Composition, ileal amino acid digestibility and nutritive value of organically grown legume seeds and conventional rapeseed cakes for pigs

Eight white-flowered pea (Pisum sativum) and two white-flowered field bean (Vicia faba) cultivars grown organically were analysed for proximate composition and amino acid content. In vivo ileal amino acid digestibilities and faecal energy digestibility were predicted from the in vitro enzymatic digestibility of nitrogen and organic matter, respectively. The crude protein (CP) content of the pea and field bean cultivars ranged from 244 to 279 and from 320 to 347 g/kg dry matter (DM), respectively. The concentrations of several essential amino acids in protein decreased as the CP content increased. In peas, predicted in vivo digestibilities did not correlate with chemical composition, and in field beans were lower than in peas. A digestibility trial was carried out on six cannulated barrows according to a 6 × 5 cyclic changeover design to determine the faecal and ileal nutrient digestibilities of organically grown leafed peas (cv. Sohvi, 199 g CP/kg DM), semileafless peas (cv. Karita, 240 g CP/kg DM), field beans (cv. Kontu, 320 g CP/kg DM), narrow-leafed lupins (Lupinus angustifolius cv. Pershatsvet, 220 g CP/kg DM), and conventional warmand cold-pressed rapeseed cakes (360 and 313 g CP/kg DM, respectively). The net energy contents of the leafed and semileafed peas, field beans, lupins, and coldand warmpressed rape seed cakes were 10.8, 11.2, 9.8, 9.7, 9.4 and 12.3 MJ/kg DM, respectively. The apparent ileal digestibilities of lysine and threonine were similar, but the digestibility of methionine was poor in all legume seeds. Cystine digestibility was highest in lupins and lowest in field beans. With the exception of phenylalanine, there was no difference in apparent ileal amino acid digestibilities between rapeseed cakes.


Introduction
In recent years, alternative livestock production methods such as organic farming have been developed to meet the demands of consumers which are concerned about environmental pollution and animal health and welfare in intensive production systems.Organic livestock farming is based primarily on home-grown feedstuffs with the objective of establishing an almost complete on-farm nutrient cycle.In pig feeding, purchased conventional feedstuffs are limited to 20% of the total amount of annual feed consumption on a dry matter (DM) basis.Moreover, the use of solvent-extracted oil seed meals, e.g.soybean and rapeseed meal, and synthetic amino acids is not allowed (EC 1999).Nitrogen fixating legumes play an important role in the crop rotation of organic farms.In addition, legume seeds, i.e. peas, field beans and sweet lupins, are valuable protein-rich feedstuffs for pigs.
The suitability of different pea cultivars for organic farming has been studied recently.As well as high yield, desirable features for pea cultivars are even maturation, resistance to lodging and high protein content.In monoculture, the lodging percentage is higher in leafed than in semileafless pea cultivars.In mixed cropping, however, leafed cultivars compete better with weeds, and their seeds contain more protein than do those of semileafless cultivars (Niskanen 2000).However, little information is available on the variation in amino acid composition and digestibility in organically grown pea cultivars for pigs.Furthermore, the nutritive value of current field bean cultivars has not been determined.
The crude protein (CP) content of peas (255 g/kg dry matter (DM)), field beans (300 g/ kg DM) and sweet lupins (320-380 g/kg DM) is intermediate between that of soybean meal and cereals.Compared to soybean meal, the protein of legume seeds, sweet lupins excepted, is rich in lysine.Legume seed protein contains threonine in a similar proportion to that in soybean meal protein; the proportions of sulphur-containing amino acids and tryptophan are, however, lower (Gatel 1994, van Barneveld 1999).Without synthetic amino acid supplementation, cereal-legume seed diets do not meet the protein requirements of pigs.Therefore, rapeseed cake, either organic or conventional, is generally included in organic pig diets.Rapeseed protein is rich in sulphur-containing amino acids and threonine and so complements well the amino acid insufficiency of legume seeds (Castell and Cliplef 1993).However, the fat content of rapeseed cakes can vary greatly, from 100 to 300 g/kg DM, depending on the crushing method.Cold-pressed rapeseed cakes with a high fat content are seldom used in conventional pig production and hence little information is available on their digestibility and nutritive value for pigs.
The first objective of this study was to investigate the variation in the chemical composition and amino acid and energy digestibilities of organically grown pea and field bean cultivars for pigs by an in vitro method.The second objective was to determine in vivo the apparent ileal amino acid digestibilities and nutritive value of organically grown grain legumes, i.e. peas, field beans and sweet lupins, and conventional rapeseed cakes for pigs.

