Underutilized Resources as Animal Feedstuffs (1983)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Industrial Food Processing Wastes INTRODUCTION As a result of changing waste-handling technology, many industrial food processing wastes now being fed to animals were once considered to be without economic value as animal feed. Other factors that have increased the interest in wastes as animal feeds include the cost of disposing of waste and increased restrictions, brought about by environmental con- cerns, on disposing of waste materials. The necessity of separating solid waste from liquid waste, as well as the necessity of removing suspended and dissolved substances from wastewater before discharging it, has re- sulted in the production of waste materials that are lower in water content and consequently more economically attractive as animal feeds. More stringent controls on the use of pesticides have also reduced the pesticide levels in food processing wastes. QUANTITY In general, information on quantities of industrial food processing wastes (residuals) is limited. Information on processing wastes from fruit, ve*~- etable, and seafood processing, collected by Katsuyama et al. (1973), is presented in Tables 1 and 2. Although these data are not current, an overall view is given of waste from fruit, vegetable, and seafood processing. The major changes since 1973 would be greater utilization of wastes for animal feed and as sources of energy. There are also data on quantities of waste in the sections on specific wastes. Table 1 summarizes food processing s

6 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS TABLE 1 Yearly Food Processing Industry Solid Residuals, by Product and Disposal Method TotalTotal Total TotalTotalNot Raw Tonsas in By-Resid-Accounted ProductProcessedSolids LiquidFeed OtherProductsualsFor Vegetables Asparagus10922 017 17383 Bean, dry2095 a2 26(-1) Bean, lima1099 09 917(-3) Bean, snap57161 358 581180 Beet24559 516 168219 Broccoli sprouts, cauliflower23619 182 82100a Cabbage20958 55 569( -1 ) Carrot25433 291 a911279 Corn2,24981 31,388 1,3881,46938 Greens, spinach2187 a22 22304 Mushroom6129 a 029( - 2) Pea52622 044 a44674 Pickle50836 1 037(-14) Potato, white3,23877 44943 9431,061154 Pumpkin, squash20020 822 2235088 Tomato6,322345 27109 109472136 Vegetable, misc.1,106100 47100 100245a Fruits Apple95282 a100 7918126327 Apricot1095 a6 28144 Berry1819 21 1132 Cherry17218 14 a4241 Citrus7,07572 12,721 32,7212,794281 Fruit, misc.13623 a7 29333 Olive772 a 9910a Peach998163 1345 4085263(-17) Pear37265 933 3310913 Pineapple81627 9326 3263630 Plum, prune245 1a 061 Specialtiesb2,26839 22190 152092720 Seafood  Clam, scallop8211 a 01 259 Oyster180 2 141416a Crab274 14 0201 Shrimp1096 3714 a156018 Salmon1090 324 25363 Sardine240 0 555a Tuna, misc. seafood4720 a62 27909082 U.S. total30,3911,514 2896,421 1196,6248,420950 NOTE: All figures x 1,000 metric tons; rounded (after adding). a4S0 metric tons or less. bBaby food, soups, ethnic foods, health food, and prepared dinners. SOURCE: Adapted from Katsuyama et al. (1973).

industrial Food Processing Wastes 7 industry solid residuals by product and disposal method. The weight not accounted for, when it is a positive number, represents materials probably leached into wastewater and other product shrinkage in the time between weighing the raw product and processing it. When the numbers are neg- ative, the weight not accounted for is probably due to errors in estimating the percentage yields of the residual tons disposed of. Table 2 shows the same data as Table 1, but on a regional basis. PHYSICAL PROPERTIES In general, food processing wastes contain a high percentage of water, are perishable, and must be processed rapidly. The dry matter of animal processing wastes tends to be high-protein, low-carbohydrate materials that are available throughout the year. The dry matter of fruit and vegetable processing wastes tends to be low-protein, high-carbohydrate materials that are seasonably available. FRUIT AND NUT PROCESSING WASTES Apple (Malus pumila) Processing Wastes Several wastes Tom apple processing are suitable animal feeds. Apple pomace, the residual material from pressing apples for juice, contains TABLE 2 Region Food Processing Industry Solid Residuals per Year, by TotalTotal Total Total Total Not Raw Tonsas in By- Resid- Accounted Region Processed Solid Liquid Feed Other Products ualsFor New England 889 8 23104 14118154 163 Mid-Atlantic 1,868 254 699 59154417 118 South Atlantic 7,546 200 622,436 132,4492,712 299 North Central 5,342 363 251,156 181,1791,560 56 South Central 1,106 49 23195 6200272 39 Mountain 218 18 342 4262 12 Northwest 3,909 136 521,263 121,2791,460 100 Alaska 145 0 453 4652 5 Southwest 9,351 499 551,186 751,2611,824 118 U.S. Total 30,374 1,527 2946,484 2016,6888,513 910 NOTE: All figures x 1,000 metric tons; rounded (after adding).  SOURCE: Adapted from Katsuyama et al. (1973).

8 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS pulp, peels, and cores. Between 250 and 350 kg wet pomace are formed from each ton of apples pressed for juice, or 25 to 35 percent of the fresh weight of the apple is retained in the pomace after pressing (Smock and Neubert, 19501. The residual material from canning, drying, and freezing, also called apple pomace, consists of the peels, cores, and discarded apples or pieces. Either type of apple pomace may be used for vinegar or other by-product production. Pomace is also used for livestock feed (Katsuyama et al., 19731. Leaves, stems, dirt, and some other wastes are disposed of as landfill or by field spreading. Apple pectin pulp is the residue following extraction of pectin from apple pomace. Pectin extraction is less commonly practiced than in the past because citrus pectin extraction is more com- petitive (Ben-Gee and Kramer, 1969~. Apple pomace from juice pressing often contains 0.5 to 1.0 percent rice hulls that are added to aid filtration (Walter et al., 19751. Dried rice hulls were added at a ratio of 1:13 on a dry-weight basis in one study (Wilson, 19711. Other filter aids include diatomaceous earth and fiber paper (Katsuyama et al., 19731. Nutritional Value Apple pomace and pectin pulp, wet, dried, or ensiled, are suitable feeds for ruminant animals (Smock and Neubert, 19501. Apple pomace is pal- atable to cattle and sheep; pectin pulp is less palatable to dairy cows than is apple pomace, and addition of molasses was suggested to increase the palatability of pectin pulps. Smock and Neubert (19503 reported that apple pomace was unsuitable for horses and of questionable value to pigs. Average digestion coefficients of wet apple pomace for ruminants are protein, 37 percent; fat, 46 percent; fiber, 65 percent; and NFE, 85 percent. Burris and Priode (1957) found apple pomace had feeding value similar to grass silage for wintering beef cattle. Addition of rice hulls increases the fiber content and lowers the feeding value of pomace. Processing Fresh pomace spoils rapidly and must be used quickly or be preserved by dehydration or ensiling. Drying to about 10 percent moisture prevents spoilage and spontaneous combustion. Drying takes place in direct-fired, rotary-drum driers, and the pomace is then ground in hammer mills (Cruess, 19581. The processing may result in some heat damage to the protein. Apple pomace is often mixed with alfalfa or corn for ensiling (Smock and Neubert, 1950~. Cull apples may also be preserved as silage by mixing

Industrial Food Processing Wastes 9 with 20 percent alfalfa hay. Apples ensiled alone result in a very high moisture product with considerable loss by drainage. Citrus Processing Wastes Wastes from the citrus industry are very well utilized as by-products, including as feed. Approximately 39 percent of the processed fruit is not used in the primary product; 97 percent of this amount is recovered as by-products (Jones, 19731. Wastewater treatment accounts for the major remaining waste problem in the citrus industry. Recent developments have centered on the use of activated sludge wastes as animal feed and are reviewed in the section, "Citrus Activated Sludge." Hendrickson and Kesterson (1965) reviewed the processing of citrus wastes, the composition of by-products, and their utilization. The three main by-product feeds from citrus processing are 1. Dried citrus pulp, which is formed by shedding, liming, pressing, and drying the peel, pulp, and seed residues to 8 percent moisture. 2. Citrus meal and fines, which are formed and separated during the drying process. A typical processing plant produces 85 percent citrus pulp, 14 percent citrus meal, and l percent fines. Citrus meal has higher density than pulp, and higher fiber, nitrogen-free extract, and ash content. 3. Citrus molasses, which is made by concentrating the press liquor from the citrus peel residue. It is usually added to the dried citrus pulp. Most of these materials are utilized as animal feed, although citrus peel liquor and citrus activated sludge are utilized to a lesser extent. Citrus peel liquor is a by-product, similar to citrus molasses, but not as concen- trated (S. Reeder, SunKist Growers, Inc., Ontario, Calif., 1979, personal communication). The citrus peel liquor studied by Lofgreen and Prokop (1979) had a dry matter content of 47.3 percent. In feeding trials with growing beef cattle, citrus peel liquor was found to have a net energy for maintenance (NEm) of 2.24 Mcal/kg and for gain (NEg) 1.48 Mcal/kg on a dry-matter basis (Lofgreen and Prokop, 19791. Citrus peel liquor fed with cane molasses contained higher net energy values than either one alone. NEm and NEg are 1.97 and 1.32 Mcal/kg, respectively, for citrus molasses' on a dry-matter basis (National Research Council, 19761. Citrus activated sludge is produced by treating the liquid wastes from citrus processing plants. Sludge recovery and the nutritional value for poultry have been studied (Damron et al., 1974; Jones et al., 19751. Dehydrated sludge was found to be profitable. Dried sludge was acceptable as a poultry feed for up to 7.5 percent of the diet (Damron et al., 19741. Its inclusion in the diet reduced the amounts of yellow corn, soybean

1O UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS meal, and phosphorus required. However, amounts of fat and methionine had to be increased to maintain energy and sulfur amino acid levels, respectively. The dried sludge also improved skin and egg yolk color, and no off flavors were detected in egg yolks or albumin. Higher levels of dried sludge in the diet had detrimental effects. Peach (Prunus persica) Processing Wastes Peaches are graded and the percent of cull fruit determined before shipment of the entire lot from orchard to processors. Batches of cull fruit may need to be sorted in order to be accepted, or the grower may choose to dump the fruit in the orchard to avoid sorting costs. At the processing plant, cull and undersized fruit are removed. The total peach processing residual is 26 percent of raw fruit. The disposal methods used are land disposal, 63 percent; liquid waste, 5 per- cent; feed, 17 percent; and other by-products, 15 percent (Katsuyama et al., 1973~. Usually peeling is done by the use of lye solution and washing. Dry caustic (lye) peeling methods have been investigated but are less commonly used. The waste from lye peeling is either soluble or in very fine particles and is highly alkaline. Dry caustic peeling waste sludge has the following properties: 9 to 10 percent solids content, pH of 13 to 14, dark brown color, and applesauce-like texture (Smith, 19761. Culls and pieces are removed at various stages of the processing operation. Dry wastes, in- cluding trash and cull fruit, are usually used for landfill or soil application; some fruit is used as feed or for alcohol production. Screened solids are removed from the wastewater and disposed of as above (Katsuyama et al., 19731. Peach peeling slurry from wet-peel methods may be recovered from the wastewater by shaker screen sedimentation treatment, or it may be discharged into the wastewater stream. Dry peeling wastes are kept out of the waste stream (Gray and Hart, 19721. The factors that hinder the use of peach wastes as animal feed include (1) costs of transportation of the high-moisture wastes from processing plants to livestock producing areas; (2) the high sugar content of peaches, which makes drying difficult (and a suitable drying technology has not been developed); and (3) the short canning season. Little information is available on the feed value of c peach wastes. Pear (Pyrus commur~is) Processing Wastes In the past, pear waste was processed into two feed products, pear pulp (or pomace) and pear molasses. This is no longer done on a commercial

