Monitoring the Freezing Point of Buffalo Milk.

The aim of this study was to evaluate the basic freezing point of buffalo milk. Bulk milk samples were collected from buffalo and cattle farms in Caserta area from 2008 to 2014. The analysis involved a total of 1886 buffalo milk samples and 1711 bovine milk samples. These were also tested for fat, protein and lactose contents by means of infrared spectrometry. The freezing point was determined by means of a thermistor cryoscope. Data underwent statistical analysis. Our research showed an average freezing point of -0.528°C for buffalo milk and -0.522°C for bovine milk. Given the lack of data on the freezing point of buffalo milk, our study provides the first indication of a basic freezing point of the milk of this species in Italy.

Introduction

The freezing point of milk, which is an important indicator of its quality, is determined by its composition. As the freezing point of bovine milk is fairly constant, the freezing point of bulk tank milk is used to determine whether water has been added. The average freezing point of raw bovine milk is between -0.522 and -0.540°C (Fox and McSweeney, 1998). In bulk tank milk studies, a freezing point of -0.528°C has been found in Poland (Kuczaj, 2001) and Italy (Coni et al., 1997) and of -0.5209°C in the Netherlands (Slaghuis, 2001; Navrátilová et al., 2006). In Egypt, the freezing point of buffalo milk ranges from -0.552 to -0.558°C (Hofi et al., 1966) and the regulatory standard ranges from -0.530 to -0.560°C (Egyptian Organization for Standardization and Quality, 2005).

Until 2004, in compliance with EU legislation (Council 92/46 92/47/EEC Directive of 16 June 1992, incorporated into Italian law by DPR 54/1997; European Commission, 1992a, 1992b; Italian Republic, 1997), the maximum permitted freezing point of both raw and heat-treated drinking milk was ≤-0.520°C: above this level, it was supposed to add water to cow’s milk. In accordance with EC Regulation N°1234/07 (European Commission, 2007), milk produced for human consumption must have a freezing point close to the mean freezing point recorded in raw milk in the original area of production. However, EC regulation N°852 and 853/2004 (European Commission 2004a, 2004b) makes no mention of the freezing point, which remains an index of the quality of raw milk. Indeed, processors use the freezing point as a quality criterion for calculating the price of raw milk purchased and processed into dairy products.

Today, although the basic freezing point of cow’s milk is clearly known in Italy (Lombardi et al., 1998) and in other countries (Navrátilová et al., 2006), there are no data on the basic freezing point of buffalo milk. The aim of the present research was to determine the basic freezing point of buffalo milk by collating and evaluating the data from analyses carried out from 2008 to 2014.

Milk samples were collected from 60 buffalo farms. As the samples analysed were collected exclusively for the purpose of the project, they were taken only from farms that adopted particular precautions to avoid the presence of water residues in their tanks and which did not add water to their milk. Variability in the freezing point of buffalo and bovine milk from 2008 to 2014 was also evaluated.

Materials and Methods

The samples were collected from buffalo farms in the Caserta area from 2008 to 2014. Bulk milk samples were transported to the laboratory of the Institute for Experimental Veterinary Medicine of Southern Italy at 4°C. The analysis involved a total of 1886 buffalo milk samples collected from 60 farms and 1711 bovine milk samples collected from 40 farms. All samples were analysed for pH and fat, protein, lactose and lactic acid contents; pH was measured at 20°C, usually 1 h after the samples had arrived at the laboratory.

The chemical composition of the milk samples – milk fat (% w/v), crude protein (% w/v) and lactose anhydride (% w/v) – was determined by means of the infrared spectrometric method (IDF, 2013), using a Milkoscan FTplus instrument (Foss Electric, Hillerød, Denmark). In accordance with Italian law (Ministerial Decree 26/03/92; Italian Republic, 1992), the freezing point of milk was measured by means of a thermistor cryoscope: Advanced Cryoscope (Advanced Instruments, Norwood, MA, USA). These instruments are commonly used for parametric measurements in dairy analyses (Sánchez et al., 2007) and the whole process is carried out under standard quality control.

Statistical analysis was performed by means of IBM SPSS® 21 (Statistical Package for Social Science, version 21 for Windows; IBM Corporation, New York, NY, USA).

Results and Discussion

The freezing point was determined only in samples that displayed the composition shown in Table 1. The basic freezing point of buffalo milk was lower than that of cow’s milk; statistical analysis of all data in the years 2008-2014 showed an average value of -0.528°C for buffalo milk and an average value of -0.522°C for cow’s milk. Figure 1 shows box-and-whisker plots for buffalo milk samples collected over the 5 years. The central line indicates the median level and the boxes represent the first and third quartiles, while the whiskers are set to ±1.5 times the inter-quartile interval. These levels should match the minimum and the maximum levels in the absence of outliers and/or extreme levels. From 2008 to 2014, the percentage of samples with a value in excess of the gold point of -0.520°C declined from 35 to 6% for buffalo milk and from 35 to 5.4% for bovine milk). Focusing on last year 2014, 125 buffalo milk samples collected from 60 farms displayed an average freezing point -0.536 ±0.0091 (standard deviation), with a 95% confidence interval from -0.534 to -0.537; the average of the samples analysed was significantly different (α=5%) from m =-0.520 (Table 2).

Table 1.

Composition of buffalo milk tested (%).

	Fat	Proteins	Lactose	Lactic acid
Mean value	8.30	4.70	4.80	<0.18
Range	8.00-8.60	4.50-4.90	4.60-5.00	-

Figure 1.

Box-and-whisker plots of median freezing point values (95% CI) for buffalo milk in different years with evidence of outliers (O=1.5 times the inter-quartile range) and extreme values (*=3 times the inter-quartile range).

Table 2.

Descriptive statistics and T_Test.

	Descriptive statistics	T_Test µ=-0.520 (°C)
	No.	Mean	SD	Df	Sig. (2-sided)	Mean difference
Freezing point	125	-0.536	.009	124	.000	-.016

SD, standard deviation; df, degree of freedom.

Conclusions

In our study, the basic freezing point of buffalo milk proved to be lower than that of bovine milk; this could be explained by the fact that buffalo milk has a higher content of non-fat solids, fat, protein and some mineral salts (Zicarelli, 2001; Masud et al., 1992).

Our data analysis showed that the percentage of both bovine and buffalo samples with a freezing point value higher than -0.520°C decreased over the years considered. This suggests that the percentage of frauds (adding water to bulk milk tanks to increase the volume) decreased. However, other factors also affect the freezing point of both buffalo and cow milk, such as the breed of the animals, stage of lactation, season, feed, water intake, climate, mastitis, and storage time and temperature. Furthermore, differences between morning and evening milkings have also been reported (Slaghuis, 2001).

When milking machines are cleaned, a small quantity of rinsing water may be retained. Thus, some water will be present in milk samples if the milking machine and/or storage tank have not been properly dried after cleaning and sterilisation. It could therefore be hypothesised that in the period considered the reduction in the percentage of samples with a freezing point value above -0.520°C may have been due to a more careful drying on the part of the producers.

As about 80% of buffaloes in Italy are concentrated in Campania region (Tufarelli et al., 2008; Zicarelli, 2001), the present study yields representative data and provides the first indication of the basic freezing of point buffalo milk.