Bacon Recipe
Bacon Recipe
Bacon Recipe
5-2004
Robles, Carmina Citlali, "THE EFFECT OF FRESH AND FROZEN BELLIES ON BACON PROCESSING CHARACTERISTICS
AND BACON QUALITY" (2004). Theses and Dissertations in Animal Science. 17.
http://digitalcommons.unl.edu/animalscidiss/17
This Article is brought to you for free and open access by the Animal Science Department at DigitalCommons@University of Nebraska - Lincoln. It has
been accepted for inclusion in Theses and Dissertations in Animal Science by an authorized administrator of DigitalCommons@University of Nebraska
- Lincoln.
THE EFFECT OF FRESH AND FROZEN BELLIES ON BACON PROCESSING
by
A THESIS
Lincoln, Nebraska
May, 2004
THE EFFECT OF FRESH AND FROZEN BELLIES ON BACON PROCESSING
The objective of the study was to evaluate the impact of fresh vs. frozen bellies on
bacon parameters. Differences among bellies from pigs of different genetic lines and
sexes were evaluated. A population of 578 pork bellies from barrows and gilts of
Berkshire, Chester White, Duroc, Landrace, Poland China and Yorkshire lines were
processed to sliced bacon and cooked bacon to be evaluated. Bellies were divided into
two storage treatments: fresh (stored at 3°C) or frozen (held at -15°C). Fat samples were
collected from each belly to establish a fatty acid profile. Differences in percent pump,
smokehouse yield, slicing yield and total yield were evaluated. The bacon slabs were
sliced and cooked to evaluate differences in fatty acid content, slicing yield, shattering,
cooking yield, distortion and proximate composition. Fresh bellies had bacon slicing
yields 1.74 percent higher (P<0.05) than frozen bellies. Total yield was not different
(P>0.05) for fresh and frozen bellies. No differences (P>0.05) were found in the
saturated and unsaturated fatty acid profile of the bellies due to storage treatment. Gilts
had a profile with a 1.08 percent higher (P<0.05) unsaturated fatty acid (palmitoleic,
oleic, linoleic, linolenic and 11-eicosenoatic) when compared to barrows. In the center of
the slab (locations B, C and D,) frozen bellies had a higher total shatter mark length
(P<0.05). Genetic lines with higher fat content had lower (P<0.05) cooking yields.
Frozen bellies had higher (P<0.05) distortion scores of the cooked bacon slice than fresh
bellies. Distortion was affected by an interaction (P<0.05) among line and sex. Higher
yields were associated with sex and lines containing higher percentage of fat in the
bellies. Overall, fresh and frozen bellies had similar characteristics when processed.
these values may be slight. Considering that bellies were stored frozen for 15 d, it is not
known if longer periods of storage could result in lower yields and significant processing
They have been great source of inspiration. If it was not for them I would never have
accomplished so much. I thank them for all the efforts they have made throughout my
entire life, for being behind me, for their support, for their guidance, and for their
I love you.
Two very special people in my life, my sister Liliana and my brother Armando,
for always being interested in what I do, for their support, guidance and their love.
I want to thank my major professor, Dr. Roger W. Mandigo, for giving me the
opportunity to work with him and for challenging me every day in many different ways.
To my committee, Dr. Dennis Burson, Dr. Chris Calkins, and Dr. Steven Jones, Thank
You!
Many thanks to Tommi Jones, Sherri Pitchie, and Calvin Schrock for their friendship,
Special thanks to Dr. Steve Kachman and Dr. Gonzalo Martinez for their help with data
analysis.
To all my fellow graduate students thank you for making me laugh, for sharing my
frustrations and for your support. I will miss you. Cheers for the good times!
Betsy Booren, Joe Baumert, Mike Backawski, Clint Schafer, Steve Coufal, Oscar
Esquivel, Brad Creamer, Ryan Baumert and Jenny Sherrill thank you for all your hard
There have been many important people in my life, with which I have shared, many
precious moments. I want to thank them for believing in me, for their support, for their
help, and for pushing me all along the way. Thank you for your unconditional friendship.
FRIENDS!
Last, but never least, to Dana J. Hanson, for his understanding and support. I thank him
for his advice, help and support, but more importantly for his Love.
"Every day you may make progress. Every step may be fruitful. Yet there will stretch out
will never get to the end of the journey. But this, so far from discouraging, only adds to
Review of Literature
The effect of fresh and frozen bellies on the fatty acid profile of bellies from pigs of
different genetic line and sex on bacon quality
I. Curing Technology
Historical Perspective
The curing of meat is not a new concept, but rather a collection of techniques that
have evolved from years of product processing. According to Lechowich et aI. (1978),
prior to 1900, curing was strictly an art and not a science, being largely a matter of secret
recipes handed down from generation to generation. Curing of meat today is based on
sound scientific principles developed since the tum of the century. Curing is a complex
1997).
Several authors define curing as the treatment of meat with substances for the
purpose of durability or to increase the keeping qualities, fix the color, and alter and
improve the flavor (Lautenschlager, 1997; Lechowich et aI., 1978). Cassens (1995)
stated that the resulting cured meat characteristics of color, flavor, and texture, are
Curing Methods
Meats are cured by bringing them into intimate contact with a curing agent ("dry
cure" or "sweet pickle cure"). The time required for curing will depend on the ability of
the curing agent to penetrate to all parts of the meat (Lechowich et aI., 1978).
water. Cure is a dry mixture of ingredients including salt and nitrites. A pickle will also
include sugar, phosphates and ascorbates (Lechowich et aI., 1978). Cure may also include
spices and seasonings according to the preference of the manufacturer and the product
(Cassens et aI., 1979). Xargayo et aI. (2001) stated that a wide range of products such as
spices, fruit extracts, aromatic liquor, oils, oriental sauces, have been used to increase
palatability of meat.
The curing ingredients can be introduced into the meat by several different
methods. A dry mixture of the ingredients may be rubbed on the surface of pieces of
meat. It will dissolve in the moisture of the meat, and with allowed time, diffusion will
carry them to the interior of the product. Alternatively, a piece of meat can be soaked in
a liquid solution or pickle. This liquid solution may also be injected into the meat under
Most cures are made as a "pickle" or brine solution. A specified amount of the
pickle is injected (pumped) into the product with the objective of delivering a
predetermined amount of each of the curing ingredients in the final product (Lechowich
et aI., 1978). According to Xargayo et aI. (2001) injection is the most widely used
method because it allows dosing an exact quantity of pickle. It also increases the
uniformity of the products and decreases the diffusion time required by the dry cure or
soaking methods. These are in agreement with Rust and Olson (1973). They reported
that with an automatic multiple needle stitch pumping machine cure distribution was so
even, tempering periods were not needed before cooking. The injection of the pickle
accelerates the curing process because it reduces the time and range required for the brine
Water
Water is the largest ingredient (on a weight basis) used in the manufacture of
cured meat. It is used to dissolve, transport and distribute non-meat ingredients into the
product. It is also used to replace natural meat juices that will be cooked out when the
product is heated. Therefore, water helps maintain profit margins by reducing shrinkage
water. Water must be suitable for drinking (potable). Additionally, it should be free of
contaminants. The hardness of water has a major impact on the solubility of certain non-
Nitrite
The salts of nitrite are soluble in water, which allows them to react in numerous
ways with functional groups and therefore have many industrial applications (Cassens et
aI., 1979). The well recognized sensory effect nitrite has on the final product (Pink color
and characteristic flavor) is proof that it reacts when it is added to meat (Cassens et aI.,
1979). Furthermore, some of the added nitrite disappears (Widdus and Busta, 1982).
Greenwood (1940) found that when nitrite was added to a meat slurry, immediate
Nitrate has been used for centuries in the process of cured meat. Sebranek (1979)
concluded that nitrite was formed by bacterial reduction on nitrate. Hoagland (1914)
reviewing previous work, mentioned that Kisskalt studied the production of red color
during cooking and found that the color appeared when meat was cooked in water
containing nitrites. Meat boiled in water with saltpeter did not take on the red color, but
if the meat was first allowed to stand several days in contact with saltpeter and then
boiled, the red color appeared. Lehman as cited by Sebranek (1979) and Hoagland
(1914) showed that the cured color was derived from nitrite. When fresh meat was boiled
in water containing nitrites and free acid or in old meat broth the surface of the meat
turned bright red in color, in contrast to the brown color which fresh meat took on when
boiled in water free from nitrites (Hoagland, 1914). Haldane (1901) found that the color
of uncooked salted meats was soluble in water and gave a spectrum characteristic of NO-
reduction of nitrite to nitric oxide and the modification of pigments by cooking. Lewis et
al. (1925) demonstrated that the direct addition of nitrite to cures reduced the cure
penetration process in whole muscle meats from 2 to 1 months. Kerr et al. (1926)
demonstrated that replacing nitrates with nitrites in ham and bacon production diminished
nitrite quantities in the final product. Brooks et al. (1940) demonstrated that sodium
nitrite could replace sodium nitrate for curing hams and bacon. The reduction of nitrites
in the final product was beneficial. They demonstrated that reactions between nitrite and
secondary amines, which are natural components of meat, may lead to formation of
Wasserman and Talley (1972) subjectively studied the influence of nitrite on the
flavor of frankfurters. Judges could detect flavor differences of franks prepared with or
without nitrite in a triangle test. The test consisted of comparing two samples to a
reference sample. The panelists were to determine which one of the two samples is
similar to the reference. They found that there was a highly significant difference in
flavor produced by the addition of nitrite. When the frankfurters were smoked, however,
there was essentially no difference in the scores ofthe untreated and nitrite-treated franks.
The opposite results were found by Cho and Bratzler (1970), who reported that the
panelists were able to detect the different sample in a triangle test. Panelist indicated that
the nitrite cured smoked sample had more cured flavor than the smoked samples without
nitrite.
The United States Department of Agriculture (USDA) has regulated nitrate and
nitrite use in meat products since the early 1900s. In 1925 the USDA authorized the use
of nitrite for curing (Kerr et aI., 1926). In general, only nitrite is now used to cure meat
(Cassens, 1995; Sebranek, 1979). Its use is restricted to a maximum of 120 ppm of
Nitrite contributes to the formation of color, specific flavor and texture of the final
(Cassens, 1995). Pivnick et aI., (1969) found that meat with a low inoculate (1 spore per
g) became toxic if salt and nitrite were omitted. In contrast, meat inoculated with log 106
spores per g remained non-toxic after the same thermal process provided sufficient salt
and nitrite in the formulation. From the point of view of immediate health protection, the
antibotulinal action is considered the most important contribution (Widdus and Busta,
1982).
In recent years the use of nitrites has stimulated some health concerns. Food can
~
be contaminated with nitroso compounds. Rywotycki (2001) established that nitroso
compounds are formed in slightly acid environments as a result of the reaction of sodium
nitrite and nitric oxide with proteins, amino acids and amines present in the foodstuff,
which are precursors for the reaction. Yagmur and Ozer (2001) reported that it is
High levels of nitrites in the meat can lead to the formation of carcinogenic nitrosamines.
Phosphates
Phosphates are the salt form of phosphoric acid. Phosphates are formed when the
acid is neutralized with alkali metal ions (sodium, potassium or calcium). Phosphates are
classified based on the amount of anions the molecule contains. Ortho or simple
alkali source. There are three types of orthophosphates based on the hydrogen atoms
available on the phosphoric acid. Monobasic orthophosphates have one hydrogen atom
replaced with an alkali metal. Dibasic orthophosphates have two hydrogen atoms
replaced. Tribasic orthophosphates have all three hydrogen atoms replaced (Shimp,
1983). Poly or condensed phosphates contain two or more phosphate anions (Dziezak,
three phosphorus atoms. Both pyro and tripolyphosphates are crystalline solids, while
sodium polyphosphates with chain lengths greater than three phosphorous atoms are not
)
crystalline, but amorphous products, commonly called "glassy" phosphates (Shimp,
1983; Ellinger, 1972). The crystalline phosphates are fully neutralized and act as bases.
unique in that it contains two hydrogen atoms and thus, is not basic but, displays a very
sequester metal ions and they act as polyanions increasing the ionic strengths of
solutions. The ability of phosphates to interact in food systems allows them to provide
the final product with specific characteristics. In meat and meat products, their functions
and stabilization and improvement of firmness (Lechowich et aI., 1978; Wagner, 1986).
The greatest benefit of the use of phosphates in meat products is to manage water.
system. Increased ionic strength will dissociate the muscle proteins and create more
binding sites for water. Therefore, less of the meat juices (water) will be released during
cooking. When phosphates are used in combination with salt, the effect on the water
binding ability is increased (Dziezak, 1990). This synergetic effect is widely used in
processed meats to improve juiciness, increase water retention and improve texture.
Phosphates are used in most pumped meats, such as ham, bacon, roast beef,
cooked corned beef, pastrami, and some similar poultry products. Their advantages
Generally, the orthophosphates have the least value in improving yields, emulsion
~ "-
stability and shelf life. Sodium acid pyrophosphate accelerates the development of cured
color, but is generally detrimental to emulsion stability and the water-binding ability of
meat proteins: hence, the alkaline polyphosphates are the phosphates that give the
greatest benefit to processed meat products (Rust and Olson, 1983). Pyrophosphates or
diphosphates work best when used for emulsion products. In curing brines, mixtures of
tripolyphosphate and sodium hexametaphosphate are used. They are dissolved in water
and injected into hams and bacons. Hexametaphosphates are slowly hydrolyzed to
recommended. High temperatures may result in faster hydrolysis and cause rapid
formation of monophosphate, which is inactive in the tissues (Pearson and Gillett, 1999).
Salt
chloride. In the manufacture of meat products, salt has the following functions: it
strengthens water binding which decreases water activity; acts as a preservative reducing
microbial growth and increasing shelf life; and improves flavor acting as a flavoring
agent and flavor enhancer (Kuhne, 1989; Lechowich et aI., 1978). According to Pearson
and Gillett (1999) the use of salt might result in harsh, dry, salty product that is not very
palatable. It could also result in a dark, undesirable colored lean that is unattractive to the
consumers. The cover pickle, used for immersing the meat, may contain about 15
percent of salt, in contrast to the pumping pickle, injected into the meat, which has a
water holding capacity and emulsifying properties of the meats. Only food grade salt
should be use in curing, since impure salt can cause flavor and color problems (Pearson
Sugar
The role of sugar in the curing process has been to reduce the harshness of salt,
and improve flavor and color (Lechowich et aI., 1978). Pearson and Gillett (1999)
reported that sugar softens the products (counteracting harshness) by preventing some of
the moisture removal and by a direct moderating action on flavor. According to Frazier
and Westhoff (1988) besides adding flavor it also serves as an energy source for nitrite
Brown or white sugars can be used (Rust and Olson, 1973). Mainly sucrose is
used, but glucose can be substituted if a curing time is employed, or no sugar may be
added (Frazier and Westhoff, 1988). Dextrose is often added to cured products. It is
about 75% as sweet as sucrose and more can be added without affecting the flavor. It
also increases the water holding capacity of the products (Pearson and Gillett, 1999).
Com syrup, molasses, honey and other natural sugar substitutes are sometimes used in
place of sugar. Sugar substitutes have been used in bacon cures to prevent excessive
Sodium Erythorbate
The use of ascorbate or isoascorbate (erythorbate) speeds the color reaction and
increases stability of the final product. As a result of the accelerated reaction, they
reduce nitrite levels in the meat product and consequently reduce potential nitrosamine
formation (Lechowich et aI., 1978). Fox et ai. (1967), studied color formation in
frankfurter emulsions and found that the lag phase, when oxygen is scavenged from the
system, was about 300 minutes in an emulsion held at 38°CIl00°F. Under the same
conditions if ascorbate was added to the emulsion, the phase took 38 minutes. The
corresponding times for the color-production phase (the pink hemochrome is formed)
were 8 hours and 25 minutes, respectively (Fox et aI., 1967). The presence of ascorbate
significantly reduced the time necessary to obtain the desired color. Fujimaki et ai.
(1975) used mixtures of myoglobin, sodium nitrite and sodium ascorbate in a buffer
solution as a meat curing model system. These researchers found that there was an
increase in gas production as the amount of ascorbate in the mix was increased.
Furthermore, they found that the gas production proceeded not only in the curing period
but at the cooking stage. When the molar ratio of ascorbate went from 10 to 100, the
decomposition of the added nitrite was remarkably enhanced. They concluded that when
the curing period is 0 day and the concentration of nitrite and ascorbate is relatively high
at the cooking stage, the color development is higher and gas is generated (Fujimaki et
aI., 1975).
Rywotycki (2001) agreed with similar results in previous reported studies. The
sodium ascorbate to the brine. It causes a fast change of nitrites and at the same time
decreases the number of nitro so groups which may react with amines. Rywotycki (2001)
observed that ascorbate dissolved in water and introduced into the meat, caused a decline
When ascorbate was added to brine containing polyphosphates and sodium nitrite the
31.5% and 36.3% increased (DMNA and DENA respectively) showed by the injection of
brine containing polyphosphates and sodium nitrite but no sodium ascorbate (Rywotycki,
2001). The addition of ascorbate as a reducing agent was able to lower the nitrosamine
content in the meat even when phosphates were used in the brine.
Smoke
(Cassens et aI., 1979). Carbonyls, organic acids, phenols, organic bases, alcohols,
hydrocarbons, COz, CO, Oz, Nz and NzO are among the chemical compounds that have
been identified in the curing smoke (Gilbert and Knowles, 1975). According to Frazier
and Westhoff (1988) the addition of smoke has two main purposes: to add desired flavors
and to aid in preservation. In a study by Janky et al. (1975) the effect of smoke
processing on product flavor, tenderness and juiciness was evaluated. They observed a
significant improvement in the sensory panel evaluation scores in all the characteristics of
poultry products manufactured using a solid smoke aerosol treatment. Panelist compared
them to non smoked oven-roasted poultry. The products were rated with a scale of 1 to 5
(1= very unacceptable and 5= very acceptable). The smoked products had higher scores
for flavor (4.36 vs 3.45, respectively) and tenderness (4.38 vs 3.88, respectively) than the
There are some concerns that smoking can influence the formation of
nitrosamines in meat products. Knowles et al. (1974) mentioned that nitrites when
Citing early work of Jordan in 1985, Potthast (1993) found no increase in nitrosamines
even when meat products were made with smoke condensates added to the meat mixture.
