Roche/ASAS Foundation Beef Cattle Nutrition Symposium: Forage

Supplementation and Grazing

Effects of Supplementation on Voluntary Forage Intake, Diet

Digestibility, and Animal Performance1,2

J. E. Moore3, M. H. Brant4, W. E. Kunkle, and D. I. Hopkins

Department of Animal Science, University of Florida, Gainesville 32611

ABSTRACT: A data base was constructed to BW. When supplements increased VFI, forage TDN:
describe and estimate supplementation effects in CP ratio was > 7 ( N deficit), and VFI when fed alone
nonlactating cattle consuming forage ad libitum. The was often low. There was little relationship between
data base included 66 publications on 126 forages (73 change in VFI and sources of supplemental CP and
harvested and 53 grazed) and a total of 444 compari- TDN. Supplements caused total diet TDN concentra-
sons between a control, unsupplemented treatment tion to deviate from expected values by −10 to +5% of
and a supplemented treatment. Daily gains were OM. When supplemental TDN intake was > .7% of
reported for 301 comparisons and voluntary intake for BW, diet TDN concentration was always less than
258. Direct measures of forage digestibility were expected. There was little relationship between devia-
reported for 202 comparisons, and total diet digestibil- tion from expected total diet TDN and type or
composition of forages or supplements. Empirical
ity for 150. Supplements did not increase gain in all
multiple regression equations were developed to esti-
cases. Change in ADG due to supplement was not
mate effects of supplements on VFI and total diet TDN
related closely to intake of supplemental TDN. Lowest
concentration. The most acceptable intake equation
increases in ADG were with native forages sup- estimated VFI when fed with supplement ( r 2 = .84)
plemented with molasses alone or with low intakes of That equation included VFI when fed alone, supple-
molasses containing high levels of NPN. Greatest ment intake, CP and TDN concentrations in forage
increases in gain were with improved forages, supple- and supplement, and classification codes describing
ments with > 60% TDN, and supplemental CP intake forages and supplemental energy. The most acceptable
> .05% of BW. Supplements decreased voluntary equation for estimating total diet TDN concentration
forage intake (VFI) when supplemental TDN intake included only the expected total diet TDN concentra-
was > .7% of BW, forage TDN:CP ratio was < 7 tion ( r 2 = .87). These equations may be used in
(adequate N), or VFI when fed alone was > 1.75% of nutritional models to account for associative effects.

Key Words: Supplementary Feeds, Weight Gain, Forage, Diet

1999 American Society of Animal Science and American Dairy Science Association. All rights reserved.

Introduction quate to meet desired rates of animal performance

(i.e., ADG or milk production). In such cases,
When cattle consume forages as their only energy supplements may be provided to attain the desired
source, intake of available energy may not be ade- performance. In many cases, animal responses to

1Florida Agric. Exp. Sta. journal series no. R-06491.

2This research was supported in part by Special Grant in 3To whom correspondence should be addressed. Current address:

Tropical Agriculture 94-34135-0653 from the USDA T-STAR Carib- 5920 W. 53rd St., Stillwater, OK 74074.
bean Program. 4Current address: 109 Amberchase Drive, Lexington, SC 29073.

122
 FORAGE SUPPLEMENTATION 123

supplements are either greater or less than expected. The objectives of this review are to examine effects
The deviations between expected and observed perfor- of supplements on daily gain, voluntary forage intake,
mance are usually explained by associative effects of and total diet TDN concentration and to develop and
supplements upon voluntary intake and available evaluate equations for estimating total diet intake and
energy concentration of the total diet. The concept of total diet TDN concentration.
associative effects refers to nonadditive interactions
among ingredients in mixed diets.
This paper focuses on associative effects that occur Effects of Supplements on Gain, Intake, and TDN
when forage intake is voluntary and supplements are
fed separately in restricted amounts. Although as- Data base Construction. A literature review
sociative effects under these conditions are well- provided 66 references that met the requirements for
documented, they are difficult to quantify and are not inclusion in the data base (i.e., voluntary forage
considered in most nutritional models. After an intake, supplements fed separately, nonlactating cat-
extensive review of associative effects in forage-based tle, and an unsupplemented control treatment); these
references are found in the Appendix. The 66 refer-
diets, Horn and McCollum (1987) concluded that
ences included studies of 126 different forages: 73
“present relationships do not permit prediction of
harvested and 53 grazed. There were 444 comparisons
effects of supplementation on forage intake and
of an unsupplemented control with a supplemented
utilization for the widely different production environ-
treatment. Table 1 summarizes the distribution of
ments.” Their challenge is addressed in this article.
these comparisons among forages and supplements in
In most nutritional models, gain is a function of the data base.
both intake and available energy concentration (e.g., Most studies involved growing calves or yearlings.
TDN) of the total diet, and these are used indepen- If cows were used in the study, intake and digestibility
dently in computations. Because, in the context of this data were included, but their daily gains were not. If
article, supplement intake is known, supplement full body weights and gains were reported, they were
effects on voluntary forage intake may be quantified. converted to the shrunk basis using equations derived
It is not possible, however, to quantify the effect of from full and shrunk weights on forage-fed cattle
supplement on the TDN concentration of the forage (Kunkle and Moore, unpublished data). Intake data
component of a mixed diet. Further, it is not known if were converted to a percentage of mean shrunk body
the associative effect applies to the supplement as well weight.
as the forage. It is necessary, therefore, to compare the Data on forage characteristics were limited to those
observed TDN of the total diet with that expected from provided in the references. Digestibility data used in
the TDN of the ingredients (Brant, 1993). the data base were limited to those from in vivo trials.

Table 1. Distribution of comparisons between unsupplemented and supplemented

treatments (444 total comparisons)

By forage type n By supplement type n

Temperate 122 Liquid 150
Tropical 125 Dry 255
Native 175 Combined 39
Straw 22
By sources of supplemental energy and protein
Energy n Protein n
Protein feeds only 43 Energy feeds only 148
Molasses 178 Non-protein nitrogen 142
Grain 129 Protein feeds 143
By-Products 35 Combinations (NPN + feeds) 11
Forages 27
Combinations 32

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
124 MOORE ET AL.

