Using an ultrafiltration membrane (molecular cut-off, 3000), low molecular weight compounds in bovine milk were collected (YM-3 filtrate). A hydrogen peroxide (H2O2)-like substance was generated in the YM-3 filtrate. This substance was undetected at 0 h, but increased in a time-dependent manner, peaking after 2 h of incubation at 38°C. After incubating the YM-3 filtrate with catalase and lactoperoxidase, the signal showing the presence of this substance disappeared. The substance was quantified using one chemiluminescence and three colorimetric H2O2 detection systems. In all systems, their estimates were within the same range. The amount of substance, as estimated by the chemiluminescence H2O2 detection system, was correlated with that estimated by the other three colorimetric systems (r=0·98, 0·95 and 0·87). The substance was eluted at the same position as H2O2 by gel filtration on Superdex 30. Thus, the substance had the same characteristics as H2O2. An H2O2-generating substance in either the YM-3 filtrate or whey had a molecular mass of about 600. In this study, we clarify that bovine milk is capable of generating H2O2 by utilizing a low molecular weight compound. Thus, we present a new type of H2O2-supplying system in bovine milk.

The solute carrier family 27 member 1 (SLC27A1) is an integral membrane protein involved in the transport of long-chain fatty acids across the plasma membrane. This protein has been implicated in diet-induced obesity and is thought to be important in the control of energy homeostasis. In previous reports, our group described the isolation and characterization of the bovine SLC27A1 gene. The bovine gene is organized in 13 exons spanning over more than 40 kb of genomic DNA and maps in BTA 7 where several quantitative trait loci for fat related traits have been described. Because of its key role in lipid metabolism and its genomic localization, in the present work the search for variability in the bovine SLC27A1 gene was carried out with the aim of evaluating its potential association with milk fat content in dairy cattle. By sequencing analysis of all exons and flanking regions 14 new single nucleotide polymorphisms (SNPs) were identified: 1 in the promoter, 7 in introns and 6 in exons. Allele frequencies of all the SNPs were calculated by minisequencing analysis in two groups of Holstein-Friesian animals with highest and lowest milk-fat content estimated breeding values as well as in animals of two Spanish cattle breeds, Asturiana de los Valles and Menorquina. In the conditions assayed, no significant differences between Holstein-Friesian groups were found for any of the SNPs, suggesting that the SLC27A1 gene may have a poor or null effect on milk fat content. In Asturiana and Menorquina breeds all the positions were polymorphic with the exception of SNPs 1 and 8 in which C allele was fixed in both of them.

Changes in chemical composition, proteolysis, lipolysis, texture, melting and sensory properties of low-fat Kashar cheese made with three different fat replacers (Simplesse® D-100, Avicel Plus® CM 2159 or β-glucan) were investigated throughout ripening. The low-fat cheeses made with fat replacers were compared with full- and low-fat counterparts as controls. Reduction of fat caused increases in moisture and protein contents and decreases in moisture-in-non fat substance and yield values in low-fat cheeses. The use of fat replacers in the manufacture of low-fat Kashar cheese increased water binding capacity and improved overall quality of the cheeses. Use of fat replacer in low-fat cheese making has enhanced cheese proteolysis. All samples underwent lipolysis during ripening and low-fat cheeses with fat replacers had higher level of total free fatty acid than full- or low-fat control cheeses. Texture attributes and meltability significantly increased with addition of fat replacers. Sensory scores showed that the full-fat cheese was awarded best in all stages of ripening and low-fat variant of Kashar cheeses have inferior quality. However, fat replacers except β-glucan improved the appearance, texture and flavour attributes of low-fat cheeses. When the fat replacers are compared, the low-fat cheese with Avicel Plus® CM 2159 was highly acceptable and had sensory attributes closest to full-fat Kashar cheese.

