Agricultural and Food Science, Vol. 15 (2006): 293–323. 293 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. © Agricultural and Food Science Manuscript received June 2006 Review article Recent developments in forage evaluation with  special reference to practical applications Pekka Huhtanen MTT Agrifood Research Finland, Animal Production Research, FI-31600 Jokioinen, Finland, e-mail: pekka.huhtanen@mtt.fi Juha Nousiainen Valio Ltd, Farm Services, PO Box 10, FI-00039 Valio, Finland Marketta Rinne MTT Agrifood Research Finland, Animal Production Research, FI-31600 Jokioinen, Finland The present re-evaluation of a dataset of systematically collected laboratory analyses and in vivo digestibil- ity information for several types of silages gives convincing evidence of the biological weaknesses of feed characterisation based on the proximate feed analysis. The problems include intrinsic failures of the analy- sis in describing cause-response relationships between forage composition and digestibility, and heavy de- pendency of the equations on forage specific and environmental factors. It is concluded that proximate analysis is not suitable for characterisation of neither forages nor concentrate feedstuffs. In vitro pepsin-cel- lulase solubility of organic matter (OMS) and concentration of indigestible neutral detergent fibre (iNDF) predicted forage organic matter digestibility (OMD) with an acceptable accuracy for practical feed evalua- tion purposes provided that forage type dependent correction equations were employed. The revised detergent system dividing forage dry matter (DM) into almost completely available neutral detergent solubles (NDS), and insoluble residue (neutral detergent fibre, NDF) shows potential for future development. The combined use of long-term in situ ruminal incubation and NDF fractionation can be used to divide forage DM into three biologically meaningful fractions: NDS, iNDF and potentially digestible NDF (pdNDF). The summative models can then be used to predict forage D-value, i.e. apparently digestible organic matter in forage (g kg-1 DM). The models sum digestible NDS, which can be determined by Lucas equation, and digestible NDF (dNDF), which is the amount of pdNDF that is actually digested during any specific fermentation or retention time. Forage type specific summative models were as good as regression equations based on OMS or iNDF in predicting forage D-value and general summative models gave better results than general equations based on iNDF and especially OMS. If the goal is to reduce prediction error of D-value below 15 g kg-1 DM, forage type specific prediction equations should be used regardless of whether they are based on OMS, iNDF or summative models. An- other option in the future may be dynamic models, which can incorporate simultaneously the two important dynamic processes constraining feed digestion in ruminants: the rates of NDF passage and degradation (kd). 294 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation However, a vital prerequisite to employ dynamic models in practical feed evaluation is that iNDF and kd can be easily and reliably determined from on-farm forages. Although a NIRS prediction equation for iNDF will be adopted in practical use in the near future in Finland, the methodology for estimating kd warrants further research. Key words: silage, prediction, cell wall quality, digestibility, near infrared reflectance spectroscopy Introduction The main objective of feed evaluation techniques is to predict the availability of nutrients and feed- ing value of feeds for animal production systems. The methods available include chemical analy- sis, in vitro digestibility with rumen bacteria or enzymes, in situ incubation in nylon bags and near infrared reflectance spectroscopy (NIRS). Feed evaluation of forages is more important than that of concentrate feedstuffs due to the large variation in the nutritive value of forages and the large contribution of forage to total diet dry matter (DM) compared to individual concen- trate ingredients. In addition to direct influence of forage quality on nutrient digestibility in ru- minant diets, it also indirectly affects total nutri- ent supply, because of the large impact of both digestibility and silage fermentation characteris- tics on silage DM intake (Rinne 2000, Huhtanen et al. 2002). Accurate estimation of forage digestibility is a prerequisite for diet formulation, economic evalu- ation of forages and prediction of animal respons- es. Determination of in vivo digestibility is time consuming and expensive for routine and even re- search use; therefore, different biological and chemical laboratory methods have been developed to estimate digestibility of forages. Advantages and disadvantages of different methods have been discussed in detail in many recent reviews (Steg et al. 1990, Weiss 1994, Coleman et al. 1999, Beever and Mould 2000, Cherney and Cherney 2003). From these reviews, it can be concluded that bio- logical laboratory methods seldom estimate di- gestibility values directly. This does not mean that these methods are not useful, because they often have close empirical relationships to in vivo di- gestibility, but it does imply that empirical correc- tion equations are required for estimating in vivo digestibility from in vitro and in situ measure- ments. These correction equations are often spe- cific for different forages, environments and even laboratories (Weiss 1994, Van Soest 1994, Nou- siainen 2004). The relationships between chemical and even biological measurements to digestibility can be markedly different for the main grass species used for silage in Finland, i.e. timothy (Phleum prat- ense) and meadow fescue (Festuca pratensis), to those estimated elsewhere. Temperature and light intensity influence lignification of the cell wall (Deinum et al. 1968, Van Soest 1994), which af- fects the relationship between fibre and digestibil- ity. Grasses grown in northern latitudes had higher digestibility at the same stage of maturity than those grown at latitudes closer to the equator (Dei- num et al. 1968). In Finland, a data set has been compiled of sys- tematically collected information on grass and le- guminous silages to evaluate laboratory methods for predicting in vivo digestibility of forages with the final aim to develop a rapid, accurate and pre- cise method based on NIRS for analysing farm samples. A series of papers has been published from this work (Nousiainen et al. 2003a, b, 2004, Nousiainen 2004, Huhtanen et al. 2005, Rinne et al. 2006). The objectives of this paper are to (1) re-evaluate and discuss the methods routinely used for forage analysis, (2) describe alternative meth- ods of predicting digestibility using regression equations and summative models, (3) present the sources of variation in estimating digestibility and (4) make implications of the data for developing practical analysis for farm samples. 295 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. Description of data The dataset used in this analysis consisted of infor- mation for Finnish silages with in vivo digestibility determined in sheep fed at approximately mainte- nance level of feeding using total faecal collection method. The silages were harvested over 9 years in 1994–2003 from mixed timothy (Phleum Prat- ense) meadow fescue (Festuca pratensis) leys in primary growth (PG, n = 33) and in regrowth (RG, n = 27) and ensiled with formic acid based addi- tives in pilot-scale tower silos or farm-scale bunker silos. The digestibility of both PG and RG silages was varied by systematically changing harvesting date. The silages are described in Nousiainen et al. (2003a, b) with a few added observations. The data set of pure legume silages including red clover (Trifolium pratense, n = 15) and galega (Galega orientalis, n = 4) is described by Rinne et al. (2006) with the exception that only feeds of Finnish origin are used in this analysis. Further, data from whole- crop silages prepared from barley (Hordeum vul- gare, n = 5) and wheat (Triticum aestivum, n = 2) were included in the data set. Characteristics of the silages used are presented in Table 1. Chemical methods in forage  characterisation Proximate analysis The proximate feed analysis has been in use for more than 100 years. The following components of DM are analysed: ash, crude protein [CP = ni- Table 1. Description of ash, crude protein (CP), neutral detergent solubles (NDS), neutral detergent fibre (NDF), lignin and indigestible NDF (iNDF) concentrations, organic matter pepsin-cellulase solubility (OMS) and in vivo digestibility of organic matter (OMD) and NDF (NDFD) of different forage types. In dry matter, g kg-1 Ash CP NDS NDF iNDF Lignina OMS OMD NDFD Primary growth grass (n = 33) Mean 72 148 357 568 79 32 757 0.733 0.739 Standard deviation 8.2 35.0 67.1 70.1 39.1 10.5 70.5 0.0606 0.0701 Regrowth grass (n = 27) Mean 94 144 376 533 106 28 757 0.694 0.701 Standard deviation 8.5 25.4 27.8 34.0 28.2 4.8 27.8 0.0339 0.0493 Legume (n = 19) Mean 99 211 532 369 109 43 754 0.707 0.627 Standard deviation 14.3 38.1 70.3 75.6 52.5 18.6 65.5 0.0623 0.0815 Whole crop (n = 7) Mean 74 114 495 432 119 27 758 0.686 0.515 Standard deviation 10.7 7.2 71.3 63.2 30.6 8.1 69.3 0.0471 0.0300 All (n = 86) Mean 85 158 413 502 97 33 757 0.711 0.684 Standard deviation 15.6 43.0 92.9 100.3 41.2 12.8 58.0 0.0552 0.0919 Minimum 49 79 265 274 17 17 628 0.581 0.477 Maximum 122 301 627 669 211 79 878 0.840 0.869 a Analysed as permanganate lignin (Robertson and Van Soest 1981); n = 31 and n = 84 for primary growth grass and all silages, respectively. 296 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation trogen (N) × 6.25], ether extract (EE) and crude fibre (CF). The nitrogen free extract (NFE) is cal- culated as: NFE = organic matter (OM) – (CP + EE + CF) [1][1] It is typically assumed that CF is the least di- gestible fraction of feed fibre and that NFE repre- sents the highly digestible carbohydrates. Howev- er, this assumption is often not correct, especially for forages because a large proportion of indigest- ible lignin and hemicellulose is solubilised during CF extraction (see Van Soest 1994). Consequently, forage NFE is comprised of components that vary from completely unavailable lignin to completely available fractions such as soluble carbohydrates and organic acids. Apparent digestibility of NFE was less than that of CF in a large number of cases (Van Soest 1975). In the data set of 52 grass silages (Nousiai- nen et al. 2004), the digestibility of CF was higher than that of NFE in 31 cases. The difference be- tween CF and NFE digestibility decreased (P < 0.01) with advancing maturity of grass ensiled, i.e. the earlier the grass was harvested, the greater the difference in the digestibility of CF and NFE. Heterogeneous availability of grass NFE frac- tion can be demonstrated by the Lucas test (see Van Soest 1994). The purpose of the Lucas test is to identify ideal nutritional entities that have uni- form digestibility over a wide range of feedstuffs by plotting the digestible nutrient concentration in DM against the nutrient concentration in DM. The slope of regression estimates the true digestibility and the intercept is an estimate of the metabolic and endogenous faecal matter (M) for the nutrient, which consists of unabsorbed digestive juices, mi- crobial debris from the rumen and microbial cells from the hindgut fermentation. The true digestibil- ity of silage NFE estimated by the Lucas test had a high standard error (±0.15 units of digestibility) and positive intercept. A positive intercept is not biologically possible, because at zero concentra- tion of a nutrient there cannot be a positive amount digested. The variable true digestibility of NFE indicates that it is not an ideal nutritional entity, which is not surprising considering that forage NFE fraction contains a range of chemical components differing in their availability. Consistent with this, silage NFE concentration had no correlation to OM di- gestibility (OMD). Faecal NFE output as grams per kg DM intake was closely related to lignin concentration (faecal NFE = 46.8±9.8 + 2.24±0.31 × Lignin, residual mean squared error (RMSE) = 18.8, R2 = 0.51). The regression coefficient of lignin suggests that one gram lignin protected 2.24 g of carbohydrates recovered as NFE fraction in faeces, most likely hemicellulose. The intercept of regression may be interpreted as the metabolic and endogenous faecal component resulting from the error of using factor 6.25 for N to calculate faecal CP (more detailed discussion later). Crude fibre also cannot be regarded as an ideal nutritional entity, because the Lucas test showed a significant (P < 0.01) positive intercept (112±19.8), and a variable and low slope (0.36±0.065) as an estimate of true digestibility. Although CF and NFE had significant positive correlations in the Lucas test, they are not ideal nutritional entities because their slopes were variable and intercepts were positive. Of the proximate analyses, only CP and EE behave as ideal nutritional entities and they typically comprise <0.20 of OM in forages. At- tempts to identify a larger ideal nutritional fraction using the proximate analysis, e.g. CF-free OM (OM – CF), were unsuccessful because the com- bined fraction of CP, EE and NFE was non-ideal due to the impact of variability in NFE among for- ages (Fig. 1). y = 1.70x - 611 R 2 = 0.852 300 350 400 450 500 550 600 550 580 610 640 670 700 OM-CF (g kg-1 DM) Digestible (OM-CF) (g kg-1 DM) Fig. 1. The uniformity of organic matter (OM) minus crude fibre (CF) concentration determined with the Lucas test; data comprising of silages made from primary growth and regrowth grass. 