EFFECTS OF FORAGE CHARACTERISTICS AND VOLUNTARY FEED INTAKE ON
RUMINAL PASSAGE OF DIGESTA FRACTIONS IN LACTATING DAIRY COWS
By
Kimberly L. Kammes

A DISSERTATION
Submitted to
Michigan State University
in partial requirements
for the degree of
DOCTOR OF PHILOSOPHY
Animal Science
2012

ABSTRACT
EFFECTS OF FORAGE CHARACTERISTICS AND VOLUNTARY FEED INTAKE ON
RUMINAL PASSAGE OF DIGESTA FRACTIONS IN LACTATING DAIRY COWS
By
Kimberly L. Kammes
Passage from the rumen is a dynamic, complicated process that involves numerous
animal and feed factors. Ruminal passage rates affect intake, digestibility, and the amount and
type of fermentation endproducts and protein available to dairy cows. The success of existing
nutrition models is limited by the lack of data for passage rate of individual digesta fractions, and
the inability to account for effects of feed intake and forage characteristics of passage rates of
specific feed fractions. These limitations can be addressed using the pool and flux method,
which allows measurement of rates of passage for fractions within feeds, and an experimental
design optimized to assess the interactions of diet characteristics and dry matter intake (DMI).
Five experiments were conducted to evaluate the effects of level of feed intake on the
response of passage rate of digesta fractions in dairy cows to forage characteristics. All
experiments utilized ruminally and duodenally cannulated cows with a wide range of DMI in a
crossover design, where DMI was measured during a preliminary period (pDMI), and rates of
digestion and passage of feed fractions were determined by the pool and flux method. The use of
pDMI, an index of nutrient demand, allowed the evaluation of treatments on animal responses in
relation to level of intake and effects of intake level independent of treatments. Level of intake
ranged from approximately 20 to 30 kg dry matter per day among cows on the five experiments,
and each experiment evaluated individual pairs of forage treatments including: 1) legume particle
size, 2) grass particle size, 3) legume maturity, 4) grass maturity, and 5) forage family. Alfalfa
and orchardgrass were selected as a representative legume and grass, respectively, and were the

sole source of forage used in diets, which were formulated to contain a similar forage neutral
detergent fiber concentration among diets within experiment. We hypothesized that the normal
variation in diet characteristics related to forage alters both the passage rates of feed fractions
and the extent to which passage rates of these fractions are affected by voluntary DMI.
In general, it has been accepted that ruminal passage rates, microbial nitrogen flow to the
duodenum, and efficiency of microbial synthesis increase with DMI; however, this is likely an
oversimplification based on our research. Results from these experiments demonstrated passage
rates of individual digesta fractions from the rumen were highly variable at a given level of
voluntary DMI across cow periods, depended upon the forage characteristic being evaluated, and
were inconsistent among the forage treatments evaluated. Additionally, microbial efficiency was
not related to level of intake but was related to other factors, which varied among the
experiments and included passage rates of starch and potentially digestible neutral detergent
fiber, digestion rate of starch, and amount of true ruminally digested organic matter.
Furthermore, there was no evidence that the filling effect of diets affected feed intake differently
for cows with high intake compared to cows with low intake for any of the forage treatments
evaluated. Finally, this research illustrated the complexity of ruminal passage and emphasized
the difficulty involved in accurately predicting ruminal passage and digestibility.
Although the effects of DMI on passage rates are not consistent, these experiments
provide absolute passage rates of digesta fractions for use in the development of equations to
predict ruminal digesta passage and the foundation for additional research in this area. Data
obtained from these experiments and others using the pool and flux method can be compiled and
used in a meta-analysis, with the potential to discover relationships. The results will improve the
accuracy of nutrition models to predict nutrient intake, passage, and utilization in dairy cows.

ACKNOWLEDGMENTS

It is with gratification that I reflect upon my graduate program and remember all those
who helped me in any way during my journey. The completion of this dissertation would not
have been possible without the help and support from many generous people. Your
contributions to my development as a scientist and an individual are greatly appreciated, and it is
my pleasure to say “THANK YOU”.
To my advisor Dr. Allen, thank you for giving me the opportunity to pursue a doctorate
degree and your mentorship, support, and patience throughout my entire program. Even though I
would get frustrated at times, I realize now that you were providing the extra push I needed to
excel to the next level and challenging me with my best interest in mind Your wealth of
knowledge and commitment to weekly individual meetings were invaluable.
To my committee members Dr. Bucholtz, Dr. Herdt, Dr. Romsos, and Dr. WeberNielsen, thank you for your time, advice, and the interest invested in my research and program.
To the lab technicians Dave Main, Dewey Longuski, and Jackie Ying, for your many
hours of work at the farm and in the lab, willingness to help any time of the day or night and any
day of the week, ability to teach as experts in your specific areas as well as life in general,
positive encouragement, and sense of humor. Dave incredibly managed up to 11 undergraduate
students at one time and was key to the completion of lab analyses for the innumerable samples
collected during this research. Dewey was instrumental in the correct pronunciation of my last
name. The singing, humming, and whistling of Dave and Dewey were always music to my ears.
Jackie had the amazing ability to recall details regarding experiments and retrieve data from past
years. Also, thanks to the numerous undergraduate students that helped with sampling at the

iv

farm, especially with overnight collections and rumen dumping, and sample processing and lab
analyses. This very labor-intensive research was accomplished only through the involvement of
several dedicated individuals.
To Bob Kreft and the staff at the Dairy Cattle Teaching and Research Center, thank you
for taking care of the daily tasks involving the cows at the farm and your flexibility and patience
as you worked around sampling schedules and special requests. Also, thanks to Dr. Ames for
performing the ruminal and duodenal cannulation surgeries.
To graduate coordinator Dr. Bursian, thank you for your continuous support of graduate
students, willingness to listen, and concern regarding both the positives and negatives of the
program. Thanks for the daily greetings and farewells you shared as you passed by my office as
well as the friendly visits and for understanding my sense of humor. You always brightened my
day.
To fellow graduate students, thank you for your assistance with research and coursework
questions, friendship, and persistent urging to leave the work behind and go out every now and
then. Thanks for the birthday cakes and decorations – you never missed a year.
To my parents, thank you for your love, support, and encouragement throughout my
many years of education. You deserve much credit for teaching me the true meaning of hard
work and not giving up, which I needed to successfully complete this degree. Thanks for the
opportunity to grow up on a dairy farm and work along side you each day – it has and will
continue to serve me well. It is amazing how milking cows and cutting grass clears my mind and
allows me to think and the home grown, home cooked meals provide brainpower. There is no
place like home.

v

To my friend Dave Main, a special thank you for being there each step of the way – the
ups and downs and everything inbetween. Thanks for providing a listening ear, shoulder to lean
on, and caring heart in times of need and for sharing your time, talents, words of wisdom, advice,
laughs, and all you have taught me, especially who I am and who I can be, and, of course, the
many facial expressions I never knew I had. Thank you for teaching me how to play the guitar,
which has become a great hobby for me, as well as your patience, understanding, and flexibility
during the lessons. I am grateful for the opportunity to meet and get to know your family. Your
parents always made me feel welcome, special, and “at home”, and I enjoyed interacting with
your children by working together at the farm, in the lab, and attending some of their horse and
sporting events. Most importantly, thank you for your friendship and the time we shared. I
could not have done this without you.
Additionally, thank you to the funding sources, which included the National Research
Initiative Competitive Grant no. 2006-35206-16708 from the USDA National Institute of Food
and Agriculture, the Animal Agriculture Initiative, and the MSU Graduate School. Thanks to
Barenbrug USA for their donation of Baridana orchardgrass seed and West Central Soy for
®

donating the SoyPlus protein supplement.

vi

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................................... xi
LIST OF FIGURES ................................................................................................................... xvi
KEY TO ABBREVIATIONS .....................................................................................................xx
CHAPTER 1
A Review of Literature
Overview..............................................................................................................................1
Ruminal Digestion and Passage...........................................................................................1
Constraints to Passage from the Rumen ..............................................................................2
Particle Size .............................................................................................................2
Particle Density........................................................................................................3
Ruminal Mat ............................................................................................................4
Selective Retention ..................................................................................................4
Ruminant Nutrition Models .................................................................................................5
Limitations of Ruminant Nutrition Models .............................................................6
Measurements of Ruminal Rates of Digestion and Passage................................................7
Pool and Flux Method..............................................................................................9
Effects of DMI on Passage Rate ..........................................................................................9
Effects of Feed Characteristics on Passage Rate ...............................................................12
Forages and Rumen Fill .........................................................................................12
Forages and Chewing Activity...............................................................................13
Experimental Model...............................................................................................14
Forage Factors Affecting Digesta Passage Rates...............................................................15
Forage NDF Concentration....................................................................................15
Forage NDF Digestibility ......................................................................................16
Forage Particle Size ...............................................................................................17
Forage Maturity .....................................................................................................18
Forage Family ........................................................................................................19
Research Objectives and Hypothesis .................................................................................20
Appendix............................................................................................................................22
References..........................................................................................................................25
CHAPTER 2
Nutrient Demand Interacts with Legume Particle Length to Affect Digestion Responses
and Rumen Pool Sizes in Dairy Cows
Abstract ..............................................................................................................................33
Introduction........................................................................................................................34
Materials and Methods.......................................................................................................36

vii

Cows and Treatments.............................................................................................36
Data and Sample Collection...................................................................................38
Sample Analysis and Calculations.........................................................................39
Statistical Analysis.................................................................................................42
Results and Discussion ......................................................................................................43
Comparison of Forages and Diets..........................................................................43
Effects of Legume FPL and pDMI ........................................................................45
Effects of pDMI on Ruminal Passage Rates..........................................................49
Effects of Treatment and pDMI on N Flux and Microbial Efficiency ..................49
Conclusions........................................................................................................................50
Appendix............................................................................................................................52
References..........................................................................................................................78
CHAPTER 3
Nutrient Demand Interacts with Grass Particle Length to Affect Digestion Responses and
Chewing Activity in Dairy Cows
Abstract ..............................................................................................................................82
Introduction........................................................................................................................83
Materials and Methods.......................................................................................................85
Cows and Treatments.............................................................................................85
Data and Sample Collection...................................................................................86
Sample Analysis and Calculations.........................................................................87
Statistical Analysis.................................................................................................91
Results and Discussion ......................................................................................................92
Comparison of Forages and Diets..........................................................................92
Effects of Grass FPL and pDMI ............................................................................93
Effects of pDMI on Ruminal Passage Rates..........................................................96
Effects of pDMI on N Flux and Microbial Efficiency...........................................98
Conclusions........................................................................................................................99
Appendix..........................................................................................................................101
References........................................................................................................................129
CHAPTER 4
Nutrient Demand Interacts with Legume Maturity to Affect Rumen Pool Sizes in Dairy
Cows
Abstract ............................................................................................................................134
Introduction......................................................................................................................135
Materials and Methods.....................................................................................................137
Cows and Treatments...........................................................................................137
Data and Sample Collection.................................................................................138
Sample Analysis and Calculations.......................................................................139
Statistical Analysis...............................................................................................142
Results and Discussion ....................................................................................................143

viii

Comparison of Forages and Diets........................................................................143
Effects of Legume Maturity and pDMI ...............................................................145
Effects of pDMI on Ruminal Passage Rates........................................................149
Effects of Treatment and pDMI on N Flux and Microbial Efficiency ................151
Conclusions......................................................................................................................152
Appendix..........................................................................................................................154
References........................................................................................................................179
CHAPTER 5
Nutrient Demand Interacts with Grass Maturity to Affect Milk Fat Concentration and
Digestion Responses in Dairy Cows
Abstract ............................................................................................................................183
Introduction......................................................................................................................184
Materials and Methods.....................................................................................................186
Cows and Treatments...........................................................................................186
Data and Sample Collection.................................................................................187
Sample Analysis and Calculations.......................................................................189
Statistical Analysis...............................................................................................192
Results and Discussion ....................................................................................................193
Comparison of Forages and Diets........................................................................193
Effects of Grass Maturity and pDMI ...................................................................194
Effects of pDMI on Ruminal Passage Rates........................................................199
Effects of Treatment and pDMI on N Flux and Microbial Efficiency ................200
Conclusions......................................................................................................................201
Appendix..........................................................................................................................203
References........................................................................................................................234
CHAPTER 6
Nutrient Demand Interacts with Forage Family to Affect Digestion Responses in Dairy
Cows
Abstract ............................................................................................................................238
Introduction......................................................................................................................239
Materials and Methods.....................................................................................................241
Cows and Treatments...........................................................................................241
Data and Sample Collection.................................................................................242
Sample Analysis and Calculations.......................................................................244
Statistical Analysis...............................................................................................247
Results and Discussion ....................................................................................................248
Comparison of Forages and Diets........................................................................249
Effects of Forage Family and pDMI....................................................................250
Effects of pDMI on Ruminal Passage Rates........................................................256
Effects of Treatment and pDMI on N Flux and Microbial Efficiency ................257
Conclusions......................................................................................................................258
Appendix..........................................................................................................................260
References........................................................................................................................293

ix

CHAPTER 7
Rates of Particle Size Reduction and Passage are Faster for Legume Compared to CoolSeason Grass Resulting in Lower Rumen Fill and Less Effective Fiber
Abstract ............................................................................................................................297
Introduction......................................................................................................................298
Materials and Methods.....................................................................................................300
Cows and Treatments...........................................................................................300
Data and Sample Collection.................................................................................301
Sample Analysis and Calculations.......................................................................302
Statistical Analysis...............................................................................................305
Results and Discussion ....................................................................................................306
Conclusions......................................................................................................................310
Appendix..........................................................................................................................311
References........................................................................................................................323
CHAPTER 8
Epilogue
Recap................................................................................................................................326
Summary of Results: Effects of Forage Characteristics, Level of Intake, and Their
Interactions.......................................................................................................................326
Application of Data..........................................................................................................329
Future Research ...............................................................................................................330
Appendix..........................................................................................................................332
References........................................................................................................................344

x

LIST OF TABLES

Table 1. Rate of passage of feeds determined by excretion pattern of rare earth markers in feces
or of feed fractions by the pool and flux method...........................................................................23
Table 2. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet........................................................................................................53
Table 3. Chemical composition, particle size distribution, and fermentation parameters of the
long (19 mm) and short cut (10 mm) alfalfa silage included in the treatment diets ......................54
Table 4. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole source of forage
........................................................................................................................................................56
Table 5. Milk production and composition, feed intake, and BW change of cows fed treatment
diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole source of
forage .............................................................................................................................................57
Table 6. Rumen kinetics of cows fed treatment diets containing either long (19 mm) or short cut
(10 mm) alfalfa silage as the sole source of forage .......................................................................58
Table 7. Rumen pools of cows fed treatment diets containing either long (19 mm) or short cut
(10 mm) alfalfa silage as the sole source of forage .......................................................................60
Table 8. Starch digestion of cows fed treatment diets containing either long (19 mm) or short cut
(10 mm) alfalfa silage as the sole source of forage .......................................................................61
Table 9. NDF digestion of cows fed treatment diets containing either long (19 mm) or short cut
(10 mm) alfalfa silage as the sole source of forage .......................................................................62
Table 10. DM and OM digestion of cows fed treatment diets containing either long (19 mm) or
short cut (10 mm) alfalfa silage as the sole source of forage.........................................................64
Table 11. Ruminal VFA concentrations and pH of cows fed treatment diets containing either
long (19 mm) or short cut (10 mm) alfalfa silage as the sole source of forage .............................66
Table 12. Nitrogen metabolism of cows fed treatment diets containing either long (19 mm) or
short cut (10 mm) alfalfa silage as the sole source of forage.........................................................67
Table 13. Characterization of 15 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet......................................................................................................102
Table 14. Chemical composition, particle size distribution, and fermentation parameters of the

xi

long (19 mm) and short cut (10 mm) orchardgrass silage included in the treatment diets..........103
Table 15. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the sole source of
forage ...........................................................................................................................................105
Table 16. Milk production and composition, feed intake, and BW change of cows fed treatment
diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the sole source
of forage .......................................................................................................................................106
Table 17. Nitrogen metabolism of cows fed treatment diets containing either long (19 mm) or
short cut (10 mm) orchardgrass silage as the sole source of forage ............................................107
Table 18. Rumen kinetics of cows fed treatment diets containing either long (19 mm) or short
cut (10 mm) orchardgrass silage as the sole source of forage .....................................................109
Table 19. Starch digestion of cows fed treatment diets containing either long (19 mm) or short
cut (10 mm) orchardgrass silage as the sole source of forage .....................................................111
Table 20. Rumen pools of cows fed treatment diets containing either long (19 mm) or short cut
(10 mm) orchardgrass silage as the sole source of forage ...........................................................112
Table 21. Chewing activity of cows fed treatment diets containing either long (19 mm) or short
cut (10 mm) orchardgrass silage as the sole source of forage .....................................................113
Table 22. Ruminal VFA concentrations and pH of cows fed treatment diets containing either
long (19 mm) or short cut (10 mm) orchardgrass silage as the sole source of forage .................115
Table 23. DM and OM digestion of cows fed treatment diets containing either long (19 mm) or
short cut (10 mm) orchardgrass silage as the sole source of forage ............................................116
Table 24. NDF digestion of cows fed treatment diets containing either long (19 mm) or short cut
(10 mm) orchardgrass silage as the sole source of forage ...........................................................118
Table 25. Characterization of 16 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet......................................................................................................155
Table 26. Chemical composition, particle size distribution, and fermentation parameters of the
early or late cut alfalfa silage included in the treatment diets......................................................156
Table 27. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either early or late cut alfalfa silage as the sole source of forage ..............................158
Table 28. Milk production and composition, feed intake, and BW change of cows fed treatment
diets containing either early or late cut alfalfa silage as the sole source of forage......................159

xii

Table 29. NDF digestion of cows fed treatment diets containing either early or late cut alfalfa
silage as the sole source of forage................................................................................................160
Table 30. Rumen kinetics of cows fed treatment diets containing either early or late cut alfalfa
silage as the sole source of forage................................................................................................162
Table 31. Rumen pools of cows fed treatment diets containing either early or late cut alfalfa
silage as the sole source of forage................................................................................................164
Table 32. Starch digestion of cows fed treatment diets containing either early or late cut alfalfa
silage as the sole source of forage................................................................................................165
Table 33. Ruminal VFA concentrations and pH of cows fed treatment diets containing either
early or late cut alfalfa silage as the sole source of forage ..........................................................166
Table 34. DM and OM digestion of cows fed treatment diets containing either early or late cut
alfalfa silage as the sole source of forage ....................................................................................167
Table 35. Nitrogen metabolism of cows fed treatment diets containing either early or late cut
alfalfa silage as the sole source of forage ....................................................................................169
Table 36. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet......................................................................................................204
Table 37. Chemical composition, particle size distribution, and fermentation parameters of the
early or late cut orchardgrass silage included in the treatment diets ...........................................205
Table 38. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either early or late cut orchardgrass silage as the sole source of forage ....................207
Table 39. Milk production and composition, feed intake, and BW change of cows fed treatment
diets containing either early or late cut orchardgrass silage as the sole source of forage............208
Table 40. Starch digestion of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................209
Table 41. Ruminal VFA concentrations and pH of cows fed treatment diets containing either
early or late cut orchardgrass silage as the sole source of forage ................................................210
Table 42. Nitrogen metabolism of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................211
Table 43. Milk fatty acid profile of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................213
Table 44. NDF digestion of cows fed treatment diets containing either early or late cut

xiii

orchardgrass silage as the sole source of forage ..........................................................................216
Table 45. DM and OM digestion of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................218
Table 46. Rumen kinetics of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................220
Table 47. Rumen pools of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................222
Table 48. Chewing activity of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage ..........................................................................223
Table 49. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet......................................................................................................261
Table 50. Chemical composition, particle size distribution, and fermentation parameters of
alfalfa silage and orchardgrass silage included in the treatment diets .........................................262
Table 51. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of forage ............264
Table 52. Milk production and composition, feed intake, and BW change of cows fed treatment
diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of forage....265
Table 53. NDF digestion of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................266
Table 54. Rumen kinetics of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................268
Table 55. Rumen pools of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................270
Table 56. Starch digestion of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................271
Table 57. Ruminal VFA concentrations and pH of cows fed treatment diets containing either
alfalfa (AL) or orchardgrass (OG) silage as the sole source of forage ........................................272
Table 58. Nitrogen metabolism of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................273
Table 59. DM and OM digestion of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................275

xiv

Table 60. Chewing activity of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................277
Table 61. Plasma metabolites of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................279
Table 62. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet......................................................................................................312
Table 63. Chemical composition and particle size distribution of the alfalfa silage and
orchardgrass silage included in the treatment diets .....................................................................313
Table 64. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of forage ............314
Table 65. Particle size distribution of intake, rumen pool, and duodenal flux for cows fed
treatment diets containing alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage ...........................................................................................................................................315
Table 66. Particle size kinetics for cows fed treatment diets containing alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage.................................................................317
Table 67. Relationship between DMI (kg/d) and ruminal NDF pool (kg) or ruminal digesta wet
weight (kg) ...................................................................................................................................333

xv

LIST OF FIGURES

Figure 1. Simple first-order model of digestion and passage in the rumen for a feed fraction. ...24
Figure 2. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for milk fat concentration. ........69
Figure 3. Interaction of long (19 mm) and short (10 mm) alfalfa particle length with preliminary
DMI for A) NDF rumen pool, B) rumen digesta wet weight, C) rumen digesta volume, and D)
DMI................................................................................................................................................70
Figure 4. Interaction of long (19 mm) and short (10 mm) alfalfa particle length with preliminary
DMI for starch A) true ruminal digestibility, B) flux from the rumen to duodenum, and C)
postruminal digestibility. ...............................................................................................................72
Figure 5. Interaction of long (19 mm) and short (10 mm) alfalfa particle length with preliminary
DMI for NDF ruminal digestibility................................................................................................74
Figure 6. Interaction of long (19 mm) and short (10 mm) alfalfa particle length with preliminary
DMI for nonammonia, nonmicrobial N flux from the rumen to duodenum..................................75
Figure 7. Interaction of long (19 mm) and short (10 mm) alfalfa particle length with preliminary
DMI for A) nonammonia, nonmicrobial N and B) microbial N flux from the rumen to duodenum
expressed as percent of duodenal NAN. ........................................................................................76
Figure 8. A) Relationship between starch ruminal digestion rate (kd) and microbial N (MN) flux
from rumen to duodenum for long and short. B) Relationship between true ruminally digested
OM (TRDOM) and MN flux from rumen to duodenum for long and short alfalfa particle
length..............................................................................................................................................77
Figure 9. Interaction of long (19 mm) and short (10 mm) orchardgrass particle length with
preliminary DMI for: A) potential digestible NDF (pdNDF) ruminal digestion rate, B) NDF
ruminal turnover rate, C) NDF rumen pool, D) DMI, E) eating time, min/kg NDF intake, and F)
eating time, min/d. .......................................................................................................................120
Figure 10. Relationship of long (19 mm) and short (10 mm) orchardgrass particle length with
preliminary DMI for: A) starch ruminal digestion rate, B) starch true ruminal digestibility, C)
starch rumen pool, D) ruminal ammonia concentration, E) branched chain VFA concentration,
and F) ammonia N flow to duodenum. ........................................................................................123
Figure 11. Relationship of long (19 mm) and short (10 mm) orchardgrass particle length with
preliminary DMI for: A) ruminating time per unit forage NDF, B) potentially digestible NDF
(pdNDF) ruminal digestibility, C) pdNDF flux to duodenum, and D) ruminal acetate
concentration................................................................................................................................126

xvi

Figure 12. A) Relationship between starch ruminal passage rate and microbial efficiency. B)
Relationship between potentially digestible NDF (pdNDF) ruminal passage rate and microbial
efficiency. Open circles denote long (19 mm) and closed circles denote short (10 mm)
orchardgrass particle length. ........................................................................................................128
Figure 13. Interaction of early and late alfalfa maturity with preliminary DMI for A) indigestible
NDF (iNDF) rumen pool and B) NDF rumen pool. ....................................................................171
Figure 14. Interaction of early and late alfalfa maturity with preliminary DMI for ruminal
acetate concentration....................................................................................................................172
Figure 15. Relationship of early (open circles) and late (closed circles) alfalfa maturity with
preliminary DMI for indigestible NDF (iNDF) ruminal passage rate. ........................................173
Figure 16. Interaction of early and late alfalfa maturity with preliminary DMI for A) starch
digestion rate, B) ruminal ammonia concentration, and C) branched chain VFA
concentration................................................................................................................................174
Figure 17. Interaction of early and late alfalfa maturity with preliminary DMI for microbial
efficiency expressed as g of microbial N/kg of true ruminally digested OM (TRDOM)............176
Figure 18. Relationship between true ruminally digested OM (TRDOM) and microbial
efficiency for early and late alfalfa maturity................................................................................177
Figure 19. Relationship between starch ruminal digestion rate and microbial efficiency for early
and late alfalfa maturity. ..............................................................................................................178
Figure 20. Interaction of early and late orchardgrass maturity with preliminary DMI for milk fat
concentration................................................................................................................................225
Figure 21. Interaction of early and late orchardgrass maturity with preliminary DMI for rumen
digesta volume. ............................................................................................................................226
Figure 22. Interaction of early and late orchardgrass maturity with preliminary DMI for
potentially digestible NDF (pdNDF) A) ruminal passage rate, B) ruminal digestibility, and C)
flux to the duodenum. ..................................................................................................................227
Figure 23. Interaction of early and late orchardgrass maturity with preliminary DMI for starch
A) ruminal passage rate, B) true ruminal digestibility, and C) postruminal digestibility............229
Figure 24. Relationship of early and late orchardgrass maturity with preliminary DMI for
indigestible (iNDF) ruminal passage rate. ...................................................................................231
Figure 25. Interaction of early and late orchardgrass maturity with preliminary DMI for
nonammonia, nonmicrobial N flux to the duodenum. .................................................................232

xvii

Figure 26. Relationship between true ruminally digested OM (TRDOM) and microbial
efficiency for early and late orchardgrass maturity. ....................................................................233
Figure 27. Interaction of alfalfa and orchardgrass with preliminary DMI for DMI. ..................280
Figure 28. Interaction of alfalfa and orchardgrass with preliminary DMI for indigestible NDF
(iNDF) A) intake, B) ruminal passage rate, and C) ruminal turnover time. ................................281
Figure 29. Interaction of alfalfa and orchardgrass with preliminary DMI for A) ruminal
digestibility of potentially digestible NDF (pdNDF), B) ruminal digestibility of NDF, C) pdNDF
flux to duodenum, and D) NDF flux to duodenum......................................................................283
Figure 30. Interaction of alfalfa and orchardgrass with preliminary DMI for A) meal bouts, B)
rumination bouts, and C) rumination interval..............................................................................285
Figure 31. Relationship of alfalfa and orchardgrass with preliminary DMI for concentrations of
plasma A) glucose, B) glucagon, and C) insulin. ........................................................................287
Figure 32. Interaction of alfalfa and orchardgrass with preliminary DMI for A) ammonia N, B)
nonammonia, nonmicrobial N (NANMN), and C) NAN flux to the duodenum.........................289
Figure 33. Interaction of alfalfa and orchardgrass with preliminary DMI for microbial efficiency
expressed as gram of microbial N produced per kilogram of true ruminally digested OM
(TRDOM). ...................................................................................................................................291
Figure 34. A) Relationship between potentially digestible NDF (pdNDF) ruminal passage rate
and microbial N flux to duodenum for alfalfa and orchardgrass. B) Relationship between pdNDF
ruminal passage rate and microbial efficiency for alfalfa and orchardgrass. ..............................292
Figure 35. Model of ruminal particle size reduction and passage. .............................................318
Figure 36. Particle size distribution of NDF in intake, rumen, and duodenum. .........................319
Figure 37. Interaction of alfalfa and orchardgrass with preliminary DMI for the proportion of
NDF intake as A) large particles (NDF L; !2.36 mm) or B) small particles (NDF S; <2.36
mm). .............................................................................................................................................320
Figure 38. Interaction of alfalfa and orchardgrass with preliminary DMI for passage rate of large
particles (!2.36 mm) of potentially digestible NDF (pdNDF L).................................................321
Figure 39. Passage rates of indigestible NDF small particles (iNDF S; <2.36 mm) and
potentially digestible NDF small particles (pdNDF S; <2.36 mm) for cows fed diets containing
alfalfa or orchardgrass as the sole source of forage.....................................................................322

xviii

Figure 40. Relationship between preliminary DMI and A) indigestible NDF (iNDF) passage
rate for alfalfa maturity, B) iNDF passage rate for orchardgrass maturity, C) potentially
digestible NDF (pdNDF) passage rate for orchardgrass maturity, D) starch passage rate for
orchardgrass maturity, and E) iNDF passage rate for forage family. .........................................334
Figure 41. Relationship between microbial efficiency expressed as g of microbial N per kg true
ruminally digested OM (TRDOM) and A) starch passage rate for orchardgrass particle size, B)
potentially digestible NDF (pdNDF) passage rate for orchardgrass particle size, C) starch
digestion rate for alfalfa maturity, D) TRDOM for alfalfa maturity, E) TRDOM for orchardgrass
maturity, and F) pdNDF passage rate for forage family..............................................................337
Figure 42. Relationship between preliminary DMI (pDMI) and DMI for: A) alfalfa particle size,
B) orchardgrass particle size, C) forage family, D) alfalfa maturity, and E) orchardgrass maturity.
......................................................................................................................................................341

xix

KEY TO ABBREVIATIONS

ADF

Acid detergent fiber

ADL

Acid detergent lignin

AL

Alfalfa

ANOVA

Analysis of variance

AOAC

Association of Official Analytical Chemists

BCS

Body condition score

BW

Body weight

CLA

Conjugated linoleic acid

CNCPS

Cornell Net Carbohydrate Protein System

CP

Crude protein

DM

Dry matter

DIM

Days in milk

DMI

Dry matter intake

DHIA

Dairy Herd Improvement Association

EARLY

Diet containing early maturity forage

FCM

Fat-corrected milk

FPL

Forage particle length

HPLC

High performance liquid chromatography

iNDF

Indigestible neutral detergent fiber

kd

Rate of digestion

kp

Rate of passage

xx

kr

Rate of particle size reduction

L

Large particle (! 2.36 mm)

LATE

Diet containing late maturity forage

LONG

Diet containing forage with theoretical length of cut of 19 mm

LSM

Least squares mean

MFD

Milk fat depression

MN

Microbial nitrogen

MUN

Milk urea nitrogen

NAN

Nonammonia nitrogen

NANMN

Nonammonia, nonmicrobial nitrogen

NDF

Neutral detergent fiber

NDFD

Neutral detergent fiber digestibility

NRC

National Research Council

NS

Not significant

OG

Orchardgrass

OM

Organic matter

pDMI

Preliminary dry matter intake

pdNDF

Potentially digestible neutral detergent fiber

2

R

Coefficient of determination

S

Small particle (< 2.36 mm)

SD

Standard deviation

SHORT

Diet containing forage with theoretical length of cut of 10 mm

SNF

Solids-not-fat

xxi

TDN

Total digestible nutrient

TLC

Theoretic length of cut

TMR

Total mixed ration

TRDOM

True ruminally digested organic matter

VFA

Volatile fatty acids

xxii

CHAPTER 1
A Review of Literature

OVERVIEW
Passage from the rumen is a dynamic, complicated process that involves numerous
animal and feed factors. Passage of liquid and particulate matter from the rumen occurs through
the reticular-omasal orifice, and the flow of particles can be restricted by particle size, particle
density, and particle entrapment within the rumen mat. Ruminal passage rates of nutrients affect
feed intake, nutrient digestibility, and the amount and type of fermentation endproducts and
protein available to dairy cows. The accuracy of existing nutrition models are hampered by
limited availability of accurate passage rate data, especially for feed fractions, and lack of
information on how passage rates are affected by dry matter intake (DMI) and dietary
characteristics. Although the importance of digesta passage and its role in the performance of
lactating dairy cows is recognized, a limited understanding of the process and lack of accurate
quantitative measures of passage rates for feed fractions greatly hinders the ability of ruminant
nutrition models to accurately predict ruminal digestion and passage.

RUMINAL DIGESTION AND PASSAGE
Feeds that enter the rumen can disappear by two processes: digestion or passage (Figure
1). Rates of digestion and passage are important determinants of ruminal digestibility (Waldo et
al., 1972), which is calculated using the following equation: ruminal digestibility (%) =
[kd/(kd+kp)]*100; where kd = rate of digestion (%/h) and kp = rate of passage (%/h). This
equation is applicable to homogenous feed fractions only as both digestion rate and passage rate

1

can vary widely among fractions within an individual feed ingredient. The processes of
digestion and passage compete against one another affecting digestibility; digestibility increases
as digestion rate increases or as passage rate decreases (i.e. retention time in the rumen
increases). Retention time is the reciprocal of fractional rate of passage. Given the impact of
passage on feed intake, ruminal digestibility, and microbial growth, a greater understanding of
the influence of animal and dietary characteristics and their interactions on ruminal passage of
particulate matter and quantitative knowledge of rates of nutrient passage from the rumen are
necessary.

CONSTRAINTS TO PASSAGE FROM THE RUMEN
Passage rate of fibrous particles largely depends on the reduction of particle size and the
increase in particle specific gravity, which are required for particles to escape the rumen mat,
sink to the ventral rumen, and exit the rumen via the reticular-omasal orifice (Sutherland, 1988).
Particle size reduction is mainly a result of chewing during eating and ruminating (Kennedy,
1985). Although digestion has little direct effect on particle breakdown (Murphy and Nicoletti,
1984), it affects the rate of reduction from chewing by increasing tissue fragility (Chai et al.,
1984). Differences in fragility of forage particles affect particle size reduction rate and retention
time in the rumen (McLeod and Minson, 1988).
Particle Size
Resistance to flow from the rumen increases with particle size (Poppi et al., 1980; Dixon
and Milligan, 1985), and passage rate from the rumen increases exponentially with decreasing
particle size (Lechner-Doll et al., 1991). Poppi et al. (1980) suggested a threshold size in which
particles greater than the defined size have increased resistance to passage, and this was

2

determined to be particles retained on sieves with apertures of 1.18 mm for sheep (Poppi et al.,
1980) and 3 to 4 mm for dairy cows (Cardoza and Mertens, 1986) and steers (Dixon and
Milligan, 1985). However, the threshold size likely varies with physical form of forages, DMI,
and neutral detergent fiber (NDF) concentration in the diet (Van Soest et al., 1988).
Particle Density
Although particle size reduction is a prerequisite to flow from the reticulorumen, it is not
always a constraint because most of the particulate matter in the rumen is smaller than the
maximum particle size of the feces (Allen, 1996). This suggests that particle size reduction is not
the only requirement for particles to flow from the rumen. Particle density is also a constraint to
flow; a negative relationship between particle density and retention time in the rumen exists
(Allen, 1996). Ruminal retention time decreased from 91 to 19 h as density of inert particles
increased from 0.9 to 1.5 g/ml across experiments reported in the literature (Lechner-Doll et al.,
1991). Despite a true specific gravity of 1.3 to 1.5 (Siciliano-Jones and Murphy, 1991), most
digesta particles in the rumen are buoyant (Sutherland, 1988) which decreases their probability
of escape. Buoyancy is the result of retention of fermentation gases in and on particles (Hooper
and Welch, 1985). Its effect lessens as digestion progresses (Hooper and Welch, 1985; Nocek
and Kohn, 1987; Siciliano-Jones and Murphy, 1991) because the digestible portion of the
particle or the ability of the cells to retain gas or both are diminished (Jung and Allen, 1995),
which increases the likelihood of passage. Forages differ greatly in buoyancy over time because
of differences in both concentration and digestion rate of potentially digestible NDF (pdNDF;
Jung and Allen, 1995) and because of differences in their anatomic structure that affect their
ability to retain gases (Wilson et al., 1989).

3

Ruminal Mat
The distribution of particles, including stratification of long fibrous particles into layers,
in the rumen forms a mat that functions effectively as a retaining mechanism (Sutherland, 1988)
with less dense, longer material on top and more dense, smaller particles sinking to the bottom.
The ruminal mat has high selectivity for large particles (Sutherland, 1988), and long forage
particles entrap smaller particles from other feeds increasing ruminal retention time and
digestibility (Grant, 1997). Escape from the ruminal mat has been identified as a rate-limiting
step to the passage of forage particles (Poppi et al., 2001).
Selective Retention
Particles within the rumen can be defined as potentially escapable or non-escapable based
on size and density, and the rate at which a non-escapable particle becomes potentially escapable
depends on the rate of particle size reduction and change in density and buoyancy (Allen and
Mertens, 1988). This suggests that passage is not random. Instead, it is mediated by selective
retention of undigested fibrous particles in the rumen, a phenomenon that evolved in ruminants
to maximize ruminal fiber digestion (Allen and Mertens, 1988), and involves the ruminal mat
previously discussed. Although fibrous particles contain both indigestible and digestible
fractions, it is reasoned that the likelihood of particles to escape the rumen should increase as the
particle increases in indigestible material; therefore, the passage rate of indigestible NDF (iNDF)
should be greater than that of pdNDF (Huhtanen et al., 1995). Most studies have reported
selective retention of pdNDF compared to iNDF (Tamminga et al., 1989; Rinne et al., 2002;
Huhtanen et al., 2007; Lund et al., 2007). However, Stensig and Robinson (1997) reported a
faster rate of passage of iNDF than that of pdNDF for timothy but not alfalfa. Passage rates of

4

pdNDF and iNDF might vary with the distribution of these fractions within plant parts and with
differences in buoyancy over time (Firkins et al, 1998).
There is no single rate-limiting component to flow from the rumen, and more research is
needed to expand our knowledge in particle dynamics and their effects on passage rate. This was
explored further in a comparison of alfalfa and orchardgrass reported in Chapter 7 of this
dissertation.

RUMINANT NUTRITION MODELS
A basic goal of ruminant research has been to improve animal production and reduce
excretion of waste, which requires unraveling the mechanisms that influence nutrient utilization
by the animal. Therefore, extensive research has been conducted over several decades to
increase our knowledge in areas including, but not limited to, feeds and feeding management,
intake, digestion and passage rates, escape of dietary protein, and microbial growth efficiency to
estimate energy, nutrient requirements and supply, and feed utilization for various feeding
situations. This research provided a large volume of scientific knowledge, and mathematical
models were generated to integrate various aspects of nutrient utilization in an effort to advance
diet formulation, predict animal responses, and explain biological mechanisms.
Ruminant nutrition models have evolved for nearly 55 years to predict requirements and
supply of energy and nutrients based on animal characteristics, diet composition, and
environmental factors. Nutrition models have been developed over time for different purposes,
such as research, diet formulation, and evaluation of management strategies (Illius and Allen,
1994). Recent advances in modeling nutrient utilization in ruminants are reported by Kebreab et
al. (2009). Models vary in complexity and may be classified as empirical or mechanistic as well

5

as static or dynamic; however, the accuracy of each is greatly affected by its ability to model
digestion in and passage from the rumen. Existing mechanistic models have limited prediction
accuracy, particularly across a wide range of conditions (Firkins et al., 1998; Offner and Sauvant,
2004; Tedeschi et al., 2005), and they must be improved, especially with an accurate prediction
of passage rate, before they will be capable of contributing to the improvement of nutrient
utilization (Allen, 2011).
Although discussion of the numerous models available and their specifics are beyond the
scope of this literature review, two major ruminant nutrient models used for diet formulation,
namely the Nutrient Requirements of Dairy Cattle (NRC, 2001) and the Cornell Net
Carbohydrate Protein System (CNCPS; Fox et al., 2004), are briefly described here as they are
referred to later in the review. The seventh edition of the NRC was published 10 years ago and
provides a summary of current reference information for dairy cattle nutrition, which has been
included in a ration evaluation program. The current CNCPS model was designed to formulate
farm specific feeding programs and evaluate nutrient excretion and efficiency of use of nutrients.
Both models contain a stand-alone computer software program.
Limitations of Ruminant Nutrition Models
Major limitations contributing to the inability of models to accurately predict passage
rates of digesta fractions (e.g. starch, indigestible fiber, and digestible fiber) have been outlined
by Allen (2011) and include the lack of available data for passage rates of feed fractions and the
effects of voluntary DMI and dietary characteristics on passage rates of individual fractions.
These limitations and methods to obtain accurate, quantitative information needed to further the
development of models are discussed.

6

MEASUREMENTS OF RUMINAL RATES OF DIGESTION AND PASSAGE
Models of nutrient digestion and utilization in dairy cows need to predict the ruminal
digestibility of specific nutrients. The ruminal digestibility of any homogeneous nutrient fraction
can be calculated by dividing its rate of digestion by its total rate of disappearance from the
rumen, which is the sum of its rate of digestion and rate of passage. Both digestion rate and
passage rate can vary widely among fractions within a single feed ingredient. Because the
various fractions are digested and utilized at different rates and through different mechanisms,
digestibility and digestion rate are measured for feed fractions rather than entire feeds, but data
for passage rates of feed fractions are lacking (Allen, 2011).
Rate of digestion of individual feed fractions is often determined by measuring nutrient
disappearance in vitro or in situ at a number of time points, and data are readily available in the
literature. However, these data are not useful for models because they provide relative
measurements, not absolute measurements of digestion rates of feed fractions, which are required
to predict ruminal digestibility (Allen, 2011). The estimation of rate of passage is done through
an independent procedure based on using external markers applied to intact forages and
concentrates because of the cost and difficulty involved in directly determining passage rate, thus
standard values are often used. Because these passage rates are inaccurate and apply to entire
feeds, they are not useful in models to predict digestibility of fractions within feeds across a wide
range of conditions (Allen, 2011).
Current models, such as NRC and CNCPS, use the digestion rates of individual fractions
but passage rates of entire feeds derived from experiments using external markers to predict
digestibility of feed fractions because passage rate of fractions are not available. This is
problematic because differences in fraction solubility, particle size, and buoyancy result in

7

different passage rates for the various fractions within a feed (Allen, 2011). The use of the same
overall passage rate for all fractions within feeds will overestimate ruminal digestibility of
soluble fractions and small particles with faster rates of passage and will underestimate ruminal
digestibility of large particles with much slower rates of passage (Allen, 2011).
In an attempt to improve model prediction, the NRC (2001) and CNCPS (Seo et al.,
2006) use separate passage rate prediction equations for concentrates and forages, and NRC
made further distinctions in passage rates for dry or wet forages. Mean passage rates of dry
forages, wet forages, and concentrates were 4.5, 5.2, and 6.7%/h, respectively, based on rare
earth markers (Seo et al., 2006, Table 1). These passage rates for forages were twice as high as
the mean passage rate of pdNDF (2.4%/h), and the passage rate for concentrates was less than
one half the mean passage rate for starch (15.3%/h; Table 1) based on 315 records in 11
experiments using the pool and flux method (Voelker Linton, 2006). The similar passage rates
obtained for forages and concentrates when using rare earth markers is likely because markers
migrate from labeled feeds (Teeter et al. 1984; Combs et al., 1992), preferentially bind to small
particles (Erdman and Smith, 1985), and therefore do not accurately represent the passage rate of
the feed originally labeled (Allen, 2011). The pool and flux method is the only method capable
of accurately measuring passage rates of individual feed fractions and is discussed below.
Krizsan et al. (2010) reported the NRC and CNCPS models overestimated ruminal
passage rate (i.e. underestimated retention time of particulate matter in the rumen) based on a
meta-analysis of 172 treatment means from 49 studies that fed a range of forages and used the
pool and flux method. Evaluation of the NRC and CNCPS models with the observed data
resulted in both significant (P < 0.001) mean biases of -2.40 and -1.70%/h and linear biases of 0.59 and -0.53, respectively (Krizsan et al., 2010). The above discrepancies stress the need for

8

accurate data for incorporation into models to improve accuracy of predictions over a wide range
of conditions. Because the digestibility of most nutrients in the rumen is limited by their
retention times, which vary widely across feeding situations, approaches that integrate the
kinetics of digestion with the kinetics of passage are required.
Pool and Flux Method
Although accurate passage rates of feed fractions are not currently available for use in
models, these rates can be accurately measured using the pool and flux method (Robinson et al.,
1987). This method allows the simultaneous measurements of rate of digestion and rate of
passage of individual feed fractions. Passage rate is calculated by dividing duodenal flux of an
individual digesta fraction by its ruminal pool size. Additionally, the total rate of disappearance
(turnover rate) of a digesta fraction from the rumen at steady state can be calculated by dividing
its rate of intake by its ruminal pool size. The digestion rate of the fraction can then be
calculated by subtracting its passage rate from its total rate of disappearance in the rumen.
Because digestion and passage rates of each fraction are calculated simultaneously, these rates
are consistent with both pool size and ruminal digestibility (Allen, 2011). Therefore, the
information is compatible with models that partition feeds into fractions with more uniform rates
of digestion and passage. Data using this method are scarce because it requires duodenally and
ruminally cannulated cows and data collection and sample analysis are labor intensive.
However, the pool and flux method can be used to measure passage rates of feed fractions and
provide the data needed to improve the various models of nutrient availability in ruminants. This
method was used for all research presented in the following chapters of this dissertation.

EFFECTS OF DMI ON PASSAGE RATE

9

Passage rate from the rumen increases with DMI (Riewe and Lippke, 1970). This likely
occurs through effects of increased distention on the rate of reticular contractions (Dado and
Allen, 1995) and on the amplitude and duration of reticular contractions (Okine and Mathison,
1991a). The proportion of large particles in the rumen increases with higher DMI (Okine and
Mathison, 1991b), probably because ruminating time per unit of DM consumed decreases as
DMI increases (Welch and Smith, 1969; Luginbuhl et al., 1989). Additionally, fecal particle size
increased with increasing DMI in several experiments suggesting that increased ruminal fill from
greater DMI probably increases passage rate from the rumen through regulation at the reticularomasal orifice (Allen, 1996).
Diet digestibility generally decreases as a result of the increased passage rate. However,
the effects of feeding level on digestibility are not constant but vary with diet composition. As
feeding level increased, high concentrate diets decreased ruminal retention time more than low
concentrate diets (Colucci et al., 1982, 1990) and concentrate particles had a faster passage rate
than forage particles (Colucci et al., 1990; Wylie et al., 2000). An indirect comparison showed
effects of feeding level on digestibility are greater for corn silage-based diets (Gabel et al., 2003)
than for grass silage-based diets (Volden, 1999). Diets with a high digestibility at maintenance
exhibited greater depressions in digestibility with increased intake (Huhtanen et al., 2009).
Depression in digestibility with increased DMI is greater for higher concentrate diets because the
rapid digestion of concentrates (primarily starch) can lower ruminal pH and subsequently reduce
fiber digestion (NRC, 2001). Additionally, recent reports of the relative decrease in organic
matter digestibility per multiple of maintenance feeding were 2.0% (Huhtanen et al., 2009) and
2.5% (Yan et al., 2002) for grass silage-based diets and 3.2% (Gabel et al., 2003) for primarily

10

corn silage-based diets. These values are smaller than the 4% reported by Tyrrel and Moe
(1975), which was adopted by NRC (2001).
Predictions for passage rate from a mechanic model developed to predict rate of passage
of forage particles in dairy cattle was most sensitive to intake (Seo et al., 2009). This indicates
the importance of including accurate DMI in models, but the inconsistencies on the extent of
depression in digestibility mentioned above in response to increased feeding level are difficult to
account for in nutrition models. Some models have adjusted energy values for the level of
feeding to improve predictions, but Huhtanen et al. (2009) reported the NRC and CNCPS models
overestimated the total digestible nutrient (TDN) discount for dairy cows with DMI greater than
maintenance intake based on a meta-analysis comprised of 497 treatment means from 92 studies
that fed mainly grass silage-based diets. Residual analysis of the NRC and CNCPS models
resulted in both significant (P < 0.001) mean biases of -0.77 and -0.55 kg/d and linear biases of 0.55 and -0.41, respectively (Huhtanen et al., 2009).
Data are lacking for the effect of nutrient intake above 4X maintenance on digestibility
and passage rate. These data are needed because nutrient intake at this level is now common
among dairy cows (Vandehaar, 1998; NRC, 2001), and the linear relationship between DMI and
passage rate should not be extrapolated beyond the limits of the DMI in the existing data (NRC,
2001). In fact, the decline in digestibility with increasing DMI may not be linear but rather
digestibility might decrease at a decreasing rate as DMI increases (Van Soest et al., 1992). Fiber
digestibility is not only affected by forage source and retention time in the rumen but also by the
fermentability of other carbohydrate sources in the ration. Therefore, any measurement of the
effects of diet characteristics on ruminal passage rate and the resulting effects on digestibility
must also account for the interactions between DMI and the effects of diet on passage rate.

11

EFFECTS OF FEED CHARACTERISTICS ON PASSAGE RATE
Rate of passage of individual feed fractions are not constant and are greatly affected by
other dietary and animal factors and their interactions. This is well summarized by Nousiainen et
al. (2009) who stated that digestibility of dairy cow diets is influenced by intrinsic digestion
characteristics of the diet and extrinsic factors that influence the extent to which the intrinsic
digestion potential is achieved. The NRC subcommittee recognized that intrinsic properties of
feeds affect passage rate but concluded “data are too sparse to make adjustments for those
factors” (NRC, 2001). Evaluation of the effects of specific forage and concentrate factors on
ruminal digestibility in studies where only one specific factor varied demonstrated that, in
addition to the effects of feeding level, many other factors had a significant effect on digestibility
(Nousiainen et al., 2009) suggesting the need to include other factors in models for accurate
prediction.
Forages and Rumen Fill
The primary constraints to passage from the rumen vary greatly among forages. Not only
is forage fiber retained in the rumen longer and more filling than other feed components because
of its bulkiness, fibrous forage particles also form a rumen mat capable of sequestration of other
feed particles and increase their retention time. The physical filling effect of diets is related to
forage NDF concentration and the digestion characteristics of forage NDF (Allen, 2000),
including the proportion of iNDF fraction, the rate of digestion of pdNDF fraction, and the
retention time of feed particles in the rumen. Most of the variation in organic matter digestibility
for cows consuming feed above maintenance level was related to changes in the concentration
and digestibility of the NDF fraction in the diet (Nousiainen et al., 2009). Enhanced NDF

12

digestibility within a forage family decreases gut fill and has greater potential effect on DMI as
milk yield increases (Allen, 2000). Neutral detergent fiber from cool-season grasses, although
more digestible than NDF from legumes, is generally more filling and has a greater benefit when
DMI is not limited by rumen distention (Voelker Linton and Allen, 2007).
Rumen fill can limit feed intake, especially for high producing cows and cows fed high
forage diets (Allen et al., 2005). Although most studies report a significant decrease in DMI as
forage NDF increased, the DMI response is variable, depending upon the degree to which intake
is limited by ruminal fill. Dry matter intake of high producing cows (earlier lactation, higher
nutrient demand) is limited by the filling effects of diets to a greater extent than for low
producing cows (later lactation, lower nutrient demand) consuming the same diet (Voelker et al.,
2002). As lactation proceeds and milk yield declines, feed intake is increasingly dominated by
metabolic signals rather than rumen fill (Allen et al., 2009). Highly fermentable diets often
decrease feed intake in mid to late lactation, likely from stimulation of hepatic oxidation by
propionate (Allen et al., 2009). When fill limits intake, passage rate can increase as a mechanism
allowing the animal to increase its intake even though ruminal and total tract digestibility may be
reduced (Jung and Allen, 1995).
Forages and Chewing Activity
Concentrations of NDF and its source are associated with chewing activities, which can
affect ruminal passage. Forage NDF is positively related to total chewing time, which stimulates
the flow of saliva necessary to neutralize fermentation acids and maintain adequate pH levels for
proper rumen function (Allen, 1997). As rumen pH decreases, ruminal motility (Ash and
Dobson, 1963) and rate of fiber digestion (Oba and Allen, 2003) are reduced. Thus, low ruminal
pH exerts negative effects on digestion and passage, rumen fill, and possibly intake.

13

Additionally, rumination contributes to particle size reduction and increases specific gravity of
forages (Welch and Hooper, 1988) increasing their probability of passage.
Because forages have a greater effect on digesta passage rate than most other feeds, the
effect of forage characteristics on ruminal passage rate is the focus of this dissertation and
specific forage characteristics are discussed later.
Experimental Model
The response of dietary characteristics, and the extent to which physical and metabolic
factors affect DMI, are dependent on individual energy balance (Mertens, 1994; Allen, 1996).
Therefore, testing only overall treatment mean differences may mask important responses in
intake, digestion, and production (Allen, 2000). Because cows are now frequently grouped and
fed according to milk yield, models that predict effects of feed intake level on response to diet
are necessary.
An experimental model to evaluate effects of preliminary DMI or milk yield, indicators
of nutrient demand, on animal responses to dietary treatments has been developed and used for
more than a decade (Oba and Allen, 1999a; Burato et al., 2001; Voelker et al., 2002; Harvatine
and Allen, 2002; Bradford and Allen, 2004; Voelker Linton and Allen, 2007, 2008). The design
is a crossover model with a preliminary period utilizing a group of cows with a wide range and
uniform distribution of preliminary DMI or milk yield. A common, intermediate diet is fed
during the preliminary period, which is used to determine preliminary DMI. This experimental
model uses relatively short periods to minimize period effects and decrease the possibility of a
treatment by period interaction while allowing sufficient time for diet adaptation. Post peak
lactation cows only are used in this model because the rapid rise in milk yield and DMI during
the first 50-60 days after calving can cause treatment by period interactions. After peak milk

14

production, a gradual decline in milk yield occurs which is not expected to result in a treatment
by period interaction. This model was used for all research presented in the following chapters
of this dissertation.

FORAGE FACTORS AFFECTING DIGESTA PASSAGE RATES
Diet characteristics related to forage type and to forage harvesting and management
practices that are expected to have the greatest effect on passage rate of digesta fractions are
discussed here. Forage NDF concentration, forage NDF digestibility, forage particle size, forage
maturity, and forage family are the chosen topics for this section; however, many other dietary
factors influence passage rates.
Forage NDF Concentration
Forage NDF concentration is a major factor that affects intake and production of dairy
cows because fiber is the least digestible component in feeds and its bulkiness occupies more
space in the rumen than other feed constituents. The effect of NDF concentration on DMI
depends on the mechanism that is controlling intake. A summarization of 15 studies by Allen
(2000) showed a general decline in DMI with increasing NDF concentrations when diets
exceeded 25% NDF. Neutral detergent fiber has rumen filling characteristics, and high forage
NDF concentration in a diet increases rumen distention (Dado and Allen, 1995), which can limit
voluntary DMI (Allen, 2000). Additionally, Voelker et al. (2002) found high forage diets limit
voluntary DMI to a greater extent as milk yield increases. Because the filling effect of a diet is
highly dependent on the rates of digestion and passage, research including measurements of
ruminal kinetics are necessary to determine the underlying reason for the decrease in DMI
observed when high NDF diets are fed.

