Wednesday, December 20, 2023

Obscure controversies in Cenozoic marine vertebrate paleontology 4: when did baleen whales become gigantic?

For earlier entries in this series, check out these links:

Obscure controversies in Cenozoic marine vertebrate paleontology 1: taxonomic feuding over the basilosaurid whales Zygorhiza, Pontogeneus, and Cynthiacetus

Obscure controversies in Cenozoic marine vertebrate paleontology 2: the whale jaw that didn't disprove evolution

Obscure controversies in Cenozoic marine vertebrate paleontology 3: Was there an early Miocene extinction in pelagic sharks?

Introduction

Baleen whales, or mysticetes, are the largest animals of all time (sorry dino weenies) – the blue whale (Balaenoptera musculus) is commonly cited in books, magazines, and TV programs as attaining lengths of 100 feet (30 meters) and 200 tons – however, there are some reports of individuals approaching 110 feet (33 meters). Fin whales (Balaenoptera physalus) are the next largest, regularly attaining or surpassing 20 meters (~65-75 feet) with some southern hemisphere individuals reaching 80 and nearly 90 feet (25-27 meters). A host of other species exceed 50 feet in length (15 meters) including humpback whales (Megaptera novaeangliae), sei whales (Balaenoptera borealis) bowheads (Balaena mysticetus), and right whales (Eubalaena spp.). The remaining species  - minkes, the Bryde’s whale complex, pygmy right whales, and gray whales – are all smaller and “only” get to 25-50 feet in length (8-15 meters).

This contrasts quite strongly with the fossil record – most fossil baleen whales are quite small. Most toothed baleen whales from the Oligocene and  late Eocene epochs (~40-33 Ma) are tiny, with skulls similar in size to small-bodied delphinid dolphins and body lengths between 2 and 4 meters, with a few outliers like Llanocetus (~8-12 meters – but see Bisconti et al., 2023) and Coronodon (5 meters). Eomysticetids are the largest stem mysticetes, with sizes of about 8-9 meters. During the Miocene, most baleen whales are quite small, albeit with a larger minimum size than toothed mysticetes – skulls no smaller than about 40 cm in width, corresponding to a body length of about 6-10 meters. Some of the only large-bodied Miocene mysticetes include Pelocetus and Uranocetus, with skulls about a meter wide – similar in size or slightly larger than minke whales. The same is generally true of the late Miocene and Pliocene.

So, what gives? Our idea of a large fossil mysticete is something only slightly larger than a minke whale – widely considered to be one of the daintiest of modern baleen whales. This problem has bothered me for fifteen years – and started bugging me after spending years in the 6-2 million year old Purisima Formation of northern California and not finding any large mysticete bones representing anything larger than a gray whale. There are two general hypotheses that have emerged in recent years – the first is that baleen whales got big very recently, perhaps as a response to global climate change and glaciation during the Pliocene and Pleistocene epochs. The second is that baleen whales were modest in size but got somewhat large before the Pleistocene began. Each hypothesis has its merits, and each has problems – but the interplay between these two ideas has really spurred some great thought into how and when baleen whales got humongous.

Why are whales big?

Marine mammals are large in general because of their life in water, releasing them from gravity. However, most cetaceans are not gigantic; the majority of species are relatively “small” – between 6 and 20 feet in length (~2-6 meters). Half of baleen whales are in the 25-45 foot (7-14 meter range) and only the other half of species are larger. So, something else is driving gigantism. What about diving? The largest odontocetes (toothed whales) are typically all deep divers – including sperm whales and beaked whales. All of these are diving taxa that feed on squid at depth. One notable exception is the killer whale – likely a giant owing to its macropredatory feeding behavior. In general, odontocete body size is largely driven by feeding – and the same is probably also true for mysticetes.

But what about mysticetes? Large body size is important for a number of reasons. The first is economy of scale: you can’t filter feed unless you’re over a certain size, and a recent study proposed that modern minke whales are at the lower theoretical body size limit. Simply put, the lunge feeding approach that balaenopterid whales use for filter feeding is energetically expensive – and given that larger sized whales can net larger volumes of prey during filter feeding, below a certain body size it is too expensive to forage for small prey and manage to acquire enough food to survive (Goldbogen et al., 2023). Large size is also beneficial for migration: suitably dense accumulations of small prey – like krill – are very patchy and isolated in the oceans and whales must travel long distances to get to them, frequently requiring fasting during travel. After a point, possessing a filter feeding apparatus in your mouth also unlocks a larger body size possibility for baleen whales. So, admittedly, there is also a bit of a chicken and the egg issue at work: are baleen whales big because they filter feed, or do they filter feed because they’re big?

How do we estimate baleen whale body size in the fossil record?

Measuring a modern baleen whale is challenging – during the days of whaling, or with a stranding, you can go and measure the whale itself from the tip of the snout to the tail with a very long measuring tape. The key word is “very long” – after all, whales are big. It’s logistically difficult enough to do this – but live whales are even worse, because they’re usually underwater, and even when they surface, you never see the whole animal. Drones with cameras flying directly overhead can take photos at a pre-designated altitude, thereby controlling the scale of what’s in the photograph and permitting ready estimation of the body length (or outright measurement from the photograph) if the whole whale is visible (alternatively, using a few landmarks from the head can be used to estimate body length). But what about fossils? We don’t actually have a lot of baleen whale skeletons that are complete: whales, even fossils, are big, and because big fossils are very easily discovered as they erode out over years or decades, they’re often discovered already partially destroyed. Headless skeletons are not frequently collected, and as a result we have a lot of tailless skeletons with skull, ribcage, flippers, and maybe some lumbar vertebrae. We have an idea of the average number of lumbars and caudals, and can reconstruct the length – but only with sizable error bars.

