The Doyle footage
According to the online Thylacine Museum, "in South Australia in 1973, Liz and Gary Doyle briefly captured a rather fascinating image on motion film. It shows a yellow brown, dog-shaped animal running in a manner which appears to be suggestive of the thylacine" (Campbell, date unknown).
As mentioned in a previous article, author Robert Paddle gives limited but significant evidence that the thylacine existed in South Australia in the early 1800s despite popular perception.
However, to date, I haven't found any critical analysis of the Doyle footage beyond that presented by Campbell.
In this article we'll make a detailed comparison between the animal in Doyles' footage and confirmed footage of a running greyhound.
Before we begin - is anyone aware of any other analysis of this footage? And in particular, was it ever reviewed by a first-hand witness of pre-extinction thylacines? Please let me know in the comments if you have an answer.
The method and the madness
For the purpose of this comparison, I took 69 still frames from the 1973 footage.
Next, I found that the GAGAH website (Giving a Greyhound a Home) had lengthy footage of a greyhound running on a beach and took a further 32 still frames.
Let me state up-front that I have no experience with evaluating body movement. If you do - or if you know someone - I welcome your feedback!
Due to the limitations of the MPEG source file for the 1973 Doyle footage, there were in fact very few usable frames.
The MPEG file format was never designed to produce screen shots. Rather, it acheives small file sizes, and hence fast data transfer, by compressing the footage. To do this, it generally only stores a complete frame every 12 frames or so. For the intervening frames, it only stores as much information as required to "overlay" the last full frame such that you get an effect as if a full frame was rendered.
It should be noted that due to the MPEG file format, the frames I have grabbed may (and almost certainly will) in fact be different to those originally captured on film. Unless the original footage can be sourced, we will have to make do with this limitation by making assessments on the general form of the animal, and not specific measurements.
Overview of the 1973 Doyle footage (click to enlarge)
Firstly, let's just take a quick overview of the main postures captured reasonably clearly in the MPEG format of the 1973 Doyle footage...
Following is a list of notable features which are exhibited, and the frame numbers which exhibit them.
Some features were selected because they may support the veiw that this animal is a thylacine (thin tail), although alternative explanations for the observed feature may well be plausible.
Other features were selected simply because they were (relatively) clearly visible - no matter how many animal species may fit the observation.
Some features appear differently in different frames (eg a heavy forepaw in some frames, and a thinner forepaw in others) and these are likely to be explained by the limitations of the MPEG file format (see notes above).
|A thin tail :||2, 5, 7, 8, 10, 11, 12, 17, 20, 22|
|Heavy forepaw :||2, 14, 15, 17|
|Position of forepaw relative to body :||2|
|Kangaroo-like posture :||5, 8, 12, 22, 23|
|Shape of hind feet :||5, 8, 12, 17|
|Head shape :||1, 2, 3, 5, 7, 8, 9, 14|
|Ear position :||9, 24|
|Bent forepaw :||24|
Note that frame 08 appears to be a duplicate of frame 05.
The most thylacine-like features in the opinion of this author, are the thin tail and the kangaroo-like posture which occurs at some points mid-stride. These will be addressed separately below.
Although I have noted the "heavy forepaw" as a notable feature, you can notice that in those frames, only one forepaw is ever visible. In other words, the appearance of a heavy-set forepaw may be due to both paws being side-by-side and appearing as a single forepaw.
Frame 2 has been noted for the position of the forepaw with respect to the body. As we will see, the 1930s footage demonstrates this same appearance when the forepaws are angled backwards. Clearly, in other frames, the effect is opposite when the forepaw is reaching forwards. That is - it does not appear to be positioned as far back along the body when the paw is reaching forwards.
We need to ask -
- is the observed range of movement characteristic of normal thylacine movement?
- What other animals, if any, might exhibit all the same features?
Kangaroo-like posture (click to enlarge)
Let's have a look at some of the 1930s footage where a thylacine rears up on its hind legs. The first thing to note is that this animal is rearing up on the spot, whereas in the 1973 footage the animal is running as fast as it can.
Note the kangaroo-like posture, the thin tail, the apparently thick forepaw (although the impression is given by both paws being very near to each other - as commented on, above) and the distance from the head at which the forepaw appears to leave the body (in this specific posture where the forepaws are near to the hind feet).
This image highlights many of the features noted in the 1973 footage, namely: the shape of the hind feet, thin tail, long neck, head shape and ear position.
Note also the general curve of the spine and the way in which the thin tail joins the body at the broad base. This broad base seems to be formed by the continuation of the curve of the leg meeting the end of the curve of the spine and is almost triangular in appearance.
This image of a thylacine walking shows various joints in both fore and hind feet as well as general head shape and ear position.
The same animal has been shown in all three images.
Overview of a greyhound running (click to enlarge)
Next we turn out attention to a running greyhound, before we move on to a side-by-side comparison between each of the three sources.
