The image of the camel, symbol of human survival in the desert, is tied to the history of the major nomadic civilizations of the hot dry areas of the northern hemisphere. The camel embodies one of the essential elements of the culture and agriculture of these regions.
Geographically, the camel is distributed throughout the tropical and subtropical dry zones of North Africa, western Asia and northwest India (Figure 1). The limits of its natural distribution are determined by wet climates and the presence of the tsetse fly. Camels were introduced into Australia on a large scale during the last century. Isolated introductions took place in parts of the United States of America, Central America, the Caribbean, southern Africa and Europe (Wilson, 1984; Wilson, Araya and Melaku, 1990).
The camel is the ideal domestic animal in deserts with long, dry, hot periods of eight months or more and scarce, erratic annual rainfalls between 50 and 550 mm.
The camel is used for several purposes for which its role is essential. It is used as a beast of burden for transporting goods and people as well as for providing milk. Milk is often the only regular food source for its owners. The camel's meat, wool and leather are also widely utilized. In some parts of East Africa, the animal is bled regularly and its blood consumed fresh or mixed with milk. The camel is universally highly valued and provides social standing for its owner.
The chief role of the camel relates directly to its remarkable adaptation to extremely harsh conditions. It can flourish where no other domestic animal can survive. This exceptional ability is the result of several anatomical and physiological characteristics. Where green forage is available in mild climates, the camel may go several months without drinking. Under very hot conditions, it may drink only every eight to ten days and lose up to 30 percent of its body weight through dehydration (Yagil and Etzion, 1980; Yagil, 1982; Wilson, 1984; Yagil, 1985; Ramet, 1987).
This remarkable attribute results from a very low basal metabolism and several water-conserving adaptations. Water losses by respiration and perspiration are low because of the camel's ability to withstand, without apparent difficulty, large variations in body temperature of up to 6°C. Excess heat accumulated during the heat of the day or after hard work is later lost by conduction, convection and radiation when the animal is at rest or when the atmosphere cools down at night. Moreover, the water lost by respiration and perspiration is low compared with the body weight of the animal. Water loss in the faeces and urine is also very limited (Wilson, 1984; Yagil, 1986). The morphology of the animal, characterized by the length of the neck and limbs and by the conical shape of the abdomen, creates a large surface area that improves heat transfer. General thermal conductivity also appears to be enhanced by the location of fatty tissues in the hump (Wilson, 1984; Yagil, 1986).
Another limitation imposed by arid conditions is the sparse and poor quality of pastures. Compared with other ruminants, the camel is distinguished by the high diversity of its diet. It can feed on herbaceous plants, shrubs, shoots, cacti and date stones. During the dry season, it often has to survive on thorny, withered plants low in protein but rich in fibre and cellulose (Peyre de Fabregues, 1989). According to the little research on the subject, it appears that the camel assimilates nitrogen and cellulose better than any other mammal (CIHEAM, 1988; Kamoun, Girard and Bergaoui, 1989; Gérard and Richard, 1989).
This high efficiency in assimilating cellulose appears to be related to a specific mastication process that induces improved impregnation of saliva into the bolus. The location of aquiferous cells on the stomach wall permits better wetting of feed during rumination and improves intake of some soluble elements. In addition, it seems that the stomach can retain coarse particles and allow only the smallest elements to pass through the intestine wall, which increases digestive recovery (Yagil, 1985; Yagil, 1986).
Another distinguishing feature is the camel's highly efficient system of recycling urea to meet its nitrogen requirements and to balance the low content of this element in desert plants. Unlike other mammals, camels have a distinctive kidney structure that considerably reduces the removal of urea in urine. The removal of blood urea is effected by selective permeability of the stomach and intestinal walls. Later, this urea is assimilated by the stomach microflora in the cavities to ensure protein synthesis (Wilson, 1984; Yagil, Saran and Etzion, 1984; Yagil, 1985).
In 1985, the world camel population was estimated at about 16.5 million, with more than 80 percent of the world herd in Africa. Somalia and the Sudan have the largest populations, with some 70 percent of the African herd. In Asia, about 70 percent are spread over the Indian subcontinent (Wilson, Araya and Melaku, 1990).
