How useful is triticale for pigs and poultry?

Abstract

91 HOW USEFUL IS TRITICALE FOR PIGS AND POULTRY? R.J. MCKENZIE* and D.J. FARRELL** SUMMARY Investigations of the nutritive value of triticale for broilers and layers are reviewed. Crude protein was from 8.3 .and there was a significant negative correlation between crude and lysine content of the protein. Rat trials indicated that threonine are the limiting amino acids in triticales. Tannin triticale samples were from 0.05% to 0.10%. pigs, to 17.2% protein lys.ine and levels in . Digestibility of components in two samples of triticale was determined with pigs., Values found were similar to those in wheat. A . serie.s of substitution trials were conducted with pigs, broilers and layers. With one exception triticale replaced wheat effectively but one sample of triticale caused growth depressionin pigs. A reduction in yolk colour in eggs was seen in triticale-based diets. A taste panel test was conducted on meat of broilers grown on triticale-based diets. A slight difference in flavour was detected but there was no ,preference for or against this flavour. INTRODUCTION . Triticale is a cereal,grain which has recently gained much interest from cereal growers and stock feeders. Originally developed to combine the winter hardiness of rye and its range 'of habitat with the milling and baking qualities of wheat, it soon became apparent that triticales were frequently high in protein, lysine and other amino acids (Villegas et al. 1970). It was the prospect of a.cereal with higher protein and lysine levels than those currently in use for stock feeding and one which would not besubject to legislative controls nor be so competitive for human consumption that stimulated work to, evaluate triticale as a feedstu.ff for livestock. This man-made cereal results from an intergeneric cross between wheat (Triticum spp.) and rye (Secale cereale). Withouth the development of the chromosome doubling technique using colchicine, triticale could not have been produced sinc.e early types were sterile, as might be expected from. an intergeneric cross. However with research progr-es in many countries to develop the genetics and agronomic features of the crop; triticale now equals or out yields modern wheat varieties. It will grow in areas where many other cereals, yield poorly. In Australia, groups .at the University of'New England, at the N.S.W. : Department of Agriculture, the University of Sydney, the Waite Institute . and the Victorian Department of Agriculture are breeding and developing, strains of triticale: The purpose of the present programme .was to examine these Australian varieties and to'evaluate them as feedstuffs for, pigs and poultry. *Animal Production Centre, Hawkesbury Agricultural College, Richmond, N.S.W. 2753. *'Department o.f Biochemistry and Nutrition, University of New England, Armidale; N.S.W. 235l. . 92 'Triticale for pigs Studies to examine the substitution of triticale for various grains s in pig diets have been conducted by several workers. This work appears ' to fall'into two categories: (1) those who found that triticale can be substituted for the cereal part . of the ratjlon completely without adverse effects e.g. 'Stothess and. Shebeski (1965) using barley; Shimada et,al (1971) using* sorghum; Allee and Hines (1972a) using sorghum although not on an isonitrogenous basis; and Shimada et al. (1974) using maize. '(2) those who found some depression in feed intake, growth rate or feed conversion efficiency e.g. Stothers and Shebeski (1965) suggested that palatability problems caused a reduction in feed intake, Harrold . et al. (1971) recommended a maximum inclusion rate of 25% of the ration for growing pigs because of reduced palatability, and Ericksen et al. (.1976) who,. in a series of three experiments, demonstrated.& reduction in growth rate at high levels of substitution of triticale. for maize. They suggested that this reduction may have been due to' either a trypsin inhibitor or to poor lysine availability. . - Attempts have been made to measure the energy value of`triticale for pigs. Cornejo et al. (1913) reported the digestible energy, . metabolisable energy-) and'nitrogen-corrected metabolisable energy of triticale to ,be 15.06, 14.72 and 13.48 Xj/kg, respectively. Shimada a n Cline (1974) reported a MEvalue-of 13.67 M-j/kg,. a figure which appears * .to agree reasonably well with the former authors and which compares . favourably with other commonly used cereal grains. mipfel (1969) studied the protein efficiency ratio (PER) of several cereal grains and reported that triticale was equal to rye in PER but superior to wheat. He concluded that the superiority of triticale was due to higher contents of lysine and sulphur amino acids. Allee and Hines (1972b) ,reported that lysine was the first limiting &mino acid'in . triticale for finishing, pigs and they also suggested that the. lysine in d 1 I 1 : triticale may not be fully available to the pig. Shimada and Cline. (1974) suggest that the two limiting amino acids in triticale for the pig are . lysine and threonine in that order. .. Sauer et 21, (1974) investigated the availability of.amino acids in triticale, wm barley and soybean meal ,and. found that the limiting amino acids lysine and threonine in triticale were more available than in wheat and barley. Taverner et al. (1978) found availability values. for lysine and threonine in tritm to be similar to those of wheat, sorghum, maize and barley. These results are at variance with the suggestions of Erikson, et al. (1976) and Allee and Hines (1972b) that poor lysine . - availability may affect the performaqce of pigs fed.triticale,' . . With regard td'ititrinsic and. extrinsic toxic substances Mad1 and .'Tsen' (1974) have observed that the trypsin inhibitor . activity of triticale. .tends' to lie-between that of the wheat and rye'parents. `Nishihuta.et'al. (1978) suggest' that a toxicsubstance igolated by wa&i*water extracs . . of triticale,. which depressed rat growth, may have been a trypsin inhibitor. . These workers. dlso isolated growth-depressing substances in ether and .hexane extracts which theysuggested may have been alkyl resorcinols (see&so Radcliffe 1979). .However, Larter (1974) discounts the presence of'. .' . ' 93' resorcinols in ,triticale as a reason fdr growth depression. * Ergot (Claviceps purpurea) has been 'shown to depress both growth and feed intake (Harrold et al. 1971), and may explain growth depression in other trials (Larter 1974). The tannin level of triticale has also been implicated in growth depression (King 1978). Thus it would seem that the area of toxic factors remains rather unclear and further work in this area is needed. Triticale for broilers When substituted on a weight or energy basis triticale has been found equal or superior to wheat 'in nutritional value (Sell et al. 1962; Bragg and Sharbey, 1970; Wilson and Mcfiab, 1975). SimilarlyX substituted on an energy and nitrogen basis Rao et al. (1975) 'found that triticale had a similar nutritive value to maizeinroiler diets. However; Bixler et al. (1968) found that triticale was inferior to maize and wheat. Fernandez and McGinnis (1974) suggested that lysine and threonine are the limiting amino acids in triticale forchick growth and showed no. response to methionine and tryptophan. Marc&e2 and Avila (1973) similarly . showed no response to methionine and tryptophan or to valine but showed a . significant response. to threonine supplementation. These results are .both in disagreement with Bragg and Sharbey (1970) who showed' an improvement with methionine supplementation but not with lysine. However, this response occurred in only one diet of one experiment and could not be repeated 'in, a subsequent experiment. It was suggested that the explanation could be the two different strains of chickens .used in the two experiments. Bragg et al. (1970) investigated the effect of ergot in triticale on growth, feed efficiency and mortality in broilers. Ergot levels above 0.8% 'were found to significantly depress growth and feed efficiency and to increase mortality. Other toxic factors appear not to have been investigated with chickens. Triticale for layers Sibbald (-1977) and Farrell (1978) found the mean metabolisable energy level of several varieties of triticale for'poultry to be 15,34 and 15-30 Mj/kg, respectively-' It would seem from these preliminary .analyses that triticale should be of equivalent or better nutritive value to other cereal grains used in layer diets. In order to assess the nutritive value of triticale compared to other cereal grains, several trials have been conducted wheretriticale was substituted for the cereal component of layer diets. Guenthner and ,Carlson (1970) compared triticale, maize, wheat and sorghum inlaying diets which were formulated to be Triticale was found to depress egg isonitrogenous and isocaloric. production in both 12% and 14% protein diets, compared, to'other grains. Cuca.and Avila ,(1973) substituted