Manufacture Of Cheddar Cheese Research Articles

Milk protein standardization by ultrafiltration for Cheddar cheese manufacture

SummarySkim milks were concentrated by ultrafiltration. Cream was added to the retentates to give cheesemilks standardized to a casein: fat ratio of ∼ 0·74 with protein levels ranging from 30 to 46 g/1. Pasteurized cheesemilks were renneted on a volume basis (22 ml single strength calf rennet/100 1) and converted to Cheddar cheese in 500 1 cheese vats. Set temperatures were reduced from 31 to 28 °C with increasing milk protein level to normalize curd firming rates and prevent curd shattering on cutting the curds. Cheesemaking was otherwise as normal for Cheddar cheese. The proportions of milk fat and protein lost in the cheese whey were not significantly influenced by milk protein level. Moisture-adjusted cheese yields increased with milk protein at a rate similar to that predicted by the Van Slyke cheese yield equation. However, owing to the negative correlation between cheese moisture and milk protein concentration, actual yields increased at a lower rate with respect to milk protein than moisture-adjusted yields. Increasing milk protein levels resulted in significant (P < 0·01–0·001) decreases in the concentrations of moisture, moisture-in-non-fat cheese solids and fat-in-dry matter in the cheese and increases (P < 0·05–0·001) in the levels of protein, salt-in-moisture, Ca and P. While increasing milk protein concentration resulted in significant (P < 0·05–0·01) reductions in the levels of water-soluble N at all stages of the 270d ripening period, it had little influence on the sensory scores awarded for aroma/flavour at 180 or 270d.

An application in cheddar cheese manufacture for a strain of Lactococcus lactis producing a novel broad-spectrum bacteriocin, lacticin 3147.

Lactococcus lactis DPC3147, a strain isolated from an Irish kefir grain, produces a bacteriocin with a broad spectrum of inhibition. The bacteriocin produced is heat stable, particularly at a low pH, and inhibits nisin-producing (Nip+) lactococci. On the basis of the observation that the nisin structural gene (nisA) does not hybridize to DPC3147 genomic DNA, the bacteriocin produced was considered novel and designated lacticin 3147. The genetic determinants which encode lacticin 3147 are contained on a 63-kb plasmid, which was conjugally mobilized to a commercial cheese starter, L. lactis subsp. cremoris DPC4268. The resultant transconjugant, DPC4275, both produces and is immune to lacticin 3147. The ability of lacticin 3147-producing lactococci to perform as cheddar cheese starters was subsequently investigated in cheesemaking trials. Bacteriocin-producing starters (which included the transconjugant strain DPC4275) produced acid at rates similar to those of commercial strains. The level of lacticin 3147 produced in cheese remained constant over 6 months of ripening and correlated with a significant reduction in the levels of nonstarter lactic acid bacteria. Such results suggest that these starters provide a means of controlling developing microflora in ripened fermented products.

Seasonal influences on Cheddar cheese manufacture: influence of diet quality and stage of lactation

The influence of diet quality and stage of lactation on Cheddar cheese moisture and yield were segregated and the effects of other potential variables (cow breed, milk protein genotype, age and level of mastitis) were minimized. An experiment was conducted using milk of cows in mid lactation or late lactation fed on high or low quality diets. Increased moisture in cheese was largely due to the influence of cow diet. Milk from cows given high quality diets produced lower moisture cheese. Cheese yield was influenced by diet and stage of lactation. Cows in late lactation fed on high quality diets always gave milks yielding the most cheese. Fat and protein recovery in cheese was not influenced by diet or stage of lactation. The results suggest that low quality diets may induce changes in the complex interactions of milk proteins and other components that affect cheesemaking.

Microbiological Safety of Blue and Cheddar Cheeses Containing Naturally Modified Milk Fat

