See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/270160140

Changes in Forage Quality During Harvest and Storage

Chapter · January 1994

DOI: 10.2134/1994.foragequality.c20

CITATIONS READS

234 4,694

2 authors, including:

C.A. Rotz
Agricultural Research Service
319 PUBLICATIONS 9,117 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Dairy Agroecosystem Working Group View project

Environmental and Economic Sustainability of U.S. Grassfed Beef Production Systems View project

All content following this page was uploaded by C.A. Rotz on 29 December 2014.

The user has requested enhancement of the downloaded file.

 Published 1994

CHAPTER 20
CHANGES IN FORAGE QUALITY DURING HARVEST AND STORAGE

C. Alan Rotz and Richard E. Muck

INTRODUCTION

In many regions of the world, the climate is not suitable for forage growth
during much of the year. In these regions, forage must be conserved through
harvest and storage to feed animals during the months when fresh forage is not
available. A primary ~oal in forage conservation is to maintain the crop dry
matter (DM) and nutnents with minimal loss. The amount of loss that occurs
is influenced by the type and size of equipment and storage facilities used,
management decisions, and weather. The major constraint for reducing losses
is the cost of the required technology. Costs of production must be balanced
with system performance (including losses) to select the best forage systems.
Many harvest and storage systems are used, but the major options are dry
hay and silage production. Hay can be harvested in bales of various sizes,
shapes, and densities. To produce dry hay, forage is normally dried in the field
to a moisture content of less than 200 g/kg. Hay may be harvested at moisture
contents up to 280 g/kg, but further processing or treatment is required for
proper storage. Field curing of dry hay in thin, wide swaths requires 3 to 5 d
of drying. In heavier windrows, field drying may require 6 to 7 d. Rain during
the curing process can prolong the drying many more days. Hay that lays in the
field more than two weeks is often not suitable for animal feed.
Losses during hay harvest and storage range from 15 to 100% of the
initial forage DM. For hay dried under relatively good drying conditions, losses
are 15 to 18% (Rees, 1982; Rotz and Abrams, 1988). Rain damage increases
the loss by up to 30% (Rotz and Abrams, 1988) and with extended poor drying
conditions, the whole crop can be lost. Average losses in hay making are
estimated between 24 and 28% (Hodgson et al., 1947; Hoglund, 1964; Waldo
and Jorgensen, 1981; Wilkinson, 1981; Buckmaster et aI., 1990). Most of this
loss occurs durin~ harvest with about 5% loss during the storage of dry hay.
To make sIlage, forage is normally wilted in the field to a moisture
content between 500 and 650 g/kg which reduces field curing time to between
1 and 4 d. Silage is stored in tower and bunker silos, stacks, bags and wrapped
large round bales. The handling of wetter material and the reduced field curing
time lead to lower harvest losses than in hay systems, but storage losses are
greater. Average losses in silage eroduction are 14 to 24% with about half of
this loss occurring during storage (Hodgson et al., 1947; Hoglund, 1964; Waldo
and Jorgensen, 1981; Wilkinson, 1981; Buckmaster et aI., 1990).

c.A. Rotz, Agricultural Engineer, U.S. Dairy Forage Research Center, USDA/
Agricultural Research Service, 206 FarraH Hall, Michigan State Univ., East
Lansing, MI 48824; R.E. Muck, Agricultural Engineer, U.S. Dairy Forage
Research Center, USDA/ ARS, 1925 Linden Drive West, Univ. of Wisconsin,
Madison, WI 53706.

828
 Losses in hay and silage production include: 1) the physical detachment
of forage material and 2) the internal depletion or degradation of plant
nutrients. Physical separation is normally caused by harvest equipment, but
rain also may disassociate forage material. This loss is primarily leaves.
Because leaves have a higher concentration of nutrients, leaf loss causes a
reduction of the nutrient concentration in the remaining forage. Nutrients
depleted from forage are primarily soluble carbohydrates, but other soluble
nutrients may be lost. Soluble carbohydrates are used by plant and microbial
respiration whereas all soluble nutrients are susceptible to leaching by rainfall.
These losses cause a substantial reduction in highly digestible DM, feed
nutrients, and ener~. Overall, the sum of these losses (physical detachment
and nutrient depletIOn) typically cause a gain in neutral detergent fiber (NDF)
concentration of 30 to 120 g/kg DM and either a small reduction or a small
gain in crude protein (CP) concentration (Hodgson et aI., 1947; Johnson et aI.,
1984; Rotz and Abrams, 1988).
The economic value of forage losses can be substantial. This value is a
function of the amount of material lost, the effect of the loss on the nutrient
concentration of the remaining forage, and the end use of the forage. As such,
a comprehensive analysis of forage systems is required to determine the value
of losses. On Michigan dairy farms with an annual milk production of 8000
kg/cow-yr, the average value lost ranges from $14/t DM of alfalfa (Medicago
sativa L.) produced in silage systems to $35/t DM in dry hay systems
(Buckmaster et aI., 1990). The lost value is even ~reater for higher producing
dairy herds or commercial hay systems where hay IS priced according to weight
and forage quality. Based upon recent forage production statistics, the value
of harvest and storage losses in the U.S. exceeds $2 billion/yr.
Fora~e losses can be divided into five major categones. These are: 1)
plant respIration loss; 2) rain damage; 3) machine-induced loss; 4) hay storage
loss, and 5) silo storage loss. Losses and quality changes occurring in each of
these categories will be discussed. The discussion includes a brief description
of the processes involved, the factors influencing loss, and typical losses and
quality changes occurring in each category.

RESPIRATION LOSS
Physical, biological, and chemical changes occur in forage as it dries in the
field. The primary biological or metabolic loss is due to plant respiration, but
other minor changes also occur. This section discusses all DM losses and
nutrient changes during field curing except those caused by rainfall or machine
treatment.

Factors Which Affect Respiration

Respiration performs a vital role in living plants. The hydrolytic and

respiratory enzymes present in living cells continue to function after the crop
is mowed until some lethal condition intervenes (Sullivan, 1973). As the crop
dries, respiration decreases because diffusion processes supplyin~ oxygen within
the plant are inhibited and the enzyme-catalyzed reactions WhICh require the
presence of water are reduced (Parkes and Greig, 1974). Respiration decreases
nearly in proportion to the decrease in moisture content expressed on a wet
basis (Wood and Parker, 1971) approaching zero between 260 and 400 gjkg
moisture (Greenhill, 1959; Wolf and Carson, 1973). Within this moisture
range, plant enzymatic activity ceases or at least drops below a detectable level.
Rewetting of the crop by dew or rain reactivates enzyme activity and thus

829 ROTZ & MUCK

prolongs respiration. Pizarro and James (1972) found that rewetted grass
respired at a similar rate as non-rewetted grass at a similar moisture content.
Prolonged field curing also may allow the development of bacteria, yeasts,
and fungi on forage (Pizarro and Warboys, 1972). Respiration of these
microbial organisms increases the overall respiration rate (Honig, 1979). Since
respiration by plant tissue and microbial flora use the same substrate, they have
similar effects on loss and nutrient change in the crop (Pizarro and James,
1972). Microbial respiration effects are small during field curing except over
extended poor drying conditions.
Plant respiration increases with temperature from no respiration at the
freezing point of the forage plant. This temperature is about -4°C for alfalfa,
but it may be slightly lower for some grass species (Wilkinson and Hall, 1966).
Respiration rate increases exponentially to a temperature of 27°C. At higher
temperatures, the rate levels off reaching a maximum at about 45°C. At 55°C,
enzymatic activity and respiration cease in alfalfa (Wolf and Carson, 1973).
Plant maturity may influence respiration rate during wilting, but consistent
effects are not reported. Pizarro and James (1972) found a marked decline in
the respiration rate of perennial ryegrass (Lolium perenne L.) with increasing
maturity. Respiration rate in grass at the emer~ence of inflorescence was
nearly five times that in grass 30 dafter anthesIs was complete. In other
studies, no variation in respiration rate was found across maturities of ryegrass
(Wood and Parker, 1971) and alfalfa (Wilkinson and Hall, 1966). The differing
results may be due to more rapid dryin~ of the younger crop. With faster
dryin~, respiration rate declines more qUIckly, offsetting some of the effect of
the hIgher initial respiration rate (Melvin and Simpson, 1963).
Respiration rate also may vary across species, but little work has directly
compared respiration among species as they dry. Morris (1972) found similar
respIration-induced losses among five grass species dried slowly in the
laboratory. Wilkinson and Hall (1966) reported that wheatgrass (Agropyron)
appeared to have a higher respiration rate than that of alfalfa.
Conditioning treatments used to speed forage drying may alter respiration
rate. Mechanical crushing can stimulate a small increase in respiration, but this
effect is not consistently reported (Simpson, 1961; Wood and Parker, 1971).
The faster drying obtained by conditioning causes a more rapid drop in
respiration rate which likely offsets any initial increase.
Under suitable conditions, photosynthesis may continue for a short time
after the crop is mowed (Greenhill, 1959). If photosynthesis occurs,
carbohydrate DM is added to the plant offsetting some of the respiration loss.
Published data suggest that very small DM increases occur when climatic
conditions promote rapid drying, but under cloudy conditions which extend
plant life, the gain may be as much as 5% (Rees, 1982).
Dry Matter and Nutrient Losses
Initial plant respiration can be considered as the complete oxidation of
hexose sugar to carbon dioxide and water (Parkes and Greig, 1974). The
exothermic relationship is:
C6H 120 6 + 6 02 -+ 6 HzO + 6 CO2 + 2870 kJ
Soluble carbohydrates (primarily sugars) in the plant tissue provide the
principal substrates for respiration. The water, carbon dioxide, and heat
produced by respiration leave the plant, causing a DM loss. Respiration rate
rapidly declines when the readily available carbohydrates are depleted.
Substrates other than carbohydrates (fat and protein) are then used, but a more
complex conversion is required (Greenhill, 1959). Depletion of available

CHANGES IN FORAGE QUALITY 830

 carbohydrates normally does not occur during field wilting unless the crop is
heavily damaged by rain.
Most studies show good agreement between theoretical and measured
values of DM loss, CO~ production, and heat production (Greenhill, 1959;
Wilkinson and Hall, 1906; Wood and Parker, 1971). In the one exception,
Parkes and Greig (1974) measured DM loss that was six times that determined
based upon measured CO 2 production. They suggested that a fermentation
process occurred that converted sugar to lactIc aCId, ethanol, and CO 2•
Dry matter losses due to plant respiration are difficult to measure during
field curing and measured values vary widely. Losses measured in field-cured
alfalfa vary between -8 and 19% of the initial crop DM (Rotz and Abrams,
1988). The loss is influenced by drying conditions with the ~reatest loss in
material dried in a warm, humid environment. For alfalfa dned under good
weather conditions with less than 4 d of field curing, average DM loss due to
respiration appears to be 3 to 4% (Rotz et aI., 1984; Rotz and Sprott, 1984;
Rotz et aI., 1987; Rotz and Abrams, 1988). Dry matter loss of ryegrass and
white clover (Trifolium repens L.) dried in the laboratory averages 5 to 7%. In
warm humid conditions, the loss in grass species can exceed 10% (Morris,
1972).
Carbohydrates are the principal compounds used in respiration with little
loss of total nitrogen (N) and fiber. In rye~rass, fructosans and total soluble
fructose residues decrease throughout drymg (Melvin and Simpson, 1963).
Sucrose content may decrease early in the drying process but increase
appreciably during the latter part of drying. Sucrose may be synthesized from
glucose or fructose in leaf tissue to provide the net increase in sucrose.
Glucose content of ryegrass does not vary consistently during drying. In alfalfa,
the greatest losses are glucose and fructose with some decrease in sucrose
(Sullivan, 1973).
Little loss of total N normally occurs during field drying, but some protein
breaks down to simpler, nonprotein nitrogen (NPN)(Brady, 1960; Melvin and
Simpson, 1963; Carpintero et aI., 1979). Protein breakdown in the presence of
suffIcient moisture is very rapid and the extent of degradation is influenced by
the time required to dry the crop. With very rapid wilting (less than 6 h),
protein breakdown can be negligible (Carpintero et aI., 1979). Based upon the
work of Macpherson (1952), NPN increases 5 to 6 gjkg of total N each hour
during field wilting until the crop dries to 600 gjkg moisture. At moisture
contents below this level, further de~radation of protein is negligible. A
moderate increase in temperature dunng drying increases the rate of protein
breakdown. However, warmer temperatures normally provide faster drying so
the net effect on degraded protein is small.
Protein degraded during wilting forms a rapid accumulation of amides,
particularly asparagine (Macpherson, 1952; Melvin and Simpson, 1963). Some
of the nitrogenous constituents are decomposed to give CO2 and ammonia with
the latter then converted to amide. Other NPN forms include free amino acids,
volatile bases, and peptides. Protein changes during wilting result in decreased
true protein solubility (Brady, 1960; Rotz and Abrams, 1988).
In summary, plant respiration during field drying results primarily in the
loss of soluble carbohydrates. Consequently, the concentration of CP, crude
fiber, acid detergent fiber (ADF), NDF, lignin, and many other forage plant
constituents not affected by respiration increase in proportion to the loss of
other DM. Within the CP fractIOn, some true protem is converted to soluble
NPN.

831 ROTZ&MUCK
 RAIN DAMAGE
When rain occurs during field curing, the loss of yield and quality of
forage can be high. Yield or OM losses of up to 30% are commonly reported
(Shepherd et al., 1954; Schukking and Overvest, 1979; Van Bockstaele et ai.,
1979; Wilman and Owen, 1982; Collins, 1985; Fonnesbeck et al., 1986; Rotz
and Abrams, 1988) with a loss of over 50% reported with heavy rain damage
(Collins, 1983). Most of the OM lost consists of highly soluble and digestible
plant nutrients so that forage quality is disproportionately reduced. With heavy
DM losses (greater than 25%), the crop is normally not acceptable for animal
feed, resulting in a complete loss as feed.

Factors Which Affect Loss

The amount of loss incurred by rain damage varies widely, dependent

upon many crop and environmental factors. The major factors WhICh affect loss
are the characteristics of the rain including amount, intensity, and duration.
Crop factors such as moisture content at the time of rain, maturity, leaf-to-stem
ratio, swath density, plant species, and the crop conditioning treatment also
influence loss. Ram can indirectly cause other losses. Additional mechanical
treatments may be needed for timely drying of the crop. Also, delayed harvest
can retard crop regrowth, further reducing DM yield over the harvest season.
Environmental Factors
Leaf loss is one form of damage caused by rain. The impact of raindrops
on the crop causes some leaves to sever from the stem and fall to the soil
surface. This loss is important in legume species but less so in grass species.
Reported values for leaf loss in legumes vary widely. In a field study, Shepherd
et al. (1954) measured increased leaf loss of 8.8% in alfalfa subjected to two
showers and 36% with three showers. Under artificial rainfall, leaf losses were
less than 4% of the crop DM (Rotz et al., 1991b). Using a combination of
artificial and natural rainfall, Collins (1983) found that leaf loss in alfalfa and
red clover (Trifolium pratense L.) increased with the amount of rainfall,
averaging 5.7,8.0, and 10.0% for 25, 41, and 62 mm of rain, respectively.
The leaching of soluble plant nutrients is the predominant loss from rain
damage. The influence of rain and crop characteristics on leaching loss is
difficult to measure in the field under natural rainfall. Field measurements
often include leaf loss as well as leaching loss and in some cases increased
mechanical losses following rain. Published data on leaching loss also include
plant and microbial respiration losses fOllowing the rain. When the plant is
rewetted, respiration of plant tissue and microflora on the plant continue until
the cro~ is again below about 300 gjkg moisture. Since both leaching and
respirabon remove many of the same plant constituents, it is impossible to
separate the two losses. Field studies show a trend for loss to mcrease in
proportion to the amount of rain. In general, the increased loss is about
0.7%/mm of rain, but it may be as high as 1%/mm (Shepherd et aI., 1954; Van
Bockstaele et ai., 1979; Collins, 1983; Rotz and Abrams, 1988).
More controlled studies using artificial rainfall indicate that leaching loss
increases with the amount and duration of rainfall and tends to decrease with
increased rain intensity. Rotz et al. (1991b) found that alfalfa exposed to rain
at a rate of 18 mm/h fost about 0.1 % DM/mm of rain. With the rain amount
held constant at 18 mm, the loss was about 3% DM for the first hour of rainfall
and 1.1 %/h for longer durations. With an increase in rain intensity, duration
for a given amount of rain decreased reducing the loss per unit of rain.
Experiments which use artificial rainfall tend to show lower losses than
those which occur under natural rain. Lower losses are due to differences in

