ANNUAL

REVIEWS Further
Annu. Rev. Fluid Mech. 1991. 23: 65-79 Quick links to online content

DRAG REDUCTION IN NATURE!

D. M. Bushnell

Fluid Mechanics Division, NASA Langley Research Center, Hampton,

Virginia 23665-5225
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

K. J. Moore
by University of Ottawa on 02/02/12. For personal use only.

Cortana Corporation, Falls Church, Virginia 22046

KEY WORDS: swimming, flying, resistance minimization

INTRODUCTION

The "energy crisis" of the 1970s rekindled interest and research in drag­
reduction techniques for all types of land, sea, and air transportation. As
in any research, the kernel problem for drag reduction is the genesis and
development of new approaches, techniques, insights, and understanding.
One source of inspiration for alternative drag-reduction approaches is a
renewed study of Avians and Nektons, i.e. fliers and swimmers in the
natural world. The presumption is that drag-reduction adaptations have
evolved for improved efficiency or speed, or both, thereby aiding species
survival in the Darwinian sense. Such a study should result in (a) identi­
fication of approaches that technology could pursue and, if successful,
optimize for practical application; (b) identification of instances where
existing human-derived technology occurs in the natural world; and (c)
improved understanding of animal form and function. This review docu­
ments the current status of such an examination conducted intermittently
over the last 10 years by the authors for ultimate application to such systems
as aircraft, submersibles, surface ships, and long-distance pipelines.
An appropriate beginning is to define and delineate the various forms
of drag affecting both natural and man-made fliers and swimmers. Poten­
tially the largest drag component is pressure or form drag, which is par-

1 The US Government has the right to retain a nonexclusive, royalty-free license in and to
any copyright covering this paper.
 66 BUSHNELL & MOORE

ticularly troublesome when flow separation occurs. The basic physics for
this drag component involves the viscous influence upon the ideal or
inviscid-flow pressure field. Some pressure drag, at a relatively benign
level, occurs even if the flow is attached, simply because of the uncambering
of the surface by viscosity-induced flow displacement. However, once flow
separation occurs, this drag component increases tremendously. There­
fore, the foremost consideration for drag control and mitigation for species
survival and efficiency is probably the avoidance of flow separation. The
two remaining drag components (for submerged swimmers and fliers)­
skin-friction drag and drag due to lift-are usually of the same order (for
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

fliers) and much smaller than the pressure- or form-drag component,

except in the attached-flow case (where the pressure drag is also small).
Skin-friction drag is the result of the no-slip boundary condition on the
by University of Ottawa on 02/02/12. For personal use only.

surface and can either be laminar (low Reynolds number) or turbulent

(high Reynolds number). In fact, the maintenance of laminar flow to higher
Reynolds number is an obvious, and currently much in vogue (again),
technique for obtaining skin-friction reduction. The various drag mech­
anisms are not independent, and as laminar flow is more easily separated,
it may be advantageous to artificially trip the flow to the turbulent case to
avoid the large pressure-drag penalty associated with separated flow. The
remaining drag component (drag due to lift) is caused by flow spillage on
lifting surfaces from high- to low-pressure regions and affects both fliers
(which require lift to remain in the air) and swimmers (many of which
utilize "lifting surfaces" for propulsion and control).
The bulk of the drag-reduction techniques identified thus far in the
natural world are discussed in the present article under their appropriate
heading (e.g. Form-Drag Reduction, Skin-Friction Drag Reduction, and
Drag-Due-to-Lift Reduction), so the reader can gain an appreciation for
the diversity of "natural" drag-reduction techniques. As stated previously,
these techniques can often be utilized for either improved efficiency or
increased speed, depending upon the ecological niche occupied by the
species. Three classes of drag-reduction approaches are described: (a)
those that have proven performance potential, i.e. careful human research
demonstrates that the method "works"; (b) approaches for which only
preliminary data exist; and (c) morphological observations that, in the
opinion of the present authors, are worth furthcr scientific study to deter­
mine their efficacy.

FORM-DRAG REDUCTION

The fundamental problem in form-drag reduction is to avoid flow separa­

tion, both steady and unsteady (vortex shedding), for three-dimensional
 DRAG REDUCTION IN NATURE 67
flows. The three-dimensional specification is important as there are few, if
any, two-dimensional or truly axisymmetric creatures (or creature parts) in
nature, although close approximations are common in human technology.
Flow separation (flow breakaway from the surface) is induced by positive
or adverse pressure gradients, i.e. pressure fields where pressure increases
in the streamwise direction. For highly three-dimensional flows, transverse
pressure gradients can also induce separation. Typically, the forward por­
tion of the body is a region of falling pressure, whereas the after portion
is subject to increasing pressure. Therefore, afterbody regions are the most
prone to large-scale flow separation, albeit perhaps less so for the three­
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

dimensional "natural" bodies prevalent for Nektons and Avians as com­

pared with the two-dimensional and axisymmetric bodies often favored in
human technology. Flow separation occurs when flow momentum near
by University of Ottawa on 02/02/12. For personal use only.

the surface is insufficient to overcome the increasing pressure. The local

flow far from the surface contains higher momentum, and the canonical
flow-separation control/pressure-drag reduction problem is to transfer
momentum toward the wall from the outer flow or, in anticipation of
current "mission-adaptive wing" technology, utilize "variable geometry"
to mitigate local pressure gradients, a ploy widely adapted by Nektons
and Avians.

