Reducing supplementation frequency for Nellore beef steers grazing tropical

Reduced supplementation frequency is a broadly applied management practice. Ruminants consuming low quality forages/pastures, supplemented less than once daily are able to maintain body weight gain (BWG), efficiency of use of dry matter, nitrogen and other nutrients, as compared with animals supplemented once daily. We evaluated the feeding behavior, dry matter intake (DMI), dry matter and organic matter digestibility (DMD and OMD), BWG, Longissimus muscle area and backfat depth of Nellore steers raised on Brachiaria brizantha cv. Marandu pastures during the dry season, with different supplementation patterns. Thirty six animals (338 ± 40.7 kg) were distributed over nine paddocks according to a completely randomized design. Treatments were based on supplementation frequency: once daily (OD), once daily except Saturdays and Sundays (SS), or on alternate days (AD), at 1.0 %, 1.4 % and 2.0 % BW, respectively. Average total DMI accounted for 1.6 % BW day-1, with no effect of supplementation frequency. Supplementation frequency had no effect on BWG or grazing time during the day. There was no difference in Longissimus muscle area animals supplemented daily, SS and AD. The backfat depth was thinner in animals supplemented AD, but even in this case, it was within the standards considered satisfactory for a finishing steer. Reducing supplementation frequency seems a good option to lower labor costs without affecting feed efficiency or carcass quality in beef cattle grazing tropical pastures.


Introduction
Forage intake is the most infl uential factor when assessing animal performance in grazing animals, and is affected by animal, pasture and weather characteristics, as well as by their interactions (Black and Griffi ths, 1975;Webster, 1992).In tropical and subtropical areas, the availability and quality (defi ned by digestibility and crude protein (CP) content) of pastures for beef cattle is subject to great variations throughout the year.These variations are greater than those occurring in temperate areas, where production is a more important constraint than quality.
Reduced supplementation frequency is a broadly applied management practice.Beaty et al. (1994), Farmer et al. (2004) and Currier et al. (2004), demonstrated that ruminants consuming low quality forages/pastures, and supplemented less than once daily, are able to maintain BWG, effi ciency of DM use, nitrogen and other nutrients, and effi ciency of microbial synthesis, as compared with animals supplemented once daily.However, studies dealing with the effects of reduced supplementation frequency in cattle grazing tropical pastures are scarce, as a result of the number and complexity of variables to be taken into account under the conditions prevailing.
Supplementation is then likely to affect intake and digestibility of the basal diet (pasture) in a substitutive, additive or mixed way (Moore, 1980).A substitution effect is expected when infrequent supplementation is provided to cattle grazing low-quality, tropical forages during the dry season, probably as a result of the greater amount of supplement provided on supplementation days as compared with more frequent supplementation rates.Feeding behavior may also be affected by the composition, amount and pattern of administration of the concentrate, which in turn affects the amount of supplement and pasture consumed, and subsequent animal performance.
This study aimed to assess the effect of supplementation frequencies on feeding behavior, DM intake and digestibility, average BWG and carcass quality (Longissimus muscle area and backfat depth), in Nellore beef steers grazing Brachiaria brizantha cv.Marandu during the dry season in Brazil.

Experimental animals, area and treatments
Thirty six Nellore castrated steers, two years old and with an average body weight (BW) of 338 ± 40.7 kg at the beginning of the experiment, were distributed over nine paddocks (2 ha each) of Brachiaria brizantha cv.Marandu (four animals per paddock).Of these 36 Nellore steers, nine animals were fi tted with ruminal and duodenal cannulas (one animal per paddock).Treatments were based on supplementation frequency, with supplement fed (at 08h00 in troughs placed in the paddocks) once daily (OD), once daily except Saturdays and Sundays (SS), or on alternate days (AD), at 1.0 %, 1.4 % and 2.0 % BW, respectively (equivalent to 1 % BW day -1 , as-fed basis, for all three treatments)¡ Three paddocks were assigned to each treatment.Steers were placed in Sci.Agric.v.71, n.2, p.105-113, March/April 2014 the paddocks two weeks before the start of the experiment, and adaptation to the supplement began at the same time.Animals remained permanently in the paddocks for the whole length of the experiment, which lasted from July to Nov 2006, totaling 144 d.
Initially, the supplement was formulated to have a minimum of 12 % CP in the dietary DM for fi nishing cattle (NRC, 1984), assuming the low CP of the available forage during the experimental period.Thus, the supplement was formulated to contain 21 % CP in July (18 % cottonseed meal, 79.5 % citrus pulp and 2.5 % urea (44.8 g N kg -1 DM) on a DM basis), 25 % CP in Aug and Sept (27.5 % cottonseed meal, 70 % citrus pulp and 2.5 % urea), and 28 % CP in Oct and Nov (35.5 % cottonseed meal, 62 % citrus pulp and 2.5 % urea.Chemical composition of cottonseed meal and citrus pulp is shown in Table 1.

