represents the generic place holder for the numbers above.

Specimen RU-EFP-00006 is a partially articulated skeleton of *Thoracosaurus neocesariensis*. The right femur (Figure 1A) was chosen to analyze for collagen I remnants because this bone is generally well-preserved in terms of gross morphology (Figure S1). Additionally, being a large limb bone, it possesses a greater amount of cortical tissue for analyses. These attributes imply this fossil could be a favorable candidate for molecular retention [25,38,39]. Stylopodial and zeugopodial elements of a juvenile American alligator (*Alligator mississippiensis*) served as a modern positive control for biomolecular assays (see Supp for treatment of extant bones prior to analyses below). Both modern and fossil molecular analyses were conducted with solely cortical tissue.

#### *2.2. Methods*

All fossil analyses were performed in a permanently dedicated, fossil-only laboratory at North Carolina State University (NCSU; see Schroeter et al. [24] for additional details on lab sterilization protocols). For the biomolecular analyses, modern control trials were conducted in a separate lab at NCSU. Over the duration of this project, demineralization was carried out by separate project personnel in three separate labs: one sample (*Thoracosaurus* femur RU-EFP-00006-11) was demineralized at NCSU, seven samples were demineralized in a designated, sterilized, fossil-only chemical fume hood at Drexel University, and four were demineralized in a similarly dedicated, sterilized chemical fume hood at Rowan University. Negative controls, which included buffer-only solutions and sediments demineralized and/or co-extracted in the same reagents, were evaluated in tandem with the fossil for all assays. For immunoassays, specificity controls were also conducted (see below). Procedures for all controls and modern/fossil samples were performed in exactly the same manner unless otherwise noted. Three replicates were completed of each assay.

**Figure 1.** Thin sections of specimens used for demineralization illustrate the varying degrees of bone microstructure preservation. (**A**) *Thoracosaurus* femur (RU-EFP-00006-11) femur showing wellpreserved bone microstructure including Sharpey's fibers (Histologic Index [HI] = 5). (**B**) Indeterminate crocodilian tibia (RU-EFP-00030) showing little bone microstructure preserved (HI = 1). (**C**) *Euclastes* costal prong (RU-EFP-00018) showing excellent preservation (HI = 5). (**D**) *Taphrosphys* peripheral (RU-EFP-02175) showed very poorly preserved microstructure (HI in this region = 1; HI for the entire thin section = 2). (**E**) Indet. Pan-Cheloniid peripheral (RU-EFP-02245) showed well preserved internal cortex but a highly altered outer cortex (HI = 4). (**F**) Juvenile *Taphrosphys* costal (RU-EFP-02222) with well-preserved microstructure (HI of this region and the entire thin section = 4–5). White arrows indicate vascular canals, yellow arrows indicate lines of arrested growth, and blue arrows indicate primary osteons. The red bracket corresponds to external rims of light color and poor histological integrity.

Collagen I, a structural protein abundant in bone tissue, was selected as the primary target for biomolecular investigation in this study because of its high preservation potential. Many factors contribute to this high preservation potential, including: (1) that it is the most abundant protein in bone [40–43]; (2) its primary, secondary, and tertiary structures make it highly stable [25,44,45] and resistant to many proteases [46], and (3) its intimate association with hydroxyapatite crystallites within the tissue structure of bone, (e.g., [25,40,41,47,48]).

#### 2.2.1. Histology

Standard petrographic thin sections were prepared for histological analysis (see [49]). Histological analysis was performed on eight of the twelve specimens that were demineralized (three crocodilian and five turtle samples; see Table 2 for specimen specifics) to acquire a representative dataset of histological variability among specimens from EFP. Bone fragments were embedded in Silmar 41 resin (U.S. Composites), thick sectioned, and mounted to frosted glass slides with Loctite® Heavy Duty 60 Minute epoxy (Westlake, OH, USA). Thin sections were then ground to an appropriate thickness (~100 μm) and polished. The Histological Index (HI) of Hedges and Millard [42] was used to qualify the degree of microstructural preservation.

#### 2.2.2. X-ray Diffraction (XRD)

XRD analyses were conducted using a Phillips X'Pert diffractometer (#DY1738) at the Department of Earth and Environmental Science at the University of Pennsylvania. Approximately 1–2 g of two fossil bone samples (one light tan, one dark purple-brown) and Hornerstown Formation sediment were each mechanically powderized to less than 10 μm in a SPEX tungsten carbide Mixer-Mills (model #8000). Each powdered sample was then loaded into a sample holder and analyzed using Cu Kα radiation (λ = 1.54178A◦) and operating at 45 kV and 40 mA. Diffraction patterns were measured from 5–75◦ 2θ with a step size of 0.017◦ 2θ and 1.3 s per step (=0.77 degrees per minute). Phillips proprietary software HighScore Plus v. 3.0e was used to interpret the resulting diffraction patterns.

#### 2.2.3. Demineralization

A roughly 0.5 cm<sup>3</sup> fragment of each fossil was initially submerged in 0.5 M disodium ethylenediaminetetraacetic acid (EDTA) to chelate calcium. This solution was changed every other day for approximately four weeks, then weekly for another 1–2 months, as needed. In some cases, demineralization of fossils from this locality was found to occur at a drastically slower rate than we have previously encountered for other similarly aged specimens, including others from marine sediments [11,34,36]. After 1–2 months in EDTA, if demineralization was still mostly ineffective, then as necessary samples were demineralized by acid dissolution in 0.2 M or 0.5 M hydrochloric acid (HCl) for several days to two weeks, with solution changes made every 1–2 days, depending on the resilience of the fossil bone matrix. Demineralization products were transferred to glass slides and visualized using a Zeiss Stemi 2000-C reflected light microscope (Oberkochen, Germany) with a linked AxioCam 506 camera (Zeiss, Oberkochen, Germany). Abundances of osteocytes, vessels, and fibrous matrix fragments from each specimen were assigned to five relative categories based on Ullmann et al. [11]: absent, rare, uncommon, frequent, and abundant. Modern alligator and Hornerstown Formation sediment samples were also demineralized following the same procedures, to serve as positive and negative controls, respectively.

#### 2.2.4. Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDX)

Osteocytes isolated by demineralization in 0.2 M HCl from RU-EFP-00006-11 were collected using a 1 μm filter (Millipore), which was then placed on an aluminum stub and allowed to dry. The resulting uncoated sample was imaged using a field emissionscanning electron microscope (FE-SEM Zeiss Supra 50VP, Oberkochen, Germany) at Drexel University, operating at a working distance of 6.6 mm and at an accelerating voltage of 1 kV. Elemental spot analyses were collected as a standardless assay using a coupled Oxford model 7430 EDS INCAx microanalyzer (Oxford Instruments, Abingdon, UK).

#### 2.2.5. Protein Extraction

Herein, we used the same protein extraction protocol as Ullmann et al. [20]. In brief, ground bone and sediments, along with a buffer/"blank" control of only the extraction solutions, were each separately extracted with 0.6 M HCl. The resulting supernatants were collected (here called the "HCl fraction"), then the remaining pellet was incubated in 0.05 M ammonium bicarbonate (AMBIC). The resulting supernatants were then also collected (here called the "AMBIC fraction"). HCl fractions were precipitated with trichloroacetic acid (TCA) to form pellets, which were then allowed to dry in a laminar flow hood. AMBIC fractions were dried in a speed vacuum. Samples were stored at either 4 ◦C, −20 ◦C, or −80 ◦C depending on the length of time until analysis. Concentrations are reported based on amounts of pre-extracted bone rather than absolute masses because post-extraction yields from this protocol contain salts from extraction buffers in addition to varying amounts of protein [50]. Fresh *Alligator* cortical bone was extracted in the same manner as the fossil using the same reagents and the same protocol, but in a separate lab at NCSU.

#### 2.2.6. Polyacrylamide Gel Electrophoresis (PAGE) with Silver-Staining

To test for the presence of organics within the extracts from *Thoracosaurus* femur RU-EFP-00006-11, we followed the silver-staining protocols of Zheng and Schweitzer [51], as also detailed in Ullmann et al. [20], but without treatment with the iron chelator pyridoxal isonicotinic hydrazide as it was not found to be necessary (refer to the Supplementary Materials for details).

#### 2.2.7. Enzyme-Linked Immunosorbant Assay (ELISA)

Herein, we used the same ELISA protocol as Ullmann et al. [20]. This included plating of AMBIC extracts at a concentration equivalent to 200 mg of pre-extracted bone per well, blocking of non-specific binding, and incubation with rabbit anti-*Alligator* purifiedskin collagen antibodies. Extracts from an extant *Alligator* were analyzed using the same protocols (in a separate lab at NCSU) as a modern control (see the Supplementary Materials for further details).

#### 2.2.8. Immunofluorescence

We followed the fossil tissue embedding and immunofluorescence procedures of Schweitzer et al. [4,15], which are detailed in Zheng and Schweitzer [51]. This included embedding of RU-EFP-00006-11 demineralization products in LR WhiteTM resin (Electron Microscopy Sciences, Hatfield, PA, USA), gathering of 200 nm sections with an ultramicrotome, and incubation with rabbit anti-chicken collagen I antibodies. Inhibition and collagenase digestion specificity controls, as well as a modern *Alligator* control, were also conducted (see the Supplementary Materials for further details).

#### **3. Results**

#### *3.1. Histology*

Histological investigation of the eight sectioned specimens revealed variable preservation of original bone microstructure among and within samples. The gross morphology of all fossil specimens examined herein is well preserved. However, the external surface of nearly half of the specimens is light in color while that of the other half is dark (Table 2). Variation in color even occurs among bones of the same individuals, such as crocodile RU-EFP-00030 (Table 2). Areas of disrupted microstructure correspond to areas of bone that are light in color under gross morphological investigation, which comprise varying amounts of the total bone thickness/volume (see below). This light alteration is common in fossil bones from this locality and has been previously noted by others ([52]). In long bones, these areas of discoloration occur on the outer cortical layer, (e.g., *Thoracosaurus* femur RU-EFP-00006-11; Figure 2) or extend the entire way across the cortex, (e.g., crocodile tibia RU-EFP-00030-1). Tabular-shaped bones exhibit the same pattern of surficial discoloration

and disrupted microstructure; as in long bones, these alteration effects may be restricted to near the cortical margin, (e.g., *Thoracosaurus* scute RU-EFP-00006-8) or pervasive through the entire cross-section of a specimen, (e.g., *Taphrosphys* peripheral RU-EFP-02175). Within these discolored regions, occasional Wedel tunnels are present and the original microstructure is completely or almost completely obliterated, (e.g., Figure 1B,D). In contrast, the darker-colored areas of fossil bones exhibit well-preserved microstructure, with primary and secondary osteons, lines of arrested growth, and osteocyte lacunae clearly visible, (e.g., Figure 1A,C,F).

**Figure 2.** Examples of the two bone colors observed in specimens from EFP: an outer rim of lightcolored bone (indicated by red brackets) and a darker interior. (**A**) Hand sample of a bone fragment from EFP. Note the light-coloration of the lower surface of the bone (directly above the scale bar). (**B**,**C**) *Thoracosaurus neocesariensis* femur (RU-EFP-00006-11) in (**B**) thick section and (**C**) thin section. Images (**B**,**C**) are not from the same position along the circumference of this bone, and the light-colored rim is not a consistent thickness along the entire circumference. Scale bars as indicated.

The microstructure of RU-EFP-00006-11 is well preserved (HI = 5) except for its outermost 50–250 μm (Figures 1A and 2). In this outer rim, the bone is light in color and the microstructure is completely obliterated (HI = 0). A few Wedl tunnels extend from this poorly preserved rim into the underlying, darker, better-preserved cortex (Figure S2A). The rest of the 5 mm of total cortical thickness is dark in color and well-preserved histologically. The outer cortex is composed of lamellar-zonal bone with at least 17 lines of arrested growth present. The innermost cortex is composed of compact coarse-cancellous bone. Osteocyte lacunae with short branching canaliculi are visible throughout the dark regions of the cortex. Small, longitudinal vascular canals are also present and in some instances are partially infilled with authigenic gypsum and pyrite.

