Abstract: This study was done to evaluate the effects of restricting Total Mixed Ration (TMR) and supplying Eragrostis curvula and Medicago sativa hay mixture adlib on lactation performance of mid-lactation multiparous Holstein cows. Twenty Holstein cows, averaging 598±73 kg body weight and 100 days in milk were assigned to either a 100% TMR diet (control) or a 75% TMR-restricted diet. Cows on the 75% TMR-restricted diet had adlib access to E. curvula and M. sativa hay mixture (1:1). The experiment included 2 weeks adaptation period and 4 weeks sampling period. Cows were milked twice daily. Total Dry Matter Intake (DMI), DMI as BW%, daily CP intake and intake of net energy for lactation were higher (p<0.05) for cows on the 100% TMR than for cows on the 75% TMR. Cows in 75% TMR consumed 12.5% less total DM and 14.2% less kg DM as of BW% than cows in 100% TMR. Intake of NDF was not affected (p>0.05) by treatment. Milk yield, milk fat% and yield (kg day^-1) did not differ between treatments and averaged 29.2 kg day^-1, 3.70% and 1.08 kg, respectively. Feed efficiency ranged from 1.22-1.37 and tended to be higher (p<0.10) with 75% TMR diet. Efficiency of nitrogen and phosphorous utilisation was not affected (p>0.05) by treatments. Results suggest that TMR restriction to 75% during mid-lactation does not negatively impact milk production.

INTRODUCTION

Confinement feeding systems are widely used on commercial dairy farms where rations are offered as Total Mixed Rations (TMR) consisting of combinations of concentrates (protein, energy and mineral-vitamin premixes) and forages. Economic condition caused by low milk prices and narrow profit margins challenge dairy farmers to find ways to increase farm profitability in order to stay competitive (Muller and Fales, 1998). In addition, environmentally-friendly practices have resulted in an increasing interest in pasture-based dairy farming.

Pasture-based dairying can be competitive in terms of profitability compared with confinement systems (White et al., 2002) but supplemental energy needs to be fed to high producing dairy cows on pasture in order to meet genetic potential for lactation performance (Vibart et al., 2008). It was reported that the lower milk production from high producing dairy cows consuming only high quality pasture compared with cows consuming a nutritionally balanced TMR was due to a lower DMI and energy intake (Kolver and Muller, 1998).

Some management challenges when pasture is the only forage include low milk production per cow, low milk fat and protein content, variations in production because of climate conditions, difficulty in budgeting pasture availability and inaccurate estimation of total and pasture DMI (Muller and Fales, 1998). Supplementation of pasture with a TMR may reduce these challenges (Bargo et al., 2002). Some of the potential advantages of feeding a partial TMR include the provision of a more uniform ration throughout the year and less chance of rumen digestive problems (Bargo et al., 2002).

A number of other studies have also compared the performance of dairy cows on TMR’s vs. TMR plus pasture (Soriano et al., 2001; Bargo et al., 2002). Research data is available where TMR is increasingly supplemented with pasture or where TMR is increasingly replaced by high-quality pasture grazing (Washburn et al., 2002; White et al., 2002; Bargo et al., 2002; Fontaneli et al., 2005). Results of the study of Vibart et al. (2008) showed that the TMR could be replaced by pasture up to the greatest amount of pasture dry matter intake tested (41%). As far as can be ascertained, no study thus far has compared TMR vs. restricted TMR plus E. curvula and M. sativa forage hay offered in confinement. E. curvula hay is available abundantly at low price in South Africa. Combination of E. curvula hay with a high nutritional legume such as of M. sativa result in a relatively good quality hay mixture which can be used when pasture is not available and especially on confinement. Researchers hypothesized that cows which receive a correctly formulated but restricted amount of TMR with additional E. curvula and M. sativa hay offered ad libitum in mid-lactation would perform as well as cows receiving a TMR ad libitum at this stage. A positive relationship exists between dietary protein content and DMI (Schor and Gagliostro, 2001; Bargo et al., 2003) which affect nitrogen excretion and milk production (Mulligan et al., 2004). Thus when investigating the effect of reducing TMR, it is important to also investigate the impact on nitrogen excreted. From an environmental point of view, it is advisable to decrease nitrogen and phosphorous excretion per milk unit produced. The objective of this study therefore, was to evaluate the impact of partial replacement of TMR by E. curvula and M. sativa hay mixture on milk production, milk composition, nitrogen and phosphorous utilization by Holstein cows during mid-lactation.

