Rare Ivory-billed Woodpecker Images



My visits to Cornell’s Kroch Library, where the Rare and Manuscript Collections are housed, have been very productive. In addition to the last letter to Tanner pertaining to the Singer Tract ivorybills quoted at length here, I’ve come across several little known ivorybill images, some better quality reproductions of the plates in Tanner, and some additional hints about ivorybill foraging excavations that I’ll discuss in a future post. I suspect that all of the images below are actually stills from the 1935 film footage that has been lost save for a few minutes. To see it, go here and start at 14:00. To the best of my knowledge, these images have not previously been published as stills, and a couple of the frames may never have been publicly available.

The first image is similar to the one that appears on Page 82 0f  The Race to Save the Lord God Bird.  This is a sequence (that apparently has been lost) in which the birds are changing places on the nest. A third image that follows the first two appears on p. 120 of The Race . . . A colorized version, at once gorgeous and crude and sadly somewhat damaged, is also included here; it’s reproduced in black and white in Jackson (p. 27).

I think the bird in the remaining frames is the male. In the second frame, he may be engaging in the motion described by Tanner, “. . . jerking as though working food from the back of its mouth.” the next frame shows the him peering into the cavity. These two images are clips from the surviving footage. The final shot may have come from a lost piece of film, since a remaining clip, filmed from a similar angle doesn’t include it.

In addition to the images posted below, two figures in Tanner’s dissertation include unpublished photos from 1938 – one of a male at the nest cavity and the other of a juvenile peering out of it. Those images may also be included in a future post. All four pictures below were taken with my iPhone. I have a high resolution scan of the fourth on order, since it is one of the best representations of presumed ivorybill excavation available. Images are Courtesy of the Division of Rare and Manuscript Collections, Cornell University Library.



IMG_1119 (1)


IMG_1116 (1)

IMG_1120 (1)IMG_1122 (1)

Strips, Flakes, Chips, Chunks, and Slabs: Squirrels, Pileated Woodpeckers, and Ivorybills, Part 4

Careful examination of bark chips found in conjunction with extensive scaling is one of the key elements in our diagnostic gestalt, but “chips”, a term I’ve been using for years, is both inaccurate and too vague for what we believe is being left behind by Ivory-billed Woodpeckers and for differentiating it from the leavings of other animals. Tanner used “pieces” of bark, ranging “from the size of a “silver dollar to the size of “a man’s hand.” A caption from the National Geographic article on the 1935 Allen and Kellogg expedition that refers to “large chunks of bark”.  The existing images of these pieces of bark suggest that chunks is the better term.

It’s important to reiterate that this discussion applies only to live and freshly dead hardwoods. Pines slough bark quickly after death. The process is slower in hardwoods, but as decay progresses, the bark loosens considerably, with the rate of loosening depending on species and environmental conditions. Once the bark has loosened sufficiently, PIWOs can and do scale bark extensively, sometimes leaving behind large chips. In the images that follow (from Allen and Kellogg and Tanner), the bark chips ascribed to ivorybills appear to come from considerably longer dead trees than some of the examples we’ve found, but the images are informative.

Ivorybill Scaling Courtesy of the Division of Rare and Manuscript Collections, Cornell University Library

Ivorybill Scaling Courtesy of the Division of Rare and Manuscript Collections, Cornell University Library

The small tree shown above, identified as a “dead gum” by the 1935 expedition, appears to be a hackberry or sugarberry not a gum, and a fairly long dead one; the pieces of bark at the base resemble ones we found beneath hackberries or sugarberries in our old search area, some of which were considerably larger (the one below is the largest).


This colorized slide reveals more about the bark at the base of these pines than the black and white print in Tanner (Plate 9).

There’s also this example, (Plate 10 in Tanner), which appears to be in a considerably more advanced state of decay, and presumably looser, than much of the work we find most interesting. I suspect most of the grubs were placed on the chip for illustrative purposes; the caption “Beetle larvae from beneath bark of Nuttall’s oak” is ambiguous as to where the larvae, which appear to be small Cerambycidae, were actually found.

What I think is most salient in Tanner’s description of bark chips is shape not size. In this regard, it seems important to come up with a more specific set of terms to replace the commonly used “chips”. I’d suggest using chunks and slabs for suspected ivorybill work (although smaller pieces of bark may also be present). Pileated bark removal can involve chips, strips, or flakes, the last when they’re doing the layered scaling discussed here and here. I suspect that squirrels remove hardwood bark primarily or exclusively in strips, and of course, their bark removal on cypresses leaves shredded bark hanging from the trees.

Let’s take a closer look at the differences among pieces of bark we have reason to believe were left by squirrels, those we have reason to believe were left behind by Pileated Woodpeckers, and those we suspect were left behind by Ivory-billed Woodpeckers.

I collected a number of bark chips from the tree we know to have been scaled by a squirrel, and while these were removed before our camera trap revealed the source, there’s strong reason to think they too were left behind by squirrels.