CHANGES IN FORAGE QUALITY 832

rainfall duration and the kinetic energy of the raindrops. Natural rainfall
normally occurs over several hours and may I?ersist for several days. Artificial
rain often is applied in a relatively short penod of 5 min to 1 h, and the crop
begins to redry soon after the water is applied (Collins, 1982; Collins, 1983;
Rotz et aI., 1991b). With a shorter duratIOn, losses are expected to be less.
The kinetic energy absorbed by the crop during rain is related to the size
and velocity of raindrops. Artificial raindrops are normally smaller in size
and/or lower in velocity than those of natural rain. A five-fold increase in drop
diameter can increase the kinetic energy of the drop by 600 times (Rotz et aI.,
1991b). Although not proven, the energy absorbed on Impact likely has a large
effect on leaf shatter and the leaching of soluble nutrients. Since droplet size
varies widely among natural rain occurrences, this also may explain some of the
difference in loss incurred among showers and storms of different intensities.
Crop Factors
Many crop characteristics affect the amount of loss that occurs with rain,
but few show consistent effects. Perhaps the greatest effect is from crop
moisture content at the time of rainfall. Most experimental data indicate that
rain early in the drying process causes little loss (Murdoch and Bare, 1963; Van
Bockstaele et aI., 1979). McGechan (1989b) noted an increase in loss with a
decrease in crop moisture. Artificial rainfall on alfalfa of 400 g/kg or less
moisture caused slightly greater leaching of plant nutrients and 15% greater
leaf loss compared to alfalfa with 650 g/kg moisture (Rotz et aI., 1991b).
Among legumes, plant species does not appear to affect leaf loss except
that loss appears to be higher in plants with a greater leaf to stem ratio.
Collins (1983) measured similar losses for both alfalfa and red clover with a
trend toward greater loss in the more leafy clover. Stage of maturity did not
affect leaf loss from either species. In another study, less rain-induced leaf loss
occurred in legume-grass IDlXtures than occurred in pure stands of alfalfa, red
clover, and birdsfoot trefoil (Lotus comiculatus L.) (Collins, 1985). Losses that
occurred in the mixtures were primarily from the legumes.
Leaching losses likely vary among forage species, but few studies have
compared species under the same conditions. Extensive comparisons ofleaching
losses among alfalfa, red clover, and birdsfoot trefoil show inconsistent results
(Collins, 1982, 1983, 1985). In the 1982 study, leaching of nonstructural
carbohydrates and cell contents was highest in red clover and lowest in alfalfa.
In the 1983 study, leaching losses were the same in alfalfa and red clover, and
in the 1985 study, leaching losses were highest in alfalfa, moderate in red
clover, and relatively low in birdsfoot trefoil. In spite of inconsistencies among
the studies, one trend was evident: forage plants with the lowest NDF content
(highest concentration of cell contents) consistently had the greatest loss.
Leaching losses have not been directly compared between grass and
legume species. Collins (1985) reported that DM losses from legume-smooth
brome~rass (Bromus inermis Leyss.) mixtures were about 20% greater than
those In respective pure legume stands. This suggests that grass was more
susceptible to leaching. In the same experiment, however, fiber concentration
increases in the forages indicated a greater leaching of nutrients from the pure
legume crops. Field losses measured in Coastal bermudagrass (Cynodon
dactylon (L.) Pers) were very low per unit of rainfall (O.13%/mm) compared to
those commonly measured for legumes and other grasses (Hart and Burton,
1967). In a senes of field trials, rain-induced losses appeared to be greater in
Italian l1egrass (Lolium multiflomm Lam.) than in the perennial ryegrass
species (Van Bockstaele et aI., 1979). Based upon the liIDlted .available data,
species effects appear most closely related to the initial concentration of cell
contents in the forage.

833 ROTZ & MUCK

 Crop maturity also influences the concentration of cell contents in the
forage and thus may influence loss. Collins (1983) found slightly less loss in
both alfalfa and red clover at a late bloom stage of maturity compared to that
at bud stage. In an extensive field study with a grass-white clover mixture,
stage of maturity did not affect the extent of reduction in digestible DM caused
by rain (Wilman and Owen, 1982). Under artificial rain, Rotz et al. (1991b)
found no correlation between alfalfa maturity and loss. These results indicate
that crop maturity effects are often small and relatively unimportant.
Conditionin~ treatments applied to hay to speed drying may increase the
crop's susceptibihty to rain loss. Crushing or crimping has a small effect,
perhaps increasing rain loss by 20% (Murdoch and Bare, 1963; Rotz et aI.,
1987). Forage cut with a flail mower or conditioner can be more susceptible
to damage. Rain-induced losses were reported to double with this type of
treatment of grass crops, but little difference was found in alfalfa (Svensson,
1978; Rotz and Sprott, 1984).
Swath density may affect loss because forage spread in a thin swath is
more exposed to the rain. Rotz and Sprott (1984) found slightly greater rain-
induced loss in alfalfa left in a wider swath. In a grass crop, Wilman and Owen
(1982) found swath thickness to have no effect on the extent of reduction in
di~estibility caused by rain. Under better controlled conditions with artificial
ramfall, leaching loss was not greatly affected by swath thickness of alfalfa, but
leaf loss in heavy swaths was about half that in light swaths (Rotz et aI., 1991b).

Other Factors
The time of day when rain occurs appears to have little effect on loss
when all other conditions are equal. Rain-induced losses were similar when
artificial rain was applied early in the day with rapid redrying following the rain
or in the evening WIth little drying overni&ht (Rotz et aI., 1991b).
Delayed regrowth of the forage crop IS an indirect loss from rain damage.
Regrowth is retarded by about the same number of days as the length of the
wilting period (Schukking and Overvest, 1979). This delay is caused by shading
of the stubble by the cut forage and dama~e to the initial re~rowth by delayed
machine operations. A delay in the apphcation of N fertihzer also can be a
factor on grass crops. Delayed regrowth may affect the quality of the following
harvest, but this effect is dIfficult, if not impossible, to predict.
Extra machinery operations required to dry the rewetted crop cause loss
indirectly related to rain damage. Machine losses and their effect on forage
quality are addressed later in this chapter. Losses caused by extra operations
used following rain may be attributed to rain damage.
In summary, major factors that 60...--------------,
affect rain-induced losses are the
crop species, crop moisture content, ~ Hay moisture content
and the amount of rain. Other minor "-' 40
influences come from crop maturity ~ 300 lIk,
and the form of conditioning applied ..s so
to the crop. Trends in the effects of ~
various factors on rain damage can ~ 20
be established considering all e
available information just discussed. 1>\ 10
Typical losses for a mechanically ""
conditioned alfalfa crop are Q
illustrated in Figure 1. This typical 20 30 .0 60
loss includes leaf shatter and Total rainfall (CIn)
leaching caused by rain and the plant
and microbial respiration caused by Figure 1. Typical rain-induced losses
the delay in drying. for mechanically conditioned alfalfa.

CHANGES IN FORAGE QUALITY 834

 Effects on Forage Quality

Dry matter losses caused by rain damage can greatly alter the quality of
the remaining fora~e. Leaf loss has a minor effect relative to other losses.
Since leaves contam a higher concentration of important nutrients for the
animal, any loss of leaves results in an overall reduction in nutrient
concentration. For example, alfalfa consisting of 50% leaves, a leaf CP content
of 280 g/kg, and a stem CP content of 120 g/kg has an overall CP content of
200 g/kg. A 10% loss of leaves reduces the CP concentration in the remaining
forage to 196 g/kg. Although a substantial loss of DM and nutrients occurs,
the nutrient concentration in the remainin~ forage is reduced only slightly.
Leaching loss and the microbial respIration resulting from rain damage
have a more marked effect on the concentration of many plant nutrients. Rain
leaches the more soluble nutrients from forage. These include readily available
carbohydrates, soluble N, soluble minerals, and lipids (Fonnesbeck et aI., 1986).
These nutrients come primarily from the cell contents of the forage plant, and
they are highly digestible nutrients for the animal. As a result, rain damage
causes a substantial decrease in digestibility and an increase in fiber concen-
tration (Collins, 1983; Rotz and Abrams, 1988).
Experimental data indicate that the portion of nutrients lost varies among
the major soluble nutrients. Nutrient loss is likely related to the solubility and
concentration of plant constituents. Crop s.{>ecies, crop maturity, the growing
environment, and other factors affect nutnent concentration and solubility.
Available data are sufficient to draw general conclusions on nutrient losses, but
the effect of various factors on specific nutrient changes cannot be predicted.
These effects are expected to be relatively small, inconsistent, and relatively
unimportant in the analysis of rain dama~e.
Dry matter leached by rain is primanly nonstructural carbohydrate. Data
collected by Fonnesbeck et a1. (1986) indicate that about half of the DM lost
is available carbohydrate. Data from Collins (1982) generally support this.
Collins' data also suggest that loss from red clover is higher in nonstructural
carbohydrates than the loss from alfalfa and birdsfoot trefoil. The lost
carbohydrates are primarily sugars with a little loss of starch (Collins, 1982).
Measurements of N or CP loss from leaching vary widely. Reported data
indicate that the loss from alfalfa consists of between 100 and 500 g CP /kg DM
(Collins, 1982, 1983, 1985; Fonnesbeck et al., 1986; Rotz and Abrams, 1988;
Rotz et al., 1991b). In two studies, Collins (1983, 1985) found a higher portion
of the loss from alfalfa to be N compared to the loss from red clover or
birdsfoot trefoil. In another study, the opposite was found (Collins, 1982).
Limited data on loss from grass indicate that CP loss is similar to that for
legumes (Murdoch and Bare, 1963). Data reported by Rotz et a1. (1991b) did
not indicate consistent effects of rain and crop characteristics on the portion of
the loss which was CPo As an average or typical value, about 300 glkg of the
leached DM can be assumed to be CPo
Nitro~en leached from the plant tissue is highly soluble, so the
concentratIOn of water-insoluble protein is increased (Rotz and Abrams, 1988).
The concentration of acid detergent insoluble protein (ADIP) also is increased.
Rotz et al. (1991b) found that the increase in ADIP was often a little greater
than that explained by elution of soluble protein. Rain damage, therefore, also
may reduce the solubility of some of the remaining protein.
The remainder of the leaching loss consists of soluble minerals and lipids.
Data of Fonnesbeck et al. (1986) indicate that for alfalfa, soluble minerals and
total lipids each make up about 100 g/kg of the leached DM. In legume and
legume-grass mixtures, Collins (1985) measured a small (1.5 g/kg) decrease in
calcium concentration with little change in phosphorus, potassium, and
magnesium.

835 ROTZ & MUCK

 MECHANICAL LOSSES

Several machine operations are normally required to harvest forage, and

each operation causes additional loss. Mechanical losses affect forage quality
in two ways. First, the DM lost contains nutrients that would have benefitted
the animal consuming the forage. Second, the loss affects the nutrient
concentration in the remaining forage. Mechanical losses often include more
leaf material than stem material. Since leaves have a greater concentration of
many important nutrients, a change in the portion of leaves causes a change in
the nutrient concentration or quality of the remaining forage.
Operations used in forage harvest include mowing and conditioning, swath
manipulation, and harvest by baling or chopping. Forage losses incurred vary
widely among the operations. The amount of loss is influenced by the type of
machines used and the adjustment of those machines. Many crop factors also
influence the loss. These include crop species, maturity, moisture content,
portion of crop that is leaves, and swath structure. The following discussion of
the major operations includes the factors that influence loss and typical DM
losses and quality changes.

Mowing and Conditioning

Most harvested forage is mowed with a mower-conditioner, a machine

which combines mowing and conditioning in one operation. Mowing is most
often done with either a cutterbar or rotary disk device. Other devices such as
rotary drum and flail mowers are sometimes used, but high power requirements
and greater losses deter their use. Many types of mechanical conditioning
devices are employed to enhance forage drying. The most common device used
to condition alfalfa is intermeshing rubber rolls, but other forms of
intermeshing and non-intermeshing rubber and steel rolls are used. Flail type
conditioning devices often are preferred for mowing ~rass forage crops. Most
conditioning devices provide similar improvements m drying when properly
adjusted (Straub and Bruhn, 1975; Rotz and Sprott, 1984; Shinners et aI., 1991).
Losses reported for mowing and conditioning generally vary between 1
and 5% of the DM yield (Savoie et aI., 1982; Koegel et aI., 1985; Savoie, 1988;
Shinners et aI., 1991). This is a typical range in loss with well adjusted
cutterbar and rotary disk mowers. Rotary mowers may at times cause a little
more loss than that from cutterbar mowers, but under most conditions, losses
are similar (Rotz and Sprott, 1984; Koegel et aI., 1985; Shinners et aI., 1991).
Losses are much higher with flail mowing machines, averaging between 6 and
11 % of yield (Barrin~ton and Bruhn, 1970; Honig, 1979; Rotz and Sprott,
1984). Flail conditionmg devices adjusted for very aggressive conditioning also
may double the loss during alfalfa mowing (Koegel et aI., 1985).
Among roll-type conditioning machines, roll design has little effect on loss
(Shinners et aI., 1991). Adjustment of roll pressure and clearance has more
effect than the configuration or material used in the rolls. Crimping rolls,
which are rarely used today, may cause a little more loss than the more
common crushing-type rolls (Boyd, 1959; Kepner et aI., 1960). The difference
in loss between a well adjusted mower-conditlOner and a similar mowing device
without a conditioner appears to be about 1% (Kepner et aI., 1960; Dobie et
aI., 1963; Goss et aI., 1964; Rotz et aI., 1987). When the machine is adjusted
for more severe conditioning, loss may increase an additional 1 to 2% of yield.
Crop maturity influences mowing and conditioning loss in both legume
and grass crops (Savoie, 1988; Shinners et aI., 1991). The loss can double as
forage matures from a late vegetative to a full bloom stage of development.
The moisture content of the standing crop decreases with maturity. Perhaps
the lower moisture and/or a weakened attachment of leaves cause greater loss.

CHANGES IN FORAGE QUALITY 836

 Crop characteristics such as species and the amount of leaves likely affect
loss, but these effects are not well documented. Savoie (1988) found mowing
and conditioning loss in timothy (Phleum pratense L.) to be about 40% less than
that in alfalfa with all other conditions similar. Loss in a mixed stand fell
between those of the pure stands. When comparing full harvest systems, Ciotti
and Cavallero (1979) found similar loss differences among alfalfa, orchardgrass
(Dactylis !Jlomerata L.), and mixed stands. In le~me crops, greater loss is
expected m crops with more leaves. Koegel et al. (1985) observed greater loss
in more leafy second and third cuttings than occurred in first cutting alfalfa.
Since mowing and conditioning loss is normally small, the loss has only a
small effect on the quality of the remaining forage. Buckmaster et al. (1990)
estimated that 75% of thIS loss is leaves in alfalfa. Because leaves are higher
in CP and lower in fiber than stem material, mowing and conditioning loss
causes a small decrease in CP concentration and a small increase in fiber
concentration in the remaining forage.

Swath Manipulation

A variety of machine treatments are sometimes used to manipulate the

forage swath in an attempt to speed the field curing process. These treatments
are generally categorized as tedding, swath inverSIOn, and raking. Tedding
machines use rotating tines to spread and fluff the swath to improve drying.
Swath inverters lift and flip the swath over to expose the more wet forage on
the bottom. The swath structure and dimensions are not affected much by
inversion, but a small amount of widening and fluffing may occur. Raking
operations are used to roll the forage off the soil surface into a more narrow
windrow for pickup by a baler or forage harvester.