In some cases, digestion trials were conducted with

sheep and these data were used without adjustment.
Data on DM digestibility of forages and mixed diets
were converted to OM digestibility by the following
formula (Moore, unpublished data): OM digestibility
( % ) = −.664 + 1.032 × DM digestibility ( % ) . Forage
TDN was assumed to be equivalent to digestible OM.
Regarding supplements, composition data were
taken from the reference, or calculated from supple-
ment ingredient formulas, as given in the reference,
and tabulated values of CP and TDN concentration.
Because of the wide variation in ash percentage
reported for many forages and supplements, all data
were converted to the OM basis by dividing concentra-
tions on the DM basis by the OM concentration as a
percentage of DM.
Effects on Daily Gain. Associative effects between
Figure 1. Comparison of daily gain by cattle when fed
supplements and forages were demonstrated clearly in
forage plus supplement with daily gain when the same
terms of ADG. In many cases, ADG was not increased
forage was fed alone.
when forages were supplemented, and was sometimes
decreased (Figure 1). Effects of supplements on ADG
were quantified as the change in shrunk ADG
(GAINchg), using the following formula: GAINchg = gain occurred primarily with grazed native forages
GAINtotal − GAINforage, where GAINtotal = shrunk supplemented with molasses alone or molasses with
ADG on total mixed diet (kg/d) and GAINforage = NPN. When GAINchg was greatest(> .4 kg/d),
shrunk ADG on forage fed alone (kg/d). A positive however, it occurred with harvested improved forages
GAINchg indicates that ADG was increased when that were supplemented with either dry feeds or
supplements were fed. Most, but not all, GAINchg molasses with added nitrogen. At intermediate ranges
values were positive. of GAINchg, there was little difference among types of
Forage and supplement types were confounded forages and supplements with respect to GAINchg.
when GAINchg was at either extreme (Table 2). There was little relationship between GAINchg and
Decreases and slight increases (< .02 kg/d) in daily supplemental TDN intake (STDNI) (Figure 2). At

Table 2. Frequency distribution of forages and supplements according to ranges

of change in daily gain due to supplements

Range in gain change due to supplementation, kg/d

Comparison < .02 .02 to .05 .06 to .10 .11 to .20 .21 to .30 .31 to .40 >.40
Total 29 33 57 61 42 41 38
Forages
Grazed 27 28 46 41 29 13 5
Harvested 2 5 11 20 13 18 33
Native+straw 18 26 36 30 3 3 4
Cool+warm 11 7 21 31 38 38 34
Supplements
Molasses 25 28 37 34 17 15 13
Alone 11 9 6 5 0 2 0
+NPN 14 18 21 29 13 4 5
+Meal 0 0 0 0 3 4 5
+NPN+meal 0 1 0 0 1 5 3
Dry feeds 4 5 20 27 25 26 25

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
 FORAGE SUPPLEMENTATION 125

Figure 2. Effect of supplemental total digestible Figure 4. Effect of supplemental total digestible
nutrient (TDN) intake on change in daily gain due to nutrient (TDN) concentration on change in daily gain
supplement, classified by type of forage. due to supplement, classified by source of supplemental
energy.

low STDNI, there was often a large positive GAINchg,

especially with native forages. Negative GAINchg negative GAINchg values were with molasses having
occurred in a few cases with native forages and warm- a high TDN percentage (Figure 4); these supplements
season improved forages. The same array of data contained nonprotein nitrogen (NPN) sources such as
classified according to energy source (Figure 3 ) urea and ammonium sulfate (Figure 5).
illustrates the confounded nature of the data base In many cases, the TDN concentration of molasses
discussed above (e.g., native forages were sup- supplements was less than 60% of OM (Figure 4);
plemented often with molasses). Also, most of the these supplements contained high percentages of NPN

Figure 3. Effect of supplemental total digestible Figure 5. Effect of supplemental crude protein (CP)
nutrient (TDN) intake on change in daily gain due to intake on change in daily gain due to supplement,
supplement, classified by source of supplemental classified by source of added supplemental protein (in
energy. addition to protein from energy feeds, if any).

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
126 MOORE ET AL.

because molasses was used simply as a carrier.

Generally, these supplements were fed in small
amounts and gave a positive but relatively small
GAINchg. When supplemental TDN percentage was
above 60% of OM and GAINchg was positive,
GAINchg was not related to energy source.
The largest increases in GAINchg occurred when
supplemental CP intake was greater than .05% of BW
(Figure 5). When supplemental CP intake was
greater than .1% of BW, GAINchg was always
positive. There was little difference among sources of
protein. Low GAINchg values were observed at low CP
intakes, but this effect is confounded with type of
forage and supplement (low GAINchg with native
pastures supplemented with molasses and NPN).
When responses to feed protein supplements were
examined (Figure 6), there was little difference Figure 7. Comparison of voluntary forage organic
among grains, by-products, and plant protein feeds, matter (OM) intake when fed with supplement to intake
but supplements with added escape protein tended to when the same forage was fed alone, classified by type
give the highest GAINchg at a given STDNI. of forage.
Effects on Voluntary Forage Intake. Voluntary intake
of forage was both increased and decreased by
supplementation (Figure 7). Most of the increases of body weight (VFIchg), using the following formula:
were with native forages and straws, whereas most of VFIchg = VFIwith − VFIalone, where VFIwith = VFI
the decreases were with improved cool and warm of forage fed with supplement ( % of BW, OM basis)
season forages. When forage intake fed alone was > and VFIalone = VFI of forage fed alone ( % of BW, OM
1.75% of BW, supplement decreased forage intake in basis). A negative VFIchg means that supplement
most cases. decreased intake of the forage. This effect has been
Effects of supplements on voluntary forage intake termed substitution (i.e., supplement substitutes for
were quantified as the change in VFI, as a percentage forage) and has been expressed as substitution rate,
the decrease in VFI per unit of supplement fed. The
use of VFIchg provides a more readily understood
expression of the effect of supplements on forage
intake; it is negative when intake is decreased, and
positive when intake is increased.
When VFIchg was compared with the ratio of TDN
to CP in forages (FTDN:CP; Figure 8), it appeared
that much of the effect due to forage type could be
explained by this characteristic of the forage. When
FTDN:CP was < 7, VFIchg was generally negative.
The straws that accounted for five of the seven
exceptions (i.e., positive VFIchg when FTDN:CP was
< 7 ) were ammoniated. When FTDN:CP was > 12,
almost all VFIchg were positive, and all forages were
native. When VFIchg was compared with STDNI and
responses classified by FTDN:CP (Figure 9), increas-
ing STDNI resulted in a more negative VFIchg with
forages having FTDN:CP < 7. There was little effect of
Figure 6. Effect of supplemental total digestible STDNI on VFIchg with those forages having FTDN:
nutrient (TDN) intake on change in daily gain due to CP > 7, except when STDNI was greater than .7% of
supplement, classified by source of feed protein (does BW. In almost all comparisons, except for ammoniated
not include non-protein nitrogen sources). straws, when VFIchg was positive, FTDN:CP was > 7.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
 FORAGE SUPPLEMENTATION 127

