| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL NUTRITION |
* Department of Animal and Food Sciences, University of Kentucky, Lexington 40546-0215;
and
Agri-Nutrients Technology Group, Petersburg, VA 23803; and
University of Aarhus, Faculty of Agricultural Sciences, Tjele, DK-8830, Denmark
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: metabolism • nitrogen • nonprotein nitrogen • ruminant • steer • urea
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Numerous means of promoting greater synchrony of N availability and carbohydrate degradation have been sought including inhibiting ruminal urease activity (Whitelaw et al., 1991
; Ludden et al., 2000a, b). Although urease activity can be substantially reduced, inhibitors have provided only short-term regulation because remaining urease capacity is still great enough to hydrolyze ruminal urea, possibly because of microbial adaptation to the inhibitors (Whitelaw et al., 1991; Ludden et al., 2000b). An alternate solution could be to modify urea to control its rate of release so that ammonia release more closely parallels carbohydrate digestion. Several compounds have been utilized in an effort to achieve this goal with mixed results (Oltjen et al., 1968; Males et al., 1979; Owens et al., 1980). Newer forms of slow-release urea (SRU) compounds use various coatings such as oil or polymers to release N more slowly than feed-grade urea. However, little research has investigated the behavior of these newer coatings on ruminal dynamics and growth performance. Therefore, 2 experiments were designed to determine the behavior of a polymer-coated SRU for ruminants. The objective of Exp. 1 was to characterize the behavior of SRU in the rumen and determine if the ruminal micro-flora would adapt during an extended feeding period of SRU. The objective of Exp. 2 was to determine the effects of supplemental SRU on the growth performance of beef steers.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Experiment 1
Animals and Treatments.
Twelve ruminally cannulated steers (529 ± 16 kg of BW) were used in a randomized complete block design with 2 treatments, urea or SRU (Agri-Nutrients Technology Group, Petersburg, VA). The experiment was 35 d, with 33 d of diet adaptation and sampling on d 34 and 35. Treatments were urea or SRU mixed into the basal diet at 1.8% of diet DM; 50% of the calculated amounts of urea were fed for the first 3 d and the full treatment amount for the remaining 32 d. Diets (Table 1) contained corn silage (89.5% of diet DM) and a ground corn-based vitamin and mineral supplement (8.7% of diet DM). Diets were offered as a total mixed ration at 1.25% of BW daily. Animals were maintained in individual stalls (2.4 x 2.4 m2) in a temperature (20°C) and light-controlled (12 h light:12 h dark) room with water available for ad libitum consumption.
|
Ruminal fluid (100 mL) was collected by suction strainer (Raun and Burroughs, 1962; 19-mm diameter, 1.6-mm mesh) on d 34 at 0, 2, 4, 6, 8, and 10 h after feeding and immediately analyzed for pH. Ruminal fluid (10 mL) was acidified with 2 mL of 25% (wt/vol) metaphosphoric acid and stored (–20°C) until analyzed for ammonia and VFA. Samples taken 4 h after feeding were immediately analyzed for urease activity (Bunting et al., 1989).
An in situ incubation trial was conducted on d 35 to determine the release of SRU from nylon bags suspended in the rumens of animals fed either SRU or urea for 35 d. Nylon bags containing 0.5 g of SRU (2 bags + blank at each time) were suspended for 0, 2, 4, 6, 8, 12, and 24 h. The bags were removed, rinsed lightly to remove rumen debris, dried at 55°C, and analyzed for N (Leco Corp., St. Joseph, MI). Zero-hour bags (6) were rinsed and dried without placing them in the rumen.
Frozen rumen fluid samples were thawed and centrifuged (39,000 x g for 20 min, 4°C), and supernatant was collected for analysis of ammonia-N and VFA. Ammonia-N concentrations were determined enzymatically (coupled with glutamate dehydrogenase) using procedures adapted for use on a Cobas Fara II centrifugal analyzer (Roche Diagnostic Systems, Montclair, NJ). Ruminal fluid samples were analyzed for VFA concentrations according to Harmon et al. (1985).