triticale fdr sorghum on a protein basis .and found a depression in -egg production and egg weight. In two.experiments Weber et al. (1972) compared sorghum to wheat and triticale- In the first experiment, one of two varieties of triticale was found to give' similar egg production and .feed conversion to sorghum while` the other variety depressed performance. Both triticale varieties. depressed egg weight compared to sorghum. In the `second experiment . triticale and wheat gave sirnil& performance to sorghum. In none of these experiments were diets formulated,to take account of varying levels of essential amino acids in the grain samples. It is likely therefore, that in mo'st cases replacement of soybean protein by triticale protein may have - 94 resulted in lysine and possibly threonine deficiencies which would in turn result in depressed performance. In an experiment where diets'were formulated to make full use of the high protein content of triticale, that is, account was taken of the amino acid composition of triticale, Fernandez et al: (1973) found that hens receiving diets containing over 80% tritic=ad produdtion parameters that were equal to those of hens receiving more conventional layer diets. Although Choudhary and Netke (1976) found that the quantitative -. substituion of triticale for maize did not adversely ,affect. egg production and egg weight they suggested that the protein quality of a triticalebased diet wa's similar or inferior to that of*a maize-based diet, e possible because of a threonine deficiency. Similarly, him etal. (1976) found in two experiments that triticale was equal to maize izporting 'egg production, egg weight and body weight. In a fre'e' choice feeding experiment Karunajeewa (1978) compared ,' triticale to wheat and found that rate of lay and egg weight were similar. on either grain; Hens on triticale diets ate more*grain than those'on wheat diets and it is suggested. that the ME content of the triticale-was . lower.than ,that of the wheat used. It is interesting to note that both groups of birds ate a similar amount of the protein concentrate offered and that their total intakes of protein were-very similar (triti'cale diet 2l.Og/day, wheat diet 20.9g/day). It is unlikely, with these Drotein intakes, that either group would have suffered from an amino acid ' deficiency. In a digestibility trial with Jaying hens McNab and Shannon (1975) . concluded that triticale should *be nutritionally, superior to wheat because of its..higher ME value and digestible lysine content.. Thus, it would seem that where allowance ,is made in diet formulation for the amino acid and ME contents of triticales, they are quite capable of being substituted for'other common cereal grains in the diets of`laying hens; . In this paper we provide. a prelimin&ry report of .results obtained in experiments'designed' to investigate the following: (1) the' levels.of protein and amino acids in various samples of triticale, (2) the level of tannins'in various samples of triticale, (3) the sequence of limiting amino acids in triticale for the growing rat, (4) ' the effect .of substituting triticale for wheat in the diets `of pigs,. layers and broilers, (5) energy values of triticale for pigs'and poultry.' . MATERIALS AND METHODS Chemical analysis . Samples of ten varieties of triticale grown at four different .. locations were analysed for nitrogen using a mic'ro-Kj,eldahl method. The crud.e protein levels were obtained-by multiplying the nitrogen content by a factor of 5.76 (Tkachuk 1969)'. Sambles of thirteen different. . 95 varieties of triticale were analysed for tannin. content by the method of. Price and Butler (1977), Individual amino acids were determined on the acid hydrolysates using a TSM-1 Technician amino-acid analyser. Other procedures used in the analysis of feed and faeces followed standard methods (AOAC 1960; Van Soest and Wine 1967). Rat trials Weanling rats of the Wistar strain were housed individually in wire .cages in an air-conditioned room. The animals and feed containers were weighed at the beginning of the experiment and after 7, 10 and. 14 days. and'water were available ad libitum. The triticale sampl'.es (above) analysed for proteinwere bulked to obtain three.different crude protein levels- These mixtures were then left unsupplemented, or supplemented. with lysine, lysine + threonine, lysine + threonine and methionine or a high-quality protein source (soybean meal and fishmeal) to supply the various recommended