supplements. When STDNI was > .7% of BW, VFIchg

was always negative. There was no apparent relation-
ship between supplemental CP intake and VFIchg,
and no apparent difference in response among sources
of added protein (Figure 11). When VFIchg was
positive, NPN and meal were equally effective.
Effects on Dietary TDN. The TDN concentration of
total diets was both greater and less than expected
(Figure 12). Expected total diet TDN was calculated
as follows: TDNexpected ( % of OM) = [(VFIwith ×
FTDN) + (SOMI × STDN)]/(VFIwith + SOMI),
where VFIwith is as defined above, FTDN = forage
TDN ( % of OM), SOMI = supplement OM intake ( %
of BW), and STDN = supplement TDN ( % of OM).
When expected TDN was greater than 60% of OM, the
observed TDN was less than expected in most cases.
Figure 8. Effect of the ratio between forage total Effects of supplementation on total diet TDN
digestible nutrient (TDN) concentration and crude concentration were quantified as deviation from ex-
protein concentration (CP) on the change in voluntary pected total diet TDN (TDNdev), calculated as
forage organic matter intake (OMI) due to supplementa- follows: TDNdev = TDNobserved − TDNexpected,
tion, classified by type of forage. where TDNobserved = observed TDN of total diet ( %
of OM) and TDNexpected is as defined above. A
negative TDNdev indicates simply that supplementa-
Perhaps FTDN:CP values greater than 7 indicate a tion resulted in an observed total diet TDN concentra-
deficit of N in relation to available energy. tion that was less than expected. It does not indicate
There was no clear distinction among energy whether supplement altered the digestibility of the
sources with respect to the effect of STDNI on VFIchg forage, the supplement, or both.
(Figure 10), except that protein feeds generally gave There was a large range in TDNdev (Figure 13),
a negative VFIchg. When VFIchg was positive, there with many values between −10 and +5% of OM (not
was no difference in response between liquid and dry TDN). Such large deviations from expected TDN

Figure 9. Effect of supplemental total digestible Figure 10. Effect of supplemental total digestible
nutrient (TDN) intake on the change in voluntary forage nutrient (TDN) intake on the change in voluntary forage
organic matter intake (OMI), classified by forage TDN: organic matter intake (OMI), classified by source of
crude protein (CP) ratio. supplemental energy.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
128 MOORE ET AL.

Figure 11. Effect of supplemental crude protein (CP) Figure 13. Effect of forage total digestible nutrient
intake on the change in voluntary forage organic matter (TDN) concentration on the deviation from expected
intake (OMI), classified by source of added supplemen- total diet TDN concentration, classified by type of
tal protein (in addition to protein from energy feeds, if forage.
any).

Development and Evaluation of Equations

concentrations would have major effects on the NE
concentration of diets, and on estimated animal The data base was divided into two subsets: one for
performance. In most cases when TDNdev was posi- equation development and the other for equation
tive, the forage was a native hay or straw having a evaluation. The development subset was used to select
TDN concentration < 55% of OM. When STDNI was variables and generate coefficients for the estimation
greater than .7% of BW, TDNdev was negative in most of total diet intake and TDN concentration. The
cases (Figure 14). There was little difference among evaluation subset was used as an independent data
energy sources with respect to TDNdev.

Figure 14. Effect of supplemental total digestible

Figure 12. Comparison of observed to expected total nutrient (TDN) intake on the deviation from expected
digestible nutrient (TDN) concentration in total diets total diet TDN concentration, classified by source of
(forage plus supplement), classified by type of forage. supplemental energy.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
 FORAGE SUPPLEMENTATION 129

set to evaluate equations. About 25% of the total included as independent variables; these are defined
comparisons were assigned to the evaluation subset. A in the following.
particular reference was assigned to either one subset Forage type codes:
or the other. Attempts were made to have each subset 1 = temperate or tropical forage
similar with respect to distributions of forages and 2 = native mixed forage, or straw
means of forage and supplement variables (Table 3).
The range in variables was narrower in the evaluation Supplement type codes:
subsets than it was in the comparable development 1 = dry
subsets. When forage TDN was included in the intake 2 = liquid (e.g., based on molasses)
subsets, the number of comparisons was decreased in 3 = combination (e.g., slurry)
the development subset more than it was in the Supplemental energy codes:
evaluation subset.
Equations were developed using the appropriate 1 = protein supplement only (e.g., soybean meal)
development data sets. Independent variables ( X ) 2 = molasses
included VFIalone, forage CP ( % of OM), forage TDN 3 = grain or by-product (e.g., corn, wheat mid-
dlings)
( % of OM), forage TDN:CP ratio, supplement OM
4 = forage (e.g., alfalfa)
intake ( % of BW), supplement CP ( % of OM), and
supplement TDN ( % of OM). All independent varia- Supplemental carbohydrate degradability codes:
bles were squared, and several products were calcu- 0 = rapid (e.g., corn)
lated (e.g., supplement TDN intake, % of BW). In 1 = slow (e.g, soybean hulls)
addition, many linear interactions among forage and
supplement variables were calculated (e.g., forage Supplemental protein codes:
TDN × supplement TDN). Classification codes 1 = energy feed (e.g., corn)
describing types of forages and supplements were 2 = non-protein nitrogen (e.g., urea)

Table 3. Description of data subsets used in development and evaluation of equations to estimate organic
matter intake (% of body weight) and total digestible nutrient concentration
(% of organic matter) of total diets

