Effect of quantity, quality, and length of alfalfa hay on selective consumption by dairy cows.

Twenty-four lactating Holstein cows were used in a replicated 6 x 6 Latin square design. Experimental periods were 6 or 7 d. Cows were housed in tie-stalls, and diets were fed ad libitum twice daily at 1100 and 1600 h. Diets contained 60% concentrate and either 40% alfalfa hay or 20% alfalfa hay and 20% alfalfa silage (dry matter basis). The effect of quantity, quality, and length of hay on sorting behavior was determined. Treatments consisted of 20% lower or higher quality long alfalfa hay, 20% lower or higher quality chopped alfalfa hay, and 40% lower or higher quality chopped alfalfa hay. Variation of sorting among cows was also determined. Particle size distribution of samples of as-fed total mixed rations and orts were determined using the Wisconsin particle size separator. Screens have square holes with diagonals of 26.9, 18, 8.98, 5.61, and 1.65 mm (screens Y1 to Y5, respectively). Sorting was calculated as the actual intake of each fraction expressed as a percentage ofthe predicted intake. Increasing the proportion of dry hay increased sorting. Quality of alfalfa hays that were offered did not affect sorting activity. Feeding long alfalfa hay increased selective consumption of fine particles. However, feeding long alfalfa hay also increased intake of longer particles because a higher percentage of long particles was offered. Across treatments, animals consistently sorted against longer particles in favor of finer particles. In particular, intake of Y1 as a percentage of the predicted intake was the most variable. Average Y1 intake, across the six treatments for each cow, was between 60 and 70% of predicted intake for four cows, 71 to 80% for 11 cows, 81 to 90% for five cows, 91 to 100% for two cows, and 101 to 110% for two cows. On one diet a cow failed to consume any of the Y1 portion of the total mixed ration. This variation among animals in sorting of very long feed particles may have practical significance.