Microstructure preservation among specimens varied, with some being well preserved, (e.g., RU-EFP-00006-11) and others poorly preserved (HI = 1 or 2). Preservation at the microscopic level does not correlate with macroscopic preservation, (i.e., well-preserved gross morphology does not always equate to high histologic integrity) or cortical thickness of a specimen. Specimens RU-EFP-02245, RU-EFP-00006-8, and RU-EFP-00002-2 exhibit a thin rim of this lighter, histologically degraded bone. In these specimens, as in RU-EFP-00006-11, the remainder of the cortex is dark in color and retains high histologic integrity. In contrast, crocodile tibia RU-EFP-00030-1 and *Taphrosphys* peripheral RU-EFP-02175 are entirely light in color with poor microstructure throughout, including large areas of bone tissue that appear completely obliterated by microbial destructions (possibly microscopical focal destructions [MFDs]; Figure S2B; [53]). These alterations are invisible to the naked eye; the gross morphology of RU-EFP-00030-1 is well-preserved with little evidence of bioerosion or damage to the periosteal surface to indicate its microstructure is destroyed. Histological analysis has yet to be conducted on crocodile humerus RU-EFP-00030-2, but thick sections of this bone (made for another study) exhibit an overall moderate level of discoloration between that of the light bone with poor microstructures and dark well-preserved bone seen in other specimens. This is in contrast to the costal

prong of RU-EFP-00018, which is dark throughout even though it has a very thin external cortex and an expansive cancellous region.

#### *3.2. X-ray Diffraction*

The diffractogram of a sample of RU-EFP-00006-11 identifies it to be composed of fluorapatite (Figure 3A), indicating it has been minorly altered from its original hydroxyapatite composition. Based on these results, combined with the generally well-preserved gross morphology and histology of RU-EFP-00006-11 and its production of relatively abundant demineralization products, we hypothesized that this specimen could be a favorable candidate for biomolecular analyses. The diffractogram of a light-colored, indeterminate crocodilian bone fragment from the upper Hornerstown Formation reveals that altered bone is also composed of fluorapatite. Both bone samples were distinct from Hornerstown Formation sediments, which were identified as glauconite.

**Figure 3.** Mineralogical analysis of sediment, dark bone, and light bone samples from EFP and SEM-EDS of a structure morphologically consistent with an osteocyte from RU-EFP-00006-11. (**A**) Diffractogram of XRD results showing distinct differences between the sediment and fossils, which were identified as glauconite and fluorapatite, respectively; (**B**) SEM micrograph of a structure morphologically consistent with an osteocyte; (**C**) EDS spectrum from center of osteocyte (square area) in (**B**). The large peak surrounding 0 keV is an artifact of the very low count rates from the osteocyte. Scale bar 4 μm in (**B**).

#### *3.3. Demineralization*

A few bone fragments, (e.g., RU-EFP-00002-2) were demineralized within a month in EDTA, but most remained relatively hard and brittle after six weeks and required further treatment in HCl before they could be imaged. In general, light-colored bone fragments demineralized more quickly than darker bone fragments and would develop a comparatively more granular and crumbly texture early in the process of breaking down (Figure S3B). Both demineralizations with EDTA and HCl yielded structures morphologically consistent with osteocytes, vessels, and fibrous matrix. Although further molecular testing is needed to confirm that these structures represent original cells/tissues, we hereafter refer to them as osteocytes and vessels for brevity. Our antibody assays identified the presence of endogenous collagen I in pieces of fibrous matrix (see below), supporting the endogenous nature of these tissue fragments. Osteocytes were observed fully embedded, partially isolated, or fully isolated from mineralized bone matrix. No sediment samples yielded any structures morphologically consistent with vertebrate cells or tissues; only, small, subrounded to angular glauconite and quartz grains were observed (Figure S3A). Several forms of osteocytes were recovered from fossil bone samples, varying in both cell body shape and complexity of branching of filopodia (Figure 4I–L). Morphologically, these osteocytes correspond to the stellate and flattened oblate morphologies as defined by Cadena and Schweitzer [27]. All morphotypes retain filopodia, some of which possess tertiary, quaternary, or even more ramifications in better preserved specimens, as present in osteocytes isolated from modern bone (Figure 4A). Most recovered "osteocytes" were red-brown in color, (e.g., Figure 4I–L), but a few samples, (e.g., RU-EFP-02175, RU-EFP-04161-8) yielded primarily dark brown to black osteocytes (from both EDTA and HCl demineralization, e.g., Figure S3C).

Following the qualitative categories of Ullmann et al. [11], seven of the twelve demineralized fossils produced Abundant osteocytes, and RU-EFP-00006-8 produced Frequent osteocytes. In the four remaining samples, three (RU-EFP-00030-2, RU-EFP-02175, and RU-EFP-02245) produced only rare osteocytes, and osteocytes were absent in RU-EFP-00030-1. In fact, RU-EFP-00030-1 did not yield any cellular or soft tissue microstructures (osteocytes, vessels, or fibrous matrix). Among all the specimens tested, only three (RU-EFP-00006-11, RU-EFP-04161-8, and RU-EFP-00002-2) yielded vessels, ranging from ~20–50 μm in diameter, and in all three cases their recovery was either frequent or rare. These same three specimens, as well as RU-EFP-00006-8, also yielded frequent to the rare fibrous matrix. Unlike many EFP samples, RU-EFP-00006-11 yielded all three types of microstructures. The fibrous matrix was more plentiful when this bone was demineralized with HCl, but both HCl and EDTA yielded only a few vessels.

#### *3.4. Scanning Electron Microscopy and Energy Dispersive X-ray Spectroscopy*

SEM of microfiltered RU-EFP-00006-11 demineralization products revealed threedimensional osteocytes with rough, fibrous surface texture over each cell body and short, branching filopodia (Figure 3B). Broken bases of filopodia appeared to exhibit brittle fractures and solid cross sections (not hollow). Regarding surface texture, similar rough, crisscrossing grooves on the surface of isolated osteocytes have previously been attributed to either the impression of collagen fibrils or microbial degradation [8,11,54]. EDX spot analyses found these microstructures to be composed of oxygen, iron, and carbon (Figure 3C), similar to the results of many other EDX studies of fossil bone demineralization products, (e.g., [8,11,55]. Under SEM, osteocytes from RU-EFP-00006-11 are indistinguishable from those of extant reptiles, (e.g., [27,56]).

**Figure 4.** Modern and fossil demineralization products. (**A**) Modern American alligator osteocyte. (**B**) Alligator blood vessels. (**C**) Alligator collagen matrix. (**D**,**E**) *Thoracosaurus* (RU-EFP-00006-11) blood vessels. (**F**) *Taphrosphys* (RU-EFP-00002-2) blood vessel. (**G**) *Thoracosaurus* (RU-EFP-00006-8) collagen matrix. (**H**) *Taphrosphys* (RU-EFP-00002-2) collagen matrix. (**I**) *Taphrosphys* (RU-EFP-02222) stellate osteocyte. (**J**) Testudines indet. (RU-EFP-04161-8) stellate osteocyte. (**K**) *Taphrosphys* (RU-EFP-04162) flattened oblate osteocyte. (**L**) *Euclastes* (RU-EFP-00018) flattened oblate osteocyte.

#### *3.5. Polyacrylamide Gel Electrophoresis with Silver-Staining*

After electrophoresis, the movement of the AMBIC fossil sample through the gel colored the lanes a light yellow-brown (Figure 5A) prior to the application of chemical staining. Incubation in fixing solution (50% methanol) reduced but did not eliminate this "pre-staining" coloration (see Supplementary Materials). Despite this, a drastic increase in color intensity was readily apparent across the entire examined a range of molecular weights after silver-staining (Figure 5B), indicating the presence of organics within the fossil bone AMBIC extracts. "Pre-staining" coloration was decreased by resuspending the portion of the bone pellet that did not dissolve into the solution used for the fossil lane and adding it as a second sample in another lane. Resuspended bone pellet lanes exhibited little to no "pre-staining" and a clear increase in signal intensity after development with silver nitrate (Figure 5A,B), again indicative of the presence of organics within this sample.

**Figure 5.** Polyacrylamide gel electrophoresis (PAGE) with silver-staining of fossil and modern samples. (**A**) PAGE of fossil samples prior to silver-staining, exhibiting pre-staining in the fossil bone lane. Fossil samples were loaded at 50mg of pre-extracted bone per lane; (**B**) same gel as (**A**) after development with silver nitrate. Both fossil sample lanes exhibit increased staining, whereas no binding is visible in sediment, Laemmli buffer, and extraction blank lanes; (**C**) silver-staining of modern *Alligator* samples, loaded at 20 μg/lane.

Extraction buffer blanks and Laemmli buffer controls exhibited no staining. Sediment controls exhibited weak "post-development" staining (after incubation with silver nitrate) only at the highest and lowest molecular weights examined (Figure 5). In a subset of replicates, sediment lanes exhibited a faint band near 50 kDa; no banding or increase in staining was discernible in any fossil or other control lane at this molecular weight, indicating it represents a type(s) of organics present only within the sediment controls. In separate gels, modern *Alligator* extracts imparted no "pre-staining" and yielded clear, distinct banding patterns with minimal smearing after development via silver staining (Figure 5C).

#### *3.6. Enzyme-Linked Immunosorbent Assay*

In all replicates, a positive signal for collagen I was identified in fossil AMBIC extracts by absorbance readings well over twice the background, (e.g., [57,58]). At 240 min, AMBIC extracts of RU-EFP-00006-11 reached an absorbance of 0.855 (Figure 6). At this same time point, sediment and extraction blank controls exhibited negligible absorbance. In all time point readings, the signal from the fossil bone was at least an order of magnitude higher than in all the negative controls. RU-EFP-00006-11 AMBIC extracts exhibited a significantly reduced signal relative to modern *Alligator* AMBIC extracts, which reached saturation (2.85) at ~150 min (Figure S4). At this same time point in the best replicate, the *Thoracosaurus* AMBIC sample reached an absorbance of 0.53.

**Figure 6.** Enzyme-linked immunosorbent assay results of fossil and sediment chemical extracts. Extraction buffers were also run as a negative control to test for contaminants. Fossil and sediment samples were loaded at 200 mg of pre-extracted weight. Blue columns represent absorbance at 405 nm at 240 min, with incubation in anti-chicken collagen I antibodies at a concentration of 1:400. Yellow columns represent the absorbance of the secondary-only controls for each sample. The fossil bone sample is over three times its secondary control and the sediment and extraction blank show low absorbance, indicating that they are not a source for the positive signal in the fossil sample. Error bars represent one standard deviation of the mean absorbance of each sample.

#### *3.7. Immunofluorescence*

Fossil demineralization products reacted positively with polyclonal anti-chicken collagen I antibodies. Fluorescence was only observed in primary antibody-incubated tissue sections and was well above the negligible background fluorescence of secondary-only controls (Figure 7A–C). Fluorescence was restricted to tissue pieces, appearing morphologically as a spotty pattern. Specificity controls show decreased to essentially no signal when primary antibodies were inhibited prior to incubation (a control for non-specific paratopes in the polyclonal primary antibodies; Figure 7D) and when tissue sections were digested with collagenase A for 3–6 h prior to exposure to primary antibodies (a control for non-specific binding of the primary antibodies to molecules other than the target protein; Figure 7E). As also found by Schroeter [59] and Ullmann et al. [20], and references therein), digestion with collagenase A for 1 h initially increased signal slightly.