Tedding and Inversion

Loss caused by tedding is reported between 1 and 3% of the crop yield,
but much greater loss can occur (Murdock and Bare, 1960; Ciotti and
Cavallero, 1979; Savoie et aI., 1982). The beating action of the tedder is most
damaging to legume crops due to a more delicate attachment of leaves and
stems. Tedding loss can be three to four times greater in alfalfa than in grass
crops (Ciotti and Cavallero, 1979; Savoie, 1988). Within legume crops, a
~reater loss is expected as the leaf portion increases. Field measurements
mdicate less loss in first cutting alfalfa which normally has less leaves than later
cuttings (Savoie et al., 1982; Davis et al., 1989). Laboratory measurements show
less potential for loss in birdsfoot trefoil than alfalfa and less loss as the crops
mature (MacAulay and Bilanski, 1968; Raghavan and Bilanski, 1973). These
effects also may be due to a difference in the portion of leaves on the plant.
Leaf loss is greatly influenced by crop moisture content, particularly in
legume crops. As the crop dries, leaf shatter increases exponentially with a
very high potential loss in forage containing less than 300 gjkg moisture
(Savoie, 1988). Laboratory measurements indicate that this loss may exceed
20% of the yield if tedding is applied to very dry legumes (MacAulay and
Bilanski, 1968; Raghavan and Bilanski, 1973; Savoie, 1988).
Tedding loss affects the quality of the remaining forage by decreasing the
portion of leaves remaining WIth the crop. Savoie (1988) found the portion of
leaves in the lost material to increase from 60 to 80% as the moisture content
of alfalfa at the time of tedding decreased from 800 to 300 gjkg. With a high
tedding loss in alfalfa, up to a 10 gjkg DM decrease in CP and a 20 gjkg DM
increase in NDF concentrations can occur as a result of the operation.
Swath inverting machines provide a more gentle method for manipulating
field curing swaths. Although many designs are used, these machines normally
include a pickup device that lifts the swath onto a platform or moving belts.

837 ROTZ&MUCK
 fOr------------------------.
The swath is turned and dropped
back to the soil surface inverted from
its original position (Savoie and Swath thickness
Beauregard, 1990). Measured ht

increases in field loss due to swath

inversion are small, varying between
oand 1.5% of yield in predominantly
alfalfa crops (Davis et aI., 1989;
Savoie and Beauregard, 1990).
When a swath is inverted onto the
same soil surface where a swath was
removed, succeeding operations may foo 200 300 400 600
pick up some of the material missed Crop moisture content (g/kg)
by the swath inverter. Thus, the
actual loss may be very small. The Figure 2. Typical dry matter losses
limited available data show little from raking forage with a side-delivery
relationship between the amount of rake.
loss and crop characteristics,
including moisture content.
Raking
Raking loss varies widely with reported losses ranging from 1 to 20% of
crop yield (Dobie et al., 1963; Gordon et al., 1969; Savoie et aI., 1982;
Buckmaster, 1993). Raking loss is influenced by crop moisture content
(Buckmaster, 1993) and the density of the forage laying in the swath (Klinner
and Shepperson, 1975; Rotz and Abrams, 1988). Loss increases as the crop
moisture content decreases, particularly below 300 gjkg (Fi~re 2). When the
crop is spread over much of the field surface, it is more diffIcult to gather with
the rake, and loss is increased. Although comparisons of raking loss in grass
and legume crops have not been made, available data indicate similar losses
among species at similar moisture contents.
The amount of raking loss varies among machine designs, but this
influence is not adequately investigated. Rice (1966) compared six rake designs
on grass and found little consistent difference in loss among machines. There
was a trend toward less loss with a wheel rake compared to the side-delivery
rake. This result is not consistent with other reports (Friesen, 1978;
Buckmaster, 1993), so the difference was likely due to machine adjustment.
Rice (1966) found a trend for greater loss with a traverse chain design, and
Savoie et al. (1982) found greater loss with a rotary windrower. These two
designs use tines to sweep hay into a windrow. The sweeping action likely
allows more crop to become entangled with the stubble and lost. Side-delivery
rake designs provide a rolling and wrapping action that reduces entanglement
with the stubble and increases entanglement among plants.
Raking loss normally has a small effect on the quality of the remaining
forage. Measured quality of forage lost during raking indicates that just a little
more leaf DM is lost than stem DM (Rotz and Abrams, 1988; Buckmaster,
1993). With similar loss of leaves and stems, the quality of the remaining
fora~e is affected little. The portion of leaves lost and the resulting effect on
quahty may be greater when the raked crop contains less than 300 gjkg
moisture (Buckmaster, 1993).
Baling and Chopping
Loss during hay baling typically varies between 2 and 5% of yield with a
little greater loss from large round balers than from small rectangular balers
(Whitney, 1966; Friesen, 1978; Koegel et aI., 1985; Rotz and Abrams, 1988;

CHANGES IN FORAGE QUALITY 838

Shinners et aI., 1992). Baler loss includes pickup and chamber losses. Pickup
loss is material dropped as the windrow is lifted into the machine. Chamber
loss is that disassociated and dropped during the formation of the bale in the
baling chamber.
Pickup loss varies between 1 and 3% of crop yield (Whitney, 1966;
Friesen, 1978; Koegel et aI., 1985; Rotz and Abrams, 1988; Shinners et aI.,
1992). This loss is primarily influenced by the structure of the swath or
windrow. Pickup loss expressed as a portion of yield decreases as the heaviness
or thickness of the swath increases. When forage is mowed and laid in a swath
by a mower-conditioner, the forage plants are generally oriented in the same
dIrection with little entanglement. Raked hay is rolled into a ti~hter windrow
where plants become more entangled. Loss is likely greater dunng the pickup
of less entangled swaths (McGechan, 1989b). Pickup loss can be very high
when the machine is pulled at a faster speed than the rotating speed of the
pickup device. The machine tends to overrun the swath causing excessive loss,
as high as 5% (Friesen, 1978).
Crop species and moisture content may affect pickup loss, but these
effects appear to be small. Although differences among grass and legume
species have not been documented, a comparison of Friesen's (1978) data to
other sources indicates that pickup loss may be slightly less in grass crops.
Data of Rotz and Abrams (1988) mdicate that the quality of material lost is
similar to that picked up, i.e., the loss has no effect on the leaf fraction or
nutrient concentration in remaining forage.
Chamber loss from a small rectangular baler varies between 1 and 3% of
yield for most conditions (Friesen, 1978; Koegel et aI., 1985; Rotz and Abrams,
1988; Shinners et aI., 1992). This loss is primarily shattered leaves. The
amount of chamber loss is largely influenced by crop moisture content with
greater loss in drier material. Shattered leaves are very dry, containing less
than 100 g/kg of moisture. When hay is baled at night, leaf moisture is higher
and similar to stem moisture. With a more uniform moisture content near 180
g/kg, chamber loss can be reduced by 50% (Friesen, 1978).
Chamber loss is higher in large round balers, and it varies considerably
among baler designs. For a variable chamber baler, the loss is about 40%
greater than that in a small rectangular baler (Koegel et aI., 1985). For a fixed
chamber baler, the loss can be three times that of a small rectangular baler.
Chamber loss is very sensitive to the feed rate of hay. Excessive loss occurs at
low feed rates because the hay must be rolled in the chamber many more
revolutions to form a bale and the rolling dislodges small particles.
Chamber loss consists of about 80% leaf material and 20% stem particles
(Buckmaster et aI., 1990). Since the loss is mostly high quality leaf material,
excessive chamber loss has more effect on the quality of remaining forage than
most other machine losses. By maintaining the chamber loss below 3%, the
effect on forage quality is relatively small.
To make silage, a fora~e harvester is used to chop the crop for easier
handling and better packing m a silo. Losses from a forage harvester include
pickup and drift losses. Pickup loss is similar to that for balers (Whitney,
1966). Drift loss occurs as the chopped material exits the spout of the
harvester and travels toward a trailing wagon or truck. As the machine moves
through the field, the spout can become momentarily misdirected causing the
flow of material to miss the trailin~ wagon or truck. A cross wind increases this
loss by diverting a portion of the fme particles away from the wagon. Machine
adjustment and operator error also contribute to more loss.
Drift loss typically varies between 1 and 3% of the harvested yield
(Whitney, 1966; Straub et aI., 1986). Dry forage is more susceptible to drift
loss than wet forage. The effect of crop moisture on drift has not been
measured, but Buckmaster et al. (1990) estlmate an exponential increase from

839 ROTZ&MUCK
 0.5 to 5% as moisture content decreases from 750 to 500 gjkg. The qualio/ of
the lost material is expected to be similar to that harvested, so the loss has httle
effect on the quality of the remaining forage.

HAY STORAGE LOSSES

Respiration causes a low level of DM and nutrient loss to continue during
hay stora~e (Wilkenson and Hall, 1966; Wood and Parker, 1971). Since hay
moisture IS below 400 gjkg, plant enzymatic activity is very low (Honig, 1979);
respiration is primarily from microorganisms (bacteria, fungi, and yeasts) on the
hay. Respiration transforms DM to heat and gases which leave the hay. In dry
hay stored under cover, little heating occurs and DM loss over 6 mo of storage
is about 5% (Collins et al., 1987; Rotz and Abrams, 1988; Rotz et al., 1991c).
Similar loss occurs with either large round bales or small bales stored in a shed
(Collins et al., 1987). Respiration reduces forage quality by removing some of
the most digestible nutrients for the animal.
Hay stored outside and unprotected experiences the same loss as hay
stored inside plus an additional loss from weathering of hay on the exposed
surface. The additional loss is often 10 to 15% DM for large round bales.
Factors which affect losses are different for the two types of storage.
Inside Storage
Dry Matter Loss
For protected hay, the major factor influencing loss during hay storage is
the moisture content of hay as it enters storage. Hay with less than 150 gjkg
moisture is relatively stable and little respiration occurs. In hay containing
more moisture, microbial respiration causes the hay to heat during the first 3
to 5 weeks of storage. The amount of heatin~ and the associated loss increase
with moisture content. Dry matter losses dunng the first month of storage vary
from 1% in hay of 150 gjkg moisture to 8% in 300 gjkg moisture hay (Nelson,
1966, 1968; Rotz et al., 1991c). For hay containing more than 300 gjkg
moisture, excessive loss and even spontaneous combustion can occur.
Greater heating occurs as the hay density in bales is increased (Nelson,
1966; Buckmaster et al., 1989a). The heat developed per unit of hay DM,
however, is nearly independent of density. The increased heating pnmarily
comes from packing more DM into the bale. The loss per unit DM tends to
increase with density, but this effect is small.
Heating during the first month of storage helps dry the hay. After the
first month, hay normally contains less than 180 gjkg moisture and thus is
relatively stable during the remaining storage period. Although a major portion
of the loss may occur in the first month, a small loss of about 0.5% DM per
month continues throughout storage (Rotz et al., 1991c). Consequently, storage
loss also is affected by the length of the storage period.
Crop species appears to have little influence on DM and nutrient losses
during inside storage. Studies conducted by Nelson (1966, 1972) show similar
moisture content and density effects on changes in alfalfa and native hays
during storage. The native hay was a mixture of grasses and clover. Greenhill
et al. (1961) found similar temperature and moisture effects on DM loss and
quality changes in white clover, alfalfa, and ryegrass hays.
Crop maturity may affect respiration and the resulting storage losses;
however, consistent effects have not been reported. Nelson (1968) reported
about 25% more heating and DM loss in bud-stage alfalfa compared to more
mature alfalfa with no difference between half-bloom and full-bloom. Other
studies found no difference in respiration rate across maturities of alfalfa and
ryegrass (Wilkinson and Hall, 1966; Wood and Parker, 1971).

CHANGES IN FORAGE QUALITY 840

 Quality Changes
Dry matter loss and heating during hay storage affect the concentration
of most nutrient constituents. Much of the lost DM is nonstructural
carbohydrate respired to carbon dioxide and water. The primary carbohydrates
lost are sugars, particularly sucrose (Greenhill et al., 1961). As carbohydrate
is depleted, proteins and fats also are used but at a slower rate. All of this loss
is highly digestible DM.
Reported changes in CP concentration during hay storage range from a
moderate loss (Davies and Warboys, 1978; Collins et al., 1987) to no change
(Nelson, 1966, 1968, 1972) and even a small gain (Rotz and Abrams, 1988). An
important factor affecting CP loss is the len~th of storage. Most protein data
are for hay monitored over short storage penods of less than 60 d. During the
first month when carbohydrate loss is greatest, protein loss is relatively small.
The change in CP concentration is often insignificant and a small increase can
occur with high carbohydrate loss in high-moisture hay (Rotz and Abrams,
1988; Rotz et aI., 1991c). As storage proceeds, the rate of carbohydrate loss
declines but protein loss continues. The differential rate of loss leads to a
decrease in CP concentration over 6 to 9 mo of storage (Davies and Warboys,
1978; Collins et aI., 1987).
The loss of CP presumably occurs due to slow volatilization of ammonia
contained in the hay or produced through microbial respiration. The loss is
nearly independent of hay moisture and DM loss. Data for untreated small hay
bales indicate a loss of about 2.5 g CP jkg DM during each month of storage
(Davies and Warboys, 1978; Collins et aI., 1987; Rotz and Abrams, 1988).
The loss of more soluble N components causes small increases in the
water insoluble N and acid detergent insoluble N (ADIN) concentrations (Rotz
and Abrams, 1988). In addition, the heating of hi~h-moisture hay causes the
formation of further ADIN through Maillard reactions. Maillard or browning
reactions are the chemical polymerization of sugars and other carbohydrates
with amino acids. The polymers produced are measured as increased ADF,
and ADIN increases as a fraction of total N. The formation of ADIN is
proportional to the number of degree-days the hay is above 35°C (Thomas et
aI., 1982). Therefore, ADIN increases with the loss of more soluble protein
and the heat damage of remaining protein (Buckmaster et aI., 1989a).
Fiber concentrations consistently show an increase during hay storage
(Greenhill et aI., 1961; Collins et aI., 1987; Rotz and Abrams, 1988; Buckmaster
et aI., 1989a). The increase primarily is due to the loss of non-fiber
constituents with little or no loss of fiber. Components not lost during storage
include ADF, NDF, crude fiber, lignin, and ash.
The digestibility of forage DM and many nutrient constituents decreases
during storage (Davies and Warboys, 1978; Collins et al., 1987; Rotz and
Abrams, 1988). The decrease occurs because the most digestible constituents
are used in respiration. Heating also reduces the digestibility of CP through
the formation of ADIN (Thomas et aI., 1982). The digestibility of fiber changes
less than that of many other plant components (Nelson, 1968). With the loss
of readily available carbohydrates and a decrease in DM digestibility, energy
content of the forage also is reduced.
Quality changes are relatively small in hay of less than 200 gjkg moisture.
Typical changes in alfalfa hay during 6 mo of storage are a 10 to 20 gjkg
decrease in digestible DM, no change in CP and ADIN concentrations, and a
10 gjkg increase in NDF concentration (Rotz and Abrams, 1988). Only in
higher moisture hay do substantial changes occur that affect the diet and
performance of animals consuming the forage. For example, in hay of 250 gjkg
moisture, a DM loss of 8% causes about a 30 gjkg decrease in dIgestible DM
content with a 35 gjkg increase in NDF content.