Pea and field bean samples
The chemical composition and in vitro digestibility of eight white-flowered pea (Pisum sativum) and two white-flowered field bean (Vicia faba) cultivars were evaluated.Seed samples for the pea cultivars (Table 1) were kindly supplied by Mr. Markku Niskanen, MTT Agrifood Research Finland, South Ostrobotnian Research Station, and were grown organically in Ylistaro, Finland, in 1997.The two field bean cultivars, Ukko (Hankkija 1984) and Kontu (Boreal 1997), were obtained from organic farms in Tarvasjoki and Orimattila, Finland, respectively.All samples were ground to pass through a 1-mm sieve and analysed for proximate composition, amino acid content, and in vitro ileal nitrogen and DM digestibilities and in vitro faecal organic matter digestibility.

Digestibility trial
Eight growing barrows (seven Finnish Landrace and one Finnish Landrace × Yorkshire) with an average body weight (BW) of 39 kg were surgically fitted with a T-shaped silicone cannulae at the caecum according to the post valve T-cannula (PVTC) method (van Leeuwen et al. 1991).Before surgery, the pigs fasted for 36 h and had no access to water for 12 h.After surgery, the pigs were allowed a 16-d recovery period.To minimise pain, they were injected with Finadyne (2.2 mg/kg BW i.m., Orion, Finland) for 3 d.To prevent infections, they were injected with Borgal (3 ml/50 kg BW i.m., Hoechst, Germany) for 1 d and thereafter were given Oriprim (10 g/d, Orion, Finland) with their feed for 6 d.During the recovery period, the daily allowance was gradually increased until the pre-surgical level of feed intake was achieved.Water was available ad libitum.The pigs were housed in 1.43 × 1.23-m metabolic pens with slatted plastic floors and transparent plastic sides at an ambient temperature of 20 to 23°C.One pig died immediately after surgery because of internal bleeding.
The experiment was carried out on six pigs according to a 6 × 5 cyclic change-over design (Davis and Hall 1969).The six experimental treatments were as follows: 1) leafed peas (Pisum sativum cv.Sohvi), 2) semileafless peas (Pisum sativum cv.Karita), 3) field beans (Vicia faba cv.Kontu), 4) narrow-leafed lupins (Lupinus angustifolius cv.Pershatsvet), 5) warm-pressed rapeseed cake (Tupla-Öpex, Mildola, Helsinki) and 6) cold-pressed rapeseed cake (Virgino, Kankaisten Öljykasvit Ltd., Hämeenlinna).The legume seeds were grown organically in 1998, the leafed peas in Vihti, the semileafless peas and field beans in Orimattila and the lupins in Piikkiö, Finland.The composition of the barleybased experimental diets is shown in Table 2.The barley and legume seeds were ground in a hammer mill to pass through a 4-mm screen (Automatic, Automatic equipment MFG, Pender, Nebrasca, USA).Chromium mordanted straw (93 g Cr/kg DM) prepared according to Udèn et al. (1980) was used as an indigestible marker (1.6 g/kg feed).The pigs were fed twice daily (0600 and 1800) and were given 100 g feed/kg BW 0.75 , based on the mean BW of the pigs at the beginning of each period.The feed allowance was kept constant for the whole 14-d period, and adjusted according to the BW at the beginning of each period.Feed was mixed with water (2 l/kg feed) immediately before feeding.
There were five 14-d experimental periods.After a 6-d adaptation period, faeces were collected quantitatively for 3 d according to van Kleef et al. (1994).Thereafter, ileal digesta were collected for a total of 12 h as follows: on day 12 from 0600 to 0800, 1000 to 1200, and 1400 to 1600, and on day 14 from 0800 to 1000, 1200 to 1400, and 1600 to 1800.The digesta were collected directly into a plastic bag fixed to the cannulae.The plastic bags were removed every 15 min, weighed, and frozen instantly at -20°C.