15

Two experiments have been conducted to evaluate the effects of forage NDF
concentration on passage rates of feed fractions using the pool and flux method. One study
compared low and high dietary NDF diets (29% and 38% NDF, respectively) by feeding a corn
silage-based diet using either brown midrib or normal corn silage; the low NDF diet increased
the passage rates of pdNDF (3.49 vs. 2.43%/h, P < 0.001) and starch (14.5 vs. 9.00%/h, P <
0.001) but did not affect the passage rate of iNDF compared with the high NDF diet (Oba and
Allen, 2000). Another study compared low and high dietary NDF diets (24% and 31% NDF,
respectively) by feeding a 2:1 mixture of corn silage and alfalfa silage-based diet that varied in
forage:concentrate ratio (45:55 and 61:39, respectively); the low NDF diet increased the passage
rates of iNDF (4.84 vs. 4.36%/h, P < 0.01) and starch (25.4 vs. 18.8%/h, P < 0.01) and tended to
increase the passage rate of pdNDF (1.57 vs. 1.04%/h, P = 0.06) compared with the high NDF
diet (calculated from data reported in Voelker Linton and Allen, 2007). The inconsistent results
on the effects of forage NDF concentration on passage rate of iNDF might be because of
differences in forage sources or NDF concentrations between these experiments. In general, low
NDF diets increased passage rates compared with high NDF diets based on these two studies,
which may allow greater DMI for cows fed low NDF diets than high NDF diets. Additional
research is needed to confirm this generalization.
Forage NDF Digestibility
Digestibility of NDF is an important forage parameter that influences intake and
performance of dairy cows. Because fiber fractions ferment slower and are retained in the rumen
longer, they have a greater effect on ruminal fill than nonfiber fractions. However, ruminal
digestibility of forage NDF is highly variable ranging from less than 35% to over 75% for
various forage types (Nocek and Russell, 1988). Oba and Allen (1999b) compiled treatment

16

means for 13 sets of forage comparisons differing in NDF digestibility (NDFD) from several
experiments reported in the literature. They concluded that enhanced NDFD of forage within
forage family improved DMI and milk yield; a one-unit increase in NDFD in vitro or in situ was
associated with a 0.17 kg increase in DMI and a 0.25 kg increase in 4% fat corrected milk yield.
Measurements of ruminal kinetics are needed to explain the fundamental reason for the increase
in DMI and milk production observed when high NDFD forages are fed.
Effects of forage NDFD on passage rates of feed fractions using the pool and flux method
have been evaluated. In a comparison of brown midrib corn silage and normal corn silage with a
9.4 percentage unit difference in NDF digestibility, the brown midrib corn silage with higher
NDFD increased the passage rates of iNDF (3.64 vs. 3.20%/h, P < 0.001) and starch (12.9 vs.
10.6%/h, P = 0.02) but did not affect passage rate of pdNDF compared to the normal corn silage
with lower NDFD (Oba and Allen, 2000). Overall, high NDFD forage increased passage rates
compared with the low NDFD forage based on this single study. The faster passage rates,
especially for iNDF, might permit greater DMI for cows fed high NDFD forage compared with
low NDFD diet; however, more research is required to verify this statement. Although it is
difficult to conduct experiments evaluating forage NDFD that are not confounded by differences
in fiber source or forage:concentrate ratio, this research is needed.
Forage Particle Size
Sufficient amounts of forage in both chemical and physical forms are necessary for
proper rumen function in dairy cows. The current NRC (2001) guidelines provide
recommendations for minimum chemical concentrations of total NDF and forage NDF but do
not account for physical form or particle size of feeds. Forage fiber, in a form that is physically
effective, is necessary in dairy cow diets to promote rumen health and cow performance.

17

Insufficient particle size may reduce rumen pH, impair fiber digestion, and increase incidence of
health disorders including ruminal acidosis and displaced abomasum. Increasing forage particle
size increases ruminal retention time (Dixon and Milligan, 1985) and promotes formation of the
rumen mat (Grant, 1997). Although impaired rumen fermentation and function can result when
cattle are fed rations lacking in physical structure, excessive amounts of long, coarse fiber may
limit intake (Allen, 2000).
Effects of particle size were evaluated for two forages including a legume (alfalfa) and a
grass (orchardgrass) because of differences in anatomical structure and digestion characteristics
affecting particle size reduction and passage (Allen, 1996) and are reported in Chapters 2 and 3
of this dissertation. Greater ruminal distention caused by longer forage particles is more likely to
affect passage rate and DMI when feed intake is more limited by fill.
Forage Maturity
Rates of digestion, passage, and particle size reduction in the rumen are influenced by the
fiber content of forage and the potential digestibility of the fiber (Mertens, 1993), which change
during the development and growth of the plant. Because fiber has been related to the filling
properties of feeds and forage fiber increases but its digestibility decreases as plants mature, the
importance of physical rumen fill in limiting feed intake increases when feeding dairy cows a
more mature forage compared to a less mature forage. Previous research showed highly lignified
forages (i.e. more mature) remained in the rumen longer due to its slow rate of digestion, and
possibly increased buoyancy, and resulted in lower DMI (Jung and Allen, 1995; Allen, 2000).
Effects of forage maturity on passage rates of feed fractions using the pool and flux
method have been evaluated previously. In a comparison of grass silages harvested at four
stages (1 week intervals), the passage rate of pdNDF tended to increase linearly and

18

quadratically (P < 0.10) and the passage rate of iNDF increased linearly (P < 0.01) as maturity
increased (Rinne et al., 2002). According to Poppi et al. (1981) and Ulyatt (1983), increased
maturity of grasses at harvest increased the rate of particle size reduction by chewing because of
greater fragility. Thus, the increasing passage rate as maturity increased noted here might be
because of greater fragility due to increased lignification.
Effects of maturity were evaluated for two forages including a legume (alfalfa) and a
grass (orchardgrass) because of differences in anatomical structure and digestion characteristics
affecting particle size reduction and passage (Allen, 1996) and are reported in Chapters 4 and 5
of this dissertation. Greater ruminal distention caused by increased maturity is more likely to
affect passage rate and DMI when feed intake is more limited by fill.
Forage Family
Fiber particles from different forage families (legumes and grasses) have different
chemical compositions, anatomical characteristics, and digestion characteristics that affect the
rate and extent of digestion (Allen, 1996; Wilson and Kennedy, 1996). Cows fed grass-based
diets had lower DMI and milk yield than cows fed legume-based diets despite greater NDF
digestibility for grass (Oba and Allen, 1999b). Voluntary DMI is probably more limited for
grass forage because of its filling effect caused by slow particle breakdown or slow passage rate.
Grass NDF is digested more slowly than alfalfa NDF, and its cell walls are more resistant to
particle breakdown than are alfalfa cell walls (Wilson and Hatfield, 1997). Because grass NDF
is more digestible but also more slowly digested, digestion is extended over a greater period of
time for grass particles, increasing their buoyancy over time compared to alfalfa particles (Allen,
1996). Because passage rate increases as density increases and as particle size decreases, alfalfa

19

particles probably escape the rumen more quickly than grass particles possibly allowing for
greater DMI.
Evaluation of the effects of forage family on passage rates of fiber fractions, measured
using the pool and flux method, yielded opposing results for relative passage rates of iNDF and
pdNDF. An experiment comparing alfalfa silage to orchardgrass silage-based diets formulated
to similar forage NDF concentrations reported alfalfa tended to increase iNDF passage rate
compared to orchardgrass (2.9 vs. 2.4%/h, P = 0.06) but there was no difference in pdNDF
passage rate between the two species (Voelker Linton and Allen, 2008). In a comparison of
alfalfa silage and timothy silage-based diets, alfalfa increased pdNDF passage rate compared
with timothy (2.4 vs. 1.8%/h, P = 0.005) but iNDF passage rates were similar (Stensig and
Robinson, 1997). The reason for this discrepancy in passage rates is not apparent. Effects of
forage family on passage rate were evaluated further and are reported in Chapter 6 of this
dissertation.

RESEARCH OBJECTIVES AND HYPOTHESIS
The research reported in this dissertation was designed to determine the effect of
voluntary DMI on ruminal passage rates of digesta fractions and how it is altered by forage
characteristics. Five experiments were conducted using the pool and flux method and
experimental model described in the literature review. The specific objective was to determine
the effects of voluntary DMI on the response of passage rate of individual feed fractions to
legume particle size (Chapter 2), grass particle size (Chapter 3), legume maturity (Chapter 4),
grass maturity (Chapter 5), and forage family (legume vs. grass; Chapter 6). An additional
objective of the forage family experiment was to the evaluate the relationships between

20

voluntary DMI and effects of forage family on rates of particle size reduction in, and particle
passage from, the rumen (Chapter 7). Each experiment allowed the effects of the interaction
between the treatment (forage characteristic) and DMI to be evaluated, direct comparison of
treatment effects, and measurement of ruminal passage rates for individual feed fractions. We
hypothesized that the normal variation in diet characteristics related to forages, including forage
particle size, forage maturity, and forage family, alters both the passage rates of feed fractions
and the extent to which passage rates of these fractions are affected by voluntary DMI.

21

APPENDIX

22

Table 1. Rate of passage of feeds determined by excretion pattern of rare earth markers in feces
or of feed fractions by the pool and flux method
Mean
Range
1
Rare earth markers
Dry forage
4.5
3.4 – 5.7
Wet forage
5.2
3.9 – 6.3
Concentrate
6.7
3.6 – 9.2
2
Pool and flux
Indigestible NDF
3.2
1.2 – 5.3
Potentially digestible NDF
2.4
0.2 – 4.3
Starch
15.3
3.4 – 33.9
3
Pool and flux
Indigestible NDF
2.6
1.1 – 5.1
1

Seo et al., 2006; 319, 63, and 139 treatment means for dry forage, wet forage, and concentrate,
respectively, from 275 published experiments.
2
Voelker Linton, 2006; 315 cow periods from 11 experiments conducted in Allen laboratory at
Michigan State University. Pool and flux calculations based on rumen pool size of indigestible
NDF (iNDF) and iNDF flux to duodenum.
3
Krizsan et al., 2010; 172 treatment means from 49 published experiments. Pool and flux
calculations based on rumen pool size of iNDF and a combination of flux of iNDF from intake or
fecal output.

23

Figure 1. Simple first-order model of digestion and passage in the rumen for a feed fraction.
The total rate of escape is the sum of the rate of digestion (kd) and the rate of passage (kp). The
fraction digested (D) is calculated by dividing the rate of digestion by the total rate of escape (D
= kd/(kd+kp)); the fraction passed (P) is calculated by dividing the rate of passage by the total
rate of escape (P = kp/(kd+kp)). This is the basic subunit of many digestion and passage models,
and it is applied to multiple feed fractions with similar rates of digestion and passage.

24

REFERENCES

25

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Allen, M. S. 1997. Relationship between fermentation acid production in the rumen and the
requirement for physically effective fiber. J. Dairy Sci. 80:1447-1462.
Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cattle.
J. Dairy Sci. 83:1598-1624.
Allen, M. S. 2011. Mind over Models. Pages 29-44 in Proceedings of the Tri-State Dairy
Nutr. Conf. Fort Wayne, IN.
Allen, M. S., B. J. Bradford, and K. J. Harvatine. 2005. The cow as a model to study food intake
regulation. Annu. Rev. Nutr. 25:523-547.
Allen, M. S., B. J. Bradford, and M. Oba. 2009. The hepatic oxidation theory of the control of
feed intake and its application to ruminants. J. Anim. Sci. 87:3317-3334.
Allen, M. S., and D. R. Mertens. 1988. Evaluating constraints on fiber digestion by rumen
microbes. J. Nutr. 118:261-270.
Ash, R. W., and A. Dobson. 1963. The effect of absorption on the acidity of rumen contents. J.
Physiol. 169:39-61.
Bradford, B. J., and M. S. Allen. 2004. Milk fat responses to a change in diet fermentability vary
by production level in dairy cattle. J. Dairy Sci. 87:3800-3807.
Burato, G. M., J. A. Voelker, and M. S. Allen. 2001. Effects of pretrial milk yield on feed intake,
production, and feeding behavior responses to forage particle size by lactating cows. J.
Dairy Sci. 84(Suppl. 1):199.
Cardoza, R. S., and D. R. Mertens. 1986. Effect of fiber source and content on threshold size for
passage and fecal particle distribution. J. Dairy Sci. 69 (Suppl. 1):134.
Chai, K., P. M. Kennedy, and L. P. Milligan. 1984. Reduction in particle size during rumination
in cattle. Can. J. Anim. Sci. 64 (Suppl.):339.
Colucci, P. E., L. E. Chase, and P. J. Van Soest. 1982. Feed intake, apparent diet digestibility,
and rate of particulate passage in dairy cattle. J. Dairy Sci. 65:1445–1456.
Colucci, P. E., G. K. MacLeod, W. L. Grovum, I. McMillan, and D. J. Barney. 1990. Digesta
kinetics in sheep and cattle fed diets with different forage to concentrate ratios at high
and low intakes. J. Dairy Sci. 73:2143–2156.

26

Combs, D. K., R. D. Shaver, and L. D. Satter. 1992. Retention of rare earths by hay particles
following incubation in fresh or autoclaved rumen fluid. J. Dairy Sci. 75:132-139.
Dado, R. G., and M. S. Allen. 1995. Intake limitations, feeding behavior, and rumen function of
cows challenged with rumen fill from dietary fiber or inert bulk. J. Dairy Sci. 78:118-133.
Dixon, R. M., and L. P. Milligan. 1985. Removal of digesta components from the rumen of
steers determined by sieving techniques and fluid, particulate and microbial markers. Br.
J. Nutr. 53:347-362.
Erdman, R. A., and L. W. Smith. 1985. Ytterbium binding among particle size fractions of
forage cell walls. J. Dairy Sci. 68:3071-3075.
Firkins, J. L., M. S. Allen, B. S. Oldick, and N. R. St-Pierre. 1998. Modeling ruminal
digestibility of carbohydrates and microbial protein flow to the duodenum. J. Dairy Sci.
81:3350-3369.
Fox, D. G., L. O. Tedeschi, T. P. Tylutki, J. B. Russell, M. E. Van Amburgh, L. E. Chase, A. N.
Pell, and T. R Overton. 2004. The Cornell Net Carbohydrate and Protein System model
for evaluating herd nutrition and nutrient excretion. Anim. Feed Sci. Technol. 112:29-78.
Gabel, M., B. Pieper, K. Friedel, M. Radke, A. Hagemann, J. Voigt, and S. Kuhla. 2003.
Influence of nutrition level on digestibility in high yielding cows and effects on energy
evaluation systems. J. Dairy Sci. 86:3992–3998.
Grant, R. J. 1997. Interactions among forages and nonforage fiber sources. J. Dairy Sci. 80:14381446.
Harvatine, K. J., and M. S. Allen. 2002. Saturation effects of rumen-inert fat sources on feed
intake, milk production, and feeding behavior in lactating cows varying in milk yield. J.
Dairy Sci. 85(Suppl. 1):141-142.
Hooper, A. P., and J. G. Welch. 1985. Effects of particle size and forage composition on
functional specific gravity. J. Dairy Sci. 68:1181-1188.
Huhtanen, P., U. Asikainen, M. Arkkila, and S. Jaakkola. 2007. Cell wall digestion and passage
kinetics estimated by marker and in situ methods or by rumen evacuations in cattle fed
hay 2 or 18 times daily. Anim. Feed Sci. Technol. 133:206-227.
Huhtanen, P., S. Jaakkola, and U. Kukkonen. 1995. Ruminal plant cell wall digestibility
estimated from digestion and passage kinetics utilizing mathematical models. Anim. Feed
Sci. Technol. 52:159-173.
Huhtanen, P., M. Rinne, and J. Nousiainen. 2009. A meta-analysis of feed digestion in dairy
cows. 2. The effects of feeding level and diet composition on digestibility. J. Dairy Sci.
92:5031-5042.

27

Illius, A. W., and M. S. Allen. 1994. Assessing forage quality using integrated models of intake
and digestion by ruminants. Pages 869-890 in Forage Quality, Evaluation, and
Utilization. G. C. Fahey, Jr., M. Collins, D. R. Mertens, and L. E. Moser, ed. American
Society of Agronomy, Crop Science Society of America, and Soil Science Society of
America, Madison, WI.
Jung, H. G., and M. S. Allen. 1995. Characteristics of plant cell walls affecting intake and
digestibility of forages by ruminants. J. Anim. Sci. 73:2774-2790.
Kebreab, E., J. Dijkstra, A. Bannink, and J. France. 2009. Recent advances in modeling nutrient
utilization in ruminants. J. Anim. Sci. 87(E. Suppl.):E111-E122.
Kennedy, P. M. 1985. Effect of rumination on reduction of particle size of rumen digesta by
cattle. Aust. J. Agric. Res. 36:819-828.
Krizsan, S. J., S. Ahvenjärvi, and P. Huhtanen. 2010. A meta-analysis of passage rate estimated
by rumen evacuation with cattle and evaluation of passage rate prediction models. J.
Dairy Sci. 93:5890-5901.
Lechner-Doll, M., M. Kaske, and W. Von Englehardt. 1991. Factors affecting the mean retention
time of particles in the forestomachs of ruminants and camelids. In: T. Tsuda, Y. Sasaki,
and R. Kawashima, ed. Physiological Aspects of Digestion and Metabolism in
Ruminants. Academic Press, New York.
Luginbuhl, J.-M., K. R. Pond, J. C. Burns, and J. C. Russ. 1989. Eating and ruminating behavior
of steers fed coastal bermudagrass hay at four levels. J. Anim. Sci. 67:3410-3418.
Lund, P., M. R. Weisbjerg, and T. Hvelplund. 2007. Digestible NDF is selectively retained in the
rumen of dairy cows compared to indigestible NDF. Anim. Feed Sci. Technol. 134:1-17.
McLeod, M. N., and D. J. Minson. 1988. Large particle breakdown by cattle eating ryegrass and
alfalfa. J. Anim. Sci. 66:992-999.
Mertens, D. R. 1993. Rate and extent of digestion. Pages 13-51 in Quantitative Aspects of
Ruminant Digestion and Metabolism. M. M. Forbes and J. France, ed. CAB Int.,
Wallingford, United Kingdom.
Mertens, D. R. 1994. Regulation of forage intake. Pages 450– 493 in Forage Quality, Evaluation,
and Utilization. G. C. Fahey, Jr, M. Collins, D. R. Mertens, and L. E. Moser, ed.
American Society of Agronomy, Crop Science Society of America, and Soil Science
Society of America, Madison, WI.
Murphy, M. R., and J. M. Nicoletti. 1984. Potential reduction of forage and rumen digesta
particle size by microbial action. J. Dairy Sci. 67:1221-1226.

28

National Research Council. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl.
Acad. Washington, DC.
Nocek, J. E., and R. A. Kohn. 1987. Initial particle form and size on change in functional
specific gravity of alfalfa and timothy hay. J. Dairy Sci. 70:1850-1863.
Nocek, J. E., and J. B. Russell. 1988. Protein and energy as an integrated system. Relationship of
ruminal protein and carbohydrate availability to microbial synthesis and milk production. J.
Dairy Sci. 71:2070–2107.
Nousiainen, J., M. Rinne, and P. Huhtanen. 2009. A meta-analysis of feed digestion in dairy
cows. 1. The effects of forage and concentrate factors on total diet digestibility. J. Dairy
Sci. 92: 5019-5030.
Oba, M., and M. S. Allen. 1999a. Effects of brown midrib 3 mutation in corn silage on dry
matter intake and productivity of high yielding dairy cows. J. Dairy Sci. 82:135-142.
Oba, M., and M. S. Allen. 1999b. Evaluation of the importance of the digestibility of neutral
detergent fiber from forage: effects on dry matter intake and milk yield of dairy cows. J.
Dairy Sci. 82:589-596.
Oba, M., and M. S. Allen. 2000. Effects of brown midrib 3 mutation in corn silage on
productivity of dairy cows fed two concentrations of dietary neutral detergent fiber: 3.
Digestibility and microbial efficiency. J. Dairy Sci. 83:1350-1358.
Oba, M., and M. S. Allen. 2003. Effects of corn grain conservation method on ruminal digestion
kinetics for lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:184-194.
Offner, A., and D. Sauvant. 2004. Comparative evaluation of Molly, CNCPS, and LES rumen
models. Anim. Feed Sci. Technol. 112:107-130
Okine, E. K., and G. W. Mathison. 1991a. Reticular contraction attributes and passage of digesta
from the ruminoreticulum in cattle fed roughage diets. J. Anim. Sci. 69:2177-2186.
Okine, E. K., and G. W. Mathison. 1991b. Effects of feed intake on particle distribution, passage
of digesta, and extent of digestion in the gastrointestinal tract of cattle. J. Anim. Sci.
69:3435-3445.
Poppi, D. P., W. C. Ellis, J. H. Matis, and C. E. Lascano. 2001. Marker concentration patterns of
labeled leaf and stem particles in the rumen of cattle grazing Bermuda grass (Cynodon
dactylon) analysed by reference to a raft model. Br. J. Nutr. 85:553-563.
Poppi, D. P., D. J. Minson, and J. H. Ternouth. 1981. Studies of cattle and sheep eating leaf and
stem fractions of grasses. III. The retention time in the rumen of large feed particles.
Aust. J. Agric. Res. 32:123-137.

29

Poppi, D. P., B. W. Norton, D. J. Minson, and R. E. Hendricksen. 1980. The validity of the
critical size theory for particles leaving the rumen. J. Agric. Sci. 94:275-280.
Riewe, M. E., and H. Lippke. 1970. Considerations in determining the digestibility of harvested
forages. Page F1 in Proc. Nat. Conf. on Forage Quality Evaluation and Utilization. Univ.
Nebraska. Lincoln.
Rinne, M., P. Huhtanen, and S. Jaakkola. 2002. Digestive processes of dairy cows fed silages
harvested at four stages of grass maturity. J. Anim. Sci. 80:1986-1998.
Robinson, P. J., S. Tamminga, A. M. Van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci. 17:37-62.
Seo, S., C. Lanzas, L. O. Tedeschi, A. N. Pell, and D. G. Fox. 2009. Development of a
mechanistic model to represent the dynamics of particle flow out of the rumen and to
predict rate of passage of forage particles in dairy cattle. J. Dairy Sci. 92:3981-4000.
Seo, S., L. O. Tedeschi, C. G. Schwab, B. D. Garthwaite, and D. G. Fox. 2006. Evaluation of the
passage rate equations in the 2001 dairy NRC model. J. Dairy Sci. 89:2327–2342.
Siciliano-Jones, J., and M. R. Murphy. 1991. Specific gravity of various feedstuffs as affected by
particle size and in vitro fermentation. J. Dairy Sci. 74:896-901.
Stensig, T., and P. H. Robinson. 1997. Digestion and passage kinetics of forage fiber in dairy
cows as affected by fiber-free concentrate in the diet. J. Dairy Sci. 80:1339-1352.
Sutherland, T. M. 1988. Particle separation in the forestomachs of sheep. In: Aspects of
Digestive Physiology in Ruminants. A. Dobson and M. J. Dobson, ed. Cornell Univ.
Press, Ithaca, NY.
Tamminga, S., P. H. Robinson, H. Vogt, and H. Boer. 1989. Rumen ingesta kinetics of cell wall
components in dairy cows. Anim. Feed Sci. Technol. 25:89-98.
Tedeschi, L. O., D. G. Fox, R. D. Sainz, L. G. Barioni, S. R. de Medeiros, and C. Boin. 2005.
Mathematical models in ruminant nutrition. Sci. Agric. 62:76-91.
Teeter, R. G, F. N. Owens, and T. L. Mader. 1984. Ytterbium chloride as a marker for particulate
matter in the rumen. J. Anim. Sci. 58:464-473.
Tyrrel, H. F., and P. W. Moe. 1975. Effect of intake on digestive efficiency. J. Dairy Sci.
58:1151–1163.
Ulyatt, M. J. 1983. Plant fibre and regulation of digestion in the ruminant. Page 103 in Fibre in

30

Human and Animal Nutrition. G. Wallace and L. Bell ed. The Royal Soc. Of New
Zealand, Wellington, N.Z.
Van Soest, P. J., M. B. Rymph, and D. G. Fox. 1992. Discounts for net energy and protein – fifth
edition. Pages 40-68 in Proc. Cornell Nutr. Conf., Ithaca, NY.
Van Soest, P. J., C. J. Sniffen, and M. S. Allen. 1988. Rumen dynamics. Pages 21-42 in Aspects
of Digestive Physiology in Ruminants. A. Dobson and M. J. Dobson, ed. Cornell Univ.
Press, Ithaca, NY.
Vandehaar, M. J. 1998. Efficiency of nutrient use and relationship to profitability of dairy farms.
J. Dairy Sci. 81:272–282.
Voelker, J. A., G. M. Burato, and M. S. Allen. 2002. Effects of pretrial milk yield on responses
of feed intake, digestion, and production to dietary forage concentration. J. Dairy Sci.
85:2650-2661.
Voelker Linton, J. A. 2006. Effects of dietary forage characteristics on digesta passage rate in
dairy cows. Ph.D. dissertation, Department of Animal Science, Michigan State
University.
Voelker Linton, J. A., and M. S. Allen. 2007. Nutrient demand affects ruminal digestion
responses to a change in dietary forage concentration. J. Dairy Sci. 90:4770-4779.
Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Volden, H. 1999. Effects of level of feeding and ruminally undegraded protein on ruminal
bacterial protein synthesis, escape of dietary protein, intestinal amino acid profile, and
performance of dairy cows. J. Anim. Sci. 77:1905–1918.
Waldo, D. R., L. W. Smith, and E. L. Cox. 1972. Model of cellulose disappearance from the
rumen. J. Dairy Sci. 55:125-129.
Welch, J. G., and A. P. Hooper. 1988. Ingestion of feed and water. Pages 108-124 in The
Ruminant Animal: Digestive Physiology and Nutrition. D.C. Church, ed. Prentice-Hall,
Englewood Cliffs, NJ.
Welch, J. G., and A. M. Smith. 1969. Effect of varying amounts of forage intake on rumination.
J. Anim. Sci. 28:827-830.
Wilson, J. R., D. E. Akin, M. N. McLeod, and D. J. Minson. 1989. Particle size reduction of the
leaves of a tropical and temperate grass by cattle. II. Relation of anatomical structure to
the process of leaf breakdown through chewing and digestion. Grass Forage Sci. 44:6575.

31

Wilson, J. R., and R. D. Hatfield. 1997. Structural and chemical changes of cell wall types during
stem development: consequences for fibre degradation by rumen microflora. Aust. J.
Agric. Res. 48:165-180.
Wilson, J. R., and P. M. Kennedy. 1996. Plant and animal constraints to voluntary feed intake
associated with fibre characteristics and particle breakdown and passage in ruminants.
Aust. J. Agric. Res. 47:199-225.
Wylie, M. J., W. C. Ellis, J. H. Matis, E. M. Bailey, W. D. James, and D. E. Beever. 2000. The
flow of forage particles and solutes through segments of the digestive tract of cattle. Br. J.
Nutr. 83:295–306.
Yan, T., R. E. Agnew, and F. J. Gordon. 2002. The combined effects of animal species (sheep
versus cattle) and level of feeding on digestible and metabolizable energy concentrations
in grass silage based diets of cattle. Anim. Sci. 75:141–151.

32

CHAPTER 2
Nutrient Demand Interacts with Legume Particle Length to Affect Digestion Responses
and Rumen Pool Sizes in Dairy Cows

1

ABSTRACT
Effects of legume particle length on dry matter intake (DMI), milk production, ruminal
fermentation and pool sizes, and digestion and passage kinetics, and the relationship of these
effects with preliminary DMI (pDMI) were evaluated using 13 ruminally and duodenally
cannulated Holstein cows in a crossover design with a 14-d preliminary period and two 19-d
treatment periods. During the preliminary period, pDMI of individual cows ranged from 22.8 to
32.4 kg/d (mean = 26.5 kg/d) and 3.5% fat-corrected milk yield ranged from 22.9 to 62.4 kg/d
(mean = 35.1 kg/d). Experimental treatments were diets containing alfalfa silage chopped to
either a) 19 mm (LONG) or b) 10 mm (SHORT) theoretical length of cut as the sole forage.
Alfalfa silages contained ~43% neutral detergent fiber (NDF); diets contained ~47% forage and
~20% forage NDF. Preliminary DMI, an index of nutrient demand, was determined during the
last 4 d of the preliminary period when cows were fed a common diet and used as a covariate.
Main effects of legume particle length and their interaction with pDMI were tested by ANOVA.
Alfalfa particle length and its interaction with pDMI did not affect milk yield or rumen pH.
LONG decreased milk fat concentration more per kilogram pDMI increase than SHORT and
increased yields of milk fat and fat-corrected milk less per kilogram pDMI increase than
SHORT, resulting in a greater benefit for LONG at low pDMI and for SHORT at high pDMI.
LONG tended to decrease DMI compared to SHORT. Ruminal digestion and passage rates of
1

Accepted for publication in the Journal of Dairy Science on December 4, 2011.

33

feed fractions did not differ between LONG and SHORT and were not related to level of intake.
LONG tended to decrease the rate of ruminal turnover for NDF but increased NDF rumen pools
at a slower rate than SHORT as pDMI increased. This indicated the faster NDF turnover rate did
not counterbalance the higher DMI for SHORT resulting in higher NDF rumen pools for SHORT
than LONG. As pDMI increased, LONG increased ruminal digestibility of potentially digestible
NDF and total NDF and SHORT decreased them, but total tract digestibilities of potentially
digestible NDF, total NDF, organic matter and dry matter were lower for LONG than SHORT.
Ruminal digestibilities of starch and organic matter interacted quadratically with level of intake.
When legume silage was the only source of forage in the diet, increasing chop length from 10 to
19 mm tended to decrease DMI but did not negatively affect productivity of cows.

INTRODUCTION
Optimal utilization of diets by dairy cows is influenced by the chemical composition and
physical characteristics of feeds. Forage fiber, in a form that is physically effective, is necessary
in dairy cow diets to promote rumen fermentation and function (Allen, 1997). Increasing forage
particle size has been shown to increase chewing activity resulting in increased saliva flow,
rumen pH, acetate-to-propionate ratio and milk fat concentration (Nørgaard 1983; Beauchemin et
al., 1997), increase ruminal retention time (Dixon and Milligan, 1985), and promote formation of
the rumen mat (Grant, 1997). Although impaired rumen function and health can result when
cattle are fed rations lacking in physical structure, excessive amounts of long, coarse fiber may
decrease ruminal digesta passage rates and limit feed intake of lactating dairy cows when feed
intake is limited by rumen fill (Allen, 2000).
Forage particle length (FPL) has been widely researched, but results of animal responses

34

to FPL are inconsistent and inconclusive. These inconsistencies are likely due to large variation
in dietary factors among studies, which make direct comparisons across studies difficult. Wide
ranges of FPL (2 to 32 mm; Tafaj et al., 2007) and differences between FPL compared within
studies (6 vs. 8 mm; Yang and Beauchemin, 2004 and 24 vs. 170 mm; Randby et al., 2008) have
been used to evaluate FPL. Additionally, studies often evaluate effect of FPL in combination
with other dietary factors including forage:concentrate ratio (Soita et al., 2000; Einarson et al.,
2004), grain processing (Yang et al., 2001), grain fermentability (Krause and Combs, 2003),
non-forage fiber sources (Mooney and Allen, 1997), and supplemental fat (Onetti et al., 2003).
Furthermore, responses to FPL vary depending on preservation methods (hay, silage) and forage
source with greater differences reported for legume and grass-based TMR compared to corn
silage-based TMR (Tafaj et al., 2007). This suggests consideration of forage family when
studying the effects of particle size is necessary. Alfalfa (AL; Medicago sativa) was selected as
a representative legume for use in this experiment because it is the predominant legume fed to
dairy cows in the United States.
Besides dietary factors, inconsistent responses to FPL may be related to animal factors.
Numerous studies have examined the effects of alfalfa FPL, but most were designed using cows
at a specific stage of lactation such as early lactation (Kononoff and Heinrichs, 2003) or mid
lactation (Krause and Combs, 2003). However, cows respond differently to treatments
depending on their level of intake (Voelker et al., 2002; Voelker Linton and Allen, 2008) and
production (Oba and Allen, 1999). Because FPL and level of intake affect ruminal passage and
digestion rates and thus digesta fill in the rumen, the response to effects of particle size and its
relationship with intake level should be assessed to determine if responses to treatment vary
among cows with a wide range in DMI. We hypothesized that responses of DMI and passage

35

rates to legume particle length are related to level of intake and shorter particle length will permit
a greater increase in passage rate than longer particle length as feed intake increases.
The objectives of this experiment were to evaluate the relationships between voluntary
DMI and effects of length of cut of legume silage on DMI, milk production, ruminal
fermentation and pool sizes, and digestion and passage kinetics in lactating dairy cows. This
study had three distinctive features to improve our understanding of the role of particle size and
interpret its effect on animal responses. First, it allowed effects of the interaction between FPL
and preliminary DMI (pDMI) to be evaluated. The use of pDMI, an index of nutrient demand,
allowed the evaluation of treatments on animal responses in relation to level of intake and
provided an indicator to test effects of intake level independent of treatments. Second, it directly
compared treatment effects of long and short cut legume as the sole source of forage without the
confounding effects of other dietary factors. Third, ruminal passage rates of individual feed
fractions, instead of entire feeds, were measured using ruminally and duodenally cannulated
cows.

MATERIALS AND METHODS
Cows and Treatments
Experimental procedures were approved by the Institutional Animal Care and Use
Committee at Michigan State University. Thirteen multiparous Holstein cows from the
Michigan State University Dairy Cattle Teaching and Research Center were assigned randomly
to treatment sequence in a crossover design experiment with one 14-d preliminary period and
two 19-d experimental periods. During the preliminary period, the first 10 d were allowed for
diet adaptation and samples were collected during the final 4 d. During each experimental

36

period, the first 12 d were allowed for diet adaptation and samples were collected during the final
7 d. Cows were 177±66 (mean±SD) DIM at the end of the preliminary period and were selected
to provide a wide range and uniform distribution of pDMI and milk yield. During the final 4 d of
the 14-d preliminary period, the average pDMI among cows ranged from 22.8 to 32.4 kg/d
(mean = 26.5 kg/d) and 3.5% FCM yield ranged from 22.9 to 62.4 kg/d (mean = 35.1 kg/d; Table
2). Prior to calving, cows were cannulated ruminally (Bar Diamond, Inc., Parma, ID) and
duodenally with a gutter-type T cannula placed approximately 10 cm distal to the pylorus (Joy et
al., 1997). Surgery was performed at the Department of Large Animal Clinical Science, College
of Veterinary Medicine, Michigan State University.
Experimental treatments were diets containing AL silage chopped to either a) 19 mm
(LONG) or b) 10 mm (SHORT) theoretical length of cut (TLC) as the sole forage. These TLC
were selected to provide a wide interval within the normal range of TLC to examine if animal
response to FPL is affected by level of feed intake.
Alfalfa was produced at the campus farm at Michigan State University (East Lansing),
harvested from the same field using a New Holland FP230 pull-type forage harvester set
according to manufacturer specifications for theoretical lengths of cut of 19 mm and 10 mm for
long and short cut AL, respectively, and ensiled in Ag-Bags (Ag-Bag Systems Inc., St. Nazianz,
WI). During the sample collection periods, long and short cut AL contained ~43% NDF (DM
basis; Table 3). Diets LONG and SHORT were formulated to contain 21% forage NDF and 18%
CP. The diet fed during the preliminary period was formulated so that long and short cut AL
each contributed 50% of forage NDF. Diets also contained dry ground corn, soybean meal (48%
®

CP), SoyPlus (West Central Soy, Ralston, IA), vitamin-mineral premix, limestone, and salt
(Table 4).

37

Data and Sample Collection
Throughout the experiment, cows were housed in tie-stalls and fed diets as total mixed
rations once daily (1130 h) at 110% of expected intake. The amount of feed offered and refused
(orts) was weighed daily for each cow. Forage samples were collected twice weekly and
analyzed to adjust diets to account for DM, NDF, and CP fluctuation. Samples of all dietary
ingredients (0.5 kg) and orts (12.5%) were collected daily from d 11 to 14 during the preliminary
period and d 13 to 17 during each experimental period. Samples were frozen immediately after
collection at !20°C and combined to one composite sample per period prior to analysis.
Cows were moved to an exercise lot twice daily (0230 and 1300 h) prior to milking in a
parlor (0400 and 1430 h). Milk yield was measured, and milk was sampled, at each milking on d
11 to 14 of the preliminary period and on d 13 to 17 of the experimental periods. Rumen-empty
BW was measured by weighing the cow after evacuation of ruminal digesta on d 14 of the
preliminary period and d 19 of each experimental period. Body condition score was determined
on the same days by 4 trained investigators blinded to treatments (Wildman et al., 1982; 5-point
scale where 1 = thin and 5 = fat).
Duodenal samples (900 mL), fecal samples (500 g), rumen fluid and particulate samples
for microbial isolation (400 g), and rumen fluid samples for pH, concentrations of VFA, lactate,
and ammonia (100 mL) were collected every 15 h from d 13 to 17 of each experimental period
so that 8 samples were taken for each cow in each period, representing every 3 h of a 24-h period
to account for diurnal variation. Rumen fluid and particulate matter for microbial isolation was
collected from the reticulum, near the reticular-omasal orifice, transported to the laboratory, and
processed. Rumen fluid for pH, VFA, lactate, and ammonia were obtained by combining digesta
from five different sites in the rumen and straining it through nylon mesh (~1 mm pore size);

38

fluid pH was recorded immediately. Samples were stored at -20°C.
Ruminal contents were evacuated manually through the ruminal cannula at 4.5 h after
feeding at the beginning of d 18 (1600 h) and 2 h before feeding at the end of d 19 (0930 h) for
each experimental period. Total rumen content mass and volume were determined. To ensure
accurate sampling, every tenth handful of digesta (10%) was separated for a subsample
throughout evacuation. This subsample was squeezed into primarily solid and liquid phases.
Both phases were weighed and sampled (350 mL) for determination of nutrient pool size. All
samples were stored at -20°C.
Sample Analysis and Calculations
Milk yield recorded at both milkings were summed for a daily total, which were averaged
for each period. Milk samples were analyzed for fat, true protein, lactose, and SNF with infrared
spectroscopy by Michigan DHIA (East Lansing). Yields of 3.5% FCM and milk components
were calculated using milk yield and component concentrations for each milking, summed for a
daily total, and averaged for each period.
Forage samples were combined to one composite sample per forage per period. Particle
size distribution was determined using the Penn State Particle Separator containing 2 sieves (19
and 8 mm) and a pan (Lammers et al., 1996). In addition, samples were wet sieved manually and
sequentially through screens with the following aperture sizes: 19.0, 9.50, 4.75, 2.36, 1.18,
0.600, 0.300, 0.150, 0.075, and 0.038 mm. The fraction of DM retained on the screens from wet
sieving was used to calculate mean particle size.
Diet ingredients, orts, and feces were lyophilized (Tri-Philizer MP, FTS Systems, Stone
Ridge, NY). All dried samples were ground with a Wiley mill (1-mm screen; Arthur H. Thomas,
Philadelphia, PA). Dried, ground fecal samples were combined on an equal DM basis into one

39

sample per cow per period. Frozen duodenal samples for each cow period (n=8) were chopped
finely using a commercial food processor (84142 Food cutter, Hobart Manufacturing Co., Troy,
OH) and subsampled in the frozen state to obtain representative samples. These duodenal
subsamples and the 350 mL of ruminal solid and liquid samples were lyophilized and ground as
described above. Dried ruminal solid and liquid samples were recombined according to the
original ratio of solid and liquid DM.
Samples were analyzed for ash, NDF, indigestible NDF (iNDF), ADF, acid detergent
lignin (ADL), ADF nitrogen (forages only), CP, and starch. Ash concentration was determined
after 5 h combustion at 500°C in a muffle furnace. Concentrations of NDF were determined
according to Mertens (2002) and ADF and ADL according to Goering and Van Soest (1970).
Forage samples were analyzed for ADF nitrogen by Cumberland Valley Analytical Services, Inc.
(Hagerstown, MD) using ADF method 973.18 (AOAC, 2000), modified for using glass microfiber filter with 1.5 µm particle retention in place of fritted glass crucible, and followed by
nitrogen analysis of ADF residue using a Leco FP-528 Nitrogen Combustion Analyzer (St.
Joseph, MI). Indigestible NDF was estimated as NDF residue after 240 h in vitro fermentation
(Goering and Van Soest, 1970); flasks were reinoculated at 120 h to insure a viable microbial
population. Forage NDF digestibility was determined by 30 h in vitro fermentation (Goering and
Van Soest, 1970). Ruminal fluid for the in vitro incubations was collected from a nonpregnant
dry cow fed dry hay only. Fraction of potentially digestible NDF (pdNDF) was calculated by
difference (1.00 – iNDF). Crude protein was analyzed according to Hach et al. (1987). Starch
was measured by an enzymatic method (Karkalas, 1985) after samples were gelatinized with
sodium hydroxide. Glucose concentration was measured using a glucose oxidase method
(Glucose kit #510, Sigma Chemical Co., St. Louis, MO), and absorbance was determined with a

40

micro-plate reader (SpectraMax 190, Molecular Devices Corp., Sunnyvale, CA). Concentrations
of all nutrients except DM were expressed as percentages of DM determined by drying at 105°C
in forced-air oven for more than 8 h.
Duodenal digesta were analyzed for purines and ammonia to estimate microbial N (MN)
flow and nonammonia, nonmicrobial N (NANMN) flow to the duodenum. Purine concentration
was used as a microbial marker, and purine to MN ratio was estimated by analysis of microbial
pellets obtained by differential centrifugation of the rumen fluid and particulate samples
collected near the reticulum. Rumen fluid and particulate matter was blended, strained through
nylon mesh, and the liquid portion was centrifuged at 500 x g for 15 min. The supernatant was
centrifuged at 18,000 x g for 15 min, and the pellet was washed with 0.9% NaCl solution and
centrifuged again at 18,000 x g for 15 min, resuspended in water, and lyophilized. Total purines
were measured by spectrophotometer (Beckman Instruments Inc., Fullerton, CA) at 260 nm
according to Zinn and Owens (1986). Ammonia concentration was determined for centrifuged
duodenal and rumen fluid samples according to Broderick and Kang (1980). Rumen fluid was
also analyzed for concentration of major VFA and lactate by HPLC (Waters Corp., Milford,
MA) according to Oba and Allen (2003).
Dry matter and nutrient intakes were calculated using the composition of feed offered and
refused. Ruminal pool sizes (kg) of OM, NDF, iNDF, pdNDF, starch, MN, and NANMN were
determined by multiplying the concentration of each component in rumen samples by the
ruminal digesta DM mass (kg). Duodenal flows (kg/d) of DM, OM, total NDF, pdNDF, starch,
MN, NANMN, and ammonia N were determined using iNDF as a flow marker; iNDF intake
(kg/d) was multiplied by the ratio between the component and iNDF in duodenal digesta.
Duodenal flow of microbial OM was determined using the ratio of purines to OM (Oba and

41

Allen, 2003), and true ruminally digested OM was calculated by subtracting duodenal flow of
nonmicrobial OM from OM intake. Indigestible NDF was used as an internal marker to estimate
nutrient digestibility in the rumen and in the total tract (Cochran et al., 1986). Turnover rate in
the rumen, passage rate from the rumen, and ruminal digestion rate of each component was
calculated by the following equations:
Turnover rate (%/h) = 100 x (Intake of component / Ruminal pool of component) / 24
Passage rate (%/h) = 100 x (Duodenal flow of component / Ruminal pool of component) / 24
Digestion rate (%/h) = Turnover rate in the rumen (%/h) – Passage rate from the rumen (%/h).
Statistical Analysis
All data were analyzed by using the fit model procedure of JMP (Version 8, SAS
Institute, Cary, NC). To determine differences between treatments and evaluate interactions of
treatment with DMI, where pDMI (calculated as the mean of DMI values on d 11 to 14 of the 14d preliminary period) was used as the covariate for treatment responses, data were analyzed
2

according to the following model: Yijk = µ + Ci + Pj + Tk + PTjk + pDMI + TkpDMI + pDMI +
2

TkpDMI + eijk where µ is the overall mean, Ci is the random effect of cow (i = 1 to 13), Pj is
the fixed effect of period (j = 1 to 2), Tk is the fixed effect of treatment (k = 1 to 2), PTjk is the
interaction of period and treatment, pDMI is the linear effect of pDMI, TkpDMI is the interaction
2

2

of treatment and pDMI (linear), pDMI is the quadratic effect of pDMI, TkpDMI is the
interaction of treatment and pDMI (quadratic), and eijk is the residual error. Statistical
2

significance for TkpDMI and TkpDMI indicated treatment differences were related to pDMI.
Covariate and interaction terms were removed stepwise from the model if P > 0.20. Treatment

42

effects and their interaction (linear and quadratic relationships) were declared significant at P "
0.05 and P " 0.10, respectively. Tendencies for treatment effects and their interactions were
declared at P " 0.10 and P " 0.15, respectively.
Sixteen cows started the experiment; however, two cows were removed during the
experiment (one cow injured a teat and the other cow went off feed). Additionally, data from
one cow was excluded prior to statistical analysis because she ate sporadically and had
inconsistent DMI, which ranged from 12.3 to 21.7 kg/d during the 4 d collection of the
preliminary period and 2.7 to 26.0 kg/d and 17.5 to 24.2 kg/d during the 5 d collection of the first
and second experimental periods, respectively. Data from 13 cows were statistically analyzed
for all response variables except those associated with N metabolism, which included 12 cows.
One cow (with highest pDMI) was considered an outlier based on large Cook’s distance values
(Cook and Weisberg, 1982) for response variables for MN flux and microbial efficiency only.
This indicated a problem with the partitioning of NANMN and MN due to purine concentration
for this cow; however, all N data from this cow was removed.

RESULTS AND DISCUSSION
Comparison of Forages and Diets
Physical characteristics of AL are listed in Table 3. Forages chopped to a TLC of 19 and
10 mm had mean particle sizes of 14.1 and 8.1 mm, respectively. The proportion of particles >
19 mm was 22.5 percentage units higher (33.2 vs. 10.7%) and particles < 8 mm was 13.7
percentage units lower (23.9 vs. 37.6%) for long than short cut AL, respectively.
Chemical analyses (Table 3) showed that AL with different lengths of cut had similar
concentrations of OM, ADF, ADL, CP, and starch. Long cut AL had higher DM concentration

43

than short cut AL due to the longer wilting time for long cut AL as the silages were sequentially
harvested and long cut AL was mowed, chopped, and ensiled last. Indigestible NDF expressed
as a percent of total NDF was high for both silages. Despite AL silages being harvested from the
same field on the same day, the proportion of iNDF of total NDF was 7.9 percentage units higher
for long cut than short cut because concentration of total NDF was 1.7 percentage units lower
and iNDF was 2.3 percentage units higher for long cut compared to short cut AL. Although the
iNDF concentration was higher for long cut AL compared to short cut AL, in vitro NDF
digestibility (30 h) of long cut AL was only 0.9 percentage unit lower than that of short cut AL.
It is possible the drier, longer particles of long cut AL did not pack as densely as the wetter,
shorter particles of short cut AL, which might have affected fermentation and storage; however,
this was not evident based on the chemical analyses or fermentation profile. The ADF nitrogen
concentrations, a measure of indigestible compounds formed by chemically linked protein and
carbohydrate used as an indicator of heat-damaged protein, were low and similar for both cuts of
AL, and they appeared to undergo favorable fermentation and be well preserved based on the
low pH and the production of mainly lactic acid.
Diet ingredients and chemical composition are shown in Table 4. The preliminary diet
contained similar proportions of forage NDF from long and short cut AL. Both treatment diets
had a 47:53 forage:concentrate ratio, contained ~20% forage NDF and had similar OM, CP, and
starch composition, which was mathematically calculated according to the proportion of each
feed ingredient in the diet and its respective analytical values. LONG had ~1 percentage unit
lower total NDF and higher iNDF than SHORT, which resulted in the proportion of iNDF of
total dietary NDF 5.9 percentage units higher for LONG. Differences in DM concentration in
diets were because of the different DM concentrations of the forages. The calculated

44

concentrations of total NDF in LONG and SHORT and forage NDF in LONG were slightly
lower than the formulated targets but similar to NRC (2001) minimum requirements. In both
diets, forage NDF provided nearly 80% of the total diet NDF.
Effects of Legume FPL and pDMI
Results of AL particle length and its interaction with pDMI on milk yields and
composition are shown in Table 5. Response of milk fat concentration to FPL was related to
pDMI, as indicated by a significant interaction between FPL and pDMI (P = 0.01); LONG
decreased milk fat concentration more per kilogram pDMI increase than SHORT (Figure 2).
This effect on concentration of milk fat influenced other treatment by pDMI interactions
including milk fat yield (P = 0.006), FCM yield (P = 0.03), and efficiency (FCM/DMI, P =
0.06); LONG increased these responses less per kilogram pDMI increase than SHORT. The
aforementioned interactions resulted in a greater benefit for LONG for cows with low pDMI and
a greater benefit for SHORT for cows with high pDMI.
LONG tended to decrease DMI (26.3 vs. 27.2 kg/d, P = 0.10, Table 5) compared to
SHORT. We expected LONG to be more filling than SHORT causing greater rumen distention
and potentially limiting DMI, particularly in cows with high DMI for which ruminal distention is
more likely to limit feed intake (Allen, 1996). Oba and Allen (1999) and Voelker et al. (2002)
found DMI responses to a more filling diet varied by production level, which is generally
correlated with DMI (NRC, 2001), where DMI was increasingly limited by high fill diets
compared to low fill diets as milk yield increased. We expected LONG to slow rates of ruminal
passage but FPL and its interaction with pDMI did not affect the rates that pdNDF, iNDF, or
starch passed from the rumen (Table 6).
LONG tended to decrease the rate of ruminal turnover of NDF (4.68 vs. 4.91%/h, P =

45

0.09, Table 6) compared to SHORT. Despite the slower turnover rate of NDF for LONG, the
rumen pool of NDF was less for LONG compared to SHORT except for cows with the lowest or
highest DMI (interaction P = 0.08, quadratic, Figure 3A). This indicated that the faster NDF
turnover rate was not sufficient to counterbalance the higher DMI for SHORT resulting in larger
NDF rumen pools for SHORT than LONG. Additionally, rumen digesta wet weight (P = 0.03,
quadratic, Figure 3B) and volume (P = 0.006, quadratic, Figure 3C) were related to pDMI (Table
7); rumen digesta wet weight and volume were less for LONG compared to SHORT except for
cows at the low and high ends of the pDMI range. Although the effect of treatment on DMI was
not related to pDMI (P # 0.18), a visual examination of a graph with pDMI and DMI (Figure 3D)
illustrated the difference in DMI between LONG and SHORT was small for cows with low
pDMI but the difference became greater as pDMI increased and then narrowed for cows with
high pDMI. Because LONG had less rumen digesta mass and volume than SHORT for cows
with the greatest reduction in DMI, it is unlikely that DMI for LONG was limited by rumen fill.
Feeding behavior was measured in another experiment of similar design (Kammes and
Allen, accepted) evaluating the effects of grass FPL. Similarly, cows consumed 0.9 kg/d less
(21.8 vs. 22.7 kg/d) when fed diets with long cut compared to short cut grass silage. Total
chewing time was greater for cows consuming long cut than short cut grass silage diets, such that
cows consuming long cut grass silage were approaching maximum chewing times reported in the
literature (Tafaj et al., 2007). Feeding behavior was not measured in this study, but it is possible
that DMI for cows consuming LONG was limited by chewing time.
Treatment interacted quadratically with pDMI to affect site of starch digestion (Table 8).
True ruminal starch digestion (kg/d, interaction P = 0.13) and true ruminal starch digestibility
(%, interaction P = 0.09, Figure 4A) were lower for LONG compared to SHORT for cows with

46

low and high pDMI. As a result, starch flux from the rumen to duodenum (interaction P = 0.05,
Figure 4B) was greater for LONG compared to SHORT for cows with low and high pDMI.
Postruminal starch digestion (kg/d, interaction P = 0.05) followed a similar pattern as starch flux,
and postruminal starch digestibility (%, interaction P = 0.08, Figure 4C) had a pattern that was
the inverse of true ruminal starch digestibility. Although the cow at each end of the pDMI range
(< 23 and > 32 kg DM/d) was not identified as an outlier based on Cook’s distance values (Cook
and Weisberg, 1982), both cows amplified the quadratic effects. Therefore, data was statistically
reanalyzed after the two cows were removed. Removal did not eliminate the quadratic effects or
notably change results and conclusions so data from both cows were included; however, caution
should be used when making inferences.
As pDMI increased, LONG increased ruminal digestibility of pdNDF and SHORT
decreased it (P = 0.10, Table 9). Ruminal digestibility of NDF was lower for LONG compared
to SHORT (P = 0.01, Table 9) and the differences were greater for cows with lower pDMI
(interaction P = 0.09, Figure 5). Total tract digestibilities of pdNDF (63.5 vs. 78.0%, P < 0.001)
and NDF (23.7 vs. 34.0%, P < 0.001) were lower for LONG than SHORT (Table 9). The lower
digestibility of NDF for LONG may in part be due to the higher concentration of iNDF for long
cut AL than short cut AL (Table 3) despite being harvested from the same field and having
similar ensiling characteristics as previously discussed.
Total tract digestibilities of NDF (and pdNDF) are lower than ruminal digestibility
because negative postruminal digestibilities were calculated for NDF (and pdNDF) in the present
experiment. We evaluated Cr2O3 [5 g dosed through the ruminal cannula at 8 h intervals (total
of 15 g Cr2O3 /d) from d 6 to d 17 with a priming dose of 2X on d 6], ADL, and ADL peroxide
(Cochran et al., 1988) as alternative flow markers. Based on comparisons of calculated flow data

47

using different markers, iNDF provided the most reasonable results and was used as the flow
marker. The higher digestibility for total tract than in the rumen is due to a net gain of fiber from
the duodenum to the feces, which has previously been reported with both the gutter-type T
duodenal cannula (Huhtanen and Jaakkola, 1993; Poore et al., 1993), which is the type used in
this study, and closed T-type duodenal cannula (Stensig and Robinson, 1997). The
underestimation of duodenal NDF flow or duodenal iNDF:NDF ratio using iNDF as a marker
creates inaccuracies of estimated flow of duodenal fiber and postruminal digestibility. These
errors may be related to unrepresentative digesta sampling due to differential separation of fluid
and particles relative to the true material flowing out of the duodenum or analytical problems in
fiber determination of duodenal samples possibly due to a component in the duodenal digesta
that interferes with the analysis. While absolute values are not biologically reasonable, relative
comparisons between treatments within the same experiment are useful.
As a result of NDF digestibility, LONG decreased total tract digestibility of DM (65.6 vs.
67.6%, P = 0.04) and OM (66.6 vs. 68.6%, P = 0.03) and total tract digestion of DM (17.3 vs.
18.2 kg/d, P = 0.01) and OM (16.4 vs. 17.2 kg/d, P = 0.02) compared to SHORT (Table 10).
Interaction of treatment and pDMI for true ruminal OM digestion (kg/d, P = 0.11) and
digestibility (%, P = 0.03, Table 10) were because of effects on starch digestion (Table 8, Figure
4).
Although differences in ruminal digestion were detected, FPL and its interaction with
pDMI did not affect rumen pH (P > 0.19), which was 6.26 for LONG and SHORT, or total VFA
concentration (P > 0.48, Table 11). However, LONG tended to increase concentrations of
butyrate (19.2 vs. 18.3 mM, P = 0.06) and valerate (2.32 vs. 2.20 mM, P = 0.06) compared with
SHORT (Table 11).