allowances for the growing rat (N-RX. 1972)'and compared to a wheat-based diet * similarly substituted. Groups of six rats, 3 males and 3 females were allocated at random . to each treatment. ' Pig trials (i) Digestibility Four Large White x Landrace castrated males weighing . approximately 30 kg were p&aced in metabolism crates in a rodm.kept at 20 + 1�C. The animals on a commercial grower diet were allowed to adjust to the experimental conditions for.7 days. Water was, freely available from drinking nipples. The design of the experiment was a cross over with two diets containing triticale of differentcrude protein levels each fed to two pigs for a 7-day adj,ustment period and a 5-day collection period. The . diets were then crossed over and fed to the remaining two pigs for a further 7-day adjustment and 5-day collection period. The diets consisted (g/kg) of 970 triticale, 23 bone meal, 5 vitamin/ mineral premix and 2 salt. The crude. protein (N x 5.76 on a DM basis) of the two diets was 139 and 154 g/kg`. Pigs were fed once daily at a level of 3% of average live-weight, the fe&d being mixed with an equal weight, of' water. Any,feed remaining one hour after feeding was removed and dried-to constantweight in a forced-draft oven at 750C. At the commencement of each col&ection period, 10 g of ferric oxice was mixed with 100 g of dry feed and given to each pig followed later by the remainder of the ration. This was repeated 5 days later. Collection of faeces was commenced and ceased .upon the appearance.of the ferric oxide. . Faeces were collected and dried to constant weight and samples stored for analysis. Total urine output was collected daily in 10 ml of concentrated H2SC4. A 10% aliquot was accumulated daily and stored at .- 10�C for chemical analysis. (ii) Substitution trials Three experiments were conducted at Hawkesburyand one at Armidale, to examine the effect of substituting triticale I .for 96' wheat in pig diets. The first of'these involved pigs from weaning to approximately 20 kg live weight, the remaining two from approximately 45 '. 'kg to 75 kg.. In experiment I, forty two mixed sex, Large White x Landrace were allocated to three -treatments rep.licated twice. Pigs weighing 4.0 to 6.0 kg were allocated at random within one replicate and pigs 6.0 to 8.8 kg to the second replicate. Each group.of seven pigs was housed in a weaner cage in a forced ventilation weaner shed. Feed water were available ad libitum'. The composition of the diets used Table 1. pigs from from and ,' is in Diets were formulated to be isonitrogenous and isocaloric with triticale supplying equivalent lysine levels to wheat .in diet 3. Pigs were individually weighed at weekly intervals and feed consum&on was recorded on a pen b&is. IThe duration of the trial was 47 days. .TABLE 1 Composition of pig. diets (g/kq)used in experiment 1 . In experiment 2, twenty four entire male Large White x Landrace pigs .of approximately 45.kg body weight were randomly allocated to four treatments. 'Pigs were ,fed in individual feeding stalls and were housed in groups of six in a naturally'ventilated grower house. Feed was offered once daily mixed with an equivalent weight of water and all'feed not consumed within one.hour was weighed back. The composition of the diets used in shown is Table 2. Diets were formulated to contain equal energy,' lysine, methionine and.threonine levels. Diets 2, '3 and 4 contained 33,3, 66.7 and 100% of the lysine level'of,diet 1 contributed by triticale (var. Quic,kgro, -9% CP)'. . 97 Pigs were weighed at 14 days intervals and feed allowances were adjusted accordingly based.on the ARC semi -ad lib scale (ARC 1967). The trial was concluded when the average live weight of all pigs reached 70 kg when backfat thickness was measured using a 'Meritronics' backfat ,tester. . TABLE 4 Ingredient (g/kg) and chemical composition of experimental diets offered to five groups each of six pigs in experiment 4 98 Experiment 3, shown in Table 3; was conducted in similar fashion to experiment 2 using a different tritical'e (var.. AT6, 12.5% CP). Experiment 4 was conducted at Armidale. Thirty Large White x . . Landrace entire male pigs (19 kg) were placed in individual pen&n a temperature-controlled (22OC) piggery.' Following a short period of adjustment five pigs were allocated to each of the five diets (Table 4); these were by calculation isocaloric and isonitrogenous. Restricted amounts of each diet were offered once daily, using the formula Pigs were weighed each week