J. Kahurananga
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia
Introduction
Methods
Results
Discussion
Acknowledgements
References
Abstract
Spaced plants of 98 accessions of perennial Trifolium spp. mainly from the Ethiopian highlands were planted at Shola, Addis Ababa, Ethiopia in June 1985 for initial screening. The objective was to select those suitable for use on Vertisols which are subject to waterlogging in highland pastures with an extended dry season and periodic frosts. Total annual rainfall for 1985 and 1986 was 1050 mm and 1198 mm respectively. There were more frosts in 1985. T. cryptopodium followed by T. burchellianum had the highest DM production and persistence during the dry season, but the latter had poor flowering. T. semipilosum including cv Safari had lower production but very good flowering. T. repens included as a control had fair performance. None of the accessions were affected by frost. Results indicate that these clovers have potential for use in pastures particularly on fallow areas and for soil conservation. Nine accessions were selected for further evaluation in mixture with grass under cutting or grazing.
The highlands of Eastern Africa, in this context land above 1800 m, cover 70,000 km² of which 60% is in Ethiopia. They contain the highest human and livestock densities and are the most intensively cultivated (Getahun, 1978; Morgan, 1982; Jahnke and Asemenew, 1983). The major function of cattle in the Ethiopian highlands is traction. In the highlands of East Africa, cattle are primarily kept for milk. In fact the largest proportion of dairy production in Africa is in these highlands with Kenya leading (Mukasa-Mugerwa, 1980). The highlands therefore represent ideal smallholder mixed farming systems and have great potential for livestock production.
Unfortunately, the highlands are plagued by the same constraints as elsewhere in Africa of inadequate livestock feed particularly during the dry season. Natural pastures provide about 65% of the feed resources, and the rest are provided by crop residues (Jahnke and Asemenew, 1983). So one way of increasing livestock feed would be through pasture improvement. However, periodic frosts, extended dry periods, high acidity, extensive areas of Vertisols and Nitosols, each with its own peculiarities present special problems to legume introduction. On the other hand highland pastures are rich in legume composition particularly the Trifolium spp. (Gillett, 1952; Gillett et al., 1971; Mengistu, 1975; Thulin, 1983). Evaluation of annual Ethiopian clovers has shown that they have good potential for hay production (Kahurananga and Tsehay, 1984). In 1985 work was started in screening perennial clovers as research in Australia has shown that some of them have high potential for improving pasture production (Mannetje, 1964; Jones and Cook, 1981).
The screening was conducted at ILCA Headquarters, Shola, Addis Ababa, Ethiopia (9° 2'N, 38° 42'E) at an altitude of 2380 m on a gently sloping site with seasonally waterlogged black clay soil typical of the bottomlands of the Ethiopian highlands traditionally used for hay production and grazing. The soil is slightly acidic (pH 5.8) and very deficient in P (8.68 ppm).
Ten seedlings, each seven to nine weeks old of 98 accessions of Trifolium africanum (2), T. burchellianum (23), T. cryptopodium (13), T. semipilosum (57), including cv Safari for comparison together with T. repens (1) (control) originally from Ethiopia (83), Kenya (11), Tanzania (1) and southern Africa (3) were transplanted into a well cultivated field at 1 m intervals on 24 and 25 June, 1985. TSP was applied at a rate of 40 kg P ha-1 Two handfuls each of sheep manure were put in the holes and mixed with soil into which seedlings were transplanted. Observations were taken once every three weeks and included vigour, cover, stolon thickness, internode length, petiole length, middle leaflet surface area, inflorescences per plant, flowers per inflorescence and peduncle length. The plants were harvested in May 1986 and again in September 1986 and the material dried in an oven at 65°C for 24 hours for estimation of dry matter.
Daily weather records were available at the ILCA Shola weather station nearby.
The climatic data is presented in Table 1. The total annual rainfall for 1985 and 1986 was 1058 mm and 1198 mm respectively with a better spread in the later part of year (Showamare, 1986). Average maximum and minimum temperatures for both years were similar but there were more frost nights in 1985 even during the rainy season which stunted annual clovers in nearby plots.