297 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. Detergent analysis The fundamental problems associated with NFE and CF fractions in the proximate feed analysis were realised by Paloheimo (1953), who initiated research to develop improved analytical methods for plant cell wall. In the pioneering work, Palo- heimo and co-workers (Paloheimo and Paloheimo 1949, Paloheimo and Vainio 1965) used weak hy- drochloric acid and a two-stage ethanol extraction to remove cellular contents and to describe vege- table fibre. Despite the appropriate criticism against fractionating feed carbohydrates into CF and NFE, these methods were too laborious, not applicable to faecal samples and the fibre residue was contaminated with protein. Based on these ideas, Van Soest (Van Soest 1967, Van Soest and Wine 1967) developed the neutral detergent frac- tionation, which used detergents to remove pro- tein and isolate dietary fibre easily in feeds and faeces. Neutral detergent fibre (i.e. NDF) is widely ac- cepted as an estimate of forage cell wall content, with the major exception that cell wall pectin is extracted. This means that the neutral detergent solubles (NDS) defined as: NDS = DM – NDF [2] contains ash, sugars, starch, organic acids, soluble proteins and lipids and also soluble cell wall car- bohydrates like β-glucans and pectin, but they are readily degraded in the rumen (Van Soest 1994). Because ash contributes no energy to the animal, ash can further be subtracted from NDS resulting in neutral detergent soluble OM. Later in this pa- per, NDS refers to ash free neutral detergent solu- bles. The original method (Van Soest and Wine 1967) was modified by Robertson and Van Soest (1981) and Van Soest et al. (1991) by including the use of a heat-stable amylase to remove starch, but they removed sodium sulphite to minimize the losses of phenolic compounds, which isolated a fraction they called neutral detergent residue. The official method approved by AOAC (Mertens 2002a) uses both heat-stable amylase and sodium sulphite. Results can be calculated in four different ways in the official NDF method (with ash, with ash and blank corrected, ash-free, and ash-free and blank corrected). The effects of blank correction are minimal (Mertens 2002a), but especially for forage samples, ash-free values are lower. To avoid confusion, it is important to describe in detail how the NDF analysis was conducted. In the present work, NDF was analysed without the use of amy- lase except for the seven whole-crop silages, but with sodium sulfite and measured ash-free without blank correction. Neutral detergent divides the feeds into a solu- ble fraction that is rapidly and almost completely available and a fibre fraction that is slowly and in- completely degraded by microbial enzymes. The neutral detergent soluble fraction has a high and relatively constant true digestibility across most feeds, which indicates that it is an ideal nutritional entity (Van Soest 1994). When a wide range of feeds (n = 504) was evaluated, Weisbjerg et al. (2004) reported a complete true digestibility for NDS fraction and an endogenous faecal output of 90.2 g kg-1 DM intake. In the present data, the true digestibility (0.963) of NDS for all silages was sig- nificantly (P = 0.04) different from unity. There were some differences between the forage types both in the estimated endogenous faecal output and the true digestibility of NDS (Table 2). The RMSE of the Lucas test was higher for all data compared with that of each forage type. The inter- cept was lower (P < 0.01) for whole-crop silages than the overall intercept, whereas the slopes for PG and whole-crop silages were higher (P < 0.01) and that of RG tended (P < 0.06) to be lower than the overall slope. The true digestibility of NDS was higher for PG silages compared with RG si- lages (1.015 vs. 0.925; P < 0.01). This difference may be explained by higher content of substances, such as waxes and cutins, in the NDS of RG com- pared to PG grass that have low availability in vivo (see Van Soest 1994). The true NDS digestibility above unity for whole-crop silages may partly be attributed to a small number of samples, but it may also be associated with reduced endogenous out- put with increased NDS concentration. Nitrogen content in faecal NDS was estimated by regression to be 72 g kg-1 (Fig. 2), a value simi- 298 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation lar to that reported by Van Soest (1994). Although there were statistically significant (P < 0.01) dif- ferences between the forage types in the faecal N to NDS ratio, the numerical differences were rela- tively small (66.2, 63.4, 74.7 and 78.7 g N kg-1 NDS for the PG and RG grass, legume and whole- crop silages, respectively). The differences in this ratio may be associated with the hind-gut fermen- tation (higher N concentration in intact microbial cells than in partially digested microbial cell walls) and variation in faecal N output of feed origin (leg- umes). The high true digestibility of NDS and close relationship between faecal N and NDS sug- gests that most of faecal NDS and N are of micro- bial and endogenous origin. Markedly lower N concentration in faecal endogenous OM than in feed protein (70 vs. 160 g kg-1 DM) results in a large error in calculating faecal NFE concentra- tion. The low N concentration in metabolic and en- dogenous faecal OM creates errors for the calcula- tion of endogenous losses for CP, which should be 14 × N instead of 6.25 × N for any fraction calcu- lated using CP. For example faecal NFE concen- tration is calculated as: Faecal NFE = OM – CF – CP (6.25 × N) – EE [3] Because the faecal NFE uses factor 6.25 to cal- culate faecal CP instead of the factor 14 based on the true N content of faecal endogenous OM, prox- imate analysis system results in an erroneously high concentration of NFE in faeces, which is sup- posedly of non-structural carbohydrate origin. However, faecal NFE contains significant undi- gested plant cell wall components, primarily hemi- cellulose and lignin, as indicated by the relation- ship between faecal NFE and lignin. The propor- tion of faecal NFE that is undigested cell wall can be calculated as: Faecal NFE (Cell wall) = 349±19.1 – 39 ± 26.6 × OMD [4] On average, 0.63 of the faecal NFE was endog- enous matter, which is overestimated because the traditional coefficient of 6.25 × N was used, and 0.37 undigested NDF. In the data of Van Soest (1994), the proportion of cell wall fraction in fae- cal NFE was more than half, which may be related Table 2. Faecal metabolic output (intercept, g kg-1 DM) and true digestibility (slope) of the neutral detergent solubles fraction (NDS) of different forage types by regressing intake of NDS against apparently digestible NDS. Forage n Intercept s.e.a Slope s.e. RMSEb Adj. R2 All 86 –92 7.4 0.963 0.018 15.1 0.972 Primary growth grass 33 –101 6.0 1.015 0.017 6.3 0.991 Regrowth grass 27 –90 10.2 0.925 0.027 4.1 0.978 Legume 19 –101 18.0 0.962 0.034 10.0 0.979 Whole-crop 7 –136 4.0 1.111 0.008 1.4 1.000 a Standard error b Residual mean squared error Adj = 0.072x - 1.5 R2 = 0.964 y = 0.069x - 0.3 R 2 = 0.801 10 15 20 25 30 35 40 200 250 300 350 400 450 500 550 Feacal NDS (g kg -1 DM) Faecal N (g kg -1 DM) Unadj. Adj. Fig. 2. The relationship between faecal neutral detergent solubles (NDS) and faecal nitrogen (N) concentrations es- timated by single regression analysis and by a mixed mod- el regression with random study effect of all forage types. 299 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. to lower digestibility of the grasses in that data set. In the present data, the concentration of the cell wall fraction in faecal NFE (g kg-1 DM intake) was strongly correlated (RMSE = 10.7; R2 = 0.808) to the apparent OMD of the silage. Analogous to NFE in the proximate analysis system a fraction can be calculated by difference in the detergent analysis that represents soluble or non-fibre carbohydrates. Because non-structural carbohydrates are typically determined analytical- ly as starch plus sugars, this fraction is commonly called non-fibre carbohydrate (NFC) or neutral de- tergent soluble carbohydrate to indicate that it is calculated from fibre analysis: NFC = OM – NDF – CP (6.25 × N) – EE [5] The true digestibility of NFC from grass silag- es (0.96±0.026) was not significantly different from unity (P = 0.11). When the PG and RG si- lages were analysed separately, the true digestibil- ity was 1.03 for both silages, but the intercept was more negative for the regrowth silages (–54 vs. –42 g kg-1 DM). These estimates of endogenous loss of NFC are erroneously high because the N correc- tion factor for faecal NFC should be 14 instead of 6.25. Theoretically, the endogenous losses of NFC should be zero because there is little carbohydrate in endogenous animal secretions and microbial de- bris. Faecal NFC (OM – NDF – CP (6.25 × N) – EE) averaged 139 g kg-1 faecal DM (or 43 g kg-1 DM intake) for 52 grass silages. The value of 43 g kg-1 DM intake for the apparent faecal output of NFC is close to the intercept of the regression be- tween lignin concentration and faecal NFE output, i.e. faecal NFE that is not related to dietary cell wall fraction. The detergent system provides conceptually sound basis for understanding the physical and bio- chemical factors that influence the digestibility of feed fractions and causal relationships behind di- gestibility. If the feed fraction has a true digestibil- ity close to unity and it behaves uniformly among feed types, the faecal output per kg DM intake should not be related to dietary concentration of the fraction or OMD. With proximate analysis, only relatively small proportion of forage OM (CP and EE) behaves uniformly compared with the NDS fraction in the detergent system. For exam- ple, in the primary growth silages (n = 27), both the dietary concentration (362 vs. 196 g kg-1 DM) and faecal output (94.8 vs. 56.5 g kg-1 DM intake) of uniformly behaving entities were markedly greater when based on the detergent system. The faecal output of NFE, CF and NDF decreased with increasing digestibility (Fig. 3), whereas faecal output of CP and NDS were not related to diet di- gestibility or dietary concentrations. The relation- ship was stronger for NDF (R2 = 0.993) than for 0 50 100 150 200 250 0.60 0.65 0.70 0.75 0.80 0.85 OM digestibility Faecal output (g kg -1 DMI) CP+EE SCHO NDS NDF NFE CF Fig. 3. The relationships between organic matter (OM) digestibility and faecal output of feed compo- nents per kg DM intake. Data from primary growth silages (n = 27). CP = crude protein, EE = ether extract, SCHO = soluble carbohydrates, NDS = neutral de- tergent solubles, NDF = neutral detergent fibre, NFE = nitrogen free extract, CF = crude fibre. 300 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation NFE and CF (R2 = 0.926 and 0.923, respective- ly). The advantages of the detergent system com- pared with the proximate analysis are: larger frac- tion which behaves uniformly (1), one instead of two fraction of which faecal output varies with di- gestibility (2) and a closer relationship of faecal fibre (CF vs. NDF) output to OMD (3). These analyses confirm the statement of Paloheimo et al. (1968), that dividing feed carbohydrate fraction into NFE and CF has no scientific justification and limited biological utility; therefore the use of NFE to evaluate forages should be ended. In spite of the limitations of CF analysis to describe the plant cell wall fraction and calculating the more easily avail- able carbohydrate fraction as a difference, proxi- mate analysis is still the basis of calculating feed energy values in most current feed evaluation sys- tems in Europe. Methods to estimate forage  digestibility Empirical relationships Chemical composition Much effort has been directed toward developing regression equations that relate various chemical components to digestibility, although these at- tempts have not been very successful because of large interspecies and environmental variation (Van Soest 1994). In the present work, the relation- ships between selected chemical parameters [CP, NDF, acid detergent fibre (ADF) and lignin] and OMD by regression analysis were evaluated as a reference for comparison to more biologically based models. Statistical significance of the pre- diction errors between the forage types was tested by one-way analysis of variance using GLM pro- cedure of SAS (SAS 1999). Because most prediction errors between forage types were significant, the next step was to esti- mate prediction accuracy of regression equations based on forage specific relationships between chemical parameters and digestibility. Finally, re- lationships within study and forage type were esti- mated using the MIXED procedure of SAS with trial (forage) as a random factor (random inter- cept). This model excludes variation resulting from differences such as animals in digestibility trials, animals in iNDF determination, enzyme activity in OMS determination, and the year effect between forage chemical composition and digestibility, i.e. the analysis describes the relationships between the independent and dependent variables within a study (Tables 5 and 6). The regression equations were considered acceptable predictors, when the prediction error was less than one-third of the standard deviation of the reference population, when the regression was biologically sound and they fit several forage types irrespective of envi- ronmental factors (see Nousiainen 2004 and Fig. 4). The concentrations of feed chemical fractions using general equations were poorly related to in vivo OMD of silages (Table 3, Fig. 4). Although the relationships were statistically significant, pre- diction error using CP, NDF and ADF as independ- ent variables was not markedly less than the stand- 0.000 0.010 0.020 0.030 0.040 0.050 0.060 CP NDF ADF Lignin RMSE s.d. A B C Fig. 4. Residual mean squared errors (RMSE) of the re- gression equations between feed components and organic matter digestibility estimated with different models (s.d. = standard deviation of the data; A = general relationship; B = forage type specific equations; C = variation from a ran- dom study effect excluded). Data contains all forage types. 