Data collection and analytical procedures
Biomass availability and pasture quality were assessed monthly from July to Nov, corresponding to d 2, d 43, d 79, d 112 and d 143 of the experimental period.Forage production was measured by randomly throwing fi ve 1 m × 1 m metallic squares into each paddock and cutting forage inside to ground level.Half of the collected biomass was separated into green leaves, stems and dead material, and the rest was used for chemical analysis, including n-alkanes concentration.
Due to the cost and diffi culty of performing the technique adopted (n-alkanes) with Nellore animals, twenty-seven animals were chosen for estimating total dry matter intake, nine less-reactive animals from each treatment (three animals from each paddock; two animals without cannula and one cannulated).Forage and supplement intake, and digestibility of the whole diet, were measured over two 12-day periods (Aug and Sept) using the n-alkanes technique.The fi rst 7 d of each period  in Sept of the experimental period) were designated for stabilization of the alkanes in the digesta.On d 8 to 12 of the alkane dosing, fecal samples were simultaneously collected twice-daily (at 7h00 and 17h00) in Aug (d 42-46 of the experimental period) and Sept (d 77-81 of the experimental period), directly from the rectum of each animal, and frozen at -20 ºC for analysis of n-alkanes concentration.At the end of each period, composite samples were formed for each animal based on the dry mass of the samples.
Samples of leaves, stems and dead material from Brachiaria, supplement ingredients and feces were milled through a 1 mm sieve.After this, 0.5 g was weighed and placed into 200 × 20 mm thick-walled screw-topped Pyrex test-tubes, adding 100 mg of internal standard (solution of heptane containing 1 mg g -1 of C 22 and C 34 ).Then n-alkanes were extracted following the technique described by Mayes et al. (1986), with the modifi cations suggested by Keli et al. (2008).
Alkane analysis was carried out by on-column injection of 0.2 µL of the eluate onto a 30 m × 0.530 mm HP-1 capillary column (1.5 µm thickness) in an Agilent 6890 gas chromatograph fi tted with an automatic injector and fl ame ionization detector.The carrier gas was helium (10 mL min -1 ) as was the make-up gas to the detector (45 mL min -1 ).The injector was programmed to track the oven's temperature program which was as follows: 230 °C for 0.2 min and a ramp of 6 °C min -1 to 300 °C, maintained for a further 18 min.Equilibrium time was set at 5 min.The detector was maintained at 350 °C throughout the whole process.Peak area data were processed using the HP ChemStation software (version A.08.03).Detector response factors for individual n-alkanes were determined by injecting onto the chromatograph a standard n-alkane mixture (C 21 -C 36 inclusive) after every eight sample extracts.
The intake of forage components (leaves, stems and dead material) and concentrate was estimated using the EatWhat program developed by Dove and Moore (1996).This requires the correction of alkane profi les of dietary components for fecal recovery, and for this purpose the values obtained by Morais et al. (2011) in a parallel experiment with the same animals fed palisade grass indoors were used.
Digestibility of DM (DMD) and OM (OMD) was calculated using the natural alkane C 31 as an internal marker as follows (Mayes et al. 1986 where I 31 and F 31 are the concentrations of C 31 in intake and feces (recovery-corrected), respectively.The former was estimated from the calculated proportions of Brachiaria leaves, stems and dead material, and supplement in the diet.Concentration of C 31 was expressed as mg kg -1 DM or mg kg -1 OM in order to obtain DMD or OMD, respectively.
Feeding behavior was studied over three consecutive days in Aug (d 24-26 of experimental period) in all animals.During these three days animals were receiving supplementation, and visual observations were taken once every 15 min from 06h00 to 18h00.Trained observers (three per paddock, working shifts of four hours each) recorded the grazing/non-grazing activity of the four animals in each paddock at the same time.Animals in SS and AD treatments were observed during the fi rst To ascertain performance and carcass characteristics, 27 Nellore steers (without cannula) were chosen, selecting three animals per paddock.Body weight was recorded at the beginning of the experiment (d 0) and every four weeks afterwards (d 29, d 57, d 86, d 115, and d 143 of the experimental period).The weight measurements were taken in the morning, before supplementation, without solid and liquid fasting.At the same time, real-time ultrasonographic measurements (Piemedical Aquila equipment fi tted with a 3.5 MHz echo sounder) of Longissimus muscle area and backfat depth (both between 12 th and 13 th ribs) were taken (Silva et al., 2005).
The N (AOAC Offi cial Method 984.13) and ash (AOAC Offi cial Method 942.05) concentrations in pasture (whole, leaves, stems and dead material), supplement components (cottonseed meal and citrus pulp) and feces (ashes only) samples were determined in accordance with AOAC (1995).The supplement ingredients were also analyzed for ether extract (EE) as described by AOAC (1995; Offi cial Method 920.39).The concentrations of neutral detergent fi ber (NDF) and acid detergent fi ber (ADF) in pasture and supplement components were determined using the method proposed by Van Soest and Robertson (1985), with samples being digested for 40 min at 110 ºC and 50.7 kPa in an autoclave (Senger et al., 2008).Acid detergent lignin (ADL) was determined according to Goering and Van Soest (1970).

