The technique of restriction fragment length polymorphism (RFLP) analysis, of chloroplastic and genomic DNA, was investigated as a means of identifying eucalypt species and cultivars which are morphologically indistinguishable from one another. In order to resolve chloroplast DNA (cpDNA) RFLPs, a method was developed to extract high yields of intact chloroplasts from Eucalyptus grandis S/N M6. Starch contamination was reduced by incubation of saplings in the dark for 48 h prior to extraction and watering with a solution containing 370 mM Na-phosphate and 296 mM KN03. Optimal chloroplast yields (25 ug chlorophyll/g fresh mass) were obtained by chopping leaf material, using a vertical homogenizer, in a buffer containing 350 mM sorbitol, 50 mM tris-HCL and 5 mM EDTA, 0.1 % (w/v) bovine serum albumin, 0.15 % (w/v) 2-mercaptoethanol, 2 mM L-ascorbic acid and 1 mM MgCI2 followed by washing of leaf pieces in a buffer containing only sorbitol, tris-HCL and EDTA. When these chloroplasts were used in an "in-organelle" DNA digestion procedure, polymorphisms were observed between the cpDNA profiles resolved for E. grandis S/N M6 and that of an outgroup species (spinach). However, the developed chloroplast extraction technique could not be used to obtain chloroplasts from various other eucalypt species, probably as a result of variability in the material at an ultrastructural or biochemical level. For the analysis of genomic DNA RFLPs, a DNA extraction procedure was optimized for use with various eucalypt species and cultivars. This included the development of a purifcation technique during which DNA was ammonium acetate-ethanol precipitated and subjected to mini-dialysis. Following Dra I restriction of DNA, the extract was electrophoresed and Southern blotted onto both nylon and nitrocellulose membranes. These were probed with a Hind-III restricted sample of the multilocus plasmid probe pV47-2. This probe was labelled using 32p as well as a non-radioactive labelling substance digoxygenin (DIG). Hybridization conditions, including the composition of the hybridization buffer, were optimized for use with these labels, and DNA RFLPs (fingerprints) were resolved for the eucalypt species E. grandis and E. macarthurii and cultivars of E. grandis (S/N M6, TAG 5 and TAG 14). An average of 8.5 bands were detected with 32p and 5.0 fragments with DIG. All the species and cultivars fingerprinted with the 32P-label could be distinguished from one another. However, as a result of the reduced sensitivity of the DIG system, two of the E. grandis cultivars, S/N M6 and TAG 5, could not be differentiated. It is concluded that the latter system would be most suitable for incorporation into a routine eucalypt screening programme, although it is suggested that the colourimetric detection assay, used in this study to resolve DNA bands, be replaced by a more sensitive one.

Thesis (M.Sc)-University of Natal, Durban, 1993.