0.37	0.38	0.017	0.683
*See footnote in table 2

pH, organic acids and NH3-N contents

The decreases in pH values were very rapid in the first week (from around 6.3 to around 3.8) (P<0.01) and continued to decrease in the second week to a final value of around 3.6 (P<0.05), and then remained low up to day 84 (P>0.05) (Table 6). However, the pH values increased when SPV level increased in the treatments after mixing (at 0 day) and the differences remained up to 84 days of ensiling (P<0.05).

There was a significant decrease in lactic acid content from treatment SP7:3 to SP3:7 at 0 and 7 days of ensiling (P<0.01), but a non-significant difference in acetic acid concentration in all treatments (P>0.05). Butyric acid was very low and changed little during ensiling in all treatments (P>0.05) (Table 5).

Table 5. Effect of sweet potato root and vine ratio on acetic and butyric acid (g/kg DM) content in sweet potato silage
Parameter	Treatment^*	Time of ensiling, days								SE	P
Parameter	Treatment^*	0	7	14	21	28	42	70	84	SE	P
Acetic acid	SP7:3	4.4^a	17.8^b	24.2^c	26.2^c	25.7^c	26.9^c	24.7^c	27.5^c	0.874	0.000
	SP6:4	4.4^a	16.5^b	24.3^c	24.1^c	27.0^c	24.2^c	25.2^c	25.8^c	1.181	0.000
	SP5:5	4.9^a	18.1^b	23.6^c	24.7^c	23.89^c	24.5^c	24.6^c	25.2^c	1.625	0.000
	SP4:6	5.2^a	17.4^b	24.2^c	26.3^c	25.2^c	23.8^c	24.8^c	24.0^c	0.543	0.000
	SP3:7	4.2^a	16.7^b	25.0^c	25.9^c	26.3^c	24.4^c	24.3^c	26.6^c	0.638	0.000
	SE	0.219	0.403	1.935	1.590	0.712	0.874	0.797	0.820
	P	0.062	0.089	0.991	0.830	0.108	0.185	0.955	0.113
Butyric acid	SP7:3	0.53	0.33	0.27	0.49	0.34	0.47	0.50	0.48	0.141	0.749
	SP6:4	0.50	0.41	0.37	0.48	0.38	0.39	0.47	0.43	0.068	0.803
	SP5:5	0.56	0.40	0.45	0.34	0.39	0.43	0.45	0.42	0.080	0.695
	SP4:6	0.49	0.46	0.45	0.57	0.40	0.37	0.47	0.48	0.107	0.932
	SP3:7	0.49	0.48	0.43	0.54	0.35	0.33	0.47	0.45	0.068	0.386
*See footnote in table 2 ^a,b,c within rows, values with different superscript letters are significantly different (p<0.05)

There were significant increases of acetic acid and lactic acid after the first week of ensiling, especially lactic acid in all treatments. This upward trend continued until 14 days (P<0.01), and then levels remained constant to 84 days of ensiling (P>0.05) (Table 6).

Table 6. Effect of sweet potato root and vine ratio on lactic acid content (g/kg DM) and pH in sweet potato silage
Parameter	Treatment^*	Time of ensiling, days								SE	P
Parameter	Treatment^*	0	7	14	21	28	42	70	84	SE	P
Lactic acid	SP7:3	^x21.7^a	^x88.5^b	123.9^c	127.6^c	126.2^c	124.5^c	124.9^c	123.0^c	1.847	0.000
	SP6:4	^xy20.7^a	^x87.7^b	123.1^c	125.0^c	123.3^c	122.9^c	124.3^c	122.0^c	1.321	0.000
	SP5:5	^y17.4^a	^y74.2^b	121.7^c	120.9^c	122.5^c	120.9^c	119.2^c	118.7^c	1.866	0.000
	SP4:6	^z11.6^a	^yz67.0^b	119.2^c	121.6^c	119.7^c	121.4^c	119.2^c	119.5^c	1.499	0.000
	SP3:7	^z10.3^a	^z64.6^b	118.3^c	120.1^c	119.6^c	120.5^c	119.5^c	119.5^c	1.265	0.000
	SE	0.702	1.556	1.706	2.005	1.516	1.049	1.766	2.202
	P	0.000	0.000	0.188	0.130	0.070	0.125	0.106	0.609
PH	SP7:3	^x6.26^a	^x3.74^b	^x3.57^c	^x3.60^c	^x3.60^c	3.59^c	^x3.59^c	^x3.59^c	0.041	0.000
	SP6:4	^xy6.35^a	^xy3.79^b	^xy3.61^c	^x3.60^c	^xy3.61^c	3.62^c	^y3.62^c	^y3.62^c	0.012	0.000
	SP5:5	^yz6.47^a	^xyz3.84^b	^xy3.60^c	^xy3.62^c	^yz3.62^c	3.63^c	^y3.64^c	^y3.64^c	0.028	0.000
	SP4:6	^z6.53^a	^yz3.93^b	^y3.64^c	^y3.63^c	^z3.64^c	3.64^c	^y3.63^c	^y3.63^c	0.010	0.000
	SP3:7	^z6.55^a	^z3.94^b	^y3.66^c	^y3.64^c	^z3.64^c	3.63^c	^y3.64^c	^y3.64^c	0.021	0.000
	SE	0.026	0.031	0.012	0.005	0.004	0.010	0.005	0.005
	P	0.000	0.008	0.006	0.002	0.001	0.074	0.002	0.001
*See footnote in table 2 ^a,b,c within rows, values with different superscript letters are significantly different (p<0.05) ^x,y,z within columns, values with different superscript letters are significantly different (p<0.05)