Industrial Food Processing Wastes 11 scale. The 1979 production of pears in the United States was 781,553 metric tons with 502,680 metric tons or 64 percent processed (U.S. De- partment of Agriculture, 19801. According to Katsuyama et al. (1973) about 29 percent of the raw pear tonnage was residual and 30 percent of the residual was recovered as feed. In 1950, when pear waste processing was being developed, 40 percent of the raw tonnage was residual (Graham et al., 19521. From one ton of waste, 55 kg of pear pulp (8 percent moisture) and 135 kg of molasses (50 percent sugar content) can be obtained (Brown et al., 19501. Pears are peeled both mechanically and chemically (lye peeling), but mechanical peeling predominates. Present methods of disposing of pear wastes include landfill and field spreading, with some of the solid wastes (peels, cores, and screened solids) being used as feed. Because of high transportation costs these wastes are fed fresh by local farmers (G. York, University of California, Davis, 1979, personal communication). Other methods of waste recovery have been considered and some implemented. These include fermentation to produce alcohol, methane production, and edible juice recovery. Nutritional Value According to Guilbert and Weir (1951), pear pulp and pear molasses are both very palatable to ruminants. The pear pulp studied contained a small amount of other fruit wastes. In feeding trials, steers were fed a fattening diet with pear pulp forming 25 percent of the concentrate, replacing mo- lasses and dried beet pulp. Pear pulp had a value of 70 to 75 percent of molasses and dried beet pulp, and pear molasses had a value of 115 to 120 percent of cane molasses. Pear molasses had a lower nitrogen and ash content and higher total digestible nutrients than cane molasses. Pear molasses was more palatable than beet molasses-so palatable that it could not be fed as free choice but instead had to be mixed with other feeds. Feeding trials with sheep were also conducted (Guilbert and Weir, 19511. The total digestible nutrients, dry-matter basis, for pear pulp and pear molasses were 60 and 86 percent, respectively. Processing According to Brown et al. (1950) pear wastes processed for feed and containing some other wastes had an average solids content of 13.8 per- cent. The wastes included variable quantities of peach, tomato, and grape wastes. The variability of the composition of the waste made processing more difficult. Pear waste could not be dehydrated directly and the slimy

12 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS character of the waste made pressing difficult. To overcome these prob- lems, a process was developed, and later used commercially, consisting of grinding the waste in a hammer mill, liming, aging for a short time, and pressing. The press juice was concentrated to form molasses and the press cake shredded to form pear pulp. Because of the low fiber content of the waste, a certain amount of dried pulp or other material was added to aid pressing. Fruit Canneries' Activated Sludge In several food processing industries, wastewater is treated by activated sludge methods, and the feeding value of some form of activated sludge was studied. Since many fruit and vegetable processors process several fruits or vegetables at one plant and could use activated sludge treatment, this topic is covered separately. Waste activated sludge treatment was evaluated at a plant processing apples, pears, peaches, plums, crabapples, and cherries (Esvett, 19761. The principal costs of the system were for additions of nitrogen and phosphorus because the wastes were low in these, and energy costs for aeration, sludge circulation, and sludge disposal. Sludge at the plant is currently thickened and disposed of by application to agricultural land. Physical Characteristics The sludge is highly viscous and can be moved by a bucket-type loader or belt conveyor (Esvett, 19761. When the sludge was stored for 10 months in a 208-liter drum, deterioration and production of ammonia were evident. Nutritional Value The composition of the sludge, dry-matter basis, was 39.1 percent crude protein, 3.2 percent crude fiber, 0.8 percent ether extract, 11.64 percent ash, 1.08 percent calcium, and 1.28 percent phosphorus (Esvett, 19761. In a digestibility study, biological solids (or concentrated activated sludge) were fed to steers as 2.3, 4.5, and 9.2 percent of the diet, on a dry-matter basis. The digestibility of the diets was not affected by inclusion of biological solids at the 5 percent level or less. A second study was done using biological solids in the finishing diet of steers at 2.3, 4.6, and 8.9 percent biological solids. Feed efficiency was not significantly dif- ferent from the control.

Industrial Food Processing Wastes 13 Winery Wastes Wine-making is a large industry in California, with lesser quantities being produced in New York and other states. In 1979, 2,461,324 metric tons of grapes were used for wine-making in the United States. Of this amount 96.5 percent was processed in California, 2.5 percent in New York, and the balance in other states (U.S. Department of Agriculture, 19801. Some wines are made from apples, pears, and other fruits; information on the quantity of nongrape wines produced is not available. The major wastes from wineries are stems, pomace, lees, stillage, and cleanup washwater. If distilled wines are made, the pomace becomes part of the stillage (also called still slops or still bottoms). The stillage is the residue remaining after all alcohol has been distilled. The lees are the sediments that settle out during the storage and aging of wine and that are removed during racking. Winery operations are seasonal; crushing and fermentation occurs from August to November, while distillation may continue throughout the year. The stems (approximately 5 percent of the original grape material) are usually disposed of by field spreading or burning (Stokes, 19671. Use of dried pomace as a feed declined to negligible amounts in 1966 (Amerine et al., 1972), but recently large amounts of pomace are being dehydrated for feed in California. The quantity of pomace produced has been estimated as 10 to 20 percent of the original grape weight (Ben-Gee and Kramer, 19691. Another study estimated a yield of 10 percent pomace, containing 45 percent solids (Pattee, 1947~. Amerine et al. (1972) estimated 12 percent pomace yield. The solids content of the pressed pomace may be 30 to 35 percent (J. Cooke, University of California, Davis, 1976, personal communication). Annual U.S. production of grape pomace was estimated at 45.4 to 72.6 million dry kg, and production of combined apple and pear pomace was estimated at 7.3 to 9.1 million dry kg (Prokop, 19791. In California, approximately 18.2 million dry kg are used for feed (G. Cooke, University of California, Davis, 1979, personal communication). Nutritional Value Only a few studies on the feeding value of grape pomace have been conducted, but pomace has been fed to ruminant animals in the United States and other countries for many years. The French use pomace as feed and consider it similar in quality to good hay (Amerine et al., 19721. Because of differences in wine production in the two countries, pomace from French wineries may have higher feed value. According to Amerine et al. (1972), the grape stems contain appreciable amounts of fermentable

14 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS carbohydrate. Grape pomace has a high fiber content and lower feed value than other fruit-wine pomaces. The seed content of wet pomace ranged from 20 to 30 percent (Amerine et al., 19721. The seeds have a fibrous hull that decreases the feeding value of the pomace. If the seeds were hulled and pressed for oil extraction; the presscake would be valuable as animal feed. The pomaces are high in moisture and generally rather fi- brous. Protein and energy are noorlv divested, the digestibility of protein being ~o per`;en~ anu tne ~ values ranging from 24 to 30 percent, air-dry basis. These values were obtained in a study in which grape pomace was fed at a level higher than 50 percent of the diet, with alfalfa hay in one trial and as 100 percent of the diet in another (Forger, 19401. Prokop (1979), in a more recent study, fed grape pomace at 20 percent of the diet. Net energy values obtained with beef cattle on finishing diets were: NEm, 0.75 Mcal/kg, and NEg, 0.41 Mcal/k~ on a dry-matter basis. In the same study, a combination of apple, pear, and grape pomaces and apple pomace were also tested at the 20 percent level. On a drY-matter ~7 1 ~ ~ 1 ~1_ _ ~% T 1 - basis, the value of apple winery pomace was comparable to beet pulp. The combination was similar in nutritive value to dried alfalfa pellets (20 percent protein). The value of pear pomace was estimated at 89 percent of the value of the apple pomace (Prokop, 19791. In a recent study in Cyprus, Hadjipanayiotou and Louca (-1979) fed dried grape pomace (also called grape mare) as 15 and 30 percent of calf fattening diets. Urea was added to compensate for the low digestibility of grape pomace protein. The composition of the grape pomace was very similar to that reported by Prokop (19791. On a dry-matter basis, crude protein was 12.3 percent and was 19.5 percent digestible. The digestion coefficient for dry matter was 28.4 percent. Metabolizable energy was 1450 kcal/kg, dry-matter basis approximately half that of barley. The feed intake and gain of the calves were not significantly different when grape pomace formed 0, 15, or 30 percent of the diet. However, there was a significant difference in feed efficiency between the 0 and 30 percent levels. The dressing percentages of the calves fed 30 percent grape pomace tended to be lower. Most of the pear and apple winery pomaces are being used as feed. Some is being dried for this purpose, but most is being fed wet, (M. J. Prokop, University of California, E1 Centro, 1979, personal communi- cation). Processing Wet grape pomace stores well in compacted piles; only the outer layer deteriorates (Stokes, 19674. Large quantities can be dried in rotary drum

Industrial Food Processing Wastes 15 driers and ground in hammer mills. Small quantities of apple and pear pomaces may be combined with the grape pomace for drying. The non- grape pomaces are more difficult to dry, and combining with grape pomace makes drying easier. The pomace not being used as feed is spread on land. Cacao (Theabroma cacao) Processing Wastes Commercial cacao beans are imported into the United States for chocolate and cocoa manufacture. Imports of cacao beans into the United States in 1979 were 167,881 metric tons (U. S. Department of Agriculture, 19801. The two major uses of cacao bean shells are as animal feed and nursery mulch. Calculated yields of shell available for by-product recovery ranged from 8 to 12 percent of the commercial beans, depending on the efficiency of winnowing, the quantity of shell-nib mixtures, and the variety of bean (Chats, 19531. Physical Characteristics The shells are brittle and, depending on the processing equipment, may be broken into fine pieces, but they are not ground intentionally at this stage. Feed manufacturers may grind the shells prior to incorporating them into feed mixtures. In California more shells are used as feed than in other areas of the country because cacao shells are not used as extensively for mulch in California, other mulches being preferred. Nutritional Value The fat content of the shells is about 3 to 3.5 percent and varies with the amount of fat transferred during roasting and the quantity of nib present (the nib is the cotyledon of the cacao from which various chocolate and cocoa products are made) (Chats, 19531. The ash content varies from 5.5 percent (Chats, 1953) to 10.8 percent (Morrison, 1956) dry-matter basis. The crude protein content is 13.5 to 16 percent but has low digestibility (Chats, 1953; Morrison, 19564. Crude fiber is 16 to 20 percent (Chats, 19531. The shells also contain considerable vitamin D (Morrison, 1956~. Fruit Pits, Fruit Pit Kernels, Nut Hulls, and Nut Shells Total U.S. nut shell and fruit pit wastes have been estimated at over 453.6 million kg annually (Mantel!, 19751. Nut shells and fruit pits are burned as fuel, made into charcoal, or used as landfill. Some of the charcoal made from nut shells and fruit pits is used in animal feeds. Shells from

16 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS peanuts, almonds, pecans, and filberts can be used as mulch, but must compete with many other waste products. Almond and peanut shells can be used as poultry litter. In general, fruit pits (except kernels) and nut shells have not been found to be useful livestock feed materials. They are now being burned as energy sources at fruit and nut processing plants. VEGETABLE PROCESSING WASTES Potato (Solanum tuberosum) Processing Wastes The potato is processed into many different forms, usually at separate plants. In 1978, 7.95 million metric tons were processed (U.S. Department of Agriculture, 19801. The products include dehydrated potatoes (19.1 percent); chips and shoestrings (22.1 percent); frozen trench fries (45.5 percent); other frozen products (8.9 percent); canned potatoes and other canned products, such as soups, hash, and stews (2.7 percent); and starch and flour (1.7 percent). Potato starch production results in quite different waste products and will be covered separately. In 1971 it was estimated that 33 kg of solid waste were generated per 100 kg of potatoes processed (Jones, 19731. Of this, 29 kg (88 percent) of solid potato waste is turned into by-products, mostly feed. The other 12 percent is handled as solid waste, and some of this is recoverable as feed. About 5 percent of the potato crop was fed to animals in the United States in 1966 (Ben-Gee and Kramer, 1969~. In 1971 only about 3 percent of the fresh crop was used as feed (Hogan and Highlands, 19751. Physical Characteristics Wastes from potato processing include ( 1 ) the large solids that are removed by screening (peels, culls, trimmings), (2) the smaller solids that are removed by primary treatment, and (3) less easily separable solids and dissolved solids that may be treated by secondary treatments or used as substrate for yeast or fungal growth. The large solids are known as screening or screen solids. The solids removed by primary treatment may be referred to as primary solids, sludge, slurry, or filtered potato sludge. The primary solids include pulp, the more finely divided pieces from which the solubles have been leached. The effluent from primary treatment contains solubilized starch, pro- teins, amino acids, and sugars (Pailthorp et al., 19751. This effluent may receive secondary treatment, the resultant sludge being called waste bi- ological solids, bacterial mass, or waste activated sludge.