841 ROTZ & MUCK

 Outside Storage
Losses and quality changes in large round bales stored outside vary widely.
Reported DM losses range from 3 to 40% (Rider et al., 1979; Verma and
Nelson, 1983; Belyea et al., 1985; Collins et al., 1987; Brasche and Russell,
1988; Huhnke, 1988, 1990; Harrigan and Rotz, 1992). Many factors affect the
loss, and most of these have not been examined in a common experiment.
Of the factors affecting loss during outside storage, weather, length of
storage, and storage method have the greatest impact. An interaction occurs
among these factors in that the storage method has less effect during short
storage periods and/or in dry climates. The loss is again primarily caused by
microorganisms in the hay, and the biological activity is greatest when the hay
is moist and warm. Loss is less in hay stored over winter periods in northern
climates or in hay stored in more arid climates where the hay remains relatively
dry. For a given set of conditions, the loss appears to be nearly proportional
to the length of storage. Reported losses in alfalfa hay range from 0.5 to 1.5%
DM/mo of storage in drier climates (Rider et al., 1979; Huhnke, 1988) and 1.0
to 3.0% DM/mo in wetter climates (Verma and Nelson, 1983; Belyea et al.,
1985; Collins et al., 1987; Brasche and Russell, 1988; Harrigan and Rotz, 1992).
Storage method affects loss by providing different levels of protection
from the environment. Major options for outside storage of round bales
include the placement of bales on the soil without covers, elevated off the soil
without covers, and elevated with covers. When stored outside, the center of
bales is preserved similar to hay stored in a shed (Harrigan and Rotz, 1992).
Much greater loss occurs in the outer 10 to 20 cm of the bale where hay is
exposed to the environment. When set on damp soil, high moisture levels
develop in the bottom of the bale causing considerable deterioration and loss
in that portion of the bale. Compared to elevated bales, setting bales on soil
causes an average additional loss of about 3% DM (Verma and Nelson, 1983;
Collins et al., 1987; Huhnke, 1988, 1990; Russell et al., 1990).
When rain occurs on an exposed bale, some of the rain is absorbed in the
outer layer of the bale. The moisture content in this portion of the bale is
increased to between 250 and 400 g/kg (Russell and Buxton, 1985; Huhnke,
1988; Harrigan and Rotz, 1992). The moisture content inside the bale also may
increase a small amount. The increased moisture leads to greater microbial
activity and loss. A plastic wrap around the bale circumference greatly reduces
the moisture accumulation, particularly in bales elevated from the soil. A
plastic wrap on an elevated bale reduces storage loss to about 35% of the loss
III an uncovered bale (Rider et al., 1979; Verma and Nelson, 1983; Belyea et
al., 1985; Collins et al., 1987; Brasche and Russell, 1988; Huhnke, 1988, 1990;
Harrigan and Rotz, 1992).
Forage species may have some effect on loss. When stored under the
same conditions, bermudagrass and smooth bromegrass hay in unprotected
bales experienced about 30% less loss than that in alfalfa bales (Rider et al.,
1979; Brasche and Russell, 1988). The apparent reason for the difference is
that bales of grass hay shed rain more easily. This is supported by lower
moisture contents in grass hay following storage. Since grass normally contains
greater amounts of fiber, grass hay also may have less soluble matter to lose.
When round bales contain more than 180 g/kg of moisture, storage loss
is increased (Verma et al., 1986). The outer bale surface dries to the moisture
conditions of the environment with little additional loss, but the interior of the
bale goes through a heating process with the associated loss described above
for bales stored inside a shed. Other factors such as bale density, bale orient-
ation, bale binding material, and the desi~n of the baler may at times affect loss
(Huhnke, 1990; Russell et al., 1990). WIth the limited available data, general

CHANGES IN FORAGE QUALITY 842

 conclusions across conditions cannot be drawn. These effects appear to be
small with good harvest and storage practices.
Quality chan&es in lar~e round bales are variable, but reported data
indicate some relatively consIstent trends. The DM lost is once again the most
digestible portion of hay resulting in a decrease in digestible DM content
(Russell and Buxton, 1985; Brasche and Russell, 1988; Huhnke, 1990; Russell
et al., 1990). Fiber concentrations generally increase. In some studies, the
increased fiber content is small, indicating that some of the lost DM is fiber
(Verma and Nelson, 1983; Harrigan and Rotz, 1992). In other studies, the
increased fiber content, particularly NDF, is very high, indicating no loss of
fiber (Collins et al., 1987; Russell et al., 1990). Changes in lignin content show
similar variations (Russell and Buxton, 1985; Brasche and Russell, 1988; Russell
et al., 1990).
Crude protein changes are even more variable than those for fiber.
Reported values range from a substantial increase to a substantial decrease in
concentration with an average of little change in concentration (Rider et al.,
1979; Verma and Nelson, 1983; Russell and Buxton, 1985; Collins et al., 1987;
Huhnke, 1990; Russell et al., 1990; Harrigan and Rotz, 1992). ADIP
concentration generally increases with the heating of hay and the loss of more
soluble protein. More must be learned about the factors that affect forage
quality III order to develop more quantitative understanding of how these
factors affect hay quality in bales stored outside.

SIW STORAGE LOSSES

In contrast to hay, a moist forage is preserved by ensiling in an anaerobic
environment at a low pH. Creating a silo environment without oxygen is
essential for stopping plant respiration, preventing aerobic microbial growth,
and stimulating the growth of lactic acid bacteria. The lactic acid bacteria
ferment sugars producin~ principally lactic and acetic acids which lower silage
pH. A low pH is essentIal for inhibiting plant enzyme activity and preventing
the growth of undesirable anaerobic microorganisms.
Ensiling Phases and Processes
The effect of a particular plant, microbial, or chemical process on silage
quality depends on the nature of the process, ensiling time, and environmental
conditions. The silo storage period can roughly be divided into four major
phases: pre-seal, active fermentation, stable phase, and feedout. Each phase
IS dominated by several major processes (Figure 3).
The pre-seal phase represents the time between the start of filling and the
sealing of the silo. Oxygen is present, and the dominant process affecting
forage quality is plant respiration. In the living plant, a portion of the energy
liberated by respIration is captured in adenosine triphosphate. Biosynthesis is
assumed to be lImited in the harvested fora~e so most of the energy is released
as heat (McDonald, 1981). In the silo, thIS heat is not readily dissipated so
forage temperature is raised. Thus, respiration in the pre-seal phase not only
causes silage DM loss, but the rates of many other ensiling processes are
affected by the temperature rise. If heating from plant respiration is sufficient
to raise temperatures above 35°C, Maillard or browning reactions may become
significant. In these reactions, principally amino acids and sugars are
polymerized, increasing ADIN content, and releasing heat.
Other plant enzymes also are active in this period. Enzymes released by
cell rupture during chopping break down protein to peptides and free amino
acids, and carbohydrates to sugars. Ruptured plant cells provide substrate for
843 ROTZ&MUCK
 Active Stable
Pre-seal fermentation phase Feedout

Plant respiration --r-,

Proteolysis i..>
Enzymatic hydrolysis
of carbohydrates
C>
Forage cell lysis ----!>
Yeasts
Molds -[>
Acetic acid bacteria -t.:> i..>
Bacilli C> i..>
Lactic acid bacteria
Clostridia ~
v
Maillard reactions -t> r----------t->
Acid hydrolysis of
hemicelluloses

Figure 3. Major phases during ensiling when various plant, microbial, and
chemical processes are most active.

aerobic microorganisms and populations of yeasts, molds, and aerobic bacteria

increase. Their activity usually causes little change in forage quality at this
stage of ensiling unless populations of approximately 108 colony fornnng units
(CFU) of yeasts or bacteria/g forage and/or 106 CFU of mold/g fora~e are
reached (Pitt et aI., 1991). Similarly, facultative anaerobes such as lactIc acid
bacteria and enterobacteria will grow.
Once the silo is sealed, plant respiration removes the remaining oxygen
in a matter of hours (Pitt et aI., 1985). After anaerobic conditions prevail, the
active fermentation phase occurs. Within several hours, plant cells besin to
lyse releasing more cell contents (Pitt et aI., 1985). In wet crops, this SIgnals
the beginning of effluent loss (including highly digestible soluble carbohydrates
and N fractions contained therein) from the silo. With cell lysis, proteolysis
and enzymatic breakdown of carbohydrates occur at their highest rates. These
rates decline with time such that little activity occurs after one or two weeks.
Plant cell lysis also provides substrate for anaerobic microbial growth.
Enterobacteria and lactic acid bacteria normally dominate all of the other
microorganisms within the first 1 to 3 d after sealing. However, once pH has
dropped to 5, the enterobacteria decline rapidly, leaving the lactic acid bacteria
the principal microorganisms in the silage (Muck, 1991). Both groups ferment
sugars primarily. The enterobacteria largely produce acetic acid whereas the
lactic acid bacteria mainly produce lactic aCId. Other products from these two
~roups include ethanol, 2,3-butanediol, succinic acid, formic acid, and mannitol
(McDonald, 1981). The lactic acid bacteria grow actively for 1 to 4 wk
lowering crop pH usually to between 3.8 and 5.0, dependent upon crop
moisture, buffering capacity, and sugar content. Once pH drops sufficiently to
inhibit microbial growth or the substrate is exhausted, lactic acid bacteria
become inactive and their population slowly declines (Muck, 1991).
The third principal anaerobic bacteria, clostridia, have a much more
detrimental effect on silage quality if silage pH is not sufficiently low. These
strict anaerobes ferment sugars, lactic acid, and amino acids producing butyric

CHANGES IN FORAGE QUALITY 844

 acid and amines. These fermentations result in significant DM loss, and the
fermentationfroducts reduce the palatability of the silage (McDonald, 1981).
Many 0 the yeasts present when the silo is sealed are solely aerobIC
microorganisms; their populations decline rapidly once oxygen is not present.
However, fermentative yeasts are normally present and may increase in
numbers sufficient to affect silage quality (Muck et al., 1992). These yeasts
typically ferment sugars to ethanol, causin~ DM loss (McDonald, 1981). Other
aerobic microorganisms rarely have a SIgnificant effect during this period.
Acetic acid bacteria and mold populations decline rapidly in the absence of
oxy~en (Muck et al., 1992). Some bacilli may ferment sugars, forming products
simIlar to the enterobacteria (McDonald, 1981).
Maillard reactions may continue during the active fermentation phase if
the temperature attained in pre-seal was sufficiently high. Once pH is low,
chemical or acid hydrolysis of hemicelluloses may occur. Rates of acid
hydrolysis are low (Dewar et al., 1963) and will continue throughout storage.
After anaerobic microbial activity has essentially ceased because of low
pH or lack of substrate, the stable phase of storage be~ins. Even if silos are
well sealed, slow diffusion of air occurs throu~h the plastlc or concrete. During
active fermentation, the oxygen may be utihzed by plant respiration. As the
plant enzymes are inactivated, oxygen is principally used by aerobic and
facultative anaerobic microorganisms. This microbial growth results in the loss
of the most digestible components of the crop. Growth of certain molds can
produce mycotoxins harmful to ruminants (Woolford, 1990). As acids are
utilized by aerobic microorganisms, silage pH increases which may allow the
development of listeria, an animal and human pathogen (Woolford, 1990).
Growth of lactic acid bacteria in the presence of oxygen can result in the
conversion of lactic acid to acetic acid (Condon, 1987).
When the silo is opened, the final phase of feedout occurs. Much more
oxygen is present at the open face, and this oxygen penetrates into the silage
mass. With greater oxygen availability, aerobic microbial growth can increase
substantially, causing heating as well as the DM loss and health hazards
described previously. The initial heating is generally caused by yeasts or acetic
acid bacteria (Courtin and Spoelstra, 1990; Woolford, 1990). Bacilli and molds
develop later if the forage is not readily consumed (Muck and Pitt, 1992).
Factors Affecting the Ensiling Processes
Plant Respiration
A major cause of silage DM loss is plant and microbial respiration. Plant
respiration effects on forage quality in the field were discussed earlier. Rates
and factors affecting respiration in the silo are similar at least during the pre-
seal ehase (Pitt et aI., 1985). Potential activity probably declines with time in
the sIlo but this has not been measured. Both proteolysIs and starch hydrolysis
activity last for approximately a week under anaerobic conditions (McKersie
and Buchanan-SIDlth, 1982; Muck, 1987, 1990). Proteolytic activity has been
measured after 21 d of ensiling, even though increases in soluble NPN in the
silage ceased after approximately 3 d (McKersie and Buchanan-Smith, 1982).
Consequently, reduced rates of plant respiration may occur even after active
fermentation has ceased.

Proteolysis
Proteolysis in grass and legume silages may convert most of the true
protein to soluble NPN (primarily peptides, free amino acids, and ammonia)
via the activity of plant proteases. The N in wilted forages entering the silo
normally consists of 100 to 350 gjkg soluble NPN (Kemble and MacPherson,
1954; Muck, 1987). At feedout, soluble NPN is 250 to 850 gjkg of total N

845 ROTZ & MUCK

 (Kemble, 1956; Muck, 1987; Albrecht and Muck, 1991). The level of
proteolytic activity depends on forage species, pH, time, temperature, and
moisture content.
Proteolysis varies among different legume silages. Among alfalfa, red
clover, and birdsfoot trefoil, activity is highest in alfalfa (McKersie, 1985).
Albrecht and Muck (1991) reported that proteolysis is inversely correlated with
tannic acid content. Soluble NPN averaged 600 g/kg of total N in non-tannin-
containing legume silages compared to 300 g/kg of total N in legumes with 30
g tannic acid equivalents/kg DM. Of the three non-tannin-containing legumes
studied, soluble NPN was 100 to 200 g/kg higher in alfalfa than in red clover
or cicer milkvetch (Astragalus deer L.).
Overall proteolytic activity in legumes has an optimum at approximately
pH 6.0 (Brady, 1961; McKersie, 1985), although the pH optima of specific
proteolytic enzymes vary considerably (McKersie, 1981). Activity at eH 4.0 is
reported to be 15 to 35% of optimum. Therefore, reduction of sIlage pH
reduces but does not eliminate proteolytic activity. As indicated earher,
proteolytic activity decreases with time with little activity observed after 1 wk.
This is reported in both alfalfa (McKersie 'and Buchanan-Smith, 1982; Muck,
1987) and ryegrass (Kemble, 1956). As a result, a rapid droy in pH (preferably
the first day) is necessary to substantially alter the level 0 soluble NPN in a
silage.
Increasing !emperatures between 10 and 40°C increases proteolysis rates
exponentially (McKersie, 1981). Temperature has less effect on the amount of
proteolysis during ensiling, however, because a temperature increase also
mcreases bacterial fermentation and the subsequent rate of pH decline.
Increasing temperature from 15 to 35°C increases soluble NPN in alfalfa silage
by only 100 g/kg of total N (Muck and Dickerson, 1988).
The mOIsture content of the crop affects proteolysis. Initial proteolysis
rates in alfalfa are correlated linearly with moisture content with a rate of 0.9
mg N/g DM/h at 800 g/kg moisture and declining to 0 at 240 g/kg moisture
content (Muck, 1987). The amount of proteolysis in alfalfa silage is not as
strongly correlated with moisture content because variation in the pH time-
course in wetter silages (> 500 g/kg moisture) often offsets moisture effects on
proteolysis.
Enzymatic and Acid Hydrolysis
Plant enzymes convert nonstructural carbohydrates to simple sugars.
These sugars then are available for fermentation by lactic acid bacteria.
Fructosans, the main storage carbohydrate in grasses, are largely hydrolyzed to
fructose during ensiling (Henderson et aI., 1972). Starch, the storage
carbohydrate in legumes, is hydrolyzed presumably by a -amylases to maltose
(Doehlert and Duke, 1983) during ensihng so that starch content is normally
only 10 to 20 g/kg DM at feedout (Muck, 1990).
The activity of the amylases is affected by pH, starch concentration,
glucose, time, and temperature. Optimum actiVIty of alfalfa leaf amylase is
reported to be at pH 7.25 with the activity at pH 4.0 approximately 10% of
optimal activity (Muck, 1990). In alfalfa ensilin~ experiments, starch hydrolysis
rates were found to be linearly correlated wIth starch concentration (little
activity below 12 mg/g DM; 1.6 mg/g DM/h at 70 mg/g DM) and inhibited by
the addition of glucose at 50 to 100 mg/g DM (Muck, 1990). Hydrolysis rates
declined linearly with time, dropping by approximately 50% over 4 d (Muck,
1990). Moisture content had no effect on rates except below 400 g/kg moisture
(Muck, 1990).
Both enzymatic and acid hydrolysis of structural carbohydrates, primarily
the hemicellulosic fraction, probably occurs in both grasses and legumes. In
both forage types, NDF reductions during storage of 10 to 30 g/kg DM were

CHANGES IN FORAGE QUALITY 846

 observed (Jones et aI., 1992). Hemicellulases have been isolated from Italian
rye~rass, perennial ryegrass, and orchardgrass (Dewar et al., 1963). Their
actIvity declined with time such that little or no activity was observed after 7
d of incubation, but activity was enhanced at neutral pH and higher
temperatures (Dewar et aI., 1963). Acid hydrolysis also was measured by
Dewar et al. (1963). Rates were low (10.5 g sugar/g hemicellulose/h),
proportional to the hydrogen ion concentration, and they increased linearly with
temperature. This study, and that of Morrison (1979), suggest that enzymatic
hydrolysis of hemicelluloses is more important in grasses than acid hydrolysis.
In alfalfa, the opposite may be true. Morrison (1989) found a small
reduction in the hemicellulosic fraction during ensiling and a greater loss of
cellulose. Jones et aI. (1992) found little evidence of cellulose degradation, but
loss of hemicellulosic sugars was closely related to pH decline. Pitt et aI.
(1985) found that hemicellulase activity in alfalfa needed to be 10% of that in
grasses for accuracy of predictions. Therefore, acid hydrolysis of hemicellulose
appears to be more significant in alfalfa than enzymatic hydrolysis. Because of
the discrepancy between Morrison (1989) and Jones et aI. (1992), losses of
cellulose during ensiling are uncertain.

