Previous Page Table of Contents Next Page

The uptake of white clover technology in farming practise

B E Frankow-Lindberg1, D-A Danielsson2, C Moore3

1 Swedish University of Agricultural Sciences, Dept. Crop Production Science, P.O. Box 7043, S-750 07 Uppsala, Sweden

2 Swedish University of Agricultural Sciences, Western District of Animal Research, P.O. Box 234, S-532 23 Skara, Sweden

3 Greenmount College of Agriculture and Horticulture, Antrim, BT41 4PU, Northern Ireland

Farmers' opinions on white clover
Materials and methods
Output/performance on beef farms
Output/performance on dairy farms


Results from two studies (in Sweden and Northern Ireland) with the aim of stimulating a greater reliance on white clover pastures on commercial farms are reported. The Swedish study was conducted during the years 1992-94, and comprised 15 farms (dairy or beef/sheep) in 1992 and nine farms thereafter. Animal performance, pasture output and clover content were monitored during the grazing season. Marginal costs for pasture production were calculated. The study in Northern Ireland begun in 1993 and comprised six beef/sheep farms. Physical and financial performance were recorded monthly by the farmer using the Agricultural Business Services (ABS) and clover content and sward height were assessed during the grazing season.

Daily liveweight gain of beef cattle varied between 589 and 1205 g head-1 (farm averages) in the Swedish study, while utilized metabolizable energy (UME) varied between 27 000 and 48 000 MJ ha-1. Clover contents in the pastures were generally high. They were c. 20% in spring and increased to a peak of 40-50% in July/August. Negligible amounts of fertilizer N were applied on one farm. There were no incidences of bloat from a total number of 243 animals. On two of the dairy farms that to a large extent depended on grass/clover pastures fertilizer N input was 0 to 70 kg ha-1. Cows were fed both forage and concentrate during the grazing season. Milk yield was on average 14 160 and 14 500 kg energy corrected milk per ha with an input of 847 and 1039 kg of concentrate cow-1, respectively. Corresponding stocking rates per ha were 6.3 and 6.6 in spring and 3.3 and 3.6 in late summer. UME varied between 39 000 and 57 000 MJ ha-1. Clover content was c. 20% in spring and increased to a peak of 20-60% in July/August. There were no cases of bloat. These results are comparable to previous studies on commercial farms with grass + N pastures, while marginal costs for pasture production from grass/clover swards were c. 50% of those for grass + N pastures.

Clover contents in the Northern Ireland study were generally within recommended levels for the time of the year. On farms where sward height was reduced more than 0.5 cm between 1993 and 1994 clover content showed the greatest increase in the swards. Cattle and sheep performed well on all farms, and no cases of bloat were encountered. The mean gross margin per ewe and per hectare on the clover development farms were 12.5% and 29% higher, respectively, than the average of ABS recorded farms.


The research community has produced a plethora of scientific papers during the past decades devoted to white clover (Trifolium repens L.). Emphasis has been placed on the superiour nutritional value and the high intake characteristics of white clover (e.g. Castle et al,. 1983; Thomson, 1994), or the reduced need of fertilizer N to sustain DM yields (see review by Frame and Newbould, 1986). In combination, it has been postulated that these two factors lead to a better economic result on individual farms (Doyle and Morrison, 1983), although the price of fertilizer N is likely to have a major impact on the difference in profitability between the two systems (Ryan, 1988). Large-scale studies suggest that animal output from grass/clover pastures is 80-85% of output from optimally N-fertilized grass pastures (Ryan, 1989; Orr et al. 1990; Bax and Thomas, 1992; Younie, 1992), or just as good as grass + moderate amounts of N fertilizer (Pflimlin et al., 1989; Giovanni, 1990; Swift and Vipond, 1992). Younie (1992) also found that the variability of the output of utilized metabolizable energy (UME) in beef production did, if anything, vary somewhat less compared to a grass + N system. It has been suggested that the clover content should be c. 30% in order to get an acceptable animal output (Pflimlin and Journet, 1983), although Laidlaw and S teen (1989) and Chestnutt (1992) have suggested that much lower levels have been efficiently utilized by stock. Orr et al. (1990) thought it possible that far less clover may be required in grazed, compared to cut swards, due to the N cycling in the former.

However, uptake of white clover technology on commercial farms in Europe has hitherto been slow. The future need for the extensification of livestock systems within the EU is now thought to be an impetus for a future change towards a greater reliance on clover-based systems.

This paper presents some results from recently finalised or on-going studies where animal performance and output from commercial farms relying on grass/clover pastures in Sweden and Northern Ireland have been monitored.