301 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. ard deviation of OMD in the data (Fig. 4). Lignin was the best single predictor of OMD, but it ex- plained proportionally only 0.43 of the variation and prediction error (42 g kg-1) was too high for practical feed evaluation and ration formulation. A large proportion of unexplained variation in global regression was related to forage type. This was demonstrated by significant differences in the prediction error among the forage types, when general relationships between a feed fraction and OMD were used (Table 4), and the use of forage specific equations decreased the prediction error markedly (Fig. 4). The decrease was greater for the cell wall components than for CP. The better rela- tionship between NDF and ADF, and especially that of lignin is related to the fact that these com- ponents are causative factors and related to the bio- logical availability, whereas CP has no direct ef- fect on digestibility provided that minimum N re- quirements of rumen microbes are met. This re- evaluation confirms previous findings and a more detailed discussion about the relationships between chemical feed components and OMD is given in the original papers (Nousiainen et al. 2003a, b, Rinne et al. 2006). Van Soest (1994) stated that cell wall fractions predict the digestibility of regrowth silages poorly because the association between lignin and cellu- lose in them is weak. The present data support this view as indicated by poor general relationships be- tween lignin and NDF. However, when the random study (= year) effect was included in the statistical model, lignin was strongly correlated to NDF. This suggests that environmental differences among years affect lignification of forage cell walls and lead to variable digestibilities at the same lignin concentration. Prediction errors were further re- duced by excluding the random trial (forage) ef- Table. 3. Predictions of in vivo organic matter digestibility of all forage types from feed chemical fractions (kg kg-1 dry matter) using a single regression equation (F) or a mixed model regression (M) with random trial(forage) effect. Component Model Intercept s.e.a Slope s.e. P-value RMSEb Adj. R2 Crude protein F 0.623 0.021 0.560 0.126 <0.01 0.0501 0.179 M 0.467 0.021 1.600 0.106 <0.01 0.0205 0.918 Neutral detergent fibre F 0.811 0.029 –0.200 0.056 <0.01 0.0518 0.12 M 1.096 0.029 –0.760 0.049 <0.01 0.0201 0.935 Acid detergent fibre F 0.846 0.028 –0.460 0.094 <0.01 0.0489 0.215 M 1.060 0.024 –1.200 0.072 <0.01 0.0192 0.925 Lignin F 0.805 0.013 –2.860 0.358 <0.01 0.0418 0.431 M 0.846 0.012 –4.190 0.274 <0.01 0.0208 0.869 a Standard error b Residual mean squared error Table 4. Residual mean squared errors of in vivo organic matter digestibility (OMD) predicted from chemical parameters assuming a general relationship between the chemical fraction and OMD. The values in bold are significantly (P < 0.05) different from zero. Primary growth grass Regrowth grass Legume Whole-crop RMSEa P-value Crude protein –0.025 0.009 0.031 0.001 0.041 <0.01 Neutral detergent fibre –0.032 0.010 0.028 0.036 0.040 <0.01 Acid detergent fibre –0.032 0.009 0.022 0.054 0.036 <0.01 Lignin –0.017 0.027 –0.022 0.037 0.032 <0.01 a Residual mean squared error 302 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation fects from the variation, i.e. within a trial and for- age type, chemical composition was closely related to in vivo OMD (Fig. 4). It is concluded that feed fractions can not be used to predict OMD with ac- ceptable precision, even when forage specific equations are used because the RSME of these equations are greater than 1/3 of the SD for the population of OMD. Organic matter cellulase solubility In vivo apparent digestibility (intake minus faecal output) determined with sheep by total faecal col- lection is the basis of most existing feed evaluation systems. For practical and often even for research purposes this method is too expensive, laborious and a large quantity of the feed is required. There- fore, laboratory in vitro methods have been devel- oped and are widely used, based on ruminal fluid (introduced by Tilley and Terry 1963; extensively reviewed by Weiss 1994) or commercial fungal cellulases. Due to difficulties in obtaining rumen fluid in commercial laboratories and standardisa- tion of the system, an enzymatic in vitro procedure in the determination of forage digestibility has been evaluated. Enzymatic digestion procedures have been de- scribed and discussed in detail in a review by Jones and Theodorou (2000). Basically the method in- cludes removing of cell solubles either by HCl- pepsin or neutral detergent followed by a 24 or 48 h incubation in buffered enzyme solution. The cellulase method differs from the in vivo digestion at least in two aspects: no endogenous matter is produced, i.e. solubility reflects true rather than apparent digestibility, and the capacity of commer- cial enzymes to degrade cell wall carbohydrates is less than that of rumen microbes (McQueen and Van Soest 1975, Nousiainen 2004). Nousiainen (2004) estimated that in vitro grass silage NDF solubility was 0.79 of the in vivo sheep NDF di- gestibility and only 0.67 of the potential NDF di- gestibility estimated by a 12 day ruminal in situ incubation. However, these differences do not pre- clude the use of enzymatic OM solubility (OMS) in predicting the in vivo digestibility provided that appropriate correction equations are used. The de- tails of the OMS method used in Finland are de- scribed by Nousiainen et al. (2003a). The present data indicates that the relationship between OMS and in vivo OMD is not uniform among the forage types (Table 5), because the pre- diction error within each forage type was markedly smaller than that estimated using the general cor- rection equation. However, compared to chemical components the prediction error was much smaller for either the general or forage-specific equations suggesting that enzymatic hydrolysis reflects the mechanisms of digestibility better than concentra- Table 5. Empirical relationships between pepsin-cellulase organic matter (OM) solubility (kg kg-1) and in vivo OM digestibility determined with fixed (F) or mixed regression analysis with random study effect (M). Forage Model Intercept s.e.a P-value Slope s.e. RMSEb Adj. R2 Primary growth grass F 0.103 0.0289 <0.01 0.83 0.038 0.0151 0.937 M 0.077 0.0211 <0.01 0.86 0.027 0.0085 0.981 Regrowth grass F –0.070 0.1030 0.50 1.01 0.136 0.0193 0.676 M –0.154 0.0627 0.05 1.12 0.082 0.0091 0.921 Legume F 0.002 0.0332 0.94 0.93 0.044 0.0122 0.962 M 0.003 0.0332 0.93 0.93 0.044 0.0121 0.962 Whole-crop F 0.182 0.0487 0.01 0.66 0.064 0.0109 0.947 M 0.290 0.0996 0.21 0.52 0.129 0.0090 0.942 All F 0.064 0.0348 0.07 0.86 0.046 0.0245 0.804 M 0.040 0.0193 0.05 0.89 0.026 0.0099 0.964 a Standard error b Residual mean squared error 303 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. tions of components from proximal analysis or de- tergent fractionation. When the forage specific equation was used, prediction error for OMD in all data decreased to 15.3 g kg-1. The prediction error (Observed-Predicted) of the general OMS equa- tion was positively related to pdNDF concentra- tion determined by 12 d ruminal in situ incubation (P < 0.01; R2 = 0.16), suggesting that the relative efficiency of the enzyme system to solubilize for- age OM decreased with increased concentration of NDF potentially digestible by rumen microbes. Using a mixed model regression, which includ- ed a random year-of-study variable, decreased pre- diction error especially for all forages, but also for grass silages. The smaller prediction error with the mixed model analysis may be associated with the variation between the trials in activity of the en- zyme and differences between the animals used for the in vivo determination. For RG grass, the asso- ciation between OMS and OMD could also depend on climatic and environmental conditions as indi- cated by the poor overall and good within year-of- study relationship between lignin and NDF, and between iNDF and NDF. Consequently, at a cer- tain OMD level, OMS for RG and legume silages is apparently higher than for PG silages (Table 1, Nousiainen et al. 2003b, Rinne et al. 2006). In addition to forage-specific equations, the laboratory-specific equations may be needed. De- spite serious attempts, the laboratories of Valio Ltd. and MTT were not able to standardise the OMS methods (Nousiainen 2004). There was a difference in the intercept, but the slope was 1.00 and R2 high (0.97). The intercept difference sug- gests particle loss during the procedures (manual filtration vs. Tecator crucibles). Further evidence for the possible contribution of the particle losses during OMS procedures is provided by a compari- son of the method described by Nousiainen et al. (2003a) and the Ankom filter bag system (Z.M. Kowalski et al., unpublished). It is also noteworthy that in the study with leg- ume silages (Rinne et al. 2006), in vitro OMD de- termined by Tilley and Terry (1963) method sig- nificantly underestimated in vivo OMD. In their In their original evaluation, Tilley and Terry (1963) specu- lated that despite a close relationship between in vivo and in vitro digestibility, these values are not identical and specific correction equations within laboratory and possibly within forage type may be needed. Weiss (1994) interpreted between-labora-Weiss (1994) interpreted between-labora- tory variations to suggest that ruminal in vitro sys- tems need laboratory-specific correction equa- tions. Organic matter digestibility can be predicted from OMS of pre-ensiled herbage as precisely as from OMS of the resultant silages provided that silages are well preserved with low or moderate ensiling losses (Huhtanen et al. 2005). For practi- cal ration formulation, sampling of herbage during silage harvesting allows more representative sam- pling and provides a better indication of the varia- tion in silage digestibility than samples taken from the silos, especially those drilled from the top of large tower silos. Advance information of silage digestibility would also be useful in the planning of rations for the feeding period. In conclusion, in vitro OMS provides more precise prediction of forage OMD than chemical feed analysis, when general equations are used (RMSE of 24 vs. 42 to 50 g kg-1 DM). However, to achieve accurate estimates (RMSE less than 20 g kg-1 DM), forage specific correction equations should be used. Solubility values may also be labo- ratory and methodology specific, which indicates that relationships between OMS and OMD must be developed for each laboratory setting. Indigestible neutral detergent fibre A part of the forage cell wall is unavailable to mi- crobial digestion in ruminants, even if total tract residence time of fibre could be extended to infi- nite time (Allen and Mertens 1988, Van Soest 1994). This forage DM fraction can be called indi- gestible fibre, here referred to as indigestible NDF (iNDF). In addition to NDS, iNDF represents by definition a uniform feed fraction with zero true digestibility. Potentially digestible fibre (pdNDF) may then be calculated as: pdNDF = NDF − iNDF [6] Several methods may be used to divide forage NDF to potentially digestible and indigestible frac- tions, e.g. end-point measurement with long-term 304 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation (up to 144 h) in vitro batch rumen fluid incubation (Traxler et al. 1998) or fitting time-dependent (0– 96 h) in vitro or in situ NDF nylon bag degradation data to single digestion pool rumen model (Wilman et al. 1996a). The ultimate extent of NDF digestion may not be reached with in vitro batch system and the in situ estimates may be biased due to crucial drawbacks of the traditional nylon bag procedure as discussed earlier (Nousiainen 2004, Nousiainen et al. 2004). The slow rate of NDF digestion within the nylon bags with small pore sizes leads to pro- longed NDF digestion (Huhtanen et al. 2006), and the difference between the extent of digestion reached at 96 and 288 h incubations increased as the digestibility of silage decreased (Rinne et al. 2002). Because forage iNDF fraction is attributa- ble to cross-linking between cell wall lignins and hemicellulose when plants mature (Van Soest 1994), several attempts to predict iNDF from lignin concentration in DM or NDF have been made (see Traxler et al. 1998). Despite this bio- logical conjecture, it has not been successful due to relatively high proportional errors in lignin and iNDF analyses, as well as differences between for- age types in lignin to iNDF ratio, which may also be prone to climatic factors. The Cornell Net Carbohydrate and Protein sys- tem uses a factor 2.4 × lignin concentration in NDF in describing iNDF of forages (Van Soest et al. 2005). This factor is presumed to be universal across forage species and growth environments. Validation of this concept with data containing several forage species (corn, alfalfa, grasses, wheat straw) resulted in satisfactory regression (R2 = 0.94) between observed and predicted (2.4 × lignin) iNDF (Van Soest et al. 2005). However, the present data does not support a generally applica- ble relationship between permanganate lignin and iNDF measured by 12 d in situ fermentation, al- though the overall slope was 2.4 (Fig. 5). The slopes for individual forages species varied be- tween 2.8 and 5.5, and a general regression equa- tion predicted iNDF with an unsatisfactory accu- racy (R2 = 0.56; RMSE = 27.4 g kg-1 DM). If for- age-specific relationships were used, the RMSE for predicted iNDF decreased to 14.9 g kg-1 DM. This confirms the previous findings (Nousiainen et al. 2004) and suggests that a universal lignin equa- tion describing the iNDF fraction did not exist as we measured them. The forage type specific lignin equations may be used to predict iNDF if in situ estimates are not available. To determine forage iNDF concentration, a long-term (12 d) in situ incubation has been used at MTT to ensure complete digestion of potentially digestible NDF. The small pore size (6 or 17 μm) combined with a relatively large open surface area of the nylon bag cloth used allows moderate mi- L: y = 2.68x - 7.6 R2 = 0.970 RG: y = 5.41x - 43.1 R2 = 0.837 WC: y = 3.72x + 18.7 R2 = 0.962 PG: y = 3.09x - 23.1 R2 = 0.708 0 40 80 120 160 200 0 20 40 60 80 Lignin (g kg-1 DM) iNDF (g kg-1 DM) PG Grass RG Grass Legume Whole crop Fig. 5. The relationship between lignin and iNDF concentrations estimated by fixed regression analysis of silages made from pri- mary (PG) or regrowth (RG) grass and leguminous (L) or whole-crop (WC) forages; over- all regression y = 2.4x + 18 (R2 = 0.555, residual squared mean er- ror = 27.4 g kg-1 DM). 