The first notable feature is that the overall body shape of the greyhound assumes a much more curved posture than that of the 1973 animal (see frames 1, 4, 6, 8, 9, 12, 13, 16, 19).
Secondly, the greyhound's tail seems much more flexible (see frames 1, 8, 9, 12, 14, 15, 16, 17, 18).
In contrast, the animal in the 1973 footage exhibits a straight tail, angled downwards in every frame where the tail is visible except one (see frames 1, 2, 5, 7, 8, 9, 10, 11, 12, 13, 16, 17, 18, , 20, 22 from the 1973 footage, and frame 4 for a horizontal position).
Robert Paddle notes (2000) that "common in the literature is the idea that the thylacine's tail was a fixed, straight, relatively immovable extension of the backbone ... Generations of zoologists and comparative psychologists, with fixed, primitive tails in mind, have looked at photographs of thylacines and seen just what they wanted to see ... in no sense was the positioning of the tail fixed."
Whilst Paddle goes on to cite numerous examples of tail flexibility in the thylacine, it is this author's contention that in general, the thylacine carried its tail in a manner very similar to that observed in the 1973 footage. This reasoning is based on observation of those thylacines depicted in the seven films presented by the online Thylacine Museum. Whilst it is true that even in that footage the thylacines did lift their tails at times, it would seem that for the majority of the time, a standing, relaxed thylacine would have its tail pointed downwards in a straight, or very slightly curved line.
Of course, part of the problem here is that we have no known footage of a running thylacine, and this author has not located any textual description of the same.
The third feature to note is that in several of the greyhound frames, the animal has only its hind feet on the ground and is in the process of launching itself forwards (see frames 3, 5, 7, 10, 19, 20). In these frames, the back of the head, the neck and the spine all form a singular, smooth curve. For want of a better description, the "direction" of the curve is comparable to a rainbow.
However the animal from the 1973 footage seems to hold its head in a much more upright position, producing an "upside down rainbow" curve (see frames 2, 5, 7, 12, 17 in the 1973 footage)
We can safely conclude that the 1973 animal is not a greyhound!
However, we cannot automatically conclude the animal is a thylacine either!
Let's look at the 1973 footage in direct comparison with the 1930s footage of a thylacine. Despite our note at the start of this analysis that the MPEG file format should prohibit us from drawing any conclusions by taking measurements, nevertheless we'll look at general body form and body part proportions.
Forepaw and tail comparison - 1973 Doyle footage and 1930s thylacine
For this comparison I have selected frame 2 from the 1973 Doyle footage for several reasons:
- it shows the forepaws positioned together, giving the impression of a single, heavy-set forepaw
- The forepaws are angled backwards towards the hind feet
- The tail, rump and head are all visible
The 1930s frame shows a thylacine in a similar posture.
Note that the 1930s thylacine was about to rear up, as can be seen from the left image. I have rotated that image so as to align the body position with the 1973 frame as much as possible.
Note also that in the 1930s image, the rear part of the rump, and the tail are both in shadow.
In comparison, the tail of the 1973 image appears to be much more abruptly connected to the body than in the 1930s thylacine. The overall head shape, ear position and ear size seem to match between both images. The general thickness of the trunk of each animal, in comparison to its length, seems to maintain the same proportions between animals. The general curve of the spine is reasonably close although, as noted, the 1930s thylacine is lifting its head (resulting in its muzzle being much further from its chest).
The general proportions and positioning of the paired forepaws seem also consistent between images. These will be examined in more detail, below.
Body proportion comparison - 1973 Doyle footage and 1930s thylacine
In this composite image, I have highlighted (in yellow) the outline of the spine, tail and underside of the rump of each animal purely for interest's sake.
The three lines were drawn as follows:
Red = the distance from the throat to the point at which the forepaws leave the trunk of the body,
Blue = the length of the forepaws,
Green = the distance from the throat to the base of the tail.
Next, I took measurements of each line and calculated the following ratios:
The length of the forepaw in proportion to the chest (blue/red) was 0.777 for the 1973 frame and 0.806 for the 1930s frame. The ratio was 3.72% larger for the 1930s frame.
The length of the forepaw in proportion to the trunk of the body (blue/green) was 0.371 for the 1973 frame and 0.341 for the 1930s frame. The ratio was 8.85% larger for the 1973 frame.
The length of the chest in proportion to the trunk of the body (red/green) was 0.478 for the 1973 frame and 0.423 for the 1930s frame. The ratio was 12.9% larger for the 1973 frame.
Whilst all this may seem well calculated, in reality this is presented only for interest's sake as we are operating under conditions far less than ideal considering the MPEG file format and low resolution of the 1973 frame, and the fact the animals were engaged in two completely different actions (running and rearing up) and as a result are holding slightly different postures.