Turbulent Flow

As stated in the introduction, laminar flow is much more separation-prone

than the turbulent case, where dynamic eddying motions cause outer- to
inner-region (near-wall) momentum transfer that can delay separation.
Therefore, one form of flow-separation control is the establishment of
turbulent wall flow. Adverse pressure gradients themselves constitute a
major turbulizing factor on the flow; however, several species also deploy
roughness, using either projecting bands near the position of maximum
body girth (e.g. Walters 1962, Bone 1975, Aleyev 1977, Webb 1978) or
embedded (and unstable) separated-flow regions [such as on the dragonfly
wing (Newman et aI1977»), presumably to ensure the presence of turbulent
flow over the afterbody. This particular ploy is only usable in cases where
(a) the forward portion of the animal is kept smooth and laminar for low
drag (Aleyev 1977), and (b) the Reynolds number (ratio of dynamic to
viscous forces) is large enough to allow turbulence production using "trips"
(geometric irregularities). The efflux from fish gills, positioned near the
maximum girth, may also provide a turbulence-enhancement function
(Magnuson 1978), as could the mass addition from subdermal canals and
passive porous surfaces (Bone 1972, Bone & Brook 1973) and feather
porosity-induced surface bleed. Turbulization of the caudal or propulsive
 68 BUSHNELL & MOORE

fin is probably hastened by the small-scale turbulence shed by the caudal

finlets due to swimming body motions (Aleyev 1977). For the fully laminar
lower Reynolds number case, the "natural" approach of choice for pres­
sure-drag reduction is "design" for reduced pressure gradient [i.e. longer,
more gradual afterbody regions (e.g. Aleyev 1977), leaf or branch defor­
mation under load (Meroney 1968, King & Loucks 1978), and use of
auxiliary body features to exploit favorable interference, such as the alula
near the leading edge of bird wings (Hertel 1963, McMasters 1986), which
is similar to the leading-edge slot utilized on aircraft].
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

Vortex Generators

An alternative approach for transferring momentum toward the wall for

by University of Ottawa on 02/02/12. For personal use only.

flow-separation control is the establishment of "stationary" longitudinal

vortex motions. Such "vortex generators" were "discovered" in the lab­
oratory and researched in the 1940s and 1950s; they are now utilized
extensively and routinely in human fluid-flow technology. Their appear­
ance in nature considerably antedates their human discovery. Vortex­
generator realizations in the natural world include the owl leading-edge
comb (Hertel 1963, Blick et al 1975), certain bird feathers that "pop up"
under critical loading (McMasters 1986), shark dermal denticles and other
scales that deform differentially upon body motion ( Bechert et al 1986,
Pershin et al 1976), and the small separately set finlets behind the second
dorsal and anal fins on many fast-swimming fish (Walters 1962, Steer 1963,
Aleyev 1977). For three-dimensional bodies and/or angles of attack, large­
scale nose-region-induced vortices can control leeside separated flows, and
fighter-aircraft designers are adapting the "shark-nose" configuration as
a result of studies that indicate that this nose shape produces vortices that
are more effective in separation control than "conventional" axisymmetric
noses. Large surface grooves, both longitudinal and transverse, have also
proven successful in reducing flow separation through vortex production
and alteration (Goodman & Howard 1985). Studies at the NASA Langley
Research Center of typical cactus shapes indicate sizable pressure-drag
reductions from alteration of the well-known Karman vortex sheet by the
transverse body grooves. Examples of longitudinal body grooves in the
natural world include shell indentations (Vogel 1981). These two cases
(cacti and shells) are examples of organisms that, while being quite station­
ary, still "live fast," i.e. cacti in high-desert windstorms and shells in
underwater currents in the benthic boundary layer. As a final note, swim­
ming body motions can locally induce dynamic favorable pressure gradi­
ents (along with vortices), which can both delay separation and reduce
turbulence production.
 DRAG REDUCTION IN NATURE 69

Mass Transfer
Another alternative approach to "energizing" the near-wall flow to obviate
separation drag is to simply add momentum directly by blowing at high
speed along the wall. This is again widely used in industry and is possibly
utilized in nature by fish that close down the inboard gill during turns and
shunt the gill efflux into the outboard, separation-prone region of the body
(Lighthill 1969). The orientation and placement of the gill openings are
particularly well suited to this jet-blowing technique (Babenko & Koval'
1982), which is at the heart of much of the high-lift aviation technology.
An alternative mass-transfer technique is passive "bleed," or utilization of
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

porous surfaces and subdermal channels in combination with the bodyjwing

pressure field to remove the inner (near-wall) low-momentum portion of
the flow in separation-prone regions (Bone 1972, Bone & Brook 1973).
by University of Ottawa on 02/02/12. For personal use only.