305 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. crobial activity within the bags (Huhtanen et al. 1998) and prevents particle in- and out-flow from the bags. Indigestible NDF determined by long in situ incubation has been used to describe ruminal cell wall kinetics (Tamminga et al. 1989), and as digestibility marker for estimating total and rumi- nal digestibility (Huhtanen et al. 1994, Ahvenjärvi et al. 2000). The details of the procedures are de- scribed by Huhtanen et al. (1994) and Ahvenjärvi(1994) and Ahvenjärvi et al. (2000). After(2000). After in situ incubation, the residues are washed with water and treated with neutral de- tergent solution to remove microbial matter. An inter-laboratory ring test in Nordic coun- tries showed large differences in the iNDF esti- mates (Lund et al. 2004) leading to standardization of the method, which was succesful in removing the between-laboratory differences (J. Nousiainen et al. unpublished). Currently the method includes the use of polyester bags with 10–17 μm pore size and 10–20 mg cm-2 sample to surface ratio. Rumi- nal incubations (288 h) should be conducted with two cows fed with forage based diets (forage to concentrate ratio at least 60:40). The use of NIRS in predicting grass silage iNDF has also been eval- uated with promising results (Nousiainen et al. 2004), but the standardization of the reference method is a vital prerequisite in developing robust calibrations. Previous results for grass silages (Nousiainen et al. 2003b, Nousiainen 2004) suggested that iNDF can be used in a general linear regression equation to predict forage OMD relatively univer- sally over a range of species and harvesting condi- tions. The intercept of this equation represents a theoretical maximum of forage OMD provided that all NDF is potentially digestible and that the rate of pdNDF digestion (kd) is the only factor lim- iting digestibility when the forage is fed to sheep at maintenance level of feed intake. The slope of the regression describes the decline in OMD with in- creasing iNDF concentration. However, when for- age-specific equations for PG, RG, legume and whole-crop silages are compared, the relationship between iNDF and OMD was not uniform (Table 6). This can be judged both by the variable inter- cepts and slopes between the different forage types. For PG grass and whole-crop silages the slope of the iNDF equation seems to be equal (about −1.5) irrespective of the model used (fixed vs. mixed) and suggests that one gram iNDF protects 1.5 gram NDF (or OM) from digestion in sheep Table 6. Empirical relationships between forage indigestible neutral detergent fibre concentration (kg kg-1) and in vivo organic matter digestibility determined with fixed (F) or mixed regression analysis and corrected for the random study effect (M). Forage Model Intercept s.e.a P-value Slope s.e. RMSEb Adj. R2 Primary growth grass F 0.852 0.0064 <0.01 –1.52 0.07 0.0159 0.932 M 0.851 0.0062 <0.01 –1.51 0.05 0.0086 0.979 Regrowth grass F 0.802 0.0137 <0.01 –1.03 0.13 0.0180 0.718 M 0.829 0.0108 <0.01 –1.30 0.08 0.0072 0.963 Legume F 0.831 0.0095 <0.01 –1.14 0.08 0.0175 0.921 M 0.832 0.0097 <0.01 –1.15 0.07 0.0129 0.956 Whole-crop F 0.867 0.0134 <0.01 –1.52 0.11 0.0082 0.970 M 0.867 0.0134 <0.01 –1.52 0.11 0.0082 0.970 All F 0.834 0.0053 <0.01 –1.26 0.05 0.0190 0.883 M 0.839 0.0051 <0.01 –1.32 0.04 0.0106 0.964 a Standard error b Residual mean squared error 306 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation fed at maintenance. The respective slope, however, is lower for RG grass (−1.3) and especially for leg- ume (−1.15) silages. This may be explained by the variable relationship between iNDF concentration and rate of pdNDF digestion (kd) among the forage types resulting in different apparent OMD at the same iNDF concentration (see Rinne et al. 2006) and also suggests that pdNDF is not a uniform nu- tritional entity. Especially legumes at high iNDF concentration are still relatively highly digestible as compared to grasses (Wilman et al. 1996b, Van Soest 1994). Nevertheless, despite the lack of uni- form behaviour, the iNDF regression equation may be very useful in predicting forage OMD, espe- cially if forage specific OMS equation cannot be used (see Tables 5 and 6). Summative models Background and methods Most of the existing feed evaluation systems use the total amount of digestible nutrients expressed as grams in feed DM to determine feed metabolis- able energy value (MAFF 1975, MTT 2006). However, the analytical procedures and equations to estimate digestible nutrients vary between sys- tems. Van Soest (1967) developed a comprehen- sive system of feed analysis and its application to forages. He divided the feed into NDS fraction which is essentially completely available but its digestibility is apparently incomplete, because of faecal endogenous and microbial material. The second fraction corresponds to fibre (NDF) and its availability is controlled by structural features that link cellulose, hemicellulose and lignin. The fibre fraction is not uniform between forages. Goering and Van Soest (1970) presented a summative mod- el to describe availability of forage DM: dDM = NDFD × NDF + 0.98 × NDS – M [7] where dDM = digestible DM, NDFD = coefficient of NDF digestibility, and M = microbial and en- dogenous faecal DM losses. Theoretically this model is sound, but generally NDFD is not known. Conrad et al. (1984) modified this model by divid- ing feeds into NDS and potentially digestible NDF. They applied surface area law (mass raised to pow- er 0.67) to calculate NDF that is covered by lignin, and this proportion was multiplied by lignin-free NDF to estimate available NDF. Their available, lignin-free NDF component, i.e., (NDF–L) × (1– L2/3 / NDF2/3) was an attempt to estimate pdNDF in the current terminology. They assumed that the di- gestibility of available lignin free NDF was 0.75 to calculate TDN at maintenance level of intake. Weiss et al. (1992) revised the Conrad et al. (1984) model and it was adopted by NRC (2001) to esti- mate total digestible nutrients (TDN). Huhtanen (2003) evaluated the NRC (2001) model using in vivo sheep digestibility data. The predicted and ob- served digestible OM concentrations (D-value, g kg-1 DM) were relatively well correlated, but there was a considerable slope bias. The NRC (2001) system clearly underestimated the D-value of high quality grass silages. This suggests that this system is not uniform for forages grown in different envi- ronmental conditions. The major problem was that the potential maximum of 0.75 for the digestibility of lignin free NDF is clearly too low for high qual- ity grasses grown in northern latitudes. Because the in vivo pdNDF digestibility is markedly less variable than the total NDF digesti- bility, and because the fraction subjected to this variation (pdNDF vs. NDF) is smaller, the accura- cy of the summative systems based on three frac- tions (NDS, pdNDF and iNDF) could be improved compared to systems dividing feeds only to total NDF and ND solubles. In the present data, the co- efficient of variations of pdNDF and NDF digesti- bility were 0.064 and 0.135, and respective con- centrations 403 and 500 g kg-1 DM. Exclusion of the whole-crop silages from the data decreased the coefficient of variation in pdNDF digestibility to 0.041. In the Lucas test for the pdNDF fraction, the overall coefficient of determination was high (R2 = 0.95) and the intercept was close to zero (Fig. 6), but obviously the high R2 reflected partly a large range in the pdNDF concentration. The intercept was significantly positive for PG (P = 0.01) and legume (P = 0.08) silages and negative (P = 0.001) for the whole-crop silages. Both the negative and positive intercepts are biologically impossible, be- cause the amount absorbed can not be positive at 307 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. L = 0.74x + 34 R 2 = 0.872 RG = 0.81x + 29 R 2 = 0.773 PG = 0.71x + 68 R 2 = 0.820 WC = 0.88X - 54 R 2 = 0.996 Total = 0.86x - 4 R 2 = 0.951 0 100 200 300 400 500 600 200 250 300 350 400 450 500 550 600 pdNDF (g kg -1 DM) dNDF (g kg-1 DM) . PG Grass RG Grass Legume Whole cropFig. 6. Lucas test for digestible neutral detergent fibre (dNDF) of silages made from primary (PG) or regrowth (RG) grass and legu- minous (L) or whole-crop (WC) forages. zero intake and there is no faecal endogenous and microbial excretion of digestible fibre. Excluding the whole-crop silages resulted in the following equation by the Lucas test: dNDF(g kg-1 DM) = 16.9±7.0 + 0.821± 0.017 × pdNDF (g kg-1 DM) (R2 = 0.97) [8] where dNDF is digestible NDF and pdNDF poten- tially digestible NDF calculated as NDF − iNDF. This equation meets all other criteria of uniformity presented by Lucas (see Van Soest 1994) except that the intercept was slightly, although signifi- cantly (P = 0.02) positive. However, when legume silages were excluded from the data the intercept increased to 65 g kg-1 DM (P < 0.01) suggesting that pdNDF is not a uniform entity. Despite high R2 values for the dNDF in the Lucas test, it can not be considered as fundamental biochemical cause- and-effect relationship, and therefore the summa- tive approach in determining forage availability based on tdNDS and dNDF estimated by the Lucas concept must be essentially interpreted as an em- pirical approach. Three different summative approaches were used to estimate the silage D-value. All methods had the same basic structure but the method used in estimating dNDF differed: D-value (g digestible OM kg-1 DM) = tdNDS + dNDF – M [9] where tdNDS (g kg-1 DM) is truly digestible NDS and M is faecal microbial and endogenous output of OM (g kg-1 DM). NRC (2001) estimated dNDF (g kg-1 DM) as follows: dNDFNRC (g kg -1 DM) = 0.75 × (NDF –Lignin) × [1 – (Lignin/NDF) 0.667] [10] where NDF and lignin are expressed as g kg-1 DM. In this equation, the first part may be interpreted as potentially digestible NDF (i.e. pdNDF) and the latter part [1 – (Lignin/NDF) 0.667] digestibility of pdNDF. Both the original parameter values and those estimated form the present data were used. Mertens (2002b) derived a simple equation in which dNDF is a linear function of NDF and lignin: dNDFMertens = a × NDF (g kg -1 DM) + b × Lignin (g kg-1 DM) [11] where a and b can be estimated by regression. Constant a is the digestibility coefficient of pdNDF and constant b is the product of the digestibility coefficient of pdNDF and the proportion of NDF 308 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation protected by lignin. This equation has no intercept, i.e. neither endogenous nor microbial excretion of NDF. The parameter values for both NRC and Mertens equations were estimated by the Solver tool (Fylstra et al. 1998) in Microsoft® Excel, which employs the Generalized Reduced Gradient (GRG2) non-linear optimization code (Lasdon et al. 1978). The third summative equation in estimating D- value was based on applying the Lucas test both for NDS and pdNDF: D-value (g kg-1 DM) = tdNDS (g kg-1 DM) + dNDF (g kg-1 DM) [12] Because the effects of lignin and iNDF on di- gestibility and output of M were forage type spe- cific, the summative models were tested both using all data and separately for each forages. The mod- els of NRC (2001) and Mertens (2002b) were also tested by using iNDF instead of lignin. For all models, both general equations derived from all data and forage specific equations were used for dNDF and dNDS. The models were compared on the basis of residual mean squared errors were cal- culated as: RMSE = √ ∑ (Observed – Predicted) 2/ n [13] Mean squared prediction error (MSPE) was di- vided to components resulting from mean bias, slope bias and random variation around the regres- sion line (Bibby and Toutenburg 1977). Results and discussion The NRC (2001) system clearly underestimated in vivo dNDF (Fig. 7), which agrees with Huhtanen (2003). The mean bias (observed – model predict- ed) was 48 g kg-1 DM, but it varied from 75 (PG silages) to –34 g kg-1 DM (whole-crop). The major problem was that the first part of the NRC equation [0.75 × (NDF – Lignin)] clearly underestimated the concentration of potentially digestible NDF (351 vs. 404 g kg-1 DM). However, the precision of the prediction was good (R2 = 0.89), mainly be- cause of the close relationship between forage NDF and pdNDFNRC concentrations (R 2 = 0.89). In vivo digestibility of NDF was higher than the max- imum potential NDF digestibility of the NRC (2001) system (0.75) in 19 cases and that of lignin- free NDF in 39 cases out of 86. In the present study, lignin was analysed as permanganate lignin, which results in higher val- ues than ADL. The mean bias of the NRC (2001) system would probably have been smaller, if ADL had been used. Variation in dNDFNRC was more y = 1.16x + 1.5 R 2 = 0.850 100 200 300 400 500 100 200 300 400 Predicted pdNDF (g kg -1 DM) Observed pdNDF (g kg-1 DM) PG RG L WC y=x Fig. 7. The relationship between digestible neutral detergent fibre (dNDF) concentration predicted according to NRC (2001) and ob- served dNDF determined with sheep fed at maintenance level of feeding. 