the use of sorghum, oat, wheat, triticale, and pearl millet ...

The Pennsylvania State University

The Graduate School

College of Agricultural Sciences

THE USE OF SORGHUM, OAT, WHEAT, TRITICALE, AND PEARL MILLET

SILAGES IN LACTATING DAIRY COW DIETS

A Dissertation in

Animal Science

by

Michael Thomas Harper

2018 Michael Thomas Harper

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Doctor of Philosophy

May 2018

The dissertation of Michael Thomas Harper was reviewed and approved* by the following:

Alexander N. Hristov

Professor of Dairy Nutrition

Dissertation Advisor

Chair of Committee

Gregory W. Roth

Professor of Agronomy

Associate Department Head of Agronomy

Kevin J. Harvatine

Associate Professor of Nutritional Physiology

Marvin H. Hall

Professor of Forage Management

Terry Etherton

Distinguished Professor of Animal Nutrition

Head of the Department of Animal Science

*Signatures are on file in the Graduate School

iii

ABSTRACT

Cropping decisions are important to the economic and environmental

sustainability of dairy farms. Being able to feed alternative forage silages in addition to

corn silage and alfalfa haylage to lactating dairy cows may increase the resilience of dairy

farms.

A series of 3 experiments were conducted to investigate the effects of partially

replacing corn silage with the following alternative forage silages (AFS): brown midrib

dwarf forage sorghum (Sorghum bicolor L. Moench)) or fall grown oat (Avena sativa L.)

silages (Exp. 1), winter wheat (Triticum aestivum L.) or triticale (X Triticosecale) silages

(Exp. 2), or brown midrib dwarf pearl millet (Pennisetum glaucum L.) silage (Exp. 3)

The AFS were included at 10% of the diet dry matter of lactating dairy cows. The

experiments investigated the effect on dry matter intake, milk yield, milk components and

fatty acid profile, apparent total-tract nutrient digestibility, N utilization, and enteric

methane emissions. Additionally, we analyzed the in situ dry matter and neutral detergent

fiber disappearance of the AFS vs corn silage and alfalfa haylage. Sorghum was grown

in the summer and harvested in the milk stage. Oat was grown in the fall and harvested in

the boot stage. Wheat and triticale were planted in the fall as cover crops and harvested in

the spring at the boot stage. Pearl millet was harvested at the flag leaf visible stage. Corn

was harvested at one half milkline. All forages were ensiled. Neutral detergent fiber and

acid detergent fiber concentrations were higher in all AFS than in corn silage. Lignin

concentrations were less consistent with sorghum, wheat, and triticale silages having

higher lignin content than the corn silage, while the oat and pearl millet silages had lignin

iv

concentrations similar to that of corn silage. All AFS had < 1% starch compared with ≥

35% starch in corn silage.

Experiments 1 and 2 were replicated 3 × 3 Latin square design experiments with

3, 28-d periods and 12 lactating Holstein cows. The control diets contained 44% (DM

basis) corn silage for both experiments. Experiment 3 was a crossover design experiment

with 2 periods of 28 d each and 17 lactating Holstein cows. The control diet for

experiment 3 contained 50% (DM basis) corn silage.

In experiment 1, sorghum silage inclusion decreased dry matter intake, milk yield,

and milk protein content, but increased milk fat and maintained energy corrected milk

yield similar to the control. Oat silage had no effect on dry matter intake, milk yield or

components compared with the control. The oat silage diet increased apparent total-tract

digestibility of dietary nutrients, except starch, whereas the sorghum diet slightly

decreased digestibility. Cows consuming the oat silage diet had higher milk urea N and

urinary urea N concentrations. Milk N efficiency was decreased by the sorghum diet.

Diet did not affect enteric methane emissions. This study showed that oat silage can

partially replace corn silage at 10% of the diet DM with no effect on milk yield, but BMR

sorghum silage harvested at the milk stage with < 1% starch will decrease dry matter

intake and milk yield in dairy cows.

In experiment 2, dry matter intake was not affected by diet, but both wheat and

triticale silage decreased yields of milk (41.4 and 41.2 vs 42.7 kg/d) and milk

components, compared with corn silage. Milk fat from cows fed the wheat and triticale

diets contained higher concentrations of 4:0, 6:0, and 18:0 and tended to have lower

v

concentrations of total trans fatty acids. Apparent total-tract digestibility of DM and

organic matter was decreased in the wheat silage diet, and digestibility of neutral-and

acid-detergent fiber was increased in the triticale silage diet. The wheat and triticale

silage diets resulted in higher excretion of urinary urea, higher milk urea N, and lower

milk N efficiency, compared with the corn silage diet. Enteric CH4 emission per kg of

energy corrected milk was highest in the triticale silage diet. This study showed that, at

milk production of around 42 kg/d, wheat silage and triticale silage can partially replace

corn silage DM and not affect dry matter intake, but milk yield may decrease slightly.

In experiment 3, diet did not affect dry matter intake or energy corrected milk

yield, which averaged 46.7 kg/d. The pearl millet silage diet tended to increase milk fat

concentration, had no effect on milk fat yield, and increased milk urea N. Concentrations

and yields of milk protein and lactose were not affected by diet. Apparent total-tract

digestibility of DM decreased from 66.5% in the control diet to 64.5% in the pearl millet

silage diet. Similarly, organic matter and crude protein digestibility was decreased by

pearl millet silage, whereas neutral- and acid-detergent fiber digestibility was increased.

Total milk trans fatty acid concentration was decreased by pearl millet silage with a

particular decrease in trans-10 18:1. Urinary urea and fecal N excretion increased with

pearl millet silage compared with corn silage. Milk N efficiency decreased with pearl

millet silage. Pearl millet silage increased enteric methane emission from 396 to 454

g/cow/d and increased methane yield and intensity. Substituting corn silage with brown

midrib dwarf pearl millet silage at 10% of the diet dry matter supported high milk

production in dairy cows.

vi

Overall, the alternative forages fed in these experiments had fiber of greater

digestibility than the corn silage they replaced. However, AFS dry matter was less

digestible than corn silage due to the lower starch content and milk yields were

decreased. Alternative forages have the potential to increase dairy farm resilience in part

because of their varied use of water and nutrients and their positive effects on soil health,

including reducing soil erosion and increasing soil organic matter. Alternative forages

can be excellent forages for farms that need more forage fiber to feed more cows or

desire more highly digestible forage fiber for higher producing cows.

Keywords: forage, brown midrib, dairy cow, sorghum silage, oat silage, wheat

silage, triticale silage, pearl millet silage

vii

TABLE OF CONTENTS

List of Figures .......................................................................................................................... x

List of Tables ........................................................................................................................... xi

Abbreviations ........................................................................................................................... xiii

Acknowledgements .................................................................................................................. xiv

Chapter 1 Introduction ............................................................................................................. 1

References ........................................................................................................................ 6

Chapter 2 Literature Review .................................................................................................... 10

Role of forages in a dairy cow ration ............................................................................... 10 Importance of highly digestible fiber from forages ......................................................... 10 Warm Season Annual Forages ......................................................................................... 12

Sorghum ................................................................................................................... 12 Pearl Millet ............................................................................................................... 18

Cool Season Annual Forages ........................................................................................... 21 Oats .......................................................................................................................... 21 Wheat and Triticale .................................................................................................. 24 Other Alternative Forages ........................................................................................ 27

References ........................................................................................................................ 29

Chapter 3 Using brown midrib 6 dwarf forage sorghum silage and fall grown oat silage in

lactating dairy cow rations. .............................................................................................. 35

ABSTRACT ..................................................................................................................... 35 INTRODUCTION ........................................................................................................... 37 MATERIALS AND METHODS ..................................................................................... 39

Crops and Silages ..................................................................................................... 39 Animals and Diets .................................................................................................... 41 Sampling and Analyses ............................................................................................ 42 In Situ ....................................................................................................................... 45 Income Over Feed Costs .......................................................................................... 47 Statistical Analysis ................................................................................................... 48

RESULTS AND DISCUSSION ...................................................................................... 49 CONCLUSIONS .............................................................................................................. 63 ACKNOWLEDGEMENT ............................................................................................... 64 REFERENCES................................................................................................................. 64

Chapter 4 Inclusion of wheat and triticale silage in the diet of lactating dairy cows ............... 81

ABSTRACT ..................................................................................................................... 81

viii

INTRODUCTION ........................................................................................................... 83 MATERIALS AND METHODS ..................................................................................... 85

Crops and Silages ..................................................................................................... 85 Animals and Diets .................................................................................................... 86 Sampling and Analyses ............................................................................................ 87 In Situ ....................................................................................................................... 91 Income Over Feed Costs .......................................................................................... 92 Statistical Analysis ................................................................................................... 93

RESULTS AND DISCUSSION ...................................................................................... 94 Forages ..................................................................................................................... 94 Dry Matter Intake and Milk Yield ............................................................................ 95 Milk Composition and Yield .................................................................................... 97 Milk Fatty Acid ........................................................................................................ 97 Nutrient Intake and Digestibility .............................................................................. 98 In Situ ....................................................................................................................... 99 N Utilization ............................................................................................................. 100 Enteric CH4 and CO2 Emissions............................................................................... 101 Income Over Feed Cost ............................................................................................ 102

CONCLUSIONS .............................................................................................................. 103 ACKNOWLEDGEMENT ............................................................................................... 104 REFERENCES................................................................................................................. 104

Chapter 5 Inclusion of brown midrib dwarf pearl millet silage in the diet of lactating

dairy cows. ....................................................................................................................... 120

ABSTRACT ..................................................................................................................... 120 INTRODUCTION ........................................................................................................... 122 MATERIALS AND METHODS ..................................................................................... 124

Crops and Silages ..................................................................................................... 124 Animals and Diets .................................................................................................... 125 Sampling and Analyses ............................................................................................ 126 In Situ DM and NDF Degradation ........................................................................... 130 Statistical Analysis ................................................................................................... 131

RESULTS AND DISCUSSION ...................................................................................... 132 Forages ..................................................................................................................... 132 Dry Matter Intake, Body Weight, and Milk Yield ................................................... 133 Milk Composition and Yield .................................................................................... 134 Nutrient Intake and Digestibility .............................................................................. 135 Milk Fatty Acid ........................................................................................................ 136 In Situ DM and NDF Degradation ........................................................................... 137 N Utilization ............................................................................................................. 139 Enteric Methane and Carbon Dioxide Emissions ..................................................... 140

CONCLUSIONS .............................................................................................................. 141 ACKNOWLEDGEMENT ............................................................................................... 141 REFERENCES................................................................................................................. 142

Chapter 6 Conclusions and Future Research ........................................................................... 159

Conclusions ...................................................................................................................... 159

ix

Future Research ................................................................................................................ 161 Appendix Indigestible NDF1 of various ensiled forages ................................................. 163

x

LIST OF FIGURES

Figure 3-1. Ruminal in situ DM disappearance of forage sources. .......................................... 79

Figure 3-2. Ruminal in situ NDF disappearance of forage sources. ........................................ 80





xi

LIST OF TABLES

Table 3-1. Ingredient and chemical composition of the diets fed in the experiment ............... 70

Table 3-2. Nutrient composition and fermentation profile of oat and sorghum silages (%

of DM or as indicated)1 .................................................................................................... 72

Table 3-3. Ruminal in situ DM and NDF degradability of ensiled forages1 ............................ 73

Table 3-4. Effect of oat and sorghum silage on DMI, milk production, and feed efficiency

in lactating dairy cows ..................................................................................................... 74

Table 3-5. Effect of oat and sorghum silage on milk fatty acid composition (g/100 g of

total fatty acids) in lactating dairy cows .......................................................................... 75

Table 3-6. Effect of oat and sorghum silage on nutrient intake and apparent total-tract

digestibility in lactating dairy cows ................................................................................. 77

Table 3-7. Effect of oat and sorghum silage on nitrogen utilization, urinary purine

derivatives and carbon dioxide (CO2) and methane (CH4) emissions1 in lactating

dairy cows ........................................................................................................................ 78


Table 4-2. Nutrient composition and fermentation profile of triticale and wheat silages (%

of DM or as indicated)1 .................................................................................................... 111

Table 4-3. Effect of triticale and wheat silage on DMI, milk production, and feed

efficiency in lactating dairy cows .................................................................................... 112

Table 4-4. Effect of triticale and wheat silage on milk fatty acid composition (g/100 g of

total fatty acids) in lactating dairy cows .......................................................................... 113

Table 4-5. Effect of triticale and wheat silage on nutrient intake and apparent total-tract


Table 4-6. Effect of triticale and wheat silage on nitrogen utilization and urinary purine

derivatives in lactating dairy cows ................................................................................... 116

Table 4-7. Effect of triticale and wheat silage on carbon dioxide (CO2) and methane

(CH4) emissions1 in lactating dairy cows ......................................................................... 117


Table 5-2. Nutrient composition and fermentation profile of pearl millet and corn silages

(% of DM or as indicated)1 .............................................................................................. 150

Table 5-3. Effect of pearl millet silage on DMI, milk production, and feed efficiency in

lactating dairy cows ......................................................................................................... 151

xii

Table 5-4. Effect of pearl millet silage on nutrient intake and apparent total-tract


Table 5-5. Effect of pearl millet silage on milk fatty acid composition (g/100 g of total

fatty acids) in lactating dairy cows ................................................................................... 153

Table 5-6. Effect of pearl millet silage on nitrogen utilization and urinary purine

derivatives in lactating dairy cows ................................................................................... 155

Table 5-7. Effect of pearl millet silage on carbon dioxide (CO2) and methane (CH4)

emissions1 in lactating dairy cows ................................................................................... 156

xiii

ABBREVIATIONS

ADF Acid Detergent Fiber

ADG Average Daily Gain

BMR Brown Midrib

BW Body Weight

CP Crude Protein

DIM Days in Milk

DM Dry Matter

DMD Dry Matter Digestibility

DMI Dry Matter Intake

ECM Energy Corrected Milk

ESC Ethanol Soluble Carbohydrates

FA Fatty Acid

FCM Fat Corrected Milk

FPCM Fat and Protein Corrected Milk

MUFA Mono-unsaturated Fatty Acid

NDF Neutral Detergent Fiber

NDFD Neutral Detergent Fiber Digestibility

NEL Net Energy for Lactation

OM Organic Matter

PUFA Poly-unsaturated Fatty Acid

peNDF Physically Effective Neutral Detergent Fiber

VFA Volatile Fatty Acid

WSC Water Soluble Carbohydrates

xiv

ACKNOWLEDGEMENTS

I am extremely grateful for the opportunity to learn and conduct research here in the

Department of Animal Science at The Pennsylvania State University. Divine providence brought

me to Dr. Alexander Hristov in the central parklet of State College on New Year’s Eve 2013. He

offered me a job as his lab tech and later decided to offer me a position as a PhD student. I am

very grateful for his ability to see potential in me that I did not even see in myself. He continued

to work with me over the past few years to develop me into the dairy nutritionist I am today. I

thank Dr. Greg Roth for his guidance from the agronomic side of working with alternative

forages. I thank Drs. Kevin Harvatine and Marvin Hall for their input into my research and

speaking with me on multiple occasions.

Many others have helped me in my work, thought, and character. Among them is Dr.

Joonpyo Oh. He has been an example to me of perseverance that is required for the unique

hardships that we face. Audino Melgar has always been a smiling face of encouragement. Dr.

Krum Nedelkov always kept safety first and showed me what a young animal scientist can be

like. Tyler Frederick was instrumental in training me in the ways of the lab. Dr. Xianjiang Chen

was a great office mate to finish out my studies with and taught me my first Chinese characters.

Fabio Giallongo helped me formulate my early dairy cattle diets. Susanna Räisänen, Sergio,

Christian Martins, Katieli, Gilberto, Laiz Flores, Juliana Lopes, and many undergraduates were

very important in helping me conduct my research. Many other people and labs have made my

academic pursuits possible. Foremost among them are Dr. Harvatine’s lab (Dr. Michel Baldin,

Isaac Salfer, Elaine Barnoff, and others) and the milk fatty acid analysis that they helped me

perform. I greatly appreciate the hard work of the dairy barn crew under the leadership of Nadine

Houck and Travis Edwards. I would have had nothing to feed were it not for farm services that

grew, harvested, and ensiled the forages.

xv

Most importantly, I am extremely appreciative for the support of my beautiful wife,

Stacie Ann Harper, and my Creator, Jesus Christ, who have encouraged and strengthened me.

They believed in me when I lost hope. They were faithful to see me to the end of my PhD, which

is only the beginning.

xvi

EPIGRAPH

“All things were made by Him; and without Him was not any thing made that was made.”

John 1:3

Chapter 1

Introduction

Dairy cattle are ruminants designed to digest forages in a symbiotic relationship with

rumen microbes. Forages are a critical component of dairy cow diets around the world and

homegrown forages form the foundation of dairy rations in the northeastern U.S. (Wolf, 2003).

The amount and quality of the harvested forage positively affects the economic sustainability of

dairy farms and particularly of smaller farms (< 200 cows) which may be less specialized in milk

production and have less capital to purchase additional cropland than larger farms (Gillespie et

al., 2010). Many factors affect farmers decisions on which forage to grow including: yield,

palatability, digestibility, persistence, cost, and nutrient composition (high protein, high energy,

or high fiber).

Currently, in the northeastern U.S., the 2 principal dairy cattle forages are corn silage and

alfalfa haylage (Thoma et al., 2013). Since the late 1980s, corn silage production in the U.S. has

been increasing due to increasing yields per hectare while alfalfa and grass forage production has

decreased dramatically primarily due to reduced planted hectares of those crops (Martin et al.,

2017). According to a survey by Thoma et al. (2013), which generated a theoretical composite

diet based on the amounts fed daily in various farms, lactating cows in the northeast consumed

7.94 kg DM of corn silage, 3.15 kg DM of alfalfa silage and 1.29 kg DM of alfalfa hay, grass hay,

wheat silage, grass silage, and ryegrass silage combined. There are several reasons for the shift

towards more corn silage production and less legume and grass forages. Corn silage has a high

DM yield (over 21 t/ha), requires just one harvest per year which is simpler and leads to a more

nutritionally uniform forage supply, and is highly digestible and palatable containing over 30%

starch. The increase in corn silage utilization in dairy cow diets has followed the increased energy

2

demand of the modern dairy cow due to genetic advances that have had a dramatic 4-fold increase

in milk production per cow from 1944 to 2007 (Capper et al., 2009).

Although corn silage is high yielding, increased reliance on corn silage as a feedstuff

promotes crop rotations with more years of corn-after-corn. Continuous corn cropping is known

to result in yield declines following the first year (Bennett et al., 2012) and rotation to another

crop, such as soybeans, raises corn yields the following year (Thomison et al., 2011). Indeed,

more diverse systems may be more productive and resilient. Grassland that was planted with

greater diversity produced 2.7 times greater biomass than monocultures in a study by Tilman et

al. (2001). More complex pasture mixes were more productive and more stable in their yield from

year to year than monocultures as reported by Deak (2007). Growing alfalfa (Medicago sativa L.)

with berseem clover (Trifolium alexandrinum L.) at an 80:20 mixture was shown to improve

yields compared with the crops in monoculture by over 10% in an irrigated field trial (Al-

Suhaibani, 2010) and growing annual grasses as cover crops with crimson clover (Trifolium

incarnatum) increased DM yields by 18% compared with annual grasses alone (Brown et al.,

2018). Duffy et al. (2017) showed that increased biodiversity increased biomass production for

multiple ecosystems around the globe. Biologically diverse ecosystems are more productive but

monocultures are easier to manage and can also be high yielding (e.g., corn silage) at least in the

short term.

One practice that adds some diversity into a corn silage cropping system that can be

practically managed is double cropping. The agronomic practice of double cropping has been

increasing on dairy farms in the U.S. and future climate change is projected to increase the area

suitable for this practice (Seifert and Lobell, 2015). The northeast U. S. in particular is projected

to have the growing season increased by 12-17 more frost-free days in the spring and an

additional 11-20 frost-free days in the fall (Hristov et al., 2018). This change would greatly

increase the viability of double cropping. A common double cropping rotation includes a winter

3

annual successively paired with a summer annual. For dairy farms, this is commonly cereal rye or

triticale planted in the fall to be harvested in the spring at the flag leaf stage followed by planting

of corn for silage and subsequent harvest in the fall before the cycle repeats. Perhaps the clearest

motivation for dairy farmers to plant winter cover crops is to prevent soil erosion (Kasper et al.,

2001). It is often necessary to plant a winter cover crop after corn silage to protect the bare soil

during the winter and maintain long-term productivity of their cropland. However, other potential

benefits of double cropping include higher annual forage yields, healthier soils (Fae et al., 2009),

and increased nutrient utilization (Di and Cameron, 2002; Carey et al., 2016; Coblentz et al.,

2016). Managing manure nutrients for environmentally sound crop utilization can be challenging

for dairies that rely primarily on corn silage and alfalfa haylage forage. A survey conducted in

Minnesota showed that manure was applied to corn, after alfalfa, in excess of the N crop

requirement by more than 50% of dairy farmers (Yost et al., 2014). Double cropping with small

grains immobilizes manure N applied in the fall by incorporating it into plant tissue. This

decreases N leaching and subsequent water quality impairment.

Double cropping with cereal grains does not come without challenges. To ensure high

yields, cereal grains should be planted by mid-September on average in the Northeast U.S. which

is during or immediately after a demanding corn silage harvest. Likewise, the spring harvest

window for boot stage cereal grains is just before or during busy spring corn planting.

Additionally, much of the available manure should be spread at those times in the spring and fall

for efficient nutrient utilization, particularly N. This can result in labor shortages for a dairy farm.

To overcome this challenge, custom operators might be utilized for some of the field operations

so that delays in harvesting or planting are avoided.

Labor shortages exacerbate the risk of missing the narrow (5-7 d) harvest window for

good yield and quality of cereal crops at the flag leaf to boot stage. This presents a great

challenge particularly because these crops must be cut and wilted from 20-25% DM to around 30-

4

35% DM while this season of the year tends to have frequent rain events. A potential strategy to

offset this risk is to plant several species of cover crop with different maturity dates in separate

fields. For instance, winter barley matures 10-14 d earlier than triticale or winter wheat (Jemison

et al., 2012). It would also be beneficial to practice what has been termed ‘haylage in day’ which

is cutting, wilting and ensiling haylage crops all in one day as the name implies to fit haylage

harvest into narrow (1-2 d) spring dry periods. This is accomplished primarily by cutting into a

wide swath of 70-80% of the cutter bar width for rapid drying. Conditioning is not useful as it

does not affect crop dry down until 40% DM has been reached and foregoing conditioning has the

benefits of using less fuel, a smaller tractor, and simpler, less expensive mowing equipment

(Undersander and Saxe, 2013). The wilted crop of 30-35% DM is then raked into windrows,

chopped, and ensiled. Research from the University of Wisconsin actually reported higher non-

fiber carbohydrate concentrations in rapidly dried wide swath alfalfa haylage versus a narrow

swathed treatment (Undersander and Saxe, 2013).

The effect of these risks on cereal grain silage quality can be illustrated using the

Integrated Farm System Model (IFSM v. 4.3) running yearly iterations for 25 years of simulated

weather data. For instance, a 100 lactating cow dairy located in State College, Pennsylvania with

32 ha of alfalfa and 48 ha of corn for silage with 12 of those corn hectares double cropped with

wheat for silage is modeled by IFSM to yield a wheat silage with an average CP% of 14.9 ± 2.7

and an NDF% of 53.1 ± 2.5. When that same farm is modeled with twice the area (24 hectares)

of double cropped wheat silage, quality is decreased to 14.2 ± 2.8% CP and 54.2 ± 2.8% NDF

because the harvest timeframe is longer on average. But, when the model, with twice the area of

wheat, includes double the amount of available labor to represent custom harvesting, wheat silage

quality is equal to the first scenario for NDF (53.1 ± 2.7%) and slightly better in CP (15.2 ±

2.5%). Each scenario showed similar large amounts of variation from year to year with CP

5

ranging from 8.9-23.6% and NDF ranging from 49.0-57.0% showing the difficulty in harvesting

at the targeted maturity stage from year to year. Feeding plans will need to account for the

unpredictability in forage quality of cereal grain silage from year to year; though this issue is not

unique to cereal grain silage, it is generally more variable than corn silage.

Even with the aforementioned strategies of custom harvesting, planting multiple cereal

grains, and rapid wilting, there is a risk that boot stage harvest of cereal grains may be missed;

however, farmers have another good harvest opportunity at the soft dough stage. Numerous

studies indicate that a soft dough stage harvest is the next best maturity stage to boot stage harvest

(Acosta et al., 1991; Ashbell et al., 1997). Soft dough stage harvest of cereal grains typically

yields about 50% more DM per hectare than boot stage and, although NDF digestibility

decreases, starch concentration increases to provide a rapidly available energy source (Jemison et

al., 2012). This type of forage could be used as a physically effective NDF source, but the

negative effects on DMI would limit its use in peak and mid-lactation dairy cows. A potentially

better nutritional fit would be in late lactation diets to aid in preventing cows from gaining too

much body condition or in dry cow and heifer diets which have lower energy requirements.

Another challenge that dairy farmers relying on corn silage may face is more

unpredictable rainfall patterns as a result of potential climate change; though, exactly how the

weather patterns will change remains uncertain (IPCC, 2014). Specific to the northeast U. S.,

climate change models over the next 80+ yrs forecast higher summer temperatures with lower

summer rainfall, even though annual precipitation is anticipated to increase (Hristov et al., 2018).

Sorghum and pearl millet are two summer annual forage crops that have a higher water use

efficiency than corn silage and could also breakup continuous corn cropping (Aydin et al., 1999;

Zegada-Lizarazu and Iijima, 2005). Even without climate change, more water efficient crops may

have a greater yield than corn silage in certain areas because rainfall and water availability are

very spatially specific and differ even within watershed (Matlock et al., 2013).

6

Because of the popularity of corn silage and alfalfa haylage as dairy cow forages in the

U. S., other forage options are termed alternative forages. Common alternative forages include

the winter annual cereal grains (e.g. wheat, oats, rye, barley, and triticale) and warm season

annuals (e.g. sorghum, sudangrass, and pearl millet). Less common alternatives include other

legumes (e.g. soybeans, peas, and red clover) and cool season grasses (e.g. tall fescue, reed

canary grass).

Relatively little research has looked at alternative forages for inclusion in the diets of

high producing lactating cows. Lactating cows consume the most feed on a dairy and must be

considered in any forage program. Additionally, there is a need for crop diversity on dairy farms

to produce sustainably resilient yields from year to year that promote soil health and efficient

utilization of soil nutrients and water under changing weather conditions.

Therefore, the effect of partial replacement of corn silage with alternative forage silages

in diets of high producing lactating dairy cows was investigated. The following chapters of this

dissertation encompass a literature review on the use of alternative forages to feed lactating dairy

cows (Chapter 2), my published research that was conducted with BMR sorghum and oat silage

(Chapter 3), wheat and triticale silage (Chapter 4), and pearl millet silage (Chapter 5), and finally,

a conclusion (Chapter 6) with thoughts on the future use and research of alternative forages in

lactating dairy cow rations.

References

Acosta, Y. M., C. C. Stallings, C. E. Polan, and C. N. Miller. 1991. Evaluation of barley

silage harvested at boot and soft dough stages. J. Dairy Sci. 74:167-176.

7

Al-Suhaibani, N. A. 2010. Estimation Yield and Quality of Alfalfa and Clover for

Mixture Cropping Pattern at Different Seeding Rates. American-Eurasian J. Agric. & Environ.

Sci. 8:189-196.

Ashbell, G., Z. G. Weinberg, I. Bruckental, K. Tabori, and N. Sharet. 1997. Wheat silage:

Effect of cultivar and stage of maturity on yield and degradability in situ. J. Agric. Food Chem.

45:709–712.

Aydin, G., R. J. Grant, and J. O’Rear. 1999. Brown midrib sorghum in diets for lactating

dairy cows. J. Dairy Sci. 82:2127–2135.

Brown, A. N., G. Ferreira, C. L. Teets, W. E. Thomason, and C. D. Teutsch. 2018.

Nutritional composition and in vitro digestibility of grass and legume winter (cover) crops. J.

Dairy Sci. 101:2037-2047.

Carey, P. L., K. C. Cameron, H. J. Di, G. R. Edwards, and D. F. Chapman. 2016. Sowing

a winter catch crop can reduce nitrate leaching losses from winter-applied urine under simulated

forage grazing: a lysimeter study. Soil Use Manag. 32:329-337. doi:10.1111/sum.12276

Coblentz, W., W. Jokela, and J. S. Cavadin. 2016. Production and nitrogen use efficiency

of oat forages receiving slurry or urea. Agron. J. 108:1390–1404.

doi:10.2134/agronj2016.01.0009

Deak, A. 2007. Benefits of forage species diversity in grazing systems in Pennsylvania.

PhD Dissertation. Pennsylvania State Univ., University Park.

Di, H. J., and K. C. Cameron. 2002. Nitrate leaching in temperate agroecosystems:

sources, factors and mitigating strategies. Nutrient Cycling in Agroecosystems. 64:237–256.

Duffy, J. E., C. M. Godwin, and B. J. Cardinale. 2017. Biodiversity effects in the wild are

common and as strong as key drivers of productivity. Nature. 549:261-264.

Faé, G. S., R. M. Sulc, D. J. Barker, R. P. Dick, M. L. Eastridge, and N. Lorenz. 2009.

Integrating winter annual forages into a no-till corn silage system. Agron. J. 101:1286–1296.

8

Gillespie, J., R. Nehring, C. Sandretto, and C. Hallahan. 2010. Forage outsourcing in the

dairy sector: The extent of use and impact on farm profitability. Agric. Resour. Econ. Rev.

39(3):399-414.

Hristov, A. N., A. T. Degaetano, C. A. Rotz, E. Hoberg, R. H. Skinner, T. Felix, H. Li, P.

H. Patterson, G. Roth, M. Hall, T. L. Ott, L. Baumgard, W. Staniar, R. M. Hulet, C. Dell, A. F.

Brito, and D. Y. Hollinger. 2018. Climate change effects on livestock in the northeast U.S. and

strategies for adaptation. Climatic Change 146:33-45.

IFSM v 4.3. 2016. The integrated farm system model. C. A. Rotz, M. S. Corson, D. S.

Chianese, F. Montes, S. D. Hafner, H. F. Bonifacio, and C. U. Coiner. Pasture systems and

watershed management research unit. Agricultural Research Service. United States Department of

Agriculture.

IPCC, 2014: Climate Change 2014: Synthesis Report. Contribution of Working Groups I,

II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change

[Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, 151 pp

Jemison, J. M., H. M. Darby, and S. C. Reberg-Horton. 2012. Winter grain-short season

corn double crop forage production for New England. Agron. J. 104:256-264.

Kasper, T. C., J. K. Radke, and J. M. Laflen. 2001. Small grain cover crops and wheel

traffic effects on infiltration, runoff, and erosion. J. Soil and Water Conservation 56(2):160-164.

Martin, N. P., M. P. Russelle, J. M. Powell, C. J. Sniffen, S. I. Smith, J. M. Tricarico, and

R. J. Grant. 2017. Invited review: Sustainable forage and grain crop production for the US dairy

industry. J. Dairy Sci. 100:9479-9494.

Matlock, M., G. Thoma, E. Cummings, J. Cothren, M. Leh, and J. Wilson. 2013.

Geospatial analysis of potential water use, waterstress, and eutrophication impacts from dairy

production. Int. Dairy J. 31(Suppl.1):S78–S90.

9

Seifert, C. A., and D. B. Lobel. 2015. Response of double cropping suitability to climate

change in the United States. Environ. Res. Lett. 10:024002.

Tilman, D., P. B. Reich, J. Knops, D. Wedin, T. Mielke, and C. Lehman. 2001. Diversity

and productivity in a long-term grassland experiment. Science 294:843-845.

Thoma, G., J. Popp, D. Shonnard, D. Nutter, M. Matlock, R. Ulrich, W. Kellogg, D. Soo

Kim, Z. Heiderman, N. Kemper, F. Adom, and C. East. 2013. Regional analysis of greenhouse

gas emissions from USA dairy farms: A cradle to farm-gate assessment of the American dairy

industry circa 2008. Int. Dairy J. 21:S29–S40.

Thomison, P., P. Raul, R. Hammond, and R. Mullen. 2011. Managing risks in continuous

corn. C.O.R.N. Newsl. 2011-02. https://agcrops.osu.edu/newsletters/2011/02 (accessed 28 Dec.

2017).

Undersander, D. and C. Saxe. 2013. Field drying forage for hay and haylage. Focus on

Forage. Vol.12: No. 5. University of Wisconsin.

Wolf, C. A. 2003. The economics of dairy production. Vet. Clin. Food Anim. 19:271–

293.

Yost, M. A., J. A. Coulter, M. P. Russelle, and M. A. Davenport. 2014. Opportunities

exist to improve alfalfa and manure nitrogen crediting in corn following alfalfa. Agron. J.

106:2098–2106.

Zegada-Lizarazu, W., and M. Iijima. 2005. Deep root water uptake ability and water use

efficiency of pearl millet in comparison to other millet species. Plant Prod. Sci. 8:454-460.

10

Chapter 2

Literature Review

Role of forages in a dairy cow ration

Forages are the major source of peNDF, and that is arguably their most important

contribution to a dairy cow ration (Mertens, 1997). Forages can also be a significant source of

rapidly degradable energy in the form of starch or sugar (e.g. corn silage and oat silage,

respectively) and/or protein (e.g. alfalfa haylage, triticale silage) depending on the crop and time

of harvest. All forages grown on a dairy farm must provide value through a combination of the

yield of nutrients produced and the ecosystem services provided (e.g. soil cover). High yielding

forages are generally preferred due to the positive effect on short-term profitability. However,

forage cropping choices must be made from a whole farm system (soils, water, rations, manure,

etc.) perspective including both short-term (1-5 yr) and long-term (10-20+ yr) time frames. For

example, the need to plant a triticale cover crop to reduce soil erosion (a long-term issue) and

utilize manure nutrients in the fall and spring season (whole farm system thinking) with a slightly

shorter season corn silage may be a better choice than planting only a longer season (i.e. higher

yielding) corn silage crop. Farmers should identify the ration components (e.g. peNDF, starch,

CP) that the forage should supply to determine which crops to plant and in what rotation. The

value of alternative forages is usually in their ability to provide highly digestible peNDF.

Importance of highly digestible fiber from forages

Long fiber (8 mm) in forages is necessary for proper health and function of the rumen,

primarily through the physical effects of the fiber to promote rumination and rumen mat

11

formation (Mertens, 1997; Zebeli et al., 2012). Forages are essential for the maintenance of an

adequate rumen pH level, which in turn allows for normal fiber digestion and animal health

(Yang and Beauchemin, 2009). Kmicikewycz and Heinrichs (2014) demonstrated that

supplemental long hay helped maintain DMI in lactating dairy cows during a subacute ruminal

acidosis challenge. The rumen mat of long fiber particles helps retain smaller feed particles for

fermentation and enhances rumen motility to move VFA to the rumen wall for absorption (Zebeli

et al., 2012).