Adaptations for Body Intersection Regions

This section describes adaptations for handling more localized flow-sep­
aration control problems engendered by body-appendage intersections.
The pressure field associated with intersection regions typically produces a
horseshoe or necklace vortex that wraps around the base of the appendage
(wing, fin, etc) and streams back along the body. This vortex usually
constitutes a net drag increase, as, in this case, it is not generally employed
to control an even larger separated-flow region. A further separation-drag
problem with appendages occurs during maneuvering when the inter­
secting body is no longer aligned with the flow and therefore tends to create
and shed large separated flows. Natural adaptations for drag mitigation in
intersection regions act to reduce the causative transverse pressure gradi­
ents and include (a) filleting; (b) sweeping, often in a far better and more
sophisticated manner than current practicc in man-made tcchnology; (c)
elastic deformation to achieve optimum shape (Pershin et al 1979); and
(d) concentration of mucin-producing cells (Pershin 1978). Details of these
natural "fairings" have been little studied, and such research could yield
very valuable insight into passive techniques for three-dimensional vortex
control. Intersection separation induced by angle-of-yawjmaneuvering is
approached through variable geometry-in the shark case the rear portion
of the appendage near the surface flaps to the side as the body is moved,
providing a "turning vane" to guide the flow. An adaptation about which
there is current speculation is the caudal finlets and keels that occur near
the caudal fin-body intersection. Along with providing muscle attachments
for side-to-side movement of the caudal fin, the keel appendages rotate
the major body axis 90° (from "vertical" to "horizontal") and may thereby
promote attached flow over the caudal fin (Aleyev 1977). The flow field
 70 BUSHNELL & MOORE

associated with caudal keels requires considerable further study. The cau­
dal finlets along the margin of the body probably serve to segment the
organized vorticity shed by the body due to swimming motions and, when
in pairs, accelerate the flow into the caudal (propulsive) fin ( Walters 1962).
One of the most obvious methods of dealing with protuberance drag is to
eliminate the protuberance, and this is accomplished on some species by
folding back various fins into the body at high speeds (8urdak 1957,
Aleyev 1977, Magnuson 1978).

SKIN-FRICTION DRAG REDUCTION

Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

Skin-friction drag reduction is accomplished in nature through two basic

approaches: (a) maintenance of laminar (low-drag) flow as long as possible
by University of Ottawa on 02/02/12. For personal use only.

through use of smooth surfaces and favorable pressure gradients, and/or

(b) body smoothness/alteration of the structure of turbulent motions once
the near-wall flow becomes turbulent. Obviously, decreasing the skin fric­
tion by large amounts in the presence of adverse pressure gradients is
inadvisable, as this would invite flow separation and attendant large drag
increases. Much of the skin-friction reduction technology identified thus
far in the natural world is for the underwater case.

Sur/ace "Additives"
In water, except for maintenance of laminar flow, surface additives provide
the largest skin-friction drag-reduction payoff. These additives are of three
types: (a) polymers, (b) surfactants, and (c) bubbles. Detailed studies
indicate that most fish slimes exhibit significant drag-reduction behavior
(Rosen & Cornford 1971, Hoyt 1975), with maximum effectiveness occur­
ring upon deposition into the very near-wall region. Drag reductions well
above 50% are commonly measured for the polymer case (Povkh et al
1979). The appearance of ctenoid scales in the turbulent-flow (and usually
only in the turbulent-flow) regions of fish (Pershin et al 1976, Aleyev 1977;
Figure 1) suggests that the toothlike structure of these scales may aid in
the deposition of the slime-polymer into the critical near-wall region. These
"teeth" are a subroughness and therefore should not, by themselves, affect
the turbulence directly. It should be pointed out that while fish slime does
contain high-molecular-weight polymer compounds, such as mucopoly­
saccharides, nucleic acids, and proteins (Pershin et al 1976), it also
contains surfactants in the form of lipids, phospholipids, and lipoproteins
(Lewis 1970, Mittal & Agarwal 1977, Lebedeva & Chernyakov 1978,
Zaccone 1979). It is only recently that surfactants were recognized as
producing drag-reduction effects similar to those of aqueous solutions of
 DRAG REDUCTION IN NATURE 71
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org
by University of Ottawa on 02/02/12. For personal use only.

