Index | Search | Home | Table of Contents

link to pdf file

Bhardwaj, H.L. 2002. Evaluation of lupin as a new food/feed crop in the US Mid-Atlantic region. p. 115–119. In: J. Janick and A. Whipkey (eds.), Trends in new crops and new uses. ASHS Press, Alexandria, VA.

Evaluation of Lupin as a New Food/Feed Crop in the US Mid-Atlantic Region*

Harbans L. Bhardwaj

*Contribution of Virginia State University, Agricultural Research Station. Journal Article Series No. 227. The use of any trade names or vendors does not imply approval to the exclusion of other products or vendors that may also be suitable.


White lupin (Lupinus albus L. Fabaceae) is one of four economically important species of the genus Lupinus, which consists of over 300 annual species (Hondelmann 1984). The other three species are: L. angustifolius, L. luteus, and L. mutabilis. Approximately 100 species of Lupinus occur in the US and Canada and most are native.

White lupin, an old crop for the southern United States, is receiving increasing attention due to its high potential in both conventional and sustainable production systems (van Santen et al. 1994; Reeves 1991). Lupin is known to potentially fix 150 to 200 kg/ha N for the use of a succeeding crop. It has been estimated that if lupin replaced a quarter of wheat area in the Southeastern US it could save 86,184 tonnes (t) of nitrogen fertilizer worth $50 million to $60 million per year (Reeves et al. 1990). This would not only reduce input costs but also help protect water from nitrogen pollution.

The New Crops Program of Virginia State University is interested in developing bitter lupin cultivars, with high contents of alkaloids, to use as green manure for environment-friendly production of summer crops; and developing sweet lupin cultivars, with non-existent or low levels of alkaloids, to provide protein-rich grains for food and feed to support ever-increasing human population. It has been estimated that grain production needs to increase by 40% to 50% to support addition of approximately 75 million human beings per year until 2020 (IFPRI 1999).

Four pre-requisites have been identified for successful development of lupin as a crop in Virginia and the mid-Atlantic region of US. These consist of: availability of high yielding, winter-hardy cultivars; characterization of lupin’s nutritional quality and alkaloid content; characterization of lupin’s nitrogen fixing potential; and management of anthracnose disease.


Seed Yield and Winter-hardiness

A diverse material consisting of French cultivars and accessions from USDA-ARS lupin collection, maintained at Pullman, Washington, were evaluated at Petersburg, Virginia (located at approximately 37.14°N and 77.24°W). Replicated experiments were conducted during 1998–1999 and 1999–1900 seasons, to compare performance of a determinate (‘Lucyanne’) and an indeterminate cultivar (‘Lunoble’) by planting in early and late October, and mid-November and by using three inter-row spacings of 30, 60, and 90 cm. A collection consisting of 124 accessions was evaluated for winter-survival during 2000–2001 season in single row observation plots.

Nutritional Quality

The effects of growing environment and genotypes on nutritional quality of lupin were evaluated by comparing the composition of seed produced in Virginia during 1994 to that produced in Maine during 1989 and 1990. During the 1999–2000 season, effects of varying alkaloid content on lupin performance were evaluated at Petersburg, Virginia. A replicated field experiment with 19 lupin lines was planted on Oct. 11, 1999 and harvested in mid June, 2000. The data were obtained for plant height and seed yield. The seeds of these 19 varieties from three replications (57 samples) were analyzed in the laboratory for determination of alkaloid and protein contents. The alkaloid content was extracted with 70% isopropanol and determined colorimetrically using bismuth oxynitrate reagent. Based on the alkaloid content, the lupin lines were given a score from 1 to 6, with 1 indicating sweet and 6 indicating bitter seeds, and the alkaloid content increasing from 1 to 6. The protein content was determined colorimetrically after acid digestion using Nessler reagent.

Nitrogen Fixation

In order to study the interaction between Bradyrhizobium strains and lupin genotypes, two greenhouse experiments were conducted. In the first experiment, 60 Bradyrhizobium strains were evaluated with three lupin genotypes. In the second experiment, 80 lupin genotypes were evaluated with three Bradyrhizobium strains selected based on their performance in the first experiment. Data on plant vigor, nodulation on crown and roots, and shoot and root dry weights were recorded and analyzed. In separate replicated field experiments, yields of cantaloupes and sweet corn, following lupin and other legumes as winter cover crops, were evaluated.