Farmers' opinions on white clover

There is very little published information about what farmers think of white clover pastures. Successful growers in France state that their main interest in white clover stems from the reduced need for purchases of fertilizer N and concentrates, and hence a reduced economic vulnerability (Allezard et al., 1983). Swedish farmers value the reduced need for fertilizer N, but also mention the good intake characteristics of white clover (Arnesson, 1991), as well as the ability of white clover to fill gaps in the sward, as positive traits of the species. Both French and Swedish farmers appreciate the good summer growth of white clover.

Perceived problems among successful growers are the slow establishment (Allezard et al., 1983), slow spring growth and weed problems (Allezard et al., 1983; Arnesson, 1991). Many farmers quote lack of advisory expertise on questions related to the establishment and management of white clover pastures (Arnesson, 1991), and would like to know more about suitable cultivars, seed mixtures, stocking rates, how to manipulate botanical composition and how to avoid bloat.

Materials and methods


A study was initiated in 1992 with the aims 1) to monitor animal performance and pasture output during the grazing season and 2) to stimulate a greater reliance on biological nitrogen fixation in grazed swards. A selection of a number of farms in the south-west of Sweden (around lat. 58°N) were made according to their expressed interest in white clover. Their production were either dairy or beef (18-month bulls with one grazing season). Initial clover content in the pastures varied a lot. Bulls were weighed prior to turn-out and at turn-in. Body weights of dairy cows were estimated from heart girth measurements. Dairy cows in Sweden are generally fed both concentrate and forage during the grazing season. Farmers recorded the amounts fed, ME concentration in home grown feed stuffs were analysed and the ME intake from these sources and bought feed stuffs were calculated. Milk production figures were estimated from recordings from the dairy company. Pasture intake was estimated from animal ME need for maintenance + pregnancy + production level achieved +/- body weight changes - intake from other feed sources. Farmers kept daily records of which paddock was used, animal numbers and management inputs (fertilization, topping, irrigation etc.). During the first year the farms were merely monitored, and data on animal performance and clover content were collected. For the following two years the number of farms was reduced. The farmers were visited monthly from spring to autumn and the state of the swards, as well as clover content was determined. In spring, every farmer received individual guidelines on suitable management of his pastures and on supplemental feeding for the dairy cows. The study unfortunately suffered from very dry summer weather during 1992 and 1994, while 1993 showed normal precipitation. In 1994 most grasslands withered completely in July and August, and dairy cows were fed very high levels of supplemental feeding to sustain milk production.

Northern Ireland

Clover development work was initiated on beef/sheep farms during 1993. The project aims to develop low input systems based on grass/white clover swards. The targets are to 1) establish six development farms with at least 50% of the project area devoted to grass/clover swards within three years and 2) achieve a gross margin within the top 33% of Agricultural Business Services (ABS) recorded beef/sheep farms by saving money on fertilizer N and improving returns from improvements in cattle and sheep performance. The Agricultural Development Service (ADS) identified potential farms which were visited by technologists. From these, six were selected giving a good geographical spread across the country with one farm in each county. On each farm a major portion of the land was incorporated into the project area. Soil samples were taken to determine soil P and K status. An annual management plan for each farm was drawn up by the technologist in conjunction with the farmer. In the majority of cases predominantly grazed clover-based swards are integrated with other areas of grass, receiving higher levels of N fertilizer, which are utilized for silage production. Physical and financial performance is recorded monthly by the farmer using the ABS system. Greenmount College staff assess clover content of swards and sward height in July and September each year. Clover percentage in the DM is assessed by correlation of subjective visual determinations taken on each farm with a pre-determined relationship of visual determination with DM content determined by separation and drying of clips. The farmers in the group along with their development adviser attend a farm walk arranged by the technologists at one of the development farms each year.

Output/performance on beef farms


Results from three farms are presented (Table 1). Farm A and Farm C had newly established very clover-rich pastures. These farms participated in a governmental supported scheme aimed at extensive production. Thus no fertilizer dressings were allowed, and it was not allowed to conserve any grass surplus as all the grass was to be grazed in situ. Surpluses in spring were either grazed by other animal species (farm A) or topped during good weather. The bulls appreciated the dried grass, even if the quality was bound to be very low, as it was a good complement to the high levels of clover leaves in the sward. There were no indications that the clover was hampered by this practice.

Table 1. Animal performance (beef cattle), utilized metabolizable output (UME) output, stocking rate and sward characteristics during the grazing season. (Mean of three years).

