ABSTRACT: Ensiling corn for longer periods is a strategy used to increase rumen bacteria access to starch. In fact, when corn is ensiled for insufficient periods, starch digestibility decreases, as evidenced by excreted starch. This study investigates the effects of corn silage ensiling time on starch digestibility of dairy cows through fecal starch analysis. The trial was conducted during the spring of 2013 and the fall of 2014 on twenty dairy herds located in the state of Paraná, southern Brazil. Total mixed ration (TMR), whole-plant corn silage (WPCS) and fecal samples were collected for determination of dry matter (DM), neutral detergent fiber (NDF), acid detergent fiber (ADF), lignin and starch. Apparent total tract starch digestibility (ATTSD) was calculated using equations developed by Fredin et al. (2014) and Bal et al. (1997). Data were analyzed using the CORR, REG and GLM procedures of SAS. Spring and fall WPCS were ensiled for 260 and 132 days, respectively (p < 0.01). There was no difference (p > 0.05) in ATTSD-Fredin from season to season though there was a trend (p = 0.11) towards greater ATTSD-Bal for spring samples. ATTSD-Bal was positively correlated with ensiling days (r = 0.31). Starch digestibility was not negatively affected by WPCS harvesting maturity. Differences in post-ruminal starch digestion, variation between farms in DM intake and diets, limitations of the equations, influence of uncontrolled factors, and the small number of experimental units might have contributed to the absence of significant results. Overall, Bal et al. (1997) equation was more efficient in distinguishing ATTSD from different ensiling periods.

ABSTRACT The objective of this study was to assess the accuracy of the content of total mixed rations (TMR) offered to high-producing cows from 20 dairy herds of Castro, Paraná State, Brazil. The average milk yield during the sample collection period was 38.0±6.8 kg/day, with 3.47±0.25% milk fat and 3.05±0.18% milk protein. The particle size distribution of samples of fresh diet and leftovers was evaluated using the Penn State Particle Separator (PSPS) and chemical analysis was also conducted. The TMR homogeneity was estimated by the coefficient of variation (CV) of particles retained on the two intermediary sieves (<19 and >1.18 mm), in which a CV higher than 5% indicated a poorly mixed TMR. The TMR offered exceeded the formulated TMR only for NDF (+7.3%) and was lower than the formulated TMR for CP (−6.5%), ADF (−10.1%), and ash (−18.6%) contents. These differences are due to daily variations in feed quality, mixing equipment, and ingredient mixing order, mostly for forage. Differences between leftovers and TMR offered were substantial: −9.4% for CP, +25.1% for NDF, +31.6% for ADF, and +13.1% for ash, suggesting a feed sorting effect against long forage particles and in favor of small concentrate particles. The proportion of the offered TMR retained in the top PSPS sieve showed positive correlations with NDF (r = 0.58) and ADF (r = 0.54) contents of the refusals, which indicates that cows sort against long particles. Positive correlations were also found between the CV in the second sieve and NDF (r = 0.56) and ADF (r = 0.47) contents of refusals, suggesting that poorly mixed diets are more easily sorted by cows. Therefore, evaluating the chemical and physical properties of TMR, especially for forage, is extremely important for providing a consistent diet.