and their daily feed allowance adjusted accordingly. The experiment was terminated when each pig weighed 45 kg, and backfat thickness was then measured using an ultrasonic instrument (Scanoprobe, Ithaca). ' Apparent digestibility of feed 'dry matter, nitrogen and energy.of each pig diet was determined using chromic oxide (0.1%) in the feed and its concentration measured incorresponding faeces to determine. faecal output. Layer trial . Fifty White Leghornx Australorp hens of' the Hyline strain aged 31 weeks with an average liveweight of 1.99 kg were allocated at random to single bird laying cages in an open sided gable roofed shed. .Pens were then allocated .at random to one of the five treatments. Birds had access * to feed (in individual troughs) and water at all times and were on a 16 hour lighting programme. The compositio,n of the diets used in shown in Diets were formulated to contain equal amounts of energy, lysine, Table.5. methionine and threonine. TABm 5 Composition of diets, (g/kg) in the lay.i.ng trial The birds were weighed at the beginning o-f the experiment, at weeks of age, and at the end of the experiment at 41 weeks of age., production of each'hen was recorded daily and feed consumption was recorded over 14 days. During the last 14 days of the experiment, from each hen were collected and used to determine average weight, colour, specific gravity and albumen quality. . 31 EgGI 5 eggs yolk 99 Broiler trials . ' A preliminary experiment wasconducted in,battery brooder cages to examine the substitution of two samples of triticale for wheatFour replicate.groups of eight treatments, each containing six male chicks were housed in battery brooder cages at one week of age.and received the test diets for 15 days. Weight gains and food conversion . ratios were measured for the 15 day trial period. The diets used and their calculated composition are presented in Table 6i These diets were formulated on a Qeleq analog computer. TABLE 6. Composition of diets'(g/kg) used in chick trial (air-dry basis) . 100 thousand mixed sex broiler .chickens were allocated at random to 20 pens . in a conventional style broiler shed. These. birds were fed on a commercial broiler starter ration for 28 days and the pens were randomly allocated to four treatments given 5 replicates of each treatment. The birds were..weighed ,at this stage and feeding of the experimental diets was commenced. The birds were again weighed at the conclusion of the -. trial, 25 days, later. Feed consumption was recorded.for this period. The composition of. the diets used is shown in Table 7.' 'Diets were.formulated to contain equal amounts of energy, lysine and methionine. . A taste panel test was conduced on representative birds from each treatment to determine any differences in flavour of cooked meat from any ' of the treatments. Twenty six untrained .subjects were offered coded samples of meat cooked under i'dentical conditions from.each of the treatments. They were. asked to determine the presence of different' flavours between the samples and the level of these differences on a graduated scale. They were also asked to rank the samples in order of preference for flavour. RESULTS ,Chemical analysis . The crude protein values obtained for the triticale samples are shown in Table 8 and the amino acid analysis of selected sample& is shown in Table 9. .There was a significant negative correlation between lysine content of protein and crude protein level (r. = -0.70). _ The values for tannin content in the triticale samples analysed 'are `shown.in Table 10. 101 TABLE 9 Amino acid composition of 'triticale samples ('air-dry basis) grams of .amino acid p.er 100 grams of sample .Rat trial 'The mean values obtained for 14 .day weight gain and feed conversion ratio in the rat grmth trial are shown -in Tab1e.s 11 and 12 respectively, The addition of lysine to all diets increased,growth rate and improved feed efficie.ncy,,.significantly (PiO.05) in the case of growth rate on the low protein tri,ticale and wheat, and in .all'cases of feed Similarly the addition of threonine resulted in significant efficiency.. improvements in growth rate in all but the highest protein triticale diet with a similarbut not significant trend in feed efficiency. The use of soybean meal and fishmeal as a source of essential amino acids gave no further improvement.in performance over the triticale diets supplemented with lysine and threonine. There is a suggestion from the feed . efficiency data that the addition of methionine to a low protein triticale diet'may give some improvement. 