Milk containing naturally modified fat was obtained by feeding lactating dairy cows a Control diet and two experimental diets containing either extruded soybeans or sunflower seeds. Milk from cows fed the experimental diets contained higher levels of both long chain (C18-C18:2) and unsaturated fatty acids than the milk from cows fed the Control diet. Each milk was pasteurized, standardized to 3.6% milk fat, and inoculated with Listeria monocytogenes (strains Scott A and V7), Salmonella typhimurium and Salmonella senftenberg, before manufacturing into Blue or stirred-curd Cheddar cheeses. Populations of L. monocytogenes and Salmonella spp. were monitored during manufacture and aging using Oxford and Xylose Lysine Desoxycholate agars, respectively. During the manufacture of Blue and Cheddar cheese, and during the aging of Blue cheese, behavior of Salmonella spp. and L. monocytogenes in the experimental cheese was similar to the Control cheese. During aging of Cheddar cheese, the rate and extent of decline of Salmonella spp. and L. monocytogenes varied among the cheeses. Declines correlated with the accumulation of specific fatty acids, namely C12, C14, C18:1 and C18:2. These fatty acids were also found to be inhibitory to S. typhimurium and L. monocytogenes when incorporated into tryptic soy agar plates at 37°C. Therefore, the natural fat modification of Blue and Cheddar cheeses enhanced the safety of these cheeses.

Rennet: Current trends and future research

Abstract A worldwide shortage of calf rennet for cheese production has existed for several decades. Bovine pepsin and to a certain extent porcine pepsin and plant coagulants have been used as rennet substitutes, but these have not been commercially successful owing to their extensive proteolytic nature and other inherent drawbacks. Investigations to develop other alternatives have resulted in the introduction of various microbial coagulants which have found markets in several countries despite certain shortcomings when compared with traditional calf rennet. However, in some countries, microbial coagulants have not been accepted for regular cheese manufacture because they are believed to result in a reduced yield and a lower quality product; this is particularly applicable to cheddar cheese manufacture. Immobilization of proteases for milk coagulation has received renewed thrust; this may convert renneting of milk to a continuous operation. However, this technique has yet to be applied on an industrial scale; such a process is not likely to be successful since immobilized enzyme technology is dependent on the ability of a small, rapidly diffusing substrate to move quickly around the immobilized enzyme. In contrast, casein micelles are large and diffusion is slow, so the proteolysis rate is very slow. Microbial rennets offer an attractive target for genetic engineers since it may be possible to alter their structure/function characteristics to match those of calf rennet; application of site‐directed mutagenesis could be particularly rewarding. The term “calf rennet,” in the cheese industry, generally refers to an enzyme extract obtained from the fourth stomach (abo‐masum) of 10‐ to 30‐day‐old calves and used to coagulate milk for cheese production. The purified milk‐clotting enzyme present in crude rennet preparations is known as “rennin” or “chymosin.” The designation “chymosin” is now recommended in the international enzyme nomenclature (renin is associated with hypertension and is derived from the kidney). Calf rennet is referred to as “animal rennet” frequently, the term “chymase” appears in the older literature as a synonym for rennet. In more general usage, however, any milk‐clotting enzyme preparation yielding a relatively stable curd is designated as rennet. Christian Hansen is credited with the first industrial production of rennet in 1874 (1). This enzyme was called chymosin (EC 3.4.23.4); it belongs to the group of aspartic proteases and is the standard against which all other types of milk‐clotting enzymes are compared. Since the origin of cheese production, the manufacturing process has always needed soluble enzymes to clot the milk; it has been adapted to the properties of calf rennet, an essential enzyme for this purpose. Rennet coagulates milk rapidly at its natural pH with little further degradation of the milk proteins. The zymogen, prorennin, is converted to chymosin at pH below 5.0, but optimally at pH 2.0 (2). Chymosin, a strong protease, has been crystallized (3) and detailed crystallographic studies of the enzyme have been reported by Bunn et al. (4). Besides a slightly high proteolytic activity, another disadvantage of this enzyme is that it is extracted from the abomasum of the unweaned calf. As the calf ages, chymosin is replaced by pepsin, although in cattle, the secretion of chymosin never comes to a complete stop. Although pepsin can clot milk, it has a tendency to result in higher fat losses because the curd formed has a more open, looser structure than that formed with chymosin; the cheese produced also has a softer body than desired. Pepsin and a commercial product under the trade name Metroclot (5, 6) have, however, been used for the production of a variety of cheeses in the past. The kinetic properties and amino acid composition of chymosin and pepsin have been extensively studied. There is more or less a consensus of opinion in favor of chymosin as the enzyme for cheese making; presently three companies are producing calf chymosin through recombinant DNA technology. The cloned chymosin preparations produced by different microorganisms have been tested in various countries for cheese manufacture. No major differences could be detected among cheeses made with cloned chymosin and those made with the natural enzyme. This review describes the past, present, and future status of rennet substitutes utilized as milk coagulants and applications of modern biological tools to strain improvement and process development.