Unexpectedly, modern *Alligator* bone demineralized with HCl exhibited relatively dimmer fluorescence than previous *Alligator* samples demineralized with EDTA and treated with the same primary antibodies (Figure 7F–H; cf. [59], Figure 4D). The expected binding pattern, showing visible bone tissue structures, (e.g., Haversian systems), also appeared patchier than has been observed in previous studies using this antibody (cf. [59], Figure 4D). This patchy binding obliterated most Haversian systems and other recognizable histologic features, leaving instead an irregular fluorescence pattern (Figure 7G,H) somewhat similar to that observed in the fossil *Thoracosaurus* bone (Figure 7B,C). However, despite these apparent artifacts of protocol-induced degradation (see Discussion), the modern *Alligator* tissue sections still exhibited stronger fluorescence than did the fossil samples. Inhibition and digestion controls for the modern samples (see Supplementary Materials), also each

dramatically diminished the signal (including after only 1 h of digestion; Figure 7I,J), confirming the specificity of antibody binding in our assays.

**Figure 7.** In situ immunofluorescence results for fossil and modern *Alligator* samples. All sections imaged at 200 ms exposure. (**A**,**F**) Secondary only negative-control tissue sections never exposed to primary antibodies; (**B**,**G**) tissue sections incubated with anti-chicken collagen I antibodies at 1:40 concentration; (**C**,**H**) overlays of fluorescence images in (**B**,**G**) on light microscope images, showing localization of fluorescence to tissue within sections (fluorescence shown as green coloration); (**D**,**I**) tissue sections treated with the same anti-chicken antibodies but exposed to *Alligator* collagen prior to incubation with sections ("inhibition control"); (**E**,**J**) tissues sections incubated with collagenase for 3 h prior to incubations with the same primary antibodies. Negative controls and specificity controls exhibit reduced or absent signals, indicating a lack of spurious binding and the presence of endogenous collagen I in each bone sample. Scale bars as indicated.

#### **4. Discussion**

Our findings cumulatively support the recovery of endogenous soft tissues and biomolecules from 63–66-million-year-old vertebrate fossils from EFP. As osteocytes were found only in fossil samples and still embedded within bony tissues, it is parsimonious to infer that these structures originated from the fossil material and not the environment. Additionally, multiple assays found only fossil samples to yield organics that reacted positively with antibodies raised against *Alligator* and chicken collagen I. Therefore, the unique depositional environment recorded by sediments of the Hornerstown Formation is the latest in the expanding list of paleoenvironmental settings shown to allow molecular preservation over geologic time.

Wiemann et al. [10] previously demineralized a vertebra from an indeterminate crocodilian (YPM 656) that was collected from the Navesink Formation, which underlies the Hornerstown Formation. In stark contrast to our results, demineralization of YPM 656 failed to yield any cells or soft tissues [10]. There are several possible reasons for this difference. No vertebrae were analyzed in the present study, but even the cancellous turtle shell bones examined herein yielded demineralization products (Table 2), implying that it is unlikely that factors related to skeletal element type alone could account for the lack of recovery of soft tissues from YPM 656 [10]. Although the Navesink Formation also contains iron-rich glauconite, it is in a lower percentage than the Hornerstown Formation [30]. It is thus possible that sediments of the Navesink Formation did not form as conducive of a diagenetic environment for molecular preservation as those of the Hornerstown Formation (cf., [33]). It is also possible that, as observed in the disparity of microstructural preservation between light and dark fossils recovered from the Hornerstown Formation, preservation in the Navesink may be variable as well. Histological examination of YPM 656 may show that its microstructure was poorly preserved despite its intact gross morphology, similar to the light tan fossils of the Hornerstown Formation that also often failed to yield demineralization products herein (Table 2). Additionally, Wiemann et al. [10] acquired their sample from

collections, meaning it was not collected fresh for the purposes of paleomolecular analyses. Though historical specimens in collections have yielded results, (e.g., [6]), it has been previously suggested that fresh samples yield better soft tissue and biomolecular recovery ([60] and references therein). However, at this time, it is not possible to concretely resolve which of these processes are responsible for the lack of recovery reported by Wiemann et al. [10].

At EFP, the majority of the Hornerstown Formation currently lies below the natural water table, (e.g., the MFL is positioned ~20 ft beneath it), and studies of regional sequence stratigraphy [61] imply that fossils within the Hornerstown Formation at this locality have likely spent tens of millions of years under saturated conditions. Despite this, we successfully recovered cells and soft tissues from numerous MFL fossils and collagen I from RU-EFP-00006-11. Thus, our findings corroborate other recent studies [3,6,10,16,35] which refute the traditional hypothesis that marine paleoenvironments are inconducive to biomolecular preservation due to hydrolysis caused by constant exposure to water [25,26]. As iron is hypothesized to aid in molecular preservation in many cases [3,32,33], it is possible that the high concentration of glauconite, a mineral rich in iron, aided in the preservation of these endogenous organics, (e.g., via iron free radical-induced molecular crosslinking [33]). It is also possible that the presence of abundant dissolved iron was responsible for some of the analytical challenges encountered herein (see below). Other authors [62] have alternatively suggested that iron may precipitate around cells and other soft tissue microstructures preserved via other pathways during early diagenesis, (e.g., alumino-silicification), forming a mineralized coating later in diagenesis. At this time, there is no evidence that these preservation pathways are mutually exclusive; indeed, at some localities, (e.g., [62]) both may contribute to long-term preservation at the molecular level.

#### *4.1. Overall Preservation*

Our histological analyses found light bone to be heavily degraded, having largely lost its original microstructure (HI = 0–2). This histologic alteration appears responsible for light bones yielding few cells and soft tissues upon demineralization (Table 2). For example, the only specimen to not yield any soft tissue products (crocodile tibia RU-EFP-00030-1) is entirely light in cross-section. The other specimen with this condition (*Taphrosphys* peripheral RU-EFP-02175) only yielded rare osteocytes, and crocodile humerus RU-EFP-00030-2 also only yielded rare osteocytes likely because it appears to be preserved in a state of partial degradation. Additionally, when present, the osteocytes recovered from light bones were typically poorly preserved, (i.e., with only short, stubby filopodia) compared to those from dark bones retaining well-preserved histology (HI = 4–5), which more frequently retain long, branching filopodia with multiple ramifications. The only other specimen to yield rare osteocytes (and no vessels or fibrous matrix) was turtle peripheral RU-EFP-02245. Although this specimen only exhibits a thin external rim of light bone, it is primarily composed of highly porous cancellous bone; its accordingly low volume of bone material may have reduced the chances of recovering soft tissues from this specimen, regardless of its level of preservation. All other specimens with a thin layer of light bone or composed entirely of dark bone yielded a consistently greater recovery of cellular and soft-tissue microstructures (Table 2).

Based on our results, there appears to be no association (for these 12 specimens) between soft tissue preservation and either taxon, skeletal element, quality of macroscopic preservation, cortical thickness, or bone tissue type, (i.e., cortical vs. cancellous bone). Of the limb bones sampled, only two yielded abundant cellular and soft tissue microstructures even though all are well-preserved at the gross-morphology level. At this locality, and based on current information, the best predictor of histologic and soft tissue preservation appears to be bone color: bone tissues that are dark are histologically well-preserved bones and generally yield far greater soft tissue recovery upon demineralization. A correlation between morphologic quality and molecular preservation has been suggested previously [25,38,39,63,64], and our results further support this hypothesis. However, it remains unclear what is causing differential degradation among bones preserved within

the same probable mass death assemblage [28,30] in the same horizon of the same geologic stratum. It is possible that the light degradation is a late-diagenetic artifact resulting from the modern acidic groundwater of southern New Jersey. Alternatively, these regions of poorly preserved histology within bone could result from microbial activity, as suggested by the presence in some specimens of potential MFDs and Wedl tunnels extending from the light outer layer into the underlying, better-preserved portions of dark bone (cf., [65–67]). This taphonomic question is currently under investigation as part of a separate study.

#### *4.2. Soft Tissue Preservation*

Demineralization in EDTA in this study proceeded more slowly and generally yielded fewer microstructures than from other similarly aged fossil bones we have tested [11,59]. Slow demineralization in EDTA was also found by Norris et al. [68] for a Permian *Dimetrodon* bone. As in our study, these authors were able to successfully demineralize their specimen in a solution containing HCl. The cause of such cases of slow demineralization in EDTA remains unknown. Powder XRD analyses identified RU-EFP-00006-11 as being primarily composed of fluorapatite (Figure 3A), indicating that micro-scale permineralization by secondary mineral phases containing non-divalent cations, (e.g., iron oxides with Fe3+ ions or quartz with Si4+ ions) which cannot be chelated by EDTA is an unlikely explanation. It is possible that significant substitution of trivalent cations for divalent Ca2+ ions in bone apatite could hinder the demineralization process (as EDTA only chelates divalent cations; [50]), but further analyses, (e.g., trace element analyses) would be required to evaluate this potential cause. HCl provides harsher, more acidic conditions for demineralization; therefore, demineralization trials employing this solution required significantly less time and yielded a greater recovery of soft tissues. However, as HCl incubation is also a step in our protein extraction protocol (see Supplementary Materials), incubation of fossil bone fragments in this acid may result in solubilization of proteins and other biomolecules which lack divalent cations. Although microstructures isolated by demineralization with HCl did not appear under transmitted light to visually differ from those recovered using EDTA, the effects of HCL versus EDTA demineralization warrants further investigation.

Osteocytes were the most abundant microstructures recovered from our EFP specimens. The common lack of recovery of microstructure morphologically consistent with blood vessels, at least in our crocodilian samples, may be due to the rarity of vessels in the original cortical/cancellous bone tissue. The cortex in modern crocodilians is not as vascular as that of non-avian dinosaurs [69–71] or modern birds; as a result, fossil crocodile bone would not be expected to yield as many structures morphologically consistent with vessels as would dinosaur bone. It should also be noted that although the external cortex of femur RU-EFP-00006-11 has been perforated by occasional Wedl tunnels, the vessel structures we recovered do not appear to represent biofilm coatings of these tunnels because they retained their structure after demineralization, and each possess a clear lumen [72]. Low recovery of the fibrous matrix may relate to the depositional environment, but more testing would be required to elucidate such a causal connection.

#### *4.3. Biomolecular Preservation*

Taken together, our molecular assays support the conclusion that the soft tissues and collagen I recovered from this specimen are endogenous. Though a PAGE with silver-stain assay is not a specific test for collagen I or proteins, it can identify the presence and molecular weights of organic compounds in a sample, making it an informative screening assay for paleomolecular studies [4,20,73]. Both fossil AMBIC extract lanes showed a distinct increase in coloration after silver-staining, whereas all negative control lanes (sediment, extraction blank, and Laemmli buffer) exhibited little to no coloration (Figure 5A,B). However, the prestaining of the gel likely points to diagenetic humic substances being present in this sample [74]. Humics can bind silver [75,76], which may have caused the increase in staining. Though this pattern of silver nitrate binding cannot directly support the presence of protein in the sample, this pattern does not falsify the presence of protein in this

fossil (as no staining would). Thus, we continued to analyze this sample with assays of greater specificity that are not susceptible to humics. As predicted, the modern *Alligator* extracts exhibited less smearing and better banding than the fossil extracts (Figure 5C; see Supplementary Materials for further discussion of our silver-staining results).

Given the recognition of unique, albeit non-specific, organics in RU-EFP-00006-11 by silver-staining, we next performed ELISA and in situ immunofluorescence as independent assays to identify if the primary structural protein collagen I was present in both wholebone extracts and demineralization products, respectively. These assays complement one another as ELISA is approximately an order of magnitude more sensitive than immunofluorescence [73,77], whereas immunofluorescence allows in situ localization of epitopes in native tissues [15,73]. Fossil extracts in ELISA assays yielded well over double the absorbance of the sediment and extraction buffer controls (Figure 6), supporting the presence of endogenous collagen I in RU-EFP-00006-11. Negative absorbance values for sediment and extraction controls indicate these samples were less reactive than the group blank (PBS only), demonstrating the collagen signal in bone wells is not attributable to these sources of possible contamination. Additionally, the absorbance values of all samples incubated with only secondary antibodies were negligible, (i.e., significantly less than twice the absorbance of the fossil samples incubated with both primary and secondary antibodies), removing non-specific secondary antibody binding (such as to humic substances) as a cause for the high positive signal for the fossil bone extracts. Thus, this assay supports the presence of endogenous collagen I in the femur of this *Thoracosaurus*.