In these experiments the result was more reduction in the nitrosamine concentration when
phenol fractions were added to the meat mixtures. In recent research, Rywotycki (2001)
established the possibility that smoke (generated by a glow smoke generator) increased
nitrosamine levels in beef. When raw meat was smoked there was an increase in
conjunction with nitrites to beef ham, the raise was DMNA 80.8% and 80.9% DENA,
respectively.
The use of frozen bellies, followed by a thawing step prior to curing is a common
practice in the production of bacon. The freezing step has been a convenient procedure
for those processing plants were storage periods are necessary prior to curing (Rust and
Olson, 1973). Cassens (1994) mentioned that the use of frozen meat facilitates inventory
Freezing provides an excellent means for storing meat for long periods of time.
But the quality of meat is not improved by frozen storage (Cassens, 1994). Fennema
(1971) established that quality of meat could be affected greatly by reactions such as
gradual deterioration in meat quality during frozen storage (Jeremiah, 1980; Miller et aI.,
1980; Huber and Stadelman, 1970). The conditions under which the meat is frozen have
been reported to playa significant role on the final product. Berne (1994) reported that
the freezing process affects the quality and shelf life of frozen foods. Special attention
should be paid to the speed of freezing. Bjerklie (1996) opinion is that the faster the meat
goes from the fresh into the frozen state, the higher the quality of the meat product.
Freezing Meat
Meat freezes at -2°C (28°F) (Cassens, 1994). The quality and shelf life of frozen
foods are greatly affected by the freezing process by which they are prepared (Berne,
1994). Bjerklie (1996) explained that the speed of freezing is important. As water cools
and turns to ice, it expands. Ice crystals form within the food product. The slower the
transition from liquid to solid, the larger the ice crystals become (Berne, 1994).
Ramsbottom et al. (1949) reported that the size and location of ice crystals in frozen meat
depended, in large measure, on the rate at which the temperature of the meat was dropped
from just above the freezing point (-1.4°C/29.5°F to -1.1°CI30°F for beet) to a
temperature of -3.9°C (25°F). The size and location of the crystals determine the damage
made to the tissue. Calvelo (1981) reported that the shape of the ice in the frozen tissue,
the size of the crystals formed and their distribution among the intra or extracellular
spaces are particularly important in a whole series of macroscopic effects which occur in
frozen meat. These effects are, (a) change in the water holding capacity of muscle, once
thawed; (b) changes in texture; and (c) changes in surface color (Calvelo, 1981).
Ramsbottom et aI. (1949) found that when freezing beef rounds in air blast at -32°C (-
25°F), very small intracellular ice crystals on the surface were formed, but large ice
crystals were produced between fibers, 10.16 cm (4 in) into the muscle. Large crystals
rupture the structure of the cells around them (Bjerklie, 1996). Nusbaum et aI. (1983)
reported that slower freezing rates increased cooking losses while faster freezing rates
decreased cooking losses. These researchers found larger ice crystal size in patties frozen
at -lOoC (14°F) than in patties frozen at faster rates (-80, -30 and -15°C) (-112, -22, 5°F,
respectively). They concluded that larger ice crystals size observed in slower frozen
patties might account for higher cooking losses (Nusbaum et aI., 1983). Ngapo et aI.
(1999) studied the effect of freezing and thawing rate on drip loss of pork. Ngapo et al.
(1999) reported that at faster freezing rates the drip loss obtained was not significantly
different from that obtain from the fresh samples. These results indicate that the freezing
process did not affect drip formation. However, at the slower freezing rates studied, drip
loss was significantly different from drip loss obtained from the fresh sample.
Nusbaum et al. (1983) reported that sensory tenderness scores were lower and
corresponding shear values higher for patties frozen at -lOoC (14°F) than for fresh or
faster frozen patties. They accounted that difference in sensory characteristics to the
Ngapo et al. (1999) found no evidence to suggest that freezing, frozen storage or
Several researchers have studied the effect of freezing on the color of meat.
Ramsbottom et al. (1949) reported that slow-frozen meats are dark, and therefore,
unattractive in appearance. They observed that steaks that were frozen at -29°C (-20°F)
in an air blast approached the fresh color. The steaks that were quick-frozen (-79°C/-
110°F) were much lighter in color. Steaks that were slow-frozen at -7°C (+20°F) were
much darker. Tuma (1971) compared freezing cycles and their effect on beef steaks.
Treatments that were composed of higher freezing temperatures (-26, -29 and -40°C) and
longer times produced darker color steaks after freezing than the fresh pre-frozen steaks.
Rapid short-term systems using -57°C (-70°F) or lower temperatures resulted in brighter
red colors immediately after freezing. Guenther and Hendrickson (1962) reported that
the colder the freezing temperature, the lighter the frozen steak colors. However, these
differences disappeared once the steaks were thawed. The differences in color when the
meat is frozen are caused by differences in size and location of the ice crystals
(Ramsbottom et aI., 1949). According to these researchers, there could be a wide variety
of cycles to freeze meat. Ramsbottom et ai. (1949) recommended that the rate of freezing
should be fast enough so that the color of the lean meat is at least as light as it was before
freezing.
Ramsbottom et ai. (1949) reported that under the same conditions, beef, pork,
lamb and veal freeze at similar rates if their fat content was similar. The total time that
well marbled beef, pork and lamb took to go from 4.4°C/40°F to -17.8°C/OoF was 4.7, 4.7
Freezing Methods
Many methods were used to freeze meats, including freezing in air with low
velocity movement, in high velocity air blast, direct contact with refrigerated plates, and
immersion or spraying with sodium chloride brine or other solutions (Ramsbottom et aI.,
1949). More recently, Berne (1994) reported that there are several methods and a variety
of refrigerants used for cooling and freezing. Mechanical systems are used for products
of relatively low value but produced in high volumes. Mechanical systems use
(HCFCs), or liquid carbon dioxide in a closed loop system. The cryogenic technique
offers a more precise freezing method. Therefore, they are used for higher value and
quality foods. Cryogenic methods are preferred for smaller size food. The cryogenic
system utilizes the cooling capabilities of gases such as nitrogen (N2), carbon dioxide
(C02), and air. Systems combining the advantages of cryogenics and traditional freezing
method, but one advantage is that they can also be used for storage (Bjerklie, 1996).
Frozen Storage
Frozen storage of meat takes place at temperatures ranging from -lODC to -30D C.
Frozen storage can be detrimental in the quality of meat. According to Calvelo (1981)
phenomena that take place during frozen storage are: denaturation of proteins,
that reduction of the latter two depend fundamentally on the use of suitable packing.
The oxidation of lipids is dependent of the access of oxygen and light. If there
were adequate protection, rancidity would not appear until after 15 months of storage at
_20De. Ramsbottom (1947) reported that extensive production of peroxides and free
fatty acids occurred after 60 days of storage with pork held at -3°C. Very little evidence
of rancidity development was noted at -20°C (Ramsbottom, 1947). After seven years of
fat. Ramsbottom (1947) concluded that with proper packaging, meat quality was not
Sublimation or freezer bum occurs when moisture evaporates from the surface of
meat. This takes place as a result of poor packaging were air is allowed in the package.
The outcome is a blanched color and a dehydrated layer on the surface of the meat
(Cassens, 1994).
generally evaluated in terms of the quantity of homogenized tissue that can be extracted
with a 5% solution of NaCl. The formation of ice crystals during freezing concentrates
the remaining solution, which accelerates the denaturation of proteins during storage.
The longer the storage time, the smaller the quantity of proteins removed; thus there is an
increase in denaturation (Calvelo, 1981). Khan et al. (1963) indicated that during frozen
storage chicken muscle proteins became less extractable and the rate of loss of protein
McBride and Richardson (1979) reported that a trained panel found that for all the
sensory attributes (color, texture and flavor) in beef patties, increased storage time
moisture loss and softening of the meat (Bjerklie, 1996). Nichols and MacKintosh
(1952) reported increased fragmentation as a result of thawing and freezing cycles. They
concluded that both intracellular and intercellular ice crystal formation contributed to the
fiber fragmentation. As a result of fiber damage, more drip loss was recorded with each
Thawing involves melting all of the ice crystals in the product. Phenomenas such
as exudate loss, microbiological growth, evaporation loss and deterioration could occur
during thawing (Gonzalez-Sanguinetti et aI., 1985; Marriott et aI., 1980; Beck and
Milone, 1972).
According to Bezanson (1975) from an economic point of view, the drip loss that
meat could be in the order of 5%. This drip has a protein content of about 10%. In
addition to the economic loss, drip loss is related to quality factors such as reduced bind
factor, loss of flavor, color changes and dried out texture. Furthermore, if the lost protein
gets into the drain, it adds to the BOD load of the wastewater system (Bezanson, 1975).
Therefore, it is important that thawing be quick enough to pass through this temperature
range as rapidly as possible (Calvelo, 1981). Thawing should be completed at low
temperatures (above 2°C), relatively high heat transfer coefficients and relative humidity
Thawing may be done in cold or warm air. A flow or movement of air will speed
the process. Thawing by immersion in water give an improved thawing rate due to the
direct contact. Nearly all the meat thawed in processing plants is either air-thawed or
water-thawed (Bezanson, 1975). The use of microwave equipment also has appeared to
standard practice for hams, bellies, briskets, poultry, tuna, and shrimp. It is much faster
and consumes less space than air thawing. However, regulations concerning water usage
Some of the disadvantages of the thawing methods include the time that is
consumed during the thawing process, the large amounts of floor space required, and the
opportunity for undesirable changes to occur to the product (Bezanson, 1975; Satchell
Breed
According to Cisneros et al. (1996) the genetic potential of commercial pigs has
changed dramatically over recent years, particularly in terms of lean growth rates.
A study by Tess et al. (1986) reported that at three different ages pigs from a
Hampshire X Large White (CX) cross (industry selection based on several traits) were
heaviest and pigs from Beltsville Highfat (HF) (selected for increased backfat thickness)
stock were the lightest. The weights were taken on weeks 10, 17 and 24; ex pigs
weighed 17.59, 56.10 and 100.59 kg, respectively. Meanwhile HF pigs weighed 13.65,
40.05 and 71.28 kg for 10, 17 and 24 week respectively. Other stocks of pigs was also
compared in this study, a Lowfat (LF) (selected for decrease backfat thickness) Duroc-
Yorkshire composite. The LF stock had intermediate weights but more similar to HF
(14.91,42.55 and 81.79 kg for 10, 17 and 24 week, respectively). The HF pigs contained
the most fat at all three ages, with ex pigs being intermediate (at 10, 17 and 24: HF 2.29,
13.12 and 33.58 kg; LF 1.49, 8.32 and 15.39 kg; ex 1.68, 11.64 and 28.25, respectively).
At 24 weeks of age, HF pigs were 13% lighter than LF pigs, while ex pigs were 23%
heavier than LF pigs. For non-fat components, ex pigs were heaviest and HF lightest.
The fat-free mass weights at week 24 were 37.09, 55.62 and 71.03 kg for HF, LF and
ex, respectively. The HF pigs exceeded LF for chemical fat (47 vs 31 %) and were lower
for protein (11 vs 14%). The ex pigs contained 28% fat and 15% protein. They
suggested that selection for rapid lean growth has an effect upon growth and production
efficiency. Selection against backfat should decrease total empty body fat, but may also
Friesen et al. (1994) reported that medium-lean pigs had greater backfat thickness
and carcass lipid accumulation than high-lean pigs. High lean pigs had an average
backfat thickness of 3.3 cm. Pigs from the medium lean genotype, had an average
backfat thickness of 3.5 cm. Lipid accretion was 272.3 vs 226.4 gld for high-lean and
medium-lean pigs, respectively. Friesen et al. (1994) reported that crude protein
accretion was greater in high-lean pigs than in medium-lean pigs (106.3 vs 89.7 gld,
respectively). The differences observed in growth performance up to 104 kg remained
evident at 127 kg for both genotypes, although a main effect of genotype was not
detected in the finishing phase (from 104 to 127 kg). Their data suggest that high-lean
pigs did not produce carcasses superior to those of medium-lean pigs, but that the desired
Chiba et al. (2002) reported that pigs selected for lean growth efficiency had less
10th rib backfat (22.3 vs 31.5 mm), larger longissimus muscle area (31.6 vs 29.2 cm2)
and greater lean accretion rate (273.7 vs 241.6 gld) when compared with control pigs.
In a study by Gu et al. (1992) it was observed that growth coefficients for lean,
fat, bone and skin were not significantly different among genotypes. The genotypes
compared in this research were (1) Hampshire x Hampshire Duroc; (2) Synthetic terminal
sire line; (3) Hampshire Duroc x Landrace [Yorkshire Duroc]; (4) Landrace x Yorkshire
Duroc; (5) Yorkshire x Landrace. The coefficients for lean growth in genotypes 1 to 5
were: .814, .792, .807, .754, .836, respectively. For weight fat growth the average
coefficient was 1.584, which indicates that fat tissue grew at a higher rate than did total
carcass. Since the coefficients within each trait did not differ among genotypes, it was
suggested that growth of these traits, relative to carcass weight, were basically the same
in all genotypes. This study is in disagreement with Chiba et al. (2002), Friesen et al.
(1994) and Tess et al. (1986). Gu et al. (1992) suggested that the reason for a lack of
genotype effect might be due to insufficient numbers and (or) smaller genotype
divergence.
Selection by genotype has shown positive results in reducing fat and increasing
lean in pig carcasses. Furthermore, selected genotypes tend to grow faster and more
efficiently. The selection for leaner pigs has an effect on bacon production.
Wenther (1999) reported that genetic line had the most dramatic effects on bacon
processing parameters and proximate composition. Lines with the higher backfat content
and the smallest loin eye area had lower pumping yields. They had the highest percent of
Sex
In a study with pigs representing two crossbred genotypes and five slaughter
weights barrows grew faster than gilts (Cisneros et aI., 1996). They found an interaction
between genotype and sex for days on test and daily gain. This interaction resulted from
no difference in growth rate (daily gain = .811 kg for barrow and .809 kg for gilts)
the case of a two crossbred genotype (a breeding company hybrid) barrows grew faster
than gilts (daily gain = .930 and .817 kg for barrows and gilts respectively). Cisneros et
aI. (1996) results suggest that sex differences in growth rate may vary with genotype. In
relation to the percentage of weight loss from the carcass in the first 24 h postmortem
(carcass cooler shrink), Cisneros et aI. (1996) reported that the shrink was higher for gilts
than for barrows (3.08 and 2.55% loss for gilts and barrows respectively). They also
found a genotype x sex interaction for this trait. The breeding company hybrid barrows
lost less weight than gilts (2.09 and 3.27%, respectively). However there was no
difference between cooler shrink between barrows and gilts for the three crossbred
genotype (3.0 and 2.99%, respectively) (Cisneros et aI., 1996). Sexes differed for the
wholesale cut weight and percentage of ham and carcass trim and the wholesale
percentage of shoulder and picnic. But sex did not have an effect on the weight of the
belly (7.70 and 7.54 kg for barrows and gilts respectively). Furthermore there was no
difference for the belly on the wholesale cut percentage of barrows and gilts (16.29 and
16.05%, respectively). When curing yields were analyzed, they reported no differences
between sexes for belly yield (98.36 and 97.48 % for barrow and gilts respectively)
(Cisneros et aI., 1996). They found no differences between the sexes in some muscle
quality traits measured on the longissimus thoracis et 1umborum muscle. The traits with
no differences were drip loss, cooking loss, tenderness, juiciness, off flavor Wamer-
Bratzler shear and muscle composition (moisture and fat). However barrows had higher
scores for subjective color, firmness and marbling (2.75, 2.85 and 3.07 respectively for
Larzu1 et aI., (1997) in a study with Large White purebreds, reported that gilts
grew more slowly than barrows (average daily gain of 791 gld for gilts and 836 gld for
barrows). Gilts also had leaner carcasses than barrows. Gilts lean percentage was 52.4
compare to 49.8% of barrows. An average backfat thickness for gilts (29.8 mm) and
barrows (31.6 mm) was also reported. Larzul et aI. (1997) tested for meat quality traits
on the longissimus muscle. They reported that sex had an effect on the intramuscular fat
content (1.22 and 1.38% for gilts and barrows respectively). When color was measured,
Friesen et aI. (1994) reported that dressing percent was lower in barrows than in
gilts. An average backfat thickness and 10th rib fat depth were less in gilts than in
barrows (2.65 vs. 3.02 and 2.43 vs. 3.15 cm, respectively) for carcasses of 104 kg pigs.
Similarly results were reported for carcasses of 127 kg pigs. Gilts had less average
backfat thickness (3.28 vs 3.60 cm, respectively) and 10th rib fat depth (3.05 vs. 3.73 cm,
respectively) than barrows. In both of the pigs weights groups, carcasses from gilts had a
higher percentage of crude protein (104 kg group: 17.17 vs. 15.77%, respectively; 127 kg
group: 15.83 vs. 14.40%, respectively). A difference in ash and lipid content was only
found among sexes in the 104 kg pig group. Ash values were 2.85 vs. 2.46%, for gilts
and barrows respectively. Values of27.60 and 31.05% of lipid percent were reported for
Wenther (1999) determined that the sex of the animals significantly affected
slicing yields and total yield. Barrows had higher slicing yield than gilts.
Pringle and Williams (2001) compared pork carcasses varying in fatness and
muscling. They reported that side weight was greater in the fattest backfat category than
in either of the leaner categories. In the lower 10th rib backfat depth «2.03 cm) category,
the side weight mean was 37.1 kg. The intermediate category (2.03-2.54 cm) had a 38.5
kg mean. Finally, the thick backfat category (>2.54 cm) was significantly higher than the
other two side weights with a 43.8 kg mean. However, they found that the belly
thickness and belly pocket fat thickness were only different between carcasses in the
leanness category. Belly thickness in the leanness category was 1.96 cm in comparison
with 2.51 and 2.77 cm for the intermediate and fattest categories respectively. For belly
pocket fat depth, the values were 2.49, 3.02 and 2.90 cm for leanness, intermediate and
fattest respectively (Pringle and Williams, 2001). The yield of trimmed belly increased
as backfat category increased (11.67, 12.37 and 13.21 for leanness, intermediate and
fattest respectively).