industrial Food Processing Wastes 17 Potato by-product meal is produced from the screenings and other wastes of products for human consumption. It also may include cull potatoes and potato pulp from starch manufacture (Dickey et al., 19711. Limestone is added to mixed wastes to aid drying and the mixture is dehydrated. Each waste material is metered into the mixture so that the resultant by-product . ~ . . ~ IS lair. y UIlllOrm. Alkaline potato peels from dry caustic peeling have a pH of about 12, and solids content of 12 to 15 percent. The peelings can be used to neu- tralize straw that has been steam treated under pressure. This is a patented process (R. P. Graham, U.S. Patent 3,692,530, September 19, 19721. Potatoes have been sun-dried on abandoned concrete airstrips in Cali- fornia (Hogan and Highlands, 1975) and have been "freeze-dried', in the Red River valley and in Colorado by spreading on pastures in the winter. Cattle are allowed to feed on the partially dried potatoes in the field. The Symba-yeast process is a method of deriving a food or feed product from waste effluents including primary solids (Skogman, 1976~. The dry matter content of the wastewater should be over 2 percent. This can be increased by wastewater recirculation or adding solid waste. The produc- tion season is fairly long. In the Aspergillus niger protein process, starchy material is homoge- nized to increase surface area; inorganic nitrogen and other nutrients are added; and the material is heated, inoculated with fungus, and aerobically cultured for 24 to 36 hours. The product is then sterilized again and filtered (Gillies, 1978) before being used in the preparation of feeds. About 200 to 225 kg of pulp solids are produced per ton of starch produced (Treadway, 1975~. Protein water, also known as fruit water and starch washwater, usually contains 1 to 3 percent solids (Treadway, 1975~. About 225 to 325 kg of protein water solids are produced per ton of potato starch. Nutritional Value Primary solids and slurry are usually dilute, containing 4 to 7 percent solids. These may be concentrated by belt-type vacuum filters (preferred method) or by centrifuges to 15 to 18 percent solids (Pailthorp et al., 19751. When fed to calves as 25 to 50 percent of concentrate mixture, primary solids were equivalent to ground barley (Hogan and Highlands, 19751. The dry matter was 73.5 percent digestible and contained 1.498 kcal ME/kg when fed at 19.2 percent of dry matter of the diet and 1.259 kcal ME/kg when fed at 37.5 percent of dry matter of the finishing diet of steers. It can be fed at 40 to 50 percent of dry matter of a steer diet before a reduction in daily gain occurs (Hogan and Highlands, 1975~.

18 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS Primary solids from caustic peeling plants have a high sodium hydroxide content and cannot be fed at high levels (Grames and Kueneman, 19691. The sludge can be stored prior to filtration, and fermentation will cause the pH to drop and increase filterability. The product can then be fed in combination with other feeds, as is presently being done at some plants. Some plants run their own feedlots to dispose of the sludge. The price obtained for primary sludge is sufficient to pay for processing. The alkaline peel-straw product has a pleasant molasses-like odor. The solids content is about 20 percent, and represents a combination of 1.9 parts steamed straw and 1 part potato peels, on a dry-matter basis. The product has a pH of about 7 and a total-soluble-after-enzyme in vitro digestibility of 80 to 85 percent. The product from the Symba-yeast process has been tested in feeding trials with calves as 40 percent replacement for milk proteins (Skogman, 19761. The limit is about 25 percent of poultry diets. It has also been tested as a portion of the diet of rats, pigs, mink, and house pets. The yeast product is low in nucleic acids, 4.25 percent dry-matter basis, and has high B-vitamin content. There is no difference in the composition of the yeast made from different waste materials. The ash content in yeast from lye peel water may be higher. The Aspergillus niger protein process fungal product contains 37 percent protein (Rogers and Coleman, 19784. Fresh potatoes are fed to ruminants and horses. Swine and other mon- ogastrics should have potatoes cooked to increase digestibility (Hogan and Highlands, 19754. Fresh potatoes can be stored by ensiling with a mini- mum of 20 percent dry corn stover or hay. Cooked potatoes can be ensiled without roughage (Hogan and Highlands, 19751. Potato silage that includes 20 percent corn fodder or alfalfa hay is approximately equal in value to corn silage for cattle and sheep. Raw potato has lower apparent digestibility of nitrogen and lower efficiency of nitrogen utilization in pigs. Chymo- trypsin inhibitor activity is high in raw potato and absent when cooked (Whittemore et al., 19751. Cooked potatoes are equivalent to corn in nutritive value for fattening pigs; however, the high moisture content reduces feed intake. Cooked potato flakes have been studied in feeding trials with chicks and pigs (D 'Mello and Whittemore, 19751. The ME energy value of cooked potato flakes in chicks is 3.26 Mcal/kg, dry-matter basis. Growth rates and efficiency of feed conversion were lower in chicks fed cooked potato flakes than in those fed glucose or corn. Up to 20 percent cooked potato flakes in pelleted diets allowed satisfactory performance; 30 to 40 percent had adverse effects on growth and efficiency. Feed intake was not de- pressed by potato flakes. The protein quality of cooked potato flakes was comparable to that of ground corn.

Industrial Food Processing Wastes 19 When cooked potato flakes were fed to piglets, better results were obtained than when fed to poultry (Hillyer and Whittemore, 19751. The digestibility of energy was 96 percent and of nitrogen, 89 percent. Cooked potato flakes contained a DE of 3.94 Mcal/kg, dry-matter basis. Diets containing 50 to 58 percent cooked potato flakes were equal in nutritive value to those containing corn. The potato flakes contain slightly less sulphur amino acids and slightly more lysine than cereals. The digestibilities of dry matter and nitrogen by calves decline when cooked potato flour replaces spray-dried whey in liquid diets (Hinks et al., 1975~. Wet pulp is commonly fed fresh to cattle in Europe. Pulp silage is very palatable to cattle (Brugman and Dickey, 19551. Dried potato pulp is high in sugar, low in fat, moderate in fiber, and low in vitamins. Manufacturers often add 2 to 6 percent molasses to the pulp. Digestible crude protein is 6 percent and TDN is 79 percent for cattle. It is equivalent to yellow hominy feed at 22.5 percent of the grain mixture. Dried potato pulp may have too high a fiber content for hogs but could be used as a portion of grain. It is assumed that potato pulp would have a value for hogs similar to that for cattle and other ruminants but little information is available. The fiber content of dried potato pulp is too high for poultry, but the pulp may be used to limit energy intake for replacement stock (Brugman and Dickey, 19614. Dried pulp can be used to improve grass silage and as a preservative in silage. It has excellent water-holding capacity, 210 kg of water per 100 kg of pulp. About 60 percent of the total solids of protein waste water are nitro- genous compounds, two-thirds of which is free amino acids, and one- third protein (Treadway, 19751. The nonnitrogenous solids include sugars, organic acids, inorganic salts, and a few other minor constituents. A feeding study was done with chicks using whole fruit water concen- trate meal that contained 50 percent protein, dry-matter basis (Rosenau et al., 19771. This product slowed chick growth rates if used above the 6 percent level. This was thought to be caused by lysine loss due to Maillard browning. The presence of trypsin inhibitor in potato products requires that the product be heat-treated in order to inactivate the inhibitor. Processing Methods The waste effluent from primary treatment can be treated by numerous types of secondary treatments (Guttormsen and Carlson, 1970; Richter et al., 19731. Most potato processors have primary treatment but fewer have secondary treatment, although increasing numbers are applying secondary treatment as pollution standards are raised. More study has been done on ;

20 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS starch-plant soluble waste recovery than on potato processing plant soluble wastes. The yield of biological solids from activated sludge process is large; about one unit of biological solids was produced per unit of biological oxidation demand (BOD) removed (Richter et al., 19731. Wet pulp is difficult to dewater by pressing because of its slippery texture and high water holding capacity. Treating the wet pulp with 0.3 to 0.5 percent Ca(OH)2 for 20 minutes prior to pressing improves water removal. In Germany it was reported that pulp is pressed to reduce water content to 70 to 75 percent by using a double drum centrifugal press. This pulp was then dried and used as cattle feed (Hogan and Highlands, 19751. In Maine pulp was pressed or centrifuged to 79.5 to 83 percent water (Brug- man and Dickey, 19551. Pulp can be dried in a steam tube drier or direct- fired rotary drier. Pulp is commercially dried in Maine by means of an add-back process and is not pressed prior to drying (Hogan and Highlands, 19751. Wet pulp, if stored, ferments rapidly. After 8 days in storage, pH drops to 3.7 (Brugman and Dickey, 19551. It will drop further over time to 3.5 and store well for extended periods. Silage may be enriched with protein that has been extracted from fruit water by precipitation with organic acids (Ben-Gee and Kramer, 19691. Studies have been concentrated on the recovery of the various constit- uents of protein water from production of potato starch (Osten, 19761. Proteins are in fairly low concentrations in the water, usually 1.1 percent solids. This can be increased to 3.1 percent solids by using a starch recovery process that uses less water. Efforts have been made to decrease the amount of water used by changing the starch recovery process, and efforts have also been made in numerous methods of protein extraction and/or protein water concentrations. Rosenau et al. (1976) described a starch recovery process using greatly reduced amounts of water, yielding a protein water with 4 percent solids and 50 percent protein, dry basis. Rosenau et al. (1977) mentioned a process yielding a juice stream with 5 percent solids. Four methods of constituent recovery of protein are evaporation, ul- trafiltration, coagulation of protein by steam injection heating and con- centration of filtrate, and the five-step method of constituent recovery. Evaporation was used to concentrate fruit water solids (Stabile et al., 19711. The protein water is evaporated in a triple-effect evaporator to 60 percent solids. This slurry could be mixed with dried potato pulp for use as feed. A yield of 24 tons per day of mixed feed is postulated from a 27 ton-per-day starch plant.

Industrial Food Processing Wastes 21 Della Monica et al. (1975) studied the stability of protein water slurry. It was found that the product would require final drying. Whole juice concentration and spray or drum drying yields 50 percent crude protein meal. Sufficient heating is necessary to inactivate the trypsin inhibitor. Rapid heating of fruit water by steam injection and acid precipitation can be used to coagulate about 35 percent of the protein (Rosenau et al., 19771. The protein product does not have antitrypsin activity. The end product contains 8 percent moisture and 75 to 80 percent protein on a dry-matter basis. This process has a high energy requirement (Osten, 19761. The deproteinated juice is then concentrated to 70 percent solids. Because of lysine destruction, Rosenau et al. (1977) suggested mixing it with pulp and feeding it to ruminants. By-product yields were estimated at 1.1 kg of protein meal (10 percent moisture) and 8.9 kg of pulp (10 percent moisture) from 100 kg of potatoes processed for starch. Starch production from 100 kg of potatoes would be 12.2 kg of starch ( 1 8 percent moisture). The juice may be processed further by heating to 60°C in a plate-type regenerative heater and then to 121 °C by steam injection. This coagulates about 35 percent of the crude juice protein so that it can be removed as a 20 percent solids (70 percent protein, dry basis) sludge by a nozzle-type disk centrifuge and dried in a spray drier. The remaining deproteinated juice (43 percent crude protein, dry basis) is concentrated by reverse osmosis to 10 percent solids and by multiple-stage evaporation to 65 percent solids. It is stable in this condition and could be useful as a molasses substitute for ruminant feeding (Rosenau et al., 19781. Sweet Potato (lpomeea batatus) Processing Wastes Sweet potatoes are grown mostly for human consumption today, although they have been grown for animal feed. They are sold fresh, canned, or frozen. The production of waste from conventional sweet potato canning can be divided as follows (Colston and Smallwood, 19721: 5 percent of the weight of incoming sweet potatoes is dirt, which is removed by screening and washing and is disposed of as landfill; 9.5 percent of original potatoes is culled (cull potatoes can be used as cattle feed); 40 percent of the original becomes the canned product; 45.5 percent of original is solid waste. According to Smallwood et al. ( 1974), snips and trimmings are removed by 10-mesh screen, are handled as solid waste, and account for 20.5 percent of the original potatoes. These have been disposed of as landfill but could be removed as feed. Twenty-five percent of the original potato appears in the waste stream after screening. The waste effluent has a pH

