If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Milk production of cows grazing pasture supplemented with grain mixes containing canola meal or corn grain or both over the first 100 days of lactation
Grain mixes varying in proportions of wheat grain, barley grain, canola meal and corn grain were fed to grazing dairy cows in early lactation to determine the contribution of canola meal and corn grain to milk yield, body weight (BW), body condition score (BCS), eating behavior and blood serum metabolite concentrations. The experiment used 80 multiparous, seasonally calving Holstein-Friesian dairy cows during the first 100 d of lactation, the treatment period, and over the subsequent carryover period of 100 d, during which all cows were fed a common diet. Cows were divided into 4 cohorts (blocks) based on calving date and within each cohort, 5 cows were randomly allocated to each of the 4 treatments. Dietary treatments included disc milled grain mixes comprising (on a dry matter (DM) basis) 1) a control treatment of wheat (25%) and barley (75%); 2) wheat (25%), barley (50%) and canola meal (25%); 3) wheat (25%), barley (50%) and corn (25%), and 4) wheat (25%), barley (25%), canola meal (25%), and corn (25%). Treatment diets were introduced at 19 d in milk (DIM) ± 4.7 d which included a 7-d adaptation period and were applied up until 100 DIM. Each grain mix was fed at 9 kg DM/cow per d, offered twice daily, in equal proportions in the parlor at milking times. In addition to the grain mix, all cows grazed perennial ryegrass pasture at a daily allowance of approximately 35 kg DM/cow per d (measured to ground level). Results were analyzed in terms of corn and canola presence or absence in the diet. Including canola meal in grain mixes increased grain intake and pasture intake by 0.6 and 2.1 kg DM/cow per d, respectively, resulting in an increased milk yield of 2.6 kg/cow per d during the first 100 d of lactation. Including canola meal also increased yields of milk fat and protein, and concentrations of milk fat, as well as increasing mean BW and BCS over the 100 d. The inclusion of canola meal in the grain mixes also resulted in greater blood serum β-hydroxybutyrate and urea concentrations, compared with feeding grain mixes that did not contain canola meal. The inclusion of corn grain provided no milk production benefits and did not change BW, BCS or any feeding behavior variables. There were no carryover effects on milk production from either canola meal or corn grain after the treatment period. In summary, the results demonstrate that the provision of canola meal in grain mixes can improve milk production and increase mean BCS. Further, there are no benefits to milk yield when a proportion of barley is substituted for corn, in a wheat and barley grain mix fed to grazing dairy cows in early lactation. However, these results are dependent on the level of inclusion and the feeding system employed.
Interpretive summary: Supplying canola meal and corn grain to grazing dairy cows has previously been shown to have a positive impact on milk production. The objective of this study was to determine the individual and combination effects of feeding canola meal and corn grain to early lactation dairy cows that were grazing perennial ryegrass pasture. The results showed that feeding canola meal can improve milk production and body condition, while no benefits were found from feeding corn grain.
INTRODUCTION
Grazed pasture is a major source of nutrients for dairy cattle in many parts of the world, including Australia, New Zealand and Ireland, because of its inherent low cost (
). However, feeding pasture alone to high producing dairy cows often limits milk production as the seasonal variations in pasture growth and nutritive characteristics mean that pasture nutrient availability is often not sufficient to meet animal demand (
). This is particularly true in early lactation. Early lactation is a critical period for the successful nutritional management of dairy cattle. Following calving, dietary intake is often unable to meet the nutrient demands of high milk production (
). During this time, cows often enter a state of negative energy balance (Gross et al., 2011) that can persist until around 100 DIM (wk 14 of lactation;
Dietary intervention to minimize loss of body condition and support milk production during early lactation typically involves increasing the energy density of diet by supplementation with concentrates and or fats (
). Previous research has demonstrated that corn grain and canola meal have proven benefits to milk production as sources of dietary starch and protein in TMR systems, respectively (
, conducted in early lactation, also showed that cows grazing pasture and fed a grain mix containing wheat grain, canola meal, and corn grain increased ECM yield by up to 4.9 kg/cow per d compared with cows offered an equivalent amount of wheat grain. The authors suggested that this response could be attributed to an increase in DMI associated with the canola meal providing additional CP, the corn grain providing a more slowly digestible starch source, and the overall diet resulting in stabilized milk fat concentrations when the amount of supplement increased. However, the experiment of
replaced the canola meal in a wheat, barley, corn and canola meal supplement, with urea and a fat supplement to mimic the qualities of canola meal, the same improvements to milk solids were not seen, highlighting that the effects of canola meal were driven by factors more complex than a simple increase in the intake of fat and N. Understanding the relationship between corn grain, canola meal and milk responses over a longer duration than undertaken in the studies by
, is needed to provide important information to farmers to make more informed decisions about supplementary feeding and economically optimizing milk production. In addition, research has shown that concentrates supplied to grazing dairy cows during early lactation not only impact immediate milk production, but also influence performance through the remainder of lactation (
The effect of herbage allowance and concentrate supplementation on milk production performance and dry matter intake of spring-calving dairy cows in early lactation.