The observations from the 980 plants are still being entered into the computer for cluster analysis in order to group plants according to their growth characteristics. Data so far recorded for the most important observations for each species are summarised in Table 2. The results show that T. cryptopodium had the highest DM production followed by T. burchellianum var. johnstonii. All varieties of T. semipilosum had lower DM with var. semipilosum, glabrescens, including cv Safari, intermedium and brunellii in that order. T. repens, T. africanum and T. burchellianum var. blongum had the lowest DM production. Again T. cryptopodium and T. burchellianum had the best vigour or green leaf retention during the dry season. T. semipilosum varieties shed most of their leaves during the dry season and those which remained turned copper red.
Table 1. Climatic records at Shola for 1985 and 1986.
|
|
Rainfall |
Evapotr. |
Air temperature °C |
No. of |
Sunshine |
|||||||||
|
(mm) |
(mm) |
Av. |
Max. |
Av. |
Min grass lev. |
Frost night |
Duration |
|||||||
|
1985 |
1986 |
1985 |
1986 |
1985 |
1986 |
1985 |
1986 |
1985 |
1986 |
1985 |
1986 |
1985 |
1986 |
|
|
Jan. |
29 |
0 |
198 |
176 |
23.0 |
23.0 |
8.4 |
6.5 |
-5.5 |
-3.0 |
16 |
21 |
9.7 |
10.4 |
|
Feb. |
0 |
44 |
185 |
139 |
23.6 |
23.3 |
8.3 |
10.6 |
-2.5 |
1.0 |
5 |
0 |
9.5 |
6.4 |
|
Mar. |
24 |
71 |
187 |
188 |
24.6 |
23.5 |
10.0 |
10.6 |
-4.0 |
1.5 |
5 |
0 |
8.0 |
7.8 |
|
Apr. |
110 |
217 |
160 |
129 |
22.7 |
22.3 |
11.0 |
11.8 |
-2.0 |
5.5 |
1 |
0 |
4.9 |
5.9 |
|
May |
119 |
51 |
170 |
154 |
23.0 |
23.7 |
11.2 |
11.6 |
-0.5 |
1.5 |
2 |
0 |
7.3 |
8.2 |
|
June |
94 |
226 |
149 |
100 |
22.6 |
20.9 |
10.5 |
10.8 |
-0.5 |
1.0 |
2 |
0 |
8.9 |
4.5 |
|
July |
228 |
168 |
89 |
99 |
19.3 |
20.4 |
10.4 |
10.9 |
0.5 |
6.5 |
0 |
0 |
3.7 |
4.4 |
|
Aug. |
280 |
260 |
82 |
111 |
19.7 |
20.7 |
10.7 |
10.4 |
0.5 |
2.5 |
0 |
0 |
4.1 |
4.5 |
|
Sept |
138 |
124 |
133 |
127 |
20.6 |
21.1 |
10.5 |
9.9 |
3.5 |
-2.5 |
0 |
1 |
6.5 |
6.5 |
|
Oct. |
37 |
37 |
175 |
177 |
21.4 |
22.0 |
9.4 |
8.1 |
-0.5 |
-2.0 |
1 |
5 |
8.9 |
8.9 |
|
Nov. |
0 |
0 |
176 |
185 |
22.1 |
22.8 |
6.9 |
7.4 |
-3.5 |
-3.5 |
15 |
10 |
10.7 |
10.6 |
|
Dec. |
0 |
0 |
175 |
1.7 |
22.0 |
22.9 |
7.1 |
6.5 |
-3.5 |
-4.0 |
10 |
10 |
9.8 |
9.7 |
|
Total |
1059 |
1198 |
|
|
|
|
|
|
|
|
57 |
47 |
|
|
Source: Showamare (1986).