Historical trends in worldwide numbers are difficult to track because of lack of reliable data. It appears, however, that a decrease in numbers was observed from 1950 to 1980. Several causes were responsible, including mechanization of transport, sedentarization of nomads and exceptional droughts. Over the last decade, with the exception of a few isolated cases, a new phase in the development of the camel can be noted (Table 1). This is the result of several factors, mainly an increasing demand for milk and meat as a consequence of large human population increases in the areas involved. Other factors are associated with the extension of the desert in the Sahel region and increased utilization of camels as pack and work animals in countries where the cost of fuel is high. A further reason for this resurgence is the effect of recent technical and scientific research (Wilson, Araya and Melaku, 1990; Farah, 1993). This work has shown that the camel is the most efficient domestic animal for converting vegetative matter into work, milk and meat in hot arid areas. Recent advances in understanding camel pathology and physiology in relation to its products have led to better understanding of breeding and processing methods (Hoste, Peyre de Fabregues and Richard, 1985; Higgins, 1986; Marie, 1987; OIE, 1987; IEMVT, 1989; CIHEAM, 1989; Wilson, Araya and Melaku, 1990; Farah, 1993).
Existing data on the milk yield of camels are numerous but highly variable. According to results from several authors, lactation periods vary from 9 to 18 months, with annual milk yields of between 800 and 3 600 litres. Mean daily milk production is reported to range from 2 to 6 litres under desert conditions and up to 12 to 20 litres under more intensive breeding systems. These large differences can be explained by the fact that measurements have often been made under local conditions without taking into account local factors that might influence milk production. Furthermore, camel breeds or individual animals probably exist with significantly different milk-producing potential that has not been fully exploited because the selective pressure of humans on the camel has been minimal compared with other domestic animals (Richard and Gérard, 1989).
Nutritional factors also influence milk production. Diets enriched with green forages such as alfalfa, bersim or cabbage greatly increase milk yield (Knoess, 1977; Knoess et al., 1986; Richard and Gérard, 1989). The amount of milk is only marginally decreased when drinking-water is restricted, while total solids are significantly lowered (Yagil and Etzion, 1980; Yagil, Saran and Etzion, 1984; Ramet, 1987; Farah, 1993). This milk dilution is a physiological response to heat and could be a natural adaptation to provide much-needed water to the dehydrated calf (Yagil, Saran and Etzion, 1984; Farah, 1993).
Studies concerning the development of milk quantity as a function of stage of lactation indicate little correlation. Lactation curves in fact indicate large differences compared with other lactating mammals. Some curves indicate low yields during the first half of the lactation period and an increase in the second. Other results report higher yields at the beginning, followed by falls towards the end. Occasionally, one or two distinct peaks can be observed or, conversely, steady production throughout the lactation (Field, 1979; Bachmann and Schulthess, 1987; Ellouze and Kamoun, 1989; Richard and Gérard, 1989; Martinez, 1989). The high disparity between these various sets of data can probably be explained by differences in genetic potential, climate, feeding conditions and sampling techniques.
Milking practice also affects the amount of milk. Generally, the calf is allowed to suckle for a few minutes before hand milking. The actual volume of milk secreted is therefore difficult to measure. If milking is performed without any previous mechanical stimulation of the mammary gland, lower yields are observed. Milking must be done by a person who is well known to the camel. When the usual milker is changed, significant milk retention is often observed. It also appears that milking frequency influences daily milk yield. Generally, animals are milked two to four times a day (Hartley, 1980; Ramet, 1987; Martinez, 1989; Abeiderrahmane, 1994) but sometimes as many as six or seven times (Knoess, 1977). Changing the milking frequency from two to four operations increased milk production from one to 1.5 litres a day (Evans and Powys, 1980).
Publications dealing with the composition of camel milk are relatively scarce and much of the information is approximate and fragmentary. Table 2 therefore indicates only the more important data published in review articles by several authors (Yagil, 1982; Wilson, 1984; Wilson, Araya and Melaku, 1990; Farah, 1993). More recent results have been added.
Table 2 shows a fairly wide range of values for the main constituents of camel milk. As mentioned above for milk yields, this diversity could be mainly related to the different genetic potential of breeds, to varying physiological and feeding conditions or to stage of lactation. The computed mean value indicates that the total solids content of camel milk is slightly lower than cow's milk.