Calculations and Statistical Analysis. Data were analyzed as a completely random design using the GLM procedure (SAS Inst. Inc., Cary, NC). For samples collected over time (ruminal pH, ammonia-N, VFA), the data were analyzed as a split-plot in time with the main plots as a completely random design and time as the subplot. Animal (treatment) was used as the error term for treatment. Treatment effects and treatment x time interactions were declared significant at P < 0.05, and tendencies were declared at P < 0.10.
Experiment 2
Animals and Treatments.
One hundred eighty Angus cross steers (330 ± 2.3 kg) were used in a randomized complete block design with a 2 x 4 + 1 factorial arrangement of treatments; animals were blocked by BW and feeding location and assigned randomly to 1 of 9 treatments (4 steers per pen, 5 pens per treatment). Treatments were supplied in a ground corn-based supplement and included no supplemental urea (control) or 0.4, 0.8, 1.2, or 1.6% of diet DM consisting of supplemental urea or SRU (Agri-Nutrients Technology Group). Ingredient and nutrient compositions of experimental diets are shown in Table 2; diets contained 90% corn silage (45% of corn silage produced in 2004, 45% of corn silage produced in 2005) and 10% supplement (DM basis). The corn silage produced in 2005 was drought stressed and had minimal grain content. Intermediate treatments (0.4, 0.8, and 1.2% urea or SRU) were obtained by blending proportions of the control and the 1.6% urea or SRU supplements. Dietary CP concentrations were consistent between the 2 urea sources at a given supplementation concentration with diets ranging from 9.1 to 12.3% CP. Diets were fed daily as a completely mixed ration to achieve ad libitum intakes.
|
. Calculations and Statistical Analysis. Data were analyzed using the GLM procedure (SAS Inst. Inc.) as a randomized complete block design with a 2 x 4 + 1 factorial arrangement of treatments using pen as the experimental unit; data were first analyzed as a randomized complete block design with 9 treatments using a model that included block and treatment effects. Data were then analyzed as a 2 x 4 factorial structure by eliminating the control treatment, and the treatment sums of squares were partitioned into urea source, concentration of urea, and their interaction. These factors were tested using the error terms from the initial analysis (including all 9 treatments), and F-tests and probabilities were computed by hand. The effects of urea concentration were tested for linear, quadratic, and cubic effects using contrast statements; linear, quadratic, and cubic effects within each urea source were determined using contrast statements when a concentration x urea source interaction occurred. Main treatment effects and interactions of main treatment effects were declared significant at P < 0.05, and tendencies were declared at P < 0.10.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Treatment did not affect DMI (mean 6.80 kg/d; P = 0.64; data not shown) or BW (mean 529 kg; P = 0.62; data not shown). Effects of urea or SRU on ruminal pH, ruminal metabolite concentrations, urease activity, and in situ degradation of SRU are shown in Table 3. There were no significant treatment x time interactions for ruminal pH or ammonia concentrations (P > 0.15), therefore only treatment means are presented. Slow-release urea did not affect ruminal pH (P = 0.36) but did decrease ruminal ammonia concentrations (P = 0.02) and tended to increase ruminal urease activity (P = 0.06) compared with urea treatment. The in situ rate of SRU degradation did not differ when incubated in the rumen of animals fed urea for 35 d versus those fed SRU for 35 d (P = 0.48).
|
. Across treatments, total ruminal VFA concentrations were 101.5 mM and were not affected by treatment (P = 0.57). Likewise, concentrations of acetate, propionate, butyrate, valerate, and isovalerate were not affected by treatment (P > 0.20). Isobutyrate concentrations were greater for the urea treatment than SRU at 2 h but were less for subsequent times (treatment x time P = 0.03; data not shown). Similarly, molar proportions of VFA except acetate were also not affected by treatment (P > 0.47), but SRU increased the molar proportion of acetate at 2 h but decreased it for subsequent times compared with urea (treatment x time P = 0.05; data not shown). Experiment 2
Body weights, ADG, DMI, and G:F of steers are shown in Table 4. As designed, initial BW did not differ among treatments. Final BW was not affected by urea source but increased quadratically (P = 0.001) with increasing urea supplementation, primarily because the control group weighed less than groups supplemented with any concentration of urea; increasing supplementation concentration above 0.8% of diet DM did not further increase final BW.
