The Role of Vortices in Drumlin Formation 

Drumlins are streamlined; they generally have a remarkably symmetrical, oval form. They have been compared to the shape of snowdrifts, and fast swimming fishes.  [Charlesworth, 1957]. Their profile is comparable to the aerodynamic design of airplane wings. Could the drumlins have somehow been formed by a scraping, shearing and grinding action as vast, hypothetical sheets of ice slid across the land? 

This incongruous mechanism has been seriously proposed in some recent papers which attack and criticize the subglacial flood model of drumlin origin proposed by John Shaw, who argues that drumlins were formed by currents of enormously powerful floods beneath an ice sheet. [Eyles & Boyce, 1998, 1999; Shaw, 1999]. 

The reasoning is that faults in volcaniclastics and breccias can gouge the rock surfaces, producing characteristic marks, including striations, ridges and grooves, crescentic scours around projections, and "stoss and lee" effects such as "rat tails" behind boulders, and these marks are comparable to drumlins and elongated low profile flutings.  So by analogy, a mechanism of gouging and shearing between the base of the ice and the drift or bedrock that it over rode is believed responsible for forming patterns of drumlins, as the hypothetical ice sheet scraped over the landscape. 

The catastrophic subglacial flood idea of Shaw is thus rejected by these writers. They point out some defects in Shaw's arguments, such as Shaw's postulated flow of a subglacial megaflood in the region west of Lake Ontario that seems incorrect. There was more likely a flow towards the west that fanned out to the northwest in the Guelph area, as they suggest [Eyles & Boyce, 1999, p. 159]. But the real issue is the question about the nature of the agent of streamlining that caused drumlins; was it shearing and grinding of fault surfaces, or the action of vortices in a rapidly flowing fluid? 

One wonders if some of the criticism of Shaw's model is in defense of a philosophical principle of uniformitarianism, in the sense that Charles Lyell advocated it in the early days of geology. Lyell ruled out catastrophic mechanisms, even the glacial theory of Agassiz, which he regarded as too catastrophic. Eyles and Boyce complain that "the megaflood model is based on inferred processes that  have never been observed or measured on the scale required at modern glaciers or ice sheets"  [Eyles & Boyce, 1999, p. 160]. Apparently Lyell's detrimental influence lingers, and the idea of any kind of a flood continues to be philosophically distasteful to some. 

Another paper about drumlin origin cited studies of deformed debris scoured by a glacier at Breidamerkurjökull, in Iceland, apparently with little resemblance to true drumlins [Boulton, 1987, p. 57]. Defining grooves and ridges of debris over-ridden by ice as drumlins seems to constitute flawed logic, yet this error persists. Elevating hypotheses to fact will not likely lead geologists to an understanding of how drumlins really formed. Boulton suggested that drumlins formed by plastic deformation of the drift and rock that the ice sheets passed over, but this seems improbable for real drumlins since there is a great difference between the strengths of  the various materials comprising typical drumlins; how could drumlins of bedrock be explained by deformation of the material comprising them? 

J.K. Charlesworth discredited reports of drumlins being formed by glaciers, when he wrote: "An appeal to modern glaciers is unavailing, since drumlins in statu nascendi are unknown though accumulations somewhat resembling them have been described from time to time." [Charlesworth, 1957, p. 395]  An experienced observer, Charlesworth rejected the claims that the ridges of glacier molded debris were actually drumlins. 

What is ignored by many who investigate drumlins is that the drumlins are not caused by ice, and could never be formed by the direct action of a glacier. The effect of a glacier scraping over the ground resembles the effect of a bulldozer more than blowing snow, or the effects of a swift flowing stream on its sandy bed, both of which cause streamlined bed forms that resemble typical drumlins, as well as ripples and dunes. 

A bulldozer may indeed be capable of making straight, long marks in the dirt, but can not build streamlined forms; that is the work of fluids, and streamlined bed forms are evidence for vortices, as is shown by the common transition of drumlins to transverse bed forms that resemble ripples. The rogen ridges associated with drumlins in some regions are similar in form to giant current ripples, which are transverse bedforms generated by vortices that acted as rollers in former currents. Both longitudinal and transverse vortices act to minimize friction at the sediment-fluid interface [Folk, 1976]. 

There is also a question about whether it is valid to compare the effects of faulting with the movements of hypothetical ice sheets, which would only have a very slow motion, whereas displacements during faulting occur with lightning speed and so the shearing effects may not be at all comparable to those of ice. 