Chemical analyses
Feed samples and freeze-dried digesta and faecal samples were ground to pass through a 1-mm sieve.Dry matter content was determined by drying at 103°C for 16 h.Ash, crude fibre and ether extract (after acid hydrolysis) were determined by standard methods (AOAC 1990).
Starch was analysed after ethanol extraction according to Bach Knudsen et al. (1987).Nitrogen was determined by the Dumas method with a Leco FP 428 N analyser (Leco Corp., St Joseph, USA).Amino acids were analysed according to the official method of the EC (1998).Minerals were determined by ICP emission spectrophotometry (Luh Huang and Schulte 1985) and chromium by atomic absorption spectrophotometry (Williams et al. 1962).In vitro ileal DM and ni- 1 The premix supplied per kg of feed: 2.3 g of Ca, 0.8 g of P, 0.5 g of Mg, 3.3 g of NaCl, 103 mg of Fe, 22 mg of Cu, 91 mg of Zn, 23 mg of Mn, 0.28 mg of Se, 0.28 mg of I, 5170 IU of vitamin A, 517 IU of vitamin D 3 , 50 mg of vitamin E, 2 mg of thiamin, 5 mg of riboflavin, 3 mg of pyridoxine, 20 µg of vitamin B 12 , 0.2 mg of biotin, 14 mg of pantothenic acid, 20 mg of niacin, 2 mg of folic acid, and 2 mg of vitamin K.
trogen digestibilities were determined according to Boisen and Fernández (1995) and in vitro faecal organic matter digestibility according to Boisen and Fernández (1997).

Calculations and statistical analysis
The in vivo ileal digestibilities of nitrogen and amino acids were predicted from in vitro nitrogen and dry matter digestibilities as described by Boisen and Fernández (1995) and the in vivo faecal digestibility of energy was predicted from in vitro organic matter digestibility according to Boisen and Fernández (1997).Correlation coefficients were calculated between proximate composition and amino acid composition and digestibility coefficients by the CORR procedure of SAS (SAS 1998).
Apparent ileal and total tract digestibilities were calculated using chromium as an indigestible marker as follows: Apparent ileal or total tract digestibility = [(N/Cr) d -(N/Cr) f ] / (N/Cr) d where (N/Cr) d = the dietary ratio of nutrient to Cr and (N/Cr) f = the ratio of nutrient to Cr in faeces or ileal digesta.The digestibilities of the investigated protein feedstuffs were calculated by difference using previously determined digestibility coefficients for barley (Valaja et al. 1999).It was assumed that synthetic amino acids were completely absorbed in the small intestine.The net energy content of protein feedstuffs was calculated from chemical composition and determined faecal digestibility coefficients according to Tuori et al. (1996).
Statistical analysis of digestibility data was carried out using the GLM procedure of SAS (SAS 1998) and the following model (Snedecor and Cochran 1989): Y ijk = µ + A i + P j + D k + e ijk , where A, P and D are effects of animal, period and dietary treatment, respectively.Residuals were checked for normality and plotted against fitted values.Differences between treatments were tested with the Tukey test when appropriate.

Chemical composition and in vitro digestibility of pea and field bean cultivars
The chemical composition and in vitro digestibilities of the eight organically grown pea and two field bean cultivars studied is presented in Tables 3 and 4, respectively.The average CP content of the pea cultivars (254 ± 11.7 g/kg DM) is similar to values reported recently for whiteflowered peas (Gdala et al. 1992, Igbasan et al. 1997, Fan and Sauer 1999).The protein content of peas is known to vary greatly both between and within cultivars, as shown in the reviews of Savage and Deo (1989) and Gatel and Grosjean (1990), who reported CP ranges from 156 to 325 and from 181 to 346 g/kg DM, respectively.Here, the leafed cultivars, Sohvi and Scorpio, had a higher CP content than did the semileafless ones (271 ± 8 vs. 248 ± 3.4 g/kg DM), a finding that is in agreement with the results of official variety trials carried out in Finland (Järvi et al. 2000).The variations in CP content were not related to seed colour, since green-and yellow-seeded cultivars had similar average protein contents (255 ± 12.6 vs. 253 ± 7.5 g/kg DM).There was also no correlation between protein content and seed size (Table 5), which confirms the results of Igbasan et al. (1997).Both ether extract and crude fibre contents were low in pea cultivars, as found previously (Savage andDeo 1989, Gatel andGrosjean 1990).The crude fibre content increased (r = 0.73) and the ether extract content decreased (r = -0.74)with CP content.Field bean cultivars contained more CP and crude fibre than did peas, but the ether extract content was similar to that of peas.The higher fibre content of beans is probably due to the higher proportion of hulls in their seeds than in those of peas.The proportion of hulls in pea cultivars has ranged from 8.0 to 12.0% of total seed weight (Igbasan et al. 1997) and that of field beans from 10.5 to 15.5% (Brufau et al. 1998).