Wenther (1999) results suggest that the belly composition had an effect on the
processing parameters. Leaner bellies exhibited higher pumping yields. Bellies with
higher fat content had better smokehouse yields and total yields. These results are in
agreement with previous research that reported that the yield of the green and cured belly
is inversely related to lean content (Kemp et aI., 1969; Freeden et aI., 1975a,b; Stiffler et
aI., 1975; Jabaay et aI., 1976; McMillan et aI., 1977). In relation to proximal composition
Wenther (1999) reported that as the fat in the belly increased moisture, protein and ash
Fat
Fats are organic chemical compounds made up fatty acids and glycerol. The fatty
acids consist of chains of carbon atoms with different amounts of hydrogen and oxygen
attached. The most common fatty acids found in food have 16 or 18 carbon atoms in a
straight chain. Other are shorter or longer, ranging from 4 to 26 carbon atoms. In food
and in the body, fatty acids are usually combined with molecules of glycerol, forming
glycerides. One, two, or three fatty acid molecules can be attached to one molecule of
glycerol. Most natural fats, from animal and vegetable sources, are triglycerides
(Schneeman, 1986).
designations refer to the number of double bonds present in the fatty acid chain of the
(polyunsaturated) double bonds. In the case of saturated fatty acids every carbon atom
Consumer demand for cheap lean pork has led to a remarkable improvement of
the muscle to fat ratio. This has been achieved by breeding and a reduction in the age
and weight of slaughter animals (KUhne, 1984). Unfortunately not only the quality of
pork meat but also the properties of fat have change (Cameron and Enser, 1991;
Kempster et aI., 1986; Wood et aI., 1985; Wood and Enser, 1982).
Lack of consistency of adipose tissues is one of the main problems the manufacturers of
meat products have to face (Davenel et aI., 1999). An example of unacceptable physical
characteristics is usually called "Soft fat". According to KUhne (1984) fat of that
consistency cannot be used successfully for the processing of high quality meat products.
Soft fat is characterized by defects such as insufficient drying, oily appearance, rancidity
development and lack of cohesiveness between muscle and adipose tissue on cutting
(Bailey et aI., 1973). According to Ellis and McKeith (1999) soft fat is of major concern
to the meat processor because it can cause significant problems during cutting, grinding
and slicing operations and can result in lower processing yields and reduced value.
tissue consistency, while other components including collagen and water do not have
significant effects (Enser et aI., 1984; Whittington et aI., 1986). Fat consistency has been
attribute to the ratio of monounsaturated to saturated fatty acids (Lea et aI., 1970), the
percentage of all the unsaturated fatty acids (Elliot and Bowland, 1969) and the
percentage of stearic acid (Wood et aI., 1978). Maw et aI. (2003) established that the
physical characteristics of fat samples are related to their fatty acid composition. In their
trial, hardness of the fat samples was related to fatty acid composition with a very high
significance. His results show that increased softness was associated with increased
linoleic and a-linolenic acid, and a decreased in stearic and palmitic acid percentages.
Research has also been done relating carcass characteristics and fatty acid profile.
Wood et aI. (1989) compared the composition of the backfat in pigs with different
backfat depths and showed that leaner pigs had a higher proportion of polyunsaturated
fatty acids (Linoleic and Linolenic). Other studies have linked lean pigs with higher
content of unsaturated fatty acids. Chant et aI. (1976) evaluated the overall fatty acid
composition of bacon. Bellies were classified SUbjectively based on leanness into five
categories. Leanest bellies were scored as 1 and fatter bellies were 5. They reported that
the saturated acids (palmitic and stearic) increased as leanness of bacon decreased,
whereas linoleic acid increased with bacon leanness. Total saturated fatty acids
decreased with leanness of bacon. Their results suggested that leaner bacon had softer,
more unsaturated fat with greater susceptibility to rancidity development (Chant et aI.,
1976). It was also reported that leaner bellies absorb more salt than fatter ones.
packaged rindless bacon slices where soft fat resulted in loss of definition of the
Geri (1984) reported differences between castrated males and intact females in the
cellularity of adipose tissue. In gilts the number of adipocytes in subcutaneous fat was
significantly lower than in barrows from 40 to 200 kg live weight. In the perirenal
deposit adipocytes were significantly larger in barrows than in gilts. There were sexual
differences for total lipid concentration and for water content of the m. logissimus dorsi.
There was lower proportion of water and higher proportion of lipid in barrow muscle
made at the same anatomical position between pigs receiving a similar diet. Fatty acid
composition varies with fatness of the carcass and thus may be related to sex and
castration in various breeds. The fatty acid composition of pig fat is more unsaturated
than cattle and sheep. While the level of unsaturation varies between breeds (Wood,
1973), differences in the fatty acid composition of subcutaneous fat from boars, gilts and
castrated males has been shown by many authors (Koch, 1968a; Newel and Bowland,
1972; Smithard et aI., 1980). At 90 kg live weight, more than 60% of the total fatty acids
are unsaturated in all three groups although the unsaturation was slightly higher in entire
males. The high level of unsaturation was mainly due to the high concentration of oleic
acid (44%) and linoleic acid (10-12%) (Smithard et aI., 1980). Koch et ai. (1968b) found
that backfat from barrows contained more palmitic and stearic acids and less linoleic acid
Allen (1984) established that there is an effect of sex on the relationship between
fat composition and carcass composition. He suggested that this relationship between fat
composition and carcass composition indicates that very lean carcasses of boars, castrates
or gilts will be more likely to suffer from "floppiness" (soft fat, in which firmness and
lean and fat carcasses and separate them, then it could be that they can supply the bacon
A total of S78 bellies from barrow and gilt of six different breeds (Berkshire,
Chester White, Duroc, Landrace, Poland China and Yorkshire) were utilized in the study.
Bacon Manufacture
Hogs were slaughtered at Geo, Hormel Inc., Austin, MN. Each carcass was
fabricated at Geneva Meats, Inc., Geneva, MN. One belly from each pig was cut to
1997). Fresh bellies were shipped to the Loeffel Meat Laboratory at the University of
Lincoln, Lincoln, NE. The bellies were randomly divided into two treatments (Fresh or
Frozen). Bellies for the frozen treatment were frozen and stored -ISoC (SOP) at UNL
Meat Laboratory for processing at a later date. Bellies in the fresh treatment were held in
refrigerated (3°C/37OP) storage for curing on the ISth day following slaughter.
Defrosting
stainless steel thawing tank, immersed in cold, air agitated water to allow complete
thawing. Fresh bellies received similar water treatment. Once thawed, bellies were
checked and, trimmed for skin and other fabrication defects. After pre-curing assessment,
31
these bellies were weighed. Prior to pumping, a 7 mm slice of fat from the anterior edge
of the belly was removed. The sample was label with the belly ID number. It was
After collecting the fat sample, the belly dimensions were taken as follows:
Pickle Production
Pickle was prepared fresh each day of bacon production. A batch of 40 bellies
required 90 kg (sufficient for pumping operation) of pickle. Ingredients for the pickle
were weighed in advance. Table 1 lists the formulation for the pickle solution. The
ingredients were mixed using a Rotostat Model 80XP63SS (Admix, Inc., Londonderry,
NH). The Rotostat was set at 2500 rpm, the speed reduced to 100 rpm before addition of
the liquid smoke, to avoid any production of foam. Ingredients were added one at a
time in the following order: water, phosphate, salt, cure, sugar, erythorbate and liquid
smoke. The pickle was transfer to. the Townsend (Model 1450, Townsend Engineering,
Pickle Injection
The bellies were pumped to a target percent of green weight (12.0% pick-up), fat
side down on the Townsend multi-needle bacon injection pump. Green weight and post-
Following injection, bellies were hung on a bacon comb at the posterior end,
approximately 2.54 cm (1 in) from end. Combed bellies were placed on smokehouse
32
trucks. The bellies had a resting time of 1 hour before the thennal process to facilitate
cure equalization.
Thermal Processing
Bellies were thennally processed in and Alkar single truck smokehouse unit
(Alkar, Inc., Lodi, WI). The thennal schedule for the smokehouse is shown in Table 2.
Following showering and drying, cooked bellies were chilled overnight in a 3°C/37<>P
cooler.
ChillinglTempering
Following the overnight chilling (3°C/37<>P), the individual bellies were weighed
to detennine smokehouse yield. Bellies were vacuum packaged and stored for subsequent
Commercial Pressing
The vacuum packaged bacon slabs were stacked into pallets. The pallets were
transferred to a commercial bacon plant (MPS, Omaha, NE). The slabs were tempered at
-4°C/24<>P for 24 hrs prior to pressing. The bellies were removed one at a time from the
pallets and packaging removed. The tags were removed for pressing and the temperature
was recorded (Omega Eng. Inc., Stamford, CT). Bellies were pressed (Anco Model
1411, Cherryburrell, Lousville, KY). They were retagged with the original tag and
stacked into new pallets. Pallets were transported back to UNL Meat Laboratory. Pallets
Slicing
Slabs of bacon were sliced one at a time. The slabs weight and temperature
(Omega Eng. Inc., Stamford, CT) were recorded. The slabs were sliced in a Grote Bacon
33
Slicer (Model 613, Grote Company, Columbus, OR) starting from the posterior end. The
bacon slicer was set to slice 10 slices/inch (10slices/2.54 cm). The following information
was recorded for each individual slab: slicing date, slab identification number, slab
weight, slab length, slab width, weight of bacon ends and pieces (incomplete slices) and
the number of slices in the slab. Slab length corresponded to the measure of the center of
the belly from anterior to posterior. Slab width was the measured at the center of the
After slicing, incomplete slices from both the posterior and anterior ends of the
slab were removed. The slices that were affected by the comb (posterior to comb mark
and including combing marks) were also removed. End pieces and incomplete slices
from posterior and anterior ends of the belly were weighed and recorded separately. The
bacon slices were reassembled in original order. Tags were inserted every 20 slices to
allow easy counting of the total number of slices in the slab. The slab was divided in 5
zones, each with an even number of slices. The 5 zones were assigned the letters A, B,
C, D and E (each representing the next 20% of slices) indicating their location from
posterior to anterior.
Each of the five locations were sampled. The first two slices of each zone were
evaluated for shattering. Both slices were wrapped in white polyethylene coated, meat
wrapping paper (Loxol freezer paper, James River Corp., Parchment, MI) and stored in a
cooler (3°C/37°F) for subsequent visioning analysis and proximate analysis. The next
34
five slices in each zone were assigned for cooking. Slices were vacuum packaged and
stored in a cooler (3°C/37<>P). The next five slices were also vacuum packaged and kept
The samples for fatty acid analysis were collected before pickle injection. A thick
strip of fresh belly adipose tissue was removed from the anterior edge of the fresh belly.
The sample was vacuum packaged and stored in a -4°C (24<>P) freezer. They were
removed from the freezer and the fatty acid composition of the adipose tissue was
determined by gas chromatography. Two samples (0.5-1 g) from the adipose tissue of
each fresh belly strip were analyzed. Triglycerol fatty acids were hydrolyzed by
saponification and then methylated to form fatty acid methyl ester (FAME). The methyl
Metcalfe et a!. (1966). The gas chromatograph was equipped with an automatic injector
(HP 7673) and a flame ionization detector (HP-5890 Series II, Hewlett-Packard
Company, Avondale, PA). The methyl esters were separated on a 30 m long x .25 mm
internal diameter x .20 f.1m film thickness Supelco model SP2330 capillary column
(Supelco, Bellafonte, PA) utilizing helium gas as a carrier. The injector, oven and
detector temperatures were set at 270, 190 and 300°C respectively. The individual
methyl esters were identified by their retention time. Peak areas were integrated and
Company, Avondale, PA). The results were expressed as a percentage of the total area
for all the peaks analyzed. Procedures and analysis conditions are outlined in greater
and 68D-Check Prep) were used to determine retention times for specific fatty acids,
under GLC conditions utilized. The retention times of known FAME were compared to
those from belly samples and an in house standard was developed. The standard was
used to identify peaks for the rest of the samples and calibrate GLC equipment. The fatty
acids studied were: myristic (CI4:0), palmitic (CI6:0), palmitoleic (CI6:1), stearic
(CI8:0), oleic (CI8:1), linoleic (CI8:2), linolenic (CI8:3), and ll-eicosenoatic (C20:1).
These fatty acids were selected, since they account for more than 95% of the total amount
present IDavanel et al., 1999; Houben and Krol, 1984). Other fatty acids, although
present in small quantities, were not considered as important to our objectives, thus, were
Shattering
Shatter marks were defined as breaks/shatters of the fat portion of the slice that
occurred perpendicular to the length of the slice (Salas-Perez, 2002). Shatter marks did
not include the natural separation of fat tissue or the separation between fat and lean
tissue.
In order to obtain information about the variation within the bacon slab as well as
geographical location of the shatter marks within the bacon slab, a grid within the sliced
bacon slab was made. The grid consisted of dividing the bacon slice into four quadrants
(1 st, 2nd , 3rd , or 4th), within the five locations (A, B, C, D, or E) (see Appendix 2). Each
quadrant represented 1/4 of the distance from the dorsal side to the ventral side of the
bacon slice; whereas each zone represents 1/5 of the distance from the anterior to
posterior ends of the sliced bacon slab. This procedure was established to accommodate
36
variable width and length of the bellies. A diagram of this division is presented in
Appendix 2.
The shatter marks were classified into the following five categories depending on
their length: 1-10 mtn, 11-20 mtn, 21-30 mtn, 31-40 mtn, >41 mtn.
After recording weights for the sliced bacon slab, the trays were placed on a table
with the posterior end closer to the person doing the evaluation, leaving the ventral edge
(quadrant 4) to the left side and dorsal to the right side (quadrant 1). The first two slices
from each zone were evaluated for shatter marks. Starting from the first complete slice,
the slices were separated carefully. Then the slice was rolled over the finger, from
quadrant 1 to 4, for a more evident appearance of the shatter marks. Any shatter marks
were measured and recorded according to their corresponding quadrant and size category.
After shatter marks were recorded for a given slice, the slices were name and wrap
appropriately and kept in a cooler until subsequent analysis. The same two slices were
Composition-Proximate Analysis
Proximate analysis was perfonned for each location of the belly. Ether
1990; section 27.006a, p159. Moisture and Ash were determined using a
thennogravimetric analyzer (TGA-60 1, Leco Corp., St. Joseph, MI). Protein was
PA) double belt conveyor cooker. Cook yield, dimensional changes in length and width,
and distortion scores were determined and recorded for these samples. Preliminary trials
were conducted to determine the correct settings for the double belt cooker to yield the
color described as "golden brown" but not crisp as described by Ross (1999). Cooked
yield was required to be less than 40% of raw weight to comply with USDAIFSIS
regulations (USDA, 1996). The target cook yield was 37-39% based on cooked color. As
a result of these trials, the temperature of the top and bottom platens was set at
204.4°C/400°F with a clearance between platens of 0.33 cm during the cooking process.
The preheat temperature was also set at 204.4°C/400Gp. The belt speed of the cooker was
adjusted to control cooking time. After trials and with the objective to reach the cooked
yield and color required, the time was set to 40 seconds from the time the sample entry
Cooked samples were evaluated for percent change in length (at the midpoint of
the slice), width (at the midpoint of the slice), and weight from the raw weight. A
distortion scale developed by Ross (1999) was used to evaluate the change in bacon
shape (Figure 1). The scale has five distortion scores. A distortion score of one
The samples were removed from the cooler, the vacuum package was opened and
the slices from location A were removed and separated. Their length and width was
measured individually. The slices were placed on the paper plate in order from 1 to 5 and
weighed. The plate marked "A" was then taken to the belt cooker for cooking. The same
procedure was follow for locations B, C, D and E in alphabetical order. The slices were
placed on the cooker belt in order from 1 to 5 with spaces of two inches (5 cm). Gaps of
eight inches (20.3 cm) were left between locations to prevent any mixing of slices from
different locations. As the slices exited the belt cooker, they were placed in numerical
order on a clean and identified paper plate. The slices were allowed to cool for 30
seconds before further analysis. All slices from location A were evaluated individually
with the distortion scale. Following length and width measures of the five slices were
Statistical Analysis
the belly from a pig with a given genetic line and sex and with one storage treatment
(fresh and frozen). The sample size for each one of the treatments was different.
Treatments were allocated in a complete randomize array. Data were analyzed using the
Mixed Model procedure of SAS with a level of significance of P<0.05 (SAS, 1999).
Means were separated using the Fisher's least significant difference test.
The effects of genetic line and sex on backfat, loin eye area and carcass length
were evaluated using slaughter weight (ranged from 92.98 kg/205 lb - 138.34kg/305Ib)
as a covariant.
39
The effects of genetic line, sex and storage treatment on processing parameters
were also analyzed for main effects and two-way interactions. Smokehouse yield was
analyzed using percent pump (ranged from 6-16%) as a covariant. The slicing yield was
analyzed using the temperature (-6°C 1210P - O°C/32OP) of the belly before slicing as a
covariant. When the zone or quadrants were considerate, the analysis splits, or spilt-split
a. Phosphate was added to the cold water (~oC/40op) agitated until phosphates are
dissolved.
b. With continued agitation, the remainder of the non-meat ingredients were added in the
following order: salt, sugar, nitrite, natural liquid smoke flavor as required and lastly the
ascorbate.
Wenther, 1999.
41
Ross, 1999.
43
References
Allen, P. 1984. Sex differences in fat composition. In Fat Quality in Lean Pigs.,
Workshop in The CEC Programme, Brussels, Belgium 145-160.
AOAC, 1990. Official Methods of Analysis, 14th ed. Association of Official Analytical
Chemists, Washington, DC.