48

Effects of pDMI on Ruminal Passage Rates
Experimental data on rates of passage from the rumen, particularly for individual feed
fractions, are scarce. Given the impact of passage on ruminal digestibility and pool sizes and
microbial growth, quantitative knowledge on rates of nutrient passage from the rumen are needed
to better understand nutrient availability in ruminants and improve nutrition models.
Furthermore, because passage rates from the rumen generally increase with increased intake,
measurements of ruminal passage rates of nutrients over a wide range of DMI are necessary. We
measured the effects of DMI on rates of passage of feed fractions from the rumen using the pool
and flux method (Robinson et al., 1987).
We expected ruminal passage rates to increase with pDMI, but passage rates of pdNDF,
iNDF, and starch were not related to level of intake either independent of or dependent upon
treatment (Table 7). These results are consistent with the previously mentioned experiment of
similar design evaluating the effects of grass FPL (Kammes and Allen, accepted). Although
passage rates were not related to pDMI in either study, rates of ruminal digestion of starch and
pdNDF were related to pDMI in that study, which were not observed in the present experiment.
Effects of Treatment and pDMI on N Flux and Microbial Efficiency
Flux of NANMN passed from the rumen to the duodenum was related to pDMI, which
increased at a slower rate for LONG than SHORT (P = 0.03, Figure 6). The increase in
NANMN flux contributed to increased NAN flux as pDMI increased (P = 0.02, Table 12), as
level of intake did not affect MN flux. Despite the increase in NAN flux with greater intake,
postruminal digestion (g/d) and digestibility (%) were not related to pDMI (P # 0.21, Table 12).
When expressed as a percent of duodenal NAN, NANMN and MN fluxes to the duodenum were
related to pDMI (P = 0.008, Table 12). As pDMI increased, LONG decreased NANMN flux and

49

SHORT increased it (Figure 7A), and the reverse was observed for MN flux (Figure 7B).
Microbial N flux was highly related to the rate of ruminal digestion of starch for LONG (P =
2

2

0.004, R = 0.71) but not SHORT (P = 0.78, R = 0.05, Figure 8A). Similarly, MN flux was
2

highly related to true ruminal OM digestion (kg/d) for LONG (P = 0.002, R = 0.75) but not
2

SHORT (P = 0.25, R = 0.26, Figure 8B). Microbial efficiency was not related to level of intake
either independent of or dependent upon treatment (P # 0.25, Table 12). LONG tended to
increase ruminal ammonia concentration (16.4 vs. 14.3 mg/dl, P = 0.06, Table 12) compared to
SHORT. Despite being drier, long cut AL had higher ammonia (6.32 vs. 4.95 mM, Table 3) than
short cut AL, which may be the source for the greater ruminal ammonia concentration observed
for cows consuming LONG.

CONCLUSIONS
In addition to treatment differences in particle length, forages differed in iNDF
concentration (iNDF as a proportion of total NDF was 7.9 percentage units higher for long cut
AL than short cut AL) for unknown reasons, despite our efforts to prevent potentially
confounding errors. Legume particle length and its interaction with pDMI did not affect milk
yield or rumen pH. LONG decreased milk fat concentration more per kilogram pDMI increase
and increased yields of milk fat and fat-corrected milk less per kilogram pDMI increase than
SHORT. LONG tended to decrease DMI compared to SHORT. Ruminal digestion and passage
rates of feed fractions did not differ between LONG and SHORT and were not related to level of
intake. LONG tended to decrease rate of ruminal turnover for NDF but increased NDF rumen
pools at a slower rate than SHORT as pDMI increased. This indicated the faster NDF turnover

50

rate was not sufficient to counterbalance the higher DMI for SHORT resulting in larger NDF
rumen pools for SHORT than LONG. As pDMI increased, LONG increased ruminal
digestibilities of pdNDF and total NDF and SHORT decreased them, but total tract digestibilities
of pdNDF, total NDF, OM and DM were lower for LONG than SHORT. When legume silage
was the only source of forage in the diet, increasing chop length from 10 to 19 mm tended to
decrease DMI but did not negatively affect productivity of cows.

51

APPENDIX

52

Table 2. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet
Standard
Parameter
Median
Mean
deviation
Minimum
Maximum
Parity
3
3.1
0.9
2
5
1
609
612
61
508
750
BW , kg
BCS
2.75
2.7
0.5
1.9
3.6
DIM
202
177
66
50
250
Milk, kg/d
33.1
35.5
10.7
21.9
59.4
3.5% FCM, kg/d
34.4
35.1
10.4
22.9
62.4
DMI, kg/d
26.7
26.5
2.6
22.8
32.4
1
Empty BW (ruminal digesta removed).

53

Table 3. Chemical composition, particle size distribution, and fermentation parameters of the
long (19 mm) and short cut (10 mm) alfalfa silage included in the treatment diets
Alfalfa silage
Item
Long
Short
Chemical composition
DM, %
42.2
36.9
OM, % DM
92.0
90.9
NDF, % DM
42.3
44.0
1
29.1
26.8
iNDF , % DM
iNDF, % of NDF
68.8
60.9
ADF, % of DM
36.5
37.2
ADF nitrogen, % of DM
1.70
1.65
2
8.91
9.05
ADL , % of DM
CP, % DM
20.4
19.6
Starch, % DM
1.89
1.92
3
34.1
35.0
IV NDF digestibility , %
4

1

Particles size distribution
Wet sieving, % DM retained
19.0 mm
9.50 mm
4.75 mm
2.36 mm
1.18 mm
0.600 mm
0.300 mm
0.150 mm
0.075 mm
0.038 mm
5
Mean particle size , mm
Penn State Particle Separator,
% DM retained
> 19.0 mm
19.0 to 8.0 mm
< 8.0 mm
Fermentation
pH
Acetic acid, % DM
Propionic acid, % DM
Butyric acid, % DM
Lactic acid, % DM
Lactic:Acetic
Ethanol, % DM
Ammonia, mM

23.2
39.6
21.0
6.29
3.49
2.24
1.86
1.20
0.76
0.44
14.1

8.57
20.4
30.3
26.6
6.14
3.56
1.93
1.19
0.88
0.58
8.1

33.2
42.9
23.9

10.7
51.7
37.6

4.62
1.29
0.08
<0.01
5.40
4.19
0.10
6.32

4.42
1.49
0.07
<0.01
6.46
4.34
0.08
4.95

iNDF = indigestible NDF.

54

Table 3 (cont’d)
2

ADL = acid detergent lignin.

3

30 h in vitro NDF digestibility.

4

Particle size distributions of silages were measured each period (n = 2).

5

Mean particle size calculated from particle size distribution determined by wet sieving.

55

Table 4. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole source of forage
Preliminary
Long
Short
Ingredients, % DM
Alfalfa silage, long cut
23.2
46.3
--Alfalfa silage, short cut
23.2
--47.0
Dry ground corn
36.4
36.4
36.0
Soybean meal (48% CP)
7.99
8.03
7.84
®
4.00
4.02
3.92
SoyPlus
1
4.68
4.68
4.68
Vitamin mineral mix

1

Limestone
Salt
Chemical composition
DM, %
OM, % DM
NDF, % DM
% forage NDF
% NDF from forage
2
iNDF , % DM
iNDF, % of NDF
CP, % DM
Starch, % DM

0.39
0.19
53.1
93.4
26.9
21.7
80.9
3
NA
NA
18.0
30.6

0.39
0.19

0.39
0.19

58.8
93.4
24.5
19.6
80.0
15.5

53.5
92.8
25.5
20.7
81.0
14.6

63.2
19.3
30.8

57.3
18.9
30.5

Vitamin mineral mix contained (DM basis) 17.1% sodium bicarbonate, 3.9% dicalcium
phosphate, 2.6% magnesium oxide, 1.9% salt, 1.9% trace mineral premix, 0.4% vitamin A, 0.4%
vitamin D, 0.2% vitamin E, and 71.6% dry ground corn as a carrier.
2
iNDF = indigestible NDF.
3

NA = no analysis for preliminary diet.

56

Table 5. Milk production and composition, feed intake, and BW change of cows fed treatment diets containing either long (19 mm) or
short cut (10 mm) alfalfa silage as the sole source of forage
1

2

Treatment LSM
Long
Short
Yield, kg/d
Milk

1

FCM (3.5 %)
Milk fat
Milk protein
Milk lactose
SNF
Milk composition, %
Fat
Protein
Lactose
SNF
DMI, kg/d
3.5% FCM/ DMI
BW change, kg/19 d
BCS change/19 d

P
SE

Trt

37.7

37.4

3.3

38.8
1.28
1.20
1.85
3.01

38.8
1.29
1.19
1.83
2.99

3.77
3.24
4.86
7.98
26.3
1.36
15.3
0.11

3.80
3.26
4.87
8.01
27.2
1.34
11.2
0.07

Trt x
Period

pDMI

0.67

NS

0.03

3.2
0.09
0.08
0.18
0.26

0.88
0.57
0.71
0.66
0.67

NS
NS
NS
NS
NS

0.15
0.13
0.07
0.09
0.5
0.09
2.2
0.06

0.58
0.38
0.41
0.14
0.10
0.40
0.22
0.68

NS
NS
NS
NS
NS
NS
0.009
NS

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.04

NS

0.03
0.11
0.008
0.03
0.02

NS
0.03
0.006
NS
NS
NS

0.06
0.20
0.04
0.03
0.03

NS
NS
NS
NS
NS

0.08
0.32
0.003
0.18
<0.001
0.08
0.15
NS

0.01
NS
NS
NS
0.34
0.06
0.19
NS

0.05
0.09
0.005
NS
0.03
0.07
NS
NS

NS
NS
NS
NS
0.18
NS
NS
NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

57

Table 6. Rumen kinetics of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole
source of forage
1

2

Treatment LSM
Long
Short
Ruminal turnover rate, %/h
DM
OM
NDF

4

pdNDF
Starch
Ruminal turnover time, h
DM
OM
NDF
pdNDF
5

1

iNDF
Starch
Ruminal passage rate, %/h
pdNDF
iNDF
Starch
Ruminal digestion rate, %/h
pdNDF
Starch

P
SE

Trt

Trt x
Period

pDMI

11.0

10.8

0.3

0.59

0.03

11.2
4.68
13.8

11.0
4.91
16.7

0.3
0.16
2.3

0.61
0.09
0.41

54.3

52.2

4.8

9.28
9.09
21.9
7.58
30.3

9.36
9.16
20.6
6.75
31.0

2.08

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS

NS

NS

0.04
0.08
NS

NS
NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

0.73

0.03

NS

NS

NS

NS

0.26
0.27
0.7
0.58
1.2

0.77
0.81
0.10
0.34
0.55

0.02
0.03
0.07
NS
0.16

NS
NS
NS
0.24
NS

NS
NS
NS
NS
NS

NS
NS
NS
0.15
NS

NS
NS
NS
NS
NS

2.11

0.15

0.85

0.006

NS

NS

NS

NS

1.44
3.39
23.6

1.99
3.28
23.4

0.45
0.13
3.1

0.33
0.36
0.97

0.16
0.17
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

12.3
33.3

14.7
31.4

2.3
3.4

0.51
0.56

NS
0.05

NS
NS

NS
NS

NS
0.18

NS
NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
58

Table 6 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

59

Table 7. Rumen pools of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole
source of forage
1

2

Treatment LSM
Long
Short

SE

Trt

Trt x
Period

Wet weight, kg

72.6

77.6

2.4

0.05

Volume, L
Density, kg/L
Rumen pool, kg
DM
OM
NDF

86.8
0.84

95.1
0.82

3.6
0.02

10.2
9.35
5.57
0.74

10.8
9.85
5.98
0.85

4.89
0.72

pdNDF

4

5

1

iNDF
Starch

P
pDMI

Trt x
pDMI

pDMI x
pDMI

0.03

0.07

0.32

0.02
0.30

NS
NS
0.18

Trt x
pDMI x
pDMI
0.03

0.07
0.96

0.01
0.14

0.44
0.71

0.006
0.15

0.4
0.41
0.25
0.06

0.07
0.11
0.04
0.27

0.13
0.16
NS
NS

0.02
0.02
0.02
0.004

0.33
0.22
0.09
NS

0.16
0.17
0.19
NS

0.15
0.15
0.08
NS

5.18

0.24

0.12

0.19

0.04

0.29

0.08

0.09

0.72

0.06

0.97

0.01

0.09

NS

NS

NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

60

Table 8. Starch digestion of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole
source of forage
1

Treatment LSM
Long
Short
Starch
Intake, kg/d

1

Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
% of duodenal passage
Apparent total tract digestion
kg/d
%

2

P
SE

Trt

Trt x
Period
3

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

<0.001

0.53

0.03

0.14

8.49

8.51

0.18

0.92

NS

5.35
62.9

4.94
57.8

0.35
4.0

0.43
0.43

NS
NS

0.01
0.11

0.98
0.81

0.004
0.008

0.14
0.10

5.54
65.1
3.14

5.12
59.9
3.57

0.35
4.0
0.33

0.43
0.43
0.40

NS
NS
NS

0.009
0.11
0.99

0.99
0.78
0.83

0.004
0.01
0.03

0.13
0.09
0.05

2.57
30.5
83.3

3.09
36.7
84.0

0.33
4.1
1.9

0.32
0.36
0.80

NS
NS
NS

0.66
0.08
0.10

0.78
0.76
NS

0.03
0.01
0.18

0.05
0.08
NS

7.92
93.6

8.03
94.1

0.13
0.5

0.45
0.52

NS
NS

<0.001
0.12

0.32
NS

0.009
NS

0.04
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

61

Table 9. NDF digestion of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the sole
source of forage
1

Treatment LSM
Long
Short
NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Potentially digestible NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Indigestible NDF
Intake, kg/d
1
Treatment least squares means.

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

6.17

6.72

0.13

<0.001

0.16

0.001

NS

3

0.08

NS

1.99
33.1
4.03

2.50
38.1
4.07

0.09
1.3
0.12

<0.001
0.01
0.78

0.02
0.08
NS

0.13
0.90
0.02

NS
0.09
0.15

NS
NS
NS

NS
NS
NS

-0.58

-0.27

0.14

0.09

NS

0.37

0.15

NS

NS

1.40
23.7

2.22
34.0

0.07
1.2

<0.001
<0.001

NS
NS

NS
0.07

NS
NS

NS
NS

NS
NS

2.31

2.95

0.05

<0.001

0.006

<0.001

0.03

0.04

0.10

1.99
87.7
0.28

2.50
87.1
0.37

0.09
3.2
0.08

<0.001
0.88
0.43

0.02
NS
NS

0.13
0.95
0.64

NS
0.10
0.10

NS
NS
NS

NS
NS
NS

-0.58

-0.27

0.14

0.09

NS

0.37

0.15

NS

NS

1.40
63.5

2.22
78.0

0.07
3.1

<0.001
0.009

NS
NS

NS
0.10

NS
NS

NS
NS

NS
NS

3.84

3.79

0.08

0.56

NS

0.001

NS

0.07

NS

62

Table 9 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

63

Table 10. DM and OM digestion of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the
sole source of forage
1

2

Treatment LSM
Long
Short
DM
Intake, kg/d

1

Apparent total tract digestion
kg/d
%
OM
Intake, kg/d
Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
Apparent total tract digestion
kg/d
%

P
SE

Trt

Trt x
Period

26.3

27.2

0.5

0.10

NS

17.3
65.6

18.2
67.6

0.3
0.6

24.6

25.3

11.6
47.8

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

3

<0.001

0.34

0.03

0.18

0.01
0.04

0.13
NS

<0.001
0.07

NS
NS

0.02
NS

NS
NS

0.5

0.18

NS

<0.001

0.36

0.03

0.18

11.8
45.8

0.4
1.8

0.65
0.47

NS
NS

0.002
0.17

NS
0.32

<0.001
0.004

NS
0.18

15.2
61.8
12.9

14.9
58.9
13.7

0.6
1.9
0.5

0.68
0.31
0.23

NS
NS
NS

0.005
0.21
0.09

0.51
0.21
0.25

0.006
0.01
0.24

0.11
0.03
0.09

4.62
20.1

5.74
21.7

0.45
1.6

0.10
0.45

NS
NS

0.76
0.06

0.21
NS

0.03
0.005

0.14
NS

16.4
66.6

17.2
68.6

0.2
0.5

0.02
0.03

0.11
NS

<0.001
0.05

NS
NS

0.01
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).

64

Table 10 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

65

Table 11. Ruminal VFA concentrations and pH of cows fed treatment diets containing either long (19 mm) or short cut (10 mm)
alfalfa silage as the sole source of forage
1

Treatment LSM
Long
Short
Total VFA, mM

1

Acetate
Propionate
Butyrate
Lactate
Isobutyrate
Valerate
Isovalerate
Branch chain VFA
Acetate:Propionate
Ruminal pH

139
84.4
29.9
19.2
0.37
1.37
2.32
2.10
3.54
2.85
6.26

138
84.8
29.3
18.3
0.81
1.39
2.20
2.13
3.50
2.90
6.26

2

P
SE

Trt

Trt x
Period

2

0.48

1.3
0.6
0.4
0.24
0.07
0.05
0.09
0.12
0.04
0.04

0.57
0.40
0.06
0.21
0.89
0.06
0.78
0.82
0.32
0.99

NS
NS
0.03
0.05
NS
NS
NS
NS
NS
0.008
0.01

3

pDMI

Trt x
pDMI

pDMI x
pDMI

0.15

NS

NS

Trt x
pDMI x
pDMI
NS

NS
0.04
0.11
NS
0.43
NS
0.55
NS
0.08
0.06

NS
NS
NS
NS
0.19
NS
0.15
NS
NS
0.94

NS
NS
NS
NS
0.49
NS
NS
NS
NS
0.86

NS
NS
NS
NS
0.19
NS
NS
NS
NS
0.19

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

66

Table 12. Nitrogen metabolism of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) alfalfa silage as the
sole source of forage
1

Treatment LSM
Long
Short
N intake, g/d

793

Ruminal ammonia, mg/dl
Flow to duodenum
Ammonia N, g/d
NAN
g/d
% of N intake
4

NANMN
g/d
% of N intake
% of duodenal NAN
Microbial N
g/d
% of duodenal NAN
5

SE

Trt
0.86

13

Trt x
Period

pDMI

0.13

0.001

Trt x
pDMI
3

pDMI x
pDMI
NS

Trt x
pDMI x
pDMI
NS

NS

NS

16.4

14.3

0.8

0.06

NS

NS

NS
NS

12.4

11.7

0.6

0.42

NS

NS

NS

NS

NS

559
70.2

568
71.9

21
1.8

0.71
0.58

NS
NS

0.02
NS

NS
NS

NS
NS

NS
NS

243
30.9
42.0

241
30.3
43.1

13
1.9
3.5

0.90
0.78
0.68

NS
NS
NS

0.03
0.39
0.63

0.03
0.04
0.008

NS
NS
0.79

NS
NS
0.13

316
58.0
22.5

328
56.9
22.9

23
3.5
1.2

0.60
0.68
0.77

NS
NS
NS

0.40
0.63
NS

0.20
0.008
NS

NS
0.79
NS

NS
0.13
NS

319
40.5
56.1

19
2.0
1.7

0.38
0.29
0.21

NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

541
68.6

11
0.9

0.37
0.27

0.06
NS

0.02
0.17

NS
NS

NS
NS

NS
NS

g/kg TRDOM
NAN apparent postruminal digestion
g/d
297
% of N intake
37.4
% of duodenal passage
52.8
N apparent total tract digestion
g/d
531
%
67.1

1

790

2

P

Treatment least squares means.

67

Table 12 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

NANMN = nonammonia, nonmicrobial nitrogen.

5

TRDOM = true ruminally digested OM.

68

Figure 2. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for milk fat concentration (P =
0.01). The preliminary DMI on the x-axis are the mean DMI of individual cows during the final
4 d of the preliminary period when all cows were fed a common diet.

69

Figure 3

70

Figure 3 (cont’d)

Figure 3. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for A) NDF rumen pool (P =
0.08), B) rumen digesta wet weight (P = 0.03), C) rumen digesta volume (P = 0.006), and D)
DMI (interaction not significant). The preliminary DMI on the x-axis are the mean DMI of
individual cows during the final 4 d of the preliminary period when all cows were fed a common
diet.

71

Figure 4

72

Figure 4 (cont’d)

Figure 4. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for starch A) true ruminal
digestibility (P = 0.09), B) flux from the rumen to duodenum (P = 0.05), and C) postruminal
digestibility (P = 0.08). The preliminary DMI on the x-axis are the mean DMI of individual
cows during the final 4 d of the preliminary period when all cows were fed a common diet.

73

Figure 5. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for NDF ruminal digestibility (P
= 0.09). The preliminary DMI on the x-axis are the mean DMI of individual cows during the
final 4 d of the preliminary period when all cows were fed a common diet.

74

Figure 6. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for nonammonia, nonmicrobial N
flux from the rumen to duodenum (P = 0.03). The preliminary DMI on the x-axis are the mean
DMI of individual cows during the final 4 d of the preliminary period when all cows were fed a
common diet.

75

Figure 7. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) alfalfa particle length with preliminary DMI for A) nonammonia,
nonmicrobial N (P = 0.008) and B) microbial N (P = 0.008) flux from the rumen to duodenum
expressed as percent of duodenal NAN. The preliminary DMI on the x-axis are the mean DMI
of individual cows during the final 4 d of the preliminary period when all cows were fed a
common diet.

76

Figure 8. A) Relationship between starch ruminal digestion rate (kd) and microbial N (MN) flux
from rumen to duodenum for long (open circles, dashed line; MN flux, g/d = [205 + (1.76 x
2
2
starch ruminal kd, %/h) + (0.306 x (starch ruminal kd, %/h – 30.1) )]; P = 0.004, R = 0.71) and
2

short (closed circles, solid line; P = 0.78, R = 0.05) alfalfa particle length. B) Relationship
between true ruminally digested OM (TRDOM) and MN flux from rumen to duodenum for long
2
(MN flux, g/d = [-145 + (30.0 x TRDOM, kg/d) + (6.27 x (TRDOM, kg/d – 14.2) )]; P = 0.002,
2

2

R = 0.75) and short (P = 0.25, R = 0.26) alfalfa particle length.

77

REFERENCES

78

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Allen, M. S. 1997. Relationship between fermentation acid production in the rumen and the
requirement for physically effective fiber. J. Dairy Sci. 80:1447-1462.
Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cows.
J. Dairy Sci. 83:1598-1624.
AOAC. 2000. Official Methods of Analysis. 17th ed. Association of Official Analytical
Chemists. Arlington, VA.
Beauchemin, K. A., L. M. Rode, and M. V. Eliason. 1997. Chewing activities and milk
production of dairy cows fed alfalfa as hay, silage, or dried cubes or silage. J. Dairy Sci.
80:324-333.
Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64-75.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting digestibility of
different diets with internal markers: Evaluation of four potential markers. J. Anim. Sci.
63:1476-1483.
Cochran, R. C., E. S. Vanzant, and T. DelCurto. 1988. Evaluation of internal markers isolated by
alkaline hydrogen peroxide incubation and acid detergent lignin extraction. J. Anim. Sci.
66:3245-3251.
Cook, R. D., and S. Weisberg. 1982. Residuals and influence in regression. Chapman
and Hall, New York.
Dixon, R. M., and L. P. Milligan. 1985. Removal of digesta components from the rumen of
steers determined by sieving techniques and fluid, particulate and microbial markers. Br.
J. Nutr. 53:347-362.
Einarson, M. S., J. C. Plaizier, and K. M. Wittenberg. 2004. Effects of barley silage chop length
on productivity and rumen conditions of lactating dairy cows fed total mixed ration. J.
Dairy Sci. 87:2987-2996.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.
Grant, R. J. 1997. Interactions among forages and nonforage fiber sources. J. Dairy Sci. 80:14381446.

79

Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More powerful peroxide
Kjeldahl digestion method. J. AOAC 70:783-787.
Huhtanen, P., and S. Jaakkola. 1993. The effects of forage preservation method and proportion of
concentrate on digestion of cell wall carbohydrates and rumen digesta pool size in cattle.
Grass Forage Sci. 48:155-165.
Joy, M. T., E. J. DePeters, J. G. Fadel, and R. A. Zinn. 1997. Effects of corn processing on the
site and extent of digestion in lactating cows. J. Dairy Sci. 80:2087-2097.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019-1027.
Kononoff, P. J., and A. J. Heinrichs. 2003. The effect of reducing alfalfa haylage particle size on
cows in early lactation. J. Dairy Sci. 86:1445-1457.
Krause, K. M., and D. K. Combs. 2003. Effects of forage particle size, forage source, and grain
fermentability on performance and ruminal pH in midlactation cows. J. Dairy Sci.
86:1382-1397.
Lammers, B. P., D. R. Buckmaster, and A. J. Heinrichs. 1996. A simplified method for the
analysis of particle sizes of forage and total mixed rations. J. Dairy Sci. 79:922-928.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:12171240.
Mooney, C. S., and M. S. Allen. 1997. Physical effectiveness of the neutral detergent fiber of
whole linted cottonseed relative to that of alfalfa silage at two lengths of cut. J.
Dairy Sci. 80:2052-2061.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl.
Acad. Washington, DC.
Nørgaard, P. 1983. Saliva secretion and acid-base status of ruminants: A review. Acta. Vet.
Scand. Suppl. 89:93-100.
Oba, M., and M. S. Allen. 1999. Effects of brown midrib 3 mutation in corn silage on dry
matter intake and productivity of high yielding dairy cows. J. Dairy Sci. 82:135-142.
Oba, M., and M. S. Allen. 2003. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174–183.
Onetti, S. G., R. D. Shaver, S. J. Bertics, and R. R. Grummer. 2003. Influence of corn silage

80

particle length on the performance of lactating dairy cows fed supplemental tallow. J.
Dairy Sci. 96:2949-2957.
Poore, M. H., J. A. Moore, T. P. Eck, R. S. Swingle, and C. B. Theurer. 1993. Effect of fiber
source and ruminal starch degradability on site and extent of digestion in dairy cows. J.
Dairy Sci. 76:2244-2253.
Randby, Å. T., T. Garmo, M. Eknæs, and E. Prestløkken. 2008. Effect of grass silage chop
length on intake and milk production by dairy cows. Grassl. Sci. in Europe. 13:768-770.
Robinson, P. J., S. Tamminga, and A. M. van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci. 17:37-62.
Soita, H. W., D. A. Christensen, and J. J. McKinnon. 2000. Influence of particle size on the
effectiveness of the fibre in barley silage. J. Dairy Sci. 83:2295-2300.
Stensig, T., and P. H. Robinson. 1997. Digestion and passage kinetics of forage fiber in dairy
cows as affected by fiber-free concentrate in the diet. J. Dairy Sci. 80:1339-1352.
Tafaj, M., Q. Zebeli, Ch. Baes, H. Steingass, and W. Drochner. 2007. A meta-analysis examining
effects of particle size of total mixed rations on intake, rumen digestion and milk
production in high-yielding dairy cows in early lactation. Anim. Feed Sci. Technol.
138:137-161.
Voelker, J. A., G. M. Burato, and M. S. Allen. 2002. Effect of pretrial milk yield on responses
of feed intake, digestion, and production to dietary forage concentration. J. Dairy Sci.
85:2650-2661.
Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. J. Dairy Sci. 65:495–501.
Yang, W. Z., and K. A. Beauchemin. 2004. Grain processing, forage-to-concentrate ratio, and
forage length effects on ruminal nitrogen degradation and flows of amino acids to the
duodenum. J. Dairy Sci. 87:2578-2590.
Yang, W. Z., K. A. Beauchemin, and M. L. Rode. 2001. Effects of grain processing, forage to
concentrate ratio, and forage particle size on rumen pH and digestion in dairy cows. J.
Dairy Sci. 84:2203-2216.
Zinn, R. A., and F. N. Owens. 1986. A rapid procedure for purine measurement and its use for
estimating net ruminal protein synthesis. Can. J. Anim. Sci. 66:157-166.

81

CHAPTER 3
Nutrient Demand Interacts with Grass Particle Length to Affect Digestion Responses and
Chewing Activity in Dairy Cows

1

ABSTRACT
Effects of grass particle length on dry matter intake (DMI), milk production, ruminal
fermentation and pool sizes, digestion and passage kinetics, and chewing activity and the
relationship of these effects with preliminary DMI (pDMI) were evaluated using 15 ruminally
and duodenally cannulated Holstein cows in a crossover design with a 14-d preliminary period
and two 18-d treatment periods. During the preliminary period, pDMI of individual cows ranged
from 22.6 to 29.8 kg/d (mean = 25.8 kg/d) and 3.5% fat-corrected milk yield ranged from 29.2 to
56.9 kg/d (mean = 41.9 kg/d). Experimental treatments were diets containing orchardgrass silage
chopped to either a) 19 mm (LONG) or b) 10 mm (SHORT) theoretical length of cut as the sole
forage. Grass silages contained ~46% neutral detergent fiber (NDF); diets contained 50%
forage, 23% forage NDF, and 28% total NDF. Preliminary DMI, an index of nutrient demand,
was determined during the last 4 d of the preliminary period when cows were fed a common diet
and used as a covariate. Main effects of grass particle length and their interaction with pDMI
were tested by ANOVA. Grass particle length and its interaction with pDMI did not affect milk
yield, milk composition, or rumen pH. LONG tended to decrease DMI compared to SHORT,
which might have been limited by rumen fill or chewing time or both. Passage rates of feed
fractions did not differ between LONG and SHORT and were not related to level of intake. As
pDMI increased, LONG decreased ruminal digestion rate of pdNDF at a faster rate than SHORT.
1

Accepted for publication in the Journal of Dairy Science on October 18, 2011.

82

As a result, LONG decreased or tended to decrease rates of ruminal turnover for NDF, organic
matter, and dry matter and increased their rumen pools compared to SHORT for cows with high
pDMI. LONG increased eating time, which affected cows with high intake to the greatest extent,
and total chewing time compared to SHORT. As intake increased, ruminal digestion (kg/d) and
digestibility (%) of starch decreased, rumen pool size of starch increased, and post ruminal
digestion and digestibility of starch increased quadratically. When grass silage was the only
source of forage in the diet, increasing chop length from 10 to 19 mm tended to decrease DMI
but did not negatively affect productivity of cows, which were fed adequate fiber.

INTRODUCTION
Forage particle size impacts various aspects of rumen function and digestion kinetics.
Ruminal digesta passage rates decrease with increasing particle size due to greater retention time
in the rumen (Dixon and Milligan, 1985) and mat formation by long forage particles, which
increases digestibility of smaller particles (Grant, 1997). Greater ruminal distention caused by
longer forage particles is more likely to affect passage rate and feed intake of lactating dairy
cows when feed intake is more limited by rumen fill. Decreasing particle size permits rapid
removal of digesta from the rumen allowing increased feed intake when intake is limited by
distention, but ruminal pH might be reduced for several reasons including a reduction in buffer
capacity of ruminal digesta mass, a reduction in rate of VFA absorption from decreased motility,
and a reduction in salivary buffer secretion from decreased rumination.
Forage particle length (FPL) has been widely researched, but the effects of FPL on
animal responses are inconsistent and inconclusive. Some of the inconsistency on responses to
particle size may be due to forage type. Tafaj et al. (2007) reported that effects of forage particle

83

size were less when corn silage was included in the TMR and greater for grass silage-based
TMR. Furthermore, grasses and legumes differ in in vitro cell wall digestion rates (Smith et al.,
1972; Robles et al. 1980) and anatomical structure and digestion characteristics affecting particle
size reduction and passage (Allen and Mertens, 1988). These differences suggest consideration
of forage family is necessary when evaluating the effects of particle size. Orchardgrass (OG;
Dactylis glomerata L.) was selected as a representative cool-season grass for use in this
experiment.
Besides dietary factors, another reason for inconsistent responses to FPL may be related
to animal factors. Cows respond differently to treatments depending on their level of intake
(Voelker Linton and Allen, 2008) and production (Oba and Allen, 1999). Because FPL and level
of intake affect ruminal passage and digestion rates and thus digesta fill in the rumen, the
response to effects of particle size and its relationship with intake level should be assessed to
determine if responses to treatment vary among cows with a wide range in DMI. We
hypothesized that responses of DMI and digesta passage rates to grass particle length are related
to level of intake and shorter particle length will permit a greater increase in passage rate than
longer particle length as feed intake increases.
The objectives of this experiment were to evaluate the relationships between voluntary
DMI and effects of length of cut of grass silage on DMI, milk production, ruminal fermentation
and pool sizes, digestion and passage kinetics, and chewing behavior in lactating dairy cows.
This study had three unique features to improve our understanding of the role of particle size and
interpret its effect on animal responses. First, it allowed effects of the interaction between FPL
and preliminary DMI (pDMI) to be evaluated. The use of pDMI, an index of nutrient demand,
allowed the evaluation of treatments on animal responses in relation to level of intake and

84

provided an indicator to test effects of intake level independent of treatments. Second, it directly
compared treatment effects of long and short cut OG as the sole source of forage without the
confounding effects of other dietary factors. Third, ruminal passage rates of individual feed
fractions, instead of entire feeds, were measured using ruminally and duodenally cannulated
cows.

MATERIALS AND METHODS
Cows and Treatments
Experimental procedures were approved by the Institutional Animal Care and Use
Committee at Michigan State University. Fifteen multiparous Holstein cows from the Michigan
State University Dairy Cattle Teaching and Research Center were assigned randomly to
treatment sequence in a crossover design experiment with one 14-d preliminary period and two
18-d experimental periods. The first 10 d of each period were allowed for diet adaptation and
samples were collected during the final 4 d of the preliminary period and 8 d of each
experimental period. Cows were 164±56 (mean±SD) DIM at the end of the preliminary period
and were selected to provide a wide range and uniform distribution of pDMI and milk yield.
During the final 4 d of the 14-d preliminary period, the average pDMI among cows ranged from
22.6 to 29.8 kg/d (mean = 25.8 kg/d) and 3.5% FCM yield ranged from 29.2 to 56.9 kg/d (mean
= 41.9 kg/d; Table 13). Prior to calving, cows were cannulated ruminally (Bar Diamond, Inc.,
Parma, ID) and duodenally with a gutter-type T cannula placed approximately 10 cm distal to the
pylorus (Joy et al., 1997). Surgery was performed at the Department of Large Animal Clinical
Science, College of Veterinary Medicine, Michigan State University.
Experimental treatments were diets containing OG silage chopped to either a) 19 mm

85

(LONG) or b) 10 mm (SHORT) theoretical length of cut (TLC) as the sole forage. These TLC
were selected to provide a wide interval within the normal range of TLC to examine if animal
response to FPL is affected by level of feed intake.
Orchardgrass (Baridana cultivar, Barenbrug USA, Tangent, OR) was produced at the
campus farm at Michigan State University (East Lansing), chopped from the same field, and
ensiled in Ag-Bags (Ag-Bag Systems Inc., St. Nazianz, WI). During the sample collection
periods, long and short cut OG contained ~46% NDF (DM basis; Table 14). Diets LONG and
SHORT were formulated to contain 21% forage NDF, 28% total NDF, and 18% CP. The diet
fed during the preliminary period was formulated so that long and short cut OG each contributed
50% of forage NDF. Diets also contained dry ground corn, soybean meal (48% CP), SoyPlus

®

(West Central Soy, Ralston, IA), vitamin-mineral premix, and limestone (Table 15).
Data and Sample Collection
Throughout the experiment, cows were housed in tie-stalls and fed diets as total mixed
rations once daily (1130 h) at 110% of expected intake. The amount of feed offered and refused
(orts) was weighed daily for each cow. Forage samples were collected twice weekly and
analyzed to adjust diets to account for DM, NDF, and CP fluctuation. Samples of all dietary
ingredients (0.5 kg) and orts (12.5%) were collected daily from d 11 to 14 during the preliminary
period and d 11 to 15 during each experimental period. Samples were frozen immediately after
collection at !20°C and combined to one composite sample per period prior to analysis.
Cows were moved to an exercise lot twice daily (0230 and 1300 h) prior to milking in a
parlor (0400 and 1430 h). Milk yield was measured, and milk was sampled, at each milking on d
11 to 14 of the preliminary period and on d 11 to 15 of the experimental periods. Rumen-empty
BW was measured by weighing the cow after evacuation of ruminal digesta on d 14 of the

86

preliminary period and d 18 of each experimental period. Body condition score was determined
on the same days by 3 trained investigators blinded to treatments (Wildman et al., 1982; 5-point
scale where 1 = thin and 5 = fat). Chewing activity was monitored and recorded by observation
every 5 min for 24 h on d 16 of each experimental period. Activity was noted as eating,
ruminating, drinking, or idle for each cow at each time.
Duodenal samples (900 mL), fecal samples (500 g), rumen fluid and particulate samples
for microbial isolation (400 g), and rumen fluid samples for pH, concentrations of VFA, lactate,
and ammonia (100 mL) were collected every 15 h from d 11 to 15 of each experimental period
so that 8 samples were taken for each cow in each period, representing every 3 h of a 24-h period
to account for diurnal variation. Rumen fluid and particulate matter for microbial isolation was
collected from the reticulum near the reticular-omasal orifice, transported to the laboratory, and
processed. Rumen fluid for pH, VFA, lactate, and ammonia was obtained by combining digesta
from five different sites in the rumen and straining it through nylon mesh (~1 mm pore size);
fluid pH was recorded immediately. Samples were stored at -20°C.
Ruminal contents were evacuated manually through the ruminal cannula 4 h after feeding
at the beginning of d 17 (1530 h) and 2 h before feeding at the end of d 18 (0930) for each
experimental period. Total rumen content mass and volume were determined. To ensure
accurate sampling, every tenth handful of digesta (10%) was separated for a subsample
throughout evacuation. This subsample was squeezed into primarily solid and liquid phases.
Both phases were weighed and sampled (350 mL) for determination of nutrient pool size. All
samples were stored at -20°C.
Sample Analysis and Calculations

87

Milk yield recorded at both milkings were summed for a daily total, which were averaged
for each period. Milk samples were analyzed for fat, true protein, lactose, SNF, and MUN with
infrared spectroscopy by Michigan DHIA (East Lansing). Yields of 3.5% FCM and milk
components were calculated using milk yield and component concentrations for each milking,
summed for a daily total, and averaged for each period.
Forage samples were combined to one composite sample per forage per period. Particle
size distribution was determined using the Penn State Particle Separator containing 2 sieves (19
and 8 mm) and a pan (Lammers et al., 1996). Due to the high moisture content of the silages, the
particles tended to cling together and remain on the top sieve during the shaking process yielding
inaccurate measurements. Therefore, samples were dried to a constant weight with forced air (no
added heat) prior to separation. In addition, samples were wet sieved manually and sequentially
through screens with the following aperture sizes: 19.0, 9.50, 4.75, 2.36, 1.18, 0.600, 0.300,
0.150, 0.075, and 0.038 mm. The fraction of DM retained on the screens from wet sieving was
used to calculate mean particle size.
Diet ingredients, orts, and feces were lyophilized (Tri-Philizer MP, FTS Systems, Stone
Ridge, NY). All dried samples were ground with a Wiley mill (1-mm screen; Arthur H. Thomas,
Philadelphia, PA). Dried, ground fecal samples were combined on an equal DM basis into one
sample per cow per period. Frozen duodenal samples for each cow period (n=8) were chopped
finely using a commercial food processor (84142 Food cutter, Hobart Manufacturing Co., Troy,
OH) and subsampled in the frozen state to obtain representative samples. These duodenal
subsamples and the 350 mL of ruminal solid and liquid samples were lyophilized and ground as
described above. Dried ruminal solid and liquid samples were recombined according to the
original ratio of solid and liquid DM.

88

Samples were analyzed for ash, NDF, indigestible NDF (iNDF), ADF, acid detergent
lignin (ADL), CP, and starch. Ash concentration was determined after 5 h combustion at 500°C
in a muffle furnace. Concentrations of NDF were determined according to Mertens (2002) and
ADF and ADL according to Goering and Van Soest (1970). Indigestible NDF was estimated as
NDF residue after 240 h in vitro fermentation (Goering and Van Soest, 1970); flasks were
reinoculated at 120 h to insure a viable microbial population. Ruminal fluid for the in vitro
incubations was collected from a nonpregnant dry cow fed dry hay only. Fraction of potentially
digestible NDF (pdNDF) was calculated by difference (1.00 – iNDF). Crude protein was
analyzed according to Hach et al. (1987). Starch was measured by an enzymatic method
(Karkalas, 1985) after samples were gelatinized with sodium hydroxide. Glucose concentration
was measured using a glucose oxidase method (Glucose kit #510, Sigma Chemical Co., St.
Louis, MO), and absorbance was determined with a micro-plate reader (SpectraMax 190,
Molecular Devices Corp., Sunnyvale, CA). Concentrations of all nutrients except DM were
expressed as percentages of DM determined by drying at 105°C in forced-air oven for more than
8 h.
Duodenal digesta were analyzed for purines and ammonia to estimate microbial N (MN)
flow and nonammonia, nonmicrobial N (NANMN) flow to the duodenum. Purine concentration
was used as a microbial marker, and purine to MN ratio was estimated by analysis of microbial
pellets obtained by differential centrifugation of the rumen fluid and particulate samples
collected near the reticulum. Rumen fluid and particulate matter was blended, strained through
nylon mesh, and the liquid portion was centrifuged at 500 x g for 15 min. The supernatant was
centrifuged at 18,000 x g for 15 min, and the pellet was washed with 0.9% NaCl solution and
centrifuged again at 18,000 x g for 15 min, resuspended in water, and lyophilized. Total purines

89

were measured by spectrophotometer (Beckman Instruments Inc., Fullerton, CA) at 260 nm
according to Zinn and Owens (1986). Ammonia concentration was determined for centrifuged
duodenal and rumen fluid samples according to Broderick and Kang (1980). Rumen fluid was
also analyzed for concentration of major VFA and lactate by HPLC (Waters Corp., Milford,
MA) according to Oba and Allen (2003a).
Dry matter and nutrient intakes were calculated using the composition of feed offered and
refused. Ruminal pool sizes (kg) of OM, NDF, iNDF, pdNDF, starch, MN, and NANMN were
determined by multiplying the concentration of each component in rumen samples by the
ruminal digesta DM mass (kg). Duodenal flows (kg/d) of DM, OM, total NDF, pdNDF, starch,
MN, NANMN, and ammonia N were determined using iNDF as a flow marker; iNDF intake
(kg/d) was multiplied by the ratio between the component and iNDF in duodenal digesta.
Duodenal flow of microbial OM was determined using the ratio of purines to OM (Oba and
Allen, 2003a), and true ruminally degraded OM was calculated by subtracting duodenal flow of
nonmicrobial OM from OM intake. Indigestible NDF was used as an internal marker to estimate
nutrient digestibility in the rumen and in the total tract (Cochran et al., 1986). Turnover rate in
the rumen, passage rate from the rumen, and ruminal digestion rate of each component was
calculated by the following equations:
Turnover rate (%/h) = 100 x (Intake of component / Ruminal pool of component) / 24
Passage rate (%/h) = 100 x (Duodenal flow of component / Ruminal pool of component) / 24
Digestion rate (%/h) = Turnover rate in the rumen (%/h) – Passage rate from the rumen (%/h).
Manually observed chewing activity was summarized by a logic script in Igor Pro
(Version 6.12, WaveMetrics Inc., Lake Oswego, OR) to generate meal and rumination bout
information according to previously established criteria (Dado and Allen, 1994). Variables

90

determined included frequency of meal bouts per day, interval between meals, frequency of
ruminating bouts per day, interval between ruminating bouts, eating time per day, ruminating
time per day, and total chewing time per day.
Statistical Analysis
All data were analyzed by using the fit model procedure of JMP (Version 8, SAS
Institute, Cary, NC). To determine differences between treatments and evaluate interactions of
treatment with DMI, where pDMI (calculated as the mean of DMI values on d 11 to 14 of the 14d preliminary period) was used as the covariate for treatment responses, data were analyzed
2

according to the following model: Yijk = µ + Ci + Pj + Tk + PTjk + pDMI + TkpDMI + pDMI +
2

TkpDMI + eijk where µ is the overall mean, Ci is the random effect of cow (i = 1 to 15), Pj is
the fixed effect of period (j = 1 to 2), Tk is the fixed effect of treatment (k = 1 to 2), PTjk is the
interaction of period and treatment, pDMI is the linear effect of pDMI, TkpDMI is the interaction
2

2

of treatment and pDMI (linear), pDMI is the quadratic effect of pDMI, TkpDMI is the
interaction of treatment and pDMI (quadratic), and eijk is the residual error. Statistical
2

significance for TkpDMI and TkpDMI indicated treatment differences were related to pDMI.
Covariate and interaction terms were removed stepwise from the model if P > 0.20. Treatment
effects and their interaction (linear and quadratic relationships) were declared significant at P "
0.05 and P " 0.10, respectively. Tendencies for treatment effects and their interactions were
declared at P " 0.10 and P " 0.15, respectively.

91

Sixteen cows started the experiment, however, one cow went off feed during the second
experimental period 2 d prior to the start of sample collection and was removed. Thus, data from
15 cows were statistically analyzed.

RESULTS AND DISCUSSION
Comparison of Forages and Diets
Physical characteristics of OG are listed in Table 14. Forages chopped to a TLC of 19
and 10 mm had mean particle sizes of 15.3 and 11.3 mm, respectively. The proportion of
particles > 19 mm was 20 percentage units higher (46 vs. 26%) and particles < 8 mm was 17
percentage units lower (25 vs. 44%) for long than short cut OG, respectively. Particle size
distribution of short cut OG was similar to the guidelines recommended by Heinrichs (1996) that
the portion of haylage retained on the 19 mm sieve, the 8 mm sieve, and the bottom pan of the
Penn State Particle Separator should be 15-25%, 30-40%, and 40-50%, respectively.
Chemical analyses (Table 14) showed that OG with different lengths of cut had similar
concentrations of OM, total NDF, pdNDF, iNDF, ADF, ADL, CP, and starch. Both silages were
wetter than expected, and long cut OG had lower DM concentration than short cut OG due to the
shorter wilting time for long cut OG as the silages were sequentially harvested and long cut OG
was mowed, chopped, and ensiled first. Both OG silages underwent favorable fermentation and
were well preserved based on the low pH and high lactic acid concentrations. However, the
concentrations of acetic acid were higher than that typical for grass silages, which is likely due to
the high moisture content of both OG (Kung and Shaver, 2001).
Diet ingredients and chemical composition are shown in Table 15. The preliminary diet
contained similar proportions of forage NDF from long and short cut OG. Both treatment diets

92

had a 50:50 forage:concentrate ratio, contained 23% forage NDF and had similar chemical
composition, which was mathematically calculated according to the proportion of each feed
ingredient in the diet and its respective analytical values. The calculated percent forage NDF in
the diet was slightly higher than the formulated target but was similar for both LONG and
SHORT and above NRC (2001) minimum requirements. In both diets, forage NDF provided
over 82% of the total diet NDF. Differences in DM concentration in diets were because of the
different DM concentrations of the forages.
Effects of Grass FPL and pDMI
Forage particle length and its interaction with pDMI did not affect yields of milk or milk
components or milk composition (Table 16). LONG increased MUN concentrations (P < 0.001;
Table 16) compared to SHORT. This is consistent with higher ruminal ammonia concentration
and flow of ammonia N to the duodenum (P < 0.01; Table 17) for LONG than SHORT.
Although long cut OG silage was wetter, similar ammonia concentrations (Table 14) were
measured in both silages and, therefore, long cut OG silage was not the source for increased
ammonia observed in cows fed LONG.
LONG tended to increase starch ruminal rate of digestion (19.0 vs.14.9%/h, P = 0.07,
Table 18) and true ruminal digestibility (60.3 vs. 49.8%, P = 0.09, Table 19) and tended to
decrease starch flux from the rumen to the duodenum (2.84 vs. 3.51 kg/d, P = 0.09, Table 19)
compared to SHORT. The mechanism for increased starch digestion rate is unclear but one
explanation is that longer forage particles may promote greater numbers or activity of starch
digesting bacteria in the rumens of cows consuming LONG. Some starch-digesting bacteria in
the rumen (e.g. Streptococcus bovis) also have high proteolytic activity (Russell et al., 1981)
resulting in deamination of amino acids and production of ammonia, which could contribute to

93

the increased ammonia concentrations for LONG.
LONG tended to decrease DMI (21.8 vs. 22.7 kg/d, P = 0.06, Table 16) compared to
SHORT. We expected LONG to be more filling than SHORT causing greater rumen distention
and potentially limiting DMI particularly in cows with high DMI for which ruminal distention is
more likely to limit feed intake (Allen, 1996). Oba and Allen (1999) and Voelker et al. (2002)
found DMI responses varied by production level, which is generally correlated with DMI (NRC,
2001), where DMI was increasingly limited by high fill diets compared to low fill diets as milk
yield increased. Although we expected the longer particles of LONG to slow rates of ruminal
passage, FPL and its interaction with pDMI did not affect the rates that pdNDF, iNDF, or starch
passed from the rumen (Table 18). However, rate of ruminal digestion of pdNDF was related to
pDMI, which decreased at a faster rate for LONG than SHORT as pDMI increased (interaction P
= 0.08, Figure 9A). As a result, LONG decreased or tended to decrease rates of ruminal turnover
of pdNDF, NDF (interaction P = 0.07, Figure 9B), OM, and DM (Table 18) and increased rumen
pools of NDF (interaction P = 0.04, Figure 9C), OM, and DM (Table 20) compared to SHORT
for cows with high pDMI, but effect of treatment on DMI was not related to pDMI. Although
the treatment by pDMI interaction was not significant (interaction P > 0.40), a visual
examination of a graph with pDMI and DMI (Figure 9D) illustrated the differences in DMI
between LONG and SHORT were small for cows with low pDMI but the difference became
greater as pDMI increased with the greatest divergence for cows with high pDMI. Therefore,
rumen fill as a constraint limiting DMI for cows with high intake fed LONG compared to
SHORT is possible but results did not provide conclusive evidence.
Chewing activity results (Table 21) suggest that DMI for LONG was possibly limited by
chewing time. LONG increased meal length (39.1 vs. 33.0 min/bout, P = 0.008) and meal size

94

(2.52 vs. 2.28 kg DM/meal, P = 0.05) resulting in greater eating time (16.5 vs. 15.1 min/kg DMI,
P = 0.02) and ruminating time (25.4 vs. 23.2 min/kg DMI, P = 0.05) and thus increased total
chewing time (42.0 vs. 38.3 min/kg DMI and 867 vs. 827 min/d, P = 0.02) compared to SHORT.
LONG increased total time spent chewing to reduce particle size compared to SHORT, but the
effect of FPL on eating time was related to pDMI. LONG increased and SHORT decreased
eating time expressed as min/kg DMI, min/kg NDF intake (interaction P = 0.006, Figure 9E),
and min/kg forage NDF intake as pDMI increased. LONG increased eating time expressed as
min/d with increasing pDMI compared to SHORT, which remained constant for SHORT across
the range of pDMI (interaction P = 0.004, Figure 9F). Because the total amount of time spent
chewing per day is likely limited (Van Soest, 1994), cows with high intake consuming LONG
might have reached the upper limit for time spent chewing. In this study, LONG had greater
time chewing (mean = 867 min/d) compared to the mean chewing time (694 min/d) across 72
treatments and 19 experiments (Tafaj et al. 2007), and total chewing time for individual cows
ranged from 735 to 1055 min/d. Cows consuming LONG spent 42 min/kg DM and 152 min/kg
NDF, which were approaching the maximum chewing time per unit of DM and NDF (47 and
160 min, respectively) reported by Tafaj et al. (2007).
Forage particle length and its interaction with pDMI did not affect ruminal pH (Table
22). We expected LONG to potentially increase ruminal pH through greater chewing and
salivary buffer flow (Allen, 1997), but this was not observed. This might be because there were
no main effects of treatment on rumen pool sizes (Table 20), total rumination time (Table 21), or
OM truly digested in the rumen (Table 23). Differences were detected for concentrations of
VFA (Table 22), but these differences were quite small and likely not biologically significant.
LONG decreased rumen empty BW for cows with pDMI < 26 kg/d but increased BW for

95

cows with pDMI > 26 kg/d compared to SHORT (interaction P = 0.02, Table 16). The BW gain
for LONG with high pDMI occurred despite digesting less DM in total tract than SHORT
(interaction P = 0.04, Table 23). The reason for the BW changes observed in relation to FPL and
pDMI is not clear.
Direct comparisons of animal responses across individual studies evaluating the effects of
FPL should be interpreted with caution. There are multiple reasons for this, which may help
explain why results from particular experiments may or may not be in agreement. There is a
wide range of FPL (2 to 32 mm) reported for studies from 1997 to 2005 (Tafaj et al., 2007) and
differences between FPL compared within studies (6 vs. 8 mm; Yang and Beauchemin, 2004 and
24 vs. 170 mm; Randby et al., 2008). Furthermore, the lack of a consistent method of measuring
and reporting physical characteristics complicates interpretation because similar TLC and
physically effective NDF do not necessarily yield the same particle size distribution
(Beauchemin and Rodes, 1998). Responses to FPL vary depending on forage source with greater
differences reported for grass-based TMR compared to corn silage-based TMR (Tafaj et al.,
2007). Additionally, feeding conditions differ from offering forage and concentrate separately
and limit fed (Zebeli et al., 2007) to numerous studies where cows are fed total mixed rations ad
libitum. Studies often evaluate effect of FPL in combination with other dietary factors including,
but not limited to, forage:concentrate ratio (Soita et al., 2000; Einarson et al., 2004), grain
processing (Yang et al., 2001), grain fermentability (Krause and Combs, 2003), non-forage fiber
sources (Mooney and Allen, 1997), and supplemental fat (Onetti et al., 2003).
Effects of pDMI on Ruminal Passage Rates
Experimental data on rates of passage from the rumen, particularly for individual feed
fractions, are scarce. Given the impact of passage on ruminal degradation and microbial growth,

96

quantitative knowledge on rates of nutrient passage from the rumen are needed to better
understand nutrient availability in ruminants and improve nutrition models. Furthermore,
because passage rates from the rumen generally increase with increased intake, measurements of
ruminal passage rates of nutrients over a wide range of DMI are necessary. We measured the
effects of DMI on rates of passage of nutrients from the rumen using the pool and flux method
(Robinson et al., 1987).
Although we expected ruminal passage rates to increase with pDMI, passage rates of
pdNDF, iNDF, and starch were not related to level of intake independent of or dependent upon
treatment (Table 18). However, rate of starch digestion decreased quadratically (interaction P =
0.02, Figure 10A) and true ruminal digestibility of starch tended to decrease quadratically
(interaction P = 0.13, Figure 10B) as pDMI increased. Two cows with high pDMI (> 29 kg
DM/d) amplified the quadratic effects. The reductions in starch rate of digestion and
digestibility are consistent with increased starch rumen pool (interaction P = 0.11, Figure 10C)
and likely related to the increased liquid dilution rate associated with increased intake (not
measured) decreasing populations of starch degrading microbes. Concentrations of ruminal
ammonia (P = 0.002, Figure 10D) and branch chained VFA (interaction P = 0.05, Figure 10E),
which are derived from degradation of branched chain amino acids, also decreased with
increased pDMI and are consistent with less proteolytic activity in the rumen due to greater
removal of ruminal microbes through passage and lysis. As pDMI increased, ammonia N flux to
the duodenum (P = 0.05, Figure 10F) increased and MUN concentration (P = 0.07, Table 16)
tended to increase.
Additionally, the rate of pdNDF digestion decreased as pDMI increased (Table 18) and
had a greater effect on LONG than SHORT (interaction P = 0.08, Figure 9A) as previously

97

mentioned. The reduction in pdNDF digestion rate is likely related to differences in mechanical
processing (TLC) and mastication, which reduce particle size and increase the surface area
available for microbial attachment and enzymatic attack (Bowman and Firkins, 1993). The
proportion of large particles in the rumen increases with higher DMI (Okine and Mathison,
1991) because ruminating time per unit of DM consumed decreases as DMI increases (Welch
and Smith, 1969). In this experiment, cows tended to decrease amount of time spent ruminating
per unit of forage NDF consumed as level of intake increased (P = 0.06, Figure 11A). As a
result of the reduction in pdNDF digestion rate, cows tended to decrease ruminal pdNDF
digestibility (%, P = 0.06, Figure 11B) and increase pdNDF flux from the rumen to the
duodenum (P = 0.02, Figure 11C) as pDMI increased. Furthermore, ruminal acetate
concentration decreased as pDMI increased (P = 0.02, Figure 11D), and this is consistent with
lower ruminal pdNDF digestibility because acetate is the predominant VFA produced from fiber
digestion.
Effects of pDMI on N Flux and Microbial Efficiency
Fluxes of NAN and NANMN from the rumen to the duodenum tended to increase as
pDMI increased, which increased or tended to increase postruminal NAN digestion (g/d) and
digestibility (%) with increased pDMI (P " 0.10, Table 17). Microbial N flux and efficiency (g
of MN produced per kg true ruminally digested OM) were not related to level of intake (Table
17) or quantity of OM truly digested in the rumen (not shown). However, a positive relationship
2

was observed between microbial efficiency and ruminal passage rates of starch (P < 0.001, R =
2

0.44, Figure 12A) and pdNDF (P < 0.001, R = 0.52, Figure 12B). This indicated that energy
from ruminal fermentation was more efficiently utilized for microbial growth as passage rates for
starch and pdNDF increased, and the greater passage rates possibly decreased microbial lysis and

98

turnover in the rumen because many microbial organisms flow from the rumen attached to
fibrous particles.
Microbial N flux and efficiency are low for both treatments in the present experiment.
This is unlikely associated with the method used because this same method was used in other
experiments that reported higher MN flux and efficiency (Oba and Allen, 2003b; Taylor and
Allen, 2005). Microbial N flux has not been consistent among studies comparing orchardgrass
and alfalfa with MN flux ranging from low (Voelker Linton and Allen, 2009) to high (Kammes
and Allen, submitted). This indicates low microbial yield is not specific to orchardgrass. The
reason for low MN production in this study is not clear but appears to be related to the
treatments.