Emuent
As the plant cells lyse during silo storage, effluent is produced. The
amount of effluent that leaves the silo is dependent primarily on the moisture
content of the crop and the type of silo. Most studies of effluent have been on
unwilted or lightly wilted grasses in bunker or clamp silos. The moisture
content at which no effluent leaves these silos ranges between 670 and 710 g/kg
(McDonald, 1981). Based on a literature review, Waldo (1977) indicated that
no effluent should be released from corn silage of 700 g/kg moisture in tower
silos of less than 12 m. The moisture content should be reduced by 10 g/kg for
every 3 m of additional height to prevent effluent loss. Thus, higher pressures
and/or densities increase effluent production at a given moisture content.
Effluent volumes tend to increase quadratically WIth increasing moisture
content. At 800 g/kg moisture, effluent production of 70 to 220 L/Mg from
grass may be expected (McDonald, 1981).
Effluent production is highest during the first week of ensiling and
declines thereafter (Mayne and Gordon, 1986). The DM content of effluent
varies inversely with flow rate. Effluent collected from Italian ryegrass silage
with 810 g/kg moisture after three days of ensiling contained 50 g DM/kg and
increased to 90 g DM/kg after 63 d (McDonald, 1981). Mayne and Gordon
(1986) found 80 to 100 g DM/kg in effluent from grass silage with 780 g/kg
moisture. These results suggest that a silage containing 800 g/kg moisture
would lose approximately 40 g DM/kg of silage DM from effluent, based on
100 L of effluent/Mg of silage containing 80 g DM/kg of effluent.
Effluent contains many soluble compounds: sugars, fermentation products,
soluble protein and NPN fractions, and ash. In a study reported by McDonald
(1981), the DM of ryegrass effluent after 3 d of ensiling consisted of 220 g/kg
CP, 250 g/kg sugar, and 300 g/kg ash. By 63 d, the CP content increased to
310 g/kg DM whereas sugar and ash declined to 40 and 210 g/kg DM,
respectively. Much of the change over the 60 d was accounted for by increases
in fermentation products. Effluent is high in minerals. Potassium, calcium,
phosphorus, and magnesium averaged 86,31,8, and 6 g/kg DM, respectively,
III grass silage effluent (Purves and McDonald, 1963).

Aerobic Microorganisms
The most significant effect of aerobic microorganisms on silage quality is
respiration. The substrate for microbial respiration is dependent on the
organism. Yeasts and acetic acid bacteria typically consume only soluble

847 ROTZ&MUCK
 compounds such as sugars and fermentation products (Courtin and Spoelstra,
1990). Bacilli use proteins and some polysaccharides. Molds degrade the
widest range of substrates including the structural carbohydrates and lignin
(McDonald, 1981).
Aerobic microbial respiration, although not desirable, produces fewer
immediate losses than plant respiration in the harvested crop. Aerobic
microbes respire approximately half of the sugars consumed. The remainder
is converted into cell mass which may be lost after the microorganisms die
(Muck et al., 1991). Of the respired sugars, only 58% of the released energy
ends up as heat in contrast to plant respiration in the harvested forage where
virtually all energy is released as heat (McDonald, 1981).
Growth rates of these microorganisms are dependent on species, pH,
temperature, substrate, and fermentation products. In general, yeasts and
molds are less sensitive to low pH (Le., 4.0) than bacteria (Muck et al., 1991).
Molds are much slower growing than yeasts (Muck et al., 1991) and have much
lower populations during silo storage due to their greater sensitivity to the
absence of oxygen (Muck et al., 1992). Bacilli are apparently more sensitive
to either low pH and/or fermentation products (Muck and PItt, 1992) so that
they seldom initiate aerobic deterioratIOn.
Fermentation products are known to inhibit various aerobic
microorganisms. Fungi are sensitive to the undissociated concentrations of
volatile fatty acids (VFA), particularly the longer chain VFA (Woolford, 1975).
In well-fermented silages, yeast and mold growth rates are reduced in
proportion to the undissociated concentration of acetic acid (Muck et al., 1991).
Clostridial silages contain significant levels of butyric acid and thus are known
to be very stable aerobically (Woolford, 1990). Lactic acid is inhibitory to
yeasts and even more so to acetic acid bacteria (Courtin and Spoelstra, 1990).
The apparent succession of aerobic microorganisms in sila~e exposed to
air appears to be as follows (Muck and Pitt, 1992). Yeasts and/or acetic acid
bacteria use sugars and fermentation products as substrates, potentially raise
silage temperatures to 40 to 45°C, and cease to be a factor when those
substrates are used up. Use of the fermentation products raises silage pH.
Bacilli then become the dominant microorganisms, consuming more complex
substrates and maintaining temperatures in the 40's. Finally, molds complete
the deterioration process and they may raise temperatures above SO°c.
Aerobic microorganisms may cause other changes in forage quality.
Molds such as Aspergillus and other species can produce mycotoxins during
storage (Woolford, 1990). In higher pH silages (above 5.0 generally), Listeria
monocytogenes, a human and animal pathogen, may multiply (Woolford, 1990).

Anaerobic Bacteria
Utilization of su~ars by lactic acid bacteria produces the least adverse
change in forage qUalIty of any of the processes. In contrast to the aerobic
organisms, only 10 to 20% of the sugars assimilated go into cell mass with the
rest being fermented for energy (Pitt et al., 1985). Table 1 shows various
fermentation pathways and the subsequent DM and energy losses. Lactic acid
bacteria cause little loss of gross energy. Homofermentative pathways (i.e.,
those producing only lactic acid) result in no DM loss whereas hetero-
fermentative pathways produce up to 24% DM loss. In most natural lactic acid
bacterial fermentations with a mix of hetero- and homofermentative pathways,
energy density per unit silage DM increases slightly.
Dry matter loss depends on the strain of lactic acid bacteria and the
environmental conditions. Most natural lactic acid bacteria in silage are either
heterofermentative or facultative homofermenters (Sneath et al., 1986). The
facultative strains normally are homofermentative but become heterofermenters

CHANGES IN FORAGE QUALITY 848

 Table 1. Selected silage fermentation pathways and accompanying DM and
energy losses (McDonald, 1981).

Pathway Dry matter Energy

loss (%) loss (%)

Homofermentative lactic acid bacteria

1 ~lucose .... 2 lactate 0.0 0.7
1 mctose .... 2 lactate 0.0 0.7

Heterofermentative lactic acid bacteria

1 ~lucose.... 1 lactate + 1 ethanol + 1 CO 2 24.0 1.7
3 mctose.... 1 lactate + 3 acetate +
2 mannitol + 1 CO2 4.8 1.0

Clostridia
1 glucose.... 1 butyrate + 2 CO2 + 2 H2 51.1 20.9
2 lactate.... 1 butyrate + 2 CO2 + 2 H2 51.1 18.4

Yeasts
1 glucose .... 2 ethanol + 2 CO2 48.9 0.2

when sugar content and growth rate are low (Christensen et aI., 1958).
However, homofermentation in Streptococcus bovis at slow growth rates has
been observed when pH is low (Russell and Hino, 1985) and, similarly,
fermentation in silage has been observed to become more homofermentative
with decreasing pH (Pitt et al., 1985).
The amount of sugars converted by lactic acid bacteria to fermentation
products is dependent on the sugar content, moisture level, and buffering
capacity of the crop. Lactic acid bacteria stop growing from either the lack of
substrate or low pH. This is illustrated by alfalfa silage made at several
moisture levels with and without glucose addition (Figure 4). At low moisture
levels, added sugar had no effect on final pH, indicating that low pH stopped
bacterial growth. Under wetter conditions, the glucose addition lowered silage
pH, suggesting that fermentation in the untreated silages was limited by the
sugar content. The pH at which bacterial growth ceased also decreased with
increasing moisture content (Muck, 1990).
The buffering capacity of the crop (equivalents of acid needed to drop pH
from 6.0 to 4.0) determines approximately the amount of fermentation acids
necessary to reach a given pH. This has been demonstrated in alfalfa where
silage pH was negatively correlated with the sugar-to-buffering capacity ratio
(Melvin, 1965). In general, the buffering capacities are lowest for corn (150 to
250 meqjkg DM), intermediate for grasses (250 to 500), and highest for
legumes (400 to 600) (McDonald, 1981). This suggests that the highest levels
of fermentation products would be seen in alfalfa and other legumes and the
lowest in corn. However, low sugar contents in alfalfa relative to other species
means that grass silages often have the highest levels of fermentation products,
occasionally reaching 200 gjkg DM (e.g., Morgan et ai., 1980).
Most lactic acid bacteria ferment only sugars and some organic acids.
However, some strains ferment serine and arginine, producing ammonia
(Ohshima and McDonald, 1978). More importantly, many lactic acid bacteria
multiply in the presence of oxygen. They do not respire oxygen as aerobic
bacteria, but various silage strains can use oxygen, often converting lactic acid

849 ROTZ&MUCK
 e.or--------------,
to acetic (Condon, 1987). This is
one cause of high acetic acid content
:5.5 .. ,
silages.
Poor fermentation is normally == , .
associated with the development of Q, 5.0
...
clostridia. Clostridia can be divided ~
into three groups: those that produce ~ •. 5
, .
butyric acid by fermenting sugars and fi.i "-------------
lactic acid; those that ferment amino 4.0
Glucose added -Q""
acids producing various acids, 'a

ammonia, and amines; and those that

ferment both sugars and amino acids 3·g0~0~-~:7:-0~-:5~O-:-O~-:8...l.00-~?.J..OO~---.J800
(McDonald, 1981). As shown in Moisture content (g/kg)
Table 1, clostridial fermentation of
glucose and lactic acid results in Figure 4. Alfalfa silage pH with or
butyric acid and substantial DM and without the addition of glucose at the
energy losses. Amino acids are rate of 70 to 120 g/kg of DM (Muck,
fermented via several mechanisms: 1990).
deamination, decarboxylation, and
oxidation/reduction. Deamination
increases ammonia content but results in no DM loss. Decarboxylation causes
DM loss from carbon dioxide release and produces amines which are
detrimental to forage intake by animals. Oxidation/reduction reactions
produce or~anic acids, ammonia, and carbon dioxide (McDonald, 1981).
Clostndia prefer warmer silage temperatures with most silage species
havin~ optima near 37°C (McDonald, 1981). The principal means of inhIbiting
clostndial growth is through rapid lowering of sdage pH. The pH at which
clostridial growth is completely mhibited is dependent on the water activity of
the silage, which is a function of the crop species and its moisture content. The
wetter the crop, the lower the pH necessary to prevent clostridial growth. The
water activity of legumes at a given moisture level is lower than that of grasses
so a lower pH is required to prevent clostridial growth in grasses. From a
practical standpoint, clostridial growth is normally prevented by ensiling the
crop with less than 700 ~/kg of moisture. Exceptions to this general
observation can occur if baSIC compounds such as ammonia or urea are added
at ensiling, raising silage pH.
Enterobacteria normally do not have a large effect on silage quality.
These facultative anaerobes compete with the lactic acid bacteria for sugars
early in ensiling, producing primarily acetic acid. Their fermentation pathways
are similar to the heterofermentative lactic acid bacteria, producing DM loss
but little loss of gross energy. Unlike the lactic acid bacteria, enterobacteria
have pH optima around 7.0, and their populations generally decline rapidly
below pH 5.0 (McDonald, 1981). Spoelstra (1987) found that some strains
reduce nitrate to nitrite and nitric oxide, a form of silo gas. Consequently,
some loss of CP could result from their activity.

Maillard Reactions
As in the storage of high-moisture hay, Maillard reactions can increase
silage ADF and ADIN. Hemicelluloses can be complexed in these reactions
(Fi~re 5) so that the difference between NDF and ADF declines. As
indIcated in Fi~re 5, Maillard reaction rates increase exponentially with
temperature. little activity is expected at temperatures below 35°C. Since the
Maillard reaction releases heat, the process can further increase silage
temperature and cause more extensive rates of browning.
Moisture content is another factor affecting browning. For normal
ensiling moisture contents of 400 to 700 g/kg, moisture does not affect the

CHANGES IN FORAGE QUALITY 850

 600,------------,220.-..
Maill~rd reaction significantly.--.. ~
(Goenng et aI., 1973). However, z 600 ~
moisture affects the heat capacity of a; 200 .IIi
the silage. Based on a herbage heat ~ .00 .J;.I_~~,~~_e.l_l,U_,l_O_,s_e,S ---.
capacity of 1.89 kJ/kg DM;oC +> ___ :;
(McDonald, 1981), the heat ~300 160 i
capacities of a silage at 700, 500, and . . . . .
;g
0
300 g/kg moisture are 11.67, 6.08, ~200
and 3.69 kJ/kg DM;oC, respectively. Z -,,_ 160 ~
Thus a given amount of browning ~ 100 ADI -,~'s
raises temperature more than three III
times as much at 300 g/kg moisture ~LO~4-LO~S""'0"""'--'60~7.LO"""'---:'6':-O~9-LO~10.L.O~1.J1h40:=
content than at 700 g/kg. As a
result, Maillard products are most Silage temperature (c)
common in drier silages. Figure 5. Effect of temperature on the
Temperatures high enough to cause extent of browning assayed as ADIN
spontaneous combustion can be and .the associated decline in
reache~. Silo fire~ are .a potential hemicell loses (Goering et a1. 1973).
hazard III crops ensIled WIth less than u ,
450 g/kg mOIsture.
Susceptibility to browning varies among forages. The cause of this
variation does not appear to be related to total N content, sl?ecies, or initial
ADIN (Goering et aI., 1973). Analyses of forage silages by vanous researchers
indicate that 18 to 40% of the silages had signs of overheating (Goering, 1976).
With the move toward ensiling wetter silages over the last 20 yr, fewer silages
with Maillard products are expected.

Overall Losses and Quality Changes in the Silo

The importance of the various processes in the silo relative to losses and
silage quality is difficult to assess due to the number of processes occurring
simultaneously, and their interactions. One means of integrating all factors and
assessing their importance is through computer modeling.
Most computer programs have dealt with only one or two phases of
ensiling. The earliest models principally focused on fermentation (Neal and
Thornley, 1983; Pitt et al., 1985; Leibensperger and Pitt, 1987; Meiering et aI.,
1988). These models looked primarily at the growth of lactic acid bacteria and
clostridia and at plant enzyme activity during active fermentation. Recently,
modeling activity has turned to aerobic losses during storage and feedout. Pitt
(1986) and McGeehan (1989a) studied losses due to air infiltration during
storage. Current efforts are on simulation of aerobic losses at feedout
(Parsons, 1991; Muck and Pitt, 1992; Pitt and Muck, 1993). These modeling
efforts have indicated that individual processes during silo storage can be
reasonably simulated.
A comprehensive model of silo storage was developed to simulate losses
from filling through feeding (Buckmaster et aI., 1989b). This silo model was
integrated in a simulation model of the dairy forage system (DAFOSYM) to
study losses as affected by farm operations and environmental conditions (Rotz
et aI., 1989). Losses during pre-seal, the stable phase, and feedout were
assumed to be solely by respiration. Oxygen movement into the silo was by
diffusion. Changes in DM, fiber content, and N fractions during active
fermentation were based on simulations of the Pitt et a1. (1985) fermentation
model. The comprehensive silo model reasonably eredicted whole silo DM
losses comp'ared to actual studies. The effects of sIlo size and type on DM
losses for sIlos emptied over 360 d are shown in Figure 6. A bottom-unloaded,
oxygen-limited silo provided the lowest losses whereas the top-unloaded,

851 ROTZ&MUCK
 16r------------------------.
concrete stave silo had approximately ,..,.
two percentage units greater losses ~ 16
over most capacities. Except at small !ll
silo capacities, the bunker silo had ~ H
the greatest DM losses. -
Recently, effluent production $ 12
was added to the model to compare ~
direct-cut ensiling of alfalfa with 10 S
normal practices of ensiling alfalfa
wilted to 650 g/kg moisture (Rotz et
al., 1993). Harvesting over 26 yr of
£'
6 Bottom-unloaded
.ea1d
weather conditions for East Lansing, 60'--~1.J...OO~-2....LO-0~3-'0-O~-40.LO-==5""-OO~--'600
Michigan was simulated. Both Silo capacity (ton DM)
harvest systems placed the alfalfa in
bunker silos and direct-cut alfalfa Figure 6. Effect of silo size and type on
was treated with formic acid to predicted DM loss for silos emptied
control fermentation. Distribution of over 360 d (Buckmaster et al., 1989).
the average silo losses are shown in
Table 2. Normal wilted silage
production had no effluent loss, and most losses were split between that caused
by oxygen infiltration during storage and aerobic losses at feedout. Ensiling of
dIrect-cut material increased overall losses and shifted losses more to events in
pre-seal and active fermentation.
Currentlx, there are no data to confirm the relative losses in each of the
phases of ensIling listed in Table 2. However, data show that failure to cover
bunker silos increased organic matter loss in the top 50 cm from 270 to 410
g/kg in a farm silo survey (Dickerson et al., 1990). Losses 2 m below the
surface were substantially less under both covered and uncovered silos.
Fermentation losses predicted by the model also appear reasonable. Bacterial
inoculants, which provide a homofermentative silage fermentation, have been
reported to improve DM recovery from silos by approximately 2.5 percentage
units (Muck and BoIsen, 1991). Because some reduction in aerobic losses also
may result from inoculant use, the fermentation losses in Table 2 are
reasonable. Overall, these results tend to corroborate the importance of
aerobic microbial respiration on silage DM losses suggested by the model of
Buckmaster et al. (1989b).

