Effect of corn shredlage on feed intake, rumen fermentation, and lactation performance of dairy cows fed a low-fibre diet

Abstract Corn shredlage is a whole-plant corn silage with a greater proportion of long particles and intensively processed stalks (by shredding and peeling) and kernels (by thorough crushing). Corn shredlage may improve rumen function and milk performance of dairy cows, especially in low-fibre diets. Paradoxically, little is known about the specific effects of corn shredlage on rumen fermentation. This study aimed to understand how the dietary replacement of corn silage with corn shredlage changes feed intake, rumen fermentation, rumination time, and milk performance of dairy cows fed a low-fibre diet. Thirty-four lactating cows were allocated to two dietary treatments in a crossover design with two 30-day periods. The dietary treatment was a total mixed ration with either 1) 25% conventionally processed corn silage (CON; 14-mm theoretical length of cut, TLOC) or 2) 25% corn shredlage (SHR; 26-mm TLOC). Cows on the SHR diet had increased DMI and body weight. Although milk yield was unaffected by diet, cows on the SHR diet tended to have increased yields of 4% FCM, ECM, and milk fat. The yields of milk protein and lactose were not affected by diet. Similarly, the proportion of milk components remained unaffected. The feed efficiency (milk/DMI) was higher in cows fed the CON diet, whereas it remained unaffected when expressed as ECM/DMI. Neither rumination nor rumen fermentation parameters were affected by the diet. Overall, the positive effects of corn shredlage on milk performance were relatively small and a result of increased DMI rather than improved rumen fermentation or rumination. HIGHLIGHTS Corn shredlage increased DMI Corn shredlage tended to increase yields of 4% FCM, ECM, and milk fat Corn shredlage did not affect rumination and rumen fermentation


Introduction
High-producing lactating dairy cows encounter enormous energy requirements to support milk production, especially during early lactation.To meet these requirements, a diet usually contains a high proportion of concentrate (up to 60%) and therefore, a relatively low concentration of fibre (Beauchemin et al. 2003).Although this strategy supports high milk production, it can also result in poor ruminal health with possible negative effects on overall health, digestion, and production, including subclinical ruminal acidosis, reduced fibre digestion, and milk fat depression (NRC 2001).Increasing the theoretical length of cut (TLOC) of forage might be a strategy to increase chewing activity and, therefore, increase saliva flow, rumen pH, the proportion of acetate, and milk fat levels (Kononoff et al. 2003;Salvati et al. 2021).
Whole-plant corn silage is worldwide a major source of forage in the dairy cow diet (Khan et al. 2015;Ferraretto et al. 2018).Thus, corn silage is an important source of physically effective neutral detergent fibre (peNDF) in these diets.A sufficient amount of peNDF in the diet is required to maintain proper rumen health and functioning (Yang and Beauchemin 2006), as peNDF stimulates chewing, saliva production, and rumen buffering (Zebeli et al. 2012).The concept of peNDF integrates information on dietary particle size and the chemical NDF content of the diet (Mertens 1997) that act jointly and interdependently to stabilise rumen fermentation (Zebeli et al. 2012).The peNDF can be increased by increasing the TLOC (i.e.particle size), which might be useful, especially in diets with a low proportion of forage and low NDF content (e.g.early lactation diet).However, increasing particle size may reduce the DMI of the diet (Zebeli et al. 2012).In recent years, harvesting whole-plant corn silage as shredlage has gained widespread interest among dairy producers and nutritionists (Vanderwerff et al. 2015).Shredlage is produced using a special forage harvester fitted with cross-grooved crop-processing rolls.The production process involves chopping the plants to greater lengths than usual (up to 30 mm), longitudinal stem shredding, and corn kernel breakage (Ferraretto and Shaver 2012;Vanderwerff et al. 2015;Bach et al. 2021).More intensive and complex processing during the production of corn shredlage may reduce the negative effects of long forage particles on the DMI (Ferraretto and Shaver 2012).
Responses to replacing corn silage with corn shredlage in dairy cow diets have been variable but relatively small in general.Ferraretto and Shaver (2012) reported a tendency for higher DMI and 3.5% FCM yield in cows fed TMR with 50% corn shredlage than in cows fed the same amount of conventional corn silage.A higher DMI in the last weeks of the experimental period, but without improvement in milk yield and quality, was reported in a recent study of dairy cows fed a TMR with 32.7% corn shredlage (Bach et al. 2021).This study also reported no effects of corn shredlage on the rumen microbiome (Bach et al. 2021).No effects on DMI, milk yield, and milk quality were reported in dairy cows fed corn shredlage from brown midrib corn hybrid (Vanderwerff et al. 2015).The small effects of corn shredlage in these studies might be due to the relatively high concentration of NDF (well above the minimum recommended by NRC 2001) in the experimental diets.Moreover, limited data are available on the effects of corn shredlage on rumen fermentation.
Therefore, the objective of this study was to understand how the dietary replacement of corn silage with corn shredlage changes the DMI, rumen fermentation, rumination time, and milk performance of dairy cows fed a low-fibre diet.We hypothesised that feeding a diet with corn shredlage would increase rumination time, rumen pH, acetate-to-propionate ratio, milk yield, and milk fat proportion without reducing DMI.