CONCLUSIONS
Grass particle length and its interaction with pDMI did not affect milk yield, milk
composition, or rumen pH. Increasing grass particle length tended to decrease DMI, which
might be limited by rumen fill or chewing time or both. Passage rates of feed fractions did not
differ between LONG and SHORT and were not related to level of intake. As pDMI increased,
LONG decreased ruminal digestion rate of pdNDF at a faster rate than SHORT. As a result,
LONG decreased or tended to decrease rates of ruminal turnover for NDF, OM, and DM and
increased their rumen pools compared to SHORT for cows with high pDMI. LONG increased
eating time, which affected cows with high intake to the greatest extent, and total chewing time
compared to SHORT. Ruminal starch digestibility decreased, starch rumen pool increased, and
post ruminal starch digestibility increased quadratically as feed intake increased. Sorting of feed
particles was minimal in this experiment due to the wet forages and individual feeding of cows,

99

but sorting would likely increase for cows fed diets using drier forages or group fed. When grass
silage was the only source of forage in diet, increasing chop length from 10 to 19 mm tended to
decrease DMI but did not negatively affect productivity of cows, which were fed adequate fiber.

100

APPENDIX

101

Table 13. Characterization of 15 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet
Standard
Parameter
Median
Mean
deviation
Minimum
Maximum
Parity
3
3.1
1.2
2
6
1
581
579
58
469
687
BW , kg
BCS
2.1
2.3
0.7
1.5
3.7
DIM
168
164
56
83
258
Milk, kg/d
43.1
41.5
10.6
24.2
62.2
3.5% FCM, kg/d
40.6
41.9
8.9
29.2
56.9
DMI, kg/d
25.4
25.8
2.1
22.6
29.8
1
Empty BW (ruminal digesta removed).

102

Table 14. Chemical composition, particle size distribution, and fermentation parameters of the
long (19 mm) and short cut (10 mm) orchardgrass silage included in the treatment diets
Orchardgrass silage
Item
Long
Short
Chemical composition
DM, %
26.5
31.4
OM, % DM
88.2
88.4
NDF, % DM
46.9
45.2
1
12.2
11.6
iNDF , % DM
iNDF, % of NDF
26.1
25.7
ADF, % of DM
31.3
32.1
2
2.93
2.99
ADL , % of DM
CP, % DM
21.1
20.5
Starch, % DM
1.21
1.24
3
Particles size distribution
Wet sieving, % DM retained
19.0 mm
30.6
27.1
9.50 mm
28.4
24.4
4.75 mm
15.5
28.1
2.36 mm
5.81
14.3
1.18 mm
3.24
4.57
0.600 mm
2.19
2.33
0.300 mm
1.09
1.41
0.150 mm
0.57
0.72
0.075 mm
0.34
0.37
0.038 mm
0.36
0.34
4
15.3
11.3
Mean particle size , mm
5

Penn State Particle Separator ,
% DM retained
> 19.0 mm
19.0 to 8.0 mm
< 8.0 mm
Fermentation
pH
Acetic acid, % DM
Propionic acid, % DM
Butyric acid, % DM
Lactic acid, % DM
Lactic:Acetic
Ethanol, % DM
Ammonia, mM
1
iNDF = indigestible NDF.

46.1
29.0
24.9

26.2
31.7
42.1

4.63
3.36
0.12
<0.01
10.5
3.14
0.12
5.09

4.69
5.26
0.49
0.09
10.6
2.01
0.40
4.84

2

ADL = acid detergent lignin.

3

Particle size distributions of silages were measured each period (n = 2).
103

Table 14 (cont’d)
4

Mean particle size calculated from particle size distribution determined by wet sieving.

5

Silages were dried to constant weight with forced air (no heat) prior to separation using Penn
State Particle Separator due to high moisture content.

104

Table 15. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the sole source of
forage
Preliminary
Long
Short
Ingredients, % DM
Orchardgrass silage, long cut
26.0
49.8
--Orchardgrass silage, short cut
26.0
--49.7
Dry ground corn
36.9
38.4
37.9
Soybean meal (48% CP)
1.99
2.89
3.45
®
3.47
3.39
3.39
SoyPlus
1
3.97
3.97
3.97
Vitamin mineral mix

1

Limestone
Chemical composition
DM, %
OM, % DM
NDF, % DM
% forage NDF
% NDF from forage
2
iNDF , % DM
iNDF, % of NDF
CP, % DM
Starch, % DM

1.60

1.60

1.60

41.7
89.9
28.5
23.8
83.4
6.25

40.8
90.0
28.3
23.3
82.6
8.20

46.5
90.1
27.4
22.5
82.1
7.90

21.9
17.1
29.2

29.0
17.9
29.8

28.8
17.9
29.5

Vitamin mineral mix contained (DM basis) 16.5% sodium bicarbonate, 14.2% magnesium
sulfate, 7.1% salt, 5.8% dicalcium phosphate, 2.4% trace mineral premix, 0.4% vitamin A, 0.4%
vitamin D, 0.2% vitamin E, and 53.1% dry ground corn as a carrier.
2
iNDF = indigestible NDF.

105

Table 16. Milk production and composition, feed intake, and BW change of cows fed treatment diets containing either long (19 mm)
or short cut (10 mm) orchardgrass silage as the sole source of forage
1

2

Treatment LSM
Long
Short
Yield, kg/d
Milk

1

FCM (3.5 %)
Milk fat
Milk protein
Milk lactose
SNF
Milk composition, %
Fat
Protein
Lactose
SNF
MUN, mg/dl
DMI, kg/d
3.5% FCM/ DMI
BW change, kg/18 d
BCS change/18 d

P
SE

Trt

Trt x
Period

39.9

40.0

2.3

0.80

39.8
1.40
1.17
1.88
2.26

39.5
1.40
1.18
1.89
2.27

1.5
0.05
0.06
0.12
0.14

3.61
3.04
4.68
5.62
12.3
21.8
1.64
6.88
-0.08

3.68
3.06
4.67
5.61
11.2
22.7
1.60
2.57
-0.06

0.12
0.06
0.07
0.08
0.3
0.9
0.06
3.91
0.05

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

<0.001

0.59

0.71

0.15

0.62
0.99
0.57
0.77
0.73

NS
NS
NS
NS
NS
NS

<0.001
0.004
0.006
<0.001
<0.001

NS
NS
0.71
0.42
0.41

NS
NS
0.93
0.68
0.68

NS
NS
0.18
0.18
0.16

0.22
0.44
0.75
0.74
<0.001
0.06
0.39
0.54
0.84

NS
NS
NS
NS
NS
NS
NS
NS
NS

0.03
0.001
0.11
0.16
0.07
0.02
0.01
0.88
NS

NS
NS
NS
NS
NS
NS
0.42
0.02
NS

NS
NS
NS
NS
NS
0.20
0.21
0.46
NS

NS
NS
NS
NS
NS
NS
0.09
0.16
NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

106

Table 17. Nitrogen metabolism of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage
as the sole source of forage
1

Treatment LSM
Long
Short
N intake, g/d

620

Ruminal ammonia, mg/dl
Flow to duodenum
Ammonia N, g/d
NAN
g/d
% of N intake
4

NANMN
g/d
% of N intake
% of duodenal NAN
Microbial N
g/d
% of duodenal NAN
5

8.11

6.47

SE

Trt
0.02

25

Trt x
Period
3

0.26

<0.001

NS
NS

pDMI

Trt x
pDMI

pDMI x
pDMI

0.05

NS

0.15

Trt x
pDMI x
pDMI
NS

0.002

NS

0.15

NS

12.5

10.7

0.8

0.004

NS

0.05

NS

0.16

NS

523
80.3

514
75.6

40
3.0

0.64
0.14

NS
NS

0.06
0.19

NS
NS

0.18
NS

NS
NS

296
45.2
52.0

269
42.3
51.6

27
3.1
1.9

0.18
0.24
0.83

NS
NS
NS

0.10
0.20
NS

0.23
NS
NS

0.09
0.13
NS

0.18
NS
NS

230
48.0
19.7

230
48.4
18.8

15
1.9
1.3

0.98
0.83
0.55

NS
NS
NS

0.11
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

295
45.0
56.8

29
3.2
2.1

0.69
0.37
0.87

NS
NS
NS

0.05
0.07
0.12

NS
NS
NS

0.11
0.10
0.05

NS
NS
NS

431
65.8

15
1.3

0.008
0.24

NS
NS

0.03
NS

0.12
NS

0.06
NS

NS
NS

g/kg TRDOM
NAN apparent postruminal digestion
g/d
302
% of N intake
47.8
% of duodenal passage
57.2
N apparent total tract digestion
g/d
399
%
63.9

1

650

2

P

Treatment least squares means.

107

Table 17 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

NANMN = nonammonia, nonmicrobial nitrogen.

5

TRDOM = true ruminally digested organic matter.

108

Table 18. Rumen kinetics of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the
sole source of forage
1

Treatment LSM
Long
Short
Ruminal turnover rate, %/h
DM
OM
NDF

4

pdNDF
Starch
Ruminal turnover time, h
DM
OM
NDF
pdNDF
5

1

iNDF
Starch
Ruminal passage rate, %/h
pdNDF
iNDF
Starch
Ruminal digestion rate, %/h
pdNDF
Starch

2

P
SE

Trt

Trt x
Period
3

7.55

7.93

0.37

0.29

7.62
4.40
7.31

8.03
4.44
7.04

0.39
0.22
0.36

0.27
0.83
0.48

NS
NS
NS
NS

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.34

0.11

NS

NS

0.34
0.18
0.35

0.08
0.07
0.11

NS
NS
NS

NS
NS
NS

32.9

31.0

2.6

0.49

NS

0.16

0.89

0.46

0.09

13.6
13.5
23.6
14.1
48.2

13.0
12.9
23.3
14.8
45.2

0.6
0.6
1.1
0.7
2.6

0.32
0.31
0.83
0.37
0.38

NS
NS
NS
NS
NS

0.29
0.28
0.14
0.28
0.08

0.07
0.05
0.05
0.13
0.09

NS
NS
NS
NS
NS

NS
NS
NS
NS
NS

3.12

3.43

0.25

0.21

NS

0.29

0.85

0.48

0.06

0.89
2.23
14.3

0.75
2.29
15.8

0.22
0.13
1.9

0.53
0.70
0.41

NS
NS
NS

0.12
0.14
0.50

0.45
NS
NS

0.45
NS
0.13

0.19
NS
NS

6.63
19.0

6.28
14.9

0.29
1.8

0.38
0.07

NS
NS

0.04
0.01

0.08
0.34

NS
0.01

NS
0.02

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
109

Table 18 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

110

Table 19. Starch digestion of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the
sole source of forage
1

Treatment LSM
Long
Short
Starch
Intake, kg/d

6.49

1

Apparent ruminal digestion
kg/d
3.91
%
58.9
True ruminal digestion
kg/d
4.01
%
60.3
Passage to duodenum, kg/d
2.84
Apparent postruminal digestion
kg/d
2.53
% of intake
36.5
% of duodenal passage
89.2
Apparent total tract digestion
kg/d
6.19
%
95.4

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

6.62

0.22

0.41

NS

3

0.02

NS

NS

NS

3.31
48.4

0.34
4.9

0.19
0.09

NS
NS

0.80
0.10

0.23
0.30

0.10
0.02

0.15
0.14

3.40
49.8
3.51

0.34
4.8
0.35

0.19
0.09
0.09

NS
NS
NS

0.75
0.10
0.03

0.22
0.30
0.34

0.10
0.02
0.02

0.14
0.13
0.13

3.19
47.0
89.9

0.34
4.8
1.3

0.11
0.11
0.63

NS
NS
NS

0.03
0.10
0.39

0.43
0.40
NS

0.02
0.02
0.02

0.16
0.17
NS

6.32
95.5

0.20
0.4

0.38
0.76

NS
NS

0.02
0.54

NS
0.15

NS
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

111

Table 20. Rumen pools of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the
sole source of forage
1

2

Treatment LSM
Long
Short

SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

Wet weight, kg

94.3

92.5

3.4

0.47

0.17

0.18

NS

Volume, L
Density, kg/L
Rumen pool, kg
DM
OM
NDF

108
0.87

114
0.84

5
0.02

0.11
0.30

0.19
NS

0.09
NS

NS
0.09
NS

Trt x
pDMI x
pDMI
NS

0.47
NS

0.07
NS

12.4
11.1
5.64
2.43

12.3
11.0
5.67
2.58

0.6
0.5
0.22
0.12

0.89
0.83
0.89
0.32

0.13
0.14
NS
NS

0.03
0.03
0.004
0.02

0.09
0.05
0.04
NS

0.17
0.17
NS
NS

NS
NS
NS
NS

3.22

3.10

0.14

0.48

0.17

0.03

0.12

NS

NS

0.87

0.97

0.06

0.17

NS

0.008

0.84

0.14

0.11

pdNDF

4

5

1

iNDF
Starch

P

3

pDMI x
pDMI

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

112

Table 21. Chewing activity of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the
sole source of forage
1

Treatment LSM
Long
Short
Meals
Bouts/d
Length, min/bout
Interval, min
Meal size, kg
DM
OM
NDF
4
pdNDF
5

iNDF
Starch
Eating time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake
Rumination
Bouts/d
Length, min/bout
Interval, min
Ruminating time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake

9.28
39.1
138

10.2
33.0
123

2

P
SE

Trt

Trt x
Period

0.67

0.13

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS
NS
NS

0.46

0.49

0.61

0.08

0.008
0.14

2.3
10

pDMI

0.85
0.67

0.007
0.38

0.98
0.95

0.03
0.09

3

2.52
2.26
0.70
0.50

2.28
2.06
0.62
0.44

0.24
0.21
0.06
0.04

0.05
0.05
0.02
0.02

NS
NS
NS
NS

0.52
0.51
0.57
0.55

0.58
0.58
0.50
0.62

0.69
0.70
0.65
0.64

0.01
0.01
0.01
0.01

0.20

0.17

0.02

0.04

NS

0.61

0.38

0.67

0.03

0.78

0.69

0.07

0.03

NS

0.41

0.75

0.72

0.02

342
16.5
59.9
70.9

326
15.1
55.9
67.2

8
0.5
2.0
2.4

0.08
0.02
0.06
0.13

NS
NS
NS
NS

0.08
0.47
0.67
0.39

0.004
0.006
0.006
0.005

NS
NS
NS
NS

NS
NS
NS
NS

14.5
36.8
55.1

14.1
34.3
57.3

0.9
1.4
3.6

0.44
0.06
0.43

NS
NS
NS

0.45
NS
0.19

0.04
NS
0.41

0.51
NS
0.74

0.05
NS
0.13

525
25.4
92.0
109

502
23.2
86.1
103

13
0.8
3.1
4

0.17
0.05
0.12
0.23

NS
NS
NS
NS

NS
0.07
0.15
0.06

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

113

Table 21 (cont’d)

1

Total chewing time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake

867
42.0
152
180

827
38.3
142
170

16
1.2
5
5

0.02
0.02
0.04
0.12

NS
NS
NS
NS

0.16
0.13
NS
0.10

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

114

Table 22. Ruminal VFA concentrations and pH of cows fed treatment diets containing either long (19 mm) or short cut (10 mm)
orchardgrass silage as the sole source of forage
1

2

Treatment LSM
Long
Short
Total VFA, mM

1

Acetate
Propionate
Butyrate
Lactate
Isobutyrate
Valerate
Isovalerate
Branch chain VFA
Acetate:Propionate
Ruminal pH

SE

Trt

Trt x
Period

152

2

0.62

0.9
1.0
0.8
0.18
0.05
0.07
0.09
0.13
0.06
0.03

0.52
0.11
0.002
0.42
0.08
0.32
0.02
0.01
0.09
0.88

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS

91.8
35.3
18.5
0.18
1.26
1.87
2.15
3.42
2.62
5.84

151
91.3
33.7
19.9
0.34
1.18
1.91
1.98
3.16
2.74
5.84

P

3

pDMI

Trt x
pDMI

pDMI x
pDMI

NS

NS

NS

Trt x
pDMI x
pDMI
NS

0.02
NS
0.92
0.10
0.55
0.54
0.36
0.41
0.03
NS

NS
NS
NS
0.04
0.11
NS
0.78
0.32
0.15
NS

NS
NS
0.15
0.12
0.79
0.19
0.39
0.51
NS
NS

NS
NS
NS
NS
0.18
NS
0.07
0.05
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

115

Table 23. DM and OM digestion of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage
as the sole source of forage
1

2

Treatment LSM
Long
Short
DM
Intake, kg/d

1

Apparent total tract digestion
kg/d
%
OM
Intake, kg/d
Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
Apparent total tract digestion
kg/d
%

P
SE

Trt

Trt x
Period

21.8

22.7

0.9

0.06

NS

15.5
70.6

16.4
71.6

0.6
0.9

0.02
0.42

19.6

20.4

0.8

9.45
47.4

10.2
48.5

11.7
59.2
10.4

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.02

NS

0.20

NS

NS
NS

0.009
0.98

0.04
0.09

0.07
NS

NS
NS

0.06

NS

0.02

NS

0.20

NS

0.41
2.7

0.25
0.64

NS
NS

0.20
0.17

0.16
0.16

NS
0.06

NS
NS

12.4
59.9
10.6

0.4
2.4
0.8

0.22
0.71
0.74

NS
NS
NS

0.08
0.19
0.06

0.15
0.12
0.16

NS
0.04
0.08

NS
NS
NS

5.06
25.4

5.09
25.0

0.55
2.4

0.95
0.87

NS
NS

0.04
0.08

NS
NS

0.02
0.009

NS
NS

14.2
72.0

14.9
72.6

0.5
0.9

0.03
0.56

NS
NS

0.009
0.87

0.04
0.08

0.07
NS

NS
NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).

116

Table 23 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

117

Table 24. NDF digestion of cows fed treatment diets containing either long (19 mm) or short cut (10 mm) orchardgrass silage as the
sole source of forage
1

Treatment LSM
Long
Short
NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
% of intake
Total tract digestion
kg/d
%
Potentially digestible NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
% of intake
Total tract digestion
kg/d
%
Indigestible NDF
Intake, kg/d

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

6.02

6.12

0.24

0.39

NS

3

0.03

NS

0.16

NS

3.85
63.3
2.25

3.91
64.9
2.14

0.14
1.8
0.16

0.63
0.37
0.33

NS
NS
NS

0.18
0.07
0.03

NS
0.16
0.20

0.19
0.49
0.28

NS
0.16
0.08

0.04
0.63

-0.04
-0.23

0.12
2.06

0.60
0.70

NS
NS

0.11
0.10

NS
NS

0.07
0.04

NS
NS

3.89
64.8

3.87
63.8

0.13
1.5

0.87
0.56

NS
NS

0.01
0.99

0.14
NS

0.01
0.04

NS
NS

4.32

4.41

0.16

0.29

NS

0.03

NS

0.14

NS

3.85
90.7
0.52

3.91
89.9
0.44

0.14
1.8
0.12

0.63
0.69
0.47

NS
NS
NS

0.18
0.07
0.05

NS
NS
0.23

0.19
NS
0.44

NS
NS
0.19

0.04
0.91

-0.04
-0.21

0.12
2.89

0.60
0.73

NS
NS

0.11
0.08

NS
NS

0.07
0.04

NS
NS

3.89
90.6

3.87
88.7

0.13
2.0

0.87
0.44

NS
NS

0.01
0.97

0.14
NS

0.01
0.03

NS
NS

1.64

1.65

0.06

0.80

NS

0.12

NS

NS

NS

118

Table 24 (cont’d)
1

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

119

Figure 9

120

Figure 9 (cont’d)

121

Figure 9 (cont’d)

Figure 9. Interaction of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) orchardgrass particle length with preliminary DMI for: A) potential digestible
NDF (pdNDF) ruminal digestion rate (P = 0.08), B) NDF ruminal turnover rate (P = 0.07), C)
NDF rumen pool (P = 0.04), D) DMI (interaction not significant), E) eating time, min/kg NDF
intake (P = 0.006), and F) eating time, min/d (P = 0.004). The preliminary DMI on the x-axis
are the mean DMI of individual cows during the final 4 d of the preliminary period when all
cows were fed a common diet.

122

Figure 10

123

Figure 10 (cont’d)

124

Figure 10 (cont’d)

Figure 10. Relationship of long (19 mm; open circles, dashed line) and short (10 mm; closed
circles, solid line) orchardgrass particle length with preliminary DMI for: A) starch ruminal
digestion rate (interaction P = 0.02), B) starch true ruminal digestibility (interaction P = 0.13), C)
starch rumen pool (interaction P = 0.11), D) ruminal ammonia concentration (P = 0.002), E)
branched chain VFA concentration (interaction P = 0.05), and F) ammonia N flow to duodenum
(P = 0.05). The preliminary DMI on the x-axis are the mean DMI of individual cows during the
final 4 d of the preliminary period when all cows were fed a common diet.

125

Figure 11

126

Figure 11 (cont’d)

Figure 11. Relationship of long (19 mm; open circles) and short (10 mm; closed circles)
orchardgrass particle length with preliminary DMI for: A) ruminating time per unit forage NDF
(P = 0.06), B) potentially digestible NDF (pdNDF) ruminal digestibility (P = 0.07), C) pdNDF
flux to duodenum (P = 0.05), and D) ruminal acetate concentration (P = 0.02). The preliminary
DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the preliminary
period when all cows were fed a common diet.

127

Microbial efficiency,
g of microbial N / kg of TRDOM

35

A

30
25
20
15
10

5

10

15

20

25

30

35

Starch ruminal passage rate, %/h

Microbial efficiency,
g of microbial N / kg of TRDOM

35

B

30
25
20
15
10
-0.5

0

0.5

1

1.5

2

2.5

pdNDF ruminal passage rate, %/h

Figure 12. A) Relationship between starch ruminal passage rate and microbial efficiency.
Microbial efficiency, g of microbial N/kg of true ruminally digested OM (TRDOM) = 11.5 +
2
0.595 x starch ruminal passage rate, %/h (P < 0.001, R = 0.44). B) Relationship between
potentially digestible NDF (pdNDF) ruminal passage rate and microbial efficiency. Microbial
efficiency, g of microbial N/kg of TRDOM = 14.8 + 6.26 x pdNDF ruminal passage rate, %/h
2
(P < 0.001, R = 0.52). Open circles denote long (19 mm) and closed circles denote short (10
mm) orchardgrass particle length. Starch and pdNDF ruminal passage rates were also positively
2
correlated (P < 0.001, R = 0.43, not shown).

128

REFERENCES

129

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Allen, M. S. 1997. Relationship between fermentation acid production in the rumen and the
requirement for physically effective fiber. J. Dairy Sci. 80:1447-1462.
Allen, M. S., and D. R. Mertens. 1988. Evaluating constraints of fiber digestion by rumen
microbes. J. Nutr. 118:261-270.
Beauchemin, K. A., and L. M. Rode. 1998. Effective fibre in barley-based diets. Adv. Dairy
Technol. 10:151-165.
Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64-75.
Bowman, J. G. P., and J. L. Firkins. 1993. Effects of forage species and particle size on bacterial
cellulolytic activity and colonization in situ. J. Anim. Sci. 71:1623-1633.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting digestibility of
different diets with internal markers: Evaluation of four potential markers. J. Anim. Sci.
63:1476-1483.
Dado, R. G., and M. S. Allen. 1994. Variation in and relationships among feeding, chewing,
and drinking variables for lactating dairy cows. J. Dairy Sci. 77:132-144.
Dixon, R. M., and L. P. Milligan. 1985. Removal of digesta components from the rumen of
steers determined by sieving techniques and fluid, particulate and microbial markers. Br.
J. Nutr. 53:347-362.
Einarson, M. S., J. C. Plaizier, and K. M. Wittenberg. 2004. Effects of barley silage chop length
on productivity and rumen conditions of lactating dairy cows fed total mixed ration. J.
Dairy Sci. 87:2987-2996.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.
Grant, R. J. 1997. Interactions among forages and nonforage fiber sources. J. Dairy Sci. 80:14381446.
Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More powerful peroxide
Kjeldahl digestion method. J. AOAC 70:783-787.
Heinrichs, J. 1996. Evaluating particle size of forages and TMRs using the Penn State Particle

130

Size Separator. Technical Bulletin of The Pennsylvania State University, College of
Agriculture Science, Cooperative Extension. DAS 96-20.
Joy, M. T., E. J. DePeters, J. G. Fadel, and R. A. Zinn. 1997. Effects of corn processing on the
site and extent of digestion in lactating cows. J. Dairy Sci. 80:2087-2097.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019-1027.
Krause, K. M., and D. K. Combs. 2003. Effects of forage particle size, forage source, and grain
fermentability on performance and ruminal pH in midlactation cows. J. Dairy Sci.
86:1382-1397.
Kung, L., and R. Shaver. 2001. Interpretation and use of silage fermentation analysis reports.
University of Wisconsin-Extension Publ. Series. Focus on Forage 3(13):1–5.
Lammers, B. P., D. R. Buckmaster, and A. J. Heinrichs. 1996. A simplified method for the
analysis of particle sizes of forage and total mixed rations. J. Dairy Sci. 79:922-928.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:12171240.
Mooney, C. S., and M. S. Allen. 1997. Physical effectiveness of the neutral detergent fiber of
whole linted cottonseed relative to that of alfalfa silage at two lengths of cut. J.
Dairy Sci. 80:2052-2061.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl.
Acad. Washington, DC.
Oba, M. and M. S. Allen. 1999. Effects of brown midrib 3 mutation in corn silage on dry
matter intake and productivity of high yielding dairy cows. J. Dairy Sci. 82:135-142.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174–183.
Oba, M., and M. S. Allen. 2003b. Effects of diet fermentability on efficiency of microbial
nitrogen production in lactating dairy cows. J. Dairy Sci. 86:195-207.
Okine, E. K., and G. W. Mathison. 1991. Effects of feed intake on particle distribution, passage
of digesta, and extent of digestion in the gastrointestinal tract of cattle. J. Anim. Sci.
69:3435-3445.
Onetti, S. G., R. D. Shaver, S. J. Bertics, and R. R. Grummer. 2003. Influence of corn silage

131

particle length on the performance of lactating dairy cows fed supplemental tallow. J.
Dairy Sci. 96:2949-2957.
Randby, Å. T., T. Garmo, M. Eknæs, and E. Prestløkken. Effect of grass silage chop length on
intake and milk production by dairy cows. Grassl. Sci. in Europe. 13:768-770.
Robinson, P. J., S. Tamminga, and A. M. van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci.17:37-62.
Robles, A. Y., R. L. Belyea, F. A. Martz, and M. F. Weiss. 1980. Effect of particle size upon
digestible cell wall and rate of in vitro digestion of alfalfa and orchard grass forages. J.
Anim. Sci. 51:783-790.
Russell, J. B., W. G. Bottje, and M. A. Cotta. 1981. Degradation of protein by mixed cultures of
rumen bacteria: Identification of Streptococcus bovis as an actively proteolytic rumen
bacterium. J. Anim. Sci. 53:242-252.
Smith, L. W., H. K. Goering, and C. H. Gordon. 1972. Relationships of forage compositions with
rates of cell wall digestion and indigestibility of cell walls. J. Dairy Sci. 55:1140-1147.
Soita, H. W., D. A. Christensen, and J. J. McKinnon. 2000. Influence of particle size on the
effectiveness of the fibre in barley silage. J. Dairy Sci. 83:2295-2300.
Tafaj, M., Q. Zebeli, Ch. Baes, H. Steingass, and W. Drochner. 2007. A meta-analysis examining
effects of particle size of total mixed rations on intake, rumen digestion and milk
production in high-yielding dairy cows in early lactation. Anim. Feed Sci. Technol.
138:137-161.
Taylor, C. C., and M. S. Allen. 2005. Corn grain endosperm type and brown midrib 3 corn
silage: Ruminal fermentation and N partitioning in lactating cows. J. Dairy Sci. 88:14341442.
Van Soest, P. J. 1994. Nutritional Ecology of The Ruminant. Second Edition. Cornell University
Press, Ithaca, NY.
Voelker, J. A., G. M. Burato, and M. S. Allen. 2002. Effect of pretrial milk yield on responses
of feed intake, digestion, and production to dietary forage concentration. J. Dairy Sci.
85:2650-2661.
Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Voelker Linton, J. A., and M. S. Allen. 2009. Nutrient demand interacts with forage family to
affect nitrogen digestion and utilization responses in dairy cows. J. Dairy Sci. 92:15941602.

132

Welch, J. G., and A. M. Smith. 1969. Effect of varying amounts of forage intake on rumination.
J. Anim. Sci. 28:827-830.
Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. J. Dairy Sci. 65:495–501.
Yang, W. Z., and K. A. Beauchemin. 2004. Grain processing, forage-to-concentrate ratio, and
forage length effects on ruminal nitrogen degradation and flows of amino acids to the
duodenum. J. Dairy Sci. 87:2578-2590.
Yang, W. Z., K. A. Beauchemin, and M. L. Rode. 2001. Effects of grain processing, forage to
concentrate ratio, and forage particle size on rumen pH and digestion in dairy cows. J.
Dairy Sci. 84: 2203-2216.
Zebeli, Q., M. Tafaj, I. Weber, J. Dijkstra, H. Steingass, and W. Drochner. 2007. Effects of
varying dietary forage particle size in two concentrate levels on chewing activity, ruminal
mat characteristics, and passage in dairy cows. J. Dairy Sci. 90:1929-1942.
Zinn, R. A., and F. N. Owens. 1986. A rapid procedure for purine measurement and its use for
estimating net ruminal protein synthesis. Can. J. Anim. Sci. 66:157-166.

133

CHAPTER 4
Nutrient Demand Interacts with Legume Maturity to Affect Rumen Pool Sizes in Dairy
Cows

1

ABSTRACT
Effects of legume maturity on dry matter intake (DMI), milk production, ruminal
fermentation and pool sizes, and digestion and passage kinetics, and the relationship of these
effects with preliminary DMI (pDMI) were evaluated using 16 ruminally and duodenally
cannulated Holstein cows in a crossover design with a 14-d preliminary period and two 17-d
treatment periods. During the preliminary period, pDMI of individual cows ranged from 22.9 to
30.0 kg/d (mean = 25.9 kg/d) and 3.5% fat-corrected milk (FCM) yield ranged from 34.1 to 68.2
kg/d (mean = 43.7 kg/d). Experimental treatments were diets containing alfalfa silage harvested
either a) early cut, less mature (EARLY) or b) late cut, more mature (LATE) as the sole forage.
Early and late cut alfalfa contained 40.8 and 53.1% neutral detergent fiber (NDF) and 23.7 and
18.1% crude protein, respectively. Forage:concentrate ratios were 53:47 and 42:58 for EARLY
and LATE, respectively; both diets contained ~22% forage NDF and 27% total NDF.
Preliminary DMI, an index of nutrient demand, was determined during the last 4 d of the
preliminary period when cows were fed a common diet and used as a covariate. Main effects of
alfalfa maturity and their interaction with pDMI were tested by ANOVA. Alfalfa maturity and
its interaction with pDMI did not affect milk yield but EARLY increased DMI compared with
LATE; thus, EARLY had lower efficiency of milk production than LATE. EARLY decreased
milk fat concentration more per kilogram pDMI increase than LATE, but milk fat yield was not
1

Accepted for publication in the Journal of Dairy Science on December 11, 2011.

134

affected. The lower concentration and faster passage rate of indigestible NDF (iNDF) for
EARLY resulted in lower rumen pools of iNDF, total NDF, and dry matter for EARLY than
LATE, which EARLY increased at a slower rate than LATE as pDMI increased. EARLY
decreased starch intake and increased ruminal pH compared with LATE. Rate of ruminal starch
digestion was related to level of intake, but this did not affect ruminal or postruminal starch
digestion. Total tract digestibility of NDF, organic matter, and dry matter was higher for
EARLY than LATE. Microbial efficiency tended to decrease for EARLY and increase for
LATE as pDMI increased. When alfalfa silage was the only source of forage in the diet, cows
supplemented with additional concentrate to account for decreased protein and increased fiber
concentrations associated with more mature alfalfa produced similar FCM yields with greater
efficiency than cows fed less mature alfalfa.

INTRODUCTION
The supply and utilization of nutrients in dairy cows fed forage-based diets are affected
by forage maturity at harvest. Generally, postponing harvest increases DM yield, but nutritive
quality declines as the concentration of NDF increases and its digestibility decreases because of
greater lignification. Dairy cows fed less mature forage generally produce more milk with less
supplemental concentrate than cows fed more mature forage. Thus, harvesting and feeding
quality forages may improve animal performance, reduce purchased feed costs, and increase
profitability.
Fiber concentration and digestibility affect feed intake by contributing to physical fill in
the rumen. As forages mature, the indigestible fraction of fiber increases and the rate of

135

digestion of the potentially digestible fraction decreases (Smith et al., 1972). Additionally,
highly lignified (i.e. more mature) forages increase retention time in the rumen due to a slow rate
of digestion (Allen, 2000). This suggests that more mature forages result in slower ruminal
turnover and greater ruminal distention. Therefore, the degree to which rumen fill limits feed
intake increases when dairy cows are fed more mature forage compared with less mature forage.
Alfalfa (AL; Medicago sativa) was selected as a representative legume for use in this experiment
because it is the predominant legume fed to dairy cows in the United States.
In addition to dietary factors affecting ruminal passage, digestion, and distention, the
individual cow’s appetite will also affect the responses of passage rate and intake to forage
maturity. Cows respond differently to treatments depending on their level of intake (Voelker
Linton and Allen, 2008) and production (Oba and Allen, 1999). Because legume maturity and
level of intake affect ruminal passage and digestion rates and, thus, the physical filling effects in
the rumen, the response to effects of legume maturity and its relationship with intake level
should be assessed to determine if responses to treatment vary among cows with a wide range in
DMI. We hypothesized that responses of DMI and digesta passage rates to legume maturity are
related to level of intake and earlier maturity will permit a greater increase in passage rate than
later maturity as feed intake increases.
The objectives of this experiment were to evaluate the relationships between voluntary
DMI and effects of legume maturity on DMI, milk production, ruminal fermentation and pool
sizes, and digestion and passage kinetics in lactating dairy cows. This study had three distinctive
features to improve our understanding of the role of legume maturity and interpret its effect on
animal responses. First, it allowed effects of the interaction between legume maturity and
preliminary DMI (pDMI) to be evaluated. The use of pDMI, an index of nutrient demand,

136

allowed the evaluation of treatments on animal responses in relation to level of intake and
provided an indicator to test effects of intake level independent of treatments. Second, it directly
compared treatment effects of early and late cut AL as the sole source of forage. Third, ruminal
passage rates of individual feed fractions, instead of entire feeds, were measured using ruminally
and duodenally cannulated cows.

MATERIALS AND METHODS
Cows and Treatments
Experimental procedures were approved by the Institutional Animal Care and Use
Committee at Michigan State University. Sixteen multiparous Holstein cows from the Michigan
State University Dairy Cattle Teaching and Research Center were assigned randomly to
treatment sequence in a crossover design experiment with one 14-d preliminary period and two
17-d experimental periods. The first 10 d of each period were allowed for diet adaptation and
samples were collected during the final 4 d of the preliminary period and 7 d of each
experimental period. Cows were 137±45 (mean±SD) DIM at the end of the preliminary period
and were selected to provide a wide range and uniform distribution of pDMI and milk yield.
During the final 4 d of the 14-d preliminary period, the average pDMI among cows ranged from
22.9 to 30.0 kg/d (mean = 25.9 kg/d) and 3.5% FCM yield ranged from 34.1 to 68.2 kg/d (mean
= 43.7 kg/d; Table 25). Prior to calving, cows were cannulated ruminally (Bar Diamond, Inc.,
Parma, ID) and duodenally with a gutter-type T cannula placed approximately 10 cm distal to the
pylorus (Joy et al., 1997). Surgery was performed at the Department of Large Animal Clinical
Science, College of Veterinary Medicine, Michigan State University.

137

Experimental treatments were diets containing AL silage from the same field harvested
either a) early cut, less mature (EARLY) or b) late cut, more mature (LATE) as the sole forage.
Alfalfa was produced at the campus farm at Michigan State University (East Lansing), harvested
using a New Holland FP230 pull-type forage harvester set according to manufacturer
specifications for a theoretical length of cut of 10 mm, and ensiled in Ag-Bags (Ag-Bag Systems
Inc., St. Nazianz, WI). During the sample collection periods, early and late cut AL contained
40.8 and 53.1% NDF and 23.7 and 18.1% CP, respectively (DM basis; Table 26). Diets EARLY
and LATE were formulated to contain 22% forage NDF, 30% total NDF, and 18% CP. We
acknowledge these treatments affect dietary starch concentration but maintaining similar forage
and total NDF concentrations for both treatments was of primary interest. The diet fed during
the preliminary period was formulated so that early and late cut AL each contributed 50% of
®

forage NDF. Diets also contained dry ground corn, soybean meal (48% CP), SoyPlus (West
Central Soy, Ralston, IA), vitamin-mineral premix, and limestone (Table 27).
Data and Sample Collection
Throughout the experiment, cows were housed in tie-stalls and fed diets as total mixed
rations once daily (1130 h) at 110% of expected intake. The amount of feed offered and refused
(orts) was weighed daily for each cow. Forage samples were collected twice weekly and
analyzed to adjust diets to account for DM, NDF, and CP fluctuation. Samples of all dietary
ingredients (0.5 kg) and orts (12.5%) were collected daily from d 11 to 14 during the preliminary
period and d 11 to 15 during each experimental period. Samples were frozen immediately after
collection at !20°C and combined to one composite sample per period prior to analysis.
Cows were moved to an exercise lot twice daily (0230 and 1300 h) prior to milking in a
parlor (0400 and 1430 h). Milk yield was measured, and milk was sampled, at each milking on d

138

11 to 14 of the preliminary period and on d 11 to 15 of the experimental periods. Rumen-empty
BW was measured by weighing the cow after evacuation of ruminal digesta on d 14 of the
preliminary period and d 17 of each experimental period. Body condition score was determined
on the same days by 4 trained investigators blinded to treatments (Wildman et al., 1982; 5-point
scale where 1 = thin and 5 = fat).
Duodenal samples (900 mL), fecal samples (500 g), rumen fluid and particulate samples
for microbial isolation (400 g), and rumen fluid samples for pH, concentrations of VFA, lactate,
and ammonia (100 mL) were collected every 15 h from d 11 to 15 of each experimental period
so that 8 samples were taken for each cow in each period, representing every 3 h of a 24-h period
to account for diurnal variation. Rumen fluid and particulate matter for microbial isolation was
collected from the reticulum, near the reticular-omasal orifice, transported to the laboratory, and
processed. Rumen fluid for pH, VFA, lactate, and ammonia was obtained by combining digesta
from five different sites in the rumen and straining it through nylon mesh (~1 mm pore size);
fluid pH was recorded immediately. Samples were stored at -20°C.
Ruminal contents were evacuated manually through the ruminal cannula 4.5 h after
feeding at the beginning of d 15 (1600 h) and 2 h before feeding at the end of d 17 (0930 h) for
each experimental period. Total rumen content mass and volume were determined. To ensure
accurate sampling, every tenth handful of digesta (10%) was separated for a subsample
throughout evacuation. This subsample was squeezed into primarily solid and liquid phases.
Both phases were weighed and sampled (350 mL) for determination of nutrient pool size. All
samples were stored at -20°C.
Sample Analysis and Calculations

139

Milk yield recorded at each milking was summed for a daily total, which were averaged
for each period. Milk samples were analyzed for fat, true protein, lactose, and SNF with infrared
spectroscopy by Michigan DHIA (East Lansing). Yields of 3.5% FCM and milk components
were calculated using milk yield and component concentrations for each milking, summed for a
daily total, and averaged for each period.
Forage samples were combined to one composite sample per forage per period. Particle
size distribution was determined using the Penn State Particle Separator containing 2 sieves (19
and 8 mm) and a pan (Lammers et al., 1996). In addition, samples were wet sieved manually and
sequentially through screens with the following aperture sizes: 19.0, 9.50, 4.75, 2.36, 1.18,
0.600, 0.300, 0.150, 0.075, and 0.038 mm. The fraction of DM retained on the screens from wet
sieving was used to calculate mean particle size.
Diet ingredients, orts, and feces were lyophilized (Tri-Philizer MP, FTS Systems, Stone
Ridge, NY). All dried samples were ground with a Wiley mill (1-mm screen; Arthur H. Thomas,
Philadelphia, PA). Dried, ground fecal samples were combined on an equal DM basis into one
sample per cow per period. Frozen duodenal samples for each cow period (n=8) were chopped
finely using a commercial food processor (84142 Food cutter, Hobart Manufacturing Co., Troy,
OH) and subsampled in the frozen state to obtain representative samples. These duodenal
subsamples and the 350 mL of ruminal solid and liquid samples were lyophilized and ground as
described above. Dried ruminal solid and liquid samples were recombined according to the
original ratio of solid and liquid DM.
Samples were analyzed for ash, NDF, iNDF, ADF, acid detergent lignin (ADL), CP, and
starch. Ash concentration was determined after 5 h combustion at 500°C in a muffle furnace.
Concentrations of NDF were determined according to Mertens (2002) and ADF and ADL

140

according to Goering and Van Soest (1970). Indigestible NDF was estimated as NDF residue
after 240 h in vitro fermentation (Goering and Van Soest, 1970); flasks were reinoculated at 120
h to insure a viable microbial population. Forage NDF digestibility was determined by 30 h in
vitro fermentation (Goering and Van Soest, 1970). Ruminal fluid for the in vitro incubations
was collected from a nonpregnant dry cow fed dry hay only. Fraction of pdNDF was calculated
by difference (1.00 – iNDF). Crude protein was analyzed according to Hach et al. (1987).
Starch was measured by an enzymatic method (Karkalas, 1985) after samples were gelatinized
with sodium hydroxide. Glucose concentration was measured using a glucose oxidase method
(Glucose kit #510, Sigma Chemical Co., St. Louis, MO), and absorbance was determined with a
micro-plate reader (SpectraMax 190, Molecular Devices Corp., Sunnyvale, CA). Concentrations
of all nutrients except DM were expressed as percentages of DM determined by drying at 105°C
in forced-air oven for more than 8 h.
Duodenal digesta were analyzed for purines and ammonia to estimate microbial N (MN)
flow and nonammonia, nonmicrobial N (NANMN) flow to the duodenum. Purine concentration
was used as a microbial marker, and purine to MN ratio was estimated by analysis of microbial
pellets obtained by differential centrifugation of the rumen fluid and particulate samples
collected near the reticulum. Rumen fluid and particulate matter was blended, strained through
nylon mesh, and the liquid portion was centrifuged at 500 x g for 15 min. The supernatant was
centrifuged at 18,000 x g for 15 min, and the pellet was washed with 0.9% NaCl solution and
centrifuged again at 18,000 x g for 15 min, resuspended in water, and lyophilized. Total purines
were measured by spectrophotometer (Beckman Instruments Inc., Fullerton, CA) at 260 nm
according to Zinn and Owens (1986). Ammonia concentration was determined for centrifuged
duodenal and rumen fluid samples according to Broderick and Kang (1980). Rumen fluid was

141

also analyzed for concentration of major VFA and lactate by HPLC (Waters Corp., Milford,
MA) according to Oba and Allen (2003a).
Dry matter and nutrient intakes were calculated using the composition of feed offered and
refused. Ruminal pool sizes (kg) of OM, NDF, iNDF, pdNDF, starch, MN, and NANMN were
determined by multiplying the concentration of each component in rumen samples by the
ruminal digesta DM mass (kg). Duodenal flows (kg/d) of DM, OM, total NDF, pdNDF, starch,
MN, NANMN, and ammonia N were determined using iNDF as a flow marker; iNDF intake
(kg/d) was multiplied by the ratio between the component and iNDF in duodenal digesta.
Duodenal flow of microbial OM was determined using the ratio of purines to OM (Oba and
Allen, 2003a), and true ruminally digested OM was calculated by subtracting duodenal flow of
nonmicrobial OM from OM intake. Indigestible NDF was used as an internal marker to estimate
nutrient digestibility in the rumen and in the total tract (Cochran et al., 1986). Turnover rate in
the rumen, passage rate from the rumen, and ruminal digestion rate of each component was
calculated by the following equations:
Turnover rate (%/h) = 100 x (Intake of component / Ruminal pool of component) / 24
Passage rate (%/h) = 100 x (Duodenal flow of component / Ruminal pool of component) / 24
Digestion rate (%/h) = Turnover rate in the rumen (%/h) – Passage rate from the rumen (%/h).
Statistical Analysis
All data were analyzed by using the fit model procedure of JMP (Version 8, SAS
Institute, Cary, NC). To determine differences between treatments and evaluate interactions of
treatment with DMI, where pDMI (calculated as the mean of DMI values on d 11 to 14 of the 14d preliminary period) was used as the covariate for treatment responses, data were analyzed
2

according to the following model: Yijk = µ + Ci + Pj + Tk + PTjk + pDMI + TkpDMI + pDMI +

142

2

TkpDMI + eijk where µ is the overall mean, Ci is the random effect of cow (i = 1 to 16), Pj is
the fixed effect of period (j = 1 to 2), Tk is the fixed effect of treatment (k = 1 to 2), PTjk is the
interaction of period and treatment, pDMI is the linear effect of pDMI, TkpDMI is the interaction
2

2

of treatment and pDMI (linear), pDMI is the quadratic effect of pDMI, TkpDMI is the
interaction of treatment and pDMI (quadratic), and eijk is the residual error. Statistical
2

significance for TkpDMI and TkpDMI indicated treatment differences were related to pDMI.
Covariate and interaction terms were removed stepwise from the model if P > 0.20. Treatment
effects and their interaction (linear and quadratic relationships) were declared significant at P "
0.05 and P " 0.10, respectively. Tendencies for treatment effects and their interactions were
declared at P " 0.10 and P " 0.15, respectively.

RESULTS AND DISCUSSION
Comparison of Forages and Diets
Chemical analyses and physical characteristics of AL silages are listed in Table 26. As
expected, delaying harvest of AL decreased the concentration of CP (23.7 vs. 18.1%) and
increased concentrations of total NDF (40.8 vs. 53.1%), iNDF (27.7 vs. 39.3%), and ADL (8.41
vs. 11.9%). Indigestible NDF, expressed as a percent of NDF, was lower for early cut AL than
for late cut AL (67.8 vs. 74.1% of NDF), but was unusually high for both silages. In vitro NDF
digestibility (30 h) was 4.4 percentage units higher for early cut AL than for late cut AL (28.8 vs.
24.4%). Early cut AL had lower DM concentration, higher pH, and contained more lactic and
acetic acids than late cut AL. Based on wet sieving, early cut AL had a greater mean particle

143

size than late cut AL (8.36 vs. 6.92 mm). Additionally, early cut AL contained a larger
proportion of particles between 8 and 19 mm (67.3 vs. 35.7%; middle sieve) and a smaller
proportion of particles less than 8 mm (25.5 vs. 58.6%; bottom pan) than late cut AL when
sieved with the Penn State Particle Separator. Although AL was chopped to the same theoretical
length of cut for both silages, the smaller particle size for late cut AL may be due to greater
fragility from increased maturity or greater DM concentration or both.
Diet ingredients and chemical composition are shown in Table 27. The preliminary diet
contained more early cut AL than late cut AL so each forage supplied similar concentrations of
forage NDF. Because treatment diets were formulated to contain similar forage NDF
concentrations, forage:concentrate ratios were different between diets with ratios of 53:47 and
42:58 for EARLY and LATE, respectively. Besides forage source, differences in diets included
source and concentration of protein supplement and concentration of corn grain, which were both
lower for EARLY than LATE, to account for differences between the AL silages. Additionally,
limestone was added to LATE to compensate for lower Ca concentration in more mature AL
silage. The chemical composition of each ration was mathematically calculated according to the
proportion of each feed ingredient in the diet and its respective analytical values, and was similar
between rations for OM, total NDF, and CP concentrations. Based on the calculated diet
composition, both diets contained lower NDF and higher CP concentrations than the formulated
targets but were similar for EARLY and LATE and NDF concentration was above NRC (2001)
minimum requirements. Starch concentration was lower for EARLY because of more forage and
less concentrate in the diet for EARLY compared with LATE. Differences in DM and iNDF
concentrations in diets, which were lower for EARLY than LATE, were reflective of the

144

concentrations in the forages, which were lower for early cut AL than late cut AL. In both diets,
forage NDF provided over 80% of the total diet NDF.
Effects of Legume Maturity and pDMI
EARLY increased DMI (28.6 vs. 26.8 kg/d, P = 0.003, Table 28) and total NDF intake
(7.63 vs. 7.30 kg/d, P = 0.03, Table 29) compared with LATE, but these responses were not
related to level of intake (P # 0.53). We expected LATE to be more filling than EARLY causing
greater rumen distention and potentially limiting DMI, particularly in cows with high DMI for
which ruminal distention is more likely to limit feed intake (Allen, 1996). Oba and Allen (1999)
and Voelker et al. (2002) found DMI responses to a more filling diet varied by production level,
which is generally correlated with DMI (NRC, 2001), where DMI was increasingly limited by
high fill diets compared with low fill diets as milk yield increased. As fill became more limiting,
we expected ruminal passage rate of iNDF to increase more for EARLY than for LATE.
However, response of passage rate of iNDF to treatment was not related to pDMI (P # 0.37), but
EARLY increased ruminal passage rate of iNDF compared with LATE (3.86 vs. 3.57%/h, P =
0.01, Table 30).
The lower concentration and faster rate of passage of iNDF for EARLY is reflected in the
lower rumen pool sizes of iNDF (5.49 vs. 6.24 kg, P < 0.001) and NDF (6.23 vs. 6.96 kg, P <
0.001) for EARLY compared with LATE (Table 31), which increased at a slower rate for
EARLY than LATE as pDMI increased (interaction P " 0.10, Figure 13). It is also partially
responsible for lower rumen pool sizes of OM (10.3 vs. 11.4 kg, P = 0.006) and DM (11.3 vs.
12.4, P = 0.007) for EARLY compared with LATE (Table 31), which tended to increase at a
slower rate for EARLY than LATE as pDMI increased (interaction P " 0.14). Furthermore,
EARLY decreased rumen digesta wet weight (76.9 vs. 80.8 kg, P = 0.04) and volume (91.3 vs.