Table 2. Average and range of the important observations of perennial Trifolium at Shola in 1985 and 1986.
|
Species and variety |
Number of accessions |
Dry season vigour (1-10) |
DM weight per plant (gm) |
Days to first flowering |
% plants in flower |
|
T. africanum |
3 |
4 (3-4) |
69 (20-176) |
192 (176-207) |
60 (50-70) |
|
T. burchellianum var. johnstonii |
23 |
6 (4-9) |
192 (11-342) |
185 (182-252) |
23 (0-50) |
|
T. burchellianum var. oblongum |
1 |
5 (5) |
49 (31-71) |
139 (139) |
10 (10) |
|
T. cryptoodium |
13 |
6 (3-8) |
224 (42-569) |
212 (181-233) |
49 (30-80) |
|
T. repens |
1 |
5 (5) |
61 (21-160) |
183 (183) |
30 (30) |
|
T. semipilosum var. brunelli |
1 |
1 (1) |
68 (4-14) |
221 (221) |
50 (50) |
|
T. semipilosum var. glabrescens |
12 |
3 (1-7) |
111 (11-240) |
156 (119-184) |
98 (90-100) |
|
T. semipilosum var. intermedium |
2 |
3 (2-3) |
87 (55-120) |
223 (217-228) |
50 (50) |
|
T. semipilosum var. semipilosum |
42 |
4 (1-7) |
147 (41-222) |
205 (166-231) |
88 (40-100) |
|
Total |
98 |
|
|
|
|
The positions were reversed regarding flowering. T. semipilosum had profuse flowering with varieties semipilosum and glabrescens excelling. T. africanum had good flowering and T. cryptopodium was average. The one accession of T. repens had rather fair flowering. Both varieties of T. burchellianum had the worst flowering with some accessions not flowering at all. None of the accessions were affected by frost.
Besides the above characteristics, all the clovers showed good soil binding characteristics, demonstrated by the fact that most of the soil cracking during the dry season occurred mostly on spots where there was no clover growth. T. cryptopodium with its numerous matt-forming rhizomatous stems was most outstanding in this respect.
This is the first time that such a large collection of perennial Trifolium species from Ethiopia have been screened. It was rather surprising that the smallish often overlooked T. cryptopodium being evaluated for the first time surpassed the others in DM. This is partly due to its dense matt of stems with short internodes. It grows well in natural pastures with grasses such as Andropogon abyssinica, A. amethystinus, A. distachyos, Pennisetum thunbergii and P. clandestinum or Kikuyu grass. At one time it was briefly referred to as forming a valuable close sward under heavy grazing in association with Digitaria scalarum on Mt. Elgon (Gillett, 1952). The prostrate growth habit and its small leaves and short internodes makes it ideal for sheep grazing at high altitudes from about 2600 m - 3000 m. The matt-forming growth habit which binds soil together also makes it ideal for soil conservation. Its only problem is that its pods shatter easily after ripening. There is therefore need to study the seed ripening characteristics in order to devise efficient means of harvesting.
The other species which showed good productivity and which persisted well during the dry season was T. burchellianum var. johnstonii. Its ability during the dry season was also noted in early observations in Kenya and Australia (Bogdan, 1956; Britten, 1962: Mannetje, 1964). This characteristic is rather surprising as this species is normally distributed in moist grasslands and open glades in forests where rainfall is above 1000 mm (Gillett et al., 1971; Thulin, 1983). This species is found growing with T. cryptopodium. Its major limitation is the poor flowering and hence seed production. This was also observed in one of the early glasshouse experiments where it failed to flower altogether (Mannetje and Pritchard, 1968). However, in this trial there were two accessions, one from Ethiopia and the other from Tanzania which had good flowering. There are therefore possibilities for selection. This species has good potential for use in pasture for both cattle and sheep. It is also suitable for soil conservation.