MATERIALS AND METHODS

The study was conducted at the Animal Production Institute, Irene of the Agricultural Research Council of South Africa. Latitude 25.89907 South and longitude 28.20063 East at 1475 m altitude. Average annual minimum temperature is 14°C with mean annual maximum temperature of 24°C.

Twenty multiparous Holstein cows in 2nd and 3rd lactation, averaging 598±73 kg Body Weight (BW) and 100 Days in Milk (DIM) were used in a completely randomized block design. They were assigned to one of ten blocks of two cows based upon parity, BW and previous milk production. The treatments were either a 100% TMR diet or a 75% TMR plus hay diet (restricted TMR). A TMR diet (Table 1) was formulated to meet nutritional requirement of a 600 kg cow at 120 DIM and yielding 35 kg milk day^-1 with 3.7% fat and 3.2 CP. The TMR was fed ad libitum (100% TMR) or restricted to 75% of TMR. Cows on the restricted diet received TMR at 12:00 h and had ad libitum access to chopped grass (Eragrostis curvular) legume (Medicago sativa) hay (1:1) (Table 2). The same chopped forages were used in preparing the TMR. The 75% TMR restriction was calculated from the average of the 2 weeks TMR consumption during the adaptation period.

The cows were individually housed and fed to monitor individual feed intake. Fresh water was continuously available. The cows were exposed to continuous lighting and were milked at 06:00 and 17:00 h daily in a 10 point DeLaval herring bone parlour equipped with an Alpro Herd management system (DeLaval. (Pty.) Ltd., Heilbron, 9650, South Africa).

Table 1:	Ingredient and chemical composition of the TMR diet (DM basis)

Table 2:	Chemical composition of the forage mixture
aCalculated as Metabolizable Energy (ME) using the database of Van der Merwe and Smith (1991) and converted to NEL as: ME X 0.67 (NRC, 2001)

Animal care was consistent with the guide for the care and use of animals in agriculture research and teaching and animal use was approved by the animal ethics Committee of the Agricultural Research Council of South Africa.

Measurements and analyses: Animals were adapted to their diets for 2 weeks and measurements and samples collected over 4 weeks. Cows were weighed and scored using a five point BCS scale; very thin and very fat (Wildman et al., 1982) at inception and termination of the experiment.

Individual milk yield was recorded daily. Weekly samples of TMR, hay mixture and orts, within cow were collected and pooled by treatment and frozen pending chemical analyses.

Dry matter was determined by oven drying at 60°C for 48 h. Dried TMR were ground and analysed for Organic Matter (OM) by ashing in a muffle furnace at 600°C for 6 h. Crude Protein (CP) was determined according to AOAC (2000) method 968.06 and ether extract according to AOAC (2000) procedure 920.39. Calcium (Ca), potassium (K) and Magnesium (Mg) were determined according to Giron (1973) using a Perkin elmer atomic spectrophotometer. Phosphorus (P) was assayed according to AOAC (2000) procedure 965.17. Neutral Detergent Fibre (NDF) was determined according to Van Soest et al. (1991) and Acid Detergent Fibre (ADF) according to Goering and van Soest (1970). Non-Fibre Carbohydrate (NFC) was calculated from the assayed nutrients (Hall, 1998). Dried refusals were only analysed for CP and NDF. The DM, CP and NDF of the forage mixture were determined with the above mentioned methods.

A milk analysis was done at Lacto Lab (Pty.) Ltd., Irene using infrared analyser (4000 Foss Electric, Hillerod, Denmark). Composite milk samples were prepared from consecutive morning and afternoon milking once weekly and analyzed for fat, CP, lactose and Milk Urea Nitrogen (MUN). Faecal grab samples were collected from all cows at approximately 11:00 on day 26 or 27th and dried at 60°C in a forced-air oven for 96 h. The faecal samples were ground to pass a 1 mm Wiley mill screen and analyzed for DM, N (Kjeldahl method) (AOAC, 2000) and Phosphorus (P) was assayed according to AOAC (2000) procedure 965.17.

Urine samples were collected from cows by mechanical stimulation of the vulva, immediately after collecting fecal grab samples. A sub-sample of 15 mL was preserved in 60 mL 0.072 N H₂SO₄ and stored at -20°C pending analysis (Broderick et al., 2008) for N and P. Total Manure excretion (Me) was determined using equations of Nennich et al. (2005) and urine excretion as described by Nennich et al. (2006), respectively based on DMI and MUN:

(1)

(2)

Statistical analysis: Analysis of a Completely Randomized block Design (CRD) with repeated measurements was used to test for differences between treatments. The data was normally distributed with homogeneous treatment variances. Treatment means were separated using Fisher’s protected t-test and mean differences considered significant at p<0.05 and tendencies were noted at p≤0.10.