22 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS of 9.5 to 1 1 .5. A large amount of the 25 percent original could be recovered by a two-stage screening system. Nutritional Value Sweet potato meal was equivalent to corn when fed to cattle at less than 50 percent of the grain portion of the diet (Bond and Putnam, 19671. If sweet potato meal is used to replace all of the grain in the diet, lower performance results. As a large portion of the diet, sweet potato meal is less palatable than grain and can be laxative (Morrison, 19561. If cottonseed meal is added to sweet potato meal to bring its protein content up to the level in corn meal, the sweet potato meal, in general, is equivalent to 90 percent of corn meal when fattening steers (Edmond and Ammerman, 19711. Studies with lactating cows also showed sweet potato meal to have 90 percent of the value of corn meal. Feeding sweet potato meal to cattle and sheep has produced satisfactory results (Morrison, 19561. When used at levels up to 50 percent of the grain mixture, this feed has been equal or nearly equal to corn for dairy cows. When it replaced up to 50 percent of the grain for fattening cattle, it has been found to be worth 95 to lOO percent of the value of corn. Blanched sweet potato meal may be equivalent to corn meal for pigs, but not enough research has been done to draw definite conclusions. Studies have shown that sweet potatoes are less economical to grow for feed than corn; however, culls and processing wastes, when dehydrated or dried, can be used as feed (Edmond and Ammerman, 19711. Bond and Putnam ( 1967) studied the feeding of dehydrated sweet potato trimmings to steers. A diet with 51 percent trimmings was compared to a similar diet with 51 percent cracked corn. Sweet potato trimmings had a value equivalent to 80 percent that of corn and were not equivalent to dried whole sweet potato meal, which had a value nearly equal to that of corn. The trimmings were coarse and hard, lowering the feed intake. Grinding improved feed intake and subsequently improved the weight gain of the steers. Trimmings should be ground for best results. The steers fed trimmings had higher carcass grades but lower dressing percentage and less marbling. The digestibilities of dry matter and crude protein of the trimmings were lower than for corn (Bond and Putnam, 19671. Processing Methods Solid wastes from canning, cull sweet potatoes, snips, and trimmings are easily recovered (Smallwood et al., 1974J. They are handled as a group and made into dry sweet potato meal, or the snips and trimmings are handled separately and made into a product called sweet potato trimmings.

Industrial Food Processing Wastes 23 Other wastes are carried as liquid waste from which solids can be recovered by screening. The screened solids include caustic peelings, abrasive peel- ings, and solids from washings and spillage. Screened solids may also contain snips and trimmings if they have not been handled separately. Trimmings and screened wastes are less well utilized than sweet potato meal. It was found that the cost of drum drying conventional waste was too high (Smallwood et al., 19741. Wastes from the "dry peelers" can be mixed with trim wastes and fermented to reduce the high pH. It was also noted that the cost of drying sludge from sweet potatoes was higher than for drying sludge from white potatoes. In sweet potato processing, peel losses are higher, the dry-matter content of the potatoes is higher, and more lye is used. Tomato (Lycopersicon esculentum) Processing Wastes Tomatoes are used for many canned or heat-processed products. The quantity of tomatoes processed has increased greatly in the past few de- cades, tripling in the 25-year period prior to 1975 (Brands et al., 1978~. Fifteen percent of processed tomatoes are used for canned whole tomatoes, a process which requires peeling, and the remaining 85 percent are used for pulped products (Schultz et al., 19761. Tomato processing wastes can be divided into three categories according to the type of by-product that can be recovered. Cull tomato feed is made from cull tomatoes processed into tomato pulp or tomato pomace. A second category is peel residue, a waste product from canning whole tomatoes; about 12 percent of the original tomato is removed as peel and adhering pulp. Peeling residue in California in 1974 was estimated to be 170,000 tons (Schultz et al., 19761. The third category is tomato pomace, the residue from the manufacture of juice, paste, puree, sauce, and catsup. Pomace contains the skin, seeds, fiber, and some adhering pulp. Physical Characteristics The peel residue from caustic rubber-disc peeling has a pH of 13 to 14; a solids content of 7 to 8 percent; a bright red color; and an easy-flowing texture like a tomato puree with many skins. The pH of the wastewater from dry caustic peeling is 6.2 compared to 9.4 for wet caustic peeling (Smith, 19761. Nutritional Value The ash and nitrogen-free extract levels are higher in cull tomato feed than in tomato pomace, and fiber is lower (Ammerman et al.,1965~. Cull

24 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS tomato feed has been used to replace citrus pulp as 10 to 30 percent of the concentrate in steer diets and has produced similar results (Ammerman et al., 19631. It has been studied in the dry form. In another study, cull tomato feed has been fed as 70 and 100 percent of steer diets. Digestibility decreased slightly when cull tomato feed was fed alone. The average TDN for the two diets was 64.8 percent, dry-matter basis. No problems were encountered, although the steers seemed to be getting insufficient fiber from cull feed alone. In a study with lambs the protein of cull tomato feed at 33 percent of the diet was less digestible and of lower biological value than that of soybean meal (Ammerman et al., 19631. On a dry-matter basis, cull tomato feed had 24 percent crude protein and 12.2 percent apparent digestible protein. In steers the apparent digestible protein was 13.5 percent. Tomato cull feed satisfactorily replaced alfalfa meal as 3 percent of the diet for poultry (Ammerman et al., 19651. Carotenoid pigmentation in skins and shanks, however, was reduced. A level of 12 to 15 percent tomato pomace in cattle feed produced good results (Esselen and Fellers, 19391. Morrison (1956) reported that tests had been conducted with dairy cows in which pomace was included as 15 percent of the concentrate mixture with satisfactory results. Tomato pomace is palatable to hogs, cows, and chicks as 10 to 15 percent of the diet, but fed at higher levels, bitterness may make pomace unpalatable. Dried tomato pomace has a content of about 25 percent crude protein, 15 percent ether extract, 22 percent crude fiber, and 3 percent ash, dry-matter basis. It is a good source of some B-vitamins and a fair source of vitamin A (Esselen and Fellers, 1939~. More recently, feeding trials have been conducted with sheep (Hinman et al., 19784. Tomato pomace (wet) was fed at levels up to 77.5 percent of the diets, dry-matter basis. After the sheep became used to the pomace they consumed it readily. The crude protein level of all diets was about 20 percent, dry basis. The pomace had slightly lower digestion coefficients for dry matter, organic matter, and crude fiber than alfalfa, but had slightly higher coefficients for cellulose, nitrogen-free extract, and ether extract in diets with large amounts of pomace. In a second trial, lower digestibility of protein resulted when dried pomace was fed. This may have been caused by heat damage of the protein during drying. One problem found with pomace was the variability of dry-matter content; it ranged from 11.9 to 27.5 percent. Some dried, bagged pomace is used in pet food (Katsuyama et al., 1973) and for fur-bearing animals (Schultz et al.,1976), although the portion recovered and used as animal feed in the United States is small. Tomato pomace has been used in dog foods and fox and mink diets (dry

Industrial Food Processing Wastes 25 pomace at 5 percent of wet ration) to prevent diarrhea (McCoy and Smith, 1940). Tomato pomace from tomato juice production was fed to rats to study the protein quality (Esselen and Fellers, 1939~. The dry tomato pomace contained 20.9 percent crude protein, 17 percent ether extract, 12 percent crude fiber, 46.4 percent nitrogen-free extract and 3.6 percent ash, on a dry-matter basis. The protein efficiency ratio (PER) of dry tomato pomace unsupplemented with amino acids was 2.18, and net protein utilization (NPU) was 0.55. Tomato seed presscake has been studied as a feed ingredient. Rats grew at a normal rate using presscake as the sole protein source. Amino acid studies have been conducted on tomato seed presscake, tomato seeds, and tomato skins (Tsatsaronis and Boskou, 19751. Other Vegetables In general, when vegetables are harvested, as much of the human nonfood plant material as possible, including leaves, stalks, stems, and pods, is left in the field. Some of these are unavoidably brought into the packing or processing plant and become part of the waste residual, along with cull, broken, or unacceptable vegetables. Also included in vegetable wastes are stems from leaf vegetables, cobs and husks from corn (Zea mays), and cores from cabbage (Brassica oleracea capitata). Most vegetable wastes are disposed of and not dried because the moisture content is too high to enable the wastes to be burned. The main value of leaf meals appears to be as a pigment and vitamin source for poultry. Ben-Gera and Kramer (1969) mentioned the use of dried celery (Apium graveolens) tops as cattle feed. Of the wastes discussed only pimento (Pimenta officinalis) wastes and possibly artichoke (Cynara scolymus) wastes are currently being dried (Livingston et al., 19761. The following is a review of information on dehydrated vegetable wastes (Katsuyama et al., 19731. Although many of these wastes are fed to livestock, information is not readily available on their feeding value. Asparagus (Asparagus officinalis) waste includes butt ends and broken spears. Most wastes are carried by water, and the screened wastes are used for animal feed. Asparagus waste was dehydrated, ground in a ham- mermill, and fed to sheep and dairy cows (Forger, 19401. The composition of the meal was 17.2 percent crude protein, 39 percent nitrogen-free extract, 1.1 percent ether extract, and 35.1 crude fiber, on a dry-matter basis. Folger found that the dried waste could be substituted for fair- quality roughage. It was palatable to sheep but less palatable to dairy cows. TDN was 51.7 percent, on a dry-matter basis.

26 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS  11 Artichoke (Cynara scolymus) waste was dehydrated in a pilot plant study (Livingston et al., 19761. The solids content of fresh trimmings during the harvest peak was 16 percent. The protein content of unpressed dried meal ranged from 17.4 to 23.1 percent and that of the pressed dried meal from 14.2 to 21 percent. Pressing increased crude fiber content of the meal. The carotene and xanthophyll contents were too low for use as a poultry pigmentation supplement but would be adequate for most animal feeds. The amino acid and mineral compositions were also studied. Lima beans (Phaseolus limensis) are threshed either by mobile field viners or by stationary vipers. Vines and pods from field viners are plowed under, but at stationary viners these may be collected for feed use. Quan- tities of vines and pods are not included in residual figures (Table 11. The processing plant wastes, including leaves, pods, and pieces of vines, are handled dry and disposed of on land, but screenings from the wastewater may be fed to animals. Snap beans (Phaseolus vulgaris), usually of the bush type, are machine harvested. In the processing plant the remaining field trash dirt is removed by shaker sieves and updraft blowers. Snipped ends and culled beans are carried in the wastewater and are recovered by screening. Screened solids may be mixed with other wastes and used as feed or field spread. Table beets (Beta vulgaris) are machine harvested and topped in the field. At the processing plant, dirt and remaining leaves are removed. Beets are steam peeled. Fragments, undersize beets, and screened solids form the waste. Only a small portion of beet waste is used as feed. Cabbage (Brassica oleracea capitata) for sauerkraut is harvested both mechanically and by hand. Most of the outer leaves are left in the field; the remaining outer leaves and cores are removed in the processing plant and are usually handled dry. Material spilled during various processes is included in the waste. Only a small amount is used as feed. Carrots (Daucus carota) are mechanically harvested, topped, and often sorted in the field. At the processing plant the carrots are graded and sorted, with small, split, and woody carrots being discarded. The carrots are lye or steam peeled. Trimmed crowns are carried by water and removed by screening. The discarded carrots and screened solids are spread on fields and cattle are allowed to consume the wastes. Cauliflower (Brassica oleracea botrytis) leaf waste has been dehydrated  n a pilot-scale alfalfa dehydrator. The dry-matter content of the fresh waste is 7 to 10 percent. The dried meal is separated by air classification into poultry and cattle feed fractions. The poultry meal fraction contains 26 to 31 percent protein and 375 to 620 mg/kg xanthophyll. Individual xanthophyll ratios are similar to those found in alfalfa meal. The poultry