Table 2. Simulated average DM losses during storage of wilted

(650 g/kg moisture) and direct-cut silage predicted by
the comprehensive silo model of DAFOSYM (Rotz et
al., 1993).

DM lost (%)
Wilted Direct-cut

Effluent 0.0 4.7

Aerobic respiration during filling 0.8 1.3
Fermentation 0.7 1.5
Aerobic respiration during storage 5.0 4.7
Aerobic respiration during emptying 5.2 3.8
Total 12.1 17.7

CHANGES IN FORAGE QUALITY 852

 Respiration not only has a dominating effect on DM losses in the silo but
also on forage quality. Digestible carbohydrates are largely lost through
respiration, increasing the concentration of other forage components. For
example, little loss of N occurs in the silo unless effluent losses andjor nitrate
in the incoming crop are significant. Consequently, a 10 to 20 gjkg DM
increase in CP should occur dependent on the CP of the crop entering the silo
and the DM lost during silo storage. Changes in cell wall or NDF content,
however, also are dependent on the amount of enzymatic and acid hydrolysis
of structural carbohydrates during stora~e. In laboratory studies where DM
losses are minimized, NDF typically declInes 10 to 60 gjkg DM in grasses and
legumes (Spoelstra, 1990; Jones et aI., 1992). Thus, under real farm conditions,
NDF levels in silages may range from a 10 gjkg DM decline to a 40 gjkg DM
increase, dependent on the respiration loss relative to the amount of cell wall
hydrolysis. With good silo management, ADF changes little, so the
concentration of ADF increases 20 to 50 gjkg DM with the loss of other
carbohydrates. Carbohydrates lost are highly digestible, so DM digestibility and
total dIgestible nutrient (TDN) concentration decline 20 to 70 gjkg DM.

METHODS TO REDUCE LOSSES AND IMPROVE QUALITY

Many options are available to reduce the DM and nutrient losses that
occur dunng forage harvest and storage. These include equipment modifica-
tions to reduce loss, new equipment systems to speed field curing, and chemical
treatments to enhance field drying and preservation in storage. The primary
restraint to implementing new technologies that reduce losses or enhance
quality is economic feasibility. Most technologies increase the costs of forage
production. The improvement in yield and quality must have a value greater
than the added cost to encourage adortion by growers. Other constraints
include labor availability, environmenta impact, and public perception.
Equipment Systems
Reduction of losses and preservation of forage quality always have been
among the major considerations in the development and design of forage
equipment. Other considerations include cost of production, labor
requirements, power or fuel requirements, and operator safety. A design to
reduce or eliminate loss may be compromised at times to meet these other
equallx or more important considerations. Progress has been made over the
past fIfty years in designing machines to reduce losses. An example is the
development of the mower-conditioner. Much effort was devoted to balancing
the effectiveness in improving drying with the loss caused by the operation.
Further improvements are expected III the future.
Small improvements can be made to reduce the loss caused by various
machine treatments. Such improvements may be done by designing machines
for more gentle handling of forage or the capture and recycling of shattered
forage material. One of the greatest opportunities for improvement is the
reduction of chamber losses in large round balers. For most machine
operations, the associated forage loss is low so the opportunity for reducing loss
is small.
The largest harvest losses and quality changes occur during field curing,
primarily from rain damage. Thus, new systems which speed field curing have
the greatest potential for benefit. One technique under development is
maceration and mat drying of forage. Maceration is the most severe form of
mechanical conditioning. Plant stems are shredded, fully exposing the internal
moisture. Drying restraints imposed by the internal cell structure, the

853 ROTZ&MUCK
 epidermis, and the cuticle are removed. Macerated alfalfa can dry to a
moisture content suitable for baling in 4 to 6 h of favorable drying conditions
(Rotz et aI., 1990).
Macerated forage contains many fine particles that are very susceptible
to loss during field curing and harvest. To reduce the potential loss, the
shredded material can be pressed into a mat that is laid on the field surface for
rapid curing (Koegel et aI., 1988). The dried mat then can be picked up with
nunimalloss. Commercial equipment for maceration and mat drying is not yet
available, but research and development of the process is continuing. In order
to implement the process on the farm, new equipment and procedures are
required for fora~e harvesting, handling, and storage.
Fast drying IS a major advantage of the mat process. The drying rate of
alfalfa mats is two to three times greater than the rate of conventional swaths
so that hay can be made with one day of field curing. Maceration of forage
appears to greatly reduce respiration through very rapid drying and severe
rupturing of plant tissue (Rotz et aI., 1991b). Macerated fora~e is very
susceptible to rain damage with three times the loss of alfalfa conditIOned with
intermeshing rubber rolls (Rotz et aI., 1991b). However, with rapid drying, rain
damage should not occur often.
The mat process can potentially provide a substantial improvement in
forage quality. A long-term analysis indIcates a 10% improvement in harvested
yield through reduced field losses (Rotz et aI., 1990). The reduction of loss
provides a 20 to 30 ~/kg DM average decrease in NDF concentration and a
small improvement III harvested protein. In addition, shredding the forage
increases fiber digestibility up to 15%. The improvement in digestion allows
the animal to obtain more energy from the forage and perhaps to increase its
intake of forage. With this quality enhancement, macerated forage produces
a higher quality forage for hIgh producing dairy cows than obtained through
conventional forage systems.
Disadvantages of the mat system include high power requirements and
equipment costs. The long-term benefits of faster drying, reduced loss, and
improved forage quality appear to outweigh the added fuel and equipment costs
(Rotz et al., 1990). In hay production, the system can return up to $4 for each
dollar spent on increased equipment and fuel costs. The system appears less
economical in silage production, but additional benefit may be obtained
through reduced silo losses acquired by faster and greater packing in the silo.

Chemical Drying Aids

Chemicals can be used to speed the field curing of both legume and grass
forage crops. Chemical treatments improve drying by opening stomata,
desiccating the plant prior to mowing or by modifying the epicuticular waxes.
Several chemical treatments are known to enhance forage drying, but most are
not practical or economical for field use (Harris and Tulberg, 1980).
The first practical method for using chemicals to speed the field curing of
forage came with the use of aqueous potassium carbonate. The exact
mechanism by which this treatment improves drying is not known. The solution
is thou~ht to form a continuous film over the plant surface which extends down
the caVIties between the wax platelets. The film joins the liquid phase moisture
within the parenchyma, pectin, and cuticular membrane, and thus enables
moisture transfer in the h9,.uid phase through the wax layer by capillary forces
(Harris and Tulberg, 1980). When sprayed on the crop as it is mowed, the
solution can increase drying rate. This process of chemical conditioning is
effective on legume forage species but not on grass species.
Potassium carbonate is the most common active ingredient in chemical
drying agents. A 28 giL solution of potassium carbonate in water applied at

CHANGES IN FORAGE QUALITY 854

 an appropriate rate speeds drying under good dryin~ conditions. Replacing half
of the potassium carbonate wIth sodium carbonate IS less expensive, but equally
effective to potassium carbonate alone. The treatment is more effective as
more solution is applied, but the economically optimal amount is about 300
L/ha for yields under 3.5 Mg/ha and 470 L/ha for greater yields. Mixing more
chemical per unit of water does not improve drying performance nor
compensate for a decrease in solution application rate (Rotz and Davis, 1986).
Chemical conditioning can improve the drying of all cuttings of alfalfa, but
it is most effective on summer harvests (Rotz et aI., 1987). In the Midwest,
chemical conditioning provides about a 40% increase in the drying rate of first
cutting alfalfa with up to a 120% increase on second and third cuttings. Poor
drying conditions in the fall limit the increase to about 20%. The typical
reduction in field curing time for chemical conditioning compared to
mechanical conditioning is 5 daytime hours on first cutting and 7 to 8 h on later
cuttings.
Chemical conditioning of alfalfa can provide an avera~e increase in
harvested yield of over 10% in hay production (Rotz et aI., 1989). The process
appears to have little effect on respiration rate and the resultant losses (Rotz
et al., 1984; Rotz et aI., 1987). When rain occurs during field curing, the
treatment has little or no influence on rain-induced losses (Rotz et aI., 1987;
Rotz et al., 1991b). The primary benefit comes through the occasional
avoidance of rain damage through faster drying. The improvement in yield and
quality is of marginal economic benefit in hay production with little benefit in
sIlage production.

Hay Preservatives
In hay making systems, field losses can be reduced by baling hay at a
moisture content near 250 gjkg. Baling moist hay reduces baler chamber losses
providing a small (up to 2%) improvement in harvested yield and a small
Improvement in harvested quality. Raking and pickup losses also may be
reduced a small amount. Field curing time on the average is reduced about 1
d which reduces the potential for rain damage. With all of these factors
combined, harvested yIeld is increased an average of 7%. However, the moist
hay deteriorates rapidly in storage, offsetting the benefit of reduced field losses
unless treated to enhance preservation.
Materials used for the preservation of high-moisture hay include propionic
acid, organic acid mixtures (may include propionic, acetic, fumaric, citric,
benzoic, lactic, and formic acids), buffered acid mixtures, anhydrous ammonia,
and bacterial inoculants. Propionic acid (or an effective organic acid mixture)
normally reduces mold growth (Rotz et aI., 1991c). With application rates of
10 to 20 gjkg of hay weight, acid treatments can reduce the heating of high-
moisture hay (Knapp et al., 1976; Davies and Warboys, 1978; Rotz et aI.,
1991c). Some report similar heating in treated, damp hay- and dry « 180 gjkg
moisture) hay while others report a little more heatmg (Rotz et aI., 1991c).
Propionic acid treatment reduces storage loss in damp hay during the first
few months of storage, but the loss is higher than that in dry hay (Knapp et al.,
1976; Davies and Warboys, 1978; Rotz et aI., 1991c). Over 6 mo of storage,
reported losses are similar in treated and untreated hays at similar moisture
levels (Rotz et al., 1991c). Acid-treated hay maintains a higher moisture
content throughout storage. Apparently the more moist environment in the hay
maintains a little higher level of microbial activity even with the acid treatment.
Over a 6 mo storage period, the loss in acid-treated hay catches up, providing
little difference in losses and nutrient changes between treated and untreated
high-moisture hays.

855 ROTZ&MUCK
 Dry matter lost during storage is very di~estible (Buckmaster et al.,
1989a). When compared to hay at similar mOIsture levels, propionic acid
treatment provides smaller decreases in in vitro DM and cell wall digestibilities
and lower levels of ADF, NDF, and ADIN after short storage periods (Knapp
et aI., 1976; Rotz et aI., 1991c). Consistent improvement in hay quality is not
reported with propionic acid treatment of hay stored for more than 4 mo
(Davies and Warboys, 1978; Rotz et aI., 1991c). When compared to dry hay,
acid-treated damp hay is often higher in fiber content and less green in color
(Rotz et aI., 1991c).
Propionic and similar acids promote corrosion in balers and bale handling
equipment. To reduce corrosion, buffered acid products have largely replaced
the use of straight acids. The acid is blended with ammonia or other
compatible chemical to increase the pH of the treatment. The less volatile
buffered products may be as effective as propionic acid when equivalent
amounts of propionate are retained in the hay. One buffered product was
ineffective at application rates below 5 g/kg of hay (Rotz et al., 1991c).
Routine use of acid treatment on high moisture hay appears
uneconomical. A thorough analysis of various harvest strategies indicate that
the limited benefit received does not justify the cost of the treatment (Rotz et
aI., 1992). The treatment can only be economical when it is used to avoid
heavy rain damage of the crop. The economic benefit can be improved by
greatly increasing the effectiveness of the treatment over long storage periods
and/or reducing the treatment costs.
Anhydrous ammonia is perhaps the most effective hay preservative (Rotz
et al., 1986; Rotz et aI., 1992). Storage DM loss is reduced or eliminated in
hay of up to 350 g/kg moisture when wrapped in plastic and treated with
ammonia at 10 g/kg of hay weight or more. Ammonia treatment prevents
heating, and it may eliminate mold development while the hay is covered.
Ammonia treatment increases CP concentration by adding NPN. With less
storage loss, the increase in ADF and NDF which normally occurs during
storage is reduced. Increases in in vitro DM, hemicellulose, and cellulose
digestion and energy content are reported.
Although anhydrous ammonia provides the most effective and economical
preservation of forage, animal and human safety concerns deter its use.
Ammonia treatment of forage has caused toxicity to animals (Rotz et aI., 1986).
Toxicity most often occurs when ammonia is used on high quality hay at hi~her
than recommended application rates (greater than 30 g/kg of DM). DIrect
exposure to anhydrous ammonia can cause severe burns, blindness, and death.
Bacterial inoculants are sometimes applied to hay. Inoculation with a few
strains of Lactobacillus had no effect on mold, color, heating, DM loss, and
quality change in high-moisture hay (Rotz et aI., 1988). In another study, both
Lactobacillus and Bacillus inoculants improved hay appearance with little effect
on DM loss and quality compared to untreated hay of similar moisture (Tomes
et aI., 1990). Until a more tangible benefit is shown, the economic value of
these products cannot be addressed.

Silage Additives

A wide variety of additives is used in silage making. The principal

additives include bacterial inoculants, enzymes, and NPN. Each class of
products has different effects on silage quality.
The most common silage additives in the U.S. are bacterial inoculants
which supplement the natural lactic acid bacteria of the crop. Inoculant
bacteria have been selected from forages for fast growth rate and
homofermentativeness. When the inoculant bacteria dominate fermentation, the
resulting silage has less acetic acid and ethanol, more lactic acid, and a lower

CHANGES IN FORAGE QUALITY 856

 pH than expected from the unaided natural fermentation. This shift in
fermentation should improve DM recovery as indicated in Table 1. A survey
of inoculant studies found an average 2.5 percentage unit improvement in DM
recovery at feedout when the inoculant was successful (Muck and BoIsen,
1991). This should be due principally to the shift in fermentation products;
however, some reduction in aerobic respiration also may be represented. An
additional benefit from inoculant use is a small reduction in proteolysis,
particularly in the ammonia fraction (Muck and BoIsen, 1991).
Inoculant success is primarily related to the size of the natural lactic acid
bacterial population. Inoculants were successful in approximately two-thirds of
the reported studies on grasses and le~mes since 1985; these products have
been successful less than half the time In corn and sorghum silages (Muck and
Bolsen,1991). This corresponds well with natural lactic acid bacterial numbers,
which are typically higher on corn than on alfalfa or grass.
Most enzyme products consist of a mixture of cellulases, hemicellulases,
pectinases, and amylases and are targeted for reducing the fiber content of a
silage. A survey of published studies since 1985 found that these products
reduced ADF and NDF in 80% of the grass experiments but in only 40% of the
alfalfa trials (Muck and BoIsen, 1991). The failure on alfalfa was presumably
from differences in cell wall structure. Typical reductions in fiber contents
range from 10 to 50 g/k~ DM. These products also appear to improve DM
recovery; six of nine studIes reported improvements, averaging 6.6 percentage
units (Muck and BoIsen, 1991). The cause of reduced aerobic loss during silo
storage is uncertain. When effective, enzyme-treated silages consolidate more
than untreated silages, which should reduce porosity and subsequent oxy~en
movement into the silage mass. On the negatIve side, increased consolidatIOn
could cause additional effluent loss in wetter sila~es.
Both ammonia and urea are common addItives to corn silage. These
additives boost the CP content and make silages more aerobically stable by
killing aerobic microorganisms. With lower volatile losses, urea is more
efficient in increasing CP, whereas ammonia is more effective in improving
aerobic stability (Muck and BoIsen, 1991). These additives increase crop pH
at ensiling and thereby cause more fermentation and fermentation products,
particularly' acetic acid. Both compounds, but especially ammonia, improve
DM and fIber digestibility and reduce proteolysis. In spite of these benefits,
animal performance has been enhanced infrequently (Muck and BoIsen, 1991).
Also, DM recoveries with NPN additives have been reduced in 11 of 16 recent
trials, presumably due to increased fermentation losses by either lactic acid
bactena or clostridia (Muck and BoIsen, 1991).