The most important factor affecting the overall composition of camel milk is water content. It has been clearly demonstrated that experiments which restricted drinking-water caused an increase in water content and a subsequent decrease in total solids (Yagil and Etzion, 1980; Yagil, 1986; Yagil, Amir, Abu-Ribaya and Etzion, 1986). Seasonal climatic variations and water and feed availability had a similar effect (Knoess et al., 1986; Ramet, 1987; Ramet, 1994a).
Although the overall composition of camel milk is similar to cow's milk, some differences exist in the molecular composition of proteins and lipids and in the mineral balance.
Protein. The average mean composition of the protein and nitrogen fractions of camel milk are generally similar to those of cow's milk. The average values for the casein and whey protein contents vary from 1.9 to 2.3 percent and 0.7 to 1.0 percent respectively. The nitrogen content of casein is a little lower than cow's milk, reaching 71 to 79 percent of total protein nitrogen compared with 77 to 82 percent (Jenness and Sloan, 1969; Mehaia, 1987; Farah, 1993).
Casein fractions have been isolated in camel milk and found to be homologous with bovine casein. The balance between the different casein fractions is very different, however, and chiefly identified by a low amount of kappa casein of only about 5 percent of the total casein, compared with about 13.6 percent in bovine casein (Table 3; Jardali, 1988; Jardali and Ramet, 1991; Farah, 1993). The molecular weights and amino acid composition of the casein fractions differ from those of cow's milk (Table 4; Sawaya et al., 1984; Larsson-Raznikiewicz and Mohamed, 1986; Farah and Ruegg, 1989; Mohamed, 1990; Farah, 1993).
The state of the casein micelle structure has seldom been investigated. Most results, however, conclude that the size distribution of casein particles in camel milk is significantly broader than in cow's milk, exhibiting a greater number of large particles. The average micelle diameter of camel milk was found to be about double that of cow's milk at 320 nm and 160 nm respectively (Table 4; Sawaya et al., 1984; Larsson-Raznikiewicz and Mohamed, 1986; Farah and Ruegg, 1989; Jardali and Ramet, 1991; Jardali, 1994).
The quantity of whey proteins is higher in camel milk than cow's milk, at 0.9 to 1.0 percent and 0.7 to 0.8 percent respectively. Individual fractions have been identified according to chromatographic and electrophoretic mobility and to the primary sequence of their amino acid chains. Two types of alpha-lactalbumin similar to bovine milk have been isolated. Beta-lactoglobulin has not been clearly identified (Conti et al., 1985; Beg et al., 1987; Farah, 1986). Two novel camel whey proteins, unlike any known bovine milk whey proteins, have been separated and characterized (Beg et al., 1987). The heat stability of camel milk whey proteins was found to be considerably higher than in cow's milk (Farah,1986; Farah and Atkins, 1992).
Lactose. Table 2 indicates that the average lactose content of camel milk is slightly lower (4.62 percent) than cow's milk (4.80 percent). It seems, however, that the variability is higher, with extreme values between 2.90 to 5.80 percent in camel milk compared with 4.40 to 5.80 percent in cow's milk (Webb, Johnson and Alford, 1974).
Lipids. A bibliographic review indicates that the fat content of camel milk varies greatly from 1.10 to 5.50 percent depending on the breed and feeding conditions, the average being the same as for cow's milk (Table 2). Studies on the structure and composition of fat globules revealed two main characteristics:
Minerals. Table 7 indicates the mineral content of camel milk of various origins as measured by several authors. The mean values show that the concentrations of the major salts are slightly lower than cow's milk.
It appears that the salt balance between the soluble and the colloidal forms of calcium, phosphorus and magnesium is very similar to that measured in cow's milk. The percentage of the soluble fractions reaches 30 percent of the total content (Farah and Ruegg, 1989). It also seems that the proportion of soluble calcium and phosphorus increases up to 61 and 75 percent respectively when milk is collected in the hot season from animals managed along traditional extensive lines.