Industrial Food Processing Wastes 27 meal fraction, fed to chicks at 2 to 10 percent of the diet, was found to be palatable. Pigmentation was comparable to that provided by corn gluten or dehydrated alfalfa meal when fed at similar xanthophyll levels. The cattle feed fraction contained 17 to 20 percent crude protein and 14 to 16 percent fiber (Livingston et al., 19721. Sweet corn (Zea mays) is machine harvested for human consumption, and the stalks and leaves are left in the field. Ears are air cleaned at the processing plant, ends mechanically trimmed, and husks and silks me- chanically removed. The unusable portions of the ears are then trimmed away. The corn is washed and kernels are cut from the cob. Broken kernels, silks, and fragments are removed by water-assisted screening, shaker sieves, and air cleaners. These wastes are screened from the wastewater. The cobs, husks, leaves, and stalks are chopped, mixed with screened solids, and ensiled or stacked. Corn waste is sold as feed. The corn waste at one plant contained 90 percent husk and leaf, 8 percent cob, and 2 percent kernel and had 30 to 40 percent dry matter. The corn screening had only 5 percent solids content. The wastewater contains a significant organic load and has potential for recovery by methods similar to those used for potato wastes. Peas (Pisum sativum) are mechanically harvested and vines and pods are removed by mobile or stationary viners and used for feed. At the processing plant the-peas are dry cleaned, then washed. Remaining leaves, vines, pods, and pea shells form the waste. During grading, unacceptable peas are removed. The wastewater is screened for solids. The quantity of wastes handled dry or allowed to enter the waste stream varies between plants. Dry and screened wastes are fed to animals or field spread when animal feeding is not feasible. Pimento (Pimenta o~icinalis) waste consists of pulp, skins, and seeds. The waste is dried and has been found to be rich in red xanthophylls. The xanthophyll content ranges from 896 to 1114 mg/kg. The wastes contain 1 1.9 percent protein, 9.48 percent fat, and 44.5 percent fiber, on a dry- matter basis. The antioxidant ethoxyquin is effective in preventing caro- tenoid losses in storage. Levels of 1 percent pimento meal fed in com- bination with alfalfa meal are considered adequate in poultry diets. Red xanthophylls that contain ketone groups when fed alone impart a pink or reddish color to egg yolks and poultry skin (Livingston et al., 19741. Pumpkins (Cucurbita pepo) are harvested by hand or machine, and vines are left in the field. At the processing plant the pumpkins are washed and the vine ends trimmed. Pumpkins are split or chopped into large pieces and washed by machine; the seeds are removed manually. Unusable pieces are discarded. The pieces are then steamed and pressed; the solids from

28 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS the pressed liquid may be centrifuged and returned to the product. The pressed pumpkin is pulped and finished. The residual from the finisher contains seeds, skins, and fiber; seeds and fiber are also screened from the wastewater. Some seeds are processed for human consumption, but the other solid wastes are used for animal feed. A significant amount of fine particles passes through screens into the wastewater. Spinach (Spinacia oleracea) and other greens are machine harvested. The dirt is removed by rotating screens, and weeds, discolored leaves, and roots are manually removed and disposed of. These are not used as feed. Following washing, the screened solids are fed to livestock. ANIMAL BY-PRODUCTS Dairy Whey Whey is the largest waste in the dairy industry and results from the production of cheese. Other dairy wastes, which are insignificant by com- parison are (1) adulterated milk, which may be disposed of by dumping or irrigating, or may be powdered and used as animal feed; (2) rinsewater from tanks, lines, and equipment containing milk solids, which are re- cycled into nonfluid food products. Previously rinsewater was disposed of as liquid waste. Marketing of whey products and economic considerations seem to be the reason whey continues to be a waste problem (Jones, 1974~. Quantity Whey processing and utilization was reviewed by Clark (19791. Whey production was calculated from data on cheese production in the United States. Manufacture of cheddar cheese results in 9 kg sweet whey/kg cheese produced. Cottage cheese manufacture results in production of 6 kg acid whey/kg cheese produced. In 1976 it was estimated that 15.6 million tons of liquid whey were produced, and about 88 percent of this was sweet whey; the equivalent whey solids were 1,011 million kg. About 57 percent of the whey solids were further processed into human food or animal feed products; 438 million kg were wasted. About 33 percent of the whey solids that were further processed were used in animal feed. Physical Characteristics Commercial whey products are numerous and include condensed whey from acid and sweet whey; dry whey products, which include dry whole

Industrial Food Processing Wastes 29 whey, partially delactosed dry whey, and partially demineralized dry whey; and lactose and whey solids wet blends. Chemical Composition Raw whey is dilute and contains 93 to 94 percent water, 0.7 to 0.9 percent crude protein, 0.5 to 0.6 percent fat, 4.5 to 5 percent lactose, 0.2 to 0.6 percent acid, and 0.5 to 0.6 percent ash (Jones, 19741. Dried whole whey contains 13.1 percent crude protein, 76.9 percent lactose, 9.0 percent ash, 0.98 percent calcium, and 0.76 percent phosphorus, dry-matter basis (Schingoethe, 19761. Nutritional Value Liquid whey has been used as an animal feed for centuries. It can be fed at levels up to 30 percent of ruminant diets and 20 percent of swine diets, dry-matter basis. Because of its high water content it is usually used near the cheese plants where it is produced. Alternatively, it may be processed into more concentrated forms, but a major problem is the cost of dehy- drating, especially at smaller cheese plants. Schingoethe (1976) gave an excellent review of the feeding value of whey and whey products. Con- densed whey has been fed to ruminants and swine, and fermented am- moniated condensed whey has been fed successfully to cattle and sheep (Schingoethe, 1976~. Dried whey products have been fed to all types of livestock, including poultry; however, poultry have a lower lactose tol- erance than other animals, and dried whey should be limited to a maximum of 10 percent of the diet. Levels of 3 to 4 percent dried whey are considered optimal. Ruminants appear to be able to consume very large quantities of liquid or dry whey products. Results from feeding trials determining the upper limit for feeding whey products to ruminants have varied. Calves can be fed milk replacers con- taining up to 89 percent dried whey. Dried whey is also useful for im- proving silage quality. Whey protein concentrates can be recovered from whey; whey proteins are very high quality proteins. The limiting amino acids are phenylalanine and tyrosine. Whey protein concentrates, prepared by various methods, usually contain 50 to 75 percent protein, 20 to 30 percent lactose, and 5 percent or less ash, dry-matter basis. The production of whey protein concentrates leaves a large residual of deproteinized whey. Deproteinized whey has been formed into blocks containing 0.6 percent nitrogen, 70 to 72 percent lactose, and 12 percent ash. The blocks have been used as licks for cattle and calves.

30 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS Processing Methods Bough and Landes (1976) reviewed methods of recovering proteins from whey and investigated the use of chitosan in coagulating whey proteins. Following freeze drying, the coagulated solids contained 72.3 percent protein, 6 percent lactose, 9.5 percent ash, 6.8 percent moisture, and 2.15 percent chitosan. The coagulated solids were fed to rats and the PERs obtained were similar to those for casein and whey solids without chitosan. Fermented, ammoniated whey (FACW) is formed by fermenting the lactose in sweet or acid whey while maintaining the pH with ammonia to produce ammonium lactate (Juengst, 19791. The FACW is condensed to 55 to 65 percent solids; a 60 percent solids product blends well, handles easily, and is stable. It contains 45 percent crude protein, 37 percent lactic acid as ammonium lactate, and 4.7 percent ash, dry~matter basis. In the past few years research has been conducted on FACW as a feed supplement for dairy and beef cattle. Results of these trials were reviewed by Juengst (19791. In feeding trials, FACW was found superior to urea as a source of nitrogen. Seafood Processing Wastes The fish processing industry uses many types of fish and shellfish and is widely spread along coastlines and estuaries. The quantity of waste varies tremendously among plants and among fish and shellfish types, from O percent for whole rendered fish to 85 percent for some crabs; the average for all types of seafood wastes is about 30 percent. Soderquist and Wil- liamson (1975) have presented a review of these wastes. Some wastes are recovered in fish meal or fish rendering plants, and there are also numerous by-products that are not used for animal feeds. The remoteness and small size of some plants reduces the feasibility of processing the wastes into by-products. Quantity The by-products produced at rendering plants are fish meal, fish oil, and fish solubles. In 1977, 257 million kg fish meal were produced in the United States (Pennington and Husby, 19791. Fish meal is widely used for animal feeds, and its processing and nutritional value has been thor- oughly studied. Condensed fish solubles are concentrated from the water removed at fish meal and oil plants. In 1968, 65 million kg fish solubles were produced (Soderquist and Williamson, 19751. Fish solubles are used in feeds and fertilizers.

Industrial Food Processing Wastes 31 The technology for processing solid wastes is available but for various reasons is not used, and great quantities of wastes are disposed of on land or at sea. The demand for fish meals is greater than U.S. production can supply. Nutritional Value Patton and Chandler (1975) investigated chitinous products by in vivo fermentation. Chitin constitutes 12.3 percent of freshwater crayfish meal, 12.9 percent of crab meal, and 7.6 percent of shrimp meal. Samples of shrimp meal, crab meal, and purified chitin were placed in the rumen of fistulated steers. Average rumen solubilities were 17.4, 35.7, and 21.5 percent, respectively. Solubility of crab meal in the rumen was 15 percent over that in water alone. This research investigating their degradation in the rumen was the first step in determining the feasibility of feeding . . . . . ch~t~nous maters .s to ruminants. Processing Methods Wastes from processing fish and crabs have been dehydrated and ground to produce fish meal and crab meal. However, for small processors this method of utilization is not feasible. Fish silage has been produced with the addition of formic acid (Backhoff, 1976) or a combination of propionic and formic acids (Gildberg and Raa, 1977~. A stable silage was produced by ensiling a 2:1 mixture of fish waste and barley straw with formic and propionic acids (Gildberg and Raa, 19771. Satisfactory ensiling was re- ported for mixtures of fish waste and corn stover or peanut hulls with the addition of a small amount of molasses (Samuels et al., 19821. Ensiling of crab waste was not as satisfactory. Poultry Processing Wastes Poultry by-products are well utilized; therefore, poultry processing and by-product recovery will be briefly summarized. Egg processing, specif- ically egg breaking plants, have some wastes that are less well utilized. Eleven percent of the eggs in the United States were processed at egg breaking plants, producing 362.9 million kg liquid egg products in 1972. In 1973 the total egg production in the continental United States was 5,544 million dozen eggs. Egg production and egg breaking plants are wide- spread throughout the United States, with the highest concentrations oc- curring in the Southeast and California. Egg breaking plants use surplus or undergraded eggs; maximum production occurs in late spring and early summer.