This fungal disease caused by Colletrotrichum gloeosporiodes (Penz) is a serious disease of lupins with worldwide importance. The resistance to anthracnose in white lupin is unknown. Rovral fungicide, as a seed treatment, is known to be effective against this fungus. However, there are indications that seed treatment with Rovral reduces plant vigor. In a field experiment during 2000–2001 season, 124 accessions were evaluated in pairwise combinations to compare the vigor of lupin plants grown from Rovral-treated seed with untreated seed. This experiment was replicated twice. Number of surviving plants in treated and untreated rows and their vigor were compared.


Seed Yield and Winter-hardiness

The results of replicated experiments indicated that the indeterminate cultivar yielded more than the determinate cultivar (Table 1). The seed yield of 5623 kg/ha, following a closer row spacing (30 cm) was superior to the wider row spacings (4059 kg/ha for 60 cm and 3164 kg/ha for 90 cm) for both determinate and indeterminate cultivars. These experiments also established that early October planting time was more conducive to higher seed yields. The severe winter during 2000–2001 season also established that our previous observation that ‘Lunoble’ and ‘Lucyanne’ are winter-hardy in Virginia was ill-founded. Both these cultivars suffered severe winter kill, ‘Lunoble’ more than the ‘Lucyanne’. However, considerable variation existed among the 124 accessions for winter-survival. We are currently multiplying the seed of winter-hardy lines for further evaluations.

Table 1. Performance of determinate and indeterminate lupin cultivars at Petersburg, Virginia in 1998–1999 and 1999–2000.

Comparison Seed yield (kg/ha)
Lunoble (Indeterminate) 4787 az
Lucyanne (Determinate) 3777 b
Row spacing  
30 cm 5623 a
60 cm 4059 b
90 cm 3164 b
Planting date  
Early Oct. 5861 a
Late Oct. 3895 b
Mid Nov. 3089 c

zMeans followed by similar letters were not different according to Duncan’s Multiple Range test at 5% level.

Nutritional Quality

Growing environment significantly affected the nutritional quality of lupin (Bhardwaj et al. 1998). The protein content of 12 lupin genotypes produced in Virginia during 1993–1994 varied from 32% to 43% with a mean of 37% (Table 2). Significant variation also existed among 12 lupin genotypes for other seed composition traits. The results indicated that site-specific evaluation of adapted lupin genotypes for chemical composition should be included in efforts to evaluate lupin’s overall potential. Based on a comparison of lupin’s seed composition with that of other legumes (Table 2), lupin seed have potential as human food and livestock feed. The seeds of sweet lupin do not contain alkaloids or trypsin inhibitors and do not require high temperature cooking before use as a livestock feed (Hill 1977; Lopez-Bellido and Fuentes 1986; Larson et al. 1989). A comparison between sweet and bitter lines indicated that bitter lines had protein content similar to the sweet lines (32% each) but higher values for plant height (68 vs. 58 cm) and plant yield (333 vs. 207 g). However, the considerable variation among 19 lines for seed yield could be exploited to develop high yielding sweet lines (Table 3).

Table 2. Composition of lupin and seeds of other legumes.

Variable Content (%)
Lupin Kidney bean Pinto bean Soybean
Ash 2.99z 3.37y 3.63b 4.87b
Protein 36.59 22.53 20.88 36.49
Amino acids        
Aspartic acid 2.96 2.72 2.53 4.59
Threonine 0.83 0.95 0.88 1.58
Serine 0.99 1.23 1.14 2.11
Glutamic acid 6.08 3.44 3.18 7.07
Glycine 0.92 0.88 0.81 1.69
Alanine 0.83 0.94 0.87 1.72
Valine 1.06 1.18 1.09 1.82
Methionine 0.35 0.34 0.31 0.49
Isoleucine 1.26 1.00 0.92 1.77
Leucine 1.92 1.80 1.67 2.97
Tyrosine 1.32 0.63 0.59 1.38
Phenylalanine 1.02 1.22 1.13 1.90
Histidine 0.51 0.63 0.58 0.98
Lysine 1.19 1.55 1.43 2.43
Arginine 2.98 1.39 1.29 2.83
Proline 1.31 0.95 0.88 2.13
Oil 4.86 1.06 1.13 19.94
Fatty acids        
Palmitic 6.60 0.14 0.23 2.12
Stearic 0.78 0.02 0.01 0.71
Oleic 50.95 0.08 0.23 4.35
Linoleic 23.49 0.23 0.17 9.92
Linolenic 9.68 0.36 0.24 1.33
Saturated 11.29 0.15 0.23 2.88
Unsaturated 88.72 99.85 99.77 97.12
Polyunsaturated 33.30 0.59 0.41 11.25
Monounsaturated 55.43 0.08 0.23 4.40