Our aim was to determine the contribution of canola meal and corn grain in isoenergetic supplement mixes to milk yield in early lactation cows grazing perennial ryegrass (Lolium perenne L.) pasture. These 2 feed components were chosen because they are widely available, albeit expensive, in the dairy industry of southeast Australia and elsewhere, and because of the previously observed increases in DMI and productivity, described above. The supplement mixes used in the experiment varied in proportions of wheat grain, barley grain, corn grain and canola meal. The research hypotheses tested were that during the first 100 d of lactation: (1) milk yield would be greater for cows consuming a grain mix that contains canola meal compared with grain mixes that do not contain canola meal (main effect of canola); (2) milk yield would be greater for cows consuming a grain mix that contained corn compared with cows consuming a grain mix that did not contain corn (main effect of corn); and (3) there would be an interaction effect (either crossover or non-crossover) of canola meal and corn on milk yield. In the period from 101 to 200 d when cows were fed a common diet, the carryover effect of the canola meal and corn on milk yield was also determined.
MATERIALS AND METHODS
The experiment was conducted at the Agriculture Victoria Research Centre, Ellinbank, Victoria, Australia (38°14′S, 145°56′E). All procedures were conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes (
). Approval to proceed was obtained from the Department of Energy Environment and Climate Action's Agricultural Research and Extension Animal Ethics Committee.
Experimental Design and Management
The experiment commenced in July 2017 and used 80 multiparous Holstein-Friesian dairy cows that had calved in late winter/early spring. All cows were in their 2nd to 5th lactation, with an average BW of 565 kg ± 6.1 (SEM) and were milked twice daily at approximately 0600 and 1500 h through a common parlor. Following calving, cows were allocated into 1 of 4 cohorts of 20 cows; the first 20 cows that calved were recruited into the first cohort, and so on. Each cohort was treated as a block and each treatment was allocated 5 of 20 cows in each block resulting in a randomized complete block design. The treatments were allocated to cows randomly by balancing for age, parity, BW, calving date, BCS and average milk yield for 14 d postpartum using the method of Harville (1974). The study design was implemented in the software GenStat for Windows as the procedure COVDESIGN (GenStat 18th edition, VSN International Ltd., Hemel Hempstead, UK). A pre-experimental period began at calving, during which cows spent 4 d in a colostrum herd where they received grazed pasture (30 kg DM/cow per d measured to ground level), clover hay (4 kg DM/cow per d) offered on a feed pad, and a wheat (2.8 kg DM/cow per d) and barley grain mix (2.8 kg DM/cow per d) fed individually over 2 feeds during milking times. This was followed by a minimum of 10 d where cows received the same pasture and grain mix as the colostrum herd but the hay was replaced with a mixed ration containing vetch hay (1.3 kg DM/cow per d), corn silage (5 kg DM/cow per d), wheat grain (1.5 kg DM/cow per d), canola meal (0.75 kg DM/cow per d, corn grain (0.75 kg DM/cow per d, limestone (100 g/cow per d), magnesium oxide (15 g/cow per d) and minerals (250 g/cow per d), offered once per d on a feed pad.
Cows were transitioned gradually to treatment diets commencing at 19 ± 4.7 (mean ± SE) DIM and reached the full treatment diet at 26 ± 4.7 DIM. The treatment period concluded when each cohort reached an average of 100 DIM. As each cohort began their treatment period, they were added to the existing cohorts and were managed as a single herd. When each cohort reached approximately (63 ± 5.9 DIM) cows underwent a measurement period of 6-d during which individual pasture intakes were estimated using the n-alkane technique. During the measurement period, each cohort grazed in a separate paddock. All cows grazed perennial ryegrass (Lolium perenne L.) pasture. During the treatment period pre-grazing pasture mass was 3900 kg DM/ha, post-grazing pasture mass was 2100 kg DM/ha and daily pasture allowance was 31 kg DM/cow per d, all measured to ground level. During the measurement period, pre-grazing pasture mass was 4500 kg DM/ha, post-grazing pasture mass was 2300 kg DM/ha and daily pasture allowance was 36 kg DM/cow per d, all measured to ground level. Pasture allowance was set to ensure cows had ad libitum intake.
Cows were individually offered 1 of 4 grain mix treatments at 9 kg DM/cow per d:
The composition of each of the grain mix treatments is presented in Table 1, all grains were finely disc milled and all components thoroughly mixed together. Included in the each grain mix was an additional vitamin and mineral supplement comprised of a mineral mix (Barastoc Dairysure 250 plus Rumensin, Barastoc, Melbourne, VIC, Australia), 246 g/cow per d (236 mg rumensin/cow per d); magnesium oxide (E Mag 523, Swancorp, Rocklea, QLD, Australia), 22 g/cow per d; limestone, 69 g/cow per d; canola oil, 56 g/cow per d; sodium bicarbonate, 151 g/cow per d and virginiamycin (Eskalin 500 feed premix), 15 g/cow per d, formulated using the NDS Professional (version 6.55; RUM&N, Reggio Nell'Emilia, Emilia-Romagna, Italy) to ensure the requirements for vitamins and minerals were satisfied for a production level of 35 L.