309 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. closely associated with the NDF concentration than with the observed in vivo dNDF (R2 0.91 vs. 0.75). This suggests that the components of the equation predicting pdNDF and its digestibility did not describe biological cause-and-effect rela- tionships explaining the variation in these basal at- tributes of digestibility. This evaluation using the in vivo data for several forage species did not vali- date successfully the assumptions about the sur- face area law describing the effect of lignin on di- gestibility. The results of the evaluation of the general summative models are shown in Table 7. Estimat- ing the parameters from the present data decreased the prediction error of the NCR (2001) equation due to reduced mean bias. The parameter value de- scribing the maximum digestibility of lignin free NDF increased from 0.75 to 0.966, which overes- timated the concentration of pdNDF by 48 g kg-1 DM (452 vs. 404 g kg-1 DM). However, the preci- sion of the prediction was acceptable (R2 = 0.89). The power in the second component of function which assumes that lignin affects NDF digestibili- ty according the surface area law, decreased from 0.67 to 0.515. As for the original NRC equation, simulated pdNDF digestibility by the model was not correlated with the in vivo data. The more com- plex function of lignin adopted in the NRC (2001) model did not improve the precision of the predic- tion compared to purely empirical prediction based on NDF concentration and linear relationship be- tween NDF digestibility and lignin concentration (RMSE = 33.9). Mertens (2002b) equation result- ed in a similar error to that of NRC and empirical approach. Interestingly, the prediction errors of NRC and Mertens equations were strongly corre- lated (R2 = 0.99) indicating that the form of lignin function had no influence on D-value prediction. When lignin was replaced with iNDF, the pre- diction error of pdNDF digestibility reduced mark- edly. This improvement can be attributed to the fact that direct determination of iNDF by 12 day in situ describes the fibre fraction that is completely unavailable for microbial digestion better than lignin concentration. The ratio between lignin and iNDF was not uniform between the forage types, a prerequisite for accurate and precise prediction of dNDF from simple or complex functions of lignin. The error in lignin analysis is absolute rather than proportional (Van Soest 1994), which can lead to Table 7. Prediction of the digestible NDF (dNDF) and D-value using different summative equations for all forage types; general equations were used to predict both dNDF and digestible neutral detergent solubles. Trait/method Independent variable Intercept Slope R2 MSPEa Distribution of MSPE Bias Slope Random dNDF (g kg-1 DM) NRC (2001)NRC (2001) Lignin 11 0.971 0.853 33.6 0.000 0.005 0.994 NRC (2001) iNDF 9 0.977 0.954 18.7 0.001 0.012 0.988 Mertens (2002b) Lignin 9 0.975 0.849 34.0 0.000 0.004 0.996 Mertens (2002b) iNDF 8 0.978 0.957 18.5 0.001 0.011 0.988 Lucas test dNDF 0 1.001 0.952 19.1 0.000 0.000 1.000 D-value (g kg-1 DM) NRC (2001) Lignin –104 1.162 0.551 35.3 0.000 0.023 0.976 NRC (2001) iNDF –80 1.124 0.904 17.0 0.001 0.103 0.896 Mertens (2002b) Lignin –121 1.188 0.542 35.7 0.000 0.029 0.971 Mertens (2002b) iNDF –78 1.121 0.911 16.4 0.001 0.106 0.893 Lucas test dNDF –152 1.235 0.908 18.3 0.000 0.264 0.736 a Mean squared prediction error 310 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation high proportional errors in lignin analysis of for- ages of low lignin concentration. Predicting dNDF assuming pdNDF as a uniform nutritional entity described pdNDF almost as precisely as the mod- els of NRC (2001) and Mertens (2002b). The prediction error of D-value was not higher than that of dNDF in spite of relatively large (16 g kg-1 DM) error in predicting dNDS from Lucas equation. The non-additivity of the errors was mainly due to the negative correlation between the errors in dNDF and dNDS, i.e. the errors were partly counterbalanced. For example, all models underestimated dNDF for whole-crop silages, but because faecal endogenous output was underesti- mated, the overall D-value was predicted fairly ac- curately for the whole-crop silages. The slope bias in D-value predictions also suggests interactions between dNDF and dNDS components. When the forage specific equations were ap- plied to predict both dNDF and tdNDS (Table 8), prediction errors were markedly reduced compared with the general equations. Prediction errors for dNDF were only 11–12 g kg-1 DM for the three models based on iNDF, and the models describing the mechanisms of digestion were slightly better. The more complex NRC (2001) model was not better than the simpler Mertens (2002b) model. This provides further evidence that the theoretical surface law of lignin protection does not predict digestibility of pdNDF more accurately than the empirical relationships between iNDF and pd- NDF. The Mertens (2002b) equation can be formu- lated in three different ways: dNDF = a × NDF + b × iNDF [14] dNDF = a × (NDF – iNDF) + b × iNDF [15] dNDF = a × (NDF – iNDF) [16] Equation [14] describes dNDF as a function of NDF and iNDF. Coefficient a can be interpreted as a maximum potential NDF digestibility and coef- ficient b representing a discount for dNDF related to iNDF. This equation also allows NDF and iNDF interact in such a way that possible effects of iNDF concentration on pdNDF digestibility can be ac- counted for. In equation [15], the fraction (NDF- iNDF) describes by definition potentially digesti- Table 8. Prediction of digestible NDF (dNDF) and D-value using different summative equations from data comprising of silages made from primary or regrowth grass and leguminous or whole-crop forages; forage specific equations were used to predict both dNDF and digestible neutral detergent solubles. Independent variable Intercept Slope R2 MSPEa Distribution of MSPE Bias Slope Random dNDF (g kg-1 DM) NRC(2001)NRC(2001) Lignin 15 0.960 0.944 20.9 0.000 0.005 0.994 NRC(2001) iNDF 4 0.989 0.984 11.2 0.001 0.012 0.988 Mertens (2002b) Lignin 13 0.964 0.946 20.5 0.000 0.004 0.996 Mertens (2002b) iNDF 2 0.993 0.983 11.4 0.001 0.011 0.988 Lucas test dNDF 0 1.001 0.980 12.4 0.000 0.000 1.000 D-value (g kg-1 DM) NRC (2001) Lignin –50 1.078 0.802 23.4 0.003 0.021 0.976 NRC (2001) iNDF –16 1.026 0.932 13.6 0.002 0.008 0.990 Mertens (2002b) Lignin –33 1.053 0.809 23.0 0.002 0.010 0.987 Mertens (2002b) iNDF –13 1.020 0.930 13.8 0.001 0.005 0.994 Lucas test dNDF –50 1.078 0.922 15.0 0.001 0.057 0.942 a Mean squared prediction error 311 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. ble NDF, coefficient a is digestibility of pdNDF and b is a discount factor for iNDF allowing pd- NDF digestibility to differ with iNDF concentra- tion. The third equation [16] is a simplification from equation [15] and it assumes a constant di- gestibility for pdNDF (NDF – iNDF), i.e. the equa- tion is Lucas model without M for pdNDF. The three equations were compared and the re- sults for PG grass and legume silages are shown in Table 9. Coefficient a and RMSE of the models were similar for equations [14] and [15], and rep- resent the digestibility of pdNDF. In equation [14] the coefficient b is associated both with iNDF and the effect of iNDF on the digestibility of pdNDF, whereas in equation [15] coefficient b describes the additional effect of iNDF on pdNDF digestibil- ity. Indigestible NDF had a strong negative effect on pdNDF digestibility of PG grasses (–0.317). In contrast, iNDF had only a minor effect on pdNDF digestibility of legume silages and consequently, the simple equation [16] did not increase markedly the prediction error. However, for the PG silages, equation [15] resulted in a smaller prediction error due to the strong impact of iNDF on pdNDF di- gestibility. The results suggest that equation [15] includes the basic nutritional concepts of fibre digestion: it separates NDF into potentially digestible and indi- gestible fractions (1), and that the equation is flex- ible allowing interactions between pdNDF and iNDF to influence the digestibility of pdNDF (2). Compared with equation [16] or the Lucas equa- tion allowing an intercept, equation [15] markedly reduced the prediction error. This effect may be as- sociated to the curvilinear relationship between maturity and pdNDF concentration, whereas iNDF increases linearly with advancing maturity. Al- though equation [14] predicts pdNDF equally well to equation [15], interpretation of the coefficients is biologically more difficult. The summative approach based on uniform nu- tritional entities and biochemical cause-and-effect relationships for non-uniform entities, predicted silage D-value at least as accurately as the best em- pirical equation using either OMS or iNDF as in- dependent variables. When the general relation- ships were used, the summative approach was markedly better than OMS (Table 10). This can mainly be attributed to forage specific relation- ships between OMS and OMD. Results in Table 10 suggest that only minor reductions in RMSE are gained by the use of forage-specific equations compared to general equations for summative models and those using iNDF. These equations would reduce by a factor of four the number of parameters that must be estimated and are consist- ent with the uniform nutritional availability of the Lucas test. Re-evaluation of the different approaches re- veals that for accurate and precise prediction of D- Table 9. Comparison of three versions of Mertens (2002b) equation (for description of equations, see text) in predicting forage D-value using data of silages made from primary growth grass or legumes. Forage Equation Coefficient RMSEa Regression for potentially digestible NDF a b Intercept Slope R2 Primary growth grass [14] 0.901 –1.218 10.14 30 0.928 0.912 [15] 0.901 –0.317 10.14 30 0.928 0.912 [16] 0.849 0.000 15.23 68 0.838 0.820 Legume [14] 0.886 –0.927 9.72 30 0.867 0.871 [15] 0.886 –0.042 9.72 30 0.867 0.871 [16] 0.868 0.000 9.90 34 0.852 0.872 a Residual mean squared error 312 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation value forage specific equations are needed irre- spective the method used (empirical vs. summa- tive, see Table 10). Basically, this is because among the forages faecal output of NDS is not constant (1), and because the relationship between iNDF concentration and the rate of pdNDF digestion is variable (2). An advantage of the summative sys- tems is that they are based on physical and bio- chemical factors that influence the availability of various feed fractions. It is possible that sometimes the errors counterbalance each other, the case be- ing especially likely for the methods that use iNDF as independent factor. The strong empirical rela- tionship between iNDF concentration and OMD reported by Nousiainen (2004) was also confirmed by the results obtained from this larger dataset. Dynamic models Several reviews have discussed the mathematic modelling of ruminal cell wall digestion and strengths and weaknesses of the experimental methods used to determine the parameter values required in the models (Mertens 1993, Illius and Allen 1994, Ellis et al. 1999, Huhtanen et al. 2006). The recent knowledge of digestion and passage ki- netics has been incorporated into the Nordic dairy cow model Karoline (Danfær et al. 2005). This model predicted accurately and precisely the amount of NDF digested. The sensitivity analysis demonstrated that iNDF is a key parameter in esti- mating nutrient supply from the digestive tract (Huhtanen et al. 2006), which is consistent with the close relationship between forage iNDF con- centration and D-value. Digestion in the ruminant digestive tract is the competition between the rates of digestion (kd) and passage (kp). When the rate of digestion in relation to passage increases, digesti- bility of pdNDF increases. The variation in pdNDF digestibility must therefore be associated with dif- ferences in the rates of digestion and passage. Previous discussion of the methods to describe feed availability clearly demonstrated that the scope to decrease the prediction error of D-value is rather limited with traditional regression equation and summative approaches. An additional source of variation is the variable faecal NDS secretion, and a more thorough understanding of the underly- ing biological mechanisms causing this variability (in this data from 81 (whole-crop) to 121 (leg- umes) g kg-1 DM intake) is needed to improve the models in predicting forage D-value. Both the empirical and summative approaches were limited in their ability to explain the variation in pdNDF digestibility related to rates of passage and digestion, which had a range of 0.11 to 0.15 units for grass and legume silages. A large propor- tion of this variation is related to differences in the rate of digestion attributable to intrinsic feed fac- Table 10. Prediction of D-value (g kg-1 DM) from organic matter pepsin cellulase solubility (OMS) and indigestible neutral detergent fibre concentration (iNDF) using empirical relationships or the summative approach according to Mertens (2002b). Method Equationa Intercept Slope R2 MSPEb Distribution of MSPE Bias Slope Random OMS G –29.2 1.05 0.816 22.3 0.000 0.008 0.992 S –5.1 1.01 0.929 13.8 0.000 0.001 0.999 iNDF G –1.3 1.00 0.893 16.9 0.000 0.000 1.000 S –49.7 1.08 0.802 14.3 0.000 0.000 1.000 Mertens (2002b) G –78.3 1.12 0.912 16.3 0.001 0.106 0.893 S –13.6 1.02 0.930 13.7 0.001 0.005 0.994 a G = General equation for all forages; S = Forage type specific equations. b Mean squared prediction error 313 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. tors, since at maintenance level the differences in compartmental residence time are unlikely to be large enough to explain the observed differences in pdNDF digestibility. For example, if kd is 0.05 per h, mean compartmental residence time should in- crease from 50 to 90 h to increase pdNDF digesti- bility from 0.80 to 0.90. Similarly, with 50 h com- partmental residence time kd should increase from 0.05 to 0.09 per h to increase pdNDF digestibility from 0.80 to 0.90, respectively. Attempts to estab- lish relationships between feed chemical fractions and kd of fibre have had little success (for review see Huhtanen et al. 2006). The relationships may be reasonable within a forage type, but overall re- lationships are poor. There are two prerequisites for the dynamic models to improve predictions of D-value: the method must be accurate in predict- ing the true pdNDF digestion rate (1) and it must be more precise than the current empirical ap- proaches (2). Until now the progress in this area has been limited by the lack of in vivo validation data. Most of the kd studies have compared differ- ent laboratory and in vitro methods and the data has been mainly qualitative ranking of feedstuffs. The studies conducted at MTT have suggested that in vitro gas production technique (for review see Schofield 2000) is a promising tool for estimat- ing kd of NDF. When the parameter values derived from gas production kinetics of isolated NDF were used in dynamic rumen models, in vivo NDF di- gestibility was predicted both accurately and pre- cisely (Huhtanen et al. 2001, Rinne et al. 2006). The data from in vivo digestion trials can be used to estimate digestion rate by solving the equation of Allen and Mertens (1988) for kd by assuming a fixed compartmental residence time (Huhtanen et al. 2006). Digestion rates estimated from isolated silage NDF with in vitro gas production technique and those calculated from the in vivo data were strongly correlated (R2 = 0.90) without mean bias. In contrast, ruminal in situ incubation markedly underestimated the in vivo digestion rate (Huh- tanen et al. 2006). The current