Experimental design and statistical analysis
A completely randomized design was employed with three treatments (supplementation frequency) and three replications (paddocks).Herbage allowance and quality (chemical composition), dry matter intake (DMI), digestibility (DM and OM), body weight gain animal (BWG), carcass characteristics (LMA and BD) and grazing time data were subjected to an analysis of variance as repeated measures, using the PROC MIXED in SAS (Statistical Analysis System, version 9.0).Data were analyzed as a mixed model with the fi xed effects of supplementation frequency, month of study and their interaction, and random residual error.Various variance and covariance structures of errors were tested for each variable.The selection of the structure was based on the lowest Bayesian information criterion (BIC).
The covariance structures FA(1) (DMI pasture, kg day -1 ); CS (DMI pasture, kg 100 kg -1 BW and DMI supplement, kg day -1 ); UN(1) (DMI supplement, kg 100 kg -1 BW and DMI total, kg day -1 or kg 100 kg -1 BW); VC (DM and OM digestibility, and BWG) and CSH (LMA and BD), were used.Time around which grazing activity concentrated was analyzed using Rayleigh's uniformity test of the Oriana (Kovach Computing Services, version 2.02).Mean separation between mean values were tested using Tukey's test, and signifi cance was declared at p ≤ 0.05.Initial live weight was used as a covariate when analyzing BWG and carcass characteristics data.