Vitamins. The vitamin content of camel milk differs from cow's milk in that it includes a higher level of vitamin C and niacin (Table 8). Conversely, the amount of vitamin A is much lower, varying between 12.9 IU/100 g (Ahmed, Awad and Fahmy, 1977) and 50.0 IU/100 g (Sawaya et al., 1984). Because only incomplete information is available on the vitamin content of camel milk, the figures mentioned above should be treated with caution.
General principles. Milk is a biological substance that is particularly susceptible to degradation through the action of microbes and enzymes. This situation mainly arises from its complex composition, where spoilage organisms can find a large variety of nutrients. The high water content and neutral pH also facilitate undesirable changes.
Cheese-making technology aims to preserve milk so that consumption can be postponed for periods from a few days to several months. The preservation of the product is obtained mainly through lactic acidification and limited dehydration. These operations take place during the two first steps of processing, the setting (or coagulation) and draining phases. For many cheese types, a third phase known as ripening then takes place. This induces changes in the coagulum, separated during draining, caused by complex microbial and enzyme reactions.
In cheese making, control of the preservation process allows these changes to be manipulated in order to obtain a large number of cheese types with distinct physical-chemical and microbial composition and sensory characteristics. The resulting cheeses can be divided into four main categories depending on the level of preservation brought about by acidification (pH) and reduction of the water content (Aw or water activity).
4.3 - 4.5
0.980 - 0.995
4.5 - 4.8
0.970 - 0.990
4.8 - 5.2
0.940 - 0.970
5.0 - 5.2
0.885 - 0.905
Milk clotting coincides with the destabilization of the original micellar state of milk casein. In practice, destabilization is effected by two methods:
The mechanisms of the clotting methods are completely different but both lead to the formation of a coagulum called a curd or clot. The physical and rheological properties of the curd depend on the clotting method used (Table 9).
In typical cheese making, the two methods are never used separately but the balance of each is well defined for a particular cheese variety. The different cheese categories can be identified on this basis as follows:
The fresh coagulum is physically unstable, which leads to a progressive and spontaneous separation of the curd and whey. This development is characterized by segregation of the different components of the milk solids. Most of the water and lactose and a small fraction of the fat and protein accumulate in the whey. Most of the protein and fat are progressively concentrated into the cheese curd according to the method used to drain the whey.
In addition to its clotting effect, the acidification process plays a key role in eliminating the colloidal minerals of the casein micelles. The final solubility level of calcium and phosphorus determines the draining rate of the curd and, in turn, the texture and total solids content of the cheese.
The processing parameters for each type of cheese aim to develop the curd and, at the same time, an acidity profile which induces a specific acidity level and physical-chemical composition. Typical acidity development profiles must be followed during the draining process in order to produce different varieties successfully. This includes the need to know the strength of the lactic acid bacteria and to understand and control the development of lactic starters.
At the end of draining, the composition, volume and shape of the curd are well defined. At this stage, most cheese varieties are placed in ripening rooms. The purpose of this final processing phase is to modify and improve the appearance, composition, texture, flavour and nutritive value of the cheese.
From a chemical standpoint, ripening corresponds to an enzymic development of the curd in which proteolysis and lipolysis are mainly dominant. Casein is hydrolysed during ripening into fractions of low molecular weight: polypeptides, peptides, amino acids and ammonia. The fat is less modified in the majority of cheeses but, conversely, more hydrolysed in some blue types of soft cheeses. As a result, fatty acids, glycerol, aldehydes and ketones are liberated and accumulated in each type of cheese according to a typical profile.
Proteolysis and lipolysis are caused by numerous enzymes of various origin: endogenous milk enzymes, the residual activity of milk clotting enzymes, microbial enzymes produced by moulds and bacteria and yeasts growing into or on the surface of the cheese. This last category is dominant in cheese varieties ripened by these microflora. For cheese without external or internal flora, hydrolysis is much lower.
The optimum pH for the enzymes is generally near neutral (pH 7.0). At the end of draining, the cheese pH, around 4.5 to 5.2 depending on the variety, is too low and unsuitable for optimal development. It is thus necessary in practice to increase the pH, which may be done as follows:
The cheese ripening processes are complicated and specific to each cheese variety. From a practical standpoint, several factors such as regulation of room climate (temperature, humidity, air flow), time and handling (turning and cleaning) are used to obtain a standard product in accordance with the required composition and taste characteristics.
