SOMATIC-CELL COUNTS IN GOAT MILK AND THEI R RELATION TO MILK-COMPOSITION AND PROPERTIES

Milk composition and properties were studied in 20 goats of the White Polled Short-wooled breed in the first lactation from the 6th to 255th day in nine time intervals. Goat milk samples were examined for somat ic cell counts (Fossomatic and mastitis N-test) and the following prop erties were investigated at the same time: basic milk components (fat, crude protein, lactose, ash), N-matters (true proteins, casein, whey proteins, nonprotein nitrogen, urea, ammonia), minerals (Ca, Mg, total P, inorg. P, Na, K, Cl), physical and chemical properties (titratable acidity, pH, buffering capacity, electrical conductivity) and renneta bility and alcoholic stability as processing properties. Correlation a nalysis was used to assess the relations of milk composition and prope rties to somatic cell (SC) counts. The results were grouped according to SC counts into classes of 250,000 SC per 1 ml milk, means and varia bility of all values concerned were calculated for every class and sig nificance of differences in the means between the classes was tested. SC differential counts were determined after staining microscopic prep arations by Giemsa method. The assessment involved 146 milk samples in total from animals in which the occurrence of acute mastitis was not demonstrated. SC counts varied in a wide range of 78,000 to 4,520,000 SC per 1 mi milk with an average value of 860,947 (Tab. I). A mastitis N-test used for rapid diagnosis of secretory disorders in dairy cows was verified at the same time. A statistically highly significant corr elation (r = 0.56(++)) was found out between SC counts and N-test valu es (Tab. II). Samples with low N-test values exhibited high SC counts and vice versa. No relationship was demonstrated between the lactation pattern and SC counts. Increased SC counts were occasionally observed in the 3rd to 4th month of lactation, which could be partly due to tr ansition to machine milking. Tab. III shows the values of correlation coefficients for relations between SC counts and the characteristics u nder study. The classes from which the average values against the prec eding class are significantly different can be seen for the milk compo nents and properties in Tab. IV. A decrease in potassium content (r = -0.17(+)) and in a K/Na ratio (r = -0.15(+)) was recorded for SC count s higher than 0.5. 10(6) per 1 ml goat milk. With SC counts higher tha n 0.75.10(6) lactose content is decreasing (r = -0.31(++)), the ratio chlorides/lactose - chlorine-sugar number is increasing (r = 0.16(+)), whey protein content is rising (r = 0.17(+)) and there are significan t variations in casein, sodium and chloride contents. With SC counts h igher than 1.10(6) there is a statistically significant decrease in go at milk titratable acidity (r = -0.13(+)) and alcohol number and signi ficant variations in milk pH value and buffering capacity. SC counts e xceeding the limit 1.25.10(6) result in significant variations of aver age values in a majority of other N-matters under observation, in an i ncrease in ash content (r = 0.30(++)), in a decrease in inorganic phos phorus content (r = -0.42(++)) and its percentage in total phosphorus content (r = -0.38(++)), in prolonged time of milk rennet clotting (r = 0.42(++)) and in variations of specific weight. Total phosphorus con tent in milk is significantly decreasing (I = -0.15(+)) with SC counts higher than 1.5.10(6) per 1 ml milk. Average values in the establishe d classes are significantly different from the preceding classes for f at, ammonia nitrogen, milk conductivity (r = -0.13(+)), calcium (r = - 0. 21(+)) and magnesium only when SC counts are higher than 1.75 and/o r 2.0.10(6) per 1 ml milk. To determine the percent representation of the SC groups in goat milk. differential SC counts were informatively determined in 16 milk samples at various stages of lactation and with SC counts ranging from 123,000 to 4.52.10(6) per 1 ml, out of which 10 samples bad SC counts maximally up to 1.10(6) per 1 ml milk. Polymorp honuclears showed the highest percent representation from 40 to 70% wi th peak in the 3rd to the 4th month of lactation. Macrophage counts ra nged between 10 and 20% without any apparent relation to lactation sta ge. Lymphocyte counts were highest at the beginning of the period of o bservation (about 20%) and they were slightly decreasing to 5 to 7% un til the end of lactation. The representation of epithelial cells range d from about 10 to 15% at the beginning of lactation, showing a clear drop until the 2nd month of lactation, followed by a moderate increase to 15% since then until the end of the period of observation. It was not possible to assess the relationship between the representation of the SC groups and total SC counts due to a small number of randomly se lected samples. Nevertheless, some statistically highly significant re lations between somatic cell counts and milk composition could be dete rmined. The results document that with SC counts higher than 1.10(6) p er 1 ml milk significant variations in values of many basic components and properties of milk can occur in individual goat milk samples, the re are variations in the averages in the particular classes or the var iability of the average is increasing. Therefore we believe that hygie nic requirements for goat milk to be processed should comprise the est ablishment of a limit value of SC counts amounting to 1.10(6) per 1 ml for milk from individual animals and to 750,000 SC per 1 ml for bulk goat milk.