Intake subsets
Ignoring forage TDN Including forage TDN TDN subsets
Item Develop Evaluate Develop Evaluate Develop Evaluate
References, n 31 15 24 15 23 9
Comparisons, n 187 59 144 56 111 37
Temperate 68 18 56 15 26 8
Tropical 56 21 40 21 36 13
Native 48 16 38 16 35 16
Straw 187 4 10 4 14 0
Forages
OM intake, % BW 1.87 ± .64 1.83 ± .62 1.85 ± .66 1.80 ± .61 1.87 ± .77 1.82 ± .50
CP, % OM 10.2 ± 5.6 9.1 ± 4.3 10.4 ± 5.8 9.2 ± 4.4 9.7 ± 5.9 8.3 ± 3.4
TDN, % OM 55.1 ± 12.2 51.9 ± 8.9 52.8 ± 12.2 52.2 ± 8.1
Supplements
OM intake, % BW .53 ± .31 .51 ± .28 .56 ± .32 .52 ± .28 .51 ± .28 .47 ± .27
CP, % OM 23.0 ± 22.3 23.2 ± 18.8 19.8 ± 12.9 21.9 ± 18.5 21.0 ± 15.8 21.4 ± 16.3
TDN, % OM 81.3 ± 10.7 83.5 ± 7.9 82.1 ± 10.0 83.5 ± 8.1 80.6 ± 9.2 85.0 ± 8.4
Total diets
OM intake, % BW 2.25 ± .47 2.29 ± .54 2.23 ± .50 2.27 ± .54 2.32 ± .57 2.28 ± .44
TDN, % OM
Observed 57.3 ± 9.0 57.3 ± 6.6
Expected 59.4 ± 10.1 58.9 ± 7.8

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
130 MOORE ET AL.

3 = protein feed (e.g, soybean meal, meat meal) unacceptable difference > 5.7
4 = combination of NPN and feed
Equations were evaluated on the basis of the percen-
Supplemental protein degradability codes:
tage of differences that were acceptable, marginal, or
0 = ruminally degraded (e.g., urea, soybean unacceptable.
meal) Intake Equations. There were three approaches to
1 = escape (e.g., fish meal, meat meal) computing estimated total diet OM intake (ETOMI, %
General relationships between dependent and in- of BW):
dependent variables (including squares, products, and
interactions) were explored using PROC REG of SAS 1. VFIchg was estimated and ETOMI = VFIalone
with stepwise selection at P < .15. Many combinations + estimated VFIchg + supplement OM intake.
2. VFIwith was estimated and ETOMI = esti-
of independent variables were examined to determine
mated VFIwith + supplement OM intake.
which variables were most often included. Final
3. ETOMI was estimated directly.
selection of variables was done using the R2 selection
with the CP statistic to minimize bias and avoid
Some equations were developed using the entire
overfitting (MacNeil, 1983). In no case was the
development set. In addition, the development subset
correlation between interaction variables permitted to
was divided into two additional subsets having forage
exceed .7. Independent variables were tested in
TDN:CP ratios either above or below 7. Equations
various combinations. The final selection of variables
based on the entire data set were more acceptable
was made from a set that included variables that had
than those based on the two subsets. The code for
appeared frequently in previous runs. In this proce-
forage type was included in most equations, and this
dure, the combinations of variables tested are selected
variable may have accounted for differences associated
by the researcher rather than the computer.
with forage TDN:CP ratio.
After deciding which variables to include, coeffi-
There were high correlations between forage CP
cients for multiple regression equations were com-
puted using PROC REG. Linear variables were added concentration and forage TDN:CP, and between sup-
to the model if they occurred as squared terms or in plement OM intake and TDN intake. Therefore, in the
interactions. These equations were then used to final stages of equation development, either forage CP
estimate intake or TDN variables for each comparison or forage TDN:CP ratio, and either supplement OM
in the appropriate evaluation subset. Equations were intake or TDN intake, were used in a 2 × 2 factorial
evaluated by regressing observed values of dependent arrangement. In all, 47 intake equations were deve-
variables ( Y ) on the comparable estimate ( X ) and loped. Each equation was evaluated by regressing
recording the coefficient of determination ( r 2) and observed or actual TOMI on ETOMI, the latter being
root mean square error (RMSE). calculated as described above for the three ap-
The major criterion used to evaluate equations was proaches.
the difference between estimated and observed values The three best equations for each option gave very
(difference = estimated − observed). A negative similar statistical parameters and differences between
difference indicates that the estimate was less than observed and estimated values (Table 4), but the
the observed value. The following criteria of accepta- VFIchg and VFIwith equations were slightly superior
bility of differences were based on common assump- in terms of the difference criteria. In fact, the same
tions about the variability among animals fed alike for variables were included in the VFIchg and VFIwith
intake (10%) and digestibility (5%). equations, even though they were developed indepen-
Mean total OM intake = 2.3% of BW dently. All coefficients were identical in the two
equations, except for the coefficient for VFIalone. The
acceptable difference = 2.3 × .1 = .23 coefficients for VFIalone were .0101 for the VFIchg
marginal difference = 2.3 × .2 = .46 equation and 1.0101 for the VFIwith equation; such a
unacceptable difference > .46 difference would be expected because VFIchg is
calculated as the difference between VFIalone and
Mean total diet TDN = 57% of OM VFIwith. Because it would be the simplest equation to
use in nutritional computations, the VFIwith equation
acceptable difference = 57 × .05 = 2.9 was chosen as the “best” equation, instead of the
marginal difference = 57 × .1 = 5.7 VFIchg equation; it is as follows:

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
 FORAGE SUPPLEMENTATION 131

Table 4. Evaluation of equations for estimating total diet organic matter intake
and TDN concentration

Intake, % of BW TDN, % of OM
With
Item Change supp. Total Complex Simple
Variables 15 15 13 7 2
Develop R2 .86 .90 .88 .84 .77
Evaluate r2 .84 .84 .82 .78 .87
RMSEa .22 .22 .23 3.2 2.4
Differenceb
Mean −.07 −.07 −.07 −1.0 −.9
±SD .23 .23 .24 3.2 2.4
Percentage
Acceptable 67.9 67.9 66.1 64.9 75.7
Marginal 30.3 30.3 30.3 24.3 21.6
Unacceptable 1.8 1.8 3.6 10.8 2.7
aRoot mean square error.
bDifference = estimate − observed.

