Bark Scaling and Woodpecker Anatomy

I wrote the article that follows (below the images) in April 2010. It reflects my belief that it is possible to distinguish a certain narrow category of feeding sign as being the work of Ivory-billed Woodpeckers. Since that time, I have spent many days in the field in potential IBWO habitat and also examining feeding sign in areas well-outside the historic range. My basic view has not changed.

My criteria are:

1) The trees must be hardwoods, alive or very recently dead.

2) The scaling must be ‘clean’ – bark stripped to the cambium with little or no damage to the sapwood.

3) The bark must be generally tight (this is not always easy to ascertain.)

4) The scaling must be extensive.

5) The chips must be large with at least some 3″x6″ or more.

6) The edges of the scaled area must appear to be chiseled, and there should be little or no sign that bark has been removed in layers, something that is typical of Pileated Woodpecker scaling.

I have only seen concentrations of this type of work in two locations. Some images from both locations are attached to illustrate. Scroll down for the original article.


Suspected IBWO scaling on two sweet gums.


Detail of scaled sweet gum.

This is a more ambiguous image, since it is possible that the bark on this recently dead hardwood was sloughing naturally ; however, this came from an area with a concentration of scaling, and the tree (believed to be a sweet gum) was very freshly dead.


Scaling on a live willow oak.young oak

Detail of scaling on a freshly dead small sweet gum or oak showing clean edges and evidence of lateral strikes.


These bark chips are an extreme example, among the largest I’ve ever found (boot is a size 13.) These were fresh when found, removed from a living but diseased oak. They weighed approximately 1 pound each when collected.

This article was inspired by my longstanding interest in identifying IBWO foraging sign (a problem I believe I’ve solved, at least in part,) lengthy discussions with Frank Wiley and several others, field observations, and most recently a lot of research. I’m grateful to Bret Tobalske for directing me to a 1965 article by Lowell Spring entitled “Climbing and Pecking Modifications in Some North American Woodpeckers,” which is available for purchase here:

Spring references an earlier article by Walter Bock and Waldron DeWitt Miller, “The Scansorial Foot of the Woodpeckers, With Comments on the Evolution of Perching and Climbing Feet in Birds”

The article is Bock’s elaboration on and revision of an unpublished paper on the anatomy of the Ivory-billed Woodpecker foot written by Miller in the 1910s. Scansorial means adapted to or specialized for climbing, and the article posits that the feet of Campephilus woodpeckers are specialized in this manner, representing a more advanced evolutionary form.

“The Scansorial Foot. . .” does more than just describe the anatomy of IBWO feet; it also compares anatomical features among various woodpecker species, including the Pileated Woodpecker (Dryocopus pileatus). Bock posits that Dryocopus woodpeckers evolved from ground dwellers and that the more specialized Campephilus woodpeckers may be descendants of Dryocopus. More recent DNA-based taxonomic research has shown that the genera are not related in this way:

Nevertheless, the anatomical differences that Bock highlights are significant.

To summarize, Bock suggests that foot structure, climbing ability, and foraging behavior are closely linked in woodpeckers, although, for the purposes of this analysis, climbing ability is not particularly relevant. Pileated Woodpecker foot structure is somewhat similar to that of the Flicker, which feeds almost exclusively on the ground, doing little or no excavation, except of roosts and nests. Similarities in basic foot structure notwithstanding, there can be no doubt that pileateds are strong climbers and are well adapted to direct excavation through bark and underlying wood. By contrast, the available anatomical data suggest that Campephilus woodpeckers have a morphology that favors the technique of removing relatively large pieces of bark and also of delivering more powerful blows when excavating harder wood.

In “Climbing and Pecking Adaptations in Some North American Woodpeckers,” Spring states:

“The stance employed by Campephilus is most advantageous for peeling bark. With the feet spread far apart in an anterior position, the widest possible base and greatest stability is provided for the delivery of sidewise blows. What effect this stance has on climbing ability is not clear.”