Modern *Alligator* tissue samples analyzed by in situ immunofluorescence exhibited fluorescence when exposed to polyclonal antibodies raised against chicken collagen I (Figure 7G,H). Together with the reduced signals observed in our specificity controls (Figure 7I,J; Supplementary Materials), these results agree with those of Schroeter [59] who concluded that the collagen I epitopes in these two extant archosaurs are conserved enough in structure to each be recognized by the primary antibodies used herein. Because these antibodies can successfully detect collagen I in modern *Alligator*, they were predicted to also bind to fossil crocodilian collagen I. The fluorescence signal exhibited by the fossil tissues was lower in intensity compared to the modern *Alligator* (Figure 7B,C), yet still far brighter than in secondary-only controls (Figure 7A). Fluorescence in fossil tissue sections decreased in all specificity controls (Figure 7D,E; Supplementary Materials), again supporting the specific binding of the primary antibodies to epitopes of collagen I. Both modern and fossil tissues exhibited patchy binding patterns, possibly due to tissue degradation from demineralization with HCl or its ability to solubilize proteins (as in our extraction protocol), as well as decay inherent in fossilization for RU-EFP-00006-11.

#### **5. Conclusions**

Our paleomolecular investigations of fossils preserved in the glauconitic, shallowmarine depositional environment of the Hornerstown Formation at EFP add a unique paleoenvironment to the growing list of those which have yielded biomolecular and soft tissue preservation within fossil bones. This depositional environment was rich in iron at the time of burial and remained rich in iron due to the glauconite-dominated composition of the sediments and influx of dissolved iron from recent groundwaters. We hypothesize that this abundant supply of iron may have facilitated soft tissue and biomolecular preservation in these specimens via the free radical-induced crosslinking reactions elucidated by Boatman et al. [33]. Soft tissues were recovered by demineralization with HCL and EDTA, though HCL treatments took less time and were required for some samples. Though all specimens we examined are well-preserved in terms of gross morphology, multiple exhibit varying amounts of alteration in the form of light tan-colored bone which was found to be histologically degraded and thus yielded minimal cells and soft tissues upon demineralization. Association of poor histological preservation with poor soft tissue recovery is logical, and it appears that at EFP this alteration may have occurred, at least in part, by microbial degradation. All biomolecular assays completed on RU-EFP-00006-11 support the presence

of endogenous collagen I in this *Thoracosaurus* femur. PAGE with silver-staining exhibited evidence for organics in this fossil, and ELISA and in situ immunofluorescence results each independently support retention of collagen I in this specimen. Collectively, our results and those of previous authors [3,6,10,16,35] support the conclusions of Nielsen-Marsh and Hedges [78] and Hedges [38] that relatively constant immersion in water may not preclude endogenous molecular preservation in fossil bones. As a result, there are many additional paleoenvironments yet to be explored that may preserve endogenous biomolecules and soft tissues.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/biology11081161/s1, Supplementary Materials narrative: a DOCX Word file including additional description of methods and continued discussion. Figure S1: Gross morphology of representative fossil specimens examined herein. Figure S2: Examples of microbial bioerosion in histologic thin section. Figure S3: Additional demineralization results. Figure S4: Comparison of enzyme-linked immunosorbent assay results of fossil bone, sediment, and modern *Alligator* chemical extracts. References [79–84] are cited in the Supplementary Materials narrative.

**Author Contributions:** Conceptualization, K.K.V. and K.J.L.; methodology, K.K.V. and E.R.S.; formal analysis, K.K.V., Z.M.B. and P.V.U.; investigation, K.K.V., Z.M.B., P.V.U. and W.Z.; data curation, K.K.V., Z.M.B. and P.V.U.; writing—original draft preparation, K.K.V.; writing—review and editing, K.K.V., Z.M.B., P.V.U. and E.R.S.; visualization, K.K.V. and Z.M.B.; project administration, K.K.V.; funding acquisition, K.K.V. and P.V.U. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by NSF GRFP, DGE Award 1002809, and Rowan University.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** All data generated by this study are available in this manuscript and the accompanying Supplementary Materials.

**Acknowledgments:** Thank you to M. Schweitzer for fruitful discussions during planning and data collection as well as access to facilities and J. Tabacco for assistance in data collection. Additionally, we are grateful for the support and assistance provided by the Inversand Mining Company (Clayton, NJ, USA) for the collection of specimens prior to 2016. We would also like to thank the editor and four reviewers for their suggested improvements to the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Environmental Factors Affecting Feather Taphonomy**

**Mary Higby Schweitzer 1,2,3,4,\*, Wenxia Zheng <sup>1</sup> and Nancy Equall <sup>5</sup>**


**Simple Summary:** This study seeks to test the effect of burial/exposure, sediment type, the addition of feather-degrading microbes, and the addition of minerals on feather preservation, and for the first time, compares these states in ambient vs. elevated CO2 atmospheres to test the effect of CO2 on degradation and/or preservation under various depositional settings.

**Abstract:** The exceptional preservation of feathers in the fossil record has led to a better understanding of both phylogeny and evolution. Here we address factors that may have contributed to the preservation of feathers in ancient organisms using experimental taphonomy. We show that the atmospheres of the Mesozoic, known to be elevated in both CO2 and with temperatures above present levels, may have contributed to the preservation of these soft tissues by facilitating rapid precipitation of hydroxy- or carbonate hydroxyapatite, thus outpacing natural degradative processes. Data also support that that microbial degradation was enhanced in elevated CO2, but mineral deposition was also enhanced, contributing to preservation by stabilizing the organic components of feathers.

**Keywords:** feather; taphonomy; degradation; keratin; microbes; CO2; apatite; melanin

#### **1. Introduction**

Feathers are arguably the most complex integumentary structures in the entire animal kingdom. The evolutionary origins of feathers are still debated, but growing evidence from both molecular studies in extinct theropods [1–8] and living birds (e.g., [9–18]), as well as numerous fossil discoveries of structures morphologically consistent with feathers (e.g., [4,19–25]) indicate that feathers arose from filamentous structures first identifed in some theropod dinosaurs and birds more than 160 million years ago (e.g., [2,26,27]). However, some data suggest that integumentary structures similar to those from which feathers derived may have been present at the base of Dinosauria [28,29] or perhaps, the base of Archosauria ([30,31] and references therein). Because modern feathers are not biomineralized in life (contra [32,33]) their persistence in the fossil record is counterintuitive, but critical. The impressions of feathers in sediments surrounding skeletal elements led to the identification of *Archaeopteryx* as the first bird [34,35], but there was no organic trace with this specimen to suggest that any original material remained. However, the first specimen attributed to *Archaeopteryx* was a single, isolated feather [36]. This specimen presented differently from feather impressions surrounding the skeletal remains, instead visualized as a carbonized trace clearly distinct from the embedding sediments, suggesting that taphonomic processes resulting in preservation differed between the isolated feather and the skeletal specimen. The environmental factors resulting in these different modes of preservation remain relatively unexplored.

For any organic remains to be preserved in deep time, they must be stabilized before they can degrade [37]. Although many taphonomic modes may result in the preservation

**Citation:** Schweitzer, M.H.; Zheng, W.; Equall, N. Environmental Factors Affecting Feather Taphonomy. *Biology* **2022**, *11*, 703. https://doi.org/ 10.3390/biology11050703

Academic Editor: Andreas Wagner

Received: 27 February 2022 Accepted: 29 April 2022 Published: 3 May 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

of feathers (e.g., carbonized film [36,38], sediment impressions [34,35], three-dimensional filaments [3,39], amber preservation [40,41], bacterial mediation [42,43], or other stabilizing processes (e.g., [42,44–46])), few of these have been subjected to rigorous experimentation, particularly at the molecular level (but see [47]). It is likely that preservation processes differ for every fossil element and/or environment and arise from a complex interplay between the molecular composition of the original structures and the geochemical properties of the surrounding depositional matrix. Factors contributing to feather preservation are testable by approximating naturally occurring conditions in laboratory experiments.

Here, we examine multiple environmental factors that may, to varying degrees, affect the preservation of feathers in the fossil record. We discontinued the experiments after six weeks, because previous experiments (feathers degraded in sandy settings with added microbes) have shown that within this time period, the degradation of feathers was almost complete (unpublished data). We tested feathers buried vs. unburied; soaked in natural pondwater, sterile water, or pondwater after incubation with feather-degrading bacteria; the addition of solubilized hydroxyapatite (HA) to waters; and burial in sand vs. mud. Finally, we repeated the experiment in atmospheres elevated in CO2. Although the value we employed is higher than proposed for the entire Mesozoic, it is consistent with values proposed for the Ypresian [48]. Thus, this value is in line with what has occurred without human intervention, allowing us to test the direct effect of CO2 in a shorter time period. We observed that in at least one case, there was no material remaining to be tested at six weeks; degradation was complete.

#### **2. Methods**

We subjected the black and white feathers of an extant magpie (*Pica hudsonia*) to these environmental conditions to test their effect on preservation and/or degradation (see Table 1). These feathers allowed us to also test the hypothesis that melanized feather regions would show greater preservation than un-melanized ones [49]. See supplemental documents for additional experimental details.

**Table 1.** Experimental conditions to test their effect on preservation and/or degradation.


We hypothesized that rate, degree, and pattern of the degradation of feathers might differ in atmospheres elevated in CO2 relative to today's levels, so we divided all experimental conditions into "elevated CO2" and "ambient" atmospheres. Within these two environments, we tested the effect of burial vs. surface exposure; native microbial populations in surface waters (pondwater) or the addition of feather-degrading microbes (*B. licheniformis*) to those microbes naturally present in pondwater; clean porous sand (No. 1113 Premium Play Sand, Quikrete) vs. natural pond sediments, which were a silty-to-clay mix; and an addition of 6 mM calcium hydroxylapatite (HA; Ca10(PO4)6(OH)2), (0.1 MCaCl2 solution mixed with NaH2PO4 (0.1 M) with volume ratio 10:6)) [50] to mimic the higher concentration of this mineral within pore waters that occurs under elevated CO2 and concurrent lower pH [51–53]. This would be more likely to occur in Mesozoic than present-day waters because of the increased acidification and resulting mineral solubilization brought on by elevated CO2 [52–55]. To test the role of microbes in mediating HA mobilization, we added this mineral to both E-Pure water and pondwater before adding to feathers. Although we used E-Pure water in the last two conditions as controls, some microbes were present, arising from the natural sediments and/or the feathers themselves.

Table 1 illustrates the conditions tested and the abbreviations used in further descriptions. Feathers were cut into sections, marked for periodic recovery, and sampled every two weeks (see supplemental information and Figure S1 for more information). Only data from week six are shown herein, except for buried feathers with added microbes in elevated CO2 (CPBB). Under this condition, no material remained for testing; we show data from the four-week timepoint for this condition in all figures except SEM images for elevated CO2 conditions. When we could discern original color, it was noted in the figure legend.

In addition to micromorphological changes, we tested the effect of degradation on antibody binding using a polyclonal antibody raised against extant mature white feathers, which are comprised almost completely of feather corneous β-protein (CβK [39]). In all cases, there was a bacterial component to the experimental conditions, including the E-Pure water conditions, because feathers still rested on unsterilized sediments infiltrated with bacteria in pondwaters or present on the feathers themselves. This is appropriate, as there are no naturally occurring environments that are devoid of bacterial influence. Normal microbial flora would be expected to be present in waters, sediment/sand, and the feathers themselves in all conditions.