102 . TABLE 11 Mean 14 day weight gain (g) of rats fed diets based on triticales'of three protein levels and wheat TABLE 12 Mean feed conversion ratios of rats fed diets based on triticales of diffe'rent protein levels and wheat . Pig trials (i) Digestibility trial The results obtained from the digestibility : trial are shown'in Table 13. Each value is the mean of four individual pig values. There appears to'be little difference between the values obtained for the two triticale sampkes. 103 TABLE 13 Means and standard errors of digestibility parameters for pigs fed die.ts based on triticale (ii) Substitution trials The mean daily weight gains and group'feed conversion ratios of pigs obtained inexperiment 1 are shown ,in Table 14. TABLE 14 Mean daily weight gain (g/day) and feed conversion ratio (FCR) of pigs (Experiment 1) There were no significant (P>O.O5) treatment difference between diets when pigs (4-8 kg) were grown for 47 days on diets shown in Table 1. The mean daily weight gains, feed conversion. ratios and backfat thickness obtained for experiment 2 are'shown in Table 15. TABLE '15 Mean daily liveweight gain (g/day), feed conversion ratios (FCR)' and b ackfat thickness (BF mm) of groups of pigs grown from 45 to 70 kg (Experiment 2): 104 A. significant (PcO.05) depression in growth occurred on the two diets (3 and.4) containing high levels of triticale. This' trend also appeared in FCR but it was not significant. The'values obtained for daily weight gain, feed conversion ratio and backfat thickness from experiment 3 are shown in Table 16. TABLE16 Mean daily liveweight gain'(g/day), feed conversion ratios (FCR) and backfat thickness (mm) of pigs grown from 45 to 70 kg (Experiment 3) There were, no differences in growth and FCR between wheat- and triticale-based diets (1 and 2) thereby indicating that the variety of triticale (var. Quickgro) used was equal to that of wheat in nutritive value (Table 17). Analysis of diets (Table 4) showed them to be ' isonitrogenous butnot isocaloric as calculated.. Digestible energy . contents of diets 1 and 2 were higher than the remainder. Because pigs on all treatments were*on the same feeding scale it was not surprising that differences in performance occurred. Had pigs on diets 3 and 4 been given amounts of feed commensurate with the actual digestible energy contents of the diets differences in performance between groups did not necessarily reflect differences in protein quality of the diets. TABLE.17' Growth rate, feed conversion ratio and backfat thickness of groupsof 6 pigs per treatment grown from 20-45 kg Layer trialsThe results obtained for hen.day production (%), daily feed intake per hen (g/day), egg weight (g), bodyweight change (kg), specific gravity, Haugh Index and.yolk colour are shown in Table.'l8. 105 TABLE 18 Mean hen day production, daily feed intake, egg weight, body weight changet specific gravity, 'Haugh index and yolk colour There were no significant differences.in any parameters exceptyolk colour. There was a significant depression,of yolk colour at high levels of triticale substitution. Broiler trials (i) Chick trial The values obtained for weight gains and feed conversion ratios are shown in Table 19. TABLE 19 Mean weight gain (g) and .feed conversion ratio of broiler chicks Qrown from 7 to 22 days . . (ii) Broiler finisher trial The values obtained 'for weight gains and feed conversion ratios are shown in Table 20. TABLE 20 .Mean weight gain (g) and feed conversion ratios of broiler . chickens from 28 to. 43 days 106 (iii) Taste panel Tables 21 and 22. TABLE 21 The results of the taste panel test are shown in . Test for differences in flavour, comparing broiler meat from treatment 1 (all wheat) to meat from triticale diets (numbers of respondents) TABLE 22 Flavour preferences comparing meat from treatment 1 (all wheat) to meat from triticale diets There were some differences in flavour between the samples of ,meat but there was no preference for the flavour of any meat. DISCUSSION The results of the chemical analysis indicate a wide variation in ,crude protein content betweenvarieties and in all samples tested; a range of 8.3% to 17.2% was obtained. This is a similar range to that found by Villegas et aid (1970) working on Mexican varieties. Similarly with lysine content of the 'protein, the above workers found a significant negative correlation between lysine content and crude protein level. However, it