Eighteen lactating goats were allocated to three treatments to evaluate diets based on protein-rich leaves, cassava roots, rice bran and a molasses-urea block as alternatives to the conventional grass and cereal-rich concentrates for lactating goats. The protein-rich leaves were provided by mixtures of foliage from: Jackfruit (Artocarpus heterophyllus) and Trichanthera gigantea (TG-JF); and Flemingia macrophylla and Trichanthera gigantea (TG-FM). The control diet was Guinea grass (Panicum maximum) ad libitum and 600 g/day of concentrates. The experimental diets were introduced in week 5 of lactation, following a period on a standard diet (Guinea grass and cereal-rich concentrates) from parturition to end week 4. The mixed leaves were supplemented with chopped sugar cane stalk, dried cassava root (400 g/day), rice bran (400 g/day) and molasses-urea (10% urea) block. The does sucked their kids 24 hours/day for the first 3 weeks. In the 4th week once daily hand milking was introduced, with suckling after milking in the morning and suckling in the afternoon without milking. Milk intake by the kids was estimated by weighing before and after suckling. Yield was measured as milk taken at once daily milking and as the total milk (milking plus that consumed by the kids). Milk yield during the 4th week on the standard diet was used as covariate to correct milk yields recorded from weeks 6 though 12.
The goats fed with protein rich leaves had a higher average milk yield during the 7 weeks of experiment (750 and 754 g/day for total milk on TG-JF and TG-FM, respectively) than those on the control treatment of concentrates and Guinea grass (641 g/day) (SE of means ±15) Corresponding values for milk at "milking" were 464, 455 and 341 g/day (SE of means ±7.8). There were no differences in growth rates of the kids, or in weight changes in the does, due to the dietary treatments imposed on the does.
The traditional feed resources for goats in intensive systems in experimental stations in Vietnam have been grasses and concentrates, the latter derived from cereal grains, protein meals and by-products. These feeding systems have had little impact with farmers because the green grasses are difficult to grow in winter and cereal grains and their by-products are steadily increasing in cost and are in competition with the nutritional needs of people.
It has been shown that sugar cane can replace guinea grass as the basal forage for growing goats in the winter season with no effect on growth rate (Le Diep Long Bien 1998). Molasses and cassava roots are highly digestible sources of carbohydrate that are readily available but have a negligible protein content. Urea can supply the nitrogen needed for microbial fermentation in the rumen (Preston and Leng 1987) but sources of "bypass" or "escape" protein, needed to complement microbial protein especially for lactating ruminants, are more difficult to find. Groundnuts and soya beans can be grown in North Vietnam but yields are moderate and the protein-rich seeds are used preferentially in human nutrition. Protein-rich leaves from multi-purpose trees and from water plants, plant species which are high yielding in the tropics, have been proposed as more appropriate protein sources for livestock in the tropics (Preston 1998). In North Vietnam it has been observed that leaves derived from foliage of trees such as Jackfruit (Artocarpus heterophyllus), Trichanthera gigantea, and Flemingia macrophylla are readily consumed by goats and are available all the year round, and especially in the dry season (Dinh Van Binh, personal communication). In vitro studies (Keir et al 1997a,b; Lai 1998) indicate that leaves of Jackfruit and Trichanthera gigantea are potentially of high nutritive value.
This study aimed to evaluate the use for lactating goats of combinations of leaves from three trees / shrubs (Jackfruit [Artocarpus heterophyllus], Trichanthera gigantea, and Flemingia macrophylla) as the major source of protein in diets based on carbohydrate-rich sugar cane stalk, molasses block (with urea) and cassava roots.
This study was located at the Goat and Rabbit Research Centre, Bavi district, HaTay province which is located 50 km to the West of Hanoi. The land area of the Centre is 60 ha, two thirds of which are hilly and cultivated land. The average temperature is 20-25 oC and annual rainfall is 1,800 mm.
Eighteen lactating goats (Local Bach Thao breed and Barberi breed introduced from India) were used to evaluate three feeding systems. These were:
There were 6 goats housed in individual cages on each treatment.
On treatments TG-JF and TG-FM the foliages were offered at the rate of 1,000 g/goat/day. In addition, these goats were offered chopped sugar cane stalk (2,200 g/day), dried cassava root (400 g/day), rice bran (400 g/day) and a multi-nutritional block (molasses and minerals with 10% urea) (100 g/day). The control goats were fed the Guinea grass ad libitum and the concentrate at 600 g/day. It was estimated that all diets supplied the total crude protein requirements of lactating goats of 140-150 g/day (Devendra and Burns 1983).
From kidding to the end of week 4 all does in the three treatments were fed with the control diet of 600 g/doe/day of concentrate (divided in two meals/day) and ad libitum guinea grass (given 5 times per day). The experimental feeds were introduced from the beginning of the 5th week and continued through to the end of week 12. Week 5 was to adapt to the new feeding system in the case of treatments TG-JF and FM-TG.
