Use of a multi-exponential model to assess the effect of fermentation in the reticulorumen on particle fragmentability as simulated from artificially macerating leaf and stem fractions of two tropical grasses

Abstract

This study reports the use of a mechanical macerating device to characterise the patterns of, and resistance to, fragmentation of leaf and stem fractions of spear grass ( Heteropogon contortus) and sabi grass ( Urochloa mosambicensis). The four substrates, cut to a length of 1.5–2 cm, were exposed to microbial fermentation in the rumen of cattle for periods of between 0 and 96 h. A novel macerating device was used to grind substrates in a water medium for a series of predetermined number of rotor revolutions of the macerator. The extent of fragmentation was then measured by wet sieving. Statistical modelling, using phase-type distributions, was used to describe the numbers of revolutions required to fragment the substrates into three compartments defined by particles retained on screens of 1180, 600, and 50 μm pore size. The study identified multi-exponential patterns of fragmentation of tropical grass fractions. Resistance to fragmentation of large particles, estimated by the number of rotor revolutions required for 0.75 of particles to be removed from that pool, was high for both stem fractions, and was not markedly affected by time of microbial fermentation. Spear grass leaf had high resistance to fragmentation until exposed to microbial fermentation for 24 h, after which it attained the low resistance to fragmentation typical of sabi grass leaf at all microbial fermentation times. Fragmentability of large particles, defined as the proportion of small particles originating from the large particle pool which did not pass through the medium particle pool, was always high for leaf (> 0.66), but for stem fractions it was quite variable. Sabi grass stem achieved maximum fragmentability values typical of leaf (0.70), but for spear grass stem fragmentability was lower at 0.52. The potential digestibility of particle fractions following various degrees of maceration, was assessed after in vitro fermentation for 120 h. Extensive maceration of samples not exposed to rumen fermentation increased sample digestibility of the leaf fractions by 3 to 7%, and of stem of spear grass and sabi grass by 380 and 120% respectively. For sabi grass stem, the relative increase in potential digestibility after maceration progressively declined in samples after rumen fermentation; no such decline was seen for spear grass stem. The implications of the results to nutritive value for the ruminant eating forage diets are discussed. It is suggested that the phase-type statistical modelling techniques used are of value in the study of particle movements and fragmentation in the reticulorumen.