However, NDF is a bulky feed component with a relatively slow rate of digestion, 2-

4%/h (Mertens, 1987). Additionally, the rumen selectively retains (prevents passage of) buoyant

large fiber particles for continued fermentation. Together, the slow rates of NDF digestion and

passage create an upper limit to DMI due to rumen fill sensed by rumen stretch receptors as the

combination of ruminal digesta weight and volume (Allen, 1996). The effect of NDF content in

the diet on DMI is influenced by particle size and fragility, iNDF content, and fermentation rate

of the potentially digestible NDF fraction. Neutral detergent fiber that has a faster rate of

digestion and/or passage allows for increased DMI due to a decrease in rumen fill. Oba and Allen

(1999a) statistically evaluated the results of 13 forage comparisons with differing NDFD values

and found that for every unit increase in NDFD (measured in vitro or in situ) there was a 0.17 kg

increase in DMI and a 0.25 kg increase in 4% fat-corrected milk. Ivan et al. (2005) compared

high and low NDFD corn silage in dairy cow diets with substitution of the corn silage on a DM

and NDF basis. In both cases, milk and milk fat yield increased for the high NDFD corn silage

treatment. Of particular interest is the fact that when the low NDFD corn silage was substituted

with high NDFD corn silage on an NDF basis, there was a greater milk response for higher

producing cows. Therefore, it may be advisable to feed high NDFD forages to peak lactation

cows because those cows are more likely to have DMI limited by rumen fill than other cows in

the herd, as suggested by Oba and Allen (1999b).

12

Warm Season Annual Forages

Sorghum

Sorghum (Sorghum bicolor L. Moench) is a tall growing warm season annual grass. The

crop can yield more DM and grain per hectare than corn under certain environmental conditions

(e.g. low rainfall and low soil fertility) but the reverse is often true under irrigation or high

rainfall. The northeast U. S. usually receives enough rainfall for good corn silage yields, but some

soils are shallow with poor water-holding capacity and may be more suitable to growing

sorghum. Abdelhadi and Santini (2006) in Argentina, for instance, reported a whole-plant yield of

15.5 vs 7.0 t DM/ha and a grain yield of 6.5 vs 2.5 t DM/ha for grain sorghum vs corn,

respectively. Aydin et al. (1999) in Nebraska reported DM yields of 15.0 t/ha for standard

sorghum, 12.6 t/ha for BMR sorghum, 13.6 t/ha for corn, and 12.1 t/ha for alfalfa. In contrast,

under irrigation in Israel, DM yields were highest for corn (17.8 t/ha), intermediate for standard

sorghum (13.0 t/ha), and lowest for BMR sorghum (10.8 t/ha) where corn received more

irrigation water and N fertilizer than either sorghum crop (Miron et al., 2007). A sorghum

sudangrass cross (Sorghum bicolor x Sorghum vulgare var. sudangrass) grown in the Po valley of

northern Italy yielded 7.2 t DM/ha under a single cut system with irrigation vs corn silage that

yielded 15.4 t DM/ha, though the corn did receive an extra irrigation application and more

fertilizer (Colombini et al., 2010).

Sorghum silage digestibility (DMD and NDFD) has typically been poorer than corn

silage but BMR sorghum varieties have increased NDF digestibility comparable to corn silage.

Dry matter digestibility, measured by total fecal collection, of standard sorghum silage was

decreased compared with corn silage in a diet fed to heifers by Lusk et al. (1984) whereas a BMR

sorghum silage (a BMR 12 mutant, specifically) had similar DMD. The authors reported similar

13

ADF, NDF, and lignin for the BMR sorghum silage and corn silage but higher lignin in the

standard sorghum silage. Likewise, Grant et al. (1995), in a study with mid-lactation dairy cows,

reported lower ADF and NDF total-tract digestibility, estimated with acid insoluble ash, for a

normal sorghum silage-based diet vs either a corn silage or a BMR sorghum silage-based diet.

Standard sorghum silage had a lower total-tract NDF and ADF digestibility than corn silage in

dairy cows, whereas BMR sorghum silage had only a numerically lower NDF digestibility but a

significantly lower ADF digestibility in a study by Aydin et al. (1999). Oliver et al. (2004)

reported similar in situ NDFD at 48 h for BMR 6 sorghum (62.4%), BMR 18 sorghum (61.0%),

and corn (59.1%) silages but a lower NDFD for normal sorghum silage (56.4%). Colombini et

al. (2010) reported lower in situ DM degradability (72.4%) of a BMR sorghum sudangrass hybrid

silage compared with corn silage (79.4%). Colombini et al. (2010) also found a similar decrease

for the in situ NDF degradability between BMR sorghum sudangrass silage (66.6%) and corn

silage (70.9%). Though, the rate of NDF degradation was higher in the BMR sorghum sudangrass

silage and the end result was that calculated effective degradability of NDF was similar.

Abdelhadi and Santini (2006) reported a lower in vitro DMD for grain sorghum silage than corn

silage. In vitro DMD of corn and BMR sorghum silages were similar (67%) and that of standard

sorghum silage (64%) was lower in a study by Miron et al. (2007). The same authors reported

higher in vitro NDFD in BMR sorghum silage (60%) than in either standard sorghum silage or

corn silage (56%). In a total fecal collection study with Corriedale lambs, Di Marco et al. (2009)

compared the digestibility of grain, sweet and BMR sorghum silages harvested at the start of

grain filling with little starch accumulation. Starch was slightly higher in the grain sorghum silage

while lignin was slightly lower in the sweet sorghum silage. No statistical differences were found

among the forages for DM or OM though the sweet sorghum silage had numerically higher

values. Neutral detergent fiber digestibility was significantly higher in the lower lignin sweet

14

sorghum silage. Pino and Heinrichs (2017) measured higher in situ DMD of corn silage vs a

BMR sorghum silage.

In animal experiments feeding corn silage vs normal sorghum silage or BMR sorghum

silage, normal sorghum silage often decreases and BMR sorghum silage maintains animal

performance similar to corn silage. Lusk et al. (1984) found no difference in milk yield, averaging

24.6 kg/d, or milk components in two lactation trials comparing BMR sorghum silage (a mutant

of gene locus 12) vs corn silage in dairy cows.

Brown midrib sorghum silage was compared with normal sorghum, alfalfa, and corn

silages fed at 65% of the diet DM to mid-lactation Holstein cows in a 4 × 4 Latin square

experiment (Grant et al., 1995). Dry matter intake was higher for BMR sorghum silage (25.3

kg/d) than normal sorghum (20.4 kg/d) or alfalfa (19.6 kg/d) silages whereas corn silage DMI

(23.1 kg/d) was not different from any of the treatments. Milk fat concentration was similar for

BMR sorghum silage and corn silage but decreased in the normal sorghum silage diet. Fat

corrected milk yield was similar among BMR sorghum, alfalfa, and corn silages (25.8 kg/d) and

lower for the normal sorghum silage (17.9 kg/d). Compared with corn silage, milk protein

concentration was decreased significantly for the normal sorghum silage but only numerically for

the BMR sorghum silage. Milk lactose concentration was also decreased for the normal sorghum

silage compared with the corn silage but not for the BMR sorghum silage.

Aydin et al. (1999) compared standard sorghum, BMR sorghum, and corn silages at

35.3% of the diet DM in a 10-wk lactation study using Holstein cows. These authors kept all the

diets identical except for the target forage. Treatment silage nutrient composition was similar in

NDF and CP. Dry matter intake averaged 24.5 kg/d and was not affected by treatment. Milk yield

was increased by BMR sorghum (36.0 kg/d) over standard sorghum (33.8 kg/d) and was similar

to corn silage (34.6 kg/d). Milk composition was not affected by diet but feed efficiency was

15

higher for BMR sorghum than the other treatments likely due to a greater extent of NDF

digestion.

In a short-term Latin square experiment, Aydin et al. (1999) found a diet based on corn

silage, at 65% of diet DM, produced more milk (29.5 kg/d) than other silages; alfalfa (25.2 kg/d),

BMR sorghum (24.3 kg/d), or standard sorghum (21.5 kg/d). Milk protein percentage was highest

for corn silage (3.36%) vs the sorghum silages (3.23%). Dry matter intake, averaging 23.4 kg/d,

was similar among treatments. Neutral detergent fiber content of the sorghum silage treatments

were at least 5% units higher than the corn silage treatment; starch content was not reported.

Oliver et al. (2004) compared a normal sorghum silage, 2 BMR sorghum silages (a BMR

mutant of gene locus 6 and a mutant of locus 18), and a corn silage at 40% of the diet DM in

lactating Holstein cows. They found no difference in DMI (24 kg/d) but a decrease in milk yield

in the normal sorghum silage (31.0 kg/d) compared with the other treatments (33.4 kg/d).

Interestingly, milk fat concentration and yield were decreased in the normal sorghum silage

treatment despite a higher diet NDF content, which might be due to a lower NDFD in that diet

compared with the other treatments. The authors found no difference in milk protein or lactose

concentrations or yields among the treatments.

Grain sorghum silage was compared with corn silage as a pasture supplement for

growing steers and did not affect ADG, though it must be noted that the starch content of the

grain sorghum (22.9%) was higher than that of the corn silage (13.7%) (Abdelhadi and Santini,

2006). When the animal production data were combined with the crop production data it was seen

that the grain sorghum silage treatment had the highest carrying capacity.

A 3 x 3 Latin square design lactation study with 7 wk periods was conducted with

Holstein cows in Israel to compare a BMR sorghum silage, a standard sorghum silage, and a corn

silage (Miron et al., 2007). The authors replaced the corn silage, 41% of the diet DM, with

sorghum silage and ground corn at 35% and 6% of the diet DM, respectively. The highest milk

16

yield was with corn silage (42.1 kg/d), a similar yield with BMR sorghum silage (41.4 kg/d), and

the lowest yield was with standard sorghum silage (40.7 kg/d). Yield and concentration of milk

fat increased and milk protein decreased in the sorghum silage treatments vs the corn silage. Body

weight loss was increased in the sorghum silage treatments compared with the corn silage.

Colombini et al. (2010) fed mid-lactation cows either corn silage or BMR sorghum

sudangrass silage with corn meal in a change-over design with 35d periods. The authors

measured similar DMI (24.9 kg/d), milk yield (32.9 kg/d), milk component percentages and

yields as well as FCM yield.

Amer et al. (2012) replaced alfalfa silage with sweet sorghum silage at 35% of the diet

DM in early lactation dairy cows. The authors reported no change in DMI, averaging 25 kg/d, but

a reduction in milk yield from 36.8 to 33.0 kg/d with an increase in milk fat yield from 1.36 to

1.43 kg/d but decreases in milk crude protein (1.18 to 1.06 kg/d) and lactose (1.69 to 1.52 kg/d).

Energy corrected milk yield was not different. Probably the most interesting result is the decrease

in milk lactose concentration from 4.61 to 4.55%, though the mechanism for this outcome was

not described. The exact reasons for the milk production changes are difficult to elucidate

because the concentrate ingredient proportions were altered along with the forage change and the

resulting diets did not have similar NDF, CP, or starch contents. However, the authors comment

that the lower starch and higher NDF contents of the sorghum silage diet is likely the cause of the

lower milk yield and higher milk fat concentration.

Colombini et al. (2015) looked at the methane production of corn, grain sorghum, and

forage sorghum silages in diets fed to lactating Italian Friesian cows. Diets were balanced for

starch and NDF with supplemental ground corn and resulted in 41.5, 36.7, and 28.0% of target

forage diet DM for corn, grain sorghum, and forage sorghum silages, respectively. Forage

sorghum silage DMI and FPCM yield (18.2 and 24.1 kg/d) tended to be lower than either corn or

grain sorghum silages (20.0 and 25.8 kg/d, respectively). Milk component concentrations,

17

reported in a separate paper, (Colombini et al., 2012), were not different among the treatments.

Enteric methane emission was not different among the treatments (339 g/cow/d) but methane

yield (g CH4/kg DMI) tended to be higher for forage sorghum silage (18.7) than for grain

sorghum silage (17.3) or corn silage (16.9). The authors explained the higher methane yield in the

forage sorghum silage diet as the result of greater NDFD due to a slower passage rate in that diet.

Mid-lactation Holstein cows in Georgia were used in a randomized complete block

design by Bernard and Tao (2015) to compare diets based on 38.7% diet DM of corn silage or

dwarf BMR sorghum silage. Corn silage crops were either the first planting harvested in the

summer or the second planting which was then harvested in the fall. Similarly, the BMR sorghum

silage crops were either from the first summer harvest or a regrowth harvested in the fall. No

difference in DMI (22.0 kg/d) or milk yield (33.0 kg/d) was found among treatments but milk fat

concentration was increased in the BMR sorghum silage diets (3.47%) compared with the corn

silage diets (3.06%). This trial was balanced for diet nutrient composition, so again, specific

effects of the BMR sorghum silage cannot be identified; however, the study does show the ability

of dairy cows to produce comparable amounts of milk on a diet including BMR sorghum silage

similar to a diet based on corn silage.

Cattani et al. (2017) replaced corn silage with forage sorghum silage and corn meal in the

diet of Holstein cows in a study conducted in northern Italy. Diets were similar in nutrient

composition. Dry matter intake (24.7 kg/d) was the same between treatments, but milk yield (31.6

vs 29.8 kg/d) was decreased by the sorghum silage. However, milk fat concentration was

increased for sorghum silage (4.26%) vs corn silage (3.98%) which resulted in no difference for

ECM yield between the treatments. The sorghum silage diet did have a higher NDF content than

the corn silage diet (39.7 vs 36.5%, respectively), which the authors suggested may explain the

higher milk fat concentration in the sorghum silage diet. Milk FA profile showed lower PUFA

concentrations in milk from the sorghum silage treatment.

18

A meta-analysis by Sanchez-Duarte et al. (2017) looked at lactating dairy cow studies

comparing conventional sorghum silage, corn silage, and BMR sorghum silage, including many

of the aforementioned studies. Brown midrib sorghum vs conventional sorghum silage tended to

increase DMI and increased milk yield, milk fat concentration, and milk component yields. In

comparing BMR sorghum silage with conventional corn silage, it was found that milk fat

concentration increased but milk protein concentration decreased while having no change in

DMI, milk yield, or yields of milk components.

Results of studies comparing sorghum and BMR sorghum with corn silage have been

inconsistent though in general BMR sorghum silage yields similar animal performance as corn

silage. Many of the previous studies altered ingredient concentrations in the diets in addition to

the target silages which prevents a clear cause and effect relationship with the silage.

Furthermore, BMR sorghum silage grown in the northeast has not been compared to corn silage

in high producing cows but appears to be a viable option from research published to date.

Pearl Millet

Pearl millet (Pennisetum glaucum L.) is a warm season annual C4 grass. Its shorter

growing season than full season corn (65 d vs. 130 d) may make it more practical for double

cropping strategies in certain geographies with shorter growing seasons and in specific years (e.g.

abnormally wet spring seasons which delay corn planting). Additionally, pearl millet is drought

tolerant with a high water use efficiency which is a particularly important trait for crops planted

after cover crops that may decrease the available soil moisture (Maman et al., 2003; Zegada-

Lizarazu and Iijima, 2005). Pearl millet is traditionally harvested 2 or more times at around the

boot stage (Cherney et al., 1990; Mustafa et al., 2004), but a non-BMR variety of pearl millet

reportedly yielded 19.9 t DM/ha of silage in a single cut system in a 2016 New York field trial

19

(Kilcer, June 2017). There are BMR varieties of pearl millet that have reduced lignin and

increased digestibility but also decreased yield (Cherney et al., 1990; Mustafa et al., 2004;

Hassanat et al., 2007). This BMR gene has been paired with a brachytic dwarfing trait in some

hybrids for highly digestible forage that has reduced lodging. Cherney et al. (1990) demonstrated

that BMR pearl millet was preferred to normal pearl millet in grazing lambs. They also reported

that wethers, on an all forage diet, showed no difference in DMI for BMR and normal pearl millet

hay of first cutting pearl millet, but DMI was decreased to a greater extent in the second cutting

for normal pearl millet hay vs BMR pearl millet hay.

Mid-20th century research reported decreases in milk fat content of dairy cows grazing

pearl millet (Miller et al., 1965; Bucholtz et al., 1969) and consuming pearl millet greenchop vs

sudangrass (Harner et al., 1969; Schneider et al., 1970). More recent studies with pearl millet

silage, however, do not show decreases in milk fat concentration.

Ward et al. (2001) compared the in vivo digestibility of pearl millet, forage sorghum, and

tropical corn silages grown as the summer annual portion of an annual ryegrass-based double

cropping system in the southeastern U.S. Pearl millet and forage sorghum were harvested in the

vegetative stage and tropical corn was harvested at one-half milkline. Holstein heifers ate more of

the pearl millet silage than the other 2 treatments, but the pearl millet silage was less digestible.

Therefore, the authors recommended feeding either forage sorghum or tropical corn which had

similar DMI and digestibilities.

Amer and Mustafa (2010) investigated pearl millet silage in comparison to corn silage on

milk production of mid-lactation Holstein cows; the crops were grown in Quebec, Canada. The

pearl millet silage had higher NDF (66.9 vs 40.7%), 48h in situ NDFD (52.0 vs 39.1%), and CP

(13.0 vs 9.4%) than the corn silage. The diets contained 35% of the target forage and were

balanced to be isonitrogenous. There was no effect on DMI (23.9 kg/d) or milk yield (38.0 kg/d);

however, milk fat concentration was increased from 3.78% in the corn silage diet to 4.35% in the

20

pearl millet diet. Consequently, ECM and 4% FCM yields were both higher in the pearl millet

silage diet. The pearl millet silage diet was uniquely supplemented with a Ca salt of palm fatty

acids which may have contributed to the difference in milk fat concentration.

A regular pearl millet silage and a sweet pearl millet silage were compared with corn

silage in a study by Brunette et al. (2014) in Quebec, Canada. Early lactation Holstein cows in a 3

x 3 Latin square design experiment were fed the treatment forages at approximately 37% of diet

DM with additional high-moisture corn and less soybean meal in the pearl millet silage diets

along with being uniquely supplemented with a Ca salt of palm fatty acids. Even after these diet

modifications, starch remained lower, and NDF content higher, in both pearl millet silage diets.

Dry matter intake decreased in the pearl millet diets (22.8 vs 24.4 kg/d); milk yield was highest in

the corn silage diet (35.2 kg/d) and lowest in the regular pearl millet silage diet (32.7 kg/d)

whereas sweet pearl millet had intermediate milk yield (34.0 kg/d) and did not differ from either

diet. In situ effective degradability of DM was higher for the corn silage (65.3 vs 53.8%) than the

pearl millet silages with the opposite occurring for in situ effective degradability of NDFD. Pearl

millet silage had higher values (32.1 vs 18.9%) than corn silage. Total-tract digestibilities of the

diets, on the other hand, did not show any differences.

Brown midrib pearl millet silage has not yet been fed to high producing dairy cows in a

publish research trial, but the forage shows signs of being suitable due to the high digestibility of

its fiber portion in particular. Its shorter growing season also make it a possibly attractive double

cropping pairing with winter cereals (e.g. triticale).

21

Cool Season Annual Forages

Oats

Oat (Avena sativa L.) is a quick growing cool season annual cereal grain that can be used

as a whole crop silage. It can be planted either in the spring or late summer for harvest in early

summer or late autumn, respectively, but will not overwinter in the northern U.S. Studies using

spring grown oat forage have typically ranked the crop’s feeding value below corn, alfalfa,

barley, and wheat. More recent lactating cow studies with fall grown oats have reported high milk

yields.

In eastern Canada, Burgess et al. (1973) compared the milk producing ability of corn,

barley, wheat, and oat silages in Holstein cows. Corn was harvested in the dent stage and small

cereals were harvested in the soft dough stage. Diets consisted of one of the silages and a

concentrate mix. The corn silage-based diet resulted in the lowest DMI but the highest 4% FCM

yield (22.6 kg/d). The small cereal silages as a group performed similarly but had higher DMI and

lower milk yields than the corn silage.

Oltjen and Bolsen (1980) looked at the effects of corn, winter barley, winter wheat, and

spring oat silages on DMI and ADG of growing steers. Corn was harvested at the hard-dent stage

and small cereals were harvested at the dough stage. The authors reported the lowest DMI (6.67

kg/d) and ADG (0.48 kg/d) for the oat silage and regression analysis determined that increased

silage ADF content decreased ADG.

Barley, oat, triticale, and corn silages were compared by Khorasani et al. (1993) in an

experiment with lactating Holstein dairy cow TMR consisting of 50% target forage and 50%

concentrate mix. The cereal crops were planted in the spring in Alberta, Canada and harvested at

the early to mid-dough stage. The alfalfa was harvested at mid-bloom from the second cutting.

22

Alfalfa had a higher protein content and a lower NDF content (21.3% and 32.2%, respectively) vs

the cereals (17.6% and 36.6%, respectively). The study had a 3 wk covariate period followed by a

12 wk treatment period. Dry matter intake was lower in the oat (17.0 kg/d) and triticale (17.6

kg/d) diets than the barley (18.9 kg/d) and alfalfa diets (18.9 kg/d). Milk yield (26.7 kg/d) and 4%

FCM yield (25.0 kg/d) were only numerically decreased in the cereal silage diets compared with

the alfalfa diet. Total-tract diet OM and NDF digestibility was higher in the alfalfa and barley

diets than the oat and triticale diets.

Leonardi et al. (2005) evaluated the effect of chop length of oat silage on the DMI and

milk production of Holstein cows. Diets contained 25% corn silage, 25% oat silage, and 50%

concentrate (DM basis). Overall, DMI averaged 21.5 kg/d and milk yield averaged 39.1 kg/d;

however, both parameters decreased linearly with increased oat silage chop length.

A diet of 24% oat silage, 24% alfalfa silage, 42% barley grain-based energy supplement,

and a 10% protein supplement (DM basis) fed to Holstein dairy cows supported milk yield of

36.1 kg/d (Bhandari et al., 2008). Treatments were forage chop lengths of 19 and 6 mm. Dry

matter intake increased from 19.4 to 21.2 kg/d when the oat silage chop length was shortened. No

differences in concentrations of milk fat (3.00%) or milk protein (3.16%) were measured.

Contreras-Govea and Albrecht (2006) compared early summer (July) vs autumn

harvested (October) oat for forage. There was a 1 t DM/ha decrease in yield for the autumn oat

forage (6.7 t DM/ha) compared with the early summer oat (7.7 t DM/ha). However, when the

treatments were harvested 77 d after sowing, the autumn harvested oat was less mature in the

mid-boot stage vs the mid-milk stage for the early summer harvested oat. This resulted in lower

NDF (52.1%) and ADF (29.7%) in the autumn oat forage vs the early summer oat forage, NDF

(59.6%) and ADF (35.0%). Furthermore, leaf WSC (10.3%), stem WSC (22.1%), CP (18.0%),

and NDFD (61.2%) were higher in the autumn oat forage compared with the early summer

treatment containing 6.4% leaf WSC, 6.7% stem WSC, 13.5% CP, and 44.5% NDFD.

23

Coblentz et al. (2014) measured acceptable ADG of 0.85 kg/d in gravid Holstein heifers

grazing fall-grown oat for 6h each day and fed a supplemental TMR of alfalfa and rye silages.

The study was conducted in Wisconsin and pastures were planted in mid-August. Levels of WSC

in the Forage Plus variety of oat forage increased from 4.8 to 18.2 % DM from September to

November, respectively. Neutral detergent fiber concentrations (% of DM) of the same cultivar

increased quadratically from 44.1%, in September, to a peak of 53.7%, in October, and down to

47.2%, in November. Crude protein (% of DM) decreased from a high of 24.7% to a low of

12.6%.

Nitrogen fertilization of fall grown oat forage (Forage Plus), with rates ranging from 0 to

100 kg of N/ha, was found to increase DM yield along with NDF and CP content (Coblentz et al.,

2017). These changes also resulted in lower WSC concentrations and in vitro DMD but no

statistical changes for in vitro NDFD.

Hall and Coblentz (2017) partially replaced BMR corn silage in the diet of high

producing lactating Holstein cows with fall grown oat silage in an 8 wk randomized block design

experiment. The control diet contained 20% alfalfa silage, 35% BMR corn silage and 45%

concentrates (% of diet DM); the experimental diets partially replaced BMR corn silage to

contain 8 or 16% oat silage. Dry matter intake (29.4 kg/d) was high but not affected by treatment;

however, 3.5% FPCM decreased from 50.5 kg/d (control) to 48.5 kg/d (8% oat silage) to 46.2

kg/d (16% oat silage). Total-tract NDFD had a quadratic response: 56.3%, 54.9%, and 58.0% for

control, 8% oat silage, and 16% oat silage treatments, respectively.

The nutritional profile of fall grown oat silage indicate that it would be an ideal high

digestible fiber source for lactating dairy cows. Few animal studies have been conducted with fall

grown oat silage and the author is no aware of any studies comparing fall grown oat silage to

conventional corn silage in high producing dairy cows.

24

Wheat and Triticale

Wheat (triticum aestivum L.) and triticale (X Triticosecale) are cereal grains that can be

planted in the fall, after corn is harvested for silage, and harvested the following spring. These

winter crops, along with cereal rye (Secale cereale), have been popular for a number of years to

use as cover crops to reduce soil erosion, increase soil organic matter, and capture available soil

nutrients, but more recently dairy farmers have started to harvest these crops for forage as well.

The fiber of these crops can be highly digestible with in vitro NDFD values of 73-78% (% of

NDF) reported by Brown et al. (2018).

Jemison et al. (2012) compared crop yields and nutrient characteristics of winter grains

(harvested at the boot or soft dough stage) double cropped with short season corn vs full season

corn in Maine and Vermont. Winter grains were winter barley, triticale, and winter wheat. The

cold winters of northern New England caused around 40% of the winter barley to winter kill

which decreased yields compared with triticale or winter wheat. Waiting the 18 to 24 d from boot

stage to soft dough stage in the winter cereals increased DM yields by about 50%. Total DM

yields from double cropping with winter grain harvest at the boot or soft dough stage were 20 and

33% higher, respectively, than the 13.6 t/ha yield of full season corn single cropping.

A survey of winter cereals (cereal rye and triticale) double cropped with corn silage on

New York dairy farms reported DM yields of 3.62 and 4.88 t/ha for cereal rye and triticale,

respectively (Ketterings et al., 2015). Yields commonly ranged from 2-5 t/ha for the winter cereal

production.

Many studies have looked at the effect of harvest timing on the feeding value of cereal

grains. Sutton et al. (2001) summarized 3 experiments aimed to increase the efficiency of milk

production from urea treated whole-crop wheat silage in the UK by using molasses, concentrates,

or sodium hydroxide. They concluded that whole-crop wheat harvested at DM levels of 55-60%

25

(mature dough stage) was a relatively low energy feed and suggested that future work be focused

on harvesting the wheat at earlier stages of maturity.

Arieli and Adin (1994) compared early (mid-flowering) and late (end of milk stage)

harvest of wheat for silage on milk production in lactating Israeli-Friesian cows on a commercial

dairy. Wheat silage contributed 30.5 and 33.2% of the diet DM for the early and late harvest

treatments, respectively. Treatments were balanced to 33% NDF, 16.5% CP, and 1.73 NEL/kg.

Milk and milk protein yield were higher for the early treatment cows but with similar milk fat

yields between treatments due to a higher milk fat % in the late treatment. The authors attributed

these differences to a higher wheat silage NDFD for the early harvest wheat silage as measured in

situ. The DM yield/ha was 30% lower for the early harvest wheat but the resulting economics of

the early treatment diet were more profitable despite being a more expensive diet.

In Israel, wheat silage is a major crop for dairy cattle and roughages are relatively

expensive. Ashbell et al. (1997) compared 2 wheat cultivars (an early maturing and late maturing

variety) and 4 maturity stages at harvest (shooting, flowering, milk and dough) to assess the

optimum cultivar and harvest timing for use in dairy cattle diets. Whole plant DM yields

increased from 7.4 t/ha to 15.8 t/ha from shooting to dough stages, respectively. In situ DM

degradability decreased from shooting to flowering and milk stages but increased to similar levels

as the shooting stage for the dough stage whereas NDFD generally decreased with each

advancing maturity stage. The early maturing variety had higher DM degradability values

(72.6%) and similar NDFD values (38.6%) as the late maturing wheat for the shooting stage

sample. Including DM yields, the authors calculated the highest degradable DM yields for the

early maturing crop at the dough stage but the highest degradable NDF yields for the late

maturing variety.

Wheat silage grown in northern Italy was studied by Crovetto et al. (1998) to determine

optimum maturity stage for harvest. The authors investigated boot, mid-bloom, milk, and dough

26

maturity stages fed to Bergamasca wethers as the sole feed source. In vivo DM digestibility was

high for boot (73.4%) and mid-bloom (67.1%) stages and significantly lower for milk (59.2%)

and dough (59.6%) stages. Neutral detergent fiber digestibility decreased similarly (71.8, 65.5,

52.9, and 34.5% from boot through dough stages). The authors concluded that, although yield is

lower for earlier stage harvests, in a double cropping system with corn silage, targeting harvest

for mid-bloom would give the best balance of yield, digestibility, and corn planting timing.

Fonseca et al. (2005) tested dough stage wheat silage against a late cute ryegrass silage in

diets of lactating Holstein cows in a Latin square design experiment conducted in Portugal. Diets

were 55% concentrate and 45% test forage, fed separately. In situ DM and NDF degradabilities

were lower for the wheat silages compared with the ryegrass silage. Dry matter intake (20.2 kg/d)

milk yield (28.7 kg/d) and milk components were not affected.

Cosentino et al. (2015) compared diets of 50% corn silage or triticale silage for 105d in

lactating Holstein cows to identify which forage feeding strategy would have a smaller ‘water

footprint’. Diet did not affect DMI (21.7), milk yield (38.1 kg/d), milk fat (3.50%), or milk

protein (3.32%). The authors concluded that the triticale silage diet had the smaller water

footprint due to the similar production effects but a smaller water footprint for triticale silage

production.

Many studies have focused on the effects of harvest timing on the nutrients and

digestibility of wheat silage. A number of animal experiments feeding wheat silage to lactating

dairy cows have also been published. Few animal studies have fed winter wheat that was

harvested at the high digestible boot stage. Winter triticale has received even less attention than

winter wheat though its yield can be higher with comparable quality parameters.

27

Other Alternative Forages

There are some other alternative forages that have received less research attention. Tall

fescue silage (Festuca arundinacea Schreb.) was compared with orchardgrass (Dactylis

glomerata L.) and alfalfa (Medicago sativa L.) silages in lactating cows by Cherney et al. (2004).

Dry matter intake (26.4 kg/d) and milk yield (40.4 kg/d) for first cutting tall fescue, orchardgrass,

and alfalfa silages were similar, but for the higher NDF second cuttings of tall fescue and

orchardgrass silages, DMI (22.5 kg/d) and milk yield (35.7 kg/d) decreased. It should be noted

that the grass silage treatments had increased levels of concentrate in the ration. Bender et al.

(2016) investigated tall fescue hay as a replacement for either corn silage or alfalfa silage in dairy

cows milking an average of 40.4 kg/d. Dry matter intake was reduced slightly but milk yield only

numerically decreased. While corn silage was replaced, corn grain was increased to maintain

starch levels among diets. Total-tract NDF digestibility in vivo was increased when tall fescue

hay replaced corn silage.

A 40:60 mixture of kura clover (Trifolium ambiguum M. Bieb.) and reed canarygrass

(Phalaris arundinacea L.) silage was compared with alfalfa silage for feeding lactating cows due

to better persistence of the crop in the northern U.S. and greater adaptability to poorly drained

soils (Kammes et al. 2008). Diets contained 57% of the target forages. Neutral detergent fiber,

33.1 vs 25.6% was high for the kura clover/reed canarygrass silage based diet compared with the

alfalfa silage based diet. Dry matter intake (24.2 vs 22.8 kg/d) and 4% FCM yield (32.8 vs 30.9

kg/d) were higher for alfalfa silage vs kura clover/reed canarygrass silage, respectively.

Barley (Hordeum vulgare) silage, often fed in western Canada, was compared with corn

silage in lactating cow diets, at 54.4% of the diet DM, by researchers in Quebec (Benchaar et al.,

2014). The barley silage used in the experiment was higher in NDF (52.3 vs 36.7%) and lower in

starch (13.9 vs 32.2%) than the corn silage. In situ OM degradability was lower for barley silage

28

than corn silage. Total-tract DM digestibility was decreased in the barley silage diet from 69.3 to

63.2%. The barley silage diet had a lower DMI (22.0 kg/d) and milk yield (31.9 kg/d) than the

corn silage diet (27.2 and 37.0 kg/d, respectively). Milk fat concentration was higher in the barley

silage diet but milk fat yield was not increased. Furthermore, milk protein concentration and yield

increased in the corn silage diet. Enteric methane emissions increased with the increased milk

yield of the corn silage diet, but methane intensity (13.8 g CH4/kg ECM) was not different. The

higher NDF content and lower OM digestibility of the barley silage limited DMI and decreased

milk yield.

In the U. K., Moorby et al. (2003) evaluated a whole-crop barley (Hordeum vulgare)/kale

(Brassica oleracea) bi-crop silage (80:20 ratio) in comparison with a first-cut perennial ryegrass

silage as feed for lactating dairy cows. Dry matter intake (18.2 vs 16.3 kg/d) and milk yield (24.0

vs 22.6 kg/d) was higher for cows fed the bi-crop silage than the ryegrass silage. Milk fat yield

was higher for the ryegrass silage diet but milk protein and lactose yields were higher for the bi-

crop silage diet. The bi-crop silage diet had a more efficient use of N which the authors suggest

was the result of the higher starch content in the bi-crop silage.

Forage soybean (Glycine max L. Merr.) silage was compared with alfalfa haylage as a

forage for lactating dairy cows in Quebec, Canada (Vargas-Bello-Pérez et al., 2008). The

treatment silages were fed at 36% of the ration DM along with 12% corn silage. Soybean silage

decreased DMI (23.5 vs 25.1 kg/d) and milk yield (35.5 vs 37.2 kg/d) compared with alfalfa

haylage but did not affect milk fat yield (1.34 kg/d). Total-tract diet digestibilities of DM, OM,

CP, and NDF were not different between the treatments. Contrastingly, in situ degradability of

the silages showed that alfalfa haylage had higher effective degradabilities of DM, CP, and NDF

than soybean silage.

29

Many alternative forages could be fed to lactating cows. Additional forage species

options and feeding strategies should be investigated to reduce reliance on any one crop and,

thereby, make dairy farm forage production more consistent and robust.

References

Abdelhadia, L. O., and F. J. Santini. 2006. Corn silage versus grain sorghum silage as a

supplement to growing steers grazing high quality pastures: Effects on performance and ruminal

fermentation. Anim. Feed Sci. Technol. 127:33–43.

Allen, M. S. 1996. Physical constraints on voluntary intake of forages by ruminants. J.

Anim. Sci. 74:3063-3075.

Amer, S., P. Seguin, and A. F. Mustafa. 2012. Short communication: Effects of feeding

sweet sorghum silage on milk production of lactating dairy cows. J. Dairy Sci. 95:859–863.

Arieli, A., and G. Adin. 1994. Effect of wheat silage maturity on digestion and milk yield

in dairy cows. J. Dairy Sci. 77:237–243.


Effect of cultivar and stage of maturity on yield and degradability in situ. J. Agric. Food Chem.

45:709–712.



Benchaar, C., F. Hassanat, R. Gervais, P. Y. Chouinard, H. V. Petit, and D. I. Massé

2014. Methane production, digestion, ruminal fermentation, nitrogen balance, and milk

production of cows fed corn silage- or barley silage-based diets. J. Dairy Sci. 97:961-974.

30

Bender, R. W., F. Lopes, D. E. Cook, and D. K. Combs. 2016. Effects of partial

replacement of corn and alfalfa silage with tall fescue hay on total-tract digestibility and lactation

performance in dairy cows J. Dairy Sci. 99:5436–5444.

Bernard, J. K., and S. Tao. 2015. Short communication: Production response of lactating

dairy cows to brachytic forage sorghum silage compared with corn silage from first or second

harvest. J. Dairy Sci. 98:8994–9000.

Bhandari, S. K., S. Li, K. H. Ominski, K. M. Wittenberg, and J. C. Plaizier. 2008. Effects

of the chop lengths of alfalfa silage and oat silage on feed intake, milk production, feeding

behavior, and rumen fermentation of dairy cows. J. Dairy Sci. 91:1942–1958.