Table 1Composition of the 4 grain mixes; WB, WBCM, WBCR, and WBCMCR (kg DM/cow per d)
When the treatment period finished at 100 DIM, all cows were transitioned onto a common diet of grazed perennial ryegrass pasture at an approximate allowance of 20 kg DM/cow per d, supplemented with 6 kg DM/cow per d of a wheat and barley grain mix (50:50).
Grain Mix Intake, Pasture Measurements and Nutritive Characteristics
During the treatment period, grain mixes were fed individually via automatic feed heads in the parlor during milking times. Samples of each of the 4 grain mixes (bulked per treatment group) were collected once a week (PM and AM) during the treatment period. Each sample was subsampled twice after bulking. The first was used to determine DM concentrations by drying at 105°C for 24 h, and the second subsample was frozen at −18°C, freeze-dried, ground through a 0.5-mm sieve and analyzed at a commercial laboratory (Dairy One Forage Laboratory, Ithaca, NY) for nutritive characteristics via wet chemistry (
The amount of grain mix refused was measured twice daily for each cow during the entire treatment period. The difference between the amount of grain mix offered and refused was used to calculate individual DMI of the supplement. In the carryover period, only the amount of grain mix offered was recorded.
Throughout the treatment period, cows were offered half of their daily pasture allowance after each milking and were prevented from re-grazing areas grazed on previous days with the use of back-fencing. To ensure appropriate pasture allocation so as to not limit intake, pre- and post-grazing pasture mass was estimated using a rising plate meter (
). During the treatment period, when cows grazed as a single herd, pre- and post-grazing pasture mass was determined once every 2 weeks. Rising plate meter readings (150 plonks) were taken once pre-grazing and once post-grazing in a zig-zag pattern over the allocated plot of pasture. The pasture meter was calibrated for each new set of paddocks the cows entered by using the meter height and pasture DM mass within 18 quadrats to construct calibration equations plotting actual pasture mass against pasture meter reading. The difference between pre- and post-grazing pasture mass was used to calculate average pasture DMI for a group of cows.
Samples of pasture cut to grazing height (∼5 cm) were collected before grazing for analysis of nutritive characteristics and total fatty acid concentrations. These samples were collected every 2 weeks outside the measurement period and daily during the measurement period. All samples were collected using electric shears at several points along a transect of the grazing area. Pasture samples were thoroughly mixed, sub-sampled, washed, freeze-dried, and ground through a 0.5-mm sieve. Dried feed samples were sent to a commercial laboratory (Dairy One Forage Laboratory, Ithaca, NY) for analysis of CP (method 990.03 (
WB: Disc milled wheat and barley grain mix; WBCM: Disc milled wheat, barley grain and canola meal mix; WBCR: Disc milled wheat, barley grain and corn grain mix; and WBCMCR: Disc milled wheat, barley grain, canola meal and corn grain mix.
and pasture samples collected during the treatment period, as well as total nutrient intake for each of the treatments. Nutritive characteristics are % of DM and nutrient intakes are kg/cow per day, unless otherwise indicated
. The timing of each measurement period was staggered for each cohort according to calving date to ensure individual intake was estimated over a similar DIM interval. The measurement period began at an average of 63 ± 5.9 DIM. At this point all cows were adapted to their treatment diets. Cows were separated into each cohort of 20 cows at least 6 d before the commencement of the n-alkane dosing period to ensure their intakes had stabilized. During this time pre- and post-grazed pasture mass was determined each d. During the measurement period the offered grain mix was weighed manually and offered to cows individually at each milking for 12 d, commencing 6 d before the measurement period. The amount of grain mix refused was determined twice daily, for each cow, during this period. Samples of each of the 4 grain mixes offered and refused were collected each d (PM and AM) during the measurement period.
Cows were orally dosed twice daily following each milking for 12 d with paper pellets (Carl Roth GmbH and Co.KG, Karlesruhe, Germany) impregnated with dotriacontane (378 mg/pellet). Beginning on the seventh d of the dosing period, fecal, grain mix and pasture samples cut to grazing height were collected twice daily for 6 d. Grain mix and pasture samples were immediately stored at −18°C following collection, then freeze-dried at −55°C for a minimum of 72 h. Following freeze drying, feed samples were ground through a 1-mm screen and analyzed for n-alkane concentrations. Fecal samples were stored at −18°C and at the completion of collection were defrosted and oven-dried for 72-h at 40°C. Once dried, samples were ground to 1-mm and equal proportions were composited into a 10 g sample per cow and analyzed for n-alkane concentrations. The concentrations of n-alkanes were analyzed by GC using the methods described by
Protocol for the analysis of n-alkanes and other plant-wax compounds and for their use as markers for quantifying the nutrient supply of large mammalian herbivores.