|
Overall BW gain was less for d 29 to 56 than the initial 28 d, possibly because of the influences of gastrointestinal tract fill during the initial 28 d; however, patterns of response to treatment were similar. There was a tendency for an interaction between concentration of supplementation and urea source (P = 0.06; Figure 1A); with increasing supplementation of urea from 0 to 1.6% of diet DM, BW gain increased from 0 to 0.4% supplementation and did not increase further until urea was supplemented at 1.6% of diet DM (cubic P = 0.01). In contrast, increasing supplementation of SRU from 0 to 0.8% of diet DM increased BW gain, but supplementation with 1.2% SRU did not further increase BW gain and 1.6% supplementation reduced BW gain slightly (quadratic P = 0.01). Slow-release urea reduced DMI during d 29 to 56 (source P = 0.01) compared with urea, but there was no interaction of urea source and concentration (P = 0.64) nor was there an effect of supplementation concentration (P = 0.35). During d 29 to 56, G:F exhibited a pattern similar to ADG (source x concentration, P = 0.05; Figure 1B). The G:F was increased when supplementation of urea was increased from 0 to 0.4% of diet DM, but did not increase further until 1.6% of diet DM was supplemented (cubic P = 0.01). The G:F was increased when supplementation of SRU was increased from 0 to 0.8% of diet DM, but supplementation with 1.2% SRU did not further increase G:F and 1.6% SRU supplementation reduced G:F slightly (quadratic P = 0.001).
|
). Urea supplementation at 0.4% of diet DM increased steer BW gain, but additional supplementation did not further increase BW gain (cubic P = 0.03). In contrast, SRU supplementation up to 0.8% of diet DM increased gain, but supplementation of 1.2% did not further increase BW gain and supplementation of 1.6% decreased BW gain (quadratic P = 0.001). Over d 0 to 56, DMI was not affected by urea source but increased with the addition of supplemental N and was consistent across all concentrations of N supplementation (quadratic P = 0.001). Because feed intake was consistent across source and concentration of N supplementation, G:F closely paralleled the changes in BW gain. There was an interaction of urea source x concentration (P = 0.01) for G:F. Increasing supplementation of urea from 0 to 0.4% of diet DM increased G:F, but additional increases in supplementation did not further increase G:F (cubic P = 0.04). Increasing supplementation of SRU up to 0.8% of diet DM increased G:F, but supplementation of 1.2% SRU did not further affect G:F and supplementation of 1.6% SRU decreased G:F (quadratic P = 0.001). Blood samples were collected at d 28 to assess serum urea concentrations as an indicator of N intake and availability. There was no interaction of urea source and concentration of supplementation on serum urea concentrations. Supplementation with SRU reduced serum urea concentrations compared with urea supplementation (P = 0.001). Additionally, with increasing supplementation, serum urea concentrations increased substantially (quadratic P = 0.03).