32 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS Shell egg processing plants and egg breaking plants have similar wastes. Egg losses due to shell damage are 9.3 percent from shell egg processing plants and 6 to 10 percent from egg breaking plants. Some waste eggs are collected for use in pet food. Some of the egg content is lost to waste water. Shells and adhering egg content form a large portion of the waste from egg breaking plants. At an ear breaking olant. starting with the in- shell eggs (100 percent), the following products and wastes are derived: edible food product, 78.2 percent; inedible egg product, 3.4 percent; losses to sewer, 6.3 percent; shell, l l percent; and egg liquid adhering to shell, 1.75 percent. The quantity of inedible egg product available for sale as feed can be increased by in-plant conservation practices to 85.5 kg per 1000 kg eggs processed. The egg shells from most plants are disposed of in landfills. Egg shells can be dried, but processing them is not economically feasible except at larger plants (Jewel!, 19751. Some studies have been done on increasing recovery of by-products from egg breaking plants, especially egg shell wastes and egg contents (Jewel!, 19751. cam Red Meat Processing Wastes The red meat industry is well developed in terms of by-product utilization. By-products have been recovered for many years, and efficiency of by- product production is increasing. The term red meat industry will be used to differentiate beef, sheep, and hog processing from poultry processing. The following sections will describe the processes involved in feed by- product recovery and the composition and feed value of underutilized by- products. Subjects covered include: paunch content, wastewater treatment, tannery wastes, and hair. Paunch Content Paunch content is the material from the rumen of beef or sheep or the stomach of swine. Beef paunch content is the major waste in the United States and has been studied as a feed source. Physical Characteristics The paunch content contains undigested feeds, is yellowish brown in color, has an obnoxious odor, and has a water content of 85 percent (Witherow and Lammers, 19761. Because it is poorly utilized, it is a disposal problem to small plants. Quantity The quantity of paunch content in the United States was esti- mated as 0.771 billion kg from the 35 million beef cattle slaughtered annually, or by another estimate, 24.5 kg wet weight per animal or 3.8

industrial Food Processing Wastes 33 kg dry weight per animal (Witherow and Lammers, 19761. Some paunch content is dried for feed use, but landfill or field spreading are the most common disposal methods. Methods of disposal have been enumerated by Witherow (1974) and Witherow and Lammers (19761. Nutritional Value Paunch content varies in composition with feeding practices used prior to slaughter and with the type of processing. In one study, composition was determined on a dry-matter basis as 12.2 percent crude protein, 25 percent crude fiber, 5.2 percent ether extract, 7.9 percent ash, and 49.6 percent nitrogen-free extract (Ricci, 1977 3. Summerfelt and Yin (1974) claim that the variability of paunch content is less than that of many commonly used feeds and that paunch content from cattle finished on high protein formulated feeds would be even less variable. The range of values for dehydrated paunch content was 12 to 15 percent protein and 12 to 39 percent crude fiber. Messersmith (1973) studied the feeding value of paunch feed (dehy- drated paunch content) with cattle. In one study with high concentrate beef cattle finishing diets, paunch feed had a depressing effect on feed intake when fed as the only roughage. This effect was not seen when the roughage portion of the diet contained both paunch feed and brome hay. In these tests, paunch feed made up 5, 7.5, 10, and 15 percent of the diets. Inclusion of these levels had no significant effect on average daily gain, daily feed consumption, or efficiency of feed conversion. The report indicated that paunch feed had a value similar to that of poor-quality alfalfa hay when fed to ruminants, or 80 percent of the value of grass hay. In the same study, physical and/or chemical treatments were tested on paunch feed, but the response was not as good as the response of other poor-quality roughages. Summerfelt and Yin (1974) fed paunch feed to catfish at 10, 20, and 30 percent levels. At the 20 percent level or less there was no significant difference in the final weights of pond-reared fish between those fed diets containing paunch feed and those receiving commercial feed. However, fish fed 30 percent paunch feed were smaller. With 10 and 20 percent paunch feed levels, feed costs were similar to the costs for commercial feeds; feed costs per unit of gain were greater with 30 percent paunch feeds. Processing Methods Paunch content is sufficiently similar to cattle man- ure that systems developed for processing manure have potential appli- cability to paunch content. Witherow (1974) described several methods for processing and feeding paunch content, including using paunch content fresh, ensiled, and dried by gas-fired rotary driers, fluid bed driers, solar driers, or by pressing to reduce the water content. Odor during rotary

34 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS drying can be controlled by installation of a scrubber. One problem that has been encountered is the limited market for the dried feed. Some plants have installed rotary driers and pelletized the feed to improve marketa- bility; they have successfully sold the feed for use with both swine and cattle. Mixing blood with paunch content and drying the mixture yields a product containing 43 percent protein and handles two waste problems at once. Studies have been conducted on the other systems of processing paunch content mentioned above, but it is difficult to determine the extent of implementation of various systems. The two methods of feed use that appear most feasible are drying or ensiling. Ensiling has been tried with paunch content mixed with cornstalks, corn and beet pulp pellets, or combinations of these (Witherow, 1974; Witherow and Lammers, 19761. Some problems resulted because of the limited acceptability of the ensiled materials by cattle, apparently due to the feed's acidity; the addition of sodium bicarbonate seemed to solve the problem. Meat Processing Wastewater Treatment There are numerous methods for treating wastewater from meat packing operations; however, only those with feed by-product potential will be described. Grant (1976) described a process using ion exchange resins to recover proteins from wastewater. He reported that 2 to 5 percent of the total carcass protein is lost in effluents from meat packing plants and poultry plants. The dried protein product recovered contains 68 percent true protein and 3.7 percent ash, dry-matter basis. There are no significant deficiencies in the amino acid pattern. In processing plants where cooking or rendering takes place, fat may become complexed with the protein and will appear in the dried floe protein by-product. Feeding trials were conducted in New Zealand with dried effluent pro- tein from a meat plant (Grant, 19761. Chicks were fed diets in which 50 percent of the protein was supplied by dried effluent protein. The dried effluent protein was found to be equal in nutritive value to meat meals and casein, slightly inferior to fish meal, and superior to meat and bone meal and protein extracted from grass. Performance of pigs fed 5 percent dehydrated floe proteins was similar to that of control pigs. Hallmark et al. (1978) reported on the use of lignosulfonic acid (LSA) and dissolved air flotation for recovery of a usable feed protein product. Known as the Alwatech process, this method is being used in several plants and reduces BOD of wastewater by 60 to 90 percent. Solids content of the sludge is 6 to 12 percent. The sludge is neutralized with calcium hydroxide, and surplus blood may also be added. The sludge is dewatered

Industrial Food Processing Wastes 35 to 30 to 50 percent water and then dried separately or mixed with other meat by-products. Alwatech protein concentrate usually contains 10 percent LSA, dry- matter basis. Composition varies with the type of effluent treated; meat and fish effluents have been studied. Because the effluent is treated within 1 or 2 hours, the bacteriological condition is excellent. The product has been substituted for 50 percent fish meal or soybean meal in chick and swine diets with satisfactory results. LSA is a purified derivative from wood pulping liquors. Paulson and Lively (1979) reported on the use of activated sludge treatment and the use of activated sludge as animal feed. The initial solids content of the sludge was 0.3 to 0.6 percent. Pilot studies on thickening sludge by air flotation and centrifugation were conducted. Sludge samples were dried and analyzed. Nitrite and nitrate content were found to be less than 0.1 percent. The protein content, dry-matter basis, from several samples was 47 to 57 percent. Feeding trials were conducted with rats where sludge replaced up to 100 percent of the dietary protein. Good results were obtained when sludge provided 25 percent of the dietary protein. The rats gained weight when sludge was fed as the only protein source but at a slower rate than normal. This was probably due to the amino acid imbalance of the sludge. The sludge was analyzed for amino acids and found to have a high level of methionine compared to soybean meal. Tannery Wastes In recent years the United States has been producing about 40 million cattle hides per year, 50 percent of which are tanned by domestic tanners (M. Komanowsky and J. C. Craig, USDA Eastern Regional Research Center, Philadelphia, Pa., 1979, personal communication). When pro- cessed in the tannery only about 55 percent of the weight of the hide goes into making leather. Most of the remaining 45 percent is wasted, 5 to 10 percent as waste hair, 5 to 10 percent as dissolved proteins, 15 percent as flashings and trimmings, and 15 percent as splits. A fresh cattle hide contains 64 percent water, 33 percent protein, 2 percent fat, 0.5 percent mineral salts, and 0.5 percent other substances. The 33 percent protein is composed of 87.8 percent collagen, 6.1 percent keratin, 5.2 percent non- structural proteins (albumins, globulins, etc.), and 0.9 percent elastin. To make leather, the tanner removes most of the noncollagenous materials. Nutritional Value Wisman and Engel (1961) prepared two tannery by- product meals, referred to as partially hydrolyzed tannery by-product meal

36 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS and unhydrolyzed, acetone extracted, tannery by-product meal. Both were made from limed hide flashings. The hydrolyzed meal contained 93.5 percent dry matter, 68.3 percent crude protein, 1.52 percent ether extract, and 23.2 percent ash. The unhydrolyzed meal contained 79.2 percent dry matter, 67.8 percent crude protein, 7.7 percent ether extract, and 4.2 percent ash. The protein of the meals contained 8.1 percent lysine, 20.9 percent glycine, 0.13 percent tryptophan, and 2.4 percent methionine. The meals were tested in poultry diets as replacements for up to 75 percent of soybean oil meal protein. Diets had constant levels of protein, energy, calcium, and phosphorus. Maximum levels of meals in poultry diets appeared to be 12.5 to 25 percent of the soybean meal protein. Responses to both meals were similar. Adding tryptophan to the diet did not improve growth rates; thus tryptophan was not considered the first limiting amino acid (Wisman and Engel, 19611. Waldroup et al. (1970) found that chicks fed 2 to 3 percent hydrolyzed leather meal replacing soybean meal performed as well as the controls even without adding supplemental amino acids. Chicks could be fed up to 8 percent leather meal when supplemented with amino acids with no significant differences between treatments. Metabolizable energy content of the leather meal was 2,920 kcal/kg, dry matter basis. Compared to soybean meal, methionine, lysine, and tryptophan were found to be low in leather meal. Dilworth and Day ~ 1970) conducted a similar experiment and also found that chicks fed 1 to 3 percent leather meal had equal or greater growth than those fed the basal diet, with or without minimum amino acid levels being specified in the diets. A study was conducted by Knowlton et al. (1976) using hydrolyzed leather scrap to replace 0 to 75 percent of soybean meal crude protein on an isonitrogenous basis in sheep diets. On a dry-matter basis the leather meal contained 75.4 percent crude protein, 1.8 percent ether extract, 18.6 percent ash, and 3.0 percent chromium. Of the many parameters measured in the study, few were affected by the inclusion of hydrolyzed leather meal in the diet at the 50 and 75 percent substitution levels. There was a decrease in apparent digestibility of crude protein at these levels. Hydro- lyzed leather meal protein had digestibility values of 81.2, 71.6, and 71.2 percent in the 25, 50, and 75 percent substitution-level diets, respectively. Several reasons suggested for the lower digestibilities were incomplete hydrolysis during processing, heat damage incurred during flash drying, tanning, or a combination of these.

Industrial Food Processing Wastes 37 Hair Hair is available from two sources: from tannery operations that use the "save-hair system" and from slaughterhouse hog hair, which is removed by scalding and scraping. Nutritional Value Several studies have been conducted on the feeding of raw and processed hog and cattle hair. Moran et al. (1967a,b) found that raw hog hair fed to roosters was very poorly digested, with a meta- bolizable energy of only 0.58 Mcal/kg on a dry-matter basis. The energy value of the raw hog hair appeared to be derived mainly from the fat content. Hog hair contains 88.1 percent protein, 6.7 percent fat, and 2.2 percent ash, dry-matter basis. Processing under pressure (3.5 kg/cm2) at 148°C for 30 minutes greatly improved the digestibility. Metabolizable energy was 2.14 Mcal/kg on a dry-matter basis. They found that both hydrogen bonding and disulfide bonding involving cystine were respon- sible for the low digestibility of protein in raw hair. The cystine content of protein in raw hair was found to be lO to 15 percent. When the hog hair was processed, cystine was reduced to 3.5 percent of the protein and glycine was noticeably increased from 4.5 to 6.4 percent. Processing hair is very similar to processing feathers, with a slightly higher temperature being required, 148°C compared to 142°C. In chick- growing diets, up to 5 percent of the soybean protein in a 20 percent protein corn-soybean diet could be replaced by processed hog hair with little effect on growth or feed efficiency. If processed hog hair is substituted for soy protein completely, amino acid supplementation is necessary to prevent severe growth depression. Growth depression was completely overcome by supplementation with lysine, methionine, tryptophan, and glycine, which were the first through fourth limiting amino acids, re- spectively (Moran et al., 1967a,b). Moran and Summers (1968) also studied cattle hair in chick diets. The hair had been removed from the hide following calcium hydroxide and sodium sulfide treatment. The metabolizable energy of the raw cattle hair was 1.69 Mcal/kg and was increased to 2.25 Mcal/kg, dry-matter basis, when processed in the same manner as hog hair. The cystine level in raw cattle hair was 5.34 percent of protein and was reduced to 2.92 percent by processing; glycine was increased. Changes in the amino acid content, which had not been observed when processing hog hair or feathers, showed decreases in histidine, lysine, and tyrosine. The authors suggested that this may have been caused by nonenzymatic browning. Substituting the processed cattle hair for soy protein in a 20 percent protein diet did not affect chick performance. However, if processed cattle hair was substituted