Note the uniformly elongated shape and the ragged appearance at the tops and bottoms of these strips of bark. This is not typical of bark that we infer or know to have been removed by woodpeckers, and it’s consistent with chewing, not scaling. The presumed squirrel strips I collected had the following dimensions:






The downed sweet gum from which they had been removed was a fairly young tree, and the bark is much thinner than on more mature ones. These strips were approximately 1/8″ thick. While this is a very small sample, we suspect (along with Houston from IBWO.net) that approximately 3″ is the upper limit for width when a squirrel is doing the bark removal.

Our research and observations suggest that Pileated Woodpeckers have two strategies for removing tight bark; one involves pecking around the edges until they can gradually pry off small pieces, and the other involves scaling away strips, sometimes in layers. Their physical structure precludes them from doing the extensive, clean scaling of tight bark that Tanner associated with ivorybills.

We suspect that this collection of chips, from a honey locust near a known Pileated nest, reflects the range of what the species is capable of doing on a tight-barked hardwood (and honey locust bark is relatively thin). The upper limit appears to be hand size, with many-quarter sized or smaller.


The following are measurements of some fairly typical suspected Pileated strips from a sweet gum:





The strips shown below, suspected Pileated Woodpecker leavings from a high branch, are on the large end of the spectrum for this category of work. The Peterson Guide is 9.5″ x 6.5″. I can’t rule squirrel out completely for these.


Flakes resemble strips, but they are removed in layers, so that reaching the sapwood is a gradual process. Pileated scaling frequently has this appearance, something that seems frequently to be the case with congeners, including the larger-billed Black Woodpecker (Dryocopus martius).



The chunks and slabs we suspect to be ivorybill work are significantly larger and thicker than strips, flakes and chips, although strips and chips may be present in the mix at the base of suspected feeding trees. Chunks are usually more irregular and varied in size and shape, and both chunks and slabs sometimes have what appear to be strike marks from a broad bill.

I kept one of the chunks scaled from the hickory tree on the homepage, a fairly typical example. It is 8.5″x3.5″ and .375″ thick. (It has undoubtedly lost some of its thickness after drying for over two years.)


The sweet gum chunk with the apparent bill mark Frank is holding is 7.5″x3″ and .25″ thick. On mature, thicker barked trees most or all suspected ivorybill chunks, chips, and slabs will have been removed cleanly, all the way down to the sapwood.

Frank adds:

This particular bark “chunk” is intriguing on several levels. We have found that markings many describe as “bill marks” are really truncated galleries between the bark and sapwood. Marks made by woodpecker bills are distinctive, but somewhat subtle, and easily overlooked. This chunk actually has two interesting markings – markings that were left by the animal that removed the bark. The first is near the end of my left thumb – my right index finger is pointing toward it. It is about a quarter inch wide, a bit over a half inch long, and three sixteenths of an inch or so thick. The other is a “V” shaped “notch” at the end of the chunk, near the center of the photo. These places look as if they’ve been struck with a chisel – hard enough to rip the bark away from the sapwood/cambium. This suggests that, even though this bark was very tight, very few strikes were required to loosen and remove it. Granted that these marks are bill strikes, this suggests that the bird removing bark is indeed a powerful animal for its size. Back to Mark.

DSC00031The two preceding examples are on the smaller side for suspected ivorybill work; in the first, the density, tightness, and grain of hickory bark seem to be a limiting factor on size. Some of the larger examples are shown in the Bark Chip Gallery (as are several of the images shown above). A couple of additional examples of larger slabs are below. In the first, the oak was approximately 8 months dead (leaves attached), and the bark was still tight. (The fractured slab was damaged in transit.)




Rapid Evolution – I Think Not: More on Game Cams and Challenges

Cyberthrush’s blog has been an important resource for people with an interest in the IBWO for nearly a decade. I’ve agreed with him at times and have taken strong issue with him at others. His latest post falls into the latter category and strikes me as being somewhat irresponsible; it’s worthy of submission to The Annals of Armchair Quarterbacking. I responded in the comments but have some additional observations that seem better suited to posting here.

The inspiration for Cyberthrush’s post is an image someone sent him of suspected bark scaling low on a tree. The image is not shown in the post, but I suspect I’ve seen it and, if so, am sure it was the work of a mammal. In any case, he hypothesizes that the IBWO has rapidly evolved to live almost exclusively in the canopy. He uses the specialization of certain species – seabirds, Chimney Swifts, and warblers – to draw parallels, ignoring entirely that none of these species evolved nearly so rapidly. The only basis for this theory is the failure to obtain clear photographs.

He digresses to dismiss virtually all the photographs of feeding sign, cavities, and putative IBWOs and virtually every recording of putative kents or DKs he’s seen over the past 9 years. He writes: “I don’t want to discourage people from sending such things along, but I do want folks to know that the chance of getting a positive response from me is extremely slim . . .”

