AbstractAspects of engineering—oriented and geo-morphological runup studies are discussed in relation to the fact that there is little field data against which to check theoretical and model-derived conclusions concerning swashzone flow. Also, it is suggested that there has been little consideration of the effects of variability in wave trains on beach face changes. The“phase-difference”model of Kemp (1958) is examined and shown to be useful in this regard.The results of 21 experiments in which swash (2133) and backwash (1631) velocities were measured together with other surf and runup properties are presented and discussed. The data are forHb0.3 to 2.44 m;Tb7.5–10.0 sees, and characterise non-barred, high energy, mesotidal, mixed sand and gravel beaches of mild slope (5–12°) at Kaikoura, New Zealand.Breaker height is shown to be the primary control of swash length as predicted by Kemp while phase of flow exerts important effects on the temporal structure and hence the net circulation of sediments entrained by the flow. Velocities were determined with a force-plate dynamometer having an accuracy of±15 cm sec-1. Average maximum runup velocity was 168.0 cm sec-land the average duration was 2.98 sees. Backwash velocities averaged 140 cm sec-1with a mean duration of 4.25 sees. Runup velocities conformed to a near-Gaussian probability distribution and 20–60 % of the incident wave velocity is transmitted to runup velocity. The relationship between relative runup velocity and breaker energy is negative—exponential in form for flow between the breakers and still water level. This result is the opposite of one obtained on steep, slopes in a model tank by earlier workers.