38 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS for all of the soy protein, supplementation with methionine, lysine, tryp- tophan, histidine, and glycine was required to prevent growth depression. Kornegay and Thomas (1973) found that diets containing 2 to 3 percent processed hog hair meal could be substituted for soybean meal on a digestible protein basis without depressing growth or feed efficiency. At levels above 6 percent processed hog hair, however, feed intake was depressed. Amino acid deficiencies, imbalance, or poor availability were suggested as causes. ALTERNATIVE USES FOR FOOD PROCESSING WASTES Other than for animal feed, uses for food processing wastes have developed because of the increased cost of fossil fuels for energy. Some food pro- cessing wastes are now being used for fuel or for alcohol production. Depending upon cost of fuel, some materials (e.g.' almond hulls, fruit pits, and nut shells) may be burned as fuel rather than utilized as animal feed. Also, food processing wastes high in fermentable carbohydrates and sugars may be utilized in alcohol production. ANIMAL AND HUMAN HEALTH PROBLEMS AND REGULATORY ASPECTS Pesticide Residues The possibility of harmful pesticide residues must be considered when using crop material wastes. Pesticide use and consequent residues on crops for human consumption are regulated and monitored by federal and state agencies. Food processors have information on the pesticides used on the crops that they process. It is important that this information be obtained from the processor by those intending to use food processing waste materials for animal feed. In turn, it may often be necessary to analyze the waste material for pesticides to determine that tolerances are not exceeded. Pesticides may be present in higher concentrations on the waste material than on the total raw product received by the food processor. The reason for this is that the pesticide residues are usually on the surface of the commodity and are removed by washing, peeling, and trimming; thereby, they are concentrated in the waste material. Variable levels of pesticides were reported in apple pomace by Rumsey et al. (19771. Feeding the pomace caused significant accumulation of pesticide in depot fat of pregnant beef cows. Pesticide residues in potatoes

Industrial Food Processing Wastes 39 were reviewed by McCoy et al. (19751. However, by-product feeds were not covered, only the whole potato and the potato parts. As with most fruits and vegetables, the residue content is higher in the peels. McCoy et al. (1975) state that potatoes, as a root crop, are less likely to carry toxic residues than above-ground crops because the pesticides must either be in the soil or translocated from the aerial part of the plant to reach the tuber. Feed use of tomato waste may be limited because insecticide levels are often higher than residue standards set for feeds (Schultz et al., 1976, 19771. Toxaphene is one pesticide involved, and the residue can be present in the waxy layer of the skin. Removal of the tomato skin would increase the value of the pomace. The adhering pulp could then be recovered for food use, as has been similarly studied with peel residue from canning. Heavy Metals Heavy metals were not found in significant quantities in the biological solids from fruit-cannery activated sludge or the carcasses of animals fed the sludges (Esvett, 19761. Chromium accumulated in tissues, particularly kidney tissue, and some in fat in chicks fed hydrolized leather meal (Dilworth and Day, 19701. Waldroup et al. (1970) reported that chromium tended to accumulate in kidneys of chicks fed 8 percent leather meal, but not in all tissues ex- amined. Chromium levels are restricted to 2.75 percent in tannery by- products fed to swine (Knowlton et al., 19761. Animal Health Apple Pomace and Nonprotein Nitrogen A serious reproductive problem was encountered when apple pomace was fed with nonprotein nitrogen (NPN) (Fontenot et al., 1977~. Apple pomace fed with urea or biuret, or a combination of these, lowered feed con- sumption and increased body weight losses when compared to corn silage and NPN, and also when compared to apple pomace and protein supple- ment. Feeding apple pomace and NPN had several serious detrimental effects including high incidence of dead, weak, or deformed calves (Bov- ard et al., 19771. No explanation has been given for the effects of feeding apple pomace and NPN. Reproductive problems were not encountered when feeding apple pomace with protein supplements. Feeding apple pomace with NPN in any form should be avoided.

40 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS Cacao Processing Wastes The factors that limit cacao processing wastes as feed are theobromine (0.75 to 1.3 percent in shell) and caffeine (Chats, 19531. Plain chocolate contains 3 percent theobromine and 0.1 percent caffeine (Curtis and Grif- fiths, 19721. Chatt (1953) reported adverse effects if theobromine intake exceeds 0.025 g/kg body weight. In horses, theobromine intake at the level of 0.027 g/kg body weight has caused death. No waste containing theobromine should be fed to racehorses because it may cause reactions similar to doping. Cacao shells should only be fed to mature cattle at a maximum level of 2.5 percent of the diet or a maximum of 0.907 kg/day. It should not be fed to pigs, poultry, or calves because the cumulative effect is detri- mental (Chats, 19531. Calves fed 5 to 10 percent chocolate waste exhibited hyperexcitability, exaggerated gaits, and excessively alert appearance; one calf died (Curtis and Griffiths, 19721. SUMMARY Many researchers emphasize the variability of food processing wastes. The reasons for this variability are the variability of the raw food material being processed, the differences in production processes employed by different plants, and the different food products produced from the same raw material. Most of the food processing wastes have substantial nutritional value. A characteristic of most of these wastes is the high moisture, which results in high transportation and dehydration costs per unit of nutrient. Fruit and vegetable processing wastes are generally low in protein and may be limited in energy value; they are probably best suited for feeding to rum- inants. Animal processing wastes are generally high in protein, and the protein is usually high in quality. Feeding of wastes usually does not adversely affect animal performance if appropriate levels are included in the diet. Wastes can be processed by dehydration, but frequently this is not economically feasible. For many wastes, ensiling appears to be feasible. High-moisture materials should be combined with drier materials for good ensiling; the dry materials could consist of poor quality hay or crop res- idues. Although caution should be exercised, the feeding of food processing wastes does not appear to pose a serious threat to animal and human health. Pesticide levels need to be monitored. Caution should be used not to feed apple pomace in combination with nonprotein nitrogen to avoid

Industrial Food Processing Wastes 41 reproductive problems. Wastes such as cacao processing waste, which may contain harmful levels of certain drugs, should be limited to safe levels. LITERATURE CITED Amerine, M. A., H. W. Berg, and W. V. Cruess. 1972. Technology of Wine Making. Westport, Conn.: AVI. Ammerman, C. B., L. R. Arrington, P. E. Loggins, J. T. McCall, andG. K. Davis. 1963. Nutritive value of dried tomato pulp for ruminants. J. Agric. Food Chem. 11:347. Ammerman, C. B., R. H. Harms, R. A. Dennison, L. R. Arrington, and P. E. Loggins. 1965. Dried Tomato Pulp, Its Preparation and Nutritive Value for Livestock and Poultry. Fla. Agric. Exp. Stn. Bull. No. 691. Backhoff, H. P. 1976. Some chemical changes in fish silage. J. Food Technol. 11:353. Ben-Gera, I., and A. Kramer. 1969. The utilization of food industries wastes. Adv. Food Res. 17:77-152. Bond, J., and P. A. Putnam. 1967. Nutritive value of dehydrated sweet potato trimmings fed to beef steers. J. Agric. 15:726. Bough, W. A., and D. R. Landes. 1976. Recovery and nutritional evaluation of protein- aceous solids separated from whey by coagulation with chitosan. J. Dairy Sci. 59(11): 1874. Bovard, K. P., T. S. Rumsey, R. R. Oltjen, J. P. Fontenot, and B. M. Priode. 1977. Supplementation of apple pomace with nonprotein nitrogen for gestating beef cows. II. Skeletal abnormalities of calves. J. Anim. Sci. 46:523-531. Brandt, J. A., B. C. French, and E. V. Jesse. 1978. Economic Performance of the Pro- cessing Tomato Industry. Giannini Foundation Information Series No. 78- 1. Univ. Calif. Div. Agric. Sci. Bull. 1888. Brown, A. H., W. D. Ramage, and H. S. Owens. 1950. Progress in processing pear canning waste. Food Packer 31(7):30 and 31(8):50. Brugman, H. H., and H. C. Dickey. 1955. Potato Pulp as Feed for Livestock. Maine Agric. Exp. Stn. Bull. No. 539. Brugman, H. H., and H. C. Dickey. 1961. Potato Pulp as a Feed for Livestock. Maine Agric. Exp. Stn. Bull. No. 599. Burris, M. J., and B. M. Priode. 1957. The Value of Apple Pomace as a Roughage for Wintering Beef Cattle. Va. Agric. Exp. Stn. Res. Rep. 12. Chatt, E. M. 1953. Cocoa: Cultivation, Processing, Analysis. New York: Interscience. Clark, W. S. 1979. Our industry today: Whey processing and utilization, major whey product markets 1976. J. Dairy Sci. 62:96-98. Colston, N. V., and C. Smallwood, Jr. 1972. Waste control in the processing of sweet potatoes. Pp. 85-98 in Proc. Third Natl. Symp. Food Process. Wastes. EPA-R2-72- 018. Cruess, W. V. 1958. Commercial Fruit and Vegetable Products, 4th ed. New York: McGraw- Hill. Curtis, P. E., and J. E. Griffiths. 1972. Suspected chocolate poisoning of calves. Vet. Rec. 90:313. Damron, B. L., A. R. Eldred, S. A. Angalet, J. L. Fry, and R. H. Harms. 1974. Eval- uation of activated citrus sludge as a poultry feed ingredient. Pp. 142-154 in Proc. Fifth.  Natl. Symp. Food Process. Wastes. EPA 660/2-74-0s8. Della Monica, E. S., C. H. Huhtanen, and E. O. Strolle. 1975. Stability of protein water concentrates from potato starch factory waste effluents. J. Sci. Food Agric. 26:617-623.

42 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS Dickey, H. C., H. A. Leonard, S. D. Musgrave, and P. S. Young. 1971. Nutritive char- acteristics of dried potato by-product meal for ruminants. J. Dairy Sci. 54:876-879. Dilworth, B. C., and E. J. Day. 1970. Hydrolyzed leather-meal in chick diets. Poult. Sci. 49(4): 1090. D'Mello, J. P. F., and C. T. Whittemore. 1975. Nutritive value of cooked potato flakes for the young chick. J. Sci. Food Agric. 26:261. Edmond, J. B., and G. R. Ammerman. 1971. Sweet Potatoes: Production, Processing, Marketing. Westport, Conn.: AVI. Esselen, W. B., Jr., and C. R. Fellers. 1939. The nutritive value of dried tomato pomace. Poult. Sci. 18(1):45. Esvett, L. A. 1976. Fruit Cannery Waste Activated Sludge as a Cattle Feed Ingredient. Environ. Prot. Technol. Ser. EPA/2-76-253. Folger, A. H. 1940. The Digestibility of Ground Prunes, Winery Pomace, Avocado Meal, Asparagus Butts, and Fenugreek Meal. Univ. Calif. Exp. Stn. Bull. 635. Fontenot, J. P., K. P. Bovard, R. R. Oltjen, T. S. Rumsey, and B. M. Priode. 1977. Supplementation of apple pomace with nonprotein nitrogen for gestating beef cows. I. Feed intake and performance. J. Anim. Sci. 45:513-522. Gildberg, A., and J. Raa. 1977. Properties of a propionic acid/formic acid preserved silage of cod viscera. J. Food Technol. 28:647. Gillies, M. T. 1978. Animal Feeds from Waste Materials. Food Technology Review No. 46. Park Ridge, N.J.: Noyes Data Corporation. Graham, R. P., A. D. Shepherd, A. H. Brown, and W. D. Ramage. 1952. Advanced fruit- waste recovery. Food Eng. 24:82. Grames, L. M., and R. W. Kueneman. 1969. Primary treatment of potato processing wastes with by-product feed recovery. Water Pollut. Control Fed. 41(7):1358-1367. Grant, R. A. 1976. Protein recovery from meat, poultry and fish processing plants. In Food from Waste, G. G. Birch, K. J. Parker, and J. T. Worgan, eds. London: Applied Sci- ence. Gray, L. R., and M. R. Hart. 1972. Caustic Dry Peeling of Cling Peaches to Reduce Water Pollution: Its Economic Feasibility. USDA ERS Agric. Econ. Rep. 234. Guilbert, H. R., and W. C. Weir. 1951. Pear pulp and pear molasses: Nutritional value for cattle and palatability to sheep tested in feeding trials with commercial products. Calif. Agric. 5:6. Guttormsen, K., and D. A. Carlson. 1970. Status and research needs of potato processing wastes. Pp. 27-38 in Proc. First Natl. Symp. Food Process. Wastes. EPA-12060-04/ 70. Hadjipanayiotou, M., and A. Louca. 1979. A note on the value of dried citrus pulp and grape mare as barley replacements in calf fattening diets. Anim. Prod. 23:129-132. Hallmark, D. E., J. C. Ward, H. C. Isaksen, and W. Adams. 1978. Protein recovery from meat packing effluent. Pp. 288-305 in Proc. Ninth Natl. Symp. Food Process. Wastes. EPA-600/2-78- 188. Hendrickson, R., and W. Kesterson. 1965. By-Products of Florida Citrus, Composition, Technology, and Utilization. Univ. of Florida, Gainesville, Agric. Exp. Stn. Bull. 698. Hillyer, C. M., and C. T. Whittemore. 1975. Intake by piglets of diets containing cooked  potato flake. J. Sci. Food Agric. 26:1215. Hinks, C. E., D. G. Peers, and I. W. Moffat. 1975. The nutritive value of cooked potato in milk replacers for young calves. J. Sci. Food Agric. 26:1219-1224. Hinman, N. H., W. N. Garrett, J. R. Dunbar, A. K. Swenerton, and N. E. East. 1978. Tomato pomace scores well as sheep feed. Calif. Agric. 32(8):12.