. Peak areas were converted to amounts (mg/kg DM) of n-alkane by reference to the internal standard (tetratriacontane, C34). Individual pasture intake was estimated using the following equation:
[1]
Where Fi, Pi and Gi are the fecal, pasture and grain mix concentrations of odd-chain n-alkane (C33), Fj, Pj and Gj are the fecal, pasture and grain mix concentrations of even-chain length alkane C32, and Dj is the daily dose of even-chain C32 alkane. Grain mix intakes (I) were entered.
Milk Yield and Composition
Milk yield of every cow was measured twice daily using a DeLaval Alpro milk metering system (DeLaval International, Tumba, Sweden). Using in-line milk samplers (DeLaval International, Tumba, Sweden), composite milk samples (PM and AM) were collected from each cow once/week during the treatment period, 3 times per week during the measurement period and once per month during the carryover period. Samples were analyzed for fat and protein using an infrared milk analyzer (Model 2000, Bentley Instruments, Chaska, MN, USA).
Body Weight and Body Condition Score
Individual cow BW were recorded each morning using walkover scales as cows exited the milking parlor (DeLaval Automatic Weight System AWS100, Tumba, Sweden). Every cow was assessed for BCS once every 2 weeks throughout the experiment by 2 trained assessors according to the 8-point scale of
Feeding behavioral characteristics were determined with pressure and movement sensors (RumiWatch, Itin + Hoch GmbH, Liestal, Switzerland) attached to head halters. Before the attachment of the RumiWatch halters, ordinary cattle halters were attached to each cow for 24 h to allow cows to become familiarized with wearing a halter. Each cow was then fitted with RumiWatch halters for their 6-d measurement period. Data were analyzed using the RumiWatch manager 2 version 2.2.0.0 and RumiWatch converter software version 0.7.4.13. Data were converted into hourly summaries and then summed for each 24-h period before analysis. The RumiWatch halters determined time spent eating and ruminating.
Blood Sampling
Samples of blood were collected from each cow via the coccygeal vessels once every 2 weeks following the morning milking; a total of 9 samples were collected per cow during the experiment. Samples were not staggered for individual cows; they were collected once every 2 weeks on the same d for all cows. Each blood sample was collected into a 10-mL serum clot activator vacutainer tube (Beckton Dickson) and allowed 2 h to clot at 24°C before being centrifuged at 1,200 x g for 15 min at 24°C. Serum was then decanted and stored at −20°C before analysis of concentrations of BHB, nonesterified fatty acids (NEFA), urea, total Ca, Mg, total protein, and albumin at Regional Laboratory Services (Benalla, Victoria, Australia) using a Kone 20 XT clinical chemistry analyzer (Thermo Fisher Scientific, Waltham, MA). Reagents for the analyses of fatty acids, urea, Ca, and Mg were sourced from Randox Laboratories (Crumlin, UK) and those for the analyses of BHB, albumin, and total protein were provided by Regional Laboratory Services (Benalla, Victoria, Australia).
Statistical Analyses
Daily data were averaged over treatment and or measurement periods to give a mean for each cow for each period, before analysis by analysis of covariance (ANCOVA) using the software Genstat (GenStat 18th edition, VSN International Ltd., Hemel Hempstead, UK). In the case of behavior data, counts and durations of activities were summed within each 24-h period (midday to midday) before averaging over d for each cow. Average rates of change in BW and BCS for each cow were calculated as linear regression slopes over the 100 d of the treatment period and analyzed in the same way by ANCOVA. The individual average pasture DMI measured by n-alkane technique were also analyzed by ANCOVA. The study was designed with a 2*2 factorial arrangement of treatments and the resulting 4 treatment diets were coded for the presence (+) or absence (-) of corn and canola meal in the diet.
•
WB: canola (-), corn (-)
•
WBCM: canola (+), corn (-)
•
WBCR: canola (-), corn (+)
•
WBCMCR: canola (+), corn (+)
The analysis of covariance (ANCOVA) model in the Genstat had a treatment structure canola * corn which expands to canola (main effect) + corn (main effect) + canola × corn (interaction effect). F-tests in the ANCOVA were used to test the main effects and interaction of canola and corn. The main effect of canola is given by the contrast between treatment means that have canola meal and do not have canola meal. Similarly, the main effect of corn is given by the contrast between treatment means that have corn and that do not have corn. The interaction term canola.corn tests whether the effect of corn depends on the presence of canola. This is mathematically equivalent to the contrast for whether the effect of canola depends on the presence of corn. The blocking structure was cow within cohort, and covariates were pre-calving age, parity, BW, calving date, BCS and average MY for 14 d postpartum and were used to balance treatment groups within cohorts. The full ANCOVA model fitted was as follows:
where yij was the outcome variable for cow i in cohort j and μ the overall mean; Ck(ij) was the main effect of canola meal in the diet, k(ij); Nl(ij) was the main effect of corn in the diet, l(ij); CNkl(ij) was the interaction effect between canola meal and corn in the diet. For cow i in cohort j,aij was the pre-calving age with coefficient β1; pij was the parity with coefficient β2; wij was the body weight with coefficient β3; dij was the calving date with coefficient β4; sij was the BCS with coefficient β5;mij was average milk yield for 14 d postpartum with coefficient β6; Kj was an effect of cohort j and εij was the residual error.