Component Values for Milk Used in Cheddar Cheese

Interest in valuing the solids components of milk has continued to rise in recent years. A component valuation scheme for milk is devised to be used in cheddar cheese manufacture. Using daily plant-level data, and employing a flexible functional form, a plant-specific cheese yield function is estimated. This yield relationship serves as the basis for a valuation scheme that divides the value of all end products among three solids components of milk. These values are compared to a pricing scheme that recently appeared in the Great Basin Federal Milk Order. The plan presented here places a higher value upon milk of high fat test than does the order plan.

Effect of milk protein standardization, by ultrafiltration, on the manufacture, composition and maturation of Cheddar cheese

SummarySkim milks were pre-acidified to pH 6·4 and concentrated by ultra-filtration to give retentates with protein levels of 210 g/1. Retentates were blended with skim milk and cream to give standardized milks with protein levels ranging from 30 to 82 g/1. These were used for the manufacture of Cheddar cheese in conventional equipment. Increasing milk protein level resulted in reduced gelation times, increased curd firming rates and a decrease in the set-to-cut time when cutting at equal firmness values (i.e. elastic modulus, G′, ∼ 16 Pa). As the curd firming rates increased with milk protein level, it became increasingly difficult to cut the curd cleanly, without tearing, before the end of the cutting cycle. Reflecting the tearing of curd, and consequent curd particle shattering, fat losses in the running wheys were greater than those predicted on the basis of volume reduction (due to ultrafiltration) for milks with protein levels > 50 g/1. Reduction of setting temperatures, in the range 31–27 °C, and the level of added rennet brought the set-to-cut times and curd firming rates of concentrated milks closer to those of the control milk. While increasing milk protein level in the range 30–70 g/1 had little effect on cheese composition, it resulted in slower proteolysis and maturation.

Comparisons between the electrophoretic pattern and textural assessment of aged Cheddar cheese made using various levels of calf rennet or microbial coagulant (Rennilase 46L)

The set-to-cut time, rennet concentration and set temperature were varied during the commercial manufacture of Cheddar cheese in nine Damrow cheese vats on each of three consecutive days. In addition, in two separate pilot plant trials, the setting conditions and coagulant type (calf rennet or Rennilase) were varied during early and mid-season manufacture of Cheddar cheese. The effect of variations on proteolysis were assessed by measuring the non-casein nitrogen (NCN) and the non-protein nitrogen (NPN) content of the cheese. Protein degradation was also estimated by determining the breakdown of αS1 and β-caseins using polyacrylamide gel electrophoresis (PAGE), at intervals over a six-month period. At the same time, changes in cheese texture were assessed by a trained panel. NPN levels were similar in all cheeses and tended to increase slightly over the six month period. The NCN level increased markedly in cheese from both pilot plant trials but less so in cheese from the commercial trial due to a lower moisture content and a higher pH. The specificity and activity of Rennilase differed from that of calf rennet and this was reflected in sharply elevated NCN levels during early maturation. The breakdown of αS1-casein was both rapid and extensive when calf rennet was used as the coagulant. The degradation of β-casein was greater in cheese produced with Rennilase than with calf rennet. Higher residual coagulant levels in the cheese due to the increasing concentrations of rennet (10·6, 16·0 and 24·0 ml/100 litres milk) and Rennilase (4·6, 7·0 and 10·5 ml/100 litres milk) used initially were reflected in the significant (P < 0·001) early breakdown of αS1-casein. In spite of the significant variation in the early breakdown of αS1-casein, sensory evaluation of the textural characteristics of the cheese was not sensitive to a variation in coagulant concentration. Cheese made using Rennilase was significantly (P < 0·001) less rubbery, less smooth, firmer and grittier than cheese made using calf rennet but differences were small, suggesting that the type of coagulant had little effect on sensory scores. Independent of rennet concentration and set temperature during commercial cheese manufacture, the rate of breakdown of αS1-casein was higher for a 50-min set-to-cut time than for a 30-min set-to-cut time. The degradation of β-casein was not influenced to the same extent. As the variation in αS1-casein breakdown was of a similar magnitude to that found for varying levels of coagulant, the strong correlations (r2 = +0·800, −0·914 and −0·651) between αS1-casein breakdown and changes in curdiness, stickiness and smoothness with time do not fully explain the increase in smoothness and decrease in curdiness with increasing set-to-cut time. However, increasing the set-to-cut time from 30 to 50 min significantly (P < 0·1) increased the cheese moisture content and significantly (P < 0·001 to P < 0·1) reduced the cheese salt content, the pH and the calcium level. Consequently, cheese texture is determined primarily by its pH and chemical composition. Subsequent, early changes in texture are determined by αS1-casein breakdown. Downgrading for body defects such as curdiness at 35 days does not accurately predict what the cheese texture will be after (6 months) storage. The results of this work have shown that the body of a cheese can be singled out from others on the basis of a curdy texture at 1 month, but is indistinguishable at 6 months.