Fresh (cottage) cheese. Fresh cheeses are distinguished by various technological characteristics giving each variety its individual character.
Coagulation during fresh cheese production is mainly acid in character. The cheesemaker reinforces the production of lactic acid first by inoculating the milk with measured amounts of lactic starters (0.5 to 3.0 litres/100 litres) and by adjusting the milk temperature for optimum growth (18 to 27°C). Second, the activity of the milk-clotting enzyme is limited by the use of very low amounts (1 to 5 ml/100 litres) and by setting the temperature as far as possible from the optimum.
Consequently, the progress of coagulation depends closely on acidity development and the subsequent decrease of pH. This development is quite slow. The clotting time varies from 6 to 15 hours and the cutting time increases to around 16 to 48 hours. This long coagulation time encourages fat creaming when full or standardized milk is processed. In order to avoid this, the use of homogenized or skimmed milk is recommended.
At the end of the coagulation period, the acidity is high (0.65 to 1.00 percent), the pH value is low (pH 4.5 to 4.8) and the curd takes on its particular rheological properties, such as high firmness and brittleness and good whey permeability.
The ability of acid curd to drain is extremely limited. The final cheese solids content is thus generally less than 30 percent, with a range between 12 and 22 percent.
Spontaneous syneresis remains slow and weak because of the high demineralization of the casein micelles and subsequent low curd elasticity. To obtain reasonable wheying-off times and well drained cheese, it is often necessary in practice to apply some physical treatments to the curd, which must always be carried out carefully because the curd is fragile.
In traditional processing, these treatments consist of cutting, pressing and mixing the curd. The process takes place while moulding the clot in draining bags or hoops and during turning. The total draining period lasts for 24 to 36 hours at room temperature (20 to 30°C). With modern centrifugal processing techniques, removal of whey occurs instantaneously in the separator. This harsher mechanical treatment requires a firmer curd, achieved by increasing the amount of clotting enzyme and the temperature of renneting.
At the end of draining, the cheese is characterized by a low dry-matter content and low pH values and mineralization (0.1 percent calcium and 0.2 percent phosphorus). As a result, the cheese lacks cohesion and looks like a soft watery paste. For further preservation, the product must be packaged into rigid, airtight cups in order to prevent wheying off and external contamination.
Consumption of fresh cheese generally occurs without ripening immediately after draining. The basic acid taste can be modified by adding a large variety of ingredients, such as cream, salt, sugar, spices or jams. Shelf-life is limited to a few days under refrigerated conditions (0 to 4°C) but can be increased by applying heat treatment or air drying.
Soft cheese. Soft cheese making is typified by:
Coagulation takes place using the equal action of milk-clotting enzymes and lactic acid. The average amounts are 15 to 25 ml/100 litres and 1 to 3 litres/100 litres for enzyme and starter respectively. Milk temperature is adjusted for good enzyme and starter activity. The clotting conditions impart specific physical properties into the curd: medium firmness, elasticity and brittleness.
Soft cheese is drained using mild mechanical treatment in accordance with the rheological behaviour of the clot. These conditions lead to the formation of a cheese of average solids content (45 to 55 percent) and residual mineral content (0.2 to 0.3 percent calcium) and a low pH (4.7 to 4.9). The curd is of average cohesiveness. The size and weight of the cheese is also average.
Ripening of soft cheese is fairly rapid (two to eight weeks) depending on the water content and the occurrence of microflora with high enzymic activity growing on the surface or inside the cheese. On the basis of the dominant type of ripening organism, three soft cheese categories may be distinguished:
Semi-hard and hard cheese. The processing of these cheese classes is defined by the following characteristics:
The main enzyme coagulation action is obtained by using high concentrations of clotting enzymes (20 to 40 ml/100 litres) and by adjusting the temperature to a level which is moderate for enzyme activity (32 to 40°C). For the same reason, lactic acid development remains very limited through the use of low initial amounts of mesophilic and/or thermophilic bacteria (0.5 to 1 litre/100 litres).