145

97.8 L, P = 0.03) compared with LATE (Table 31). The faster rate of passage of iNDF and
lower rumen pools for EARLY are consistent with the faster rates of ruminal turnover of NDF,
OM, and DM (5.25 vs. 4.45, 10.8 vs. 9.15, and 10.7 vs. 9.07%/h, respectively, P < 0.001) for
EARLY compared with LATE (Table 30). The decreased feed intake of LATE was
accompanied by increased size of rumen pools suggesting rumen fill as a constraint limiting DMI
for cows consuming LATE is possible, but there was no evidence in this experiment that ruminal
distention is more likely to limit feed intake for cows with high intake compared to cows with
low intake as we were unable to detect a treatment by pDMI interaction for DMI.
Although EARLY increased DMI compared with LATE, forage maturity and its
interaction with pDMI did not affect milk, milk component, or FCM yields (Table 28).
Therefore, EARLY decreased efficiency of milk production (FCM/DMI) compared with LATE
(1.45 vs. 1.51, P = 0.01), as cows consuming EARLY ate more feed than LATE to produce
similar yields of FCM. Response of milk fat concentration to treatment was related to pDMI
(interaction P = 0.09, Table 28) such that EARLY decreased milk fat concentration more per
kilogram pDMI increase than LATE. However, this effect was small and milk fat yield was not
affected. The aforementioned results are likely because of the greater concentration of
supplemental concentrate in the diet for LATE.
The different forage:concentrate ratios in the diets were necessary to account for changes
in chemical composition of AL with increasing maturity and maintain the same concentration of
forage NDF in the diets. As a result, EARLY decreased starch intake (8.16 vs. 8.74 kg/d, P =
0.007, Table 32), decreased ruminal propionate (29.5 vs. 31.0 mM, P = 0.04) and butyrate (16.1
vs. 18.1 mM, P < 0.001) concentrations, and increased ruminal acetate concentration (86.3 vs.
81.9 mM, P = 0.005) compared with LATE (Table 33). EARLY increased the acetate to

146

propionate ratio compared with LATE (2.94 vs. 2.65, P < 0.001), and the difference in ruminal
acetate concentration between EARLY and LATE was greatest for cows with low pDMI
(interaction P = 0.09, Figure 14). These results are reflective of the diets.
EARLY increased mean ruminal pH (6.36 vs. 6.29, P = 0.04, Table 33) compared with
LATE, which was expected because of lower starch intake for EARLY than LATE, but the
reason for the difference in pH is not clear based on our results. Apparent ruminal digestion of
OM was greater for EARLY than LATE (9.76 vs. 7.93 kg/d, P < 0.001, Table 34), which would
suggest greater production of VFA and lower pH for EARLY, but total VFA concentration in the
rumen was similar for EARLY and LATE (P = 0.48, Table 33) while pH was higher for EARLY
than LATE. Additionally, rumen content mass was smaller for EARLY than LATE (Table 31)
suggesting less buffering capacity of digesta. The ruminal pH difference observed might be due
to chewing activity and buffering through saliva secretion but these responses were not
measured. In an experiment of similar design evaluating grass maturity in which chewing
behavior was measured, cows spent less time ruminating per day when consuming early cut
orchardgrass compared with late cut orchardgrass (Kammes and Allen, submitted). However, it
is not known if differences in ruminating time because of maturity would be the same for
legumes and grasses.
EARLY increased pdNDF intake (2.71 vs. 2.06 kg/d, P < 0.001) and decreased iNDF
intake (4.92 vs. 5.24 kg/d, P = 0.008) compared with LATE (Table 29) because of differences in
chemical composition of forages. Despite the differences in the proportion of iNDF and pdNDF
within the total NDF of the forage, both treatments had nearly complete digestion of pdNDF in
the rumen leading to essentially no passage of pdNDF to the duodenum (Table 29), and
therefore, ruminal passage rate of pdNDF was negligible (Table 30). Ruminal digestibility of

147

pdNDF in the current experiment is unusually high because the pdNDF fraction was abnormally
low. EARLY increased ruminal (35.7 vs. 30.2%, P = 0.02) and total tract (28.5 vs. 19.5%, P <
0.001) digestibility of NDF compared with LATE (Table 29), which is because of a greater
proportion of pdNDF and lower proportion of iNDF in NDF for EARLY compared with LATE
(Table 26).
Total tract digestibilities of NDF (and pdNDF) are lower than ruminal digestibility
because negative postruminal digestibilities were calculated for NDF (and pdNDF) in the present
experiment. This is due to a net gain of fiber from the duodenum to the feces, which has
previously been reported with both the gutter-type T duodenal cannula (Huhtanen and Jaakkola,
1993; Poore et al., 1993), which is the type used in this study, and closed T-type duodenal
cannula (Stensig and Robinson, 1997). The underestimation of duodenal NDF flow or duodenal
iNDF:NDF ratio using iNDF as a marker creates inaccuracies of estimated flow of duodenal
fiber and postruminal digestibility. These errors may be related to unrepresentative digesta
sampling due to differential separation of fluid and particles relative to the true material flowing
out of the duodenum or analytical problems in fiber determination of duodenal samples possibly
due to a component in the duodenal digesta that interferes with the analysis. While absolute
values are not biologically reasonable, relative comparisons between treatments within the same
experiment are useful.
EARLY increased true ruminal OM digestibility (52.5 vs. 46.4%, P = 0.002) and
digestion (13.4 vs. 11.1 kg/d, P < 0.001) compared with LATE (Table 34). Because postruminal
OM digestibility and digestion were similar for EARLY and LATE, EARLY also increased total
tract OM digestibility (62.8 vs. 58.3%, P < 0.001) and digestion (16.1 vs. 14.1 kg/d, P < 0.001)

148

compared with LATE. This resulted in greater total tract DM digestibility (62.5 vs. 58.0%, P <
0.001) and digestion (17.4 vs. 15.1 kg/d, P < 0.001) for EARLY than LATE (Table 34).
Despite less supplementation with soybean products in EARLY compared with LATE,
the high CP concentration in the early cut AL resulted in greater N intake for EARLY than
LATE (918 vs. 812 g/d, P < 0.001, Table 35). EARLY increased ruminal ammonia
concentration compared with LATE (22.5 vs. 20.4 mg/dl, P < 0.001, Table 35). Early cut AL
had higher ammonia concentration than late cut AL (8.09 vs. 6.04 mM, Table 26), which may be
one source for the greater ammonia concentration observed for cows consuming EARLY.
Another source might be protein degradation in the rumen as indicated by greater concentrations
of isovalerate (2.82 vs. 2.51 mM, P = 0.003) and branched chain VFA (4.41 vs. 4.03 mM, P =
0.005) for EARLY than LATE (Table 33). This is consistent with the faster rate of ruminal
starch digestion previously mentioned for EARLY compared with LATE (Table 30); early cut
alfalfa might promote greater numbers or activity of starch-digesting bacteria in the rumen of
cows consuming EARLY. Because some starch-digesting bacteria in the rumen (e.g.
Streptococcus bovis) also have high proteolytic activity (Russell et al., 1981) resulting in
deamination of amino acids and production of ammonia, this could contribute to the increased
ammonia concentration for EARLY.
Effects of pDMI on Ruminal Passage Rates
Experimental data on rates of passage from the rumen, particularly for individual feed
fractions, are scarce. Given the impact of passage on ruminal digestion and microbial growth,
quantitative knowledge of rates of nutrient passage from the rumen is needed to better
understand nutrient availability in ruminants and improve nutrition models. Furthermore,
because passage rates from the rumen generally increase with increased intake, measurements of

149

ruminal passage rates of nutrients over a wide range of DMI are necessary. We measured the
effects of DMI on rates of passage of feed fractions from the rumen using the pool and flux
method (Robinson et al., 1987).
Rate of ruminal passage for iNDF was related to level of intake independent of treatment
(P = 0.07), which increased quadratically as pDMI increased with a greater increase for cows
with lower pDMI (Figure 15). This is in contrast to results in the previously mentioned
experiment of similar design evaluating effects of grass maturity in which cows in the middle of
the pDMI range had the slowest rate of iNDF passage (Kammes and Allen, submitted). As
previously discussed, ruminal digestion of pdNDF was nearly complete for all cows regardless of
treatment and varying levels of pDMI; therefore, rate of ruminal passage for pdNDF was
negligible and not related to level of intake.
Level of intake had no effect on the rate of starch passage, but it tended to influence rate
of ruminal starch digestion (Table 30). As pDMI increased, EARLY slightly increased ruminal
starch digestion rate and LATE quadratically affected it (interaction P = 0.13, Table 30). The
largest difference in starch digestion rate between EARLY and LATE was observed for cows in
the middle of the pDMI range, such that EARLY had a greater starch digestion rate than LATE
(Figure 16A). However, this did not affect starch ruminal turnover rate (Table 30), digestibility
(%), or digestion (kg/d; Table 32), which were similar for EARLY and LATE. EARLY
decreased rumen pool of starch compared with LATE (0.964 vs. 1.13 kg, P = 0.02, Table 31)
because of the lower intake and faster digestion rate of starch for EARLY but was not related to
pDMI. The reduction in starch rate of digestion for LATE might be related to increased liquid
dilution rate (not measured) associated with greater increases in rumen pool sizes for LATE
compared with EARLY as intake increased, decreasing populations of starch-digesting microbes.

150

Ruminal ammonia concentration tended to decrease with increased pDMI (P = 0.14, Figure 16B)
and response of isovalerate (interaction P = 0.002, quadratic, Table 33) and branched chain VFA
(interaction P = 0.01, quadratic, Figure 16C) concentrations, which are derived from degradation
of branched chain amino acids, followed a pattern similar to starch rate of digestion for LATE,
but not EARLY. This is consistent with less proteolytic activity in the rumen for LATE due to
greater removal of ruminal microbes through passage or lysis.
Effects of Treatment and pDMI on N Flux and Microbial Efficiency
Results for N metabolism are shown in Table 35. EARLY decreased NANMN flux (32.5
vs. 41.1% of N intake, P = 0.008) and NAN flux (79.5 vs. 88.3% of N intake, P < 0.001) to the
duodenum as a percent of N intake and tended to increase MN flux (58.2 vs. 52.4% of duodenal
NAN, P = 0.09) to the duodenum as a percent of duodenal NAN compared with LATE, but the
amounts (g/d) of NANMN, NAN, and MN that passed from the rumen to the duodenum were not
different (P # 0.13). However, fluxes of NANMN (P = 0.03), NAN (P = 0.001), and MN (P =
0.11) to the duodenum were positively related to pDMI independent of treatment. This is in
agreement with the results in a review by Clark et al. (1992) that reported positive linear
relationships between OM intake and fluxes of NAN, NANMN, and MN as OM intake increased
over a very wide range (3 to 23 kg/d).
Based on studies with continuous culture fermenters, increases in solid and liquid dilution
rates, which can be associated with increased intake, resulted in greater microbial efficiency
(Crawford et al., 1980; Shriver et al., 1986). In this experiment, microbial efficiency tended to
be related to pDMI (interaction P = 0.14, Table 35), but the response differed by treatment;
EARLY tended to decrease microbial efficiency and LATE tended to increase it as pDMI
increased (Figure 17). Microbial efficiency of cows consuming EARLY quadratically decreased

151

2

as true ruminally digested OM increased (P = 0.03, R = 0.42, Figure 18); however, there was no
relationship between microbial efficiency and true ruminally digested OM for cows consuming
LATE (P = 0.65). This is the opposite of that demonstrated in the experiment of similar design
previously mentioned that evaluated the effects of grass maturity; there was no relationship
between microbial efficiency and true ruminally digested OM for cows consuming diets with
early cut grass silage and a linear decrease in microbial efficiency as true ruminally digested OM
increased for cows consuming diets with late cut grass silage (Kammes and Allen, submitted).
Additionally, microbial efficiency linearly decreased independent of treatment as ruminal
digestion rate of starch increased (P = 0.002, Figure 19). Similarly, others have reported that
efficiency of MN production was inversely related to true ruminally digested OM (kg/d; Oba and
Allen, 2003b) and starch digestion rate (Oba and Allen, 2003b; Voelker and Allen, 2003) in
experiments evaluating carbohydrate source, concentration, and fermentability of diets in dairy
cattle. These negative relationships indicated factors other than availability of energy limited
efficiency of MN production and energy from OM fermentation was uncoupled from microbial
growth (Russell and Cook, 1995).

CONCLUSIONS
Delaying the harvest of alfalfa resulted in lower CP and higher NDF and iNDF
concentrations. Alfalfa maturity and its interaction with pDMI did not affect milk or FCM yields
but EARLY increased DMI compared with LATE; thus, EARLY had lower efficiency of milk
production than LATE. EARLY decreased milk fat concentration more per kilogram pDMI
increase than LATE, but milk fat yield was not affected. The lower concentration and faster
passage rate of iNDF for EARLY resulted in lower rumen pools of iNDF, total NDF, and DM

152

for EARLY compared with LATE, which EARLY increased at a slower rate than LATE as
pDMI increased. EARLY decreased starch intake and increased ruminal pH compared with
LATE. Rate of ruminal starch digestion was related to level of intake; however, this did not
affect ruminal or postruminal starch digestion. Total tract digestibility of NDF, OM, and DM
was higher for EARLY than LATE. When alfalfa silage was the only source of forage in the
diet, cows supplemented with additional concentrate to account for decreased protein and
increased fiber concentrations associated with more mature alfalfa produced similar FCM yields
with greater efficiency than cows fed less mature alfalfa.

153

APPENDIX

154

Table 25. Characterization of 16 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet
Standard
Parameter
Median
Mean
deviation
Minimum
Maximum
Parity
3
2.88
0.79
2
4
1
557
569
43
528
692
BW , kg
BCS
2.22
2.24
0.38
1.63
3.06
DIM
155
137
45
56
208
Milk, kg/d
40.0
41.7
9.6
29.6
66.1
3.5% FCM, kg/d
42.4
43.7
8.6
34.1
68.2
DMI, kg/d
26.1
25.9
2.1
22.9
30.0
1
Empty BW (ruminal digesta removed).

155

Table 26. Chemical composition, particle size distribution, and fermentation parameters of the
early or late cut alfalfa silage included in the treatment diets
Alfalfa silage
Item
Early
Late
Chemical composition
DM, %
30.2
35.6
OM, % DM
89.5
92.1
NDF, % DM
40.8
53.1
1
27.7
39.3
iNDF , % DM
iNDF, % of NDF
67.8
74.1
ADF, % of DM
37.4
44.8
2
8.41
11.9
ADL , % of DM
CP, % DM
23.7
18.1
Starch, % DM
1.16
1.48
3
28.8
24.4
IV NDF digestibility , %
4

Particles size distribution
Wet sieving, % DM retained
19.0 mm
9.50 mm
4.75 mm
2.36 mm
1.18 mm
0.600 mm
0.300 mm
0.150 mm
0.075 mm
0.038 mm
5
Mean particle size , mm
Penn State Particle Separator,
% DM retained
> 19.0 mm
19.0 to 8.0 mm
< 8.0 mm
Fermentation
pH
Acetic acid, % DM
Propionic acid, % DM
Butyric acid, % DM
Lactic acid, % DM
Lactic:Acetic
Ethanol, % DM
Ammonia, mM
1
iNDF = indigestible NDF.

8.61
20.3
33.4
19.2
7.32
4.34
2.81
2.14
1.10
0.86
8.36

8.84
11.7
33.9
27.1
7.52
4.99
2.57
1.66
0.97
0.77
6.92

7.19
67.3
25.5

5.69
35.7
58.6

4.61
4.42
0.45
0.26
9.01
2.04
0.14
8.09

4.42
2.17
0.12
<0.01
8.40
3.87
0.11
6.04

2

ADL = acid detergent lignin.

156

Table 26 (cont’d)
3

30 h in vitro NDF digestibility.

4

Particle size distributions of silages were measured each period (n = 2).

5

Mean particle size calculated from particle size distribution determined by wet sieving.

157

Table 27. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either early or late cut alfalfa silage as the sole source of forage
Preliminary
Early
Late
Ingredients, % DM
Alfalfa silage, early cut
27.0
53.4
--Alfalfa silage, late cut
20.2
--42.3
Dry ground corn
38.1
34.4
39.3
Soybean meal (48% CP)
6.94
--13.0
®
2.75
7.57
--SoyPlus
1
4.68
4.68
4.68
Vitamin mineral mix

1

Limestone
Chemical composition
DM, %
OM, % DM
NDF, % DM
% forage NDF
% NDF from forage
2
iNDF , % DM
iNDF, % of NDF
CP, % DM
Starch, % DM

0.39
44.8
94.1
26.8
22.0
82.0
3
NA
NA
18.4
32.1

---

0.78

43.4
91.8
27.1
21.8
80.4
17.5

54.2
92.8
27.7
22.4
81.1
19.9

64.6
20.2
28.2

71.8
19.0
32.1

Vitamin mineral mix contained (DM basis) 17.1% sodium bicarbonate, 3.9% dicalcium
phosphate, 2.6% magnesium oxide, 1.9% salt, 1.9% trace mineral premix, 0.4% vitamin A, 0.4%
vitamin D, 0.2% vitamin E, and 71.6% dry ground corn as a carrier.
2
iNDF = indigestible NDF.
3

NA = no analysis for preliminary diet.

158

Table 28. Milk production and composition, feed intake, and BW change of cows fed treatment diets containing either early or late cut
alfalfa silage as the sole source of forage
1

2

Treatment LSM
Early
Late
Yield, kg/d
Milk

1

FCM (3.5 %)
Milk fat
Milk protein
Milk lactose
SNF
Milk composition, %
Fat
Protein
Lactose
SNF
DMI, kg/d
3.5% FCM/ DMI
BW change, kg/17 d
BCS change/17 d

P
SE

Trt

Trt x
Period

42.5

41.7

2.6

0.14

44.0
1.59
1.31
2.06
3.76

43.0
1.55
1.30
2.03
3.71

2.4
0.08
0.05
0.15
0.22

0.24
0.33
0.62
0.30
0.40

NS
NS
NS
NS
NS
NS

3.78
3.19
4.82
8.76
28.6
1.45
4.70
0.065

3.74
3.23
4.85
8.83
26.8
1.51
4.61
0.067

0.12
0.07
0.08
0.07
0.7
0.05
3.03
0.044

0.49
0.18
0.30
0.06
0.003
0.01
0.99
0.98

NS
NS
NS
NS
0.02
0.02
NS
0.11

3

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.001

NS

0.09

NS

0.003
0.01
<0.001
0.001
<0.001

NS
NS
NS
NS
NS

0.13
0.19
0.04
0.08
0.06

NS
NS
NS
NS
NS

<0.001
0.03
0.002
NS
<0.001
0.03
0.02
<0.001

0.09
NS
NS
NS
NS
NS
NS
0.72

0.06
NS
0.05
NS
0.14
0.17
0.03
0.001

NS
NS
NS
NS
NS
NS
NS
0.08

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

159

Table 29. NDF digestion of cows fed treatment diets containing either early or late cut alfalfa silage as the sole source of forage
1

Treatment LSM
Early
Late
NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Potentially digestible NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Indigestible NDF
Intake, kg/d
1
Treatment least squares means.

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

7.63

7.30

0.20

0.03

0.15

<0.001

NS

3

0.14

NS

2.63
35.7
4.97

2.15
30.2
5.12

0.13
1.5
0.17

0.003
0.02
0.48

NS
NS
0.03

0.07
NS
<0.001

NS
NS
NS

NS
NS
0.08

NS
NS
NS

-0.50

-0.74

0.14

0.19

NS

NS

NS

NS

NS

2.13
28.5

1.40
19.5

0.09
1.0

<0.001
<0.001

NS
NS

0.02
NS

NS
NS

NS
NS

NS
NS

2.71

2.06

0.07

<0.001

NS

<0.001

NS

0.16

NS

2.63
99.9
0.01

2.15
107.8
-0.15

0.12
5.1
0.11

0.003
0.29
0.29

NS
NS
NS

0.07
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

-0.50

-0.74

0.14

0.19

NS

NS

NS

NS

NS

2.13
80.0

1.40
69.2

0.09
3.3

<0.001
0.007

NS
NS

0.02
NS

NS
NS

NS
NS

NS
NS

4.92

5.24

0.14

0.008

0.12

<0.001

NS

0.14

NS

160

Table 29 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

161

Table 30. Rumen kinetics of cows fed treatment diets containing either early or late cut alfalfa silage as the sole source of forage
1

2

Treatment LSM
Early
Late
Ruminal turnover rate, %/h
DM
OM
NDF

4

pdNDF
Starch
Ruminal turnover time, h
DM
OM
NDF
pdNDF
5

1

iNDF
Starch
Ruminal passage rate, %/h
pdNDF
iNDF
Starch
Ruminal digestion rate, %/h
pdNDF
Starch

P
SE

Trt

10.7

9.07

0.31

10.8
5.25
17.2

9.15
4.45
14.1

36.5

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

<0.001

0.002

0.07

0.10

NS

0.09
0.05
0.80

NS
NS
NS
NS

0.33
0.15
1.5

<0.001
<0.001
0.15

0.002
0.04
NS

0.11
0.04
0.19

NS
NS
NS

33.1

2.2

0.21

0.02

NS

NS

NS

NS

9.56
9.54
19.3
6.70
26.6

11.1
11.0
22.7
8.90
28.2

0.33
0.34
0.71
0.66
1.1

<0.001
0.002
<0.001
0.05
0.11

0.004
0.003
0.10
NS
0.09

0.03
0.04
0.03
NS
0.03

NS
NS
NS
NS
NS

0.08
0.09
0.03
NS
0.07

NS
NS
NS
NS
NS

2.93

3.19

0.15

0.21

0.005

NS

NS

NS

NS

-0.06
3.86
19.7

-0.91
3.57
18.4

0.77
0.13
2.2

0.42
0.01
0.57

NS
0.05
NS

NS
0.05
NS

NS
NS
NS

NS
0.07
NS

NS
NS
NS

16.1
17.1

13.9
13.0

1.5
1.8

0.37
0.03

NS
0.01

NS
0.57

NS
0.06

NS
0.58

NS
0.13

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
162

Table 30 (cont’d)
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

163

Table 31. Rumen pools of cows fed treatment diets containing either early or late cut alfalfa silage as the sole source of forage
1

2

Treatment LSM
Early
Late

SE

Trt

Trt x
Period

Wet weight, kg

76.9

80.8

2.2

0.04

Volume, L
Density, kg/L
Rumen pool, kg
DM
OM
NDF

91.3
0.84

97.8
0.83

3.0
0.01

11.3
10.3
6.23
0.75

12.4
11.4
6.96
0.73

5.49
0.964

pdNDF

4

5

1

iNDF
Starch

P
pDMI

Trt x
pDMI

pDMI x
pDMI

0.08

NS

NS

0.03
0.29

NS
NS
NS

Trt x
pDMI x
pDMI
NS

0.07
NS

NS
NS

NS
NS

NS
NS

0.3
0.3
0.17
0.05

0.007
0.006
<0.001
0.83

NS
NS
NS
NS

0.07
0.08
0.06
0.007

0.13
0.14
0.10
NS

NS
NS
NS
NS

NS
NS
NS
NS

6.24

0.18

<0.001

NS

0.18

0.08

NS

NS

1.13

0.051

0.02

0.06

0.20

NS

NS

NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

164

Table 32. Starch digestion of cows fed treatment diets containing either early or late cut alfalfa silage as the sole source of forage
1

Treatment LSM
Early
Late
Starch
Intake, kg/d

1

Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
% of duodenal passage
Apparent total tract digestion
kg/d
%

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI
3

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.18

NS

8.16

8.74

0.23

0.007

0.04

0.001

3.87
48.2

3.74
44.6

0.32
3.8

0.67
0.29

0.11
NS

NS
NS

NS
NS

NS
NS

NS
NS

4.01
49.9
4.07

3.90
46.5
4.78

0.32
3.8
0.35

0.73
0.31
0.04

0.12
NS
NS

NS
NS
0.13

NS
NS
NS

NS
NS
NS

NS
NS
NS

3.46
44.2
84.5

4.01
46.5
82.9

0.34
3.8
1.3

0.13
0.53
0.36

NS
NS
NS

0.20
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

7.33
92.3

7.74
91.1

0.15
0.5

0.02
0.05

0.03
NS

0.002
0.11

NS
NS

NS
NS

NS
NS

NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

165

Table 33. Ruminal VFA concentrations and pH of cows fed treatment diets containing either early or late cut alfalfa silage as the sole
source of forage
1

Treatment LSM
Early
Late
Total VFA, mM

1

Acetate
Propionate
Butyrate
Lactate
Isobutyrate
Valerate
Isovalerate
Branch chain VFA
Acetate:Propionate
Ruminal pH

141
86.3
29.5
16.1
1.79
1.61
2.68
2.82
4.41
2.94
6.36

139
81.9
31.0
18.1
1.37
1.54
2.47
2.51
4.03
2.65
6.29

2

P
SE

Trt

Trt x
Period

2.0

0.48

1.0
0.6
0.4
0.33
0.02
0.07
0.09
0.10
0.05
0.02

0.005
0.04
<0.001
0.37
0.02
0.05
0.003
0.005
<0.001
0.04

NS
NS
NS
0.06
NS
NS
NS
NS
NS
NS
NS

3

pDMI

Trt x
pDMI

pDMI x
pDMI

NS

NS

NS

Trt x
pDMI x
pDMI
NS

0.74
NS
NS
0.31
0.11
0.13
0.07
0.02
NS
NS

0.09
NS
NS
NS
NS
0.95
0.04
0.09
NS
NS

NS
NS
NS
0.06
NS
0.60
0.02
0.01
NS
NS

NS
NS
NS
NS
NS
0.02
0.002
0.01
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

166

Table 34. DM and OM digestion of cows fed treatment diets containing either early or late cut alfalfa silage as the sole source of
forage
1

2

Treatment LSM
Early
Late
DM
Intake, kg/d

1

Apparent total tract digestion
kg/d
%
OM
Intake, kg/d
Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
Apparent total tract digestion
kg/d
%

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

28.6

26.8

0.7

0.003

0.02

<0.001

NS

17.4
62.5

15.1
58.0

0.4
0.5

<0.001
<0.001

0.005
0.05

26.2

24.8

0.7

0.009

9.76
37.9

7.93
33.2

0.41
1.6

13.4
52.5
16.5

11.1
46.4
17.0

6.30
24.8
16.1
62.8

pDMI x
pDMI

Trt x
pDMI x
pDMI

3

0.14

NS

0.003
0.60

NS
0.12

NS
NS

NS
NS

0.02

<0.001

NS

0.14

NS

<0.001
0.007

0.009
0.14

NS
NS

NS
NS

NS
NS

NS
NS

0.5
1.7
0.7

<0.001
0.002
0.40

0.03
NS
NS

NS
NS
0.001

NS
NS
NS

NS
NS
0.10

NS
NS
NS

6.12
25.1

0.44
1.7

0.73
0.90

NS
NS

0.05
NS

NS
NS

NS
NS

NS
NS

14.1
58.3

0.3
0.5

<0.001
<0.001

0.003
0.03

0.003
NS

NS
NS

NS
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).

167

Table 34 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

168

Table 35. Nitrogen metabolism of cows fed treatment diets containing either early or late cut alfalfa silage as the sole source of forage
1

Treatment LSM
Early
Late
N intake, g/d

918

Ruminal ammonia, mg/dl
Flow to duodenum
Ammonia N, g/d
NAN
g/d
% of N intake
4

NANMN
g/d
% of N intake
% of duodenal NAN
Microbial N
g/d
% of duodenal NAN
5

SE

Trt

pDMI

Trt x
pDMI

<0.001

23

Trt x
Period
0.07

<0.001

0.14

Trt x
pDMI x
pDMI
NS

NS

NS

3

pDMI x
pDMI

22.5

20.4

0.6

<0.001

NS

0.58

NS
0.14

17.8

16.9

0.9

0.31

NS

0.004

NS

NS

NS

727
79.5

715
88.3

27
2.4

0.49
<0.001

NS
NS

0.001
0.13

NS
NS

0.07
0.17

NS
NS

292
32.5
41.8

321
41.1
47.6

19
2.3
2.5

0.14
0.008
0.09

0.16
NS
NS

0.03
NS
0.31

0.17
NS
0.17

NS
NS
NS

NS
NS
NS

397
58.2
30.2

357
52.4
32.2

19
2.5
1.4

0.13
0.09
0.27

NS
NS
0.03

0.11
0.31
0.54

NS
0.17
0.14

NS
NS
NS

NS
NS
NS

394
46.8
54.2

22
1.8
1.1

0.53
0.06
0.56

0.15
NS
0.10

0.01
NS
NS

NS
NS
NS

0.15
NS
NS

NS
NS
NS

480
60.0

13
0.7

<0.001
<0.001

0.02
0.02

0.005
0.05

NS
0.20

NS
0.08

NS
NS

g/kg TRDOM
NAN apparent postruminal digestion
g/d
405
% of N intake
42.6
% of duodenal passage
55.1
N apparent total tract digestion
g/d
586
%
64.7

1

812

2

P

Treatment least squares means.

169

Table 35 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

NANMN = nonammonia, nonmicrobial nitrogen.

5

TRDOM = true ruminally digested OM.

170

Figure 13. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
alfalfa maturity with preliminary DMI for A) indigestible NDF (iNDF) rumen pool (P = 0.08)
and B) NDF rumen pool (P = 0.10). The preliminary DMI on the x-axis are the mean DMI of
individual cows during the final 4 d of the preliminary period when all cows were fed a common
diet.

171

Figure 14. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
alfalfa maturity with preliminary DMI for ruminal acetate concentration (P = 0.09). The
preliminary DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the
preliminary period when all cows were fed a common diet.

172

Figure 15. Relationship of early (open circles) and late (closed circles) alfalfa maturity with
preliminary DMI for indigestible NDF (iNDF) ruminal passage rate (P = 0.07). The preliminary
DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the preliminary
period when all cows were fed a common diet.

173

Figure 16

174

Figure 16 (cont’d)

Figure 16. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
alfalfa maturity with preliminary DMI for A) starch ruminal digestion rate (P = 0.13), B) ruminal
ammonia concentration (P = 0.14 linear), and C) branched chain VFA concentration (P = 0.01).
The preliminary DMI on the x-axis are the mean DMI of individual cows during the final 4 d of
the preliminary period when all cows were fed a common diet.

175

Figure 17. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
alfalfa maturity with preliminary DMI for microbial efficiency (P = 0.14) expressed as g of
microbial N/kg of true ruminally digested OM (TRDOM). The preliminary DMI on the x-axis
are the mean DMI of individual cows during the final 4 d of the preliminary period when all
cows were fed a common diet.

176

Figure 18. Relationship between true ruminally digested OM (TRDOM) and microbial
efficiency for early (open circles, dashed line; microbial efficiency, g of microbial N/kg of
2
2
TRDOM = [47.3 – (1.43 x TRDOM, kg/d) + (0.213 x (TRDOM, kg/d – 12.3) )]; P = 0.03, R =
2

0.42) and late (closed circles, solid line; P = 0.65, R = 0.02) alfalfa maturity.

177

Figure 19. Relationship between starch ruminal digestion rate and microbial efficiency.
Microbial efficiency, g of microbial N/kg of true ruminally digested OM (TRDOM) = 40.1 –
2
0.568 x starch ruminal digestion rate, %/h (P = 0.002, R = 0.27). Open circles denote early and
closed circles denote late alfalfa maturity.

178

REFERENCES

179

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cows.
J. Dairy Sci. 83:1598-1624.
Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64-75.
Clark, J. H., T. H. Klusmeyer, and M. R. Cameron. 1992. Microbial protein synthesis and flows
of nitrogen fractions to the duodenum of dairy cows. J. Dairy Sci. 75:2304-2323.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting digestibility of
different diets with internal markers: Evaluation of four potential markers. J. Anim. Sci.
63:1476-1483.
Crawford, R. J., Jr., W. H. Hoover, and L. L. Junkins. 1980. Effects of solids and liquid flows on
fermentation in continuous cultures. II. Nitrogen partitioning and efficiency of microbial
synthesis. J. Anim. Sci. 51:986-995.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.
Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More powerful peroxide
Kjeldahl digestion method. J. AOAC 70:783-787.
Huhtanen, P., and S. Jaakkola. 1993. The effects of forage preservation method and proportion of
concentrate on digestion of cell wall carbohydrates and rumen digesta pool size in cattle.
Grass Forage Sci. 48:155-165.
Joy, M. T., E. J. DePeters, J. G. Fadel, and R. A. Zinn. 1997. Effects of corn processing on the
site and extent of digestion in lactating cows. J. Dairy Sci. 80:2087-2097.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019-1027.
Lammers, B. P., D. R. Buckmaster, and A. J. Heinrichs. 1996. A simplified method for the
analysis of particle sizes of forage and total mixed rations. J. Dairy Sci. 79:922-928.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:12171240.

180

National Research Council. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl.
Acad. Washington, DC.
Oba, M., and M. S. Allen. 1999. Effects of brown midrib 3 mutation in corn silage on dry
matter intake and productivity of high yielding dairy cows. J. Dairy Sci. 82:135-142.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174–183.
Oba, M., and M. S. Allen. 2003b. Effects of diet fermentability on efficiency of microbial
nitrogen production in lactating dairy cows. J. Dairy Sci. 86:195-207.
Poore, M. H., J. A. Moore, T. P. Eck, R. S. Swingle, and C. B. Theurer. 1993. Effect of fiber
source and ruminal starch degradability on site and extent of digestion in dairy cows. J.
Dairy Sci. 76:2244-2253.
Robinson, P. J., S. Tamminga, and A. M. van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci. 17:37-62.
Russell, J. B., W. G. Bottje, and M. A. Cotta. 1981. Degradation of protein by mixed cultures of
rumen bacteria: Identification of Streptococcus bovis as an actively proteolytic rumen
bacterium. J. Anim. Sci. 53:242-252.
Russell, J. B., and G. M. Cook. 1995. Energetics of bacterial growth: balance of anabolic and
catabolic reactions. Microbiol. Rev. 59:48-63.
Shriver, B. J., W. H. Hoover, J. P. Sargent, R. J. Crawford, and W. V. Thayne. 1986.
Fermentation of a high concentrate diet as affected by ruminal pH and digesta flow. J.
Dairy Sci. 69:413-419.
Smith, L. W., H. K. Goering, and C. H. Gordon. 1972. Relationships of forage compositions with
rates of cell wall digestion and indigestibility of cell walls. J. Dairy Sci. 55:1140-1147.
Stensig, T., and P. H. Robinson. 1997. Digestion and passage kinetics of forage fiber in dairy
cows as affected by fiber-free concentrate in the diet. J. Dairy Sci. 80:1339-1352.
Voelker, J. A., and M. S. Allen. 2003. Pelleted beet pulp substituted for high-moisture corn: 3.
Effects of ruminal fermentation, pH, and microbial protein efficiency in lactating dairy
cows. J. Dairy Sci. 86:3562-3570.
Voelker, J. A., G. M. Burato, and M. S. Allen. 2002. Effect of pretrial milk yield on responses
of feed intake, digestion, and production to dietary forage concentration. J. Dairy Sci.
85:2650-2661.

181

Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. J. Dairy Sci. 65:495-501.
Zinn, R. A., and F. N. Owens. 1986. A rapid procedure for purine measurement and its use for
estimating net ruminal protein synthesis. Can. J. Anim. Sci. 66:157-166.

182

CHAPTER 5
Nutrient Demand Interacts with Grass Maturity to Affect Milk Fat Concentration and
Digestion Responses in Dairy Cows

1

ABSTRACT
Effects of grass maturity on dry matter intake (DMI), milk production, ruminal
fermentation and pool sizes, digestion and passage kinetics, and chewing activity and the
relationship of these effects with preliminary DMI (pDMI) were evaluated using 13 ruminally
and duodenally cannulated Holstein cows in a crossover design with a 14-d preliminary period
and two 18-d treatment periods. During the preliminary period, pDMI of individual cows ranged
from 23.5 to 28.2 kg/d (mean = 26.1 kg/d) and 3.5% fat-corrected milk (FCM) yield ranged from
30.8 to 57.2 kg/d (mean = 43.7 kg/d). Experimental treatments were diets containing
orchardgrass silage harvested either a) early cut, less mature (EARLY) or b) late cut, more
mature (LATE) as the sole forage. Early and late cut orchardgrass contained 44.9 and 54.4%
neutral detergent fiber (NDF) and 19.9 and 14.8% crude protein, respectively.
Forage:concentrate ratio was 58:42 and 46:54 for EARLY and LATE, respectively; both diets
contained ~25% forage NDF and 30% total NDF. Preliminary DMI, an index of nutrient
demand, was determined during the last 4 d of the preliminary period when cows were fed a
common diet and used as a covariate. Main effects of grass maturity and their interaction with
pDMI were tested by ANOVA. EARLY decreased milk yield and increased milk fat
concentration compared with LATE, which was likely due to lower inclusion of concentrate for
EARLY than LATE. Grass maturity and its interaction with pDMI did not affect FCM yield,
1

Submitted for publication in the Journal of Dairy Science on September 21, 2011.

183

DMI, rumen pH, or microbial efficiency. EARLY increased rates of ruminal digestion of
potentially digestible NDF and passage of indigestible NDF (iNDF) compared with LATE. The
lower concentration and faster passage rate of iNDF for EARLY resulted in lower rumen pools
of iNDF, total NDF, organic matter, and dry matter for EARLY than LATE. Ruminal passage
rates of potentially digestible NDF and starch were related to level of intake (quadratic
interactions) and subsequently affected ruminal digestion of these nutrients. EARLY decreased
eating, ruminating, and total chewing time per unit of forage NDF intake compared with LATE.
When grass silage was the only source of forage in the diet, cows supplemented with additional
concentrate to account for decreasing protein and increasing fiber concentrations associated with
more mature grass produced similar FCM yields as cows fed less mature grass.

INTRODUCTION
Forage maturity at harvest affects the supply and utilization of nutrients in dairy cows fed
forage-based diets. Maturity affects the yield and quality of forages, and time of harvest is a
compromise between these factors. As plants grow and mature, total yield of DM increases but
nutritive quality generally decreases due to increasing fiber, and lignification of fiber, and
decreasing protein concentrations. Therefore, producing high quality forage is largely dependent
on harvesting at the optimum maturity. While dairy cows can be fed supplemental concentrate to
improve milk production when cows are fed low quality forage, feeding quality forages can
optimize milk production, improve cow health, reduce purchased feed costs, and increase dairy
profitability.
Increasing maturity in perennial forages reduces digestibility and intake potential of

184

forages due to increased concentration of NDF and greater lignification of the NDF. This not
only decreases the potentially digestible NDF (pdNDF) concentration but also decreases
digestion rate of the remaining pdNDF (Smith et al., 1972), which might allow particulate matter
to remain buoyant longer, decreasing specific gravity and rate of passage from the rumen (Jung
and Allen, 1995). This suggests that more mature forages will result in slower passage rates
from the rumen and greater ruminal distention. However, increased maturity at harvest increased
rate of particle size reduction by chewing for grass stems and leaves (Poppi et al., 1981) and for
ryegrass (Ulyatt, 1983) because of greater fragility. This more rapid particle size reduction could
increase passage rate for more mature forages, acting in opposition to the expected effects of
buoyancy. Orchardgrass (OG; Dactylis glomerata L.) was selected as a representative coolseason grass for use in this experiment.
In addition to the combination of dietary factors affecting ruminal distention and rate of
particle breakdown, the individual cow’s appetite will also affect the responses of passage rate
and intake to forage maturity. Cows respond differently to treatments depending on their level of
intake (Voelker Linton and Allen, 2008) and production (Oba and Allen, 1999). Because grass
maturity and level of intake affect ruminal passage and digestion rates and, thus, digesta fill in
the rumen, the response to effects of grass maturity and its relationship with intake level should
be assessed to determine if responses to treatment vary among cows with a wide range in DMI.
We hypothesized that responses of DMI and digesta passage rates to grass maturity are related to
level of intake and less mature grass will permit a greater increase in passage rate than more
mature grass as feed intake increases.
The objectives of this experiment were to evaluate the relationships between voluntary
DMI and effects of grass maturity on DMI, milk production, ruminal fermentation and pool

185

sizes, digestion and passage kinetics, and chewing behavior in lactating dairy cows. This study
had three unique features to improve our understanding of the role of grass maturity and interpret
its effect on animal responses. First, it allowed effects of the interaction between grass maturity
and preliminary DMI (pDMI) to be evaluated. The use of pDMI, an index of nutrient demand,
allowed the evaluation of treatments on animal responses in relation to level of intake and
provided an indicator to test effects of intake level independent of treatments. Second, it directly
compared treatment effects of early and late cut OG as the sole source of forage. Third, ruminal
passage rates of individual feed fractions, instead of entire feeds, were measured using ruminally
and duodenally cannulated cows.

MATERIALS AND METHODS
Cows and Treatments
Experimental procedures were approved by the Institutional Animal Care and Use
Committee at Michigan State University. Thirteen multiparous Holstein cows from the
Michigan State University Dairy Cattle Teaching and Research Center were assigned randomly
to treatment sequence in a crossover design experiment with one 14-d preliminary period and
two 18-d experimental periods. The first 10 d of each period were allowed for diet adaptation
and samples were collected during the final 4 d of the preliminary period and 8 d of each
experimental period. Cows were 164±85 (mean±SD) DIM at the end of the preliminary period
and were selected to provide a wide range and uniform distribution of pDMI and milk yield.
During the final 4 d of the 14-d preliminary period, the average pDMI among cows ranged from
23.5 to 28.2 kg/d (mean = 26.1 kg/d) and 3.5% FCM yield ranged from 30.8 to 57.2 kg/d (mean
= 43.7 kg/d; Table 36). Prior to calving, cows were cannulated ruminally (Bar Diamond, Inc.,

186

Parma, ID) and duodenally with a gutter-type T cannula placed approximately 10 cm distal to the
pylorus (Joy et al., 1997). Surgery was performed at the Department of Large Animal Clinical
Science, College of Veterinary Medicine, Michigan State University.
Experimental treatments were diets containing OG silage from one field harvested either
a) early cut, less mature (EARLY) or b) late cut, more mature (LATE) as the sole forage.
Orchardgrass (Baridana cultivar, Barenbrug USA, Tangent, OR) was produced at the campus
farm at Michigan State University (East Lansing), chopped to 10 mm theoretical length of cut,
and ensiled in Ag-Bags (Ag-Bag Systems Inc., St. Nazianz, WI). During the sample collection
periods, early and late cut OG contained 44.9 and 54.4% NDF and 19.9 and 14.8% CP,
respectively (DM basis; Table 37). Diets EARLY and LATE were formulated to contain 25%
forage NDF, 30% total NDF, and 18% CP. We acknowledge these treatments affect dietary
starch concentration but maintaining similar forage and total NDF concentrations for both
treatments was of primary interest. The diet fed during the preliminary period was formulated so
that early and late cut OG each contributed 50% of forage NDF. Diets also contained dry ground
®

corn, soybean meal (48% CP), SoyPlus (West Central Soy, Ralston, IA), vitamin-mineral
premix, and limestone (Table 38).
Data and Sample Collection
Throughout the experiment, cows were housed in tie-stalls and fed diets as total mixed
rations once daily (1130 h) at 110% of expected intake. The amount of feed offered and refused
(orts) was weighed daily for each cow. Forage samples were collected twice weekly and
analyzed to adjust diets to account for DM, NDF, and CP fluctuation. Samples of all dietary
ingredients (0.5 kg) and orts (12.5%) were collected daily from d 11 to 14 during the preliminary
period and d 11 to 15 during each experimental period. Samples were frozen immediately after

187

collection at !20°C and combined to one composite sample per period prior to analysis.
Cows were moved to an exercise lot twice daily (0230 and 1300 h) prior to milking in a
parlor (0400 and 1430 h). Milk yield was measured, and milk was sampled, at each milking on d
11 to 14 of the preliminary period and on d 11 to 15 of the experimental periods. Rumen-empty
BW was measured by weighing the cow after evacuation of ruminal digesta on d 14 of the
preliminary period and d 18 of each experimental period. Body condition score was determined
on the same days by 3 trained investigators blinded to treatments (Wildman et al., 1982; 5-point
scale where 1 = thin and 5 = fat). Chewing activity was monitored and recorded by observation
every 5 min for 24 h on d 16 of each experimental period. Activity was noted as eating,
ruminating, drinking, or idle for each cow at each time.
Duodenal samples (900 mL), fecal samples (500 g), rumen fluid and particulate samples
for microbial isolation (400 g), and rumen fluid samples for pH, concentrations of VFA, lactate,
and ammonia (100 mL) were collected every 15 h from d 11 to 15 of each experimental period
so that 8 samples were taken for each cow in each period, representing every 3 h of a 24-h period
to account for diurnal variation. Rumen fluid and particulate matter for microbial isolation was
collected from the reticulum, near the reticular-omasal orifice, transported to the laboratory, and
processed. Rumen fluid for pH, VFA, lactate, and ammonia were obtained by combining digesta
from five different sites in the rumen and straining it through nylon mesh (~1 mm pore size);
fluid pH was recorded immediately. Samples were stored at -20°C.
Ruminal contents were evacuated manually through the ruminal cannula 4 h after feeding
at the beginning of d 17 (1530 h) and 2 h before feeding at the end of d 18 (0930 h) for each
experimental period. Total rumen content mass and volume were determined. To ensure
accurate sampling, every tenth handful of digesta (10%) was separated for a subsample

188

throughout evacuation. This subsample was squeezed into primarily solid and liquid phases.
Both phases were weighed and sampled (350 mL) for determination of nutrient pool size. All
samples were stored at -20°C.
Sample Analysis and Calculations
Milk yield recorded at both milkings were summed for a daily total, which were averaged
for each period. Milk samples were analyzed for fat, true protein, lactose, SNF, and MUN with
infrared spectroscopy by Michigan DHIA (East Lansing). Yields of 3.5% FCM and milk
components were calculated using milk yield and component concentrations for each milking,
summed for a daily total, and averaged for each period. Milk samples used for analysis of fatty
acid profile were composited based on milk fat yield and centrifuged at 17,800 " g for 30 min at
8°C. Fat cake (300 to 400 mg) was extracted according to Hara and Radin (1978), and methyl
esters were formed according to Christie (1982) as modified by Chouinard et al. (1999). Fatty
acids were quantified by gas chromatography (model 8500; Perkins-Elmer Corp, Norwalk, CT)
according to Kramer et al. (1997) using a SP-2560 capillary column (100 m " 0.20 mm id with
0.02-µm film thickness; Supelco, Bellefonte, PA). Oven temperature was 70°C for 4 min, then
increased 13°C/min to 175°C and held for 27 min before being increased again at 4°C/min to
215°C and held for 31 min. Helium flow was 20 cm/s, and the total run time was 80 min.
Forage samples were combined to one composite sample per forage per period. Particle
size distribution was determined using the Penn State Particle Separator containing 2 sieves (19
and 8 mm) and a pan (Lammers et al., 1996). In addition, samples were wet sieved manually and
sequentially through screens with the following aperture sizes: 19.0, 9.50, 4.75, 2.36, 1.18,
0.600, 0.300, 0.150, 0.075, and 0.038 mm. The fraction of DM retained on the screens from wet
sieving was used to calculate mean particle size.

189

Diet ingredients, orts, and feces were lyophilized (Tri-Philizer MP, FTS Systems, Stone
Ridge, NY). All dried samples were ground with a Wiley mill (1-mm screen; Arthur H. Thomas,
Philadelphia, PA). Dried, ground fecal samples were combined on an equal DM basis into one
sample per cow per period. Frozen duodenal samples for each cow period (n=8) were chopped
finely using a commercial food processor (84142 Food cutter, Hobart Manufacturing Co., Troy,
OH) and subsampled in the frozen state to obtain representative samples. These duodenal
subsamples and the 350 mL of ruminal solid and liquid samples were lyophilized and ground as
described above. Dried ruminal solid and liquid samples were recombined according to the
original ratio of solid and liquid DM.
Samples were analyzed for ash, NDF, indigestible NDF (iNDF), ADF, acid detergent
lignin (ADL), CP, and starch. Ash concentration was determined after 5 h combustion at 500°C
in a muffle furnace. Concentrations of NDF were determined according to Mertens (2002) and
ADF and ADL according to Goering and Van Soest (1970). Indigestible NDF was estimated as
NDF residue after 240 h in vitro fermentation (Goering and Van Soest, 1970); flasks were
reinoculated at 120 h to insure a viable microbial population. Ruminal fluid for the in vitro
incubations was collected from a nonpregnant dry cow fed dry hay only. Fraction of pdNDF was
calculated by difference (1.00 – iNDF). Crude protein was analyzed according to Hach et al.
(1987). Starch was measured by an enzymatic method (Karkalas, 1985) after samples were
gelatinized with sodium hydroxide. Glucose concentration was measured using a glucose
oxidase method (Glucose kit #510, Sigma Chemical Co., St. Louis, MO), and absorbance was
determined with a micro-plate reader (SpectraMax 190, Molecular Devices Corp., Sunnyvale,
CA). Concentrations of all nutrients except DM were expressed as percentages of DM
determined by drying at 105°C in forced-air oven for more than 8 h.

190

Duodenal digesta were analyzed for purines and ammonia to estimate microbial N (MN)
flow and nonammonia, nonmicrobial N (NANMN) flow to the duodenum. Purine concentration
was used as a microbial marker, and purine to MN ratio was estimated by analysis of microbial
pellets obtained by differential centrifugation of the rumen fluid and particulate samples
collected near the reticulum. Rumen fluid and particulate matter was blended, strained through
nylon mesh, and the liquid portion was centrifuged at 500 x g for 15 min. The supernatant was
centrifuged at 18,000 x g for 15 min, and the pellet was washed with 0.9% NaCl solution and
centrifuged again at 18,000 x g for 15 min, resuspended in water, and lyophilized. Total purines
were measured by spectrophotometer (Beckman Instruments Inc., Fullerton, CA) at 260 nm
according to Zinn and Owens (1986). Ammonia concentration was determined for centrifuged
duodenal and rumen fluid samples according to Broderick and Kang (1980). Rumen fluid was
also analyzed for concentration of major VFA and lactate by HPLC (Waters Corp., Milford,
MA) according to Oba and Allen (2003a).
Dry matter and nutrient intakes were calculated using the composition of feed offered and
refused. Ruminal pool sizes (kg) of OM, NDF, iNDF, pdNDF, starch, MN, and NANMN were
determined by multiplying the concentration of each component in rumen samples by the
ruminal digesta DM mass (kg). Duodenal flows (kg/d) of DM, OM, total NDF, pdNDF, starch,
MN, NANMN, and ammonia N were determined using iNDF as a flow marker; iNDF intake
(kg/d) was multiplied by the ratio between the component and iNDF in duodenal digesta.
Duodenal flow of microbial OM was determined using the ratio of purines to OM (Oba and
Allen, 2003a), and true ruminally digested OM was calculated by subtracting duodenal flow of
nonmicrobial OM from OM intake. Indigestible NDF was used as an internal marker to estimate
nutrient digestibility in the rumen and in the total tract (Cochran et al., 1986). Turnover rate in

191

the rumen, passage rate from the rumen, and ruminal digestion rate of each component was
calculated by the following equations:
Turnover rate (%/h) = 100 x (Intake of component / Ruminal pool of component) / 24
Passage rate (%/h) = 100 x (Duodenal flow of component / Ruminal pool of component) / 24
Digestion rate (%/h) = Turnover rate in the rumen (%/h) – Passage rate from the rumen (%/h).
Manually observed chewing activity was summarized by a logic script in Igor Pro
(Version 6.12, WaveMetrics Inc., Lake Oswego, OR) to generate meal and rumination bout
information according to previously established criteria (Dado and Allen, 1994). Variables
determined included frequency of meal bouts per day, interval between meals, frequency of
ruminating bouts per day, interval between ruminating bouts, eating time per day, ruminating
time per day, and total chewing time per day.
Statistical Analysis
All data were analyzed by using the fit model procedure of JMP (Version 8, SAS
Institute, Cary, NC). To determine differences between treatments and evaluate interactions of
treatment with DMI, where pDMI calculated as the mean of DMI values on d 11 to 14 of the 14d preliminary period was used as the covariate for treatment responses, data were analyzed
2

according to the following model: Yijk = µ + Ci + Pj + Tk + PTjk + pDMI + TkpDMI + pDMI +
2

TkpDMI + eijk where µ is the overall mean, Ci is the random effect of cow (i = 1 to 13), Pj is
the fixed effect of period (j = 1 to 2), Tk is the fixed effect of treatment (k = 1 to 2), PTjk is the
interaction of period and treatment, pDMI is the linear effect of pDMI, TkpDMI is the interaction
2

2

of treatment and pDMI (linear), pDMI is the quadratic effect of pDMI, TkpDMI is the
interaction of treatment and pDMI (quadratic), and eijk is the residual error. Statistical
192

2

significance for TkpDMI and TkpDMI indicated treatment differences were related to pDMI.
Covariate and interaction terms were removed stepwise from the model if P > 0.20. Treatment
effects and their interaction (linear and quadratic relationships) were declared significant at P #
0.05 and P # 0.10, respectively. Tendencies for treatment effects and their interactions were
declared at P # 0.10 and P # 0.15, respectively.
Sixteen cows started the experiment, however, one cow was removed from the study due
to circumstances unrelated to treatments. Additionally, data from two cows were excluded prior
to statistical analysis; one cow had a broken duodenal cannula during the last collection of the
second experimental period that was replaced but affected her DMI and subsequently rumen
evacuation measurements, and the other cow was considered an outlier based on large Cook’s
distance values (Cook and Weisberg, 1982) for several response variables of primary interest.
Thus, data from 13 cows were statistically analyzed.