Use of a Nisin-Producing Starter Culture Suitable for Cheddar Cheese Manufacture

A nisin-producing starter culture that produced acid at rates suitable for Cheddar cheese manufacture was developed by combining the naturally occurring lactose-fermenting, nisin-producing, and proteinase-positive strain Lactococcus lactis ssp. lactis NCDO 1404 and the lactose-positive, nisin-positive, proteinase-positive transconjugant Lactococcus lactis ssp. cremoris JS102. Cheddar cheese manufactured with 1% NCDO 1404 and .5% JS102 as starter cultures exhibited normal acid development; the cheese was of normal composition and contained ca. 700 IU of nisin/g. Experiments using a nisin resistance factor from plasmid pFG010 to increase resistance to nisin by L. lactis ssp. lactis C2 revealed that nisin resistance and acid development by the recipient increased in the presence of nisin. However, the mechanism of resistance encoded by pFG010 resulted in inactivation of nisin in the test system.

Effects of speed and duration of cutting in mechanized Cheddar cheesemaking on curd particle size and yield

SummaryThe speed and duration of cutting the coagulum were varied during the commercial manufacture of Cheddar cheese in nine Damrow cheese vats on each of 3 consecutive days. The effects of the variations were assessed by determining the curd particle size distribution and fat and fines losses into the whey at draining. Short durations of cutting at slow speeds produced small curd particles and high fat losses into the whey. As the duration and/or speed of cutting was increased, average curd particle size increased while fat losses into the whey decreased. A maximum particle size and minimum fat level were reached after the Damrow's knife panels had completed approximately 37 revolutions. Further increases in speed and/or duration of cutting decreased curd particle size. Fat losses did not increase significantly. The hypothesis suggested to explain the results of these trials is that curd particle size is determined not by the cutting programme alone, but by a combination of the speed and duration of cutting and the subsequent speed of stirring prior to cooking. Fat losses in the whey are also influenced by this combination. Based on this hypothesis, a model is proposed which explains how variation in cutting speed and duration, followed by a constant stirring speed, determines curd particle size distribution in a Damrow cheese vat. Implications of the present findings in relation to cheese yield (losses of fat and fines into the whey) are discussed.

Influence of Pasteurization of Milk on Protein Breakdown in Cheddar Cheese During Aging

Cheddar cheese was made from pasteurized (63°C for 30min) and raw milk. The extent and characteristics of proteolysis were monitored during 6 mo of aging. Pasteurization decreased the rate of production of 12% TCA-soluble nitrogen during cheese aging. The breakdown of β-casein and the high molecular weight proteolytic products of αs-caseins that comigrated with β-casein (as measured by SDS-PAGE) was slower in cheese made from pasteurized milk. The total amount of water-soluble peptides measured by HPLC at wavelengths of 214nm and 280nm was similar in cheeses made from pasteurized and raw milk. However, there was a higher ratio of water-soluble hydrophobic to hydrophilic peptides in Cheddar cheese made from pasteurized milk. Thus, milk pasteurization prior to Cheddar cheese manufacture influenced both the extent and characteristics of proteolysis during Cheddar cheese aging. Milk pasteurization caused heat-induced interaction of whey proteins with casein and resulted in retention of additional whey protein in the cheese beyond that normal amount which is soluble in the aqueous phase of the cheese. The presence of heat denatured whey protein in cheese may influence the accessibility of caseins to proteases during cheese aging and cause differences in proteolysis during aging. These differences may be one factor that contributes to differences in flavor development in Cheddar cheese made from pasteurized and raw milk.