Under these conditions, clotting time is short (10 to 30 minutes). The curd possesses the typical properties of an enzymic gel: good elasticity, low brittleness and high syneresis ability, all of which are directly induced by high casein mineralization.
Drainage of the curd is fast and thorough. The high solids content (45 to 70 percent), is reached by applying physical treatments such as cutting, stirring, washing, cooking and pressing. These actions are improved by moderate parallel acid development of the curd. This development is aimed at obtaining high curd mineralization (1.2 to 1.8 percent calcium) and a large cheese. The pH value at the end of draining ranges from 5.0 to 5.2.
Draining takes from 20 to 48 hours, with the greater part of the whey running off during the first two hours. The final pressing is carried out mainly to compact the curd grains rather than to induce further draining.
The process of cheese ripening is conditioned by internal neutralization determined mainly by the interaction between lactic acid and calcium. In some cheeses, complementary neutralization takes place through development of microflora on the surface, as in soft cheeses.
These interactions induce progressive neutralization and subsequent protein and fat breakdown in the cheese. Among the various sources of enzymes involved in ripening, the clotting proteases and those of microbial origin are most involved. For some hard cheese types (Gruyère and Emmenthal), specific propionic fermentation develops during the second part of the ripening period, providing the typical holes and flavour.
The average ripening time is three weeks to six months under factory conditions and can extend to 6 to 12 months for traditional production according to dry- matter content. The temperature of the ripening rooms is normally between 12 and 14°C. When propionic fermentation is required, the temperature is increased to about 20°C during the latter part of ripening.
During the ripening period, it is necessary to cure the cheese surface in order to regulate growth of certain microbial flora or to inhibit adverse growth on products with dry coats or crusts.
Cheeses made from whey. Whey generally contains significant amounts of whey proteins (0.75 to 0.95 percent), which consist largely of lactalbumin and lactoglobulin (Webb, Johnson and Alford, 1974). These proteins coagulate when heat is applied (Lyster, 1979) and are easily collected after precipitation. Heat denaturation begins when the temperature is close to 65°C and increases with the time-temperature combination used during heating.
In practice, these properties are exploited to produce the specific dairy products known as "whey cheeses", which are traditionally manufactured largely in the countries around the Mediterranean (Pernodet, 1979; Ramet, 1985c; Kandarakis, 1986). These products are not genuine cheeses, because they are not obtained directly from the coagulation of milk.
Processing consists of selecting whey rich in soluble proteins and, preferably, drained from hard cheeses made from raw or thermized milk. The whey is slowly heated for 20 to 45 minutes to between 78 and 95°C and held at that temperature for a further 15 to 30 minutes. The first clotted particles appear at around 78 to 80°C, depending on acidity and whey protein concentration. The clusters of coagulated proteins gather at the surface of the whey and can be easily collected using simple techniques such as filtration through cloth or skimming with a ladle.
The degree of heating influences the quality of the cheese. Above 88°C, the texture tends to become drier, harder and more granular. A cooked flavour makes the cheese less acceptable. Below these temperatures, the clusters are small, crumbly and difficult to gather; the precipitate is watery and drains slowly. Cheese yields are variable and closely dependent on the richness of the whey and the water content of the final product. For example, the yield from cow's milk cheese whey can reach 3.5 kg/100 kg. The solids content of whey cheese is around 20 to 25 percent; the percentage of fat in dry matter is generally between 11 and 45 percent.
The keeping quality of the product is poor because of the high water content and reduced acidity (pH 5.5 to 6.2). The cheese can be eaten like other cheese or used in cooking and pastry.
Numerous alternatives to the above-mentioned process are used. These processes are aimed at improving recovery of whey proteins and increasing the total solids of the cheese to 40 to 50 percent, as with ricotta or brucciu. The best method consists of adding from 20 to 30 percent whole milk to the whey or increasing the acidity to pH 4.6 to 5.8 (Kandarakis, 1986). Acid development may be supplemented with organic acids such as acetic, citric, lactic or tartaric, with mineral acids like phosphoric acid or by using acid whey. The addition of salts (0.1 to 0.5 percent calcium and/or sodium chloride) is sometimes used for the same purpose.
