| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 División Producción Animal. E. P. S. O. Universidad Miguel Hernández Ctra. Beniel, km 3,2 - 03312 Orihuela-Alicante, Spain
2 Department of Animal Science Universitat Politècnica Camí de Vera, 14 46071 València, Spain
Corresponding author: N. Fernandez; e-mail: nfernandez{at}dca.upv.es.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: machine milking • pipeline height • free fatty acid • ewe
Abbreviation key: HL = high level, LL = low level, ML = midlevel, MM = machine milk, MSM = machine stripping milk
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Milkline height selection criteria in a milking machine usually encompass certain aspects relating to cost of the installation, organization, and efficiency of milking, as well as possible effects on udder health and milk quality. In this sense, the advantages that may be achieved by installation of HL or ML compared with LL arise from the reduced installation costs (25 to 35%, depending on manufacturer) and the greater efficiency in milking for the same number of operators (Johnson et al., 1977; Le Du, 1983; Fernández et al., 1991). On the other hand, an LL system allows milking to occur with a lower nominal vacuum level without impacting the rate of teat cup slippage (Fernández et al., 1997). The LL milking system may also maintain a more stable vacuum at the level of the udder (Notsuki et al., 1979; Osteras and Lund, 1980; Le Du, 1983, 1985; Klos and Woyke, 1985; Lobotka et al., 1992). It is also reported that fluctuations in vacuum, through different associated mechanisms (inverse flow, impact, and reverse pressure gradients), can be a predisposing factor to mastitis (Mein, 1992; Billon et al., 1998). In fact, although no differences have been detected between both milking systems in sheep (Le Du, 1983), ML in dairy cows has occasionally been associated with a higher milk SCC (Notsuki et al., 1979; Czediwoda, 1991).
Others have reported that an LL system used with dairy cows may also lower lipolysis rates in milk (Gilson and Cousins, 1985; Heuchel and Chilliard, 1988; Meffe, 1994). In small ruminants, this aspect has only been studied in caprine (Morand-Fehr et al., 1983), with no significant differences reported between both milking systems (ML vs. LL).
Given the scarcity of information available in ovine, it was decided to carry out an experiment to study the effect of milk pipeline height on milk yield, milking efficiency, milk composition and the degree of lipolysis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
).
|
Milking Material and Equipment
Ewes were milked in a milking parlor (2 x 12), where 2 milk pipelines were installed: ML and LL (Figure 1). Both milklines terminated in 2 different receivers, and had 6 milking units (ML) or 12 milking units (LL), but always with one milker. The ML milkline, dead end type, was 52 mm in diameter and 518 cm in length and located at 214 cm height above the milker stands (112 cm above ewe standing level). The LL milkline, looped type, was 40 mm in diameter, 1110 cm in length (435 cm length from final connection milk inlet to receiver on each side) and was situated at 62 cm height above the trough floor (40 cm below ewe standing level). Both milklines were stainless steel, with 0.2% slope toward the receiver and a 61-cm distance between connection milk inlet. The milk pump was similar in both milking systems (PL37, De Laval Agri, Tumba, Sweden). The milk delivery line to the refrigeration tank was made of transparent flexible PVC with an interior diameter of 20 mm and length approximately 8 (LL) and 7 m (ML).
|
In both lines, 3 electromagnetic pulsators driven by an electronic control box supplied pulsation to the clusters (2 clusters per pulsator). In each milking unit a cylindrical glass recorder jar was installed with a floating level indicator and 1500-mL capacity. The milk inlets to the recorder jar were located at 102 cm above ewe standing level in ML, whereas in LL they were positioned at a similar height to the milk inlet in the milkline. The milking machine performance was monitored 3 times during the study, at the start of the preexperimental stage and each of the experimental periods. Pulsation (180 cycles/min), vacuum level (36 kPa), pulsation ratio (50%), and effective reserve (800 L/min) remained unchanged throughout the experiment. Vacuum level at short milk tube and pulsation were checked with an Alfatronic IV pulsation recorder (De Laval Agri) at the end of the second period. Mean flow rate was 720 mL/min at this time. Milking routine, with no prior udder preparation, included machine stripping by vigorous udder massage for 7 to 10 s just before the teatcups were removed. In addition, vacuum was always shut off at the claw before teatcup removal. In the experiment, all teats were dipped in iodine (0.5%) after milking.