Materials and methods
The experiment was conducted at the experimental farm (Netluky, Prague, Czech Republic) of the Institute of Animal Science.The experimental protocol was approved by the Animal Ethics Committee of the Institute of Animal Science.Cows were handled following the applicable national and European legislation (Directive 2010/63/EU, on the protection of animals used for scientific purposes; European Union, 2010).

Harvesting and processing of whole corn plant
A corn silage hybrid (WALTERINIO KWS, KWS Osiva Ltd., Czech Republic) was planted in the Institute of Animal Science field at a density of 80 000 seeds/ha and a 75-cm row spacing.One half of the field was harvested (at about one-half milk line stage of kernel maturity) as conventional corn silage, and the other half as corn shredlage on September 7, 2017.Corn silage was harvested using a forage harvester (Class Jaguar 850, CLAAS, Harsewinkel, Germany) equipped with conventional processing rollers with a 30% roll speed differential, a 1.6-mm roll gap, and a 14-mm theoretical length of cut (TLOC).The corn shredlage was harvested using a forage harvester (Class Jaguar 950, CLAAS, Harsewinkel, Germany) equipped with a multi-crop cracker Shredlage processor with a 50% roll speed differential, 1.6-mm roll gap, and 26-mm TLOC.The shredlage processor was fitted with two counterrotating spiral-grooved rollers.The plant material passing through the rollers is pulled apart by the sideways movement of teeth.The stalks are shredded effectively in the longitudinal direction, and the bark is peeled off.Moreover, corn cobs are fully broken, and the kernels are crushed and split thoroughly (Jan c ık et al. 2022).Both chopped fresh forage, corn silage (95 000 kg), and corn shredlage (99 000 kg) were treated with a commercial silage inoculant containing Lactobacillus buchneri, Lactobacillus plantarum, and Pediococcus pentosaceus (Bonsilage Mais, Schaumann Ltd., Czech Republic) at a rate of 1 g/ton and ensiled in silage bags until the start of the experiment in dairy cows (approximately 7 months).

Experimental design, animals, and feeding
The experiment was conducted as a crossover design with two consecutive experimental periods.Each experimental period lasted for 30 days and consisted of 20 days of adaptation to the diet (from day 1 to day 20) and 10 days of sampling (from day 21 to day 30).A total of 34 lactating dairy cows (initial average BW of 694 ± 58 kg, 76 ± 29 DIM, 43 ± 7 kg milk yield, and 3 ± 1 parity) of two breeds (30 Holstein and 4 Czech Pied cows) were blocked into pairs according to breed, parity, DIM, and milk yield at the start of the experiment.Within pairs, cows were randomly allocated to one of two groups.In each group, there were 13 multiparous and 2 primiparous Holstein cows and 2 multiparous Czech Pied cows.In the first experimental period, the groups were randomly assigned to one of two experimental diets.In the second period, diets were crossed over between the two groups.The experimental diets were the same total mixed rations (TMR) except that 1) CON, a control diet contained 25% conventionally processed corn silage (on DM basis) and 2) SHR, a shredlage diet contained 25% corn shredlage (on DM basis; Tables 1 and 2).TMR was offered ad libitum.Fresh TMR was prepared and delivered to the barn twice a day at approximately 0400 and 1600 h, and feeding troughs were refilled with a shovel at least five times per day.The cows were housed in a free-stall barn with free access to water and milked twice daily at approximately 0530 and 1630 h.

