Dunkleosteus terrelli (fossil fish) by James St. John - CC 2.0

Diagram featuring the multiple size estimations of Dunkleosteus terrelli - Public Domain

In my previous post, I mentioned that one method of calculating a figure's scale is to divide the length of the full-size animal by the length of the miniature, with the result being its scale. The Dunkleosteus (2007) figure from Safari Ltd (whose length is easily measured), would seem to be a good candidate for this straightforward approach. This method, however, can be problematic for a variety of reasons, and the Dunkleosteus figure serves as a perfect example of these challenges.

The Dunkleosteus was an ancient armored fish known as a placoderm, characterized by heavy bony plates around its head and shoulder area. These bony plates preserved well, yielding numerous fossil remains of the fish, which have been discovered over the years. Unfortunately, the rest of the animal's skeletal cartilage did not fare as well, leaving little information about the Dunkleosteus' body from its shoulders aft to its tail. Due to this lack of fossil evidence, paleontologists must rely on comparative anatomy of known fishes to make educated guesses about the appearance of the Dunkleosteus, as well as to estimate its overall length. This estimation has varied widely over the last 20 years. According to Wikipedia, the most recent estimate places it between 4.1 and 5 meters, while previous estimates have reached as long as 10 meters.

Using the longest estimate of ten meters for your calculations produces a resulting scale of 1:55 (1000 cm / 18 cm = 1:55.5 ), while using the shortest estimated length of 4.1 meters produces a resulting scale of 1:23 (410 cm / 18 cm = 22.7). While on the surface, each of these may seem reasonable, under closer scrutiny, you will realize both calculations produce skull and shoulder dimensions substantially larger than the largest fossil specimens yet discovered. Because of this, in the case of the Dunkleosteus (2007) from Safari Ltd., it is more appropriate to use known measurements of the skull and shoulder area (in this case of the well documented CMNH 5768) to calculate its scale, rather than the estimated full body lengths achieved through comparative analysis and extrapolation.