The blood measurements were analyzed using linear mixed models which were similar to Equation 1, but with an added time factor and its 2-way and 3-way interaction canola and corn, with individual cow as the unit of analyses relevant to treatment diet effects. The fixed effects included the main effect of canola meal, the main effect of corn, main effect of measurement time (DIM interval), and all 2-way and 3-way interactions among canola meal, corn and measurement time. The effect of cow and - measurement date within cow were fitted as random effects and the measurement date within cow was used as a residual error term. The random effects were assumed to follow a normal distribution with zero mean and constant variance.
Distributional assumptions of normal distribution and constant variance were checked graphically using plots of residuals against fitted values, normal quantile plots and histograms of residuals. The alkane measurement of pasture intake for a cow offered the WBCMCR diet, feeding behavior of 2 cows on the WBCMCR diet, blood serum BHB of 3 cows (1 on WBCR, 2 on WB), serum calcium of 1 cow on the WBCR diet, serum protein of 2 cows on WB diet, serum urea of 1 cow on WBCR diet, serum albumin of 1 cow on WB diet and serum albumin:globulin ratio of 1 cow on WB diet were missing and treated as missing values in the analyses. The linear mixed model was used to analyze the blood serum measurements because of these missing values. The serum NEFA values were logarithmically transformed to satisfy the assumption of normality with constant variance. P-values of <0.05 were considered statistically significant.
The total nutrient intake for each diet treatment was calculated as
Where pj and gj are percentages of the nutrient in pasture and grain from Table 2 and the DMIs from Table 3. The standard errors for total nutrient intake were calculated using a formula for the variance of a product of independent random variables.
Table 3Mean daily milk yield, concentrations and yields of milk fat and protein, DM intake (DMI), eating behavior, BW, BCS, and rate of change of BW (kg/100 d) and BCS over the first 100 d of lactation for cows offered either WB, WBCM, WBCR, or WBCMCR diets
Dry Matter Intake and Nutritive Characteristics of Grain Mixes and Pasture
Cows fed grain mixes that contained canola meal had greater (P < 0.001) total DMI (grain mix DMI plus pasture DMI) than cows consuming grain mixes without canola meal (Table 3). This was due to greater intakes of both grain (P < 0.001) and pasture (P < 0.001). Cows offered canola meal consumed 0.6 kg DM more grain and 2.1 kg DM more pasture each d than cows fed the WB mix. The inclusion of corn grain in the mix did not affect the total DMI, nor did it affect the individual DMI of grain (P = 0.501) or pasture (P = 0.887).
The cows offered canola meal generally consumed about 1 kg extra CP, ∼1.4 kg extra aNDF, ∼1 kg less starch, at least 24 MJ extra ME and 0.2 kg extra total fatty acids each day than cows who did not receive canola meal (Table 2).
Milk Yield and Composition
Mean yields of milk and milk composition for cows on the 4 dietary treatments are presented in Table 3. Observed milk production over both the treatment and carryover periods are presented in Figure 1. Feeding grain mixes that included canola meal resulted in greater (P < 0.001) milk and milk protein yields, compared with mixes without canola meal. Milk fat concentrations were also greater (P < 0.001) when canola meal was included. There was an interaction effect between canola meal and corn grain on milk fat yield, where milk fat yields increased (P = 0.035) with the inclusion of both canola meal and corn, but the increase was greater when canola meal was added alone. There was also an interaction between canola meal and corn grain for milk protein concentration, whereby the inclusion of canola meal and corn grain resulted in an increase (P = 0.015) in milk protein concentration. However, the increase in protein concentration was the highest when canola meal was present alone in the mix. Including corn alone in the grain mixes did not influence milk yield or composition (P > 0.05). Milk yield, fat yield, protein yield, and fat and protein concentrations were the lowest in cows on the WB diet.
Figure 1Observed milk production over both the experimental period (15–100 DIM) and the carryover period (101–200 DIM) for cows offered 1 of 4 supplementary mixes during the experimental period. WB: disc milled wheat and barley grain mix; WBCM: disc milled wheat, barley grain and canola meal mix; WBCR: disc milled wheat, barley grain and corn grain mix; or WBCMCR: disc milled wheat, barley grain, canola meal and corn grain mix. Error bars indicate the SEM on the relative d.