Figure 1 Photograph of a typical ctenoid scale.

polymer. While optimum concentrations of drag-reducing surfactants are

generally higher than those for polymers, surfactants can be more robust,
i.e. less fractionation/mechanical degradation of the molecular complex
compared with the high-molecular-weight polymers.
The other additive, bubbles, decreases the average density near the wall,
and careful laboratory studies again indicate drag reductions above 50%.
Both the polymer and surfactant mechanisms are unique to water appli­
cations, since, in the air case, all gases are approximately Newtonian, and
there is no injectable substance readily available that has a much smaller
density. The use of bubbles for drag reduction in nature is still speculative
and concerns sailfish, seals, and penguins, whose travel near and through
the air-water interface can trap air within body surface layers, which is
then observed to "outgas" from the surface in bubble form (Ovchinnikov
1971). This conclusion is tenuous, as the bubble sheet is neither massive
nor continuous as required in the laboratory experiments with man-made
devices.
 72 BUSHNELL & MOORE

A further additive consists of antifouling compounds, which, by obvi­

ating formation of biofouling in Nektons, ensure smooth, low-drag sur­
faces (Burdak 1973, Baier et al 1984).

Morphology

Considerably smaller [0(10%)] but still interesting levels of skin-friction

reduction are available from surface and body-geometry modifications.
Laboratory studies in the USSR indicate that the swordfish sword serves
as a drag-reducing device (Aleyev 1977, Kozlov & Babenko 1978). The
mechanism involves establishment of turbulent flow on the sword itself
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

through special roughness provided by tubercles, folds, and spinelets

(Ovchinnikov 1966) and takes advantage of the fact that skin friction
decreases as the viscous or boundary layer thickens on the body along the
by University of Ottawa on 02/02/12. For personal use only.

flow direction. In the swordfish case, the high drag associated with early
thin turbulent layers occurs on the sword, which has a small wetted area
and hence a small total drag. By the time the turbulent flow reaches the
main body of the animal, the viscous layer is thick and, therefore, the drag
per unit area is smaller over the main body region, which contains the
major portion of the wetted surface.
Another geometrical alteration associated with skin-friction drag
reduction is the ridge feature occurring on shark dermal denticles (Pershin
et al 1976, Reif 1978, 1982, Reif & Dinkelacker 1982, Raschi & Musick
1986, Bechert et al 1986; Figure 2). These ridges are lined up with the flow
and are of a size and shape similar to the NASA "riblets" (Walsh &
Lindemann 1984) utilized on the 12-m yacht Stars and Stripes in the 1987
America's Cup finals in Australia and in crew races in the 1984 Olympics.

DRAG-DUE-TO-LIFT REDUCTION

Historically, drag-due-to-lift reduction in nature has been studied far less

than either pressure or friction-drag reduction, and even then mostly for
the Avian/air case. This is curious, as studies identify this drag component
as generally of greater import than skin-friction drag for some fish species
(Magnuson & Weininger 1978). The basic problem in drag-due-to-lift
reduction is to either make use of or reduce the tip bleed flow that occurs
from the high- to low-pressure regions of a lifting surface. The first and
most obvious ploy is to increase the span-to-chord ratio ("aspect ratio")
of the lifting surface, as this makes the tip flow increasingly less important
in the overall dynamics. Such an approach is limited by structure/strength
of materials considerations but is employed in nature to the extent possible
[e.g. the albatross has a wing aspect ratio of 17 (Cone 1962b)]. Other well-
 DRAG REDUCTION IN NATURE 73
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org
by University of Ottawa on 02/02/12. For personal use only.

Figure 2 Photomicrograph of a shark dermal denticle.

documented "natural" drag-due-to-lift reduction techniques include the

use of tip sails or tip feathers to segment the vortex that occurs at the tip
as a result of the high- to low-pressure bleed (e.g. Cone 1962a, Spillman
1978). The general concept is to "utilize" the angled tip flow by deploying
embedded surfaces (split tips) that supply a thrust component from their
lift vector. These tip feathers are used (for example, on the condor) evi­
dently in lieu of larger aspect ratio and are also useful in partially "block­
ing" the wing-tip bleed. Another drag-due-to-lift reduction technique stud­
ied extensively is wing upsweep (Cone I 962c). Optimized (elliptical)
loading is also evidently employed on at least some bird wings and is
achieved via wing twist (Oehme 1971).
More recently, various underwater creatures have been reexamined for
morphological features possibly associated with drag-due-to-lift reduction.
These studies have resulted in several new avenues of research and some
early successes. The initial observations involved qualitative comparisons
of caudal-fin geometry as a function of structural makeup. In general, for
the caudal fin, as one proceeds from bone (fish) to a cartilaginous structure
(sharks) and finally to simple muscle (whale), the caudal-fin aspect ratio
tends to become smaller (understandably from a structural point of view),
but the fin itself becomes more geometrically complex. The supposition
 74 BUSHNELL & MOORE