Biomass availability and composition
Biomass availabilities per ha (total, green leaves, stems and dead material) and per 100 kg BW (total and green leaves), during the months studied, are shown in Table 2. Differences between months (p < 0.05) were observed for all variables studied.Total biomass availability was, on average, 4.5 t DM ha -1 , with the highest (p < 0.05) values in July (5.9 t DM ha -1 ).On average, 18% of the available biomass was green leaves, 59% stems and 24% dead material.These values, however, changed between months, with the forage collected in Oct and Nov having more leaves (31% and 32%, respectively), and less stems (50%) and dead material (19 % and 17%), than the pasture harvested in July (9% green leaves, 59% stems and 32% dead material, respectively), Aug (7%, 69% and 24%) or Sept (9%, 65% and 26%).These last two months showed the lowest biomass availability (3.8 t DM ha -1 ) together with the highest proportions of stems in it (67 % on average).The high biomass availability observed in July (5.9 t DM ha -1 ) matched with the highest proportion of dead material (32%).Herbage allowance (kg 100 kg -1 BW) was also higher (p < 0.05) in July, as a consequence of the greater biomass production and the lower weight of the animals.On the other hand, offer of green leaves (kg DM 100 kg -1 BW) was also higher in Oct and Nov (p < 0.05).As expected, stocking rate (not shown), calculated as kg BW 450 -1 (weight of the standard animal) per ha, increased from July (1.55) to Nov (1.74), although differences were signifi cant only for this last month.
Table 2 shows the chemical composition of available pasture from July through Nov. All analyzed parameters, except lignin (p > 0.05), were infl uenced by the month studied (p < 0.05).Crude protein content increased in Oct and Nov, whereas DM, NDF and ADF decreased, probably as a result of higher rainfall (Oct = 184.5 mm; Nov = 166.8mm) than in other months (July = 3.2mm; Aug = 19.1 mm; Sept = 37.6 mm), which led to pasture regrowth with a higher proportion of green leaves, and hence better quality.
The n-alkanes concentration in green leaves, stems and dead material from Brachiaria brizantha cv.Marandu, and the concentrate used in the present experiment is shown in Table 3.Only data for Aug and Sept are given as these were the two months when intake studies were performed.Leaves had the highest concentration in total alkanes (332.4 mg kg -1 DM on average), with a predominance of C 31 and C 33 .These two hydrocarbons were also the most abundant in the dead material and the stems, with an average concentration in total alkanes of 180.6 mg kg -1 DM and 91.7 mg kg -1 DM, respectively.As expected, the concentrate showed the lowest content in paraffi ns (53.9 mg kg -1 DM), and in these case C 25 and C 23 were the alkanes in higher proportions.The observed marked decrease in concentrations from Aug to Sept for green leaves is in accordance with the decline (although not signifi cant) in NDF content of the Brachiaria as a whole (Table 2).The chemical composition of stems (1.2 % vs. 1.6 % CP, and 81.2 % vs. 81.7 % NDF for Aug and Sept, respectively) and dead material (2.3 % vs. 2.5 % CP, and 75.1 % vs. 74.7 % NDF) changed to a much lesser extent than that of leaves (8 % vs. 14.4 % CP, and 63.2 % vs. 61.4% NDF), and hence the decline in alkane concentrations was much lower (stems) or did not exist all (dead material).