The most modern processes involve more sophisticated techniques such as ultrafiltration to concentrate whey proteins prior to heating or centrifugation to enhance separation of precipitates.
Whey cheese is generally accepted by local consumers because of its typical taste and smooth texture. Nutritional value is high because of the richness of whey proteins in essential amino acids, mainly cystine and methionine (Porter, 1978).
The fat globule. The fat in cow's milk is emulsified as fat globules 3 to 5 micrometres in diameter. The fat globules have a heterogenous structure composed of three parts: an external membrane, a central core of triglycerides with high melting points and an intermediate stratum of triglycerides with low melting points. The stability of the fat emulsion is dependent on the integrity of the globular structure.
Butter making consists of destabilizing the emulsion in order to concentrate the fat content from 3.5 to 4.5 percent in milk to 82 percent in butter. This transformation proceeds through several mechanical and chemical stages.
The phases of butter making. The first step of butter processing is to separate the milk to obtain cream with a fat content about ten times higher than milk. The process can be carried out by natural creaming of milk in the traditional way or by modern centrifugal techniques.
Sometimes, when the acidity is in excess of 0.2 percent, it is necessary to neutralize the cream in order to avoid coagulation during heating and development of disagreeable flavours in the butter. Neutralization is carried out either by dilution with water and further separation or chemically using sodium hydroxide.
Heat treatment of the cream is recommended to eliminate the bacteria and enzymes that could cause quality problems and spoilage. For this reason, the cream is heat treated at a temperature of 90 to 95°C for between 30 seconds and two minutes. Heating takes place in containers (vats, cauldrons, etc.) or in tubular or plate exchangers. Optional de-gassing may be used to remove disagreeable flavours which may be dissolved in the water or fat. After heating, the cream is cooled to between 8 and 14°C for ripening. Ripening, or ageing, is used to lower the pH slightly, to develop flavour and to regulate fat crystallization.
Biological ripening is effected by adding a lactic starter for 10 to 16 hours at 8 to 14°C. The cream acidity required for traditional churning methods is 0.40 to 0.45 percent but only 0.20 to 0.35 percent for continuous butter making. During ripening, lactic acid bacteria also produce flavouring molecules such as diacetyl, which are important in building up the typical cooked or hazelnut flavour.
Physical ripening is carried out to regulate the proportion of solid and liquid fat. At low temperatures all the fat is crystalline, leading to long churning times and very hard butter texture. If the temperature is too high, all the fat melts and the butter will be very soft, resulting in considerable fat losses in the buttermilk. Another important factor in physical ripening is control of the cream cooling rate. If this is too slow, large fat crystals will form, leading to a fractured, sandy texture in the butter. If the cooling rate is too fast, the crystals are small and not detectable in the mouth. As a result, the texture of the butter is significantly improved.
The final phase of butter making consists of breaking a limited number of fat globules in order to expel a small amount of liquid fat that will ensure a continuous bond with adjacent globules. When this bonding takes place, grains of butter appear. In order to facilitate this development, the fat globules have to come together during churning. This is brought about by the formation of foam as churning of the cream begins. As churning continues, the bubbles get smaller, making the foam more compact and so applying pressure on the fat globules. As the bubbles become increasingly dense, more liquid fat is squeezed out and the foam becomes so unstable it collapses. The discharge of the liquid fat is caused by the impact of globules against each other and against the surface of the churn. Immediately the butter particles appear, the foam collapses and buttermilk separates from the butter.
When the butter particles become large enough for separation, the buttermilk is run off and churning, or working, continues. If the butter is intended for storage, it is washed several times with water that is bacterially and chemically pure. After churning and washing, working the butter in the churn ensures dispersion of any residual buttermilk droplets into the butter. Churning adjusts the water content of the butter to the legal maximum of 16 percent, when such regulations exist.
After working, the butter is removed from the churn and packed. Packaging materials and containers must provide effective protection against microbial contamination and must be opaque, as light increases fat oxidation. The keeping quality of the product depends on the residual content of microbes and enzymes and on the storage temperature. For storage over several months or years, deep freezing to -20 to -35°C is required. In traditional production methods, the butter is melted and boiled to destroy spoilage organisms and enzymes and kept for several months in cans or glass or earthenware jars.