The California Mastitis Test has previously been adapted for use in an inline, cow-side sensor and relies on the fact that the viscosity of the gel formed during the test is proportional to the somatic cell concentration. In this paper, the use of capillary and rotational viscometry was compared in light of the expected rheology of the gel formed during the test. It was found that the gel is non-Newtonian, but the initial phase of viscosity increase was not due to shear dependence, but rather due to the gelation reaction. The maximum apparent viscosity of the gel was shear dependent while the time it took to reach the maximum was not truly shear dependent, but was rather dependent on the degree of mixing during gelation. This was confirmed by introducing a delay time prior to viscosity measurement, in both capillary and rotational viscometry. It was found that by mixing the reagent and infected milk, then delaying viscosity measurement for 30 s, shortened the time it took to reach maximum viscosity by more than 60 s. The maximum apparent viscosity, however, was unaffected. It was found that capillary viscometry worked well to correlate relative viscosity with somatic cell count, but that it was sensitive to the reagent concentration. It can therefore be deduced that the rheology of the gel is complicated not only by it being non-Newtonian, but also by the strong dependence on test conditions. These make designing a successful sensor much more challenging.

The principal chemical components of milk from the Arabian camel (Camelus dromedarius) were monitored in Jordan over one year. The analyses included total solids, fat, protein, vitamins, minerals and organic acids. Large seasonal variations in total solids and fat were apparent, with maxima in mid-winter of 139 and 39·0 g/l, respectively, and minima in August of 102 and 25·0 g/l. These differences may be sufficient to alter the sensory properties of the milk, and the fat: casein ratio may need standardisation for cheesemaking. The mean values of trace elements like zinc (5·8 mg/l), iron (4·4 mg/l) and manganese (0·05 mg/l) in Jordanian camel milk could provide valuable additions to the diet of urban populations, as could the mean concentration of vitamin C (33 mg/l). The levels of organic acids were generally higher than in bovine milk and, as with all the constituents of the milk, there were discernible patterns linking concentration and season of the year.

The concentration of lactoferrin (LTF) in milk varies during lactation, rising sharply during involution. We proposed that LTF might have a regulatory role in involution and investigated its effects in vitro on the viability of bovine mammary epithelial cells (BMEC) and on casein expression in bovine mammospheres. Mammospheres capable of milk protein expression were formed by culturing primary BMEC on extracellular matrix in the presence of lactogenic hormones. Exogenously added LTF decreased β-casein and κ-casein mRNA expression in mammospheres while transfection of a short interfering RNA (siRNA) to suppress LTF expression resulted in increased casein mRNA expression. We believe that LTF exerts its effect on casein gene expression by up-regulating interleukin-1β (IL-1β) as IL-Iβ gene expression was elevated in mammospheres treated with LTF. LTF also decreased viability of BMEC grown as monolayers and as mammospheres. Interestingly, LTF was only effective in reducing casein mRNA expression and viability in mammospheres when added at concentrations found during early involution but was inactive when used at concentrations found in milk. We suggest that LTF has a regulatory role during early involution, decreasing casein expression and reducing BMEC viability.

The objective of the present study was to determine the effect of high pressure (HP) processing (200, 450 and 650 MPa) at various temperatures (20, 40 and 55°C) on the total plasmin plus plasminogen-derived activity (PL), plasminogen activator(s) (PA) and cathepsin D activities and on denaturation of major whey proteins in bovine milk. Data indicated that transfer of both PL and PA from the casein micelles to milk serum occurred at all pressures utilized at room temperature (20°C). In addition to the transfer of PL and PA from micelles, there were reductions in activities of PL (16–18%) and PA (38–62%) for the pressures 450 and 650 MPa, at room temperature. There were synergistic negative effects between pressure and temperature on residual PL activity at 450 and 650 MPa and on residual PA activity only at 450 MPa. Cathepsin D activity in the acid whey from HP-treated milk was in general baroresistant at room temperature. The residual activity of cathepsin D decreased significantly at 650 MPa and 40°C and at the pressures 450 and 650 MPa at 55°C. Synergistic negative effects on the amount of native β-lactoglobulin were observed at 450 and 650 MPa and on the amount of native α-lactalbumin at 650 MPa. There were significant correlations between enzymatic activities (PL, PA and cathepsin D) and the residual native β-lactoglobulin and α-lactalbumin in bovine milk. In conclusion, HP significantly affected the activity of indigenous proteolytic enzymes and whey protein denaturation in bovine milk. Reduction in activity of indigenous enzymes (PL, PA and cathepsin D) and transfer of PL and PA from the casein to milk serum induced by HP is expected to have a profound effect on cheese yield, proteolysis during cheese ripening and quality of UHT milk during storage.