Bailey, C., C.L. Cutting, M.B. Enser, and D.W. Rhodes. 1973. The influence of
slaughter weight on the stability of pork sides in frozen storage. J. of Sci. Food
and Agri. 24:1299-1304.
Berne, S. 1994. New techniques yield a tastier freeze. Prepared Foods. June. 109-115.
Chicago, IL.
Bezanson, A. 1975. Thawing and tempering frozen meat. 11th Meat Industry Research
Conference. Amer. Meat Inst. Found. Chicago, IL.
Bjerklie, S. 1996. PC QC, Maintaining frozen portion quality control. Meat and
Poultry. February. 16-18.
Brooks, J., R.B. Haines, T. Moran, and J. Pace. 1940. The function of nitrate, nitrite and
bacteria in the curing of bacon and hams. Food Invest. Special Report No. 49.
His Majesty's Stationery Office, London. England, UK.
Cameron, N.D. and M. Enser. 1991. Fatty acid composition of lipid in longissimus dorsi
muscle of Duroc and British Landrace pigs and its relationship with eating
quality. Meat Sci. 29:295-307.
Cassens, R.G., M.L. Greaser, T. Ito, and M. Lee. 1979. Reactions of nitrite in meat.
Food Tech. 33(7):46-57.
Cassens, R.G. 1994. Meat Preservation. Preventing Losses and Assuring Safety. Food
and Nutrition Press, Inc. Trumbull, CT.
Cassens, R.G. 1995. Use of sodium nitrite in cured meats today. Food Tech. 49(7):72-
115.
Chant, J.L.Jr., D.M. Stiffler, D.M. Kinsman and A.W. Kotula. 1976. Chemical and
sensory aspects of commercial bacons. J. Anim. Sci. 43:989-993.
44
Chiba, L.I., D.L. Kuhlers, L.T. Frobish, S.B. Jungst, E.J. Huff-Lonergan, S.M Lonergan,
and K.A. Cummins. 2002. Effect of dietary restrictions on growth perfonnance
and carcass quality of pigs selected for lean growth efficiency. Livest. Prod. Sci.
74:93-102.
Cho, I.C. and LJ. Bratzler. 1970. Effect of sodium nitrite on flavor of cured pork. J.
Food Sci. 35:668-670.
Cisneros, F., M. Ellis, F.K. McKeith, J. McCaw, and R.L. Fernando. 1996. Influence of
slaughter weight on growth and carcass characterisitics, commercial cutting and
curing yields, and meat quality of barrows and gilts from two genotypes. J. Anim.
Sci. 74:925-933.
Davenel, A., A. Riaublanc, P. Marchal, and G. Gandemer. 1999. Quality of pig adipose
tissue: relationship between solid fat content and lipid composition. Meat Sci.
51:73-79.
Desmoulin, B. 1984. Sex effects on the deposition and quality of fat in lean pigs. In Fat
Quality in Lean Pigs., Workshop in The CEC Programme, Brussels, Belgium
130-144.
Dziezak, J.D. 1990. Phosphates improve many foods. Food Tech. 44(4): 80-92.
Ellinger, R.H. 1972. Phosphates as Food Ingredients. CRC Press. Cleveland, OH.
Elliot, J.I., and J.B. Bowland. 1969. Correlation of melting point with the sum of the
unsaturated fatty acids in samples of porcine depot fat. Can. J. of Anim. Sci.
49:397-398.
Ellis, M., and McKeith, F. 1999. Nutritional influences on pork quality. In Pork Facts.
National Pork Producers Council. Des Moines, IA.
Enser, M. 1984. The relationship between the composition and consistency of pig
backfat. In Fat Quality in Lean Pigs., Workshop in The CEC Programme,
Brussels, Belgium 53-57.
Enser, M., E. Dransfield, P.D. Jolley, R.C.D. Jones, and M. Leedman. 1984. The
composition and consistency of pig backfat as it affects the quality of vacuum-
packaged rindless bacon rashers. J. of Sci. Food and Agri. 35:1230-1240.
Fennema. O.R. 1971. Rates of chemical deterioration in frozen foods. Meat Industry
Research Conf. Proceedings. Amer. Meat Inst. Found. Chicago, IL. 35-44.
Frazier, W.C. and D.C. Westhoff. 1988. Food Microbiology. Fourth ed. McGraw-Hill.
New York, NY.
45
Fredeen, H.T., A.H. Martin, D.S. Harbison, and J.G. McAndrews. 1975a. Criteria of
belly bacon desirability. I. Fat-lean ratio of the pork belly in relation to backfat,
carcass weight and specific gravity. Can. J. Anim. Sci. 55:641-648.
Fredeen, H.T., A.H. Martin, D.S. Harbison, and J.G. McAndrews. 1975b. Criteria of
belly bacon desirability. III. Within-belly variance in chemical and physical
characteristics. Can. J. Anim. Sci. 55:661-672.
Friesen, K.G., J.L. Nelssen, J.A. Unruh, R.D. Goodband, and M.D. Tokach. 1994.
Effects of the interrelationship between genotype, sex, and dietary lysine on
growth performance and carcass composition in finishing pigs fed to either 104 or
127 kilograms. J. Anim. Sci. 72:946:954.
Fox, J.B.Jr., W.E. Townsend, S.A. Ackerman and C.E. Swift. 1967. Cured color
development during frankfurter processing. Food Tech. 21(3):386-392.
Fujimaki, M., M. Emi, and A. Okitani. 1975. Fate of nitrite in meat-curing model
systems composed of myoglobin, nitrite and ascorbate. Agr. BioI. Chern. 39:
371-377.
Geri, G. 1984. Genetic and sex effects on fat deposition and quality. In Fat Quality in
Lean Pigs., Workshop in The CEC Programme, Brussels, Belgium 126-129.
Gilbert, J. and M.E. Knowles. 1975. The chemistry of smoked foods: a review. J. Food
Techno!. 10:245-261.
Gonzalez-Sanguinetti, S., M.C~ Anon, and A. Calvelo. 1985. Effect of thawing rate on
the exudates production of frozen beef. 1. Food Sci. 50:697-706.
Greenwood, D.A. 1940. Some further studies on the destruction of sodium nitrite by
heating. AMI. Proceedings. Amer. Meat Inst. Found. Washington, DC.
Gu, Y., A.P. Schinckel, and T.G. Martin. 1992. Growth, development and carcass
composition in five genotypes of swine. J. Anim. Sci. 70: 1719-1 729.
Guenther, J.J. and R.L. Hendrickson. 1962. Temperatures, methods used in freezing
determine tenderness, color of meat. Quick Froz. Foods. 25 (11):115-117.
Hoagland R. 1908. The action of saltpeter upon the color of meat. In 25 th Ann. Rep.
Bureau of Animal Industry. U.S. Dept. of Agriculture, Washington, DC. 301-
314.
46
Hoagland R. 1914. Coloring matter of raw and cooked salted meats. J. Agri. Res.
3:211-225.
Huber, C.S., and W.J. Stadelman. 1970. Effect of freezing rate and freeze drying on the
soluble proteins of muscle. 1. Chicken muscle. J. Food Sci. 35:229-232.
Jabaay, R.W., J.C. Forrest, E.F. Aberle, H.V. Courtenay and M.D. Judge. 1976. Bacon
quality criteria and associated carcass traits. J. Food Sci. 41:431-437.
Janky, D.M., J.L. Oblinger, and J.A. Koburger. 1975., Sensory and microbiological
evaluation of smoked Cornish game hens. Poultry Sci. 54 (6):1942-1945.
Jeremiah, L.E. 1980. Effect of frozen storage and protective wrap upon the cooking
losses, palatability and rancidity of fresh and cured pork products. J. Food Sci.
45:187-196.
Kemp, J.D., W.G. Moody and J.D. Fox. 1969. Effect of fatness and fresh pork quality
on yield and quality of bacon and yield of ham. J. Anim. Sci. 28:612-619.
Kempster, A.J., A.W. Dilworth, D.G. Evans and K.D. Fisher. 1986. The effects of fat
thickness and sex on pig meat quality with special reference to the problems
associated with overleaness. 1. Butcher and consumer panel results. Anim.
Produc. 43:517-533.
Kerr, R.H., C.T.N. Marsh, W.F. Schroeder and E.A. Boyer. 1926. The use of sodium
nitrite in the curing of meat. J. Agri. Res. 33:541-551.
Khan, A.W., L. van den Berg, and C.P. Lentz. 1963. Effects of frozen storage on
chicken muscle proteins. J. Food Sci. 28:425-430.
Knowles, M.E., J. Gilbert, and DJ. McWeeny. 1974. Nitrosation of phenols in smoked
bacon. Nature. 249: 672-673.
Koch, D.E., A.F. Parr, and R.A. Merkel. 1968a. Fatty acid composition of the inner and
outer layers of porcine backfat as affected by energy level, sex and sire. J. Food
Sci. 33:176-180.
Koch, D.E., A.M. Pearson, W.T. Magee, J.A. Hoefer and B.S. Schweigert. 1968b.
Effect of diet on the fatty acid composition of pork fat. J. Anim. Sci. 27:360-365.
KUhne, D. 1984. Fat quality in lean pigs. In Fat Quality in Lean Pigs., Workshop in The
CEC Programme, Brussels, Belgium 36-42.
Lea, C.H., P.A.T. Swododa, and D.P. Gatherum. 1970. A chemical study of soft fat in
cross-bred pigs. J. Agri. Sci. Cambridge. 74:279-289.
Lechowich, R.V., W.L. Brown, R.H. Deibel, and 1.1. Somers. 1978. The role of nitrite in
the production of canned cured meat products. Food Tech. 32:45-58.
Lewis, W.L., R.S. Vose, and C.D.Jr. Lowry. 1925. Ind. and Eng. Chern. 17:1243.
Mandigo, R.W. 1999. Bacon processing procedures. AMI Cured Meat and
Manufacturing Short Course. Iowa State Univ. Ames, IA.
Marriott, N.G., R.A. Garcia, M.E. Kurland, and D.R. Lee. 1980. Appearance and
microbial quality of thawed retail cuts of beef, pork and lamb. J. Food Prot.
43:185-189.
Maw, S.J., V.R. Fowler, M. Hamilton, and A.M. Petchey. 2003. Physical characteristics
of pig fat and their relation to fatty acid composition. Meat Sci. 63:185-190.
McBride, R.L. and K.C. Richardson. 1979. The time-temperature tolerance of frozen
foods: sensory methods of assessment. J. Food Tech. 14:57-67.
McMillan, K.W., M.D. Judge, J.C. Forest, V.L. Anderson and E.D. Aberle. 1977.
Relationships of bacon quality to pork carcass traits. J. Anim. Sci. 45: 1023-1 031.
Metcalfe L.D., Schmitz, A.A. and Pelka, J.R. 1966. Rapid preparation of fatty acid
esters from lipids for gas chromatographic analysis. Anal. Chern. 38:514-?15.
Miller, A.J., S.A. Ackerman and S.A. Palumbo. 1980. Effects of frozen storage on
functionality of meat for processing. 1. Food Sci. 45:1466-1471.
NAMP. 1997. North American Meat Processors Association Meat Buyer's Guide.
North American Meat Processors Assoc. Reston, VA.
Ngapo, T.M., I.H. Baber, J. Reynolds and R.F. Mawson. 1999. Freezing and thawing
rate effects on drip loss from samples of pork. Meat Sci. 53:149-158.
Nichols, J.B., and D.L. MacKintosh. 1952. Structural changes occurring in muscle
tissue during repeated freezing and thawing. Food Tech. 6: 170-174.
Nusbaum, R.P., J.G. Sebranek, D.G. Topel and R.L. Rust. 1983. Structural and
palatability relationships in frozen ground beef patties as a function of freezing
treatments and product fonnulation. Meat Sci. 8:135-146
Pearson, A.M. and T.A. Gillet. 1999. Processed Meats. Third Ed. Aspen Publishers,
Inc. Gaitherburg, MD.
Pivnick, H., H.W. Barnett, H.R. Nordin, and L.J. Rubin. 1969. Factors affecting the
safety of canned, cured, shelf-stable luncheon meat inoculated with Clostridium.
botulinum. Can, Inst. Food Technol. J. 2:141-148.
Pringle, T.D. and S.E. Williams. 2001. Carcass traits, cut yields, and compositional end
points in high-lean-yielding pork carcasses: effect of 10th rib backfat and loin eye
area. J. Anim. Sci. 79:115-121.
Ramsbottom, J.M., Goeser, P.A. and Strandine, EJ. 1949. The effect of different
factors on the freezing rate of meats. Proc. of the Conference on Research.
Amer. Meat Inst. Found. Chicago,IL. 96-109.
Ramsbottom, J.M. 1947. Freezer storage effect on fresh meat quality. Ref. Eng. 53:19-
23
Romans, J.R., K.W. Jones, W.J. Costello, C.W. Carlson, and P.T. Ziegler. 1985. The
Meat We Eat. 13 Ed. Interstate Publishers, Inc. Danvile,IL.
Ross, R. 1999. Precooked bacon manufactured by microwave and double belt conveyor
processing systems. M.S. Thesis. University of Nebraska-Lincoln, Lincoln, NE.
Rust, R.E. and D.G. Olson. 1973. Meat curing principles and modem practice. Koch
Supplies, Inc. Kansas City, MO.
Rust, R.E. and D.G. Olson. 1983. Processing Workshop. Phosphates. Meat Industry.
August. Mill Valley, CA.
Salas-Perez, PJ. 2002. Differences among breeds, diets, slaughter weights and sex type
in the shattering of bacon slices. M.S. Thesis. University of Nebraska-Lincoln,
Lincoln, NE.
SAS. 1999. SAS/STAT User Guide (Version 8.0). SAS Institute Inc. Cary, NC.
Satchell, F.E. and D.M. Doty. 1951. High frequency dielectric heating for defrosting
frozen pork bellies. Bulletin No. ·12. Amer. Meat Inst. Found. Chicago,IL.
Schneeman, B.O. 1986. Fats in the diet: why and where? Food Tech. 115-120.
Sebranek, J.G. 1979. Advances in the technology of nitrite use and consideration of
alternatives. Food Tech. 33(7):58-93.
Smithard, R. R., W. C. Smith, and M. Ellis. 1980. A note on the fatty acid composition
of backfat from boars in comparison with gilts and barrows carcass
characteristics. Anim. Prod. 31 :217-219.
Stiffler, D.M., J.L.Jr. Chant, D.M. Kinsman and A.W. Kotula. 1975. Indices of leanness
in commercial bacons .. J. Anim. Sci. 41: 1611-1617
Tess, M.W., G.E. Dickerson, J.A. Nienaber and C.L. Ferrell. 1986. Growth,
development and body composition in three stocks of swine. J. Anim. Sci.
62:968-979.
Tuma, H.J. 1971 Processing technology for freezing retail meta cuts. Meat Industry
Research Conf. Proceedings. Amer. Meat Inst. Found. Chicago, IL. 53-62.
USDA. 1996. Food Standards and Labeling Policy Book. Food Safety and Inspection
Service; Washington, DC.
Wasserman, A.E. and F. Talley. 1972. The effect of sodium nitrite on the flavor of
frankfurters. J. Food Sci. 37:536-538.
Wenther, J.B. 1999. The effect of genetic line, diet, sex, slaughter weight and type of
bacon on bacon processing parameters. M.S. Thesis. University of Nebraska-
Lincoln, Lincoln, NE.
50
Whittington, F.M., N.J. Prescott, J.D. Wood, and M.Enser. 1986. The effect of dietary
linoleic-acid on the firmness of backfat in pigs of 85 kg live weight. J. Sci. of
Food and Agri. 37:753-761.
Widdus, R., and F.F. Busta. 1982. Antibotulinal alternatives to the current use of nitrite
in foods. Food Tech. 36(12):105-106.
Wood, J. D. 1973. The fatty acid composition of backfat from Pietrain and Large White
pigs. Anim. Prod. 17:281-285.
Wood, J.D., M.B. Enser, H.J. MacFie, W.C. Smith, J.P. Chadwick, M. Ellis, and R.
Laird. 1978. Fatty acid composition of backfat in large white pigs selected for
low backfat thickness. Meat Sci. 2:289-300.
Wood J.D. and M. Enser. 1982. Comparison of boars and castrates for bacon
production. 2. Composition of muscle and subcutaneous fat, and changes in side
weight during curing. Anim. Prod. 35:65-74.
Wood, J.D., R.C. Jones, J.A. Bayntun and E. Dransfield. 1985. Backfat quality in boars
and barrows at 90 kg live weight. Anim. Prod. 40:481-487.
Wood, J.D., M. Enser, F.M. Whittington, C.B. Moncrieff, and A.J. Kempster. 1989.
Backfat composition in pigs: differences between fat thickness groups and sexes.
Livestock Prod. Sci. 22:351-362.
Yagmur, C., and A. Ozer. 2001. Residual nitrate and nitrite levels in different meat
products. Analysis of dry fermented sausage, salami and sausage produced and
marketed in Turkey. Fleischwirtschaft International. 3:47-50.
51
1. Authors Robles and Mandigo are currently affiliated with the University of
Nebraska-Lincoln, Lincoln, NE 68503-0908
52
Abstract
The objective of the study was to evaluate the impact of fresh vs. frozen bellies on
bacon parameters. Physical and chemical differences among bellies from pigs of
different genetic lines and sexes were evaluated. A population of 578 pork bellies from
barrows and gilts of Berkshire, Chester White, Duroc, Landrace, Poland China and
Yorkshire lines were processed to sliced bacon and evaluated. Bellies were divided into
two storage treatments: fresh (stored at refrigeration temperature, 3°C) or frozen (held at-
15°C). Raw bellies were processed into bacon under the same conditions in order to
weight), slicing yield (weight of the slices) and total yield (sliced weight/green weight).