Partanen, K. et al. Organically grown legume seeds and conventional rapeseed cakes for pigs
the pea and field bean cultivars (Table 3).There was no evidence to show that this variation was related to seed weight or colour.The protein of semileafless pea cultivars contained more lysine and other essential amino acids than did that of leafed pea cultivars.This could be explained by the lower CP content in the former.Previous studies have shown that amino acid concentrations in pea protein are negatively correlated with pea protein content (Gatel andGrosjean 1990, Igbasan et al. 1997) and the same observation was made in this study (Table 5).Only arginine (r = 0.71) had a positive correlation with protein content.Field bean protein contained less lysine, methionine, cystine and threonine than did pea protein, which is consistent with the studies reviewed by Gatel (1994).Of the non-essential amino acids, the average concentrations of alanine, aspartic acid, glutamic acid, glycine, proline, and serine were slightly higher for pea than for field bean protein (results not shown).
In this study, only two field bean cultivars were evaluated, and thus no conclusions can be drawn about the relationship between CP and amino acid contents in field beans.According to Gatel (1994), the proportions of essential amino acids, particularly those of lysine, sulphur-containing amino acids, tryptophan and threonine, decrease in field bean protein as protein content of seeds increases.The concentration of amino acids in legume seed protein is a function of the amino acid composition of the storage proteins, albumins and globulins.The albumin is relatively rich in sulphur-containing amino acids (Casey et al. 1993).The bulk of the pea and field bean seed protein comprises, however, globulins, i.e. legumin, vicilin and convicilin, which have low content of sulphur-containing amino acids.
The pea cultivars showed little variation in in vitro nitrogen and organic matter digestibilities and the respective predicted in vivo ileal amino acid and faecal energy digestibilities (Table 4).Boisen and Fernández (1995) reported an in vitro nitrogen digestibility of 95.6% for peas, which is slightly higher than that observed in this study.Boisen and Fernández (1997) reported an in vitro organic matter digestibility of 89.3% and predicted an in vivo faecal energy digestibility of 84.1% for peas.These values are lower than those observed here.In peas, the predicted digestibility coefficients did not correlate with proximate composition or seed weight.The predicted in vivo digestibilities were higher than those reported for different pea cultivars in in vivo studies (Gdala et al. 1992, Fan and Sauer 1994, 1999).This is in agreement with the results of Cone and van der Poel (1993), who used a two-step in vitro method (pepsin-HCl digestion followed by pancreatin and a-amylase digestion) to predict the in vivo digestibility of different pea cultivars.They found no linear relationship between the in vivo and in vitro digestibility of pea samples.Here, lower predicted in vivo ileal amino acid digestibilities were

Partanen, K. et al. Organically grown legume seeds and conventional rapeseed cakes for pigs
found in field beans than in peas, particularly for sulphur-containing amino acids.In addition, the predicted in vivo faecal digestibility of energy was lower in the former.This confirms with the findings of previous in vivo digestibility studies (Gatel 1994).