Yoghurt and starter culture producers are still searching strains of Lactobacillus delbrueckii subsp. bulgaricus and Streptococcus thermophilus to produce healthier yogurt with longer shelf life, better texture, taste and quality. However, selective identification of Lb. delbrueckii subsp. bulgaricus and Strep. thermophilus from a mixed population using microbiological and biochemical methods is difficult, time consuming and may not be accurate. In this study, a quick, sensitive and accurate method is proposed to identify both Lb. delbrueckii subsp. bulgaricus and Strep. thermophilus using PCR. The method is comprised of two parts. In the first part, methionine biosynthesis genes, known to be present in both species were partially amplified by designed primers (cysmet2F and cysmet2R). Partial amplification of the methionine biosynthesis gene which gives 700 bp fragment resulted in selective identification of Lb. bulgaricus and Strep. thermophilus. All 16 Lb. bulgaricus and 6 Strep. thermophilus isolates assessed by this method gave the expected amplification. On the other hand, further analysis of other closely related species with the same primers have indicated that the same product was also amplified in two more lactobacilli namely, Lb. delbrueckii subsp. lactis and Lb. helveticus species. Thus, in the second part of the method, further differentiation of Lb. delbrueckii subsp. bulgaricus and Strep. thermophilus from each other and these species was achieved using restriction analysis of 16S rRNA gene with EcoRI.

The dynamics of non-esterified fatty acid (NEFA) metabolism in lactating dairy cows requires quantification if we are to understand how dietary treatments and disease influence changes in body condition (adipose reserves) and the production of milk fat. We present here a novel compartmental model that employs the pattern of plasma glucose concentrations to predict the dynamic changes that occur in plasma NEFA concentrations during an intravenous glucose tolerance test (IVGTT) in lactating dairy cows. The model was developed using data obtained from ten early-lactation, Holstein-Friesian cows given a standard IVGTT. The model described all of the major features of the NEFA response to an IVGTT; it was consistent with physiological processes and provided a number of parameters that can be used to quantify NEFA production and utilization. For all of the individual cows, all model parameters were well identified and usually had CV<10% of their estimated values. In the model, elevated plasma glucose concentrations cause an increase in the level of glucose in a remote compartment, which in turn suppresses the rate of entry of NEFA to the plasma compartment. The means (±sd) for the five adjustable parameters of the model were: rate of entry of NEFA to the plasma pool (SFFA) 183±71 [μmol l−1 min−1], rate of removal (oxidation, sequestration in adipose tissue and uptake by the mammary gland for milk production) of NEFA from the plasma pool (KFFA) 0·140±0·047 [min−1], a threshold parameter (gs) representing a plasma glucose concentration above which elevated levels of plasma glucose result in entry of glucose into a ‘remote’ or inaccessible glucose compartment, 3·30±0·52 [mmol/l], a rate constant (K) describing the movement of plasma glucose (above gs) into a remote compartment 0·063±0·033 [min−1] and a parameter Φ which is a Michaelis Menten type affinity constant which modulates the extent to which remote glucose inhibits the provision of NEFA to the plasma pool, 0·812±0·276 [mmol/l]. It is concluded that the model is suitable to describe NEFA kinetics in lactating dairy cows and it may have application in other species.

Olive tree leaves (OTL) and grape marc (GM) are by-products with high linolenic (LNA) and linoleic (LA) acid content, respectively, which can be used as dietary ingredients to increase the cis-9 trans-11 conjugated linoleic acid (CLA) content of milk fat in sheep and goats. An experiment was conducted with 16 Friesian ewes and 16 Alpine goats to study the effect of OTL and GM inclusion in sheep and goat diets on their milk fatty acid profile, with emphasis on cis-9, trans-11 CLA and vaccenic acid (VA). Ewes and the goats were fed the control (C) diets from parturition to the 90 days in milk (DIM) and then both groups were divided into two sub-groups (treatments). The control groups of both species continued to be fed the C diets, whereas the treated groups were gradually switched over a 2-week period (DIM=91–105) from the C diets to that of treatment 1, which contained air-dried OTL. These OTL diets were fed ad libitum for 1 month (DIM=106–135). After that period, the same treated groups, after 2 weeks of gradual adaptation (DIM=136–150), were switched to treatment 2, which contained air-dried GM. The GM diets were fed ad libitum for 1 month (151–180 DIM). Concentrations of polyunsaturated fatty acids (PUFA) and mono-unsaturated fatty acids increased significantly in milk fat of sheep fed OTL v. C. For goats, only the PUFA in milk fat was increased by feeding OTL compared with C. Relative to C, GM increased significantly the concentration of PUFA only in milk fat of sheep. OTL and GM diets increased the cis-9, trans-11 CLA and VA content in milk fat, compared with C, only in sheep. GM caused a sharp increase in 18:0 only in sheep milk fat, while the OTL diet increased significantly the 18:0 in milk fat of goats. GM and OTL diets also had opposite effects on the 18:1/18:0 ratio of sheep milk fat. In conclusion, OTL and GM, when included in sheep diets altered the milk fatty acid profile with a pronounced increase in cis-9, trans-11 CLA and VA contents. The results show that the response of sheep and goats to OTL and GM diets was different, suggesting a species difference that needs further investigation.