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The objective of Exp. 1 was to characterize the behavior of SRU in the rumen and determine if the ruminal microflora would adapt during an extended feeding period of SRU. Urea undergoes rapid hydrolysis in the rumen to ammonia as confirmed by the substantially greater ruminal ammonia concentrations for urea versus SRU. Urea increased mean ruminal ammonia concentrations by 58% compared with SRU. Compared with SRU treatment over the 10-h sampling period, urea treatment increased ruminal ammonia concentrations by 25% at 2 h with a maximal increase of 147% at 8 h. These results demonstrate that in vivo SRU does indeed have a slower ammonia release rate than urea and can effectively reduce ruminal ammonia concentrations when substituted for urea. Past research has attempted to limit rapid ammonia appearance by inhibiting ruminal urease activity, including using the urease inhibitors acetohydroxamic acid, phenylphosphorodiamidate, and N-(n-butyl) thiophosphoric triamide (Streeter et al., 1969; Whitelaw et al., 1991; Ludden et al., 2000a). However, research with these compounds has shown that urease inhibitors are effective at inhibiting some ruminal urease activity, but the remaining urease activity is not retarded and is still sufficient to rapidly hydrolyze ruminal urea (Voigt et al., 1980; Whitelaw et al., 1991; Ludden et al., 2000b). The use of compounds such as biuret or combinations of urea with coatings, gelatinized starches, or molasses has successfully reduced ruminal ammonia concentrations compared with urea (Oltjen et al., 1968; Thompson et al., 1972; Owens et al., 1980). However, a sustained release of NH3-N was not achieved in vitro or in vivo with molasses-urea, gelatinized starch-urea, or slow-release liquid in another experiment (Males et al., 1979). More recent SRU products have not been well characterized in the rumen, but results from this experiment and those of Taylor-Edwards et al. (2009) demonstrate that this polymer-coated SRU product is able to successfully modulate the appearance of ammonia in the rumen environment.
In theory, a possible concern could be that SRU releases ammonia too slowly in the rumen to provide enough ammonia for efficient bacterial utilization. Although ammonia concentrations of 50 mg/L rumen fluid (2.9 mM) were sufficient to support maximal rates of microbial growth in vitro (Satter and Slyter, 1974), which is substantially less than the ruminal ammonia concentrations observed in either treatment of Exp. 1, others have concluded that ammonia concentrations of 235 mg/L (13.8 mM) are optimal for rumen fermentation in vivo (Mehrez et al., 1977). Regardless, Hespell and Bryant (1979) concluded that it is unlikely that the ammonia concentration in the rumen would limit growth of ammonia-requiring bacteria to such an extent that significant uncoupling of fermentation from microbial cell growth would occur. Instead, it is likely that the optimal ruminal ammonia concentration needed for microbial utilization is instead dependent on diet composition and intake. This topic has been reviewed by several authors (Clark and Davis, 1983; Clark et al., 1992; Bach et al., 2005). Therefore, it might be expected that ammonia availability from SRU may be more limiting for animals with high ruminal N requirements, such as those with high intakes or production demands or those consuming highly degradable feedstuffs, than for those with decreased ruminal N requirements.
Ruminal concentrations of total and individual VFA were not affected by treatment, and molar proportions of VFA were also unaffected by treatment. Although replacement of true feed protein with urea often results in reduced concentrations of isobutyrate, isovalerate, and valerate (Griswold et al., 1996; Köster et al., 1997; Sannes et al., 2002), replacing urea with a SRU in a diet rarely affects any ruminal metabolite concentrations other than ammonia, at least in situations where reduced ammonia concentrations presumably do not limit microbial growth. Data of Garrett et al. (2005) showed that total VFA production in continuous culture fermenters was not affected by urea source (urea or oil-coated SRU). These results and the present study suggest that urea source does not affect total production or ruminal concentrations of VFA.
The ability of a slow-release product to maintain efficacy over time is important. In the past, most SRU products have either released ammonia too quickly or N was so tightly complexed that little ammonia was released into the ruminal environment (Males et al., 1979). In contrast, Exp. 1 demonstrates that SRU was able to maintain a slow-release rate over 35 d because the in situ degradation rate of SRU did not differ between steers that had been fed urea or SRU for 35 d and across treatments averaged 6.28%/h. In contrast, the degradation rate of feed-grade urea is too fast to measure using this system because it rapidly dissolves. Therefore, the SRU product used in this experiment can effectively modulate the appearance of ammonia in the rumen over 35 d; more research is needed to determine if the efficacy of SRU is maintained over longer feeding periods. The slower rate of in situ degradation of SRU also more effectively matches availability of other nutrients necessary for microbial growth. The degradation rate of SRU (6.28%/h) was similar to rates of starch degradation, which range from 3.1 to 23.5%/h depending on starch source (Herrera-Saldana et al., 1990). For corn grain, starch degradation rate is approximately 4.6 to 6.4%/h (Herrera-Saldana et al., 1990; Philippeau et al., 1999). Therefore, SRU is able to match release of N to energy availability of common energy sources in ruminant feedstuffs.