Sampling and analysis
The individual daily feed intake of cows was automatically recorded using a roughage intake control (RIC) system (Hokofarm Group BV, Marknesse, The Netherlands).Diets were offered using 20 electronic feeding troughs.Each treatment group with 17 cows had access to 10 feeding troughs.The groups were kept separated in two identical sections of one barn.Each cow had an ear tag with a unique radio frequency transponder that allowed access to the transponder-controlled access gates of troughs.The troughs recorded every visit of each cow together with the start and end time of the visit (to the nearest 1 s) and the start and end weight (to the nearest 0.1 kg) of the feed ration in the trough.Orts were removed, and troughs were refilled with fresh TMR during the morning milking (between 05:00 and 06:00) when cows did not have access to the troughs.Dry matter intake data were acquired by correcting the as-fed intake for the DM content of the feed.Body weight was measured twice a day after each milking using an electronic livestock scale (AfiWeigh scale; Afimilk Ltd., Kibbutz Afikim, Israel) placed in a common exit alley of the milking parlour.Milk yield was recorded (AfiMilk MPC Milk Metre, Afimilk Ltd., Kibbutz Afikim, Israel) daily at the cow level at each milking and daily milk yield was calculated as a sum of the morning and evening milkings.Feed intake, body  e ProMel (a blend of liquid protein concentrate and homogenised beet molasses in a 1:1 ratio; VVS Verm e rovice, s. r. o., Verm e rovice, Czech Republic).f NFC (non-fibre carbohydrates) ¼ 1000 -NDF (g/kg DM)crude protein (g/kg DM)ether extract (g/kg DM)ash (g/kg DM).
weight, and milk yield were recorded throughout the experiment; however, only data collected from the last 10 days of each period were used for statistical analysis.Milk samples were collected from each cow during two consecutive milkings (morning and evening) on days 22 and 29 of each experimental period.The samples from morning and evening milkings were pooled in proportion according to the individual milk yield of each milking and analysed for milk fat, protein, lactose, and urea concentrations by infra-red spectroscopy (Foss FT2, MilkoScan, Foss Electric, Hillerod, Denmark).
Feed samples were collected weekly and composited by period.The composite samples were analysed according to AOAC International (2005) for DM (method 934.01), crude protein (method 954.01), crude fat (method 920.39), and ash (method 942.05).Neutral detergent fibre (NDF) was assayed with a heat-stable amylase and expressed exclusive of residual ash (Mertens 2002), whereas acid detergent fibre (ADF) was determined according to method 973.18 of AOAC International (2000).The starch content was determined polarimetrically using the Ewers method ISO 10520 (ISO 1997).
The pH and concentrations of VFA, lactic acid, and ammonia-N were determined in the rumen fluid to assess the effects of shredlage on rumen fermentation.Rumen fluid was sampled on the last day (day 30) of each experimental period from 16 cows (n ¼ 8) 4 h after morning feeding using the oral stomach tubing technique (Muizelaar et al. 2020).The head of the oral stomach tube fitted with a strainer was inserted to a depth of 180 cm, and the sample ($250 mL) was collected in a 500-mL Erlenmeyer flask using a vacuum pump.The samples were placed on ice and transported to the laboratory, where the rumen fluid pH was measured (pH 700, Eutech Instruments, Singapore).Subsamples were stored at À20 C until analysis of the fermentation products.The VFA (acetic, propionic, butyric, and valeric acids) and lactic acid concentrations were determined using the single column isotachophoresis analyser unit Ionosep 2003 (Recman Laboratory Equipment, Ostrava, Czech Republic), according to Fil ıpek and Dvo r ak (2009).Ammonia-N concentrations were determined using the phenol-hypochlorite method, as described by Weatherburn (1967).Rumen fluid samples (1 ml) were acidified with 20 ml of 9 M H 2 SO 4 before ammonia-N analysis to prevent ammonia volatilisation.
To collect daily rumination data for individual cows, each cow was fitted with a rumination monitoring system (Vitalimetr 5 P, Farmtec a. s., Jistebnice, Czech Republic).The system consisted of a collar 3-axis accelerometer sensor (62 Â 53 Â 35 mm; 200 g) positioned below the neck, a data logger with on-board data analysis, and software for processing electronic data (Farmsoft, Farmtec a. s., Jistebnice, Czech Republic).The on-board data logger identified a specific pattern of rumination using a generic algorithm and summarised data at hourly intervals.Then, 24 one-hour summaries were totalled to form one-day summaries.Consequently, there were 10 one-day summaries in each cow in each period.Before statistical analysis, these one-day summaries were averaged per cow and period.
The particle sizes of the corn silage and corn shredlage samples were determined using the Penn State Particle Separator, as described by Kononoff et al. (2003).The sieve arrangement was as follows: 19-mm sieve on top, 8-mm sieve second, 4-mm sieve third, and plastic pan on the bottom.Approximately 300 g of undried and unground samples were spread out on the top 19-mm sieve, and the separator was manually shaken 40 times in the horizontal direction.The separator was rotated by 90 after every five shakes (one shake ¼ forward and backward motion over approximately 17 cm).Energy-corrected milk (ECM) yield was calculated according to Sjaunja et al. (1991):