Biologists like to use body mass, as this communicates a bit of additional information relative towards assessing overall health and body condition but body mass is 1) frequently meaningless to paleontologists (since we just have skeletal remains to work with) and 2) varies considerably during the year as fat stores increase during foraging season and are winnowed away during migration. As a result, paleontologists prefer skeletal metrics – when possible, rare measures of skeletal length are preferred, and can be estimated quite well even with nearly complete skeletons if you’re confident about how many vertebrae you’re missing (possible only when ~90% of the vertebral column is preserved – it’s more difficult, nigh impossible, to do so with a 50% complete vertebral column).

Skull size often corresponds well with body length. The first study to formally propose an equation to estimate body length from skull size was by Lambert et al. (2010), who reported a simple linear equation that best fit a large sample size of modern and fossil mysticetes where both skull width and total skeletal length were known. In other words, the width of the skull seems to linearly scale with skeleton length. A somewhat more complicated suite of equations for estimating cetacean body length from skull measures was published by Pyenson and Sponberg (2011), which used either a combination of five skull measurements or an alternative equation based only on maximum skull width across the zygomatic arches. There are some problems with this latter method – for example, the exact unit of measurement (centimeters, millimeters) was not published, nor were the raw measurements (again) – and because the equations are logarithmic, you get quite different estimates ldepending upon where you stick your decimal point in the measurement (covered in detail in Boessenecker et al., 2023). Another issue is that these equations have frequently underestimated the length of stem Odontocetes and stem mysticetes, outlined in my papers on the eomysticetid Tokarahia (Boessenecker and Fordyce, 2015), the giant dolphin Ankylorhiza (Boessenecker et al., 2020), and the toothed mysticete Coronodon (Boessenecker et al., 2023). In each of these studies, the actual skeletal length was much longer than the estimate from the skull alone.

The first scatterplot of baleen whale size through time (right), compared with mysticete diversity (left) - from Lambert et al. (2010).
 

Hypothesis 1 – Baleen whales evolved to be giants after the loss of megapredators from ocean ecosystems

In the paper reporting the absolutely monstrous sperm whale Livyatan melvillei, Lambert et al. (2010) reported the first scatterplot of baleen whale body size through time – generally showing smaller than modern minimum body size, much smaller than modern maximum body size, and a slight increase in body size during the Pliocene – as well as an “expansion” into smaller body sizes with the dwarf whale Nannocetus (Which really belongs in the Messinian/Late Miocene time period, rather than Pliocene). These authors proposed, albeit indirectly, that the extinction of giant macrophagous predators like Livyatan and Carcharocles megalodon may have permitted baleen whales to exceed lengths of 10-15 meters for the first time. This hypothesis has not been outlined quite as explicitly as here, but touched upon by Lambert et al. (2010), Collareta et al. (2017), and Boessenecker et al. (2019).

Skull size of large mysticetes through time - while it's not clearly labeled in the paper, I *think* the left is supposed to be the North Pacific and the right is supposed to be the North Atlantic. From Pyenson and Vermeij (2016).
 

Hypothesis 2 – Baleen whales became gigantic recently

A short paper by Pyenson and Vermeij in 2016 reported skull width data through time for fossil mysticetes. In this study, marine mammal body size was tracked as it may be a rough proxy for marine primary productivity – earlier papers by Vermeij indicate marine invertebrate shell size is corresponds closely to primary productivity. Pyenson and Vermeij (2016) indeed find that maximum diameter of baleen whale skull size increases dramatically across the Pliocene-Pleistocene boundary. They hypothesized that the increase in primary productivity owing to glacioeustatic changes in sea level caused a bottom-up driver that led to rapid increases in body size in baleen whales.

Body size scatterplot as compared to global oxygen isotope curve, supporting the Pleistocene gigantism hypothesis - from Slater et al. (2017). I've intentionally left the caption here for easier interpretation of the methods.

An important study by Graham Slater and others in 2017 attempted to answer this for the first time. This study conducted two analyses: a survey of baleen whale body size records through the fossil record, and then an analysis of body size evolution within a phylogenetic context. In the first analysis, the study finds that modern mysticetes obviously have a larger maximum body length but also a larger minimum body length, with the smallest extant mysticete – Caperea – being much larger than the smallest extinct mysticetes. In the phylogenetic analyses (using the dataset from Marx and Fordyce, 2015), these authors found that 1) gigantism evolved convergently within the balaenopterids and balaenids (hardly groundbreaking, but good to demonstrate) and 2) analysis of evolutionary rate indicating a very recent, late Pleistocene (~200,000 years ago) shift from an “unbiased random walk” mode of evolution to “upward biased random walk” – essentially meaning an abrupt and geologically very recent increase in body size. Frustratingly, the actual dataset upon which the study is founded is a bit opaque: none of the raw measurements used for the study were published.

Body size evolution of mysticetes within a phylogenetic context - from Slater et al. (2017). Yellow = small, orange and red = medium and large/gigantic. I wish this had a better gradient on it, the orange is difficult to pull out.

However, Slater et al. noted that this surprisingly recent shift is driven by a lack of fossils dating to the last three million years – an artifact that ends up lending disproportionate ‘weight’ (if you will) to the extant species data. They attempted to correct this by adding a couple of middle-late Pleistocene records of extant species including a fossil humpback from Japan (Megaptera novaeangliae) and a fossil gray whale from southern California (Eschrichtius robustus – or, more appropriately, Eschrichtius sp., cf. E. robustus after the original authors). These extended the geologic range for these two taxa to the late Pleistocene (0.125 Ma) and middle Pleistocene (0.2-0.5 Ma), respectively – and pulled the evolutionary rate and mode shift earlier by about 100,000 years, to 0.3 Ma. Lastly, these authors performed some simulations in an attempt to evaluate whether or not preservation bias or sampling bias might explain the size difference between modern and fossil mysticetes. They assigned varying values here that would indicate gigantic/large taxa would have an equal probability of being discovered and collected relative to smaller whale fossils, and also lesser chances. They found no evidence of size-based preservation or sampling bias – admittedly these methods are a bit of a ‘black box’ and the devil is in the details, so I‘m unable to comment further. But, I do have some commentary (below) on sampling biases, perhaps not taken into account by Slater et al.