Industrial Food Processing Wastes 43 Hogan, J. M., and M. E. Highlands. 1975. Potato and potato products for livestock feed. Pp. 639-645 in Potato Processing. Westport, Conn.: AVI. Jewell, W. J. 1975. Egg Breaking and Processing Waste Control and Treatment. Environ. Prot. Technol. Ser. EPA-660/2-75-019. Jones, H. R. 1973. Waste Disposal Control in the Fruit and Vegetable Industry. Pollution- Technology Review No. 1. Park Ridge, N.J.: Noyes Data Corp. Jones, H. R. 1974. Pollution Control in the Dairy Industry. Pollution Technology Review No. 7. Park Ridge, N.J.: Noyes Data Corp. Jones, R. H., J. T. White, and B. L. Damron. 1975. Waste Citrus Activated Sludge as a Poultry Feed Ingredient. EPA-660/2-75-001. Juengst, F. W., Jr. 1979. Use of total whey constituents Animal feed. J. Dairy Sci. 62(1): 106- 111. Katsuyama, A. M., N. A. Olson, R. L. Quirk, and W. A. Mercer. 1973. Solid Waste Management in the Food Processing Industry. National Canners Association. EPA PB219- 019. Available from NTIS. Knowlton, P. H., W. H. Hoover, C. J. Sniffen, C. S. Thompson, and P. C. Belyea. 1976. Hydrolyzed leather scrap as a protein source for ruminants. J. Anim. Sci. 43(5):1095. Kornegay, E. T., and H. R. Thomas. 1973. Evaluation of hydrolyzed hog hair meal as a protein source for swine. J. Anim. Sci. 36(2):279. Livingston, A. L., R. E. Knowles, J. Page, D. D. Kuzmicky, and G. O. Kohler. 1972. Processing of cauliflower leaf waste for poultry and animal feed. J. Agric. Food Chem. 20(1):277-281. Livingston, A. L., R. E. Knowles, R. H. Edwards, and G. O. Kohler. 1974. Processing of pimento waste to provide a pigment source for poultry feed. J. Sci. Food Agric. 25:483-490. Livingston, A. L., R. E. Knowles, R. H. Edwards, and G. O. Kohler. 1976. Processing of fresh artichoke trimmings for use in animal feeds. J. Agric. Food Chem. 24(6): 1158- 1161. Lofgreen, G. P., and M. Prokop. 1979. Citrus peel liquor as an energy source in cattle growing rations. In Calif. Feeders Day Rep. Univ. Calif. Dept. Anim. Sci., Coop. Ext. Imperial Valley Field Stn. Mantell, C. L. 1975. Nut Shells and Fruit Pits in Solid Wastes: Origin, Collection, Pro- cessing and Disposal. New York: Wiley-Interscience. McCoy, C. M., and S. E. Smith. 1940. Tomato pomace in the diet. Science 91:388. McCoy, C. M., J. B. McCoy, and O. Smith. 1975. The nutritive value of potatoes. Pp. 235-273 in Potato Processing. Westport, Conn.: AVI. Messersmith, T. L. 1973. Evaluation of dried paunch feed as a roughage source in ruminant finishing rations. M.A. thesis. Department of Animal Science, University of Nebraska. Moran, E. T., Jr., and J. D. Summers. 1968. Keratins as sources of protein for the growing chick. 4. Processing of tannery by-product cattle hair. Poult. Sci. 47(2):570. Moran, E. T.' Jr., J. D. Summers, and S. J. Slinger. 1967a. Keratins as sources of protein for the growing chick. 2. Hog hair, a valuable source of protein with appropriate pro- cessing and amino acid balance. Poult. Sci. 46(2):456-465. Moran, E. T., Jr., H. S. Bayley, and J. D. Summers. 1967b. Keratins as sources of protein  for the growing chick. 3. The metabolizable energy and amino acid composition of raw and processed hog hair meal. Poult. Sci. 46(3):548. Morrison, F. B. 1956. Feeds and Feeding, 22nd ed. New York: Morrison Publishing. National Research Council. 1976. Nutrient Requirements of Beef Cattle. Washington, D.C.: National Academy of Sciences.

44 UNDERUTILIZED RESOURCES AS ANIMAL FEEDSTUFFS Osten, B. J. 1976. Protein from potato starch mill effluent. P. 196 in Food From Waste, G. G. Birch, K. J. Parker, and J. T. Worgan, eds. London: Applied Science. Pailthorp, R. E., J. W. Filbert, and G. A. Richter. 1975. Waste disposal. P. 646 in Food Processing. Westport, Conn.: AVI. Pattee, E. C. 1947. Winery Waste Recovery. Cincinnati: Research Division, National Distillers Products Corp. Patton, R. S., and P. T. Chandler. 1975. In vivo digestibility evaluation of chitinous materials. J. Dairy Sci. 58:397-403. Paulson, W. L., and L. D. Lively. 1979. Oxidation Ditch Treatment of Meatpacking Wastes. EPA-600/2-79-030. Pennington, H., and F. Husby. 1979. University of Alaska fishmeal research. Univ. of Alaska, Alaska Seas and Coasts 7(1):6-8. Prokop, M. 1979. Dried winery pomace as an energy source in cattle finishing rations. In Calif. Feeders Day Rep. Univ. Calif. Dept. Anim. Sci., Coop. Ext. Imperial Valley Field Stn. Ricci, R. 1977. A Method of Manure Disposal for a Beef Packing Operation. First Interim Tech. Rep. EPA-600/2-77-103. Richter, G. A., K. L. Sirrine, C. I. Tollefson. 1973. Conditioning and disposal of solids from potato waste water treatment. J. Food Sci. 38:218-224. Rogers, C. J., and W. E. Coleman. 1978. Protein from Aspergillus niger grown on starchy waste substrate in animal feeds. Waste Mater. Food Technol. Rev. 46:139. Rosenau, J. R., L. F. Whitney, and R. A. Elizondo. 1976. Low waste water potato starch/ protein production process Concept, status, and outlook. Pp. 118- 128 in Proc. Seventh Natl. Symp. Food Process. Wastes. EPA-600/2-76-304. Rosenau, J. R., L. F. Whitney, and J. R. Haight. 1977. Potato juice processing. Pp. 284- 291 in Proc. Eighth Natl. Symp. Food Process. Wastes. EPA-600/2-77-184. Rosenau, J. R., L. F. Whitney, and J. R. Haight. 1978. Economics of starch and animal feed production from cull potatoes. Pp. 89-99 in Proc. Ninth Natl. Symp. Food Process. Wastes. EPA-600/2-78- 188. Samuels, W. A., J. P. Fontenot, K. E. Webb, Jr., and V. G. Allen. 1982. Ensiling of seafood waste and low quality roughages. VPI and State Univ. Anim. Sci. Res. Rep. 2: 175. Schingoethe, D. J. 1976. Whey utilization in animal feeding: A summary and evaluation. J. Dairy Sci. 59(3):556. Schultz, W. G., R. P. Graham, and M. R. Hart. 1976. Pulp recovery from tomato peel residues. Pp. 105-117 in Proc. Sixth Natl. Symp. Food Process. Wastes. EPA-600/2- 76-224. Schultz, W. G.' H. J. Neumann, J. E. Schade, J. P. Morgan, P. F. Hanni, A. M. Kat- suyama, and H. J. Maagdenberg. 1977. Commercial feasibility of recovering tomato peeling residuals. Pp. 119- 136 in Proc. Eighth Natl. Symp. Food Process. Wastes. EPA- 600/2-77- 184. Skogman, H. 1976. Production of symba-yeast from potato wastes. In Food From Waste, G. G. Birch, K. J. Parker, and J. T. Worgan, eds. London: Applied Science. Smallwood, C., R. S. Whitaker, and N. V. Colston. 1974. Waste Control and Abatement in the Processing of Sweet Potatoes. Environ. Prot. Technol. Ser. EPA-660/2-73-021.  Smith, T. J. 1976. Dry peeling tomatoes and peaches. Pp. 194-203 in Proc. Sixth Natl. Symp. Food Process. Wastes. EPA-600/2-76-224. Smock, R. M., and A. M. Neubert. 1950. Apples and Apple Products. New York: Inter- science. Soderquist, M. R., and K. J. Williamson. 1975. Fish and shellfish wastes. In Solid Wastes: Origin, Collection, Processing and Disposal. New York: Wiley-Interscience.

Industrial Food Processing Wastes 45 Stabile, R. L., V. A. Turkot, and N. C. Aceto. 1971. Economic analysis of alternative methods for processing potato starch plant effluents. Pp. 185-202 in Proc. Second Natl. Symp. Food Process. Wastes. EPA-12060-03/71. Stokes, R. D. 1967. An evaluation of current practices in the treatment of winery wastes. M.A. thesis. University of New South Wales, Australia. Summerfelt, R. C., and S. C. Yin. 1974. Paunch manure as a feed supplement in channel catfish farming. Pp. 246-257 in Proc. Fifth Natl. Symp. Food Process. Wastes. EPA- 660/2-74-046. Treadway, R. H. 1975. Potato starch. P. 546 in Potato Processing. Westport, Conn.: AVI. Tsatsaronis, G. C., and D. G. Boskou. 1975. Amino acid and mineral salt content of tomato seed and skin waste. J. Sci. Food Agric. 26:421-423. U.S. Department of Agriculture. 1980. Agricultural Statistics. Washington, D.C.: U.S. Department of Agriculture. Waldroup, P. W., C. M. Hilliard, W. W. Abbott, and L. W. Luther. 1970. Hydrolyzed leather meal in broiler diets. Poult. Sci. 49(5):1259. Walter, R. H., J. B. Bourke, R. M. Sherman, R. G. Clark, E. George, Jr., A. B. Karasz, R. Pollman, S. E. Smith, and G. Lake. 1975. Apple pomace in the dairy regimen. Cornell Univ., Geneva, New York. Food Life Sci. 8:12-13. Whittemore, C. T., A. G. Taylor, I. W. Moffat, and A. Scott. 1975. Nutritive value of raw potato for pigs. J. Sci. Food Agric. 26:255. Wilson, L. L. 1971. Adipose tissue concentrations of certain pesticides in steers fed apple waste during different parts of the finishing period. J. Anim. Sci. 33:1356-1360. Wisman, E. L., and R. W. Engel. 1961. Tannery by-product meal as a source of protein for chicks. Poult. Sci. 40(6): 1761. Witherow, J. L. 1974. Paunch handling and processing techniques. Nat. Provis. 10:14. Witherow, J. L., and S. Lammers. 1976. Paunch and viscera handling. Pp. 37-66 in  Workshop (1973) on In-plant Waste Reduction in the Meat Industry, compiled by J. L. Witherow and J. F. Scaief. Environ. Prot. Technol. Ser. EPA-600/2-76-214.