Duke (1981) listed white and yellow sweetclovers as flowering during summer and fall. Munz (1973) listed white sweetclover as flowering from May through September, and yellow from May through August. McLeod (1982) stated that the yellow species blooms earlier than the white.

Annual white sweetclover (cv 'Hubam') sown on May 7 attained full flower during the first week of August (coastal Massachusetts) (Bugg and Ellis, 1990).

As related by Duke (1981), white and yellow sweetclovers are insect pollinated; pollination triples the seed set obtained for yellow sweetclover without pollination. For both species, 2-5 colonies of bees per acre are recommended, but 25 per acre gives maximum pollination. White sweetclover should be harvested for seed just as the pods turn black or dark brown.

Annual white sweetclover (cv 'Hubam') was avoided by one farmer because of its tendency to reseed (Schonbeck, 1988).

White sweetclover attains a height of of 1 m or more, and yellow sweetclover attains heights of from 0.3 m to 2.8 m (Duke, 1981). Munz (1973) listed white sweetclover as attaining 1-2 m.

Annual white sweetclover (cv 'Hubam') slightly exceeded 1.00 m (Bugg and Ellis, 1990).

White sweetclover grows over 6 ft. high (Johnny's Selected Seeds, 1983). Sweetclovers can attain heights of 4-8 feet, and growth is vigorous (Schonbeck, 1988). In Mendocino County, height in May of October-sown mixed stands of white and yellow sweetclovers was 22.23+/-5.81 cm (Mean +/- S.E.M.) (Bugg et al., unpublished data).

The biennial form of white sweetclover has a deep taproot; however, in the north, mowing or grazing during the first year prevents good root growth (Duke, 1981). Johnny's Selected Seeds (1983) wrote that biennial sweetclover has a deep, strong root system that can penetrate through soil layers.

White and yellow sweetclovers have deep taproots, according to Bugg (pers. comm.).

Kutschera (1960) reported that yellow sweetclover (Melilotus officinalis) generally roots to a depth of 96 cm and white sweetclover (Melilotus albus) generally roots to a depth of 65-240 cm.

In the north, mowing or grazing biennial white sweetclover during the first year prevents good root growth (Duke, 1981). Grazing, and presumably, mowing can begin once plants have attained a height of 20-25 cm. For the biennial forms of the white species, mow during bud stage (before flowering). Hay yields the first year range from 2.2-3.5 Mg/ha, and in the second year 2.2-8.1 Mg/ha. In India, the annual form ('Hubam') can yield 9-10.5 Mg/ha in three cuttings.

Duke (1981) related that yellow sweetclover matures 10-14 days earlier than biennial white sweetclover. First-year top growth should be removed before August or after mid October, either by mowing or grazing. Grazing (or presumably, mowing) can begin when the stand height is from 20-25 cm, and frequent removal is needed to prevent forage from becoming coarse and unpalatable. Yields of cv 'Madrid' can be 4.2-4.5 Mg/ha in the first year, and 5.5-8.5 Mg/ha in the second.

Sweetclovers can be difficult to incorporate, and regrowth can be a problem (Schonbeck, 1988). White sweetclover may be turned under at any time, according to Duke (1981). However, McLeod (1982) advised cultivating in the early spring, rather than the fall, because biennial forms will resume growing the second year if turned under in the fall, but not in the spring. Duke (1981) recommended that yellow sweetclover is best incorporated during the second spring, after the stand is more that 15 cm tall.

Mouldboard plough, rotavator, tandem disc, offset disc, and heavy duty cultivator were evaluated as tools for incorporating yellow sweetclover prior to spring wheat planting. There were no significant differences due to implement for amount of soil nitrogen in the 0-60 cm profile 4285(Foster, 1990).

Foster (1990) also evaluated three dates of incorporation for incorporating yellow sweetclover prior to spring wheat planting: June 15, July 1, and July 15. June 15 incorporation gave the best result in terms of the amount of soil nitrogen in the 0-60 cm profile (59, 66, and 70 kg/ha in the three years of the study), and for grain yield of the wheat.

White sweetclover should be harvested for seed just as the pods turn black or dark brown (Duke, 1981). Yellow sweetclover is best harvested if windrowed when damp and threshed 4-5 days later with a combine having a pickup attachment. The binder or combine should be set high.

If the companion grain crop of sweetclover is to be harvested the first year, set the binder on the combine high (Duke, 1981).

Duke (1981) wrote that white sweetclover could be used as forage, hay, and ensilage, and for pasture and soil imporvement. It is also used as a green manure and cover crop and is a top honey plant. Cortical fiber can be used for paper manufacture, and seed oil is used in paint and varnish. Seed meal is detoxified and used as a protein supplement in cattle feed. Yellow sweetclover is used for forage, hay, pasture, and soil improvement and erosion prevention. As an herb, it is used to flavor cheeses and tobacco snuff, and seeds used as a substiture for Tonka beans. It is also a honey plant, and Kalmuks use the roots for food.