Head and ears comparison - 1973 Doyle footage and 1930s thylacine
The general appearance of the head and ears in the 1973 footage is perhaps best captured by this frame. The 1930s thylacine frame has been rotated to best align the spine and head with the 1973 frame.
The animal in the 1973 frame is more perpendicular to the camera than the thylacine in the 1930s frame.
Accounting for the thylacine not being perpendicular to the camera, the overall head shape, ear size, ear positioning and ear spacing seem consistent between frames.
In addition, the overall "weight" of the trunk of the body (length and thickness - especially at the chest and rump) seem consistent between frames.
Likewise the curve in the spine seems to be a close match.
Fourthly, each frame provides some insight into the joints in each animal's legs. The thylacine in the bottom part of the image, clearly shows the two joints in its hind feet, and the joint at its wrist in the forepaw.
Similarly, the joint in the wrist of the 1973 animal is visible on the forward forepaw.
Head and ears comparison, enlargement - 1973 Doyle footage and 1930s thylacine (click to enlarge)
This image is a copy of the previous one and has been magnified x2. Click to enlarge.
Three way comparison - 1973 Doyle footage and 1930s thylacine (click to enlarge)
To conclude the analysis, the following composite image has been created in order to demonstrate the few remaining physiological comparisons between all three sources.
Rear feet planted comparison - 1973 Doyle footage, 1930s thylacine and a greyhound (click to enlarge)
Note the shape of the hind foot in the 1973 frame and compare it to the 1930s thylacine, as well as to the first greyhound frame.
Note also the shape of the rump at the base of the tail.
As noted above, the 1973 footage is available at the online Thylacine Museum. In summarising his own conclusion on the film, Campbell says that "despite performing a comprehensive analysis of the film frame-by-frame, I myself remain unable to form any solid opinion of it as of yet. The image quality is simply too poor, and the motion too unsteady, to perceive any distinguishing physical features that would conclusively identify the animal as a thylacine. However, the manner in which the animal runs is indeed quite interesting, and in several frames, stripes appear to be vaguely visible on the hindquarters."
Campbell goes on to quote the author of the online publication "Magnificent Survivor - Continued Existence of the Tasmanian Tiger" as saying "the physical dimensions of the animal shown in the Doyles' footage are not consistent with a fox or dog - particularly the back legs, which look identical to those of a thylacine. The animal also appears to be bigger than a fox, and the tail seems to be longer and certainly straighter than that of a fox or dog. When running, most of the animal's driving force comes from the back legs, and some of the stills show it in a stance like that of the kangaroo. It is fairly simple to identify a hoaxed thylacine image, but I can't see anything in this film to suggest it is a hoax. The footage seems convincing to me - consistent with the running juvenile thylacine I saw in 2002, and there was no uncertainty in that case. The juvenile that I witnessed also ran primarily using the power of the back legs, and appeared to grab and pull at the ground with its front feet"
With regard to Campbell's conclusion, I agree that the image quality is too poor to identify distinguishing features in all but a very few frames. The general information we can obtain from individual stills is the overall body shape and dimensions. Many frames appeared to be distorted by the MPEG file format to the point of being unusable, however those frames I have selected for this analysis seemed to render usable information despite the limitations.
Initially I also felt that some frames hinted at a striped pattern on the animal's hindquarters, but I think in all fairness I have to discard such a conclusion on the basis of the overall poor image quality. There just is not enough detail to make that conclusion.
With regard to the Magnificent Survivor author (Tigerman)'s conclusions, I agree that the back legs do look consistent with a thylacine, but I disagree that they are not consistent with a dog. In the final image of the above analysis it can be clearly seen that the joints on the hind leg of a greyhound are proportionate with both the 1973 footage and the 1930s thylacine. I would be very interested to hear a professional opinion on the morphology of the hind legs of thylacines with respect to dogs.
This analysis has heavily rested on still frames, and Tigerman refers to the manner in which the animal generates momentum as evidenced in the animated frames. Although this author does not claim to have seen a living thylacine, I agree with Tigerman's description of how the 1973 animal generates momentum; that is, primarily by driving the hind feet whilst appearing to grab and pull at the ground with the front feet.
Separate to the above conclusions, this analysis has demonstrated that the overall body proportions of the 1973 animal - including the dimensions of the trunk of the body, the head shape, ear shape, ear position, tail thickness, base of the rump, tail posture, forepaw position, forepaw proportion and curviture of the spine - are consistent with that of a thylacine.
Whilst I too cannot conclusively state that the animal can only be a thylacine, I feel that all observable features of the 1973 animal remain consistent with the morphology of a thylacine.
However, whether I too am seeing "just what [I] want to see" (Paddle, 2000) - is also open to debate!
Campbell, C. (undated) 'The natural history of thylacinus cynocephalus: alleged mainland thylacine sightings', accessed 10/9/2006
Paddle, R. (2000) 'The Last Tasmanian Tiger: The History and Extinction of the Thylacine', Cambride University Press, 22-24, 50.