Brown, A. N., G. Ferreira, C. L. Teets, W. E. Thomason, and C. D. Teutsch. 2018.

Nutritional composition and in vitro digestibility of grass and legume winter (cover) crops. J.

Dairy Sci. 101:2037-2047.

Bucholtz, H. F., C. L. Davis, D. L. Palmquist, and K. A. Kendall. 1969. Study of the low-

fat milk phenomenon in cows grazing pearl millet pastures. J. Dairy Sci. 52:1388-1394.

Burgess, P. L., J. W. G. Nicholson, and E. A. Grant. 1973. Yield and nutritive value of

corn, barley, wheat and forage oats as silage for lactating dairy cows. Can. J. Anim. Sci. 53:245-

250.

Cattani, M., N. Guzzo, R. Mantovani and L. Bailoni. 2017. Effects of total replacement

of corn silage with sorghum silage on milk yield, composition, and quality. J. of Anim. Sci. and

Biotech. 8:15

Cherney, D. J. R., J. H. Cherney, and L. E. Chase. 2004. Lactation performance of

Holstein cows fed fescue, orchardgrass, or alfalfa Silage. J. Dairy Sci. 87:2268-2276.

Coblentz, W.K., G. E. Brink, P. C. Hoffman, N. M. Esser, and M. G. Bertram. 2014. Fall-

grown oat to extend the fall grazing season for replacement dairy heifers. J. Dairy Sci. 97:1645–

1660.

31

Coblentz, W. K., M. S. Akins, J. S. Cavadini, and W. E. Jokela. Net effects of nitrogen

fertilization on the nutritive value and digestibility of oat forages. 2017. J. Dairy Sci. 100:1739-

1750.

Colombini, S., L. Rapetti, D. Colombo, G. Galassi, and G. M. Crovetto. 2010. Brown

midrib forage sorghum silage for the dairy cow: nutritive value and comparison with corn silage

in the diet. Ital. J. of Anim. Sci. 9:273-277.

Colombini, S., G. Galassi, G. M. Crovetto and L. Rapetti. 2012. Milk production,

nitrogen balance, and fiber digestibility prediction of corn, whole plant grain sorghum, and foarge

sorghum silages in the dairy cow. J. Dairy Sci. 95:4457-4467.

Colombini, S., M. Zucali, L. Rapetti, G. M. Crovetto, A. Sandrucci, and L. Bava. 2015.

Substitution of corn silage with sorghum silages in lactating cow diets: In vivo methane emission

and global warming potential of milk production. Agricultural Systems 136:106-113.

Contreras-Govea, F. E., and K. A. Albrecht. 2006. Forage production and nutritive value

of oat in autumn and early summer. Crop Sci. 46:2382-2386.

Cosentino, C., F. Adduci, M. Musto, R. Paolino, P. Freschi, G. Pecora, C. D’Adamo, and

V. Valentini. 2015. Low vs high “water footprint assessment” diet in milk production: A

comparison between triticale and corn silage based diets. Emir. J. Food Agric. 27:312-317.

Crovetto, G. M., G. Galassi, L. Rapetti, A. Sandrucci, and A. Tamburini. 1998. Effect of

the stage of maturity on the nutritive value of whole crop wheat silage. Livest. Prod. Sci. 55:21–

32.

Di Marco, O. N., M. A. Ressia, S. Arias, M. S. Aello, and M. Arzadun. 2009.

Digestibility of forage silage from grain, sweet and bmr sorghum types: Comparison of in vivo, in

situ and in vitro data. Anim. Feed Sci. and Technol. 153:161-168.

32

Fonseca, A. J. M., A. R. J. Cabrita, C. S. S. Nogueira, D. S. P. Melo, Z. M. C. Lopes, and

J. M. F. Abreu. 2005. Lactation responses of dairy cows to whole-crop wheat or ryegrass silages.

Anim. Feed Sci. Technol. 118:153-160.

Hall, M. B. and W. K. Coblentz. 2017. Substitution of fall-grown oat forage for corn

silage affects lactating dairy cow performance. J. Dairy Sci. 100(Suppl. 2):397. (Abstr.)

Harner, J. P., R. W. Hemken, J. H. Vandersall, N. A. Clark, and B. A. Schneider. 1969.

Relationship of pearl millet to milk fat depression in dairy cows. I. Cation fertilization. J. Dairy

Sci. 52:1244-1248.

Ivan, S. K., R. J. Grant, D. Weakley, and J. Beck. 2005. Comparison of a corn silage

hybrid with high cell-wall content and digestibility with a hybrid of lower cell-wall content on

performance of Holstein cows. J. Dairy Sci. 88:244–254.

Jemison, J. M., H. M. Darby, and S. C. Reberg-Horton. 2012. Winter grain-short season

corn double crop forage production for New England. Agron. J. 104:256-264.

Kammes, K. L., G. B. H. Heemink, K. A. Albrecht, and D. K. Combs. 2008. Utilization

of kura clover-reed canarygrass silage versus alfalfa silage by lactating dairy cows. J. Dairy Sci.

91:3138–3144.

Ketterings, Q. M., S. Ort, S. N. Swink, G. Godwin, T. Kilcer, J. Miller, and W. Verbeten.

2015. Winter cereals as double crops in corn rotations on New York dairy farms. J. of Ag. Sci.

7:18-24.

Khorasani, G. R., E. K. Okine, J. J. Kennelly, and J. H. Helm. 1993. Effect of whole crop

cereal grain silage substituted for alfalfa silage on performance of lactating dairy cows. J. Dairy

Sci. 76:3536–3546.

Kilcer, T. 2017. Crop Soil News. June 2017. advancedagsys.com/wp-

content/uploads/2017/06/wet-conditions.pdf (accessed Jan. 7, 2018).

33

Kmicikewycz, A. D. and A. J. Heinrichs. 2014. Feeding lactating dairy cattle long hay

separate from the total mixed ration can maintain dry matter intake during incidents of low rumen

pH. J. Dairy Sci. 97:7175-7184.

Leonardi, C., K. J. Shinners, and L. E. Armentano. 2005. Effect of different dietary

geometric mean particle length and particle size distribution of oat silage on feeding behavior and

productive performance of dairy cattle. J. Dairy Sci. 88:698–710.

Mertens, D. R. 1987. Predicting intake and digestibility using mathematical models of

ruminal function. J. Dairy Sci. 64:1548– 1558.

Mertens, D. R. 1997. Creating a system for meeting the fiber requirements of dairy cows.

J. Dairy Sci. 80:1463-1481.

Miller, R. W., R. W. Hemken, J.H. Vandersall, D. R. Waldo, M. Okamato, and L. A.

Moore. 1965. Effect of feeding buffers to dairy cows grazing pearl millet or sudan grass. J. Dairy

Sci. 48:1319-1323.

Moorby, J. M., P. R. Evans and N. E. Young. 2003. Nutritive value of barley/kale bi-crop

silage for lactating dairy cows. Grass and Forage Sci. 58:184-191.

Oba, M., and M. S. Allen. 1999a. Evaluation of the importance of the digestibility of

neutral detergent fiber from forge: Effects on dry matter intake and milk yield of dairy cows. J.

Dairy Sci. 82:589–596.

Oba, M., and M. S. Allen. 1999b. 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.

Oltjen, J. W., and Bolsen K. K. 1980. Wheat, barley, oat and corn silages for growing

steers. J. Anim. Sci. 51:958-965.

Pino, F. and A. J. Heinrichs. 2017. Sorghum forages in precision-fed dairy heifer diets. J.

Dairy Sci. 100:224-235.

34

Sanchez-Duarte, J. I., K. F. Kalscheur, A. D. Garcia, and F. E. Contreras-Govea. 2017.

Performance of dairy cows fed conventional sorghum or corn silages compared to brown midrib

sorghum silage: A meta-analysis. J. Dairy Sci. 100(Suppl. 2):109-110. (Abstr.)

Schneider, B. A., N. A. Clark, R. W. Hemken, and J. H. Vandersall. 1970. Relationship

of pearl millet to milk fat depression in dairy cows. II. Forage organic acids as influenced by soil

nutrients. J. Dairy Sci. 53:305-310.

Sutton, J. D., R. H. Phipps, S. B. Cammell, and D. J. Humphries. 2001. Attempts to

improve the utilization of urea-treated whole-crop wheat by lactating dairy cows. Anim. Sci.

73:137-147.

Vargas-Bello-Pérez, E., A. F. Mustafa, and P. Seguin. 2008. Effects of feeding forage

soybean silage on milk production, nutrient digestion, and ruminal fermentation of lactating Dairy

Cows. J. Dairy Sci. 91:229-235.

Yang, W. Z., and K. A. Beauchemin. 2009. Increasing physically effective fiber content

of dairy cow diets through forage proportion versus forage chop length: Chewing and ruminal

pH. J. Dairy Sci. 92:1603-1615.

Zebeli, Q., J. R. Aschenbach, M. Tafaj, J. Boguhn, B. N. Ametaj, and W. Drochner 2012.

Invited review: Role of physically effective fiber and estimation of dietary fiber adequacy in high-

producing dairy cattle. J. Dairy Sci. 95:1041-1056.

Chapter 3

Using brown midrib 6 dwarf forage sorghum silage and fall grown oat silage

in lactating dairy cow rations.

Journal of Dairy Science Vol. 100, 7, 5250-5265, 2017

https://doi.org/10.3168/jds.2017-12552

M. T. Harper, J. Oh, F. Giallongo, J. C. Lopes, G. W. Roth, and A. N. Hristov

ABSTRACT

Double cropping and increasing crop diversity could improve dairy farm

economic and environmental sustainability. In this experiment, corn silage was partially

replaced with 2 alternative forages, brown midrib-6 brachytic dwarf forage sorghum

(Sorghum bicolor) or fall-grown oat (Avena sativa) silage in the diet of lactating dairy

cows. We investigated the effect on dry matter (DM) intake, milk yield (MY), milk

components and fatty acid profile, apparent total tract nutrient digestibility, N utilization,

enteric methane emissions, and income over feed cost. We analyzed the in situ DM and

neutral detergent fiber disappearance of the alternative forages versus corn silage and

alfalfa haylage. Sorghum was grown in the summer and harvested in the milk stage. Oats

were grown in the fall and harvested in the boot stage. Compared with corn silage,

neutral detergent fiber and acid detergent fiber concentrations were higher in the

alternative forages. Lignin content was highest for sorghum silage and similar for corn

silage and oat silage. The alternative forages had less than 1% starch compared to the

36

approximately 35% starch in the corn silage. Ruminal in situ DM effective degradability

was similar, though statistically different, for corn silage and oat silage, but lower for

sorghum silage. Diets with the alternative forages were fed in a replicated 3 × 3 Latin

square design experiment with 3, 28-d periods and 12 Holstein cows. The control diet

contained 44% (DM basis) corn silage. In the other 2 diets, sorghum or oat silages were

included at 10% of dietary DM, replacing corn silage. Sorghum silage inclusion

decreased DM intake, MY, and milk protein content, but increased milk fat and

maintained energy corrected MY similar to the control. Oat silage had no effect on DM

intake, MY or components compared with the control. The oat silage diet increased

apparent total tract digestibility of dietary nutrients, except starch, whereas the sorghum

diet slightly decreased DM, organic matter, crude protein, and starch digestibility. Cows

consuming the oat silage diet had higher milk urea N and urinary urea N concentrations.

Milk N efficiency was decreased by the sorghum diet. Diet did not affect enteric methane

or carbon dioxide emissions. This study shows that oat silage can partially replace corn

silage at 10% of the diet DM with no effect on MY. Brown midrib sorghum silage

harvested at the milk stage with < 1% starch will decrease DM intake and MY in dairy

cows.

Keywords: dairy cow, forage, oat silage, sorghum silage

37

INTRODUCTION

Forage is the most important feed component on dairy farms and forage shortages

can restrict the number of cows that can profitably be milked on a dairy. A fixed land

base and annual variation in climatic conditions (i.e., rainfall) are often the reason for a

limited amount of forage on farms in the northeastern U.S. Additionally, reliance on a

few forage crop species, such as corn silage, grown continuously may reduce yields

through weeds, pests, and diseases (Vencill et al., 2012; Gentry et al., 2013). Increasing

forage yield by double cropping and improving year-to-year yield stability through crop

rotation strategies using a variety of plant species that reduce pest, disease, and climatic

risk may increase farm sustainability (Faé et al., 2009; Sindelar et al., 2016). Due to its

high concentration of starch, it is difficult to match the energy content of corn silage;

therefore, variety selection (e.g., brown midrib; BMR) and harvest timing (e.g., boot or

soft dough stage) are critical for alternative forage quality as plant OM digestibility can

change rapidly. To be adopted on a large scale, alternative forages must be suitable for

inclusion not only in heifer and dry cow diets, but also in diets for lactating cows because

they consume over 50% of the feed on a dairy farm. Therefore, forages must be highly

digestible to meet the nutrient needs of the modern, high-producing dairy cow.

Sorghum and oat silages are 2 forages that have shown potential as alternative

forages for lactating dairy cows. Sorghum (Sorghum bicolor) is a C4 warm season annual

grass similar to corn silage except that it has a panicle type seed head with smaller grain

kernels, a higher lignin content and greater yields in low moisture conditions (Miron et

al., 2007). Sorghum roots are toxic to western corn root worm (Diabrotica virgifera

38

virgifera) which can reduce pest pressure on corn if used in a crop rotation (Branson et

al., 1969). Brown midrib varieties of sorghum have been developed that have decreased

lignin content and increased NDF digestibility (NDFD) over traditional varieties (Grant

et al., 1995; Oliver et al., 2004). The use of brachytic dwarfing decreases lodging in the

low lignin BMR varieties while increasing the leaf-to-stem ratio. The BMR-6 variant of

forage sorghum has shown NDFD values higher than the BMR-12 variety and equal to

corn silage (Oliver et al., 2004). Yields of sorghum are usually lower than corn, in good

soil with available moisture, but can match or exceed corn yields on marginal ground

particularly in water stressed conditions (Aydin et al., 1999; Abdelhadi and Santini,

2006).

Oats (Avena sativa) are a C3 cool season annual grass that grows well in the

cooler temperatures of the spring and fall as part of a double cropping strategy to increase

annual forage yield per unit area. Earlier studies have not found spring grown oat silage

to be as high quality as corn silage (Burgess et al., 1973; Oltjen and Bolsen, 1980).

However, fall grown oats grow quickly and can be harvested in a highly digestible state

with relatively high CP content of around 18% (Contreras-Govea and Albrecht, 2006).

Oats do not typically survive northeastern winters and must be harvested in the fall if the

goal is inclusion in animal diets. Additionally, oats have the potential to efficiently utilize

fall-applied manure and reduce nitrate leaching (Shephard, 1999; Di and Cameron, 2002;

Carey et al., 2016).

A resilient cropping strategy would include a diverse variety of alternative forages

though corn silage might still yield over 50% of the annual forage harvest. Therefore, the

39

hypothesis of this study was that both BMR-6 brachytic dwarf forage sorghum and fall-

grown oats could serve as alternative forages to feed in addition to corn silage in lactating

dairy cow rations in the northeastern U.S. The objective of the experiment was to

partially replace corn silage with either BMR-6 brachytic dwarf sorghum silage or oat

silage at 10% of the diet DM, to reflect a theoretical proportion of whole farm alternative

forage crop yield, and investigate the effects on DMI, MY, milk components and fatty

acid (FA) profile, nutrient digestibility, N utilization, enteric methane emissions, and

income over feed costs (IOFC) in lactating dairy cows.

MATERIALS AND METHODS

Crops and Silages

Brown midrib-6 brachytic dwarf forage sorghum (Alta AF 7202; King’s

Agriseeds, Ronks, PA) and oats (ForagePlus; Seedway, Hall, NY) were grown in Centre

County, PA at approximately 40° N latitude on Hagerstown and Hublersburg soils during

the summer and fall of 2014. Both crops were planted with a no-till drill (John Deere

1590; Moline, IL) into fields fertilized with 44.8 t/ha of dairy manure prior to planting,

contributing 42 kg/ha of ammonium N. Sorghum was planted with 38 cm row spacing

and oats were planted with 19 cm row spacing. A John Deere 946 mower with a roll

conditioner was used to mow both crops and, after wilting to around 30% DM, the

forages were gathered and chopped using a John Deere 6750 harvester. Both crops were

ensiled without inoculant in 3-m diameter plastic silage bags (Up North Plastics, Cottage

40

Grove, MN). Sorghum was planted on June 30th, 2014 after barley and triticale harvested

for forage, at a seeding rate of 7.3 kg/ha and fertilized with 67 kg of N/ha from a 30%

urea and ammonium nitrate liquid fertilizer on August 18th, 2014. It was mowed on

November 10th, 2014 at the milk stage of grain development after being partially frost-

killed and harvested on November 11th, 2014 with a 16 mm theoretical chop length. Oats

were planted at a seeding rate of 108 kg/ha on August 16th, 2014 after wheat harvested

for grain. The oats were mowed in the boot stage on November 8th, 2014 and harvested

on November 14th, 2014 with a 12 mm theoretical chop length. The corn silage, which

was the control in this experiment, was a mixture of the following hybrids: Mycogen

TMF2R737 (112 d relative maturity), Dekalb DKC 52-61 (102 d relative maturity), and

NK N60F-3111 (107 d relative maturity). Corn silage was grown in Centre County, PA at

approximately 40° N latitude on Hagerstown and Hublersburg soils and planted between

May 1st and May 10th, 2014 at a rate of 79,000 seeds/ha. It was planted with a no-till drill

(John Deere 1590; Moline, IL) into fields fertilized with 44.8 t/ha of dairy manure prior

to planting, contributing 42 kg/ha of ammonium N. An additional 43 kg/ha of N was

applied as 30% urea and ammonium nitrate liquid prior to planting and 100 kg/ha of N in

the same form as a sidedress application. Corn silage harvest was conducted between

September 15th and September 30th at a target DM of 38% with a 19 mm chop length and

ensiled in an upright concrete silo.

41

Animals and Diets

All animals were cared for according to procedures approved by The

Pennsylvania State University’s Institutional Animal Care and Use Committee. Twelve

mid-lactation Holstein dairy cows, 6 primiparous (MY 37 ± 2.6 kg; DIM 100 ± 6 d; BW

592 ± 51 kg) and 6 multiparous (MY 47 ± 5.8 kg; 2.3 ± 0.5 lactations; DIM 61 ± 16 d;

BW 639 ± 39 kg at the beginning of the experiment) were used in a replicated 3 × 3 Latin

square design balanced for residual effects. Each 28 d period consisted of 18 d of

adaptation and 10 d of data and sample collection. Cows were placed in 4 groups of 3

cows each based on DIM, MY, and parity. Cows within a group were randomly assigned

to 1 of 3 diets, as described below. All cows were housed in the tie stall barn of The

Pennsylvania State University’s Dairy Research and Teaching Center. Diets were mixed

and fed from a Rissler model 1050 TMR mixer (I.H. Rissler Mfg. LLC, Mohnton, PA).

Cows were fed once daily around 8 a.m. to yield approximately 5-10% refusals. Feed was

pushed up 3 times throughout the day. The cows were milked twice daily at 7 a.m. and 6

p.m.

Three different diets, as in Table 1, were fed to the cows during the experiment as

follows: a control diet (CS), based on corn silage and alfalfa haylage; an oat silage diet

(OS), oat silage included at 10% of dietary DM, replacing 22.7% of the control diet corn

silage DM; and a sorghum silage diet (SS), sorghum silage included at 10% of dietary

DM, replacing 22.7% of the control diet corn silage DM. The CS diet was formulated to

meet or exceed the NRC (2001) requirements for NEL and MP of a cow with 650 kg BW,

44 kg/d MY, 3.8% fat, 3.2% true protein, and at 27 kg/d DMI.

42

Sampling and Analyses

Refusals were collected and weighed individually for each cow just prior to the

morning feeding to measure daily as-fed intake. Total mixed ration, refusals, and forage

(sorghum, oat, alfalfa, and corn silages) samples were collected twice weekly,

composited by week and diet (i.e., silage type), stored at -20°C, and then oven dried at

55°C for 72 h. At least 2 separate TMR samples were collected during each period and

processed individually for particle size analysis using the Penn State Particle Separator

with 19, 8 and 4 mm sieves. The procedure described in the extension publication by

Heinrichs (2013) was used for analysis. All TMR samples were collected within 1 h of

feeding. The weekly DM content of the TMR and refusals was used to calculate the

individual daily DMI. Concentrate feeds were sampled weekly and stored at -20°C until

analysis. Sorghum and oat silages were first ground through a 4-mm screen (for in situ

disappearance measurements), then, along with alfalfa haylage, corn silage, and TMR

samples, were ground through a 1-mm screen in a Wiley mill (Thomas Scientific,

Swedesboro, NJ) and further composited by period on an equal weight basis. Dried

composite samples of sorghum, oat, and corn silages were sent to Cumberland Valley

Analytical Services Inc. (Maugansville, MD) to be analyzed by wet chemistry methods

for amylase-treated NDF (Van Soest et al., 1991), ADF (973.18; AOAC International,

2000), lignin (Goering, H.K. and P.J. Van Soest., 1970), fat (2003.05; AOAC

International, 2006), CP (990.03; AOAC International, 2000), soluble protein

(Krishnamoorthy et al., 1982), starch (Hall, 2009), ethanol soluble carbohydrates (Dubois

et al., 1956), ash (942.05; AOAC International, 2000), and minerals (985.01; AOAC

43

International, 2000). Fermentation profiles of the corn, oat and sorghum silages were

analyzed by Cumberland Valley Analytical Services Inc. using near infrared reflectance

spectroscopy for lactic, acetic and butyric acids, titratable acidity, and pH. Concentrate

feed samples were ground and composited once for the entire experiment. Dried

composite concentrate ingredients were analyzed by Cumberland Valley Analytical

Services Inc. by wet chemistry methods for CP, amylase-treated NDF, ADF, fat, CP,

starch, ash, and minerals, (procedures as referenced above) and calculated NFC using the

equation, NFC = 1 – CP – fat – NDF – ash and NEL using the equation, NEL = 0.0245 ×

TDN - 0.12. Concentrations of CP, NDF, ADF, NFC, NEL, starch, fat, ash, Ca and P in

the TMR were calculated based on the individual feed ingredient values and their

inclusion levels in the TMR. The diet values of RDP, RUP and NEL balance were

calculated based on NRC (2001) at actual DMI, MY, BW, and milk composition of the

cows.

Milk weights were automatically recorded at each milking using the Afimilk

system (Kibbutz Afikim, Israel). Milk samples for components and FA analysis were

collected on 2 consecutive days (4 consecutive milkings) during wk 4 of each period

from the p.m. and a.m. milkings. Milk component samples were collected into tubes

containing 2-bromo-2-nitropropane-1,3-diol and analyzed individually by Dairy One

Laboratory (Ithaca, NY) for fat, true protein, MUN, and lactose content using infrared

spectroscopy (Milkoscan 4000; Foss Electric, Hillerød, Denmark). Milk samples for FA

analysis from the 4 milkings for each period and cow were collected without preservative

and frozen at -20°C until composited based on MY so that a single composited sample

44

was analyzed per cow and per period following the procedure described by Rico and

Harvatine (2013). Body weight was recorded daily upon exiting the milking parlor using

an AfiFarm 3.04E scale system (S.A.E. Afikim, Rehovot, Israel) for periods 1 and 2.

During period 3, BW was not measured due to a scale system malfunction.

During wk 4 of each period, urine and fecal samples were collected for

digestibility and N utilization estimates. Spot urine and fecal samples (approximately 300

mL and 500 g per sample, respectively) were collected 8 times over 3 d at (d 1) 5 a.m., 12

p.m., 6 p.m., (d 2) 12 a.m., 9 a.m., 3 p.m., 9 p.m., and (d 3) 3 a.m. to obtain a

representative sample of a 24 h period. A full description of the urine and fecal sample

processing and analyzing can be found in Lee et al. (2012). Briefly, raw urine from each

sampling was acidified, diluted, composited by cow and period and frozen at -20°C for

later analysis of allantoin, uric acid, creatinine, urea N and total N. Allantoin was

analyzed following the procedure by Chen et al. (1992). Stanbio Laboratory (Boerne,

TX) kits were used to analyze uric acid (Uric Acid Kit 1045), creatinine (Creatinine Kit

420), and urea N (Urea Nitrogen Kit 580). Total N was analyzed in freeze dried urine

samples of 1:10 diluted and acidified urine using a Costech ECS 4010 C/N/S elemental

analyzer (Costech Analytical Technologies Inc., Valencia, CA). Fecal samples were oven

dried at 65°C, ground through 1-mm screen in a Wiley mill and analyzed for DM, OM,

CP, starch, NDF and ADF. A Mixer Mill MM 200 (Retsch GmbH, Haan, Germany) was

used to pulverize a 0.5 g aliquot of fecal sample for CP analysis (N × 6.25) using the

Costech ECS 4010 C/N/S elemental analyzer. Starch analysis of fecal DM for apparent

total tract digestibility was performed using a procedure based on the method which

http://www.sciencedirect.com/science/article/pii/S0022030212004857#bib0050

45

included acetate buffer described by Hall (2009). Briefly, starch was gelatinized with

50% NaOH, incubated for 16 h at 55°C with acetate buffer and amylase, centrifuged,

plated on a 96-well plate, and then reacted with a PGO (Glucose Oxidase/Peroxidase)

enzyme solution (P7119; Sigma-Aldrich, Saint Louis, MO) for 45 min before being read

at 450 nm. Neutral detergent fiber and ADF were analyzed with an Ankom 200 fiber

analyzer (Ankom Technology Corp., Macedon, NY) based on the procedures of Van

Soest et al. (1991) with α-amylase and sodium sulfite in the NDF analysis. A 10-d

ruminal incubation was used to determine the indigestible NDF (iNDF; Huhtanen et al.,

1994 as modified by Lee et al., 2012) of both feces and TMR, which was used as a

marker to estimate apparent total-tract digestibilities of dietary nutrients.

Enteric CH4 and CO2 emissions were measured during wk 4 of each period with

the GreenFeed system (C-Lock Inc., Rapid City, SD). Measurements were collected 8

times over 3 d at 9 a.m., 3 p.m., 9 p.m., 3 a.m., 12 p.m., 6 p.m., 12 a.m. and 5 a.m. to

obtain a representative sample of a 24 h period. Gas sampling procedures followed those

recommended by Hristov et al. (2015). Gas measurements of at least 2 minutes in length

per sampling were used in the final statistical analysis which occurred for 81.3% of the

measurements. Gas emission data were averaged by cow and period for the statistical

analysis.

In Situ

Ruminal disappearance of DM and NDF from the sorghum and oat silages, and

separate alfalfa haylage and corn silage samples, was determined in situ. The sorghum

46

and oat silage samples were from the current experiment whereas the corn silage and

alfalfa haylage forage samples were from a similar experiment conducted 4 mo after the

current experiment (Harper et al., 2017, companion study). Corn silage was 38.5% DM

and contained (DM basis) 6.4% CP, 46.6% NFC, 34.5% starch, and 41.0% NDF. Alfalfa

haylage was 46.0% DM and contained (DM basis) 21.0% CP, 24.0% NFC, and 44.2%

NDF. The nutrient composition of the sorghum and oat silages is in Table 2. Six

ruminally cannulated lactating Holstein cows averaging: DMI 24.4 ± 2.4 kg; MY 36.8 ±

2.9 kg; 2.2 ± 0.4 lactations; DIM 148 ± 10 d; BW 616 ± 40.7 kg were used for the in situ

incubations. Cows were fed (% DM basis): corn silage 38.33, alfalfa haylage 13.83, grass

hay and straw mixture 4.17, ground corn 9.58, canola meal 9.58, cookie meal 5.33,

roasted soybeans 5.00, molasses 5.00, whole cotton seed 4.58, cracked corn 2.50, and

mineral mix 2.10. Oven dried forages were ground through a 4-mm sieve in a Wiley mill.

Approximately 7 g of sample were weighed into 10 × 20 cm nylon bags with 50-µm

porosity (Ankom Technology Corp., Macedonia, NY) and closed with a ziptie after

folding. Triplicate bags were sequentially incubated in each cow for 12, 24, 48, 72, and

96 h and simultaneously removed. Two bags per forage were made for the 0 h time point

and processed as the incubated samples except the rumen incubation step. Upon removal

from the rumen, bags were rinsed 3 times with cold water in a washing machine set to

agitate for 6 min each rinse. The zipties were cut off and any remaining particles rinsed

off with cold tap water. Rinsed bags were then oven dried for 72 h at 55° C before

weighing for DM determination. Samples were composited by silage, time point, and

cow before NDF analysis as previously described. Ruminal disappearance was calculated

based on initial dry weight of the incubated sample, residue dry weight, and NDF

47

concentration of initial sample and bag residue. Degradation curves were fit to the

equation p = a + b (1 - e-ct) where p is the degraded fraction (of DM or NDF) at time t,

constant a is the soluble fraction, b is the potentially degradable fraction and c is the rate

of degradation of the b fraction (Ørskov and McDonald, 1979). The effective

degradability (ED) was determined with the following equation (Ørskov and McDonald,

1979): ED = a + b {c ÷ (c + k)}, where k is the rate of passage assumed to be 0.03/h.

Corn silage NDF degradability did not fit the Ørskov and McDonald model and so was fit

with a linear model. Therefore, corn silage NDF degradability was not statistically

compared to the other forages.

Income Over Feed Costs

Income over feed costs for the 3 diets was calculated using the Pennsylvania State

Extension Dairy Team IOFC Tool (Penn State Extension, 2015). The cash flow

spreadsheet from the Pennsylvania State Extension Dairy Team (Penn State Extension,

2016) was used to calculate forage monetary values for the IOFC tool. The model dairy

included 34.4 ha cropland, 65 lactating cows, 10 dry cows, 52 heifers, and 12 calves. It

was assumed that only the forages were grown on the farm whereas concentrates were

purchased. The lactating cow ration was changed in the scenarios to reflect the treatment

diet whereas diets for other cow groups (e.g., dry cows, heifers, and calves) were kept the

same among scenarios. First, the total amount of the different forages required for each

scenario was calculated. Next, the hectares needed to produce that amount was found by

dividing the total amount of each crop needed by the per hectare crop yields obtained for

48

the forages used in the trial. The corn silage yield when double cropped with oats was

decreased by 4.9 t DM/ha to account for the lower yield of short season corn, which

would have to be planted before oats. An additional scenario was run where a sorghum

yield of 13.4 t DM/ha was used to show a more typical yield based on timely planting.

Then, the variable costs of seed, fertilizer, and herbicide per acre for each crop during

2015 was entered into the spreadsheet. Along with the input costs and the yield

information for each crop, the fixed costs were allocated among the forages based on the

labor used to produce them to determine price per ton. Milk and components yield from

the current study was used with the average milk pricing in PA for 2015 to generate the

income side of the IOFC equation.

Statistical Analysis

Statistical analyses for all but the in situ data were run using the MIXED

procedure of SAS v9.4 (SAS Institute Inc., Cary, NC). Cow was the experimental unit.

Milk yield and DMI from the last 10 d of the experiment were analyzed with day as a

repeated measure. The statistical model included cow, block, day, period, diet, and period

× diet and diet × day interactions. Block and cow within block were random effects with

all others fixed. Milk composition and FA, nutrient intake, digestibility, N utilization,

methane and CO2 emissions data used the same model without day and diet × day

interaction. Milk composition data were weighted averages based on the MY at each

milking. Particle size distribution of the 3 TMR was analyzed by sieve size with the

MIXED procedure including diet, period, and diet × period interaction in the model.

49

Silage nutrient composition was compared using the MIXED procedure with silage type

as the model. Significance was declared at P ≤ 0.05 and tendency was declared at 0.05 <

P ≤ 0.10. If not indicated otherwise, data are presented as least squares means.

Ruminal in situ degradation of DM and NDF was analyzed using the NLMIXED

procedure of SAS. The overall regression curve and the individual parameters (a, b, c,

and ED) were contrasted among forages and significance was declared at P ≤ 0.05.

RESULTS AND DISCUSSION

Differences among diets (Table 1) resulted from compositional differences in oat,

sorghum, and corn silages (Table 2). As corn silage was replaced on a weight basis, the

diets were not isonitrogenous or isocaloric. As indicated earlier, the control, CS diet was

formulated before the study to meet or exceed NRC (2001) requirements of the cows at

the beginning of the experiment. Cows, however, produced less milk with a lower true

protein content during the experiment. Therefore, MP supply exceeded requirements for

all diets. Crude protein concentration was, numerically, highest for OS, intermediate for

SS, and lowest for CS, following the statistical ranking of CP concentration in oat,

sorghum, and corn silages. Soluble CP was also higher (P < 0.001) in the oat and

sorghum silages than the corn silage. Neutral detergent fiber and ADF content

numerically increased from CS to OS to SS. Starch was numerically lower in OS and SS

compared with CS, but NEL balance was similar and positive for all diets.

50

In the current study, the intention was to harvest the sorghum silage at the soft

dough stage after starch was deposited in the grain. However, a late planting prevented

adequate crop development before a killing frost, which forced a harvest at the milk stage

before starch deposition. This resulted in a lower than expected yield of 8.13 t DM/ha and

almost nonexistent starch concentrations in the sorghum silage. In contrast, Oliver et al.

(2004) reported a yield of 9.7 t DM/ha with 16.8% starch for BMR-6 sorghum harvested

at a late dough stage. This highlights the importance of management, particularly prompt

planting, in the use of alternative forages. Oats yielded 4.79 t DM/ha and corn silage

varieties had an average yield of 18 t DM/ha. The DM content of all silages was between

30 and 40%. Oat and sorghum silages had pH of around 4.5, which is typical of grass

silages. Titratable acidity matched the levels of total acid in both silages. Lactic acid was

low in the sorghum silage. Lactic and acetic acids in the oat silage fell within normal

values according to the Dairy One Interactive Feed Composition Library

(http://dairyone.com/analytical-services/feed-and-forage/feed-composition-

library/interactive-feed-composition-library/; accessed February 28, 2017). Butyric acid

was detected above 0.5% in the oat silage possibly indicating an extended fermentation

process partly due to wetter harvested material. Drying oats is a challenge in cool fall

weather. Both sorghum and oat silages contained starch concentrations below 1%. Oat

silage had higher (P < 0.001) K and sorghum silage had a higher concentration (P = 0.02)

of ethanol soluble carbohydrates (i.e., sugars) than the other silages.

Dry matter intake was not different between CS and OS diets; however, DMI was

decreased (P = 0.02) in the SS diet. In this experiment, we substituted part of the ration

51

corn silage with 1 of 2 low-starch forages which did not contain developed grain. Corn

silage routinely contains starch concentrations above 30% and could be considered part

concentrate and part forage (Boivin et al., 2013). Dry matter intake of high producing

dairy cows can be limited by rumen fill, which may present a challenge in meeting

energy demands (Allen, 2000). Rumen fill is also positively related to NDF content of

the diet (Allen, 2000). Nichols et al. (1998) demonstrated that increasing dietary NDF

concentrations for dairy cows decreased DMI. Oltjen and Bolsen (1980) showed that

increasing diet ADF content in growing steers decreased DMI. Diets with lower

concentrations of digestible fiber cause rumen fill at lower DMI than diets with higher

concentrations of digestible fiber (Allen, 2000; Mertens, 2009).