empirical and summative models are probably accurate and precise enough to predict the D-value at maintenance level and hence are suit- able for calibration of NIRS equipment for practical feed evaluation of farms samples. However, the fu- ture feed and ration evaluation models such as Karoline (Danfær et al. 2005) need accurate and precise estimates of the kinetic parameters of NDF digestion. The existing energy values predicted from feed digestibility at the maintenance level still form the sound basis for feed evaluation systems, but dynamic models are needed to cope with the interactions between dietary components at differ- ent feeding levels. Near infrared reflectance spectroscopy Implementation of computerised chemometrics based on near infrared reflectance spectroscopy (NIRS) of homogenised feed samples was a major innovation that made the recent developments of forage evaluation research available for practical farmers. Since Norris et al. (1976) first introduced the NIRS equations for predicting forage quality, much success has been achieved in developing NIRS for the forage analysis (for a review, see Deaville and Flinn 2000). The parallel develop- ment of computers, optical devices and calibration software have stimulated the progress of NIRS ap- plications in feed analysis. The purpose of this chapter, however, is not to review the theory be- hind NIRS, instrumentation, sample treatment and presentation, mathematical treatments of spectral data and calibration methods; instead, the reader is referred to the numerous textbooks and reviews (see e.g. Williams and Norris 1987, Windham et al. 1989, Reeves 2000). Herein the developments in NIRS equations for predicting D-value of forages that are typically produced in Finland are dis- cussed. Near infrared reflectance spectrum (usually from 1100 to 2500 nm) of forages contains spe- cific absorbance regions e.g. for water and protein (see Deaville and Flinn 2000), which both can be predicted relatively accurately by NIRS. In con- trast, the predictions of forage fibre characteristics and OMD in particular are more challenging be- cause these traits are not definite chemical entities and do not have specific absorbance bands in the NIR spectrum. Therefore, NIRS has been criti- 314 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation cized as a “black box” method for predicting feed characteristics of value for animal nutrition. How- ever, as indicated by Deaville and Flinn (2000), interpretation of published NIRS equations reveal that OMD of forages is often associated to spectral regions near to 1650–1670 and 2260–2280 nm (see Fig. 8). Nousiainen et al. (2004) demonstrated that these regions were negatively correlated to grass silage iNDF as determined with long (288 h) ruminal in situ incubation and that the standard normal variate and de-trended (see Barnes et al. 1989) correlation spectrum for lignin and iNDF showed much resemblance. Previous findings by Russell et al. (1989) also relate these spectral re- gions to lignin bonding, thus providing scientifi- cally valid background for determining forage fi- bre characteristics by NIRS. The essential advantage of NIRS is the speed and economy of forage evaluation. The accuracy of NIRS results is related to the scope of the data set used to calibrate for forage digestibility predic- tions. A wide range of reference values are needed for NIRS calibration data set to predict forage di- gestibility, even when intended to be applied to a specific forage type. For forage D-value predic- tion, in vivo digestibility would be the most logical reference method, as was the first application in Finland (Hellämäki 1992). Though the reference data set included a reasonable number of samples (n = 90) and the performance of calibration was satisfactory [standard error of calibration 15.3 g kg-1], it resulted in biased predictions when applied to unknown samples. The evident reason for this was too narrow a range of D-values in the refer- ence data (SD 36 g kg-1). Further, much spectral variation is caused by the unhomogeneous nature of forages attributable to species and variety differences, ensiling meth- ods, harvest (primary vs. regrowth) and possibly climatic factors. As a consequence, several hun- dred reference samples are required for a multi- species forage population (Deaville and Flinn 2000), which makes the use of in vivo digestibility data as a reference method essentially unpractical and expensive. Hence, a biologically valid in vitro reference method is needed, i.e. validation against in vivo data. Due to obvious advantages for a com- mercial forage laboratory, the pepsin-cellulase method as described by Friedel (1990) was chosen for the reference method (Klemetti et al. 1995), and in addition to in vivo samples, the calibration data set was extended with data from on-farm si- lages to increase the D-value range and spectral variation. The resulting calibration performed ac- ceptably (validation R2 = 0.752 and SEP/SD = 2.1), but produced unrealistically high D-value predictions for silages made from regrowth grass. Later it appeared that the single correction equa- tion for OMS introduced by Friedel (1990) does not generally apply to different forage types (dis- cussed earlier in this paper). In conclusion, the total prediction error of a NIRS D-value calibration is strongly dependent on the biological validity of the in vitro reference method. Nousiainen (2004) compared different reference methods for grass silage D-value and demonstrated that when the proportion of refer- ence error increases, the total NIRS prediction er- ror (observedin vivo vs. predictedNIRS) increased sig- nificantly. Thus a good NIRS calibration and vali- dation statistics does not automatically guarantee acceptable total prediction performance, and if not 0 0.1 0.2 0.3 0.4 0.5 1000 1200 1400 1600 1800 2000 2200 2400 Wave length (nm) log/(1/R) -0.02 0.00 0.02 0.04 0.06 0.08 Difference Early cut PG Late cut PG Difference Fig. 8. Near infrared reflectance spectrum of very early (D-value 764 g kg-1 DM) and late (D-value 586 g kg-1 DM) cut silages made from primary growth (PG) grass in 1996; the arrows in the difference spectrum indicate the impor- tant wave length areas that are associated with digestibili- ty. 315 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. recognised, this may lead to serious misuse of NIRS in nutritional applications. The calibration and validation statistics of NIRS D-value equations for forages used between the years 2003–2005 in Finland is shown in Table 11. The calibrations were produced either using general or forage type specific (2003 vs. 2004 and 2005) OMS prediction equations. The number of samples also differed between the calibrations as well as math treatment of the spectral data (first vs. second order derivatization in calibration 2003 and 2004 vs. 2005, respectively). The calibrations were applied to experimental silages, and the total pre- diction errors (Observedin vivo vs. PredictedNIRS) were calculated (Table 12). Despite the best cross- validation results for the 2003 calibration (i.e. best precision), it produced the lowest accuracy of the D-value estimates compared to in vivo, mainly ow- ing to over- and under-prediction of silages made from regrowth and primary growth grass with gen- eral OMS equation, respectively. This problem was only partly solved in the 2005 calibration as Table 11. Near infrared reflectance spectroscopy calibration and validation statistics for silage D-value (J. Nousiainen et al. unpubl.). Calibration OMS equationc Nb Mean s.d.d Calibrationa Cross validation Math SECe R2 SECVf R2 SD/SECV 2003 General 750 672 34.7 1,4,4,0 10.8 0.903 11.7 0.887 2.97 2004 Specific 994 660 45.6 1,4,4,1 18.3 0.839 19.1 0.824 2.38 2005 Specific 1159 658 46.4 2,4,4,1 16.5 0.874 17.6 0.857 2.64 a For description of equipment, scanning, sample and spectral treatment and calibration methods, see Nousiainen et al. 2004 b Including on-farm produced grass, legume and whole crop silages c Reference D-values were calculated as D-value = OM × OMD and pepsin-cellulase organic matter solubility was used to predict OMD either with general or species specific correction equation (see Table 4) d Standard deviation of the reference population e Standard error of calibration f Standard error of cross validation (see Nousiainen et al. 2004) Table 12. Comparison of total prediction performance of three near infrared reflectance spectroscopy D-value calibrations applied to primary growth (PG) grass, regrowth (RG) grass and legume silages and within forage species prediction performance with a calibration using forage-specific OMS equation as reference method. Intercept Slope R2 MSPEb Distribution of MSPE Bias Slope Random Calibrationa 2003 –19 1.02 0.623 31.9 0.04 0.00 0.96 2004 97 0.85 0.689 29.3 0.00 0.06 0.94 2005 9 0.99 0.783 23.7 0.00 0.00 1.00 Calibration 2005 PG grass 30 0.97 0.902 19.9 0.28 0.01 0.71 RG grass –68 1.08 0.688 22.8 0.50 0.01 0.49 Legume 100 0.85 0.661 29.6 0.03 0.06 0.92 a See Table 11 b Mean squared prediction error = √(∑(Observedin vivo−PredictedNIRS)2/n) 316 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation indicated by decreases in under-prediction of PG silage from 13.6 to 10.7 and in over-prediction of RG silages from 34.5 to 16.3 g kg-1 DM. Conse- quently the proportion of bias for both PG (0.28) and RG (0.50) silage is still large (Table 12). The harvesting year had a significant influence on NIRS prediction errors for D-value (Fig. 9). Apparently there are at least three sources or errors behind the year effects; animal differences in the in vivo experiments conducted in different years (1), variations in laboratory analyses between years (2) and variation in environmental conditions that may affect forage composition (3). Moreover, the ef- fects of number of harvest and harvesting year seem to be additive for grass silages. With the last calibration (2005, Table 6) prediction error in D- value decreased to 14.8 g kg-1 DM when correcting results for year within forage effect. This figure is only slightly higher than the residual variation in digestibility trials of this data (13.8; Nousiainen 2004), but it is questionable whether the year ef- fects within forage types can be totally excluded. However, it should be possible to reduce the pre- diction error to 16–17 g kg-1 DM with grass silag- es, i.e. to that attained without the mean bias error for the PG and RG silages, respectively. It appears that despite attempts to use forage specific OMS equations, the bias between forage types still remained rather high. This may at least partly be associated with errors in coding the har- vest (primary vs. regrowth) of on-farm samples used in the calibration data. Reference methods less dependent on forage type such as iNDF and summative models, or dynamic models in the fu- ture, may reduce this problem. To evaluate the potential of different methods in calibrating NIRS for D-value prediction of on- farm forages, a comparison between in vivo, OMS, iNDF and summative model was made (Table 13). The reference values were based on either general or specific equations. All reference methods re- sulted in good calibration statistics (calibration R2 > 0.96 and for cross-validation R2 > 0.91; results not shown). This is in good agreement with the re- sults presented previously by Nousiainen (2004), and describes the good precision of NIRS. The to- tal prediction error (Observedin vivo minus Predict- edNIRS) was lowest for specific OMS and highest for general OMS. A noteworthy feature of the cali- brations is that the difference in prediction error compared to OMS calibrations based on large data of on-farm samples (see Tables 12 and 13) is due to lower bias error. This describes the potential in improving the existing OMS calibration (2005, Ta- ble 12) either by correction the errors in coding the number of harvest or using iNDF or summative models as a reference. Interpretation of sources of  errors in determining D-value  of forages Provided that the true digestibility of NDS is unity and that the metabolic OM is constant, the varia- tion in OMD is related only to pdNDF digestibil- 0 10 20 30 40 50 A B C D Prediction error (g kg-1 DM) 2003 2004 2005 Fig. 9. Total D-value prediction errors (g kg-1 DM) of the three NIRS calibrations (see table 12): A = standard devia- tion of the reference population, B = NIRS calibration [=√(∑(Observedin vivo−PredictedNIRS)2/n)], C = B and varia- tion between forage types excluded, D = B and variation between year(forage) excluded. 