Growth and Activity of Lactococcus lactis ssp. cremoris in Ultrafiltered Skim Milk

Ultrafiltration of milk causes physico-chemical and compositional changes in the concentrated product. Ultrafiltration of skim milk to give concentration factors 1 to 5 caused an increased buffer capacity as concentration increased. Milk proteins (casein and whey proteins) and minerals (calcium, phosphorus, magnesium, potassium, and sodium) increased with the increased concentration, NPN remained fairly constant, and lactose content decreased slightly. The effect of concentration of UF retentate on growth, lactic acid production, and proteolytic activity of Lactococcus lactis ssp. cremoris strains 103, 108, and 208 was determined. By using large inocula to overcome high buffer capacity of higher concentration retentates, there was an uncoupling of growth from lactic acid production. Although large amounts of lactic acid were produced in higher concentration retentates, the high buffer capacity resisted the change of pH. A simulated Cheddar cheese manufacturing test was used to determine inoculum size of starter cultures necessary to give the desired pH change in UF retentates. It was possible to overcome the effect of buffer capacity up to concentration factor 4 by use of increased inoculum size of strongly proteolytic strains, but for retentate of concentration factor 5 the use of a 10% inoculum failed to overcome the buffer effect.

Size‐Exclusion HPLC Separation of Bitter and Astringent Fractions from Cheddar Cheese Made with Added Lactobacillus Strains to Accelerate Ripening

ABSTRACT Lactobacillus strains were added as an adjunct to the regular lactic starter in Cheddar cheese manufacture in order to accelerate ripening. Microbial cheese proteolysis resulted in the release of free amino acids which were extracted with the astringent and bitter fractions and separated by size‐exclusion and reversed‐phase HPLC chromatography. Lactobacillus strains generally increased the degree of proteolysis. L. plantarum and L. brevis produced off‐flavors possibly due to an accumulation of medium‐size peptides. The control cheese (without lactobacilli) had the most peptides with a mean molecular‐ weight of &lt; 1000 daltons and had a flavor described as slightly bitter. Addition of L. casei‐casei L2A accelerated ripening and yielded a well‐aged Cheddar cheese without any bitterness even after 7 months at 6°C.

Effects of Bacterial Protease on the Quality of Cheddar Cheese

A heat-stable protease secreted by Pseudomonas fluorescens T25 was used to accelerate the ripening of cheddar cheese. Additions of the bacterial protease to pasteurized milk prior to the addition of a renneting agent (porcine pepsin) in the manufacture of cheddar cheese resulted in a product which showed a higher intensity of cheddar flavour (P < 0.05) and a better preference score (P < 0.01) within six months of aging when compared to that of the control cheese not containing the bacterial protease. Neither off-flavour nor bitterness was detected in the protease-treated cheese throughout the ripening period. The results of this study indicate that protease-T25 from Ps. fluorescens T25 may be a suitable agent to accelerate the ripening of cheddar cheese and also to enhance cheddar flavour.

Permeate Flux During Ultrafiltration of Whey: Influence of Milk Coagulant Used for Cheese Manufacture

A pilot-scale plate and frame UF system was used to fractionate Cheddar cheese whey and study the effects of different commercial milk coagulants on permeate flux. Coagulants used in this study were calf rennet, Mucor pusillus protease, and Mucor miebei protease. Whey UF performance studies were conducted at a commercial Cheddar cheese plant and at Cornell under controlled conditions. Ultrafiltration was done in a continuous mode and initial concentration factor was set at 2× to simulate the first stage of a multistage whey UF system.Permeate flux decline was rapid in the first 30min of UF for all wheys studied. More important, the type of milk coagulant used in cheese making had a profound effect on permeate flux during whey UF. No differences in the gross composition of the various wheys were correlated with differences in permeate flux. The highest permeate flux was measured for UF of whey produced during manufacture of Cheddar cheese using coagulant derived from Mucor pusillus. Lowest permeate flux was measured for UF of whey produced during manufacture of Cheddar cheese using calf rennet. Whey from cheese manufactured using Mucor miebei coagulant had flux performance intermediate to Mucor pusillus and calf rennet. The impact of milk coagulants on whey UF process efficiency should be considered by cheese makers.