Slater et al. conclude that a relatively recent change in oceanographic conditions must have factored into the size increase in baleen whales, given that there seems to be no increase or real change in body size trends earlier during the Miocene epoch (23-5.3 Ma), and that no size increase corresponds to the onset of Antarctic glaciation, the evolution of baleen, or coevolution with megapredators (e.g. Livyatan and C. megalodon). Likewise, these authors find no evidence for cooling climate to have driven increases in body size – which is somewhat surprising. Instead, they reason that dramatic increases in primary productivity, caused by upwelling and especially at high latitudes, permitted mysticete body size to increase. Critically, these authors consider the highly fragmentary and undersampled record of Pliocene, and especially Pleistocene mysticetes to be an impediment to testing their hypothesis.


 The giant whale from the Italian Pleistocene - an early Pleistocene blue whale reported by Bianucci et al. (2019).

Hypothesis 3 – Baleen whales have been large for a while – Bianucci et al. 2019

A follow-up study by Bianucci et al. (2019) reported a massive skull of a fossil blue whale, Balaenoptera sp., cf. B. musculus, from early Pleistocene strata of Italy; the skull itself consists of a fragmentary but enormous braincase measuring nearly three meters in width and corresponding to a length of 23.4-26.1 meters (77-85 feet) – similar in size relative to modern blue whales. This specimen originates from precisely the time period that Slater et al. identified as a problem: the early to middle Pleistocene; it dates to approximately 1.5-1.25 million years. Bianucci et al. also report a number of data points based on large mysticetes from the Pisco Formation of Peru – a large specimen of the middle Miocene “cetothere” Pelocetus with a body length estimated at 12-13 meters (39-42 feet), a somewhat younger Tortonian-aged rorqual (Balaenopteridae) with an estimated length of 16-18 meters (52-59 feet), and an even younger Messinian-aged balaenopterid estimated at 15-17 meters (49-56 feet).


Updated version of the analysis with the new datapoints from the Italian Pleistocene and the Peruvian Miocene - from Bianucci et al. (2019).

Bianucci et al. incorporated these four new big datapoints (sorry for the pun) into the same analysis run earlier by Slater et al. Owing to reproducibility issues with reporting uncollected fossils documented during fieldwork, Bianucci et al. ran the analysis with and without the Peruvian “field whales”. Based only on the inclusion of the 1.5-1.25 Ma fossil blue whale from Italy, the “mode shift” – e.g. the timing of a trend towards gigantism and change in the mode of evolution – occurs quite a bit earlier, 3.6 Ma (during the middle Pliocene; originally 0.3 Ma according to Slater et al.), and the sharpness of the mode shift is decreased considerably. In plain English, this means that the increase in body size occurred earlier and was more gradual than originally highlighted in the analyses of Slater et al. Bianucci et al. consider modern baleen whale sizes to have evolved during the early Pliocene or even late Miocene – in stark contrast to the Pleistocene evolution of gigantism from Slater et al.


Some of the specimens measured in the field by Bianucci et al. (2023) in the Peruvian desert - and left behind.

A caveat with this approach is that the three Peruvian specimens were not collected, but rather documented in the field with measurements and photographs taken – and *left* in the field. On one hand, this is a bit of an issue as it violates the typical requirement that all fossils be catalogued within a museum collection in order to qualify for publication. This is not universally mandated, of course, but one issue is that as these fossils erode away, it will not be possible to double check the measurements or identification – beyond that data that is already published by Bianucci et al. I will be the first to admit that I am awfully anal retentive about this norm in paleontological publishing – voucher specimens are needed to ensure that published studies are testable. However, there are two caveats here – not additional caveats per se, but caveats to my caveat outlined above. Or, rather, let’s call them justifications. The first of these is actually at the very heart of this issue: whales, after all, are BIG. And these Peruvian whales are big skeletons. Big skeletons are difficult to excavate, difficult to prepare, curate*, and eventually, study – even photography of gigantic specimens has its own problems (at U. Otago Ewan Fordyce, Cheng-Hsiu Tsai, Felix Marx, and I would have to photograph mysticete skulls and mandibles from the second story of the Geology building; at the UCMP Berkeley oversize facility I’ve climbed atop a 15 foot ladder to do the same). These logistical issues are compounded by the fact that Peruvian museums were basically filled up to the brim about 10 years ago, and it’s been difficult to add any more additional skeletons. So, Giovanni Bianucci, Olivier Lambert, and others have resorted to publishing field observations like this – not my favorite, but a solution to a serious problem. The second justification is that it actually helps prove a point I will make below regarding sampling bias (see below).

*And that’s only if the museum curator cares about fossil whales under their care.

Change in body size in Mysticeti over the course of the past 40 million years - from Bisconti et al. (2023).
 


 Body size evolution of mysticetes within a phylogenetic context -from Bisconti et al. (2023).


Body size evolution in Mysticeti as relating to various global, regional, and evolutionary events and trends. Quite possibly the busiest figure I've ever seen - actually combined from three figures of Bisconti et al. (2023).

More recent comprehensive studies – a combination of hypotheses 2 and 3?

A rather detailed analysis of mysticete body size was carried out earlier this year by Bisconti et al. (2023), and I expect this to be the final word for a long while as it seems quite exhaustive. According to this analysis, toothless mysticetes increased gradually in body size throughout the Miocene – with an increase in mean body size around the Pliocene-Pleistocene boundary about 2-3 million years ago. Notably, Bisconti et al. consider the Pliocene-Pleistocene interval to be the period of time with the highest diversity in body size among mysticetes – from tiny herpetocetines up to the gigantic species of Balaenoptera. Further, this study notes that the gigantic size of fin whales (B. physalus) and blue whales (B. musculus) are actually outliers when it comes to the range of body size amongst modern mysticetes in general. Another interesting finding is that, owing to the much exhaustive sampling of fossil balaenopterids relative to Slater et al., Bisconti et al. indicate that the small body size of minke whales (Balaenoptera acutorostrata and B. bonaerensis) is likely caused by dwarfism rather than plesiomorphic retention of ancestrally small body size.*

*However, the phylogenetic placement of the minke-like species I named, Balaenoptera bertae, is incorrect in this study and results from some miscodings. If this taxon is placed as an ancestral minke whale as in some other phylogenies, it might very well change this interpretation.