The sorghum silage in this experiment had greater NDF, ADF, and lignin content

than the corn silage which was also observed in other studies comparing BMR sorghum

with corn silage (Oliver et al., 2004; Bernard and Tao, 2005). The higher concentration of

less digestible fiber in the sorghum silage likely explains the lower DMI of the SS diet,

resulting from slower digestion and passage rate and increased ruminal fill. The OS diet,

likewise, was higher in NDF and ADF than the CS diet, but contained less lignin and had

more digestible NDF. The increased digestibility of the oat silage NDF versus corn silage

NDF potentially limited the effects of ruminal fill in OS. This resulted in similar DMI

between OS and CS. However, OS resulted only in a slight numerical increase in intake

NDF as a percent of BW over SS. Furthermore, OS and SS had the same intake of forage

NDF as a percent of BW (1.08%) at different DMI. This agrees with Mertens (2009),

who proposed that NDF content was the major intake limiting factor at high DMI with

52

NDFD altering intake regulation to a lesser extent. Therefore, the difference in NDF

concentration between oat silage and sorghum silage may have contributed more to the

change in DMI than the difference in NDFD.

The ruminal in situ results (Table 3) compare the alternative forages of this study

with corn silage and alfalfa haylage from another study, thus direct conclusions cannot be

made regarding the corn silage and alfalfa haylage from the current experiment.

However, the nutrient profiles of the 2 sets of conventional forages were similar, and the

results seem to support the argument that sorghum silage caused rumen fill limitation of

DMI at lower intakes partly because of its lower degradability. In contrast, oat silage

intake may have been regulated by energy demand since its greater degradability should

have allowed a higher NDF intake, though as discussed above, the NDF intakes for OS

and SS were not different. Dry matter in situ disappearance regression curves for alfalfa

haylage, corn silage, oat silage, and sorghum silage are in Figure 1. The DM

disappearance (i.e., degradability) curve differed (P < 0.001) among forages (Fig. 1).

Corn silage had the highest (P < 0.01) percent soluble DM, caused by its high starch

content, followed by oat silage, alfalfa haylage, and sorghum silage (Table 3). The

potentially degradable fraction of DM (b) was higher (P ≤ 0.002) for oat and sorghum

silage than for alfalfa haylage (Table 3). Alfalfa haylage had the highest (P < 0.001) rate

of degradation of the b fraction followed by oat silage, sorghum silage, and corn silage

(Table 3). Corn silage also had the highest (P = 0.002) ED for DM followed closely by

oat silage, then alfalfa haylage, and lastly, sorghum silage (Table 3). The oat silage had a

more rapid rate of DM degradation than corn silage and similar ED, which may have

53

caused similar rumen fill and DMI. The lower rumen disappearance of sorghum silage

DM, may have led more rapidly to rumen fill and lower DMI.

Neutral detergent fiber in situ disappearance curves for alfalfa haylage, oat silage,

sorghum silage, and corn silage is shown in Figure 2. Alfalfa had the highest (P < 0.001)

soluble NDF fraction, compared with oat and sorghum silages (Table 3). The highest (P <

0.001) potentially degradable NDF fraction was in the oat silage followed by sorghum

silage, and was lowest in the alfalfa haylage (Table 3). There was a tendency (P = 0.06)

for alfalfa haylage to have a higher degradation rate than sorghum silage (Table 3).

Effective degradability of NDF was higher (P < 0.001) in oat silage than either alfalfa

haylage or sorghum silage (Table 3). The higher disappearance of oat silage NDF

resulted in higher ED of NDF compared with the sorghum silage, which was reflected in

higher DMI for OS compared with SS.

The increased degradability of the oat silage may have come from the delayed

maturity related to their fall growth which has been shown to have 38% greater NDFD

than spring-grown oats (Contreras-Govea and Albrecht, 2006). Additionally, the fall

weather provided cooler growing temperatures than temperatures during a spring growing

season. Forage grown at cooler temperatures deposit less lignin and have increased in

vitro NDFD compared with the same forages grown at higher temperatures (Buxton,

1996).

Ration particle size can also affect DMI intake, but responses are not always

consistent (Allen, 2000; Bhandari et al., 2008). Diets only had a slightly, but significantly

(P = 0.02) different particle size distribution on the top, 19 cm, sieve as measured by the

54

Penn State particle separator. Reported on an as-fed basis, OS had more long particles

(10.6%) than CS (7.8%). The SS diet (9.1%) did not differ from the other 2 diets. The

other screens averaged 43.6, 17.5, and 29.7% (8 mm, 4 mm, and bottom pan,

respectively) of the sample. The particle distribution data are in general agreement with

the recommendations of Heinrichs (2013). Forage particle buoyancy and specific gravity

can affect rumen fill and DMI (Allen, 2000), but this did not appear to be the case in the

current study.

Complete MY data are in Table 4. Milk yield was similar for CS and OS, but

decreased (P = 0.006) for SS. Feed efficiency was not affected by diet. Dry matter intake

directly affects MY and is the likely reason for the observed MY differences in the

current study. Lusk et al. (1984) reported no difference in MY in 2 experiments when

BMR-12 sorghum silage completely replaced corn silage in the ration. Milk yield was

lower (around 25 kg/d) in that study and, in contrast to the current study, they did not see

a decrease in DMI. Additionally, the sorghum silage used by Lusk et al. (1984) had lignin

and NDF concentrations similar to the corn silage, whereas in the current study, the

sorghum silage had considerably higher lignin and NDF content than the corn silage.

Aydin et al. (1999) reported decreased DMI and MY when comparing corn silage versus

BMR sorghum diets that were 6%-units different in NDF. This agrees with our finding

comparing corn silage to BMR sorghum diets with 2%-units difference. Contrastingly,

Aydin et al. (1999), in a second experiment, reported no decrease in DMI or MY when

comparing corn silage versus BMR sorghum diets with equal NDF concentration. Grant

et al. (1995) and Oliver et al. (2004) both found no difference in DMI or MY between

55

BMR sorghum silage and corn silage diets. Their BMR sorghum silages, however, were

harvested in the dough stage and contained significant amounts of starch whereas ours

had less than 1% starch. Miron et al. (2007) reported similar DMI and MY between diets

containing BMR sorghum silage or corn silage but, unlike the current study, corn grain

was added to the sorghum silage diet to increase dietary starch. Colombini et al. (2015)

replaced corn silage with sorghum sudangrass silage plus corn grain and, similar to our

results, reported a tendency for a reduction in DMI along with a significant decrease in

MY. They added grain to the sorghum diet to match the starch concentrations in the corn

silage diet, but the sorghum was not a BMR variety. In general, differences in MY

responses to sorghum silage between the current and previous studies can largely be

explained by the relatively high MY in the current study and the replacement of corn

silage by low starch sorghum silage without additional energy supplementation.

The OS diet had no effect on DMI or MY in the current experiment. Earlier

studies reported poor results of feeding oat silage to cattle (Christensen et al., 1977;

Khorasani et al., 1993; McCartney and Vaage, 1994). Oat silage harvested in the early

dough stage was compared with corn silage and barley silage by Burgess et al. (1973) in

dairy cows. These authors reported an increased DMI per unit of BW of the oat silage

over the other treatments. However, total DMI (grain supplement plus silage) was not

different in that experiment and MY was lowest for the oat silage diet. Oltjen and Bolsen

(1980) found decreased DMI when comparing oat to corn silage in an 84% silage, 16%

supplement diet fed to growing steers. Both of those studies used dough stage spring

grown oats which have more ADF than the boot stage fall grown oats used in the current

56

experiment. More recently, TMR containing oat silage has been shown to successfully

support MY above 35 kg/d (Leonardi et al., 2005; Bhandari et al., 2008). Milk yield of

around 39 kg/d was reported when oat silage replaced alfalfa haylage at a 25% inclusion

rate in a corn silage-based diet (Leonardi et al., 2005). Their oat silage had similar

nutrient composition to the oat silage used in the current trial. Bhandari et al. (2008),

investigating the effect of silage particle length, used milk stage oat silage at 24% of the

diet DM along with 24% alfalfa haylage and reported MY of 36 kg/d. Milk yield and

intake data from the current study for OS agree with these more recent reports.

Compared with CS, milk fat concentration was not altered by OS but was

increased (P = 0.02) by SS. Milk fat yield did not differ among diets. Milk true protein

concentration was decreased (P = 0.03) by SS compared with OS or CS. Yield of milk

true protein followed the same pattern. Oliver et al. (2004) did not report a change in

either milk fat or protein concentration between BMR-6 sorghum and corn silage-based

diets likely due to the similarly high NDF (> 38% DM) and moderate starch (≤ 21% DM)

concentrations in their diets. The effect of SS on milk components is in agreement with

Miron et al. (2007) who reported an increase in milk fat content and a decrease in protein

content for both traditional hybrid sorghum and BMR sorghum when replacing corn

silage. Their sorghum diets had higher NDF concentrations than the corn silage diet as

was the case in the current study. Bernard and Tao (2015) also observed an increase in

milk fat concentration of a forage sorghum silage based diet versus corn silage, but there

was no change in milk protein. Once again there was a difference in the fiber contents of

the diet, which explained the increased milk fat content (Bernard and Tao, 2015). In that

57

study, additional ground corn and soybean meal added to the sorghum diets may have

removed differences in milk protein content whereas in our experiment no additional

concentrates were fed.

The increase in milk fat concentration in SS is likely multifactorial. Miron et al.

(2007) suggested that body fat mobilization could help explain the high milk fat content

observed in their sorghum treatments. Our milk FA data (Table 5) suggest that body fat

mobilization may have played a role in the increased milk fat content of SS in the current

study. We observed a decrease (P = 0.03) in 12:0, a de novo synthesized FA, for OS and

SS and an increase (P = 0.01) in 18:0, a preformed FA that originates either from the diet

or from lipolysis of body fat reserves (Palmquist et al., 1993). Supplemental sugar has

been shown to quadratically increase milk fat concentration with optimums of 4.8 and

6.3% total sugar in the diet in 2 studies by Broderick and Radloff (2004). The higher

ethanol soluble carbohydrate concentration in the sorghum silage versus corn silage in the

current study may have contributed to the increased milk fat, though the increase in sugar

was only around 0.25% of the diet DM due to the sorghum silage, so the effect may have

been small. Razzaghi et al. (2016) reported higher fat content and lower total milk trans-

18:1 FA with a sucrose treatment. It is known that increases in trans-10 18:1 milk FA are

positively related to milk fat depression for cows fed a low-fiber high-oil diet (Rico and

Harvatine, 2013). The SS diet had lower trans-9 18:1 and trans-10 18:1 FA (P ≤ 0.01)

which would seem to indicate a rumen environment that was more supportive of milk fat

production, but the same FA effect was observed for OS, which resulted in no change in

milk fat concentration. Along with these factors, lower milk fat concentration in CS and

58

OS may have been partially a result of the so-called dilution effect due to higher MY for

these 2 treatments compared with SS. Based on this line of thought, protein should also

have been concentrated, but we did not observe this effect for SS. Milk protein content is

elevated by increased energy intake (Emery, 1978) and milk protein concentration may

have been decreased by SS because of lower dietary digestible energy intake resulting

from lower DMI with less starch and higher lignin in the sorghum silage. This would also

explain why numerically less lactose was produced in SS. Less available energy for

ruminal microbes decreases microbial protein synthesis. Though there is only a tendency

(P = 0.06) for decreased urinary uric acid excretion and numerically lower allantoin

excretion in SS to provide support for this hypothesis. The high digestibility of the oat

silage may have provided similar amounts of energy as the corn silage to support

microbial protein and milk protein synthesis. Lactose concentration and yield, MUN, and

ECM yield were not statistically affected by diet.

Several individual milk FA were statistically affected by diet but when grouped as

saturated, mono-unsaturated, poly-unsaturated, total trans FA or odd- and branch-

chained FA no effects were observed. Odd- and branch- chained milk FA are positively

related to microbial flow to the duodenum (Vlaeminck et al., 2006) and no differences

among diets for this parameter matches the lack of effect observed for purine derivatives.

There was an increase (P <0.001) in iso 15:0 for OS and SS which has been positively

related with rumen acetate concentrations (Fievez et al., 2012). Milk FA 15:0 was

decreased (P = 0.005) in SS indicating increased proportions of rumen butyrate and

acetate and decreased propionate based on the relationships put forth by Fievez et al.

59

(2012). These rumen VFA changes are expected with increased dietary NDF and

decreased starch such as in OS and SS.

Nutrient intake and digestibility data are in Table 6. The intake data are a

reflection of the differences in DMI and nutrient content of the oat, sorghum and corn

silages. Crude protein intake was greater (P = 0.04) for OS compared with CS or SS.

Intakes of NDF and ADF were higher (P ≤ 0.02) for both OS and SS versus CS.

Conversely, starch intake was highest (P < 0.001) for CS, intermediate for OS and lowest

for SS. Digestibility of DM, OM, CP, NDF, and ADF were all increased (P < 0.001) by

OS compared with CS and SS. The largest increases were observed in NDF and ADF

digestibilities. The SS diet resulted in lower (P < 0.001) DM, OM, and CP digestibilities

than the other 2 diets, but fiber digestibility equal to CS. Higher lignin concentration in

SS from the sorghum silage and lower lignin concentration in OS from the oat silage

compared with CS can partially explain these effects as lignin concentration is negatively

correlated to forage digestibility (Mertens, 1985). This is supported by results from 2

experiments by Aydin et al. (1999) reporting on digestibility of BMR sorghum,

traditional sorghum and corn silage in lactating dairy cows. In their first experiment,

sorghum silages had higher lignin content and decreased total tract ADF digestibility. In

their second experiment, lignin content was similar between BMR sorghum and corn

silage and the 2 forages had similar in vitro potentially digestible NDF and 30 h NDFD.

It is known that fiber digestion is negatively affected by higher starch

concentrations causing a reduction in rumen pH (Firkins, 1997; Lechartier and Peyraud,

2011). In the current study, both OS and SS had lower starch concentrations, compared

60

with CS. This may have promoted fiber degradation in the rumen because the decrease in

NFC (2.5 to 2.8%-units) was in the range reported to have a large digestibility response

(Sarwar et al., 1992). Increased fiber digestion from lower dietary starch concentration

and lower DMI with slower passage rate in SS may have been counter-balanced by

decreased DM digestibility due to higher lignin content from the sorghum silage. For the

OS diet, lower starch and lower lignin concentrations supported higher fiber digestion.

Starch is more digestible than NDF, which may explain the decrease in DM and OM

apparent digestibility for the SS diet, but does not explain the increase in DM and OM

apparent digestibility for the OS diet. It should be noted that total tract apparent

digestibility was estimated using iNDF as an internal marker. Though significant, DM,

OM, and CP apparent digestibilities are less than 2%-units different from CS and reflect

differences in the TMR iNDF (iNDF, % ± SD: CS, 12.0 ± 0.45; OS 11.2 ± 0.37; SS 12.8

± 0.54). The higher CP apparent digestibility in the OS diet could have resulted from

increased protein solubility (and perhaps ammonia concentration, which was not

analyzed in this study) in the oat silage. The CS diet had the highest (P < 0.001) starch

digestibility followed by OS and SS. It is possible that starch in the corn silage was more

digestible than starch in the ground corn which was the other major starch source in the

diets. Indeed, Lanzas et al. (2007) reported a higher ruminal starch degradation rate for

processed corn silage (0.32/h) compared to finely ground corn (0.15/h). The CS diet in

the current experiment contained proportionately more starch from corn silage than OS or

SS.

61

Nitrogen utilization data is in Table 7. The OS diet resulted in higher (P = 0.04)

intake of N than either SS or CS because of the higher CP content of the oat silage. There

was a tendency for higher (P = 0.08) total excreta N for OS compared with CS, which

reflects the higher N intake and similar milk N secretion. Compared with CS, OS

increased (P = 0.03) urinary urea N excretion. The higher N intake and higher soluble

protein in the oat silage was the likely reason for increased urinary urea N excretion with

OS compared with the other diets (Van Soest, 1994; Broderick, 2003). The sorghum

silage had a higher CP content than corn silage, but, with the decrease in DMI, N intake

for SS was not different from CS. There was a tendency (P = 0.07) for increased urinary

urea N in SS versus CS, which may be explained with lower energy availability in the

rumen with the former diet. Milk N secretion and use efficiency were decreased (P =

0.05) by SS compared with CS due to the decreased milk and milk protein yields with SS

discussed earlier. As a percent of intake, urine N and total excreta N were unaffected by

diet, however, fecal N excretion was highest (P < 0.001) for SS, intermediate for CS, and

lowest for OS. This agrees with the lower and higher total tract apparent CP digestibility

of these diets, respectively.

Enteric emissions of CO2 and CH4 in Table 7 were not different among diets.

Methane emissions were not affected when presented as yield (i.e., per kg of DMI) or

intensity (i.e., per kg of MY). It has been suggested by Hristov et al. (2013) that

increasing forage digestibility is expected to decrease enteric CH4 emission intensity and

is 1 of the promising mitigation strategies. Hart et al. (2009) fed low and high digestible

grass to cows and measured CH4 emissions. They reported higher CH4 emissions in cows

62

fed higher digestible grass but emission intensity was lower. We did not observe an effect

on CO2 or CH4 emissions by diet despite the differences in apparent digestibility. The

likely explanation for this lack of effect is the small magnitude of differences in OM

digestibility, DMI, and MY. Dry matter intake is by far the greatest driver of enteric CH4

emission in ruminants and only differed by 5% in this study. Colombini et al. (2015)

similarly reported only tendencies of higher CH4 emissions when feeding sorghum silage

in place of corn silage to lactating dairy cows. In their study, DMI differed by 9%. Higher

DMI in the current study explains why daily CH4 emissions results were higher than

those reported by Colombini et al. (2015; 523 versus 342 g/d, respectively), yet emission

intensity was lower, likely due to higher MY in the current experiment.

The economic outcome of the use of alternative forages is critical for their

adoption. The IOFC of CS and OS were comparable at $9.49 and $9.43/cow/d,

respectively (data not in tables). The SS diet resulted in slightly lower IOFC,

$9.32/cow/d. A disadvantage in double cropping fall oats in central Pennsylvania is that

they must be planted in mid-August to yield well. To plant at that time, a short season

corn (< 85 d relative maturity) must be used. Short season corn usually has a decreased

yield compared with longer season varieties and this raises corn crop production costs.

The SS diet had the lowest IOFC due to a lower MY and a lower BMR sorghum crop

yield even though input costs were lower. When we ran the IOFC analysis with a 65

milking cow dairy, we had to account for rental costs of additional arable land to produce

the necessary forage due to the low yield from a late planting date. Sorghum would have

an advantage of using less irrigation water, but irrigation is not very common in the

63

northeastern U.S and, therefore, was not included in the IOFC analysis. Sorghum can

perform better than corn silage on soils with low water holding capacity which would

positively affect the IOFC of SS. Using a scenario of a higher yield of 13.4 t/ha that

would be more typical with a proper planting, we found the IOFC of SS to increase to

$9.43/cow/d. This is equal to the OS scenario and only $0.06/cow/d lower than the CS

scenario. The reported results are only a model and individual farm results would vary,

but they do demonstrate that, financially, these forages deserve consideration in dairy

farm crop rotations and lactating cow feeding programs.

CONCLUSIONS

In this study, we demonstrated that fall grown oat and BMR-6 dwarf sorghum

silages can support MY above 38 kg/d when included at 10% of the diet DM replacing

corn silage. The OS diet gave similar DMI, milk, and milk component yields as the

control corn silage diet. The higher milk and urinary urea N excretion with OS reveals a

potential for reducing dietary RDP from other feed sources when replacing corn silage

for oat silage. The SS diet decreased DMI, milk, and milk protein yields, which indicates

a need for additional rumen digestible energy sources when feeding low-starch sorghum

silage in place of corn silage. Production data from this experiment provide useful

information but should be interpreted with caution due to the lower number of

experimental units. The alternative forages tested in this study have potential in an

integrated cropping strategy and nutritional program for high-producing dairy cows.

64

ACKNOWLEDGEMENT

This project was supported by the Northeast Sustainable Agriculture Research and

Education (SARE) program. SARE is a program of the National Institute of Food and

Agriculture, U.S. Department of Agriculture. The authors thank the staff of Farm

Operations and Services of the Pennsylvania State University for growing and harvesting

the crops fed in our experiment. We also thank the staff of the Pennsylvania State

University’s Dairy Teaching and Research Center for their conscientious care of the

experimental cows. Additionally, we thank Virginia Ishler for her help with the IOFC

analysis.

REFERENCES

AOAC International. 2000. Official Methods of Analysis. 17th ed. AOAC International,

Arlington, VA.


Arlington, VA.

Abdelhadia, L. O., and F. J. Santini. 2006. Corn silage versus grain sorghum silage as a

supplement to growing steers grazing high quality pastures: Effects on performance

and ruminal fermentation. Anim. Feed Sci. Technol. 127:33–43.

Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating

dairy cattle. J. Dairy Sci. 83:1598-1624.



Bernard, J. K., and S. Tao. 2015. Short communication: Production response of lactating

dairy cows to brachytic forage sorghum silage compared with corn silage from first

or second harvest. J. Dairy Sci. 98:8994–9000.

Boivin, M., R. Gervais, and P. Y. Chouinard. 2013. Effect of grain and forage fractions of

corn silage on milk productions and composition in dairy cows. Animal. 7:245-254.

65

Bhandari, S. K., S. Li, K. H. Ominski, K. M. Wittenberg, and J. C. Plaizier. 2008. Effects

of the chop lengths of alfalfa silage and oat silage on feed intake, milk production,

feeding behavior, and rumen fermentation of dairy cows. J. Dairy Sci. 91:1942–

1958.

Branson, T. F., P. L. Guss and E. E. Ortman. 1969. Toxicity of sorghum roots to larvae of

the western corn rootworm. J. Econ. Entom. 62:1375-1378.

Broderick, G. A. 2003. Effects of varying dietary protein and energy levels on the

production of lactating dairy cows. J. Dairy Sci. 86:1370–1381.

Broderick, G. A., and W. J. Radloff. 2004. Effect of molasses supplementation on the

production of lactating dairy cows fed diets based on alfalfa and corn silage. J. Dairy

Sci. 87:2997-3009.

Buxton, D. R. 1996. Quality-related characteristics of forages as influenced by plant

environment and agronomic factors. Anim. Feed Sci. Technol. 59:37-49.

Carey, P. L., K. C. Cameron, H. J. Di, G. R. Edwards, and D. F. Chapman. 2016. Sowing

a winter catch crop can reduce nitrate leaching losses from winter-applied urine

under simulated forage grazing: a lysimeter study. Soil Use Manag. 32:329-337.

Chen, X. B., Y. K. Chen, M. F. Franklin, E. R. Ørskov, and W. J. Shand. 1992. The effect

of feed intake and body weight on purine derivative excretion and microbial protein

supply in sheep. J. Anim. Sci. 70:1534–1542.

Christensen, D. A., B. D. Owen, G. Steacy, J. P. Mtimuni, and W. L. Crowlen. 1977.

Nutritive value of whole crop silage made from seven cereal cultivars. Can. J. Anim.

Sci. 57:537-542.


Substitution of corn silage with sorghum silages in lactating cow diets: In vivo

methane emission and global warming potential of milk production. Agric. Syst.

136:106–113.

Contreras-Govea, F. E., and K. A. Albrecht. 2006. Forage production and nutritive value

of oat in autumn and early summer. Crop Sci. 46:2382–2386.

Di, H. J. and K. C. Cameron. 2002. Nitrate leaching in temperate agroecosystems:

sources, factors and mitigating strategies. Nutrient Cycling in Agroecosystems.

64:237–256

Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith. 1956. Colorimetric

method for determination of sugars and related substances. Anal. Chem. 28:350.

Emery, R. S. 1978. Feeding for increased milk protein content. J. Dairy Sci. 61:825-828.

Faé, G. S., R. M. Sulc, D. J. Barker, R. P. Dick, M. L. Eastridge, and N. Lorenz. 2009.

Integrating winter annual forages into a no-till corn silage system. Agron. J.

101:1286–1296.

66

Fievez, V., E. Colman, J.M. Castro-Montoya, I. Stefanov, and B. Vlaeminck. 2012. Milk

odd- and branched-chain fatty acids as biomarkers of rumen function- An update.

Anim. Feed Sci. and Technol. 172:51–65.

Firkins, J. L. 1997. Effects of feeding nonforage fiber sources on site of fiber digestion. J.

Dairy Sci. 80:1426–1437.

Gentry, L. F., M. L. Ruffo, and F. E. Below. 2013. Identifying factors controlling the

continuous corn yield penalty. Agron. J. 105:295–303.

Goering, H. K. and P .J. Van Soest. 1970. Forage Fiber Analysis. USDA Agricultural

Research Service. Handbook number 379. U.S. Dept. of Agriculture. Superintendent

of Documents, US Government Printing Office, Washington, D.C. 20402.

Grant, R. J., S. G. Haddad, K. J. Moore, and J. F. Pedersen. 1995. Brown midrib sorghum

silage for midlactation dairy cows. J. Dairy Sci. 78:1970-1980.

Hall, M. B. 2009. Determination of starch, including maltooligosaccharides, in animal

feeds: Comparison of methods and a method recommended for AOAC collaborative

study. J. AOAC Int. 92:42–49.

Harper, M. T., J. Oh, F. Giallongo, G. W. Roth, and A. N. Hristov. 2017. Inclusion of

wheat and triticale silage in the diet of lactating dairy cows. J. Dairy Sci. Submitted.

Hart, K. J., P. G. Martin, P. A. Foley, D. A. Kenny, and T. M. Boland. 2009. Effect of

sward dry matter digestibility on methane production, ruminal fermentation, and

microbial populations of zero-grazed beef cattle. J. Anim. Sci. 87:3342–3350.

Heinrichs, A. J. 2013. The Penn State particle separator. Penn State Coop. Ext.

Publication. DSE 13-186.

Hristov, A. N., J. Oh, J. L. Firkins, J. Dijkstra, E. Kebreab, G. Waghorn, H. P. S. Makkar,

A. T. Adesogan, W. Yang, C. Lee, P. J. Gerber, B. Henderson, and J. M. Tricarico.

2013. SPECIAL TOPICS—Mitigation of methane and nitrous oxide emissions from

animal operations: I. A review of enteric methane mitigation options. J. Anim. Sci.

91:5045–5069.

Hristov, A. N., J. Oh, F. Giallongo, T. Frederick, H. Weeks, P. R. Zimmerman, R. A.

Hristova, S. R. Zimmerman, and A. F. Branco. 2015. The use of an automated

system (GreenFeed) to monitor enteric methane and carbon dioxide emissions from

ruminant animals. J. Vis. Exp. 103:e52904.

Huhtanen, P., K. Kaustell, and S. Jaakkola. 1994. The use of internal markers to predict

total digestibility and duodenal flow of nutrients in cattle given six different diets.

Anim. Feed Sci. Technol. 48:211–227.


cereal grain silage substituted for alfalfa silage on performance of lactating dairy

cows. J. Dairy Sci. 76:3536-3546.

67

Krishnamoorthy, U., Muscato, T. V., Sniffen, C. J., Van Soest, P. J. 1982. Nitrogen

fractions in selected feedstuffs. J. Dairy Sci. 65:217–225.

Lanzas, C., C. J. Sniffen, S. Seo, L. O. Tedeschi, and D. G. Fox. 2007. A revised CNCPS

feed carbohydrate fractionation scheme for formulating rations for ruminants. Anim.

Feed Sci. Technol. 136:167–190.

Lechartier, C., and J. L. Peyraud. 2011. The effects of starch and rapidly degradable dry

matter from concentrate on ruminal digestion in dairy cows fed corn silage-based

diets with fixed forage proportion. J. Dairy Sci. 94:2440–2454.

Lee, C., A. N. Hristov, T. W. Cassidy, K. S. Heyler, H. Lapierre, G. A. Varga, M. J. de

Veth, R. A. Patton, and C. Parys. 2012. Rumen-protected lysine, methionine, and

histidine increase milk protein yield in dairy cows fed a metabolizable protein-

deficient diet. J. Dairy Sci. 95:6042–6056.

Leonardi, C., K. J. Shinners, and L. E. Armentano. 2005. Effect of different dietary

geometric mean particle length and particle size distribution of oat silage on feeding

behavior and productive performance of dairy cattle. J. Dairy Sci. 88:698–710.

Lusk, J. W., P. K. Karau, D. O. Balogu, and L. M. Gourley. 1984. Brown midrib sorghum

or corn silage for milk production. J. Dairy Sci. 67: 1739-1744.

McCartney, D., and A. Vaage. 1994. Comparative yield and feeding value of barley, oat

and triticale silage. Can. J. Anim. Sci. 74:91-96.

Mertens, D. R. 1985. Effect of fiber on feed quality for dairy cows. Pages 209-224 in 46th

Minnesota Nutrition Conference Proc.

Mertens, D. R. 2009. Maximizing forage use by dairy cows. Pages 303-319 in Proc. 27th

Annual Western Canadian Dairy Seminar. Red Deer, Alberta.

Miron, J., E. Zuckerman, G. Adin, R. Solomon, E. Shoshani, M. Nikbachat, E. Yosef, A.

Zenou, Z. G. Weinberg, Y. Chen, I Halachmi, and D. Ben-Ghedalia. 2007.

Comparison of two forage sorghum varieties with corn and the effect of feeding

their silages on eating behavior and lactation performance of dairy cows. Anim.


National Research Council. 2001. Nutrient requirements of dairy cattle. 7th rev. ed. Natl.

Acad. Press, Washington, DC.

Oliver, A. L., R. J. Grant, J. F. Pedersen, and J. O’Rear. 2004. Comparison of brown

midrib-6 and -18 forage sorghum with conventional sorghum and corn silage in diets

of lactating dairy cows. J. Dairy Sci. 87:637–644.

Oltjen, J. W., and Bolsen K. K. 1980. Wheat, barley, oat and corn silages for growing

steers. J. Anim. Sci. 51:958-965.

68

Ørskov, E. R., and I. McDonald. 1979. The estimation of protein degradability in the

rumen from incubation measurements weighted according to rate of passage. J.

Agric. Sci. (Camb.) 92:499–503.

Palmquist, D. L., A. D. Beaulieu, and D. M. Barbano. 1993. Feed and animal factors

influencing milk fat composition. J. Dairy Sci. 76:1753-1771.

Penn State Extension. 2015. Income over feed cost. Accessed October 10, 2016.

http://extension.psu.edu/animals/dairy/business-management/financial-tools/income-

over-feed-cost

Penn State Extension. 2016. Cash flow spreadsheet. Accessed October 10, 2016.

http://extension.psu.edu/animals/dairy/business-management/financial-tools/cash-

flow-planning

Razzaghi, A., R. Valizadeh, A. A. Naserian, M. Danesh Mesgaran, A. J. Carpenter, and

M. H. Ghaffari. 2016. Effect of dietary sugar concentration and sunflower seed

supplementation on lactation performance, ruminal fermentation, milk fatty acid

profile, and blood metabolites of dairy cows. J. Dairy Sci. 99:3539–3548.

Rico, D. E., and K. J. Harvatine. 2013. Induction of and recovery from milk fat

depression occurs progressively in dairy cows switched between diets that differ in

fiber and oil concentration. J. Dairy Sci. 96:6621–6630.

Sarwar, M., J. L. Firkins, and M. L. Eastridge. 1992. Effects of varying forage and

concentrate carbohydrates on nutrient digestibilities and milk production by dairy

cows. J. Dairy Sci. 75:1533-1542.

Shepherd, M. A. 1999. The effectiveness of cover crops during eight years of a UK

sandland rotation. Soil Use Manag. 15:41–48.

Sindelar, A. J., M. R. Schmer, V. L. Jin, B. J. Wienhold, and G. E. Varvel. 2016. Crop

rotation affect corn, grain sorghum, and soybean yields and nitrogen recovery.

Agron. J. 108:1592-1602.

Sjaunja, L. O., L. Baevre, L. Junkkarinen, J. Pedersen, and J. Setälä. 1990. A Nordic

proposal for an energy corrected milk (ECM) formula. Proc. 27th Session Int.

Comm. Breeding Product. Milk Anim., Paris, France. Wageningen Acad. Publ.,

Wageningen, the Netherlands.

Van Soest, P. J., J. B. Robertson, and B. A. Lewis. 1991. Methods for dietary fiber,

neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition.

J. Dairy Sci. 74:3583–3597.

Van Soest, P. J. 1994. Nutritional ecology of the ruminant. 2nd Ed. Cornell University

Press, Ithaca, NY.

Vencill, W.K., R. L. Nichols, T. M. Webster, J. K. Soteres, C. Mallory-Smith, N. R.

Burgos, W. G. Johnson, and M. R. McClelland. 2012. Herbicide resistance: Toward

69

an understanding of resistance development and the impact of herbicide-resistant

crops. Weed Science. 60:Sp1. 2-30.

Vlaeminck, B., V. Fievez, A. R. J. Cabrita, A. J. M. Fonseca, and R. J. Dewhurst. 2006.

Factors affecting odd- and branched-chain fatty acids in milk: A review. Anim. Feed

Sci. Technol. 131:389–417.

70

Table 3-1. Ingredient and chemical composition of the diets fed in the experiment

1CS = Corn silage control diet; OS = Oat silage diet; SS = Sorghum silage diet. 2Alfalfa haylage was 36.0% DM and contained (DM basis) 22.1% CP, 23.4% NFC, and

41.4% NDF. 3SoyPLUS (West Central Cooperative, Ralston, IA). 4Molasses (Westway Feed Products, Tomball, TX). 5The mineral/vitamin premix (Cargill Animal Nutrition, Cargill Inc., Roaring Spring, PA)

contained (%, as-is basis) trace mineral mix, 0.86; MgO (56% Mg), 8.0; NaCl, 6.4;

vitamin ADE premix (Cargill Animal Nutrition, Cargill Inc.), 0.48; limestone, 37.2;

selenium premix (Cargill Animal Nutrition, Cargill Inc.), 0.07; and dry corn distillers

grains with solubles, 46.7. Ca, 14.1%; P, 0.39%; Mg, 4.60%; K, 0.45%; S, 0.38%; Se,

Diet1

Item CS OS SS

Ingredient, % of DM

Corn silage 44 34 34

Alfalfa haylage2 7.5 7.5 7.5

Oat silage - 10 -

Sorghum silage - - 10

Hay/straw mixture 4 4 4

Cottonseed hulls 4 4 4

Ground corn 11 11 11

Heat-treated whole soybeans 7.5 7.5 7.5

Solvent-extracted canola meal 7 7 7

SoyPLUS3 7.5 7.5 7.5

Molasses4 4.5 4.5 4.5

Mineral/vitamin premix5 3 3 3

Composition, % of DM

CP6 16.3 16.8 16.6

RDP8 9.7 9.7 9.5

RUP8 6.6 7.1 7.0

NDF6 32.0 33.4 34.2

ADF6 22.2 23.3 23.7

NFC7 44.3 41.5 41.8

Starch6 24.3 20.8 20.9

Fat6 4.7 4.8 4.6

NEL,7 Mcal/kg 1.56 1.55 1.55

NEL intake,7 Mcal/d 41.6 42.1 40.2

NEL balance,7 Mcal/d 4.0 4.1 3.2

MP balance,7 g/d 406 512 486

Ash6 6.7 7.5 7.0

Ca6 0.8 0.8 0.8

P6 0.4 0.4 0.4

71

6.67 mg/kg; Cu, 358 mg/kg; Zn, 1,085 mg/kg; Fe, 188 mg/kg, vitamin A, 262,656 IU/kg;

vitamin D, 65,559 IU/kg; and vitamin E, 1,974 IU/kg. 6Values calculated using the chemical analysis (Cumberland Valley Analytical Services

Inc., Maugansville, MD) of individual feed ingredients of the diet. 7Estimated based on NRC (2001).

72

Table 3-2. Nutrient composition and fermentation profile of oat and sorghum silages (%

of DM or as indicated)1

1Three composite samples per silage, one for each experimental period were analyzed

(Cumberland Valley Analytical Services Inc., Maugansville, MD). 2Largest SEM published in table; n = 9 (n represents the number of observations used in

the statistical analysis). 3Analyzed by NIR. 4NA = Not analyzed.