#### **3. Results**

#### *3.1. Transmitted Light Microscopy (LM)*

Figure 1 shows feathers subjected to the above conditions and subsequently imaged using LM under ambient atmosphere (normal air, at room temperature), subjected to the above conditions. Images were taken after six weeks of degradation. For each condition (row), the two left-most panels represent buried feathers, and the right two panels are feathers exposed at the surface.

Under ambient conditions, feathers showed little obvious degradation after two weeks (not shown), but at six weeks, in most cases, degradation was obvious in both buried and exposed feathers with natural pondwater, and greater when *B. licheniformis* was added. In the former (PB, PE; Figure 1A–D), white feather regions that were buried showed more extensive fraying than observed for black feathers, and the buried feathers were more degraded than those exposed at the surface. Both black and white regions of the feathers were significantly degraded in the buried condition (Figure 1A,B), but in the unburied condition, the black regions of the feather were generally less degraded than the white regions (Figure 1C,D). In some regions of the white buried feather, no barbs could be seen. The white regions of the rachis were crumbling and covered in a fine, crystalline material. When *B. licheniformis* was added to the feather setup (Figure 1E–H), degradation was increased over the pondwater-only condition. In general, degradation, as measured by fraying, absence of barbs, and loss of integrity, was greater when feathers were buried, and visual inspection showed that white feather regions suffered greater degradation than did black ones, although white rachises were relatively intact.

When HA was added to pondwaters, preservation was enhanced in both buried (PMB) and unburied (PME) states (Figure 1I–L). In all cases, the barbs and barbules were intact and color was still discernible. Similarly, there was virtually no degradation visible in feathers exposed to E-Pure water with added HA (ESMB/E), even after six weeks (Figure 1M–P). The barbs, barbules, and vane structures were intact in both the white and black regions, both buried (M,N) and unburied (O,P), despite the presence of native microbes in the sediments and water. This was also seen in the feathers incubated with E-Pure water

without added minerals (ESB/E, Figure 1Q–T), although some fraying of the white barbs was visible in the exposed regions (Figure 1T).

**Figure 1.** Feathers degraded in ambient atmospheres after 6 weeks. (**A**,**B**) Buried and (**C**,**D**) exposed in natural pondwater; (**E**,**F**) buried and (**G**,**H**) exposed feathers with pondwater and added microbes; (**I**,**J**) buried and (**K**,**L**) exposed feathers with pondwater and added HA; (**M**,**N**) buried and (**O**,**P**) exposed feathers with E-Pure water and HA; (**Q**,**R**) buried and (**S**,**T**) exposed feathers with E-Pure water only. Scale is 500 μm for the 1st and 3rd columns, and 200 μm for the 2nd and 4th columns.

Figure 2 shows an experimental set-up identical to Figure 1, but degradation was conducted in an elevated CO2 atmosphere (5000 ppm), reflecting estimates of some of the highest levels of naturally occurring (i.e., non-anthropogenic) atmospheric CO2 of the Phanerozoic (e.g., [48]). In each row of images taken after six weeks (except panel E,F, at four weeks), the first two panels were buried and the second two exposed on the surface, as above.

Generally, under light microscopy, degradation appeared more intense in the elevated CO2 atmosphere than ambient in CP or CPB conditions. Buried feathers were degraded to completion in CPBB and CPBE, possibly indicating an upregulation of urease or carbonic anhydrase enzyme production by these microbes under elevated CO2 [56]. Because no feathers were recovered at the six-week endpoint, here we include the four-week buried data point (Figure 2E,F). In all other cases, both black and white regions of the feather could still be differentiated under transmitted light; barbs and barbules appeared frayed by the sixth week but were still intact. Buried feathers appeared slightly more degraded than unburied ones, as seen in ambient atmospheres as well, and black regions of feathers in

CPB/CPE (Figure 2A–D) and CPBB/CPBE (Figure 2E,H) were slightly less degraded than white feathers.

**Figure 2.** Feathers degraded in elevated CO2 atmospheres after 6 weeks. (**A**,**B**) Buried and (**C**,**D**) exposed with natural pondwater; (**E**,**F**) buried (4 weeks) and (**G**,**H**) exposed feathers with pondwater and *B. licheniformis*; (**I**,**J**) buried and (**K**,**L**) exposed feathers with pondwater and added HA; (**M**,**N**) buried and (**O**,**P**) exposed feathers with E-Pure water and HA; (**Q**,**R**) buried and (**S**,**T**) exposed feathers with E-Pure water only. Scale bar for 1st and 3rd columns is 500 μm, and for the 2nd and 4th, 200 μm.

#### *3.2. Scanning Electron Microscopy (SEM)*

We used SEM to test the hypothesis that black feathers are more resistant than white feathers under the conditions described above. The SEM images in Figures 3 and 4 were taken after six weeks of degradation. The conditions under which data were collected followed what is shown in Figures 1 and 2. Feathers were differentiated according to color, when possible to discern. Figure 3 represents degradation in ambient atmosphere; Figure 4 is feathers degraded in elevated CO2 atmospheres.

**Figure 3.** Feathers degraded under ambient conditions. (**A**–**D**) Buried in pondwater; (**A**,**B**) are black and (**C**,**D**) are white. (**E**–**H**) are feathers exposed at the surface; colors are not discernable. (**I**–**L**) Feathers buried in pondwater to which *B. licheniformis* has been added. (**M**–**P**) Surface-exposed feathers in pondwater with *B. licheniformis*; color is not discernible in (**I**–**P**). Arrows show microbial tunneling. (**Q**–**T**) represent feathers buried in pondwater with added HA; (**Q**,**R**) are black and (**S**,**T**) are white. A film (#) covers some regions; also (**U**–**X**) are exposed feathers with pondwater and HA; colors are not discernible. (**Y**–**BB**) are feathers buried with E-Pure water and HA; (**Y**,**Z**) are black and (**AA**,**BB**) are white. (**CC**–**FF**) are exposed feathers in E-Pure water with added HA; (**CC**,**DD**) are black and (**EE**,**FF**) are white. Finally, (**GG**–**JJ**) are buried and (**KK**–**NN**) are exposed in E-Pure water only; (**GG**,**HH**,**KK**,**LL**) are black and (**II**,**JJ**,**MM**,**NN**) are white. Scale as indicated in μm. kf, keratin filaments; bf, biofilm; m, microbial bodies; v, voids; F, fungal hyphae.

**Figure 4.** Feathers degraded for 6 weeks under elevated CO2.4(**A**–**D**) are buried in pondwater, color is not discernible. (**E**–**H**) are surface-exposed; (**E**,**F**) are black and (**G**,**H**) are white. Feathers buried in pondwater to which *B. licheniformis* was added did not persist to 6 weeks and no data are shown. (**M**–**P**) Surface-exposed feathers in pondwater with *B. licheniformis*. Color is only discernible in (**N**), which is black. (**Q**–**T**) represent feathers buried in pondwater with added HA; all feathers shown are black. (**U**–**X**) are surface-exposed with pondwater and HA; (**U**,**V**) are black and (**W**,**X**) are white. (**Y**–**BB**) are feathers buried with E-Pure water and HA; all are white. (**CC**–**FF**) are exposed feathers in E-Pure water with added HA; (**CC**,**DD**) are black and (**EE**,**FF**) are white. (**GG**–**NN**) Incubated in E-Pure water only; (**GG**–**JJ**) are buried and (**KK**–**NN**) are surface-exposed. (**GG**,**HH**,**KK**,**LL**) are black, (**II**,**JJ**) are white, and (**MM**,**NN**) are indeterminate. Scales as indicated, in μm. kf, keratin filaments; bf, biofilm; m, microbial bodies; v, voids; F, fungal hyphae; ˆ, amorphous film.

In ambient atmosphere, feathers degraded for six weeks in pondwater showed little difference between buried (PB, Figure 3A–D) and exposed (PE, Figure 3E–H). Highly aligned, confluent microbial mats (m) could be seen on the surface of the fragmented keratinous outer cover in both black (3A–B) and white (3C–D) feathers in both PB and PE samples. Fungal hyphae (F) could also be seen. Occasionally, mineral crystals (Figure 3A, arrowheads) could be seen.

When feathers were exposed on the surface (Figure 3E–H), degradation appeared to be slightly less, but confluent, ordered microbial bodies (m) were seen on the surface of the feathers in all cases. The keratinous outer sheaths were less fragmented, but invasive microbial populations were still visible. Microbial impressions, or "voids" (V), were visible on the surface of the keratinous sheath (Figure 3F,G). Figure 3H shows a bundle of fraying keratin fibers (kf). Mineral crystals and/or diatoms were also visible (Figure 3G, arrowheads). The keratin sheath was densely pitted (Figure 3F, left).

When feathers were soaked in pure cultures of *B. licheniformis* and then added to the normal pondwater flora, degradation was greatly intensified in both the buried (PBB, I–L) and unburied (PBE, M–P) feathers. The keratinous material from the buried samples was fragmented and riddled with holes. Microbial bodies (m) were visible in all panels. In both PBB and PBE, degradation followed a different pattern, clearly visible as microbial tunneling (arrows, Figure 3M–P). This was not observed unless *B. licheniformis* were added, and may be specific to this microbial group. Microbial bodies were not as readily visible in the exposed feathers relative to the buried samples. Bundles of keratin fibers (kf) were loose and ropy. Small regions of apparent crystal growth (arrowheads) can be seen in in direct association with the microbes in Figure 3I.

Buried feathers in natural pondwaters to which HA had been added (Figure 3Q–T) showed better preservation than previous conditions. Barbs (B) and barbules (BL) were still visible (Figure 3Q) and accumulations of geometric crystal growth (arrowheads) were seen on all feather surfaces, both black (Figure 3Q,R) and white (Figure 3S,T). In some cases, an amorphous film could also be seen coating feather surfaces (Figure 3S,Y,Z, (#)). However, deep in the crystal growth and amorphous film, fibers of keratin were still aligned and appeared intact. The unburied feathers with added HA (Figure 3U–X) also showed better preservation than the previous conditions. Feather barbs (B) and barbules (BL) were visible, and showed virtually no degradation; they appeared to be coated in mineral crystals, which may have stabilized the organic structures [57–59]. Fungal hyphae could be seen (Figure 3V,W (F)) but microbial bodies were rare in this condition. Figure 3X shows a fibrous mat of material (fm), the identity of which is uncertain. It was not possible to differentiate black and white feathers.

Buried feathers incubated with E-Pure water with added HA (ESMB) under ambient conditions showed virtually no degradation (Figure 3Y–BB). In some regions of these feathers, fungal hyphae were prevalent (Figure 3Y, (f)), and the same amorphous deposits (#) and mineral crystals (arrowheads) were visible on the feather surfaces. In Figure 3Z, this amorphous and slightly crystalline film (#) appeared to completely coat regions of the buried feather. Barbs, barbules, and hooklets (hl) were visible (Figure 3AA). Degradation did not differ measurably between black (Figure 3Y–Z) and white (Figure 3AA,BB) feathers.

Feathers exposed on the surface and incubated with E-Pure water and added HA (ESME) showed exceptional preservation, with virtually no degradation, although mineral deposition was noted on the surfaces of some regions. There was no obvious difference in preservation between black (Figure 3CC,DD) and white (Figure 3EE,FF) feathers. Unaltered barbs and barbules (BL) could be seen, and filaments of keratin comprising these were visible as aligned fibers (KF). Flattened fungal hyphae (F) were visible, in some cases with outgrowths of bushy, finely crystalline structures (Figure 3DD, (Cr)). These "floret-like" structures were also visible in other regions of unburied feathers and were associated with fungal hyphae. Very few microbial bodies (m) were observed in either buried or unburied feathers in this condition, but a few could be seen in association with this floret-like structure, and they appeared both as bacilliform and coccoid structures (Figure 3FF, (m)).