The present study tested the hypothesis that supplemental dietary fatty acids (FA) affect the energy corrected milk yield in proportion to the milk production level of dairy cows, and increase both long chain FA proportion of milk FA and milk fat globule diameter. Sixteen Danish Holstein cows were divided into four 4×4 Latin squares with two squares of medium yielding cows (32·2 kg energy corrected milk (ECM)/d; 158 days in milk (DIM)) and two squares of high yielding cows (40·0 kg ECM/d; 74 DIM). Experimental length was 12 weeks, with three weeks for each of the four periods. The four treatments were no supplementation (17 g FA/kg dry matter (DM)) and three diets with supplemented FA (29, 40, and 52 g total FA/kg DM, respectively) obtained by substituting barley with Palm Fatty Acid Distillate (PFAD) fat. Diets were offered as total mixed rations with 63% grass/clover silage (DM basis). Dry matter intake decreased with increasing FA supplementation, but net energy intake was not affected. The general linear responses to 10 g/kg DM increase in FA level were 1·1 kg ECM (P<0·0001), 0·061 kg milk fat (P<0·0001), 0·012 kg milk protein (P=0·09) and 0·052 kg lactose (P=0·0002) per day, and linear responses in milk composition were 0·39 g fat (P=0·07), −0·71 g protein (P<0·0001) and 0·05 g lactose (P=0·3) per kg milk, and 0·092 μm (P<0·0001) in milk fat average globule diameter. Fatty acid supplementation decreased short- and medium-chain FA and C16:0 and increased C18:1 proportions of total FA in milk. Supplemental dietary FA increased ECM yield but not in proportion to production level as anticipated, and increased average FA chain length and milk fat globule diameter.

The aim of this research was to enhance the nutritional quality of ewe milk fat by increasing potentially healthy fatty acids (FA) through diet supplementation with unprotected oil rich in linoleic acid, and without detrimental effects on animal performance. Twenty-four ewes were assigned to two high concentrate diets, control or supplemented with 6% sunflower oil (SO), for 4 weeks. No differences between treatments were found in milk production and dry matter intake. Although the SO diet increased milk fat percentage and tended to reduce milk protein concentration, it did not affect milk fat, protein or total solid yield. Most of the modifications in milk FA composition were addressed toward a potentially healthier profile: a decrease in C12:0 to C16:0 and a remarkable increase in the contents of cis-9 trans-11 C18:2 (from 0·94 to 3·60 g/100 g total FA) and trans-11 C18:1 (from 2·23 to 8·61 g/100 g total FA). Furthermore, the levels reached were maintained throughout the period monitored. However, the SO diet increased other trans C18:1 isomer percentages, too. The lack of differences between treatments in the in vitro ruminal fermentation parameters, studied with batch cultures of rumen microorganisms, would indicate no negative effects on ruminal fermentation.

Ca-Fe interactions are known, but no studies are available about the effects of Ca-enriched goat or cow milk on Fe status in nutritional ferropenic anaemia (NFA). To examine this matter, control and Fe-deficient rats were fed for 14 d with goat or cow milk diets containing either normal or high Ca content (5000 or 10 000 mg/kg diet), and different indices and parameters related to iron status were measured. The apparent digestibility coefficient (ADC) and the Fe retention/intake (R/I) ratio were higher in control and anaemic rats fed goat milk diet (G diet), despite high-Ca content. Ca enrichment decreased Fe stores in liver and sternum in anaemic rats fed cow milk diet (C diet), however G diet did not modify Fe content in the organs studied in control and anaemic rats. In anaemic rats, Ca-supplementation decreased haematocrit, but platelets and serum Fe were not affected, however, in control rats platelets increased except for Ca-enriched G diet, this fact reveals that Ca-Fe interaction is minimized with G diet. Serum ferritin was always higher in rats fed G vs. C diet, both in control and anaemic rats fed either normal or Ca-enriched diets. Ca-supplementation decreased ferritin levels in control and anaemic rats fed C diet and also, though to a lesser extent, in those given the G diet. This indicates that with this G diet there is a better recovery of body Fe stores in anaemic rats, despite Ca-supplementation. In this study it is noteworthy that despite high Ca content, a goat milk diet resulted in minimal Ca-Fe interactions and did not adversely affect Fe status in rats with NFA.