Silage P-Value

Item Corn Oat Sorghum SEM2 Silage

DM, % 39.1 31.6 31.4 - -

NDF 40.2c 54.7b 62.7a 2.25 0.001

ADF 25.9b 36.3a 40.8a 1.49 0.001

Lignin 3.70b 2.86b 4.89a 0.319 0.01

Fat 3.38a 3.86a 1.70b 0.172 <0.001

CP 6.83c 11.7a 9.50b 0.199 <0.001

Soluble protein 4.10c 8.07a 5.40b 0.332 <0.001

Starch 34.7a 0.27b 0.80b 0.863 <0.001

Ethanol soluble

carbohydrates 1.17b 1.77b 3.70a 0.455 0.02

Ash 3.47c 11.2a 5.70b 0.600 <0.001

Ca 0.20c 0.60a 0.41b 0.005 <0.001

P 0.24b 0.39a 0.24b 0.014 <0.001

K 1.21b 4.73a 1.62b 0.337 <0.001

pH3 3.83c 4.71a 4.31b 0.054 <0.001

Fermentation acids

Lactic3 5.43b 7.27a 2.28c 0.347 <0.001

Acetic3 0.98 2.10 1.57 0.526 0.38

Butyric3 0.0b 0.88a NA4 0.144 0.01

Titratable acidity3,

meq/100g 5.32 5.96 3.66 0.455 0.06

73

Table 3-3. Ruminal in situ DM and NDF degradability of ensiled forages1

Forage

Item Alfalfa

Haylage Corn Silage Oat Silage

Sorghum

Silage

DM

Soluble, % 36.4 ± 1.04c 54.0 ± 0.91a 41.0 ± 1.03b 32.0 ± 0.95d

Potentially degradable

(fraction b), % 32.0 ± 1.26b

46.0 ±

11.80ab 54.0 ± 3.05a 48.0 ± 4.82a

Rate of degradation of b, %/h 4.54 ± 0.46a 0.86 ± 0.34c 1.94 ± 0.27b 1.40 ± 0.27bc

Effective degradability2 55.7 ± 0.48c 64.3 ± 0.46a 62.2 ± 0.47b 47.3 ± 0.46c

NDF

Soluble, % 10.4 ± 1.47a - 3.15 ± 1.44b 3.16 ± 1.42b

Potentially degradable

(fraction b), % 38.4 ± 2.3c - 84.3 ± 3.3a 65.0 ± 4.5b

Rate of degradation of b, %/h 2.83 ± 0.46x 0.57 ± 0.023 2.15 ± 0.21xy 1.78 ± 0.28y

Effective degradability2 29.0 ± 0.69b - 38.3 ± 0.69a 27.4 ± 0.68b a,b,c Means within the same row without a common superscript differ (P < 0.05). x,y Means within the same row without a common superscript differ (P < 0.10). 1Values are model estimates ± SE of disappearance curves fit using SigmaPlot 10.0

(Systat Software, Chicago, IL) to the equation p = a + b (1 - e-ct), where p is the degraded

fraction (of DM or NDF) at time t, a is the soluble fraction, b is the potentially

degradable fraction, and c is the rate of degradation of the b fraction (Ørskov and

McDonald, 1979); DM disappearance, n = 144; NDF disappearance, n = 108 (n

represents the number of observations used in the statistical analysis). 2Effective degradability (ED) was estimated as: ED = a + b {c ÷ (c + k)}, where a, b, and

p are as above and k is the rate of passage (Ørskov and McDonald, 1979) assumed to be

0.03/h in this study. 3Corn silage NDF degradation data were fit to a linear model with an R2 = 0.94.

74

Table 3-4. Effect of oat and sorghum silage on DMI, milk production, and feed efficiency

in lactating dairy cows

Diet1 P-Value

CS OS SS SEM2 Diet

DMI, kg/d 26.7a 27.1a 26.0b 1.69 0.02

Milk yield, kg/d 39.6a 40.2a 38.7b 3.58 0.006

Milk ÷ DMI, kg/kg 1.48 1.49 1.49 0.05 0.86

Milk fat, % 3.58b 3.60b 3.74a 0.12 0.02

Milk fat,3 kg/d 1.42 1.42 1.39 0.07 0.78

Milk true protein, % 2.85a 2.83a 2.77b 0.06 0.03

Milk true protein,3 kg/d 1.13a 1.13a 1.04b 0.06 0.05

Lactose, % 5.01 4.98 4.98 0.05 0.58

Lactose,3 kg/d 2.00 2.00 1.88 0.12 0.17

MUN, mg/dL 13.5 13.8 13.9 0.63 0.73

ECM,3,4 kg/d 36.9 36.9 35.1 1.95 0.32

BW, kg 617 618 596 - - a,bMeans within the same row without a common superscript differ (P ≤ 0.05). 1CS = Corn silage control diet; OS = Oat silage diet; SS = Sorghum silage diet. 2Largest SEM published in table; DMI, n = 348; milk yield, n = 351; milk yield ÷ DMI, n

= 339; BW, n = 36; milk composition data, n = 36 (n represents the number of

observations used in the statistical analysis). 3Calculated using the milk yield of the 2 consecutive milk sampling days. 4Energy-corrected milk (kg/d) = kg of milk × [(38.3 × % fat × 10 + 24.2 × % true protein

× 10 + 16.54 × % lactose × 10 + 20.7) ÷ 3,140] (Sjaunja et al., 1990).

75

Table 3-5. Effect of oat and sorghum silage on milk fatty acid composition (g/100 g of

total fatty acids) in lactating dairy cows

Diet1 P-Value

Fatty Acid CS OS SS SEM2 Diet

4:0 2.59 2.58 2.74 0.088 0.22

6:0 2.12 2.14 2.12 0.041 0.88

8:0 1.25 1.25 1.24 0.023 0.76

10:0 2.99 2.93 2.85 0.072 0.11

10:1 0.25 0.24 0.24 0.012 0.22

11:0 0.05a 0.05a 0.04b 0.004 <0.001

12:0 3.41a 3.29ab 3.20b 0.087 0.03

13:0 iso 0.03 0.03 0.03 0.001 0.07

13:0 anteiso 0.06a 0.06b 0.05b 0.003 0.04

13:0 0.10a 0.10a 0.09b 0.004 0.001

14:0 iso 0.11 0.11 0.11 0.011 0.88

14:0 11.0 10.8 10.6 0.171 0.16

14:1 0.72a 0.66b 0.66b 0.044 0.03

15:0 iso 0.20b 0.22a 0.22a 0.006 <0.001

15:0 anteiso 0.41 0.40 0.40 0.009 0.90

15:0 0.94a 0.93a 0.88b 0.026 0.005

16:0 iso 0.23 0.24 0.24 0.021 0.42

16:0 26.6 25.7 25.8 0.418 0.07

16:1 0.92 0.85 0.89 0.046 0.12

17:0 iso 0.28 0.29 0.29 0.011 0.09

17:0 anteiso 0.38 0.38 0.38 0.007 0.41

17:0 0.49 0.50 0.52 0.008 0.06

17:1 0.14 0.14 0.15 0.007 0.31

18:0 13.3b 14.2a 14.6a 0.450 0.01

trans-4 18:1 0.03 0.02 0.03 0.001 0.19

trans-5 18:1 0.02 0.02 0.02 0.001 0.29

trans-6,8 18:1 0.33 0.32 0.32 0.010 0.54

trans-9 18:1 0.26a 0.25b 0.25b 0.007 0.01

trans-10 18:1 0.49a 0.45b 0.45b 0.019 <0.001

trans-11 18:1 1.21 1.24 1.24 0.042 0.68

trans-12 18:1 0.50 0.46 0.48 0.032 0.52

cis-9 18:1 18.9 19.3 19.5 0.480 0.28

trans-15 18:1 0.38 0.38 0.37 0.015 0.35

cis-11 18:1 0.70 0.66 0.67 0.028 0.50

cis-12 18:1 0.51a 0.49b 0.47c 0.020 <0.001

Linoleic acid 3.70 3.66 3.57 0.072 0.36

α-Linolenic acid 0.62b 0.67a 0.63b 0.014 0.005

20:0 0.16b 0.17a 0.17a 0.005 0.005

20:1 0.04 0.04 0.04 0.002 0.13

cis-9,trans-11 0.52 0.51 0.51 0.028 0.69

76

CLA

20:2 0.04 0.04 0.03 0.004 0.41

20:3 0.15 0.14 0.15 0.010 0.80

20:4 0.17 0.16 0.16 0.010 0.05

20:5 0.04 0.04 0.04 0.002 0.68

22:0 0.06b 0.06a 0.06a 0.002 <0.001

24:0 0.03b 0.03a 0.03a 0.002 0.001

24:1 <0.01 <0.01 <0.01 <0.001 0.39

22:4 0.03 0.03 0.03 0.004 0.18

22:5 0.07 0.07 0.07 0.005 0.86

Σ SFA 66.7 66.5 66.6 0.641 0.86

Σ MUFA 25.4 25.5 25.8 0.566 0.67

Σ PUFA 5.35 5.33 5.18 0.093 0.32

Σ trans FA 3.21 3.16 3.17 0.100 0.62

Σ OBCFA4 3.36 3.39 3.37 0.056 0.76

Unknown 2.52 2.64 2.46 0.080 0.16 a,bMeans within the same row without a common superscript differ (P < 0.05). 1CS = Corn silage control diet; OS = Oat silage diet; SS = Sorghum silage diet. 2Largest SEM shown; n = 36 for all variables (n represents number of observations used

in the statistical analysis). Data are presented as LSM. 3ND = Not detected. 4Sum of odd and branched chain fatty acids (iso13:0, anteiso13:0, 13:0, iso14:0, iso15:0,

anteiso15:0, 15:0, iso16:0, iso17:0, 17:0, 17:1).

77

Table 3-6. Effect of oat and sorghum silage on nutrient intake and apparent total-tract

digestibility in lactating dairy cows

Diet1 P-Value

Item CS OS SS SEM2 Diet

Intake, kg/d

DM3 26.1 26.9 25.9 1.15 0.23

OM 24.3 24.9 24.1 1.06 0.37

CP 4.25b 4.52a 4.30b 0.19 0.04

Starch 6.22a 5.45b 5.06c 0.24 <0.001

NDF 8.34b 9.00a 8.87a 0.38 0.02

NDF, % of

BW 1.43 1.51 1.50 0.05 0.17

Forage NDF,

% of BW 0.99b 1.08a 1.08a 0.04 <0.01

ADF 5.79b 6.28a 6.15a 0.27 0.01

Apparent digestibility, %

DM 68.1b 69.8a 66.2c 0.29 <0.001

OM 69.1b 70.8a 67.3c 0.28 <0.001

CP 66.0b 67.6a 64.6c 0.52 <0.001

Starch 98.3a 98.0b 97.7c 0.08 <0.001

NDF 43.8b 50.8a 44.8b 0.56 <0.001

ADF 41.6b 49.6a 42.1b 0.64 <0.001 a,b,c Means within the same row without a common superscript differ (P < 0.05). 1CS = Corn silage control diet; OS = Oat silage diet; SS = Sorghum silage diet. 2Largest SEM published in table; n = 36 (n represents the number of observations used in

the statistical analysis). 3DM intake reported is during the collection period for digestibility.

78

Table 3-7. Effect of oat and sorghum silage on nitrogen utilization, urinary purine

derivatives and carbon dioxide (CO2) and methane (CH4) emissions1 in lactating dairy

cows

Diet2 P-Value

CS OS SS SEM3 Diet

N intake, g/d 680b 724a 689b 30.4 0.04

N excretion or secretion, g/d 644 677 649 25.9 0.09

Urine N, g/d 218 245 221 12.0 0.17

Urinary Urea-N, g/d 139b 162a 154ab 7.8 0.03

Fecal N, g/d 248 255 265 11.3 0.11

Total excreta N, g/d 466 500 486 19.0 0.08

Milk N, g/d 178a 177a 163b 9.4 0.05

N excretion or secretion, as % of N intake

Urine N 32.3 34.1 32.7 1.80 0.76

Fecal N 36.6b 35.2c 38.6a 0.56 <0.001

Total excreta N 68.9 69.2 71.2 1.99 0.63

Milk N 26.1a 24.4ab 23.6b 0.75 0.05

Urine output3, kg/d 26.1b 33.0a 25.9b 2.20 0.002

Urinary PD4 excretion, mmol/d

Allantoin 567 585 531 38.2 0.28

Uric acid 66.2 68.3 60.9 4.83 0.06

Total PD 633 654 592 41.7 0.23

Rumen gas emissions

CO2, kg/d 12.6 12.3 12.3 0.42 0.28

CH4, g/d 495 488 523 25.7 0.15

CH4, g/kg DMI5 18.9 18.4 20.8 1.29 0.32

CH4, g/kg milk5 12.4 12.4 13.4 1.06 0.64 a,bMeans within the same row without a common superscript differ (P < 0.05). 1Rumen gas emissions were measured using GreenFeed (C-Lock Technology Inc., Rapid

City, SD). Data were derived from 8 individual measurements staggered over a 3-d

period. 2CS = Corn silage control diet; OS = Oat silage diet; SS = Sorghum silage diet. 3Largest SEM published in table; n = 36 (n represents the number of observations used in

the statistical analysis). 3Estimated from urine creatinine concentration, assumed to be excreted at 29 mg/kg of

BW. 4Purine derivatives. 5Based on milk yield and DMI data during the sampling periods.

79

Oat and Sorghum DM Disappearance

In situ incubation length, h

0 20 40 60 80 100

DM

dis

app

eara

nce

, %

30

40

50

60

70

80

90

Alfalfa Haylage

Corn Silage

Oat Silage

Sorghum Silage

Figure 3-1. Ruminal in situ DM disappearance of forage sources.

Data are means ± SE (n = 6). Disappearance curves were fit using SigmaPlot 10.0 (Systat

Software, Chicago, IL) to the equation p = a + b (1 - e-ct), where p is the degraded

fraction (of DM) at time t, a is the soluble fraction, b is the potentially degradable

fraction, and c is the rate of degradation of the b fraction (Ørskov and McDonald, 1979).

80

Oat and Sorghum NDF Disappearance


0 20 40 60 80 100

ND

F d

isap

pea

ran

ce,

% o

f N

DF

0

10

20

30

40

50

60

70

80

Alfalfa Haylage

Corn Silage

Oat Silage

Sorghum Silage

Figure 3-2. Ruminal in situ NDF disappearance of forage sources.

Data are means ± SE (n = 6). Disappearance curves of alfalfa haylage, triticle silage, and

wheat silage were fit using SigmaPlot 10.0 (Systat Software, Chicago, IL) to the equation

p = a + b (1 - e-ct), where p is the degraded fraction (of NDF) at time t, a is the soluble

fraction, b is the potentially degradable fraction, and c is the rate of degradation of the b

fraction (Ørskov and McDonald, 1979). Corn silage NDF disappearance data were fit to a

linear model.

Chapter 4

Inclusion of wheat and triticale silage in the diet of lactating dairy cows

Journal of Dairy Science Vol. 100, 8, 6151-6163, 2017

https://doi.org/10.3168/jds.2017-12553

M. T. Harper, J. Oh, F. Giallongo, G. W. Roth, and A. N. Hristov

ABSTRACT

The objective of this experiment was to partially replace corn silage with 2

alternative forages, wheat (Triticum aestivum) or triticale (X Triticosecale) silages, at

10% of the diet dry matter (DM) and investigate the effects on dairy cow productivity,

nutrient utilization, enteric CH4 emissions, and farm income over feed costs. Wheat and

triticale were planted in the fall as cover crops and harvested in the spring at the boot

stage. Neutral- and acid-detergent fiber and lignin concentrations were higher in the

wheat and triticale silages compared with corn silage. The forages had similar ruminal in

situ effective degradability of DM. Both alternative forages had 1% starch or less

compared with the approximately 35% starch in corn silage. Diets with the alternative

forages were fed in a replicated 3 × 3 Latin square design experiment with 3, 28-d,

periods and 12 Holstein cows. The control diet contained 44% (DM basis) corn silage. In

the other 2 diets, wheat or triticale silages were included at 10% of dietary DM, replacing

corn silage. Dry matter intake was not affected by diet, but both wheat and triticale silage

decreased yield of milk (41.4 and 41.2 versus 42.7 ± 5.18 kg/d) and milk components,

82

compared with corn silage. Milk fat from cows fed the alternative forage diets contained

higher concentrations of 4:0, 6:0, and 18:0 and tended to have lower concentrations of

total trans fatty acids. Apparent total tract digestibility of DM and organic matter was

decreased in the wheat silage diet, and digestibility of neutral-and acid-detergent fiber

was increased in the triticale silage diet. The wheat and triticale silage diets resulted in

higher excretion of urinary urea, higher MUN, and lower milk N efficiency, compared

with the corn silage diet. Enteric CH4 emission per kg of energy corrected milk was

highest in the triticale silage diet, whereas CO2 emission was decreased by both wheat

and triticale silage. This study showed that, at milk production of around 42 kg/d, wheat

silage and triticale silage can partially replace corn silage DM and not affect DM intake,

but milk yield may decrease slightly. For dairy farms in need of more forage, triticale or

wheat double cropped with corn silage may be an appropriate cropping strategy.

Keywords: dairy cow, forage, triticale silage, wheat silage

83

INTRODUCTION

Dairies in the northeastern United States typically grow their own forages. The

most utilized forage, corn silage, leaves bare soil after fall harvesting until spring

planting. Cover crops, such as small grains and clovers, have been used to prevent soil

erosion during bare soil periods. Preserving the soil is critically important for continued

crop productivity and therefore has long-term benefits. Cover crops have the potential to

efficiently utilize fall applied manure and reduce nitrate leaching (Shephard, 1999; Di

and Cameron, 2002; Carey et al., 2016). However, planting a cover crop requires a short-

term investment of labor, equipment, and other inputs. The use of cover crops as an

alternative forage has increased in popularity as a way to offset planting costs, increase

the annual forage yield per acre, and thereby harvest more forage from the same land

base. Recent plot studies conducted at The Pennsylvania State University showed a 4.5-

6.5 t DM/ha average annual forage yield increase when double cropping corn silage with

rye or triticale cover crop harvested as silage in the flag leaf stage (G. W. Roth, The

Pennsylvania State University, University Park, PA, personal communication). However,

the corn silage portion of annual forage yields typically decrease between 10-20%,

depending on planting date, under double cropping management due to delayed planting

(The Penn State Agronomy Guide, 2015). Less corn silage inventory leads to the

question: Can cover crop silages replace a portion of corn silage in dairy cattle diets? A

number of studies in the U.K. have reported similar milk yield (MY) responses to corn or

wheat silages harvested after kernel development (Hameleers, 1998; Sinclair et al., 2005).

In those studies, however, the wheat silage contained higher starch concentrations than

84

the corn silage. In Canada, Khorasani et al. (1993) compared cereal grain silages,

including triticale, to alfalfa haylage. They reported a decrease in DMI for triticale silage

but no significant difference in MY and suggested that triticale silage could be used in

dairy cow rations. The dough stage harvest schedule increases starch content but delays

harvest past the planting window for corn in the northeast U.S. and as such cannot be

used with corn silage double cropping. Harvesting cereal crops at the boot stage yields

similar NDF content to the dough stage (Khorasani et al., 1997), but with increased NDF

digestibility (NDFD) (Arieli and Adin, 1994). Additionally, at the flag leaf or boot

stages, cereal crops can have CP concentrations above 12% (Crovetto et al., 1998; Fearon

et al., 1990). Wheat and triticale are two cereal grain cover crops suited to the northeast

U.S. that are used as lactating dairy cow forage in other areas of the world. Furthermore,

both forages grow well in cool weather and survive cold winters.

Therefore, we hypothesized that both wheat and triticale, when harvested in the

boot stage, could serve as alternative forages to augment corn silage use in lactating dairy

cow rations in the northeastern U.S. The objective of the experiment was to replace corn

silage with either triticale silage or wheat silage at 10% of the diet DM and investigate

the effects on DMI, MY, milk components and fatty acid (FA) profile, nutrient

digestibility, N utilization, enteric CH4 emissions, and income over feed costs (IOFC) in

lactating dairy cows.

85


Crops and Silages

Wheat (Triticum aestivum ‘Malabar’; King’s Agriseeds, Ronks, PA) and triticale

(X Triticosecale ‘Hyoctane’; Seedway, Hall, NY) were grown in Centre County, PA at

approximately 40° N latitude on Hagerstown and Hublersburg soils during the fall of

2014. Both crops were planted with a no-till drill (John Deere 1590; Moline, IL) into

fields fertilized with 44.8 t/ha of dairy manure prior to planting contributing 42 kg/ha of

ammonium N. Forages were planted next to each other in the same field with 19 cm row

spacing on October 10th, 2014 after wheat harvested for grain. Seeding rate was 151

kg/ha for triticale and wheat. On April 4th, 2015, both wheat and triticale were fertilized

with 67 kg of N/ha from a 30% urea and ammonium nitrate liquid fertilizer. A John

Deere 946 mower with a roll conditioner was used to mow both crops and, after wilting

to target 30% DM, the forages were gathered and chopped using a John Deere 6750

harvester. Mowing was conducted on May 13th and 19th, 2015 at the boot stage for

triticale and wheat, respectively, and chopping occurred on May 15th and 20th,

respectively. Chop length was set to 12 mm. Both crops were ensiled without inoculant

in 3-m diameter plastic silage bags (Up North Plastics, Cottage Grove, MN). The corn

silage, which was the control in this experiment, was from the forage source normally fed

to the tie stall cows on an experiment. The corn silage was a mixture of the following

hybrids: Mycogen TMF2R737 (112 d relative maturity), Dekalb DKC 52-61 (102 d

relative maturity), and NK N60F-3111 (107 d relative maturity). Corn silage was grown

in Centre County, PA at approximately 40° N latitude on Hagerstown and Hublersburg

86

soils and planted between May 1st and May 10th, 2014 at a rate of 79,000 seeds/ha. It was

planted with a no-till drill (John Deere 1590; Moline, IL) into fields fertilized with 45 t/ha

of dairy manure prior to planting contributing 42 kg/ha of ammonium N. An additional

43 kg/ha of N was applied as 30% urea and ammonium nitrate liquid prior to planting and

67 kg/ha of N in the same form as a sidedress application. Corn silage harvest was

conducted between September 15th and September 30th at a target DM of 38% with a 19

mm chop length and ensiled in an upright concrete silo.

Animals and Diets


Pennsylvania State University’s Institutional Animal Care and Use Committee. Twelve

mid-lactation Holstein dairy cows (MY, 42 ± 10.1 kg; 2.5 ± 1.38 lactations; DIM 38 ± 5.7

d; BW 632 ± 101.6 kg at the beginning of the experiment) were used in a replicated 3 × 3

Latin square design balanced for residual effects. The experiment had 3 periods and each

period was 28 d with 18 d for adaptation to the diet and 10 d for data and sample

collection. Cows were allocated to 4 groups of 3 cows each based on DIM, MY, and

parity. Cows within a group were randomly assigned to one of 3 diets, as described

below. All cows were housed in the tie stall barn of The Pennsylvania State University’s

Dairy Research and Teaching Center. Diets were mixed and fed from a Rissler model

1050 TMR mixer (I.H. Rissler Mfg. LLC, Mohnton, PA). Cows were fed once daily

around 8 a.m. to yield approximately 5-10% refusals. Feed was pushed up 3 times

throughout the day. The cows were milked twice daily at 7 a.m. and 6 p.m.

87

Three different diets (Table 1), were fed to the cows during the experiment as

follows: a control diet (CS), based on corn silage and alfalfa haylage; a triticale silage

diet (TS), triticale silage included at 10% of dietary DM, replacing 22.7% of the control

diet corn silage DM; and a wheat silage diet (WS), wheat silage included at 10% of

dietary DM, replacing 22.7% of the control diet corn silage DM. The only difference

between the control and alternative forage diets was the replacement of 22.7% corn silage

DM with either wheat or triticale silage. The CS diet was formulated to meet or exceed

the NRC (2001) NEL and MP requirements of a Holstein cow with 680 kg BW, 41.7 kg

MY, 3.8% fat, 3.2% true protein, and at 26.3 kg DMI.


Refusals were collected and weighed individually for each cow prior to the

morning feeding to measure daily as-fed intake. Total mixed ration, refusal and forage

(triticale, wheat, alfalfa, and corn silage) samples were collected twice weekly,

composited by wk and diet (i.e., silage type), stored at -20°C and then oven dried at 55°C

for 72 h. The TMR was sampled within 1 h of feeding. The weekly DM content of the

TMR and refusals was used to calculate the individual daily DMI. Concentrate feeds

were sampled weekly and stored at -20°C until analysis. Wheat, triticale, corn and alfalfa

silages were ground through a 4 mm screen (for in situ degradability measurements),

then, along with TMR samples, were ground through a 1 mm screen in a Wiley mill

(Thomas Scientific, Swedesboro, NJ) and composited by period on an equal weight basis.

Dried composite samples of sorghum, oat and corn silages were sent to Cumberland

88

Valley Analytical Services Inc. (CVAS; Maugansville, MD) to be analyzed by wet

chemistry methods for CP, amylase-treated NDF, ADF, lignin, fat, soluble protein,

starch, ethanol soluble carbohydrates, ash, and minerals. Fermentation profiles of fresh

frozen samples of the corn, wheat and triticale silages from each period were analyzed by

CVAS by wet chemistry for pH, titratable acidity, and lactic, acetic, propionic, butyric,

and isobutyric acid concentrations. Concentrate feed samples were ground through a 1-

mm screen and composited for the entire experiment. Dried composite concentrate

ingredients were analyzed by wet chemistry methods by CVAS for CP, amylase-treated

NDF, ADF, fat, starch, ash, and minerals, and estimated NFC and NEL. Analytical

methods for all analyses conducted by CVAS are available at:

http://www.foragelab.com/Resources/Lab-Procedures (accessed September 14, 2016).

Concentrations of CP, NDF, ADF, NFC, NEL, starch, fat, ash, Ca and P in the TMR were

calculated based on the individual feed ingredient values and their percent inclusion in

the TMR. The diet values for RDP, RUP and NEL balance were calculated based on NRC

(2001).



collected on two consecutive days (4 consecutive milkings) during wk 4 of each period

from the p.m. and a.m. milkings. Milk component samples were collected into tubes

containing 2-bromo-2-nitropropane-1,3-diol and analyzed individually by Dairy One

Laboratory (Ithaca, NY) for fat, true protein, MUN, and lactose content using infrared

spectroscopy (Milkoscan 4000; Foss Electric, Hillerød, Denmark). Milk samples for FA

89

analysis from the 4 milkings for each period and cow were collected without

preservative, stored chilled at 4°C until composited based on milk production. The

composited milk samples were centrifuged, the milk fat was skimmed off and then stored

frozen at -20°C until analyzed using the procedure described by Rico and Harvatine

(2013). Cows were weighed 2 days at the beginning and end of the 1st and 2nd periods

with a Tru-test Eziweigh 2 scale (Mineral Wells, TX). During the 3rd experimental

period, BW was recorded daily upon exiting the milking parlor using an AfiFarm 3.04E

scale system (S.A.E. Afikim, Rehovot, Israel).

During wk 4 of each period, urine and fecal samples were collected for

digestibility and N utilization analyses. Spot urine and fecal samples (approximately 300

ml and 500 g per sample, respectively) were collected 8 times over 3 d at (d 1) 5 a.m., 12

p.m., 6 p.m., (d 2) 12 a.m., 9 a.m., 3 p.m., 9 p.m., and (d 3) 3 a.m. to obtain a


processing and analyzing can be found in Lee et al. (2012). Briefly, raw urine from each

sampling was acidified, diluted, composited by cow and period, and frozen at -20°C for

later analysis of allantoin, uric acid, creatinine, urea N and total N. Allantoin was

analyzed following the procedure by Chen et al. (1992). Stanbio Laboratory (Boerne,



samples of approximately 60 µl of 1:10 diluted and acidified urine using a Costech ECS

4010 C/N/S elemental analyzer (Costech Analytical Technologies Inc., Valencia, CA).

Fecal samples were oven dried at 65°C, ground through a 1-mm screen in a Wiley mill


90

and analyzed for DM, OM, CP, starch, NDF and ADF. A Mixer Mill MM 200 (Retsch

GmbH, Haan, Germany) was used to pulverize a 0.5 g aliquot of fecal sample for CP

analysis (N × 6.25) using a Costech ECS 4010 C/N/S elemental analyzer. Starch analysis

of fecal DM for apparent total tract digestibility was performed using a procedure similar

to the method including acetate buffer described by Hall (2009). Briefly, starch was

gelatinized with 50% NaOH, incubated for 16 h at 55°C with acetate buffer and amylase,

centrifuged, plated on a 96-well plate and then reacted with a PGO (Glucose

Oxidase/Peroxidase) enzyme solution (P7119; Sigma-Aldrich, Saint Louis, MO) for 45

min before being read at 450 nm. Neutral- and acid-detergent fiber were analyzed with an

Ankom 200 fiber analyzer (Ankom Technology Corp., Macedon, NY) based on the

procedures of Van Soest et al. (1991) with alpha amylase and sodium sulfite in the NDF

analysis. A 10-d ruminal incubation was used to analyze indigestible NDF (iNDF;

Huhtanen et al., 1994 as modified by Lee et al., 2012) in both feces and TMR, which was

used as a marker to estimate apparent digestibilities of dietary nutrients.

Enteric CH4 and CO2 emissions were analyzed during wk 4 of each period with

the GreenFeed system (C-Lock Inc., Rapid City, SD). Measurements were collected 8

times over 3 d at 9 a.m., 3 p.m., 9 p.m., 3 a.m., 12 p.m., 6 p.m., 12 a.m. and 5 a.m. to

obtain a representative sample of a 24-h period. Gas sampling procedures followed those

recommended by Hristov et al. (2015). Gas emission data were averaged by cow and

period for the statistical analysis.

91

In Situ

Ruminal disappearance of DM (alfalfa haylage, wheat, triticale, and corn silages)

and NDF (wheat and triticale silages and alfalfa haylage) was determined in situ. Six

ruminally cannulated lactating Holstein cows averaging: DMI 24.4 ± 2.4 kg; MY 36.8 ±

2.9 kg; 2.2 ± 0.4 lactations; DIM 148 ± 10 d; and BW 616 ± 40.7 kg were used for in situ

incubations. Cows were fed (% DM basis) corn silage 38.3, alfalfa haylage 13.8, grass

hay and straw mixture 4.2, ground corn 9.6, canola meal 9.6, cookie meal 5.3, roasted

soybeans 5.0, molasses 5.0, whole cotton seed 4.6, cracked corn 2.5, and mineral mix 2.1.

Oven dried forages were ground through a 4-mm screen in a Wiley mill. Approximately

7 g of sample were weighed into 10 × 20 cm nylon bags with 50-µm porosity (Ankom

Technology Corp.) and closed with a ziptie after folding. Triplicate bags were

sequentially incubated in each cow for 12, 24, 48, 72, and 96 h and simultaneously

removed. Two bags per forage were made for the 0 h time point and processed as the

incubated samples except the rumen incubation step. Upon removal from the rumen, the

bags were rinsed 3 times with cold water in a washer machine set to agitate for 6 min

each rinse. The zipties were cutoff and any remaining particles rinsed with cold tap water.

Rinsed bags were then oven dried for 72 h at 55°C before weighing for DM

determination. Samples were composited by silage, time point, and cow before NDF

analysis as previously described. Ruminal disappearance was calculated based on initial

dry weight of the incubated sample, residue dry weight, and NDF concentration of initial

sample and bag residue. Degradation curves were fit to the equation p = a + b (1 - e-ct)

where p is the degraded fraction (of DM or NDF) at time t, constant a is the soluble

92

fraction, b is the potentially degradable fraction and c is the rate of degradation of the b

fraction (Ørskov and McDonald, 1979). The effective degradability (ED) was determined

with the following equation (Ørskov and McDonald, 1979): ED = a + b {c ÷ (c + k)},

where k is the rate of passage assumed to be 0.03/h. Corn silage NDF degradability did

not fit the Ørskov and McDonald model and so was fit with a linear model. Therefore,

corn silage NDF degradability was not statistically compared to the other forages.

Income Over Feed Costs

Income over feed costs (IOFC) for the 3 diets was calculated using

the Pennsylvania State Extension Dairy Team IOFC Tool (Penn State Extension, 2015).

The cash flow spreadsheet from the Pennsylvania State Extension Dairy Team (Penn

State Extension, 2016) was used to calculate forage monetary values for the IOFC tool.

We used a 34.4 ha model dairy farm with 65 lactating cows, 10 dry cows, 52 heifers, and

12 calves. It was assumed that only the forages were grown on the farm whereas

concentrates were purchased. The lactating cow ration was changed in the scenarios to

reflect the treatment diet whereas diets for other cow groups (e.g., dry cows, heifers, and

calves) were kept the same among scenarios. First, the total amount of the different

forages required for each scenario was calculated. Next, the acres needed to produce that

amount was found by dividing the total amount of each crop needed by the per acre crop

yields typical for central PA. The corn yield for the double cropped scenarios was

decreased by 2.35 t DM/ha to account for the lower yield due to delayed corn planting.

Then, the variable costs of seed, fertilizer, and herbicide per acre for each crop during

93

2015 was entered into the spreadsheet. Along with the input costs and the yield

information for each crop, the fixed costs were allocated among the forages based on the

labor used to produce them to determine price per ton. Milk and components yield from

the current study was used with the average milk pricing in PA for 2015 to generate the

income side of the IOFC equation.


Statistical analyses for all but the in situ data were run using the MIXED

procedure of SAS v9.4 (SAS Institute Inc., Cary, NC). Cow was the experimental unit.

Milk yield and DMI from the last 10 d of the experiment were analyzed with day as a

repeated measure. The statistical model included cow, block, day, period, diet, and period

× diet and diet × day interactions. Block and cow within block were random effects with

all others fixed. Milk composition and FA, nutrient intake, digestibility, N utilization,

CH4 and CO2 emissions data used the same model without day and diet × day interaction.

Milk composition data were weighted averages based on the milk production at each

milking. Forage nutrient composition was compared using the MIXED procedure with

forage type as the model. Significance was declared at P ≤ 0.05 and tendency was

declared at 0.05 < P ≤ 0.10. If not indicated otherwise, data are presented as least squares

means.

Ruminal in situ degradation of DM and NDF was analyzed using the NLMIXED

procedure of SAS. The overall regression curve and the individual parameters (a, b, c,

94

and ED; see above) were contrasted among forages and significance was declared at P ≤

0.05.


Forages

Triticale yielded 2.79 t DM/ha and wheat yielded 2.57 t DM/ha. A survey of

triticale double cropped with corn in New York reported an average yield of 4.9 t DM/ha

from 2012-2014 (Ketterings et al., 2015). Yields could potentially have been increased in

the current study by earlier planting in late September and an extra 43 kg/ha of N

fertilizer in the spring. The cover crops were harvested at the same growth stage but, due

to different maturation patterns, wheat was harvested 5 d after triticale. From a practical

standpoint, different harvest windows can reduce risk and decrease the daily workload

during harvest which can be used to a farmer’s advantage. Triticale normally has greater

yields than wheat (Giunta et al., 1993; Estrada-Campuzano et al., 2012) as was the case

in this study. Corn silage varieties had an average yield of 18 t DM/ha. Nutrient

composition and fermentation data of the 3 silages are presented in Table 2. The triticale

silage fermentation may have benefited from a higher target ensiling DM because there

were elevated total VFA and lactic acid concentrations along with some butyric acid

production. Both silages did reach a final pH level below 4.5 but were higher (P <0.001)

than the pH of 3.68 in the corn silage. Titratable acidity followed the pattern of total

fermentation acid production. Overall, the alternative silages had higher (P ≤ 0.01) NDF,

95

ADF, lignin, CP, and soluble CP as a percent of CP content than corn silage whereas

having starch concentrations at or below 1%. The lignin content of wheat was slightly

higher (P < 0.01) than triticale while CP was higher (P < 0.001) in triticale than wheat.

