The movement of a solute through the soil provides an opportunity for porous networks in the soil to be characterised. Soil pores are flow pathways that act as conduits for solute. Together they form a porous network which is the soil structure. Typically most solute transport studies have focused on the fate of chemicals or the transport process itself. Few studies have quantified both structure and solute transport parameters together, so understanding is still developing in this area. Therefore the aim of this study was to ascertain the relationship between solute flow and the pore volume through which the solute flows - the macroporous structure, - in soil under different tillage treatments. The study was part of a broader evaluation of changes to key soil physical properties of two tillage treatments - Raised Beds (RB) and a control treatment consisting of Conventional Cultivation (CC). The site was near Geelong in south-western Victoria, Australia (38o 10„S S, 144 o 05„S E). The soil was a texture contrast soil and was classified as a Xeric Alfisol. For the solute transport experiment, large undisturbed soil cores (24 cm diameter) were excavated in July 2004 to a depth of 20 cm. In the laboratory a background solution of 0.02M KCl was applied followed by a pulse of 2M KCl. This was first performed under slightly unsaturated conditions (-30 mm tension) and then close to saturation (-5 mm tension) using a disc permeameter. The out-flowing solute was collected from under each core. Prior to the solute transport experiment, image analysis of resin-impregnated soil was undertaken on small undisturbed cores (10 cm diameter). This revealed pore and solid size distributions and several descriptive parameters of structure. At - 30 mm tension the drainage flux density was similar for the two tillage treatments, but at -5 mm tension the RB drainage was much faster. BreakThrough Curves (BTC) were plotted of solute concentration against the volume eluted and revealed that the peak concentration appeared earlier in the CC cores, but the BTC for the RB were much more stable. A Convective Lognormal Transfer function (CLT) was fitted to the BTC, and the first and second moments (ƒÝ and ƒã2) were used to assess the solute transport characteristics. There was no difference in ƒÝ between the treatments which indicated the mean travel pathlength was similar. But ƒã2, which describes the amount of solute spreading, was sensitive to tillage treatment. At - 5 mm tension, ƒã2 in the CC cores was significantly larger than in the RB cores, but no significant difference was found at -30 mm tension. This shows that the connected nature of pores was different. Less solute spreading in the RB cores suggested they contained a better connected pore network. Another descriptive parameter was derived, the transport volume (ƒást), to determine the fluid volume of solute participating in transport. During flow at - 5 mm tension, the RB cores had a significantly smaller ƒást than the CC cores, indicating more preferential flow in the former; however there was no difference at -30 mm tension. Because there was no significant difference in the total pore volume between tillage treatments, the observed differences in solute transport were probably due to the pore architecture within each treatment and not the porosity. The quantitative image analysis of macropore structure showed that both the pore and solid size distributions were similar between the treatments; also there was little change with depth for any structural parameter. The surface area of pores was larger for RB, which relates to the area where fluid may flow. Pore genus (a measure of pore connectivity) was greater for RB and the size of the pores in the RB soil was on average smaller than the CC soil. Thus, overall the RB soil was better structured than the CC soil. The image analysis results agreed with the solute transport analysis and indicated how soil structural changes can influence solute transport. The lack of difference in the transport parameters at -30 mm tension reflected the incomplete filling of fluid within the pore volume. Thus, differences in pore arrangement were only detected when the pore network was at -5 mm tension.