was that various morphological complexities may tend to compensate for

the reduction in aspect ratio. Specific features identified for laboratory
study included (a) swept-back tapered tips (which also occur on birds); (b)
serrated trailing edges, both locally near the tip (shark) and along the
entire trailing edge (humpback whale, many birds); (c) leading-edge bumps
(pectoral fin on humpback whale, head on hammerhead shark; Figure 3);
and (d) "fin rays" or wavy fl ow-a ligned surface relief, including
the optimal alignment of denticles near the tip. These features are listed in
decreasing order of research attention received thus far. The swept-back
tapered tips evidently do reduce drag-due-to-lift, at least partialIy as a
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

result of a vertical distribution of the lift vector (Finch 1984, van Dam
1987, Vijgen et al 1989). Tests at the NASA Langley Research Center
indicate beneficial effects from serrated trai li ng edges (Vijgen et al 1990)
by University of Ottawa on 02/02/12. For personal use only.

but as yet theoretical justification is lacking because of the complex nature

of the flow. The fin rays and leading-edge bumps are, up to this point,
unexamined. Limited data (Mair 1955) indicate that spanwise gradients
(such as those induced by b ump s) can alter stagnation-point pl acement ,
rotate the lift vector into the thrust direction, and lead to drag reduction.
The shark tip is particularly interesting, in that the outboard portion of

Figure 3 Planform of the forebody of a hammerhead shark.

 DRAG REDUCTION IN NATURE 75
the indentation evidently flips from side to side during caudal-fin motions,
thereby forming a combination "winglet" and serration.
Viable drag-due-to-lift reduction techniques are of utmost importance
in aircraft applications. Research over the last 10 years on skin-friction
reduction (laminar-flow control on wings, turbulent skin-friction reduction
for fuselages) has yielded flight-verified drag-reduction techniques capable
of reducing overall friction drag up to 0(40%). However, since con­
ventional aircraft optimize near the condition where skin-friction drag is
approximately equal to drag due to lift, full utilization of this skin-friction
drag-reduction technology may not be possible without concomitant
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

reductions in drag due to lift-hence, the interest in candidate drag-due­

to-lift reduction devices.
by University of Ottawa on 02/02/12. For personal use only.

STATUS OF PORPOISE DRAG-REDUCTION

STUDIES

The present authors, when speaking on the subject of drag reduction in

nature, are invariably queried concerning dolphin drag reduction. This
wide public interest in the dolphin case evidently results primarily from a
combination of "Gray's Paradox" and the "compliant-wall" literature of
the 1950s and 1960s. In the late 1930s, Sir James Gray suggested, based
upon energetics (steady-state energy balance), that the drag of various
underwater creatures, including the dolphin, had to be inordinately low to
correspond to speed claims (Gray 1936). This was followed in the early
1960s by an article by Max Kramer claiming that the "compliant" dolphin
skin damped turbulent motions (Kramer 1961).
Research conducted in the US and the USSR since the Kramer article
has considerably clarified the situation. The Soviets have published (a)
fluctuation measurements obtained in the boundary layer of a free-swim­
ming dolphin and telemetered back to a shore station (Romanenko &
Yanov 1973, Kozlov & Shakalo 1973), and (b) energetics calculations. In
the US, Lang has inferred dolphin drag during "coast-down" tests (Lang
& Pryor 1966, Lang 1975). This research indicates that during coasting
(absence of swimming body motions), the dolphin drag and boundary­
layer behavior are nominally what one would expect without any special
drag-reducing feature. In tests involving swimming at low speeds , the
Soviets observed a lower turbulence fluctuation level, which they attribute
to local pressure gradients induced by the swimming body motion
(Romanenko 1981).
At high speeds, the energetics do indicate large apparent drag reductions.
A possible explanation of the high-Reynolds-number situation was given
by Au & Weihs (1980), who suggested that the dolphin, which must breathe
 76 BUSHNELL & MOORE

air, simply "porpoises" when traveling at high speed, i.e. it leaps out of the
water and thereby reduces its dragjorce by a factor of 800 (density ratio
of air to water). Au & Weihs argue that this more than pays for the inter­
face or wave drag and accounts for the abnormally low apparent drag
coefficients inferred from the assumption of fully submerged travel. Further,
they argue that body surfing on bow waves accounts for certain near-ship
qualitative dolphin speed observations. Controversy remains concerning
the drag-reduction effectiveness of dolphin mucin (Uskova et al 1974).
The status of research concerning the compliant-skin effect upon the
near-wall flow is that both extensive theory and limited experimental data
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

indicate a correctly designed soft or compliant surface can delay transition

from laminar to turbulent flow. There is a lack of experimental replication
by independent investigators to bolster scattered claims of apparent com­
by University of Ottawa on 02/02/12. For personal use only.

pliant-wall drag reduction under turbulent flows; alternative explanations

have, in fact, been advanced to explain many of the observations (Bushnell
et al 1977).

