Forage quality usually is evaluated on a total‐herbage basis although in vitro digestible dry matter (IVDDM) and crude protein (CP) concentrations can vary greatly among plant parts. This study was conducted to compare IVDDM and CP concentrations of leaf blades (leaves) and remaining plant parts (stems) and the relative amount of these plant parts present within plant‐canopy segments of alfalfa (Medicago sativa L.), birdsfoot trefoil (Lotus corniculatus L.), and red clover (Trifolium pratense L.). Unseparated herbage of white clover (Trifolium repens L.) was included for comparison. Plants were sampled three or four times during two 6‐week growth periods, starting when plants were in the prebud stage. At each harvest, alfalfa and birdsfoot trefoil stems were separated above 6th and 12th nodes (from cut stem base), and leaf blades were removed. Red clover was divided above the 3rd and 6th nodes, and leaf blades also were removed. The IVDDM concentration in total herbage was greatest in white clover and frequently least in alfalfa. The IVDDM concentration of alfalfa and birdsfoot trefoil stems decreased about twice as rapidly as total herbage plant IVDDM concentration during advancing maturity. Stem bases had less IVDDM concentration than stem tops with a rate of change among stem segments of about 20 g kg‐1 dry matter (DM) node‐1. Red clover stems had greater IVDDM concentration than did leaves until flowering. The rate of change among red clover stem segments was less than 8 g kg‐1 DM node‐1. In contrast to IVDDM, range in CP concentration among the plant parts was nearly as large at the beginning as at the end of each period because of a nearly parallel rate of CP decline in the plants. Leaf mass decreased and stem mass increased in the basal segment during each period, which decreased the leaf‐to‐stem mass ratio of basal segments to 0.06 or less for alfalfa, 0.16 for birdsfoot trefoil, and 0.23 or less for red clover by the end of the two periods. We conclude that attempts to improve forage quality of alfalfa and birdsfoot trefoil should be directed toward the lower portion of the plant canopy.

Much of the Variability in digestibility and voluntary intake of herbage is closely associated with variation in cell‐wall (CW) concentration and components. This study was conducted to compare CW characteristics of total herbage and plant parts of alfalfa (Medicago sativa L.) birdsfoot trefoil (Lotus corniculatus L.), red clover (Trifolium pratense L.), and white clover (Trifolium repens L.), and to evaluate relationships between CW characteristics and in vitro digestible dry‐matter (IVDMD) concentration. Plants were sampled four times during spring growth and three times during summer regrowth. With the exception of white clover, subsamples were divided into leaves (leaf blades) and stems (remaining plant) and grouped by stem‐node segment. The CW concentration in total herbage generally was least in white clover and greatest in alfalfa. Within species, CW concentration was least for leaves and greatest for stem segments from plant bases. Stems of red clover had a lower CW concentration than stems of alfalfa or birdsfoot trefoil. Cell‐wall lignin concentration was greatest in birdsfoot trefoil and least in the two clover species. Within species, CW‐lignin concentrations of leaves and stems were similar. Both CW and CW‐lignin concentrations were inversely related to IVDMD concentration in stem segments and total herbage with correlation coefficients ranging from –0.40 to –0.98. The relationships was inconsistent in leaves. Inclusion both CW and CW‐lignin concentrations in linear multiple‐regression equations acounted for 95 and 84% of the observed variation in IVDMD concentration in stems and total herbage, respectively.

Near infrared reflectance Spectroscopy (NIRS) technology needs verification as to its utility for forages that represent diverse species, environments, and managements. Our objectives were to determine whether single NIRS calibration equations could reliably assess components of the feeding quality of alfalfa (Medicago saliva L.) grown to a range of maturities in several environments and whether single equations could be developed to simultaneously analyze components of the quality of total forage and plant parts of alfalfa, birdsfoot trefoil (Lotus cornieulattis L.), red clover (Trifolium pratense L.), and white clover (Trifolium repens L.). In the first experiment, we assayed four forage quality components in 234 alfalfa samples from eight cutting schedules of two cultivars grown in two locations in each of 2 years. In Exp. II, we assayed five quality components in 167 samples of total forage and plant parts selected from seven harvests from two growth periods of two cultivars of each of four perennial legume species. Calibration equations for in vitro digestible dry matter (IVDDM), crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), and acid detergent lignin (ADL) were obtained by multiple linear regression of known conventional laboratory assay values on NIR spectra from up to 204 samples using a scanning monochromator NIR spectre‐computer system. Equations were validated with up to 49 additional samples that had been selected from the original populations. From three to nine wavelengths were required to develop the best calibration equations for the various quality components. The squared coefficients of determination (R2) of these equations ranged from 0.93 to 0.99. These equations allowed NIRS analysis of quality values in verification samples within standard errors (dag/kg) of 1.56 to 2.06 for IVDDM, 0.42 to 1.0 for CP, 1.46 to 2.23 for NDF, 1.30 to 1.70 for ADF, and 0.63 for ADL. We concluded that our NIRS system can quickly evaluate the feeding quality of perennial legumes to a degree of accuracy similar to that of laborious conventional analyses and that several species can be evaluated simultaneously with single calibration equations nearly as well as can single species.