RESULTS AND DISCUSSION

Intake, milk yield, milk composition and efficiency of feed conversion to milk: Dry matter intake and milk production are shown in Table 3 and 4, respectively. Total DMI was higher (p = 0.008) for cows on the 100% TMR than for cows on the 75% TMR as reported by Bargo et al. (2002) and Vibart et al. (2008) when they compared 100% TMR to restricted TMR plus pasture. Cows on the restricted diet consumed all partial TMR plus 61% of the forage mixture.

Table 3:	Dry matter, CP, NDF and NEL intake of cows on the 100 and 75% TMR plus forage mixture
abMeans in the same row with different superscripts differ (p≤0.05)

Table 4:	Milk yield, milk composition and efficiency of feed conversion to milk of cows on the 100 and 75% TMR plus forage mixture
abMeans in the same row with different superscripts differ (p≤0.05); 1Energy Corrected Milk: ECM = ((0.327xkg milk)+(12.95xkg fat)+(7.2xkg protein)) (Orth, 1992); 2Feed Efficiency: FE = ECM/DMI

Intake of the forage mixture by the later group constituted 0.64% of BW and hence, higher (p = 0.002) total forage intake as of BW%. Cows in 75% TMR consumed 12.5% less total DM and 14.2% less kg DM as percentage of BW than cows in 100% TMR. Intake of NDF was not affected by treatment (p>0.05). Intake of ME was higher for cows on the full TMR diet resulting in more (p<0.05) NE_L compared to the restricted diet. Restricting TMR also reduced intake of crude protein (3.7 vs. 3.1). Milk yield (Table 4) did not differ (p>0.05) between treatments and averaged 29.2 kg day^-1; also milk fat concentration and yield were similar (p>0.05) averaging 3.70 and 1.08% kg day^-1, respectively. Thus, despite the higher DMI and NE_L supply in the 100% TMR treatment, milk yield remained unaffe cted. Milk protein concentration and yield, respectively were lower (p≤0.05) and tended to be lower (p = 0.10) for the 75% TMR (Table 4) diet.

The differences in milk protein were attributed to differences in daily protein intake. Reduced total protein intake lowers metabolizable protein available for milk protein synthesis (NRC, 2001). Milk lactose was similar in both treatments and averaged 4.47%. Lactose concentration in milk usually is not affected or is only slightly altered by dietary manipulation (Casper et al., 1990; Kim et al., 1991) which is confirmed in the present study. Milk urea nitrogen was lower (p = 0.001) in the 75% TMR group.

Reduced MUN is typically found with decreased dietary CP and/or decreased by pass protein. Factors that affect MUN include dietary CP (%), dietary CP intake, proportion of RDP and RUP and protein/energy ratio in the diet (Nousiainen et al., 2004; Roseler et al., 1993; Baker et al., 1995).

Energy-corrected milk, however was similar (p = 0.11) between treatments. Feed Efficiency (FE) ranged from 1.22-1.37 and tended (p = 0.10) to be higher in 75% TMR when compared to 100% TMR. Vibart et al. (2008) also reported a similar trend when TMR was partially substituted with pasture. Feed conversion efficiencies were 1.54, 1.43, 1.46 and 1.27 with 55, 70, 80 and 100% TMR, respectively in the 1st study and 1.62, 1.54, 1.65 and 1.33 in the 2nd study, respectively. Controversially, Bargo et al. (2002) reported entire-lactation gross feed efficiencies (3.5% FCM) of 1.25, 1.23 and 1.37 for lactating cows fed pasture plus concentrate, pasture plus TMR (approximately, 70% TMR and 30% pasture) and TMR, respectively.

Dry matter intake and milk yield were also compared within weeks and results are shown in Table 5 and 6. During week 1, no differences in total daily DMI were observed but DMI tended (p<0.1) to be higher in the 100% TMR while it was consistently higher (p<0.05) in the 100% TMR group from week 2 until the end of the experimental period. Weekly milk yield was not affected (p>0.05) by treatments from week 1 until week 4 (Table 5).