ATP-binding cassette (ABC) transporters play a pivotal role in human physiology, and mutations in these genes often result in severe hereditary diseases. ABC transporters are expressed in the bovine mammary gland but their physiological role in this organ remains elusive. Based on findings in the context of human disorders we speculated that candidate ABC transporters are implicated in lipid and cholesterol transport in the mammary gland. Therefore we investigated the expression pattern of selected genes that are associated with sterol transport in lactating and nonlactating mammary glands of dairy cows. mRNA levels from mammary gland biopsies taken during lactation and in the first and second week of the dry period were analysed using quantitative PCR. Five ABC transporter genes, namely ABCA1, ABCA7, ABCG1, ABCG2 and ABCG5, their regulating genes LXRα, PPARγ, SREBP1 and the milk proteins lactoferrin and α-lactalbumin were assessed. A significantly enhanced expression in the dry period was observed for ABCA1 while a significant decrease of expression in this period was detected for ABCA7, ABCG2, SREBP1 and α-lactalbumin. ABCG1, ABCG5, LXRα, PPARγ and lactoferrin expression was not altered between lactation and dry period. These results indicate that candidate ABC transporters involved in lipid and cholesterol transport show differential mRNA expression between lactation and the dry period. This may be due to physiological changes in the mammary gland such as immigration of macrophages or the accumulation of fat due to the loss of liquid in the involuting mammary gland. The current mRNA expression analysis of transporters in the mammary gland is the prerequisite for elucidating novel molecular mechanisms underlying cholesterol and lipid transfer into milk.

Somatic cell count (SCC) in milk is considered to be a valuable indicator of cow mastitis. For assessment of SCC in milk, the bioluminescent assay based on determination of ATP from somatic cells ([ATPsom]) in milk was proposed earlier. However, this assay is still not widely used in practice owing to lower reliability compared with conventional methods such as direct microscopy and flow cytometry. We revised the bioluminescent SCC assay and developed a simple protocol based on determination of the total non-bacterial ATP concentration in milk. It was shown that the novel ATP-releasing agent Neonol-10 (oxy-ethylated iso-nonyl phenol) has superior performance providing 100% lysis of somatic cells while not disrupting bacterial cells of milk at a concentration of 1·5% w/w. There was high correlation (R2=0·99) between measured bioluminescence and SCC as measured by direct microscopy. The observed detection limit of the bioluminescent milk SCC assay was as low as 900 cell/ml, time of analysis was 2–3 min per sample. The proposed method has high potential for on-site mastitis diagnostics.

This study was conducted to determine the effect that relocation to a new free stall barn had on locomotion and cleanliness of two breeds of dairy cows. The original facility before relocation had cows housed in an 8-row free stall barn. Cows were allocated in a new 4-row free stall facility: cows of two breeds (n=22 Holsteins and 22 Jerseys) were intermixed in the northwest section. Locomotion (scale 1–5) and cleanliness were scored (scale 1–4). Holsteins and Jerseys had no difference in locomotion score throughout 12 weeks following relocation. A lactation number by date interaction showed cows in third and greater lactations had significantly higher locomotion scores (more lameness) by day 86. Locomotion scores increased across breeds during the 86-d observation period, suggesting cows became lamer. Jerseys had cleaner lower legs than Holsteins (2·9±0·1 v. 3·5±0·1, respectively). Lactation number affected lower leg cleanliness, with scores decreasing as lactation number increased (3·4 v. 3·3 v. 2·9±0·10 for first, second and third and greater lactations, respectively; P<0·01). All cows were cleaner (lower scores) after relocation, suggesting that the new facility improved hygiene.