The NH₃-N content in all treatments fluctuated around 2-3 % of total nitrogen and tended to increase with increasing ensiling time (P>0.05). There was no significant difference in NH₃-N between treatments (P>0.05) (Table 3).

Discussion

The SPR are rich in energy due to the high carbohydrates content (between 80 to 90% of the DM) (Dominguez 1992), and SPV are relatively high in protein and contain only 8% starch and 4% sugars in DM (Onwueme 1978 cited by Winarno 1982). Based on these data, the 5 different mixtures of SPR and SPV in present study should have carbohydrate contents from 32.4 to 59.6% of the DM at least, which would have provided good conditions for fermentation, because ensiling is a preservation method based on the fermentation of carbohydrates by microorganisms under anaerobic conditions to produce organic acids (An 1999), and normally a minimum of 6 to 12 per cent water-soluble carbohydrates are required for proper silage fermentation(Bjorge 1996). In our study all silages of the various mixtures of SPR and SPV represented a well preserved feed resource for pigs. These results are in agreement with Tinh et al (2000 and 2001), who concluded that fermented SPV and SPR were very good and economical feed resources for fattening pigs and suitable for feeding in the uncooked form. However, trypsin inhibitors in the raw SPR decrease protein digestibility in the mixed feed, and the uncooked starch of sweet potatoes is very resistant to hydrolysis by amylase. When cooked, these trypsin inhibitors are destroyed, and the starch's susceptibility to the enzyme increases (Dominguez 1992). However, Peter et al (2001) reported that the ensiling process minimizes the trypsin inhibitors in SPR and eliminates the need to cook the feed. The most appropriate ratio between SPR and SPV in practice depends on the relative amounts of SPR and SPV produced. If dual-purpose varieties are planted in paddy rice fields as a winter-spring crop, the root production will be higher, but if the sweet potatoes are planted on upland fields which lack sufficient soil moisture for harvestable roots to form, the vine and leaf production will be more important than root production. When a variety contains a high amount of total dry matter in both roots and vines, it is recommended as a dual-purpose variety (Peter et al 2001). In the present study, all of the 5 different ratios of SPR and SPV, namely 70, 60, 50, 40 and 30 % of SPR with 30, 40, 50, 60 and 70% of SPV, respectively, on dry matter basis, were successfully ensiled without any additives. These results also show that ensiling without additives did not affect the quality of the silage, which is agreement with the results of Kayouli and Lee (1998), who found that a mixture of waste bananas, cassava roots and sweet potato tubers and leaves could be ensiled effectively without the need for additives.

During the 84 days of ensiling, there were only very slight changes in the chemical composition of SP silage, such as for DM, CP, NDF, Ca and P, which is in agreement with several previous studies. McDonald et al (1995) reported that losses of DM of less than 5% during ensiling are acceptable. The slight decrease in CP content was because there is normally some deamination of amino acids that occurs during fermentation (McDonald et al 1995). Lin et al (1988) also concluded that the general nutrient values (including metabolizable energy), fatty acid composition and amino acid contents (including the proportion of essential amino acids) in silages of mixtures of sweet potato roots and maize meal did not change during ensiling. Ruiz (1982) showed that the dry matter content of sweet potato foliage silages did not change by adding roots (up to 1.2%) or urea (up to 1.6%). Tinh et al (2000 and 2001) studied changes of DM, CP, ether extract, CF and ash of sweet potato vines silage and sweet potato roots silage with different additives, and found that there were no significant differences over time (after 14, 30, 60 and 90 days of fermentation).

In principle, a good silage should have a high lactic acid content, which is dominant to other acids such as acetic, propionic and butyric acids, and therefore is usually responsible for most of the drop in silage pH (Kung and Shaver 2001). In our study the lactic acid content in the silages of all the mixtures of sweet potato roots and vines increased rapidly in the first weeks (concentrations ranged from 64.6 to 88.5 g/kg DM) and continued increasing until 14 days of ensiling (concentrations from 118.3 to 123.9 g/kg DM), which resulted in pH values falling quickly (from around 6.3 to around 3.8 at 14 days of ensiling). These results agree with McDonald et al (1995), who concluded that good silages are characterized by having low pH values, usually between pH 3.7 and 4.2, and containing high concentrations of lactic acid (ranging from 80 to 120 g/kg DM in grass silages), which depend on water-soluble carbohydrate levels. The increases of lactic acid and the decreases of pH values of the sweet potato silages when SPR levels increased from 30% to 70% can be explained by the fact that water-soluble carbohydrates in SPR are higher than in SPV. Tinh et al (2000) reported that pH values of SPV silages (values ranged from 3.52 to 4.20) with different additives decreased rapidly within the first two weeks. That the NH₃-N content in the sweet potato silages fluctuated around 2-3% of total nitrogen was acceptable, as according McDonald et al (1995) it should be less than 100g NH₃-N/kg of total nitrogen.