does appear that the lysine content of the prote.in is higher than that of'wheat which have a reported maximum lysine level of 3.8 g/16 g N (Ivan 1974). Variation in these factors therefore makes it difficult . to have available standard values which can be used for feed formulation. The results of tannin estimations indicate relatively low levels of tannin present in all samples tested with only one variety (321176) exceeding 0.1%. Since a-level of approximately 0.5% is required before a growthdepressionoccurs in chicks (Vohra et al. 1966), it appears ,unlikely that tannins in the samples tested would cauSe growth depression. The data obtained from the rat growth study seem to confirm the conclusions of Shimada and Cline (1974) to the effect thatlysine and threonine are the essential amino acidslimiting to rat growth in triticale. There is also a suggestion that supplemental methionine may benefit in triticales of relatively low protein content. These results also indicate that essential amino acids other than lysine and threonine are in sufficient supply in triticales to support adequate growth since no further improvements occurred when a high quality protein source (fishmeal 107 . and soybean meal) was used to supplement the protein of the diet. The ME values for triticale obtained from the,pig measurements are somewhat higher than those of.Cornejo et al. (1973) and Shimada and Cline (1974) although the'digestible energy values are in closer agreement. The ME values however, do agree quite closely with those of Farrell (1978) using Australian varieties of triticale in chickens. It may be that the more.recent varieties of triticale have higher energy values than the earlier varieties. It was also apparent that there was very little differences in energy values for ,the two samples tested. The relatively low values obtained for nitrogen retention are probably due to amino `acid imbalance arising from the use of complete grain diets. The data obtained from the substitution experiments indicate that for pigs from weaning to ,20 .kg liveweight, and from 20 to 45 kg liveweight the triticales used in experiments 1 and 4 respectively can effectively substitute for wheat on a protein and energy basis without affecting growth rate and feed efficiency. However, because of the small scale of the experiments and the fact that only two samples of triticale were tested' it is not possible to make any general feeding recommendations. Similarly, the data from experiments 2 and 3 appear to conflict since severe growth depression occurred in experiment 2 at levels of triticale inclusion above 48% of the total diet, whereas no growth depression was observed in experiments 3 and 4. Thus it would seem that there may be some factor present in thesample of triticale used in experiment 2 which led to growth depression. Since daily feed allowances were set according to Iiveweight, pigs fed on the higher triticale diets ate less due to lighter body weights. However, there was no rejection of feed by pigs on these diets which suggests that acceptability was not a problem in this case, as has been suggested by Harrold et al. (1971). King (personal communication 1980)` has preliminary resxwhich indicate that a trypsin inhibitor is probably involved. It would be desirable to establish if there is a growth inhibitor in triticale and to define it in order that firm recommendations may.be made regarding inclusion level of this grain in pig diets. The results from the layer trial show that even when there was 100% substitution for wheat, the' sample of triti'cale used here did not'affect laying performance. It is interesting to note that depression of yolk. colour does occur at ievels of inclusion above 59% of the total diet. This observation has also been reported by Chondhary and Netke (1976) and Karunajeewa (1978); the latter author suggested an unknown factor inhibiting the utilization .of..xanthophyll pigments. Since the level of depression occurring in the present experiment was only 0.8 of one unit on the Roche Colour Fan, this 'is likely only to be a problem in cases where producers are just above the yolk .colour level set by marketing . . authorities. _The results for both broiler trials indicate that the substitution of triticale for wheat in broile? rations does not depress growth or feed efficiency. The relatively poor growth rates and feedconversion ratios obtained in experiment 2 can probably be attributed to the combined 'effects'of usin.g a mash-type