RESULTS AND DISCUSSION
Comparison of Forages and Diets
Chemical analyses of OG silages are listed in Table 37. As expected, delaying harvest of
OG decreased the concentration of CP (19.9 vs. 14.8%) and increased concentrations of total
NDF (44.9 vs. 54.4%) and iNDF (10.8 vs. 14.5%). Indigestible NDF, expressed as a percent of
NDF, was slightly lower (24.1 vs. 26.6% of NDF) for early cut OG than for late cut OG. Early
cut OG had greater DM concentration, higher pH, and contained less lactic and acetic acids than
late cut OG silage. Physical characteristics (Table 37) of early and late cut OG silages were
similar for mean particle size and particle size distribution.
Diet ingredients and chemical composition are shown in Table 38. The preliminary diet

193

contained more early cut OG than late cut OG so each forage supplied similar concentrations of
forage NDF. Because treatment diets were formulated to contain similar forage NDF
concentrations, forage:concentrate ratios were different between diets with ratios of 58:42 and
46:54 for EARLY and LATE, respectively. Besides forage source, the main differences in diets
were the concentrations of corn grain and soybean meal, which were both lower for EARLY than
LATE, to account for differences between the grass silages. Both diets had a similar chemical
composition, which was mathematically calculated according to the proportion of each feed
ingredient in the diet and its respective analytical values, except for starch which was lower for
EARLY due to more forage and less concentrate in the diet for EARLY compared with LATE.
In both diets, forage NDF provided over 82% of the total diet NDF.
Effects of Grass Maturity and pDMI
Results of grass maturity and its interaction with pDMI on milk yield and composition
are shown in Table 39. EARLY decreased milk yield (36.6 vs. 39.2 kg/d, P < 0.001) and
subsequently yields of protein (1.14 vs. 1.22 kg/d, P = 0.008), lactose (1.71 vs. 1.84 kg/d, P =
0.001), and SNF (2.06 vs. 2.22 kg/d, P < 0.001) compared with LATE, as concentrations of
protein, lactose, and SNF were similar. EARLY decreased MUN concentration (7.70 vs. 10.5
mg/dl, P < 0.001) and increased milk fat concentration (3.70 vs. 3.38%, P < 0.001) compared
with LATE. The lower milk yield, higher milk fat concentration for EARLY and higher milk
yield, lower milk fat concentration for LATE resulted in similar FCM and milk fat yields.
Response of milk fat concentration to treatment was related to pDMI (interaction P = 0.07,
Figure 20) such that EARLY increased milk fat concentration and LATE decreased it as pDMI
increased.
The aforementioned results are related to different concentrate levels in the diets, which

194

were necessary to account for changes in chemical composition of OG with increasing maturity
and maintain the same concentration of forage NDF in the diets. EARLY decreased starch
intake (5.48 vs. 5.98 kg/d, P < 0.001, Table 40) and ruminal propionate concentration (37.6 vs.
41.5 mM, P = 0.01, Table 41) compared with LATE. Because there was no difference between
treatments for acetate concentration, EARLY increased acetate to propionate ratio (2.50 vs. 2.30,
P = 0.04, Table 41) compared with LATE.
Although N intake was similar for EARLY and LATE (P = 0.89, Table 42), differences
in N sources might have affected N digestion as greater quantities of N were contributed from
grass silage for EARLY and soybean meal for LATE (Table 38). Lower MUN concentration for
EARLY is consistent with lower ruminal ammonia concentration (11.0 vs. 12.9 mg/dl, P =
0.001) and flow of ammonia N to the duodenum (15.0 vs. 17.5 g/d, P = 0.01) for EARLY than
LATE (Table 42). The increased ammonia for LATE was not because of ammonia
concentration in silage, which was higher for early cut OG than late cut OG (Table 37), nor due
to excessive degradation of amino acids in the rumen, which would result in the production of
ammonia and branched chain VFA, because isovalerate and branched chain VFA were higher for
EARLY than LATE (Table 41). The source of increased ruminal ammonia for LATE is not
known but might be from greater degradation of protein in soybean meal compared with protein
in early cut OG. EARLY decreased total tract N digestion (376 vs. 438 g/d, P = 0.003) and
digestibility (56.1 vs. 65.1%, P < 0.001) compared with LATE (Table 42), which is likely
associated with the differences in N sources in diets.
Differences in starch intake (Table 40) and milk fatty acid profile (Table 43) suggest
reduction in milk fat concentration exhibited by cows consuming LATE might be related to dietinduced milk fat depression (MFD). Three conjugated linoleic acid (CLA) isomers, trans-10,

195

cis-12 CLA; trans-9, cis-11 CLA; cis-10, trans-12 CLA, produced as intermediates in rumen
biohydrogenation have been shown to reduce milk fat to date (Bauman et al., 2008). There was
no difference in concentrations of trans-10, cis-12 CLA between treatments (P = 0.76), and the
latter two isomers were not detected by our method of analysis. Besides CLA isomers, the
reduction in milk fat during diet-induced MFD is highly correlated with increases in the milk fat
concentration of many trans-C18:1 isomers (Kadegowda et al., 2008). EARLY decreased
concentrations of rumen biohydrogenation intermediates including several milk trans C18:1 fatty
acid isomers and specifically trans-10 C18:1 (0.24 vs. 0.32%, P = 0.009) compared with LATE.
Although a relationship between the trans-10 C18:1 increase in milk fat and the reduction in milk
fat concentration has been demonstrated (Loor et al., 2005), abomasal infusions of pure trans-10
C18:1 in dairy cows showed this isomer did not reduce milk fat synthesis and was not the direct
cause of diet-induced MFD (Lock et al., 2007). While trans-10 C18:1 is not causative, it might be
an indicator of other isomers that could play a role in the reduction of milk fat but have not yet
been identified.
EARLY increased the concentration of C4 (3.26 vs. 2.96 g/100 g total fatty acids, P =
0.002) in the milk compared with LATE, which is consistent with the higher concentration of
ruminal butyrate (22.5 vs. 20.6 mM, P = 0.02, Table 41) for EARLY than LATE. Despite the
greater concentration of C4 fatty acid in the milk, EARLY tended to decrease the proportion of
de novo synthesized (< C16) milk fatty acids (24.3 vs. 25.2 g/100 g total fatty acids, P = 0.09)
compared with LATE. The proportion of milk fatty acids from preformed (> C16) fatty acids was
not different (P = 0.80) between EARLY and LATE. Therefore, de novo synthesized (short and
medium chain) and preformed (long chain) fatty acids were similarly responsible for decreased
milk fat observed for LATE.

196

EARLY increased pdNDF intake (4.88 vs. 4.54 kg/d, P = 0.002) and decreased iNDF
intake (1.83 vs. 2.00 kg/d, P = 0.003) compared with LATE (Table 44) because of differences in
chemical composition of forages due to maturity. However, grass maturity and its interaction
with pDMI did not affect total NDF intake (P $ 0.17, Table 44) or DMI (P $ 0.70, Table 45).
These results are contrary to our expectation that LATE would be more filling than EARLY
causing greater rumen distention and potentially limiting DMI, particularly in cows with high
DMI for which ruminal distention is more likely to limit feed intake (Allen, 1996). Oba and
Allen (1999) and Voelker et al. (2002) found DMI responses varied by production level, which is
generally correlated with DMI (NRC, 2001), where DMI was increasingly limited by high fill
diets compared with low fill diets as milk yield increased. Additionally, as DMI became more
limited by fill, we expected passage rate of iNDF to increase more for EARLY than for LATE.
Although EARLY increased ruminal passage rate of iNDF compared with LATE (2.76 vs.
2.10%/h, P < 0.001, Table 46), effect of treatment on passage rate of iNDF was not related to
pDMI (P $ 0.32).
The lower concentration and faster rate of passage of iNDF for EARLY are reflected in
the lower rumen pool sizes (Table 47) of iNDF (2.76 vs. 4.17 kg, P < 0.001) compared with
LATE, which also resulted in lower rumen pools of NDF (6.03 vs. 7.30 kg, P < 0.001), OM
(10.4 vs. 11.8 kg, P < 0.001), and DM (11.4 vs. 12.8 kg, P < 0.001) for EARLY than LATE.
Furthermore, rumen content wet weight tended to be lower for EARLY than LATE (P = 0.06,
Table 47) and volume increased at a faster rate for EARLY than LATE as pDMI increased
(interaction P = 0.10, Figure 21) such that cows with high pDMI had similar volumes for
EARLY and LATE.
EARLY increased ruminal digestion rate of pdNDF (4.95 vs. 3.98%/h, P = 0.007) and

197

decreased ruminal passage rate of pdNDF (0.66 vs. 1.32%/h, P = 0.04) compared with LATE
(Table 46). The difference in ruminal passage rate of pdNDF between EARLY and LATE
tended to be greatest for cows in the middle of the pDMI range (interaction P = 0.11, Figure
22A), and the reason for this is not known. These rates affected fiber digestion such that
EARLY had greater pdNDF ruminal digestion (4.27 vs. 3.34 kg/d, P = 0.01) and digestibility
(88.4 vs. 74.5%, P = 0.02) and lower pdNDF flux to the duodenum (0.55 vs. 1.13 kg/d, P = 0.03)
than LATE (Table 44). The patterns for treatment by pDMI interactions for pdNDF ruminal
digestibility (interaction P = 0.10, Figure 22B) and pdNDF flux (interaction P = 0.08, Figure
22C) correspond to pdNDF ruminal passage rate; such that the faster rate of pdNDF passage
yielded lower pdNDF digestibility in the rumen and higher pdNDF flux from the rumen and vice
versa. Despite differences in ruminal pdNDF digestion, there was no difference in total tract
digestibility of pdNDF between treatments. Similar to pdNDF, EARLY had greater true ruminal
digestibility of OM (67.4 vs. 60.2%, P = 0.01) than LATE, and the difference was greater for
cows in the middle of the pDMI range (interaction P = 0.05), but EARLY tended to decrease
total tract OM digestibility (62.3 vs. 66.2%, P = 0.06) compared with LATE (Table 45).
EARLY decreased DM total tract digestibility (61.2 vs. 65.5%, P = 0.04) compared with LATE
(Table 45), which is likely due to the lower digestibility of forage compared with concentrate and
lower inclusion of supplemental concentrate for EARLY than LATE.
As expected, ruminal turnover rate of NDF (4.43 vs. 3.54%/h, P < 0.001) was greater for
EARLY than LATE due to faster rates of passage of iNDF and digestion of pdNDF and despite
slower passage of pdNDF (Table 46). Although rate of particle size reduction was not directly
measured in this experiment, the faster passage rate of pdNDF observed for LATE could be the
result of greater fragility and breakdown of the more mature grass. Additionally, greater

198

chewing time for LATE may be responsible for greater breakdown of particles resulting in faster
passage rate of pdNDF because EARLY decreased eating time (42.1 vs. 50.9 min/kg forage
NDF, P = 0.03), ruminating time (88.6 vs. 101 min/kg forage NDF, P = 0.003), and thus total
chewing time (132 vs. 150 min/kg forage NDF, P < 0.001) per unit forage NDF intake compared
with LATE (Table 48). There was no difference in eating time per day (P = 0.31) for EARLY
and LATE, but EARLY decreased ruminating time per day (523 vs. 562 min/d, P = 0.04)
compared with LATE (Table 48). The lower ruminating time observed for EARLY may be
because of the smaller rumen pool size (Table 47) for EARLY compared with LATE.
Grass maturity and its interaction with pDMI did not affect ruminal pH (Table 41) despite
lower starch intake for EARLY compared with LATE (Table 40). This is likely because
EARLY tended to increase OM truly digested in the rumen (13.6 vs. 12.1 kg/d, P = 0.07, Table
45) compared with LATE and because increased salivary buffer secretion through greater
chewing time for LATE (Table 48) may have offset the expected reduction in pH because of
greater starch intake for LATE (Table 40).
The lack of interactions between DMI, rumen pools, chewing activity and level of intake
indicate that rumen fill and chewing time were not constraints limiting DMI for cows with high
intake, especially cows consuming EARLY. Based on the results obtained from the data and
samples collected during this experiment, the limiting constraint for DMI is not clear.
Metabolites or hormones (not measured) or both might have provided additional clues to
determine what limited intake.
Effects of pDMI on Ruminal Passage Rates
Experimental data on rates of passage from the rumen, particularly for individual feed
fractions, are scarce. Given the impact of passage on ruminal digestion and microbial growth,

199

quantitative knowledge of rates of nutrient passage from the rumen is needed to better
understand nutrient availability in ruminants and improve nutrition models. Furthermore,
because passage rates from the rumen generally increase with increased intake, measurements of
ruminal passage rates of nutrients over a wide range of DMI are necessary. We measured the
effects of DMI on rates of passage of feed fractions from the rumen using the pool and flux
method (Robinson et al., 1987).
Rate of ruminal passage for starch was related to level of intake. As pDMI increased,
EARLY slightly decreased ruminal passage rate of starch and LATE increased it (interaction P =
0.03, Figure 23A). This is consistent with expected effects of passage rate on true rumen
digestibility of starch; EARLY slightly increased true ruminal starch digestibility and LATE
decreased it as pDMI increased (interaction P = 0.13, Figure 23B). The reverse was observed for
postruminal starch digestibility (interaction P = 0.10, Figure 23C).
Fiber passage rates were also related to level of intake. Rate of ruminal passage for
pdNDF tended to be related to pDMI (interaction P = 0.11) and influenced pdNDF ruminal
digestion as previously discussed (Figure 22). Rate of ruminal passage of iNDF tended to be
quadratically related to pDMI (P = 0.08, Figure 24) independent of treatment response.
Although we expected iNDF passage rate to increase as pDMI increased, we observed cows with
low pDMI had faster iNDF passage rates than cows in the middle of the pDMI range. The
reason for this is unclear.
Effects of Treatment and pDMI on N Flux and Microbial Efficiency
Flux of NANMN passed from the rumen to the duodenum was related to pDMI and the
response differed by treatment; EARLY increased NANMN flux and LATE decreased it as
pDMI increased (interaction P = 0.04, Figure 25). This interaction contributed to the tendency

200

for a treatment by pDMI interaction for NAN flux (interaction P = 0.12, Table 42) as level of
intake did not affect MN flux. This is in contrast to the results in a review by Clark et al. (1992)
that reported positive linear relationships between OM intake and fluxes of NAN, NANMN, and
MN as OM intake increased over a very wide range (3 to 23 kg/d) compared with this study.
Based on studies with continuous culture fermenters, increases in solid and liquid dilution rates,
which can be associated with increased intake, resulted in greater microbial efficiency (Crawford
et al., 1980; Shriver et al., 1986). In this experiment, microbial efficiency was not related to
pDMI (P $ 0.29, Table 42) despite the linear increase in true ruminally digested OM with
increasing pDMI (P = 0.009, Table 45). There was no relationship between microbial efficiency
2

and true ruminally digested OM for cows consuming EARLY (P = 0.68, R = 0.02); however,
microbial efficiency of cows consuming LATE decreased dramatically as true ruminally digested
2

OM increased (P = 0.03, R = 0.37, Figure 26), indicating that factors other than availability of
energy limited efficiency of MN production and energy from OM fermentation was uncoupled
from microbial growth (Russell and Cook, 1995). Similarly, Oba and Allen (2003b) reported
that efficiency of MN production was inversely related to true ruminally digested OM (kg/d) in
an experiment comparing starch concentration and fermentability of diets in dairy cattle.

CONCLUSIONS
Delaying the harvest of grass resulted in lower CP and higher NDF and iNDF
concentrations. EARLY decreased milk yield and increased milk fat concentration compared
with LATE, which was likely because of a lower inclusion of concentrate for EARLY than
LATE. Grass maturity and its interaction with pDMI did not affect FCM yield, DMI, rumen pH,
or microbial efficiency. EARLY increased rates of ruminal digestion of pdNDF and passage of

201

iNDF compared with LATE. The faster passage rate of iNDF for EARLY resulted in lower
rumen pools of iNDF, total NDF, OM, and DM for EARLY than LATE. Ruminal passage rates
of pdNDF and starch were related to level of intake (quadratic interactions) and subsequently
affected ruminal digestion. EARLY decreased eating, ruminating, and total chewing time per
unit of forage NDF intake compared with LATE. When grass silage was the only source of
forage in the diet, cows supplemented with additional concentrate to account for decreasing
protein and increasing fiber concentrations associated with more mature grass produced similar
FCM yields as cows fed less mature grass.

202

APPENDIX

203

Table 36. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet
Standard
Parameter
Median
Mean
deviation
Minimum
Maximum
Parity
3
2.77
1.07
2
6
1
598
588
55
498
665
BW , kg
BCS
2.42
2.48
0.68
1.50
4.00
DIM
126
164
85
73
329
Milk, kg/d
43.2
43.2
7.7
29.1
55.7
3.5% FCM, kg/d
44.9
43.7
7.7
30.8
57.2
DMI, kg/d
26.2
26.1
1.4
23.5
28.2
1
Empty BW (ruminal digesta removed).

204

Table 37. Chemical composition, particle size distribution, and fermentation parameters of the
early or late cut orchardgrass silage included in the treatment diets
Orchardgrass silage
Item
Early
Late
Chemical composition
DM, %
37.3
31.9
OM, % DM
90.7
92.2
NDF, % DM
44.9
54.4
1
10.8
14.5
iNDF , % DM
iNDF, % of NDF
24.1
26.6
ADF, % of DM
28.3
33.5
2
4.08
4.67
ADL , % of DM
CP, % DM
19.9
14.8
Starch, % DM
2.92
2.10
3
Particles size distribution
Wet sieving, % DM retained
19.0 mm
18.1
20.6
9.50 mm
22.9
24.3
4.75 mm
32.9
24.6
2.36 mm
18.7
21.5
1.18 mm
3.98
5.17
0.600 mm
1.59
2.00
0.300 mm
0.90
0.91
0.150 mm
0.52
0.50
0.075 mm
0.26
0.25
0.038 mm
0.26
0.17
4
11.4
12.0
Mean particle size , mm
Penn State Particle Separator,
% DM retained
> 19.0 mm
37.8
39.7
19.0 to 8.0 mm
38.5
37.6
< 8.0 mm
23.7
22.7
Fermentation
pH
4.90
4.57
Acetic acid, % DM
1.37
2.10
Propionic acid, % DM
0.38
0.24
Butyric acid, % DM
<0.01
<0.01
Lactic acid, % DM
5.93
6.30
Lactic:Acetic
4.33
3.00
Ethanol, % DM
0.20
0.11
Ammonia, mM
4.31
3.12
1
iNDF = indigestible NDF.
2

ADL = acid detergent lignin.

3

Particle size distributions of silages were measured each period (n = 2).

205

Table 37 (cont’d)
4

Mean particle size calculated from particle size distribution determined by wet sieving.

206

Table 38. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either early or late cut orchardgrass silage as the sole source of forage
Preliminary
Early
Late
Ingredients, % DM
Orchardgrass silage, early cut
27.5
58.2
--Orchardgrass silage, late cut
23.1
--45.9
Dry ground corn
33.3
31.1
35.2
Soybean meal (48% CP)
7.49
2.16
10.3
®
3.39
3.40
3.39
SoyPlus
1
3.99
4.00
3.99
Vitamin mineral mix

1

Limestone
Chemical composition
DM, %
OM, % DM
NDF, % DM
% forage NDF
% NDF from forage
2
iNDF , % DM
iNDF, % of NDF
CP, % DM
Starch, % DM

1.20
48.7
91.5
29.6
25.3
85.5
3
NA
NA
21.8
26.9

1.20

1.20

49.5
90.8
30.4
26.1
85.9
8.18

49.2
91.9
30.3
25.0
82.4
9.08

26.9
18.7
24.0

30.0
18.7
26.5

Vitamin mineral mix contained (DM basis) 16.5% sodium bicarbonate, 14.2% magnesium
sulfate, 7.1% salt, 5.8% dicalcium phosphate, 2.4% trace mineral premix, 0.4% vitamin A, 0.4%
vitamin D, 0.2% vitamin E, and 53.1% dry ground corn as a carrier.
2
iNDF = indigestible NDF.
3

NA = no analysis for preliminary diet.

207

Table 39. Milk production and composition, feed intake, and BW change of cows fed treatment diets containing either early or late cut
orchardgrass silage as the sole source of forage
1

2

Treatment LSM
Early
Late
Yield, kg/d
Milk

1

FCM (3.5 %)
Milk fat
Milk protein
Milk lactose
SNF
Milk composition, %
Fat
Protein
Lactose
SNF
MUN, mg/dl
DMI, kg/d
3.5% FCM/ DMI
BW change, kg/18 d
BCS change/18 d

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

36.6

39.2

1.8

<0.001

0.08

0.19

37.6
1.35
1.14
1.71
2.06

38.1
1.31
1.22
1.84
2.22

1.6
0.06
0.04
0.08
0.10

0.49
0.34
0.008
0.001
<0.001

0.06
0.07
0.07
0.05
0.05

3.70
3.15
4.68
5.65
7.70
22.5
1.49
5.06
-0.05

3.38
3.14
4.68
5.67
10.5
22.4
1.58
9.81
0.05

0.09
0.07
0.07
0.09
0.38
0.4
0.06
3.11
0.05

<0.001
0.82
0.99
0.65
<0.001
0.70
0.005
0.36
0.20

NS
NS
NS
NS
NS
0.12
0.16
0.03
0.02

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS

NS

0.17
0.18
0.19
NS
NS

NS
NS
0.18
NS
NS
NS

NS
NS
NS
NS
NS

NS
NS
NS
NS
NS

0.96
NS
NS
NS
0.81
0.002
0.83
NS
NS

0.07
NS
NS
NS
0.45
NS
0.14
NS
NS

NS
NS
NS
NS
0.10
NS
NS
NS
NS

NS
NS
NS
NS
0.17
NS
NS
NS
NS

3

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

208

Table 40. Starch digestion of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of
forage
1

Treatment LSM
Early
Late
Starch
Intake, kg/d

1

Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
% of duodenal passage
Apparent total tract digestion
kg/d
%

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

5.48

5.98

0.14

<0.001

NS

3

0.02

NS

NS

NS

3.85
71.5

4.09
69.8

0.16
2.8

0.21
0.56

0.002
0.002

0.51
0.15

NS
0.16

0.01
0.01

NS
NS

4.05
75.2
1.52

4.30
73.3
1.76

0.16
2.8
0.22

0.21
0.52
0.20

0.004
0.002
0.007

0.50
0.15
0.04

NS
0.13
0.17

0.01
0.007
0.02

NS
NS
NS

1.24
23.4
84.1

1.52
26.0
86.5

0.22
2.9
1.6

0.12
0.35
0.15

0.01
0.005
NS

0.08
0.26
0.57

0.11
0.10
0.15

0.03
0.01
NS

NS
NS
NS

5.18
94.7

5.71
95.6

0.14
0.4

<0.001
0.17

NS
0.02

0.03
0.14

NS
NS

NS
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

209

Table 41. Ruminal VFA concentrations and pH of cows fed treatment diets containing either early or late cut orchardgrass silage as
the sole source of forage
1

Treatment LSM
Early
Late
Total VFA, mM

1

Acetate
Propionate
Butyrate
Lactate
Isobutyrate
Valerate
Isovalerate
Branch chain VFA
Acetate:Propionate
Ruminal pH

158
92.1
37.6
22.5
0.070
1.81
2.92
2.30
4.11
2.50
5.66

161
92.8
41.5
20.6
0.028
1.62
2.81
1.89
3.51
2.30
5.64

2

P
SE

Trt

3

0.18

1.0
1.8
0.8
0.046
0.09
0.19
0.12
0.19
0.10
0.05

0.54
0.01
0.02
0.54
0.13
0.44
0.01
0.03
0.04
0.51

Trt x
Period
3

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS

pDMI

Trt x
pDMI

pDMI x
pDMI

NS

NS

NS

Trt x
pDMI x
pDMI
NS

NS
0.63
0.64
NS
NS
NS
NS
NS
0.57
0.54

NS
0.15
0.99
NS
NS
NS
NS
NS
0.12
0.30

NS
NS
0.17
NS
NS
NS
NS
NS
NS
0.14

NS
NS
0.16
NS
NS
NS
NS
NS
NS
0.19

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

210

Table 42. Nitrogen metabolism of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of
forage
1

Treatment LSM
Early
Late
N intake, g/d

670

Ruminal ammonia, mg/dl
Flow to duodenum
Ammonia N, g/d
NAN
g/d
% of N intake
4

NANMN
g/d
% of N intake
% of duodenal NAN
Microbial N
g/d
% of duodenal NAN
5

SE

Trt
0.89

13

Trt x
Period
0.01

pDMI
0.002

Trt x
pDMI
3

pDMI x
pDMI
NS

Trt x
pDMI x
pDMI
NS

NS

NS

11.0

12.9

0.4

0.001

NS

0.04

NS
NS

15.0

17.5

1.3

0.01

NS

0.95

0.15

0.07

NS

535
80.9

531
86.6

22
4.3

0.88
0.31

0.09
NS

0.53
0.14

0.12
0.06

NS
0.18

NS
0.16

135
20.8
24.5

167
26.0
31.2

13
1.9
2.2

0.05
0.05
0.07

0.18
NS
NS

0.59
0.16
0.29

0.04
0.04
NS

0.07
0.03
0.04

NS
NS
NS

418
75.5
31.0

382
68.8
30.4

22
2.2
1.7

0.31
0.07
0.85

NS
NS
0.02

NS
0.29
NS

NS
NS
NS

NS
0.04
NS

NS
NS
NS

343
52.0
55.0

31
4.9
2.5

0.02
0.02
0.002

NS
NS
NS

0.54
0.19
NS

0.20
0.18
NS

0.18
0.21
NS

0.15
0.17
NS

438
65.1

11
1.2

0.003
<0.001

0.10
0.15

0.005
NS

NS
NS

NS
NS

NS
NS

g/kg TRDOM
NAN apparent postruminal digestion
g/d
250
% of N intake
37.1
% of duodenal passage
44.5
N apparent total tract digestion
g/d
376
%
56.1

1

672

2

P

Treatment least squares means.

211

Table 42 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

NANMN = non-ammonia non-microbial nitrogen.

5

TRDOM = true ruminally digested OM.

212

Table 43. Milk fatty acid profile of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of
forage
1

Fatty acid, g/100g total fatty
acids
C4:0
C5:0
C6:0
C7:0
C8:0
C9:0
C10:0
C11:0
C12:0
C13:0 iso
C13:0 anteiso
C13:0
C14:0 iso
C14:0
C15:0 iso
C14:1n5t
C15:0 anteiso
C14:1n5c
C16:0 iso
C16:0
C16:1n7t
C16:1n7c
C17:0
C17:1n7c
C18:0

Treatment LSM
Early
Late

2

P
SE

Trt

3.26
0.037

2.96
0.036

0.06
0.003

0.002
0.82

1.74
0.026
0.972
0.030
2.19
0.264
2.83
0.026
0.090
0.195
0.113
10.0
0.216
0.008
0.440
0.976
0.227
29.1
0.329
1.89
0.484
0.231
11.3

1.68
0.029
0.975
0.037
2.33
0.282
3.10
0.035
0.107
0.239
0.115
10.3
0.229
0.009
0.485
1.09
0.237
28.8
0.356
2.07
0.529
0.266
10.1

0.02
0.002
0.021
0.003
0.07
0.009
0.10
0.001
0.004
0.009
0.01
0.2
0.005
0.001
0.010
0.037
0.018
0.8
0.007
0.10
0.016
0.021
0.4

0.06
0.12
0.87
0.008
0.07
0.09
0.02
<0.001
0.006
<0.001
0.78
0.15
0.004
0.63
<0.001
0.02
0.54
0.46
<0.001
0.02
<0.001
0.04
0.007

213

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

0.01

0.81
0.29

0.13
0.03

0.55
NS

Trt x
pDMI x
pDMI
0.11
NS

NS
0.28
NS
0.29
NS
NS
0.16
NS
NS
0.83
0.85
0.05
NS
0.68
0.20
NS
NS
0.79
0.40
0.22
0.52
0.69
NS

NS
0.02
NS
0.09
NS
NS
NS
NS
NS
0.09
0.06
NS
NS
0.18
0.02
NS
NS
NS
NS
NS
0.09
NS
NS

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.02
NS
0.14
NS
NS
NS
0.08
0.001
0.13
NS
0.17
NS

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS

3

NS
NS
NS
NS
NS
0.14
NS
0.08
NS
NS
NS
NS
0.10
NS
NS
NS
NS
NS
NS
NS
0.20
NS
NS
NS

Table 43 (cont’d)
C18:1 t6-8
C18:1 t9
C18:1 t10
C18:1 t11
C18:1 t12
C18:1 c9
C18:1 c11
C18:1 c12
C18:2n6c
C20:0
C18:3n6c
C18:3n3c
C20:1n12c
CLA (c9, t11)
CLA (t10, c12)
C20:2n6c
C20:3n6c
C20:3n3c
C20:4n6c
C24:0
C22:5n18c
C22:5n15c
C22:5n3c
Unidentified

0.205
0.164
0.235
0.871
0.314
22.9
0.547
0.305
2.06
0.147
0.024
0.978
0.037
0.448
0.016
0.024
0.125
0.055
0.175
0.063
0.030
0.047
0.079
2.08

0.230
0.185
0.316
1.03
0.280
23.2
0.660
0.263
2.41
0.137
0.035
0.792
0.039
0.586
0.016
0.027
0.139
0.052
0.212
0.063
0.026
0.049
0.091
1.57

0.009
0.004
0.032
0.062
0.010
0.70
0.052
0.010
0.06
0.005
0.001
0.03
0.003
0.029
0.002
0.001
0.008
0.002
0.011
0.003
0.002
0.003
0.006
0.06

<0.001
<0.001
0.009
0.005
<0.001
0.67
0.008
<0.001
<0.001
0.02
<0.001
<0.001
0.53
<0.001
0.76
0.01
<0.001
0.44
<0.001
0.80
0.002
0.58
0.02
<0.001

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.12
0.12
0.08
NS
NS
0.02
NS

0.24
0.14
NS
NS
NS
0.80
0.71
0.20
0.23
NS
NS
NS
0.90
NS
NS
NS
0.84
0.14
0.30
0.16
NS
0.53
0.61
NS

0.27
NS
NS
NS
NS
NS
NS
NS
0.16
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.31
NS
NS
0.05
0.35
NS

0.05
0.09
NS
NS
NS
0.01
0.15
NS
NS
NS
NS
NS
0.18
NS
NS
NS
0.07
0.16
0.42
0.03
NS
NS
0.11
NS

0.05
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
0.009
NS
NS
NS
0.17
NS

C18:1 trans isomers
< C16
C16
> C16

1.76
24.3
31.5
41.9

2.06
25.2
31.4
41.7

0.08
0.5
0.8
1.0

<0.001
0.09
0.80
0.80

NS
0.19
NS
0.17

0.41
0.16
0.69
0.29

0.32
NS
NS
NS

0.83
0.09
0.11
0.05

0.11
NS
NS
NS

214

Table 43 (cont’d)
1

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

215

Table 44. NDF digestion of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of forage
1

Treatment LSM
Early
Late
NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Potentially digestible NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Indigestible NDF
Intake, kg/d
1
Treatment least squares means.

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI
3

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS

NS

6.71

6.53

0.11

0.17

0.13

<0.001

NS

4.27
64.1
2.38

3.34
52.1
3.08

0.17
2.0
0.15

0.01
0.009
0.01

0.08
<0.001
0.001

<0.001
0.07
0.25

0.61
0.51
0.62

0.04
0.03
0.23

0.15
0.12
0.14

-0.91

-0.29

0.18

0.06

NS

0.04

0.99

0.02

0.15

3.29
49.1

2.97
45.6

0.13
1.8

0.12
0.23

NS
0.14

0.06
NS

NS
NS

NS
NS

NS
NS

4.88

4.54

0.07

0.002

NS

<0.001

NS

NS

NS

4.27
88.4
0.55

3.34
74.5
1.13

0.17
2.9
0.13

0.01
0.02
0.03

0.08
0.003
0.003

<0.001
0.10
0.38

0.61
0.48
0.49

0.04
0.03
0.05

0.15
0.10
0.08

-0.91

-0.29

0.18

0.06

NS

0.04

0.99

0.02

0.15

3.29
67.4

2.97
65.4

0.13
2.6

0.12
0.61

NS
NS

0.06
NS

NS
NS

NS
NS

NS
NS

1.83

2.00

0.04

0.003

0.007

0.002

NS

NS

NS

216

Table 44 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

217

Table 45. DM and OM digestion of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source
of forage
1

2

Treatment LSM
Early
Late
DM
Intake, kg/d

1

Apparent total tract digestion
kg/d
%
OM
Intake, kg/d
Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
Apparent total tract digestion
kg/d
%

P
SE

Trt

Trt x
Period

22.5

22.4

0.4

0.70

0.12

0.002

NS

13.8
61.2

14.6
65.5

0.4
1.2

0.09
0.04

NS
0.03

20.4

20.5

0.4

0.78

9.61
47.5

8.40
41.7

0.51
2.0

13.6
67.4
10.5

12.1
60.2
11.8

3.02
15.0
12.7
62.3

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

3

NS

NS

0.01
NS

NS
NS

NS
NS

NS
NS

0.11

0.002

NS

NS

NS

0.18
0.14

0.03
<0.001

0.01
0.67

0.22
0.23

0.32
0.44

0.10
0.09

0.5
1.6
0.5

0.07
0.01
0.13

0.16
<0.001
<0.001

0.009
0.49
0.02

0.29
0.20
0.25

0.32
0.61
0.79

0.14
0.05
0.09

5.20
25.7

0.42
2.0

0.01
0.02

0.01
0.009

0.43
0.48

0.39
0.40

0.30
0.10

0.04
0.05

13.6
66.2

0.3
1.1

0.06
0.06

NS
0.03

0.01
NS

NS
NS

NS
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).

218

Table 45 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

219

Table 46. Rumen kinetics of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of
forage
1

Treatment LSM
Early
Late
Ruminal turnover rate, %/h
DM
OM
NDF

4

pdNDF
Starch
Ruminal turnover time, h
DM
OM
NDF
pdNDF
5

1

iNDF
Starch
Ruminal passage rate, %/h
pdNDF
iNDF
Starch
Ruminal digestion rate, %/h
pdNDF
Starch

2

P
SE

Trt

Trt x
Period
3

7.85

6.90

0.64

0.002

7.83
4.43
6.18

6.86
3.54
5.52

0.65
0.35
0.35

0.002
<0.001
0.03

NS
NS
NS
NS

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.71

NS

0.14

NS

0.70
0.73
NS

NS
NS
NS

0.14
0.14
NS

NS
NS
NS

37.7

37.3

4.5

0.92

NS

0.69

NS

0.18

NS

12.1
12.2
21.5
16.9
35.9

13.9
13.9
27.0
19.1
51.5

0.8
0.9
1.7
1.2
4.4

0.009
0.009
<0.001
0.03
0.005

NS
NS
NS
NS
NS

NS
NS
NS
NS
0.58

NS
NS
NS
NS
0.55

NS
NS
NS
NS
0.14

NS
NS
NS
NS
0.20

NS

0.60

NS

0.12

NS

2.86

3.01

0.29

0.57

0.66
2.76
10.3

1.32
2.10
11.0

0.16
0.25
2.2

0.04
<0.001
0.64

0.002
NS
0.06

0.05
0.31
0.19

0.64
NS
0.03

0.09
0.08
0.01

0.11
NS
NS

4.95
27.2

3.98
26.1

0.50
2.4

0.007
0.71

NS
0.12

0.92
NS

NS
NS

0.19
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
220

Table 46 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

221

Table 47. Rumen pools of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of forage
1

2

Treatment LSM
Early
Late

SE

Trt

Trt x
Period

Wet weight, kg

88.4

92.7

4.6

0.06

Volume, L
Density, kg/L
Rumen pool, kg
DM
OM
NDF

104
0.85

111
0.87

5
0.02

0.07
0.32

NS
NS
NS

11.4
10.4
6.03
3.46

12.8
11.8
7.30
3.60

0.8
0.7
0.44
0.25

<0.001
<0.001
<0.001
0.27

2.76

4.17

0.32

0.64

0.72

0.06

pdNDF

4

5

1

iNDF
Starch

P
pDMI

Trt x
pDMI

pDMI x
pDMI

NS

NS

NS

0.36
0.23

0.10
0.10

NS
0.30

NS
0.05

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

<0.001

NS

0.95

0.62

0.14

0.18

0.18

NS

0.80

NS

0.12

NS

3

Trt x
pDMI x
pDMI
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

222

Table 48. Chewing activity of cows fed treatment diets containing either early or late cut orchardgrass silage as the sole source of
forage
1

Treatment LSM
Early
Late
Meals
Bouts/d
Length, min/bout
Interval, min
Meal size, kg
DM
OM
NDF
4
pdNDF
5

iNDF
Starch
Eating time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake
Rumination
Bouts/d
Length, min/bout
Interval, min
Ruminating time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake

9.07
28.9
143

8.75
31.4
150

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.42

0.64

0.03

NS

NS

NS

0.009
0.11

NS
NS
NS

1.5
8

0.28
0.58

NS
NS

NS
NS

NS
NS

3

2.57
2.33
0.76
0.56

2.60
2.39
0.76
0.53

0.13
0.12
0.04
0.03

0.85
0.72
0.97
0.56

NS
NS
NS
0.11

0.05
0.05
0.05
0.05

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

0.21

0.23

0.01

0.21

NS

0.05

NS

NS

NS

0.62

0.69

0.03

0.13

NS

0.09

NS

NS

NS

255
11.0
36.7
42.1

268
12.7
43.4
50.9

12
1.0
3.2
3.9

0.31
0.08
0.05
0.03

NS
NS
NS
NS

NS
0.21
0.18
0.20

NS
0.88
0.90
0.93

NS
0.92
0.93
0.89

NS
0.14
0.13
0.14

14.9
35.4
57.3

15.3
36.4
50.2

0.7
2.2
2.1

0.56
0.58
0.04

NS
NS
0.14

0.57
0.16
NS

0.08
0.02
NS

0.32
0.26
NS

0.04
0.13
NS

523
23.2
77.7
88.6

562
25.3
86.5
101.2

16
0.8
2.7
3.1

0.04
0.02
0.01
0.003

NS
NS
NS
NS

0.18
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

223

Table 48 (cont’d)

1

Total chewing time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake

778
34.6
116
132

830
37.4
128
150

19
1.1
4
4

0.006
0.006
0.003
<0.001

NS
NS
NS
NS

NS
0.18
0.15
0.17

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

224

Figure 20. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
orchardgrass maturity with preliminary DMI for milk fat concentration (P = 0.07). The
preliminary DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the
preliminary period when all cows were fed a common diet.

225

Figure 21. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
orchardgrass maturity with preliminary DMI for rumen digesta volume (P = 0.10). The
preliminary DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the
preliminary period when all cows were fed a common diet.

226

Figure 22

227

Figure 22 (cont’d)

Figure 22. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
orchardgrass maturity with preliminary DMI for potentially digestible NDF (pdNDF) A) ruminal
passage rate (P = 0.11), B) ruminal digestibility (P = 0.10), and C) flux to the duodenum (P =
0.08). The preliminary DMI on the x-axis are the mean DMI of individual cows during the final
4 d of the preliminary period when all cows were fed a common diet.

228

Figure 23

229

Figure 23 (cont’d)

Figure 23. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
orchardgrass maturity with preliminary DMI for starch A) ruminal passage rate (P = 0.03), B)
true ruminal digestibility (P = 0.13), and C) postruminal digestibility (P = 0.11). The
preliminary DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the
preliminary period when all cows were fed a common diet.

230

Figure 24. Relationship of early (open circles) and late (closed circles) orchardgrass maturity
with preliminary DMI for indigestible (iNDF) ruminal passage rate (P = 0.08). The preliminary
DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the preliminary
period when all cows were fed a common diet.

231

Figure 25. Interaction of early (open circles, dashed line) and late (closed circles, solid line)
orchardgrass maturity with preliminary DMI for nonammonia, nonmicrobial N flux to the
duodenum (P = 0.04). The preliminary DMI on the x-axis are the mean DMI of individual cows
during the final 4 d of the preliminary period when all cows were fed a common diet.

232

Figure 26. Relationship between true ruminally digested OM (TRDOM) and microbial
2
efficiency for early (open circles, dashed line; P = 0.68, R = 0.02) and late (closed circles, solid
line; microbial efficiency, g of microbial N/kg of TRDOM = 72.0 – 3.22 x TRDOM, kg/d; P =
2
0.03, R = 0.37) orchardgrass maturity.

233

REFERENCES

234

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Bauman, D. E., J. W. Perfield II, K. J. Harvatine, and L. H. Baumgard. 2008. Regulation of fat
synthesis by conjugated linoleic acid: Lactation and the ruminant model. J. Nutr.
138:403-409.
Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64-75.
Chouinard, P. Y., L. Corneau, D. M. Barbano, L. E. Metzger, and D. E. Bauman. 1999.
Conjugated linoleic acids alter milk fatty acid composition and inhibit milk fat secretion in
dairy cows. J. Nutr. 129:1579–1584.
Christie, W. W. 1982. A simple procedure for rapid transmethylation of glycerolipids and
cholesteryl esters. J. Lipid Res. 23:1072–1075.
Clark, J. H., T. H. Klusmeyer, and M. R. Cameron. 1992. Microbial protein synthesis and flows
of nitrogen fractions to the duodenum of dairy cows. J. Dairy Sci. 75:2304-2323.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting digestibility of
different diets with internal markers: Evaluation of four potential markers. J. Anim. Sci.
63:1476-1483.
Cook, R. D., and S. Weisberg. 1982. Residuals and influence in regression. Chapman
and Hall, New York.
Crawford, R. J., Jr., W. H. Hoover, and L. L. Junkins. 1980. Effects of solids and liquid flows on
fermentation in continuous cultures. II. Nitrogen partitioning and efficiency of microbial
synthesis. J. Anim. Sci. 51:986-995.
Dado, R. G, and M. S. Allen. 1994. Variation in and relationships among feeding, chewing,
and drinking variables for lactating dairy cows. J. Dairy Sci. 77:132-144.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.
Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More powerful peroxide
Kjeldahl digestion method. J. AOAC 70:783-787.
Hara, A., and N. S. Radin. 1978. Lipid extraction of tissues with a low-toxicity solvent. Anal.
Biochem. 90:420–426.

235

Joy, M. T., E. J. DePeters, J. G. Fadel, and R. A. Zinn. 1997. Effects of corn processing on the
site and extent of digestion in lactating cows. J. Dairy Sci. 80:2087-2097.
Jung, H. G., and M. S. Allen. 1995. Characteristics of plant cell walls affecting intake and
digestibility of forages by ruminants. J. Anim. Sci. 73:2774-2790.
Kadegowda, A. K., L. S. Piperova, and R. A. Erdman. 2008. Principle component and
multivariate analysis of milk long-chain fatty acid composition during diet-induced milk
fat depression. J. Dairy Sci. 91:749-759.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019-1027.
Kramer, J. K., V. Fellner, M. E. Dugan, F. D. Sauer, M. M. Mossoba, and M. P. Yurawecz.
1997. Evaluating acid and base catalysts in the methylation of milk and rumen fatty acids
with special emphasis on conjugated dienes and total trans fatty acids. Lipids. 32:12191228.
Lammers, B. P., D. R. Buckmaster, and A. J. Heinrichs. 1996. A simplified method for the
analysis of particle sizes of forage and total mixed rations. J. Dairy Sci. 79:922-928.
Lock, A. L., C. Tyburczy, D. A. Dwyer, K. J. Harvatine, F. Destaillats, Z. Mouloungui, L.
Candy, and D. E. Bauman. 2007. Trans-10 octadecenoic acid does not reduce milk fat
synthesis in dairy cows. J. Nutr. 137:71-76.
Loor, J. J., A. Ferlya, A. Ollier, and Y. Chillard. 2005. Relationship among trans and conjugated
fatty acids and bovine milk fat yield due to dietary concentrate and linseed oil. J. Dairy
Sci. 88:726–740.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:12171240.
National Research Council. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl.
Acad. Washington, DC.
Oba, M., and M. S. Allen. 1999. Effects of brown midrib 3 mutation in corn silage on dry
matter intake and productivity of high yielding dairy cows. J. Dairy Sci. 82:135-142.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174–183.
Oba, M., and M. S. Allen. 2003b. Effects of diet fermentability on efficiency of microbial
nitrogen production in lactating dairy cows. J. Dairy Sci. 86:195-207.

236

Poppi, D. P., D. J. Minson, and J. H. Ternouth. 1981. Studies of cattle and sheep eating leaf and
stem fractions of grasses. III. The retention time in the rumen of large feed particles.
Aust. J. Agric. Res. 32:123-137.
Robinson, P. J., S. Tamminga, and A. M. van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci. 17:37-62.
Russell, J. B., and G. M. Cook. 1995. Energetics of bacterial growth: balance of anabolic and
catabolic reactions. Microbiol. Rev. 59:48-63.
Shriver, B. J., W. H. Hoover, J. P. Sargent, R. J. Crawford, and W. V. Thayne. 1986.
Fermentation of a high concentrate diet as affected by ruminal pH and digesta flow. J.
Dairy Sci. 69:413-419.
Smith, L. W., H. K. Goering, and C. H. Gordon. 1972. Relationships of forage compositions with
rates of cell wall digestion and indigestibility of cell walls. J. Dairy Sci. 55:1140-1147.
Ulyatt, M. J. 1983. Plant fibre and regulation of digestion in the ruminant. In: G. Wallace and L.
Bell (ed.) Fibre in Human and Animal Nutrition. p 103. The Royal Soc. Of New Zealand,
Wellington, N.Z.
Voelker, J. A., G. M. Burato, and M. S. Allen. 2002. Effect of pretrial milk yield on responses
of feed intake, digestion, and production to dietary forage concentration. J. Dairy Sci.
85:2650-2661.
Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. J. Dairy Sci. 65:495–501.
Zinn, R. A., and F. N. Owens. 1986. A rapid procedure for purine measurement and its use for
estimating net ruminal protein synthesis. Can. J. Anim. Sci. 66:157-166.

237

CHAPTER 6
Nutrient Demand Interacts with Forage Family to Affect Digestion Responses in Dairy
Cows

1

ABSTRACT
Effects of forage family on dry matter intake (DMI), milk production, ruminal pool sizes,
digestion and passage kinetics, and chewing activity and the relationship of these effects with
preliminary DMI (pDMI) were evaluated using 13 ruminally and duodenally cannulated Holstein
cows in a crossover design with a 14-d preliminary period and two 18-d treatment periods.
During the preliminary period, pDMI of individual cows ranged from 19.6 to 29.5 kg/d (mean =
25.9 kg/d) and 3.5% fat-corrected milk yield ranged from 24.3 to 60.3 kg/d (mean = 42.1 kg/d).
Experimental treatments were diets containing either a) alfalfa silage (AL) or b) orchardgrass
silage (OG) as the sole forage. Alfalfa and orchardgrass contained 42.3 and 58.2% neutral
detergent fiber (NDF) and 22.5 and 11.4% crude protein, respectively. Forage:concentrate ratios
were 60:40 for AL and 43:57 for OG; both diets contained ~25% forage NDF and ~30% total
NDF. Preliminary DMI, an index of nutrient demand, was determined during the last 4 d of the
preliminary period when cows were fed a common diet and used as a covariate. Main effects of
forage family and their interaction with pDMI were tested by ANOVA. Forage family and its
interaction with pDMI did not affect milk yield or composition. Alfalfa decreased efficiency of
milk production compared with OG because of numerically greater DMI for AL. Alfalfa
increased indigestible NDF (iNDF) intake and decreased potentially digestible NDF (pdNDF)
intake compared with OG because of differences in chemical composition of forages. Alfalfa
1

Submitted for publication in the Journal of Dairy Science on October 5, 2011.

238

increased ruminal pH, digestion rates of pdNDF and starch, and passage rates of pdNDF and
iNDF compared with OG, which affected ruminal digestibility. Passage rate of iNDF was
related to pDMI; AL increased iNDF passage rate and OG decreased it as pDMI increased.
Alfalfa decreased rumen pool sizes of pdNDF, starch, dry matter, and rumen digesta wet weight
and volume compared with OG. Alfalfa decreased ruminating time per unit of forage NDF
consumed compared with OG indicating that alfalfa provided less effective fiber than
orchardgrass. Alfalfa, but not OG, increased ammonia N, nonammonia nonmicrobial N, and
nonammonia N fluxes as pDMI increased. Microbial efficiency was positively related to pdNDF
passage rate for OG, but not AL. When alfalfa or orchardgrass silage was the only source of
forage in diets formulated to contain similar concentrations of forage NDF, forage family did not
affect productivity.

INTRODUCTION
Utilization of diets by dairy cows is largely influenced by the nutrient composition and
physical characteristics of the forage in the ration. Large differences exist among forage families
(grasses and legumes) including chemical composition, anatomical characteristics, and digestion
characteristics that affect digestibility (Allen, 1996; Wilson and Kennedy, 1996). The
relationship between NDF and its digestibility with feed intake varies across forages. Coolseason grasses and legumes differ in concentration and the rate and extent of digestion of fiber
(Van Soest, 1982). Grasses generally contain higher total NDF and potentially digestible NDF
(pdNDF) concentrations, which have a slower rate of digestion but greater extent of digestion
than legumes. The greater extent of digestion for grasses offers the potential for greater energy
availability, but slower digestion rates can result in greater ruminal retention times and

239

subsequently lower intake, possibly offsetting gains from higher digestibility (Allen, 2000).
Several lactation studies comparing legumes with grasses are reported in the literature;
however, many are confounded by the NDF differences between the two species. When diets are
formulated to contain an equal amount of forage DM, total and forage NDF concentrations of
diets generally will be higher for diets containing grasses compared with legumes. Increasing
dietary NDF concentration often has a negative impact on the amount of DM consumed by
lactating dairy cows (Allen, 2000). In this experiment, rations were formulated to contain
similar forage NDF concentrations in order to specifically measure the effects of forage fiber.
Alfalfa (Medicago sativa) and orchardgrass (Dactylis glomerata L.) were selected as a
representative legume and cool-season grass, respectively.
In addition to the combination of dietary factors affecting ruminal digestion and
distention, the individual cow’s appetite will also affect the responses of passage rate and intake
to forage family. Voelker Linton and Allen (2008) found that the response of DMI to forage
family depended on the appetite of individual cows as intake was more restricted by orchardgrass
than alfalfa as level of intake increased. Because forage family and level of intake affect ruminal
passage and digestion rates and, thus, digesta fill in the rumen, the response to effects of forage
family and its relationship with intake level was assessed to determine if responses to treatment
vary among cows with a wide range in DMI. We hypothesized that responses of DMI and
passage rates to forage family are related to level of intake and legumes will permit a greater
increase in passage rate than grasses as feed intake increases.
The objectives of this experiment were to evaluate the relationships between voluntary
DMI and effects of forage family on DMI, milk production, ruminal fermentation and pool sizes,
digestion and passage kinetics, and chewing behavior in lactating dairy cows. This study had

240

three unique features to improve our understanding of the role of forage family and interpret its
effect on animal responses. First, it allowed effects of the interaction between forage family and
preliminary DMI (pDMI) to be evaluated. The use of pDMI, an index of nutrient demand,
allowed the evaluation of treatments on animal responses in relation to level of intake and
provided an indicator to test effects of intake level independent of treatments. Second, it directly
compared treatment effects of alfalfa and orchardgrass as the sole source of forage. Third,
ruminal passage rates of individual feed fractions, instead of entire feeds, were measured using
ruminally and duodenally cannulated cows.