Although the average productivity of T. semipilosum was lower than the two preceeding species, there were some accessions which did well. It is rather surprising that var. glabrescens including cv Safari faired poorly. The latter is doing well in subtropical areas of Australia (Jones and Cook, 1981). Although most of the accessions lost their leaves and those leaves which remained turned copper red during the dry season, they responded very fast to the onset of rain. This species had excellent flowering, and the pods do not shatter so easily making it more suitable for seed production. It also grows well naturally with grasses in pasture, and it is found in a very wide range of altitudes, edaphic conditions for both cattle and sheep and for soil conservation. It seems to withstand very heavy grazing in communal natural pasture. The ability to persist under grazing has also been noted under Australian conditions (Sproule et al., 1983).
This screening trial has shown that the perennial Trifolium species have good potential for use in pasture. They could be used in establishing new pastures or on fallow areas including marginal areas. They could also be useful in reseeding degraded areas. It is important to continue making selections as evaluations continue. In this screening, four accessions of T. cryptopodium (ILCA Nos. 7621, 9391, 9398, 9461), three of T. burchellianum var. johnstonii (ILCA Nos. 10179, 10207, 12481) and of T. semipilosum var. semipilosum (ILCA No. 7642, 8025) were selected for the next stage of evaluation in mixture with grasses under cutting management and grazing.
I am grateful to ILCA field staff who have assisted me in the screening and to Dr. J.C. Tothill for giving useful guidelines. Also I thank ILCA's Director of Research, Dr. K. Peters and the Director General, Dr. J. Walsh, for allowing me to attend the workshop and present this paper.
Bogdan, A.V. 1956. Indigenous clovers of Kenya. East African Agricultural Journal 22:40-45.
Britten, E.J. 1962. Evaluation for pasture purposes of some African clovers in a plant introduction program. Journal of Range Management 15:329-333.
Getahun, A. 1978. Zonation of the highlands of tropical Africa ILCA, Addis Ababa, Ethiopia.
Gillett, J.B. 1952. The genus Trifolium in southern Arabia and in Africa south of the Sahara. Kew Bulletin 7:367-404.
Gillett, J.B., Polhill, R.M. and Verdcourt, B. 1971. Leguminosae (Part 4) Subfamily Papilionoidea (2). In: E. Milne-Redhead and R.M. Polhill (eds), Flora of tropical East Africa. Crown Agents, London. pp. 1016-1036.
Jahnke, H.E. and Asemenew, G. 1983. Ethiopian highlands reclamation study. An assessment of the recent past and present livestock situation in the Ethiopian highlands. Land Use Planning and Regulatory Department, Ministry of Agriculture and Food and Agriculture Organization. Addis Ababa, UTF/ETH/037/ETH Working Paper 8.
Jones, R.M. and Cook, B.G. 1981. Agronomy of Kenya white clover-cultivar. Safari Information Services sheet No. 41-3, CSIRO Division of Tropical Crops and Pastures.
Kahurananga, J. and Tsehay, A. 1984. Preliminary assessment of some annual Ethiopian Trifolium spp. for hay production. Tropical Grasslands 18:215-217.
Mannetje, L.T. 1964. The use of some African clovers as pasture legumes in Queensland. Australian Journal of Experimental Agriculture and Animal Husbandry 4:22-25.
Mannetje, L.T. and Pritchard. 1968. The effects of photoperiod and night temperature on flowering and growth in some African Trifolium species. New Phytologist 67:257-263.
Mengistu, A. 1975. Grassland condition in the Chilalo Awraja, Arussi Province, Ethiopia. M.Sc. thesis, Department of Plant Husbandry, Agricultural College of Sweden, Uppsala.
Morgan, W.T. 1982. East Africa. Longmans, London. pp. 290-331.
Mukasa-Mugerwa, E. 1980. Small holder dairy development in the highlands of tropical Africa. ILCA, Addis Ababa, Ethiopia.
Sproule, R.J., Shelton, H.M. and Jones, R.M. 1983. Effect of summer grazing on growth habit of Kenya white clover (Trifolium semipilosum cv Safari in a mixed sward). Tropical Grasslands 17:25-30.
Showamare, M. 1986. Climatic records for ILCA research sites 1985. ILCA, Addis Ababa, Ethiopia.
Thulin, M. 1983. Leguminosae in Ethiopia. Opera Botanica 68:92-223.