Estimated forage OM intake with supplement = TDN Equations. There were two approaches taken
−1.9875 for calculating estimated total diet TDN (ETTDN, %
of OM): 1 ) TDNdev was estimated and ETTDN =
+ 1.0101 × VFIalone expected TDN + estimated TDNdev and 2 ) ETTDN
+ .0587 × (VFIalone) 2 was estimated directly and expected TDN was an
− .0195 × forage CP concentration input variable. The first approach, estimating
− .0408 × forage TDN concentration TDNdev, was not successful because all equations
− .911 × supplement TDN intake were unacceptable by all criteria. The second ap-
+ .0204 × supplement CP concentration proach, estimating total TDN directly, gave 15 equa-
+ .0699 × supplement TDN concentration
− .000569 × (supplement TDN concentration)2
+ 5.87 × supplement CP intake
− 9.74 × (supplement CP intake) 2
− .221 × VFIalone × supplement TDN intake
− .0143 × VFIalone × supplement CP concentration
+ .000509 × forage TDN × supplement TDN
+ .211 × forage type code
− .0638 × supplemental energy code

Intake values are OM as % of BW and concentration

variables are as % of OM.
As mentioned above, all intake equations were
evaluated relative to total OM intake. The regression
of observed on estimated total diet OM intake is
plotted in Figure 15. Even though the forage type code
was included, there appeared to be a discrepancy
related to forage type. Intake of diets containing cool
and warm season forages was often underestimated,
and intake of diets containing native forages and
straws was often overestimated. Nevertheless, 68% of Figure 15. Regression of observed on estimated total
estimates were acceptable, and only 2% (one compari- diet organic matter (OM) intake with the evaluation data
son) was unacceptable (Table 4). set, classified by type of forage.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
132 MOORE ET AL.

tions that were evaluated by regressing observed total

diet TDN on estimated total diet TDN. Evaluations of
two equations are shown in Table 4. The “complex”
equation was the one that best fit the development
set; it included the following variables: expected TDN
concentration, (expected TDN) 2, forage TDN concen-
tration, forage TDN × expected TDN, forage code,
supplement type code, and energy code. The simple
equation fit the evaluation set best and was as follows:
Estimated total diet TDN = 59.71

− .8948 × expected TDN concentration ( % of OM)

+ .01399 × (expected TDN concentration)2

The regression of observed on estimated total diet

TDN is plotted in Figure 16. Based on the difference
between observed and estimated diet TDN, 76% of Figure 16. Regression of observed on estimated total
estimates were acceptable and 3% (one comparison) digestible nutrient (TDN) concentration in the total diet
were unacceptable (Table 4). The evaluation data (forage plus supplement) with the evaluation data set,
subset was rather small (Table 3 ) and may not have classified by type of forage.
represented all the sources of variation in total diet
TDN that were present in the development data
subset. This equation does, however, have application at a given intake of supplemental TDN than did other
in accounting for at least part of the associative effects sources of protein. The least ADG response to supple-
on digestibility that occur in mixed diets. ment was seen with native forages supplemented with
Limits of Input Variables. Equations presented here molasses alone or with low intakes of molasses
should be used with caution when input variables are containing high levels of NPN. The greatest response
outside the range of variables in the development data was seen with improved forages, when supplemental
set. Those ranges are as follows: TDN was > 60% of OM (either dry feeds or molasses
plus added protein), and when supplemental CP
Forage OM intake fed alone ( % of BW): .46 to 3.11 intake was > .05% of BW.
Forage CP ( % of OM): 2.1 to 23.0 Effect of Supplements on Voluntary Forage Intake.
Forage TDN ( % of OM): 34.9 to 78.4 The change in VFI due to supplement ranged from −1
Supplement OM intake ( % of BW): .04 to 1.85 to +1% of BW. Generally, supplements decreased
Supplement CP ( % of OM): 6.7 to 98.4 intake with improved forages, but with native forages
Supplement TDN ( % of OM): 52.7 to 95.4 and straws, supplements both increased and
decreased forage intake. This discrepancy may be
related to the ratio of TDN to CP in forages, an
Summary indicator of the amount of N relative to available
energy. When supplements increased forage intake,
The data base constructed here provided ample forage TDN:CP ratio was > 7 (deficit of N relative to
evidence that associative effects in forage-based diets available energy). Supplements decreased intake
occur and are often important quantitatively. Also, the when the TDN:CP ratio was < 7 (adequate N ) except
data base provided the opportunity to develop equa- for ammoniated straws, when forage intake fed alone
tions to account for these associative effects in was > 1.75% of BW, or when supplemental TDN
nutritional models. intake was > .7% of BW. There was little difference
Effect of Supplements on ADG. Supplements gener- between sources of supplemental CP or TDN relative
ally but not always increased ADG. There was little to changes in forage intake. When forage intake was
relationship between supplemental TDN intake and increased by supplement, liquid and dry feeds were
the change in ADG due to supplement. In many cases, equivalent as energy sources as long as the supple-
small amounts of supplemental TDN increased gains, ment contained added protein. As protein sources,
especially with native forages and straws. The use of NPN and protein meals were apparently equivalent
escape protein tended to give greater increases in gain for increasing intake.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
 FORAGE SUPPLEMENTATION 133