Wheat silage had higher (P < 0.001) ethanol soluble carbohydrates than triticale or corn

silage.

The simple by-weight replacement of corn silage in our study was chosen to

clearly demonstrate, without confounding effects, how inclusion of wheat or triticale

cover crops used as forages in a double cropping strategy might affect cow productivity

without other changes on the farm such as increased corn grain purchases. Previous

research with cereal silages have focused on more mature crops in the dough stage when

yield and starch content are high (Sutton et al., 2001). However, wheat and triticale must

be harvested at the boot stage or earlier to produce highly digestible forage and allow

growing season for double cropping. At this stage, however, these plants accumulate little

to no starch (Fearon et al., 1990; Crovetto et al., 1998).

Dry Matter Intake and Milk Yield

Dry matter intake is known to have a large influence on milk production. In the

current study, DMI was not different between diets (Table 3); however, MY was

decreased (P = 0.01) in TS and WS compared with CS. Arieli and Adin (1994) reported

equal DMI but different MY for diets containing wheat silages harvested at flower versus

milk stage maturities. The authors explained the difference in MY by differences in

NDFD of the two silages but this would not describe the results of the current study.

96

Sinclair et al. (2005) found an increase in DMI but similar MY comparing urea-treated

wheat silage harvested in the dough stage versus corn silage. Hameleers (1998) reported

an increase in DMI with corn silage or wheat silage replacing grass silage, but no effect

on MY of around 27 kg/d.

Milk yield in the current study was high for all diets but decreased to a similar

extent for TS and WS compared with CS. A decrease in feed efficiency has been

observed in other experiments utilizing wheat silage (Sutton et al., 1998) but a significant

period by diet interaction for feed efficiency in the current study prevents conclusive

agreement with their results. The most likely cause of the decreased MY is the

replacement of starch with fiber in the alternative forage diets. When these alternative

forage cover crops partially replace corn silage in a ration, starch content is decreased

and, with it, dietary available energy (Mertens, 2009). Even though when the rations

where entered into NRC (2001), the NEL concentrations were very similar. Ground corn

or other starch sources could be added into the ration to compensate for the decreased

starch supply from corn silage, but that would likely have to be purchased on most farms.

Bernard et al. (2002) replaced corn silage with annual ryegrass silage in lactating cows

along with increased ground corn and reported no effect on DMI but an increase in MY.

Even at the early harvest date of this study, alternative forages had higher NDF, ADF,

and lignin concentrations compared with corn silage possibly causing increased rumen

fill. This may have prevented an increase in DMI for the alternative forage diets to

maintain milk production.

97

Milk Composition and Yield

Milk composition was unaffected by diet which agrees with results from

Hameleers (1998) and Sutton et al. (1998) who compared wheat silage to corn and grass

silages, respectively. However, those studies were with a high starch, dough stage wheat

silage. O’Mara et al. (1998) likewise found no effect of grass silage versus corn silage

though, again starch content was kept similar across diets. We did observe higher milk

protein yield in CS even with higher RUP levels in TS and WS, which suggests that there

was less available energy for microbial protein synthesis in the rumen with the alternative

forage diets. The tendency (P = 0.07) for a decrease in the sum of odd and branched

chain milk FA in TS and WS, compared with corn silage (see below), would also support

the expectation of lower milk true protein in those diets; Vlaeminck et al. (2006) showed

a positive correlation between odd and branched chain milk FA and bacterial N flow to

the duodenum. Despite lower starch and higher fiber intake with the alternative silage

diets, compared with the control corn silage diet, milk fat concentration or yield were

unaffected.

Milk Fatty Acid

Milk FA analysis (Table 4) revealed increased (P ≤ 0.004) concentrations of de

novo synthesized FA 4:0, and 6:0 for TS and WS, compared with CS. Stearic acid (18:0)

was also increased (P = 0.008) in TS and WS whereas trans-10 18:1 was increased (P <

0.001) in CS with a tendency (P =0.10) for greater total trans FA. Increases in trans-10

18:1 in milk fat have been associated with milk fat depression albeit at greater

98

concentrations than in the current study (Rico and Harvatine, 2013). Trans-10 18:1 is a

ruminal biohydrogenation intermediate from an alternate pathway responsible for

decreased de novo milk FA synthesis (Harvatine et al., 2009). The higher (P ≤ 0.04) odd

chain FA 15:0, 17:0, and 17:1 in CS along with the lower iso 14:0 (P = 0.04) suggests a

lower rumen pH, based on the relationship with SARA reported by Fievez et al. (2012),

though we have no direct rumen pH measurements to confirm this data. The higher fiber

and lower starch contents of TS and WS, compared with CS, may have caused

differences in ruminal biohydrogenation.

Nutrient Intake and Digestibility

Intakes of CP and ADF were higher (P ≤ 0.01) for both TS and WS, compared

with CS (Table 5). Starch intake was higher (P < 0.001) for CS, compared with either of

the alternative silage diets. These results are an outcome of the higher CP and ADF and

lower starch contents in the triticale and wheat silages than in the corn silage and the

equal DMI among diets. Apparent total tract digestibility of DM and OM was decreased

(P < 0.01) in WS, compared with CS, but not in TS. Meanwhile, NDF and ADF

digestibility was increased (P ≤ 0.005) in TS but not WS. The decreased DM and OM

digestibility of WS was likely a result of the lower starch, and higher fiber and lignin

content in the wheat silage compared with corn silage. The TS diet had lower starch and

higher fiber as WS did, but somewhat surprisingly did not decrease DM or OM

digestibility. Likely the lower lignin level in the triticale silage over the wheat silage

improved TS digestibility. It has been shown that DM and NDF digestibility are

99

negatively correlated with lignin content (Jung et al., 1997). O’Mara et al. (1998)

reported higher diet digestibilities of DM, OM, and NDF but not N when grass silage

replaced corn silage in lactiating cow diets which match the results of TS in the current

experiment. Less starch in TS and WS probably resulted in higher rumen pH (though it

was not measured directly in this experiment) which improves fiber digestibility (Firkins,

1997) although the high lignin in the wheat silage may have counteracted that effect.

Starch digestibility was high for all diets but highest (P < 0.001) for CS followed by TS

and then WS. The rate of starch digestibility in corn silage is usually high and can be

faster than that of fine ground corn (Lanzas et al., 2007). The CS diet had the highest

proportion of starch from corn silage and this was likely the reason for the increased

starch digestibility in CS.

In Situ

In situ DM disappearance curves are shown in Figure 1. Numerical ranking of in

situ ED of DM among triticale, wheat, and corn silages matches apparent total tract DM

and OM digestibility of TS, WS, and CS. Because of its high starch content and despite

its low NDF degradation rate, corn silage still had one of the highest ED of DM along

with triticale silage. Wheat silage ED of DM was slightly lower (P < 0.001) than triticale

and corn silage, but higher (P < 0.001) than alfalfa haylage. Although ED of DM was not

drastically different among triticale, wheat, and corn silages, the differences in the

nutrient composition of the silage DM may have affected cow performance among the

experimental diets. The wheat and triticale silage contained higher ash and CP content

100

whereas corn silage had higher starch content, potentially resulting in more digestible

energy for supporting higher milk production in CS.

Ruminal in situ NDF degradability data (Figure 2) revealed that triticale silage

had the highest (P < 0.001) ED of NDF followed by wheat silage and alfalfa haylage.

Triticale is a cross between wheat and rye, therefore, it was expected that the NDF

degradability pattern would be similar to wheat. Triticale and wheat silages had nearly

identical NDF and ADF contents but wheat silage had higher lignin concentrations. This

appeared to have negatively affected its in situ ED of NDF and the DM and OM apparent

total tract digestibilities of WS. It may be that the wheat was actually more

physiologically mature than the triticale, since we harvested the wheat a few days after

the triticale even though we targeted the same maturity at harvest. Corn silage NDF

degradability data did not fit the Orskov and McDonald model, but it did fit to a linear

model with a rate of 0.57%/h and an R2 = 0.94. Filya (2003) reported corn silage in situ

NDF degradability values at 48 h of approximately 22.4%, which is similar to our 28.6%

value. For triticale or wheat silage at the boot stage to successfully replace part of the

corn silage in a lactating cow diet, NDF degradability has to be high since it is the main

source of energy in these silages.

N Utilization

Both TS and WS had a similar effect on N utilization (Table 6) with higher (P =

0.006) N intake than CS. Cows fed TS and WS excreted more urinary urea N (P = 0.001)

and total excreta N (P = 0.006) while having a tendency (P = 0.08) for less N secreted in

101

milk than CS. Together, this led to a lower (P = 0.005) milk N efficiency for TS and WS,

compared with CS. Hameleers (1998) also reported a decrease in milk N efficiency when

grass silage was replaced with urea-treated wheat silage. Similar to that experiment, we

formulated our CS diet to meet MP requirements for the level of production of the cows

and did not adjust protein content of the alternative forage diets. Therefore, TS and WS

supplied excess CP, as a result of the higher CP content of the wheat and triticale silages,

which was not utilized but mainly excreted in urine as urea. This presents an opportunity

to reduce purchased protein feeds and thereby reduce feed costs in the TS and WS diets

(Bernard et al., 2002; O’Mara et al., 1998). Purine derivatives excretion in urine, which is

positively correlated with microbial protein flow to the intestine (Chen, 1989), was not

different among treatments. Although this did not result in equal milk true protein yields

among diets.

Enteric CH4 and CO2 Emissions

Enteric CH4 production was not different among diets (Table 7). Methane yield

(i.e., per kg of DMI) or intensity (i.e., per kg of MY) were also not different among diets.

However, when calculating CH4 intensity per kg of ECM, CS yielded significantly less

(P = 0.04) CH4 than TS. Increasing starch in a diet favors propionate production and

thereby usually reduces CH4 production (Moe and Tyrrell, 1979; Moss et al., 2000)

though there have been reports of increased CH4 production with increased dietary starch

(Beever et al., 1988). Decreased CH4 yield in beef cattle was clearly demonstrated by

McGeogh et al. (2010) when whole crop wheat silage diets of increasing grain (i.e.

102

starch) contents were fed. Higher starch in CS was likely the cause of the lower expected

CH4 emission intensity per kg of ECM. The enteric CH4 yield and intensity values we

report, of around 17.5 g/kg DMI and 12 g/kg MY, are similar to data from our Penn State

dairy herd (Hristov et al., 2015) and others (Colombini et al., 2015). Carbon dioxide

emissions were decreased (P = 0.006) by WS and TS. Kirchgessner et al. (1991) reported

a relationship of CO2 emissions to MY of 0.14 kg CO2/d for every kg of milk produced.

Using this relationship, a decrease of around 0.2 kg CO2/d would have been expected for

the current study, yet a 0.6 kg CO2/d decrease in emissions was measured. However, the

direction of the effect was in agreement with the findings of Kirchgessner et al. (1991).

Kinsman et al. (1995) also reported a strong correlation between CO2 emissions and milk

production (r = 0.74; P < 0.001). They noted that CO2 production is primarily from

cellular respiration of the cow and only secondarily from ruminal fermentation. There is

some interest in using the ratio of enteric CH4 emission to CO2 emission as a way to

measure feed efficiency (Madsen et al., 2010). Lower values would indicate less CH4

production and more complete metabolization of the C to CO2 (Madsen et al., 2010).

Ratios reported for this study are lowest (P <0.01) for CS and not different between TS

and WS indicating CS may have been a more completely metabolizable diet in the

rumen.

Income Over Feed Cost

The IOFC of CS was $11.05 and decreased to $10.39 and $10.26 for WS and TS,

respectively. Decreased per acre corn silage yield due to later corn planting and decreased

103

MY caused the decrease in IOFC for WS and TS. The higher IOFC for WS over TS was

due to the numerically higher milk and milk fat yield resulting in higher calculated

income. The WS and TS diets were not least cost formulations and did not fully utilize

the protein value of the alternative forages as indicated by the higher MUN and urinary

urea nitrogen losses. Likely, the supplemental protein content of WS and TS could be

decreased to lower costs of on-farm rations.

CONCLUSIONS

In this study, we demonstrated that triticale and wheat cover crops harvested as

silage at the boot stage can support MY above 41 kg/d when included at 10% of the diet

DM replacing corn silage. Triticale and wheat silage inclusion did not affect DMI but

decreased MY likely due to replacing starch with fiber. Higher CP content in the

alternative forages along with lower starch resulted in higher urinary urea excretion,

higher MUN concentration and lower milk N efficiency. Enteric CH4 emission per kg of

ECM was increased by TS. Triticale silage had higher in situ effective degradability of

NDF and a slightly higher crop yield than wheat silage, although IOFC was slightly more

favorable for wheat silage due to numerically higher MY and true milk protein content.

Both alternative forages provide a highly digestible source of fiber that can successfully

replace corn silage at low inclusion rates. For dairy farms in need of more forage, triticale

or wheat double cropped with corn silage may be an appropriate cropping strategy.

104

ACKNOWLEDGEMENT

This project supported by the Northeast Sustainable Agriculture Research and






experimental cows. Additionally, we thank Virginia Ishler for her help with the IOFC

analysis.

REFERENCES

Arieli A., and G. Adin. 1994. Effect of wheat silage maturity on digestion and milk yield

in dairy cows. J. Dairy Sci. 77:237-243.


Effect of cultivar and stage of maturity on yield and degradability in situ. J. Agric.

Food Chem. 45:709-712.

Beever, D. E., S. B. Cammell, C. Thomas, M. C. Spooner, M. J. Haines, and D. L. Gale.

1988. The effect of date of cut and barley substitution on gain and on the efficiency

of utilization of grass silage by growing cattle 2. Nutrient supply and energy

partition. Br. J. Nutr. 60:307-319.

Bernard, J. K., J. W. West, and D. S. Trammell. 2002. Effect of replacing corn silage

with annual ryegrass silage on nutrient digestibility, intake, and milk yield for

lactating dairy cows. J. Dairy Sci. 85:2277-2282.

Carey, P. L., K. C. Cameron, H. J. Di, G. R. Edwards, and D. F. Chapman. Sowing a

winter catch crop can reduce nitrate leaching losses from winter-applied urine under

simulated forage grazing: a lysimeter study. Soil Use Manag. 32:329-337.

105


Substitution of corn silage with sorghum silages in lactating cow diets: In vivo

methane emission and global warming potential of milk production. Agric. Syst.

136:106–113.

Chen, X. B. 1989. Excretion of purine derivatives by cattle and sheep and its use for the

estimation of absorbed microbial protein. Ph.D. Dissertation. Aberdeen Univ.. UK.



supply in sheep. J. Anim. Sci. 70:1534–1542.

Crovetto, G. M., G. Galassi, L. Rapetti, A. Sandrucci, and A. Tamburini. 1998. Effect of

the stage of maturity on the nutritive value of whole crop wheat silage. Livest. Prod.

Sci. 55:21-32.

Di, H. J., and K. C. Cameron. 2002. Nitrate leaching in temperate agroecosystems:

sources, factors and mitigating strategies. Nutrient Cycling in Agroecosystems.

64:237–256.

Estrada-Campuzano, G., G. A. Salfer, and D. J. Miralles. 2012. Differences in yield,

biomass and their components between triticale and wheat grown under contrasting

water and nitrogen environments. Field Crops Res. 128:167-179.

Fearon, A. L., A. Felix, and V. T. Sapra. 1990. Chemical composition and in vitro dry

matter and organic matter digestibility of triticale forage. J. Agron. Crop Sci.

164:262-270.

Fievez V., E. Colman, J.M. Castro-Montoya, I. Stefanov, and B. Vlaeminck. 2012. Milk

odd- and branched-chain fatty acids as biomarkers of rumen function- An update.

Anim. Feed Sci. and Technol. 172:51–65.

Filya, I. 2003. The effect of Lactobacillus buchneri, with or without homofermentative

lactic acid bacteria, on the fermentation, aerobic stability and ruminal degradability

of wheat, sorghum and maize silage. J. Appl. Microbiol. 95:1080-1086.

Firkins, J. L. 1997. Effects of feeding nonforage fiber sources on site of fiber digestion. J.

Dairy Sci. 80:1426–1437.

Giunta, F., R. Motzo, and M. Deidda. 1993. Effect of drought on yield and yield

components of durum wheat and triticale in a Mediterranean environment. Field

Crops Res. 33:399-409.

106


feeds: comparison of methods and a method recommended for AOAC collaborative

study. J. AOAC Int. 92:42-49.

Harvatine, K. J., Y. R. Boisclair, and D. E. Bauman. 2009. Recent advances in the

regulation of milk fat synthesis. Animal. 3:40–54.

Hameleers, A. 1998. The effects of the inclusion of either maize silage, fermented whole

crop wheat or urea-treated whole crop wheat in a diet based on a high-quality grass

silage on the performance of dairy cows. Grass Forage Sci. 53:157-163.

Hristov, A. N., J. Oh, F. Giallongo, T. W. Frederick, M. T. Harper, H. L. Weeks, A. F.

Branco, P. J. Moate, M. H. Deighton, S. Richard, O. Williams, M. Kindermann, and

S. Duval. 2015. An inhibitor persistently decreased enteric methane emission from

dairy cows with no negative effect on milk production. Proc. Natl. Acad. Sci. USA.

112:10663–10668.




Jung, H. G., D. R. Mertens, and A. J. Payne. 1997. Correlation of acid detergent lignin

and klason lignin with digestibility of forage dry matter and neutral detergent fiber.

J. Dairy Sci. 80:1622–1628.

Ketterings, Q. M., S. Ort, S. N. Swink, G. Godwin, T. Kilcer, J. Miller, and W. Verbeten.

2015. Winter cereals as double crops in corn rotations on New York dairy farms. J.

Agric. Sci. 7:18-25.


cereal grain silage substituted for alfalfa silage on performance of lactating dairy

cows. J. Dairy Sci. 76:3536-3546.

Khorasani, G. R., P. E. Jedel, J. H. Helm, and J. J. Kennelly. 1997. Influence of stage of

maturity on yield components and chemical composition of cereal grain silages.

Can. J. Anim. Sci. 77:259-267.

Kinsman, R., F. D. Sauer, H. A. Jackson, and M. S. Wolynetz. 1995. Methane and carbon

dioxide emissions from dairy cows in full lactation monitored over a six-month

period. J. Dairy Sci. 78:2760-2766.

Kirchgessner, M., W. Windisch, H. L. Muller, and M. Kreuzer. 1991. Release of methane

and of carbon dioxide by dairy cattle. Agribiol. Res. 44:91-102.

107

Lanzas, C., C. J. Sniffen, S. Seo, L. O. Tedeschi, and D. G. Fox. 2007. A revised CNCPS

feed carbohydrate fractionation scheme for formulating rations for ruminants.





deficient diet. J. Dairy Sci. 95:6042–6056.

Madsen, J., B. S. Bjerg, T. Hvelplund, M. R. Weisbjerg, and P. Lund. 2010. Short

communication: Methane and carbon dioxide ratio in excreted air for quantification

of the methane production from ruminants. Livest. Sci. 129:223-227.

McGeough, E. J., P. O’Kiely, K. J. Hart, A. P. Moloney, T. M. Boland, and D. A. Kenny.

2010. Methane emissions, feed intake, performance, digestibility, and rumen

fermentation of finishing beef cattle offered whole-crop wheat silages differing in

grain content. J. Anim. Sci. 88:2703–2716.

Mertens, D. R. 2009. Impact of NDF content and digestibility on dairy cow performance.

Pages 191-201 in Proc. 27th Annual Western Canadian Dairy Seminar. Red Deer,

Alberta.

Moe, P. W., and H. F. Tyrrell. 1979. Methane production in dairy cows. J. Dairy Sci.

62:1583-1586.

Moss, A. R., J. Jouany, and J. Newbold. 2000. Methane production by ruminants: its

contribution to global warming. Ann. Zootech. 49:231-253.



O’Mara, F. P., J. J. Fitzgerald, J. J. Murphy, and M. Rath. 1998. The effect on milk

production of replacing grass silage with maize silage in the diet of dairy cows.

Livest. Prod. Sci. 55:79-87.



Agric. Sci. (Camb.) 92:499–503.

The Pennsylvania State University. 2015. The Penn State Agronomy Guide. University

Park, PA.



fiber and oil concentration. J. Dairy Sci. 96:6621–6630.

108

Shepherd, M. A. 1999. The effectiveness of cover crops during eight years of a UK

sandland rotation. Soil Use Manag. 15:41-48.

Sinclair, L. A., M. A. Jackson, J. A. Huntington, and R. J. Readman. 2005. The effects of

processed, urea-treated whole-crop wheat or maize silage and supplementation of

whole-crop wheat on the performance of dairy cows. Livest. Prod. Sci. 95:1 –10.





Sutton, J. D., S. B. Cammell, D. E. Beever, D. J. Humphries, and R. H. Phipps. 1998.

Energy and nitrogen balance of lactating dairy cows given mixtures of urea-treated

whole-crop wheat and grass silage. Animal Science. 67:203-212.

Sutton, J. D., R. H. Phipps, S. B. Cammell, and D. J. Humphries. 2001. Attempts to

improve the utilization of urea-treated whole-crop wheat by lactating dairy cows.

Animal Science. 73:137-147.



J. Dairy Sci. 74:3583–3597.

Vlaeminck, B., V. Fievez, A. R. J. Cabrita, A. J. M. Fonseca, and R. J. Dewhurst. 2006.

Factors affecting odd- and branched-chain fatty acids in milk: A review. Anim.


109


1CS = Corn silage control diet; TS = Triticale silage diet; WS = Wheat silage diet. 2Corn silage was 38.5% DM and contained (DM basis) 6.4% CP, 46.6% NFC, 34.5%

starch, and 41.0% NDF. 3Alfalfa haylage was 46.0% DM and contained (DM basis) 21.0% CP, 24.0% NFC, and


contained (%, as-is basis) trace mineral mix, 0.86; MgO (56% Mg), 8.0; NaCl, 6.4;

vitamin ADE premix (Cargill Animal Nutrition, Cargill Inc.), 0.48; limestone, 37.2;

selenium premix (Cargill Animal Nutrition, Cargill Inc.), 0.07; and dry corn distillers

Diet1

Item CS TS WS

Ingredient, % of DM

Corn silage2 44 34 34

Alfalfa haylage3 8 8 8

Triticale silage - 10 -

Wheat silage - - 10

Hay/Straw mixture 5 5 5

Cottonseed hulls 4 4 4

Ground corn 9 9 9

Heat-treated whole soybeans 7.5 7.5 7.5

Solvent-extracted canola meal 8 8 8

SoyPLUS4 7 7 7

Molasses5 4.5 4.5 4.5

Mineral/vitamin premix6 3 3 3


CP7 16.1 17.2 16.9

RDP8 9.7 10.4 10.0

RUP8 6.4 6.8 6.9

NDF7 33.5 34.5 34.5

iNDF9 12.0 11.7 12.8

ADF7 21.7 22.6 22.5

NFC8 43.1 40.3 40.8

Starch7 22.7 19.3 19.3

Fat7 4.9 4.9 4.9

NEL,8 Mcal/kg 1.56 1.57 1.55

NEL intake,8 Mcal/d 42.5 43.6 42.7

NEL balance,8 Mcal/d 1.7 3.5 3.2

Ash7 6.7 7.3 7.2

Ca7 0.8 0.8 0.8

P7 0.4 0.4 0.4

110

grains with solubles, 46.7. Ca, 14.1%; P, 0.39%; Mg, 4.60%; K, 0.45%; S, 0.38%; Se,

6.67 mg/kg; Cu, 358 mg/kg; Zn, 1,085 mg/kg; Fe, 188 mg/kg, vitamin A, 262,656 IU/kg;

vitamin D, 65,559 IU/kg; and vitamin E, 1,974 IU/kg. 7Values calculated using the chemical analysis (Cumberland Valley Analytical Services

Inc., Maugansville, MD) of individual feed ingredients of the diet. 8Estimated based on NRC (2001). 9Determined by 10d ruminal incubation.

111

Table 4-2. Nutrient composition and fermentation profile of triticale and wheat silages (%

of DM or as indicated)1

1Three composite samples per silage, one for each experimental period were analyzed by

wet chemistry (Cumberland Valley Analytical Services Inc., Maugansville, MD). Mean ±

SE is reported. 2Largest SEM published in table; n = 9 (n represents the number of observations used in

the statistical analysis). 3ND = Not detected.

Forages P-Value

Item Corn Triticale Wheat SEM2 Forage

DM, % 38.5a 30.7b 40.7a 1.42 <0.01

NDF 41.0b 51.1a 51.0a 0.96 <0.001

ADF 23.7b 32.9a 32.5a 0.55 <0.001

Lignin 2.82c 3.47b 3.83a 0.103 <0.01

Fat 3.38 3.89 3.57 0.281 0.48

CP 6.4c 17.3a 14.6b 0.32 <0.001

Soluble CP, % of CP 61.8c 80.8a 74.6b 1.07 <0.001

Starch 34.5a 0.3b 1.0b 0.96 <0.001

Ethanol soluble

carbohydrates 1.0b 2.1b 4.6a 0.32 <0.001

Ash 3.76c 9.85a 8.35b 0.277 <0.001

Ca 0.18c 0.43a 0.28b 0.016 <0.001

P 0.24c 0.42a 0.33b 0.013 <0.001

K 1.06c 4.34a 2.99b 0.077 <0.001

pH 3.68b 4.48a 4.46a 0.026 <0.001

Fermentation acids

Lactic 4.17b 7.03a 6.43a 0.309 <0.01

Acetic 1.20b 3.34a 2.29ab 0.435 0.04

Propionic 0.02b 0.49a 0.09b 0.043 <0.001

Butyric ND3 0.85 ND - -

Isobutyric ND 0.53 ND - -

Titratable acidity,

meq/100g 5.82b 8.26a 5.00b 0.486 <0.01

112

Table 4-3. Effect of triticale and wheat silage on DMI, milk production, and feed

efficiency in lactating dairy cows

Diet1 P-Value

Item CS TS WS SEM2 Diet

DMI, kg/d 27.2 27.7 27.6 1.80 0.37

Milk yield, kg/d 42.7a 41.2b 41.4b 5.18 0.01

Milk ÷ DMI, kg/kg 1.61a 1.55b 1.52b 0.16 0.033

Milk fat, % 3.77 3.80 3.80 0.14 0.93

Milk fat, kg/d 1.60 1.52 1.53 0.11 0.11

Milk true protein, % 2.96 2.95 2.97 0.04 0.91

Milk true protein, kg/d 1.27a 1.20b 1.20b 0.10 0.02

Lactose, % 4.94 4.88 4.88 0.05 0.11

Lactose, kg/d 2.14a 2.00b 1.98b 0.17 0.008

MUN, mg/dL 10.8b 12.7a 13.1a 0.53 <0.001

ECM4, kg/d 40.9a 38.6b 38.5b 2.97 0.05

BW, kg 634 629 633 30.3 0.30 a,bMeans within the same row without a common superscript differ (P ≤ 0.05). 1CS = Corn silage control diet; TS = Triticale silage diet; WS = Wheat silage diet. 2Largest SEM published in table. DMI, n = 360; milk yield, n = 335; milk yield ÷ DMI, n

= 335; BW, n = 36; milk composition data, n = 36 (n represents the number of

observations used in the statistical analysis). 3Period*Diet interaction P-Value = 0.009. 4Energy-corrected milk (kg/d) = kg of milk × [(38.3 × % fat × 10 + 24.2 × % true protein


113

Table 4-4. Effect of triticale and wheat silage on milk fatty acid composition (g/100 g of

total fatty acids) in lactating dairy cows

Diet1 P-Value

Fatty Acid CS TS WS SEM2 Diet

4:0 4.02b 4.44a 4.52a 0.102 0.008

6:0 2.24b 2.41a 2.42a 0.051 0.004

8:0 1.26b 1.33a 1.34a 0.036 0.033

10:0 2.86 2.90 2.87 0.101 0.83

10:1 0.26 0.28 0.27 0.013 0.08

11:0 0.07a 0.06b 0.05b 0.006 0.001

12:0 3.21 3.20 3.14 0.112 0.65

13:0 iso 0.03 0.03 0.03 0.001 0.48

13:0 anteiso 0.07 0.07 0.07 0.004 0.45

13:0 0.12a 0.10b 0.09b 0.008 0.001

14:0 iso 0.08b 0.10a 0.09ab 0.007 0.04

14:0 10.3 10.3 10.1 0.200 0.44

14:1 0.84 0.83 0.79 0.054 0.13

15:0 iso 0.19b 0.21a 0.20a 0.006 <0.001

15:0 anteiso 0.38 0.40 0.39 0.009 0.06

15:0 1.06a 0.96b 0.90b 0.046 0.003

16:0 iso 0.18b 0.21a 0.21a 0.015 <0.001

16:0 27.0a 26.5a 25.7b 0.458 0.02

16:1 1.19 1.09 1.10 0.068 0.09

17:0 iso 0.28 0.28 0.29 0.007 0.35

17:0 anteiso 0.38 0.38 0.39 0.007 0.28

17:0 0.54a 0.50b 0.50b 0.011 0.03

17:1 0.18a 0.16b 0.17ab 0.010 0.043

18:0 11.7b 12.5a 12.6a 0.416 0.008

trans-4 18:1 0.03 0.03 0.03 0.001 0.94

trans-5 18:1 0.02 0.02 0.02 0.001 0.79

trans-6,8 18:1 0.34a 0.31b 0.31b 0.009 0.003

trans-9 18:1 0.24 0.23 0.23 0.006 0.80

trans-10 18:1 0.68a 0.45b 0.44b 0.051 <0.001

trans-11 18:1 1.07 1.12 1.16 0.052 0.27

trans-12 18:1 0.28 0.32 0.27 0.024 0.31

cis-9 18:1 19.2 18.9 19.7 0.493 0.09

trans-15 18:1 0.27 0.28 0.26 0.012 0.17

cis-11 18:1 0.81a 0.69b 0.71b 0.036 <0.0013

cis-12 18:1 0.45a 0.42b 0.42b 0.011 <0.001

trans-16 18:1 0.40 0.41 0.39 0.010 0.18

Linoleic acid 3.42b 3.39b 3.55a 0.145 0.01

α-Linolenic acid 0.54c 0.60b 0.62a 0.021 <0.001

20:0 0.13b 0.15a 0.15a 0.004 <0.001

114

20:1 0.090b 0.094a 0.095a 0.003 0.013

cis-9,trans-11

CLA 0.54 0.52 0.52 0.024 0.69

trans-10,cis-12

CLA 0.002 0.001 0.000 0.001 0.21

20:2 0.04 0.05 0.04 0.002 0.13

20:3 0.14 0.14 0.14 0.008 0.69

20:4 0.17 0.16 0.16 0.005 0.22

20:5 0.03b 0.04a 0.04a 0.001 <0.001

22:0 0.05b 0.06a 0.06a 0.002 <0.001

24:0 0.026b 0.032a 0.033a 0.002 <0.001

24:1 0.001 0.001 0.001 0.001 0.64

22:4 0.04 0.03 0.04 0.002 0.38

22:5 0.10 0.09 0.09 0.007 0.57

Σ 16 28.3a 27.8a 27.0b 0.474 0.02

Σ SFA 66.1 67.0 66.1 0.642 0.09

Σ MUFA 26.4 25.6 26.3 0.543 0.11

Σ PUFA 5.01b 5.02b 5.20a 0.178 0.03

Σ trans FA 3.86 3.68 3.62 0.125 0.10

Σ OBCFA4 3.53 3.45 3.39 0.051 0.07

Unknown 2.50 2.43 2.37 0.108 0.63 a,bMeans within the same row without a common superscript differ (P < 0.05). 1CS = Corn silage control diet; TS = Triticale silage diet; WS = Wheat silage diet. 2Largest SEM shown; n = 36 (n represents number of observations used in the statistical

analysis). Data are presented as LSM. 3Period*Diet interactions: 8:0 = 0.05; 17:1 = 0.03; cis-11 18:1 = 0.04; 20:1 = 0.02. 4Sum of the odd and branched chain fatty acids (iso13:0, anteiso13:0, 13:0, iso14:0,

iso15:0, anteiso15:0, 15:0, iso16:0, iso17:0, 17:0, 17:1).

115

Table 4-5. Effect of triticale and wheat silage on nutrient intake and apparent total-tract


Diet1 P-Value


Intake, kg/d

DM3 26.6 27.0 27.2 1.53 0.62

OM 24.8 25.0 25.2 1.42 0.76

CP 4.28b 4.65a 4.60a 0.26 0.006

Starch 7.33a 5.77b 5.62b 0.36 <0.001

NDF 8.85 9.18 9.25 0.52 0.17

ADF 5.61b 5.99a 6.01a 0.33 0.01


DM 66.9a 67.0a 65.1b 0.75 0.01

OM 67.9a 68.2a 66.3b 0.73 0.01

CP 64.4 65.0 63.0 1.04 0.14

Starch 99.5a 99.3b 99.1c 0.05 <0.001

NDF 44.2b 47.0a 42.9b 1.07 0.005

ADF 37.1b 41.9a 36.8b 1.27 0.003 a,b,c Means within the same row without a common superscript differ (P < 0.05). 1CS = Corn silage control diet; TS = Triticale silage diet; WS = Wheat silage diet. 2Largest SEM published in table; n = 36 (n represents the number of observations used in

the statistical analysis). 3DM intake reported is during the fecal collection periods.

116

Table 4-6. Effect of triticale and wheat silage on nitrogen utilization and urinary purine

derivatives in lactating dairy cows

Diet1 P-Value


N intake, g/d 685b 743a 736a 40.9 0.006

N excretion or secretion, g/d 594b 639a 633a 38.2 0.03

Urine N, g/d 157b 191a 173ab 12.7 0.009

UUN, g/d 121b 157a 155a 11.0 0.001

Fecal N, g/d 241b 261ab 275a 15.7 0.013

Total excreta N, g/d 398b 452a 447a 25.2 0.006

Milk N, g/d 196 188 186 15.7 0.08


Urine N 23.8 26.6 24.4 1.7 0.15

Fecal N 35.6 34.9 37.3 1.05 0.08

Total excreta N 59.4 61.5 61.7 2.51 .49

Milk N 29.4a 26.4b 26.2b 2.21 0.005

Urine output4, kg/d 17.9b 23.5a 22.4a 2.13 0.03


Allantion 452 502 493 48.0 0.61

Uric acid 65.4 68.4 67.7 6.72 0.87

Total PD 517 571 561 53.3 0.62 a,bMeans within the same row without a common superscript differ (P < 0.05). 1CS = Corn silage control diet; TS = Triticale silage diet; WS = Wheat silage diet. 2Largest SEM published in table; n = 36 (n represents the number of observations used in

the statistical analysis). 3Period*Diet interaction P-Value = 0.04. 4Estimated from urine creatinine concentration, assumed to be excreted at 29 mg/kg of

BW. 5Purine derivatives.

117

Table 4-7. Effect of triticale and wheat silage on carbon dioxide (CO2) and methane

(CH4) emissions1 in lactating dairy cows

Diet2 P-Value


CO2 kg/d 13.5a 12.9b 12.8b 0.57 0.006

CH4, g/d 455 491 463 23.4 0.16

CH4, g/kg of DMI4 16.7 18.0 17.2 0.91 0.21

CH4, g/kg of MY4 11.6 12.3 12.0 1.03 0.61

CH4, g/kg of ECM4 11.7b 13.0a 12.5ab 0.74 0.04

CH4/CO2, g/kg 33.8b 38.1a 36.2a 0.88 0.005 a,bMeans within the same row without a common superscript differ (P < 0.05). 1Rumen gas emissions were measured using GreenFeed (C-Lock Technology Inc., Rapid


period. 2CS = Corn silage control diet; TS = Triticale silage diet; WS = Wheat silage diet. 3Largest SEM published in table; n = 36 (n represents the number of observations used in

the statistical analysis). 4Based on milk yield and DMI data during the sampling periods.