Body Weight (BW) and Body Condition Score (BCS): Body Weight (BW) and Body Condition Score (BCS) changes are shown in Table 7. Body weight change and body weight change as of BW% did not vary; averaging -1.5 kg and -0.25%, respectively suggesting that restricting TMR to 75% had no significant impact on body weight. Contrary, Vibart et al. (2008) reported that cows on full TMR lost 15 kg of BW and cows receiving the 55, 70 and 85% TMR plus pasture gained, respectively 19, 18 and 22 kg. Cows on the full TMR, however produced more milk.

Table 5:	Total dry matter intake during weeks 1-4 of cows on100 and 75% TMR plus forage mixture

Table 6:	Milk yield during weeks 1-4 of cows on 100 and 75% TMR plus forage mixture

Table 7:	Body weight and body condition score of cows on the 100 and 75% TMR
a, bMeans in the same row with different superscripts differ (p≤0.05)

The 100% TMR diet was formulated to support 35 kg day^-1 but cows produced 16% less milk than predicted (29.5 kg day^-1). Energy from BW changes (Gain or loss) was reported (Table 7) as either extra energy required or added to that provided by diet, depending on BW gain or loss throughout the study, respectively (daily BW gain = 5.10 Mcal kg^-1 BW whereas, average daily BW loss = 4.02 Mcal kg^-1 BW added to that provided by dietary intake) (NRC, 2001). Results indicate that while cows in 100% TMR had 21.4 Mcal kg^-1 (0.76 day^-1) surplus energy to replenish body reserves, cows in 75% TMR have mobilised a total of 26.5 Mcal kg^-1 (0.94 day^-1) from reserve to support milk production. The difference in DMI appeared to be the major factor responsible for a lower energy supply in the 75% TMR group.

Nitrogen and phosphorous excretion: Nitrogen (N) and Phosphorous (P) excretion data are shown in Table 8. Effect of restricting TMR on N and P utilization was evaluated with reference to total dietary intake. Intake of both N and P were lower (p = 0.001) in 75% TMR. Faecal DM was higher (p<0.05) with 100% TMR.

Table 8:	Urinary and faecal Nitrogen (N) and Phosphorous (P) excretion of cows on the 100 and 75% TMR plus forage mixture
a, bMeans in the same row with different superscripts differ (p<0.05); Ne: Nitrogen excretion; Pe: Phosphorous excretion; Nint: Nitrogen intake; Pint: Phosphorous intake

Estimated mean excreted urine in 100% TMR (22.1 kg day^-1) and 75% TMR (21.5 kg day^-1) were similar to data reported by Sannes et al. (2002) (21.8-25.7 kg day^-1) and Broderick (2003) (20.8-27.3 kg day^-1) entailing metabolic stability. Nennich et al. (2005) reported a mean urine excretion of 24.1 kg day^-1. Urinary and faecal N did not vary (p>0.05) between treatments but total Ne was higher (p<0.05) in 100% TMR. In the study of Hristov et al. (2004), high Nitrogen intake did not have effects on fecal and urine N, as observed in this study but controversially, total N excretion was not affected. Kebreab et al. (2002) reported a greater and exponential effect of Nitrogen intake on urinary nitrogen with limited effects on faecal N excretion. From an environmental point of view, it is advisable to decrease N excretion per milk unit produced. In this sense, improved milk production decreases the partial contribution of maintenance N requirements (Rotz, 2004) which directly helps to improve the N utilisation efficiency and decrease N excretion per litre of milk (Arriaga et al., 2009).

Intake of P, faecal P and total P were higher (p<0.05) in 100% TMR but no differences in Urinary P were observed. Proportion of N and P excretion relative to N and P intake (Ne/Nint and Pe/Pint) did not differ (p>0.05) between treatments and averaged, respectively 75.6 and 72.8% suggesting that restriction of TMR did not significantly affect the efficiency of N and P utilisation.

CONCLUSION

Partial substitution of TMR with M. sativa and E. curvula hay supplied adlib lowered total DMI, crude protein intake and NE_L. Milk yield and feed efficiencies were similar in both treatments although, daily milk production on 75% TMR was about 1.7 kg less then the full TMR, feed efficiencies were better on the restricted diet. Results suggest that restricting the TMR to 75% does not significantly impact milk production in mid-lactation. However, further studies are necessary to determine the long term effects of restricting energy and N intake during mid-lactation on subsequent lactation and longevity.

ACKNOWLEDGEMENT

This study was supported by the Agricultural Research Council, South Africa which is gratefully acknowledged.