Body size evolution in cetaceans - scatterplots of skull width, from Boessenecker and Geisler (2023) - mysticetes are in 'C'.

A few weeks ago I published a massive monograph with Jonathan Geisler on the early dolphin Xenorophus – and included within are a bunch of ‘mini-analyses’ and datasets including a fairly reasonable, but not entirely exhaustive survey of body size data points for cetaceans. This was done with specific focus on odontocetes – but we also included toothed and toothless mysticetes, though we did not discuss them in any capacity beyond Oligocene taxa. But, there is some interesting stuff in here!

1) The first is that during the Paleogene, toothed and toothless mysticetes don’t really overlap at all in body size. Late Eocene Llanocetus pulls the mean toothed mysticete body size pretty far up, but then again, there are no late Eocene toothless mysticetes, and by the early Oligocene, all toothless mysticetes are quite a bit larger than toothed taxa. This has some interesting implications for niche differentiation. Are toothed mysticetes small because of a difference in feeding ecology from toothless mysticetes? Are eomysticetids larger because they can migrate?

2) There is a serious bottleneck in the Serravallian (late middle Miocene) – this is because some of the only reported mysticetes from this time period are all closely related and anatomically similar.

3) During the Messinian (latest Miocene) we have a different problem caused by small sample size: only a handful of mysticetes are known from the Messinian, and all basically from California and Peru: the tiny herpetocetines Nannocetus and Piscobalaena and the balaenopterid Parabalaenoptera. While there are loads of unpublished Messinian balaenopterids, and other taxa like the right whale Eubalaena shinshuensis – I restricted myself to published records only – and E. shinshuensis doesn’t have a skull complete enough to measure in the same way (maximum width of the skull).* As a result, two out of the three data points make the mysticete mean body size dip quite a bit. Informed by unpublished specimens from California, I predict improved sampling to show a similar situation as during the Tortonian.

4) There is a gradual increase in body size that begins in the early Pliocene and seems to really pick up during the late Pliocene – supporting a more muted version of the dramatic trend towards gigantism by Slater et al.

5) There also seems to be a late Miocene decrease in minimum body size – late Miocene cetotheriids seem to be a bit smaller than the smallest early and middle Miocene “cetotheres”. See below.

6) Lastly, there is a subtle increase in mean body size over the course of the entire Neogene and Oligocene – with that brief dip in the Messinian.

*However, the skull is complete in terms of its length and the body length has been estimated at 12-13 meters.

 

My favorite tiny baleen whale: Herpetocetus. From El Adli et al., 2014.

Why are so many Miocene and Pliocene baleen whales tiny?

One question that continually gets left by the wayside is the flipside of the main question asked by Slater et al. and Bianucci et al., and that is “why were baleen whales so damned tiny before the Pleistocene?” A corollary question which emerges is “why are there no modern mysticetes smaller than 6 meters as adults?” I’m afraid I don’t have any answers, just some speculation, but let’s dive in anyway. Oligocene mysticetes (which I need to write about in much more detail) range in size from harbor porpoise size (e.g. toothed mysticete Fucaia spp., ~2 meters long) to minke whale sized (e.g. Eomysticetidae; 6-8 meters), and the rest somewhere in between. Most Miocene age mysticetes are not much larger – most are in the 5-10 meter range; taxa like Parietobalaena, Isanacetus, Diorocetus, and true cetotheriids. There are a few outliers like Pelocetus sp. reported by Bianucci et al. and Pelocetus calvertensis (8-12 meters). By the late Miocene and Pliocene, some cetotheriids have evolved shockingly tiny sizes, especially the herpetocetines: Nannocetus was approximately the size of a bottlenose dolphin (2.4 meters according to Slater et al.) and Herpetocetus was perhaps as large as a pilot whale or Risso’s dolphin (3.7 meters according to Slater et al. – though the source of this measurement is unclear). In fact, the data from our 2023 Xenorophus monograph indicates that the herpetocetines are somewhat smaller in minimum adult body size than the earlier “cetotheres” sensu lato.

Several hypotheses come to mind: the first, outlined by Collareta et al. (2017), is that perhaps small baleen whales may have been kept small by predation pressure from Carcharocles megalodon and maybe also killer sperm whales like Livyatan and Albicetus. Perhaps small body size made these whales more agile and able to evade macropredators. The second is that perhaps greater upwelling at temperate latitudes resulted in a greater mass of available prey to filter feed for; this is frequently understood to be an evolutionary driver of dwarfism. With the change in upwelling regime, body sizes increased and the herpetocetines went extinct (recently: Boessenecker, 2013). The third possibility is that these dwarf mysticetes fed in a manner completely different from modern mysticetes, and perhaps unlocked locally abundant food sources that were inaccessible to other baleen whales. In 2014, Joe El Adli and I published a paper naming the new mysticete Herpetocetus morrow from the San Diego Formation of California, and proposed that its unusual craniomandibular joint was adapted for only opening up to 15-30 degrees and that it was likely well-adapted for longitudinally rotating the mandibles – evidence that it may have been a dedicated benthic filter feeder along the many now-lost highly productive embayments along the California coast during the Pliocene. Lastly, and discounted by Slater et al. (perhaps too hastily) is the possibility that warmer Pliocene waters permitted somewhat smaller body masses for filter feeders like Herpetocetus.