BEHAVIORAL TECHNIQUES FOR

REDUCING DRAG

The "porpoising," or leaping out of the water, discussed in the previous

section is an obvious example of drag reduction through behavioral
modification. Another example analyzed by Weihs (1973b, 1974) is an
alternating gliding and swimming behavior for underwater creatures that
lack swim bladders and therefore must "swim to live" (i.e. to avoid
sinking). As the drag while swimming is estimated to be much larger than
that while gliding (largely because of propulsive drag-due-to-lift/vortex­
shedding drag), there may be a favorable energy benefit in swimming
intermittently and maximizing gliding time. Similar behavior for birds, i.e.
"bounding flight," has been analyzed by Lighthill ( 1977), Rayner ( 1977),
and Ward-Smith (1984). Another behavioral modification resulting in drag
reduction involves cooperative behavior. The "upwash" associated with
drag due to lift from one individual can be utilized by his/her neighbors
(while swimming/flying in formation) to ease their propulsion energy
requirements (Lissaman & Shollenberger 1970, Weihs 1973a, Hummel
1978). The V-shaped formation of migrating Canadian geese is a typical
example. Flying in the "ground-effect" regime also significantly reduces
drag due to lift (Withers & Timko 1977, Blake 1983, Hainsworth 1988).
It should be obvious from the examples of natural drag reduction cited
herein that mankind has much to gain from continued study of natural
hydro- and aeromechanics.
 DRAG REDUCTION IN NATURE 77
Literature Cited

Aleyev, Y. 1977. Nekton. The Hague, Neth: 33(3): 40-41

W.Junk Goodman, W. L., Howard, F. G. 1985.
Au, D., Weihs, D. 1980. At high speeds dol­ Axisymmetric bluff-body drag reduction
phins save energy by leaping. Nature 284: through geometrical modifications. J.
548-50 Aircr. 22(6): 516-22
Babenko, V. V., Koval', A. P. 1982. Hydro­ Gray, J. 1936. Studies in animal locomotion.
dynamic functions of swordfish gill sys­ VI. The propulsive powers of the dolphin.
tem. Bionika 1982(16): 11-15 J. Exp. Bioi. 13: 192-99
Baier, R., Meenaghan, M., Wirth, J., Gucin­ Hainsworth, F. R. 1988. Induced drag
ski, H., Nakeeb, S. 1984. Porpoise and savings from ground effect and formation
killer whale skin as natural examples of flight in brown pelicans. J. Exp. Bioi. 135:
low drag, low adhesion biomaterial sur­ 431-44
faces. Trans. World Congr. Biomater., 2nd, Hertel, H. 1963. Structure, Form and Move­
Washington, D.C., 7: 190. San Antonio, ment. New York: Reinhold
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

Tex: Soc. Biomater. Hoyt, 1. W. 1975. Hydrodynamic drag

Bechert, D. W., Bartenwerfer, M., Hoppe, reduction due to fish slimes. In Swimming
G., Reif, W.-E. 1986. Drag reduction and Flying in Nature, ed. T. Y.-T. Wu, C.
mechanisms derived from shark skin. Pre­ J. Brokaw, C. Brennen, 2: 653-72. New
by University of Ottawa on 02/02/12. For personal use only.