Farm A

Farm B

Farm C

Daily LWG at grass (g head-1)




Total LWG at grass (kg ha-1)




UME output (MJ ha-1)




Overall stocking rate (no. ha-1)




Length of grazing season (d)




Clover content (% of DM in July)




Sward height at the end of the season (cm)




*grazed by sheep until the beginning of winter

Average dates for turn-out were in the middle of May and turn-in the first week of October. The young bulls were c. 6 months old and weighed between 141 and 269 kg at turn-out, and their daily LWG at grass varied between 589 and 1205 g head-1 (farm averages). All farms increased their stocking rates during the study, while output per ha peaked in 1993 when pasture production was highest. Weed content was low in all cases. Clover content varied with season. It was generally c. 20% of DM in spring, increasing to a peak in July/August and thereafter remaining high or falling slightly. An exception was farm C, which had very drought-prone soils, where white clover appeared to be wiped out by the extremely dry weather in 1994. There were no incidences of bloat on a total number of 243 animals.

Northern Ireland

Of the land committed to the project the proportion under grass/clover swards averaged 51% and ranged from 25 to 100% on the six farms, representing 17 ha per farm. Clover contents (Table 2) were generally within recommended levels for the time of the year at which they were recorded. On farms where sward height was reduced more than 0.5 cm between 1993 and 1994 clover content showed the greatest increase in the swards.

Table 2. Clover content and sward height during the grazing season. (Mean of two years).


Clover % in DM

Sward height (cm)








































Farmers reported that cattle and sheep performed well on all farms in both years. No cases of bloat in cattle were encountered. Annual reports are not yet available for all enterprises on all farms due mainly to influence of calving pattern on the start of the recording period. Gross margin analysis for the past year reveal gross margins per ewe ranging from £41 to £71 and gross margins per hectare from £501 to £824. The mean gross margin per ewe and per hectare on the clover development farms were 12.5% and 29% higher, respectively, than the average of ABS recorded farms, the majority of which are based on swards with little clover receiving moderate levels of N fertilizer.

Output/performance on dairy farms


Results from only two farms are presented here, as most of the farms had a mixture of white clover based and grass + N pastures. Farm D relied entirely on white clover pastures that were both conserved and grazed, and there was no fertilizer N given to the grazed swards during two of the three years studied. Farm E, on the other hand, had some grass + N pastures. Average N fertilization input varied between 7 and 70 kg ha-1. Both farms had a majority of autumn calving cows. Farm D had a herd of c. 60 cows of the Swedish Red landrace and farm E had a herd of c. 30 cows of Friesian origin. The average milk yield of their herds were c. 9 000 kg ECM cow-1 year-1 on both farms. The pastures were newly established when the study began. Both farms practised rotational grazing (Table 3).

Table 3. Animal performance (dairy cows), UME output, concentrate fed and clover content during the grazing season. Mean of three years.

Farm D

Farm E

Milk yield (kg ECM cow-1)



Milk yield (kg ECMha-1)

14 160

14 500

UME output (MJ ha-1)



Concentrate fed on grass (kg cow-1)



Stocking rate (cows ha),




late summer



Clover content (% of DM in July)



Length of grazing season (d)



Sward height at the end of the season (cm)



*grazed by horses during the autumn

Gross production was estimated on farm E during 1993 by harvesting to c. 5 cm stubble height in one paddock immediately prior to the entrance of the animals. It amounted to 73 900 MJ ha-1 in 1993 of which 76% was eaten by the cows. Animal intake was generally highest in May and June and was typically 8-10 kg DM cow-1 d-1. Milk yield varied between 17 and 27 kg ECM cow-1 d-1. On both farms hay was daily fed, and also some silage during dry spells (mainly on farm D). Sward utilization could have been higher in 1993 as judged by the frequency of rejected areas, but not generally so during the dry years. Clover content was c. 20% in May, peaked in July/August and thereafter fell to c. 30%. After the dry summer of 1994, clover content was reduced to c. 6% on farm E (drought-prone soils) and 15% on farm D. There were no cases of bloat.



Based on cut plot experiments, the achievable UME yield level of grass/clover pastures in the Swedish environment is c. 54 000 MJ ha-1, which is comparable to grass + 150 kg ha-1 N. Farm E actually achieved this level two of the three years under study. Farms with beef production generally had poorer soils with a probably poorer production potential. UME output and daily LWG of the bulls were comparable to results from farms with grass + N pastures (Danielsson et al., 1992). However, stocking rates were higher and individual animal intake was lower in the present study. This was also true for the dairy farms (Magnusson and Landfeldt, 1991). Calculations of marginal costs for the white clover pastures showed that these were about half of grass + N pastures (Frankow-Lindberg et al., 1993). Bloat was not a problem on any of the farms in our study, despite the generally high levels of clover in the pastures. When asked, the farmers were all very pleased with the production levels and economic results achieved. Those who in the beginning doubted that it was possible to produce herbage without N fertilizer applications are now confident that it is indeed possible. None will abandon white clover, while most of them with a mixture of pastures will try to increase those with a white clover component. Questions that remains to be solved are mainly related to persistence (both to winter cold and to drought) and optimal proportion of white clover necessary in the sward.