Growth Performance During Feeding Trial
The objective of Exp. 2 was to determine the effects of supplemental SRU on the growth performance of beef steers. It was hypothesized that if SRU was used more efficiently than urea it might support equivalent growth rates to urea at a decreased concentration of inclusion. In this experiment, N availability was clearly limiting growth as animals responded to increasing supplemental N with increased BW gain and G:F.
Source of urea affected DMI only during the second half (d 29 to 56) of the experiment; the reduction in DMI when SRU replaced urea ranged from 1.7% at the 1.2% inclusion concentration to 7.4% in the 0.4% inclusion concentration. Because DMI did not differ during the first half (d 0 to 28) of the experiment, DMI over the entire experimental period did not differ between urea sources. Replacing urea with a SRU source has not affected DMI in past experiments (Thompson et al., 1972; Tedeschi et al., 2002; Galo et al., 2003). Of interest was the decreased DMI at the 0.4% SRU supplementation concentration during d 29 to 56 but not d 0 to 28. If intake of protein is insufficient to support microbial fermentation, DMI can be reduced (Faverdin, 1999). One could speculate that the decreased DMI for the 0.4% SRU treatment occurred only during the second 28 d of the trial because there was a cumulative N deficiency, which caused a depletion of available substrate to recycle back to the rumen over time. As discussed by Reynolds and Kristensen (2008), there are pools of labile proteins available for recycling when protein intake is low. However, it would be expected that during long-term N deficiency these pools are depleted. Therefore, at the 0.4% supplementation concentration, SRU may have reduced ruminal N availability compared with urea, which eventually caused a reduction in DMI once the labile pool of proteins available for recycling were depleted.
Over the entire experiment, source of urea did tend to affect gain and G:F, but these effects were more apparent during the second half of the experiment and were also dependent on concentration of urea supplementation. Average daily gain was reduced by SRU at the 0.4 and 1.6% inclusion concentrations, but did not differ from urea at the 0.8 and 1.2% inclusion concentrations. Because this effect was marked during the second half of the experiment (d 29 to 56), animal BW at the end of the experiment also numerically followed this trend. Because of the relative lack of treatment differences for DMI, G:F therefore followed a similar pattern.
Inclusion of SRU at 0.4% of diet DM reduced BW gain and G:F in addition to DMI, plausibly because of a limitation in N availability. Body weight gain of steers fed 0.4% SRU was intermediate to the control diet (no supplementation) and the 0.4% urea diet. This would suggest that N from SRU was not used more efficiently and that the availability of N may have been too low to support ruminal fermentation. These results are similar to those reported by Tedeschi et al. (2002), where urea or Optigen 1200, a slow-release polymer-coated urea, was supplied at either 0.4 or 1.2% of diet DM in a diet for growing steers. At the 0.4% supplementation concentration, Optigen 1200 reduced ADG by 21% compared with urea and increased the amount of feed required for gain (DMI/ADG) without affecting DMI, demonstrating a reduced efficiency of Optigen 1200 use for BW gain at the 0.4% supplementation concentration compared with urea. However, there were no differences in that experiment at a 1.2% supplementation concentration between urea or Optigen 1200, similar to results in the current experiment. The results from Exp. 2 suggest that at 0.4% of DM, SRU may limit N availability and reduce gain both directly and indirectly via effects on DMI.