In my model [Cox, 1979] the sediment involved in the formation of drumlins was probably not unconsolidated drift, but unconsolidated sediment of the type that occurs in the bedrock of the regions where drumlins occur, which must have been unconsolidated when the patterns of drumlins formed, and lithified after the streamlining of the land surface. 

Drumlin near Meaford, OntarioIt is not only the rounded form of  the hills in drumlin areas that needs to be explained; the troughs between drumlins are also an integral part of the drumlinized landscape; a shearing mechanism cannot explain the rhythmic nature of the drumlin spacing and their smooth, symmetrical forms, as shown in the photo at right of one of the drumlins near Meaford, Ontario. 

The disintegration theory of the drift makes possible a new interpretation of the drumlins. The sediment was streamlined as fast currents swept the land. Currents were generated by uplift of areas that were submerged. As the land was being eroded and streamlined, disintegration occurred. The disintegration process penetrated to various depths, forming layers of sand, clay, and gravel. This is the origin of the drift mantle. This disintegration process in drumlinized areas resulted in drumlins with varying composition. Drumlins  with the same orientation, and similar form and dimensions, can have quite different composition, from bedrock, to part bedrock and part drift, and those entirely composed of drift, including layers of cross stratified sand and gravel. 

Where the drumlins were disintegrated entirely, drift drumlins formed. Where there was only a minimal amount of disintegration, a rock cored drumlin with a thin veneer of drift could be formed. Where no disintegration occurred, the drumlins are rock drumlins. All these types occur in the same drumlin field, sometimes in close proximity. The disintegration process that formed the drift, was not the mechanism which caused streamlining, but both processes could operate in the same environment. 

Where the drift covering of some rock-cored drumlins was eroded by currents and swept away, the roches moutonnes were formed. They generally have a smooth stoss end composed of bedrock and the lee end, where pressure was less, became disintegrated and the drift was removed by the currents. 

The idea of a grinding mechanism, comparable to a fault causing drumlins is discredited further by the variations in form of drumlins, which is consistent with fluids causing them, but not ice sheets. In northwestern New York the drumlins vary in form, according to elevation. They are large, rounded and flat topped in the lowlands south of Lake Ontario. Further south they became better streamlined, and tended to be grouped in clusters. Still further south, on the rising slopes of the Allegheny Upland, the drumlins were smaller, more intricately streamlined, with steep sides and narrow crests. In the region of the Cayuga trough the drumlins became elongated flutings [Miller, 1972]. 

This reflects the increase in the velocity of the fluids forming the drumlins as they flowed over higher ground. Where depth decreased, velocity increased, by the principle of continuity. But the shearing and grinding of a hypothetical ice sheet moving over the region could not explain these variations in drumlin form; an ice sheet being pushed uphill cannot increase its velocity at higher elevations, but could only move as fast as the ice providing the push, so this mechanism fails miserably to account for the variation in form of New York drumlins. The appeal to faulting as an analogy for drumlin formation appears misguided and futile.


Boulton, G.S. A theory of drumlin formation by subglacial sediment deformation, in J. Menzies and J. Rose (eds.) Drumlin Symposium, A.A. Balkema, Rotterdam, 1987, pp. 25-80. 

Charlesworth, J. K. 1957. The Quaternary era, with special reference to its glaciation. Volume 1. Edward Arnold, London. 

Cox, D.E. 1979. Drumlins and Diluvial Currents. Creation Research Society Quarterly, 16(3):154-162 

Eyles, N. and J. I. Boyce, 1998, Kinematic indicators in fault gouge: tectonic analog for soft-bedded ice sheets-reply. Sedimentary Geology 116: 1-12. 

Eyles, N. and J. I. Boyce, 1999, Kinematic indicators in fault gouge: tectonic analog for soft-bedded ice sheets-reply. Sedimentary Geology 123:157-168. 

Folk, R. L. 1976. Rollers and ripples in sand, streams and sky: rhythmic alteration of transverse and longitudinal vortices in three orders. Sedimentology, 23(5):649-669. 

Miller, J.W. Jr. 1972. Variations in New York drumlins. Annals of the Association of American Geographers, 62(3):418-423. 

Shaw, J., 1999, Kinematic indicators in fault gouge: tectonic analog for soft-bedded ice sheets-comment. Sedimentary Geology 123:153-155. 

For references to published papers by Shaw, see:

Drumlins and Subglacial Meltwater Floods

Copyright © 1999 by Douglas E. Cox
The Creation Concept
HotBot Search for