Bolus propulsion during the normal oral phase of swallowing is thought to be characterised by the sequential elevation of the front, middle, and posterior regions of the dorsum of the tongue. However, the coordinated orchestration of lingual movement is still poorly understood. This study examined how pressures generated by the tongue against the hard palate differed between three points along the midline of the tongue. Specifically, we tested three hypotheses: (1) that there are defined individual patterns of pressure change within the mouth during liquid swallowing; (2) that there are significant negative pressures generated at defined moments during normal swallowing; and, (3) that liquid swallowing is governed by the interplay of pressures generated in an anteroposterior direction in the mouth. Using a metal appliance described previously, we measured absolute pressures during water swallows in six healthy volunteers (4 male, 2 female) with an age range of 25-35 years. Participants performed three 10-ml water swallows from a small cup on five separate days, thus providing data for a total of 15 separate water swallows. There was a distinct pattern to the each of the pressure signals, and this pattern was preserved in the mean obtained when the data were pooled. Furthermore, raw signals from the same subjects presented consistent patterns at each of the five testing sessions. In all subjects, pressure at the anterior and hind palate tended to be negative relative to the preswallow value; at mid-palate, however, pressure changes were less consistent between individuals. When the pressure differences between the sites were calculated, we found that during the swallow a net negative pressure difference developed between anterior and mid-palate and a net positive pressure difference developed between mid-palate and hind palate. Large, rapid fluctuations in pressure occurred at all sites and these varied several-fold between subjects. When the brief sharp reduction in pressure that occurred early in each swallow was used to determine the sequence of events, we found that activity occurred first at the anterior of the palate followed by the mid-palate and then the hind palate. There was a considerably longer and more variable delay between the start of activity at the front of the palate than at the rear of the palate. To obtain an index of the "effort" involved in generating the pressures at each site regardless of direction (positive or negative), we obtained the product of the root mean square (RMS) pressure change during each swallow (kPa) and its duration (s). Overall, the most effort appears to have occurred at the front of the palate and the least at mid-palate. Our results also showed that some participants exerted a small amount of midline pressure when swallowing, while others used a relatively large amount of tongue pressure. We conclude that while tongue behaviour during swallowing follows a classical sequence of rapid shape changes intended to contain and then propel the bolus from the oral cavity to the pharynx, there is a large range of individual variability in how this process is accomplished.