Another consideration – or rather conundrum – is the recent hypothesis that extant minke whales are at the lower size limit for baleen whale body size (Goldbogen et al., 2023). One serious problem with this conclusion is that there is no shortage of tiny, “sub-minke” whale fossil mysticetes in the fossil record that were either somewhat smaller and closely related (within the Balaenopteridae) or were more distantly related and considerably tinier (e.g. Cetotheriidae – Nannocetus, Piscobalaena, Herpetocetus). Normally I would wave off the cetotheriids as not being relevant because, so far as we can understand, they had a different feeding ecology: either benthic feeding in the case of Herpetocetus (El Adli et al., 2014) and even fish eating, as in the case of Piscobalaena preserved with gut contents (Collareta et al., 2015). But, aside from herpetocetines, there are several balaenopterids that were a bit smaller than minke whales including Balaenoptera bertae (61 cm wide skull corresponding to a length of 4.5-5 meters!), an undescribed species of “Burtinopsis” from Japan, and a handful of others from Japan and Europe. So, there are numerous fossil lunge-feeding rorquals that are even tinier than the extant minke whales.

 

Other records of Pliocene-Pleistocene mysticetes and baleen whale fossil “neospecies”

Slater et al. (2017) included two Pleistocene records of extant taxa in their analysis – a late middle Pleistocene (~0.125-0.15 Ma) specimen of the extant humpback whale Megaptera novaeangliae, initially reported by Nagasawa and Mitani (2004), and a late middle Pleistocene gray whale from the San Pedro Sand of Los Angeles County – Eschrichtius sp., cf. E. robustus – originally reported by McLeod and Barnes (1984). These are not the only usable records of extant species; some other records are published, but not used by Slater et al. (2017) or Bianucci et al. (2019).


A blue whale like mandible from the Pliocene of Japan - from Oishi (1997).

The first is “Sibbaldus sp.”, a mandible better identified as an early blue whale – Balaenoptera sp., cf. B. musculus – by Oishi (1997), from the lower Pliocene Tatsunokuchi Formation. The caveat is that this specimen, while it is identical in morphology to extant Balaenoptera musculus, it is not exactly enormous – it is a maximum of three meters long. Note that this is the same as the width of the entire skull of the Pleistocene Italian specimen reported by Bianucci et al. (2019). Regardless, the specimen is large for a Pliocene whale and methods do exist for estimating body length from balaenopterid mandibles (Pyenson et al., 2013; though again, no raw measurements were ever published for this study).

While not really permitting a body length estimate, a fossil bulla of Balaenoptera sp., cf. B. physalus was reported by Cheng-Hsiu Tsai and myself in 2017; this specimen dates to 0.95-1.3 Ma, and pulls an additional extant and gigantic mysticete whale species into the early Pleistocene. This record was not included in Bianucci et al. (2019; these authors only added their new specimens to the phylogeny and did not update it otherwise) though I think it would further reinforce an earlier evolution of gigantism.

Eschrichtius akishimaensis was named by Kimura et al. (2017) from the early Pleistocene Komiya Formation of Japan, dating to 1.95-1.77 Ma. This taxon is a ‘neospecies’ (extinct species in an extant genus) and pulls Eschrichtius even earlier. Additional Pliocene specimens of Eschrichtius reported from the North Sea and from Japan pull the extant gray whale further back in time than this Pleistocene species (Ichishima et al., 2006; Tsai et al., 2020). Note however that the  “Teshio” gray whale, Eschrichtius sp. from the Pliocene of Japan, is considerably smaller than extant Eschichtius robustus, with a likely skull width of less than a meter (I’d wager about 75-85 cm).

The extinct right whale Eubalaena ianitrix was named in 2017 by Bisconti et al., representing a medium-sized balaenid with a skull width of 1.7 meters – but only a maximum length of 8 meters. Additional extinct Pliocene balaenids include Antwerpibalaena liberatlas with an estimated body length of 9.5-11.9 meters, and the even more recently named Charadrobalaena valentinae with a body length of ~11 meters.

 

Perspective from a field paleontologist – observations from the Plio-Pleistocene of the Pacific and Atlantic coasts, and collections bias

So what do I think about all of this? Both Slater et al. (2017) and Bianucci et al. (2019) bring up excellent counterpoints. The fact that the analysis of Slater et al. was so sensitive to a few additional data points tells me that the Bianucci hypothesis – a gradual and sort-of-recent evolution of gigantism – as opposed to an abrupt and extremely recent origin a la Slater – tells me the former is much more likely, even though the Slater et al. (2017) hypothesis, as extreme as it is, appeals to my own field observations.


 An example of a rather large vertebra found in the field: a posterior lumbar or anterior caudal vertebra, with a centrum diameter of approximately 22-28 cm. Judging from a cursory glance at some papers on extant
Balaenoptera, this vertebra represents a whale at least 12 meters (~40 feet) and possibly up to about 15 meters (~50 feet) in length - a large, but not gigantic mysticete. This specimen is from an early Pliocene horizon in the Purisima Formation. From Boessenecker (2013).

What sort of field observations? Well, I’ve spent about 20 years looking for whales in the Purisima Formation of northern California (7-2.5 Ma). The majority of mysticete taxa are small to medium size; only a few fossils of something approaching a gray whale or humpback in overall size have ever been spotted in outcrop, let alone collected. The largest Purisima mysticete I am aware of is a right whale (Balaenidae) mandible measuring about 4 meters in length (in the collection of Frank Perry) and suggesting a body length of 12 meters (similar in size to Eubalaena shinshuensis from Japan) – nearly as large as extant Eubalaena japonica (13-15 meters). Mandibles and vertebrae are commonly encountered, and perhaps if utilized intelligently, might provide more data than restricting analyses to skulls. Vertebrae are commonly collected – modern and fossil – and we have a good handle on the range of vertebral counts in modern mysticetes. You could even just measure up the average size of thoracic and lumbar vertebrae for each modern species (these are the most common fossil vertebrae) and go out and sample fossil assemblages for vertebral size through time, compared with vertebral sizes of modern taxa: an apples to apples comparison. This might give a sample that is much, much less biased towards small sample sizes like sticking with skulls (see below) – given that a blue whale sized vertebra is much more easily collected than a blue whale sized skull. In general, I think the average body size of a Purisima whale would be something the size of a minke whale, with a few smaller, and a few that are slightly larger – and rare examples of whales approaching humpback whale sizes (mostly isolated vertebrae, and a few non-collected mandibles I’ve watched erode out over the past 20 years that perhaps represent balaenopterids with 3-4 meter long mandibles). These are admittedly ‘gestalt’ approaches and not data driven – but, suggests to me that there is another way to get at this problem by supplementing the skull size dataset with isolated mandibles (e.g. Pyenson et al., 2013) and vertebrae. Therefore, much larger datasets could be marshalled.