Conclusions

It can be concluded from the present study that all of the 5 mixtures of different ratios of SPR and SPV, namely 70, 60, 50, 40 and 30 % of SPR with 30, 40, 50, 60 and 70% of SPV, respectively, on dry matter basis, were successfully ensiled without any additives, resulting in good quality products that could be stored at least 3 months. Which ratios of SPR and SPV in the silage can be applied under practical conditions would depend on the ratios of SPR and SPV produced at harvest.

Acknowledgements

We are very grateful to the Swedish International Development Authority (Sida/SAREC) and the Swedish University of Agricultural Sciences, Department of Animal Nutrition and Management, for their financial support of this study. Special thanks to the staff in the Department of Feed Analysis, the National Institute of Animal Husbandry, for analysis of samples and to our colleagues Dr.Viet and Mrs. Len for their help.

References

An L V 1999 Ensiling of shrimp by-product and its utilization in diets for pigs under farm conditions. M.Sc. Thesis, Swedish University of Agricultural Sciences, Department of Animal Nutrition and Management, Uppsala, Sweden.

Association of Official Agricultural Chemists (AOAC) 1990. Official Methods of Analysis (15^th edition). Washington, DC 1, 69-90.

Bjorge M 1996 Ensiling process. http://www.agric.gov.ab.ca/crops/forage/silage/silag2.html INACTIVE

Dominiguez P L 1992 Feeding of sweet potato in monogastrics. In: Roots, tubers, plantains and bananas in animal feeding (Editors: David Machin and Solveig Nyvold). FAO: Rome 1992. pp. 217-233. http://www.fao.org/ag/aga/agap/frg/AHPP95/95-217.pdf

Kayouli C and Lee S 1998 Silage from by-products for smallholders. http://www.fao.org/DOCREP/005/X8486E/x8486e0l.htm

Kung L and Shaver R 2001 Interpretation and use of silage fermentation analysis reports. http://ag.udel.edu/departments/anfs/faculty/kung/articles/interpretation_and_use_of_silage.htm

Lien L V, Sansoucy R and Thien N 1994 Preserving shrimp heads and animal blood with molasses and feeding them with a supplement for pigs. In: Proceedings of National Seminar-Workshop "Sustainable Livestock Production on Local Feed Resources) (Editors. T.R. Preston, Brian Ogle, Le Viet Ly and Luu Trong Hieu). Ho Chi Minh City. November 22-27. 1993. pp 59-62.

Lin Y H, Huang T C and Huang C 1988 Quality improvement of sweet potato (Ipomoea batatas L. Lam.) roots as feed by ensilage. The British Journal of Nutrition, volume 60, Issue 1, July 1988. pp 173-184.

McDonald P, Edwards R A, Greenhalgh J F D and Morgan C A 1995 Animal Nutrition. Fifth edition. Longman Scientific and Technical, Longman Group Limited, Longman house, Burn Mill, Harlow, Essex CM20, 2JE, England.

MINITAB 1998 GLM. In Minitab Reference Manual Release 12. 1998.

National Institute of Animal Husbandry (NIAH) 2001 Composition and Nutritive value of Animal Feeds in Vietnam. Hanoi Agricultural Publishing House, Hanoi 2001.

Peter D, Tinh N T, Thai T M, Phan, H T, Nguyen T Y and Mai T H 2001 Pig Feed Improvement through Enhanced Use of Sweet Potato Roots and Vines in Northern and Central Vietnam. In: International Potato Center (CIP), Hanoi., Vietnam

Statistical Year Book 1999 Statistical Publishing House Hanoi 2000

Ruiz M E 1982 Sweet potatoes (Ipomoea batatas (L) Lam) for Beef Production: Agronomic and conservation aspects and animal responses. In: Sweet potato. Asian Vegetable Research and Development Center 1982. pp. 439-452.

Tinh N T, Thuy T P T, Thach P N and Peter D 2000 Ensiling sweet potato vines as feed to fattening pigs. Information of Science and Technologies on Animal production, No 4/2000. (In Vietnamese language). Hanoi Agricultural Publishing House, Hanoi 2001. pp 18-30.

Tinh N T, Thuy T P T, Thach P N and Peter D 2001 Processing, storing sweet potato root by fermentation and using the fermented feed to fatten pigs. Information of Science and Technologies on Animal production, No 2/2001. (In Vietnamese language). Hanoi Agricultural Publishing House, Hanoi 2001. pp 18-30.

Winarno F G 1982 Sweet potato processing and by-product utilization in the tropics. In: Sweet potato. Asian Vegetable Research and Development Center 1982. pp. 373-384.