One of the main factors in the development of pediplains is parallel retreat of slopes. Such retreat depends on the spacing of streams and bedrock structure and occurs mainly on upper and middle slopes.
As with other planation surfaces, the question of scale or size is important. Extreme smallness of such surfaces does not represent the facts of structure and process and even stage satisfactorily. Thus pediments or pediplains may be on continental or sub-continental scale as in South Africa or Brazil or Peninsular India or Australia or Antarctica. But there may be 'toy' pediments adjoining gullies.
The pediment whose slope is believed to range from 5° to 7° is a faintly concave cutrock surface. It is a hydraulic curve, i.e., shaped mainly by the action of running water. It is thought that such a surface is most suited to deal with the discharge of short-lived torrential cloudbursts characterizing arid and semi-arid regions. During such downpours strongly erosive, turbulent, thick sheet flow occurs which may modify pediments within a few hours. Pediments usually have a veneer of sediments and gravel. There may be a layer of weathered rock or the bedrock may be totally devoid of weathered matter or sediment. Pediments too may have sometimes, convex elements in their slopes where soil creep occurs and performs rounding of slope angularities.
Twidale is skeptical about the formation of pediments outlined by L.C. King. He doubts scarp retreat by linear erosion through streams and pediment formation by laminar flow of water during downpours.
The processes of pediment formation are complex and naturally involve weathering, rill wash, stream erosion, mass wasting, sheet wash, sheet flood, etc.
It has been claimed that pediment can be distinguished from other similar planation surfaces by the nature of the junction of the pediment with the upland occurring behind it. It is generally abrupt and sharp particularly where hard rocks like granite, or hard capping rocks or porous laterites occur.
Twidale does not accept that the abruptness of the junction is due to faulting or undercutting by lateral streams. He also doubts King's view that the sharpness of the junction is due to linear stream erosion of the scarps in contrast with laminar or sheet erosion on the pediment. On the pediment surface residual hills in massive coarsed-grained and jointed gneiss-granites and other crystalline rocks have been variously named—domes-gneiss in India, bornhardts or inselbergs in South Africa and inselberg in Ghana.
Pecsi was inclined to regard Tamilnadu plain as a pediplain from the observed characteristics of the area. Although it is difficult to prove the existence of all the diagnostic features, which will entitle a surface either as a pediplain or as a peneplain, some characteristics may be identified. In most cases, the identification is partial.
Thus, the backing slope may be abrupt here and gentle there, it will be convex here and concave there in the same area. The veneer of sediment may be thin here and thick there in the same region. The sediment may or may not contain gravels which would characterise a pediplain because of the abrupt backing slope.
As mentioned earlier some geographers were inclined to regard Chotanagpur as a dead erosion surface or a fossil pediplain. This means that formerly during an arid phase the area was a pediplain.
During the succeeding moister climate much waste has been produced by weathering that it could not be removed and the subaerial processes are not working on the bedrock which has been insulated from the upper surface by the thick layer of waste. As a support for this view these geographers pointed out the occurrence of gravels directly above the bedrock where it was exposed in the bed of a stream. One cannot arrive at such a major conclusion from one solitary example.
King visualised a simple pediplanation cycle marked by rill, gully and river incision and formation of scarps in the earlier stages followed by parallel retreat of scarps leaving ever-expanding pediments in front of them, during the later stages of the cycle. The residuals are eliminated by the intersection of the pediments. Then the weathering of the intersecting slopes or soil creep may produce some convexity in these last stages of the cycle. The points which weigh in favour of Chotanagpur being regarded more as a series of pediplains than peneplains include the occurrence of laterite, which is commonly absent from peneplains, and its treppen nature.