 The tiny rorqual
Balaenoptera bertae - one of the oldest well-established 'neospecies' of any balaenopterid whale. From Boessenecker (2013).

I’ve only ever excavated relatively small baleen whales from the Purisima – the holotype of the “sub-minke” whale sized Balaenoptera bertae (itself possibly an ancestral minke) being the largest specimen I’ve excavated – and even still, it only has a skull width of 61 cm! That’s only 50% larger than the largest Herpetocetus. I’ve excavated three Herpetocetus skulls since, and carted off a fourth in a 100 lb concretion. It is a pain in the ass to do this – and I am an extremely motivated field paleontologist. Many other paleontologists and collections managers who like to study things like rodents and foraminifera groan whenever you ask to see some fossil whale in their collection or help dig up new whales (dinosaur paleontologists have similar stories about sauropod fossils).


Here I am digging out the skull of Herpetocetus bramblei in 2007 when I was 21 years old (and much, much scrawnier).

Even little whales are a pain in the ass: in 2007 I found a partial skull of Herpetocetus, a species that would be named the following year as Herpetocetus bramblei. The rostrum had eroded away but a large fragment around the blowhole sat in a block I was able to pick up; I prepped out the block that afternoon at home and determined that the braincase was still in the cliff (rather than the tip of the rostrum) and I returned the following day to excavate it. I spent one afternoon removing overburden, digging a 2-3’ deep cylindrical hole with about 4” of space on all sides, and then returned the day after with my amateur colleague Chris Pirrone. We jacketed the specimen at 4pm, had it undercut, flipped, and the other side plastered by 6pm. We moved it down the beach and realized we were going to have a helluva time getting it off the beach in the face of the incoming tides. The jacket spent about 20 minutes in the surf while I reconsidered a route up the slippery rocks and also my life choices. By 8pm we had it up the first ledge – a precarious situation, since we had foot-sized spots of rock without algae which, if we slipped, would mean the jacket would certainly fail and the block inside would crack – let alone guarantee broken bones for us (either from falling or from the jacket falling on either of us). For the time being we were safe – but had a 20’ long, 18” wide ledge about 12’ above the waves to carry the 24” wide jacket across. Chris got out a floor mat from the trunk of his old honda, which we used as a hammock and – very carefully – inched the jacket down the perilous ledge. Our last feat was to lift the ~150 lb jacket up three or four surfer’s stairs – surfer’s stairs are usually quite tall, the tallest being about 4’ tall. (Surfers are pretty much always in good shape and can deftly get up or down these stairs with one hand, the other arm cradling their board). It took us an hour – and that last one, a 4’ tall ledge – took us several attempts. We eventually had to lift it a few inches at a time, and then pin it to the wall by leaning up against it for a few minutes to rest, and then inching it up again slightly. We eventually got it into my old hatchback – and I went home and slept for ten hours. I couldn’t lift anything more than a pound for a couple of days.

I can’t think of a plausible taphonomic filter to exclude giant specimens from the fossil record, but I can think of some very realistic collection bias issues caused by human behavior: caution. There is no shortage of fossil mysticetes that have simply been too large to excavate. I personally am aware of several in the Purisima Formation, and several from the east coast. The specimens from Peru are now published examples of such whales left in the field. A different problem is highlighted by fossils from southern California – there are several large mysticetes in museum collections that were excavated but never prepared; one such example is a fin or blue whale sized balaenopterid from the Capistrano Formation of Orange County – an account of the impressive excavation was written about this skull and skeleton, but no follow-up paper was ever published and the specimen may not even be prepared yet. This is not unusual: there is a large backlog of unstudied marine mammal fossils at museums like LACM (Natural History Museum of Los Angeles County), OCPC (the Cooper Center), SDNHM (San Diego Natural History Museum) owing to the vagaries of preparing – and studying – such gigantic fossils.  

Whether or not these very real biases in sampling were adequately controlled for in the methods-heavy Slater et al. study is, in a way, immaterial: the three Peruvian specimens from Bianucci et al. very clearly shift the story much earlier in time: that difference in results IS the measure of sampling bias, in a way. 


 A large balaenopterid skull from the San Diego Formation of southern California - identified in the 1970s as "Plesiocetus" - beautifully preserved, and quite large (1+ meter skull width) yet undescribed. Many such examples from the late Miocene and Pliocene of California exist. UCMP collections, photo by me.

Another possible source of bias is publishing effort. A minority of fossil mysticetes from the late Miocene and Pliocene of the eastern Pacific have actually been named: small taxa like Herpetocetus bramblei, Herpetocetus morrowi, Balaenoptera davidsonii, and Balaenoptera bertae have been named from California, Kennedycetus from Baja, Piscobalaena nana, Archaebalaenoptera euseboi, and Incakujira anillodefuego from Peru, and virtually nothing from Chile (including all those whales from the spectacular assemblage from Cerro Ballena that such a big deal was made of ten years ago). In the case of Bianucci et al. (2019), a couple of additional but unstudied mysticetes from the upper Miocene of Peru were included. Otherwise, we’re missing much of the fauna for these localities; in the case of the San Diego Formation and the Purisima Formation of California, for example, only two out of seven or eight mysticete taxa from each have been formally published, so we’re missing between 2/3 and ¾ of the fauna (this was covered, and complained about, in detail by Boessenecker, 2013).

Concluding Remarks

Several studies have surveyed the body size of baleen whales through time. The study by Slater et al. is perhaps the most methods-heavy, yet the sample is rather limited, and the conclusions of it did not stand up very long - four new datapoints shifted the rapid onset of gigantism several million years earlier into the Pliocene (Bianucci et al., 2019). More comprehensive datasets published by Bisconti et al. and yours truly this year stuck with scatterplots and mean body size trends, and dispensed with the more involved methods employed in earlier studies - at minimum, supporting Pliocene-Pleistocene increase in body size, and also a late Miocene decrease in minimum body size lost after the extinction of herpetocetines during the past one million years.