317 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. Table 13. Comparison of near infrared reflectance spectroscopy calibration methods in predicting forage D-value based on experimental foragesa. Reference for calibrationb PredictedNIRS vs. observedin vivo Mean bias, g kg-1 DM MSPE distribution Forage-specific MSPE g kg-1 DM Slope Intercept R2 MSPEc, g kg-1 DM Bias Slope Random PG grass RG grass Legume Whole crop In vivo 0.97 17 0.946 12.0 –0.1 0.000 0.012 0.988 10.9 9.5 10.3 16.9 OMS Sd 1.00 3 0.903 16.0 –0.4 0.001 0.000 0.999 15.9 16.1 16.0 17.2 OMS Gd 1.01 –4 0.780 24.2 –1.9 0.006 0.000 0.994 26.9 23.8 25.6 16.1 iNDF S 1.01 –7 0.891 17.0 0.9 0.003 0.001 0.997 15.7 17.9 16.7 18.3 iNDF G 1.00 3 0.863 19.1 –0.2 0.000 0.000 1.000 17.1 21.1 19.1 19.6 Summative S 1.02 –14 0.894 16.8 0.9 0.003 0.003 0.994 15.3 18.5 16.8 16.9 Summative G 1.12 –75 0.886 18.1 –0.5 0.001 0.077 0.923 19.0 17.3 18.2 18.9 a Data comprising of silages made from primary (PG) or regrowth (RG) grass and leguminous or whole crop forages b For description of equipment, scanning, sample and spectral treatment and calibration methods see Nousiainen et al. 2004 c Mean squared prediction error = √(∑(Observedin vivo−PredictedNIRS)2/n) d S = Forage type specific equations; G = General equation for all forages. ity. If pdNDF concentration would be predicted accurately, prediction error of OMD in the present data set would be 0.0162. This error is partly re- lated to the systematic and significant differences in faecal metabolic OM output between the forage types (Table 2), and partly to random variation in endogenous faecal output. Using forage specific equations decreased the prediction error of the faecal endogenous OM to 8.6 g kg-1. This value may be considered as the potential minimum error of the laboratory methods in estimating forage OMD. According to Van Soest (1994), the minimum variability in carefully conducted digestibility tri- als is 0.020. Nousiainen (2004) reported a value of 0.0138 from studies included in the present data. As suggested by Van Soest (1994), a difference of 0.020 in digestibility can be taken as the lower limit of biological significance of digestibility of feeds. This difference corresponds to a difference of about 1 kg d-1 in milk yield or that almost 2 kg d-1 more concentrates should be fed to compensate for the lower silage digestibility (Rinne 2000). Errors in the in vivo OMD predicted with dif- ferent methods can result from systematic errors between the forage types, random errors between the trials and random errors in determination of in vivo digestibility within trials. Contribution of the forage type on the prediction errors of OMD were analysed by one-way ANOVA using the GLM pro- cedure of SAS (1999). The significance of random study effect was tested using a mixed model analy- sis with a fixed effect of forage type and a random study effect. Possible contribution of the random variability of the in vivo trials was estimated as a relationship between the errors using a mixed model regression analysis with a random study ef- fect. It may be assumed that a strong correlation of the errors in OMD predictions is at least partly re- lated to random variability of the in vivo data. Data estimated using forage specific equations was used for this analysis. When the general prediction equations were used for all forage types, the residual mean squared prediction errors were in many cases significantly different from zero, i.e. the prediction accuracy was dependent on the forage type (Table 14). The OMS method underestimated the in vivo D-value of PG silages and overestimated that of RG and whole-crop silages. Lignin either markedly over- estimated (RG grass and whole-crop silages) or underestimated (PG grass and legume silages) D- 318 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation value. The differences in the mean bias were large between the forage types irrespective of the pre- diction being based on empirical relationship or summative approaches. The greater effect of for- age type with lignin may be related to differences in the ratio between lignin and iNDF among the forage types, as discussed by Nousiainen et al. (2004). The behaviour of iNDF and summative models was more uniform among forage types. The random study effect was significant for OMS, iNDF and Mertens (2002b) summative equation based on iNDF, when the variation result- ing from forage type was excluded. These effects may be attributed to differences between the ani- mals used in the trials, differences in the activity of enzymes used in the determination of OMS and to differences in microbial activity in the rumen of the cows used for determination of the iNDF con- centration of the forages. The study effect was not significant for lignin equations, which may be re- lated to a greater variability of the errors being pre- sumably attributable to high random variation in lignin analysis and/or climatic factors influencing cell wall lignification. All prediction errors were significantly (P < 0.01) correlated with each other within a trial (Ta- ble 15). The relationship was strongest between the methods based on iNDF and summative sys- tem (Mertens 2002b; iNDF), probably because of the strong influence of iNDF on predicted OMD in the summative system. Significant relationships between the prediction errors within a trial even when the most contrasting systems (e.g. OMS vs. iNDF and OMS vs. summative system) were com- pared, suggests that random errors of the in vivo digestibility determinations had some contribution to the overall prediction error. If one method over- estimated the in vivo OMD of a feed, the probabil- ity that another method also overestimated it, was high. Corresponding conclusions were made by Rinne et al. (2006) from similar analysis of legume silage data. Based on the mean standard error (0.0138) of OMD of the present data and 4 sheep per feed, confidence interval of P = 0.90 will be ±0.023. Calculating a reference value as a weighted mean of in vivo OMD (0.50) and the mean of three other laboratory methods (0.50), i.e. excluding the method being evaluated, led to markedly reduced prediction errors of OMD. The mean squared pre- diction errors were 0.013, 0.011, 0.021 and 0.011 for OMS, iNDF, lignin and the summative system, respectively, when the values were based on forage specific equations. Except for lignin, these values are even slightly lower than the standard error of the in vivo data (Nousiainen 2004) and close to the theoretical minimum of about 0.008, when all the variation results from random variation in faecal Table 14. Mean prediction errorsa of D-value for primary growth and regrowth grass, legume and whole crop silages predicted using different laboratory techniques. Method Primary growth Regrowth Legume Whole crop RMSE P-value OMSb –19.9f 16.1 1.8 24.8 14.9 <0.01 iNDFc 1.4 5.6 –9.6 –2.2 16.3 <0.05 Lignin 20.8 –29.6 22.8 –39.2 28.0 <0.01 NRC (2001) 75.8 45.9 39.2 –34.0 23.5 <0.01 Mertens (2002b)d 28.0 –20.2 3.0 –48.1 25.0 <0.01 Mertens (2002b)e 6.1 –1.9 –0.5 –10.5 16.0 <0.10 a (Observed – Predicted) b Pepsin-cellulase solubility of organic matter c Indigestible fibre (determined by 12 d in situ incubation) d Estimated from lignin e Estimated from iNDF f Values printed in bold are statistically different from zero (P < 0.05) 319 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. Table 15. Relationships between the prediction errors of organic matter digestibility estimated by organic matter pepsin- cellulase solubility (OMS), indigestible neutral detergent fibre (iNDF) and lignin using the forage specific equations. Y Variable X Variable A s.e.a B s.e. P-value Adj. R2 OMS iNDF 0.6 2.09 0.77 0.446 <0.01 0.391 OMS Lignin 0.2 2.19 0.93 0.268 <0.01 0.228 OMS Summativeb 0.8 2.07 0.69 0.592 <0.01 0.578 iNDF Lignin –0.2 2.07 0.93 0.311 <0.01 0.459 iNDF Summativeb 0.5 0.63 0.39 0.969 <0.01 0.890 Lignin Summativeb 2.1 2.79 0.45 1.038 <0.01 0.514 a Standard error b Mertens (2002b) equation with iNDF endogenous OM. Correlation between the errors originating from different prediction methods (1), relatively large confidence interval of OMD even in carefully conducted digestibility trials (2) and markedly lower prediction errors when the refer- ence value was based on a weighted mean of in vivo and other laboratory methods (3) all suggest that the true prediction error of the laboratory methods is likely to be smaller than the calculated errors suggest. Interestingly, the prediction errors of NIRS D-values were strongly correlated with the prediction errors of laboratory methods (Fig. 10). For OMS, this is partly attributed to using OMS predicted D-values as a reference method for NIRS calibrations, i.e. errors in reference values would automatically reflect errors in predicted val- ues. However, the highly significant (P < 0.01) re- lationship between the errors within forage(year) strongly supports the earlier suggestion about ran- dom errors of the in vivo values. Consequently, the true errors of both the laboratory methods and NIRS are likely to be smaller than the estimated errors. Implications The present re-evaluation based on a systemati- cally collected dataset confirmed the weaknesses of the proximate feed analysis. The revised deter- Unadj. = 0.85x - 1.2 R2 = 0.298 Adj. = 0.62x - 0.4 R2 = 0.411 -80 -60 -40 -20 0 20 40 60 -40 -20 0 20 40 60 Obs-Pred Mertens (g kg-1 DM) Obs-Pred NIRS (g kg -1 DM) Unadj Adj. Fig. 10. Relationships between errors of D-value (Ob- served – Predicted) estimated by the summative model (Mertens 2002b) and by NIRS calibrated with forage-spe- cific organic matter pepsin-cellulase solubility (calibration 2005). gent system should be used instead. Predicting in vivo organic matter digestibility with the empirical equations using chemical parameters gave unsatis- factory results. Pepsin-cellulase solubility predict- ed forage OM digestibility with an acceptable ac- curacy but the drawback of the method is the for- age type, environmental and laboratory dependen- cy. To reduce the D-value prediction error further, regression equations based on indigestible NDF or 320 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation summative models using uniform feed fractions from the detergent analysis and long-term in situ ruminal incubation may be used. These methods are also an interesting alternative to pepsin-cellu- lase solubility as a reference for NIRS applications in practical feed evaluation. Another option in the future may be dynamic models. However, a vital prerequisite for using dynamic models in practical feed evaluation is that iNDF and kd can be easily and reliably determined from on-farm forages. Al- though a NIRS prediction equation for iNDF will be adopted in practical use in the near future in Finland, the respective methodology for kd war- rants further research. References Ahvenjärvi, S., Vanhatalo, A., Huhtanen, P. & Varvikko, T. 2000. �eter�ination o�� reti�ulo�ru�en an�� �hole��eter�ination o�� reti�ulo�ru�en an�� �hole� sto�a�h ��igestion in la�tating �o�s by o�asal �anal or ��uo��enal sa�pling. British Journal of Nutrition 83: 67– 77. Allen, M.S. & Mertens, �.R. 1988. �valuating �onstraints o���valuating �onstraints o�� fibre ��igestion by ru�en �i�robes. Journal of Nutrition 118: 261–270. Barnes, R.J., �hanoa, M.S. & Lister, S.J. 1989. Stan��ar��Stan��ar�� nor�al variate trans��or�ation an�� ��e�tren��ing o�� near� in��rare�� ��i����use refle�tan�e spe�tra. Applied Spectros- copy 43: 772–777. Beever, �.�. & Moul��, F.L. 2000. Forage evaluation ��or e��fi� �ient ru�inant livesto�k pro��u�tion. In: Givens, �.I. et al. (e��s.). Forage evaluation in ruminant nutrition. CABI Publishing, Oxon. p. 15–42. Bibby, J. & Toutenburg, H. 1977. Prediction and improved estimation of linear models. �iley, Lon��on, �.�. 188 p.�iley, Lon��on, �.�. 188 p. Cherney, J.H. & Cherney, �.J.R. 2003. Assessing silage quality. In: Buxton, �.R. et al. (e��s.). Silage s�ien�e an�� te�hnology. ASA, CSSA, SSSA, Ma��ison, �I. ASA Monograph Series No. 42. p. 141–198. Cole�an, S.�., Lippke, H. & Gill, M. 1999. �sti�ating the�sti�ating the nutritive value o�� ��orages. In: Jung, H.J.G. & Fahey, G. C., Jr. (e��s.). Nutritional ecology of herbivores. Pro� �ee��ings o�� the 5th International Sy�posiu� on the Nutrition o�� Herbivores. A�eri�an So�iety o�� Ani�al S�ien�e, Savoy, �SA. p. 647– 695. Conra��, H.R., �eiss, �.P., O��gon�ongon, �.O. & Sho�key, �.L. 1984. �sti�ating net energy o�� la�tation ��ro� �o�� ponents o�� �ell solubles an�� �ell �alls. Journal of Dairy Science 67: 427–436. �an��ær, A., Huhtanen, P., ���én, P., Sveinbjörnsson, J. & Vol��en, H. 2005. �aroline – a Nor��i� �o� �o��el ��or ��ee�� evaluation – �o��el ��es�ription. In: �ebreab, �. etIn: �ebreab, �. et al. (e��s.). Nutrient digestion and utilization in farm ani- mals: modelling approaches. CAB International. p. 407–415. �eaville, �.R. & Flinn, P.C. 2000. Near�in��rare�� (NIR) spe�� tros�opy: an alternative approa�h ��or the esti�ation o�� ��orage quality an�� voluntary intake. In: Givens, �.I. et al. (e��s.). Forage evaluation in ruminant nutrition. CABI Publishing, Oxon. p. 301–320. �einu�, B., Van �s, A.J.H. & Van Soest, P.J. 1968. Cli�ate,Cli�ate, nitrogen an�� grass. II. The influen�e o�� light intensity, te�perature an�� nitrogen on in vivo ��igestibility o�� grass an�� pre��i�tion o�� these e����e�ts ��ro� so�e �he�� i�al pro�e��ures. Netherlands Journal of Agricultural Science 16: 217–223. �llis, �.C., Poppi, �.P., Matis, J.H., Lippke, H., Hill, T.M. & Rouguette, F.M., Jr. 1999. �ietary – ��igestive – �eta� boli� intera�tions ��eter�ining the nutritive potential o�� ru�inant ��iets. In: Jung, H.J.G. & Fahey, G.C., Jr. (e��s.).In: Jung, H.J.G. & Fahey, G.C., Jr. (e��s.).(e��s.). Nutritional ecology of herbivores. Pro�ee��ings o�� the 5th International Sy�posiu� on the Nutrition o�� Herbi� vores. A�eri�an So�iety o�� Ani�al S�ien�e. p. 423– 481.. Frie��el, �. 1990. �ie S�hätzung ��es energetis�hen Futter� �ertes von Grob��utter �it Hil��e einer Cellulase�etho� ��e. (The esti�ation o�� the energeti� ��ee��ing value o��The esti�ation o�� the energeti� ��ee��ing value o�� roughages by �eans o�� a �ellulase �etho��). Wissen- schaftliche Zeitschrift Universitet Rostock, N-Reihe 39: 78–86. Fylstra, �., Las��on, L., �atson, J. & �aren, A. 1998. �esign an�� use o�� the Mi�roso��t �x�el Solver. Interfaces 28: 29–55. Goering, H.R. & Van Soest, P.J. 1970. Forage fiber analysis.Forage fiber analysis. �S�A, �ashington. Agricultural Handbook No. 379. 22 p. Hellä�äki, M. 1992. �sti�ation o�� the nutritive value o�� si� lage using NIR ��igestibility. In: Murray, I. & Co�e, I.A. (e��s.). Making light work: advances in near infrared spectroscopy. �evelope�� ��ro� the 4th International Con��eren�e on Near In��rare�� Spe�tros�opy, Aber��een, S�otlan��, 19–23 August 1991. VCH Publishers, �ein� hei�; Ne� York; Basel; Ca�bri��ge. p. 264–268. Huhtanen, P. 2003. Fa�tors influen�ing on voluntary intake o�� silage�base�� ��iets, an�� the responses o�� silage qual� ity in �ilk pro��u�tion. In: Gar�o, T. (e��.). Proceedings of the International symposium: Early harvested forage in milk and meat production. �ringler, Nannesta��, Nor��ringler, Nannesta��, Nor� �ay, �s. p. 59–82.�s. p. 59–82. Huhtanen, P., Ahvenjärvi, S., �eisbjerg, M.R. & Nørgaar��, P. 2006. �igestion an�� passage o�� fibre in ru�inants. In:In: Sejrsen, �. et al. (e��s.). Ruminant physiology. �agenin��agenin� gen A�a��e�i� Publishers, The Netherlan��s. p. 87– 135. Huhtanen, P., �austell, �. & Jaakkola, S. 1994. The use o�� internal �arkers to pre��i�t total ��igestibility an�� ��uo��e� nal flo� o�� nutrients in �attle given six ��i����erent ��iets. Animal Feed Science and Technology 48: 211–227. Huhtanen, P., �halili, H., Nousiainen, J.I., Rinne, M., Jaak� kola, S., Heikkilä, T. & Nousiainen, J. 2002. Pre��i�tionPre��i�tion o�� the relative intake potential o�� grass silage by ��airy �o�s. Livestock Production Science 73: 111–130. Huhtanen, P., Nousiainen, J. & Rinne, M. 2005. Pre��i�tionPre��i�tion 321 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. o�� silage �o�position an�� organi� �atter ��igestibility ��ro� herbage �o�position an�� pepsin��ellulase solu� bility. Agricultural and Food Science 14: 154–165. Huhtanen, P., Ots, M., Ahvenjärvi, S. & Rinne, M. 2001. Pre�Pre� ��i�tion o�� the in vivo ��igestibility o�� grass silage ��ro� gas pro��u�tion kineti�s. Journal of Dairy Science 84, Suppl. 