| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Agriculture and Agri-Food Canada, 2560 Hochelaga Bd., Sainte-Foy QC, G1V 2J3, Canada
* Corresponding author (belangergf{at}em.agr.ca)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: DM, dry matter • FDM, forage DM yield • FNA, forage N accumulation • FNC, forage N concentration • IVCWD, in vitro cell wall digestibility • IVTD, in vitro true digestibility • LSD, least significant difference • LWR, leaf weight ratio • NDF, neutral detergent fiber • NUE, N use efficiency • RWR, root weight ratio • PCA, principal component analysis • PC, principal component • TBDM, total biomass DM yield • TBNA, total biomass N accumulation • TBNC, total biomass N concentration
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
It is well accepted that nutritive value characteristics such as forage N concentration (FNC) and DM digestibility decrease with increases in DM yield (Bélanger and Richards, 1997; Bélanger and McQueen, 1998; Casler and Vogel, 1999). This suggests that cultivars with greater DM yield may have lower nutritive value. Hence, the improvement of both DM yield and nutritive value requires the identification of genotypes exhibiting a weaker negative relationship between DM yield and nutritive value.
Genetic variability for FDM and FNC was found among 40 half-sib families of timothy grown under field conditions (Michaud et al., 1998). Some families yielded as much as 20% more than the reference cultivar, Champ, suggesting that they were more efficient in using available soil N. Furthermore, high-yielding families with high and low N concentrations indicated genetic variability for the relationship between DM yield and N concentration and the possibility of increasing both DM yield and N concentration. Variability in the relationship between DM yield and N concentration was also observed for two genotypes selected for high and low FNC (Brégard et al., 2000). The greater N concentration of the former was attributed to a greater proportion of leaves in the forage; the greater forage DM yield of the genotype with low N concentration was the result of changes in biomass partitioning between shoots and roots because no difference in total biomass (shoot + root) DM yield (TBDM) was found between the two genotypes.
We hypothesized that differences in FDM among genotypes characterized in a previous field study (Michaud et al., 1998) are not related to TBDM differences, but to changes in the root to total biomass ratio (root weight ratio; RWR). We also hypothesized that high-yielding genotypes have low leaf to forage ratio (leaf weight ratio; LWR), and, therefore, high NDF concentration, low in vitro true digestibility (IVTD), low in vitro cell wall digestibility (IVCWD), and low N concentration. Our objectives were to test the above hypotheses under controlled conditions with limiting and nonlimiting N nutrition, and to establish relationships among biomass partitioning, forage nutritive value, and DM yield of contrasting genotypes of timothy. Results will reveal the potential for selecting independently for increased yield and NUE without sacrificing nutritive value. This trial was conducted in a controlled environment to allow complete root recovery and to minimize environmental variation effects on interrelated growth components.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Plant Material and Growth Conditions
The seven half-sib families used in this study were previously characterized in a field study and classified in three groups (Michaud et al., 1998). Three half-sib families (1318, 1263, and 1258) had a higher FDM than the reference cultivar Champ and high forage N concentration (FNC). Two half-sib families (1254 and 1313) also had a high FDM but were characterized by a low FNC. Two half-sib families (709 and 674) had a low FDM equivalent to that of the reference cultivar Champ but a high FNC. Along with the seven half-sib families, two populations of timothy obtained from two cycles of a divergent selection for high (N+) and low (N-) FNC and, as a reference, the cultivar Champ, were included. Populations or half-sib families are thereafter called genotypes.
The plants were grown in a Turface (AIMCOR, Deerfield, IL) medium, which is similar to vermiculite but with a higher cationic retention. Plants were started in a growth room by placing two seeds per pot (150 cm2 of soil area; 1 L of soil volume). Pots were 15 cm apart. Four weeks after seeding, plants were cut at a 5-cm height to promote tillering. Day and night temperatures were kept at 25 ± 5 °C and 15 ± 5°C, respectively. Light was provided for a 16-h day by cool white fluorescent and incandescent bulbs with a photosynthetic photon flux density of 350 µmol m-2 s-1 (between 400 and 700 nm) at plant height. Pots were watered to capacity every 2 d. The plants were fertilized twice a week with a complete nutrient solution containing a limiting (5 mg N plant-1 wk-1) or nonlimiting (25 mg N plant-1 wk-1) N rate. These N application rates were chosen based on a previous study conducted under controlled conditions (Brégard et al., 2000). Thus, at each fertilization, the plants received 50 mL of a nutrient solution containing 1.75 (limiting) or 8.75 (nonlimiting) mM NH4NO3, 4.4 mM MgSO4, 2.2 mM K2HPO4, 2.2 mM KH2PO4, 2.2 mM CaCl2, 0.45 mM ferric citrate and micronutrients (102 µM H3BO3, 20 µM MnCl2 · 4H2O, 1.683 µM ZnSO4 · 7H2O, 0.705 µM CuSO4 · 5H2O, 1.202 µM H2MoO4 · H2O, 0.037 µM CoCl2 · 6H2O). Forage was cut at heading stage at a 5-cm height and discarded for two successive regrowth cycles, that is 60 d and 90 d after seeding. Sampling was done at the end of the third regrowth cycle, that is 120 d after seeding.