Digestibility trial
The pigs remained healthy and consumed their feed allowances throughout the experiment.One pig was removed from the trial in the last period because its cannula came off.Postmortem examinations, carried out at the conclusion of the experiment, revealed no intestinal adhesions or other abnormalities.
The chemical composition of the legume seeds and rapeseed cakes investigated in the digestibility trial are given in Table 6.The proximate composition of legume seeds was within the range reported in the literature (Gatel 1994, van Barneveld 1999).The CP contents of leafed peas and lupins were lower than was expected from the results of cultivation trials (Mehto 1986, Järvi et al. 2000).The CP content of the leafed pea cultivar, Sohvi, was only 199 g/kg DM whereas it was 279 g/kg DM in the sample used in the in vitro trial, which was grown in 1997.The 1998 growing season was cool and rainy, which may have influenced plant maturation and thus the chemical composition of seeds, particularly the protein content (Savage andDeo 1989, Wasilewko andBuraczewska 1999).The carbohydrate composition of lupins differed from that of peas and field beans, with negligible levels of starch and a high level of crude fibre.Lupin seeds are particularly rich in hemicellulose when compared to peas and field beans (van Barneveld 1999).The amino acid concentrations of pea, field bean and lupin seed protein were in the range of values observed in the in vitro trial and reported in the literature (Igbasan et al. 1997, Brufau et al. 1998, Wasilewko and Buraczewska 1999).The concentration of lysine in protein was highest in peas, followed by field beans and lupins.The methionine content was relatively low in all legume seeds.In the pea cultivars Karita and Sohvi, it was lower than in the samples used in the in vitro trial.It was not possible to obtain a result for in vitro ileal digestibility of nitrogen and DM for legume seeds; because of the viscosity, the legume seed samples did not filtrate after enzymatic digestion.This difficulty in the filtering could have been caused by the incomplete degradation of starch by pancreatin and/or pectins present in legume seeds.The in vitro faecal organic matter digestibilities of peas and field beans were within the range of values determined in the in vitro trial (Table 6).
The ether extract content in cold-pressed rapeseed cake was very high, over twice that in warm-pressed rapeseed cake.Consequently, the CP content was lower in the former.Nyström et al. (1996) tested the nutritive value of rapeseed cakes from a large oil mill.After the prepressing stage, the fat content amounted to 232 g/kg DM, whereas the final heat-pressure-moisturetreated product contained 118 g crude fat/kg DM.Keith and Bell (1991) reported ether extract contents of 212 and 39 g/kg DM for rapeseed press cake and solvent-extracted rapeseed meal.In the protein of rapeseed cakes, the concentrations of all amino acids, except lysine, were similar.The lower lysine content in warm-pressed rapeseed cake may have been caused by heat applied during the crushing process (Nyström et al. 1996).Lysine concentrations were lower than previously reported for protein in rapeseed cake (Keith and Bell 1991).Otherwise, the chemical composition was in the range of values reported previously for rapeseed cakes (Keith andBell 1991, Nyström et al. 1996).
The faecal nutrient digestibilities of legume seeds and rapeseed cakes calculated by difference are shown in Table 7. Faecal organic matter digestibility was higher in legume seeds than in rapeseed cakes (P < 0.05).Among legume seeds, the faecal digestibility of organic matter was higher in peas than in field beans or lupins (P < 0.05).The faecal CP digestibility was highest in the pea cultivar Karita and lowest in the warm-pressed rapeseed cake.The faecal CP digestibility of the pea cultivars is in the range of Vol. 10 (2001): 309-322.values reported previously (Gdala et al. 1992, Jondreville et al. 1992, Helander et al. 1996).The faecal digestibility of CP was higher in semileafless than in leafed peas (P > 0.05).The crude fibre content was higher in leafed than in semi-leafless peas, which may be one reason for the poorer protein digestibility of the former.A high fibre content has been reported to have an adverse effect on CP digestibility in peas (Gdala et al. 1992).Tannins are also known to have an