zThe values are means from 12 lupin lines: Kali, Kalina, L1027N, L127N, L133N, L139N, L251N, L389N, PI-469095, PI-483074, PI-481545, and Ultra grown at Petersburg, Virginia during 1993–1994 (Bhardwaj et al. 1998).
yThese values are from U.S. Department of Agriculture, Agricultural Research Service. 2001. USDA Nutrient Database for Standard Reference, Release 14. Nutrient Data Laboratory Home Page, Release 14 (July, 2001).

Table 3. Variation among 19 lupin lines for seed composition and agronomic traits.

Line Alkaloid
UT92 5.7 33 70 175
T108 5.7 36 68 282
UT109 5.7 32 59 152
T93 5.3 32 76 359
T86 5.3 31 78 219
UT66 5.3 31 75 367
UT69 5.3 30 69 373
UT60 5.0 29 83 493
UT11 5.0 33 71 443
T70 5.0 31 79 369
T100 4.7 35 41 192
T79 4.7 28 69 411
UT67 4.3 34 71 581
UT63 4.3 34 80 468
T65 3.7 32 76 573
T146 2.7 33 41 97
UT119 1.3 33 43 158
T58 1.0 28 65 260
UT62 1.0 34 73 289
LSD (.05) 1.8 4 13 142

zBased on alkaloid content, lupin lines were given a score from 1 to 6 with one indicating sweet and 6 indicating bitter seeds. The alkaloid content increases from 1 to 6.

Nitrogen Fixation

Significant Bradyrhizobium strain by lupin genotype interaction existed for nodulation score, and shoot and root dry weights (Table 4). Comparison of relative ranks indicated that nodulation effectiveness was dependent upon specific strain and lupin genotype combination. It was concluded that specific bradyrhizobial strain and lupin genotype combinations would need to be identified if lupin is to be a successful component of sustainable crop production systems (Robinson et al. 2000).

Table 4. Analysis of variance (mean squares) for lupin plant characteristics following inoculation of three lupin cultivars with 60 Bradyrhizobium strains.

Source Plant vigor Nodulation score Dry weight
Crown Roots Shoot Root
Strains (S) 0.91** 7.59** 3.31** 0.52** 0.06
Cultivars (C) 6.71** 62.07** 22.20** 11.68** 0.77**
S C 0.22 1.03** 0.88** 0.23* 0.05**
Error 0.25 0.49 0.55 0.18 0.03
Mean 2.54 2.57 2.36 1.27 0.36

*, **:Significant at 5 and 1 percent levels, respectively (Robinson et al. 2000).

Lupin as a winter cover crop resulted in excellent yields of cantaloupe and sweet corn (Table 5). The results indicated that use of either lupin or hairy vetch as a winter cover crop for production of cantaloupe and use of lupin as cover crop for production of sweet corn results in their performance being better than if 112 kg of N/ha fertilizer were used.

Table 5. Performance of sweet corn and muskmelon following legume cover crops and nitrogen fertilizer application during 1998–1999 and 1999–2000 seasons at Petersburg, Virginia.

Treatment Sweet corn
(ears, t/ha)
(fruit, t/ha)
Lupin as winter cover crop 6.8 az 40.7 a
Hairy vetch as winter cover crop 4.8 b 35.7 ab
Austrian winter pea as a winter cover crop 3.6 b 15.8 bc
112 kg N/ha 3.2 b 9.4 c
Control 0.9 c 3.9 c

zMeans followed by similar letters were not different according to Duncan’s Multiple Range test, 5% level.


We observed considerable variation among plant material for plant vigor during 2000–2001 season. Not all accessions suffered reduced vigor following seed treatment with Rovral fungicide. These results need to be repeated but may provide a basis for development of locally-adapted lupin cultivars whose seed could be treated with this fungicide to manage anthracnose.


Based on seed yield and success in identifying winter-hardy genotypes, the prospects of developing lupin, as an alternative winter crop in Virginia, are good. Lupin’s nitrogen fixation capability may provide an additional incentive for these efforts.