In addition to further decreasing the size of Perucetus, the below paper is the first extensive peer reviewed analysis of the body forms of giant extant whales (also see
https://www.biorxiv.org/content/10.1101/2022.08.28.505602v1 -- no marine artist is correctly illustrating blues -- or fins -- which is why all their pictures all differ in shape which should not be the case). We further reduced the size of Perucetus because a reviewer who knew basilosaur biology informed us that heavy boned examples have shorter trunks than the others, so we shortened the profile-skeletal and obtained an even lower mass more in accord with a basal whale.
A main point of the paper is the need for the vrtpaleo community to, when sufficient skeletal material is on hand and an atypical (usually really high) mass estimate are being arrived at, always without exception produce a profile-skeletal of the specimen/s (either analog or digital, the results are similar) and get the volume to generate a reliable mass with a modest +/- factor that can be used to properly test whatever other methods were used. If the volume of the model is very different then is time for a reconsideration and probable rejection of the results of the alternative scheme. Had that happened in this case then the nonsensical 85-340 tonne estimate range would never have seen publication and the public misinformed on the matter. After all, the profile-skeletal in the original Nature paper was in the area of 60 tonnes, so why did they not get a volume off of that and realize that their novel method needed a major redo?
Both Nature and Science have run the oversized Big papers that should have been modest items in the paleoliterature on marine amniotes of late -- both including those fancy digital projections of the supposed super productivity of the oceans of those ancient times that make for "major" paleoeco papers. The other being the Sander et al. 2021 paper than overestimated the bulk of basal ichthyosaurus by a factor of four or more (40-80 t) when skeletals show none known exceeded 20 or so. Again, why not just do the bloody profile-skeletals and abort the obvious overestimates?
Our paper emphasizes (less emphatically than it should because of reviewers) that the required skeletals are not, repeat not, interpretive art. They are technical drafts than any one can do if they take the time, they can even be done digitally. There is no need for artistic interpretation that impacts the final mass range. The basic methods were developed starting with myself in the 1980s in the vain hope it appears to preventing major future mass boo-boos (such as 80 t Brachiosaurus, and 35 t Brontosaurus). There should never again be such big overestimates. The way to prevent those is easy enough. There needs to be a basic requirement that all extraordinary mass claims, or use of a novel technique for estimation of such, require testing and confirmation via profile-skeletals when such is possible. Papers that do not do so should never be presented for review and if any are should be rejected. That the technique is "old fashioned" is an advantage, in that it is a tested method that produces reliable results against which alternatives and big claims need to be compared. No more excuses on this requirement what with the last two mass disasters.
As for citing my Princeton field guide for marine reptiles concerning its mass estimates, the new Sanchez-Fenollosa study on Dacentrurus casually cites my dinosaur guide, as have other peer-reviewed papers. And PDW still pops up. As do other nonreviewed academic books. And conference abstract citations are the norm. If a reviewer or editor balks at the citation of one of the field guides then point to its prior citation in the technical literature.
In the paper I handled the profile-skeletals, Asier the bone density work, so if you have questions about the latter ask him.
(PE does not give authors advance notice of publication date FYI)
Gregory S. Paul and Asier Larramendi (2025)
Further trimming down the marine heavyweights: Perucetus colossus did not come close to, much less exceed, the tonnage of blue whales, and the latter are not ultra-sized either.
Palaeontologia Electronica 28(1): a6
Extrapolating from skeletal/total mass ratios, the gigantic Paleogene whale Perucetus colossus has been estimated to have massed 85-341 tonnes, approaching and even exceeding the blue whale (Balaenoptera musculus). Such very large body mass variation lacks the precision needed for scientific rigor and is therefore not of technical value. Despite ample remains for a much more accurate volumetric model, none was produced for robust testing their procedure. A subsequent paper downscaled Perucetus to 60-70 tonnes as the most probable estimate. To better assess the question, we have produced multiview profile-skeletals for Perucetus and other large marine mammals in the most extensive such effort to date. These include the first accurate restorations of a number of large marine mammals. Perucetus is restored using the proportions of heavy boned pachycetine basilosaurs. Cross comparisons and attempts to illustrate the extinct whale's realistic volume leave no doubt that the Perucetus holotype did not reach the cetacean heavyweight category. A length of 15 to 16 m and a mass of 35-40 tonnes is more in line with its known anatomy. This result was affirmed by recalculations of skeletal/total mass relationships in large pachyosteosclerotic marine mammals, which suggest the method can produce useful estimates if conducted properly. Although their initial size expansion was remarkably rapid, basal cetaceans did not balloon to super whale dimensions just a few million years after the initial evolution of the fully marine forms, that did not occur until the late Neogene. The biggest whale of all time, the blue, is not likely to exceed ~30 meters and 200 tonnes. It is emphasized that anatomical knowledge translated into technical volumetric models remains the most critical means of restoring the mass of extinct organisms.
https://palaeo-electronica.org/content/2025/5431-trimming-down-perucetus