Virginia Kirby’s 1980 article, “An Adaptive Modification in the Ribs of Woodpeckers and Piculets (Picidae)” is also significant:

Kirby examined 61 different species of Picidae, measured the rib width to femur length ratios and compared them to foraging methods. Her data on foraging methods are at best incomplete and out of date, but her hypothesis that the power with which woodpeckers can deliver blows is shaped by anatomy seems sound. In her analysis, PIWOs ranked 54 and IBWOs ranked 59. According to Kirby:

“Burt (1930) showed that Picidae that excavate for burrowing prey as opposed to those that take primarily superficial prey have skulls that are highly modified for pounding. He found a continuum in skull structure from the least to the most specialized for pounding: Colapres auratus, Melanerpes lewis, M. erythrocephalus, M. formicivorus, M. carolinus, Dryocopus pileatus, Sphyrapicus varius, Picoides villosus, P. arcticus, and P. tridactylus (the last two species showing the same amount of modification) that corresponds with an increase in pounding in the species’ foraging behavior. The last four species had noticeably more specialized skulls than the others.”

Burt’s study did not include C. Principalis, but P. villosus ranked 56 in Kirby’s study, so it seems reasonable to infer from Burt that the skull of C. Principalis should show a level of specialization close or equal to the last two species in Burt’s sequence, which rank 60 and 61 in Kirby’s.

Significantly for our purposes, Kirby also characterizes Spring’s results as follows:

“According to Spring (1965), S. varius delivers blows by using only its neck. Picoides villosus and P. arcticus deliver more powerful blows by employing a delivery stance that creates a greater distance between the body and the tree trunk for developing acceleration and by superimposing “lower body movements upon neck action” (Spring 1965: 467). In effect then, these woodpeckers accelerate greater mass through longer distances to produce more forceful impacts.”

Note that the skull structure of S. varius is better suited to pounding that of D. pileatus.

So, what are the implications of this information?

First, David Sibley’s inaccurate depiction of IBWO feet is not a trivial error:

Sibley depicts the IBWO foot as zygodactylous, that is having the toes arranged in opposite pairs. (Thanks to Bill Benish for pointing out the mistake.) In fact, according to Bock, the IBWO’s foot:

” . . .presents a striking contrast to the foot of the flicker or of the pileated woodpecker, yet the morphological changes are relatively minor. The hind toes, both the first and fourth, have rotated to the outer side of the foot until in the extreme condition all four toes point forward-a “pamprodactyl” foot.”

Conversely, the foot of the PIWO remains zygodactylous, albeit with evolutionary adaptations that make it more suitable for climbing: larger and stronger claws than a Flicker’s and the capacity to rotate the fourth toe to the lateral side of the foot.

More importantly, the anatomical data lead me to suspect that structure imposes an upward limit on the hardness of the wood that pileateds can excavate and on their ability to scale bark. This goes a long way toward explaining why PIWOs are often observed feeding on the ground and why they prefer decaying wood. It also may point toward ways to distinguish IBWO sign from PIWO sign and explain why we have observed no typical pileated excavation on live or recently dead persimmons or honey locusts in our search area. Both these species have very hard wood, and pileateds seem to avoid them until they are in an advanced state of decay.

Frank Wiley pointed out that tail structure may also be important in this regard, and I found a small item in a paper that supports his insight:

While there are many problems with this paper, Table 2 is important because it suggests that Campephilus and Dryocopus tails have a different structure.

On a superficial level, these anatomical details explain why Campephilus woodpeckers display a “reared-back” posture and possibly why their movements on trees have a less graceful, jerkier appearance, as David Martin pointed out in a recent personal communication.

Appearances aside, the anatomy of Campephilus woodpeckers enables them to place their legs both farther apart and higher in relation to their bodies when foraging. This provides more stability and hence more power when they scale bark using lateral blows. It also allows them to deliver more forceful direct blows. As Frank Wiley pointed out, the tail structure is an integral part as well, since it creates a tripod. I’ve looked at all the videos of foraging Campephilus woodpeckers that are readily available online and have compared them with footage of foraging PIWOs. The videos show that Campephilus woodpeckers have a greater range of motion and presumably the ability to deliver more forceful blows. They frequently rear back and use their entire bodies, while the lower body of the PIWO generally remains close to the tree and the blows are delivered with the upper body only. These two videos are illustrative:

Note how the lower leg and foot morphology enables the Magellanic Woodpecker to gain greater leverage when excavating and provides stability for the delivery of lateral blows.

