Thesis (MScAgric (Agronomy)--University of Stellenbosch, 2008. / A two year experiment (2004-2005) was conducted at the Tygerhoek Experimental Farm near Riviersonderend in the Western Cape Province of South Africa. The effect of different crop rotation systems on soil properties, disease and insect pests, weed populations, wheat growth, yield and quality in the wheat crop phase, included in these crop rotation systems, was determined. This trial was part of a long term crop rotation experiment started in 2002. This trial was laid out as a block design with four replications. Crop rotation systems included wheat, barley, canola, lupins and pasture phases which consisted of medics and clovers planted collectively. Soil samples were taken at each replication for N-incubations for determination of mineral N (NO3 - -N plus NH4 + -N) at 0-150 mm soil depth. A basic soil chemical analysis was done at 0-150 mm and 150-300 mm soil depths, respectively. Each sub-plot (replication) consisted of a 3 m2 block that was divided into a 1.5 m2 block for harvest and smaller 0.25 m2 blocks for samples that were taken at different growth stages throughout both seasons. Dry mass and nitrogen (N) content of different plant components, leaf area index, disease symptoms and pest damage were recorded from each sample. Trends in basic soil chemical properties mostly differed between crop rotation systems during different seasons while similar trends in soil mineral nitrogen occurred. Highest soil mineral N levels occurred after one or two consecutive years of pasture while levels after a lupin phase were disappointingly low in both seasons. These high soil mineral N levels showed similar trends to wheat grain quality and some wheat yields, while the most influencing factors on wheat grain yield were probably soil physical properties. Soil mineral N after canola was high during plant after which levels were much lower than many other crop rotation systems. This occurrence will probably need a re-evaluation of N fertilizing programs if the same trends are found in similar, but longer trials. Lolium spp. was the most prominent weed that occurred in both seasons at some crop rotation systems seemingly with no direct effect from crop rotation. Highest disease incidence mainly from Septoria spp. and Puccinia spp. occurred, particularly in wheat/wheat rotations, except for Puccinia which showed high ratings of disease symptoms in all crop rotations in the drier 2004 season. Lower ratings occurred in crop rotation systems when wheat was preceded by nonwheat crops. Insect pest damage showed no similar trends indicating no direct effect of crop rotation on these pests and/or effective control from applied pesticides in both seasons. It was concluded that climate was one of the most influencing factors affecting differences and seem to be the main cause for different trends found between these two seasons in similar crop rotation systems. A similar trial with longer duration than this one is thus needed to obtain conclusive trends. This also indicates the importance of integration of crop rotation and management practices that are most optimal during dry and wet seasons, thus limiting risk.

Thesis (PhD(Agric) (Agronomy))--University of Stellenbosch, 2003. / 108 leaves printed on single pages, preliminary pages i-vi and numbered pages 1-1 to 1-5 to 9-1 to 9-3. Includes bibliography, list of abbreviations and 22 figures in color. Digitized at 300 dpi grayscale and 300 dpi 24-bit Color to pdf format (OCR), using an HP Scanjet 8250 Scanner. / ENGLISH ABSTRACT: A long term experiment was conducted at the Langgewens Experimental Farm, near Malmesbury in the Western Cape province of the Republic of South Africa. The effect of soil tillage, crop rotation and nitrogen fertiliser rates on mineral-N levels in the soil, nitrogen levels in plants, wheat growth and yield components, grain yield as well as quality parameters of spring wheat were determined. Although the tillage treatments tested were initiated in 1976, present crop rotations and nitrogen application rates were only applied since 1990. Most of the data that are discussed however was recorded during the 1997 to 2001 period. The trial was designed as a randomised complete block with a split-plot arrangement and four replicates. Main plots were tillage methods namely conventional tillage (CT), tine tillage (TT), minimum tillage (MT) and no tillage (NT). Crop rotations used were continuous wheat (WW) and wheatl lupin/wheat/canola (WLWC). Both soil and crop data were recorded. The results of this study showed the differences in total mineral-N content of the soil. N-contents of wheat, wheat growth and yield components as well as wheat quality parameters were found between tillage treatments, crop rotations used as well as N- fertiliser rates applied. Response however, varied largely between years due to annual variation in especially total precipitation and distribution of rainfall. The inclusion of a legume crop (lupin) and canola in the rotation with wheat was found to have only a small effect probably due to the fact that lupins were grown once in a four year cycle only. Application of different nitrogen rates did increase the mineral-N in the soil, but the effect did not last very long in most years due to either N-leaching or plant uptake. To ensure sufficient mineral-N levels, late application of N-fertiliser will therefore be needed. Minimum tillage or reduced tillage performs better than conventional tillage in low rainfall years. In general these tillage systems should be combined with crop rotation to ensure that yields are comparable to that obtained with conventional tillage. / AFRIKAANSE OPSOMMING: Die ontwikkeling, graanopbrengs en bakkwaliteit van koring (Tritium aestivum L.) in reaksie teenoor gewasratasie, metode van grondbewerking en N-bemestingspeil is in 'n lang tennyn sllldie op Langgewens Proefplaas, naby Malmesbury in die Weskaap provinsie van die Republiek van Suid-Afrika, bepaal. Hoewel die verskillende grand bewcrkingsmetodes sedert 1976, en gewasrotasies en slikstofpeile sedert 1990 toegepas is, is daar in hierdie studie gekonsentreer op data wal gcdurcnde die periode 1997-2001 ingesamel is. Die eksperiment is as 'n randomiseerde blok ontwerp met vier bewerkingsmetodes nl. konvcnsionele bewerking (CT), tand bewerking (TT). minimum bewerking (MT) en geen bewerking (NT), twee gewasratasies nl. monokuhuur koring (WW) en koring/lupienelkoringiCanola (WLWC) en drie N-peie (60 kg N ha- ', 100 kg N ha- 1 , 140 kg N ha- I uitgevoer. Aile stikstof is in die vorm van kalksteen-ammoniumnitraat toegedien. Die reaksie van beide grondfaktore soos die minerale N inhoud en gewaskomponente 5005 biomassa produksie, opbrengs en kwaliteit teenoor bogenoemde faktore het '0 goeie korrelasie getoon met die heersende klimaatstoestande. Lae reenval jare (gebiede) sal volgens hierdie stlldie die grootste voordeel verkry met minder intensiewe grand bewerkingstelsels. terwyl 'n frekwensie van meer as een peulgewas per siklus van 4 jaar nodig sal wees om grondvrllgbaarhcidsvlakke betekenisvol te verhoog. Stelsels van rninder intensiewe grondbewerking is ook tot 'n grater mate deur gewaswisseling bevoordeel as konvensionele melOdes van grondbewerking. Hocwel minerale-N vlakke in grond deur verskillende N-peile be'lnvloed is, was die efTektiwiteit van toedienings laag en het verdeelde tocdienings groat voordele ingehou.