sented at Congr. Int. Counc. Aeronaut. York: Plenum

Sci., 15th, London Hummel, D. 1978. Recent aerodynamic con­
Blake, R. W. 1983. Mechanics of gliding in tributions to problems of bird flight. Proc.
birds with special reference to the influence Congr. Int. Counc. Aeronaut. Sci. (ICAS),
of the ground effect. J. Biomech. 16(8): 11th, Lisbon, ed. 1. Singer, R. Staufenbiel,
649-54 I: 115-29
Blick, E., Watson, D., Belie, G., Chu, H. King, D., Loucks, O. L. 1978. The theory of
1975. Bird aerodynamic experiments. In tree bole and branch form. Radiat. En­
Swimming and Flying in Nature, ed. T. Y.­ viron. Biophys. 15: 141-65
T. Wu, C. J. Brokaw, C. Brennen, 2: 939- Kozlov, L. F., Babenko, V. V. 1978. Eksper­
73. New York: Plenum imental Issledovaniya progranichogo sloya.
Bone, Q. 1972. Buoyancy and hydrodynamic Kiev: Inst. Hydromech.
functions of integument in the castor oil Kozlov, L. F., Shakalo, V. M. 1973. Certain
fish, Ruvettus pretiosus (Pisces: Gempy­ results of the determination of pulsation
lidae). Copeia 1972(1): 78-87 of velocities in the boundary layer of dol­
Bone, Q. 1975. Muscular and energetic phins. Bionika 1973(7): 37-40
aspects of fish swimming. In Swimming Kramer, M. O. 1961. The dolphins' secret.
and Flying in Nature, ed. T. Y.-T. Wu, C. J. Am. Soc. Nav. Eng. 73: 103-7
J. Brokaw, C. Brennan, 2: 493-528. New Lang, T. G. 1975. Speed, power, and drag
York: Plenum measurements of dolphins and porpoises.
Bone, Q., Brook, C. E. R. 1973. On Schedo­ In Swimming and Flying in Nature, ed. T.
phi/us Medusophagus (Pisces: Stromateo­ Y.-T. Wu, C. J. Brokaw, C. Brennen, 2:
idei). J. Mar. Bioi. Assoc. UK 53: 753-61 553- 72. New York: Plenum
Burdak, V. D. 1957. Peculiarities of the onto­ Lang, T. G., Pryor, K. 1966. Hydrodynamic
genetic development and phylogenetic performance of porpoises (Stenalla atten­
relationships of black sea mullets. Sevas­ uata). Science 152: 531-33
topol Bioi. Stud. 9: 243-73 Lebedeva, N. E., Chernyakov, Y. L. 1978.
Burdak, V. D. 1973. Scale types as stages in Chemical danger signal in the predator­
the historical development of the hydro­ victim system among fish. Zh. Evol. Bio­
dynamic function of fish integument. khim. Fiziol. 14(4): 392-97
Zool. Zh. 1973(8): 1208-13 Lewis, R. W. 1970. Fish cutaneous mucus: a
Bushnell, D. M., Hefner, J. N., Ash, R. L. new source of skin surface lipids. Lipids
1977. Effects of compliant wall motion on 5(11): 947--49
turbulent boundary layers. Phys. Fluids Lighthill, M. 1. 1969. Hydromechanics of
10: S31-48 aquatic animal propUlsion. Annu. Rev.
Cone, C. D. Jr 1962a. The soaring fiight of
. Fluid Mech. I: 413-46
birds. Am. Sci. 50: 130-40 Lighthill,l. 1977. Introduction to the scaling
Cone, C. D. Jr. 1962b. Thermal soaring of of aerial locomotion. In Scale Effects in
birds. Am. Sci. 50: 180-209 Animal Locomotion, ed. T. J. Pedley, pp.
Cone, C. D. Jr. 1962c. The theory of induced 365-404. London: Academic
lift and minimum induced drag of non­ Lissaman, P. B. S., Shollenberger, C. A. 1970.
planar lifting systems. NASA TR-J39 Formation flight of birds. Science 168
Finch, R. 1984. Wingtip design. Sport Aviat. (3934); 1003-5
 78 BUSHNELL & MOORE

Magnuson, J. 1978. Locomotion by Scom­ CR-3963

brid fishes: hydromechanics, morphology, Rayner, J. 1977 . The intermittent flight of
and behavior. Fish Physio/. 7: 239-310 birds. In Scale Effects in Animal Loco­
Magnuson, J. J., Weininger, D. 1978. Esti­ motion, ed. T. J. Pedley, pp. 437-43. Lon­
mation of minimum sustained speed and don: Academic
associated body drag of Scornbirds. In The Reif, W.-E. 1978. Protective and hydro­
Physiological Ecology of Tunas, ed. G. D. dynamic function of the dermal skeleton
Sharp, A. E. Dizon, pp. 293-311. New of Elasmobranchs. Neues Jahrb. Geol.
York: Academic Paliiontol. Abh. 157: 133-41
Mair, W. A. 1955. The distribution of pres­ Reif, W.-E. 1982. Morphogenesis and func­
sure on an aerofoil in a stream with a tion of the squamation in sharks. Neues
spanwise velocity gradient. Aeronaut. Q. Jahrb. Geol. Pallionlol. Abh. 164: 172-83
6: 1 12
- Reif, W.-E., Dinkelacker, A. 1982. Hydro­
McMasters, J. 1986. Reflections of a paleo­ dynamics of the squamation in fast-swim­
aerodynamicist. Perspect. Biul. Med. 29(3): ming sharks. Neues Jahrb. Geol. Palaontol.
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org

331-84 Abh. 164: 184-87

Meroney, R. N. 1968. Characteristics of Romanenko, Yeo V. 1981. Distribution of
wind and turbulence in and above model dynamic pressure on the body of actively
forests. J. App/. Meteorol. 7: 780-88 swimming dolphin. Dokl. Akad. Nauk
Mittal, A. K., Agarwal, S. K. 1977. Histo­ SSSR 21(2): 310-12
by University of Ottawa on 02/02/12. For personal use only.

chemistry of the unicellular glands in Romanenko, Yeo V., Yanov, V. G. 1973.