Frozen stored bellies were able to pickup 0.73% more pickle (P<O.OI) than bellies stored
fresh. Fresh bellies had bacon slicing yields 1.74 percent better (P<O.05) than bacon
made from frozen bellies. The results for total yield were not different (P>0.05), 82.61 %
and 82.10% for fresh and frozen, respectively. Genetic line and sex had the most
dramatic effects on the bacon processing parameters. Higher yields (smokehouse, slicing
and total) were associated with sex and lines containing higher percentage of fat in the
bellies. Overall, fresh and frozen bellies had similar characteristics when processing
bacon. Although statistical difference among treatments existed, the practical advantages
of these values may be slight. Considering that bellies were only stored for 15 d, it is not
Introduction
demand for sliced bacon. From 1999 to 2000 supermarket bacon sales increased 45
percent. Restaurant sales grew 7 percent every year from 1997 to 1999. Due to this high
demand, the number of U.S. pork bellies in cold storage hit a historic low in April 2001
(National Pork Board, 2001). The live hog market varies due to cyclical and seasonal
differences. As a result, the price is determined by supply and demand. Belly prices also
The variation in price, supply and demand of bellies can drive manufacturing
practices of the bacon industry. The bellies are often stored frozen at times when the
prices are low. Bacon is then made as the market demands it.
Freezing provides an excellent means for storing meat for long periods of time,
but the quality of meat is not improved by frozen storage. Ice crystals are formed within
the food product during freezing. The size and location of the crystals will determine the
extent of damage to the tissue. Damage can include changes in the water holding
capacity, texture, and surface color (Berne, 1994; Calvelo, 1981; Ramsbottom et al.,
1949).
loss, oxidative deterioration reactions of lipids and protein denaturation could take place
(Gonzalez-Sanguinetti et al., 1985; Marriott et al., 1980; Bezanson, 1975). There is little
data that reports the effect of the damage on frozen bellies when processing to sliced
bacon.
54
today's pigs are bred and fed to be leaner and with less fat, compared to the pig of 1950s.
Today's model is leaner, with 50 percent less fat. Lean-to-fat ratio is one of the biggest
factors in a shopper's selection of bacon. West et a1. (1973) reported that consumers
would purchase leaner bacon as a premium product. Consequently, packers prefer lean
pork and producers are raising leaner, heavier muscled pigs to satisfy these demands
The emphasis on lean hogs has changed bacon. Leaner bellies produce higher
percentage of slices, while fatter bellies produce a lower percentage of "premium" bacon
slices (Fredeen et aI., 1975; Stiffler et aI., 1975). However, with the leaner bellies, bacon
processing characteristics such as smokehouse yield and total yield are inversely related
to carcass characteristics desired by the producer, packer and consumer (Wenther, 1999).
smokehouse yields, slicing yields and total yields when manufacturing bacon from
refrigerated and frozen stored bellies. The study also evaluated differences among
barrows and gilts from genetic lines consisting of Berkshire, Chester White, Duroc,
A total of 578 bellies from barrow and gilt bellies of six different genetic lines
(Berkshire, Chester White, Duroc, Landrace, Poland China and Yorkshire) were utilized
in the study.
55
Hogs were commercially slaughtered at Geo. Honnel Inc., Austin, MN. Each
carcass was fabricated at Geneva Meats, Inc., Geneva, MN. One belly from each pig was
cut to specifications ofNAMP #409 - PORK BELLY, SKINLESS (NAMP, 1997), and
Upon receipt, the bellies were divided into two storage treatments: fresh or
frozen. Bellies (275) in the frozen treatment were frozen in a still atmosphere freezer at
UNL Meat Laboratory and held at -15°C (5°F). Bellies (303) for fresh treatment were
Defrosting
Fresh and frozen bellies were then fully thawed in water. The bellies were placed in a
stainless steel thawing tank and immersed in cold (S4°C/400F), air-agitated water to
assure that all frost was removed (usually 30 min). Thawed bellies were re-inspected,
trimmed for skin and other fabrication defects, and weighed. Prior to pumping, a 7 mm
slice of fat from the anterior edge of the belly was removed. The fat sample was label
with the belly ID number. It was vacuum packaged and frozen until further analysis.
56
After collecting the fat sample, the belly dimensions were taken as follows:
Pickle Production
Fresh pickle was prepared the day of bacon production (Table 1). A batch of 40
bellies required 90 kg of pickle. The water and dry ingredients were mixed using a
Rotostat Model 80XP63SS (Admix, Inc., Londonderry, NH). The Rotostat was set at
2500 rpm, the speed reduced to 100 rpm before addition of the liquid smoke, to avoid any
production of foam. Ingredients were added one at a time in the following order: water
(~oC/40<>P), phosphate, salt, 6.25% sodium nitrite cure, sugar, erythorbate and liquid
smoke. The pickle was transfer to the pickle tank of the Townsend® injector (Model
Pickle Injection
The bellies were pumped to a'target percent of green weight (12.0%), fat side
down using a multi-needle bacon injection pump (Model 1450, Townsend Engineering,
Des Moines, IA). Green weight and post-pump weight were recorded.
Following injection, each belly was hung on a smokehouse truck using a bacon
comb approximately 2.54 cm (1 in) from the posterior end. The bellies were allowed to
Thermal Processing
Bellies were thennally processed in and Alkar single truck smokehouse unit
(Alkar, Inc., Lodi, WI). The thennal schedule for the smokehouse is shown in Table 2.
Following showering and drying, cooked bellies were chilled overnight in a 3°C/37tlp
cooler.
Chilling/Tempering
detennine smokehouse yield. Bellies were vacuum packaged and stored for subsequent
Commercial Pressing
The vacuum packaged bacon slabs were transferred to a commercial bacon plant
(MPS, Omaha, NE) to be pressed. The slabs were held at -4°C/24tlp 24 h prior to
pressing. Temperature was recorded (Omega Engineering Inc., Stamford, CT) prior to
pressIng. The slabs were pressed (Anco Model 1411, Cherryburrell, Lousville, Ky),
stacked on new pallets, then transported and stored at -4°C/24tlp at UNL Meat
Laboratory.
Slicing
Cooked belly weight and temperature (Omega Engineering Inc., Stamford, CT)
were recorded prior to slicing. The slab was fed in to the bacon slicer (Model 613, Grote
Company, Columbus, OR) starting at the posterior end. The slicer was set to slice 10
58
slices/inch (10 slices/2.54 cm). The following information was recorded for each
individual slab: slab identification number, slab weight, slab length, slab width, weight of
bacon ends and pieces (incomplete slices) and the number of slices in the slab. Slab
length corresponded to the measure of the center of the belly from anterior to posterior.
Slab width was the measured at the center of the belly from dorsal to ventral.
Statistical Analysis
the belly from a pig with a given genetic line and sex and with one storage treatment.
The sample size for each one of the treatments was different. Treatments were allocated
in a complete randomize array. Data were analyzed using the Mixed Model procedure of
SAS with a level of significance. of P<O.05 (SAS, 1999). Means were separated using the
Fisher's least significant difference test. The. effects of genetic line and sex on backfat,
loin eye area and carcass length were evaluated using slaughter weight (ranged from
The effects of genetic line, sex and storage treatment (fresh and frozen) on
processing parameters were also analyzed for main effects and two-way interactions.
Smokehouse yield was analyzed using percent pump (ranged from 6-16%) as a covariant.
The slicing yield was analyzed using the temperature (-6°C /21 OF - O°C/32OP) of the belly
Frozen bellies retained 0.73% more brine (P<0.01) than fresh bellies (Table 3).
These results differ from the expected. It was thought that during storage proteins in
frozen bellies would have been damaged. Therefore, fresh bellies would have had the
Ngapo et al. (1999a) studying the cross-sectional areas structure of fresh, frozen,
and thawed pork, observed cavities. The areas of the cavities in the frozen samples were
approximately ten times larger than similar areas in fresh and thawed samples. Indicating
frozen meat had suffered greater damage on its ultrastructure due to freezing.
Damage can possibly be reflected in different ways. Among those, drip loss is
meat, when thawed, has higher drip loss than fresh meat (Ngapo et al., 1999b; Bjerklie,
It is suspected that the frozen bellies in this study may have lost more water
during thawing than fresh bellies. It is also believed that muscle cells in frozen bellies
had suffered distortion during storage. As a result, frozen bellies could have cells that
were dryer and more flexible. The bellies were able to hold more brine after pumping.
The difference in pumping percent was small. On a small scale production it may
not be of practical significance (10.49 and 11.22% fresh and frozen, respectively).
However, considering a large scale of bacon production, this small improvement in yield
To evaluate the thermal processing properties of fresh and frozen bellies, the
effect of pump retention was included in the statistical model as a covariate. Fresh and
frozen bellies had similar yields after heat process in the smokehouse (Table 3).
Fresh bellies had 1.74% higher (P<0.05) slicing yields than frozen bellies (Table
3). A possible cause of the differences in slicing characteristics could be given for the
A study by Rahelic' and Puac' (1985) observed that the ice crystals formed during
freezing meat damaged muscle fibers. They reported that meat frozen at -22°C was
damaged more severe than bacon frozen at -10°C. The distribution of ice crystals and the
intercellularly and at -22°C the formation was inter and intracellular in their work. The
crystals formed between fibers will generate pressure that will separate fibers, while
crystals formed intracellularly will exert pressure in the opposite direction. As a result of
the opposing pressure, tearing of the fibers will be greater (Rahelic' and Puac', 1985).
those reported by Rahelic' and Puac' (1985). It is possible that a phenomena similar to
the one that Rahelic' and Puac' (1985) found at -22°C could have occurred. Thus, the
difference in yields could be given by the less damaged cell structure of fresh bellies. A
better integrity of the bacon slab could have a lower tendency to break or shatter during
Total yield measures the overall processing efficiency of the belly from its green
weight through to sliced weight and sliced bacon ready to be commercialized. Fresh
bellies had higher total yields (82.61 %), although, they are not significantly different than
61
the yields of frozen bellies (82.10%). This difference is very small not statistically
different, and might not be of commercial significance. It might suggest that that frozen
bellies have lost quality and thus have lower processing characteristics than fresh bellies.
There was a significant effect of genetic line on pumping percent. Bellies from
Poland China pigs had greater (P<0.05) pickle retention (11.78%) than Yorkshire,
Berkshire, Landrace and Duroc lines (10.78, 10.74, 10.66 and 10.33% respectively). But
pump percent in bellies from Poland China was similar to Chester White (10.87%). The
ability of the belly to retain pickle has been correlated with the fat and lean content of the
carcass. In this study, it was found that Poland China and Chester White barrows had
higher (3.40, 3.17 cm) backfat thickness than Landrace, Yorkshire and Duroc barrows
(2.61, 2.49 and 2.41 cm, respectively). They also had more backfat than gilts from all
genetic lines. Gilts of these two lines (Poland and Chester White) had more backfat when
compared to the leaner (Landrace, Yorkshire and Duroc) lines of the study (Table 4).
Loin eye area measured from Poland China (37.09 cm2) and Chester White (38.25 cm2)
carcasses was smaller than those of Duroc and Yorkshire (41.80 and 40.64 cm2,
respectively). Overall, our results show that fatter lines had better cure or pickle pick up
during pumping. These results are different than work reported by Wenther (1999) who
reported that bellies from light muscle and fat lines had the lowest percent pump. Same
and Bratzler (1959) also reported that bellies from fatter pork carcasses had lower
amounts of salt during processing. In this study, the lowest pumping yield was observed
in Duroc pigs, which had the largest loin eye area (41.80 cm2), and a heavier muscled
62
carcass. Pigs in this study were much leaner than those in the Saffle and Bratzler (1959)
study. This could possible explain the differences in these results. It was difficult to
compare pigs in this study to those in Wenther (1999) because they reported loin eye area
and backfat thickness as rates of change in the live animal and not as final values.
Genetic line had an effect (P<O.OI) on smokehouse yields (Table 3). Bellies from
Berkshire pigs had the highest yield (98.20%) (P<0.05) as compared to all other lines.
Previous studies report that yield is positively related to fat content in the belly. Table 4
illustrates that within sex, Berkshire pigs had the highest backfat content (3.40 and 2.66
cm for barrows and gilts, respectively). Landrace hogs had the lowest smokehouse
yields, 2.29% lower than the highest yield achieved with the Berkshires (Table 3).
Comparing backfat of Landrace barrows and gilts (2.61 and 2.28) with the rest of the
breeds were similar to the leaner Yorkshire (2.41 barrows and 2.05 gilts) and Duroc pigs
(2.49, 1.95 barrow and gilts, respectively). These results suggest that fatter bellies have
better yield during cooking in the smokehouse. The results are in agreement with
previous studies that reported that fatter bellies tended to loose less weight during the
heating process (Wenther, 1999; Kemp et aI., 1969; Carpenter et aI., 1963; Saffle and
Bratzler, 1959).
Total yield provides an estimate of the losses during bacon manufacturing from
the green weight of the belly to the total number of marketable slices of bacon. The
bellies with better perfonnance in the study were those that belong to carcasses with
thicker backfat (Tables 3 and 5). Bellies from Berkshire pigs had the highest yield with
85.90%. Poland China followed, with similar yields (83.44%). Chester White hogs had
yields (83.14%) that were similar to Poland China, but they were significantly different
63
than Berkshire pigs (Table 3). Results suggest that fat content is positively related to
total yield. Results are in agreement with Wenther (1999), Carpenter et aI., (1963) and
The length and width of the bellies (Table 6) were affected by genetic line
(P<0.05). Genetic lines that had longer bellies tended to have smaller widths, Chester
White pigs had the longest bellies (52.07 cm) and the smallest width (25.37 cm).
Yorkshire bellies resulted in the widest width (26.44 cm). Their length was among the
shortest of the study (50.39 cm). The differences in dimension of the bellies did not seem
to have an effect on slicing yield or total yield. Lines with longer bellies did not
Sex effect
The sex of the pig did not have an impact on the percent pump injection solution
retained by the belly (Table 3). The data also indicated no significant effect of sex in
Sex had an effect on slicing yield parameter (Table 3). Barrows had
significantly higher (P<O.05) slicing yields than gilts (86.86 and 85.36%, respectively).
Results in this study are consistent with prior reports that indicate barrows tend to be
fatter than gilts (Wenther, 1999; Cisneros et al., 1996; Lonergan et aI., 1992). They also
had smaller (P<0.01) loin eye area (36.25 cm2) than gilts (40.77 cm2). This suggests
barrow carcasses had less muscle content than gilts (Table 5). Although due to
interactions between sex and genetic line, main effects cannot be analyzed. A view of the
means for back fat thickness of barrows and gilts for each line are given in Table 4.
64
From these results it can be seen that within a genetic line, barrows had higher backfat
Sex had a significant effect (P<0.05) on total yield (Table 3). Barrows had
higher (P<0.05) yields than gilts. Since barrows had higher slicing yields, it was
expected that they would have higher total yields. This can also be attributed to the fact
The effect of sex on belly dimension was also analyzed. Sex had no effect
on the length, but had an effect on the width of the belly. Gilts had wider bellies than
barrows by an average of 0.48 cm (Table 6). This is in agreement with Freeden (1980)
who found that bellies from females were greater in the dorso-ventral dimension.
However, the difference in belly width did not seem to have had an effect on the any of
Conclusions
Bacon has become an important part of the increased consumer preferences due to
Commercial bacon processors consider sliced yield, smokehouse yield and total yield
These results suggest that the use of fresh or frozen bellies in the manufacture of
bacon would have very similar processing characteristics and yields. In the present study
bellies were in frozen storage for only 15 days prior to processing. Perhaps longer
could have greater impact in bacon processing operations, where volumes of production
Genetic line and sex had great impact on processing yields of bacon. The present
study agrees with previous reports that indicate that the yield of the green and cured
bellies are inversely related to the lean content of the pork carcass. Thus the potential
economic benefits associated with improved lean content in live hogs may come with a
reduction in yield when manufactwing bacon (Wenther, 1999; Freeden, 1980; Jabaay et
Reference
Berne, S. 1994. New techniques yield a tastier freeze. Prepared Foods. June. 109-115.
Chicago, IL.
Bezanson, A. 1975. Thawing and tempering frozen meat. 11 th Meat Industry Research
Conference. Amer. Meat Inst. Found. Chicago,IL.
Bjerklie, S. 1996. PC QC, Maintaining frozen portion quality control. Meat and
Poultry. February. 16-18.
Gonzalez-Sanguinetti, S., M.C. Afton, and A. Calvelo. 1985. Effect of thawing rate on
the exudates production of frozen beef. J. Food Sci. 50:697-706.
Kemp, J. D., W. G. Moody and J. D. Fox. 1969. Effect of fatness and fresh pork quality
on yield and quality of bacon and yield of ham. J. Anim. Sci. 28:612-619.
Marriott, N.G.,R.A. Garcia, M.E. Kurland, and D.R. Lee. 1980. Appearance and
microbial quality of thawed retail cuts of beef, pork and lamb. J. Food Prot.
43:185-189.
NAMP. 1997. North American Meat Processors Association Meat Buyer's Guide.
North American Meat Processors Assoc. Reston, VA.
National Pork Board. 2002. Pork Facts 2001/2002. Des Moines, IA.
http://www.porkboard.org.
Rahelic', S. and S. Puac'. 1985. Structure of beef longissimus dorsi muscle frozen at
various temperatures: part 1- histological changes in muscle frozen at -10, -22,
-33, -78, -115 and -196°C. Meat Sci. 14:63-72.
Saffle, R. L. and L. J. Bratzler. 1959. The effect of fatness on some processing and
palatability characteristics of pork carcasses. Food Techno!. 13:236-239.
SAS. 1999. SAS/STAT User Guide (Version 8.0). SAS Institute Inc. Cary, NC.
Wenther, J. B. 1999. The effect of genetic line, diet, sex, slaughter weight and type of
bacon on bacon processing parameters. M.S. Thesis. University of Nebraska-
Lincoln~ Lincoln, NE.
West, R. L., W. L. Sawyer, G. C. Smith and Z. L. Carpenter. 1973. What does consumer
expect in bacon? Meat Processing. April, p.62-68.
68
a. Phosphate was added to the cold water (~OC/40OP) agitated until phosphates are
dissolved.
b. With continued agitation, the remainder of the non-meat ingredients were added in the
following order: salt, sugar, nitrite, natural liquid smoke flavor as required and lastly the
ascorbate.