Sampling and Analyses
Plants from each experimental unit were separated into leaves, stems (including sheaths of leaves surrounding the stems), stubble (0–5 cm above the soil surface), and roots. Root samples were stored between 0 and 4°C, for a maximum of 7 d following sampling, before being washed free of Turface with water. All samples were dried at 55°C for 2 d for determination of DM yield. Dried samples were ground using a Wiley mill (Wiley Laboratory mills, Philadelphia, PA) fitted with a 20-mesh (0.85 mm) screen before N analyses. Samples for NDF and digestibility analyses were ground twice to obtain a more uniform particle size. The FDM was calculated by adding leaf and stem DM yields. The TBDM was calculated by adding forage, stubble, and root DM yields.
Total N concentrations of leaves, stems, stubble, and roots were determined using a QuikChem Automated Analyzer (Zellweger Analytics, Inc., Lachat Instruments, Milwaukee, WI) after the H2SO4-H2O2 digestion method of Kjeldahl (adapted from Richards, 1993). Total N concentration determined in roots was probably underestimated because of leaching of soluble components during root washing. The FNC was estimated from stem and leaf concentrations and the total biomass N concentration (TBNC) was estimated from forage, stubble, and root concentrations, weighted on the basis of their respective DM yield. Forage N accumulation (FNA) and total biomass N accumulation (TBNA) were calculated as N concentration multiplied by DM yield. The RWR was calculated as root DM yield divided by TBDM. The LWR was calculated as leaf DM yield divided by FDM.
The IVTD was measured in leaves and stems using the method based on rumen fluid digestion, followed by a NDF determination of the post-digestion residues with the ANKOM filter bag system according to Wilman and Adesogan (2000). The rumen fluid digestion was done with the ANKOM Rumen Fermenter (Model No: Daisy II, Ankom Technology, Fairport, NY). Rumen fluid was obtained from a lactating ruminally fistulated dairy cow, fed a good quality cool-season grass mixture silage, corn (Zea mays L.) and barley (Hordeum vulgare L.) grain, and a protein concentrate according to its requirements. The animal was cared for according to the guidelines of the Canadian Council on Animal Care (1993). The NDF determinations of pre- and post-digestion samples were done using the ANKOM Fiber Analyzer (Model No: ANKOM 200, Ankom Technology, Fairport, NY). Sodium sulfite and 
–amylase were used in the NDF procedure. The initial sample dry weight for IVTD and NDF determinations was 0.25 g. The IVCWD was calculated from NDF values as follows, IVCWD (g kg-1) = 1 - (post-digestion NDF dry weight/pre-digestion NDF dry weight) x 1000. The NDF concentration, IVTD, and IVCWD of the forage were estimated from stem and leaf results weighted on the basis of their respective DM yield.
Statistical Analysis
All variables were subjected to analyses of variance using a split-plot experimental design with N rates as main plots and genotypes as subplots (GLM Procedure, SAS Institute, 1990, Cary, NC). Because of a significant interaction between N rate and genotype for all variables except two (Table 1), principal component analysis (PCA) was performed separately for each N rate, and the analysis of variance was recalculated for each N rate to estimate SEM and LSD values. Correlations between variables were computed on the mean value of the population, and the PCA was done on the means (Genstat 5 Committee, 1993). Significant differences between genotypes were determined by the LSD (5%) range, centered about the overall mean for each variable.
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
Nitrogen Concentration and Accumulation
Genotypes did not significantly differ in FNC and TBNC under limiting and nonlimiting N, although some genotypes tended to have similar characteristics to those reported in previous studies (Tables 2 and 3). Genotype 1318 tended to have a greater FNC than the mean of all genotypes under nonlimiting N as previously reported by Michaud et al. (1998). Also, the FNC of N+ was greater than that of N- by 14% under nonlimiting N and 11% under limiting N, which is similar to the findings of Brégard et al. (2000).
Genotypes significantly differ in FNA and TBNA only under nonlimiting N. The genotypes 1263, 1254, and 1318 had greater FNA and TBNA than Champ, 674, and 709 under nonlimiting N. The genotype N+ also had a superior TBNA. The greater N accumulation of N+ and 1318 can be attributed to some extent to a greater N concentration whereas the greater N accumulation of 1263 and 1254 is primarily due to a greater yield potential.