Feather preservation was also greater when incubated in E-Pure water in both buried (ESB, Figure 3GG–JJ) and unburied (ESE, Figure 3KK–NN) states, consistent with LM data, and virtually no differences were seen between black (Figure 3GG,HH) and white (Figure 3II,JJ) regions. Fungal hyphae (F) were prevalent in buried feathers (and appeared, in some cases, to be associated with a thin amorphous coating tentatively identified as biofilm (bf)) (Figure 3HH,II). This amorphous film appeared to coat buried feather barbs almost completely in some regions.

Some regions of amorphous film (bf) were associated with fungal hyphae in the unburied feathers (Figure 3KK–NN) as well. Feather barbs and barbules were clearly visible, and although crystals had deposited on the surface (arrowheads), the structure of the feather was intact. Even though no excess HA was added to these burial conditions, in some areas, mineral crystals were still visible. The fibrous texture of the keratin comprising the barbs was intact (Figure 3NN).

Figure 4 shows feathers exposed to the same conditions as Figure 3, but conducted in an elevated CO2 atmosphere. When feathers were buried and incubated with natural pondwaters (CPB, Figure 4A–D), virtually no feather structure remained after six weeks, but fungal hyphae were evident (F) and the feathers revealed highly degraded, pitted surfaces (e.g., Figure 4C,D) colonized by fungi (F). Needle-like crystals (arrowheads) were observed on the surface (Figure 4A,B). The original color of the feathers could not be discerned.

The unburied feathers in this elevated CO2 pondwater environment (CPE) fared only slightly better (Figure 4E–H). Some of the barbules (Figure 4E–F, (BL)) remained intact, but in parts they were covered with a regional overgrowth of fibrous material (fm). Mineral crystals (arrowheads) and occasionally diatoms (dt) could be seen. However, in other regions, the surface was riddled with holes (Figure 4G,H). Ropey keratin filaments (Figure 4F, (kf)) could be seen. In contrast to the comparable ambient condition, microbial bodies were rarely visible.

In a high CO2 environment, after six weeks no buried feathers were recovered from the CPBB condition (missing data, Figure 4I–L). However, the feathers exposed at the surface (CPBE) are shown in Figure 4M–P. Microbial tunneling (Figure 4M, arrows) was visible in this condition, as was seen in the corresponding ambient condition. Figure 4M is the only sample where original color could be determined; this feather was black. Figure 4N shows a flaky material, possibly representing degraded keratin or, alternatively, fine clay grains. Although Figure 4O reveals aligned keratin fibers (kf), in most samples the outermost surface of keratin was largely degraded. Contrary to the ambient condition, microbial bodies were difficult to see, but their impressions within the degraded keratin were visible as aligned voids (Figure 4M,P; (v)).

When HA was added to pondwater in the elevated CO2 environment, preservation was improved in both the buried (CPMB, Figure 4Q–T) and unburied (CPME, Figure 4U–X) conditions. Fine feather structure, including barbs and barbules, remained visible, but a coating of either granular or smooth material (ˆ) could be seen on the buried feathers, that is smoother and less granular than shown in Fig. 3; as a result we used a different symbol. A similar heterogenous material covered the surface of the unburied feathers (Figure 4U–X) to an even greater extent, but beneath this layer, feather barbs appeared intact. Diatoms and other structures were also seen. Geometric mineral crystals (arrowheads) could be seen interspersed with small, round, concave structures approximately 2 μm in diameter (arrows). In Figure 4W, mineral overgrowth on barbules was visible, and thin, plate-like features that could be degraded keratin or clay grains were visible.

Preservation was greatly enhanced in feathers in both buried (CESMB, Figure 4Y–BB) and exposed (CESME, Figure 4CC–FF) conditions in elevated CO2 when HA was added. Feather structure was almost perfectly preserved, but interspersed with the barbules in the buried feathers were small plate-like structures. These can be seen more clearly in highermagnification images (Figure. 4BB). In the unburied feathers (Figure 4CC–FF), preservation was virtually perfect, and no alteration of structure was visible in low-magnification images

(Figure 4CC,EE). At higher magnifications, small pockets of microbes appeared on the surface (Figure 4DD,FF, m). Thin, plate-like structures could also be seen.

In the high-CO2 environment, feathers buried with E-Pure water alone were almost as well preserved as the preceding condition (CESB, Figure 4GG–JJ). Barbs and barbules were preserved with no evidence of fraying or breakdown, although at higher magnifications, some regions appeared to be covered in a crystalline coating (Figure 4JJ), and rarely, microbodies were seen on the surface (Figure 4HH). In the exposed feathers incubated with E-Pure water only, virtually no degradation was seen (Figure 4KK–NN) and barbs and barbules were intact. Small crystals of mineral precipitate were occasionally seen on the surface of the feather (Figure 4LL, arrowhead), and the presence of a few microbial bodies (m) were also noted on the feather surface. Figure 4MM–NN shows a region where keratin filaments (kf) could be seen extending to, and surrounding, a region of geometric pith (P).

#### *3.3. Transmission Electron Microscopy (TEM)*

Transmission EM allowed us to study degradation with greater resolution. With ultrathin sections (~90 nm thickness), it is impossible to discern whether black or white regions of the embedded feathers were sectioned. In most cases, where electron-opaque melanosomes were not visualized and no voids were present, we assumed these to be white regions. In all cases, melanosomes, when visible, were most abundant in the feather barbules. They were uniformly opaque to electrons and always embedded in the keratinous matrix, not displayed on the outer surface, as we have previously shown [60]. They did not appear to overlap within the keratin matrix, but were well separated.

Figure 5 shows the various conditions in ambient atmospheres after six weeks of degradation; the first two panels in each row were buried and the second two exposed at the surface. Degradation in pondwater was greater in the buried feathers (PB, Figure 5A,B) than the exposed ones (PE), with a keratinous matrix developing holes. In both buried and exposed feathers, the melanosomes were relatively intact, although some degradation was seen. When *B. licheniformes* were present with pondwater (PBB, PBE; Figure 5E–H), degradation was much more advanced. The keratin matrix was highly degraded in PBB feathers (Figure 5E,F), and the melanosomes showed a great loss of integrity. In the PBE feathers (Figure 5G,H), the keratin matrix showed a loss of smoothness, taking on a "bubbly" texture, but was relatively intact. Melanosomes were not prevalent. This could be because of the region of the feather imaged, as barbules contain more melanosomes than the rest of the feather [61]. When HA was added to the pondwater (PMB, PME; Figure 5I–L), degradation was much less than in previous conditions, and mineral crystals could be seen associated with feather surfaces. Melanosomes and matrix were intact in both cases, although some fraying and loss of integrity was seen in the PME feathers (Figure 5L). Feathers incubated in E-Pure water to which HA was added (ESMB, ESME; Figure 5M–P) showed loss of mineral integrity. No melanosomes were visualized, probably indicating that this was a white feather region. Finally, Figure 5Q–T show feathers incubated in only E-Pure water. Degradation was minimal, but greater than seen when HA was added. Melanosomes were round and almost completely solid; however, in the exposed feathers, some cracking of the keratin matrix was visible.

Figure 6 shows TEM images of feathers degraded in an atmosphere elevated in CO2. In most conditions, degradation was advanced compared to that seen in ambient conditions. Figure 6A–B, depicting CPB buried feathers, show advanced degradation, with many melanosomes completely degraded, and those remaining showing loss of integrity. Exposed feathers (CPE) fared better, with elongated and round melanosomes present within the keratin matrix. When *B. licheniformis* was added to the pondwater, all feathers were completely degraded at six weeks, so as above, images (Figure E–H) were taken after four weeks. The buried feathers (CPBB) showed greater degradation, with most melanosomes no longer intact. In the exposed feathers (CPBE; Figure 6G,H), voids can also be seen where melanosomes once resided. In high-CO2 environments, when HA was added to pondwater, degradation was greatly decreased. Melanosomes showed slight degradation in the buried

condition (CPMB; Figure 6I,J), with some apparently lysed. Electron-opaque, needle-like mineral deposits (mc) were seen outside of the keratin matrix. Exposed feathers (CPME; Figure 6K–L) showed better preservation, with little degradation of either melanosomes or keratin. There was an unidentified growth on the outer surface of the keratin that was most likely organic, based on its electron translucent character.

**Figure 5.** Feathers degraded in ambient conditions. (**A**–**D**) Exposed to pondwater; (**A**,**B**) buried and (**C**,**D**) exposed. (**E**–**H**) Pondwater with *B. licheniformis* added; (**E**,**F**) buried and (**G**,**H**) exposed. (**I**,**L**) are feathers incubated with pondwater and added HA; (**I**,**J**) buried and (**K**,**L**) exposed. Electron-opaque, needle-like mineral crystal (mc) deposits are seen exterior to the keratin of the feather barbules. (**M**–**P**) show feathers incubated in E-Pure water to which HA has been added. No melanosomes are seen, probably indicating that these are white feather regions, although early degradation is a less likely explanation. (**Q**–**T**) show feathers incubated in E-Pure water only. (**S**) low magnification, (**T**) higher magnification showing some cracking of the feather matrix. Scale bars as indicated.

**Figure 6.** Transmission electron microscopy (TEM) of feathers degraded in elevated CO2 atmosphere. (**A**,**B**) are buried and (**C**,**D**) are exposed in natural pondwaters. (**E**–**H**) show feathers in natural pondwater with *B. licheniformis* added; (**E**,**F**) are buried (after 4 weeks) and (**G**,**H**) are exposed. (**I**–**L**) Pondwater to which HA has been added; (**I**,**J**) are buried and (**K**,**L**) are exposed at the surface. (**M**,**N**) are buried and (**O**,**P**) are exposed feathers degraded in E-Pure water to which HA had been added. Finally, (**Q**–**T**) are feathers in elevated CO2, degraded in E-Pure water only; (**Q**,**R**) are buried and (**S**,**T**) are exposed. Scale bars as indicated.

Buried feathers incubated in elevated CO2 with E-Pure water to which HA had been added (CESMB; Figure 6M,N) showed relatively advanced degradation of melanosomes, leaving voids, but with little degradation of the keratin matrix in which they were embedded. Exposed feathers (CESME; Figure 6O,P) showed similar degradation. The keratin matrix was relatively unaltered. Melanosomes were rare, but empty voids were present, supporting the idea that these feathers contained pigment organelles.

Feathers incubated in E-Pure water only showed little overall degradation. Buried feathers (CESB; Figure 6Q,R) showed little keratin degradation but no melanosomes or voids, and therefore probably represent white feather regions. Undegraded, hollow melanosomes were visible embedded deep within the keratin matrix in the exposed condition (CESE; Figure 6S,T).

#### *3.4. In Situ Immunohistochemistry (IHC)*

We tested the hypothesis that degradation influences antibody binding. Figure 7 shows the response of feathers in different degradation conditions when exposed to a polyclonal antiserum raised against feathers [62]. All images were taken under identical parameters. In the ambient condition, no reduction in intensity of binding was seen, even in the feathers showing the greatest morphological damage (Figure 7A,B,E,F), consistent with preservation of epitopes after structural integrity was lost. Antibody specificity is supported by the lack of binding to the embedding material, or to regions where tissues were completely missing. Negative controls of no primary antibody applied, but all other steps being the same (not shown), indicates that the signal did not arise from non-specific binding of the secondary antibody or fluorescent label. Melanized feather regions (Figure 7A,C,I,K) were more easily visualized in transmitted LM than those regions apparently lacking melanin (Figure 7E,G,M,O,Q,S), but antibodies bound with equal intensity with both types degraded for the duration of this experiment.