Data Acquisition
During both the control period and 2 experimental periods, data were collected once a week at an afternoon (Tuesday) and morning (Wednesday) milking. At these times we recorded milk yield, milk fraction yields (machine milk [MM] and machine stripping milk [MSM]). We also collected milk samples for composition analysis (DM, fat, protein, and lactose). The SCC was measured in morning milk. The residual milk and production potential was determined immediately after morning milkings on Wednesdays. In addition, incidences of teatcup falloff were recorded along with notes to indicate the reason for the fall-off, i.e., active (caused by the animal) or passive (not associated with animal activity).
After each day of morning milking monitoring each ewe was immediately administered an intravenous injection (jugular vein) of oxytocin (3 IU) and the udder emptied by hand milking. The volume of milk obtained was recorded as a measure of residual milk. Four hours later this procedure was repeated to allow calculation of production potential, i.e., a measure of the daily milk production (milk yield in 4 h multiplied by 6).
Milk composition was determined automatically (MilkoScan FT120, A/S N Foss Electric, Hillerød, Denmark) from the mixture of the MM and MSM fractions obtained for each animal at each of the morning and afternoon milkings. Samples from afternoon milking were refrigerated at 4°C until analytical determination, which was always performed after morning milking. Milk SCC was determined with a Fossomatic 90 (A/S N Foss Electric, Hillerød, Denmark) between 24 and 48 h after collection (morning milking).
At the end of the second experimental period, the milk emission kinetics of the 2 groups of ewes in the experiment were determined, in accordance with the protocol proposed by Labussiere et al. (1984). All the ewes were milked with each milking system (ML and LL) for a period of 1 d, following a crossover design (2 x 2) for 2 d. Parameters determined were: appearance time of the first streams (T0), maximum first peak flow (MF1), MF1 appearance time (T1), first peak volume (V1), delay of second peak appearance (D), maximum second peak flow (MF2), MF2 appearance time (T2), second peak volume (V2), total milking time (T), total milk volume (TV), and average flow (MdF).
Free Fatty Acid Levels in Milk
To determine FFA content, samples were taken twice daily in the preexperimental period and in each of the 2 experimental periods. The methodology followed for sample collection was as follows: after the 2 milkings (morning and evening) of each of the 2 groups of ewes in the experiment, a sample was taken at receiver level (100 mL); the milk pump was then activated manually and another sample was taken of the milk pumped towards the refrigeration tank.
The samples were transferred to the laboratory and maintained at –18°C until analysis was carried out. Extraction of fatty matter was performed following the method of Needs et al. (1983), using HCl-acidified ethyl ether as solvent. Separation of FFA from triglycerides was accomplished by shaking with an anionic resin (Amberlyst A26, Sigma-Aldrich, St. Louis, MO), according to the methodology described by Gandemer et al. (1991). The resin was activated previously, following the procedure described by Needs et al. (1983). The final step before proceeding with FFA quantification was derivatization of the samples by formation of methyl esters with Hexane (Gandemer et al., 1991). Finally, chromatographic analysis was performed using a capillary column "GC 8160" gas chromatograph, equipped with an"ECA-80" electron capture detector and an "AJ 800" sampler (Fisons-Instruments, SA, Milan, Italy). Free fatty acids were expressed in miligrams per liter of milk.
Statistical Analysis
The effect of milk pipeline height on milk yield, milk fractioning, milk composition, and SCC was determined by the following statistical model:
 |
where
| yijkl | = | dependent variables,
| µ | = | general mean,
| Ei | = | random effect of the ewe,
| PERj | = | fixed effect of the experimental period,
| Rk(PERj) | = | fixed effect of the record nested within experimental period,
| SYSl | = | fixed effect of milk pipeline height (ML and LL), and
| eijkl | = | residual error.
|
Somatic cell count data were log10-transformed (Ali and Shook, 1980) before analysis as the SCC data were not normally distributed.
Study of the effect of milkline height on the FFA was carried out using the following model:
 |
the new factors being:
| GROl | = | group of animals samples were obtained from,
| RECm | = | milk receiving point (receiver and refrigeration tank), and
| MTn | = | milking time (morning and evening).
|
Other interactions such as "SYS*MT", "REC*MT", and "SYS*REC" were removed from the model after checking that they were not significant.
For analysis of milk emission kinetics parameters the ANOVA model used was:
 |
the new factor being:
 |
All of these statistical analyses were carried out with the Mixed Procedure from the SAS program (SAS, 1996).