Calculations and statistical analysis
Before statistical analysis, DMI, milk yield, body weight, and rumination data (10 days) and milk composition data (2 days) were averaged per cow and period.These averaged values were used in the statistical analysis.Data were analysed with the statistical software package SAS (SAS Enterprise Guide version 6.1, SAS Institute Inc., Cary, USA) using PROC MIXED according to the following model: where Y ijkl is the dependent variable, m is the overall mean, G i is the group effect, C(G) ij is the effect of the cow within the group, P k is the period effect, T l is the treatment effect, and e ijkl is the residual error.Group and cow within the group were random effects, whereas all others were fixed.The results are reported as least-squares means.Statistical differences were considered significant when p < 0.05.Trends are discussed at 0.05 p < 0.10.

Results
The chemical composition and fermentation profiles of corn silage and corn shredlage were similar (Table 1).Consequently, the chemical compositions of the CON and SHR diets were also similar (Table 2).Compared with corn silage, corn shredlage had a greater percentage of particles retained on the top screen (19 mm) and a lower percentage of particles retained on the second screen (8 mm).The percentage of particles retained on the third screen (4 mm) and bottom pan was similar between the silages.DMI was higher (p < 0.002) in cows fed the SHR diet than in those fed the CON diet (Table 3).Similarly, cows fed the SHR diet had higher (p < 0.004) body weights than those fed the CON diet.Rumination time in minutes per day (p < 0.276) and minutes per kilogram of DMI (p < 0.805) did not differ between diets.
Milk yield was unaffected (p ¼ 0.181) by diet.The yield of 4% FCM (p ¼ 0.060) and ECM (p ¼ 0.055) tended to be higher in cows fed the SHR diet.Statistically, the proportions of milk fat (p ¼ 0.139), protein (p ¼ 0.339), and lactose (p ¼ 0.664) were unaffected by the diet.Similarly, the daily production of protein (p ¼ 0.588) and lactose (p ¼ 0.129) did not differ between the diets.In contrast, daily milk fat production tended to increase (p ¼ 0.073) in cows fed the SHR diet.The concentration of MUN was unaffected (p ¼ 0.336) by the diet.
Feed efficiency, expressed as milk/DMI, was higher (p ¼ 0.024) in cows fed the CON diet, whereas when expressed as ECM/DMI, feed efficiency remained unaffected (p ¼ 0.871).
Rumen fermentation parameters suggest that neither the extent nor the type of rumen fermentation is altered by diet.Namely, the total concentrations (p ¼ 0.882) and individual VFA (p > 0.05) in the rumen fluid were unaffected by diet.Lactate concentrations did not differ (p ¼ 0.709) between the diets.Similarly, the ratio of acetate:propionate (p ¼ 0.366), rumen pH (p ¼ 0.607), and ammonia-N (p ¼ 0.147) in the rumen fluid remained unaffected by diet.

Discussion
Replacement of conventionally processed corn silage with corn shredlage in a low-fibre diet of dairy cows increased the DMI and tended to increase the yield of 4% FCM, ECM, and milk fat, which is consistent with our hypothesis.Conversely, our data provide little support for the hypothesis that corn shredlage alters rumen fermentation and rumination time.