MATERIALS AND METHODS
This article is the first of a set of two from one experiment that evaluated the effects
forage family and its interaction with level of feed intake (nutrient demand). This article
discusses the effect of pDMI on responses to treatment for production, rumen parameters and
kinetics, and chewing activity. The companion article focuses on rates of particle size
breakdown in, and particle passage from, the rumen.
Cows and Treatments
Experimental procedures were approved by the Institutional Animal Care and Use
Committee at Michigan State University. Thirteen multiparous Holstein cows from the
Michigan State University Dairy Cattle Teaching and Research Center were assigned randomly
to treatment sequence in a crossover design experiment with one 14-d preliminary period and
two 18-d experimental periods. The first 10 d of each period were allowed for diet adaptation
and samples were collected during the final 4 d of the preliminary period and 8 d of each
experimental period. Cows were 157±90 (mean±SD) DIM at the end of the preliminary period

241

and were selected to provide a wide range and uniform distribution of pDMI and milk yield.
During the final 4 d of the 14-d preliminary period, the average pDMI among cows ranged from
19.6 to 29.5 kg/d (mean = 25.9 kg/d) and 3.5% FCM yield ranged from 24.3 to 60.3 kg/d (mean
= 42.1 kg/d; Table 49). Prior to calving, cows were cannulated ruminally (Bar Diamond, Inc.,
Parma, ID) and duodenally with a gutter-type T cannula placed approximately 10 cm distal to the
pylorus (Joy et al., 1997). Surgery was performed at the Department of Large Animal Clinical
Science, College of Veterinary Medicine, Michigan State University.
Experimental treatments were diets containing either a) alfalfa silage (AL) or b)
orchardgrass silage (OG) as the sole forage. Alfalfa (Pioneer 54H91, Pioneer Hi-Bred, Johnston,
IA) and orchardgrass (Baridana cultivar, Barenbrug USA, Tangent, OR) forages were produced
at the campus farm at Michigan State University (East Lansing), chopped to 10 mm theoretical
length of cut, and ensiled in Ag-Bags (Ag-Bag Systems Inc., St. Nazianz, WI). During the
sample collection periods, alfalfa and orchardgrass contained 42.3 and 58.2% NDF and 22.5 and
11.4% CP, respectively (DM basis; Table 50). Diets AL and OG were formulated to contain
25% forage NDF, 30% total NDF, and 18% CP. We acknowledge these treatments affect dietary
starch concentration but maintaining similar forage and total NDF concentrations for both
treatments was of primary interest. The diet fed during the preliminary period was formulated so
that alfalfa and orchardgrass each contributed 50% of forage NDF. Diets also contained dry
®

ground corn, SoyPlus (West Central Soy, Ralston, IA), and vitamin-mineral premix (Table 51);
soybean meal (48% CP), urea, and limestone were used to compensate for lower CP and Ca
concentrations in orchardgrass silage than in alfalfa silage.
Data and Sample Collection
Throughout the experiment, cows were housed in tie-stalls and fed diets as total mixed

242

rations once daily (1130 h) at 110% of expected intake. The amount of feed offered and refused
(orts) was weighed daily for each cow. Forage samples were collected twice weekly and
analyzed to adjust diets to account for DM, NDF, and CP fluctuation. Samples of all dietary
ingredients (0.5 kg) and orts (12.5%) were collected daily from d 11 to 14 during the preliminary
period and d 11 to 15 during each experimental period. Samples were frozen immediately after
collection at !20°C and combined to one composite sample per period prior to analysis.
Cows were moved to an exercise lot twice daily (0230 and 1300 h) prior to milking in a
parlor (0400 and 1430 h). Milk yield was measured, and milk was sampled, at each milking on d
11 to 14 of the preliminary period and on d 11 to 15 of the experimental periods. Rumen-empty
BW was measured by weighing the cow after evacuation of ruminal digesta on d 14 of the
preliminary period and d 18 of each experimental period. Body condition score was determined
on the same days by 3 trained investigators blinded to treatments (Wildman et al., 1982; 5-point
scale where 1 = thin and 5 = fat). Chewing activity was monitored and recorded by observation
every 5 min for 24 h on d 16 of each experimental period. Activity was noted as eating,
ruminating, drinking, or idle for each cow at each time.
Duodenal samples (900 mL), fecal samples (500 g), rumen fluid and particulate samples
for microbial isolation (400 g), rumen fluid samples for pH, concentrations of VFA, lactate, and
ammonia (100 mL), and blood samples for concentrations of glucose, insulin, and glucagon were
collected every 15 h from d 11 to 15 of each experimental period so that 8 samples were taken
for each cow in each period, representing every 3 h of a 24-h period to account for diurnal
variation. Rumen fluid and particulate matter for microbial isolation was collected from the
reticulum, near the reticular-omasal orifice, transported to the laboratory, and processed. Rumen
fluid for pH, VFA, lactate, and ammonia were obtained by combining digesta from five different

243

sites in the rumen and straining it through nylon mesh (~1 mm pore size); fluid pH was recorded
immediately. Blood was sampled from coccygeal vessels and collected into 2 evacuated tubes,
one containing potassium EDTA and the other containing potassium oxalate with sodium
fluoride as a glycolytic inhibitor. Both were centrifuged at 2,000 x g for 15 min immediately
after sample collection and plasma was harvested. Samples containing potassium EDTA were
preserved with benzamidine (0.05 M final concentration). Samples were stored at -20°C.
Ruminal contents were evacuated manually through the ruminal cannula 4 h after feeding
at the beginning of d 17 (1530 h) and 2 h before feeding at the end of d 18 (0930 h) for each
experimental period. Total rumen content mass and volume were determined. To ensure
accurate sampling, every tenth handful of digesta (10%) was separated for a subsample
throughout evacuation. This subsample was squeezed into primarily solid and liquid phases.
Both phases were weighed and sampled (350 mL) for determination of nutrient pool size. All
samples were stored at -20°C.
Sample Analysis and Calculations
Milk yield recorded at each milking were summed for a daily total, which were averaged
for each period. Milk samples were analyzed for fat, true protein, lactose, SNF, and MUN with
infrared spectroscopy by Michigan DHIA (East Lansing). Yields of 3.5% FCM and milk
components were calculated using milk yield and component concentrations for each milking,
summed for a daily total, and averaged for each period.
Forage samples were combined to one composite sample per forage per period. Particle
size distribution was determined using the Penn State Particle Separator containing 2 sieves (19
and 8 mm) and a pan (Lammers et al., 1996). In addition, samples were wet sieved manually and
sequentially through screens with the following aperture sizes: 19.0, 9.50, 4.75, 2.36, 1.18,

244

0.600, 0.300, 0.150, 0.075, and 0.038 mm. The fraction of DM retained on the screens from wet
sieving was used to calculate mean particle size.
Diet ingredients, orts, and feces were lyophilized (Tri-Philizer MP, FTS Systems, Stone
Ridge, NY). All dried samples were ground with a Wiley mill (1-mm screen; Arthur H. Thomas,
Philadelphia, PA). Dried, ground fecal samples were combined on an equal DM basis into one
sample per cow per period. Frozen duodenal samples for each cow period (n=8) were chopped
finely using a commercial food processor (84142 Food cutter, Hobart Manufacturing Co., Troy,
OH) and subsampled in the frozen state to obtain representative samples. These duodenal
subsamples and the 350 mL of ruminal solid and liquid samples were lyophilized and ground as
described above. Dried ruminal solid and liquid samples were recombined according to the
original ratio of solid and liquid DM.
Samples were analyzed for ash, NDF, iNDF, ADF, acid detergent lignin (ADL), CP, and
starch. Ash concentration was determined after 5 h combustion at 500°C in a muffle furnace.
Concentrations of NDF were determined according to Mertens (2002) and ADF and ADL
according to Goering and Van Soest (1970). Indigestible NDF was estimated as NDF residue
after 240 h in vitro fermentation (Goering and Van Soest, 1970); flasks were reinoculated at 120
h to insure a viable microbial population. Forage NDF digestibility was determined by 30 h in
vitro fermentation (Goering and Van Soest, 1970). Ruminal fluid for the in vitro incubations
was collected from a nonpregnant dry cow fed dry hay only. Fraction of pdNDF was calculated
by difference (1.00 – iNDF). Crude protein was analyzed according to Hach et al. (1987).
Starch was measured by an enzymatic method (Karkalas, 1985) after samples were gelatinized
with sodium hydroxide. Glucose concentration was measured using a glucose oxidase method
(Glucose kit #510, Sigma Chemical Co., St. Louis, MO), and absorbance was determined with a

245

micro-plate reader (SpectraMax 190, Molecular Devices Corp., Sunnyvale, CA). Concentrations
of all nutrients except DM were expressed as percentages of DM determined by drying at 105°C
in forced-air oven for more than 8 h.
Duodenal digesta were analyzed for purines and ammonia to estimate microbial N (MN)
flow and nonammonia, nonmicrobial N (NANMN) flow to the duodenum. Purine concentration
was used as a microbial marker, and purine to MN ratio was estimated by analysis of microbial
pellets obtained by differential centrifugation of the rumen fluid and particulate samples
collected near the reticulum. Rumen fluid and particulate matter were blended, strained through
nylon mesh, and the liquid portion was centrifuged at 500 x g for 15 min. The supernatant was
centrifuged at 18,000 x g for 15 min, and the pellet was washed with 0.9% NaCl solution and
centrifuged again at 18,000 x g for 15 min, resuspended in water, and lyophilized. Total purines
were measured by spectrophotometer (Beckman Instruments Inc., Fullerton, CA) at 260 nm
according to Zinn and Owens (1986). Ammonia concentration was determined for centrifuged
duodenal and rumen fluid samples according to Broderick and Kang (1980). Rumen fluid was
also analyzed for concentration of major VFA and lactate by HPLC (Waters Corp., Milford,
MA) according to Oba and Allen (2003a).
Dry matter and nutrient intakes were calculated using the composition of feed offered and
refused. Ruminal pool sizes (kg) of OM, NDF, iNDF, pdNDF, starch, MN, and NANMN were
determined by multiplying the concentration of each component in rumen samples by the
ruminal digesta DM mass (kg). Duodenal flows (kg/d) of DM, OM, total NDF, pdNDF, starch,
MN, NANMN, and ammonia N were determined using iNDF as a flow marker; iNDF intake
(kg/d) was multiplied by the ratio between the component and iNDF in duodenal digesta.
Duodenal flow of microbial OM was determined using the ratio of purines to OM (Oba and

246

Allen, 2003a), and true ruminally digested OM was calculated by subtracting duodenal flow of
nonmicrobial OM from OM intake. Indigestible NDF was used as an internal marker to estimate
nutrient digestibility in the rumen and in the total tract (Cochran et al., 1986). Turnover rate in
the rumen, passage rate from the rumen, and ruminal digestion rate of each component was
calculated by the following equations:
Turnover rate (%/h) = 100 x (Intake of component / Ruminal pool of component) / 24
Passage rate (%/h) = 100 x (Duodenal flow of component / Ruminal pool of component) / 24
Digestion rate (%/h) = Turnover rate in the rumen (%/h) – Passage rate from the rumen (%/h).
Plasma samples were composited into one sample per cow per period and analyzed using
commercial kits to determine concentrations of insulin (Coat-A-Count, Siemens Healthcare
Diagnostics, Deerfield, IL), and glucagon (kit #GL-32K, Millipore, Billerica, MA). Plasma
glucose concentration was analyzed using a glucose oxidase method (Sigma Chemical Co., St.
Louis, MO).
Manually observed chewing activity was summarized by a logic script in Igor Pro
(Version 6.12, WaveMetrics Inc., Lake Oswego, OR) to generate meal and rumination bout
information according to previously established criteria (Dado and Allen, 1994). Variables
determined included frequency of meal bouts per day, interval between meals, frequency of
ruminating bouts per day, interval between ruminating bouts, eating time per day, ruminating
time per day, and total chewing time per day.
Statistical Analysis
All data were analyzed by using the fit model procedure of JMP (Version 8, SAS
Institute, Cary, NC). To determine differences between treatments and evaluate interactions of
treatment with DMI, where pDMI (calculated as the mean of DMI values on d 11 to 14 of the 14-

247

d preliminary period) was used as the covariate for treatment responses, data were analyzed
2

according to the following model: Yijk = µ + Ci + Pj + Tk + PTjk + pDMI + TkpDMI + pDMI +
2

TkpDMI + eijk where µ is the overall mean, Ci is the random effect of cow (i = 1 to 13), Pj is
the fixed effect of period (j = 1 to 2), Tk is the fixed effect of treatment (k = 1 to 2), PTjk is the
interaction of period and treatment, pDMI is the linear effect of pDMI, TkpDMI is the interaction
2

2

of treatment and pDMI (linear), pDMI is the quadratic effect of pDMI, TkpDMI is the
interaction of treatment and pDMI (quadratic), and eijk is the residual error. Statistical
2

significance for TkpDMI and TkpDMI indicated treatment differences were related to pDMI.
Covariate and interaction terms were removed stepwise from the model if P > 0.20. Treatment
effects and their interaction (linear and quadratic relationships) were declared significant at P "
0.05 and P " 0.10, respectively. Tendencies for treatment effects and their interactions were
declared at P " 0.10 and P " 0.15, respectively.
Sixteen cows started the experiment, however, two cows experienced high fevers and had
depressed intake after the first experimental period and were removed. Additionally, data from
one cow was excluded prior to statistical analysis because the calculated value for duodenal flow
was extremely high for the second experimental period and resulted in unrealistically low
ruminal digestibility. Thus, data from 13 cows were statistically analyzed.

RESULTS AND DISCUSSION

248

Comparison of Forages and Diets
Chemical analyses and physical characteristics of ensiled forages are listed in Table 50.
As expected, alfalfa had lower concentration of total NDF (42.3 vs. 58.2%) but higher
concentrations of iNDF (23.0 vs. 16.1%), ADL (7.56 vs. 6.03%), and CP (22.5 vs. 11.4%) than
orchardgrass. Indigestible NDF for alfalfa expressed as a percent of NDF was nearly twice that
for orchardgrass (54.5 vs. 27.7% of NDF). In vitro NDF digestibility (30 h) was 15 percentage
units lower for alfalfa than for orchardgrass (38.3 vs. 53.3%). Alfalfa had higher DM
concentration than orchardgrass and was drier than expected. Both silages had similar pH and
underwent lactic acid fermentation, but alfalfa had a lower lactic:acetic acid ratio than
orchardgrass. Based on wet sieving, alfalfa had greater mean particle size (11.6 vs. 9.66 mm)
than orchardgrass. Additionally, alfalfa contained a larger proportion of particles > 19 mm (29.3
vs. 17.1%; top sieve) and smaller proportion of particles < 8 mm (22.2 vs. 32.7%; bottom pan)
than orchardgrass when sieved with the Penn State Particle Separator. Although forages were
chopped to the same theoretical length of cut for both silages, the differences in particle size are
likely compared with the differences in physical characteristics between the forage species and
orientation of stems in the field.
Diet ingredients and chemical composition are shown in Table 51. The preliminary diet
contained more alfalfa silage than orchardgrass silage so each forage supplied similar
concentrations of forage NDF. Because treatment diets were formulated to contain similar
forage NDF, forage:concentrate ratios were different between diets with ratios of 60:40 and
43:57 for AL and OG, respectively. Besides forage source, differences in diets included sources
and concentrations of protein supplements and concentrations of limestone and corn grain, which
were lower for AL than OG, to account for differences in chemical composition between alfalfa

249

and orchardgrass silages. The chemical composition of each diet was mathematically calculated
according to the proportion of each feed ingredient in the diet and its respective analytical values,
and was similar for forage NDF and total NDF concentrations. Despite increasing the
®

concentration of soybean meal and adding SoyPlus and urea to increase CP in OG, AL still
contained slightly higher concentrations of CP than OG because the changes in CP concentration
of forages were greater than expected. Starch concentration was lower for AL because of more
forage and less concentrate in the diet for AL compared with OG. Indigestible NDF was higher
for AL than OG and is reflective of the iNDF concentration in the forages, which was higher for
alfalfa than orchardgrass. In both diets, forage NDF provided over 82% of the total diet NDF.
Effects of Forage Family and pDMI
Forage family and its interaction with pDMI did not affect yields of milk or milk
components or milk composition (Table 52). Mean DMI was numerically, but not statistically,
greater for AL than OG (24.2 vs. 23.2 kg/d, P = 0.13, Table 52). Alfalfa decreased efficiency of
milk production compared with OG (FCM/DMI, 1.40 vs. 1.47, P = 0.005) because AL
numerically increased DMI compared with OG to produce similar yields of FCM, and the
difference was greatest for cows with high pDMI (interaction P = 0.006, Table 52). Differences
in efficiency between AL and OG might be associated with changes in body weight as AL
increased and OG decreased body weight (6.05 vs. -3.78 kg over 18 d period, P = 0.03, Table 52)
or different concentrations of concentrate in the diets.
Our results are not consistent with lower DMI and milk production for lactating dairy
cows fed grass-based diets compared with cows fed legume-based diets (Oba and Allen, 1999;
Steinshamn, 2010), but they are consistent with Voelker Linton and Allen (2008) who reported
no treatment differences for mean milk yield and DMI for cows fed alfalfa or orchardgrass diets.

250

However, Voelker Linton and Allen (2008) found that testing overall means masked important
intake differences; response of DMI to treatment varied for cows with different nutrient
demands. Cows with low nutrient demand responded more positively to grass than legume, and
cows with high nutrient demand responded more positively to legume than grass. These
differences likely depended on the extent to which rumen fill limited feed intake of individual
cows. We expected OG to be more filling than AL causing greater rumen distention and
potentially limiting DMI, particularly in cows with high DMI for which ruminal distention is
more likely to limit feed intake (Allen, 1996; Voelker Linton and Allen, 2008). Although the
response of DMI to forage family was not related to pDMI in this experiment (P # 0.21), a visual
examination of a graph with pDMI and DMI (Figure 27) illustrates that the greatest difference in
DMI was for cows with high pDMI, which responded more positively to alfalfa than
orchardgrass-based diets.
Alfalfa decreased pdNDF intake (3.43 vs. 4.89 kg/d) and increased iNDF intake (3.53 vs.
1.80 kg/d) compared with OG (P < 0.001, Table 53) because of differences in chemical
composition of forages. Intake of iNDF was related to pDMI such that AL increased intake of
iNDF at a faster rate than OG as pDMI increased (interaction P = 0.05, Figure 28A). Alfalfa
increased rate of ruminal digestion of pdNDF (7.27 vs. 4.74%/h), rate of ruminal passage of
pdNDF (2.29 vs. 1.32 %/h), and rate of ruminal passage of iNDF (3.27 vs. 2.52%/h) compared
with OG (P < 0.001, Table 54). The faster passage rates for AL compared with OG were
associated with greater rate of particle size reduction for AL compared with OG (7.16 vs.
4.67%/h, P < 0.001, Kammes and Allen, submitted). These ruminal kinetics resulted in shorter
ruminal turnover times of pdNDF (10.9 vs. 17.4 h), iNDF (32.0 vs. 41.6 h), and DM (10.5 vs.
12.8 h) for AL than OG (P < 0.001, Table 54). Additionally, responses of iNDF ruminal passage

251

rate and turnover time to treatment were related to pDMI; as pDMI increased, AL increased rate
of iNDF passage and OG decreased it (interaction P = 0.09, Figure 28B) and AL decreased iNDF
turnover time and OG increased it (interaction P = 0.06, Figure 28C). The increased turnover
time for iNDF for OG as feed intake increased is consistent with Voelker Linton and Allen
(2008).
The aforementioned results contributed to the rumen pool sizes listed in Table 55.
Despite the faster passage rate and turnover time of iNDF for AL, AL increased the rumen pool
size of iNDF compared with OG (4.62 vs. 3.11 kg, P < 0.001) because of the greater intake of
iNDF. Alfalfa decreased rumen pools of pdNDF (1.55 vs. 3.54 kg, P < 0.001) and DM (10.6 vs.
12.4 kg, P = 0.001) compared with OG because of lower pdNDF intake and faster rates of
ruminal passage and digestion of pdNDF for AL than OG. Furthermore, AL decreased rumen
digesta wet weight (82.7 vs. 92.4 kg, P = 0.008) and volume (98.5 and 108 L, P = 0.01)
compared with OG. These ruminal pool sizes indicated that OG had greater filling effects than
AL. The numerically lower feed intake for OG was accompanied by greater rumen pools
suggesting rumen fill as a constraint limiting DMI for cows consuming OG is possible, but there
was not statistically significant evidence in this experiment that ruminal distention is more likely
to limit feed intake for cows with high intake compared with cows with low intake because we
were unable to detect a treatment by pDMI interaction for DMI.
Effects of treatment on ruminal kinetics affected fiber digestion in the rumen (Table 53).
Although rate of ruminal digestion of pdNDF was faster for AL than OG, AL decreased pdNDF
digestion in the rumen compared with OG (2.47 vs. 3.70 kg/d, P < 0.001) because of lower
concentration of pdNDF and shorter retention time in the rumen for AL than OG. As expected,
AL decreased ruminal digestibility of NDF (37.7 vs. 57.2%, P < 0.001) compared with OG. As

252

pDMI increased, AL maintained relatively constant ruminal fiber digestibility but OG tended to
increase or increased ruminal digestibilities of pdNDF (interaction P = 0.13, Figure 29A) and
NDF (interaction P = 0.08, Figure 29B), respectively. Alfalfa decreased pdNDF flux to the
duodenum (0.84 vs. 1.07 kg/d, P = 0.02) but increased NDF flux to the duodenum (4.26 vs. 2.89
kg/d, P < 0.001) compared with OG. The higher NDF flux for AL than OG is because of the
greater concentration of iNDF and lower ruminal digestibility for AL than OG. As a result of
increasing pdNDF ruminal digestibility for OG with greater feed intake, pdNDF flux to the
duodenum decreased for OG as pDMI increased (interaction P = 0.07, Figure 29C) with the
greatest difference between AL and OG for cows with low pDMI. Flux of NDF to the
duodenum increased for AL as pDMI increased (interaction P = 0.001, Figure 29D) with the
largest difference between treatments for cows with high pDMI, which is related to the greater
increase in iNDF intake for AL as pDMI increased (Figure 28A).
Similar to pdNDF digestion in the rumen, AL decreased total tract digestion of pdNDF
compared with OG (2.47 vs. 3.14 kg/d, P < 0.001). Despite greater ruminal digestion of pdNDF
for OG, AL increased total tract digestibility of pdNDF compared with OG (72.0 vs. 64.7%, P =
0.02). Alfalfa decreased total tract digestibility of NDF compared with OG (35.5 vs. 47.1%, P <
0.001) because of the higher concentration of iNDF for AL than OG. Total tract digestibilities of
NDF (and pdNDF) are lower than ruminal digestibility because negative post ruminal
digestibilities were calculated for NDF (and pdNDF) in the present experiment. This is because
of a net gain of fiber from the duodenum to the feces, which has previously been reported with
both the gutter-type T duodenal cannula (Huhtanen and Jaakkola, 1993; Poore et al., 1993),
which is the type used in this study, and the closed T-type duodenal cannula (Stensig and
Robinson, 1997). Underestimation of duodenal NDF flow or duodenal iNDF:NDF ratio using

253

iNDF as a marker creates inaccuracies of estimated flow of duodenal fiber and postruminal
digestibility. These errors may be related to unrepresentative digesta sampling due to differential
separation of fluid and particles relative to the true material flowing out of the duodenum or
analytical problems in fiber determination of duodenal samples possibly due to a component in
the duodenal digesta that interferes with the analysis. While absolute values are not biologically
reasonable, relative comparisons between treatments within the same experiment are useful.
Different concentrate levels in the diets were necessary to account for changes in
chemical composition of forages and maintain the same concentration of forage NDF in the
diets. Alfalfa decreased starch intake (6.82 vs. 7.16 kg/d, P = 0.05, Table 56) and increased
starch ruminal digestion rate (45.7 vs. 29.4%, P = 0.001, Table 54) compared with OG. This is
consistent with the greater rate of ruminal turnover of starch (59.3 vs. 42.1%/h, P = 0.005, Table
54) and smaller rumen pool of starch (0.52 vs. 0.78 kg, P = 0.002, Table 55) for AL than OG.
Although there was no difference in the amount of starch digested in the rumen per day, AL
increased true ruminal starch digestibility (80.4 vs. 74.7%, P = 0.03) and decreased starch flux to
the duodenum (1.55 vs. 2.02 kg/d, P = 0.04, Table 56) compared with OG because of lower
intake and faster digestion rate of starch for AL than OG. Alfalfa tended to decrease postruminal
starch digestibility (19.6 vs. 24.2%, P = 0.07) and decreased starch postruminally digested (1.33
vs. 1.74 kg/d, P = 0.05) compared with OG (Table 56). In the total tract, AL increased starch
digestibility (97.0 vs. 95.6%, P = 0.04) but tended to decrease starch digestion (6.60 vs. 6.89
kg/d, P = 0.08) compared with OG (Table 56).
The mechanism by which AL increased ruminal starch digestion is unclear. It is possible
that alfalfa promotes greater numbers or activity of starch-digesting bacteria in the rumen than
orchardgrass. Because some starch-digesting bacteria (e.g. Streptococcus bovis) also have high

254

proteolytic activity (Russell et al., 1981), resulting in deamination of amino acids and production
of ammonia, they might have contributed to the higher ruminal concentrations of isobutyrate
(1.71 vs. 1.17 mM, P < 0.001), isovalerate (2.32 vs. 1.81 mM, P = 0.01), branch chained VFA
(4.03 vs. 2.97, P = 0.001, Table 57), and ammonia (20.0 vs. 13.5 mg/dl, P < 0.001, Table 58) for
AL compared with OG. Alfalfa silage, which had higher ammonia concentration than
orchardgrass silage (Table 50), is another possible source for the increased ruminal ammonia
observed in cows fed AL.
Alfalfa increased ruminal pH (6.07 vs. 5.90, P = 0.001, Table 57) compared with OG.
Although we expected AL to have higher pH than OG because of lower starch intake, there was
no difference in ruminal digestion of starch or OM (kg/d; Tables 56 and 59, respectively), and
AL tended to increase total VFA concentration (149 vs. 146 mM, P = 0.09, Table 57).
Additionally, rumen digesta mass was smaller for AL than OG (Table 55) potentially decreasing
buffer capacity, and ruminating time per day was not different between AL and OG (Table 60)
suggesting similar buffering through saliva secretion. The pH difference observed was likely
because the buffering capacity of the rumen contents were greater for AL than OG; buffer
capacity of legumes is greater than grasses (Jasaitis et al., 1987).
Although there was no effect of treatment on ruminating time per day, forage family
affected eating time per day (Table 60). Alfalfa tended to increase eating time per day (295 vs.
271 min/d, P = 0.10) by increasing the number of meal bouts per day (10.3 vs. 8.96 meals/d, P =
0.03), with the greatest difference for cows with high pDMI (interaction P = 0.10, Figure 30A),
and tending to decrease the interval between meals (131 vs. 152 min, P = 0.09). As pDMI
increased, AL tended to increase the number of rumination bouts per day (interaction P = 0.13,
Figure 30B) and decrease the interval between rumination bouts (interaction P = 0.14, Figure

255

30C), whereas the reverse was observed for OG. Alfalfa decreased ruminating time per unit of
forage NDF consumed (78.4 vs. 84.7 min/kg forage NDF, P = 0.02, Table 60) compared with
OG. This indicated that AL provided less effective fiber than orchardgrass.
As previously mentioned, AL increased and OG decreased body weight (Table 52).
These body weight changes are consistent with numerically higher DMI for AL but similar FCM
yield as OG; however, this occurred despite the tendency for lower plasma insulin concentrations
for AL compared with OG (10.4 vs. 12.3 µIU, P = 0.10, Table 61). As pDMI increased, plasma
concentrations of glucose (P = 0.004, Figure 31A), glucagon (P = 0.02, Figure 31B), and insulin
(P = 0.02, Figure 31C) decreased independent of treatment.
Effects of pDMI on Ruminal Passage Rates
Experimental data on rates of passage from the rumen, particularly for individual feed
fractions, are scarce. Given the impact of passage on ruminal digestibility and pool sizes and
microbial growth, quantitative knowledge on rates of nutrient passage from the rumen are needed
to better understand nutrient availability in ruminants and improve nutrition models.
Furthermore, because passage rates from the rumen generally increase with increased intake,
measurements of ruminal passage rates of nutrients over a wide range of DMI are necessary. We
measured the effects of DMI on rates of passage of feed fractions from the rumen using the pool
and flux method (Robinson et al., 1987).
We expected ruminal passage rates to increase with pDMI. The passage rate of iNDF was
related to pDMI as previously discussed, but passage rates of pdNDF and starch were not related
to pDMI either independent of or dependent upon treatment (Table 54).

256

Effects of Treatment and pDMI on N Flux and Microbial Efficiency
Results for N metabolism are shown in Table 58. Alfalfa increased N intake compared
with OG (711 vs. 635 g/d, P = 0.001) with alfalfa silage as the primary source of N for AL. As
previously mentioned, AL increased ruminal ammonia concentration and tended to decrease
NANMN flux expressed as percent of N intake (16.6 vs. 22.2% of N intake, P = 0.10) compared
with OG indicating that protein was more rapidly degraded in the rumen for AL than OG.
Ammonia N, NANMN, and NAN fluxes from the rumen to the duodenum were related to pDMI,
but the response differed by treatment. As pDMI increased, AL increased flux of ammonia N
(interaction P = 0.05, Figure 32A), NANMN (interaction P = 0.09, Figure 32B), and NAN
(interaction P = 0.02, Figure 32C), whereas OG decreased ammonia N and NANMN fluxes and
maintained relatively constant NAN flux across the range of pDMI. The NANMN interaction
contributed to the treatment by pDMI interaction for NAN flux because level of intake did not
have an effect on MN flux. These results are in contrast to those in a review by Clark et al.
(1992) in which positive linear relationships between OM intake and fluxes of NAN, NANMN,
and MN were reported as OM intake increased over a very wide range (3 to 23 kg/d) compared
with this experiment.
Based on studies with continuous culture fermenters, increases in solid and liquid dilution
rates, which can be associated with increased intake, resulted in greater microbial efficiency
(Crawford et al., 1980; Shriver et al., 1986). In this experiment, microbial efficiency tended to
be related to pDMI (interaction P = 0.15), but the response varied by treatment. Alfalfa slightly
increased and OG quadratically affected microbial efficiency as pDMI increased (Figure 33).
There was no relationship between microbial efficiency and true ruminally digested OM, which
increased linearly with increasing levels of intake (Table 59) for cows consuming AL or OG (not

257

shown), indicating that factors other than availability of energy limited efficiency of MN
production and energy from OM fermentation was uncoupled from microbial growth (Russell
and Cook, 1995).
Microbial N flux from the rumen to the duodenum increased independent of treatment as
pdNDF ruminal passage rate increased (P = 0.02, Figure 34A). Taylor and Allen (2005) reported
a tendency for a positive correlation between MN flux and iNDF passage rate, but this
relationship was not detected in the present experiment. Microbial efficiency increased as
2

2

pdNDF ruminal passage rate increased for OG (P = 0.02, R = 0.42) but not AL (P = 0.15, R =
0.18; Figure 34B). This response indicates that energy from ruminal fermentation of OG was
more efficiently utilized for microbial growth as passage rates for pdNDF increased. Others
have reported that microbial efficiency was positively related to passage rates of particulate
matter from the rumen including pdNDF (Voelker and Allen, 2003) and starch (Oba and Allen,
2003b; Voelker and Allen, 2003; Taylor and Allen, 2005) in experiments evaluating
carbohydrate source, concentration, and fermentability of diets in dairy cattle. The greater
passage rates of particulate digesta likely decrease microbial lysis and turnover in the rumen
because microbial organisms flow from the rumen primarily attached to feed particles, resulting
in improved efficiency.

CONCLUSIONS
Concentration of total NDF was lower but concentrations of iNDF, ADL, and CP were
higher for alfalfa compared with orchardgrass. Forage family and its interaction with pDMI did
not affect milk yield, milk composition, or DMI. Alfalfa decreased efficiency of milk production
compared with OG because of numerically greater DMI for AL. Alfalfa increased ruminal pH,

258

digestion rates of pdNDF and starch, and passage rates of pdNDF and iNDF compared with OG,
affecting ruminal digestibility. Passage rate of iNDF was related to pDMI such that AL
increased iNDF passage rate and OG decreased it as pDMI increased. Alfalfa decreased rumen
pools of pdNDF, starch, DM, and rumen digesta wet weight and volume compared with OG.
Alfalfa decreased ruminating time per unit of forage NDF consumed compared with OG
suggesting that alfalfa provided less effective fiber than orchardgrass. Ammonia N, NANMN,
and NAN fluxes were increased by AL, but not OG, as pDMI increased. Microbial efficiency
was positively related to pdNDF passage rate for OG, but not AL. When alfalfa or orchardgrass
silage was the only source of forage in diets formulated to contain similar concentrations of
forage NDF, forage family did not adversely affect productivity.

259

APPENDIX

260

Table 49. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet
Standard
Parameter
Median
Mean
deviation
Minimum
Maximum
Parity
3
3.31
1.16
2
5
1
591
587
51
489
710
BW , kg
BCS
2.00
2.35
0.69
1.58
4.00
DIM
132
157
90
64
337
Milk, kg/d
41.4
41.5
10.8
22.6
57.1
3.5% FCM, kg/d
43.1
42.1
11.9
24.3
60.3
DMI, kg/d
26.7
25.9
3.0
19.6
29.5
1
Empty BW (ruminal digesta removed).

261

Table 50. Chemical composition, particle size distribution, and fermentation parameters of
alfalfa silage and orchardgrass silage included in the treatment diets
Item
Alfalfa
Orchardgrass
Chemical composition
DM, %
43.5
33.7
OM, % DM
91.9
90.3
NDF, % DM
42.3
58.2
1
23.0
16.1
iNDF , % DM
iNDF, % of NDF
54.5
27.7
ADF, % of DM
35.0
36.4
2
7.56
6.03
ADL , % of DM
CP, % DM
22.5
11.4
Starch, % DM
1.87
1.37
3
38.3
53.3
NDF digestibility , %
4

1

Particles size distribution
Wet sieving, % DM retained
19.0 mm
9.50 mm
4.75 mm
2.36 mm
1.18 mm
0.600 mm
0.300 mm
0.150 mm
0.075 mm
0.038 mm
5
Mean particle size , mm
Penn State Particle Separator,
% DM retained
> 19.0 mm
19.0 to 8.0 mm
< 8.0 mm
Fermentation
pH
Acetic acid, % DM
Propionic acid, % DM
Butyric acid, % DM
Lactic acid, % DM
Lactic:Acetic
Ethanol, % DM
Ammonia, mM

21.4
18.0
30.8
17.0
5.72
3.09
1.97
1.16
0.40
0.50
11.6

12.3
18.4
37.2
21.2
6.15
2.08
1.02
0.94
0.37
0.37
9.66

29.3
48.5
22.2

17.1
50.2
32.7

4.58
2.38
0.35
<0.01
5.94
2.49
0.33
4.65

4.59
0.90
0.07
0.26
6.10
6.78
<0.01
2.86

iNDF = indigestible NDF.

2

ADL = acid detergent lignin.

3

30 h in vitro NDF digestibility.
262

Table 50 (cont’d)
4

Particle size distributions of silages were measured each period (n = 2).

5

Mean particle size calculated from particle size distribution determined by wet sieving.

263

Table 51. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of forage
Preliminary
AL
OG
Ingredients, % DM
Alfalfa silage
30.0
59.9
--Orchardgrass silage
21.5
--42.7
Dry ground corn
36.2
33.6
36.6
Soybean meal (48% CP)
5.81
--11.8
®
1.82
2.50
3.39
SoyPlus
1
3.99
3.99
3.99
Vitamin mineral mix

1

Urea
Limestone
Chemical composition
DM, %
OM, % DM
NDF, % DM
% forage NDF
% NDF from forage
2
iNDF , % DM
iNDF, % of NDF
CP, % DM
Starch, % DM

0.15
0.60
51.6
92.4
29.1
24.7
84.8
3
NA
NA
17.5
33.5

-----

0.30
1.20

54.5
92.7
29.2
25.3
86.8
14.8

52.3
91.1
30.2
24.9
82.3
8.24

50.7
18.4
27.3

27.3
17.0
29.6

Vitamin mineral mix contained (DM basis) 16.5% sodium bicarbonate, 14.2% magnesium
sulfate, 7.1% salt, 5.8% dicalcium phosphate, 2.4% trace mineral premix, 0.4% vitamin A, 0.4%
vitamin D, 0.2% vitamin E, and 53.1% dry ground corn as a carrier.
2
iNDF = indigestible NDF.
3

NA = no analysis for preliminary diet.

264

Table 52. Milk production and composition, feed intake, and BW change of cows fed treatment diets containing either alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage
Treatment LSM
AL
OG
Yield, kg/d
Milk

1

FCM (3.5 %)
Milk fat
Milk protein
Milk lactose
SNF
Milk composition, %
Fat
Protein
Lactose
SNF
MUN, mg/dl
DMI, kg/d
3.5% FCM/ DMI
BW change, kg/18 d
BCS change/18 d

1

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI
3

35.1

35.2

2.3

0.92

0.14

0.06

36.7
1.33
1.08
1.65
1.99

36.5
1.31
1.05
1.65
1.98

2.1
0.07
0.04
0.12
0.15

0.84
0.72
0.29
0.97
0.93

NS
NS
0.07
NS
NS

0.02
0.007
0.007
0.16
0.16

NS
NS
NS
NS
NS
NS

3.79
3.14
4.83
5.82
13.4
24.2
1.40
6.05
-0.06

3.77
3.10
4.81
5.80
12.7
23.2
1.47
-3.78
-0.13

0.09
0.12
0.14
0.17
0.4
0.63
0.07
3.29
0.04

0.63
0.20
0.68
0.63
0.22
0.13
0.005
0.03
0.28

0.04
NS
NS
NS
NS
0.09
NS
NS
NS

0.16
NS
0.28
0.25
NS
0.02
0.06
NS
NS

NS
NS
NS
NS
NS
NS
0.006
NS
NS

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS

NS

NS
NS
NS
NS
NS

NS
NS
NS
NS
NS

NS
NS
0.10
0.11
NS
NS
NS
NS
NS

NS
NS
NS
NS
NS
NS
NS
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

265

Table 53. NDF digestion of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage
Treatment LSM
AL
OG
NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Potentially digestible NDF
Intake, kg/d
Ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Postruminal digestion
kg/d
Total tract digestion
kg/d
%
Indigestible NDF
Intake, kg/d
1
Treatment least squares means.

1

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

6.96

6.69

0.19

0.17

0.14

0.02

NS

3

NS

NS

2.47
37.7
4.26

3.70
57.2
2.89

0.15
1.7
0.26

<0.001
<0.001
<0.001

0.005
NS
NS

0.003
0.44
0.44

0.27
0.08
0.001

0.16
NS
0.84

NS
NS
0.13

-0.08

-0.42

0.21

0.15

NS

0.10

0.02

0.26

0.19

2.47
35.5

3.14
47.1

0.10
1.43

<0.001
<0.001

0.001
0.007

0.08
NS

NS
NS

NS
NS

NS
NS

3.43

4.89

0.11

<0.001

0.003

0.02

NS

NS

NS

2.47
76.2
0.84

3.70
78.2
1.07

0.15
2.6
0.13

<0.001
0.40
0.02

0.005
NS
NS

0.003
0.33
0.64

NS
0.13
0.07

0.16
NS
NS

NS
NS
NS

-0.08

-0.42

0.21

0.15

NS

0.10

0.02

0.26

0.19

2.47
72.0

3.14
64.7

0.10
2.1

<0.001
0.02

0.001
NS

0.08
0.60

NS
0.19

NS
NS

NS
NS

3.53

1.80

0.08

<0.001

0.11

0.02

0.05

NS

NS

266

Table 53 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

267

Table 54. Rumen kinetics of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage
Treatment LSM
AL
OG
Ruminal turnover rate, %/h
DM
OM
NDF

4

pdNDF
Starch
Ruminal turnover time, h
DM
OM
NDF
pdNDF
5

1

iNDF
Starch
Ruminal passage rate, %/h
pdNDF
iNDF
Starch
Ruminal digestion rate, %/h
pdNDF
Starch

1

2

P
SE

Trt

Trt x
Period
3

9.76

8.06

0.41

<0.001

10.0
4.80
9.56

8.15
4.35
6.06

0.43
0.23
0.51

<0.001
0.02
<0.001

NS
NS
NS
NS

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS

NS

NS

NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

59.3

42.1

3.7

0.005

NS

NS

NS

NS

NS

10.5
10.2
21.4
10.9
32.0

12.8
12.7
23.9
17.4
41.6

0.6
0.6
1.2
0.9
2.2

<0.001
<0.001
0.01
<0.001
<0.001

NS
NS
NS
NS
NS

NS
NS
NS
NS
NS

NS
NS
NS
NS
0.06

NS
NS
NS
NS
NS

NS
NS
NS
NS
NS

1.82

2.59

0.16

0.003

NS

NS

NS

NS

NS

2.29
3.27
13.6

1.32
2.52
12.7

0.27
0.15
2.0

<0.001
<0.001
0.60

NS
0.17
NS

NS
0.48
NS

NS
0.09
NS

NS
NS
NS

NS
NS
NS

7.27
45.7

4.74
29.4

0.43
2.5

<0.001
0.001

NS
NS

NS
NS

NS
NS

NS
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
268

Table 54 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

269

Table 55. Rumen pools of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage
Treatment LSM
AL
OG

1

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

NS

Trt x
pDMI x
pDMI
NS

NS
NS

NS
NS

3

pDMI x
pDMI

Wet weight, kg

82.7

92.4

3.6

0.008

0.14

0.16

Volume, L
Density, kg/L
Rumen pool, kg
DM
OM
NDF

98.5
0.84

108
0.86

3.6
0.01

0.01
0.44

NS
0.04

0.50
0.008

NS
0.16
0.11

10.6
9.58
6.19
1.55

12.4
11.2
6.67
3.54

0.6
0.59
0.38
0.20

0.001
0.002
0.06
<0.001

NS
NS
NS
0.14

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

4.62

3.11

0.21

<0.001

NS

0.12

NS

NS

NS

0.52

0.78

0.06

0.002

NS

NS

NS

NS

NS

pdNDF

4

5

1

iNDF
Starch

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

270

Table 56. Starch digestion of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage
Treatment LSM
AL
OG
Starch
Intake, kg/d

1

Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
% of duodenal passage
Apparent total tract digestion
kg/d
%

1

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

6.82

7.16

0.17

0.05

0.08

0.02

0.13

NS

3

NS

5.27
77.3

5.14
72.0

0.22
2.8

0.54
0.05

0.12
NS

0.06
NS

NS
NS

NS
NS

NS
NS

5.48
80.4
1.55

5.33
74.7
2.02

0.23
2.8
0.21

0.50
0.03
0.04

0.13
NS
NS

0.06
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

1.33
19.6
84.9

1.74
24.2
84.2

0.20
2.7
1.9

0.05
0.07
0.67

NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

NS
NS
NS

6.60
97.0

6.89
95.6

0.16
0.52

0.08
0.04

0.10
NS

0.02
0.93

0.14
0.14

NS
0.55

NS
0.17

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

271

Table 57. Ruminal VFA concentrations and pH of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage
as the sole source of forage
Treatment LSM
AL
OG
Total VFA, mM

1

Acetate
Propionate
Butyrate
Lactate
Isobutyrate
Valerate
Isovalerate
Branch chain VFA
Acetate:Propionate
Ruminal pH

149
90.8
31.5
18.6
0.125
1.71
2.52
2.32
4.03
2.92
6.07

1

146
91.5
29.0
17.9
<0.001
1.17
1.65
1.81
2.97
3.19
5.90

2

P
SE

Trt

Trt x
Period

3

0.09

NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS

1.2
2.2
1.0
0.110
0.06
0.14
0.11
0.15
0.16
0.05

0.59
0.05
0.20
0.30
<0.001
<0.001
0.01
0.001
0.03
0.001

3

pDMI

Trt x
pDMI

pDMI x
pDMI

NS

NS

NS

Trt x
pDMI x
pDMI
NS

NS
0.93
0.78
0.17
NS
0.68
NS
NS
0.92
NS

NS
0.15
0.58
NS
NS
NS
NS
NS
0.25
NS

NS
0.22
0.96
0.04
NS
0.18
NS
NS
0.19
NS

NS
0.03
0.16
NS
NS
NS
NS
NS
0.07
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

272

Table 58. Nitrogen metabolism of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole
source of forage
Treatment LSM
AL
OG
N intake, g/d

711

Ruminal ammonia, mg/dl
Flow to duodenum
Ammonia N, g/d
NAN
g/d
% of N intake
4

NANMN
g/d
% of N intake
% of duodenal NAN
Microbial N
g/d
% of duodenal NAN
5

635

2

P
SE

Trt

pDMI

Trt x
pDMI

pDMI x
pDMI

0.001

18

Trt x
Period
0.16

0.02

0.17

3

Trt x
pDMI x
pDMI
NS

20.0

13.5

0.7

<0.001

NS

0.18

NS

NS
NS

18.4

14.9

1.3

0.03

NS

0.24

0.05

NS

NS

556
79.1

591
89.5

45
3.6

0.26
0.01

NS
NS

0.55
NS

0.02
NS

0.85
NS

0.14
NS

122
16.6
21.1

139
22.2
24.5

21
2.8
3.1

0.41
0.10
0.32

NS
NS
NS

0.17
0.30
0.18

0.09
0.15
0.19

0.08
0.04
0.03

NS
NS
NS

433
78.9
30.7

454
75.5
34.7

41
3
3.0

0.54
0.32
0.22

NS
NS
0.16

0.96
0.18
0.21

0.25
0.19
0.20

0.26
0.03
0.14

0.16
NS
0.15

361
56.4
60.4

40
5.2
2.8

0.22
0.03
0.08

NS
NS
NS

0.76
0.31
0.10

0.008
0.02
0.002

0.53
0.47
0.31

0.04
0.04
0.004

396
63.0

17
1.2

0.02
0.18

0.02
0.02

0.12
0.02

0.02
0.009

0.84
0.15

0.06
0.01

g/kg TRDOM
NAN apparent postruminal digestion
g/d
319
% of N intake
44.0
% of duodenal passage
55.0
N apparent total tract digestion
g/d
455
%
65.6

1

1

Treatment least squares means.

273

NS

Table 58 (cont’d)
2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

NANMN = nonammonia, nonmicrobial nitrogen.

5

TRDOM = true ruminally digested OM.

274

Table 59. DM and OM digestion of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole
source of forage
Treatment LSM
AL
OG
DM
Intake, kg/d

1

Apparent total tract digestion
kg/d
%
OM
Intake, kg/d
Apparent ruminal digestion
kg/d
%
True ruminal digestion
kg/d
%
Passage to duodenum, kg/d
Apparent postruminal digestion
kg/d
% of intake
Apparent total tract digestion
kg/d
%

1

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI
3

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS

NS

24.2

23.2

0.6

0.13

0.09

0.02

NS

15.7
64.5

15.0
66.8

0.4
1.1

0.20
0.18

0.003
0.005

0.03
0.05

0.13
0.05

NS
0.16

NS
0.07

22.5

21.2

0.6

0.04

0.08

0.02

NS

NS

NS

9.49
44.9

9.19
42.6

0.6
2.8

0.64
0.44

0.006
0.09

0.01
0.27

NS
0.06

0.09
0.28

NS
0.12

14.4
64.1
12.1

13.8
65.1
12.0

0.5
1.5
0.8

0.35
0.55
0.85

0.01
0.15
NS

0.009
0.43
0.51

NS
0.09
0.01

NS
NS
0.74

NS
NS
0.06

4.56
20.1

5.33
25.0

0.60
2.6

0.20
0.12

NS
NS

0.35
0.10

0.005
0.01

0.19
0.14

0.01
0.03

14.8
65.5

14.0
68.2

0.4
1.1

0.12
0.10

0.004
0.009

0.03
0.06

0.12
0.05

NS
0.16

NS
0.08

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).

275

Table 59 (cont’d)
3

NS = not significant (P > 0.20); term removed from statistical model.

276

Table 60. Chewing activity of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage
Treatment LSM
AL
OG
Meals
Bouts/d
Length, min/bout
Interval, min
Meal size, kg
DM
OM
NDF
4
pdNDF
5

iNDF
Starch
Eating time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake
Rumination
Bouts/d
Length, min/bout
Interval, min
Ruminating time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake

10.3
29.1
131

1

8.96
30.7
152

2

P
SE

Trt

Trt x
Period

0.49

0.03

1.6
10

0.50
0.09

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

NS
NS
NS

0.11

0.10

NS

NS

0.02
NS

NS
NS

NS
NS

NS
NS

3

2.44
2.26
0.70
0.35

2.63
2.40
0.76
0.55

0.13
0.12
0.04
0.02

0.35
0.46
0.34
<0.001

0.03
0.03
0.03
0.003

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

0.35

0.20

0.02

<0.001

NS

NS

NS

NS

NS

0.69

0.81

0.04

0.05

0.04

NS

NS

NS

NS

295
12.4
43.0
48.7

271
11.9
41.3
47.8

14
0.6
2.1
2.5

0.10
0.39
0.40
0.69

0.20
0.05
0.06
0.06

NS
0.04
0.04
0.04

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

13.8
33.3
66.1

13.6
34.3
60.7

0.8
1.5
3.9

0.71
0.33
0.25

NS
NS
NS

0.51
NS
0.60

0.13
NS
0.14

0.14
NS
0.11

NS
NS
NS

477
19.8
69.2
78.4

484
21.0
73.3
84.7

14
0.8
2.7
3.0

0.67
0.07
0.09
0.02

NS
NS
NS
NS

NS
0.10
0.08
0.09

NS
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

277

Table 60 (cont’d)

1

Total chewing time
Min/d
Min/kg DMI
Min/kg NDF intake
Min/kg forage NDF intake

771
32.3
112
127

755
33.0
115
133

21
1.1
4
4

0.51
0.44
0.47
0.15

NS
0.05
0.08
0.07

0.54
0.02
0.02
0.02

0.19
NS
NS
NS

NS
NS
NS
NS

NS
NS
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.
4

pdNDF = potentially digestible NDF.

5

iNDF = indigestible NDF.

278

Table 61. Plasma metabolites of cows fed treatment diets containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source
of forage
Treatment LSM
AL
OG
Glucose, mg/dl

1

Glucagon, pg/ml
Insulin, uIU/ml

1

2

P
SE

Trt

58.2

60.5

0.6

0.004

144
10.4

164
12.3

3
1.0

<0.001
0.10

Trt x
Period
3

NS
NS
0.10

pDMI

Trt x
pDMI

pDMI x
pDMI

0.004

NS

NS

Trt x
pDMI x
pDMI
NS

0.02
0.02

NS
NS

NS
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
NS = not significant (P > 0.20); term removed from statistical model.

279

Figure 27. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for DMI (interaction not significant). The preliminary DMI on
the x-axis are the mean DMI of individual cows during the final 4 d of the preliminary period
when all cows were fed a common diet.

280

Figure 28

281

Figure 28 (cont’d)

Figure 28. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for indigestible NDF (iNDF) A) intake (P = 0.05), B) ruminal
passage rate (P = 0.09), and C) ruminal turnover time (P = 0.06). The preliminary DMI on the xaxis are the mean DMI of individual cows during the final 4 d of the preliminary period when all
cows were fed a common diet.

282

Figure 29

283

Figure 29 (cont’d)

Figure 29. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for A) ruminal digestibility of potentially digestible NDF
(pdNDF; P = 0.13), B) ruminal digestibility of NDF (P = 0.08), C) pdNDF flux to duodenum (P
= 0.07), and D) NDF flux to duodenum (P = 0.001). The preliminary DMI on the x-axis are the
mean DMI of individual cows during the final 4 d of the preliminary period when all cows were
fed a common diet.

284

Figure 30

285

Figure 30 (cont’d)

Figure 30. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for A) meal bouts (P = 0.10), B) rumination bouts (P = 0.13),
and C) rumination interval (P = 0.14). The preliminary DMI on the x-axis are the mean DMI of
individual cows during the final 4 d of the preliminary period when all cows were fed a common
diet.

286

Figure 31

287

Figure 31 (cont’d)

Figure 31. Relationship of alfalfa (open circles) and orchardgrass (closed circles) with
preliminary DMI for concentrations of plasma A) glucose (P = 0.004), B) glucagon (P = 0.02),
and C) insulin (P = 0.02). The preliminary DMI on the x-axis are the mean DMI of individual
cows during the final 4 d of the preliminary period when all cows were fed a common diet.

288

Figure 32

289

Figure 32 (cont’d)

Figure 32. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for A) ammonia N (P = 0.05), B) nonammonia, nonmicrobial N
(NANMN; P = 0.09), and C) NAN (P = 0.02) flux to the duodenum. The preliminary DMI on
the x-axis are the mean DMI of individual cows during the final 4 d of the preliminary period
when all cows were fed a common diet.

290

Figure 33. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for microbial efficiency (P = 0.15) expressed as gram of
microbial N produced per kilogram of true ruminally digested OM (TRDOM). The preliminary
DMI on the x-axis are the mean DMI of individual cows during the final 4 d of the preliminary
period when all cows were fed a common diet.

291

Figure 34. A) Relationship between potentially digestible NDF (pdNDF) ruminal passage rate
2
and microbial N flux to duodenum (P = 0.02, R = 0.21). B) Relationship between pdNDF
2

ruminal passage rate and microbial efficiency for alfalfa (P = 0.15, R = 0.18) and orchardgrass
(microbial efficiency, g of microbial N/kg of true ruminally digested OM (TRDOM) = 19.8 +
2
7.60 x pdNDF ruminal passage rate, %/h; P = 0.02, R = 0.42). Open circles and dashed line
denote alfalfa, and closed circles and solid line denote orchardgrass.

292

REFERENCES

293

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cows.
J. Dairy Sci. 83:1598-1624.
Broderick, G. A., and J. H. Kang. 1980. Automated simultaneous determination of ammonia and
total amino acids in rumen fluid and in vitro media. J. Dairy Sci. 63:64-75.
Clark, J. H., T. H. Klusmeyer, and M. R. Cameron. 1992. Microbial protein synthesis and flows
of nitrogen fractions to the duodenum of dairy cows. J. Dairy Sci. 75:2304-2323.
Cochran, R. C., D. C. Adams, J. D. Wallace, and M. L. Galyean. 1986. Predicting digestibility of
different diets with internal markers: Evaluation of four potential markers. J. Anim. Sci.
63:1476-1483.
Crawford, R. J., Jr., W. H. Hoover, and L. L. Junkins. 1980. Effects of solids and liquid flows on
fermentation in continuous cultures. II. Nitrogen partitioning and efficiency of microbial
synthesis. J. Anim. Sci. 51:986-995.
Dado, R. G, and M. S. Allen. 1994. Variation in and relationships among feeding, chewing,
and drinking variables for lactating dairy cows. J. Dairy Sci. 77:132-144.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.
Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More powerful peroxide
Kjeldahl digestion method. J. AOAC 70:783-787.
Huhtanen, P., and S. Jaakkola. 1993. The effects of forage preservation method and proportion of
concentrate on digestion of cell wall carbohydrates and rumen digesta pool size in cattle.
Grass Forage Sci. 48:155-165.
Jasaitis, D. K., J. E. Wohlt, and J. L. Evans. 1987. Influence of feed ion content on buffering
capacity of ruminant feedstuffs in vitro. J. Dairy Sci. 70:1391-1403.
Joy, M. T., E. J. DePeters, J. G. Fadel, and R. A. Zinn. 1997. Effects of corn processing on the
site and extent of digestion in lactating cows. J. Dairy Sci. 80:2087-2097.
Karkalas, J. 1985. An improved enzymatic method for the determination of native and modified
starch. J. Sci. Food Agric. 36:1019-1027.
Lammers, B. P., D. R. Buckmaster, and A. J. Heinrichs. 1996. A simplified method for the

294

analysis of particle sizes of forage and total mixed rations. J. Dairy Sci. 79:922-928.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:12171240.
Oba, M., and M. S. Allen. 1999. Evaluation of the importance of the digestibility of neutral
detergent fiber from forage: effects on dry matter intake and milk yield of dairy cows. J.
Dairy Sci. 82:589-596.
Oba, M., and M. S. Allen. 2003a. Effects of corn grain conservation method on feeding behavior
and productivity of lactating dairy cows at two dietary starch concentrations. J. Dairy Sci.
86:174–183.
Oba, M., and M. S. Allen. 2003b. Effects of diet fermentability on efficiency of microbial
nitrogen production in lactating dairy cows. J. Dairy Sci. 86:195-207.
Poore, M. H., J. A. Moore, T. P. Eck, R. S. Swingle, and C. B. Theurer. 1993. Effect of fiber
source and ruminal starch degradability on site and extent of digestion in dairy cows. J.
Dairy Sci. 76:2244-2253.
Robinson, P. J., S. Tamminga, and A. M. van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci.17:37-62.
Russell, J. B., W. G. Bottje, and M. A. Cotta. 1981. Degradation of protein by mixed cultures of
rumen bacteria: Identification of Streptococcus bovis as an actively proteolytic rumen
bacterium. J. Anim. Sci. 53:242-252.
Russell, J. B., and G. M. Cook. 1995. Energetics of bacterial growth: balance of anabolic and
catabolic reactions. Microbiol. Rev. 59:48-63.
Shriver, B. J., W. H. Hoover, J. P. Sargent, R. J. Crawford, and W. V. Thayne. 1986.
Fermentation of a high concentrate diet as affected by ruminal pH and digesta flow. J.
Dairy Sci. 69:413-419.
Steinshamn, H. 2010. Effect of forage legumes on feed intake, milk production, and milk quality
- a review. Animal Science Papers and Reports 28:195-206.
Stensig, T., and P. H. Robinson. 1997. Digestion and passage kinetics of forage fiber in dairy
cows as affected by fiber-free concentrate in the diet. J. Dairy Sci. 80:1339-1352.
Taylor, C. C., and M. S. Allen. 2005. Corn grain endosperm type and brown midrib 3 corn
silage: ruminal fermentation and N partitioning in lactating cows. J. Dairy Sci. 88:14341442.

295

Van Soest, P. J. 1982. Nutritional Ecology of the Ruminant. O & B Books, Inc. Corvallis, OR.
Voelker, J. A., and M. S. Allen. 2003. Pelleted beet pulp substituted for high-moisture corn: 3.
Effects of ruminal fermentation, pH, and microbial protein efficiency in lactating dairy
cows. J. Dairy Sci. 86:3562-3570.
Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. J. Dairy Sci. 65:495–501.
Wilson, J. R., and P. M. Kennedy. 1996. Plant and animal constraints to voluntary feed intake
associated with fibre characteristics and particle breakdown and passage in ruminants.
Aust. J. Agric. Res. 47:199-225.
Zinn, R. A., and F. N. Owens. 1986. A rapid procedure for purine measurement and its use for
estimating net ruminal protein synthesis. Can. J. Anim. Sci. 66:157-166.

