Previous Page Table of Contents Next Page

Effect of sowing rate, sowing date, sowing method and weed control methods on yield of white clover in mixed swards

Bodil E. Frankow-Lindberg

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

Materials and methods


Reliable establishment of clover is crucial for a wider adoption of white clover based swards on commercial farms. There is, however, few guidelines on suitable establishment techniques in the Scandinavian environment.

Experience from Great Britain indicates that varying the sowing rates of white clover within rather broad limits only affects herbage yield in the early life of the sward (Laidlaw, 1978 and references in this paper). Spring sowing compared to autumn sowing has generally resulted in the highest clover yields during the first harvesting year (Younie et al., 1985, Laidlaw and McBride, 1992, Fraser and Kunelius, 1993), with differences diminishing with time (Laidlaw and McBride, 1992; Fraser and Kunelius, 1993). Undersowing in barley has been found to give much poorer clover establishment than direct sowing in spring (Younie et al., 1985). Øyen (1991), however, reported small but positive effects due to undersowing, while Hollington, Marshall and Hides (1993) found no significant differences on stolon parameters between pure sown and undersown white clover. Broad-casting white clover seeds has been found to have a slightly advantageous effect on subsequent clover production compared to drilling (Øyen, 1991, Laidlaw and McBride, 1992).

This paper reports on two field trials performed with the aim to clarify the effects of sowing rates, sowing methods, sowing time and weed control methods on clover content and clover yield in the first harvesting year.

Materials and methods

Two trials were established in 1991 at Rådde (57°36'N, 13°16'E) at 183 m above sea level. The soil contained 10% clay, 33% silt and 6.5% organic matter. Experiment 1 (Expt. 1) comprised three different sowing rates of white clover, namely 2 (L), 4 (M) and 6 (H) kg/ha together with 6 kg/ha timothy (Phleum pratense L.), 10 kg/ha meadow fescue (Festuca pratensis L.) and 4 kg/ha smooth stalked meadow grass (Poa pratensis L.). These mixtures were sown at different times; 6 May, undersown in barley (T1), 6 May (T2), 18 June (T3), 19 July (T4) and 15 August (T5). T2 to T5 were direct sown. Seeds were drilled with an Øyjord drill with rows 12.5 cm apart. T3 was harrowed once, T4 twice and T5 three times again before sowing took place. T2 and T3 were harvested on 29 August. The experimental lay-out was a split-plot design with sowing times on main plots and sowing rates on sub-plots. There were three replicates and a total of 45 plots.

Experiment 2 (Expt. 2) involved two sowing methods, namely drilling (D) and broadcasting (BC). All treatments were sown on 6 May. Six other treatments were imposed with the aim of controlling weeds; undersown in barley (A), undersown in barley + 90 kg/ha nitrogen (B), spraying with 2.75 l/ha bentazon (Basagran 480) on 18 June + cutting on 28 August (C), spraying as in C+ cutting on 17 July (D), cutting on 17 July (E) and cutting on 17 July + 8 August (F). The cutting was done with a rather high stubble height and the cut material was left on the plots except in treatment C. The same seed mixture as in treatment M in Expt. 1 was used. The experimental lay-out was a split-plot design with sowing methods on main plots and weed control methods on sub-plots. There were three replicates and a total of 36 plots.

All treatments received 30 kg/ha nitrogen (N) at sowing except treatment B in Expt. 2. In the harvest year 33 kg/ha phosphorus and 99 kg/ha potassium were applied in spring. The trials were harvested three times; early June, late July and early September. Cultivars used were Milkanova (white clover), Vanadis (timothy), Sv. Sena (meadow fescue) and Primo (smooth stalked meadow grass). Harvested plot areas were 1.5 x 8m2. Subsamples for dry matter (DM) determination and for hand separation of species composition (DM basis) were taken from each plot.