Intake, diet composition and digestibility
Supplementation frequency did not affect total, pasture or DM concentrate intake (Table 4), neither as kg day -1 nor as kg 100 kg -1 BW (p > 0.05).On the other hand, the month of study had an effect on all variables (p < 0.05) except total and DM concentrate intake expressed as kg 100 kg -1 BW (p > 0.05).The interaction between supplementation frequency and month of study was not signifi cant in any case (p > 0.05).
Intake studies using alkanes were also tried in Nov. However, estimated values for concentrate were none in most cases whereas for pasture they were too low to be compatible with either the registered BWG or the voluntary intake expected for animals of the recorded BW.Availability of biomass and proportion of green leaves in it was much higher in Nov than in Aug or Sept (Table 2).The alkanes concentration was also much higher in leaves than in the rest of the components of the diet (Table 3), and this could have led to a reduced proportion of alkanes from concentrate in feces, and then to an error in the estimation of diet composition and hence intake (Dove and Mayes, 2005).Under these circumstances, the supplement could have been labelled with an external source of hydrocarbons, such as beeswax (Dove and Oliván, 1998;Dove et al., 2002).However, in a study dealing with the estimation of intake and diet composition in sheep fed mixed grain/roughage diets, Valiente et al. (2003) concluded that the n-alkane method might be used for this purpose even though the concentrations of these hydrocarbons in their experiment was low.The difference from the present experiment was that in their case both barley straw and grain had low contents of n-alkanes, and here leaves showed much greater values.Maybe alkanes should be incorporated into supplements when their concentration in these is scarce and much lower than that in forages, but not in other cases.
With respect to the intake and diet composition values obtained in Aug and Sept (Table 4) their accuracy in grazing conditions can only be tested by indirect methods.Energy and protein requirements of the animals used in the present experiment were calculated according to the AFRC (1993), and taking into account the BWG values shown in Table 5.The nutritive value of the Brachiaria was as from Rueda et al. (2003), and those of citrus pulp, cottonseed meal and urea were obtained from the AFRC (1993) tables.Estimated requirements were compatible with pasture and supplement intakes (Table 4).Animal requirements were also estimated taking into account the extra activity needed to acquire the feed in grazing conditions (SCARM, 1990).In this case, however, the increased needs were not covered by the estimated intakes (Table 4).It is possible that either these intakes were underestimated or, more likely, that the extra requirements foreseen by the SCARM (1990) did not apply in the present experiment due to the availability and quality (Table 2) of the pasture, and to the supplement intake.
There were low pasture intake estimates (1.1 kg 100 kg -1 BW, on average; Table 4) with no effect of supplementation frequency (p > 0.05).Canesin et al. (2007) also observed low pasture intakes (0.83 % BW) during the dry season in crossbred steers grazing Brachiaria brizantha and supplemented (1 % BW) at the same frequencies as in the present experiment.Krehbiel et al. (1998) showed that dry matter intake and net portal and hepatic fl ux of nutrients in mature ewes fed low-quality forage increased in response to supplementation with 80 g CP day -1 in the form of soybean meal, with no differences due to the frequency of administration of the supplement (once daily or every three days).Also Huston et al. (1999) did not observe sdifferences in forage and supplement intake due to supplementation frequency, 1x week (1 % BW) or 3x week (0.45 % BW), in Hereford × Brangus adult beef cattle in winter.The opposite results were reported by Beaty et al. (1994), who observed that a reduction in supplementation frequency, from 7x (0.4 % BW) to 3x week (1 % BW), was refl ected in a reduction of straw and total DM intake in Angus × Hereford steers.
Average pasture DM intake values were rather low and the question why animals did not eat more should be asked (Table 4).Time spent grazing during the day (6h00 to 18h00) was scarce (3.7 h on average, with no differences between supplementation frequencies; Table 6) and it is not likely that grazing at night would have increased this fi gure by more than 35 % (Dulphy et al., 1980).Probably the high proportion of stems and their low quality (Table 2) made the process of resource acquisition a negative aspect of feeding (Tolkamp and Ketelaars, 1992).Theoretically, the animals were able to eat more concentrate and hence gain more weight but Ketelaars and Tolkamp (1996) have stated that animals have their own objectives that are possibly quite different from rapid growth or high milk production.It is likely that oxygen effi ciency of feeding behavior, i.e., net energy intake per liter of oxygen consumption (Ketelaars and Tolkamp, 1996) would have been less favorable if animals had increased their intake.
Estimated DM digestibility (DM and OM) was not affected (Table 4) by supplementation frequency (p > 0.05) but it was by the month of study (lower values in Aug than in Sept; p < 0.05).These results are in accordance with the higher proportions (although nonsignifi cant; Table 2) of leaves in the available pasture biomass.However, lower intakes (4.9 vs. 5.5 kg day -1 ) and higher supplement proportions in the diet (0.35 vs. 0.24) were estimated in Aug (Table 4), and this should have resulted in higher DMD (Waldo et al., 1972).Not only were the DMD estimates odd in Aug but also those for OMD, which were abnormally lower than DMD estimates.When a closer look into the raw data was taken, it became patently obvious that estimates of OM intake in Aug included, on average, 0.15 dead material.This was the only case where intake of this fraction was estimated, as DMI estimates (in both Aug and Sept) or OM intake (OMI) estimates in Sept did not include this fraction.
Dead material had the second highest level of C 31 concentration, only after leaves (Table 3).Hence, a relatively small estimated intake of this fraction would have greatly increased the estimated concentration of C 31 in the whole diet.That was the case in the present experiment, leading to OMD values lower than DMD in Aug.The increase in C 31 concentration in the estimated OM intake (whole diet), with regard to concentration in DM intake, was proportionally higher than the increase in fecal concentration of C 31 in OM with respect to concentration in DM.With regard to the lower DMD values found in Aug than in Sept, it is possible that animals actually ingested some dead material in the fi rst month that is not revealed by the n-alkane technique.
Paraffi n concentrations of diet components on a DM basis were lower than on an OM basis, probably leading to less accurate estimates of diet composition (not presented) on a DM basis than on an OM basis.Although DM intake estimates were compatible with estimated animal requirements, as discussed above, it seems that more accuracy is needed when estimating diet composition, as it has a signifi cant effect on digestibility.Treating supplements with external alkanes should then be a common practice when the study of this variable is sought in supplemented grazing animals.
In absolute terms, digestibility values were about 2-19 % above those reported by Pereira et al. (2008) working with beef cattle fed diets containing Brachiaria brizantha silage and concentrate at different ratios (71 % for both DMD and OMD, and 35 % concentrate in the diet).Even though silage is expected to have lower digestibility than green forage, and the concentrate used by the above authors was composed of ingredients other than those used in the present experiment, discrepancies may refl ect, one more time, the need for a more accurate use of the n-alkanes technique.
Papers dealing with the effect of supplementation frequency in animals fed tropical pastures on DM or OM digestibility are scarce.To our knowledge, this is the fi rst attempt to estimate intake and digestibility in beef steers grazing a tropical pasture and supplemented at different frequencies using the n-alkane technique.Although this latter can, and must be improved, the results obtained were within the range considered 'normal' for the stated conditions.