**Figure 7.** In situ immunochemistry of feathers degraded in ambient atmosphere, then exposed to a polyclonal antiserum raised against whole feather extract and visualized in transmitted light (1st and 3rd image of each row) or using an FITC filter (2nd and 4th image). (**A**,**B**) are buried and (**C**,**D**) are exposed in natural pondwaters. (**E**–**H**) show feathers in natural pondwater with added *B. licheniformis* at 4 weeks; (**E**,**F**) are buried and (**G**,**H**) are exposed. (**I**–**L**) Pondwater to which HA has been added; (**I**,**J**) are buried and (**K**,**L**) are exposed at the surface. (**M**,**N**) are buried and (**O**,**P**) are exposed feathers degraded in E-Pure water to which HA has been added. Finally, (**Q**–**T**) are feathers degraded in E-Pure water only; (**Q**,**R**) are buried and (**S**,**T**) are surface-exposed. Transmitted light images are modified to increase contrast; see Supplementary Materials.

When feathers were degraded under the same conditions as in Figure 7, but in elevated CO2 atmospheres, the results were different. Buried feathers exposed to pondwater only (CPB; Figure 8A,B) showed a highly degraded area covering a more intact region. Polyclonal antibodies bound with high avidity, but no melanized regions could be seen with certainty in the transmitted LM (Figure 8A). The exposed feathers (CPE) with pondwater demonstrated better microscopic integrity. Regions of melanized barbules (Figure 8C) could be seen in LM and antibodies bound with the same level of intensity (Figure 8D), as seen in Figure 7. When feathers were buried and incubated with pondwater and *B.licheniformis* in elevated CO2 (CPBB), no feathers remained for analyses; data are shown for this condition in feathers after four weeks degradation (Figure 8E,F). Integrity was greatly reduced, despite melanin in the visible barbs (Figure 8E). Surface-exposed feathers (CPBE; Figure 8G,H) showed minimal degradation, despite containing less or no melanin. Figure 8I–L shows buried (Figure 8I,J) and exposed (Figure 8K,L) feathers incubated with pondwater with added HA. Both buried and exposed feathers showed melanized regions that remained virtually intact, with no visible degradation, and both bound antibodies with equal and intense avidity. Similarly, Figure 8M–P shows buried (M,N) and exposed (O,P) feathers in E-Pure water to which HA had been added. Virtually no degradation of epitopes was seen, and no substantial differences between lightly melanized and buried (M,N) or unmelanized and exposed (O,P) were seen. Non-melanized buried (Figure 8Q,R) and melanized, exposed (Figure 8S,T) feathers in E-Pure water only were comparable, showing no tissue degradation and no reduction in antibody binding under these conditions.

**Figure 8.** In situ immunochemistry of feathers degraded in elevated CO2 atmosphere, then exposed to a polyclonal antiserum raised against whole feather extract and visualized in transmitted light (1st and 3rd image of each row) or using an FITC filter (2nd and 4th image). (**A**,**B**) are buried and (**C**,**D**) are exposed in natural pondwaters. (**E**–**H**) show feathers in natural pondwater with added *B. licheniformis*; (**E**,**F**) are buried (after 4 weeks) and (**G**,**H**) are exposed. (**I**–**L**) Pondwater to which CaPO4 has been added; (**I**,**J**) are buried and (**K**,**L**) are exposed at the surface. (**M**,**N**) are buried and (**O**,**P**) are exposed feathers degraded in E-Pure water to which HA had been added. Finally, (**Q**–**T**) are feathers in elevated CO2, degraded in E-Pure water only; (**Q**,**R**) are buried and (**S**,**T**) are surface-exposed.

#### **4. Discussion**

We showed, using multiple methods, that the degradation of organics is qualitatively and/or quantitatively different in atmospheres elevated to Mesozoic CO2 levels (or higher) than what is observed in ambient atmospheres. Some microbes are known to precipitate CO2 as carbonate minerals [63–65], and we sought to determine whether this precipitation could be increased in atmospheres elevated in CO2. We chose hydroxyapatite because microbes are known to precipitate this mineral [66], and because the solubility of HA rises with increased CO2 [67]. Because it has been shown that organic molecules can be preserved preferentially in intracrystalline regions of bone [68], we predicted that we may see better molecular preservation in extant samples subjected to elevated CO2 through rapid stabilization by mineral precipitation. We used both melanized and non-melanized feathers to test the effect of pigment on preservation [47,69]. We added feather-degrading microbes to natural microbial biomass to see whether the action of these organisms was increased in high CO2 atmospheres, and we tested the effect that increasing the concentration of HA would have on preservation and degradation in varying atmospheres. Finally, we compared preservation when feathers were exposed to waters containing a natural microbial population vs. E-Pure water.

In ambient atmospheres, degradation was greatest when feather-degrading bacteria were added to feathers in the natural microbiota of pondwaters. In general, buried feathers were more highly degraded than surface-exposed feathers, and black regions retained more structural integrity than did white feather regions. The greater degradation observed for buried feathers was not predicted, but we hypothesize that burial gave greater access to microbes by keeping the feathers uniformly damp. In the other conditions, there was little difference, but importantly, adding HA to the waters containing microbes greatly increased preservation, perhaps by slowing the microbially mediated degradation undertaken by microbes naturally present in feathers. In E-Pure water, with or without minerals, virtually no degradation was seen, and black feathers did not differ from white in preservation, supporting a specific role for microbially mediated precipitation in preservation. Feathers in elevated CO2 showed much greater degradation, whether buried or exposed, in both natural pondwaters and with added microbes; however, in the presence of added HA, degradation was essentially halted, and preservation was relatively greater in elevated CO2. SEM data for both natural waters and waters with added microbes demonstrated multiple holes and tunneling not seen in ambient atmospheres, suggesting upregulation of microbial enzymes of degradation in elevated CO2. Increased microbial activity in response to elevated CO2 was noted in other studies as well [70–73]. The increase in preservation in both atmospheric conditions when HA was added suggests the deposition of mineralarrested degradation in both buried and unburied states, as suggested elsewhere [74,75], either by adsorbing enzymes of degradation and inactivating them (e.g., [76], by encasing organics, thus making them inaccessible [77]), or both, but this process was relatively enhanced in elevated CO2. Where color could be detected, there was no measurable difference in preservation/degradation in most conditions. Keratin filaments and intact barbs, barbules, and sheaths showed little to no degradation under the elevated CO2 + mineral condition. Fungal hyphae, prevalent in the ambient conditions, were not discerned in elevated CO2 except in the pondwater-only condition.

TEM illustrated high levels of degradation of the keratin sheath in pondwater conditions, both buried and unburied, but melanosomes remained relatively intact, though some degradation of these bodies was visible. However, in buried feathers with added *B. licheniformis*, melanosome and keratin degradation were both advanced over other conditions. Crystal deposits were seen when HA was added, and in these conditions, the feathers appeared unaltered. Patterns were similar in the elevated CO2 atmospheres for both pondwater and with added microbes, again with datapoints representing only four weeks in the buried condition. In elevated CO2 with added HA, melanosomes appeared more degraded than did keratin, leaving many voids in the buried feathers.

In ambient atmosphere, antibody binding was intense in all cases, with little difference noted between all conditions shown here. Burial state, added microbes, and presence or absence of melanin did not appear to affect the recognition of keratin epitopes by these polyclonal antibodies, and binding occurred even when structural integrity was not well maintained. Although no material persisted to the six-week timepoint in buried feathers with added *B. licheniformis* in elevated CO2, the only case where antibody binding was detectably reduced was in the buried feathers after four weeks. Both structural integrity and epitope recognition were greatly diminished.

Processes resulting in the preservation of normally labile and easily degraded organic materials have been of significant interest to the paleontological community, because examples of hair, skin, feathers, and internal organs do not follow standard models of fossilization. It is significant that incidences of exceptional soft tissue preservation are not evenly distributed through time, but occur more frequently in pre-Cenozoic deposits [78,79] during time periods when atmospheric CO2 was elevated over today's levels [80,81]. In addition, Cenozoic lagerstätte are few, but generally correlate to short periods of elevated CO2 [82–84]. We propose that this uneven distribution of exceptionally preserved deposits may be due in part to the response of microbes to elevations in atmospheric CO2 and the subsequent acidification of pore waters, making mineral ions more available for precipitation. This preliminary study demonstrates the need for further investigation into the role of both microbes and CO2 in preserving organic remains in fossils.

#### **5. Conclusions**

Degradation by all measures was greatest in buried feathers in elevated CO2. However, the addition of HA slowed degradation, whether with natural pondwaters or E-Pure water, and this effect was greater, relative to level of degradation without minerals, in elevated CO2. The rapid precipitation of minerals on organics outpaced decay in these conditions, illustrating a possible role in exceptional preservation, as suggested by [85–88] and others. Similarly, antibody recognition as a proxy for molecular preservation was still present, even in feathers showing microstructural degradation. The implications for fossilization support the hypothesis that degradation proceeded differently in the elevated CO2 atmospheres of the Mesozoic. Although degradation was more rapid and more complete in buried feathers, antibody binding was not greatly affecting during the time interval of this experiment, suggesting a lack of direct correlation between histological integrity and antibody binding. In most cases, keratin preservation was equal to or greater than melanosomes under the conditions of this experiment, whatever the atmospheric conditions. We tested the role of apatite because, in atmospheres elevated in CO2, as in the Mesozoic, warmer waters and the acidification of pore waters would allow an increase in the concentration of this solubilized mineral. Experimental taphonomy designed to address questions of degradation in the past, particularly in deep time intervals, will be misleading unless atmospheric composition and the effects of greenhouse gases on degrading microbes are taken into account.

**Supplementary Materials:** The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/biology11050703/s1: Figure S1: Experimental Setup.

**Author Contributions:** M.H.S. designed the studies, provided the feathers, collected the pondwater and sediment samples, designed the figures, and wrote most of the paper. W.Z. conducted experimental setups; collected LM, TEM, and IHC data; contributed to the paper, and wrote most of the Materials and Methods section. N.E. collected the SEM data. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** All data dealing with this study are reported in the paper.

**Acknowledgments:** This work was performed in part at the Analytical Instrumentation Facility (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation (award number ECCS-2025064). The AIF is a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), a site in the National Nanotechnology Coordinated Infrastructure (NNCI). SEM images were collected at the Image and Chemical Analysis Laboratory (ICAL), Department of Physics, Montana State University. This work was also supported by the NSF (award number 1934844).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Review* **The Rising of Paleontology in China: A Century-Long Road**

**Zhonghe Zhou**

Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, 142 Xizhimenwai Dajie, Beijing 100044, China; zhouzhonghe@ivpp.ac.cn

**Simple Summary:** A brief account of the history of Chinese paleontology for about one century is provided with some perspectives for its future development. The development of Chinese paleontology is closely related to its social-ecological background as well as its connection to the outside world. On the other hand, the rising of the Chinese paleontology also benefitted from its rich fossil resources as well as the integration with other biological and geological disciplines and the use of new technologies.

**Abstract:** In this paper, the history of paleontology in China from 1920 to 2020 is divided into three major stages, i.e., 1920–1949, 1949–1978, and 1979–2020. As one of the first scientific disciplines to have earned international fame in China, the development of Chinese paleontology benefitted from international collaborations and China's rich resources. Since 1978, China's socio-economic development and its open-door policy to the outside world have also played a key role in the growth of Chinese paleontology. In the 21st century, thanks to constant funding from the government and the rise of the younger generation of paleontologists, Chinese paleontology is expected to make even more contributions to the integration of paleontology with both biological and geological research projects by taking advantage of new technologies and China's rich paleontological resources.

**Keywords:** paleontology; China; history; 20th century; 21th century

**Citation:** Zhou, Z. The Rising of Paleontology in China: A Century-Long Road. *Biology* **2022**, *11*, 1104. https://doi.org/10.3390/ biology11081104

Academic Editor: Raymond Louis Bernor

Received: 13 June 2022 Accepted: 14 July 2022 Published: 25 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

This paper aims to provide a brief account of the history of the development of paleontology in China. The year 1920 is chosen as an important starting point for Chinese paleontology. During the past century from 1920 to 2020, Chinese paleontology has gone through several difficult periods and was only able to usher in the rapid growth seen over the past several decades thanks to the continuous efforts of the first-generation paleontologists in the 20th century to the youngest generation in the 21th century. In a chronological approach, this paper also intends to reflect the major characteristics of the discipline at various stages of the history with a brief introduction of the major achievements and an overview of how Chinese paleontologists have benefitted from international collaborations and the social-economic changes in China, focusing in particular on those since 1978. Finally, this paper discusses a few brief predictions regarding the future directions and challenges for Chinese paleontology.