Northern Ireland

Clover development farms are pursuing management strategies to actively encourage an increase in the clover content of their grazing swards. These include fertilizing policy, grazing to target sward heights, implementing longer rest periods mid season on grass/clover swards, integration of sheep with cattle, reseeding and the importance of winter grazing clover swards. ABS records to date indicate that clover development farms are able to generate gross margins in their sheep enterprises which are above the average of all recorded flocks. On going ABS records will indicate the rate at which physical and financial progress can be made by increasing the reliance on grass/clover swards on beef/sheep farms. As further data becomes available development farms will be used for farm demonstrations with the objective of encouraging other farmers to make more use of clover on their farms.


ALLEZARD, V., ANNEZO, J. F., BASSET, J., BOSCHER, B., KEROUANTON, J. and PFLIMLIN, A. (1983) Trois éleveurs breton ouvrent le débat: pourquoi des prairies a base de trèfle blanc? Comment les conduire? Les difficultés rencontrées. Fourrages. No. 94, 9-17.

ARNESSON, A. (1991) Inventering av vitklöver i betesvall. Swedish University of Agricultural Sciences. (Unpublished data.)

BAX, J. A. and THOMAS, C. (1992) Developments in legume use for milk production. In: Hopkins, A. (ed.). Grass on the Move. Occasional Symposium of the British Grassland Society. No. 26, pp. 40-53.

CASTLE, M. E., REID, D. and WATSON, J. N. (1983) Silage and milk production: Studies with diets containing white clover silage. Grass and Forage Science. 38, 193-200.

CHESTNUTT, D. M. B. (1992) Effect of swards surface height on the performance of ewes and lambs continously grazed on grass/clover and nitrogen fertilised swards. Grass and Forage Science. 47. 70-80.

DANIELSSON, D.-A., JOHNSSON, S. and LINDELL, L. (1992) Ungtjursuppfödning på bete och vallfoder - en fältstudie i västra Sverige. Swedish University of Agricultural Sciences. Department of Animal Nutrition and Management, Report 208.

DOYLE, C. J. and MORRISON, J. (1983) An economic assessment of the potential benefits of replacing grass by grass-clover mixtures for 18-month beef systems. Grass and Forage Science. 38, 273-282.

FRAME, J. and NEWBOULD, P. (1986) Agronomy of white clover. Advances in Agronomy. 40.1-88.

FRANKOW-LINDBERG, B., DANIELSSON, D.-A., and ARNESSON, A. (1993) Vitklöver i betesvallen - ett sätt att sänka beteskostnaden. Swedish University of Agricultural Sciences, Fakta. Mark/växter, No. 3.

GIOVANNI, R. (1990) La prairie graminée-trèfle blanc II. Production fourragère et paturage par des jeunes taurillons laitiers. Fourrages. No. 122, 139-158.

LAIDLAW, A. S. and STEEN, R. W. J. (1989) Turnover of grass lamina and white clover leaves in mixed swards continously grazed with steers at a high- and low- N fertilizer level. Grass and Forage Science. 44, 249-258.

MAGNUSSON, G. and LANDFELDT, A. (1991) Mjölkkor på bete - en studie på fem gårdar i Skaraborgs län. Swedish University of agricultural Sciences, Department of Animal Nutrition and Management. Report 206.

ORR, R. J., PARSONS, A. J., PENNING. P. D. and TREACHER, T. T. (1990) Sward composition, animal performance and the potential production of grass/white clover swards continuously stocked with sheep. Grass and Forage Science. 45, 325-336.

PFLIMLIN, A., CHENAIS, P., ANNEZO, J.-F., KEROUANTON, J. and LE MEUR D. (1989) Paturages de ray-grass anglais - trèfle blanc par les vaches laitières. Proceedings of the XVI International Grassland Congress, Nice, France, pp. 1159-1160.

PFLIMLIN, A. and JOURNET, X. (1983) Productivité et conduite au paturage de l'association graminée-trèfle blanc. Fourrages, No. 95, 171-187.

RYAN, M. (1988) Possibilities for reduced inputs to grassland for dairying. Proceedings of the 12th General Meeting of the European Grassland Federation. Dublin. Ireland, pp. 488-492

RYAN, M. (1989) Prospects for legume-based dairying in cool temperate conditions. Proceedings of the XVI International Grassland Congress. Nice. France, pp. 1131-1132

SWIFT, G. and VIPOND, J. E. (1992) Development of a sustainable grass/clover pasture for upland sheep grazing. Herba, No. 5, 65-67.

YOUNIE, D. (1992) Reliability of output from clover-based and intensively fertilizes grassland systems. In: Hopkins, A. (ed.). Grass on the move. Occasional Symposium of the British Grassland Society. No. 26, pp. 183-185.

Previous Page Top of Page Next Page