Biomass Partitioning
Genotypes significantly differed in biomass partitioning (Tables 2 and 3). Under limiting and nonlimiting N, the genotypes 1313 and N+ had a greater LWR than 674 and 709, and the genotype N+ had a greater RWR than 674 and N-. Brégard et al. (2000) also reported that N+ had greater LWR and RWR than N-. The genotypes 1254, 1258, 1263, and 1318 also tended to have greater LWR and RWR than 674 and 709, but the difference was not significant.
Neutral Detergent Fiber Concentration and Forage Digestibility
Genotypes differed significantly in NDF concentration, IVTD, and IVCWD under limiting N rate only (Tables 2 and 3). Genotypes 1313, 1254, 1318, and N+ had significantly lower NDF concentrations than N- and 674, whereas genotypes 1318 and N+ had greater IVTD and IVCWD than Champ, N-, and 1263 under limiting N (Table 3). The lack of genotype differences in NDF concentration, IVTD, and IVCWD under nonlimiting N may be due to the initial selection of the half-sib family parents for low NDF under a limited N stress (Michaud et al., 1998). The IVTD and IVCWD of N+ tended to be greater than those of N- and 1258 under nonlimiting N (Table 2).
Nitrogen Stress
All variables were significantly affected by N rates (P < 0.01), except NDF concentration, IVTD, and IVCWD (Table 1). Nitrogen rate significantly (P < 0.01) affected the accumulation and concentration of N and DM yield (Table 1). Under limiting N, FNA and TBNA were 18 and 24%, respectively, of those under nonlimiting N, while percentages for FDM and TBDM were 38 and 65% (Tables 2 and 3). Similar levels of N stress on DM yield may be observed in field studies (Bélanger et al., 1992a; Bélanger and Richards, 1997). The N accumulation was, therefore, more affected by the N stress than was biomass production. Hence, under N deficient conditions, plants make a more efficient use of absorbed N than under nonlimiting conditions, that is, they produce more biomass per unit of absorbed N. This explains why N concentration is lower in N stressed than non-N stressed plants.
Nitrogen stress modified biomass partitioning between shoots and roots, and leaves and stems. Under nonlimiting N, the RWR and LWR were, respectively, 47 and 89% of those under limiting N (Tables 2 and 3). Hence, a greater proportion of assimilates was exported to the root system when N was deficient and, to a lower extent, less assimilates were partitioned to stem growth, as reported by Robson and Parsons (1978), Bélanger et al. (1992b), Reynolds and D'Antonio (1996), and Bélanger and McQueen (1998). The increased LWR obtained with limiting N conditions in the present study was not sufficient to significantly affect NDF, IVTD, and IVCWD (Table 1). However, Bélanger and McQueen (1998) reported that N deficiency may increase IVTD and IVCWD and reduce NDF in timothy forage, but differences in LWR did not explain entirely the differences in nutritive value.
Relationship among Variables
Forage and Total Biomass Dry Matter Yield
We hypothesized that genotypes with high FDM would not necessarily have a high TBDM, and consequently, there would not be a strong relationship between FDM and TBDM. The results of the PCA indicated both negative and positive relationships between FDM and TBDM depending on the genotypes and the level of N stress (Fig. 1 and 2). Under limiting N, FDM was negatively correlated with TBDM in the first PC and positively correlated with TBDM in the second PC. The negative relationship was illustrated by the genotypes N- and 674 with high FDM and low TBDM, and by N+ with a low FDM and a high TBDM (Table 3). Genotypes 1254 and 1263 with high FDM and high TBDM, and Champ with low FDM and low TBDM illustrated the positive relationship between FDM and TBDM. When the two variables were considered without looking at the two components, FDM and TBDM were not correlated (r = 0.04) (Table 4). This highlights the benefit of using PCA to study the relationship among variables in the study of several genotypes. By using only the correlation between two variables, we would have concluded that there was no relationship between FDM and TBDM, in agreement with our hypothesis. The PCA, however, indicated that there is some relationship between FDM and TBDM for some genotypes.
|
|
|
Forage Dry Matter Yield, Leaf Weight Ratio, Neutral Detergent Fiber, In Vitro True Digestibility and In Vitro Cell Wall Digestibility
Leaf weight ratio is reported to decrease with increases in DM yield (Lemaire and Gastal, 1997). Furthermore, NDF concentration is negatively related to LWR (Bélanger and McQueen, 1997, 1999). We hypothesized, therefore, that genotypes with high FDM would have a low LWR, thus a high NDF concentration. Under limiting N, the negative correlation between FDM and LWR, and between NDF concentration and LWR was confirmed by N- and 674 with high FDM, low LWR, and high NDF concentration, and by 1318 and N+ with low FDM, high LWR, and low NDF concentration (Tables 3 and 4). The first PC illustrates this negative relationship between FDM and NDF concentration vs. LWR (Fig. 2).