In the carryover period from 101 to 200 DIM, there was no difference in mean milk production between any of the early lactation treatment groups (Table 4). However, there was some evidence that fat and protein concentration was higher (P = 0.066) for cows that had consumed the grain mixes containing either canola meal or corn grain during the treatment period compared with cows on the WB diet or where both canola meal and corn grain had been included in the grain mix (WBCMCR).
Table 4Mean daily milk yield, and concentrations of milk fat and protein, for cows offered either WB
The inclusion of canola meal in the grain mix resulted in greater (P = 0.025) mean BW and mean BCS measured over the whole experimental period, as well as a greater (P < 0.001) rate of BW change (Table 3) compared with cows that consumed a diet that did not contain canola meal. The inclusion of corn grain did not affect BW or BCS. There was no difference in BCS change per 100 d between the dietary treatments.
Feeding Behavior
Cows that consumed the grain mixes containing canola meal spent less (P = 0.002) time ruminating and more (P = 0.050) time idling than cows that consumed grain mixes that did not contain canola meal. The amount of time cows spent eating was not affected by the inclusion of canola meal in the grain mixes (P = 0.396). The inclusion of corn grain did not influence any feeding behavior variables.
Blood Serum Metabolites
The measurements of blood serum metabolites taken every 2 weeks can be seen in Figure 2. Mean blood serum values at each of the sampling points, for each treatment and P-values associated with treatment, time and their interactions are provided in Appendix Table A1. There was an interaction between the effects of time and canola meal (P = 0.030) on serum BHB. The initial blood sample (15–21 DIM) showed similar BHB levels between cows fed canola meal and those not fed canola meal (0.58 and 0.61 mmol/L, respectively) and both showed a decline in BHB over the experimental period. However, cows fed canola meal had higher serum BHB levels from the second blood sample onwards, with the final blood samples showing 0.48 mmol/L for cows fed canola meal and 0.36 mmol/L for cows not fed canola meal. Nonesterified fatty acids concentration was not affected by grain mix but did change over time (P < 0.001). There was a significant decrease in serum NEFA between sample 1 and sample 3 (0.61 to 0.24 mmol/L) after which NEFA concentrations plateaued. Protein levels in the serum were not affected by treatment but were affected by time (P < 0.001). There was an increase in serum protein between sample 1 and sample 2 (73.9 to 76.5 g/L) after which the concentrations plateaued. There was no difference between serum urea levels for cows eating grain mixes with or without canola meal at the first blood sampling. However, an interaction between time and canola meal (P < 0.001) resulted in serum urea levels being lower for cows not consuming canola meal at all subsequent sampling points. There were no significant effects of treatment or time on serum albumin. Globulin was only affected by time (P < 0.001), increasing from the first (39.8 g/L) to the third sample (43.0 g/L) and then plateauing. The albumin:globulin ratio was not affected by treatment but was affected by time (P < 0.001). The ratio decreased between sample 1 and sample 2 but then plateaued. There was an effect (P < 0.001) of time on serum Ca. There were no differences in the first 3 blood samples collected (2.37 mmol/L), but by the fourth blood sample at 36–42 DIM serum Ca had significantly increased to 2.45 mmol/L and remained at equivalent concentrations for the subsequent samples. There was an effect (P = 0.022) of corn on serum Ca, with corn fed cows having higher concentrations (2.43 vs 2.39 mmol/L). There was also an effect of corn on serum Mg; cows that did not receive corn in their grain mix had higher concentrations (0.96 vs 0.92 mmol/L). Serum Mg was also affected by time (P < 0.001). There was a decrease in Mg serum concentrations following the initial sample (0.97 mmol/L) until sample 3 when they reached a nadir (0.90 mmol/L), the concentrations then increased over the final 3 samples reaching a peak at sample 6 (0.97 mmol/L).
Figure 2Mean (±SE) bl[INSERT Figure 001]ood serum measurements throughout the treatment period for cows offered 1 of 4 supplementary mixes. WB: disc milled wheat and barley grain mix; WBCM: disc milled wheat, barley grain and canola meal mix; WBCR: disc milled wheat, barley grain and corn grain mix; or WBCMCR: disc milled wheat, barley grain, canola meal and corn grain mix.