118

Triticale and Wheat DM Disappearance


0 20 40 60 80 100

DM

dis

app

eara

nce

, %

30

40

50

60

70

80

90

Alfalfa Haylage

Corn Silage

Triticale Silage

Wheat Silage




fraction (of DM) at time t where a is the soluble fraction, b is the potentially degradable


Effective degradability (ED) was estimated as: ED = a + b {c ÷ (c + k)}, where a, b, and

p are as above and k is the rate of passage (Ørskov and McDonald, 1979) assumed to be

0.03/h in this study: ED (model estimates ± SE) = alfalfa haylage 55.7 ± 0.48c; corn

silage 64.3 ± 0.46a; triticale silage 65.3 ± 0.49a; wheat silage 62.0 ± 0.48b. a,b,c Means

within the same row without a common superscript differ (P < 0.05).

119

Triticale and Wheat NDF Disappearance


0 20 40 60 80 100

ND

F d

isap

pea

ran

ce,

% o

f N

DF

0

10

20

30

40

50

60

70

80

Alfalfa Haylage

Corn silage

Triticale Silage

Wheat Silage


Data are means ± SE (n = 6). Disappearance curves of alfalfa haylage, triticle silage, and

wheat silage were fit using SigmaPlot 10.0 (Systat Software, Chicago, IL) to the equation

p = a + b (1 - e-ct), where p is the degraded fraction (of NDF) at time t where a is the

soluble fraction, b is the potentially degradable fraction, and c is the rate of degradation

of the b fraction (Ørskov and McDonald, 1979). Corn silage degradation data were fit to

a linear model with an R2 = 0.94. Effective degradability (ED) was estimated as: ED = a

+ b {c ÷ (c + k)}, where a, b, and p are as above and k is the rate of passage (Ørskov and

McDonald, 1979) assumed to be 0.03/h in this study: ED (model estimates ± SE) =

alfalfa haylage 29.0 ± 0.78c; triticale silage 40.8 ± 0.78a; wheat silage 36.1 ± 0.78b. a,b,c

Means within the same row without a common superscript differ (P < 0.05).

Chapter 5

Inclusion of brown midrib dwarf pearl millet silage in the diet of lactating

dairy cows.

Journal of Dairy Science Vol. 101, 2018

https://doi.org/10.3168/jds.2017-14036

M. T. Harper, A. Melgar, J. Oh, K. Nedelkov, G. Sanchez, G. W. Roth, and A. N. Hristov

ABSTRACT

Brown midrib brachytic dwarf pearl millet, Pennisetum glaucum, forage

harvested at the flag leaf visible stage and subsequentially ensiled, was investigated as a

partial replacement of corn silage in the diet of high producing dairy cows. Seventeen

lactating Holstein cows were fed 2 diets in a crossover design experiment with 2 periods

of 28 d each. Both diets had forage to concentrate ratios of 60:40. The control diet (CSD)

was based on corn silage and alfalfa haylage and in the treatment diet, 20% of the corn

silage dry matter (corresponding to 10% of the dietary dry matter) was replaced with

pearl millet silage (PMD). The effects of partial substitution of corn silage with pearl

millet silage on dry matter intake, milk yield, milk components, and fatty acid profile,

apparent total-tract digestibility of nutrients, N utilization, and enteric methane emissions

were analyzed. The pearl millet silage was higher in crude protein and neutral detergent

fiber and lower in lignin and starch than the corn silage. Diet did not affect dry matter

121

intake or energy corrected milk yield which averaged 46.7 ± 1.92 kg/d. The PMD

treatment tended to increase milk fat concentration, had no effect on milk fat yield, and

increased milk urea N. Concentrations and yields of milk protein and lactose were not

affected by diet. Apparent total-tract digestibility of dry matter decreased from 66.5% in

CSD to 64.5% in PMD. Similarly, organic matter and crude protein digestibility was

decreased by PMD, whereas neutral- and acid-detergent fiber digestibility was increased.

Total milk trans fatty acid concentration was decreased by PMD with a particular

decrease in trans-10 18:1. Urinary urea and fecal N excretion increased with PMD

compared with CSD. Milk N efficiency decreased with PMD. Carbon dioxide emission

was not different between the diets, but PMD increased enteric methane emission from

396 to 454 g/d and increased methane yield and intensity.

Substituting corn silage with brown midrib dwarf pearl millet silage at 10% of the

diet dry matter supported high milk production in dairy cows. When planning on farm

forage production strategies, brown midrib dwarf pearl millet should be considered as a

viable fiber source.

Keywords: dairy cow, methane, milk fat, pearl millet silage

122

INTRODUCTION

Dairy farms in the northeast U.S. rely on homegrown forages for the foundation

of their rations (Wolf, 2003). The amount and quality of the harvested forage affects

profitability and hence viability of dairy farms and particularly of smaller farms (< 200

cows) which may have fewer economies of scale or opportunity to purchase additional

hectares (Gillespie et al., 2010). Corn silage has qualities (e.g., high yield, simplicity of

harvest, high energy content) that make it a popular forage for dairy cows in the U.S.

However, the long growing season of corn may prevent double cropping with a winter

annual such as triticale in some areas that have shorter growing seasons.

Pearl millet is a warm season annual grass that has a shorter growing season than

corn (65 d vs. 130 d, respectively) and may be more practical for double cropping

strategies in certain northern geographic regions and years (e.g. abnormally wet spring

seasons which delay corn planting). Additionally, pearl millet is drought-tolerant with a

high water use efficiency which is a particularly important trait for crops planted after

winter annual cereals (cover crops) that may decrease the available soil moisture (Maman

et al., 2003; Zegada-Lizarazu and Iijima, 2005). Furthermore, inclusion of pearl millet

into a cropping plan may reduce corn disease pressure by rotating away from continuous

corn every few years (Thomison et al., 2011). To support high milk production of

lactating dairy cows, the harvested forage must be highly digestible (Van Soest, 1994).

There is a brown midrib (BMR) phenotype with lower lignin content and increased

digestibility in certain pearl millet varieties (Cherney et al., 1990; Mustafa et al., 2004;

Hassanat et al., 2007).

123

Messman et al. (1992) reported that milk yield (MY) of mid-lactation dairy cows

was maintained when pearl millet replaced corn silage in the diet. In another study with

lactating cows, MY was unaffected but yield of ECM was increased by feeding pearl

millet silage at 35% of the diet DM instead of corn silage (Amer and Mustafa, 2010). On

the other hand, in an experiment by Brunette et al. (2014) MY responses to 2 pearl millet

cultivars were inconsistent; feeding regular pearl millet silage decreased MY but feeding

sweet pearl millet silage did not affect MY when both were compared with corn silage. A

more recent study by Brunette et al. (2016) compared grass silage to either pearl millet

silage harvested in either the vegetative or mature stage. They reported no differences in

MY between the grass and pearl millet silage harvested in the vegetative stage, but they

did see a decrease in MY by the pearl millet harvested at the mature stage. To the

authors’ knowledge, there have been no studies investigating feeding BMR pearl millet to

lactating cows.

Therefore, the hypothesis of this study was that BMR brachytic dwarf pearl millet

silage could serve as alternative forage to partially substitute corn silage in lactating dairy

cow rations in the northeastern U.S. The objective of the experiment was to partially

replace corn silage with BMR dwarf pearl millet silage at 10% of the diet DM, and

investigate the effects on DMI, MY, milk components and fatty acid (FA) profile,

nutrient digestibility, N utilization, and enteric methane (CH4) emissions in lactating

dairy cows.

124


Crops and Silages

The forages were grown in Centre County, PA at approximately 40° N latitude on

Hagerstown soil. Brown midrib dwarf pearl millet (Pennisetum glaucum ‘Exceed’;

King’s Agriseeds, Ronks, PA) was planted on June 15th, 2016 with a no-till drill (John

Deere 1590; Moline, IL) at a seeding rate of 22.4 kg/ha and a 19 cm row spacing. The

field was sprayed with glyphosate and fertilized with 44.8 t/ha of dairy manure

contributing 42 kg/ha of ammonium N and 177 kg/ha of organic N. An additional 73 kg

of N/ha from a 30% urea and ammonium nitrate liquid fertilizer was applied prior to

planting. Soybeans were grown in the field the previous year. A John Deere 945 mower

with a flail conditioner was used to mow the crop on August 3rd, 2016 at the flag leaf

visible stage at a height of 11.5 cm. After wilting to a target 30% DM, the forage was

gathered and chopped using a John Deere 6750 harvester on August 5th, 2016. Chop

length was set to 25 mm. The millet was ensiled without inoculant in a 2.4 m diameter

plastic silage bag (Up North Plastics, Cottage Grove, MN).

The silage corn (DKC 52-61; 102 d relative maturity; DeKalb, St. Louis, MO)

was not specifically grown for the current experiment but was from the forage source

normally fed to The Pennsylvania State University dairy herd. The corn for silage was

planted between May 1st and May 10th, 2015 at a rate of 79,000 seeds/ha. It was planted

with a no-till drill (John Deere 1590; Moline, IL) into fields fertilized with dairy manure

as stated above. An additional 43 kg/ha of N was applied as 30% urea and ammonium

125

nitrate liquid prior to planting and 67 kg/ha of N in the same form as a sidedress

application. Corn silage harvest was conducted between September 24th and September

28th, 2015 at a target DM of 38% with a 19 mm chop length and ensiled in an upright

concrete silo.

Animals and Diets


Pennsylvania State University’s Institutional Animal Care and Use Committee.

Seventeen mid-lactation Holstein dairy cows (MY, 50 ± 4.2 kg; 2.5 ± 0.62 lactations;

DIM 66 ± 20 d; and BW 630 ± 71 kg at the beginning of the experiment) were used in the

feeding experiment and an additional 4 cows were used in the in situ analysis. The

experiment was a crossover design with 2 periods of 28 d each; 21 d were allowed for

adaptation to the diet and the last 7 d of each period were for data and sample collection.

Cows were allocated to 8 groups of 2 cows each, plus 1 spare cow, based on DIM, MY,

and parity. One cow got mastitis at the end of the first sampling period. Therefore, it was

decided to also collect samples from the spare cow for the second period. Cows within a

group were randomly assigned to one of 2 diets, as described below. All cows were

housed in the tie stall barn of The Pennsylvania State University’s Dairy Research and

Teaching Center. Diets were mixed and fed from a Rissler model 1050 TMR mixer (I.H.

Rissler Mfg. LLC, Mohnton, PA). Cows were fed once daily around 8 a.m. to yield

approximately 5-10% refusals. Feed was pushed up 3 times throughout the day. The cows

were milked twice daily at 7 a.m. and 6 p.m.

126

Two different diets (Table 1) were fed to the cows during the experiment as

follows: a control diet (CSD), based on corn silage and alfalfa haylage or a pearl millet

silage diet (PMD), in which pearl millet silage was included at 10% of dietary DM,

replacing 20% of the control diet corn silage DM. Thus, the only difference between CSD

and PMD was the replacement of 20% of the corn silage DM with pearl millet silage to

mimic a possible proportion of whole farm pearl millet crop yield. The CSD diet was

formulated to meet or exceed the NRC (2001) NEL and MP requirements of a Holstein

cow with 630 kg BW, 48 kg MY, 3.8% fat, 2.95% true protein, and at 28 kg DMI.


Refusals, TMR, and Feed Ingredients.

Refusals were collected into a Ranger Mate mobile tub scale (American Calan,

Northwood, NH) and weighed individually for each cow prior to the morning feeding to

measure daily as-fed intake. Total mixed ration, refusal and forage (pearl millet, alfalfa,

and corn silage) samples were collected twice weekly, composited by wk and diet (i.e.,

silage type), and stored at -20°C. The TMR was sampled within 1 h of feeding. The

weekly DM content of the TMR and refusals oven dried at 55°C for 72 h was used to

calculate the individual daily DMI. Fermentation profiles of fresh frozen samples of the

pearl millet, alfalfa and corn silages from each period were analyzed by Cumberland

Valley Analytical Services Inc. by wet chemistry for pH, and lactic, acetic, propionic,

butyric, and isobutyric acid concentrations. Pearl millet, corn and alfalfa silages were

127

oven dried at 55°C for 72 h, ground through a 4 mm screen (for in situ degradability

measurements), then ground through a 1 mm screen in a Wiley mill (Thomas Scientific,

Swedesboro, NJ) and composited by period on an equal weight basis. Dried composite

samples of pearl millet, alfalfa and corn silages were sent to Cumberland Valley

Analytical Services Inc. (Maugansville, MD) to be analyzed by wet chemistry methods

for amylase-treated NDF (Van Soest et al., 1991), ADF (method 973.18; AOAC

International, 2000), lignin (Goering and Van Soest, 1970), fat (method 2003.05; AOAC

International, 2006), CP (method 990.03; AOAC International, 2000), soluble protein

(Krishnamoorthy et al., 1982), starch (Hall, 2009), ethanol-soluble carbohydrates, which

measures mono-, di-, and oligosaccharides (DuBois et al., 1956), ash (method 942.05;

AOAC International, 2000), and minerals (method 985.01; AOAC International, 2000).

Concentrate feeds were sampled weekly and stored at -20°C until analysis.

Concentrate feed samples were ground and composited once for the entire experiment.

Dried composite concentrate ingredients were analyzed by Cumberland Valley Analytical

Services Inc. by wet chemistry methods for CP, amylase-treated NDF, ADF, fat, CP,

starch, ash, and minerals (procedures as referenced above). The percent NFC was

calculated using the equation NFC% = 100 – CP% − fat% − NDF% − ash% and NEL

using the equation NEL = 0.0245 × TDN − 0.12. Concentrations of CP, NDF, ADF, NFC,

NEL, starch, fat, ash, Ca, and P in the TMR were calculated based on the individual feed

ingredient values and their percent inclusion in the TMR. The diet values for RDP, RUP,

and NEL balance were calculated based on NRC (2001) at actual DMI, MY, BW, and

milk composition of the cows.

128

Milk.



collected on two consecutive days (4 consecutive milkings) during wk 4 of each

experimental period from the p.m. and a.m. milkings. Milk component samples were

collected into tubes containing 2-bromo-2-nitropropane-1,3-diol and analyzed

individually by Dairy One Laboratory (Ithaca, NY) for fat, true protein, MUN, and

lactose content using infrared spectroscopy (Milkoscan 4000; Foss Electric, Hillerød,

Denmark). Milk samples for FA analysis from the 4 milkings for each period were

collected without preservative, stored chilled at 4°C and later composited by cow

weighted for the milk production of each milking. The composited milk samples were

centrifuged, the milk fat was skimmed off and then stored frozen at -20°C until analyzed

for FA using the procedure described by Rico and Harvatine (2013). Cow BW was

recorded twice daily upon exiting the milking parlor using an AfiFarm 3.04E scale

system (S.A.E. Afikim, Rehovot, Israel).

Estimation of Digestibility and Gas Emissions.

During wk 4 of each period, urine and fecal samples were collected for estimation

of apparent digestibility and N utilization. Spot urine and fecal samples (approximately

300 ml and 500 g per sample, respectively) were collected 8 times over 3 d at (d 1) 0500,

1100, 1700, and 2300 h; (d 2) 0800, 1400, and 2000 h; and (d 3) 0200 to obtain a


129

processing and analyses can be found in Lee et al. (2012). Briefly, raw urine from each

sampling was acidified, diluted, composited by cow and period, and stored frozen at -

20°C for later analysis of allantoin, uric acid, creatinine, urea N and total N. Allantoin

was analyzed following the procedure by Chen et al. (1992). Stanbio Laboratory (Boerne,



samples of approximately 60 µl of 1:10 diluted and acidified urine using a Costech ECS

4010 C/N/S elemental analyzer (Costech Analytical Technologies Inc., Valencia, CA).

Fecal samples were oven dried at 65°C, ground through a 1-mm screen in a Wiley mill

and analyzed for DM, OM, CP, starch, NDF and ADF. A Mixer Mill MM 200 (Retsch

GmbH, Haan, Germany) was used to pulverize a 0.5 g aliquot of fecal sample for CP

analysis (N × 6.25) using a Costech ECS 4010 C/N/S elemental analyzer. Starch analysis

of fecal DM for apparent total tract digestibility was performed using a procedure similar

to the method including acetate buffer described by Hall (2009). Briefly, starch was

gelatinized with 50% NaOH, incubated for 16 h at 55°C with acetate buffer and amylase,

centrifuged, plated on a 96-well plate and then reacted with a PGO (Glucose

Oxidase/Peroxidase) enzyme solution (P7119; Sigma-Aldrich, Saint Louis, MO) for 45

min before being read at 450 nm. Neutral- and acid-detergent fiber were analyzed with an

Ankom 200 fiber analyzer (Ankom Technology Corp., Macedon, NY) based on the

procedures of Van Soest et al. (1991) with α-amylase and sodium sulfite in the NDF

analysis. A 12-d ruminal incubation was used to analyze indigestible NDF (iNDF;

Huhtanen et al., 1994 as modified by Lee et al., 2012) in both feces and TMR, which was

used as a marker to estimate apparent total-tract digestibilities of dietary nutrients.


130

Enteric CH4 and carbon dioxide (CO2) emissions were measured during wk 4 of

each experimental period with the GreenFeed system (C-Lock Inc., Rapid City, SD). The

GreenFeed system measures gas mass fluxes and is one of the established techniques for

measuring enteric CH4 emissions from ruminant animals (Hristov et al., 2015b;

Hammond et al., 2016). One GreenFeed unit was used to measure gas emissions from all

cows individually in a sequential manner for 5 min of breath gas sampling and 2 min of

background gas sampling at every collection time point. There were 8 collection time-

points for each cow during each experimental period. The unit was positioned in the feed

bunk in front of each cow starting at (d1) 0800, 1400, and 2000 h; (d2) 0200, 1100, 1700,

and 2300 h; and (d3) 0500 to obtain a representative sample of a 24-h period. A detailed

and visual explanation of the gas sampling procedures can be viewed in Hristov et al.

(2015a). Gas emission data were averaged by cow and period for the statistical analysis.

In Situ DM and NDF Degradation

Ruminal disappearance of DM and NDF was determined in situ for pearl millet,

alfalfa and corn silages that were fed during the experiment. Four ruminally cannulated

lactating Holstein cows averaging: DMI 28.3 ± 4.9 kg; MY 37.8 ± 9.2 kg; 2.8 ± 0.5

lactations; DIM 208 ± 26 d; and BW 639 ± 94.4 kg were used for in situ incubations.

Cows were fed (% DM basis) corn silage 51.3, alfalfa haylage 7.4, straw 3.5, canola meal

8.9, SoyPLUS (West Central Cooperative, Ralston, IA) 8.0, ground corn 7.0, roasted

soybeans 5.0, molasses 5.0, whole cotton seed 2, and mineral mix 1.9. The in situ

procedure was performed as described in Harper et al. (2017b). Briefly, triplicate samples

131

of 7 g each of dried 4-mm ground silages were weighed into Ankom nylon bags (10cm x

20cm Forage Bag; Ankom Technology Corp., Macedon, NY) which were sequentially

incubated in the ventral rumen for 96, 72, 48, 24, 12, and 0 h and simultaneously

removed. All bags were washed under cold water and then oven-dried for 72 h at 55°C.

Ruminal disappearance was calculated based on initial dry weight of the incubated

sample, residue dry weight, and NDF concentration of initial sample and bag residue.

Degradation data were fitted to a line rising exponentially to a maximum value with the

equation p = a + b (1 - e-ct), using SigmaPlot v10.0 (Systat Software Inc., San Jose, CA)

where p is the degraded fraction (of DM or NDF) at time t, constant a is the soluble

fraction (or intercept), b is the potentially degradable fraction (i.e., predicted fraction of

DM or NDF that is potentially degradable in the rumen), and c is the rate of degradation

of the b fraction (Ørskov and McDonald, 1979). The effective degradability (ED; an

estimate of the percentage of DM or NDF that would be degraded in the rumen at

specified passage rate) was determined with the following equation (Ørskov and

McDonald, 1979): ED = a + b {c ÷ (c + k)}, where k is the rate of passage assumed to be

3%/h.


Statistical analyses were run using the MIXED procedure of SAS v9.4 (SAS

Institute Inc., Cary, NC). Cow was the experimental unit. Milk yield, BW, and DMI from

the last 7 d of the experiment and the 2 d of milk composition samples were analyzed

with day as a repeated measure using an AR(1) covariance structure. The statistical

132

model included diet, day, period, and period × diet and diet × day interactions. Group and

cow within group were random effects with all others fixed. Milk FA, nutrient intake,

apparent digestibility, N utilization, and CH4 and CO2 emissions data were analyzed

using the same model without day and diet × day interaction. Body weight change was

calculated as the average BW of the last 3 d of the experimental period less the average

BW of the first 3 d of the experimental period. Milk composition data were weighted

averages per day based on the MY of the evening and morning milkings. Individual

ruminal in situ degradation parameters (a, b, c, and ED) of DM and NDF were contrasted

among forages. For all data, significance was declared at P ≤ 0.05 and tendency was

declared at 0.05 < P ≤ 0.10. If not indicated otherwise, data are presented as least squares

means.


Forages

Brown midrib dwarf pearl millet yielded 2.8 t DM/ha in the first cutting and was

used to conduct the animal experiment. Nutrient composition and fermentation profiles of

pearl millet and corn silages are shown in Table 2. The corn silage had a high starch

content of 40% DM in contrast to the <1% starch in the pearl millet silage. The pearl

millet was harvested at the flag leaf visible stage and, therefore, low starch concentrations

were expected. Pearl millet silage had a higher concentration of NDF compared with corn

silage which was similar to results of Hassanat et al. (2007) and Mustafa et al. (2004)

133

who reported 59 and 61% NDF, respectively, in BMR pearl millet forage. Lignin

concentration was similar between pearl millet silage and corn silage. Lignin is found in

the plant cell wall and its association with cellulose and hemicellulose decreases NDF

digestibility (Van Soest, 1994). Therefore, a better way to compare lignin concentrations

between the corn and pearl millet silages may be on an NDF basis. Looking at lignin this

way shows a larger difference between the forages at 7.42 vs. 4.28% lignin, as a % of

NDF, for corn silage and pearl millet silage, respectively. Crude protein content was

greater in the pearl millet silage, compared with the corn silage. Potassium concentration

was greater in the pearl millet silage which would be a disadvantage when trying to lower

diet DCAD in prepartum cows to decrease the risk of hypocalcemia (Charbonneau et al.,

2006).

Fermentation acid concentrations were numerically similar between the forages.

Butyric acid was not detected in either silage indicating that Clostridial fermentation was

likely not taking place. Silage pH was considerably lower for corn silage compared with

the pearl millet silage, which had a pH typical for grass silages. Ward et al. (2001)

reported a similar pH of 4.50 for pearl millet silage in their study.

Dry Matter Intake, Body Weight, and Milk Yield

Dry matter intake, BW, and milk production results are shown in Table 3. Diet

had no effect on DMI. Dry matter intake is driven by nutrient demand (e.g. ECM yield)

and constrained by rumen capacity (Mertens, 2009). Messman et al. (1992) reported a

decrease in DMI for a diet containing 50% non-BMR pearl millet containing 46.5% NDF

134

compared with an alfalfa and corn silage control diet containing 35.0% NDF. The current

study PMD treatment contained 32.4% NDF and does not seem to have restricted DMI.

Body weight change was not statistically different but the short term (28 d) design of this

study did not enable us to statistically perceive small differences in BW change. Yield of

ECM and ECM feed efficiency were similar between CSD and PMD. Amer and Mustafa

(2010) likewise reported no change in DMI for lactating cows fed pearl millet silage or

corn silage at approximately 35% of the diet. Neither did they see a reduction in milk

yield (averaging 38 kg/d) or feed efficiency (1.60 kg milk/ kg DMI). Unfortunately,

specific effects of pearl millet on milk production are difficult to elucidate in that study

because Megalac, a calcium salt product of palm oil fatty acids, was supplemented

uniquely to the pearl millet diet and likely influenced the results. The design of the

current study only altered the amount of corn silage and pearl millet in the diet and

therefore differences in the data between treatments can be directly attributed to the

forage change. Milk yield of the current study was decreased (P < 0.001) by PMD along

with a decrease in feed efficiency per unit of milk (P < 0.01). However, as the next

section details, milk yield differences were largely based on a lower concentration of

milk components in CSD milk which does not add value to milk producers and therefore

should not be emphasized.

Milk Composition and Yield

Milk fat content tended to increase (P = 0.06) from 3.47% with CSD to 3.71%

with PMD but milk fat yield was not different between diets. Milk true protein content

135

and yield were similar between diets. Lactose content and yield were also not different

between diets. Early research reported decreases in milk fat content of dairy cows grazing

pearl millet (Miller et al., 1965; Bucholtz et al. 1969) and consuming pearl millet

greenchop (Harner et al., 1969; Schneider et al. 1970) vs sudan grass. More recent

research from Messman et al. (1992) and Brunette et al. (2014) observed numerically

increased milk fat content when comparing corn silage with pearl millet silage.

Furthermore, increases in milk fat concentration are often reported when total dietary

fiber and fiber digestibility increases due to a shift in VFA production increasing the

acetate:propionate ratio (Oba and Allen, 1999; Ivan et al. 2005). The PMD treatment had

a higher NDF concentration, as stated earlier, and increased NDF digestibility which is

mentioned in the following section.

Nutrient Intake and Digestibility

Many of the production results discussed above relate to nutrient intake and

apparent digestibility effects, presented in Table 4. Higher intakes of OM (P = 0.03) and

starch (P < 0.001) for CSD reflect the lower ash and higher starch concentrations,

respectively, in the corn silage compared with pearl millet silage. The higher NDF (P =

0.001) and ADF (P = 0.004) intakes for PMD are a result of greater concentrations of

those components in the pearl millet silage. Amer and Mustafa (2010) likewise reported

higher NDF intake for lactating cows fed pearl millet silage vs. corn silage. Greater (P <

0.001) DM and OM apparent digestibility in CSD was a result of the over 1 kg/d higher

starch intake and lower NDF intake in that diet compared with PMD. Apparent starch

136

digestibility in this experiment was around 99% whereas NDF digestibility was

approximately 40%. There was a > 6% increase in NDF (P < 0.001) and ADF (P = 0.02)

apparent digestibility for the PMD treatment. This is likely caused by 2 factors. First,

cellulose digestion is decreased with lower rumen pH (Ørskov and Fraser, 1975) and,

even though we did not measure rumen pH directly, it is likely that rumen pH was lower

in cows fed CSD due to the higher starch content of CSD (Lechartier and Peyraud, 2011).

Second, the pearl millet silage fiber components were likely more easily digestible due to

less lignin per unit of NDF indicating the potential for fewer lignin crosslinks with

cellulose and hemicellulose in that early harvested forage as supported by the in situ data

below (Mertens, 1985; Cherney et al., 1991; Grabber et al., 2009).

Milk Fatty Acid

Milk FA data are shown in Table 5. The PMD treatment had higher (P = 0.04)

concentrations of 4:0 and 6:0 but there were no differences in total de novo FA or SFA.

There was a higher (P = 0.002) concentration of 18:0 in PMD and lower (P ≤ 0.02)

concentration of total trans FA. Increased 18:0 in PMD indicates a more complete

ruminal biohydrogenation of linoleic and linolenic unsaturated fatty acids. This is

supported by the decrease (P = 0.03) in total PUFA concentration (primarily 18 carbons)

in PMD. Although total MUFA concentrations were not different between the treatments,

trans-10 18:1 was higher (P = 0.008) in CSD. Increases in trans-10 18:1 have been

negatively related to milk fat production primarily through its association with production

of biohydrogenation intermediate trans-10, cis-12 CLA which is its precursor (Harvatine

137

et al., 2009; Rico and Harvatine, 2013). Diets high in unsaturated fats, high in rapidly

fermentable carbohydrates, and low in NDF tend to lower rumen pH and increase trans-

10 18:1 production (Rico and Harvatine, 2013; Zened et al., 2013). Other studies have

reported similar shifts in milk trans FA when comparing corn silage to grass silage diets

(Nielsen et al., 2004; Shingfield et al., 2005).

The concentrations of branched chain FA iso 14:0 and iso 15:0 were increased (P

< 0.001) in PMD. Experiments replacing grass silage with corn silage, which is similar to

the current study, have likewise reported a decrease in iso 14:0 and iso 15:0 in milk FA

(Vlaeminck et al., 2006a). Cellulolytic bacteria Ruminococcus albus and Ruminococcus

flavefaciens have higher concentrations of iso 14:0 and iso 15:0, respectively, than other

rumen bacteria particularly amylolytic species (Vlaeminck et al., 2006a). The PMD

treatment in the current study would have promoted more favorable rumen conditions for

cellulolytic bacteria than CSD. Additionally, Vlaeminck et al. (2006b) reported a positive

correlation between iso 14:0 and iso 15:0 and rumen proportions of acetate and a negative

correlation to rumen proportions of propionate indicating potentially less propionate

production by PMD.

In Situ DM and NDF Degradation

In situ DM (Figure 1) and NDF (Figure 2) disappearance data for the 3 silages fed

in the experiment help characterize pearl millet silage. Dry matter solubility (i.e., fraction

a) of pearl millet silage was lower (P < 0.001) than corn silage and alfalfa haylage, 21.5

vs. 49.7 and 39.4%, respectively (data not shown in tables). Potentially degradable DM, b

138

fraction, was highest (P < 0.001) in the pearl millet silage, 60.9%, and similar for corn

silage and alfalfa haylage, 30.7 and 29.2%, respectively. Fractional rate of disappearance

of the potentially degradable DM, c, was highest (P < 0.001) for alfalfa haylage, 5.7 %/h,

lowest for corn silage, 2.1%/h, and intermediate for pearl millet silage, 3.1%/h. Effective

degradability of DM, calculated with an estimated 3%/h passage rate, was highest for

corn silage, 62.2%, intermediate for alfalfa haylage, 58.6%, and lowest for pearl millet

silage, 52.2% (Figure 1).

Soluble NDF was highest (P <0.001) in alfalfa haylage, 8.5%, lower for corn

silage (2.3%), and 0% for pearl millet silage. Potentially degradable NDF was similarly

high for corn silage and pearl millet silage, 83.7 and 82.2% respectively, and lower (P <

0.001) for alfalfa haylage, 35.1%. Fractional disappearance rates of NDF were similar for

alfalfa haylage and pearl millet silage, 3.4 and 2.9%/h respectively, and lower (P < 0.001)

for corn silage, 1%/h. Pearl millet silage had the highest (P < 0.001) ED of NDF, 37.3%,

followed by alfalfa haylage and corn silage at 27.0 and 21.9%, respectively.

In situ DM solubility reflected differences in the NFC content of the silages, 49.9,

24.4, and 13.3% for corn silage, alfalfa haylage and pearl millet silage, respectively, and

affected ED of forage DM. The higher DM ED of corn silage vs. pearl millet silage is

consistent with the higher DM apparent digestibility of CSD vs. PMD. The higher ED of

NDF for the pearl millet silage was likely due to its lower lignin content, as a % of NDF,

compared with corn silage and agrees with the increased NDF apparent digestibility of

the PMD treatment. Mustafa et al. (2004) observed similar in situ NDF degradability

measurements for first cutting BMR pearl millet forage with a, b, and c values of 2.2,

139

77.2, and 3.6%, respectively. Diet of the cow in which in situ bags are incubated has an

effect on degradation results (Vanzant et al., 1998) which may explain why NDF ED of

BMR pearl millet silage from the current study was lower than the NDF ED of BMR

pearl millet forage reported by Mustafa et al. (2004) which averaged 44.2% calculated

using a 3%/h passage rate.

N Utilization

Nitrogen intake (Table 6) was not significantly affected by diet but N excretion

both in urine and feces appeared to be higher (P ≤ 0.01) for PMD. Milk N secretion was

higher (P = 0.02) in CSD, compared with PMD. Excretion of urinary purine derivatives,

allantoin and uric acid, was not different between diets. These urinary compounds have

been used as an indirect indicator of ruminal microbial protein synthesis (Chen, 1989).

Urinary urea N excretion was higher (P ≤ 0.001) with PMD probably due to the higher

concentrations of ammonia in the pearl millet silage not being incorporated into microbial

protein before being absorbed by the rumen wall. It is also plausible that the CP of the

pearl millet silage was truly less digestible than the corn silage CP because the neutral

detergent insoluble CP was 1.17% in the pearl millet silage vs 0.50% in the corn silage.

The PMD treatment had lower (P < 0.001) milk N use efficiency and higher (P ≤ 0.02)

total N excretion as a percent of N intake. This can be detrimental to the environment and

N use efficiency might be improved by reducing supplemental CP sources when

incorporating a higher CP forage, such as pearl millet silage, in dairy diets (O’Mara et al.,

1998; Bernard et al., 2002).

140

Enteric Methane and Carbon Dioxide Emissions

Enteric CH4 and CO2 emission results are shown in Table 7. No differences were

observed in CO2 emissions, which predominantly originate from bovine cellular

respiration and to a much lesser degree from enteric fermentation (Hammond et al.,

2016). Dairy cattle CO2 emissions have been correlated to DMI, milk production and

metabolic body weight which were similar in this study (Kirchgessner et al., 1991;

Kinsman et al., 1995). Daily enteric CH4 production, yield (i.e., per kg of DMI), and

intensity (i.e., per kg of ECM) were all increased (P < 0.01) by PMD, compared with

CSD. Total daily enteric CH4 production of the cows in the current study was high

because of their high DMI (Knapp et al., 2014) but within the range of values reported by

Hristov et al. (2015b) for high producing Holstein cows. Both diets had lower CH4 yield

and intensity than previous work conducted by the authors (Harper et al., 2017a; Harper

et al., 2017b) due to the high milk yield and early stage of lactation of the cows.

The increase in enteric CH4 production with PMD is not beneficial to the

environment and it has to be accounted for in the context of whole-farm greenhouse gas

emissions balance. Decreased diet digestibility and increased fiber content have been

shown to increase enteric CH4 emission intensity (McGeough et al., 2010). Interestingly,

O’Neill et al. (2011) reported a decrease in enteric CH4 production, yield, and intensity

from dairy cattle grazing high quality perennial ryegrass compared with a TMR. In that

study, the perennial ryegrass had a higher NDF content but also a higher OM digestibility

suggesting that diet digestibility might have a stronger influence then total diet NDF

content on enteric CH4 emissions. In the current study, PMD decreased apparent OM

141

digestibility and had a higher fiber content which explains the increase in enteric CH4

emissions. The higher enteric CH4 emissions for PMD may have decreased the available

energy for milk production in that diet (Johnson and Johnson, 1995), which may help

explain the decrease in milk yield with PMD.

CONCLUSIONS

Pearl millet silage included at 10% of dietary DM supported a high level of milk

production in peak lactation dairy cows, did not decrease DMI, and maintained ECM

yield compared to a corn silage-based control diet. The pearl millet diet tended to

increase milk fat content and increased NDF digestibility but decreased OM digestibility.

The fiber of pearl millet silage is highly digestible and replaces corn silage fiber well.

However, pearl millet silage cut early, as in this experiment, lacks the starch that is

removed from the diet when replacing corn silage. In practical rations, a diet containing

pearl millet silage would likely have to be balanced for NFC and CP to maintain milk

production and better utilize N in high producing cows. Pearl millet shows promise as

forage that could help diversify crop rotations particularly on dairy farms in the

Northeastern U.S.

ACKNOWLEDGEMENT

This project supported by the Northeast Sustainable Agriculture Research and


142





experimental cows.

REFERENCES

Amer, S. and A. F. Mustafa. 2010. Short communication: Effects of feeding pearl millet

silage on milk production of lactating dairy cows. J. Dairy Sci. 93:5921–5925.

doi:10.3168/jds.2010-3279


Arlington, VA.