Body weight gain
Supplementation frequency did not affect BWG (p > 0.05), although this was affected by the month of study (p < 0.05), with higher fi gures in Oct and Nov (Ta- ble 5).The higher biomass availability and proportions of green leaves in the pasture together with the higher CP and lower NDF content may explain the differences (Table 2), and also the increasing % CP of the supplement leading to increased forage intake.These results are in accordance with those reported by other authors (Huston et al., 1999;Schauer et al., 2005) who observed similar BWG in ruminants fed on low-quality forages and supplemented either infrequently or once daily.
Ruminants are effi cient in maintaining adequate levels of N in the rumen, even during short periods of undernutrition, i.e. infrequent supplementation in animals fed low quality roughages (Bohnert et al., 2002).Urea recycling is known to have a relevant role.These authors stated that, even in conditions of infrequent supplementation, rumen microorganisms were able to adequately ferment structural carbohydrates then coping successfully with maintaining animal performance.Canesin et al. (2007) recorded BWG of crossbred beef steers grazing Brachiaria brizantha cv.Marandu at a low stocking rate (1.32 AU ha -1 , on average), and supplemented once daily, once daily except Saturdays and Sundays, or on alternate days, at 1 % day -1 of their BW.Supplement was composed of corn, soybean meal and urea, and there were no differences in BWG between supplementation frequencies, with an average value of 0.54 kg day -1 .

Carcass characteristics
The area of the Longissimus muscle area (LMA) was similar among the supplementation frequency treatments (p > 0.05), and differences were observed in backfat depth (BD) (p < 0.05) (Table 5).BD was thinner (p < 0.05) in animals supplemented AD (3.22 mm), but even in this case it was within the standards considered satisfactory for a fi nishing steer (3-6 mm).In a study with crossbred steers grazing Brachiaria brizantha cv.Marandu Canesin et al. (2006) did not fi nd differences in LMA or BD between supplementation frequencies (once daily, once daily except Saturdays and Sundays, or on alternate days), with average values slightly higher for LMA (59.7 cm 2 ) and lesser for BD (3.3 mm), probably due to the use of a different genetic basis (Bos indicus × Bos Taurus).
Most of the information on carcass characteristics using ultrasonography has been obtained from confi ned animals, and the technique has hardly been used under grazing conditions.The reasons may be found in the more aggressive behaviour and the thinner BD of free-ranging animals, which makes it diffi cult to obtain good images.Apart from this, Williams (2002) showed similar correlations between ultrasound and carcass measurements for backfat (r = 0.81 to 0.86) and Longissimus muscle area (r = 0.61 to 0.76).