The negative impact of SRU on performance at the 1.6% inclusion rate is more difficult to explain. If N was limiting, one would expect that increased supplementation would relieve that deficiency, and indeed, SRU improved or did not affect performance at the intermediate inclusion rates, suggesting that N should not be limiting at the greatest inclusion rate. Additionally, increasing concentrations of serum urea (on d 28) with increasing dietary urea supplementation for both urea and SRU treatments also suggest that N availability was likely not limiting growth at the 1.6% SRU supplementation concentration. It is possible that by chance the animals assigned to this treatment were a poorly performing group of animals. Alternatively, in steers receiving either urea or SRU at 1.6% of diet DM, N retention (g/d) tended to be less for animals fed SRU compared with urea with a concomitant increase in fecal N excretion but no change in DMI (Taylor-Edwards et al., 2009). These results suggest that greater concentrations of SRU supplementation may increase fecal N excretion and reduce N retention and BW gain without affecting DMI. One possible explanation for this is a difference in the digestion and flow kinetics of SRU compared with urea, especially at greater supplementation concentrations. Also, as suggested by Taylor-Edwards et al. (2009), SRU may shift nutrient digestion postruminally and increase microbial N excretion. Additionally, if the release rate of ammonia from SRU is too slow in the rumen but results in greater postruminal N supply, it would be expected that gain may be negatively affected without observable effects on serum urea concentrations as both ammonia and urea freely move between the bloodstream and the hindgut (Hoover, 1978). However, it is unexpected that the 1.6% SRU group would gain less than the 1.2% SRU treatment group especially given no effect of treatment on DMI between these concentrations of SRU and the corresponding concentrations of urea, and the conclusion that this particular group of animals gained less by chance is probably the most reasonable.
In general, the results from Exp. 2 are consistent with other experiments in which urea was replaced with a slow-release form of urea. Biuret, a condensation product of urea, is probably the most common slow-release form of urea utilized in experiments over the past 30 yr. Compared with soybean meal and urea, biuret did not affect growth rate and feed efficiency of 2- to 4-mo-old calves or heifers (Campbell et al., 1963). Biuret also did not change ADG of young bulls (Chicco et al., 1971), lambs (Karr et al., 1965), or steers (Ammerman et al., 1972). Other SRU products have shown similar results. Starea, a gelatinized grain starch and urea product, did not affect BW gain or feed efficiency of either growing steers or calves in 3 separate trials (Thompson et al., 1972), and replacement of urea with Optigen either did not affect or reduced gain of growing and finishing cattle (Tedeschi et al., 2002). Indeed, the conclusion of Tedeschi et al. (2002) that "there is no clear advantage in substituting a SRU product for urea at the levels usually fed to growing and finishing beef cattle" is an appropriate conclusion for the current experiment (Exp. 2), and substitution of SRU products may actually be detrimental to animal performance in situations of N deficiency (as in the 0.4% supplementation concentration) or high supplementation concentrations (as in the 1.6% supplementation concentration). However, at intermediate supplementation concentrations, SRU products appear to affect animal performance similar to urea.
In conclusion, SRU reduces the rapidity of ammonia release in the rumen without affecting other ruminal fermentation metabolites. After 35 d of feeding there was no evidence of microbial adaptation to the SRU product, suggesting that SRU will continue to maintain slow-release properties for long-term feeding regimes. Slow-release urea did not affect ADG, DMI, or G:F when supplemented at 0.8 or 1.2% of DMI, but at decreased (0.4%) or increased (1.6%) concentrations of supplementation SRU reduced ADG and G:F without substantial changes in DMI. These interactions of concentration and source of urea were not apparent during the first 28 d of the feeding trial but did become significant during d 29 to 56. These experiments demonstrate that SRU can be utilized as an N supplement to modulate the appearance of N in the rumen and can provide equal performance to urea supplements without the potential hazards associated with feed-grade urea.
1 Corresponding author: dharmon{at}uky.edu
Received for publication January 29, 2008. Accepted for publication August 29, 2008.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. C. Taylor-Edwards, N. A. Elam, S. E. Kitts, K. R. McLeod, D. E. Axe, E. S. Vanzant, N. B. Kristensen, and D. L. Harmon Influence of slow-release urea on nitrogen balance and portal-drained visceral nutrient flux in beef steers J Anim Sci, January 1, 2009; 87(1): 209 - 221. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), and concentration of supplementation of a corn-silage based diet on BW gain and G:F of growing steers during d 29 to 56 and d 0 to 56 of Exp. 2.