Including canola meal in a grain mix fed to grazing dairy cows twice daily during the first 100 d of their lactation resulted in greater mean milk yields compared with feeding isoenergetic grain mixes that did not contain canola meal. Thus, the first hypothesis is accepted. Feeding the mixes containing canola meal also resulted in greater yields of milk fat and protein, and concentrations of milk fat. A likely explanation is that removing some of the barley and replacing it with canola meal resulted in the ruminal fluid pH remaining in a range that optimized digestion. This is due to the fermentation of cereal grains resulting in a sharp reduction in rumen pH, which can be associated with a reduction in milk fat concentration (
). While the grain mixes included a buffer and an alkalizer, which would have acted to reduce the depression in pH, cows supplemented with cereal grains while grazing high quality pasture suffer extreme drops in rumen pH. Cows consuming spring perennial ryegrass have exhibited minimum rumen pH levels as low as 5.55 on pasture alone and 5.06 when concentrates are added (
Ruminal pH and whole-tract digestibility in dairy cows consuming fresh cut herbage plus concentrates and conserved forage fed either separately or as a partial mixed ration.
The effect of calcareous marine algae, with or without marine magnesium oxide, and sodium bicarbonate on rumen pH and milk production in mid-lactation dairy cows.
). Canola meal may also have added benefits for rumen pH as research has shown a greater buffering capacity associated with the provision of high-protein feeds (
reported an increase in ECM production with a similar diet and attributed it to a greater milk fat concentration, which is consistent with our results. The treatment response observed in this experiment would have been influenced and or confounded by differences in nutrient composition of our treatment diets. The major nutritional differences were in the protein and starch content of canola meal and the starch type in corn grain, relative to the WB control treatment.
Removing some of the barley, a rapidly fermentable starch source, and replacing it with corn and or canola meal successfully increased intake. This was presumably due to the increase in total intake, driven by fewer refusals and increased intakes of both the grain mix as well as grazed pasture. This accords with previous research that demonstrated an increase in DMI when wheat grain was similarly substituted with canola meal in partial mixed ration diets (
Effects of partial mixed rations and supplement amounts on milk production and composition, ruminal fermentation, bacterial communities, and ruminal acidosis.
). A contributing factor to this increase in DMI in cows fed canola meal may have been that propionic acid is produced primarily from the fermentation of starch in the rumen and has been shown to reduce appetite in cattle (
); hence removing starch from the diet may increase the inclination of cows to eat. It is not fully understood as to why feeding canola meal improves milk production but
suggested it may be due to an increased and more adequate supply of amino acids. It is then the increase to milk production that drives the cows to eat more (‘pull effect' (
Variation in feeding behavior and milk production among dairy cows when supplemented with 2 amounts of mixed ration in combination with 2 amounts of pasture.
reported that when cows consumed a mixed ration containing canola meal, they spent more time grazing each d, compared with cows that consumed a ration without canola meal. The current research did not detect an increase in eating time when canola meal was included in the grain mixes. However, the provision of canola meal in the grain mixes resulted in a reduction in time spent ruminating. This may be partially due to the canola meal providing a more stable rumen environment, owing to a greater buffering capacity associated with the provision of high-protein feeds (
This experiment was conducted in early lactation, a period when the typical mobilization of body reserves reflects a period where energy is portioned for milk production at the highest priority (
). This research showed that including canola meal in grain mixes resulted in a greater mean BW and BCS, in addition to an increased rate of BW gain over the 100 d. This increased mean BCS demonstrates that the provision of grain mixes that contained canola meal may have helped to alleviate excessive body condition loss in early lactation. During early lactation, BCS can be used as an indicator of energy balance, as loss of BCS is correlated with fat mobilization (
). However, the inclusion of canola meal in grain mixes also resulted in greater serum BHB concentrations from approximately 22 DIM onwards. Increased BHB concentrations are typically indicative of increased mobilization of tissue reserves (
), although this was not seen in the current research. We speculate that one possible explanation for the difference in BHB concentrations between treatments that included or did not include canola meal, is the differences in the starch content of the grain mixes. Cows consuming grain mixes containing canola meal had around 1kg DM/cow per day lower starch intake when compared with cows not consuming canola meal, and both grain mixes were slug fed in the parlor at milking times. Previous research has demonstrated that the fermentation of starch in the rumen may result in elevated propionic acid, which can reduce plasma BHB concentrations (
). The provision of canola meal in the grain mixes also resulted in increased blood serum urea and albumin concentrations from 22 DIM onwards, this is likely a result of the increased intake of CP (
reported that increased dietary CP significantly increased plasma urea, albumin, and total protein concentrations and had no significant effect on NEFA. The current experiment also found no differences in NEFA with the addition of canola meal.