Use of Recombinant Chymosin in the Manufacture of Cheddar and Colby Cheese

Recombinant chymosin purified from E. coli K12 (Genencor Inc.) was examined for its suitability to manufacture Cheddar cheese. Commercial calf rennet or recombinant chymosin with and without Bacillus subtilis cellular homogenate was used to determine if enzyme sources affected Cheddar cheese quality or yield. Cheese was evaluated at 3 and 6 mo of age. Purified enzyme was mixed with cellular extracts from B. subtilis at approximately one and four times the protein concentration in the initial homogenates. Cellular materials did not affect clot times, cheese quality, or inhibit cheese acid development. No differences were noted between the yields of Colby cheese manufactured with calf rennet and recombinant chymosin. Initial Cheddar cheese manufactured with recombinant chymosin were excessively bitter, typical of cheese retaining excessive amounts of proteases. Dilution of purified prochymosin preparations increased total enzymic activity 2.7 times. Evidently, the purified recombinant preparations contained some oligomers that became active after being incorporated into the cheese mass, causing bitter flavors to develop during aging. Diluted enzyme preparations were used successfully in the manufacture of Cheddar cheese.

A comparison of the effects of storage of raw milk at 2°C and 6°C on the yield and quality of Cheddar cheese

The advantages of storing raw milk, which is to be used for Cheddar cheese manufacture, at 2°C rather than at 6°C was examined. Storage of milk at the lower temperature effectively reduced the level of psychrotroph growth, and after 4 days the psychrotroph counts in samples stored at 2°C were 100-fold lower than those found in samples stored at 6°C. There was no advantage in terms of cheese yield in storing milks at the lower temperature, but an overall improvement in cheese quality was noted in samples produced from milk stored at 2°C.

Isolation of Streptococcus lactis C2 Mutants Selected for Temperature Sensitivity and Potential Use in Cheese Manufacture

In an attempt to select streptococcal strains that could be used to accelerate the cheese ripening process, we used ethyl methanesulfonate treatment to isolate Streptococcus lactis C2 temperature-sensitive mutants exhibiting normal growth patterns at 30 to 32°C but not at the highest cooking temperatures used in Cheddar cheese manufacture. Three classes of mutants isolated included: 1) mutants unable to grow and produce acid at 38 to 40°C; 2) mutants unable to grow at 38 to 40°C but that continued to produce acid at a rate similar to temperature-sensitive strains of S. cremoris; and 3) mutants that underwent lysis when the cells were shifted to 38 to 40°C, as evidenced by a decrease in optical density of the cell suspension and the concomitant release of the intracellular enzyme, phospho-β-galactosidase. The thermolytic variants could possibly be used to accelerate cheese ripening either by releasing their intracellular enzymes into the curd at an early stage in the cheese making process or by serving as a “delivery system” of ripening enzymes, the genes of which could be cloned into the thermolytic strains.

Thermoinducible Lysis of Temperature-Sensitive Streptococcus cremoris Strains

Temperature-sensitive Streptococcus cremoris SK11 appeared to harbor a temperate phage that was induced at the maximum cooking temperatures used in Cheddar cheese manufacture, i.e., 38 to 40°C. When incubated at 30°C in M17-lactose broth, followed by a shift to 40°C, lysis of the culture occurred within 2h. Results were similar when S. cremoris SK11 was propagated in M17-lactose broth through a simultated Pearce activity test. In 11% solids reconstituted NDM, this strain also exhibited decreased cell numbers approximately 2h after the temperature had reached 40°C in the Pearce activity test. To support the hypothesis that lysis of the cells was due to induction of phage, electron microscopy of the temperature-induced lysate revealed the presence of particles resembling phage heads and tails but only a few intact phage-like particles. Examination of other temperature-sensitive or temperature-insensitive strains of S. cremoris also suggested that lysis of temperature-sensitive strains occurred after cells were incubated at 40°C, but no lysis was observed for temperature-insensitive strains.

Seasonal trends in the efficiency of recovery of milk fat and casein in cheese manufacture

A study was undertaken to define the seasonal changes in the efficiency of recovery of milk fat and protein in curd during Cheddar cheese manufacture, and to examine the milk compositional factors that may contribute to these effects. Distinct seasonal trends in the efficiency of conversion of milk fat and protein to cheese were noted. Variations in the retention of milk fat in curd could be associated with seasonal changes in the casein to fat ratio of the milk.