296

CHAPTER 7
Rates of Particle Size Reduction and Passage are Faster for Legume Compared to CoolSeason Grass Resulting in Lower Rumen Fill and Less Effective Fiber

1

ABSTRACT
Effects of forage family on rates of particle size reduction in, and passage from, the
rumen and the relationship of these effects with preliminary DMI (pDMI) were evaluated using
13 ruminally and duodenally cannulated Holstein cows in a crossover design with a 14-d
preliminary period and two 18-d treatment periods. During the preliminary period, pDMI of
individual cows ranged from 19.6 to 29.5 kg/d (mean = 25.9 kg/d) and 3.5% fat-corrected milk
yield ranged from 24.3 to 60.3 kg/d (mean = 42.1 kg/d). Experimental treatments were diets
containing either a) alfalfa silage (AL) or b) orchardgrass silage (OG) as the sole forage. Silages
were chopped to 10 mm theoretical length of cut and contained 42.3 and 58.2% neutral detergent
fiber (NDF) for alfalfa and orchardgrass, respectively. Both diets contained ~25% forage NDF
and ~30% total NDF. Feed, orts, rumen, and duodenal samples were wet sieved to fractionate
particles above (large, L) and below (small, S) 2.36 mm. Indigestible NDF (iNDF) was used as a
flow marker. Preliminary DMI, an index of nutrient demand, was determined during the last 4 d
of the preliminary period when cows were fed a common diet and used as a covariate. Main
effects of forage family and their interaction with pDMI were tested by ANOVA. Cows
consumed ~75% NDF L and 25% NDF S for both treatments, but AL consumed more iNDF and
less potentially digestible NDF (pdNDF) than OG. Alfalfa increased the rate of reduction (L to
S) compared with OG despite less rumination per unit of forage NDF for AL than OG suggesting
1

Submitted for publication in the Journal of Dairy Science on October 5, 2011.

297

alfalfa NDF was more fragile than orchardgrass NDF. Over 55% of particles in the rumen were
below 2.36 mm (S) for AL and OG indicating that particle size was not a limiting constraint to
passage. Passage rates (kp) of iNDF L and pdNDF L were similar for AL and OG, but AL
increased kp of pdNDF L and OG decreased it as pDMI increased. Alfalfa increased kp of iNDF
S and pdNDF S compared with OG resulting in lower rumen fill for AL than OG. The kp of
iNDF S and pdNDF S were similar within forage family suggesting buoyancy was not limiting
passage. Orchardgrass increased rumen pool size of NDF L compared with AL, which likely
retained NDF S contributing to the slower kp of iNDF S and pdNDF S observed for OG. Particle
size reduction was a prerequisite to ruminal passage but not a constraint. Selective retention of S
particles was less for alfalfa than orchardgrass resulting in lower rumen fill and less effective
fiber.

INTRODUCTION
Passage of digesta from the rumen is a complicated, dynamic process and is inversely
related to the extent of digestion within the rumen. It involves the selective retention of
undigested fiber, which allows ruminants to increase ruminal fiber digestion but extended
ruminal retention times of the retained fiber can reduce DMI from ruminal distention. Most
models used to predict feed intake and digestion do not include selective retention because it is
not fully understood and difficult to measure. This implies that all particles have equal
probability of escape from the rumen. However, escapable (i.e., small) and nonescapable (i.e.,
large) particles have different passage rates (Allen and Mertens, 1988; Voelker Linton and Allen,
2008). Additionally, heterogeneous feed fractions such as NDF, which includes indigestible

298

NDF (iNDF) and potentially digestible NDF (pdNDF), have different ruminal kinetics (Stensig
and Robinson, 1997).
Several factors contribute to the decreased probability of escape of digesta from the
rumen. Particle size influences passage from the rumen but is not always a constraint because a
large proportion of the particles retained in the rumen are smaller than the maximum particle size
in the feces (Allen, 1996). Furthermore, particle density and buoyancy (Jung and Allen, 1995)
and sequestration of small particles within the fibrous rumen mat (Sutherland, 1988) affect
passage of particles from the rumen. Passage rate of fibrous particles depends on the reduction
of particle size and the increase in particle specific gravity to permit particles to escape the
rumen mat, sink to the ventral rumen, and exit the rumen via the reticular-omasal orifice
(Sutherland, 1988).
Digesta passage from the rumen is impacted by numerous feed and animal factors.
Legumes and grasses have different ruminal kinetic parameters (Voelker Linton and Allen, 2008;
Bayat et al., 2010; Krizsan et al. 2010) and increases in DMI result in a decrease in the
percentage of small particles in the rumen (Okine and Mathison, 1991). Therefore, the effects of
forage family and level of feed intake on ruminal passage rates are of interest in this study.
Alfalfa (Medicago sativa) and orchardgrass (Dactylis glomerata L.) were selected as a
representative legume and cool-season grass, respectively, and the use of preliminary DMI
(pDMI), an index of nutrient demand, allowed the evaluation of treatments on animal responses
in relation to level of intake and provided an indicator to test effects of intake level independent
of treatments. Additionally, ruminal passage rates of individual digesta fractions, instead of
entire feeds, were measured using ruminally and duodenally cannulated cows and the pool and
flux method (Robinson et al., 1987).

299

We hypothesized that rates of particle size reduction in, and particle passage from, the
rumen are faster for legumes than grasses and the rate of particle passage from the rumen
increases for legumes and grasses as DMI increases. The objective of this experiment was to
evaluate the relationships between voluntary DMI and effects of forage family on rates of
particle size reduction in, and particle passage from, the rumen. This was accomplished using a
model that fractionated the pool of NDF in the rumen into iNDF and pdNDF as well as large and
small particle size pools. The rate of reduction for iNDF from large particles to small particles
and individual rates of passage for each of the 4 fractions were calculated.

MATERIALS AND METHODS
This article is the second of a set of two from one experiment that evaluated the effects of
forage family and its interaction with level of feed intake (nutrient demand). This article focuses
on rates of particle size breakdown in, and particle passage from, the rumen. The companion
article discusses the effect of pDMI on responses to treatment for production, rumen parameters,
digestion kinetics, and chewing activity.
Cows and Treatments
Experimental procedures were approved by the Institutional Animal Care and Use
Committee at Michigan State University. Thirteen multiparous Holstein cows from the
Michigan State University Dairy Cattle Teaching and Research Center were assigned randomly
to treatment sequence in a crossover design experiment with one 14-d preliminary period and
two 18-d experimental periods. The first 10 d of each period were allowed for diet adaptation
and samples were collected during the final 4 d of the preliminary period and 8 d of each
experimental period. Cows were 157±90 (mean±SD) DIM at the end of the preliminary period

300

and were selected to provide a wide range and uniform distribution of pDMI and milk yield.
During the final 4 d of the 14-d preliminary period, the average pDMI among cows ranged from
19.6 to 29.5 kg/d (mean = 25.9 kg/d) and 3.5% FCM yield ranged from 24.3 to 60.3 kg/d (mean
= 42.1 kg/d; Table 62). Prior to calving, cows were cannulated ruminally (Bar Diamond, Inc.,
Parma, ID) and duodenally with a gutter-type T cannula placed approximately 10 cm distal to the
pylorus (Joy et al., 1997). Surgery was performed at the Department of Large Animal Clinical
Science, College of Veterinary Medicine, Michigan State University.
Experimental treatments were diets containing either a) alfalfa silage (AL) or b)
orchardgrass silage (OG) as the sole forage. Alfalfa (Pioneer 54H91, Pioneer Hi-Bred, Johnston,
IA) and orchardgrass (Baridana cultivar, Barenbrug USA, Tangent, OR) forages were produced
at the campus farm at Michigan State University (East Lansing), chopped to 10 mm theoretical
length of cut, and ensiled in Ag-Bags (Ag-Bag Systems Inc., St. Nazianz, WI). During the
sample collection periods, alfalfa and orchardgrass contained 42.3 and 58.2% NDF and 22.5 and
11.4% CP, respectively (DM basis; Table 63). Diets AL and OG were formulated to contain
25% forage NDF, 30% total NDF, and 18% CP. We acknowledge these treatments affect dietary
starch concentration but maintaining similar forage and total NDF concentrations for both
treatments was of primary interest. The diet fed during the preliminary period was formulated so
that alfalfa and orchardgrass each contributed 50% of forage NDF. Diets also contained dry
®

ground corn, SoyPlus (West Central Soy, Ralston, IA), and vitamin-mineral premix (Table 64);
soybean meal (48% CP), urea, and limestone were used to compensate for lower CP and Ca
concentrations in orchardgrass silage than in alfalfa silage.
Data and Sample Collection
Throughout the experiment, cows were housed in tie-stalls, milked in a parlor twice daily

301

(0400 and 1430 h), and fed diets as total mixed rations once daily (1130 h) at 110% of expected
intake. The amount of feed offered and refused (orts) was weighed daily for each cow. Forage
samples were collected twice weekly and analyzed to adjust diets to account for DM, NDF, and
CP fluctuation. Samples of all dietary ingredients and TMR (0.5 kg) and orts (12.5%) were
collected daily from d 11 to 14 during the preliminary period and d 11 to 15 during each
experimental period. Samples were frozen immediately after collection at !20°C and combined
to one composite sample per period prior to analysis.
Duodenal samples (900 mL) were collected every 15 h from d 11 to 15 of each
experimental period so that 8 samples were taken for each cow in each period, representing every
3 h of a 24-h period to account for diurnal variation. Ruminal contents were evacuated manually
through the ruminal cannula 4 h after feeding at the beginning of d 17 (1530 h) and 2 h before
feeding at the end of d 18 (0930 h) for each experimental period. Total rumen content mass and
volume were determined. To ensure accurate sampling, every tenth handful of digesta (10%)
was separated for a subsample throughout evacuation. This subsample was squeezed into
primarily solid and liquid phases. Both phases were weighed and sampled (350 mL) for
determination of nutrient pool size. All samples were stored at -20°C.
Sample Analysis and Calculations
Analyses of diet ingredients for chemical composition of forages and treatment diets were
described in detail by Kammes and Allen (submitted). Forage samples were combined to one
composite sample per forage per period. Particle size distribution was determined using the
Penn State Particle Separator containing 2 sieves (19 and 8 mm) and a pan (Lammers et al.,
1996). In addition, samples were wet sieved manually and sequentially through screens with the
following aperture sizes: 19.0, 9.50, 4.75, 2.36, 1.18, 0.600, 0.300, 0.150, 0.075, and 0.038 mm

302

(W. S. Tyler Inc., Gastonia, NC). The fraction of DM retained on the screens from wet sieving
was used to calculate mean particle size.
Rates of particle size reduction in, and particle passage from, the rumen were determined
by using iNDF as a marker (Figure 35). Quadruplicate 25-g TMR and duplicate 25-g orts
samples were sieved. Thawed subsamples of ruminal solid and liquid phases from each of 2
rumen evacuations per period were recombined into duplicate 30-g samples based on the original
ratio of solid and liquid phases. Duodenal samples were thawed and combined (8 per cow per
period), separated into liquid and solid phases, and recombined in duplicate 350-g samples based
on the original ratio of solid and liquid phases. Duodenal samples were sieved first to determine
threshold size for passage by individually wet sieving samples sequentially through 4.75, 2.36,
and 0.038 mm screens. Because the 2.36 mm screen was the screen with the largest aperture size
that retained duodenal digesta for all cows, 2.36 mm was selected as the threshold for passage.
Particles retained on the 2.36 and 4.75 mm screens were combined and the resulting fractions
were designated as " 2.36 (large particles (L) less likely to escape the rumen) and < 2.36 (small
particles (S) more likely to escape the rumen). Orts, TMR, and rumen samples (kept separate for
the two sampling times) were wet sieved sequentially through the 2.36 and 0.038 mm screens.
Particles retained on each screen were removed, dried at 55°C and then weighed. Materials
retained on each screen from replicate sievings were combined. The 2 fractions were ground
with a Wiley mill (1-mm screen; Arthur H. Thomas, Philadelphia, PA) and analyzed for DM,
iNDF, and pdNDF. Concentrations of NDF were determined according to Mertens (2002).
Indigestible NDF was estimated as NDF residue after 240 h in vitro fermentation (Goering and
Van Soest, 1970); flasks were reinoculated at 120 h to insure a viable microbial population.
Ruminal fluid for the in vitro incubations was collected from a nonpregnant dry cow fed dry hay

303

only. Fraction of pdNDF was calculated by difference (1.00 – iNDF). Concentrations of all
nutrients except DM were expressed as percentages of DM determined by drying at 105°C in
forced-air oven for more than 8 h.
Total intakes, ruminal pool sizes, and duodenal fluxes were discussed in detail by
Kammes and Allen (submitted). In brief, nutrient intakes were calculated using the composition
of feed offered and refused. Ruminal pool sizes (kg) of NDF, iNDF, and pdNDF were
determined by multiplying the concentration of each component in rumen samples by the
ruminal digesta DM mass (kg). Duodenal fluxes (kg/d) of NDF and pdNDF were determined
using iNDF as a flow marker; iNDF intake (kg/d) was multiplied by the ratio between the
component and iNDF in duodenal digesta.
Reduction rate of iNDF (rate of transfer of iNDF from the L pool to the S pool), passage
rates of iNDF L and iNDF S, and relative size threshold for escape from the rumen were
calculated as follows:
Reduction rate (kr) from L pool to S pool:
iNDF kr L = [iNDFIntake L (kg/d) - iNDFDuodenal flux L (kg/d)] / iNDFRumen pool L (kg),
Passage rate (kp):
iNDF kp L = iNDFDuodenal flux L (kg/d) / iNDFRumen pool L (kg), and
iNDF kp S = iNDFDuodenal flux S (kg/d) / iNDFRumen pool S (kg),
Relative size threshold:
iNDFDuodenal flux L (kg/d) / iNDFDuodenal flux total (kg/d).

304

Passage rates and relative size threshold were calculated similarly for pdNDF. Rate of
particle size reduction was calculated for iNDF only because it can leave the pool only by
breakdown or by passage, whereas pdNDF can leave the pool by digestion as well as by particle
size reduction and passage.
Statistical Analysis
All data were analyzed by using the fit model procedure of JMP (Version 8, SAS
Institute, Cary, NC). To determine differences between treatments and evaluate interactions of
treatment with DMI, where pDMI (calculated as the mean of DMI values on d 11 to 14 of the 14d preliminary period) was used as the covariate for treatment responses, data were analyzed
2

according to the following model: Yijk = µ + Ci + Pj + Tk + PTjk + pDMI + TkpDMI + pDMI +
2

TkpDMI + eijk where µ is the overall mean, Ci is the random effect of cow (i = 1 to 13), Pj is
the fixed effect of period (j = 1 to 2), Tk is the fixed effect of treatment (k = 1 to 2), PTjk is the
interaction of period and treatment, pDMI is the linear effect of pDMI, TkpDMI is the interaction
2

2

of treatment and pDMI (linear), pDMI is the quadratic effect of pDMI, TkpDMI is the
interaction of treatment and pDMI (quadratic), and eijk is the residual error. Statistical
2

significance for TkpDMI and TkpDMI indicated treatment differences were related to pDMI.
Covariate and interaction terms were removed stepwise from the model if P > 0.20. Treatment
effects and their interaction (linear and quadratic relationships) were declared significant at P #
0.05 and P # 0.10, respectively. Tendencies for treatment effects and their interactions were
declared at P # 0.10 and P # 0.15, respectively.

305

RESULTS AND DISCUSSION
Results of forage family and its interaction with pDMI did not affect DMI, milk yield, or
milk composition (Kammes and Allen, submitted). The proportions of NDF L and NDF S intake
as a percent of the total NDF intake were essentially the same as cows consumed approximately
75% of total NDF as L and 25% as S for both treatments (Figure 36) despite the significant
differences detected between AL and OG (P < 0.001) and was related to pDMI. As pDMI
increased, AL decreased the proportion of NDF L intake (interaction P = 0.03, Figure 37A) and
increased the proportion of NDF S intake (interaction P = 0.03, Figure 37B), whereas both
remained relatively constant for OG across the entire range of pDMI. This indicated that cows
consuming AL were likely sorting the feed offered and selecting against long NDF particles in
favor of short NDF particles as nutrient demand increased. This might have been allowed by the
greater mean particle size and percent of particles > 19 mm (top sieve of Penn State Particle
Separator) for alfalfa silage compared with orchardgrass silage (Table 63). Alfalfa increased the
proportion of NDF consumed as iNDF L and iNDF S and decreased the proportion of NDF
consumed as pdNDF L and pdNDF S compared with OG (P < 0.001, Table 65) because of the
differences in chemical composition of forage families as alfalfa had a higher concentration of
iNDF and lower concentration of pdNDF than orchardgrass (Table 63).
Alfalfa increased the rate of particle size reduction of iNDF from L to S compared with
OG (7.16 vs. 4.67%/h, P < 0.001, Table 66). This is consistent with the greater resistance of
grass cell walls to particle breakdown than for alfalfa cell walls because of chemical and
structural differences (Wilson and Hatfield, 1997) and contributed to a greater proportion of total
NDF in the rumen as S for AL compared with OG (P < 0.001, Figure 36). The faster reduction
rate for AL occurred despite less rumination time spent per unit of forage NDF intake for AL

306

than OG (78.4 vs. 84.7 min/kg forage NDF, P = 0.02, Kammes and Allen, submitted). These
responses indicated that alfalfa NDF was more fragile than orchardgrass NDF. However, 63.1
and 55.8% of NDF particles in the rumen for AL and OG, respectively, were below the threshold
size for passage (Table 65), which suggested that particle size was not a limiting constraint to
passage for either treatment.
Passage rates of L from the rumen were less than 1.2%/h for iNDF and pdNDF for both
treatments (Table 66). Forage family and its interaction with pDMI did not affect ruminal
passage rate of iNDF L, but response of ruminal passage rate of pdNDF L was related to pDMI
(Table 66). As pDMI increased, AL increased the passage rate of pdNDF L and OG decreased it
(interaction P = 0.10, Figure 38). An increase in passage rate with greater intakes was expected,
but the reason for the reduction in passage rate of pdNDF L with increased level of intake for OG
is not known. Responses of passage rates of S were not related to pDMI (Table 66), but AL
increased the rates of ruminal passage of iNDF S (3.85 vs. 2.66 %/h, P < 0.001) and pdNDF S
(3.80 vs. 2.50 %/h, P = 0.002) compared with OG. Despite the faster passage rate of iNDF S for
AL, ~50% of the particles in the rumen for AL were iNDF S compared with ~30% for OG (P =
0.001, Table 65), which is because of the greater intake of iNDF for AL than OG. In contrast,
~13% of the particles in the rumen for AL were pdNDF S compared with ~26% OG (P = 0.001,
Table 65) because of lower pdNDF intake and faster pdNDF digestion rate for AL than OG
(Kammes and Allen, submitted). These responses contributed to the higher rumen pool size of
iNDF and lower rumen pool sizes of pdNDF, NDF, DM, and digesta wet weight and volume for
AL compared with OG, which ultimately resulted in lower rumen fill for AL than OG (Kammes
and Allen, submitted).
Passage rates for pdNDF S and iNDF S were similar within forage family (Figure 39). If

307

buoyancy was a factor limiting passage, we would expect pdNDF to be selectively retained in the
rumen and have a slower rate of passage than iNDF. Although both iNDF and pdNDF fractions
are contained in the same particle, the probability of particles to pass should increase as the
particle increases in iNDF because there is less gas from fermentation of pdNDF associated with
the particle thereby decreasing buoyancy. Although overall passage rate (not fractionated into L
and S) of iNDF was greater than pdNDF for AL and OG (Kammes and Allen, submitted), the
similar passage rates for iNDF S and pdNDF S within forage species suggested buoyancy was
likely not a constraint to the passage of S particles from the rumen in the current experiment.
Few studies in the literature report rates of passage of various size particles for iNDF and
pdNDF fractions. In a study using the same method as the one used in this experiment, passage
rate of iNDF was greater than pdNDF for particles <2.36 mm for cows fed low fiber or high fiber
diets including alfalfa silage and corn silage as forage sources (Voelker Linton and Allen, 2007).
Another study reported faster passage rates for iNDF than pdNDF for particles <2.50 mm for
cows fed a mixed timothy and meadow fescue grass silage, but this study used rumen
evacuations and fecal output, rather than duodenal flow, to calculate passage rates (Rinne et al.,
2002). These results are in agreement with the above stated logic regarding buoyancy; however,
we did not obtain similar findings in this experiment.
Alfalfa decreased rumen pool size of NDF L compared with OG (2.30 vs. 2.95 kg, P =
0.001, Table 65). The greater pools of L fibrous particles for OG likely function to entrap S
particles and prevent their sedimentation, which reduces their probability of escape. Results on
the distribution of particles within the rumen of cows (Evans et al., 1973) and sheep (Sutherland,
1982) have indicated that the ruminal mat functions very effectively as a retaining mechanism
and escape from the mat has been identified as a rate-limiting component of passage of forage

308

particles (Poppi et al., 2001). In addition to the size of particles, shape of particles within the
rumen mat is probably important. The cubodial-shaped fragments of legumes usually pass from
the rumen faster than grass particles, which are elongated and needle-like (Buxton et al., 1996).
The intertwining of long, thin grass particles within the rumen mat might be more efficient at
retaining S particles than those of legumes. These factors likely contributed to the slower
passage rate of pdNDF S and iNDF S observed for OG.
Rumen pools (kg) of NDF S were similar for AL and OG (P = 0.17, Table 65). Despite
the faster passage and digestion rates of S particles for AL compared with OG as previously
discussed, the pool of NDF S was not smaller for AL than OG because the rate of reduction was
also faster such that S particles passed from the rumen or digested in the rumen were replaced by
the reduction of L particles into S particles. The composition of the rumen mat and its effect on
particle passage is likely a balance between passage, digestion, and reduction rates. If the rates
of passage and digestion are slower than the rate of reduction, rumen pools of L particles will
decrease and S particles will increase until more L particles are consumed. The accumulation of
S particles at the expense of L particles will decrease the ability of the rumen mat to retain S
particles.
Alfalfa tended to decrease or decreased the proportion of iNDF L (15 vs. 19% iNDF L
duodenal flux/total iNDF duodenal flux, P = 0.09) and pdNDF L (17 vs. 23% pdNDF L duodenal
flux/total pdNDF duodenal flux, P = 0.02) particles that escaped the rumen compared with OG
(Table 66). This is consistent with the lower proportion of total NDF in duodenal flow as L
particles for AL compared with OG (P = 0.02, Figure 36). Alfalfa increased the proportion of
iNDF S particles that escaped the rumen compared with OG (67.1 vs. 44.3% of total NDF, P <
0.001), and this fraction comprised the greatest proportion of particles at the duodenum for both

309

treatments (Table 65). However, AL decreased the proportion of pdNDF S particles compared
with OG (17.3 vs. 34.2% of total NDF, P < 0.001, Table 65). Lower intake and greater digestion
rate of pdNDF for AL than OG resulted in the reduction of pdNDF S at the duodenum for AL
despite the faster passage rate of pdNDF S (Table 66) and lower NDF digestibility (38.3% vs.
53.3%/h, 30 h in vitro fermentation, Table 63) for AL than OG.

CONCLUSIONS
Alfalfa increased rates of reduction of iNDF L to S and passage of pdNDF S and
iNDF S compared with OG. The passage rate of particles was not likely limited for either
treatment by the rate of reduction or particle size because the proportion of NDF in the rumen
below the threshold for passage was greater than 55%. Additionally, it was not likely limited by
buoyancy because of similar passage rates for pdNDF S and iNDF S within forage family. The
slower passage rate of pdNDF S and iNDF S for OG was likely because of greater entrapment of
NDF S within the rumen mat by the larger pool of NDF L. Particle size reduction was a
prerequisite to ruminal passage but not a limiting constraint in this experiment. When alfalfa or
orchardgrass silage was the only source of forage in diets formulated to contain similar
concentrations of forage NDF, selective retention of S particles was less for legume than coolseason grass resulting in lower rumen fill and less effective fiber.

310

APPENDIX

311

Table 62. Characterization of 13 cows during the final 4 d of the 14-d preliminary period, when
cows were fed a common diet
Standard
Parameter
Median
Mean
deviation
Minimum
Maximum
Parity
3
3.31
1.16
2
5
1
591
587
51
489
710
BW , kg
BCS
2.00
2.35
0.69
1.58
4.00
DIM
132
157
90
64
337
Milk, kg/d
41.4
41.5
10.8
22.6
57.1
3.5% FCM, kg/d
43.1
42.1
11.9
24.3
60.3
DMI, kg/d
26.7
25.9
3.0
19.6
29.5
1
Empty BW (ruminal digesta removed).

312

Table 63. Chemical composition and particle size distribution of the alfalfa silage and
orchardgrass silage included in the treatment diets
Silage
Item
Alfalfa
Orchardgrass
Chemical composition
DM, %
43.5
33.7
OM, % DM
91.9
90.3
NDF, % DM
42.3
58.2
1
23.0
16.1
iNDF , % DM
iNDF, % of NDF
54.5
27.7
ADF, % of DM
35.0
36.4
2
7.56
6.03
ADL , % of DM
CP, % DM
22.5
11.4
Starch, % DM
1.87
1.37
3
38.3
53.3
NDF digestibility , %/h
4

Particles size distribution
Wet sieving, % DM retained
19.0 mm
9.50 mm
4.75 mm
2.36 mm
1.18 mm
0.600 mm
0.300 mm
0.150 mm
0.075 mm
0.038 mm
5
Mean particle size , mm
Penn State Particle Separator,
% DM retained
> 19.0 mm
19.0 to 8.0 mm
< 8.0 mm
1
iNDF = indigestible NDF.

21.4
18.0
30.8
17.0
5.72
3.09
1.97
1.16
0.40
0.50
11.6

12.3
18.4
37.2
21.2
6.15
2.08
1.02
0.94
0.37
0.37
9.66

29.3
48.5
22.2

17.1
50.2
32.7

2

ADL = acid detergent lignin.

3

30 h in vitro NDF digestibility.

4

Particle size distributions of silages were measured each period (n = 2).

5

Mean particle size calculated from particle size distribution determined by wet sieving.

313

Table 64. Ingredients and chemical composition of preliminary and treatment diets (as analyzed)
containing either alfalfa (AL) or orchardgrass (OG) silage as the sole source of forage
Preliminary
AL
OG
Ingredients, % DM
Alfalfa silage
30.0
59.9
--Orchardgrass silage
21.5
--42.7
Dry ground corn
36.2
33.6
36.6
Soybean meal (48% CP)
5.81
--11.8
®
1.82
2.50
3.39
SoyPlus
1
3.99
3.99
3.99
Vitamin mineral mix

1

Urea
Limestone
Chemical composition
DM, %
OM, % DM
NDF, % DM
% forage NDF
% NDF from forage
2
iNDF , % DM
iNDF, % of NDF
CP, % DM
Starch, % DM

0.15
0.60
51.6
92.4
29.1
24.7
84.8
3
NA
NA
17.5
33.5

-----

0.30
1.20

54.5
92.7
29.2
25.3
86.8
14.8

52.3
91.1
30.2
24.9
82.3
8.24

50.7
18.4
27.3

27.3
17.0
29.6

Vitamin mineral mix contained (DM basis) 16.5% sodium bicarbonate, 14.2% magnesium
sulfate, 7.1% salt, 5.8% dicalcium phosphate, 2.4% trace mineral premix, 0.4% vitamin A, 0.4%
vitamin D, 0.2% vitamin E, and 53.1% dry ground corn as a carrier.
2
iNDF = indigestible NDF.
3

NA = no analysis for preliminary diet.

314

Table 65. Particle size distribution of intake, rumen pool, and duodenal flux for cows fed treatment diets containing alfalfa (AL) or
orchardgrass (OG) silage as the sole source of forage
1

3

Variable

Intake
NDF L, kg/d
NDF S, kg/d
iNDF L, kg/d
iNDF S, kg/d
pdNDF L, kg/d
pdNDF S, kg/d
NDF L, % of NDF
NDF S, % of NDF
iNDF L, % of NDF
iNDF S, % of NDF
pdNDF L, % of NDF
pdNDF S, % of NDF
Rumen pool
NDF L, kg/d
NDF S, kg/d
iNDF L, kg/d
iNDF S, kg/d
pdNDF L, kg/d
pdNDF S, kg/d
NDF L, % of NDF
NDF S, % of NDF
iNDF L, % of NDF
iNDF S, % of NDF
pdNDF L, % of NDF

Treatment LSM
AL
OG

2

P
SE

Trt

Trt x
Period

pDMI

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

0.08

0.02
0.03

NS
NS

NS
NS

NS
NS

0.02
0.01
0.03
0.03
0.13
0.13
0.12
0.17
0.03
0.002

NS
0.11
NS
NS
0.03
0.03
0.03
0.06
NS
0.08

NS
NS
NS
NS
0.03
0.03
0.11
NS
0.009
0.001

NS
NS
NS
NS
0.67
0.67
NS
NS
NS
NS

0.14
NS
0.13
NS
0.15
NS
0.07
0.07
0.16
NS
0.03

NS
NS
NS
NS
NS
NS
NS
NS
0.35
NS
NS

NS
NS
NS
NS
NS
NS
NS
NS
0.69
NS
NS

NS
NS
NS
NS
NS
NS
NS
NS
0.10
NS
NS

5.49
1.94

5.81
1.94

0.16
0.05

0.05
0.97

3.24
0.77
2.25
1.16
73.7
26.3
43.9
10.4
29.8
15.9

1.58
0.53
4.23
1.40
74.7
25.3
20.6
6.91
54.1
18.4

0.07
0.02
0.09
0.03
0.2
0.2
0.2
0.07
0.2
0.1

<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001

NS
NS
0.10
0.02
NS
0.005
0.005
<0.001
NS
<0.001
<0.001

2.30
3.89
1.70
3.10
0.59
0.79
36.9
63.1
26.6
50.6
9.37

2.95
3.72
1.25
2.02
1.69
1.71
44.2
55.8
19.4
30.3
25.3

0.16
0.23
0.09
0.14
0.08
0.12
0.9
0.9
0.8
0.7
0.50

0.001
0.17
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001

NS
NS
NS
NS
NS
NS
NS
NS
NS
0.19
NS

315

4

Table 65 (cont’d)

1

pdNDF S, % of NDF
Duodenal flux
NDF L, kg/d
NDF S, kg/d
iNDF L, kg/d
iNDF S, kg/d
pdNDF L, kg/d
pdNDF S, kg/d
NDF L, % of NDF
NDF S, % of NDF
iNDF L, % of NDF
iNDF S, % of NDF
pdNDF L, % of NDF
pdNDF S, % of NDF

12.6

25.5

0.8

<0.001

NS

0.19

0.08

NS

NS

0.57
3.30
0.42
2.80
0.15
0.69
15.2
84.8
11.5
67.1
4.15
17.3

0.53
2.12
0.25
1.24
0.28
0.93
21.1
78.9
10.6
44.3
10.9
34.2

0.08
0.20
0.06
0.12
0.02
0.07
1.7
1.7
1.4
2.6
0.76
1.9

0.69
<0.001
0.05
<0.001
0.001
0.03
0.02
0.02
0.59
<0.001
<0.001
<0.001

NS
NS
NS
NS
NS
NS
0.20
0.20
NS
NS
NS
NS

0.32
0.31
NS
0.18
0.85
0.31
0.76
0.76
0.75
0.12
0.35
0.05

0.16
0.002
NS
NS
0.16
0.15
NS
NS
NS
0.18
NS
NS

NS
0.40
NS
NS
NS
NS
0.10
0.10
0.10
0.17
0.17
NS

NS
0.07
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
iNDF = indigestible NDF, pdNDF = potentially digestible NDF, L = large particles (!2.36 mm), S = small particles (<2.36 mm).
4

NS = not significant (P > 0.20); term removed from statistical model.

316

Table 66. Particle size kinetics for cows fed treatment diets containing alfalfa (AL) or orchardgrass (OG) silage as the sole source of
forage
1

2

Variable

Treatment LSM
AL
OG

SE

Trt

Trt x
Period

pDMI

Rate of reduction, %/h
iNDF (L to S)

7.16

4.67

0.35

<0.001

0.18

NS

1.18
3.85
1.05
3.80

1.01
2.66
0.75
2.50

0.16
0.23
0.11
0.28

0.38
<0.001
0.07
0.002

0.15
0.17

0.19
0.23

0.02
0.01

0.09
0.02

3

1

Passage rate, %/h
iNDF L
iNDF S
pdNDF L
pdNDF S
Duodenal flux L/total
duodenal flux
iNDF
pdNDF

P

Trt x
pDMI

pDMI x
pDMI

Trt x
pDMI x
pDMI

4

NS

NS

NS

NS
NS
NS
0.13

0.84
NS
0.88
NS

NS
NS
0.10
NS

0.13
NS
NS
NS

NS
NS
NS
NS

NS
0.03

0.91
0.06

NS
NS

0.06
NS

NS
NS

Treatment least squares means.

2

P-values for treatment (Trt), treatment by period interaction (Trt x Period), preliminary DMI (pDMI), treatment by preliminary DMI
interaction (Trt x pDMI), quadratic effect of preliminary DMI (pDMI x pDMI), and treatment by quadratic effect of preliminary DMI
(Trt x pDMI x pDMI).
3
iNDF = indigestible NDF, pdNDF = potentially digestible NDF, L = large particles (!2.36 mm), S = small particles (<2.36 mm).
4

NS = not significant (P > 0.20); term removed from statistical model.

317

Figure 35. Model of ruminal particle size reduction and passage. Reduction of particle size
during eating is included in the rate of particle size reduction (kr). Passage rates (kpi) are
calculated for indigestible NDF (iNDF) and potentially digestible NDF (pdNDF); kr is calculated
for iNDF only. Figure reprinted from Voelker Linton and Allen (2007).

318

P < 0.001
SEM = 0.2

P < 0.001
SEM = 0.9

P = 0.02
SEM = 1.7

Figure 36. Particle size distribution of NDF in intake, rumen, and duodenum. Proportion of total
NDF as large (!2.36 mm; denoted by black) or small (<2.36 mm; denoted by white) particles for
cows fed diets containing alfalfa (AL) or orchardgrass (OG) as the sole source of forage. Least
squares mean and standard error of the mean (SEM) are shown. P-values above each set of
columns indicate significance for comparison of AL and OG within NDF large and small
particles.

319

Figure 37. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for the proportion of NDF intake as A) large particles (NDF L;
!2.36 mm; P = 0.03) or B) small particles (NDF S; P = 0.03). The preliminary DMI on the xaxis are the mean DMI of individual cows during the final 4 d of the preliminary period when all
cows were fed a common diet.

320

Figure 38. Interaction of alfalfa (open circles, dashed line) and orchardgrass (closed circles,
solid line) with preliminary DMI for passage rate of large particles (!2.36 mm) of potentially
digestible NDF (pdNDF L; P = 0.10). The preliminary DMI on the x-axis are the mean DMI of
individual cows during the final 4 d of the preliminary period when all cows were fed a common
diet.

321

P = 0.81

P = 0.49

Figure 39. Passage rates of indigestible NDF small particles (iNDF S; <2.36 mm; white bars)
and potentially digestible NDF small particles (pdNDF S; <2.36 mm; black bars) for cows fed
diets containing alfalfa or orchardgrass as the sole source of forage. Least squares mean are
shown and error bars represent standard error of the mean. P-values above each set of column
indicate significance for comparison of iNDF S and pdNDF S within alfalfa and orchardgrass
based on analysis using paired t-test.

322

REFERENCES

323

REFERENCES
Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J. Anim.
Sci. 74:3063-3075.
Allen, M. S., and D. R. Mertens. 1988. Evaluating constraints of fiber digestion by rumen
microbes. J. Nutr. 118:261-270.
Bayat, A. R., M. Rinne, K. Kuoppala, S. Ahvenjärvi, A. Vanhatalo, and P. Huhtanen. 2010.
Ruminal large and small particle kinetics in dairy cows fed red clover and grass silages
harvested at two stages of growth. Anim. Feed Sci. Technol. 155:86-98.
Buxton, D. R., D. R. Mertens, and D. S. Fisher. 1996. Forage quality and ruminant
utilization. Pages 229-266 in Cool-Season Forage Grasses. L. E. Moser, D. R. Buxton,
and M. D. Casler, ed. American Society of Agronomy, Madison, WI.
Evans, E. W., G. R. Pearce, J. Burnett, and S. L. Pillinger. 1973. Changes in some physical
characteristics of the digesta in the reticulo-rumen of cows fed once daily. Br. J. Nut.
29:357-376.
Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analysis (Apparatus, Reagents,
Procedures, and Some Applications). Agric. Handbook No. 379. ARS-USDA,
Washington, DC.
Joy, M. T., E. J. DePeters, J. G. Fadel, and R. A. Zinn. 1997. Effects of corn processing on the
site and extent of digestion in lactating cows. J. Dairy Sci. 80:2087-2097.
Jung, H. G., and M. S. Allen. 1995. Characteristics of plant cell walls affecting intake and
digestibility of forages by ruminants. J. Anim. Sci. 73:2774-2790.
Krizsan, S. J., S. Ahvenjärvi, and P. Huhtanen. 2010. A meta-analysis of passage rate estimated
by rumen evacuation with cattle and evaluation of passage rate prediction models. J.
Dairy Sci. 93:5890-5901.
Lammers, B. P., D. R. Buckmaster, and A. J. Heinrichs. 1996. A simplified method for the
analysis of particle sizes of forage and total mixed rations. J. Dairy Sci. 79:922-928.
Mertens, D. R. 2002. Gravimetric determination of amylase-treated neutral detergent fiber in
feeds using refluxing in beakers or crucibles: collaborative study. J. AOAC Int. 85:12171240.
Okine, E. K., and G. W. Mathison. 1991. Effects of feed intake on particle distribution, passage
of digesta, and extent of digestion in the gastrointestinal tract of cattle. J. Anim. Sci.
69:3435-3445.
Poppi, D. P., W. C. Ellis, J. H. Matis, and C. E. Lascano. 2001. Marker concentration patterns of

324

labeled leaf and stem particles in the rumen of cattle grazing Bermuda grass (Cynodon
dactylon) analysed by reference to a raft model. Br. J. Nutr. 85:553-563.
Rinne, M., P. Huhtanen, and S. Jaakkola. 2002. Digestive processes of dairy cows fed silages
harvested at four stages of grass maturity. J. Anim. Sci. 80:1986–1998.
Robinson, P. J., S. Tamminga, and A. M. van Vuuren. 1987. Influence of declining level of feed
intake and varying proportion of starch in the concentrate on rumen ingesta quantity,
composition and kinetics of ingesta turnover in dairy cows. Livest. Prod. Sci. 17:37-62.
Stensig, T., and P. H. Robinson. 1997. Digestion and passage kinetics of forage fiber in dairy
cows as affected by fiber-free concentrates in the diet. J. Dairy Sci. 80:1339-1352.
Sutherland, T. M. 1988. Particle separation in the forestomachs of sheep. In: A. Dobson and M.
J. Dobson (ed.) Aspects of Digestive Physiology in Ruminants, Cornell Univ. Press,
Ithaca, NY.
Voelker Linton, J. A., and M. S. Allen. 2007. Nutrient demand affects ruminal digestion
responses to a change in dietary forage concentration. J. Dairy Sci. 90:4770-4779.
Voelker Linton, J. A., and M. S. Allen. 2008. Nutrient demand interacts with forage family to
affect intake and digestion responses in dairy cows. J. Dairy Sci. 91:2694-2701.
Wilson, J. R., and R. D. Hatfield. 1997. Structural and chemical changes of cell wall types during
stem development: consequences for fibre degradation by rumen microflora. Aust. J.
Agric. Res. 48:165-180.

325

CHAPTER 8
Epilogue
RECAP
Understanding how the level of feed intake and forage characteristics interact to affect
passage rate of individual feed fractions was the primary goal of the research presented in this
dissertation. Level of intake ranged from approximately 20 to 30 kg DM per day among cows on
the five experiments, and the forage characteristics evaluated included legume particle size (10
vs. 19 mm), grass particle size (10 vs. 19 mm), legume maturity (41 vs. 53% NDF), grass
maturity (45 vs. 54% NDF), and forage family (legume vs. grass). Alfalfa and orchardgrass
were selected as a representative legume and grass, respectively. We hypothesized that the
normal variation in diet characteristics related to forage alters both the passage rates of feed
fractions and the extent to which passage rates of these fractions are affected by voluntary DMI.
This chapter provides key results including some findings that challenge conventional wisdom,
application of the information gained from this research, and direction for future research.

SUMMARY OF RESULTS: EFFECTS OF FORAGE CHARACTERISTICS,
LEVEL OF INTAKE, AND THEIR INTERACTIONS
Passage rates of individual feed fractions were highly variable at a given level of
voluntary DMI across cow periods and depended upon the forage characteristic being evaluated.
Forage particle size, level of intake, and their interaction were not related to passage rates of
potentially digestible NDF (pdNDF), indigestible NDF (iNDF), or starch when cows consumed
alfalfa or orchardgrass silage-based diets. Passage rate of iNDF was greater for early compared
with late maturity alfalfa and was quadratically related to level of intake independent of
326

treatment, such that the passage rate of indigestible NDF (iNDF) increased as level of intake
increased from low to mid range of preliminary DMI (pDMI) and remained relatively constant
from mid to high range of pDMI (Figure 40A). Alfalfa maturity, level of intake, and their
interaction were not related to passage rates of potentially digestible NDF (pdNDF) or starch.
Similar to alfalfa maturity, passage rate of iNDF was greater for early compared with late
maturity orchardgrass and was quadratically related to level of intake independent of treatment.
However, the pattern for orchardgrass maturity was different than that for alfalfa maturity with
the passage rate of iNDF decreasing from low to mid range of pDMI and increasing from mid to
high range of pDMI (Figure 40B). Additionally, passage rate of pdNDF was lower for early
compared with late maturity orchardgrass and was quadratically related to level of intake
dependent upon treatment, such that passage rate of pdNDF decreased as pDMI increased for
cows consuming early maturity orchardgrass but increased as level of intake increased from low
to mid range of pDMI and decreased from mid to high range of pDMI for cows consuming late
maturity orchardgrass (Figure 40C). Furthermore, orchardgrass maturity and its interaction with
level of intake was related to passage rate of starch, which remained constant for early maturity
orchardgrass but increased linearly for late maturity orchardgrass as pDMI increased (Figure
40D). Passage rate of iNDF was faster for alfalfa compared with orchardgrass and was linearly
related to level of intake dependent upon treatment, such that the passage rate of iNDF increased
slightly for alfalfa and decreased for orchardgrass as pDMI increased with the greatest difference
for cows with high pDMI (Figure 40E). Passage rate of pdNDF was also faster for alfalfa
compared with orchardgrass but was not related to level of intake. Additionally, forage family,
level of intake, and their interactions were not related to the passage rate of starch. Although it
has become conventional wisdom that ruminal passage rate increases with feed intake (NRC,

327

2001), these results demonstrate that the extent to which passage rates of individual feed
fractions vary with level of intake depends upon forage characteristics.
It has also been widely accepted that microbial nitrogen flow to the duodenum increases
with voluntary DMI (NRC, 2001); however, this was not evident in our research. Forage
characteristics, level of intake, and their interactions were not related to microbial nitrogen flow
or efficiency. Relationships were detected between microbial efficiency and passage rates of
starch and pdNDF, digestion rate of starch, and amount of true ruminally digested OM
(TRDOM, kg/d) for various experiments but no consistent patterns were observed. In the
experiment with orchardgrass particle size, efficiency of microbial synthesis was positively
related to passage rates of starch (Figure 41A) and pdNDF (Figure 41B) for long and short
orchardgrass. When comparing alfalfa maturity, microbial efficiency was negatively related to
digestion rate of starch for early and late alfalfa (Figure 41C), and microbial efficiency
quadratically decreased as TRDOM increased for early maturity alfalfa but was not affected for
late maturity alfalfa (Figure 41D). In contrast to alfalfa maturity, microbial efficiency linearly
decreased as TRDOM increased for late maturity orchardgrass but was not affected for early
maturity orchardgrass (Figure 41E). When comparing forage family, microbial efficiency was
positively related to passage rate of pdNDF for orchardgrass but not alfalfa (Figure 41F). Based
on our results, microbial efficiency was not related to level of intake but was related to other
factors that varied among the experiments, and these inconsistencies suggest efficiency of
microbial production is affected by multiple factors.
We expected the more filling diets (i.e. increased forage particle size, advanced forage
maturity, and grass) to cause greater ruminal distention and limit feed intake of cows with high
DMI to a greater extent than cows with low DMI. However, we were unable to detect this in any

328

of the five experiments presented here. Visual assessment illustrated that ruminal distention
caused by the more filling diet may be more likely to limit feed intake for cows with high intake
compared to cows with low intake for alfalfa particle size (Figure 42A), orchardgrass particle
size (Figure 42B), and forage family (Figure 42C) but these interactions were not statistically
significant. There was no evidence that the filling effect of diets affected feed intake differently
for cows with high intake compared to cows with low intake due to alfalfa maturity (Figure 42D)
or orchardgrass maturity (Figure 42E). Lack of significant interactions are likely because of the
high variability among cows and are not surprising after evaluation of relationships between
DMI and ruminal NDF pool, the primary feed constituent associated with ruminal fill, and total
ruminal digesta wet weight, which were also inconsistent (Table 67). At best, NDF pool size and
total digesta wet weight explained 74% and 53% of the variation in DMI (Table 67). These
results indicate numerous factors affect physical fill in the rumen and control feed intake.
In conclusion, the research presented in this dissertation illustrates the complexity of
ruminal passage. Results demonstrate that the extent to which passage rates of feed fractions
vary with level of intake depends upon forage characteristics. The high variability among
individual cow responses at a given level of intake emphasizes the difficulty involved in
accurately predicting ruminal passage and digestibility: it is not a “one size fits all” concept that
can be easily incorporated into nutrition models. Some of the relationships presented are weak,
quadratic, and unexplainable at this time but important to report nonetheless. This body of work
provides the foundation for additional research in this area.

APPLICATION OF DATA

329

Although effects of DMI on passage rates are not consistent, these experiments provide
absolute measurements of passage rates of digesta fractions, which are necessary for the
development of equations to predict digesta passage from the rumen. The data provided by this
series of experiments can be summarized and used for modeling the effects of voluntary DMI
and forage characteristics on passage rate of digesta fractions. While modeling ruminal passage
can be improved, the degree of accuracy is limited because of the high variability and
inconsistencies noted among the five experiments. In addition to the data from these
experiments, data obtained in our laboratory from other experiments utilizing the pool and flux
method for estimating passage rate of digesta fractions can be compiled and used in a metaanalysis, which has the potential of discovering relationships. Results may be useful in the
development of new regression equations predicting passage rates of iNDF, pdNDF, and starch.
The data collected from these experiments allow equations to be developed using not only data
obtained in ruminal metabolism experiments (e.g. rumen pools) but also data that can be
obtained by commercial dairy farms (e.g. DMI and diet composition). Data sets containing more
easily measured parameters along with passage rates will increase the accuracy of the prediction
of passage rates in models intended for use on commercial dairy farms. These results can be
used to improve existing models to increase accuracy of prediction of ruminal digestibility and
nutrient flow to the duodenum or used in the development of new nutrition models.

FUTURE RESEARCH
Future experiments utilizing the same experimental design and pool and flux method
used in our research should be conducted to evaluate the effects of other dietary treatments
including factors related to grain type, conservation method and physical form, non-forage fiber

330

sources (high fiber byproducts), and supplemental fat on passage rates of digesta fractions. More
experiments evaluating forage characteristics are needed to validate or refute previous results.
Furthermore, compiling databases from additional studies would allow sufficient numbers of
observations to help overcome the lack of statistical power from single studies due to the large
variation among animals and forages. Further research examining rates of particle size
breakdown and particle passage from the rumen would be beneficial to determine the primary
constraint limiting passage. These research efforts will further improve the accuracy of nutrition
models to predict nutrient intake and utilization in dairy cows. Ultimately, improved accuracy of
ruminal digestion models from the incorporation of this and future research will enhance our
ability to formulate dairy cow diets to increase milk yield, decrease feed costs, and minimize
excretion of nutrients as waste products.

331

APPENDIX

332

Table 67. Relationship between DMI (kg/d) and ruminal NDF pool (kg) or ruminal digesta wet
weight (kg)
DMI and NDF pool
DMI and digesta wet weight
2
2
P
Linear (L) or
P
Linear (L) or
R
R
Quadratic (Q)
Quadratic (Q)
Legume particle size
Long alfalfa
0.13
0.19
L
0.41
0.06
L
Short alfalfa
<0.001
0.74
L
0.006
0.52
L
Grass particle size
Long orchardgrass
0.09
0.21
L
0.05
0.27
L
Short orchardgrass
0.55
0.03
L
0.14
0.16
L
Legume maturity
Early alfalfa
0.87
0.01
L
0.08
0.21
L
Late alfalfa
<0.001
0.63
L
0.002
0.53
L
Grass maturity
Early orchardgrass
0.33
0.09
L
0.16
0.17
L
Late orchardgrass
0.93
0.001
L
0.56
0.03
L
Forage family
Alfalfa
0.11
0.22
L
0.02
0.38
L
Orchardgrass
0.005
0.66
Q
0.03
0.51
Q

333

Figure 40
A) Alfalfa Maturity

iNDF ruminal passage rate, %/h

5.5
5.0

Early
Late

4.5
4.0
3.5
3.0
2.5
2.0
22

23

24

25

26

27

28

29

30

Preliminary DMI, kg/d
B) Orchardgrass Maturity

iNDF ruminal passage rate, %/h

5.0
4.5

Early
Late

4.0
3.5
3.0
2.5
2.0
1.5
1.0
23

24

25

26

27

28

29

Preliminary DMI, kg/d

334

Figure 40 (cont’d)
C) Orchardgrass Maturity

pdNDF ruminal passage rate, %/h

2.5

Early
Late

2.0
1.5
1.0
0.5
0.0
23

24

25

26

27

28

29

Preliminary DMI, kg/d

D) Orchardgrass Maturity
Starch ruminal passage rate, %/h

50

Early
Late

40
30
20
10
0
23

24

25

26

27

28

29

Preliminary DMI, kg/d

335

Figure 40 (cont’d)
E) Forage Family

iNDF ruminal passage rate, %/h

4.5

Alfalfa
Orchardgrass

4.0
3.5
3.0
2.5
2.0
1.5
18

20

22

24

26

28

30

Preliminary DMI, kg/d

Figure 40. Relationship between preliminary DMI and A) indigestible NDF (iNDF) passage
rate for alfalfa maturity (treatment: P = 0.01, preliminary DMI: P = 0.07 quadratic, interaction:
nonsignificant (NS)), B) iNDF passage rate for orchardgrass maturity (treatment: P < 0.001,
preliminary DMI: P = 0.08 quadratic, interaction: NS), C) potentially digestible NDF (pdNDF)
passage rate for orchardgrass maturity (treatment: P = 0.04, preliminary DMI: P = 0.09
quadratic, interaction: P = 0.11 quadratic), D) starch passage rate for orchardgrass maturity
(treatment: NS, preliminary DMI: P = 0.01 quadratic, interaction: P = 0.03 linear), and E) iNDF
passage rate for forage family (treatment: P < 0.001, preliminary DMI: NS, interaction: P = 0.09
linear). The preliminary DMI on the x-axis are the mean DMI of individual cows during the
final 4 d of the preliminary period when all cows were fed a common diet.

336

Figure 41
A) Orchardgrass Particle Size

Microbial efficiency,
g of microbial N / kg TRDOM

35

Long
Short

30
25
20
15
10

5

10

15

20

25

30

35

Starch ruminal passage rate, %/h
B) Orchardgrass Particle Size

Microbial efficiency,
g of microbial N / kg TRDOM

35

Long
Short

30
25
20
15
10
-0.5

0

0.5

1

1.5

2

2.5

pdNDF ruminal passage rate, %/h

337

Figure 41 (cont’d)
C) Alfalfa Maturity

Microbial efficiency,
g of microbial N / kg TRDOM

50

Early
Late

45
40
35
30
25
20

0

5

10

15

20

25

30

Starch ruminal digestion rate, %/h
D) Alfalfa Maturity

Microbial efficiency,
g of microbial N / kg TRDOM

50

Early
Late

45
40
35
30
25
20

8

10

12

14

16

18

True ruminally digested OM, kg/d

338

Figure 41 (cont’d)
E) Orchardgrass Maturity

Microbial efficiency,
g of microbial N / kg of TRDOM

55

Early
Late

50
45
40
35
30
25
20
15

9

10

11

12

13

14

15

16

17

True ruminally digested OM, kg/d
F) Forage Family

Microbial efficiency,
g of microbial N / kg TRDOM

50

Alfalfa
Orchardgrass

45
40
35
30
25
20
15

0

1

2

3

4

5

pdNDF ruminal passage rate, %/h

Figure 41. Relationship between microbial efficiency expressed as g of microbial N per kg true
ruminally digested OM (TRDOM) and A) starch passage rate for orchardgrass particle size
2
2
(Long: P < 0.001, R = 0.61; SHORT: P = 0.01, R = 0.38), B) potentially digestible NDF
2

(pdNDF) passage rate for orchardgrass particle size (Long: P = 0.006, R = 0.45; SHORT: P <
2

2

0.001, R = 0.64), C) starch digestion rate for alfalfa maturity (Early: P = 0.16, R = 0.14; Late:
339

Figure 41 (cont’d)
2

2

P = 0.02, R = 0.35), D) TRDOM for alfalfa maturity (Early: P = 0.03, R = 0.42; Late: P =
2

2

0.65, R = 0.02), E) TRDOM for orchardgrass maturity (Early: P = 0.68, R = 0.02; Late: P =
2

2

0.03, R = 0.37), and F) pdNDF passage rate for forage family (Alfalfa: P = 0.15, R = 0.18;
2

Orchardgrass: P = 0.02, R = 0.42).

340

Figure 42
A) Alfalfa Particle Size

32

Long
Short

30

DMI, kg/d

28
26
24
22
20
22

24

26

28

30

32

34

Preliminary DMI, kg/d

30
28

B) Orchardgrass Particle Size
Long
Short

DMI, kg/d

26
24
22
20
18
16
22

23

24

25

26

27

28

29

30

Preliminary DMI, kg/d

341

Figure 42 (cont’d)
C) Forage Family
30

Alfalfa
Orchardgrass

28

DMI, kg/d

26
24
22
20
18
18

20

22

24

26

28

30

Preliminary DMI, kg/d
D) Alfalfa Maturity
34
32

Early
Late

DMI, kg/d

30
28
26
24
22
20
22

23

24

25

26

27

28

29

30

Preliminary DMI, kg/d

342

Figure 42 (cont’d)
E) Orchardgrass Maturity
27
26

Early
Late

DMI, kg/d

25
24
23
22
21
20
19
23

24

25

26

27

28

29

Preliminary DMI, kg/d

Figure 42. Relationship between preliminary DMI (pDMI) and DMI for: A) alfalfa particle size
(treatment: P = 0.10, treatment by pDMI interaction: nonsignificant (NS)), B) orchardgrass
particle size (treatment: P = 0.06, treatment by pDMI interaction: NS), C) forage family
(treatment: P = 0.13, treatment by pDMI interaction: NS), D) alfalfa maturity (treatment: P =
0.003, treatment by pDMI interaction: NS), and E) orchardgrass maturity (treatment: P = 0.70,
treatment by pDMI interaction: NS). The preliminary DMI on the x-axis are the mean DMI of
individual cows during the final 4 d of the preliminary period when all cows were fed a common
diet.

343

REFERENCES

344

REFERENCES
National Research Council. 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. Natl.
Acad. Washington, DC.

345