Conversely, under nonlimiting N, FDM was positively correlated with LWR, and NDF was negatively correlated with LWR (Table 4; Fig. 1). Genotypes 674 and 709 illustrated the positive relationship with low FDM and low LWR (Table 2). Also, genotypes 1313, 1258, 1263, 1254 tended to have higher FDM and higher LWR than the mean values. The second component of the PCA, however, indicated that NDF was positively related to FDM for some genotypes.
Hence, we conclude that there is variability for the relationship between FDM and LWR in contrast to our initial hypothesis that high-yielding genotypes would necessarily have a low LWR. Under nonlimiting N, selection for low NDF concentration would lead, therefore, to selection for high LWR. Casler (1999) reported similar findings in smooth bromegrass (Bromus inermis Leyss.).
It is well accepted that NDF concentration is negatively related to IVTD and IVCWD (Bélanger and McQueen, 1996). Under limiting N, genotypes N+ and 1318 with low NDF and high IVTD and IVCWD, and N- with high NDF and low IVTD and IVCWD, illustrated this negative relationship (Tables 3 and 4). The first PC also described the opposition between NDF concentration and digestibility (Fig. 2). Under nonlimiting N, NDF was not correlated to IVTD or IVCWD (Fig. 1; Table 4).
Dry Matter Yield and Nitrogen Concentration
The N concentration in forage is typically negatively related to DM yield because of the N dilution in DM during plant growth (Lemaire and Salette, 1984; Bélanger and Richards, 1997). Under limiting N, the genotypes N+ with low FDM and high FNC, and 1254 and 1263 with high FDM and low FNC (Table 3), illustrated the negative correlation between FNC and FDM (Table 4). A negative correlation between TBDM and TBNC was also illustrated by N- with a low TBDM and a high TBNC, and 1263 with a high TBDM and a low TBNC. Under both limiting and nonlimiting N rates, the second PC illustrated the N dilution in DM (Fig. 1 and 2).
Under nonlimiting N, despite the N dilution phenomenon, the contrasting values of FDM and TBDM corresponded to similarly contrasting values of FNA and TBNA since the N concentration did not significantly vary among genotypes. Under limiting N, however, no significant variability of FNA and TBNA was found because, for most of the genotypes, the FDM or TBDM values were negatively correlated with the FNC or TBNC values such that the decrease in N concentration was compensated by an increase in biomass DM yield.
Root Weight Ratio and Leaf Weight Ratio
In addition to the above relationships, unexpected positive correlations between LWR and RWR were observed under both limiting and nonlimiting N, and this was also indicated by the first PC (Table 4; Fig. 1 and 2). This relationship was mostly due to the half-sib families. We speculate that plants indirectly selected for high LWR, and, therefore, for reduced stem biomass, had lower C partitioning to shoots, resulting in more assimilates being available for root growth.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
We confirmed the genetic variability in timothy for forage DM yield and NUE, and for several parameters of nutritive value. The variability for forage DM yield and NUE, however, was not necessarily associated with changes in biomass partitioning but rather to a greater growth potential of some genotypes. The level of N stress applied during a selection program must be carefully chosen because a significant interaction was found between genotype and N rate for DM yield and for most of the other measured parameters. Further, primarily under nonlimiting N, we observed variability for the relationship between DM yield and biomass partitioning, indicating the potential for selecting high-yielding genotypes with high LWR and RWR. Consequently, variability for the relationship between DM yield and nutritive value parameters such as N concentration, NDF concentration, IVTD, and IVCWD was observed. This indicates the possibility of selecting high yielding genotypes with superior forage nutritive value.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication February 4, 2000.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in Crop Science:
This article has been cited by other articles:
|
|
 |
|
  J. D. Holman, C. Hunt, and D. Thill Structural Composition, Growth Stage, and Cultivar Affects on Kentucky Bluegrass Forage Yield and Nutrient Composition Agron. J., January 1, 2007; 99(1): 195 - 202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Gazanchian, N. A. Khosh Kholgh Sima, M. A. Malboobi, and E. Majidi Heravan Relationships between Emergence and Soil Water Content for Perennial Cool-Season Grasses Native to Iran Crop Sci., February 1, 2006; 46(2): 544 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Belanger, A. Bregard, and R. Michaud Phosphorous Uptake and Concentration of Timothy Genotypes under Varying N Applications Crop Sci., November 1, 2002; 42(6): 2044 - 2048. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
, and the ANOVA mean square values.****
