| dc.description.abstract | This study evaluated the lactation curve shape, the yield and composition of milk from cows
of Holstein and Jersey breeds grazing annual ryegrass (Lolium multiflorum) in northwestern
Rio Grande do Sul, and aimed to analyze its response when subjected to grazing pasture
herd consisting of two genotypes. The experiment was conducted from 08.18.2010 to
10.29.2010 and are used in total, eleven tester cows of each racial group and a variable
number of regulator cows. The animals were managed in rotational grazing with variable
stocking rate and were supplemented at a rate of 1 kg per 3 liters of milk produced above 15
liters / cow / day. Pasture management allowed similar grazing conditions between the two
tested breeds and proved to be not limiting to pasture and animals production. Individual
milking performance was measured once a week throughout experimental period. The visual
body condition score (BCS, 1-5), was also registered weekly, and body weigh was measured
monthly. The experimental design for pasture variables and for individual animal production
was a randomized block design with replications in time, having been used stepwise proc
(SAS, 2009) to test regression models. Only the supplement in the months August,
September and October in analysis, besides individual productions corrected to 4% fat, day
consuption in the months of August, September and October was considered in analysis,
besides individual animal productions corrected to 4% fat, days in lactation period , body
weights, BCS, number of lactations and age. Individual samples were taken by milking in
September and October, and analyzed for the following milk components: fat, protein,
lactose, water, acidity, freezing point, density, and computed total dry extract (TDE) and
defatted dry extract (DDE). The statistical analysis for milk components was randomized
complete blocks with triple factorial arrangement for breeds (2), periods (2), and stages of
lactation (4). ANOVA was performed by GLM (SAS, 2009) and multivariate analysis, through
Software Multiv (Pillar, 1997), using the correlation between variables as a measure of
similarity, and clustering criteria was the least variance of sum of squares Jersey exceeded
Holstein (P <0.005) for fat (%) lactose (%) and protein (%); Holstein was exceeding Jersey (P
<0.0001) for fat (kg), protein (kg) and TDE ( kg). Correlation studies showed a clear
difference in the degree of association between genotypes for dry matter intake of feed, with
lower association for Jersey genotype, Holtstein presented a higher and consistent
dependence of intake throughout the experimental period, coinciding with pasture cycle. This
trend may indicate greater efficiency of Jersey genotype in the use of pastures of high
quality, obtained in the initial pasture phases. The regression studies indicated that
supplement dry matter intake (partial R ²) in September explained 65.69% of milk production
for Holstein breed, and supplement dry matter intake (partial R ²) in October explained
55.73% of milk production of Jersey breed. | |