The feeding schedule was as follows:
7.00am: Sugar cane( 500 g/doe)
8.30am: Cassava root (200 g/doe) + MUB (50 g/doe) + 50:50 mixture foliage(600 g/doe)
9.30am: Sugar cane (500 g/doe)
11.00am: Rice bran ( 200 g/doe) + mixed foliage (600 g/doe)
2.0-3.0 pm: Sugar cane (500 g/doe)+ dried cassava root (200 g/doe)
4.0-5.0 pm: Sugar cane(700 g/doe) + dried cassava root (200 g/doe) + rice bran (200 g/doe)
5.30pm: Mixed foliage (800 g/doe)
All foliages after weighing were mixed together and hung on a wire on the top of the feed trough. Under each feed trough a plastic net was placed to collect rejected feed. Sugar cane stalk was chopped into pieces of 2.0-2.5 cm before feeding. Water was always in the pen. Feed offered and refused was recorded daily.
From the first to third week the kids were with their dams all the time. Milking was not practised during this period. From week 4 to week 12, the does were milked in the morning (7.00am) after overnight (from 5.30pm onwards) separation of the kids. The kids were then suckled for a period of 30 minutes at the beginning and end of which they were weighed to estimate milk consumed. The kids were then separated until 5.00 pm when they were again suckled for 30 minutes, but without prior milking. Weighing before and after suckling was again practised to estimate the milk consumed. The kids were then separated until milking the next morning. Milk yield in week 4 on the standard (control) diet was used as covariate to correct milk yields during experimental weeks 6 through 12. Each day the does were allowed out of the cage for one hour for exercise.
The feed intake was recorded daily from the amounts offered and refused. The liveweight of the does was recorded at the beginning of the 6th week and the end of the 12th week of lactation to determine body weight changes during the experimental period. Growth rate of the kids was calculated by linear regression using as dependent variable the weights prior to morning suckling every 7th day. Daily samples of feeds were bulked at monthly intervals and analysed for dry matter and nitrogen by AOAC (1988) methods. The General Linear Model in the Minitabl 10.2 software was used to determine the significance of treatment differences for:
The data set for milk yield included mean yields per treatment for separate weeks (weeks 6 through12).
The mean values for intakes of dietary ingredients over weeks 6 to 12 of the experiment are shown in Table 1.
| Table 1: Dry matter intake and protein intake from week 6 to week 12 between treatment of lactating goats | ||||
| TG+JF | TG+FM | GG+Con. | SE/Prob | |
| Dry matter intake, g/day | ||||
| Week 6 | 1269 | 1215 | 1008 | |
| 7 | 1382 | 1217 | 1184 | |
| 8 | 1392 | 1226 | 904 | |
| 9 | 1375 | 1228 | 888 | |
| 10 | 1389 | 1297 | 845 | |
| 11 | 1360 | 1258 | 858 | |
| 12 | 1363 | 1270 | 808 | |
| Means | 1361 | 1245 | 928 | ±21/0.001 |
| Intake, N*6.25 | ||||
| Week 6 | 146 | 134 | 132 | |
| 7 | 136 | 117 | 128 | |
| 8 | 136 | 120 | 106 | |
| 9 | 133 | 121 | 105 | |
| 10 | 135 | 123 | 101 | |
| 11 | 133 | 121 | 106 | |
| 12 | 132 | 119 | 97 | |
| Means | 136 | 122 | 111 | ±1.5/0.001 |
Both treatments with the mixed tree / shrub foliages supported significantly higher intakes of dry matter and of crude protein than the control treatment of concentrates and Guinea grass. This could be interpreted as indicating a better balance of nutrients in the mixed foliage diets as voluntary feed intake is considered to reflect such a balance of nutrients (Preston and Leng 1987). The mean intakes of protein were: 100, 98 and 120 g/kg dry matter consumed and were highest for the control diet which had the lowest dry matter intake. It could be that in the diets of mixed tree foliages the proportion of the protein which escaped the rumen fermentation was higher than on the diet of Guinea grass and concentrates. It is known that the presence of tannins in tree leaves "protects' the protein thus facilitating escape from the rumen fermentation. In contrast, the protein in grasses is soluble, tannins are not present, and the protein is almost completely fermented in the rumen (Preston and Leng 1987).
Mean values for milk yields from weeks 6 to 12 and the overall means for the total period are set out in Table 2 and in Figures 1 and 2. Milk yields, assessed by both methods, were significantly higher for the dietary treatments of mixed tree / shrub foliages than for the control diet of Guinea grass and concentrate (P=0.001).
