Higher velocity portions of a stream tend to be driven to the outside of a meander (1) (Greene). On the outside of the meander, the surface of the water has a tendency to be slightly higher or super-elevated because it has gained momentum and acceleration ('The Hydrology of Streams'). Here, the flow is forced down the outer bank which results in a steeper velocity gradient and greater bed shear stresses, and it returns to the surface toward the inside of the meander where flow is less turbulent (2) ("Secondary Flow and Channel Change in Braided Rivers"). This is because the flow is helical in form (3). Because helical flow posses inertia, the maximum circulation and erosion caused by a stream are beyond the inflection of the meander (Greene). When stream flow reaches the outer bank of a meander, the increased acceleration erodes it away causing toe scour and undercut banks (4) ("Secondary Flow and Channel Change in Braided Rivers"). Sediment is deposited at the slower moving inside bend (5). Helical flow maintains the channel profile as the river erodes its way across the valley floor (6). For example, the larger the cross-sectional area of a river, the slower helical flow will be absorbed by friction. For this reason, larger rivers have meanders with longer wavelengths (Greene). The meandering of a river or stream, which maintains a steady channel gradient and matches the waterway's flow characteristics, is a result of helical flow ("Rivers").

Until the mid 1970s, it was thought that secondary flow in meander bends consisted of only a single cell of helical rotation carrying surface water toward the outer bank and bed water toward the inner bank. In the mid 1970s, studies indicated that there was an additional small cell of reverse rotation close to the outer bank, but the unsophisticated equipment used in these studies lead to inaccurate information and information lacking detail. ("Secondary Flow and Channel Change in Braided Rivers")

The introduction of the electromagnetic current meter allowed for more detailed measurements to be made. A subsequent study found evidence that a second cell of helical flow did exist at two out of three meander bends. Conclusions were made that the existence of the second cell depended on two factors. The first was that the outer bank must be perpendicular to the flow of the river. The second factor used to determine if the second cell would be present was the strength of the component of flow toward the outer bank. ("Secondary Flow and Channel Change in Braided Rivers")

A later area of study was that of helical flow in braided rivers and streams. Due to the number of channels in a braided stream, there tends to be more than one helical cell at the confluence of the channels. The area of maximum scour occurs where two cells of helical rotation attach. Sediment transport is restricted to the areas between opposing cells. Further down from the confluence, the larger of the two helical cells tends to dominate. Deposition occurs where these cells separate again. There is also evidence of helical cells of reverse rotation further down from the channel confluence. Studies indicate that in braided streams at the entrance to the confluence helical flow may be the result of horizontal separation vortices in the lee of the bank faces especially when the converging channels are of unequal depth. Also, similar to single thread meandering streams helical flow may be the result of the relationship between the outwardly directed centrifugal force and the inwardly directed pressure gradient force caused by super-elevation. ("Secondary Flow and Channel Change in Braided Rivers")