Comparison of Conventional Linted Cottonseed and Mechanically Delinted Cottonseed in Diets for Dairy Cows
Posted in Abstracts
Authors: V. R. Moreira,1 L. D. Satter,1,2 and B. Harding3
1U.S. Dairy Forage Research Center, Madison, USDA–Agricultural Research Service, and
2Department of Dairy Science, University of Wisconsin, Madison 53706
3Buckeye Technologies, Memphis, TN
Year Published: 2003
Abstract: Performance of lactating dairy cows fed diets containing either mechanically delinted whole cottonseed (DWCS; 3.7% lint) or linted whole cottonseed (LWCS; 11.7% lint) was measured. Forty primiparous (86 ± 39 d in milk) and 40 multiparous (88 ± 30 d in milk) cows were fed a total mixed ration containing 13% (dry matter basis) DWCS or LWCS in two blocks of 112 d (n = 53 and 27, respectively). Other total mixed ration ingredients (dry matter basis) were corn silage (28.1%), alfalfa silage (23%), high moisture shelled corn (27.8%), soybean meal (1.8%), expeller soybean meal (1.8%), blood meal (2%), and mineral-vitamin supplements (2.5%). Dry matter intake and milk yield were measured daily and milk composition every other week. Fecal grab samples were taken during wk 3 and 13 of each block to estimate excretion of intact whole cottonseeds. Milk yield, 3.5% fat-corrected milk, energy-corrected milk, milk composition and dry matter intake were not affected by whole cottonseed source. Body condition score tended to increase more with DWCS (0.22 vs. 0.11) for primiparous cows, although this was not reflected in body weight change. Dry matter digestibilities, based on indigestible ADF, were 63.5 and 64.8% for the DWCS and LWCS diets. It was calculated that 2.5 and 1.5% of the consumed seeds were excreted as whole cottonseeds in feces with the DWCS and LWCS diets, respectively. Although statistically significant, treatment differences in the proportion of intact seeds in the fecal DM would have little nutritional consequence. Mechanically delinted WCS performed as well as LWCS for all of the cow performance and milk composition variables measured.