Spring in 1991 was rather cold and wet, but the summer was warm. The winter 91/92 was unusually mild, but several severe frosts were experienced every month from October until April, while a snow cover was not present except for a few weeks in January and February. Summer of 1992 was warm and extremely dry from May until July. DM yield was significantly least in treatment L in Expt. 1 (Fig. la). The effect, however, was manifest only in the first two harvests. Total clover DM yield was not affected by sowing rates in treatments T1 to T3 (Fig. 1b).

Fig. 1. Effect of sowing rate and sowing date on DM yield of a) the whole sward and b) white clover. = first cut, =second cut, = third cut

Sowing in August (T5) resulted in very low total and clover DM yields in the first harvest. This was compensated for in the following harvests, mainly as a result of a significantly increase in both grass and clover production in the second and third harvests, respectively. Weed amounts were small, or 0-4% of DM, in Expt. 2. Treatments B and P were most efficient as regards weed control. Total and clover DM yields were least after treatment C, with white clover being most affected (Fig. 2a). Largest yields, especially of clover, were produced from the sward that was cut twice during the establishment year (treatment F). Broadcasting the seeds resulted, with but one (and doubtful) exception, in significantly poorer clover DM yields (Fig. 2b). The effect persisted throughout the year, and was most pronounced in treatment B. Total DM yield was not significantly affected though.

Fig. 2. Effect of sowing method and weed control method on DM yield of a) the whole sward and b) white clover. = first cut, = second cut, = third cut


Results from these trials regarding effects of sowing rate and sowing time agree well with previous findings. Thus, sowing rate and sowing time generally affected clover yield in early harvests with effects diminishing with time. Only with the August sowing did a high sowing rate of clover increase yield in all harvests. The rather good results after a late sowing is somewhat surprising, and is in sharp contrast to the production potential of red clover after late sowings (Frankow-Lindberg and Kornher, 1982). This suggests a good cold hardiness in Milkanova in early developmental stages (c.f. Laidlaw and McBride, 1992). Broadcasting produced unexpectedly low clover yield compared to drilling. There are no obvious reasons for this result, and further experiments in progress will shed more light on this phenomenon.

The weed pressure was rather limited in the trials. However, the different controlling methods affected clover yields. It is clear that clover establishment in a direct sown sward is favoured by early harvesting. Thus, cutting twice was better than only one cut, and the heavy crop produced in treatment C in August apparently had a negative effect on clover establishment. Undersowing in barley was also better than leaving the direct sown sward uncut until late in the establishment summer.


FRANKOW-LINDBERG, B. and KORNHER, A. (1982). Vallanläggning. Inverkan av såmetod, radavstånd och utsädesmängd, såtidpunkt samt insåningsgrödans täthet. Swedish University of Agricultural Sciences. Department of Plant Husbandry. Report 102.

FRASER, J. and KUNELIUS, H. T. (1993). Influence of seeding time on the yield of white clover/orchardgrass mixtures in Atlantic Canada. Journal of Agricultural Science. Cambridge. 120, 197-203.

HOLLINGTON, P. A., MARSHALL, A. H. and HIDES, D. H. (1993). The effect of row spacing and cover-crop on stolon development and the seed yield components of white clover cultivars of contrasting leaf size. Grass and Forage Science. 48, 1-10.

LAIDLAW, A. S. 1978. Control of white clover content in swards by varying sowing rates of perennial ryegrass and white clover seeds. Record of Agricultural Research. Department of Agriculture for northern Ireland. 26, 21-27.

LAIDLAW, A. S. and MCBRIDE, J. (1992). The effect of time of sowing and sowing method on production of white clover in mixed swards. Grass and Forage Science. 47, 203-210.

ØYEN, J. (1991). Methods of establishing grass/clover swards. In: White clover development in Europe, pp. 59-65. REUR Technical Series 19. FAO, Rome.

YOUNIE, D., WILSON, J. F., CARR, G and WATT, C. W. (1985). The effect of undersowing, nitrogen application and date of sowing on white clover establishment. In: Thomson, D. J. (ed.) Forage Legumes, pp. 182-183. Occasional Symposium No. 16. British Grassland Society.

Previous Page Top of Page Next Page