An article in Current Archaeology 293 (July/August 2014) (see also www.archaeology.co.uk) makes the point that Silbury Hill sits at the head of the Kennet River – which joins the Thames at Reading. Although modern maps have the head of the Thames near Lechlade, but might this be regarded as one of several tributaries. It asks, perhaps Neolithic people saw the Kennet as more properly a continuation of the Thames, rather than the long loop that goes by way of Oxford. They go further and say, Silbury Hill was positioned in a special landscape. Earlier people, in the 4th millennium BC, constructed the West Kennet long barrow nearby, for example, with its striking bull's horn entranceway, and the Windmill Hill causewayed enclosure, a series of pits and ditches arranged in a concentric circle. The author of the article, Jim Leary, makes the point the Marsden henge, the largest one to survive, lies at the head of the River Avon. A henge is quite unlike Silbury so that is perhaps an unjustified coincidence – but Avebury is nearby and is close to some of the springs that form the Kennet.
Rivers continued to be important in myth, right the way down to the Celtic Iron Age. However, the latest explanation for Silbury, as expressed by Leary and friends, is that it had a communal role – with no particular end result foreseen. This appears to have occurred to them in the manner the mound of Silbury was fashioned – in a series of stages, over four generations. It was built roughly contemporary with the Egyptian pyramids – and has a vague similarity in form – especially as we now know the flattened top was a result of occupation in the Saxon era when it briefly became a temporary defended site against Viking raids. Hence, if the top had originally been more pointed in nature it would dramatically become a sort of pyramid – made of chalk and turves and soils. The pyramids, according to Clube and Napier, imitated the enhanced zodiacal lights they claimed were a feature of the 3rd millennium BC sky. This pyramidal region of light was located on the plane of the ecliptic – where dust caught in sun light gave the appearance of a heavenly river in the sky. In Egypt this was viewed as a Nile in the sky – but in Britain and Ireland different rivers were involved – such as the Boyne, the Avon, and the Kennet/Thames. I don't know if anyone has looked for a similar association with the Ouse, or the Severn, but alternative myths may have prevailed in different parts of the land. Hence, a pyramidal mound at the end of a river may have imitated the zodiacal light towards the end of the plane of ecliptic – where dust from comets and meteors accumulated prior to wider dispersal.