Feeding behavior
Table 6 shows that grazing activity was scarce between 6h00 and 18h00, with most of the time dedi-cated to pasture intake occurring after noon, probably as a consequence of supplementation being offered at 08h00.Time spent grazing between 6h00 and 18h00 did not differ among supplementation frequencies (p > 0.05), and accounted for 3.93, 3.92 and 3.33 h for OD, SS and AD, respectively.However, the time around which grazing activity was concentrated (mean grazing time) was 15h00 for animals supplemented OD, and 15h30 and 15h45 for those on SS and AD, respectively.Differences were found only between OD and the two other supplementation frequencies (p < 0.05).These results outline the relationships between grazing time, and pasture intake (Table 4) and weight gain (Table 5), which were not affected by supplementation frequency either.Schauer et al. (2005) did not fi nd differences (p > 0.05) in grazing time between Angus × Hereford cows grazing native ranges in the northern Great Basin supplemented once daily or every six days (7.08 vs. 7.87 h day -1 ), whereas the grazing time was 2.1 h longer in nonsupplemented animals.These results are in agreement with those found by Beaty et al. (1994), who did not observe differences in grazing time between pregnant Angus × Hereford beef cows grazing dormant tallgrass prairie pastures supplemented daily or three times per week (6.8 vs. 7.0 h day -1 ).
Overall, reducing the supplementation frequency to 5x weekly or on alternate days appeared to be an efficient management practice in the present experiment, as there was no reduction in alkane-estimated total intake with respect to animals supplemented daily.There were no differences either in terms of grazing time during day hours or average daily gain, even when the amount and quality of pasture were limiting.Hence, reduced supplementation frequency can be adopted with the aim of reducing labor costs.

Table 1 -
Chemical composition of the citrus pulp and the cottonseed meal used in the supplement.
DM: dry matter; CP: crude protein; NDF: Neutral Detergent Fiber; ADF: Acid Detergent Fiber.Sci.Agric.v.71, n.2, p.105-113, March/April 2014 day, those in OD and SS treatments during the second day, and those in OD and AD treatments during the third day.

Table 3 -
n-Alkanes concentration (mg kg -1 dry matter) in green leaves, stems and dead material from Brachiaria brizantha cv.Marandu, and in the concentrate used in the present experiment.
21 , C 24 and C 28 were not detected; C 32 and C 36 have been excluded as they were dosed for purposes not related to the present experiment.Sci.Agric.v.71, n.2, p.105-113, March/April 2014

Table 4 -
Dry matter intake (DMI), and dry (DMD) and organic matter (OMD) digestibility in Nellore beef steers grazing Brachiaria brizantha cv.Marandu in two months and supplemented with three patterns.
SEM: Standard error of the mean; p: Probability of the differences; OD: Supplement given once daily; SS: Supplement given once daily except Saturdays and Sundays; AD: Supplement given on alternate days.F: Effect of supplementation frequency; M: Effect of month of study; F*M: Interaction.

Table 5 -
Average daily body weight gain (BWG), Longissimus muscle area (LMA) and backfat depth (BD) in Nellore beef steers grazing Brachiaria brizantha cv.Marandu in different months and supplemented with three patterns.
SEM: Standard error of the mean; p: Probability of the differences; OD: Supplement given once daily; SS: Supplement given once daily except Saturdays and Sundays; AD: Supplement given on alternate days.F: Effect of supplementation frequency; M: Effect of month of study; F*M: Interaction.a, b Different letters indicate differences (p < 0.05) between months; A, B Different letters indicate differences (p < 0.05) between supplementation frequencies.Sci.Agric.v.71, n.2, p.105-113, March/April 2014

Table 6 -
Feeding behaviour between 06h00 and 18h00 of Nellore beef steers grazing Brachiaria brizantha cv.Marandu and supplemented with three different patterns.
SEM: Standard error of the mean; OD: Supplement given once daily; SS: Supplement given once daily except Saturdays and Sundays; AD: Supplement given on alternate days; p: signifi cance; *Time around which grazing activity was concentrated.a, b Different letters indicate differences (p < 0.05) between supplementation frequencies.Sci.Agric.v.71, n.2, p.105-113, March/April 2014