McLeod (1982) wrote that sweetclover can be used as hay, pasture, and green manure. Sweetclovers are recommended for renovation of very poor, thin soils. They fit well in a rotation with a cereal and corn. The sweetclover is sown with the cereal, wheat, oat or barley, for example. Even flax can be used in this manner. They act as nurse crops for the sweetclover.

Madson (1951) remarked that sweetclovers are little used as winter cover crops because growth occurs too late in the spring for them to fit into the usual cover crop program. Where they fit a short rotation program, they are excellent for opening up heavy soils.

Johnny's Selected Seeds (1983) stated that white sweetclover is grown for nitrogen fixation, organic matter, and soil life enhancement.

Sweetclover is often seeded together with a nurse crop of some small grain (Duke, 1981). During establishment, sweetclovers may benefit from a companion crop of oat (Schonbeck, 1988).

Goldstein (1989) wrote that in the northern Great Plains, yellow sweetclover is useful when grown in combination with a brome grass or a winter-hardy orchard grass like cv 'Pierre.' This is better than a solid stand of yellow sweetclover for both soil improvement and grazing. Such a mixture could be grazed or stubble mulched by undercutting the plants and leaving the residue on the surface. This approach to tillage leads to better water infiltration and efficiency of nitrogen use in the following grain crops than does plowing.

Sweetclovers may be interseeded into rows of squash or overseeded into brassicas, but they are not suitable for overseeding into corn (Schonbeck, 1988).

As related by Duke (1981), for the biennial forms of the white species, hay yields the first year range from 2.2-3.5 Mg/ha, and in the second year 2.2-8.1 Mg/ha. In India, the annual form ('Hubam') can yield 9-10.5 Mg/ha in three cuttings. For yellow sweetclover, yields of cv 'Madrid' can be 4.2-4.5 Mg/ha in the first year, and 5.5-8.5 Mg/ha in the second.

Above-ground biomass yields for yellow sweetclover at 15 July (following establishment by overseeding into standing wheat during late April or early May the previous year) were 1,320, 2,117, and 3,238 kg/ha during a three-year study in Saskatchewan (Foster, 1990).

Mid-May harvest harvest of fall-sown mixed biennial sweetclovers yielded 1.656+/-0.626 Mg/ha, Mean +/-S.E.M. in an organic vineyard in Hopland, Mendocino County, California. When weeds were included, the biomass was 4.930+/-0.620 (Mg/ha, Mean +/- S.E.M.) (Bugg et al., unpublished data).

Duke (1981) wrote that 'Hubam' annual white sweetclover grown as a green manure could add about 80 kg N/ha. The review by Bugg et al. (1991c) listed white sweetclover as containing 80 - 180 kg N/ha. Bundy et al. (No date) wrote that sweetclover can contain from 80-120 lb N/acre.

During a three-year study in Saskatchewan (Foster, 1990), above-ground nitrogen yields for yellow sweetclover at 15 July (following establishment by overseeding into standing wheat during late April or early May the previous year) were 26, 52, and 65 kg/ha. Mouldboard plough, rotavator, tandem disc, offset disc, and heavy duty cultivator were evaluated as tools for incorporating yellow sweetclover prior to spring wheat planting. There were no significant differences due to implement for amount of soil nitrogen in the 0- 60 cm profile. Three dates of incorporation were evaluated for incorporating yellow sweetclover prior to spring wheat planting: June 15, July 1, and July 15. June 15 incorporation gave the best result in terms of the amount of soil nitrogen in the 0-60 cm profile (59, 66, and 70 kg/ha in the three years of the study), and for grain yield of the wheat (Foster, 1990).

Crimson clover showed a consistently-higher water use efficiency (g of dry matter produced per liter of water evapotranspired) than hairy vetch at 10, 20, and 30C. Sweetclover showed intermediate values (Zachariassen and Power, 1991).

Werenfels et al. (1963) conducted unreplicated field observations in long-term cover crop plots that had been established in 1924 and managed consistently until 1961 in an apricot orchard on a Yolo loam soil in Davis, California. The researchers suggested that infiltration rates (ring infiltrometers) were in descending order: alfalfa (Medicago sativa) > 'Hubam' white sweetclover (Melilotus alba) > sour clover (Melilotus indicus) > cereal rye (Secale cereale) > clean cultivation. The researchers stated that the alfalfa and white sweetclover alleviated a plow pan, but that cereal rye did not. Unfortunately, the unreplicated nature of the study severely limits interpretation.