Effect of Supplements on Total Diet TDN Concentra- Literature Cited

tion. When forages were supplemented, observed diet
TDN deviated from expected TDN by −10 to +5% of Brant, M. H. 1993. Predicting the effects of supplementation of
forage diets on forage intake and total diet metabolizable
OM. When diet TDN was greater than expected, the
energy concentration. M.S. thesis. Univ. of Florida, Gainesville.
forage was a native hay or straw in most cases. When Horn, G. W., and F. T. McCollum. 1987. Energy supplementation of
supplemental TDN intake was > .7% of BW, diet TDN grazing ruminants. In: M. B. Judkins (Ed.) Proc. Grazing
was generally less than expected. There was little Livest. Nutr. Conf., Univ. of Wyoming, Laramie. pp 125−136.
MacNeil, M. D. 1983. Choice of a prediction equation and the use of
effect of type or composition of forages and supple-
the selected equation in subsequent experimentation. J. Anim.
ments on the deviation from expected total diet TDN. Sci. 57:1328−1336.
Equations to Estimate Forage Intake and Diet TDN
in Mixed Diets. An acceptable equation to estimate
voluntary forage intake when fed with supplement Appendix: References Included in
was developed. Inputs to the equation included Supplementation Data Base
voluntary forage intake when fed alone ( a function of
forage quality), supplement intake, forage and supple- Anderson, S. J., J. K. Merrill, and T. J. Klopfenstein. 1988. Soybean
ment CP and TDN concentrations, and codes describ- hulls as an energy supplement for the grazing ruminant. J.
Anim. Sci. 66:2959−2964.
ing forage types and supplemental energy sources. No Balbuena, O. 1996. Effects of corn and molasses supplements with
attempt was made in this article to estimate voluntary and without feed additives on performance, voluntary intake,
forage intake when fed alone. An acceptable equation and digestive function in cattle fed bermudagrass hay. Ph.D.
to estimate total diet TDN concentration was based on dissertation. University of Florida, Gainesville.
Beames, R. M. 1959. Molasses and urea as a supplement to low
expected total diet TDN calculated from the weighted quality pasture hay for cattle. Queensl. J. Agric. Sci. 16:
average of TDN concentrations in forages and supple- 223−232.
ments. The equations could be applied in two types of Beck, T. J., D. D. Simms, R. C. Cochran, R. T. Brandt, Jr., E. S.
nutritional models: 1 ) in a static model to estimate Vanzant, and G. L. Kuhl. 1992. Supplementation of am-
moniated wheat straw: Performance and forage utilization
animal response to a known quantity of supplement characteristics in beef cattle receiving energy and protein sup-
intake and 2 ) in an iterative model to compute the plements. J. Anim. Sci. 70:349−357.
amount of supplement required to achieve a desired Biggerstaff, H. M., W. E. Kunkle, D. B. Bates, and A. C. Hammond.
rate of animal performance. The equations should be 1991. Effect of protein and molasses supplements on the perfor-
mance of calves grazing bahiagrass pasture during the fall. Fla.
used with caution outside the range of inputs in the Beef Cattle Res. Rep. pp 69−73.
data set used to develop them. Bond, J., and T. S. Rumsey. 1973. Liquid molasses-urea or biuret
(NPN) feed supplements for beef cattle: Wintering perfor-
mance, ruminal differences and feeding patterns. J. Anim. Sci.
37:593−598.
Implications Brake, A. C., A. L. Goetsch, L. A. Forster, Jr., and K. M. Landis.
1989. Feed intake, digestion and digesta characteristics of cat-
When forages are the only source of energy and tle fed bermudagrass or orchardgrass alone or with ground
protein for growing cattle, growth rates may be less barley or corn. J. Anim. Sci. 67:3425−3436.
Brown, W. F. 1993. Cane molasses and cottonseed meal supplemen-
than desired to meet production objectives. Supple- tation of ammoniated tropical grass hay for yearling cattle. J.
ments of energy and protein are often fed to increase Anim. Sci. 71:3451−3457.
growth rates, but the increase may be more or less Brown, W. F., and D. D. Johnson. 1991. Effects of energy and
than expected based on the amount and type of protein supplementation of ammoniated tropical grass hay on
the growth and carcass characteristics of cull cows. J. Anim.
supplements fed. The deviations from expectations are Sci. 69:348−357.
due to interactions among forages and supplements Brown, W. F., J. D. Phillips, and D. B. Jones. 1987. Ammoniation or
that either increase or decrease forage consumption cane molasses supplementation of low quality forages. J. Anim.
and availability of dietary energy. Several of the Sci. 64:1205−1214.
Butterworth, M. H., E. L. Aguirre, A. C. Aragon, and D. L. Huss.
interactions that affect forage intake and diet availa- 1973. Use of molasses containing urea as a supplement to
ble energy can be accounted for by equations based on pangolagrass pastures in northeast Mexico. J. Range Manage.
common characteristics used to describe forages and 26(4):269−271.
Cohen, R.D.H. 1974. Effect of molasses-urea supplements on digesti-
supplements. Estimates of performance effects and
bility of mature carpet grass ( Axonopus affinis) and liveweight
economic consequences of supplementation strategies change of beef steers. Aust. J. Exp. Agric. Anim. Husb. 14:
should be more accurate when these equations are 589−592.
included in nutritional models. Additional interactions Condon, M. J., P. C. Smith, and L. Winks. 1970. Molasses-urea
supplements for beef weaners. Queensl. Agric. J. 96:621−625.
among forages and supplements are likely, and
DelCurto, T., R. C. Cochran, D. L. Harmon, A. A. Beharka, K. A.
further research is needed to elucidate and quantify Jacques, G. Towne, and E. S. Vanzant. 1990a. Supplementation
them. of dormant tallgrass-prairie forage: I. Influence of varying sup-

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
134 MOORE ET AL.