Dry matter intake, milk production, and milk composition
In the present study, cows fed the SHR diet consumed 0.6 kg/d more DM than cows fed the CON diet.A similar result was reported by Ferraretto and Shaver (2012), who found that replacing corn silage with corn shredlage increased DMI by 0.7 kg/d.Dry matter intake was also increased by the dietary inclusion of corn shredlage in dairy cows during the last 3 weeks of 7 weeks experimental period (Bach et al. 2021).In contrast, brown midrib corn shredlage had no effect on the DMI of dairy cows (Vanderwerff et al. 2015).Although digestibility was not measured, it could be speculated that the higher DMI for cows fed the SHR diet may be due to improved nutrient digestibility resulting from greater processing intensity during corn shredlage production using a shredlage processor.Theoretically, the shredlage processing mechanism  (Sjaunja et al. 1991).d Milk urea nitrogen.DMI: dry matter intake.
increases the surface area of the chopped plant material, leading to a greater surface area upon which microorganisms can attach.A greater surface area may improve bacterial fermentation and increase nutrient digestion (Cooke and Bernard 2005).However, our results do not support the positive effect of corn shredlage on bacterial fermentation, as rumen fermentation parameters remained unaffected by SHR.Digestibility was not measured in the present study.Nonetheless, previous results indicate that the inclusion of corn shredlage increases the digestibility of starch but has no effect on the digestibility of fibre (NDF) in dairy cows (Ferraretto and Shaver 2012;Vanderwerff et al. 2015).The higher DMI for cows fed the SHR diet may also be the result of an increased passage rate of digesta through the gastrointestinal (GI) tract.Longer forage particles in the SHR diet may stimulate the ruminal wall, resulting in greater ruminal motility and an increased passage rate of digesta.Krause et al. (2002) previously reported that increasing forage particle size increased the liquid outflow rate from the rumen and decreased the transit and retention time of solids in the GI of dairy cows.
Although milk yield was numerically higher (þ0.5 kg/d) in cows fed SHR, the difference was statistically insignificant (Table 3).Cows fed SHR tended to increase the yield of 4% FCM (p ¼ 0.060), ECM (p ¼ 0.055), and milk fat (p ¼ 0.073), which resulted from the numerically higher milk yield and proportion of fat in milk.The improved yield of 4% FCM, ECM, and milk fat seems to be rather a result of increased nutrient intake with increased DMI (Table 3) or, more speculatively, nutrient digestibility (Vanderwerff et al. 2015) than the beneficial effects of shredlage on rumen fermentation (Table 4).The tendency to improve component-corrected milk yields and numerically higher milk yields in our study agrees with that of Ferraretto and Shaver (2012).In contrast, Vanderwerff et al. (2015) and Bach et al. (2021) reported no effect of shredlage on milk yield or composition.This discrepancy is likely related to the different effects of shredlage on DMI because milk parameters were improved in the present study and the study by Ferraretto and Shaver (2012), where DMI was increased by corn shredlage.Moreover, some of the discrepancies between studies in corn shredlage effects can be attributed to factors such as the phase of lactation of the cows, the variety of corn used, or the stage of growth of corn at harvest.However, the evaluation of these factors is beyond the scope of this study.
Feed efficiency (milk/DMI) was significantly greater in cows fed the CON diet than in cows fed the SHR diet, resulting from increased DMI without a corresponding increase in milk production of SHR cows.Throughout our study, the cows were in the first half of lactation (on average between 77 and 137 DIM).Therefore, the greater feed efficiency in cows fed the CON diet might be due to the higher contribution of body reserves to milk production because these cows had lower DMI than SHR cows.The contribution of body reserves to milk production in our experiment is suggested by high feed efficiency (>1.9;Table 3).When cows are in negative energy balance, body reserves contribute to milk production, which inflates feed efficiency (Arndt et al. 2015).In agreement with our results, Spurlock et al. (2012) reported feed efficiency >1.8 for the first 150 DIM.Lower body weight in cows fed the CON diet al.so suggests higher body reserve mobilisation of these cows (Table 3).Although the feed efficiency (milk/DMI) was significantly higher and body weight lower in CON than in SHR, the differences were too small (1.5% and 0.7%, respectively; Table 3) to be biologically important and practically interesting.In addition, when expressed as ECM/DMI, the feed efficiency remained unaffected.Variable results have been obtained in previous studies.While Ferraretto and Shaver (2012) and Vanderwerff et al. (2015) reported no effect of corn shredlage inclusion on feed efficiency, Bach et al. (2021) found a small but significant decrease in feed efficiency (ECM/DMI) in cows fed corn shredlage.