Wenther, 1999.
69
Table 3. Least Square Means ± S.E. of Percent Pump, Smokehouse Yield, Slicing
Yield and Total Yield as affected by Line, Sex and Storage Treatment.
abed
Means within the same column and within a main effect with similar superscripts are
not significantly different (P>O.OS) .
• P values from Analysis of Variance for each main effect within a variable.
•• Frozen stands for bellies that were frozen at -15°C prior to processing.
Percent pump = (weight of the pumped belly / initial belly weight) x 100.
f
Smokehouse yield = (weight of the cooked belly / initial belly weight) x 100.
g
Slicing yield = (weight of the center of the sliced bacon slab / weight bacon slab) x 100.
h
Total yield = (weight of the center of the sliced bacon slab / weight of the belly) x 100.
71
Table 4. Least Square Means of the Genetic Line x Sex interaction for
Backfat (cm).
Table S. Least Square Means for Loin eye area (LEA), Carcass length (CL) and
Backfat (BF).
Table 6. Least Square Means ± S.E. of the Belly Length and Belly Width as affected
by Line, Sex and Storage Treatment.
The effect of breed, sex and storage treatment on the fatty acid profile of bellies and
bacon quality
1. Authors Robles and Mandigo are currently affiliated with the University of
Nebraska-Lincoln, Lincoln, NE 68503-0908
7S
Abstract
The objective of the study was to evaluate the effect of genetic line, sex and
storage treatment (fresh vs. frozen) of bellies on bacon quality. A population ofS78 pork
bellies from barrows and gilts of Berkshire, Chester White, Duroc, Landrace, Poland
China and Yorkshire lines were processed to cooked bacon. Bellies were divided into
two storage treatments: fresh (stored at refrigeration temperature, 3°C) or frozen (held at-
ISoC for IS d). Before treatments were processed into bacon, fat samples were collected
from each belly in order to establish a fatty acid profile. The bacon slabs were sliced and
cooked in order to evaluate differences in fatty acid content, shattering, cooking yield,
saturated and unsaturated fatty acid profile of the bellies due to treatment. Fresh bellies
revealed values of 61.69 and 37.44 percent, saturated (SFA) and unsaturated fatty acids
(UFA), respectively, compared to 62.24 and 36.86 percent of SFA and UFA for frozen
bellies. Bellies' from pigs of leaner genetic lines had higher (P<O.OI) percentages of
unsaturated fatty acids. The effect of sex was observed by differences in the content of
SFA and UFA. Gilts had a profile with a 1.08 percent higher unsaturated fatty acid
profile when compared to barrows (P<O.OS). Shattering of the bacon slice was influenced
by location. Slices from the center portion (location B, C and D) of the frozen bellies had
a higher total shatter mark length (P<O.OS). Correlation suggests that 22 percent of the
variation in the total number of shatter marks in the belly is explained by the total content
of saturated fatty acid. Genetic lines with higher fat content (Berkshire and Poland
China) had lower (P<O.OS) cooking yields (31.90 and 31.14 percent). Fresh and frozen
bellies had differences on distortion (P<O.OS) of the cooked bacon slices with least square
76
means of 2.35 and 2.43, respectively. Distortion was affected by an interaction (P<O.05)
among line and sex. Overall, fresh and frozen bellies resulted in bacon with similar
quality. Although statistical differences existed, the practical advantages of these values
may be slight.
77
Introduction
It is well accepted that consumers in prefer to purchase meat with lower levels of
fat. This trend has certainly extended to cured bacon. One of the major reasons is the
possible association between high levels of saturated fat and heart disease (Wood and
Fisher, 1990). There is a strong body of evidence which maintains that fat in meat
contributes to eating quality and that reducing fat to low levels will reduce satisfaction at
a time when consumer interest in the quality of food in general is increasing (Wood and
Fisher, 1990).
Recent studies have shown that consumers are demanding more flavorful foods.
According to the Food Marketing Institute (2000), the factors that consumers feel are
importance, were: (1) taste; (2) nutrition; (3) food safety; (4) cost; (5) storability. It
appears that despite continued health consciousness, people seem to be enjoying more
Although bacon could hardly be called a new food category, it is far from been
mature. In fact, bacon is enjoying a renaissance in retail and foodservice markets, driven
packaging (Salvage, 2002). Bacon is the fastest growing foodservice pork product, at a 5
to 7 percent annual growth (Salvage, 2002; National Pork Board, 2001). Ranging from
sandwich toppings to garnishing salads, bacon has delivered versatility to the menu, as a
side dish or ingredient for all meals. Bacon has added flavor to a wide array of dishes
and is now often considered a condiment. The most unique new products are the
78
microwaveable products and salad toppings that are actual bacon slices instead of dried
bits.
Quality defects represent lost revenue opportunities for the pork industry and
potentially the entire pork marketing chain. Estimates of 10% of the total value of the
live animal are cost related to quality nonconformity (Cannon et aI., 1996). Problems
associated with bacon include separation of fat from lean, color of fat, color of lean,
increasingly important to both producer and packer (Chant et aI., 1976). Packers are
improving their processing techniques (selecting, trimming and grading) with the goal of
Understanding the role of fat on bacon success starts with an understanding of the
variables that impact the amount and quality of the lean and fat found in the bacon.
Previous studies have analyzed the effect of genetics, sex and nutrition on the quality of
backfat and longissimus muscle, as well as the palatability of longissimus muscle and
palatability of bacon (Bryhni et aI., 2002; Cameron et aI., 2000; Mazhar et aI., 1990;
Carpenter et aI., 1963). However, there have been few attempts to evaluate the quality of
fat in the belly and its effect on bacon slicing and cooking. The fatty acid profile of the
fat found in the belly will potentially impact the processing of the belly, the slicing of the
bacon and the cooking and eating quality of the bacon. Many factors can influence the
fat content as well as the fatty acid profile including the breed and sex of the pigs as well
The purpose of this study was to identify and study the relationship of chemical
characteristics of bacon to determine how they influence the structure of the bacon slice
A total of 578 bellies from barrow and gilt bellies of six different genetic lines
(Berkshire, Chester White, Duroc, Landrace, Poland China and Yorkshire) were utilized
in the study.
Bacon Fabrication
(NAMP, 1997) during fabrication at Geneva Meats, Inc., Geneva, MN. Fresh bellies
Lincoln, NE. They were received and divided into two treatments (fresh or frozen).
Bellies for the frozen treatment were frozen and stored -15°C (5°F) 15 d at UNL Meat
Laboratory for processing at a later date. Bellies in the fresh treatment were held in
refrigerated (3°C/37OP) storage for curing on the 15th day following slaughter.
complete thawing. Fresh bellies received similar water treatment. When completely
thawed, bellies were checked and trimmed for skin and other fabrication defects. Prior to
pumping, a 7 mm slice of the fresh belly was removed from the anterior edge of the belly,
80
to obtain a fatty acid profile analyses of the belly. The sample was label with the belly
Pickle was prepared fresh each day of bacon production (Table 1). The
ingredients were mixed using a Rotostat Model 80XP63SS (Admix, Inc., Londonderry,
NH). The bellies were pumped to a target percent of green weight (12.0% pick-up), fat
side down on the Townsend multi-needle bacon injection pump (Model 1450, Townsend
the posterior end of the belly, approximately 2.54 cm (1 in) from the end. The bellies had
a holding time of 1 hour before the thermal process to facilitate cure equalization. Bellies
were thermally processed in and Alkar single truck smokehouse unit (Alkar, Inc., Lodi,
WI). The thermal schedule for the smokehouse is shown in Table 2. Following
showering and drying, cooked bellies were chilled overnight in a 3°C/37°F cooler. The
slabs were vacuum packaged and stored for subsequent processing in a tempering cooler
at -4.4°C/24OP. The bacon was transferred to a commercial plant (MPS, Omaha, NE) to
be pressed. The slabs were tempered at -4°C/24°F for 24 hrs prior to pressing. Bacon
slabs were pressed using an Anco machine (Model 1411, Cherryburrell, Lousville, KY).
Pressed bellies were transported back to UNL Meat Laboratory and stored at -4°C/24OP
Stamford, CT) were recorded. The slabs were sliced in a Grote Bacon Slicer (Model 613,
Grote Company, Columbus, OR) starting from the posterior end. The bacon slicer was
set to slice 10 slices/inch (10slices/2.54 cm). The following information was recorded for
81
each individual slab: slicing date, slab identification number, slab weight, slab length,
slab width, weight of bacon ends and pieces (incomplete slices) and the number of
complete slices in the slab. Slab length corresponded to the measure of the center of the
belly from anterior to posterior. Slab width was the measured at the center of the belly
After slicing, incomplete slices from both the posterior and anterior ends of the
slab were removed. The slices that were affected by the comb (posterior to comb mark
and including combing marks) were also removed. End pieces and incomplete slices
from posterior and anterior ends of the belly were weighed and recorded separately. The
bacon slices were reassembled in original order. Tags were inserted every 20 slices to
allow easy counting of the total number of slices in the slab. The slab was divided in S
locations, each with an equal number of slices, by the S locations, which were assigned
the letters A, B, C, D and E (each representing the next 20% of slices); A at the most
Each of the five locations was sampled. The first two slices of each location were
evaluated for shattering. Both slices were wrapped in white polyethylene coated, meat
wrapping paper (Loxol freezer paper, James River Corp., Parchment, MI) and stored in a
cooler (3°C/37°F) for subsequent visioning analysis and proximate analysis. The next
five slices in each zone were assigned for cooking; they were vacuum packaged and
stored in a cooler (3°C/37OP). The next five slices were also vacuum packaged and kept
The samples for fatty acid analysis were collected before pickle injection. A thick
strip of fresh belly adipose tissue was removed from the anterior edge of the fresh belly.
The sample was vacuum packaged and stored in a -lSoC (SOP) freezer. They were
removed from the freezer and the fatty acid ~omposition of the adipose tissue was
determined by gas chromatography. Two samples (0.5-1 g) from the adipose tissue of
each fresh belly strip were analyzed. Triglycerol fatty acids were hydrolyzed by
saponification and then methylated to form fatty acid methyl esters (FAME). The methyl
Metcalfe et al. (1966). The gas chromatograph was equipped with an automatic injector
(HP 7673) and a flame ionization detector (HP-5890 Series II, Hewlett-Packard
Company, Avondale, PA). The methyl esters were separated on a 30 m long x .25 mm
internal diameter x .20 J.lm film thickness Supelco model SP2330 capillary column
(Supelco, Bellafonte, PA) utilizing helium gas as a carrier. The injector, oven, and
detector temperatures were set at 270, 190, and 300°C respectively. The individual
methyl esters were identified by their retention time. Peak areas were integrated and
Company, Avondale, PA). The results were expressed as a percentage of the total area
for all the peaks analyzed. Commercial standard solutions of FAME (GLC-standard No.
84 and 68D-Check Prep) were used to determine retention times for specific fatty acids,
under GLC conditions utilized. The retention times of known FAME were compared to
those from belly samples and an in-house standard was developed. The standard was
used to identify peaks for the rest of the samples and calibrate GLC equipment. The fatty
83
acids studied were: myristic (CI4:0), palmitic (CI6:0), palmitoleic (CI6:1), stearic
(CI8:0), oleic (CI8:1), linoleic (C18:2), linolenic (C18:3), and ll-eicosenoatic (C20:1).
These fatty acids were selected, since they account for more than 95% of the total amount
present (Davanel et al., 1999; Houben and Krol, 1984). Other fatty acids, although
present in small quantities, were not considered as important to our objectives, thus, were
not identified.
Shattering
Shatter marks were defined as breaks/shatters of the fat portion of the slice that
occurred perpendicular to the length of the slice (Salas-Perez, 2002). Shatter marks did
not include the natural separation of fat tissue or the separation between fat and lean
tissue.
In order to obtain infonnation about the variation within the bacon slab as well as
geographical location of the shatter marks within the bacon slab, a grid within the sliced
bacon slab was made. The grid consisted of dividing the bacon slice into four quadrants
(1 st, 2nd , 3rd , or 4th), within the five locations (A, B, C, D, or E) (see Appendix 2). Each
quadrant represented 1/4 of the distance from the dorsal side to the ventral side of the
bacon slice; whereas each location represents 1/5 of the distance from the anterior to
posterior ends of the sliced bacon slab. This procedure was established to accommodate
The shatter marks were classified into the following five categories depending on
their length: 1-10 rom, 11-20 rom, 21-30 rom, 31-40 rom, >40 rom.
After recording weights for the sliced bacon slab, the trays were placed on a table
with the posterior end closer to the person doing the evaluation, leaving the ventral edge
(quadrant 4) to the .left side and dorsal to the right side (quadrant 1). The first two slices
from each zone were evaluated for shatter marks. Starting from the first complete slice,
the slices were separated carefully. Then the slice was rolled over the finger, from
quadrant 1 to 4, for a more evident appearance of the shatter marks. Any shatter marks
were measured and recorded according to their corresponding quadrant and size category.
After shatter marks were recorded for a given slice, the slices were labeled for proximate
analysis.
Due to very low values of shattering, shatter was statistically analyzed as the
value of the length of the shatter marks. To obtain such a value, the number of marks
with in a length category (1-10 rom, 11-20 rom, 21-30 rom, 31-40 rom, 41-+rom) were
multiplied by the average size within the category (5.5 rom, 15.5 rom, 25.5 rom, 35.5
rom). The resulting value was called the Total Length of the Shatter Mark (TLS).
The Total Belly Shatter Marks values were obtained by adding all the shatter
marks recorded for a particular belly, disregarding zone, quadrant and size categories.
Proximate analysis was perfonned for each location of the belly. Ether Extraction
(Soxlet) determined fat percentage (Modified procedure from AOAC, 1990; section
85
27.006a, p.159). Moisture and ash were detennined using a thennogravimetric analyzer
(TGA-601, Leco Corp., St. Joseph, MI). Protein was determined by difference.
Bacon samples (five slices per location) were cooked on a Magi-Grill PGB-60
(Magikitch'n, Quakertown, PA) double belt conveyor cooker. Cook yield, dimensional
changes in length and width, and distortion scores were determined and recorded for
these samples. Preliminary trials were conducted to determine the correct settings for the
double belt cooker to yield the proper degree of doneness. The appropriate degree of
doneness was determined to be a color described as "golden brown" but not crisp as
described by Ross (1999). Cooked yield was required to be less than 40% of raw weight
to comply with USDAIFSIS regulations (USDA, 1996). The target cook yield was 37-
39% based on cooked color. As a result of these trials, the temperature of the top and
bottom platens was set at 204.4°C/400OP with a clearance between platens of 0.33 cm
during the cooking process. The preheat temperature was also set at 204.4°C/400OP. The
belt speed of the cooker was adjusted to control cooking time. After trials and with the
objective to reach the cooked yield and color required, the time was set to 40 seconds
from the time of sample entry until its exit from the belt cooker.
Cooked samples were evaluated for percent change in length (at the midpoint of
the slice), width (at the midpoint of the slice), and weight from the raw weight. A
distortion scale developed by Ross (1999) was used to evaluate the change in bacon
shape (Figure 1). The scale has five distortion scores. A distortion score of one
The samples were removed from the cooler (3°C/37<>P), the vacuum packages
were opened and the two slices from location A were removed and separated from each
sample. Their length and width was measured individually. The slices were placed on
the paper plate in order from 1 to 5 and weighed. The plate marked "A" was then taken
to the belt cooker for cooking. The same procedure was follow for locations B, C, D and
E in alphabetical order. The slices were placed on the belt cooker in order from 1 to 5
with spaces of two inches (5 cm). Gaps of eight inches (20.3 em) were left between
locations to prevent any mixing of slices from different locations. As the slices exited the
belt cooker, they were placed in numerical order on a clean and labeled paper plate. The
slices were allowed to cool for 30 seconds before being evaluated. All slices from
location A were evaluated individually with the distortion scale. Following length and
width measures, the five slices were evaluated by video imaging for another project.
Statistical Analysis
the belly from a pig with a given genetic line and sex and with one storage treatment
(fresh and frozen). The sample size for each one of the treatments was different.
Treatments were allocated in a complete randomize array. Data were analyzed using the
Mixed Model procedure of SAS (SAS, 1999) with a level of significance of P<0.05.
Means were separated using the Fisher's least significant difference test. When the zone
or quadrants were considerate, the analysis splits, or spilt-split plot, respectively were
87
utilized. The effect of genetic line, sex and treatment on fatty acid profile, shatter marks,
cooking characteristics and composition of bacon was tested. The model for the fatty
acid analysis consisted of genetic line, sex and treatment. When analyzing cooking
characteristics and composition, the design was a split-plot (the belly was the whole plot
and locations were the split). When analyzing shattering the design was a split-split plot
(the belly was the plot, and the quadrants within location were the split-split). The model
statement included pig id, line, sex, treatment location and quadrant. The random
statement terms were pig identification (line/sex/treatment). Significant main effects and
There were no significant differences in the saturated and unsaturated fatty acid
profile of the bellies due to storage treatment (Table 3). Fresh bellies contained values of
61.69 and 37.44 percent unsaturated fatty acids (UFA) and saturated fatty acids (SFA),
respectively. Frozen bellies contained 62.24 unsaturated fatty acids (UFA) and 36.86
saturated fatty acids (SFA) percent content (Table 3). Individually, fresh or frozen bellies
did not differ in the individual fatty acid profile of Myristic, Palmitoleic, Stearic, Oleic,
Linoleic, Linolenic and Eicosenoatic (Table 4). Fresh bellies had a 0.37 % higher
(P<0.05) Palmitic acid content than frozen bellies (Table 4). These results suggest that
frozen storage of 15 days or less will not change significantly the fatty acid profile of
bellies.
88
Yorkshire pigs from this study had a higher percentage of unsaturated fatty
acids, compared to all breeds (P<0.05), with a least square mean of 64.35% (Table 3).