Teatcup fall-off rate was analyzed by means of the chi-square test (
2), using the SAS program (SAS, 1996).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
).The total extracted milk values in each of the milking systems (ML: 1163 mL; LL: 1196 mL) were similar to those of the productive potential obtained by means of double injection of oxytocin (ML: 1113 mL; LL: 1165 mL). Strip yields of the 2 milking systems were equal, as were the amount and percentages of residual milk. Milking system had no significant effect on milk composition or mean milk SCC, which averaged 200,000 cells/mL per ewe.
|
, the differences in the FFA concentration were not significant between ML and LL (45.9 vs. 40.3 mg/L). Likewise, a similar ratio was observed (%) in the FFA groups (short, medium, and long chain) between both milking systems. In contrast, time of milking (morning vs. evening) significantly impacted (P < 0.001) greater level of FFA was observed in the afternoon milking than in the morning session (60.5 vs. 25.6 mg/L). This difference was doubled in the case of short- and long-chain FFA, and tripled in the medium-chain FFA. However, differences in FFA were similar for LL and ML milking systems, i.e., there was no significant interaction between milking system and milking time.
|
). The same trend was observed when analysis was performed in terms of milking system (ML: 47.0 vs. 42.5 mg/L FFA; LL: 45.2 vs. 36.7 mg/L FFA). At the same time, the increase in FFA concentration in the milk delivery line from receiver to refrigeration tank (mg/L of FFA in refrigeration tank minus mg/L of FFA in receiver) was greater in LL than in ML (8.5 vs. 4.5 mg/L, respectively).
|
shows the mean values of the milk emission kinetics parameter depending on the milking system assayed. Of the parameters studied, significant differences were only observed between ML and LL in those defining the first milk emission (T0, MF1, T1), and in the parameter defining the start of the second peak (D).
|
), and on the other, a greater fluctuation of the vacuum level was measured during milking (6.7 vs. 2.2 kPa, respectively).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The milking systems tested in this experiment (ML and LL) did not affect milk yield or milk fractioning during milking (MM and MSM), a result agreeing with that reported in other studies carried out in ovine (Le Du, 1983) and bovine (Notsuki et al., 1979). Moreover, the extracted milk yield corresponded with the productive potential of the ewes as measured by a double oxytocin injection scheme. This suggests that both milking systems were functionally sound. This is also reflected by the similarity of RM values for each of the milking systems. These results suggest that differences in milk yield parameters, e.g., MM, MSM, and RM are affected by factors other than the 2 milking systems we tested.
Although the IDF recommendations (Billon et al., 2002) on milking machines for small ruminants make no specifications for characteristics of the pulsation cycle phases that tend to diminish the risk of new IMI, in the case of dairy cows it does recommend (ISO, 1996b) that the "d" phase should have a minimum duration of 150 ms and take up at least 15% of the pulsation cycle. In this work, although at no time was the minimum duration set out by the regulation reached, due to the high pulsation rate (180 cycles/min) utilized, in all of the cases the minimum percentage of 15% was greatly exceeded (33 and 35.4%, for ML and LL, respectively). On the other hand, the shorter duration of phases "b" and "d" of the pulsation cycle in ML, caused by the increase in phases "a" and "c," may be the consequence of the greater length of the long pulse tube installed in this system, with the result that the establishment of vacuum and atmospheric pressure in the pulsation chamber was slower than in LL (Le Du, 1978).
The greater vacuum fluctuation observed in ML compared with LL in our study is similar to other studies using ovine (Le Du, 1985; Murgia and Pazzona, 1999) or bovine (Osteras and Lund, 1980; Klos and Woyke, 1985; Lobotka et al., 1992). This may be because in ML, when the milk bubbles up from the udder of the animal towards the milkline, it completely occupies, momentarily, the long milk tube section, interrupting the transmission of the vacuum between the milkline and the teatcups (Le Du, 1977; Murgia and Pazzona, 1999). Likewise, the minimum vacuum level reached lower values in ML than LL (28.6 vs. 32.8, respectively), which coincides with that described previously by Le Du (1983) and Murgia and Pazzona (1999). Le Du (1983) observed, moreover, that teatcup fall-off rate in ML was higher than in LL, as a result of the greater reduction in vacuum level during milking. In this work, although in ML the greater minimum vacuum level reduction and fluctuation produced during milking coincided, teatcup fall-off, mainly the passive instances, was not significantly higher than observed in LL. This may possibly be influenced by the low weight and features of the mouthpiece of the clusters utilized in this experiment.