Human milk samples from three healthy donors were investigated in order to evaluate the antibacterial activity during lactation against Escherichia coli ATCC 25922 and Listeria monocytogenes. The concentration of the main human-milk antimicrobial proteins (lactoferrin (LF), lysozyme (LZ) and secretory immunoglobulin A (sIgA)) was determined by ELISA. Results showed that human milk exhibited antibacterial activity against List. monocytogenes, although it was weakly active against Esch. coli ATCC 25922. The observed antilisterial activity was positively correlated with LZ concentration. In addition, the effect of gastrointestinal proteases, at different pH conditions, that prevail in the stomach of infants (pH 2·0–6·5), on antilisterial activity and protein degradation was evaluated. Hydrolysis with pepsin at pH 4·0–6·5, followed by treatment with pancreatic enzymes, resulted in a decreased hydrolysis of LZ, LF and sIgA and an enhanced antibacterial activity against List. monocytogenes. It is suggested that partial degradation of certain milk proteins at the gastrointestinal level may produce peptides that could act synergistically with the remnant intact proteins.

Two angiotensin converting enzyme (ACE)-inhibitory peptides were isolated from enzyme modified cheese (EMC) and their amino acid sequences were identified as Leu-Gln-Pro and Met-Ala-Pro. The EMC was prepared by a combination of Protease N, Umamizyme, and Flavourzyme 500L. Both peptides were derived from β-casein, f 88-90 and f 102-104, respectively. Met-Ala-Pro showed strong ACE inhibitory activity (IC50=0·8 μm) and antihypertensive activity in spontaneously hypertensive rats (SHR) after single oral administration. The IC50 value of Met-Ala-Pro was not affected by pre-incubation with ACE, suggesting that this peptide was a true ACE-inhibitory peptide. We report here, for the first time antihypertensive peptides from EMC.

When κ-casein (κ-CN) was added to milk, and the milk was subsequently pH adjusted (pH 6·5−6·9) and heated (90°C/15 min), the serum contained considerably higher levels of denatured whey proteins than the milks without added κ-CN. When milk at pH 6·5−6·9 was heated at 90°C for different times, κ-CN was found in the serum in the early stages of heating and before significant levels of whey proteins were denatured. κ-CN reached its maximum level in the serum before the whey proteins were fully denatured. When milk at pH 6·5−6·9 was heated at 20−90°C for 15 min, κ-CN dissociated from the casein micelles at all temperatures, with the level in the serum increasing with the temperature and the pH at heating. κ-CN dissociated from the micelles at temperatures below those at which significant levels of the whey proteins were denatured. When taken together, the results from these experiments strongly indicate that the dissociation of κ-CN from the micelles can precede the interaction of the denatured whey proteins with κ-CN, and that there is a preferential interaction of the denatured whey proteins with serum-phase κ-CN.

Sediment formation was investigated during UHT treatment of goats' milk, subjected to indirect treatment at 140°C for 2 s, with upstream homogenisation. Stabilisers evaluated were sodium hexametaphosphate (SHMP), trisodium citrate (TSC), disodium hydrogen orthophosphate (DSHP), and sodium dihydrogen orthophosphate (SDHP). With no added stabiliser, goats' milk produced a heavy sediment on UHT treatment. Addition of SDHP reduced pH, had little effect on ionic calcium and did not substantially reduce sediment. However, addition of SHMP, DSHP and TSC each reduced ionic calcium, increased ethanol stability and reduced sediment. Following stabiliser additions, there was a good correlation between ethanol stability and ionic calcium (R2=0·85) but not between ethanol stability and pH (R2=0·08). Overall, reducing ionic calcium reduced the amount of sediment formed for all these three stabilisers, although there was no single trend line between sediment formation and ionic calcium concentration. Sediment formation was not well correlated with pH for TSC or for SHMP, but it was for DSHP, making it the only stabiliser where sediment formation correlated well both with ionic calcium and pH, which might account for its effectiveness at higher ionic calcium levels. Sediment was much reduced when the temperature was reduced from 140°C to 125°C and 114°C. There were no further changes in sediment on storage for two weeks. Analysis of the sediment showed that it was predominantly fat and protein, with a mass ratio ranging between 1·43:1 and 1·67:1. Its mineral content was usually less than 5% of dry weight. The maximum amounts of P and Ca were found to be 2·32% and 1·63%, respectively.