Duke (1981) wrote that the biennial form of white sweetclover has a deep taproot and relieves soil compaction.

As reported by Wen-Cheh et al. (1989), leguminous cover crops cause acidification of soils. In the long run and in the absence of liming, this could lead to downward leaching of cationic nutrients and a decrease in base saturation. In terms of mg of H+ produced per g of nitrogen fixed, experiments in hydroponic tanks indicated a rating of red clover = crownvetch > alfalfa = white sweetclover. Extrapolations based on commonly-encountered field production values suggested that the amounts of acidity produced per year per hectare would be 4.2 to 9.4 kg for red clover, 5.2 to 14.0 kg for alfalfa, 4.5 kg for crownvetch, and 4.6 kg for white sweetclover.

Crimson clover showed a consistently-higher water use efficiency (g of dry matter produced per liter of water evapotranspired) than hairy vetch at 10, 20, and 30C. Sweetclover showed intermediate values (Zachariassen and Power, 1991).

Werenfels et al. (1963) conducted unreplicated field observations in long-term cover crop plots that had been established in 1924 and managed consistently until 1961 in an apricot orchard on a Yolo loam soil in Davis, California. The researchers suggested that infiltration rates (ring infiltrometers) were in descending order: alfalfa (Medicago sativa) > 'Hubam' white sweetclover (Melilotus alba) > sour clover (Melilotus indicus) > cereal rye (Secale cereale) > clean cultivation. The researchers stated that the alfalfa and white sweetclover alleviated a plow pan, but that cereal rye did not. Unfortunately, the unreplicated nature of the study severely limits interpretation.

Duke's (1981) review indicated that sweetclovers are attacked by sweetclover weevil, Sitona cylindricollis, which can destroy a crop by defoliating seedlings. Sweetclover root borer, Walshi miscecolorella, and sweetclover aphid, Therioaphis riehmi (Boerner), damage sweetclover in the Great Plains. Some resistance is available for the latter. Blister beetles, Epicauta spp., feed on foliage of low-coumarin cultivars. Grasshoppers, leafhoppers, clover leaf weevil, cutworms, webworms, green cloverworm (Platyena scabra) and other caterpillars are seldom serious problems.

In a field trial in coastal Massachusetts, Bugg and Ellis (1990) found that high densities of nectarivorous predatory wasps (Hymenoptera: Vespidae and Sphecidae) were attracted to the flowers of annual white sweetclover (11 taxa). Buckwheat floral nectar also attracted several species of predatory wasps and appeared to be a more flexible means of attracting wasps, inasmuch as it grows rapidly and flowers sooner (30 days vs about 60 days). Of course, biennial forms of white sweetclover would have flowered still later than the annual form used here, i.e., during the subsequent year. Although several Fabaceae afford extrafloral nectar, white sweetclover appears to be among the few that provide floral nectar accessible and attractive to numerous predatory wasps. Buckwheat often ceases production of nectar during afternoon hours. Afternoon activity usually appeared low on buckwheat, whereas nectarivores continued to visit white sweetclover. No Ichneumonoidea were observed imbibing at the flowers of white sweetclover, and few were observed at blooms of buckwheat. Peak densities of sweetclover aphid, Therioaphis riehmi (Boerner), were observed on July 23rd. However, compared to what was observed in faba bean, sorghum, and hairy vetch, relatively few Coccinellidae were seen on white sweetclover. It is not clear whether sweetclover aphid was less palatable than aphids available on other cover crops.

Based on observations in New Mexico, flowers of yellow sweetclover attract predaceous wasps (e.g., Ammophila sp. [Sphecidae]) (Bugg, pers. comm.).

Sweetclovers as cover crops give poor weed control (Schonbeck, 1988). White sweetclover is small seeded and rather slow to establish. Problems with weeds should be expected (Bugg, pers. comm.). Sweetclovers may benefit during establishment from a companion crop of oat (Schonbeck, 1988) or some other small grain (Duke, 1981).

Annual white sweetclover (cv 'Hubam') was avoided by one farmer because of its tendency to reseed (Schonbeck, 1988).

In early May in an organic vineyard at Hopland, Mendocino County, California, fall-seeded mixed biennial white and yellow sweetclovers had a vegetational cover of 41.25+/-13.29 % (Mean +/- S.E.M.), which was significantly less than the cover afforded by most annual cover crops. A mid-May harvest of weeds (mainly chickweed, shepherds purse, rattail fescue, and annual ryegrass) from the sweetclover plots yielded a total weed biomass of 3.274+/-0.562 Mg/ha, Mean +/- S.E.M., which is 66.6% of the figure obtained from weedy control plots (Bugg et al., unpublished data).