SUMMARY
Forage harvest and stora~e as hay or silage are an important part of
animal agriculture in many regIOns of the world. Hay and silage are used to
feed animals during portions of the year when forage cannot be grown or
throughout the year to avoid the need for pasture systems. Substantial losses
of forage DM and nutrients occur during the harvest and storage processes. An
important aspect in the selection of forage systems is their impact on the
resulting yield and quality.
Many forms of loss occur. These can be categorized as field losses
(respiration, rain damage, and mechanical damage) and storage losses. Plant
and microbial respiration remove carbohydrates from the forage causing an
increase in protein and fiber concentrations. Rain damage removes a portion
of the leaves, leaches soluble nutrients, and induces further plant and nucrobial
respiration. Mechanical losses normally cause greater loss of leaves than stems.

857 ROTZ&MUCK
 Because leaves contain higher concentrations of most nutrients important to the
animal, a decrease in the leaf to stem ratio reduces the overall quality of
fora~es. Microbial respiration continues to remove the most digestible forage
nutnents during storage. Storage loss is primarily nonstructural carbohydrates
but some loss of protein and change in protein solubility occur.

Table 3. Typical DM losses and quality changes during hay and silage
production.

Type of forage, Dry matter loss Change in nutrient

Type of loss (%DM) concentration (g/kg DM)
Range Normal CP NDF TDN

Legume crops
Respirationa 1- 7 4 9 17 -17
Rain damage a, 5mm 3- 7 5 -4 14 -15
25 mm 7 - 27 17 -17 60 -70
50mm 12 - 50 31 -35 140 -142
Mowing/ conditioning 1- 4 2 -7 12 -14
Tedding 2- 8 3 -5 9 -12
Swath inversion 1- 3 1 0 0 0
Raking 1- 20 5 -5 10 -12
Baling, small bale 2- 6 4 -9 15 -19
round bale 3- 9 6 -17 31 -38
Chopping 1- 8 3 0 0 0
Hay storage, inside 3- 9 5 -7 21 -21
outside 6 - 30 15 0 50 -70
Silo storage, sealed 6 - 14 8 14 7 -37
stave 7 - 17 10 18 17 -47
bunker 10 - 16 12 23 27 -56

Grass crops
Respiration" 2- 8 5 8 32 -18
Rain damagea, 5 mm 1- 3 2 -2 9 -5
25 mm 4 - 14 8 -13 53 -30
50mm 8 - 27 15 -27 110 -60
Mowing/conditioning 1- 2 1 0 0 0
Tedding 1- 3 1 -2 4 -4
Swath inversion 1- 3 1 0 0 0
Raking 1- 20 5 -3 5 -6
Baling, small bale 2- 6 4 -5 9 -10
round bale 3- 9 6 -10 18 -20
Chopping 1- 8 3 0 0 0
Hay storage, inside 3- 9 5 -13 32 -18
outside 5 - 22 12 0 80 -48
Silo storage, sealed 6 - 14 8 8 9 -37
stave 7 - 17 10 12 22 -47
bunker 10 - 16 12 15 36 -56

aRespiration loss includes plant and microbial respiration for crop cured
without rain damage. Rain damage includes leaf loss, nutrient leachmg, and
microbial respiration resulting from rain damage.

CHANGES IN FORAGE QUALITY 858

 Typical losses and quality changes in forage harvest and storage are listed
in Table 3. Typical ranges and the normal OM losses are estimated based on
a comprehensive review of reported losses. Changes in CP, NOF, and roN
are estimated from the expected change in leaf-to-stem ratio and the relative
rates of removal of nutrients by respiration and rain damage.
Because interactions occur among various losses, component losses should
not be simply added to obtain total system loss. A more comprehensive study
of weather and machine impacts on losses and the interaction among losses is
performed with OAFOSYM, the dairy forage system model (Rotz et aI., 1989;
Buckmaster et aI., 1990). Long-term simulations with DAFOSYM provide
typical losses for a wide variety of alfalfa harvest systems over a full range of
crop moisture conditions (Figure 7). In addition, the model integrates DM and
nutrient losses with animal performance to predict the economic return for the
farm. The change in economic return can be used to represent the value of
forage loss expressed as a portion of the initial value (Rotz et al., 1991a). The
value loss across systems also is shown in Figure 7 for a typical dairy farm with
a herd milk production of 8172 kgjcow-yr. For most dry hay and silage
systems, about 30% of the initial crop value is lost by the time it is fed.
Dry matter and nutrient losses are only one of several considerations when
evaluating and selecting forage harvest systems. The overall goal for most
producers is to harvest forage with minimal nutrient loss at the lowest cost.
Other factors like timeliness of harvest and labor availability also must be
considered. Speeding the drying process, reducing the number of machine
operations, and modification and adjustment of machines can all be used to
reduce losses. Only by balancing the losses and costs with added benefits can
the best system for forage harvest be selected. Best options vary with climate,
crops grown, farm infrastructure, and farm management styles.

40.-----~,--------~1--------~I--~--------~----~
Directl Wilted 1 Haylage I Dried: Field
cut : silage : : hay I cured
silage 1 1 (sealed silo) I bales hay
I (bunker silo) I I
bunker : : I
30 _/1ormlc I : ~
acid) r-= ;.....---; I
I
•
1 I •
1
... 1
>:' ::
,I •
_. .,"
" I
I " I:
,," .r
fIl 20 1
I ... ",
,,"
... "","
..-1
fIl
o ",'" 1,.. ......
.'
-..
I 1
~ .'
---~
I I
~ ..
I I
I 1 .'
10 .. ' i'
.... ..
1
1 ' - Value loss
~

- - Total DM loss
........ ,
I
•••. Harvest DM loss

O~~~--~~--~~--L-~~--~~--~~~
800 700 600 500 400 300 200 100
Average moisture content at harvest (g/kg)

Figure 7. Typical harvest and storage OM losses and total value loss for
various alfalfa production systems used in Michigan (Rotz et aI., 1991a).

859 ROTZ&MUCK
 REFERENCES

Albrecht, K.A, and RE. Muck. 1991. Proteolysis in ensiled forage legumes that
vary in tannin concentration. Crop Sci. 31:464-469.

Barrington, G.P., and H.D. Bruhn. 1970. Effect of mechanical forage-harvesting

devices on field-curing rates and relative harvesting yields. Trans. ASAE
13:874-878.

Belyea, RL., F.A. Martz, and S. Bell. 1985. Storage and feeding losses of large
round bales. J. Dairy Sci. 68:3371-3375.
Boyd, M.M. 1959. Hay conditioning methods compared. Agric. Eng. 40:664-667.
Brady, C.J. 1960. Redistribution of nitrogen in grass and leguminous fodder
plants during wilting and ensilage. J. Sci. Food Agric. 11:276-284.

Brady, C.J. 1961. The leaf protease of Trifolium repens. Biochem. J. 78:631-640.

Brasche, M.R, and J.R Russell. 1988. Influence of storage methods on the util-
ization of large round hay bales by beef cows. J. Anim. Sci. 66:3218-3226.

Buckmaster, D.R 1993. Alfalfa raking losses as measured on artificial stubble.

Trans. ASAE 36(3):645-651.
Buckmaster, D.R, CA Rotz, and J.R Black. 1990. Value of alfalfa losses on
dairy farms. Trans. ASAE 33:351-360.

Buckmaster, D.R, CA Rotz, and D.R Mertens. 1989a. A model of alfalfa hay
storage. Trans. ASAE 32:30-36.

Buckmaster, D.R, CA Rotz, and RE. Muck. 1989b. A comprehensive model

of forage changes in the silo. Trans. ASAE 32:1143-1152.

Carpintero, C.M., AR Henderson, and P. McDonald. 1979. The effect of some

pre-treatments on proteolysis during the ensiling of herbage. Grass Forage
Sci. 34:311-315.
Christensen, M.D., M.N. Albury, and C.S. Pederson. 1958. Variation in the
acetic acid-lactic acid ratio among the lactic acid bacteria. Appl.
Microbial. 6:316-318.
Ciotti, A, and A. Cavallero. 1979. Haymaking losses in cocksfoot, lucerne and
a cocksfoot-lucerne mixture in relation to conditioning and degree of
drying at harvest. p. 214-220. In C. Thomas (ed.) Forage conservation in
the 80's. Proc. Occasional Symposium No. 11, Brighton, UK., 27-30 Nov.,
Br. Grassl. Soc., Hurley, UK.

Collins, M. 1982. The influence of wetting on the composition of alfalfa, red

clover, and birdsfoot trefoil hay. Agron. J. 74:1041-1044.

Collins, M. 1983. Wetting and maturity effects on the yield and quality of
legume hay. Agron. J. 75:523-527.
Collins, M. 1985. Wetting effects on the yield and quality of legume and
legume-grass hays. Agron. J. 77:936-941.

CHANGES IN FORAGE QUALITY 860

 Collins, M., W.H. Paulson, M.F. Finner, N.A Jorgensen, and C.R Keuler. 1987.
Moisture and storage effects on dry matter and quality losses of alfalfa in
round bales. Trans. ASAE 30:913-917.

Condon, S. 1987. Responses of lactic acid bacteria to oxygen. FEMS Microbiol.

Rev. 46:269-280.
Courtin, M.G., and S.F. Spoelstra. 1990. A simulation model of the microbio-
logical and chemical changes accompanying the initial stage of aerobic
deterioration of silage. Grass Forage Sci. 45:153-165.
Davies, M.H., and I.B. Warboys. 1978. The effect of propionic acid on the
storage losses of hay. J. Br. Grassl. Soc. 33:75-82.

Davis, RJ., C.A Rotz, and D.R Buckmaster. 1989. Effect of swath manipu-
lation on losses and drying of alfalfa. p. 329-332. In Proc. 1989 Forage
and Grassland Conference, Guelph, Ontario, Canada. 22-25 May, Am.
Forage Grassl. Council, Georgetown, TX.
Dewar, W.A, P. McDonald, and R Whittenbury. 1963. The hydrolysis of grass
hemicelluloses during ensilage. J. Sci. Food Agric. 14:411-417.

Dickerson, J.T., KK BoIsen, e. Lin, and G. Ashbell. 1990. Rate and extent of
top spoila~e losses in horizontal silos. p. 107-108. In Proc. Ninth Silage
Conf., Umv. Newcastle upon Tyne, Newcastle upon Tyne, UK
Dobie, J.B., J.R Goss, R.A. Kepner, J.H. Meyer, and L.G. Jones. 1963. Effect
of harvesting procedures on hay quality. Trans. ASAE 6:301-303.

Doehlert, D.e., and S.H. Duke. 1983. S{'ecific determination of Ct -amylase activ-
ity in crude plant extracts contaimng B-amylase. Plant Physiol. 71:229-234.
Fonnesbeck, P.V., M.M. Garcia de Hernandez, J.M. Kaykay, and M.Y. Saiady.
1986. Estimating yield and nutrient losses due to rainfall on field drying
alfalfa hay. Anim. Feed Sci. Technol. 16:7-15.
Friesen, O. 1978. Evaluation of hay and fora~e harvesting methods. In G.R
Quick (ed.) Grain and forage harvestmg. Am. Soc. Agric. Eng., St.
Joseph, MI.
Goering, H.K 1976. A laboratory assessment on the frequency of over-heating
in commercial dehydrated alfalfa samples. J. Anim. Sci. 43:869-872.

Goering, H.K., PJ. Van Soest, and RW. Hemken. 1973. Relative susceptibility
of forages to heat damage as affected by moisture, temperature and pH.
J. Dairy Sci. 56:137-143.

Gordon, e.H., RD. Holdren, and J.e. Derbyshire. 1969. Field losses in
harvesting wilted forage. Agron. J. 61:924-927.
Goss, J.R, RA. Kepner, and L.G. Jones. 1964. Hay harvesting with self-
propelled windrower compared with mowing and raking. Trans. ASAE
7:357-361.

Greenhill, W.L. 1959. The res~iration drift of harvested pasture plants during
drying. J. Sci. Food Agnc. 10:495-501.

861 ROTZ&MUCK
 Greenhill, W.L., J.F. Couchman, and J. De Freitas. 1961. Storage of hay III -
Effect of temperature and moisture on loss of dry matter and changes in
composition. J. Sci. Food Agric. 12:293-297.

Harris, CE., and J.N. Tulberg. 1980. Pathways of water loss from legumes and
grasses cut for conservation. Grass Forage Sci. 35:1-11.

Hart, RH., and G.W. Burton. 1967. Curing Coastal bermuda grass hay: Effects
of weather, yield, and quality of fresh herbage on drying rate, yield, and
quality of cured hay. Agron. J. 59:367-371.

Harrigan, T.M., and CA Rotz. 1992. Net, plastic and twine wrapped large
round bale storage loss. Paper 921572. Am. Soc. Agric. Eng., St. Joseph,
MI.

Henderson, AR, P. McDonald, and M.K. Woolford. 1972. Chemical changes

and losses durin~ the ensilage of wilted grass treated with formic acid.
J. Sci. Food Agnc. 23:1079-1087.

Hodgson, RE., J.B. Shepherd, W.H. Hosterman, L.G. Schoenleber, H.M.

Tysdal, and RE. Wagner. 1947. Comparative efficiency of ensiling, barn
curing, and field curing forage crops. Agric. Eng. 28: 154-156.

Hoglund, CR 1964. Comparative storage losses and feeding values of alfalfa

and corn silage crops. Agric. Econ. Report 947, Agric. Econ. Dept.,
Michigan State Univ., East Lansing, MI.

Honig, H. 1979. Mechanical and respiration losses during pre-wilting of grass.

p. 201-204. In C Thomas (ed.) Forage conservation in the 80's. Proc.
Occasional Symposium No. 11, Brighton, UK., 27-30 Nov., Br. Grassl.
Soc., Hurley, UK.

Huhnke, RL. 1988. Large round bale alfalfa hay storage. Appl. Eng. Agric.
4:316-318.
Huhnke, RL. 1990. Round bale bermudagrass hay storage losses. Appl. Eng.
Agric. 6:396-400.
Johnson, D.G., D.E. Otterby, RG. Lundquist, J.A True, F.A Benson, RE.
Smith, L.K. Lindor, and RC. Stommes. 1984. Yield and quality of alfalfa
as affected by harvesting and storage. J. Dairy Sci. 67:2475-2480.
Jones, B.A, RD. Hatfield, and RE. Muck. 1992. Effect of fermentation and
bacterial inoculation on lucerne cell walls. J. Sci. Food Agric. 60:147-153.

Knapp, W.R, D.A Holt, and V.L. Lechtenberg. 1976. Propionic acid as a hay
preservative. Agron. J. 68:120-123.

Kemble, AR 1956. Studies on the nitrogen metabolism of the ensilage process.

J. Sci. Food Agric. 7:125-130.

Kemble, AR, and H.T. Macpherson. 1954. Liberation of amino acids in

perennial rye grass during wilting. Biochem. J. 58:46-49.

Kepner, RA, J.R Goss, and J.H. Meyer. 1960. Evaluation of hay conditioning
effects. Agric. Eng. 41:299-304.