1: 197. Huhtanen, P., Vanhatalo, A. & Varvikko, T. 1998. �nzy�e�nzy�e a�tivities o�� ru�en parti�les an�� ��ee�� sa�ples in�u� bate�� in situ �ith ��i����ering types o�� �loth. British Journal of Nutrition 79: 161–168. Illius, A.�. & Allen, M.S. 1994. Assessing ��orage quality us�Assessing ��orage quality us� ing integrate�� �o��els o�� intake an�� ��igestion by ru�i� nants. In: Fahey, G.C., Jr. et al. (e��s.). Forage quality, evaluation and utilization. A�eri�an So�iety o�� Agrono� �y. p. 869–890. Jones, �.I.H. & Theo��orou, M.�. 2000. �nzy�e te�hniques ��or esti�ating ��igestibility. In: Givens, �.I. et al. (e��s.). Forage evaluation in ruminant nutrition. CABI Publis�CABI Publis� hing, Oxon. p. 155–173. �le�etti, I., Hellä�äki, M. & Nousiainen, J.1995. Pre��i�tionPre��i�tion o�� ��igestible organi� �atter in grass silage by near in� ��rare�� refle�tan�e spe�tros�opy. In: Fourth International Sy�posiu� on the Nutrition o�� Herbivores, Cler�ont� Ferran��, 11–15 Sept 1995. Annales de Zootechnie 44, Suppl. 1: 42.1: 42. Las��on, L., �aren, A., Jain, A. & Ratner, M. 1978. �esign an�� testing o�� generalize�� re��u�e�� gra��ient �o��e ��or nonlinear progra��ing. ACM Transactions on Mathe- matical Software 4: 34–49. Lun��, P., �eisbjerg, M.R., Ahvenjärvi, S., Huhtanen, P., ���en, P., Ola��sson, B. & Vol��en, H. 2004. Nor��i� ringtest on IN�F �ontent an�� N�F ��egra��ation �hara�teristi�s in three ��ee��s. In: The 10th International sy�posiu� on ru�inant physiology, 30 Aug – 4 Sept 2004, Copenha� gen, �en�ark: short papers. Journal of Animal and Feed Sciences 13, Suppl. 1: 139–142. M�Queen, R. & Van Soest, P.J. 1975. Fungal �ellulase an��Fungal �ellulase an�� he�i�ellulase in pre��i�tion o�� ��orage ��igestibility. Jour- nal of Dairy Science 58: 1482–1491. MAFF 1975. Nutrient requirements of ruminants. Ministry o�� Agri�ulture, Fisheries an�� Foo��. Technical Bulletin 33. 79 p. Mertens, �.R. 1993. �ineti�s o�� �ell �all ��igestion an�� pas��ineti�s o�� �ell �all ��igestion an�� pas� sage in ru�inants. In: Jung, H.G. et al. (e��s.). Forage cell wall structure and digestibility. A�eri�an So�iety o�� Agrono�y. p. 535–570. Mertens, �.R. 2002a. Gravi�etri� ��eter�ination o�� a�yla�Gravi�etri� ��eter�ination o�� a�yla� se�treate�� neutral ��etergent fiber in ��ee��s �ith refluxing in beakers or �ru�ibles: �ollaborative. Journal of AOAC International 85: 1217–1240.85: 1217–1240. Mertens, �.R. 2002b. Fiber: �easuring, �o��elling an��Fiber: �easuring, �o��elling an�� ��ee��ing. In: Proceedings Cornell Nutrition Conference for Feed Manufacturers. 23–25 O�t 2002. �ast Syra� �use, NY. p. 1–18. MTT 2006. Rehutaulukot ja ruokintasuositukset (Fee�� ta� bles an�� ��ee��ing re�o��en��ations). Jokioinen: MTT Agri��oo�� Resear�h Finlan��. �p��ate�� 14 Feb 2006. Cit� e�� 16 June 2006. Available on the Internet: http://���. agronet.fi/rehutaulukot/. �RN:NBN:fi���e20041449.�RN:NBN:fi���e20041449. Nousiainen, J. 2004. �evelop�ent o�� tools ��or the nutritional �anage�ent o�� ��airy �o�s on silage�base�� ��iets. A�a� ��e�i� ��issertation. University of Helsinki, Department of Animal Science Publications 72, 61 p. + 5 en�l. Avail� able on the Internet: http://ethesis.helsinki.fi/julkaisut/ �aa/kotie/vk/nousiainen/ Nousiainen, J. Ahvenjärvi, S., Rinne, M., Hellä�äki, M. & Huhtanen, P. 2004. Pre��i�tion o�� in��igestible �ell �allPre��i�tion o�� in��igestible �ell �all ��ra�tion o�� grass silage by near in��rare�� refle�tan�e spe�tros�opy. Animal Feed Science and Technology 115: 295–311. Nousiainen, J., Rinne, M., Hellä�äki, M. & Huhtanen, P. 2003a. Pre��i�tion o�� the ��igestibility o�� the pri�ary gro�th o�� grass silages harveste�� at ��i����erent stages o�� �aturity ��ro� �he�i�al �o�position an�� pepsin��ellu� lase solubility. Animal Feed Science and Technology 103: 97–111. Nousiainen, J., Rinne, M., Hellä�äki, M. & Huhtanen, P. 2003b. Pre��i�tion o�� the ��igestibility o�� the pri�ary gro�th an�� regro�th grass silages ��ro� �he�i�al �o�� position, pepsin��ellulase solubility an�� in��igestible �ell �all �ontent. Animal Feed Science and Technology 110: 61–74. Norris, �.H., Barnes, R.F., Moore, �.�. & Shenk, J.S. 1976. Pre��i�ting ��orage quality by in��rare�� refle�tan�e spe�� tros�opy. Journal of Animal Science 43: 889– 897. NRC 2001. Nutrient requirements of dairy cattle. National Resear�h Coun�il. 7th revise�� ���ition. National A�a��� e�y Press, �ashington, �C. 381 p. Palohei�o, L. 1953. So�e persistent �is�on�eptions �on� �erning the �ru��e fiber an�� the nitrogen ��ree extra�t. Maataloustieteellinen aikakauskirja 25: 16–22. Palohei�o, L. & Palohei�o I. 1949. On the esti�ation o�� theOn the esti�ation o�� the total o�� vegetable �e�brane substan�es. Journal of the Scientific Agricultural Society in Finland 21: 1–16. Palohei�o, L., Syrjälä, L. & Vainio, �.A. 1968. Stu��ies onStu��ies on ��ae�es. Journal of the Scientific Agricultural Society in Finland 40: 237–246. Palohei�o, L. & Vainio, �.A. 1965. �eter�ination o�� the �o�plex o�� �ell �all substan�es in plant pro��u�ts. Jour- nal of the Scientific Agricultural Society in Finland 37: 305–312. Reeves, J.B., III. 2000. �se o�� near in��rare�� refle�tan�e spe�tros�opy. In: �’Mello, J.P.F. (e��.). Farm animal me- tabolism and nutrition: critical reviews. CAB Interna� tional. p. 185–208. Rinne, M. 2000. Influen�e o�� the ti�ing o�� the harvest o�� pri�ary grass gro�th on herbage quality an�� subse� quent ��igestion an�� per��or�an�e in the ru�inant ani� �al. University of Helsinki, Department of Animal Sci- ence. Publications 54. 42 p. + 5 en�l. A�a��e�i� ��isser� tation. Available on the Internet: http://ethesis.helsinki. fi/julkaisut/�aa/kotie/vk/rinne. Rinne, M., Huhtanen, P. & Jaakkola, S. 2002. �igestive pro��igestive pro� �esses o�� ��airy �o�s ��e�� silages harveste�� at ��our stages o�� grass �aturity. Journal of Animal Science 80: 1986–1998. Rinne, M., Olt, A., Nousiainen, J. Seppälä, A., Tuori, M., Paul, C., Fraser, M.�. & Huhtanen, P. 2006. Pre��i�tionPre��i�tion o�� legu�e silage ��igestibility ��ro� various laboratory �etho��s. Grass and Forage Science 61. (In press). Robertson, J.B. & Van Soest, P.J. 1981. The ��etergent sys� te� o�� analysis an�� its appli�ation to hu�an ��oo��s. In: Ja�es, �.P.T. & Thean��er, O. (e��s.). The analysis of 322 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Huhtanen, P. et al. Forage evaluation dietary fiber in food. Ne� York: Mar�el �ekker. p. 123– 58. Russell, J.�., Murray, I. & Frazer, A.R. 1989. Near an�� �i��� in��rare�� stu��ies o�� the �ell �all stru�ture o�� �ereal stra� in relation to its ru�en ��egra��ability. In: Chesson, A. & Ørskov, �.R. (e��s.). Physico-chemical characterisation of plant residues for industrial and feed use. �lsevier Applie�� S�ien�e, Lon��on. p. 13–24. SAS Institute In�. 1999. SAS/STAT, User�s guide.User�s guide. Version 8, SAS Institute In�., Cary, NC, �SA. S�hofiel��, P. 2000. Gas pro��u�tion �etho��s. In: �’Mello, J.P.F. (e��.). Farm animal metabolism and nutrition. CAB International, Oxon. p. 209–232. Steg, A., Spoelstra, S.F., Van ��er Meer, J.M. & Hin��le, V.A. 1990. �igestibility o�� grass silage. Netherlands Journal of Agricultural Science 38: 407–422. Ta��inga, S., Robinson, P.H., Vogt, M. & Boer, H. 1989. Ru�en ingesta kineti�s o�� �ell �all �o�ponents in ��airy �o�s. Animal Feed Science and Technology 25: 89–98.25: 89–98. Tilley, J.M.A. & Terry, R.A. 1963. A t�o stage te�hnique ��or the in vitro ��igestion o�� ��orage �rops. Journal of the Brit- ish Grassland Society 18: 104–111. Traxler, M.J., Fox, �.G., Van Soest, P.J., Pell, A.N., Las�ano, C.�., Lanna, �.P.�., Moore, J.�., Lana, R.P., Velez, M. & Flores, A.1998. Pre��i�ting ��orage in��igestible N�F ��ro� lignin �on�entration. Journal of Animal Science 76: 1469–1480. Van Soest, P.J. 1967. �evelop�ents o�� a �o�prehensive�evelop�ents o�� a �o�prehensive syste� o�� ��ee�� analysis an�� its appli�ation to ��orages. Journal of Animal Science 26: 119–128. Van Soest, P.J. 1975. Physi�o��he�i�al aspe�ts o�� fibre ��i�Physi�o��he�i�al aspe�ts o�� fibre ��i� gestion. In: M��onal��, I.�. & �arner, A.C.I. (e��s.). Di- gestion and metabolism in ruminant. Ne� �nglan�� �ni� versity Press, Ar�i��ale, Australia. p. 352–365. Van Soest, P.J. 1994. Nutritional ecology of the ruminant. Se�on�� ���ition. Co�sto�k Publishing Asso�iates, Cor� nell �niversity Press, Itha�a an�� Lon��on. 476 p. Van Soest, P.J., Robertson, J.B. & Le�is, B.A. 1991. Met� ho��s ��or ��ietary fiber, neutral ��etergent fiber an�� nonstar�h polysa��hari��es in relation to ani�al nutri� tion. Journal of Dairy Science 74: 3583–3597. Van Soest, P.J., Van A�burgh, M.�., Robertson, J.B. & �naus, �.F. 2005. Vali��ation o�� the 2.4 ti�es lignin ��a�� tor ��or ulti�ate extent o�� N�F ��igestion, an�� �urve peel� ing rate o�� ��er�entation �urves into pools. In: Proceed- ings Cornell Nutrition Conference for Feed Manufactur- ers. 24–26 O�t 2005. �ast Syra�use, NY. p. 139–149. Van Soest, P.J. & �ine, R.H. 1967. �se o�� ��etergents in the analysis o�� fibrous ��ee��s. IV.�eter�ination o�� plant �ell� �all �onstituents. Journal of AOAC International 50: 50–55. �eisbjerg, M.R., Hvelpun��, T. & Søegaar��, �. 2004. Pre��i��Pre��i�� tion o�� ��igestibility o�� neutral ��etergent solubles using the Lu�as prin�iple. Journal of Animal and Feed Sci- ences 13, Suppl. 1: 239–242.1: 239–242. �eiss, �.P. 1994. �sti�ation o�� ��igestibility o�� ��orages by laboratory �etho��s. In: Fahey, G.C., Jr. (e��.). Forage quality, evaluation and utilization. A�eri�an So�iety o�� Agrono�y, Ma��ison, �I. p. 644–681. �eiss, �.P., Conra��, H.R. & St. Pierre, N.R. 1992. A theo� reti�ally�base�� �o��el ��or pre��i�ting total ��igestible nu� trients o�� ��orages an�� �on�entrates. Animal Feed Sci- ence and Technology 39: 95–110. �il�an, �., Foulkes, G.R. & Givens, �.I. 1996a. A �o�pari� son o�� ��our �etho��s o�� esti�ating the rate an�� extent o�� �ell �all ��egra��ation in grass silages. Animal Feed Sci- ence and Technology 63: 99–109. �il�an, �., Foulkes, G.R. & Givens, �.I. 1996b. The rate an�� extent o�� �ell��all ��egra��ation in vitro ��or 40 silages varying in �o�position an�� ��igestibility. Animal Feed Science and Technology 63: 111–122. �illia�s, P. & Norris, �. 1987. Near-infrared technology in the agricultural and food industries. A�eri�an Asso�. o�� Cereal Che�ist, In�., St. Paul, Minnesota, �SA. 330 p. �in��ha�, �.R., Mertens, �.R. & Barton, F.�., II. 1989. Pro� to�ol ��or NIRS �alibration: Sa�ple sele�tion an�� equa� tion ��evelop�ent an�� vali��ation. In: Marten, G.C. et al.In: Marten, G.C. et al. (e��.). Near in��rare�� refle�tan�e spe�tros�opy (NIRS): analysis o�� ��orage quality. �S Gov. Print. O��fi�e, �ash� ington, �C. USDA-ARS Agricultural Handbook 643. p. 96–103. 323 A G R I C U L T U R A L   A N D   F O O D   S C I E N C E Vol. 15 (2006): 293–323. Systemaattisesti kerätyn säilörehuaineiston perusteella tehty yhteenveto osoittaa selvästi ns. virallisen rehu- analyysin eli Weenden analyysin biologiset puutteet re- hujen ravitsemuksellisen laadun kuvaajana. Analyysi ei kuvaa rehun kemiallisen koostumuksen ja sulavuuden välisiä syy-seuraussuhteita. Lisäksi tilastolliset yhteydet vaihtelevat huomattavasti eri kasvimateriaaleilla ja ym- päristöolosuhteissa. Weenden analyysin käyttöä ei siis voi suositella karkea- eikä väkirehujen laadun kuvaami- seen. In vitro pepsiini-sellulaasiliukoisuus (OMS) ja su- lamattoman kuidun (iNDF) pitoisuus sen sijaan ennusti- vat karkearehujen orgaanisen aineen sulavuuden riittä- vän tarkasti käytännön ruokinnansuunnittelua varten, edellyttäen että analyysitulokset muunnettiin sulavuu- deksi rehutyyppikohtaisia korjausyhtälöitä käyttäen eli erikseen ensimmäisestä sadosta ja jälkikasvusta tehdyil- le nurmisäilörehuille, palkokasvisäilörehuille ja koko- viljasäilörehuille. Detergenttikuituanalyysi, joka jakaa rehun kuiva-ai- neen liukoiseen ja lähes täysin käyttökelpoiseen solunsi- sällykseen (NDS) sekä liukenemattomaan kuituun (NDF), on Weenden analyysiä huomattavasti kehityskel- poisempi vaihtoehto. Kun kuituanalyysiin yhdistetään pitkä in situ pötsi-inkubaatio, rehun kuiva-aine saadaan jaettua kolmeen biologisesti mielekkääseen osaan: NDS, potentiaalisesti sulava kuitu (pdNDF) ja iNDF. Rehun D-arvo eli sulavan orgaanisen aineen pitoisuus kuiva-ai- neessa voidaan ennustaa ns. summatiivisella yhtälöllä. Yhtälössä lasketaan yhteen sulanut NDS, joka voidaan määrittää Lucasin yhtälöllä, ja sulanut kuitu (pdNDF-pi- toisuus × pdNDF:n sulavuus tai vaihtoehtoisesti NDF- pitoisuus × NDF:n sulavuus). Rehutyyppikohtaiset sum- matiiviset yhtälöt ennustivat karkearehujen D-arvon lä- hes yhtä hyvin kuin OMS ja iNDF. Kun koko aineistoa tarkasteltiin yhdessä, summatiiviset yhtälöt olivat pa- rempia kuin iNDF ja erityisesti OMS. Jos D-arvon ennustevirhe halutaan saada pienem- mäksi kuin 15 g/kg kuiva-ainetta, on käytettävä rehu- tyyppikohtaisia yhtälöitä riippumatta siitä, onko lasken- nan perusteena OMS, iNDF tai summatiivinen yhtälö. Toinen vaihtoehto tulevaisuudessa on dynaamisten mal- lien käyttö. Ne pystyvät samanaikaisesti huomioimaan kaksi tärkeää dynaamista prosessia, jotka rajoittavat re- hun sulatusta pötsissä eli kuidun virtaus- ja sulatusno- peuden. Dynaamisten mallien käyttö edellyttää kuiten- kin sitä, että rehuista voidaan helposti ja luotettavasti määrittää iNDF-pitoisuus ja kuidun sulatusnopeus. Maa- tilarehujen iNDF-määritys NIRS-menetelmällä toteutuu Suomessa lähiaikoina, mutta kuidun sulatusnopeuden määritys vaatii vielä lisätyötä. SELOSTUS Karkearehujen sulavuuden määrityksen viimeaikainen kehitys ja käytännön sovellukset Pekka Huhtanen, Juha Nousiainen ja Marketta Rinne Maa- ja elintarviketalouden tutkimuskeskus ja Valio Oy Recent developments in forage evaluation withspecial reference to practical applications Introduction Description of data Chemical methods in forage characterisation Methods to estimate foragedigestibility Interpretation of sources oferrors in determining D-valueof forages Implications References SELOSTUS