In bovine, the reduction in vacuum level also involves an increase in milking time (Mein, 1992). In this work, average vacuum level was similar in the short milk tube for ML and LL (34.1 vs. 34.8 kPa, respectively), and so no significant differences were found between both systems in milking time (T) nor in mean milk flow (MdF). Experiments completed with lactating cows (Notsuki et al., 1979; Ichigawa and Nofu, 1983) and sheep (Le Du, 1983) concur with these results. The greater values found for parameters defining the first emission peak (T0, MF1, T1, and D) in ML, may be due to the fact that, whereas in LL the milk circulates towards the recorder jar as it is extracted from the udder, in ML it has to bubble up, occupying the whole of the long milkline, which means that sufficient milk must be stored in the claw in order to be drawn towards the recorder jar where it is measured. All of this indicates that the comparison of milk emission kinetics results between both types of line must be done with caution, since the type of line affects the shape of the emission curves.
The vacuum fluctuations occurring below the teat tip may be a factor that predisposes towards mastitis (Worstorff and Prediger, 1991; Mein, 1992; Billon et al., 1998). Given that the greatest vacuum fluctuations are generally observed in ML, it might be concluded that milking with this system implies a greater risk of IMI than in LL. However, in this work, in which a certain reduction was also observed in phase "d" in ML, no significant differences were found in the SCC between the 2 milking systems, a result that coincides with that reported previously by Le Du (1985). In bovine, however, the results are rather contradictory, and, whereas Notsuki et al. (1979) and more recently Czediwoda (1991), concluded that when HL is replaced by LL the SCC drops rapidly, Ichigawa and Nofu (1983), carrying out CMT tests with cows milked in HL and LL, found no significant differences.
The tested milking systems did not significantly affect the DM content, fat, protein, and lactose of the milk, as was observed previously by other authors in ovine (Le Du, 1983) and bovine (Ichigawa and Nofu, 1983). Likewise, no significant differences were found between the 2 systems in FFA concentrations present in the milk, a result that concurs with those noted in bovine by O’Brien et al. (1998), between ML and LL, and in caprine by Morand-Fehr et al. (1983), when comparing HL with LL. Nevertheless, most of the works carried out in bovine (Gilson and Cousins, 1985; Jellema, 1986; Heuchel and Chilliard, 1988; Meffe, 1994; O’Brien et al., 1998) agree that lipolysis is more favored in HL than in LL, as the milk is subjected to continuous stresses, rises, drops, and falls. In this work, some of the factors that may have influenced the fact that the FFA concentration was greater in HL than LL (greater long milk tube length, vertical sections of said tube, greater vacuum fluctuations, etc.) did not have any effect sufficiently important to render the differences significant. The total FFA concentration found in both systems was lower than that reported by Chavarri (1998) in milk from Latxa breed ewes milked in LL, who established an FFA range between 55.6 and 97.3 mg/L.
As there is no alteration of the technical aspects, factors such as the cost, installation size, and hourly parlor yield acquire added importance in the decision to install a low or high milkline.
The FFA concentration in the milk from afternoon milking reached double that of the morning session, a result that coincides with observations by several authors in bovine (Saito, 1983; O’Brien et al., 1998). In the afternoon milking, due to the different interval between milkings practiced in this work, a lower volume of milk (516 vs. 665 mL) and a higher percentage of fat (7.5 vs. 6.2%) was obtained than in the morning session. The higher FFA content may be a direct consequence of this greater concentration of fatty material and/or the greater susceptibility of the fatty globules to lipolysis due to the short space of time between the morning and afternoon milkings (Jellema, 1986).
The FFA concentration was also greater in the samples taken at refrigeration tank level than in those taken at the receiver. These differences arise because the milk, as well as having to pass through the milk pump and filter, has to traverse the whole delivery pipeline to the refrigeration tank. These results are similar to those described by Jellema (1986), Coussi (1988), and Meffe (1994), who cited an increase in lipolysis when pipeline section length was increased and/or in the presence of bends and vertical sections of the same. In addition, the lipolysis induced was greater in the milk delivery pipeline coming from the LL receiver, possibly because this presented 1 m more vertical length than in ML. Although lipolysis tended to be greater at receiver level in ML, the FFA concentration of the milk collected in the refrigeration tank, which is what is actually delivered to the industry, was similar for both systems.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication April 7, 2003. Accepted for publication July 24, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Elevage Bovin 76:75–77.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