The late Neogene and Quaternary fossil record has not improved greatly - there are still a LOT of fossils out there to get published, especially from California and Japan - but there are quite a few more datapoints available that have been published since 2016. I would be interested to see what the inclusion of the more exhaustive datasets from the Bisconti et al. and Boessenecker and Geisler (2023) would do with the analyses of Bianucci et al. (2019). Additionally, mandibles and vertebrae offer some additional sources of data.

The flipside of gigantism also needs further study: if we're asking why modern baleen whales are large, what drives certain baleen whales to have stayed so small for so long? 

ADDENDUM: Some late-breaking news from Australia - a large mysticete of early Miocene age reveals yet another hypothesis

Just as I was about to start wrapping this up, I saw on twitter some posts by my friend and colleague Erich Fitzgerald that something 'big' in whaleontology was about to come out - and sure enough, a paper on a very fragmentary, but surprisingly large mysticete from the lower Miocene of Australia was published late last night in Royal Society Open Science by James Rule et al. The specimen was found in 1921 in the Mannum Formation of South Australia, and consists of the left and right anterior tips of the mandibles - the chin, so to speak - and some other skull fragments. The specimen belongs to some sort of chaeomysticete, given the lack of teeth - though the authors stop at the broader taxon Kinetomenta (which includes the toothed Aetiocetidae). The mandible is quite large - approximately 18 cm in depth at the most, nearly twice that of most eomysticetid whales I've worked on. The authors developed some predictive equations using mandibles of modern minke whales to reconstruct the size of the entire mandible, and then body length - resulting in an estimate of 9 meters. Though the specimen is quite fragmentary, I actually think this is quite conservative. For example, the eomysticetid Tokarahia (one of the largest known Oligocene toothless mysticetes) is estimated at a body length of about 7-8 meters - and the mandible measures only 8-10 cm deep at this same location. The mandibles of eomysticetids are of course quite gracile, but this underscores the fact that owing to the incompleteness and uncertainty of the identification of the new Australian specimen, 9 meters is on the conservative side and a larger size is possible (which would only further underscore the author's point - but I digress).


The new Australian mysticete specimen reported by Rule et al. (2023).

These authors execute a couple of different analyses - the first demonstrating, quite clearly, that southern hemisphere mysticetes are relatively undersampled and constitute perhaps 1/5 of the gloval sample of fossil mysticetes. They also demonstrate this on a phylogeny, and while they don't discuss it much, it does show that the sampling is also taxonomically biased: many Eomysticetidae and eomysticetid-adjacent taxa are reported from New Zealand and show up as a cluster on the phylogeny, along with a number of late Neogene balaenopterids and cetotheriids all from South America. A key note here is that earlier studies reported a lack of earliest Miocene (Aquitanian stage) mysticetes worldwide, though this may simply be northern hemisphere bias.

The new specimen represents the first "medium sized" toothless mysticete to evolve, with a body length estimated at 9 meters (slightly larger than a minke whale, slightly smaller than bryde's whales). The study further analyses southern and northern hemisphere mysticete body sizes separately, and surprisingly finds that southern hemisphere fossil mysticetes are on the whole larger bodied than their northern hemisphere counterparts, with very different "mean" values for each sample (not exactly calculated the same way as Bisconti et al., 2023, and Boessenecker and Geisler, 2023, above - a polynomial line of best fit for each scatterplot, I imagine since the age dates/ranges were not 'binned' as in the previous studies).


 Evolution of body size in Mysticeti: Rule et al. (2023) find evidence for a more gradual evolution of large and gigantic body size.

Rule et al. propose that large body size evolved gradually and earlier in the southern hemisphere - likely driven by the setup of the circum-Antarctic current, strengthening of this current, and the relatively consistent availability of food provided by upwelling and high primary productivity enabled by the current - a long-standing idea proposed by the late Ewan Fordyce, and best outlined in Fordyce (2003). These authors point out that a northern hemisphere bias is largely responsible for the Plio-Pleistocene rapid onset of gigantism hypothesis (e.g. Slater et al., 2017) - and that the southern hemisphere sample smooths this out and instead supports a more gradual evolution of giant body sizes (e.g. Bianucci et al., 2019). They also point out that the loss of tiny mysticetes, common in the late Miocene, during the Pliocene-Pleistocene interval (they say 4 Ma, but I suspect this is far too old) may be a more paleo-ecologically reliable trend than the onset of gigantism (and as alluded to above, certainly distorts the average/mean body lengths during the late Neogene, prior to their extinctions).

While the specimen reported by Rule et al. is far from anatomically satisfying (so to speak), the study itself is fabulous and I really enjoyed the hemisphere-specific breakdown. What's next? I actually think some followup study on why tiny baleen whales were so tiny - ecological reasons and evaluating supposed physiological constraints on body size - and perhaps further analysis of the timing of dwarf baleen whale extinction over the past few million years - might be a good step forward. For my Australian colleagues: I've got some ideas about herpetocetines, let's talk =]

References

Barnes, L.G. and McLeod, S. 1984. The fossil record and phyletic relationships of gray whales. In: M.L. Jones, S.L. Swartz, and S. Leatherwood (eds.), The Gray Whale, 3-32. Academic Press, San Diego.

Bianucci, G., Marx, F.G., Collareta, A., Di Stefano, A., Landini, W., Morigi, C., and Varola, A. 2019. Rise of the titans: baleen whales became giants earlier than thought. Biology Letters 15: 20190175.

Bisconti, M., Lambert, O., and Bosselaers, M. 2017. Revision of "Balaena" belgica reveals a new right whale species, the possible ancestry of the northern right whale, Eubalaena glacialis, and the ages of divergence for the living right whale species. PeerJ 5: e3464.

