Skip to main content
ARS Home » Research » Publications at this Location » Publication #174403


item Cassida, Kimberly
item Turner, Kenneth - Ken
item Foster, Joyce

Submitted to: Animal Feed Science and Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/21/2006
Publication Date: 5/4/2007
Citation: Cassida, K.A., Turner, K.E., Foster, J.G. 2007. Comparison of detergent fiber analysis methods for forages high in pectin and condensed tannin. Animal Feed Science And Technology. 135:283-295.

Interpretive Summary: Pectin and tannin interfere with fiber analysis of broadleaf forages. Accurate determination of nutritive value is essential for efficient incorporation of such alternative forages into production systems for small ruminants. We compared fiber analysis methods for grass, legume, tannin-containing legume, and brassica forage groups. Performing analyses in sequence on the same sample improved estimation accuracy of fiber concentrations in high-pectin brassicas and legumes compared to conducting analysis on a new sample for each step, but had no effect on results with low-pectin grasses. Neither method produced acceptable accuracy for tannin-containing legumes. This work is useful because it provides guidelines for improving methods of fiber analysis for high-pectin forages and corroborates analytical interferences in tannin-containing legumes. The work will benefit researchers, plant breeders, and commercial forage testing laboratories that service farmers by providing more reliable estimates of feed nutritive value to better balance livestock rations for improved nutrient-use efficiency and performance at least cost.

Technical Abstract: Pectin, biogenic silica, and condensed tannin (CT) interfere with detergent fiber analysis of broadleaf forages that are of interest in small ruminant production systems. We compared two analysis sequences for neutral (NDF) and acid (ADF) detergent fiber (non-sequential, NONSEQ; sequential, SEQ) with and without correction for acid insoluble ash (AIA). Methods were compared both for individually processed (Trial 1) and batch-processed samples (Trial 2). Hemicellulose concentration was calculated as the difference between NDF and ADF values. Samples included legumes (alfalfa [Medicago sativa L.], red clover [Trifolium pratense L. cv. Arlington]), CT-containing legumes (birdsfoot trefoil [Lotus corniculatus L. cv. Norcen], sericea lespedeza [Lespedeza cuneata L.]), brassica forages (turnip top and root [Brassica campestris var. napus cv. Purple Top], rape leaf and stem [B. napus L. cv. Winifred] and grasses (smooth bromegrass [Bromis inermis L., oats [Avena sativa L.]). Forage groups differed in response to treatments for most variables. Ash correction reduced ADF only for brassicas (205 vs. 191 g kg-1 for uncorrected and ash-free) under individual processing, but for all forage groups under batch processing (295 vs. 288 g kg-1 across all groups). Fiber analysis sequence did not interact with ash correction for ADF or hemicellulose, and had a larger effect on outcomes than did ash correction. Compared to NONSEQ, SEQ analysis reduced ADF of brassica and legume forages by 18 to 41 g kg-1 with comparable increases in hemicellulose in both individual and batch processing systems. In contrast, SEQ analysis overestimated ADF and underestimated hemicellulose of tannin-containing legumes by up to 27 g kg-1 compared to NONSEQ analysis. Grass ADF was unaffected by analysis sequence in either trial. The difference between NONSEQ and SEQ ADF increased with pectin concentration (r2 = 0.99) across the broadleaf forage group means. Sequential fiber analysis was preferred for forage brassicas and legumes, but non-sequential methods were adequate for grasses. Both NONSEQ and SEQ methods yielded misleading results for tannin-containing legumes.