Feeding a grain mix that included corn grain during the first 100 d of lactation did not increase milk yield or alter any milk production variables, nor did it provide any interactive effects when added to a mix with canola meal. Therefore, both the second and third hypotheses are rejected. However, it is important to note that these results may differ when corn is fed in different amounts or as part of a TMR as opposed to a supplement for grazing dairy cows. The current research also demonstrated no differences in the feeding behaviors, BW or BCS when cows consumed grain mixes containing corn compared with those that did not. Corn grain is often included in concentrate diets fed to lactating dairy cows, because of its high energy density, starch content, and slower rate of fermentation when compared with other cereal grains (
). For these reasons, the inclusion of corn was investigated in the current research, to determine if it could enhance milk production, in early lactation. Similar to the current results,
reported that when barley-based diets are appropriately formulated, they can result in similar milk production benefits to corn-based diets, when the protein and starch contents are considered.
suggested that the substitution of barley grain with corn grain may alter the site and the end products of digestion. Between 80 and 90% of barley starch and wheat starch is digested in the rumen; whereas the value for corn ranges from 55 to 70% (
). Because starch found in barley degrades faster than the starch in corn, a greater amount of corn starch will reach the small intestine when compared with starch from barley and wheat. Starch digestion in the small intestine is thought to be more energetically efficient than starch digestion in the rumen, and this efficiency of starch fermentation is influenced by the rumen environment (
). Cows consuming barley and wheat-based diets may compensate for inefficiencies in starch digestion when compared with a corn-based diet, if additional nitrogen is supplied to the rumen (
). These findings support the research reported here, which demonstrate that substituting corn grain, WBCR, with canola meal, WBCM, may have supplied additional nitrogen to the rumen, allowing for more efficient starch digestion.
There was no carryover effect of treatment on milk yield. However, there was some indication that milk fat and protein concentrations were higher for cows fed the treatment diets that had included canola meal or corn grain.
reviewed 9 experiments that measured immediate and carryover effects to milk production from early lactation feeding strategies. The experiments varied in pasture allowance, type and level of concentrate feeding, length of time the feeding strategy was applied, and in the measured immediate milk response.
found that the carryover effect comprised 22 to 63% of the immediate effect. However, in 2 of the studies, there was no carryover effect and the authors postulated that this was likely due to the longer carryover periods in these experiments; 24 and 27 weeks versus 3 to 12 weeks where carryover effects were measured. In other words, the carryover effects of treatments are diluted over time. The carryover period in the present experiment comprised 28 weeks and this long-time interval could be a reason no carryover effect of the treatment diets was observed.
This research demonstrated that the grain mixes that contain canola meal (WBCM and WBCMCR) had milk production advantages when fed during the first 100 d of lactation, when compared with isoenergetic WB and WBCR grain mixes. However, the economic viability of including any supplement in the diet is also an important consideration for individual farm businesses as an increase in milk yield does not necessarily equate to an increase in profit if the income from additional milk produced does not exceed the extra costs. Future research to investigate the relative profitability of feeding grain mixes containing canola meal and or corn grain would be valuable.
CONCLUSIONS
This research showed an advantage of offering grazing dairy cows a supplement containing canola meal in early lactation, and no advantage of offering a supplement containing corn grain, when compared with a supplement containing neither corn grain nor canola meal. The treatment response observed in this experiment were influenced by differences in nutrient composition of our treatment diets, specifically the protein and starch content of canola meal and the starch type in corn grain, relative to the control diet that did not include canola meal or corn grain. The data suggested that cows fed canola meal did not mobilize peripheral tissue to support milk yield to the same degree as cows not fed canola meal, and the data indicate that the increase in milk yield was supported by extra pasture intake. Given the high cost of supplements such as canola meal and corn, particularly in Australia, this research may have implications for diet formulation for early lactation diet formulation beyond nutrient intake that should be taken into account when developing rations for dairy cows in early lactation.
ACKNOWLEDGMENTS
The authors are grateful to D. Wilson, D. Stayches, A. McDonald, L. Dorling, T. Hookey, D. Mapleson, B. Ribaux, K. Rabl, L. Burns, M. Summerfield, T. Summers, K. Hoffman, S. Bourchier, J. Garner, R. Williams, C. Lewis, T. Luke and M. Jenkin (all of Agriculture Victoria), and Agriculture Victoria (Ellinbank, Victoria, Australia) farm staff for cow feeding and husbandry. This research was funded by Agriculture Victoria, Dairy Australia and Gardiner Foundation.
APPENDIX
Table A1Mean blood serum values at six measurement times (weeks in milk) for each treatment and P-values associated with treatment, time and their interactions
Protocol for the analysis of n-alkanes and other plant-wax compounds and for their use as markers for quantifying the nutrient supply of large mammalian herbivores.
Effects of partial mixed rations and supplement amounts on milk production and composition, ruminal fermentation, bacterial communities, and ruminal acidosis.
Ruminal pH and whole-tract digestibility in dairy cows consuming fresh cut herbage plus concentrates and conserved forage fed either separately or as a partial mixed ration.
The effect of herbage allowance and concentrate supplementation on milk production performance and dry matter intake of spring-calving dairy cows in early lactation.
The effect of calcareous marine algae, with or without marine magnesium oxide, and sodium bicarbonate on rumen pH and milk production in mid-lactation dairy cows.
Variation in feeding behavior and milk production among dairy cows when supplemented with 2 amounts of mixed ration in combination with 2 amounts of pasture.
00071-7&id=gr1.jpg){kind=link}
00071-7&id=gr2.jpg){kind=link}