Rumination and rumen fermentation
Cows in our study spent an average of 417 min/d ruminating (ranging from 267 to 619 min/d).This average value and variability are consistent with previous reports of average rumination times of 436 min/d (from 236 to 610 min/d; White et al. 2017) and  (Vanderwerff et al. 2015), the results are inconsistent with the general assumption that a greater proportion of long particles is positively related to chewing and rumination activity (Mertens 1997;Zebeli et al. 2012).Rumination time is influenced by the particle size of the diet, fragility of the feed, NDF intake, and digestibility of fibre, with complex interactions among these factors (Beauchemin 2018).Therefore, a higher proportion of particles !19 mm with SHR might be compensated for by the higher fragility of the shredlage particles.In addition, the content of particles !8 mm was similar between corn silage and corn shredlage because corn shredlage had a higher proportion of particles !19 mm, but a lower proportion of particles between 19 mm and 8 mm (Table 1).The particle size distribution between corn silage and corn shredlage in the present study was similar to that reported in previous studies of corn shredlage (Ferraretto and Shaver 2012;Vanderwerff et al. 2015;Bach et al. 2021).The insignificant effect of corn shredlage on rumination time might also be explained by the lower proportion of forage to concentrate in the experimental diets.It has been previously reported that increasing the particle size of the diet is more effective at promoting rumination in diets containing a greater forage-to-concentrate ratio (Beauchemin 2018).
The rumen concentrations of VFA, lactate, ammonia-N, and rumen pH were similar in cows fed the CON and SHR diets.Thus, our results suggest that the replacement of corn silage with corn shredlage had no effect on rumen fermentation.The absence of any effect on rumen fermentation was consistent with the absence of an effect on rumination time and milk fat concentration.It has been proposed that increasing the particle size of the diet can prevent acidosis by promoting chewing and saliva production, thereby increasing the flow of bicarbonate into the rumen (Beauchemin 2018).However, as mentioned above, corn shredlage increased the proportion of long particles (!19 mm) and decreased the proportion of medium particles (between 19 and 8 mm), resulting in a similar proportion of particles >8 mm between diets.The effect of corn shredlage on rumen fermentation parameters has not been previously reported.However, in keeping with our results, Bach et al. (2021) reported no effect of corn shredlage on the microbial diversity and relative abundance of microbial groups in the rumen of dairy cows.Nevertheless, it should be noted that in both studies, in our study and Bach et al. (2021), rumen fluid was collected as a spot sample at a single moment in time.Spot samples of rumen fluid cannot capture fluctuations in rumen fermentation parameters during the day.Therefore, this sampling method may not have sufficient sensitivity to compare treatments if there is little difference in the effects between the treatments.The effect of corn shredlage on rumen fermentation was negligible, even though the experimental diets were formulated to be low in fibre.We expected a low-fibre diet to decrease ruminal pH and increase the likelihood of positive effects of shredlage on rumen fermentation.The NDF (25%) and ADF (15%) contents of the diets in the present study were at the (25%) and below the (17%) respective minimum recommendations of the NRC (2001).In contrast, the content of non-fibre carbohydrates (45%) was slightly higher than the maximum recommendation of the NRC (2001).In previous studies on corn shredlage, the NDF content was well above (28-32%) recommended level (Ferraretto and Shaver 2012;Vanderwerff et al. 2015;Bach et al. 2021).Despite the low-fibre diets in our study, the rumen pH was at the higher end of the physiological pH range (5.5-6.8;Mesgaran et al. 2020) in both treatments (pH was above 6.4;Table 4).The higher pH might be explained by the low intensity of lactate fermentation, as suggested by the low levels of rumen lactate (Table 4), use of a dietary buffer (sodium bicarbonate), and high variability of pH during the day (Krause and Oetzel 2006) which may not have been captured by spot sampling.

Conclusion
The inclusion of corn shredlage (25% TMR, DM basis) in a low-fibre diet of dairy cows increased DMI and tended to increase yields of 4% FCM, ECM, and milk fat.The positive effects on milk performance were the result of increased DMI rather than improved rumen fermentation or rumination, as suggested by unchanged rumen fermentation parameters and rumination time.The mechanism by which shredlage increases the DMI remains to be elucidated.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Table 2 .
Ingredients and chemical composition (g/kg of DM unless otherwise stated) of experimental diets.Vitamin and mineral mix, declared contents (per kg): 155 g of Ca, 40 g of P, 35 g of Na, 40 g of Mg, 403,100 IU of vitamin A, 73,494 IU of vitamin D, 1,200 mg of vitamin E, 4,700 mg of Niacinamide, 3,160 mg of Zn, 4,855 mg of Mn, 630 mg of Cu, 52.5 mg of I, 21 mg of Co, 18.4 mg of Se, 5 Â 10 10 CFU of Saccharomyces cerevisiae NCYC Sc 47 (VVS Verm e rovice, s. r. o., Verm e rovice, Czech Republic).

Table 4 .
Rumen fermentation parameters of cows fed control (CON) or shredlage (SHR) diet.434 min/d (from 151 to 630 min/d; Zebeli et al. 2006) in dairy cows.Rumination time in min/d and min/kg of DMI was unaffected by diet in the present study.Although our results are in accordance with previous findings in dairy cows fed brown midrib corn shredlage