He goes on:

“I’m in the camp that believes any remaining IBWOs have evolved heightened wariness and caution, and as such DO NOT spend ANY extensive time at ground level, where they would be far more vulnerable. In fact, I believe they are now almost exclusively residents of the upper canopies, other than when flying from point A to point B and requiring a clear pathway. While they might land momentarily lower on a tree, they probably spend most of their daily lives minimally 35+ ft. high up (maybe 50+ ft.) on tree trunks/branches and inside cavities, well above levels frequented by Pileateds and other woodpecker cousins (and generally out of easy sight-line for searchers). It’s not clear to me how many of the remote automatic camera traps were ever set that high (though it’s clear several were not). Like Swifts in the air and Albatrosses at sea, I think Ivory-bills may spend most of their lives solely in the canopies… if you send me a photo of foraging sign or a cavity or a fuzzy bird lower than ~35 ft. high, I probably won’t take it too seriously (even though there are historical records of such cases), unless there are overriding additional details to catch my attention. If Ivory-bills currently lived and foraged below 35 ft. to any significant extent I believe we’d have the definitive evidence we need by now (well before now!); only perhaps as a denizen of the upper reaches might they be able to carry on successfully, while also evading encounters and detection to the degree they have . . .”

I don’t know who else is in his camp. Some have certainly argued that hunting pressure led the IBWO to become hyper-wary, and I can accept that hypothesis to some degree, although I think scarcity and the nature of the habitat have far more to do with the lack of conclusive documentation. The suggestion that the species evolved to exist almost exclusively in the canopy, when it was not infrequently seen feeding near the ground, is nowhere near as plausible, and it’s invented out of whole cloth.

As I wrote in the comments:

Your exclusion of any work below 35′ is arbitrary and wrongheaded, IMO. Forget Allen and Kellogg, Tanner even observed that they’re “not averse” to coming near the ground. Do you really think such a significant behavioral change evolved post-Tanner? An animal seeking food is going to go where the food is, and as it happens, the largest quantity of food is often to be found in the boles. I agree that cavities will be high up and also that work that’s exclusively near the base of the tree is suspect, but when the work goes higher than about 10 feet you can pretty well rule out a mammalian source (unless giraffes are around or there’s some heretofore undiscovered porcupine species in the swamps of Louisiana and Florida). When it’s very extensive, you can be pretty confident that it’s either PIWO or IBWO, and as I’ve argued at length on the Project Coyote blog, I think there is a limited range of work that is identifiable as IBWO.

This really is giraffe foraging sign, photographed in Louisiana . . . Long story.

This really is giraffe foraging sign, photographed in Louisiana . . . Long story.

One avian example of rapid evolution (though not nearly so rapid as Cyberthrush’s IBWO) is the Cahow, but that adaptation involved a change in nesting behavior due to a lack of adequate sites, something that’s far more akin to the kinds of survival strategies discussed in this recent post. There’s simply no reason to think that IBWOs could have suddenly evolved to live almost solely in the canopy. Not when Allen and Kellogg even observed one feeding on the ground; not when Kuhn observed one feeding a foot up; not when one of the clearest photos of bark scaling shows the work going to the base of the tree (the bark chips in this image are worth a close look, and they resemble many that we have found.)

With regard to aiming cameras at the canopy, I discussed the limitations of game cam technology in this post. The challenges increase dramatically when the cameras are aimed even higher, as is illustrated in this photo of a bird about 75’ up in a tree. There’s simply no way to get a conclusive ID. I think it’s suggestive of a female IBWO and that red would be visible in the crest if it were a PIWO, although it’s clearly nowhere near as intriguing as our other two Reconyx photos. Frank disagrees and thinks it could be anything.

IMG_9463 Red Box

Reconyx image and detail. Old Project Coyote search area, March 2009

Reconyx image and detail. Old Project Coyote search area, March 2010

Either way, aiming these very poor resolution, short focal length cameras high complicates matters exponentially. The likelihood of getting a clear image at such a great distance and with backlighting is infinitesimal. Aiming high is likely to be a wasted effort. One final note about game cameras, Cyberthrush attaches a great deal more significance to the lack of photos than I do. As far as I’m aware game cameras were only extensively deployed in Arkansas, South Florida, and the Choctawhatchee. They’ll only work if they’re deployed at the right time and in the right place, something that’s much easier said than done. As I’ve mentioned previously, virtually all the scaling we’ve found seems to have been done in a very limited number of visits to the tree, perhaps just one, and in the very rare instances where we’ve found evidence of a return of whatever removed the bark, there has been a gap of several months between visits.

The key to all of this is going to be finding a roost or nest site (unless someone gets extremely lucky.) I’m convinced that feeding sign is what can point searchers in that direction; it’s what led Kuhn and the Allen and Kellogg expedition to birds in the 1930s. That’s the reason for my focus on it. Cyberthrush does a disservice to all searchers by suggesting, based on unfounded speculation, that virtually anything found below 35’ feet is not worthy of attention.

Heavily scaled young oak with suspected IBWO work extending from the base to well up on the trunk.

Heavily scaled young sweet gum or oak with suspected IBWO work extending from the base to well up on the trunk. Note the large bark chips at the base.