Table 2: Mean values for milk yield for the feeding systems with mixed tree / shrub foliages (TG+JF is Trichanthera gigantea and Jackfruit; TG+FM is Trichanthera gigantea and Flemingia macrophylla) |
||||
| TG+JF | TG+FC | GG+Con | SE/Prob | |
| Yield at milking, g/day | ||||
| Week 6 | 494 | 540 | 442 | |
| 7 | 494 | 488 | 450 | |
| 8 | 507 | 526 | 372 | |
| 9 | 457 | 429 | 353 | |
| 10 | 456 | 400 | 315 | |
| 11 | 445 | 413 | 243 | |
| 12 | 397 | 391 | 210 | |
| Mean | 464 | 455 | 341 | ±7.8/0.001 |
| Total milk (includes suckling), g/day | ||||
| 6 | 806.1 | 911.5 | 875.4 | |
| 7 | 800.7 | 802 | 786.7 | |
| 8 | 820.7 | 807.2 | 695.7 | |
| 9 | 782.9 | 722.6 | 656.3 | |
| 10 | 742.5 | 674.4 | 569 | |
| 11 | 683.2 | 689.4 | 484.4 | |
| 12 | 610.7 | 672.6 | 418.1 | |
| Mean | 749.5 | 754.2 | 640.8 | ±15/0.001 |
Variation in yield was less when yield "at milking" was measured than when "total milk" was computed taking account of the estimates of the amounts of milk sucked by the kids. Coefficients of variation (measured as 100*SE/[overall means for treatments]) were 1.86 and 2.19, respectively. Estimating suckled milk, by weighing kids before and after suckling, in order to arrive at a figure for total milk yield in a milking/suckling system of management, is time consuming and open to considerable error. This is particularly so with goats, as suckling is rarely a continuous activity as is the case with cattle.
 Figure 1: Yield at milking for goats fed mixed protein-rich leaves, MUB and cassava root versus guinea grass and concentrates (SE means ±7.8; P=0.001) |
 Figure 2: Total milk, including suckling, of goats fed mixed protein-rich leaves, MUB and cassava root versus guinea grass and concentrates (SE means ±15; P=0.001) |
The mean values of liveweight of the does, at the end of the 12th week of lactation, for the three dietary treatments, are shown in Figure 3. The slight differences in favour of the mixed tree foliage treatments were not significant.
 Figure 3: Final liveweights of does fed protein rich-leaves, sugar cane, cassava root and MUB or Guinea grass and concentrates |
 Figure 4: Liveweight gain of kids from does fed protein rich- leaves, sugar cane, cassava root and MUB compared with Guinea grass and concentrates |
Data for growth rates of the kids over the experimental period (weeks 6 to 12) are shown in Figure 4. There were no differences due to dietary treatment. The absence of differences indicates that the amounts of milk taken by sucking kids is independent of the milk yield of the dam.. Similar results were reported for a dual purpose cattle production system of once daily milking and restricted calf suckling (Fernandez et al 1978). This strengthens the argument for using the amount of milk obtained at "milking" as the criterion for measuring treatment effects rather than "total" milk.
The observed growth rates of the kids are in line with results reported by Nguyen Thi Duyen et al (1997) for goats managed in a similar milking and restricted suckling system. and are only slightly less than when milking is not practised and the kids have full access to the doe (Preston 1997).
The results of this experiment confirm the feasibility of using local resources as the basis of a feeding system for dual purpose production of milk and meat from goats. Cassava, sugar cane and Jack fruit trees are traditionally grown by farmers in hilly areas of North and Central Vietnam. The tree Trichanthera gigantea, introduced from Colombia, grows well as does the shrub Flemingia macrophylla, introduced from the Philippines. Both these plants have been adopted enthusiastically by farmers in the area. Molasses and rice bran are common by-products resulting from local processing of sugar cane and rice.
Fernandez A, MacLeod N A and Preston T R 1977 Relationships between milk production and calf growth in dual purpose herd. Tropical Animal Production. Volume 2, Number 1:49-55
Keir Brenda, Nguyen Van Lai, Preston T R and Orskov E R 1997a Nutritive value of leaves from tropical trees and shrubs: 1. In vitro gas production and in sacco rumen degradability. Livestock Research for Rural Development (9) 4:24-30
Keir Brenda, Dinh v. Binh, Preston T R and Orskov E R 1997b Nutritive value of leaves from tropical trees and shrubs: 2. Intake, growth and digestibility studies with goats. Livestock Research for Rural Development (9) 4:31-34
Le Diep Long Bien 1998 Sugar cane as replacement for Guinea grass for growing goats. Thesis Paper 1
Nguyen Van Lai 1998 Evaluating the nutritive value of tree leaves as water-soluble dry matter and in sacco degradation rate. Livestock Research for Rural Development. Volume 10: in preparation
Nguyen Thi Duyen, Le Diep LongBien, Nguyen Thi Mui, Dinh v. Binh and Preston T R 1996 Foliage of Trichantera gigantea, Jack fruit (Artocarpus heterophyllus), banana (Musa sp) and Acacia mangium as protein sources for lactating goats fed a basal diet of rice straw and sugar cane tops. Livestock Research for Rural Development (8) 3:46-50
Preston T R and Leng R A 1987 Matching Ruminant Production Systems with Available Resources in the Tropics and Subtropics. PENAMBUL Books Ltd: Armidale NSW, Australia
Preston T R 1997 Improving the well being of poor farmers by developing milk production from local resources. Report to FAO (AGA Division). TCP/VIE/6613
Preston T R 1998 Animal production and use of natural resources. Proceedings 8th World Conference on Animal Production. Seoul, South Korea (June 1998)