Duroc and Landrace pigs followed with the next highest UFA levels, 62.89 and 62.40%,
respectively. Breeds with the lowest percents of unsaturated fatty acids were Poland
China, Chester White and Berkshire (61.46, 60.50 and 60.20%, respectively). As
expected, these breeds had higher saturated fatty acid levels than Landrace, Duroc and
Stearic, Oleic, Linoleic, Linolenic and Eicosenoatic) was also affected by genetic line
(Table 4). Individual profile of each genetic line followed similar patterns of the total
The current study data are in disagreement with Villegas et al. (1973) who
reported the backfat content for Yorkshire and Duroc pigs. They reported Yorkshire with
40.04% saturated (Myristic, Palmitic and Stearic) and 57.06% unsaturated (Palmitoleic,
Oleic, Linoleic, Linolenic and Eicosenoatic) fatty acid content. Durocs were reported at
39.87% saturated fatty acid and 57.38% of unsaturated fat. Pigs in Villegas et al. (1973)
had higher overall total unsaturated fatty acids. These differences could be due to the fact
that in the last 30 years pig production (including genetic improvement and diet changes)
has altered the characteristics of the carcass including fatty acid composition.
Based on our results, it appears that production changes have had a greater impact
on Yorkshire pigs than on Duroc pigs. Our study suggests that Yorkshire pigs tended to
Sex Effect
unsaturated fatty acid content of the bellies (Table 3). Barrows had 1.12% higher
(P<0.05) of saturated fatty acids than gilts. Gilts had a 1.08% higher (P<0.05)
unsaturated fatty acid content than barrows. The difference in the saturated versus
unsaturated profile may be explained by the difference in the content of Palmitic acid.
Barrows had a 24.04% of Palmitic versus 23.17% of Palmitic for gilts. The content of
the remaining fatty acids was not different (P>0.05) for gilts vs. barrows; although, a
similar pattern can be observed: barrows had higher content of the saturated fatty acids
Villegas et ai. (1973) reported that the backfat of gilts contained more linoleic
acid (18:3) than barrows. These results suggest that gilts deposit a higher percentage of
unsaturated fatty acids than barrows. However, this disagrees with Geri (1984). Their
study examined lipid extracted from the longissimus muscle and reported no differences
between sexes for fatty acid composition of single fatty acids or the sum of saturated or
unsaturated constituents. The discrepancy among the present study and Geri (1984)
Maw et ai. (2003) established that the physical characteristics of fat samples are
related to their fatty acid composition. Unsaturated fatty acids are responsible for soft fat
in the belly (Ellis and McKeith, 1999; Enser et aI., 1984; Chant et aI., 1976). The
differences we found in content of unsaturated fatty acids in the belly among breed and
sex could have an effect on bacon production, soft fat could have negative impacts on
Shatter Analysis
There was an interaction (P<O.OI) between treatment and location (Table 5). In
both treatments (fresh, frozen), locations to the posterior end of the belly (A and B)
possessed the highest total length of the shatter mark. The length of the shatter mark
within a slice of bacon is expressed as total length of the shatter mark (TLS). Fresh
bellies had a TSL of 2.76 mm for both locations. Frozen bellies had TLS of 3.72 mm and
2.59 mm for locations on the anterior end of the slab (D and E, respectively). Frozen
bellies at locations in the center of the slab (B, C and D) had larger (P<0.05) shatter
values than fresh bellies (Figure 2). Although, fresh vs. frozen treatments were not
significantly different for locations on the most posterior and most anterior part of the
belly (A and E), it appears that frozen bellies tended to shatter more than fresh bellies. It
was discovered that frozen bellies at location D had the highest incidence of shattering
(3.72 mm). The lowest incidence of shattering was found in location A, fresh (0.46 mm)
Past reports suggest that freezing and frozen storage may affect the quality of the
meat (Bjerklie, 1996; Calvelo, 1981; Jeremiah, 1980; Miller et al., 1980). The fonnation
of ice crystal during freezing is an inevitable process. These ice crystals are believed to
be responsible for the rupturing of the cells around them (Bjerklie, 1996). These studies
have looked at muscle fiber. However, there is little data on the effects of freezing on fat
cells. Base on the higher incidence of shatter marks in the frozen bellies, it could be
speculated that adipose tissue in these bellies was damage during freezing.
91
interactions. The interactions for the total length of the shatter mark were for, Line x Sex
Lines with the highest fat content based on proximate analysis results, (Berkshire,
Poland China and Chester White) had the highest incidence of shatter marks (Figure 3).
For all lines, barrows had greater (P<0.05) TLS than gilts (Table 7).
These results are in agreement with Salas-Perez (2002), who reported that in their
study genetic lines with higher fat content had higher scores for all variables analyzed for
shattering of the bacon slice (percentage of bacon slices affected, occurrence of shatter
Sex Effect
There was an interaction (Table 5) between sex and location (P<0.01). Location
D had the highest level of TLS with 4.53 mm and 1.95 mm for barrows and gilts,
respectively (Table 6). Bacon from barrows had more TLS in all locations than bacon
made from gilt bellies (Table 6). These may be attributed to the fact that bellies from
barrows were fatter than gilts. These results are in agreement with Salas-Perez (2001).
There was an interaction between sex and slice quadrant (P<0.01) (Table 5).
Barrows had a greater TLS in each quadrant as compared to gilts (Table 7). From these
results it appears that within sex, quadrants 1 and 2 are more susceptible to shattering.
This may be explained by the way bellies were handled during slicing. The machine may
have exerted greater stress on quadrants 1 and 2 as it moved through the belly.
92
than the lowest reported value for TLS (which is in quadrant 4) (Table 7). Although, it is
difficult to asses the practical value of shatter marks with this small measurements.
Correlations were run comparing the level of fatty acid saturation in the belly with
the total number of shatter marks in the whole belly. It was thought that higher saturation
of the fat would have higher tendency to disrupt and break during slicing, therefore
resulting in shatter marks in the sliced bacon.. Because saturated fat that is refrigerated
appears to be harder. According to Table 8, there is an r= .22. Suggesting that 22% of the
variation in total number of shatter marks is due to the percentage content of saturated
fatty acids.
Overall results for shattering suggested that the occurrence of shattering in the
present study was less than in the study conducted by Salas-Perez (2002). These may be
yield of the bacon slices (Table 9). Bacon slices from locations A and B had higher
cooking yields than C, D and E (Fig. 4). Location D had the lowest yields (32.00 and
31.80 % for fresh and frozen, respectively). Except for location A, within location, fresh
Treatment had an effect on the distortion (P<0.05) of the cooked bacon slices
(Table 9). Frozen bellies had higher least square mean for distortion than fresh bellies.
Frozen bellies had an average score of 2.43 compare to a 2.35 of fresh bellies (Table 10).
According to the results, bacon made from fresh bellies had better performance
during cooking than the one made from frozen bellies. Perhaps structural changes took
place in the bellies during frozen storage. Several theories could explain the nature of the
damage to the muscle. One of these theories establishes that the formation of ice crystals
during freezing could damage the cell membrane resulting in moisture loss (Ngapo et aI.,
1999; Nusbaum et aI., 1983; Nichols and MacKintosh, 1952.). A second theory suggests
that frozen storage induces deterioration of muscle thru chemical reactions, resulting in
denaturation of muscle proteins (Miller et aI., 1980). The results in the present study
demonstrated differences between the cooking abilities among bacon from fresh and
frozen bellies, although further research should be conducted to investigate the reasons.
All cooking parameters were affected by genetic line (Table 9). Table 10
illustrates differences among lines for cooking yield. Berkshire and Poland China pigs
had the lower (P<0.05) cooking yields (31.90 and 31.14, respectively) (Table 10). Of all
lines, Berkshire and Poland China had the highest percent of fat in the belly (45.09 and
45.01 %, respectively) (Table 11). Leaner pigs such as Duroc and Yorkshire had the
highest yields (36.69 and 37.49, respectively). These results support previous data that
established that cooking yields are inversely related to the fat content of the bacon slice
(McEver, 2000; Kemp et aI., 1969; Carpenter et aI., 1963; Saffle and Bratzler, 1959).
94
Jabaayet al. (1976) found that bacon cooking losses were due to lipid cooked out.
A study by Stanley et al. (1980) reported that 95% of the moisture and 75% of the lipid in
Distortion was affected by an interaction (P<O.05) among line and sex (Table 9).
The highest distortion with an average score of 2.56 corresponded to Chester White gilts.
Followed by Duroc gilts, which had an average of 2.53 (Figure 5). For the rest of the
lines, bacon from barrows exhibited more distortion than that from gilts.
Sex Effect
There was an interaction (P<O.OOOI) between sex and location for cooking yield
(Table 9). Gilts had higher (P<0.05) yields for four (B, C, D and E) of the five locations
on the bacon slab (Figure 6). Slices from the most posterior locations of the bacon slab
(A and B) had higher (P<O.05) yields when cooking than the rest of the locations (Figure
6). The higher yields could be attributed to lower fat content of the slices. As seen in
Table 11, the posterior end of the slab (locations A and B) had lower means for percent
fat.
sex and location (Table 9). For barrow and gilts, location A had the lowest shrink in
width, while locations D and E observed the highest shrink in width of the bacon slice
(Figure 7). Difference (P<O.05) between barrows and gilts were only notice in location D
Proximate Analysis
Bellies from Berkshire and Poland China pigs had higher (P<0.05) fat percent
content (Table 11). Bacon made from Duroc and Yorkshire bellies had the lowest
(P<0.05) fat percentage content (Table 11). Genetic lines with lower fat percentage had
the highest moisture content (46.78% and 46.70% for Duroc and Yorkshire,
respectively). These genetic lines also had higher protein percentage content (14.38%
Sex Effect
Barrows had higher (P<0.05) fat percentage content when compared to gilts
(Table 11). Inversely, barrows exerted lower (P<0.05) moisture percentage than gilts
(41.64% and 45.58% for barrows and gilts, respectively) (Table 11).
Location Effect
Locations on the most anterior end of the belly (D and E) had higher (P<0.05) fat
percentage and lower moisture content than locations on the center and posterior (A, B
Conclusion
Demand for bacon has increased due to consumer preference for flavor and
Commercial bacon processors consider shattering of the fat on a bacon slice and
96
distortion of the cooked slice as quality defect. Along with these two, fatty acid
composition, cooking yield and size shrinkage are important in the overall profitability
Freezing pork bellies prior to curing has been a common practice in bacon
production. Freezing provides an excellent means for storing meat for long periods of
time. Science has proved that freezing and storage of meat could damage muscle tissue.
There is little data that reports the effect of the damage on frozen bellies when processing
to sliced bacon.
The results suggest that very similar processing characteristics and yields would
be obtained when manufacturing bacon from fresh or frozen bellies. In the present study
bellies were in frozen storage for 15 days prior to processing. This period of time had
little effect on the fatty acid profile of the bellies. Genetic line and sex had greater impact
on the percentage of saturated and unsaturated fatty acids in the pork bellies. The effects
of frozen storage on shattering of the bacon slice appear to be less significant than that of
genetic line, sex, location and quadrant. However, results suggest a relationship between
fat content and shattering; therefore, it could be speculated that adipose tissue in the
bellies was damage during freezing. Results from the cooking analysis suggest that
genetic line, sex and location had greater impact on cooking parameters than treatment
(fresh, frozen).
when using frozen bellies. Although, findings might suggest that longer periods of
storage could lead to significant processing and cooking differences. These differences
97
could have greater impact in bacon processing operations, were volumes of production
The fat content as well as the fatty acid composition appear to play an important
role in the manufacturing characteristics of the belly. In bacon production, bellies with
soft fat could have negative impact on slicing yields, shelf life and texture of the final
product.
Finally, it appears that genetic line and sex had the greatest impact on processing
Reference
Bjerklie, S. 1996. PC QC, Maintaining frozen portion quality control. Meat and
Poultry. February. 16-18.
Bryhni, E.A., N.P. Kjos, R.Ofstad and M. Hunt. 2002. Polyunsaturated fat and fish oil
in diets for growing-finishing pigs: effects on fatty acid composition and meat, fat
and sausage quality. Meat Sci. 62:1-8.
Cameron, N.D., M. Enser, G.R. Nute, F.M. Whittington, J.C. Penman, A.C. Fisken, A.M.
Perry and J.D. Wood. 2000. Genotype with nutrition interaction on fatty acid
composition of intramuscular fat and the relationship with flavour of pig meat.
Meat Sci. 55:187-195.
Cannon, J.E., J.B. Morgan, F.K. McKeith, G.C. Smith, S. Sonka, J. Heavner, and D.L.
Meeker. 1996. Pork chain quality audit survey: Quantification of pork quality
characteristics. J. Muscle Foods. 7:29-44.
Carpenter, Z.L., R.G. Kauffman, R.W. Bray and K.G.Weckel. 1963. Factors influencing
quality in pork. B. Commercially cured bacon. J. Food Sci. 28:578-583.
Chant, J.L.Jr., D.M. Stiffler, D.M. Kinsman and A.W. Kotula. 1976. Chemical and
sensory aspects of commercial bacons. J. Anim. Sci. 43:989-993.
Davenel, A., A. Riaublanc, P. Marchal, and G. Gandemer. 1999. Quality of pig adipose
tissue: relationship between solid fat content and lipid composition. Meat Sci.
51:73-79.
Ellis, M., and F., McKeith. 1999. Nutritional influences on pork quality. In Pork Facts.
National Pork Producers Council. Des Moines, IA.
Enser, M., E. Dransfield, P .D. Jolley, R.C.D. Jones, and M. Leedman. 1984. The
composition and consistency of pig backfat as it affects the quality of vacuum-
packaged rindless bacon rashers. J. Sci. Food Agri. 35:1230-1240.
Food Marketing Institute. 2000. Trends in the United States: Consumer Attitudes & the
Supermarket,2000. Food Marketing Institute, Washington, DC.
Geri, G. 1984. Genetic and sex effects on fat depositions and quality. In Fat Quality in
Lean Pigs., Workshop in The CEC Programme, Brussels, Belgium 126-129.
99
Houben, J.H. and B. Krol. 1984. Pig fats and the manufacture and storage of meat
products. In Fat Quality in Lean Pigs., Workshop in The CEC Programme,
Brussels, Belgium 15-26.
Jabaay, R.W., J.C. Forrest, E.F. Aberle, H.V. Courtenay and M.D. Judge. 1976. Bacon
quality criteria and associated carcass traits. J . Food Sci. 41 :431-437
Jeremiah, L.E. 1980. Effect of frozen storage and protective wrap upon the cooking
losses, palatability and rancidity of fresh and cured pork products. J. Food Sci.
45:187-196.
Kemp, J.D., W.O. Moody and J.D. Fox. 1969. Effect of fatness and fresh pork quality
on yield and quality of bacon and yield of ham. J. Anim. Sci. 28:612-619.
Maw, S.J., V.R. Fowler, M. Hamilton, and A.M. Petchey. 2003. Physical characteristics
of pig fat and their relation to fatty acid composition. Meat Sci. 63: 185-190.
Mazhar, A., J.R. Busboom, R.A. Field, D.C. Rule, T. Heald, W.C. Russell and R.I.
McCormick. 1990. Functional characteristics, fatty acid composition, and
palatability of bacon from pigs fed canola. J. of Food Sci. 55:575-576.
McEver, E.M. 2000. Precooked bacon manufacture by microwave and double belt
conveyor cooking systems. M.S. Thesis. University of Nebraska-Lincoln,
Lincoln, NE.
Metcalfe L.D., A.A., Schmitz, and J.R., Pelka. 1966. Rapid preparation of fatty acid
esters from lipids for gas chromatographic analysis. Anal. Chem. 38:514-515.
Miller, A.J., S.A. Ackerman and S.A. Palumbo. 1980. Effects of frozen storage on
functionality of meat for processing. J. Food Sci. 45:1466-1471.
NAMP. 1997. North American Meat Processors Association Meat Buyer's Ouide.
North American Meat Processors Assoc. Reston, VA.
Ngapo, T.M., I.H. Baber, J. Reynolds and R.F. Mawson. 1999. Freezing and thawing
rate effects on drip loss from samples of pork. Meat Sci. 53:149-158.
Nichols, J.B., and D.L. MacKintosh. 1952. Structural changes occurring in muscle
tissue during repeated freezing and thawing. Food Tech. 6:170-174.
Nusbaum, R.P., J.O. Sebranek, D.O. Topel and R.L. Rust. 1983. Structural and
palatability relationships in frozen ground beef patties as a function of freezing
treatments and product formulation. Meat Sci. 8:135-146
100
Ross, R. 1999. Precooked bacon manufactured by microwave and double belt conveyor
processing systems. M.S. Thesis. University of Nebraska-Lincoln, Lincoln, NE.
Same, R. L. and L. J. Bratzler. 1959. The effect of fatness on some processing and
palatability characteristics of pork carcasses. Food Technol. 13:236-239.
Salas-Perez, P.J. 2002. Differences among breeds, diets, slaughter weights and sex type
in the shattering of bacon slices. M.S. Thesis. University of Nebraska-Lincoln,
Lincoln, NE.
Salvage, B. 2002. Bacon boom!. Meat Marketing and Technology. Sept. Chicago,IL.
50-57.
SAS. 1999. SAS/STAT User Guide (Version 8.0). SAS Institute Inc. Cary, NC.
Stanley, D.W., P.W. Voisey, and H.J. Swatland. 1980. Texture-structure relationships in
bacon lean. J. Texture Studies. 11 :217-238.
USDA. 1996. Food Standards and Labeling Policy Book. Food Safety and Inspection
Service; Washington, DC.
Villegas, F.J., H.B. Hedrick, T.L. Veum, K.L. McFate and M.E. Bailey. 1973. Effect of
diet and breed on fatty acid composition of porcine adipose tissue. J. Anim. Sci.
36:663-668.
Wenther, J. B. 1999. The effect of genetic line, diet, sex, slaughter weight and type of
bacon on bacon processing parameters. M.S. Thesis. University of Nebraska-
Lincoln, Lincoln, NE.
Wood J.D., and A.V. Fisher. 1990. Reducing Fat in Meat Animals. Elsevier Applied
Science. New York, NY.