CHANGES IN FORAGE QUALITY 862

 Klinner, W.E., and G. Shepperson. 1975. The state of haymaking technology -
A review. J. Br. Grassl. Soc. 30:259-266.
Koegel, RG., K.J. Shinners, F.J. Fronczak, and RJ. Straub. 1988. Prototype for
production of fast-drying forage mats. Appl. Eng. Agric. 4:126-129.
Koegel, RG., RJ. Straub, and RP Walgenbach. 1985. Quantification of
mechanical losses in forage harvesting. Trans. ASAE 28:1047-1051.
Leibensperger, RY., and RE. Pitt. 1987. A model of clostridial dominance in
ensilage. Grass Forage Sci. 42:297-317.
MacAulay, J.D., and W.K. Bilanski. 1968. Mechanical properties affecting leaf
loss-in birdsfoot trefoil. Trans. ASAE 11:568-571.
Macpherson, H.T. 1952. Changes in nitrogen distribution in crop conservation.
II. Protein breakdown during wilting. J. Sci. Food Agric. 3:365-367.
Mayne, C.S., and F.J. Gordon. 1986. Effect of harvesting system on nutrient
losses during silage making. 2. In-silo losses. Grass Forage Sci. 41:341-
351.
McDonald, P. 1981. The biochemistry of silage. John Wiley and Sons,
Chichester, UK.
McGechan, M.B. 1989a. Alternative models of air infiltration into silage
clamps. Dept. Note 24, Scottish Centre Agric. Eng., Penicuik, UK.
McGeehan, M.B. 1989b. A review of losses arising during conservation of grass
forage: Part 1, Field losses. J. Agric. Eng. Res. 44:1-21.
McKersie, B.D. 1981. Proteinases and peptidases of alfalfa herbage. Can J.
Plant Sci. 61:53-59.
McKersie, B.D. 1985. Effect of pH on proteolysis in ensiled legume forage.
Agron. J. 77:81-86.
McKersie, B.D., and J. Buchanan-Smith. 1982. Changes in the levels of
proteolytic enzymes in ensiled alfalfa forage. Can. J. Plant Sci. 62:111-
116.
Meiering, AG., M.G. Courtin, S.F. Spoelstra, G. Pahlow, H. Honig, RE.
Subden, and E. Zimmer. 1988. Fermentation kinetics and toxic gas
production of silages. Trans. ASAE 31:613-621.
Melvin, J.F. 1965. Variations in the carbohydrate content of lucerne and the
effect on ensilage. Aust. J. Agric. Res. 16:951-959.
Melvin, J.F., and B. Simpson. 1963. Chemical chan~es and respiratory drift
during the drying of ryegrass. J. Sci. Food Agnc. 14:228-234.
Morgan, C.A, RA Edwards, and P. McDonald. 1980. Intake and metabolism
studies with fresh and wilted silages. J. Agric. Sci., Camb. 94:287-298.
Morris, RM. 1972. The rate of water loss from grass samples during hay-type
conservation. J. Br. Grassl. Soc. 27:99-105.

863 ROTZ&MUCK
 Morrison, I.M. 1979. Changes in the cell wall of laboratory silages and the
effect of various additives on these changes. J. Agric. Sci., Camb. 93:581-
586.
Morrison, 1M. 1989. Influence of some chemical and biological additives on the
fibre fraction of lucerne on ensilage in laboratory silos. J. Agric. Sci.,
Camb. 111:35-39.
Muck, RE. 1987. Dry matter level effects on alfalfa silage quality. 1 Nitrogen
transformations. Trans. ASAE 30:7-14.
Muck, RE. 1990. Dry matter level effects on alfalfa silage quality. II.
Fermentation products and starch hydrolysis. Trans. ASAE 33:373-381.
Muck, RE. 1991. Silage fermentation. p. 171-204. In G. Zeikus and
E.A. Johnson (ed.) Mixed cultures in biotechnology. McGraw-Hill, New
York.
Muck, RE., and K.K. BoIsen. 1991. Silage preservation and silage additive
products. p. 105-126. In K.K. BoIsen (ed.) Field guide for hay and silage
management in North America, NFIA, Des Moines, IA.
Muck, RE., and J.T. Dickerson. 1988. Storage temperature effects on
proteolysis in alfalfa silage. Trans. ASAE 31:1005-1009.
Muck, RE., and RE. Pitt. 1992. Aerobic losses at the silo face. Paper 921003,
Am. Soc. Agric. Eng., St. Joseph, MI.
Muck, RE., RE. Pitt, and RY. Leibensperger. 1991. A model of aerobic
fungal growth in silage. 1. Microbial characteristics. Grass Forage Sci.
46:283-299.
Muck, RE., S.F. Spoelstra, and P.G. van Wikselaar. 1992. Effects of carbon
dioxide on fermentation and aerobic stability of maize silage. J. Sci.
Food Agric. 59:405-412.
Murdoch, J.e., and D.1. Bare. 1960. The effect of mechanical treatment on the
rate of drying and loss of nutrients in hay. J. Br. Grassl. Soc. 15:94-99.
Murdoch, J.C., and D.1. Bare. 1963. The effect of conditioning on the rate of
drying and loss of nutrients in hay. J. Br. Grassl. Soc. 18:334-338.
Neal, H.D. St C., and J.H.M. Thornley. 1983. A model of the anaerobic phase
of ensiling. Grass Forage Sci. 38:121-134.
Nelson, L.F. 1966. Spontaneous heating and nutrient retention of baled alfalfa
hay during storage. Trans. ASAE 9:509-512.
Nelson, L.F. 1968. Spontaneous heating, gross energy retention and nutrient
retention of high-density alfalfa hay bales. Trans. ASAE 11:595-600.
Nelson, L.F. 1972. Storage characteristics and nutritive value of high-density
native hay bales. Trans. ASAE 15:201-205.
Ohshima, M., and P. McDonald. 1978. A review of the changes in nitrogenous
compounds of herbage during ensilage. J. Sci. Food Agric. 29:497-505.

CHANGES IN FORAGE QUALITY 864

 Parkes, M.E., and D.J. Greig. 1974. The rate of respiration of wilted rye grass.
J. Agric. Eng. Res. 19:259-263.

Parsons, D. 1991. Modelling gas flows in a silage clamp after opening. J. Agric.
Eng. Res. 50:208-218.

Pitt, RE. 1986. Dry matter losses due to oxygen infiltration in silos. J. Agric.
Eng. Res. 35:193-205.

Pitt, RE., and RE. Muck. 1993. A diffusion model of aerobic deterioration at
the exposed face of bunker silos. J. Agric. Eng. Res. 55:11-26.
Pitt, RE., RE. Muck, and RY. Leibensperger. 1985. A quantitative model of
the ensilage process in lactate silages. Grass Forage Sci. 40:279-303.
Pitt, RE., RE. Muck, and N.B. Pickering. 1991. A model of aerobic fungal
growth in silage. 2. Aerobic stability. Grass Forage Sci. 46:301-312.
Pizarro, E.A, and D.B. James. 1972. Estimates of respiratory rates and losses
in cut swards of Lolium perenne (S321) under simulated haymaking
conditions. J. Br. Grassl. Soc. 27:17-21.

Pizarro, E.A, and I.B. Warboys. 1979. The effect of the wilting period on the
microflora of harvested pasture plants. p. 53-60. In C. Thomas (ed.)
Forage conservation in the 80's. Proc. Occasional Symposium No. 11,
Brighton, UK, 27-30 Nov., Br. Grassl. Soc., Hurley, UK

Purves, D., and P. McDonald. 1963. The potential value of silage effluent as a
fertiliser. J. Br. Grassl. Soc. 18:220-222.
Raghavan, G.S., and W.K Bilanski. 1973. Mechanical properties affecting leaf
loss in alfalfa. Can. Agric. Eng. 15:20-23.
Rees, D.V.H. 1982. A discussion of sources of dry matter loss during the
process of haymaking. J. Agric. Eng. Res. 27:469-479.

Rice, B. 1966. A comparison of six types of hay making machines. Irish J. Agric.
Res. 5:43-53.

Rider, AR, D. Batchelder, and W. McMurphy. 1979. Effects of long term

outside storage on round bales. Paper 791538. Am. Soc. Agric. Eng., St.
Joseph, MI.

Rotz, CA, and S.M. Abrams. 1988. Losses and quality changes during alfalfa
hay harvest and storage. Trans. ASAE 31:350-355.
Rotz, CA, S.M. Abrams, and RJ. Davis. 1987. Alfalfa drying, loss and quality
as influenced by mechanical and chemical conditioning. Trans. ASAE
30:630-635.

Rotz, CA, J.R Black, D.R Mertens, and D.R Buckmaster. 1989. DAFOSYM:
A model of the dairy forage system. J. Prod. Agric. 2:83-91.

865 ROTZ&MUCK
 Rotz, C.A, L.R Borton, and J.R Black. 1991a. Harvest and storage losses with
alternative forage harvesting methods. p. 210-213. In Forages: A versatile
resource. Proc. 1991 Forage and Grassland Com., Columbia, MO. 1-4
April, Am. Forage Grassl. Council, Georgetown, TX.

Rotz, C.A, D.R Buckmaster, and L.R Borton. 1992. Economic potential of
preserving high-moisture hay. Appl. Eng. Agric. 8:315-323.
Rotz, C.A, and RJ. Davis. 1986. Sprayer design for chemical conditioning of
alfalfa. Trans. ASAE 29:26-30.

Rotz, C.A., RJ. Davis, and S.M. Abrams. 1991b. Influence of rain and crop
characteristics on alfalfa damage. Trans. ASAE 34:1583-1591.

Rotz, C.A., RJ. Davis, D.R Buckmaster, and M.S. Allen. 1991c. Preservation
of alfalfa hay with propionic acid. Appl. Eng. Agric. 7:33-40.

Rotz, C.A, RJ. Davis, D.R Buckmaster, and J.W. Thomas. 1988. Bacterial
inoculants for preservation of alfalfa 'hay. J. Prod. Agric. 1:362-367.

Rotz, C.A, RG. Koegel, KJ. Shinners, and RJ. Straub. 1990. Economics of
maceration and mat drying of alfalfa on dairy farms. App!. Eng. Agric.
6:248-256.

Rotz, C.A, RE. Pitt, RE. Muck, M.S. Allen, and D.R Buckmaster. 1993.
Direct-cut harvest and storage of alfalfa on the dairy farm. Trans. ASAE
36(3):621-628.

Rotz, C.A, and DJ. Sprott. 1984. Drying rates, losses and fuel requirements for
mowing and conditioning alfalfa. Trans. ASAE 27:715-720.

Rotz, CA, D.J. Sprott, and J.W. Thomas. 1984. Interaction of mechanical and
chemical conditioning of alfalfa. Trans. ASAE 27:1009-1014.

Rotz, CA, DJ. Sprott, and J.W. Thomas. 1986. Anhydrous ammonia injection
into baled forage. Appl. Eng. Agric. 2:64-69.
Russell, J.B., and T. Hino. 1985. Regulation of lactate production in
Streptococcus bovis: A spiraling effect that contributes to rumen acidosis.
J. Dairy Sci. 68:1712-1721.

Russell, J.R, and D.R Buxton. 1985. Storage of large round bales of hay
harvested at different moisture concentrations and treated with sodium
diacetate and/or covered with plastic. Anim. Feed Sci. Technol. 13:69-81.

Russell, J.R, S.J. Yoder, and SJ. Marley. 1990. The effects of bale density, type
of binding and storage surface on the chemical composition, nutrient
recovery and digestibility of large round hay bales. Anim. Feed Sci.
Techno!. 29:131-145.
Savoie, P. 1988. Hay tedding losses. Can. Agric. Eng. 30:39-42.
Savoie, P., and S. Beauregard. 1990. Hay windrow inversion. Appl. Eng. Agric.
6:138-142.

CHANGES IN FORAGE QUALITY 866

 Savoie, P., C.A Rotz, H.F. Bucholtz, and RC. Brook. 1982. Hay harvesting
system losses and drying rates. Trans. ASAE 25:581-585.
Schukkin~, S., and J. Overvest. 1979. Direct and indirect losses caused by
wilting. p. 210-223. In e. Thomas (ed.) Forage conservation in the 80's.
Proc. Occasional Symposium No. 11, Brighton, UK, 27-30 Nov., Br.
Grassl. Soc., Hurley, UK
Simpson, B. 1961. Effect of crushing on the respiration drift of pasture plants
during drying. J. Sci. Food Agric. 12:706-712.
Shinners, KJ., RG. Koegel, and RJ. Straub. 1991. Leaf loss and drying rate of
alfalfa as affected by conditioning roll type. Appl. Eng. Agric. 7:46-49.
Shinners, KJ., RJ. Straub, and RG. Koegel. 1992. Performance of two small
rectangular baler configurations. Appl. Eng. Agric. 8:309-313.
Shepherd, J.B., H.G. Wiseman, RE. Ely, e.G. Melin, W.J. Sweetman, C.H.
Gordon, L.G. Schoenleber, RE. Wagner, L.E. Campbell, G.D. Roane,
and W.H. Hosterman. 1954. Experiments in harvesting and preserving
alfalfa for dairy cattle feed. Tech. Bull. No. 1079, USDA, Washington,
De.
Sneath, P.H.A, N.S. Mair, and M.E. Sharpe. 1986. Bergey's manual of
systematic bacteriology, Vol. 2, Williams and Wilkins, Baltimore, MD.
Spoelstra, S.F. 1987. Degradation of nitrate by enterobacteria during silage
fermentation of grass. Neth. J. Agric. Sci. 35:43-54.
Spoelstra, S.F. 1990. Effects of cell wall degrading enzymes on silage
composition at different ensiling conditions. p. 27-28. In Proc. Ninth
Silage Conference, Faculty Agric., Univ. Newcastle upon Tyne.
Straub, R.J., and H.D. Bruhn. 1975. Evaluation of roll design in hay
conditioning. Trans. ASAE 18:217-220.
Straub, R.J., RG. Koegel, and KJ. Shinners. 1986. Mechanical losses from
alfalfa harvesting strategies. Paper 861035. Am. Soc. Agric. Eng., St.
Joseph, MI.
Sullivan, J.T. 1973. Drying and storing herbage as hay. p. 1-31. In G.W. Butler
and RW. Bailey (ed.) Chemistry and biochemistry of herbage. Vol. 3.
Academic Press, London, England.
Svensson, K 1978. Yield and quality losses in forage harvesting - some results
of Swedish experiments. p. 310-314. In G.R Quick (ed.) Grain and forage
harvesting. Am. Soc. Agric. Eng., St. Joseph, MI.
Thomas, J.W., Y. Yu, T. Middleton, and e. Stallings. 1982. Estimations of
protein damage. p. 81-98. In F.N. Owens (ed.) Protein requirements for
cattle: Symposium. 19-21 Nov. 1980, MP109, Division of Agric., Oklahoma
State Univ., Stillwater.
Tomes, N.J., S. Soderlund, J. Lamptey, S. Croak-Brossman, and G. Dana. 1990.
Preservation of alfalfa hay by microbial inoculation at baling. J. Prod.
Agric.3:491-497.

867 ROTZ & MUCK

 Van Bockstaele, EJ., TJ. Behaeghe, and A.E. De Baets. 1979. Studies on the
field losses of wilting grass. p. 205-207. In C. Thomas (ed.) Forage
conservation in the 80's. Proc. Occasional Symposium No. 11, Brighton,
UK., 27-30 Nov., Br. Grassl. Soc., Hurley, UK.

Verma, L.R, and B.D. Nelson. 1983. Changes in round bales during storage.
Trans. ASAE 26:328-332.

Verma, L.R, B.D. Nelson, and c.R. Montgomery. 1986. Preservation of high
moisture ryegrass hay with ammonia. AppI. Eng. Agric. 2:76-81.

Whitney, L.F. 1966. Hay losses from baler and chopper components. Trans.
ASAE 9:277-278.

Wilkinson, J.M. 1981. Losses in the conservation and utilization of grass and
forage crops. Ann. AppI. BioI. 98:365-375.

Wilkinson, RH., and C.W. Hall. 1966. Respiration heat of harvested forage.
Trans. ASAE 9:424-427.

Wilman, D., and I.G. Owen. 1982. Effects of stage of maturity, nitrogen
application and swath thickness on the field drying of herbage to the hay
stage. J. Agric. Sci., Camb. 99:577-586.

Waldo, D.R 1977. Suggestions for improving silage. How far with forages for
meat and milk production? p. 69-91. In Proc. 10th Research-Industry
Conf., Columbia, MO. Am. Forage Grassl. Council, Georgetown, TX.
Waldo, D.R., and N.A. Jorgensen. 1981. Forages for high animal production:
Nutritional factors and effects of conservation. J. Dairy Sci. 64:1207-1229.

Wolf, D.O., and E.W. Carson. 1973. Respiration during drying of alfalfa
herbage. Crop Sci. 13:660-662.

Wood, J.G.M., and J. Parker. 1971. Respiration during the drying of hay. J.
Agric. Eng. Res. 16:179-191.
Woolford, M.K. 1975. Microbiological screening of the straight chain fatty acids
(C1-C12) as potential silage additives. J. Sci. Food Agric. 26:219-228.
Woolford, M.K. 1990. The detrimental effects of air on ensilage. J. Appl. Bact.
68:101-116.

CHANGES IN FORAGE QUALITY 868

View publication stats