Although the pileated footage does show a bird doing some scaling of bark, the tree appears to be a cypress, the bark of which is easy to scale, and regardless of species, the bark appears to be loose. It seems likely that the following set of images more accurately reflects the way in which Pileated Woodpeckers remove tight bark from hardwoods:

In image 3, the bird is foraging near a piece of recently scaled bark that is approximately the size of a quarter. Image 4, shows scaled areas that also suggest the bark has been removed in small pieces. In images 5-8, the Pileated Woodpecker removes a somewhat larger piece of bark that appears to have already been loosened; note the fissures on the lower left and upper left. This removal of small pieces of tight bark from a hardwood matches Frank Wiley’s limited observations of Pileated Woodpecker scaling in our search area.

Summary: there are significant anatomical differences between Campephilus woodpeckers and Dryocopus woodpeckers, and these differences may provide important insights into how to distinguish IBWO foraging sign from PIWO work. While it will be impossible to make these distinctions with certainty until conclusive documentation of the IBWO’s survival is obtained, it is possible for searchers to make inferences that should be helpful in identifying likely IBWO work.

About Project Coyote

In 2009, we  learned of a private landowner in East-central Louisiana who claimed to have Ivory-billed Woodpeckers on his property. This individual seemed to be very credible, correctly pointing out several inaccuracies in sketches published in the Louisiana Hunting Guide. Over the next two years, there were a number of  possible sightings and auditory contacts. Frank Wiley, Mark Michaels, and other Project Coyote volunteers gathered evidence – camera trap photosgraphs, recordings of  suggestive calls and double knocks, images of extensive bark scaling on living and recently dead trees in the vicinity. This material was originally posted on a website that is no longer operational. We have received a number of requests to revive the site and have decided to do so here. Pages that included images will be available in PDF format.

In 2011, a large parcel of adjacent forest was logged, and while we believe that ivorybills are still present in the vicinity of the original Project Coyote search area, the birds do not seem to be frequenting our old hot zone. For this reason, we have shifted our focus to other parts of the state. We remain optimistic that the species is present in multiple Louisiana locations and that conclusive documentation will eventually be obtained, but this may well take years.

Mark Michaels

Frank Wiley

April 21, 2013

Cavities and Feeding Sign Page

See attached PDF



Camera Trap Photos PDF

The attached PDF includes the text and images from the Camera Trap page on the original Project Coyote site.


Update/Correction, November 9, 2018:

Thanks to Guy Luneau for alerting me to an error in the attachment that merits a correction, though it does not materially affect the analysis. Guy pointed out:

In Frank Wiley’s analysis of the size comparison of the ivorybill and the pileated . . . [t]he ivorybill isn’t “29% larger” as he calculated and published, but 40% larger.

Someone else may have already commented on that bad math, but I have not yet uncovered it on your website.  Thus, I will explain.  And pardon my lateness if this has already been laid to rest.

The baseline measurement is the pileated at 100 pixels body length.  The ivorybill is 140 pixels.  Thus, the proper math for saying, “The ivorybill is x% longer in body length than the pileated” is 140/100 = 1.40.  Proper statements then become, “The ivorybill is a factor of 1.4 longer in body length than the pileated,” or “The ivorybill is 40% longer in body length than the pileated.”  (Two ways of saying the same thing.)  Frank took the reverse approach with his math, which is fine(!), but then he put it into words wrong.  He would have been right if he had said, “The pileated is 29% shorter in body length than the ivorybill,” or “The pileated’s body length is 71% that of the ivorybill.”  (Again two ways of saying the same thing.  And also the same thing as saying the flip side using the 1.4 factor or 40% longer.)




Mark Michaels:    MarkProjectCoyoteIBWO @