101
a. Phosphate was added to the cold water (~4°C/40°F) agitated until phosphates are
dissolved.
b. With continued agitation, the remainder of the non-meat ingredients were added in the
following order: salt, sugar, nitrite, natural liquid smoke flavor as required and lastly the
ascorbate.
Wenther, 1999.
102
Table 3. Least Square Means ± S.E. of Percent Unsaturated Fatty Acid (0/0 UFA)
and Percent of Saturated Fatty Acids (0/0 SFA) for Sex, Line and Storage
Treatment Effects.
%UFA %SFA
Effect
B31TOWS 61.43 ± 0.26a 37.71 ± 0.27a
Gilts 62.51 ± 0.22b 36.59 ± 0.23 b
abed
Means within the same column within a main effect with
similar superscripts are not significantly different (P>O.05).
Table 4. Least Square Means ± S.E. of the Fatty Acid Profile for Sex, Line and Treatment Effects.
Berkshire 1.36 ± O.Olb 25.09 ± 0.14a 2.88 ± 0.03 b 11.89 ± 0.12ab 45.78 ± 0.27be 9.45 ± 0.27b 0.42 ± 0.03 b 0.76 ± O.oob
Chester White 1.49 ± 0.02a 24.84 ± 0.27ab 3.07 ± O.06a 11.45 ± 0.24b 45.48 ± 0.48c 9.74 ± 0.54b 0.45 ± 0.06b 0.80 ± 0.01 be
Duroc 1.30:t: O.Olc 22.27 ± 0.12d 2.13 ± 0.02e 11.99 ± O.l1a 41.90 ± 0.22e 15.66 ± 0.25 a1.15 ± 0.02a 0.78 ± O.oob
Landrace 1.28 ± 0.0200 23.20 ± 0.33 c 2.61 ± 0.07c 11.27 ± 0.30b 47.28± 0.66ab 10.17 ± 0.67b 0.46 ± 0.07b 0.77 ± 0.02be
Poland China 1.30 ± 0.03 00 24.31 ± 0.35b 2.66 ± 0.08 c 11.54 ± 0.32ab 47.35 ± 0.63 a 9.25 ± 0.71b 0.46 ± 0.08b 0.80 ± 0.02be
Yorkshire 1.26 ± O.Old 21.95 ± 0.14d 2.30 ± O.03 d 10.69 ± 0.12c 42.98 ± 0.25 d 15.73 ± 0.28a 1.18 ± 0.03 a 0.81 ± O.Olae
Fresh 1.33 ± 0.01 23.79 ± 0.13 a 2.63 ± 0.03 11.59 ± 0.12 45.12 ± 0.25 11.36 ± 0.27 0.67 ± 0.03 0.78 ± 0.00
Frozen 1.34 ± 0.01 23.42 ± 0.14b 2.59 ± 0.03 11.35 ± 0.12 45.14± 0.26 11.97 ± 0.28 0.70 ± 0.03 0.78 ± 0.00
14:0-Myristic;16:0-Palmitic;16:1-Palmitoleic;18:0-Stearic;18:1-0leic;18:2-Linoleic;18:3-Linolenic;20:1-Eicosenoatic.
abed Means within the same column and within main effect with similar superscripts are not significantly different (P<0.05).
"""'"
~
105
Table S. Analysis of Variance of the Total Length of the Shatter Mark, Main
Effects and Two-Way Interactions.
Storage 2.58 NS
Treatment •
Treatment • Location 7.36 < 0.0001
Treatment • Quadrant 0.80 NS
Table 6. Least Square Means and S.E. for Total Length of the Shatter Mark for the
Interactions* with Location **.
* Superscripts may be compared in row and column for the effects (storage
treatment, sex and quadrant). However comparisons cannot be made between two
interactions.
** Location: sampling location of the bacon slab from posterior to anterior. Each
location represents 20% of the slices of the slab (see appendix 2).
*** Quadrant: sampling location of the bacon slice representing 25% of the slice
length from dorsal to ventral (see appendix 2).
107
Table 7. Least Square Means and S.E. Total Length of Shatter Mark (mm) for Line
x Sex Interaction and Quadrant x Sex Interactions*
TBS 14:00 16:00 18:00 TSF 16:1 18:1 18:2 18:3 20:1 TUF
TBS 1.000
14:00 0.18926 1.000
<0.0001
16:00 0.28746 0.51843 1.00
<0.0001 <0.0001
18:00 0.08295 0.02069 0.56964 1.00
0.0504 0.6258 <0.0001
TSF 0.22828 0.38800 0.92824 0.83045 1.00
<0.0001 <0.0001 <0.0001 <0.0001
16:1 0.20816 0.62945 0.49978 -0.3016 0.20840 1.00
<0.0001 <0.0001 <0.0001 <0.0001 <0.0001
18:1 0.15706 0.04661 0.17560 -0.0985 0.07260 0.43088 1.00
0.0002 0.2717 <0.0001 0.0200 0.0866 <0.0001
18:2 -0.2694 -0.3666 -0.7601 -0.3970 -0.6927 -0.5588 -0.7499 1.00
<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
18:3 -0.2018 -0.3472 -0.6988 -0.3736 -0.6448 -0.5410 -0.7530 0.96808 1.00
<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001
20:1 0.10881 -0.0779 -0.0996 -0.0357 -0.0801 -0.1025 0.44309 -0.2444 -0.2260 1.00
0.0102 0.0657 0.0186 0.3998 0.0586 0.0154 <0.0001 <0.0001 <0.0001
TUF -0.2193 -0.3920 -0.9233 -0.8290 -0.9974 -0.2047 -0.0554 0.68199 0.63899 0.09027 1.00
<0.0001 <0.0001 L-C:::().OOOI <0.0001 ~-<:O.OOOI .. <0.0001 0.1911~ ,--<::0.0001_ <0.0001 0.033~ L
-
-
o00
Table 9. Analysis of Variance of Percentage of Shrink in length, Percentage of Shrink in Width and Distortion of Cooked
Bacon Slices and their Two-Way Interactions.
Line 36.32 <0.0001 2.75 0.0182 14.42 < 0.0001 4.7 0.0003
Line * Sex 0.63 NS 0.23 NS 0.44 NS 2.25 0.0482
Line * Treatment 1.85 NS 0.44 NS 0.44 NS 2.18 NS
Line * Location 1.42 NS 0.77 NS 1.25 NS 1.33 NS
Location 120.2 < 0.0001 59.03 < 0.0001 215.23 < 0.0001 124.9 <0.0001
a Cooking Yield = (cooked slice wt / initial slice wt) * 100.
b % Shrink-Length = (cooked slice length - initial slice length) * 100 / cooked slice length.
C % Shrink-Width= (cooked slice width - initial slice width) * 100 / cooked slice width.
d D"IstortIon
" Score from 1 to 5 assigned to the cooked slices (see Fig. 1) (Ross, 1999).
NS Not statistically significant (P>O.05).
.....
0
\0
110
Table 10. Least Square Means ± S.E. of Cooking Yield, 010 of Shrink in Length,
0/0 Shrink in Width and Distortion score for all Main effects. *
Table 11. Least Square Means ± S.E. Proximal Composition of the Raw Bacon for
all Main Effects.
asca Means within the same column and within a main effect with similar superscripts are not
significantly different (P>O.05).
* Sampling location of the bacon slab from posterior to anterior. Each location represents 20% of
the slices of the bacon slab (see appendix 2).
112
Table 12. Least Square Means ± S.E. Proximal Composition of the Cooked Bacon
for all Main Effects.
Ross, 1999.
114
Figure 2. Total Length of the Shatter Mark Least Square Means for
Treatment x Location.
4
3.5
3
1
~
25
2
• Fresh
• Frozen
'"
~ 1.5
1
0.5
a
A B c D E
Location
Figure 3. Total Length of the Shatter Mark Least Square Means for Line x
Sex.
4
3.5
3
E 2.5 • Barrows
.!!.
l: 2 • Gilts
C>
c:
(I)
...J
1.5
1
0.5
0
Berkshire Chester Duree Landraee Poland Yorkshire
White China
38
37
~ 36
•~
~ 35
:5! 34 • Fresh
"
;;; 33 • Frozen
.:.:
g 32
U
31
30
29
A B C D E
Location
abcde
Means with a similar superscript are not significantly different (P>0.05) .
• Cooking Yield~ (cooked slice wt / initial slice wt)· 100.
117
Figure 5. Distortion Score Least Square Means for Line x Sex Interaction. *
2.60
2.55
2.50
2.45
~ • Barrows
§ 2.40
. Gilts
~ 2.35
~ 2.30
0(
2.25
2.20
2.15
2.10
Berkshire Chester Duroe Landrace Poland Yorkshire
White China
abcde
Means with a similar superscript are not significantly different (P>O.05).
• Distortion Score refers to the value assigned to the cooked slices (see Fig. I)
(Ross, 1999).
118
45
40
35
~
~
-0 30
• Barrows
">=
Oll
c
25
20
. Gihs
:;;;
0
0
15
u 10
5
0
A B C D E
Location
abcde
Means with a similar superscript are not significantly different (P>0.05) .
• Cooking Yield ~ (cooked slice wt / initial slice wt)' 100.
119
30
25
..c:
~
20
""~,
] 15 I •·Barro~
Gilts
1
(/)
10
~
0
A B C D E
Location
aocde
Means with a similar supe rscript are not significantly d ifferent (P>0.05) .
• % Shrink· Width= (cooked slice width·initial slice width)· 100/cooked slice width .
120
"The skinless belly is as described for NAMP No. 408, except that the skin is
removed, leaving a smooth skinned surface which is practically free of hair roots and
scores. The belly is prepared from the side after removal of the ham, shoulder, loin, fat
back and spareribs. All bones and cartilage shall be removed. The fat back shall be
removed by a straight cut no more than 1 12 inches (38 mm) from the outermost dorsal r
curvature of the scribe line. The anterior (shoulder) and posterior (ham) ends of the belly
shall be reasonable straight and parallel. No side of the belly shall be more than 2 inches
(51 mm) longer than its opposite side. The width of the flank muscle (rectus abdominas)
shall be at least 25% of the width of the belly on the ham end. The fat on the ventral side
of the belly and adjacent to the flank shall be trimmed to with 0/4 inch (19 mm) from the
lean. The area ventral to the scribe line shall be free of scores and snowballs (exposed
areas of fat) which measure 3 square inches (194 sq. mm) or more. The belly shall be
free of enlarged, soft, porous, dark or seedy mammary tissue. The scribe line is not
considered a score, but shall not be more than 'l4 inch (6 mm) in depth at any point."
,.
.,.~
J
121
Comb A B c D E Incomplete
End Slices
0
Quad 0
12345 .... 12345 .... 12345 .... 12345 .... 12345 ....
1 0
0
Quad 0
2 0
0
Quad 0
3 0
0
Quad 0
4 0
Ventral
GC sample preparation.
Saponification / Methylation
1. Transfer 10-1 OOOmg of fatty tissue to a 13 x 100 mm (or 16 x 100 mm) screw cap
test tube.
2. Add 0.5 mL 0.5 N NaOH in Methanol, purge with nitrogen, cap tightly, mix
thoroughly, and heat 5 minutes at 100°C in boiling water bath or heating block.
3. Remove from heat and cool to room temperature. Add 0.5 mL boron trifluoride
in methanol, purge with nitrogen, cap tightly, mix thoroughly, and heat and
additional 5 minutes at 100°C in boiling water bath or heating block.
4. Remove from heat and cool to room temperature. Add 1.0 mL hexane (or other
injection solvent) and mix thoroughly.
5. Add 1 mL saturated NaCI, cap tube, and shake vigorously for 30 seconds.
6. Centrifuge at 1500 rpm for 5 minutes to separate phases.
7. Transfer the organic layer containing the fatty acid methyl esters to a GC vial for
injection. Be careful no to transfer any water to the vial. Samples may be "dried"
prior to injection by adding about 0.5 g Na2S04 and mixing. Allow the Na2S04 to
settly about 20 minutes, then transfer the solvent to a clean vial.
Reagents
1. Fatty acid methyl ester standards.
2. 0.5 NaOH in methanol
3. Boron trifluoride in methanol (14%), Sigma B-1252
4. Saturated NaCI in H20
5. Hexane (methylene chloride or isooctane can also be used as injection solvent)
124
Sample Preparation
Bacon samples were stored in a freezer (-20°C) prior to analysis. The sample was
removed from the plastic bag with tweezers and frozen in liquid nitrogen fro
approximately 1-2 minutes. The sample was then removed with tweezers and put into a
blender cup (250 ml cup for location samples in 1 L cup for composite samples) that had
been supercooled in liquid nitrogen. Composite samples were divided into three subsets
for ease of powdering. The sample was then powdered in a Waring Commercial Blender
(Dynamics Corporation of America, New Hartford, CT) for approximately 1 minute or
until the samples was completely homogenized. The powdered sample was then
transferred into a clean plastic bag. An additional bag with the identification tag was
placed over the sample bag and ultra-low freezer (-80°C) for storage.
Extraction Procedure
A thin film of silicone grease was applied to the ground joints of the extraction
tube. The boiling flask was filled with anhydrous ether until it was approximately % full.
A boiling stone was placed in each flask. The extraction tube was then placed on the
flask. The joints were checked for firm attachment by rotating the tube 'l4 tum and back
125
again. Weighed samples were placed in the extraction tube. Each tube held twenty
samples. The flask and extraction tube were setup in the extraction room and connected
to the condenser. The apparatus was checked for a firm connection. The distillation
water was turned on by turning the spigot y.. of a turn. The temperature on the hot plate
was set between 4 Y2 and 5, and the solution was allowed to distill for 72 h. All
connection were checked daily. After the extraction was complete, the hot plate was
turned off and allowed to cool before the flask and extraction setup were moved back into
the fume hood (1 Y2 h). The water was turned off, and the ether was poured back into the
recovery container for reuse or re-distillation. The finished samples were left in the hood
for 2 h to vent off any remaining ether. Samples from each distillation setup were kept in
a separate pan. The samples were then transferred to the drying oven (105°C) for 24 h
before recording weights. This procedure removed fat and moisture from each sample.
Fat percentage was calculated using the following equation:
Fat % = [(original sample weight - extracted sample weight) / (sample weight) -
moisture)] *100
Moisture and Ash (Thermogravimetric Analyzer (TGA-601) Leco Corp., St. Joseph,
MI)
This method measures the weight loss of the samples as a function of temperature
in a controlled environment. The analyzer consists of an electronic chassis for furnace
control and data management and a mUltiple sample furnace, which allows up to 19
samples to be analyzed simultaneously.
After the analysis method was selected, empty crucibles were loaded into the
furnace carousel. This method controls the carousel, furnace, and balance operation.
Once the crucibles were tare, each crucible was individually presented to the operator for
sample loading. The starting sample weight was measured and stored automatically.
Once all the crucibles were loaded, the analysis began. The weight loss of each sample
was monitored, percent weight loss in each sample for each analysis step was printed at
the end of the analysis as the carousel turned and lowered the sample onto the balance.
The analyzer contains a menu-driven program, which allowed analysis methods to be
126
customized. Temperature, temperature ramp rate, and atmosphere could be adjusted and
saved in the TGA-601 computer.
Powered samples were removed from the ultra-low freezer. Samples were stored
in small styro-foam cooler containing liquid nitrogen to prevent thawing while loading
the liquid nitrogen. Sample identification numbers were entered into the computer. The
method of operation was then selected (User defined). Setting for Leco:
Name Covers Ramp Rate Ramp Time Start Temp End Temp
Moisture Off 6 dim :17 min 25°C 130°C
Ash Off 20 dim :30 min 130°C 600°C
Name Atmosp Flow Rate Hold Time Const. Wt. Const. wt.
Time
Moisture N High :00 min 0.05% :09 min
Ash 0 High :00 min 0.05% :09 min
General Setting
Select "Analysis" and click on "collect". Next choose the furnace to be used.
Load empty crucibles into select furnace. TGA-601 will weigh all crucibles to obtain a
tare weight. After tare is obtained the machine will call to load each sample (lg). Return
samples to ultra-low freezer. After all samples are loaded the machine will automatically
star. When analysis is finished click the "save" icon on toolbar and print a hard copy of
results. Remove crucibles after they have cooled down for 30 minutes. Wash them in
127
soapy water and allow to dry in drying oven for 4h. Remove dry crucibles and transfer
them to desiccator until future use. Before doing another run, machine must cool down
to 25°C. Margin of error for fat samples is 1.5% difference between duplicate samples.
Margin of error for moisture is 2% difference between duplicate samples. Margin of
error for ash is .5% difference between duplicate samples.
AOAC, 1990. Official Methods of Analysis, 14th ed. Association of Official Analytical
Chemists, Washington, DC.
128
and storing frozen bellies. Previous research has suggested that freezing methods as well
....
as freezing rates have different effects over muscle cell structures. Therefore, bellies
should be freeze by different methods to obtain different freezing rates, and evaluate the
Beside the method and rate by which the belly could be freeze, frozen storage
times could also play an important role. Research should look at longer storage times.
Similar evaluations of the belly performance should be measured over time, as well as ...
changes in the fatty acid composition of the belly during frozen storage.
In the present study, fat samples were collected to evaluate fatty acid composition.
The sampling was conducted on one anatomical point of the belly. To obtain a better
idea of the fatty acid composition of the whole belly, sampling should be design to
include different locations on the belly. Since the results suggested that percentage of
Attention should be directed towards the structure of the connective tissue, to the
I
quantity of the proteins comprising it, as well as to its degree of maturation. Analysis
could be conducted to compare collagen content with the level of distortion of the cooked
slices.
The present study did not establish a direct relationship between shattering of the
bacon slicing and the potential effect on cooked bacon. Further analysis could be
Finally, a successful grading system could serve to efficiently direct carcasses and
bellies into an appropriate niche (supermarket, restaurant, further processing, etc). In its
simplest form, grading is an attempt to separate a commodity into lots, each lot having a
relatively high degree of uniformity in certain traits associated with market preferences
and value.