Partanen, K. et al. Organically grown legume seeds and conventional rapeseed cakes for pigs
adverse effect on faecal protein digestibility, but their content is generally low in white-flowered peas (Gatel andGrosjean 1990, Gdala et al. 1992).Among legume seeds, the faecal digestibility of CP was lowest in lupins, lower than in the semileafless pea cultivar, Karita (P < 0.05).
The faecal digestibility of CP in field beans was similar to that of peas (P > 0.05), and is in agreement with results reviewed by Gatel (1994).
In peas, the faecal digestibility of ether extract was lower than in the study of Helander et al. (1996), who reported a faecal ether extract digestibility of 58.0% for Pika, a semileafless pea cultivar.The faecal digestibility of crude fibre was considerably higher in peas (70.1-81.5%)and sweet lupins (60.1%) than in field beans (27.8%).Both peas and sweet lupins contain only a small amount of lignin and thus the digestibility of the fibre fraction is high (Gdala et al. 1992, van Barneveld 1999).The calculated net energy contents (MJ/kg DM) of legume seeds were 10.8 for leafed peas, 11.2 for semi- a, b, c, d Means within the same row followed by the same letters do not differ significantly (P ≥ 0.05).
leafless peas, 9.8 for field beans and 9.7 for lupins.The faecal nutrient digestibilities of warmand cold-pressed rapeseed cakes did not differ significantly.The faecal digestibility of ether extract was higher than in legume seeds, whereas the crude fibre digestibility was lower than in peas and lupins (P < 0.05), but did not differ from that in field beans (P > 0.05).The determined faecal digestibilities of rapeseed cakes are in the range of values reported previously (Keith andBell 1991, Schöne et al. 1996).The cold-pressed rapeseed cake was energetically richer than the warm-pressed type (12.3 vs. 9.4 MJ NE/kg DM).This difference in net energy content was primarily due to differences in crude fat content, because there was no difference in faecal nutrient digestibilities between the rapeseed cakes.
The apparent ileal amino acid digestibilities are presented in Table 7.The ileal digestibilities of lysine, threonine, histidine, phenylalanine, tyrosine and aspartic acid did not differ between legume seeds (P > 0.05).The ileal digestibilities of arginine, isoleucine, valine and most of the non-essential amino acids were highest in sweet lupins and lowest in field beans.The greatest differences in ileal amino acid digestibility occurred for methionine and cystine.The apparent ileal digestibility of amino acids has varied greatly among pea cultivars.In recent studies (Jondreville et al. 1992, Fan and Sauer 1994, 1999), lysine digestibility has ranged from 66% to 85%, threonine digestibility from 60% to 77%, methionine digestibility from 69% to 79, and cystine digestibility from 54% to 66%.In general, the apparent ileal digestibility of methionine was very low in all legume seeds, whereas cystine digestibility was higher in lupins than in peas or field beans.The low content and poor digestibility of sulphur-containing amino acids in legume seeds makes the formulation of organic pig diets based on on-farm feedstuffs difficult.The very low apparent methionine digestibility of legume seeds may be due to the antinutritive factors and carbohydrates present in the seeds.Le Guen et al. (1995) observed considerably lower apparent methionine digestibility in raw peas (34-48%) than in pea protein isolates (65-75%) which contained small amounts of antinutritive factors and were free of pea carbohydrates.Raw peas contain pea pectins and cell walls, which increase endogenous protein secretion.Furthermore, the alpha-calactosides and pectins present in peas may stimulate bacterial fermentation.Unfavourable growing conditions may be yet another factor responsible for poor digestibility (Savage and Deo 1989).
The apparent ileal digestibility of field bean protein was lower here than the range of values (72-89%) reviewed by Gatel (1994).Reports on the apparent ileal amino acid digestibility of field beans are, however, scarce.The apparent ileal amino acid digestibilities of narrow-leafed lupins were lower than those reported by Fernández and Batterham (1995).In their trials, the apparent ileal digestibility of protein was 69-85%, that of lysine 81-82%, that of threonine 74-75% and that of methionine 81-96%.In their reviews, Gatel (1994) and van Barneveld (1999) reported apparent ileal amino acid digestibilities of 80-88% for lysine, and 62-77% for methionine and threonine.Despite their high crude fibre content, lupin seed protein and amino acids seem to be digested equally well, if not better, in the ileum than those of peas and field beans.There was no difference in the digestibilities of ileal amino acids between warm-and coldpressed rapeseed cakes, except for phenylalanine, which was higher in the former (P < 0.05).The determined apparent ileal amino acid digestibilities are within the range reported in the literature for rapeseed meal (Fan and Sauer 1995), but slightly higher than those reported for fullfat rape seeds (Fan et al. 1995).

Conclusions
The present study shows that there is little variation in the chemical composition of organically grown pea and field bean cultivars.Protein quality in terms of the concentrations of essen-Partanen, K. et al.Organically grown legume seeds and conventional rapeseed cakes for pigs tial amino acids, with the exception of arginine, decreased in peas as the protein level increased.Field bean protein contained smaller amounts of essential amino acids than did pea protein.Predicted on the basis of the in vitro enzymatic digestibility of dry matter and nitrogen, ileal nitrogen and amino acid digestibilities were similar in all pea cultivars.Predicted digestibilities were lower in field beans than in peas, particularly for sulphur-containing amino acids and energy.The apparent ileal digestibilities of lysine and threonine were similar in leafed and semileafless peas, field beans and sweet lupins.The References ileal digestibility of methionine was low in all legume seeds, particularly in field beans.The apparent ileal digestibilities of amino acids, except phenylalanine, were similar in warm-and cold-pressed rape seed cakes.

Table 1 .
Morphological features of seeds derived from pea cultivars.

Table 2 .
Composition of barley-based experimental diets in the digestibility study.

Table 3 .
Chemical composition of pea and field bean cultivars.
DM = dry matter, N = nitrogen, CP = crude protein, EE = ether extract, CF = crude fibre, SD = standard deviation.Table4.Ileal and faecal in vitro and predicted in vivo digestibilities of pea and field bean cultivars, %.

Table 5 .
Relationship between crude protein content and chemical composition and in vitro and predicted in vivo digestibilities in peas (n = 8).

Table 6 .
Analysed composition of barley and protein feedstuffs, g/kg dry matter unless stated otherwise.

Table 7 .
Apparent total tract and ileal digestibilities of grain legumes and rape seed expellers calculated by difference.