Arlington, VA.

Bernard, J. K., J. W. West, and D. S. Trammell. 2002. Effect of replacing corn silage

with annual ryegrass silage on nutrient digestibility, intake, and milk yield for

lactating dairy cows. J. Dairy Sci. 85:2277–2282.

Brunette, T., B. Baurhoo, and A. F. Mustafa. 2014. Replacing corn silage with different

forage millet silage cultivars: Effects on milk yield, nutrient digestion, and ruminal

fermentation of lactating dairy cows. J. Dairy Sci. 97:6440-6449.

doi:10.3168/jds.2014-7998

Brunette, T., B. Baurhoo, and A. F. Mustafa. 2016. Effects of replacing grass silage with

forage pearl millet silage on milk yield, nutrient digestion, and ruminal fermentation

of lactating dairy cows. J. Dairy Sci. 99:269-279. doi:10.3168/jds.2015-9619

Bucholtz, H. F., C. L. Davis, D. L. Palmquist, and K. A. Kendall. 1969. Study of the low-

fat milk phenomenon in cows grazing pearl millet pastures. J. Dairy Sci. 52:1388-

1394.

Charbonneau, E., D. Pellerin, and G. R. Oetzel. 2006. Impact of lowering dietary cation-

anion difference in nonlactating dairy cows: A meta-analysis. J. Dairy Sci. 89:537–

548.

143

Chen, X. B. 1989. Excretion of purine derivatives by cattle and sheep and its use for the

estimation of absorbed microbial protein. Ph.D. Dissertation. Aberdeen Univ. UK.



supply in sheep. J. Anim. Sci. 70:1534-1542.

Cherney, D. J. R., J. A. Patterson, and K. D. Johnson. 1990. Digestibility and feeding

value of pearl millet as influenced by the brown-midrib, low-lignin trait. J. Anim.

Sci. 68:4345-4351.

Cherney, J. H., D. J. R. Cherney, D. E. Akin, and J. D. Axtell. 1991. Potential of brown-

midrib, low-lignin mutants for improving forage quality. Advances in Agronomy

46:157-198. doi:10.1016/S0065-2113(08)60580-5

Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith. 1956. Colorimetric

method for determination of sugars and related substances. Anal. Chem. 28:350.

Gillespie, J., R. Nehring, C. Sandretto, and C. Hallahan. 2010. Forage outsourcing in the

dairy sector: The extent of use and impact on farm profitability. Agric. Resour.

Econ. Rev. 39:399-414.

Goering, H. K. and P. J. Van Soest. 1970. Forage Fiber Analysis. USDA Agricultural

Research Service. Handbook number 379. U.S. Dept. of Agriculture. Superintendent

of Documents, US Government Printing Office, Washington, D.C. 20402.

Grabber, J. H., D. R. Mertens, H. Kim, C. Funk, F. Lua, and J. Ralpha. 2009. Cell wall

fermentation kinetics are impacted more by lignin content and ferulate cross-linking

than by lignin composition. J. Sci. Food Agric. 89:122-129.


feeds: Comparison of methods and a method recommended for AOAC collaborative

study. J. AOAC Int. 92:42-49.

Hammond, K. J., L. A. Crompton, A. Bannink, J. Dijkstra, D. R. Yanez-Ruiz, P. O’Kiely,

E. Kebreab, M. A. Eugene, Z. Yu, K. J. Shingfield, A. Schwarm, A. N. Hristov, and

C. K. Reynolds. 2016. Review of current in vivo measurement techniques for

quantifying enteric methane emission from ruminants. Anim. Feed Sci. Technol.

219:13-30. doi:10.1016/j.anifeedsci.2016.05.018

Harner, J. P., R. W. Hemken, J. H. Vandersall, N. A. Clark, and B. A. Schneider. 1969.

Relationship of pearl millet to milk fat depression in dairy cows. I. Cation

fertilization. J. Dairy Sci. 52:1244-1248.

Harper, M. T., J. Oh, F. Giallongo, J. C. Lopes, G. W. Roth, and A. N. Hristov. 2017a.

Using brown midrib 6 dwarf forage sorghum silage and fall-grown oat silage in

lactating dairy cow rations J. Dairy Sci. 100:5250-5265. doi:10.3168/jds.2017-

12552

144

Harper, M. T., J. Oh, F. Giallongo, G. W. Roth, and A. N. Hristov. 2017b. Inclusion of

wheat and triticale silage in the diet of lactating dairy cows. J. Dairy Sci. 100:6151-

6163. doi:10.3168/jds.2017-12553.

Harvatine, K. J., Y. R. Boisclair, and D. E. Bauman. 2009. Recent advances in the

regulation of milk fat synthesis. Animal 3:40–54.

Hassanat, F., A. F. Mustafa, and P. Seguin. 2007. Effects of inoculation on ensiling

characteristics, chemical composition and aerobic stability of regular and brown

midrib millet silages. Anim. Feed Sci. Technol. 139:125-140.

doi:10.1016/j.anifeedsci.2007.01.005

Hristov, A. N., J. Oh, F. Giallongo, T. Frederick, H. Weeks, P. R. Zimmerman, R. A.

Hristova, S. R. Zimmerman, and A. F. Branco. 2015a. The use of an automated

system (GreenFeed) to monitor enteric methane and carbon dioxide emissions from

ruminant animals. J. Vis. Exp. 103:e52904.

Hristov, A. N., J. Oh, F. Giallongo, T. W. Frederick, M. T. Harper, H. L. Weeks, A. F.

Branco, P. J. Moate, M. H. Deighton, S. R. O. Williams, M. Kindermann, and S.

Duval. 2015b. An inhibitor persistently decreased enteric methane emission from

dairy cows with no negative effect on milk production. Proc. Natl. Acad. Sci.

112:10663-10668. doi:10.1073/pnas.1515515112



Anim. Feed Sci. Technol. 48:211-227.

Ivan, S. K., R. J. Grant, D. Weakley, and J. Beck. 2005. Comparison of a corn silage

hybrid with high cell-wall content and digestibility with a hybrid of lower cell-wall

content on performance of Holstein cows. J. Dairy Sci. 88:244-254.

Johnson, K. A., and D. E. Johnson. 1995. Methane emissions from cattle. J. Anim. Sci.

73:2483-2492.

Kinsman, R., F. D. Sauer, H. A. Jackson, and M. S. Wolynetz. 1995. Methane and carbon

dioxide emissions from dairy cows in full lactation monitored over a six-month

period. J. Dairy Sci. 78:2760-2766.

Kirchgessner, M., W. Windisch, H. L. Muller, and M. Kreuzer. 1991. Release of methane

and of carbon dioxide by dairy cattle. Agribiol. Res. 44:91-102.

Knapp, J. R.,G. L. Laur, P. A. Vadas, W. P. Weiss, and J. M. Tricarico. 2014. Invited

review: Enteric methane in dairy cattle production: Quantifying the opportunities

and impact of reducing emissions. J. Dairy Sci. 97:3231–3261. doi:

10.3168/jds.2013-7234

Krishnamoorthy, U., T. V. Muscato, C. J. Sniffen, and P. J. Van Soest. 1982. Nitrogen

fractions in selected feedstuffs. J. Dairy Sci. 65:217-225.

145

Lechartier, C., and J. L. Peyraud. 2011. The effects of starch and rapidly degradable dry

matter from concentrate on ruminal digestion in dairy cows fed corn silage-based

diets with fixed forage proportion. J. Dairy Sci. 94:2440-2454.




deficient diet. J. Dairy Sci. 95:6042-6056.

Maman, N., D. J. Lyon, S. C. Mason, T. D. Galusha, and R. Higgins. 2003. Agron. J.

95:1618-1624.

McGeough, E. J., P. O’Kiely, K. J. Hart, A. P. Moloney, T. M. Boland, and D. A. Kenny.

2010. Methane emissions, feed intake, performance, digestibility, and rumen

fermentation of finishing beef cattle offered whole-crop wheat silages differing in

grain content. J. Anim. Sci. 88:2703-2716.

Messman, M. A., W. P. Weiss, P. R. Henderlong, and W. L. Shockey. 1992. Evaluation

of pearl millet and field peas plus triticale silages for midlactation dairy cows. J.

Dairy Sci. 75:2769-2775.

Mertens, D. R. 1985. Effect of fiber on feed quality for dairy cows. Pages 209-224 in 46th

Minnesota Nutrition Conference Proc.

Mertens, D. R. 2009. Maximizing forage use by dairy cows. Pages 303-319 in Proc. 27th

Annual Western Canadian Dairy Seminar. Red Deer, Alberta.

Miller, R. W., R. W. Hemken, J.H. Vandersall, D. R. Waldo, M. Okamato, and L. A.

Moore. 1965. Effect of feeding buffers to dairy cows grazing pearl millet or sudan

grass. J. Dairy Sci. 48:1319-1323.

Mustafa, A. F., F. Hassanat, and P. Seguin. 2004. Chemical composition and in situ

ruminal nutrient degradability of normal and brown midrib forage pearl millet grown

in southwestern Québec. Can. J. Anim. Sci. 84:737-740.



Nielsen, T.S., K. Sejrsen, H. R. Andersen, P. Lund, and E. M. Straarup. 2004. Effect of

silage type and energy concentration on conjugated linoleic acid (CLA) in milk fat

from dairy cows. J. Anim. Feed Sci. 13(Suppl.1):697-700.

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.

O’Mara, F. P., J. J. Fitzgerald, J. J. Murphy, and M. Rath. 1998. The effect on milk

production of replacing grass silage with maize silage in the diet of dairy cows.

Livest. Prod. Sci. 55:79–87.

146

O’Neill, B. F., M. H. Deighton, B. M. O’Loughlin, F. J. Mulligan, T. M. Boland, M.

O’Donovan, and E. Lewis. 2011. Effects of a perennial ryegrass diet or total mixed

ration diet offered to spring-calving Holstein-Friesian dairy cows on methane

emissions, dry matter intake, and milk production. J. Dairy Sci. 94:1941–1951.

doi:10.3168/jds.2010-3361

Ørskov, E. R., and C. Fraser. 1975. The effects of processing of barley-based

supplements on rumen pH, rate of digestion and voluntary intake of dried grass in

sheep. Br. J. Nutr. 34:493-500.



Agric. Sci. (Camb.) 92:499-503.



fiber and oil concentration. J. Dairy Sci. 96:6621-6630.

Schneider, B. A., N. A. Clark, R. W. Hemken, and J. H. Vandersall. .1970. Relationship

of pearl millet to milk fat depression in dairy cows. II. Forage organic acids as

influenced by soil nutrients. J. Dairy Sci. 53:305-310.

Shingfield, K.J., C. K. Reynolds, B. Lupoli, V. Toivonen, M. P. Yurawecz, P. Delmonte,

J. M. Griinari, A. S. Grandison, and D. E. Beever. 2005. Effect of forage type and

proportion of concentrate in the diet on milk fatty acid composition in cows given

sunflower and fish oil. Anim. Sci. 80:225-238. doi:10.1079/ASC41820225





Thomison, P., P. Raul, R. Hammond, and R. Mullen. 2011. Managing risks in continuous

corn. C.O.R.N. Newsl. 2011-02. https://agcrops.osu.edu/newsletters/2011/02

(accessed 28 Dec. 2017).



J. Dairy Sci. 74:3583-3597.

Van Soest, P. J. 1994. Nutritional ecology of the ruminant. 2nd Ed. Cornell University

Press, Ithaca, NY.

Vanzant, E. S., R. C. Cochran, and E. C. Titgemeyer. 1998. Standardization of in situ

techniques for ruminant feedstuff evaluation. J. Anim. Sci. 76:2717-2729.

Vlaeminck, B., V. Fievez, A. R. J. Cabrita, A. J. M. Fonseca, and R. J. Dewhurst. 2006a.

Factors affecting odd and branched-chain fatty acids in milk: A review. Anim. Feed

Sci. Technol. 131:389-417.

147

Vlaeminck, B., V. Fievez, S. Tamminga, R. J. Dewhurst, A. van Vuuren, D. De

Brabander, and D. Demeyer. 2006b. Milk odd- and branched-chain fatty acids in

relation to the rumen fermentation pattern. J. Dairy Sci. 89:3954–3964.

Ward, J. D., D. D. Redfearn, M. E. McCormick, and G. J. Cuomo. 2001 Chemical

Composition, Ensiling Characteristics, and Apparent Digestibility of Summer

Annual Forages in a Subtropical Double-Cropping System with Annual Ryegrass. J.

Dairy Sci. 84:177–182.

Wolf, C. A. 2003. The economics of dairy production. Vet. Clin. Food Anim. 19:271-

293.

Zegada-Lizarazu, W., and M. Iijima. 2005. Deep root water uptake ability and water use

efficiency of pearl millet in comparison to other millet species. Plant Prod. Sci.

8:454-460.

Zened, A., F. Enjalbert, M. C. Nicot, and A. Troegeler-Meynadier. 2013. Starch plus

sunflower oil addition to the diet of dry dairy cows results in a trans-11 to trans-10

shift of biohydrogenation. J. Dairy Sci. 96:451-459.

148


1CSD = Corn silage control diet; PMD = Pearl millet silage diet. 2Alfalfa haylage was 34.8% DM and contained (DM basis) 22.1% CP, 24.4% NFC, and


contained (%, as-is basis) limestone, 36.75; dry corn distillers grains with solubles, 29.00;

NaCl, 24.85; MgO (54% Mg), 4.15; Bio-Phos, 2.45; zinc sulfate, 0.96; mineral oil, 0.5;

vitamin E, 0.37; manganese sulfate, 0.37; copper sulfate, 0.26; ferrous sulfate, 0.16;

Diet1

Item CSD PMD

Ingredient, % of DM

Corn silage 50 40

Pearl millet silage - 10

Alfalfa haylage2 6 6

Hay/straw mixture 4 4

Cottonseed hulls 2 2

Ground corn 10 10

Heat-treated whole soybeans 5.5 5.5

Solvent-extracted canola meal 9 9

SoyPLUS3 7.5 7.5

Molasses4 4 4

Mineral/vitamin premix5 2 2


CP6 16.6 17.2

RDP8 9.1 9.2

RUP8 7.5 7.9

NDF6 30.3 32.4

ADF6 19.3 20.5

NFC7 43.9 40.2

Starch6 28.0 24.1

Fat6 4.6 4.6

NEL,7 Mcal/kg 1.53 1.54

NEL intake,7 Mcal/d 44.6 44.6

NEL balance,7 Mcal/d -0.7 -0.5

MP balance,7 g/d 163 322

Ash6 6.77 7.67

Ca6 0.8 0.8

P6 0.4 0.4

DCAD, mEQ/kg 204 288

149

Selenium, .13; vitamin A, 0.03; vitamin D3, 0.013; calcium iodate, 0.008; cobalt

carbonate, 0.005. 6Values calculated using the chemical analysis (Cumberland Valley Analytical Services

Inc., Maugansville, MD) of individual feed ingredients of the diet. 7Estimated based on NRC (2001).

150

Table 5-2. Nutrient composition and fermentation profile of pearl millet and corn silages

(% of DM or as indicated)1

1Two composite samples per silage, one for each experimental period were analyzed by

wet chemistry (Cumberland Valley Analytical Services Inc., Maugansville, MD). Mean ±

SE is reported. 2CPE = Crude protein equivalent. 3Butyric and Isobutyric acids were not detected in either silage. 4ND = Not detected.

Forages

Item Corn Pearl Millet

DM, % 42.2 ± 1.6 30.5 ± 0.8

NDF 36.8 ± 0.4 58.4 ± 0.1

ADF 22.3 ± 0.6 34.4 ± 0.8

Lignin 2.73 ± 0.2 2.50 ± 0.1

Lignin, % of NDF 7.42 ± 0.4 4.28 ± 0.1

Fat (ether extract) 2.58 ± 0.2 3.31 ± 0.0

CP 7.45 ± 0.6 13.2 ± 0.2

Soluble CP, % of CP 61.2 ± 0.4 63.7 ± 0.1

NH3 CPE2 0.88 ± 0.1 1.94 ± 0.2

NH3 CPE, % of CP 11.8 ± 0.2 14.6 ± 1.6

Starch 40 ± 1.9 0.9 ± 0.3

Ethanol soluble carbohydrates 1.3 ± 0.2 1.95 ± 0.8

Ash 3.8 ± 0.1 12.9 ± 0.1

Ca 0.29 ± 0.0 0.96 ± 0.2

P 0.23 ± 0.0 0.32 ± 0.0

K 1.05 ± 0.1 5.13 ± 0.1

pH 3.79 ± 0.1 4.48 ± 0.1

Fermentation acids3

Lactic 5.90 ± 0.6 6.25 ± 0.2

Acetic 1.82 ± 0.3 1.28 ± 0.4

Propionic 0.13 ± 0.1 ND4

151

Table 5-3. Effect of pearl millet silage on DMI, milk production, and feed efficiency in

lactating dairy cows

Diet1 P-Value

Item CSD PMD SEM2 Diet

DMI, kg/d 29.1 29.0 0.65 0.78

Milk yield, kg/d 51.3 49.6 2.02 <0.001

Milk ÷ DMI, kg/kg 1.77 1.72 0.053 0.01

Milk fat, % 3.47 3.71 0.118 0.06

Milk fat, kg/d 1.79 1.82 0.087 0.65

Milk true protein, % 2.86 2.85 0.050 0.64

Milk true protein, kg/d 1.46 1.43 0.055 0.44

Lactose, % 5.00 4.96 0.035 0.28

Lactose, kg/d 2.55 2.47 0.116 0.23

MUN, mg/dL 11.6 13.3 0.410 <0.001

ECM3, kg/d 46.8 46.6 1.92 0.86

ECM ÷ DMI, kg/kg 1.59 1.56 0.050 0.50

BW, kg 632 628 16.8 0.01

BW change, kg 7.23 5.27 3.335 0.72 1CSD = Corn silage control diet; PMD = Pearl millet silage diet. 2Largest SEM published in table. DMI, n = 235; milk yield, n = 231; milk yield ÷ DMI, n

= 230; BW, n = 238; BW change, n = 34; milk composition data, n = 65 (n represents the

number of observations used in the statistical analysis). 3Energy-corrected milk (kg/d) = kg of milk × [(38.3 × % fat × 10 + 24.2 × % true protein


152

Table 5-4. Effect of pearl millet silage on nutrient intake and apparent total-tract


Diet1 P-Value


Intake, kg/d

DM3 29.4 28.8 0.58 0.14

OM 27.4 26.6 0.54 0.03

CP 4.86 4.94 0.097 0.23

Starch 8.22 6.94 0.153 <0.001

NDF 8.89 9.33 0.181 0.001

NDF, % of BW 1.42 1.49 0.031 0.001

Forage NDF, % of

BW 1.11 1.19 0.025 <0.001

ADF 5.67 5.90 0.115 0.004


DM 66.5 64.5 0.38 <0.001

OM 67.2 65.1 0.38 <0.001

CP 64.3 61.8 0.55 0.003

Starch 99.2 99.1 0.09 0.26

NDF 38.5 41.0 0.65 <0.001

ADF 24.9 27.5 1.15 0.02 1CSD = Corn silage control diet; PMD = Pearl millet silage diet. 2Largest SEM published in table; n = 33 (n represents the number of observations used in

the statistical analysis). 3DM intake reported is during the fecal collection periods.

153

Table 5-5. Effect of pearl millet silage on milk fatty acid composition (g/100 g of total

fatty acids) in lactating dairy cows

Diet1

P- Value

Fatty Acid CSD PMD SEM2 Diet

4:0 4.40 4.62 0.095 0.04

6:0 2.39 2.51 0.045 0.04

8:0 1.35 1.41 0.029 0.13

10:0 3.22 3.24 0.084 0.73

cis-9 10:1 0.26 0.27 0.008 0.22

11:0 0.08 0.07 0.007 0.02

12:0 3.64 3.57 0.104 0.33

13:0 iso 0.003 0.006 0.002 0.22

13:0 anteiso 0.07 0.07 0.004 0.07

13:0 0.13 0.11 0.008 0.01

14:0 iso 0.06 0.07 0.005 <0.001

14:0 11.4 11.1 0.168 0.11

15:0 iso 0.16 0.19 0.003 <0.001

15:0 anteiso 0.33 0.35 0.006 0.005

cis-9 14:1 0.84 0.79 0.046 0.06

15:0 1.06 0.99 0.046 0.12

16:0 iso 0.16 0.17 0.010 0.36

16:0 26.8 26.5 0.526 0.48

17:0 iso 0.24 0.25 0.017 0.60

cis-9 16:1 1.12 1.06 0.074 0.08

17:0 anteiso 0.36 0.36 0.010 0.92

17:0 0.47 0.47 0.010 0.92

cis-9 17:1 0.15 0.14 0.007 0.48

18:0 11.0 11.8 0.428 0.002

trans-4 18:1 0.03 0.02 0.003 0.05

trans-6,8 18:1 0.34 0.31 0.012 0.002

trans-9 18:1 0.27 0.25 0.008 <0.001

trans-10 18:1 0.61 0.47 0.051 0.008

trans-11 18:1 1.16 1.10 0.056 0.27

trans-12 18:1 0.56 0.51 0.019 0.001

cis-9 18:1 17.4 17.5 0.349 0.79

trans-15 18:1 0.47 0.47 0.014 0.84

cis-11 18:1 0.94 0.82 0.042 0.001

cis-12 18:1 0.48 0.42 0.020 <0.001

cis-9,12 18:2 3.14 3.06 0.087 0.04

154

cis-9, trans-11 18:2 0.51 0.45 0.028 0.06

cis-6,9,12 18:3 0.08 0.08 0.004 0.45

20:0 0.12 0.13 0.003 <0.001

cis-11 20:1 0.40 0.43 0.013 0.007

22:0 0.04 0.05 0.003 0.006

20:3 0.12 0.12 0.004 0.65

20:4 0.15 0.15 0.006 0.18

Σ De novo FA3 27.2 27.2 0.359 0.92

Σ C164 27.9 27.6 0.560 0.37

Σ Preformed FA5 37.8 38.1 0.678 0.62

Σ SFA 67.4 68.1 0.605 0.25

Σ MUFA 25.0 24.5 0.480 0.32

Σ PUFA 4.00 3.86 0.112 0.03

Σ trans FA6 3.95 3.57 0.153 0.02

Σ OBCFA7 3.26 3.25 0.072 0.77

Unknown 3.58 3.58 0.064 0.95 1CSD = Corn silage control diet; PMD = Pearl millet silage diet. 2Largest SEM shown; n = 33 (n represents number of observations used in the statistical

analysis). Data are presented as LSM. 3Sum of fatty acids synthesized in the mammary gland (4:0, 6:0, 8:0, 10:0, 12:0, 14:0,

14:1). 4Sum of 16 C fatty acids (16:0, and 16:1). 5Sum of ≥ 18 C fatty acids. 6Sum of trans unsaturated fatty acids. 7Sum of the odd and branched chain fatty acids (11:0, iso13:0, anteiso13:0, 13:0, iso14:0,

iso15:0, anteiso15:0, 15:0, iso16:0, iso17:0, anteiso17:0, 17:0, cis-9 17:1).

155

Table 5-6. Effect of pearl millet silage on nitrogen utilization and urinary purine

derivatives in lactating dairy cows

Diet1 P-Value


N intake, g/d 778 790 15.6 0.23

N excretion or secretion, g/d 738 778 18.0 0.01

Urine N, g/d 229 254 8.7 0.01

UUN3, g/d 161 187 6.0 <0.001

Fecal N, g/d 278 302 7.5 0.006

Total excreta N, g/d 507 555 13.8 0.001

Milk N, g/d 231 222 5.9 0.02


Urine N 29.4 32.2 0.88 0.02

Fecal N 35.7 38.2 0.55 0.003

Total excreta N 65.1 70.5 0.96 <0.001

Milk N 29.8 28.2 0.62 0.001

Urine output4, kg/d 21.4 26.6 0.70 <0.001


Allantion 667 698 21.0 0.15

Uric acid 86 88 4.2 0.51

Total PD 753 786 24.3 0.17

1CSD = Corn silage control diet; PMD = Pearl millet silage diet. 2Largest SEM published in table; n = 33 (n represents the number of observations used in

the statistical analysis). 3UUN = Urinary Urea Nitrogen. 4Estimated from urine creatinine concentration, assumed to be excreted at 29 mg/kg of

BW. 5PD = Purine derivatives.

156

Table 5-7. Effect of pearl millet silage on carbon dioxide (CO2) and methane (CH4)

emissions1 in lactating dairy cows

Diet2 P-Value


CO2 kg/d 13.6 14.0 0.42 0.24

CH4, g/d 396 454 18.4 <0.001

CH4, g/kg of DMI4 13.8 15.7 0.54 <0.01

CH4, g/kg of ECM4 8.28 9.58 0.386 <0.01 1Rumen gas emissions were measured using GreenFeed (C-Lock Technology Inc., Rapid


period. 2CSD = Corn silage control diet; PMD = Pearl millet silage diet. 3Largest SEM published in table; n = 30 (n represents the number of observations used in

the statistical analysis). 4Based on DMI and energy corrected milk yield data during the 3-d sampling periods.

157




fraction (of DM) at time t where a is the soluble fraction, b is the potentially degradable


Effective degradability (ED) was estimated as: ED = a + b {c ÷ (c + k)}, where a, b, and

c are as above and k is the rate of passage (Ørskov and McDonald, 1979) assumed to be

3%/h in this study: ED (model estimates ± SE) = pearl millet silage 52.2 ± 2.23c; alfalfa

haylage 58.6 ± 0.78b; corn silage 62.3 ± 0.50a. a,b,c Means without a common superscript

differ (P < 0.05)

158


0 20 40 60 80 100

ND

F d

issa

pp

eara

nce

, %

of

ND

F

0

20

40

60

80

Pearl Millet Silage

Alfalfa Haylage

Corn Silage


Data are means ± SE (n = 4). Disappearance curves of pearl millet silage, alfalfa haylage,

and corn silage were fit using SigmaPlot 10.0 (Systat Software, Chicago, IL) to the

equation p = a + b (1 - e-ct), where p is the degraded fraction (of NDF) at time t where a is

the soluble fraction, b is the potentially degradable fraction, and c is the rate of

degradation of the b fraction (Ørskov and McDonald, 1979). Effective degradability (ED)

was estimated as: ED = a + b {c ÷ (c + k)}, where a, b, and c are as above and k is the

rate of passage (Ørskov and McDonald, 1979) assumed to be 3%/h in this study: ED

(model estimates ± SE) = pearl millet silage 37.3 ± 3.17a; alfalfa haylage 27.0 ± 1.15b;

corn silage 21.9 ± 1.40c. a,b,c Means within the same row without a common superscript

differ (P < 0.05).

159

Chapter 6

Conclusions and Future Research

Conclusions

Dairy cattle do not have a requirement for feed ingredients (e.g. corn silage); they have a

requirement for nutrients. Yet, this must be combined with the reality that feed ingredients, not

nutrients, are grown from the land and fed to the cows. Many different forages can be fed to dairy

cattle, but ideal forages grown for lactating dairy cattle should be palatable, highly digestible, and

contain a nutrient profile that profitably compliments the other feed ingredients in the ration.

Forages for lactating cows must be palatable because the cows must voluntarily consume their

feed. Additionally, forages for lactating cows must be highly digestible to allow for high dry

matter and digestible nutrient intakes to maintain high milk production with a limited period of

negative energy balance in early lactation. Finally, forages for lactating cows should aim to

provide more expensive nutrients (e.g. protein) instead of less expensive ones (e.g. starch) which

can be economically purchased off farm. The price and/or availability of feed nutrients varies

widely by farm, year, and geographic location; however, any particular farm should identify what

nutrients the forages (and specifically the forages for lactating cows) on the farm are targeted for.

Are they to provide: Physically Effective Fiber, Highly Digestible Fiber, Energy, or Protein? The

answers to those questions help guide the cropping decisions on farms. Large quantities of high

quality forage (e.g. high digestibility, high CP, low lignin) decrease the need for purchased feeds

and increase the profitability of a dairy farm.

In addition to the above discussion, factors aside from the cow influence cropping

decisions. These include disease pressure, water availability (irrigation and rainfall), slope and

160

soil erodibility of the fields, nutrient management, and stocking density on the farm. Some

alternative forages are beneficial because of high water use efficiency (e.g. sorghum and pearl

millet). Others are able to grow in the cooler ‘shoulders’ of the growing season in addition to

providing soil cover during the winter (e. g. rye, oats, wheat, and triticale). Utilizing alternative

forages increases the crop biodiversity on a given farm and can decrease pests and diseases that

affect corn and may accumulate during corn-after-corn rotations. A main challenge of alternative

forages on dairy farms is achieving high yields with acceptable digestibility for use in diets for

high producing cows. Where alternative forages are best used depends on the forage and its

strengths (e.g. sorghum for dry ground, triticale for rapidly available NDF in high DMI cows).

The experiments presented in this dissertation measured the effects of replacing 10% of

the diet DM of corn silage with alternative forage silages of sorghum, oat, wheat, triticale, or

pearl millet. In general, diet NDF and CP were increased, but starch was decreased. The silages

were tested in 3 different lactating cow trials consisting of cows with different levels of milk

production so direct comparison is not possible, but the basic production outcomes are

summarized. Only the sorghum silage diet decreased DMI compared with corn silage. The other

AFS had similar DMI to corn silage. Milk yield was decreased for all treatments except oat

silage, but ECM yield was only decreased for wheat and triticale silages. The warm season annual

(sorghum and pearl millet) silages increased milk fat concentration, whereas the cereal silages

had no effect on milk fat concentration compared with corn silage. In situ NDF degradability was

similar for oat, wheat, triticale and pearl millet silages. Sorghum silage had the lowest NDF

degradability curve of all the alternative forages but it was still higher than corn silage. Estimated

total-tract NDF digestibility of the diets was increased for oat, triticale and pearl millet silages,

similar for sorghum silage and decreased for wheat silage compared with corn silage control

diets. Whereas, the estimated total-tract DM digestibility was decreased for wheat, sorghum, and

pearl millet silage diets, similar for the triticale silage diet and increased for the oat silage diet.

161

The increased NDF degradability of the alternative forage silages did not fully counterbalance the

decrease in starch content compared with corn silage. If alternative forage silages are included in

lactating cow diets on an NDF basis, increases in production might be seen due to the higher

NDFD of alternative forage silages leading to more nutrients absorbed by the cow.

Future Research

Research needs to continue on developing alternative forage uses in lactating dairy cows

both from an agronomic and an animal nutrition perspective. Agronomically, future studies

should focus on increasing the DM yield and NFC content of alternative forages without

sacrificing NDF digestibility. This could potentially be realized in the future through gene editing

technology, such as that used to create low lignin HarvXtra® alfalfa, if society comes to accept

such a method of plant manipulation. Crop transformations that suspend or slow maturity at the

flag leaf stage would be useful by extending the currently narrow harvest window for small grain

silage harvest. Crop alterations or harvest/ensiling technologies should be developed that allow

for direct cut harvest, as in corn silage, as opposed to the need for multiple passes with the cut

and wilt harvest methods. Further refining of planting and harvesting dates and ways to overlap

growing seasons, such as with intercropping, could be useful. Additionally, finding ways to

practically row crop polycultures could be impactful, including the use of agroforestry such as

practiced in Brazil with eucalyptus and Brachiaria. Work should continue on increasing yields of

perennial grain crops such as Kernza®, a perennial wheat being developed by The Land Institute,

to decrease the need for annual planting. Touching on protein, crop alteration to create more

rumen undegradable protein that is simultaneously digestible in the abomasum should be

considered.

162

Future alternative forage animal nutrition studies should focus on replacing corn silage

DM with both a higher digestible alternative forage NDF and a NFC source, such as sugar or

starch, for cows in peak lactation when rumen fill often limits intake in high producing cows. It

would be interesting to look at feeding highly digestible NDF of cereal grain silage to fresh cows

with the thought of reducing hepatic oxidation and its negative impact on DMI. Alternative

forages with less digestible NDF, such as soft dough sorghum or barley silages, should be tested

as feeds in rations for late lactation and dry cows. The extent to which passage rate and rumen fill

is affected by alternative forages should also be investigated to determine forage chop length

recommendations for different diets containing various forages in combination. Another curious

research topic is to look at twice a day feeding and test if the higher digestible NDF alternative

forages would be better utilized by feeding more of that forage in the morning or afternoon based

on circadian patterns of behavior and metabolism. Potential benefits of new transgenic crops yet

to be developed will also have to be validated in animal trials.

Many opportunities for future research exist relating to alternative forages. I believe the

need for alternative forage research will only grow as the world population increases and

marginal land that is not suited to corn silage becomes of increasing importance in crop and

livestock production. Our future is bright and full of promise; let’s go see what it holds for us!

163

Appendix

Indigestible NDF1 of various ensiled forages

Forage iNDF2, % of DM iNDF, % of NDF NDF, % of DM

Corn Silage (Exp. 1) 11.2 27.9 40.2

Corn Silage (Exp. 3) 13.2 35.7 36.8

Alfalfa Haylage (Exp. 2) 27.0 61.0 44.2

Alfalfa Haylage (Exp. 3) 23.0 54.4 42.2

Sorghum Silage (Exp. 1) 22.0 35.1 62.7

Pearl Millet Silage (Exp. 3) 11.7 20.1 58.4

Wheat Silage (Exp. 2) 12.1 23.6 51.0

Triticale Silage (Exp. 2) 9.2 18.0 51.3

Oat Silage (Exp. 1) 9.3 17.1 54.7 1Determined on 4 mm dry ground forage sample incubated in the rumen of a lactating

dairy cow for a period of 288h. 2iNDF = indigestible neutral detergent fiber

VITA

Michael Thomas Harper

Education

The Pennsylvania State University, University Park, PA

Ph.D. Animal Science Spring 2018

M.S. Animal Science Summer 2010

B.S. Animal Science Fall 2006

o Minors: Equine Science, Spanish, and International Studies

Selected Publications

Harper, M. T., A. Melgar, J. Oh, K. Nedelkov, G. Sanchez, G.W. Roth, and A. N.

Hristov. 2018. Inclusion of brown midrib dwarf pearl millet silage in the diet of lactating

dairy cows. J. Dairy Sci. 101. Accepted.

Harper, M. T. 2017. BMR sorghum and oat silage as dairy cow feed. Progressive

Dairyman. May 2017. www.progressivedairy.com/topics/feed-nutrition/bmr-sorghum-and-

oat-silage-as-dairy-cow-feed

Harper, M. T., J. Oh, F. Giallongo, J. C. Lopes, G.W. Roth, and A. N. Hristov. 2017.

Using brown midrib 6 dwarf forage sorghum silage and fall-grown oat silage in

lactating dairy cow rations. J. of Dairy Sci. 100(7):5250 – 5265 doi: 10.3168/jds.2017-

12552

Harper, M. T., J. Oh, F. Giallongo, G. W. Roth, and A. N. Hristov. 2017. Inclusion

of wheat and triticale silage in the diet of lactating dairy cows. J. of Dairy Sci.

100(8):6151 – 6163 doi: 10.3168/jds.2017-12553

Harper, M. T., J. Oh, F. Giallongo, J. C. Lopes, H. L. Weeks, J. Faugeron, and A. N.

Hristov. 2016. Short communication: Preference for flavored concentrate premixes by

dairy cows. J. Dairy Sci. 99(8):6585-6589. doi: 10.3168/jds.2016-11001

Hristov, A. N., J. Oh, F. Giallongo, T. W. Frederick, M. T. Harper, H. L. Weeks, A.

F. Branco, P. J. Moate, M. H. Deighton, S. R. O. Williams, M. Kindermann, and S. Duval.

2015. An inhibitor persistently decreased enteric methane emission from dairy cows

with no negative effect on milk production. Proc. Natl. Acad. Sci. 112(34):10663-10668.

doi: 10.1073/pnas.1504124112