plemental protein and(or) energy levels on forage utilization Effects on live weight gain, nutrient quality, forage intake,
characteristics of beef steers in confinement. J. Anim. Sci. 68: particulate and fluid kinetics, ruminal fermentation, and diges-
515−531. tion. J. Anim. Sci. 74:1116−1125.
DelCurto, T., R. C. Cochran, T. G. Nagaraja, L. R. Corah, A. A. Higgins, R. K., J. E. Moore, and W. E. Kunkle. 1991. Effect of
Beharka, and E. S. Vanzant. 1990b. Comparison of soybean limited grain supplement on bermudagrass hay intake and gain
meal/sorghum grain, alfalfa hay and dehydrated alfalfa pellets by steers. In: F. T. McCollum and M. B. Judkins (Ed.) Proc.
as supplemental protein sources for beef cattle consuming dor- 2nd Grazing Livest. Nutr. Conf. Oklahoma State Univ., Still-
mant tallgrass-prairie forage. J. Anim. Sci. 68:2901−2915. water (Abstr.). p 189.
Ernst, A. J., J. F. Limpus, and P. K. O’Rourke. 1975. Effect of Holderbaum, J. F., L. E. Sollenberger, J. E. Moore, W. E. Kunkle, D.
supplements of molasses and urea in intake and digestibility of B. Bates, and A. C. Hammond. 1991. Protein supplementation
native pasture hay by steers. Aust. J. Exp. Agric. Anim. Husb. of steers grazing limpograss pasture. J. Prod. Agric. 4:437−441.
15:451−455. Horn, G. W., M. D. Cravey, F. T. McCollum, C. A. Strasia, E. G.
Essig, H. W., B. R. Higgason, K. P. Boykin, and C. E. Cantrell. 1995. Krenzer, Jr., and P. L. Claypool. 1995. Influence of high-starch
Protein supplements for steers grazing bermudagrass pasture. vs high-fiber energy supplements on performance of stocker
Mississippi Agricultural and Forestry Experiment Station Res. cattle grazing wheat pasture and subsequent feedlot perfor-
Rep. 20:5. mance. J. Anim. Sci. 73:45−54.
Fike, G. D., D. D. Simms, R. C. Cochran, E. S. Vanzant, G. L. Kuhl, Horton, G.M.J., and W. Holmes. 1976. A note on the influence of a
and R. T. Brandt, Jr. 1995. Protein supplementation of am- supplement of barley and dried lucerne on the intake of barley
moniated wheat straw: Effect on performance and forage utili- straw by cattle. Anim. Prod. 22:419−421.
zation of beef cattle. J. Anim. Sci. 73:1595−1601. Hunt, C. W., J. F. Parkinson, R. A. Roeder, and D. G. Falk. 1989.
Forbes, T. J., A. M. Raven, J.H.D. Irwin, and K. L. Robinson. 1967. The delivery of cottonseed meal at three different time inter-
The utilization of grass fed indoors to young beef cattle, with or vals to steers fed low-quality grass hay: Effects on digestion
without supplementary barley. J. Br. Grassl. Soc. 22:158−164. and performance. J. Anim. Sci. 67:1360−1366.
Forster, L. A., Jr., A. L. Goetsch, D. L. Galloway, Sr., and Z. B. Jones, A. L., A. L. Goetsch, S. R. Stokes, and M. Colberg. 1988.
Johnson. 1993. Feed intake, digestibility, and live weight gain Intake and digestion in cattle fed warm- and cool-season grass
by cattle consuming forage supplemented with rice bran hay with or without supplemental grain. J. Anim. Sci. 66:
and(or) corn. J. Anim. Sci. 71:3105−3114. 194−203.
Galloway, D. L., Sr., A. L. Goetsch, L. A. Forster, Jr., A. R. Patil, W. Kunkle, W. E., D. B. Bates, D. L. Pitzer, and A. C. Hammond. 1988.
Sun, and Z. B. Johnson. 1993. Feed intake and digestibility by Effect of protein and blackstrap molasses supplements on per-
cattle consuming bermudagrass or orchardgrass hay sup- formance of calves grazing bahiagrass pasture. Fla. Beef Cattle
plemented with soybean hulls and(or) corn. J. Anim. Sci. 71: Research Report, Univ. of Florida, Gainesville. pp 87−94.
3087−3095. Lagasse, M. P., A. L. Goetsch, K. M. Landis, and L. A. Forster, Jr.
Galloway, D. L., Sr., A. L. Goetsch, L. A. Forster, Jr., W. Sun and Z. 1990. Effects of supplemental alfalfa hay on feed intake and
B. Johnson. 1991. Feed intake and digestion by Holstein steers digestion by Holstein steers consuming high-quality ber-
fed warm or cool season grass hays with corn, dried molasses, mudagrass or orchardgrass hay. J. Anim. Sci. 68:2839−2847.
or wheat middlings. J. Dairy Sci. 74:1038−1046. Lee, G. J., D. W. Hennessy, J. V. Nolan, and R. A. Leng. 1987.
Garcés-Yépez, P., W. E. Kunkle, D. B. Bates, J. E. Moore, W. W. Responses to nitrogen and maize supplements by young cattle
Thatcher, and L. E. Sollenberger. 1997. Effects of supplemental offered a low-quality pasture hay. Aust. J. Agric. Res. 38:
energy source and amount on forage intake and performance by 195−207.
steers and intake and diet digestibility by sheep. J. Anim. Sci. Lintzenich, B. A., E. S. Vanzant, R. C. Cochran, J. L. Beaty, R. T.
75:1918−1925. Brandt, Jr., and G. St. Jean. 1995. Influence of processing
Gibb, M. J., and R. D. Baker. 1987. Performance of young steers supplemental alfalfa on intake and digestion of dormant
offered silage or thermo-ammoniated hay with or without a bluestem-range forage by steers. J. Anim. Sci. 73:1187−1195.
fish-meal supplement. Anim. Prod. 45:371−381. McCollum, F. T., and M. L. Galyean. 1985. Influence of cottonseed
Gibb, M. J., and R. D. Baker. 1988. Performance of young steers meal supplementation on voluntary intake, rumen fermenta-
offered stack-treated ammoniated hay or untreated hay with or tion and rate of passage of prairie hay in beef steers. J. Anim.
without a supplement. Anim. Prod. 47:223−229. Sci. 60:570−577.
Gibb, M. J., and R. D. Baker. 1989. Performance and body composi- McCullogh, T. A. 1976. The effect on young beef cattle of sup-
tion of young steers given stack-ammoniated hay with or plementing hay with dried grass or concentrates. J. Br. Grassl.
without a supplement or untreated hay with a supplement. Soc. 31:105−109.
Anim. Prod. 48:341−351. McLennan, S. R., P. F. Dunster, P. K. O’Rourke, and G. M. Murphy.
Hall, K. L., A. L. Goetsch, and L. A. Forster, Jr. 1990. Effects of 1981. Comparison of dry season urea supplements containing
buffer or DL-methionine with different amounts of supplemen- salt, sulfur or molasses for steers grazing native pasture in the