relation to their physiological significance Results of experiments investigating the
in thc epidermis of Monopterus-Cuchia hydrodynamics of dolphins. Bionika 1973
Synbranchiformer Pisces. J. Zoo/. 182(4): (7): 21-27
429-40 Rosen, M. W., Cornford, N. E. 1971. Fluid
Newman, B. G., Savage, S. B., Schouella, D. friction of fish slimes. Nature 234(5322):
1977. Model tests on a wing section of 49-51
an Aeschna dragonfly. In Scale Effects Spillman. J. J. 1978. The use of wing tip sails
Animal Locomotion, ed. T. J. Pedley, pp. to reduce vortex drag. Aeronaut. J. 82:
445-77. London: Academic 387-95
Oehme, V. H. 1971. On the geometrical twist Steer, D. 1963. Neuses vom Schuertfisch.
of the Avian wing. Bioi. Zentralbl. 90: Aquar. Terr. 8(12): 370-72
145-56 Uskova, Yeo T., Momot, L. N., Krisal'nyy,
Ovchinnikov, V. V. 1966. A morphological V. A. 1974. Effectiveness of skin secretions
and functional characteristic of the ros­ of some marine animals in lowering hydro­
trum of Ziphioidae. In Ecomorphological dynamic resistance. Bionika 1974(8): 148-
Research in NeklOnic Animals, pp. 42-52. 51
Kiev: Naukova Dumka van Dam, C. P. 1987. Efficiency charac­
Ovchinnikov, V. V. 1971. Swordfishes and teristics of crescent-shaped wings and cau­
billfishes in the Atlantic Ocean-ecology dal fins. Nature, 325: 435-37
and functional morphology. Rep., Atl. Sci. Vijgen, P. M. H. W., van Dam, C. P.,
Res. Inst. Fish. Oceanogr., Kalingrad, Holmes, B. J. 1989. Sheared wing-tip aero­
USSR. Trans!. from Russian by Isr. Pro­ dynamics: wind tunnel and computational
gram for Sci. Trans!., Jerusalem investigation. J. Airer. 26(3): 207-13
Pershin, S. V. 1978. The biohydrodynamic Vijgcn, P. M. H. W., van Dam, C. P.,
phenomenon of the swordfish as the Holmes, B. J., Howard, F. G. 1990. Wind
maximum case of fast hydrobionts. Bio­ tunnel investigations of wings with serrat­
nika 1978(12): 40-48 ed sharp trailing edges. P roe. Conf Low­
Pershin, S. V., Chernyshov, L. F., Kozlov, Reynolds-Number Aerodyn., Notre Dame,
L. F., Koval', A. P., Zayets, V. A. 1976. Ind. Berlin: Springer-Verlag. In press
Patterns in the integuments of fast-swim­ Vogel, S. 1981. Life in Moving Fluids, The
ming fishes. Bionika 1976(10): 3-21 Physical Biology of Flow. Boston: Willard
Pershin, S. V., Sokolov, A. S., Tomilin, A. Grant
G. 1979. Regulated hydroelastic effect in Walsh, M. J., Lindemann, A. M. 1984. Opti­
thc fins of thc largest and fastest dolphin, mization and application of riblets for tur­
the killer whale. Bionika 1979(13): 35-43 bulent drag reduction. AIAA Pap. No. 84-
Povkh, I. L., Stupin, A. B., Boyarkina, G. 0347
G. 1979. Hydrodynamic resistance of Walters, V. 1962. Body form and swimming
aqueous solutions of polymers and sur­ performance in the Scombroid fishes. Am.
face-active substances in rough tubes. Zool.2: 143-49
Inzh.-Fiz. Zh. 36(1): 16-19 Ward-Smith, A. J. 1984. Analysis of the
Raschi, W. G., Musick, J. A. 1986. Hydro­ aerodynamic performance of birds during
dynamic aspects of shark scales. NASA bounding flight. Math. Biosei. 68: 137-47
 DRAG REDUCTION IN NATURE 79
Webb, P. 1978. Hydrodynamics: Non­ Withers, P. c., Timko, P. 1977. The sig­
scombroid fish. Fish Physiol. 12: 189-237 nificance of gro und effect to the aero­
Weihs, D. 1973a. Hydromechanics of fish dynamic cost of flight and energetics of
schooling. Nature 241: 290-91 the black skimmer (Rhyncops Nigra) J.
Weihs, D. 1973b. Mechanically efficient Exp. Bio!. 70: 13-26
swimming techniques for fish with nega­ Zaccone, G. 1979. Histochemistry of ker­
tive buoyancy. J. Mar. Res. 31: 194--209 atinization and epithelial mucins in the
Weihs, D. 1974. Energetic advantages of skin of the marine Teleost Muraena Helena
burst swimming offish. (L.). Anguilliformer, Pisces. Cell. Molec.
215-29 BioI. 24: 37-50
Annu. Rev. Fluid Mech. 1991.23:65-79. Downloaded from www.annualreviews.org
by University of Ottawa on 02/02/12. For personal use only.