Thesis (MScAgric)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: Grinding of feedstuffs prior to in sacco incubation results in fine particles that could escape from dacron bags without being degraded. The objective of this study was to determine the effect of the removal of these fine particles on protein degradation parameters. The feedstuffs that were used were protein feedstuffs that are commonly used in dairy cattle diets in South Africa. Feedstuffs were soybean oilcake, sunflower oilcake, maize gluten 20, maize gluten 60 and fish meal. Treatments were; 1. grinding through a 2 mm screen with no subsequent sieving, 2. grinding through a 2 mm screen followed by sieving through a 106 μm mesh, and 3. grinding through a 2 mm screen followed by sieving through a 150 μm mesh. In the first trial, the material that was lost through sieving and the chemical composition of the different treatments were determined. Protein solubility was also determined. Between 20 and 60% of the material dry matter was lost with the sieving. The chemical composition for the soybean and sunflower oilcake and maize gluten 60 was similar between the three different treatments (or fractions). The CP content of fish meal and maize gluten 20 differed somewhat between the treatments. The 106 μm mesh seemed to be most suitable for fish meal, but a suitable mesh size could not be found for maize gluten 20. In the second trial, the degradability parameters were determined according to the in sacco degradation procedure. Three lactating Holstein cows that were fitted with rumen cannulae were used. The cows received a commercial lactation diet and oat hay that was supplied ad libitum. Samples of all the protein sources were placed in dacron bags and incubated in the rumen. The following removal times were used: 0, 2, 4, 8, 12, 24 or 48 hours. Dry matter and CP disappearances were determined, and the values were used to estimate DM and CP degradability parameters using a non-linear model. Effective CP degradability was also determined. The a-values were affected most of all. On average, the a-values were 39.4 and 40.3% higher for the un-sieved treatments than for the sieved treatments, for DM and CP, respectively. The effective CP degradability was also, on average, 43% higher for the un-sieved treatments. Grinding without the subsequent sieving of samples appears to result in an overestimation of DM and CP degradation in the rumen. It is therefore recommended that after grinding, feedstuffs should be sieved through at least a 106 μm mesh in preparation for in sacco incubations. / AFRIKAANSE OPSOMMING: Die maal van grondstowwe ter voorbereiding van in sacco inkubasies lei tot ‘n potensiële verlies van klein partikels uit die dakronsakkies sonder dat hulle gedegradeer word. Die doel van hierdie studie was om die invloed van die verwydering van fynmateriaal op proteïen-degradeerbaarheidsparameters te bepaal. Die grondstowwe wat in hierdie studie gebruik is, is proteïengrondstowwe wat algemeen in melkbeesdiëte in Suid-Afrika gebruik word. Die grondstowwe was soja-oliekoek, sonneblom-oliekoek, mielie gluten 20, mielie gluten 60 en vismeel. Behandelings was; 1. maal deur ‘n 2 mm sif sonder verdere sifting, 2. maal deur ‘n 2 mm sif gevolg deur sifting deur ‘n 106 μm sif en 3. maal deur ‘n 2 mm sif gevolg deur sifting deur ‘n 150 μm sif. In die eerste proef is die hoeveelheid materiaal wat verlore gaan as gevolg van sifting van die grondstowwe bepaal en die chemiese samestelling van die verskillende behandelings. Proteïenoplosbaarheid is ook bepaal. Tussen 20 en 60% van die materiaal het verlore gegaan as gevolg van sifting. Die chemiese samestelling van die soja- en sonneblom oliekoek, asook dié van die mielie gluten 60 was soortgelyk vir al drie die behandelings. Die ruproteïeninhoud (RP) van die vismeel en mielie gluten 20 het verskil tussen die drie behandelings. Dit wil voorkom asof die 106 μm sif die mees geskikte is vir vismeel, maar ‘n geskikte sif kon nie vir mielie gluten 20 gevind word nie. In die tweede proef is die degradeerbaarheidsparameters bepaal met behulp van die in sacco-metode. Drie lakterende Holsteinkoeie met rumen kannulas is gebruik. Die koeie het ‘n kommersiële melkbeesdieët ontvang en hawerhooi ad libitum. Die monsters is in dakronsakkies in die rumen geïnkubeer. Die sakkies is na die volgende inkubasietye verwyder: 0, 2, 4, 8, 12, 24 of 48 uur. Die DM- en RP-verdwyning is bereken en die waardes is gebruik om die DM- en RP- degradeerbaarheidsparameters te bereken met behulp van ‘n nie-lineêre model. Effektiewe RP- degradeerbaarheid is ook bereken. Die waardes wat die meeste beïnvloed is, is die a-waardes. Die a-waardes was gemiddeld 39.4 en 40.3% hoër vir die ongesifte behandelings as vir die gesifte behandelings, vir DM en RP, onderskeidelik. Die effektiewe RP-degradeerbaarheid was ook gemiddeld 43% hoër vir die ongesifte behandelings. Dit wil voorkom asof DM- en RP-degradeerbaarheid oorskat word wanneer voermonsters slegs gemaal word. Dit word aanbeveel dat grondstowwe ten minste deur ‘n 106 μm sif gesif word ter voorbereiding vir in sacco-studies.