tal corn on feed intake and nutrient digestion by Holstein dry tropics of northern Queensland. Aust. J. Exp. Agric. Anim.
steers consuming bermudagrass hay. J. Anim. Sci. 68: Husb. 21:457−463.
1674−1682. McLennan, S. R., D. J. Hirst, and P. K. O’Rourke. 1996. Effect of
Hannah, S. M., R. C. Cochran, E. S. Vanzant, and D. L. Harmon. molasses and nitrogen supplements on the liveweight perfor-
1991. Influence of protein supplementation on site and extent mance of weaner heifers grazing tropical pastures. Anim. Prod.
of digestion, forage intake, and nutrient flow characteristics in Aust. 15:718 (Abstr.).
steers consuming dormant bluestem-range forage. J. Anim. Sci. McLennan, S. R., D. J. Hirst, R. K. Shepard, and K. R. McGuigan.
69:2624−2633. 1991. A comparison of various methods of feeding supplements
Hess, B. W., L. J. Krysl, M. B. Judkins, D. W. Holcombe, J. D. Hess, of urea, sulfur and molasses to weaner heifers during the dry
D. R. Hanks, and S. A. Huber. 1996. Supplemental cracked corn season in northern Queensland. Aust. J. Exp. Agric. 31:
or wheat bran for steers grazing endophyte-free fescue pasture: 153−158.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999
 FORAGE SUPPLEMENTATION 135
McLennan, S. R., M. D. Savage, J. A. Lindsay, P. K. O’Rourke, and Stateler, D. A., W. E. Kunkle, and A. C. Hammond. 1995. Effect of
R. M. Murray. 1989. Effects of sulfur and different sources and protein level and source in molasses slurries on the perfor-
levels of nitrogen and energy on the intake and liveweight mance of growing cattle fed hay during winter. J. Anim. Sci. 73:
change of steers fed tropical native pasture hay. Aust. J. Exp. 3078−3084.
Agric. 29:157−163. Stokes, S. R., A. L. Goetsch, H. H. Nejad, G. Murphy, A. L. Jones, S.
McLennan, S. R., G. S. Wright, and G. W. Blight. 1981. Effects of Mashburn, K. W. Beers, Z. B. Johnson, and E. L. Piper. 1988.
supplements of urea, molasses and sodium sulfate on the in- Effects of supplemental bermuda grass hay or corn on intake,
take and liveweight of steers fed rice straw. Aust. J. Exp. Agric. digestion and performance of cattle consuming endophyte-in-
Anim. Husb. 21:367−370. fected fescue. J. Anim. Sci. 66:204−212.
Momont, P. A., D. D. Hinman, M. V. Boggess, C. W. Hunt, and R. C. Sunvold, G. D., R. C. Cochran, and E. S. Vanzant. 1991. Evaluation
Bull. 1994. Effects of energy supplementation of ammoniated of wheat middlings as a supplement for beef cattle consuming
barley straw on performance of cull beef cows. J. Prod. Agric. 7: dormant bluestem-range forage. J. Anim. Sci. 69:3044−3054.
437−439. Tayler, J. C., and J. M. Wilkinson. 1972. The influence of level of
Owensby, C. E., R. C. Cochran, R. T. Brandt, Jr., E. S. Vanzant, L. concentrate feeding on the voluntary intake of grass and on
M. Auen, and E. M. Clary. 1995. Grain supplementation on live-weight gain by cattle. Anim. Prod. 14:85−96.
bluestem range for intensive-early stocked steers. J. Range Templeton, J. A., R. W. Swart, and O. C. Bucek. 1970. Use of biuret
Manage. 48:246−250. in liquid supplements for ruminants. Proc. West. Sect. Am. Soc.
Petit, H. V., and D. M. Veira. 1994. Digestion characteristics of beef Anim. Sci. 21:183−188.
steers fed silage and different levels of energy with or without Vadiveloo, J., and W. Holmes. 1979. The effects of forage digestibil-
protein supplementation. J. Anim. Sci. 72:3213−3220. ity and concentrate supplementation on the nutritive value of
Petit, H. V., D. M. Veira, and Y. Yu. 1994. Growth and carcass the diet and performance of finishing cattle. Anim. Prod. 29:
121−129.
characteristics of beef steers fed silage and different levels of
Vanzant, E. S., R. C. Cochran, K. A. Jacques, A. A. Beharka, T.
energy with or without protein supplementation. J. Anim. Sci.
DelCurto, and T. B. Avery. 1990. Influence of level of sup-
72:3221−3229.
plementation and type of grain in supplements on intake and
Pitzer, D. L. 1992. Effect of blackstrap molasses and grain supple-
utilization of harvested, early-growing-season, bluestem-range
ments on performance of heifers grazing bahiagrass pasture.
forage by beef steers. J. Anim. Sci. 68:1457−1468.
Master’s thesis. Univ. of Florida, Gainesville.
Winks, L., A. R. Laing, and J. Stokoe. 1972. Level of urea for grazing
Sanson, D. W., and D. C. Clanton. 1989. Intake and digestibility of
yearling cattle during the dry season in tropical Queensland.
low-quality meadow hay by cattle receiving various levels of
Proc. Aust. Soc. Anim. Prod. 9:258−261.
whole shelled corn. J. Anim. Sci. 67:2854−2862.
Winks, L., A. R. Laing, P. K. O’Rourke, and G. S. Wright. 1979.
Sanson, D. W., D. C. Clanton, and I. G. Rush. 1990. Intake and Factors affecting response to urea-molasses supplements by
digestion of low-quality meadow hay by steers and performance yearling cattle in tropical Queensland. Aust. J. Exp. Agric.
of cows on native range when fed protein supplements contain- Anim. Husb. 19:522−529.
ing various levels of corn. J. Anim. Sci. 68:595−603. Winks, L., P. K. O’Rourke, and S. R. McLennan. 1982. Liveweight of
Smith, G. H., and B. Warren. 1986. Supplementation to improve the grazing steers supplemented with molasses, urea, and sulfur in
production of yearling steers grazing poor quality forage. 1. The northern Queensland. Aust. J. Exp. Agric. Anim. Husb. 22:
effects of forage type and a cottonseed meal supplement. Aust. 252−257.
J. Exp. Agric. 26:1−6. Winks, L., R. W. Walker, P. K. O’Rourke, I. D. Loxton, A. E. Holmes,
Stafford, S. D., R. C. Cochran, E. S. Vanzant, and J. O. Fritz. 1996. and K. A. Shaw. 1983. Molasses supplementation of steers
Evaluation of the potential of supplements to substitute for grazing a tropical grass-legume pasture in north Queensland.
low-quality, tallgrass-prairie forage. J. Anim. Sci. 74:639−647. Trop. Grassl. 17(2):64−76.

J. Anim. Sci. Vol. 77, Suppl. 2/J. Dairy Sci. Vol. 82, Suppl. 2/1999