Bisconti, M., Pellegrino, L., and Carnevale, G. 2023. The chronology of mysticete diversification (Mammalia, Cetacea, Mysticeti): Body size, morphological evolution and global change. Earth Science Reviews 239: 104373.

Boessenecker, R.W. 2013. A new marine vertebrate assemblage from the Late Neogene Purisima Formation in Central California, Part II: Pinnipeds and cetaceans. Geodiversitas 35: 815-940.

Boessenecker, R.W., Beatty, B.L., and Geisler, J.H. 2023. New specimens and species of the Oligocene toothed baleen whale Coronodon from South Carolina and the origin of Neoceti. PeerJ 11: e14975.

Boessenecker, R.W., Churchill, M., Buchholtz, E.A., Beatty, B.L., and Geisler, J.H. 2020. Convergent evolution of swimming adaptations in modern whales revealed by a large macrophagous dolphin from the Oligocene of South Carolina. Current Biology 30: 3267-3273.

Boessenecker, R.W., Ehret, D.J., Long, D.J., Churchill, M., Martin, E., and Boessenecker, S.J. 2019. The early Pliocene extinction of the mega-toothed shark Otodus megalodon: a view from the eastern North Pacific. PeerJ 7: e6088.

Boessenecker, R.W. and Fordyce, R.E. 2015. A new genus and species of eomysticetid (Cetacea: Mysticeti) and a reinterpretation of "Mauicetus" lophocephalus Marples, 1956: transitional baleen whales from the upper Oligocene of New Zealand. Zoological Journal of the Linnaean Society 175: 607-660.

Boessenecker, R.W. and Geisler, J.H. 2023. New skeletons of the ancient dolphin Xenorophus sloanii and Xenorophus simplicidens sp. nov. (Mammalia, Cetacea) from the Oligocene of South Carolina and the ontogeny, functional anatomy, asymmetry, pathology, and evolution of the earliest Odontoceti. Diversity 15: 1154.

Collareta, A., Lambert, O., Landini, W., Di Celma, C., Malinverno, E., Varas-Malca, R.M., Urbina, M., and Bianucci, G. 2017. Did the giant extinct shark Carcharocles megalodon target small prey? Bite marks on marine mammal remains from the late Miocene of Peru. Palaeogeography, Palaeoclimatology, Palaeoecology 469: 84-91.

Collareta, A., Landini, W., Lambert, O., Post, K., Tinelli, C., Di Celma, C., Panetta, D., Tripodi, M., Salvadori, P.A., Caramella, D., Marchi, D., Urbina, M., and Bianucci, G. 2015. Piscivory in a Miocene Cetotheriidae of Peru: first record of fossilized stomach content for an extinct baleen-bearing whale. The Science of Nature 102: 70.

El Adli, J.J., Deméré, T.A., and Boessenecker, R.W. 2014. Herpetocetus morrowi (Cetacea:Mysticeti) a new species of diminutive baleen whale from the Upper Pliocene (Piacenzian) of California, USA, with observations on the evolution and relationships of the Cetotheriidae. Zoological Journal of the Linnaean Society 170: 400-466.

Fordyce, R.E. 2003. Cetacean evolution and Eocene-Oligocene oceans revisited. In: D.R. Prothero, L.C. Ivany, and E.A. Nesbitt (eds.), From Greenhouse to Icehouse: the Marine Eocene-Oligocene Transition, 154-170. Columbia University Press, New York.

Goldbogen, J.A., Pyenson, N.D., and Madsen, P.T. 2023. How whales dive, feast, and fast: the ecophysological drivers and limits of foraging in the evolution of cetaceans. Annual Reviews in Ecology, Evolution, and Systematics 54: 307-325.

Ichishima, H., Sato, E., Sagayama, T., and Kimura, M. 2006. the oldest record of Eschrichtiidae (Cetacea: Mysticeti) from the Late Pliocene, Hokkaido, Japan. Journal of Paleontology 80: 367-379.

Kimura, T., Hasegawa, Y., and Kohno, N. 2017. A new species of the genus Eschrichtius (Cetacea: Mysticeti) from the early Pleistocene of Japan. Paleontological Research 22: 1-19.

Lambert, O., Bianucci, G., Post, K., Muizon, C.d., Salas-Gismondi, R., Urbina, M., and Reumer, J. 2010. The giant bite of a new raptorial sperm whale from the Miocene epoch of Peru. Nature 466: 105-108.

Marx, F.G. and Fordyce, R.E. 2015. Baleen boom and bust: a synthesis of mysticete phylogeny, diversity and disparity. Royal Society Open Science 2: 140434.

Nagasawa, K. and Mitani, Y. 2004. A humpback whale, Megaptera novaeangliae (Borowski, 1781), from the  Pleistocene Kioroshi Formation of Inba-mura, Chiba Prefecture, central  Japan. Paleontological Research 8 (3): 155-165.

Oishi, M. 1997. A dentary of Sibbaldus sp. from the lower Pliocene of Hiraizumi, Iwate Prefecture, Northeast Japan. Bulletin of the Iwate Prefectural Museum 15: 1-10.

Pyenson, N.D. and Sponberg, S.N. 2011. Reconstructing body size in extinct crown Cetacea (Neoceti) using allometry, phylogenetic methods and tests from the fossil record. Journal of Mammalian Evolution 18: 269-288.

Pyenson, N.D. and Vermeij, G.J. 2016. The rise of ocean giants: maximum body size in Cenozoic marine mammals as an indicator for productivity in the Pacific and Atlantic oceans. Biology Letters 12: 20160186.

Rule, J.S., Duncan, R.J., Marx, F.G., Pollock, T.I., Evans, A.R., and Fitzgerald, E.M.G. 2023. Giant baleen whales emerged from a cold southern cradle. Proceedings of the Royal Society B 290: 20232177.

Slater, G.J., Goldbogen, J.A., and Pyenson, N.D. 2017. Independent evolution of baleen whale gigantism linked to Plio-Pleistocene ocean dynamics. Proceedings of the Royal Society B 284: 20170546.


 

No comments:

Post a Comment