#### **2. Before 1920—The Collecting and Studying of Chinese Fossils by Foreigners**

In China, fossils have been recognized as the remains of ancient animals, plants and fungi as well as seen as evidence of changes of the Earth's environment for over two thousand years. However, they had never been studied by Chinese scholars as a scientific subject until modern sciences (e.g., geology, paleontology) began to be introduced to China during the 19th century. The first scientific collection of Chinese fossils was made by western explorers, missionaries, geographers, and geologists, particularly gaining momentum since the opium war in 1940, and then studied by western paleontologists from Europe and the Americas. For instance, L. de Korunck published on two brachiopods from China in 1943, and R. Owen published the first paper on Chinese mammal fossils [1]. E. Koken and M. Schlosser published books on Chinese fossil mammals in 1885 and 1903, respectively, which represent the earliest systematic studies on Chinese vertebrate fossils [2,3].

Beginning in the 1860s, French missionary P.A. David started to collect fossil fish from the Mesozoic lake deposits in western Liaoning. These fossils were later named as *Lycoptera* and are regarded as one of the typical elements of the Jehol Biota, which is now best known for producing many exceptional feathered dinosaurs, birds, and mammals. American geologist and explorer R. Pumpelly was one of the first western scientists to carry out geological surveys in China. He collected many fossils from China from 1863~1865 and proposed the lacustrine facies of the Chinese loess [4]. German scientist F. Richthofen made extensive geographic and geological explorations in China during 1868 and 1872. He also collected many invertebrate fossils as well as recorded their stratigraphic data in the majority of Chinese provinces.

During the late 19th century and early 20th century, western explorers and scholars also made many scientific expeditions to China. Swedish geographer and explorer S.A. Hedin made several expeditions to western China starting in 1890, and the ancient city of Loulan (Kroraina) in Xinjiang was one of his major discoveries during his voyage across the Taklamakan Desert. Also notable is the Central Asian Expeditions organized by the American Museum of Natural History in 1916 and 1919, which resulted in the discovery of many fossil mammals in Inner Mongolia. In the early 20th century, Russian geologists collected many dinosaur fossils from Heilongjiang, northeastern China, including some duck-billed dinosaurs.

Before 1920, there were hardly any Chinese scientists known mainly as a paleontologist. The first Chinese scientist who published his research on paleontology was Rongguang Qi, who was among the first group of teenagers sent to America by the Qing Imperial Dynasty government in 1871. He was trained in geology and mineral resources. In 1910, he published his study on invertebrates and plants from Hebei, North China.

The first attempt to teach college-level geology in China started in 1909 at the Imperial University of Peking (known as Peking University since 1912). In 1914, Wenjiang Ding (V.K. Ting; 1887–1936), a 1911 graduate of the University of Glasgow, started to teach China's first college-level course in paleontology at the Geological Institute established in 1913. This produced the first generation of domestically trained geologists, including a few paleontologists who were then hired by the National Geological Survey of China that was officially founded in 1916 with Ding as its first director. The National Geological Survey of China has also been generally regarded as the first national scientific institution with an international reputation in Chinese history. Among many of his tremendous achievements, Ding also made contributions to many pioneering works in Chinese paleontology and stratigraphy. He discovered the first Devonian fish in Yunnan in 1913 and pioneered the study of fossil plants in China.

In summary, upon entering the 20th century, China had gone through major social changes, and its door began to open to the outside world. In 1911, the Imperial Qing Dynasty was ended and replaced by the Republic of China. A few more young Chinese were sent to Western countries to learn science and technology. The early 20th century was the preparation stage for the early development of paleontology in China [5].

#### **3. From 1920 to 1949—Founding Stages of Chinese Paleontology**

1920 was a turning point for Chinese paleontology. By the invitation of Wenjiang Ding, two prominent western-trained geologists and paleontologists joined Peking University. One of them was A.W. Grabau from Columbia University. Grabau not only taught at Peking University, but was also the Chief Paleontologist at the Chinese Geological Survey (Figure 1). The second was Siguang Li who graduated from Birmingham University with a master's degree in 1919 (Ph.D. degree in 1931). During 1920–1937, Grabau, Li, and their colleagues mentored and trained many of the first generation of geologists and paleontologists in China who would later become the leading figures in the majority of disciplines in Chinese

geology and paleontology. It is worthy to note that Grabau has been also well known for proposing the Jehol Fauna and Jehol Series in western Liaoning, northeastern China [6,7]. He also named the Changxing limestone in Zhejiang Province, where two GSSPs (Global Boundary Stratotype Sections and Points) were later established, including the boundary between the Permian and Triassic.

**Figure 1.** Founders of Chinese geology and paleontology gathering at Amadeus W. Grabau's home in Beijing in 1935. Front row, from the left: Hongzhao Zhang, Wenjiang Ding, Amadeus W. Grabau, Wenhao Weng, and Pierre Teilhard de Chardin. Middle row, from the left: Zhongjian Yang, Zanheng Zhou, Jiarong Xie, Guangxi Xu, Yunzhu Sun, Xiechou Tan, Shaowen Wang, Zanxun Yin, and Fuli Yuan. Back row, from the left: Zuolin He, Hengshen Wang, Zhuquan Wang, Yuelun Wang, Huanwen Zhu, Rongseng Ji, and Jianchu Sun (credit to IVPP).

Many of the first-generation Chinese paleontologists had been trained overseas and earned their Ph.D. degrees during the 1920s–1940s before they returned to China to pioneer their research fields. Among them were the invertebrate paleontologists Yunzhu Sun and Senxun Yue, vertebrate paleontologist Zhongjian Yang (Chung-Chien Young), and paleobotanist Xingjian Si, who received their Ph.D. degrees in Germany in 1927, 1936, 1928, and 1933, respectively. Zanxun Yin, geologist and paleontologist, and Wenzhong Pei, paleoanthropologist and archaeologist, earned their Ph.D. degrees from France in 1931 and 1937, respectively. Invertebrate paleontologists Jianzhang Yu and Hongzhen Wang received their Ph.D. degrees from England in 1935 and 1947. Zunyi Yang, an invertebrate paleontologist, received his Ph.D. degree from the USA in 1939. They have all became the backbones of Chinese paleontology after they returned to China.

It is notable that scientific expeditions organized by organizations of Western countries continued to be active in China during the 1920s and 1930s. For instance, the Central Asiatic Expeditions led by R.C. Andrews of the American Museum of Natural History during 1922 and 1930 resulted in the discovery of many dinosaur skeletons and dinosaur eggs in Inner Mongolia as well as many Mesozoic and Cenozoic mammals, including some of the earliest placental mammals. However, with the appearance of the first generation of Chinese paleontologists, international collaborations for the study of Chinese paleontology and stratigraphy became more and more prominent and productive. For instance, the Sino-Swedish Expedition which carried out multi-discipline scientific research in north and northwest China from 1927–1935 was co-led by S. Anders and Fuli Yuan, a geologist from Peking University, and many fossil reptiles and mammal fossils were collected during these field excursions. Zhongjian Yang and Wenzhong Pei joined the Central Asiatic Expedition in 1930.

One of the most successful collaborations between the first generation of Chinese paleontologists and their Western colleagues is probably the excavation of the fossil humans in Choukoudian (Figure 2) as well as the study on Cenozoic vertebrates and their stratigraphy. The Swedish geologist J.G. Andersson was invited by the Chinese government to act as consultant for the mining industry from 1914 to 1924. He discovered the Choukoudian site in 1918. During 1921–1922, Andersson assigned the Austrian paleontologist O. Zdansky to excavate at Choukoudian, which resulted in the discovery of two human teeth in addition to an abundance of mammal fossils. In 1927, with the support from the Rockefeller Foundation, D. Black from the Beijing Union Medical College Hospital (BUMCH) was able to continue the excavation of Choukoudian and study the Cenozoic cave deposits based on an agreement with the Geological Survey of China. In 1929, Wenzhong Pei led the excavation in Choukoudian and discovered the first skull of the Peking Man (*Homo erectus*), which drew some international attention for Chinese paleontology and paleoanthropology.

**Figure 2.** Paleontologists in Choukoudian, Beijing in 1934. From the left: Wenzhong Pei, Shiguang Li, Pierre Teilhard de Chardin, Meinian Bian, Zhongjian Yang, and George B. Barbour (credit to IVPP).

The collaboration between Black and the National Geological Survey of China also resulted in the founding of the Cenozoic Research Laboratory of the Geological Survey of China in 1929, which was the predecessor of the Institute of Vertebrate Paleontology and Paleoanthropology (IVPP). Black was appointed as the honorary head of the lab with Zhongjian Yang as the deputy head and P. T. de Chardin as the adviser. After Black died in 1925, his position was filled by German paleontologist F. Weidenreich. Clearly, the Cenozoic laboratory was designed and intended to be an international research unit.

In 1922, with the support of Grabau and Anderson, Ding founded the first Chinese paleontological journal *Palaeontologia Sinica*, and he became the chief editor of this journal for nearly 15 years. The journal published only English and German papers in the beginning, but later published Chinese papers as well. The majority of the important Chinese paleontological and stratigraphic studies during the 1920s and 1930 were published in this journal.

In 1928, the National Geological Survey of China formally established its Paleontology Laboratory. In the same year, the National Research Institute of Geology was founded in Nanjing with its own Stratigraphy and Paleontology Laboratory, representing the growth of paleontological research being conducted in China.

The paleontological and stratigraphic studies carried out by Chinese paleontologists started to grow in number quickly during the 1920s and 1930s. To list a few examples, Xichou Tan excavated dinosaurs in Laiyang, Shandong Province in 1923 and collected many dinosaur bones, some of which were later published and named as a duck-billed dinosaur *Tanius sinensis* [8]. In addition, a large number of fish, insect, and plant fossils were also collected. In 1923, Zanheng Zhou, a graduate from the National Research Institute of Geology in 1916 who also later trained in Sweden, published his study on the fossil plants from Shandong Province [9], representing the first paleobotanic paper by a Chinese paleontologist.

The book "Contributions to Cambrian Faunas of North China" written by Yunzhu Sun [10] marked the first paleontological book by a Chinese paleontologist. Some domestically educated paleontologists also did great work in paleontology and biostratigraphy. For instance, Yazhen Zhao, a graduate from Peking University was hired by the National Geological Survey of China in 1923. During his short but glorious career, he published several classic books on the study of brachiopods and stratigraphy before he was tragically killed by a bandit while in the field in 1929 at the age of 31.

Siguang Li published his classic monography "Fusulinidae of North China" in 1927 [11]. In 1927, Zhongjian Yang published the book "Fossil Rodents from North China" [12] based on his doctoral dissertation, which was the first monography on Chinese vertebrate paleontology. Zhongjian Yang became the father of Chinese vertebrate paleontology. His earlier work focused on Zhoukoudian and other Late Cenozoic mammalian faunas of northern China. After 1938, his main research interest shifted to Mesozoic dinosaurs and synapsids. His collaborations with P. Teilhard de Chardin were very successful and contributed greatly to the establishment of the geochronology and the stratigraphic succession of the Tertiary and Quaternary periods in China.

Xingjian Si (Hsing-Chien Sze) came back to China after he was awarded a Ph.D. degree for his research on Chinese fossil plants [13] and contributed greatly to the study of Paleozoic and Mesozoic plants in China, which earned him international fame.

Thanks to the efforts of Zhongjian Yang and Yunzhu Sun, the Paleontological Society of China was founded in 1929. Yunzhu Sun elected the first president, and he was elected the vice chairman of the International Paleontological Association in 1948.

Although the 1920s and 1930s witnessed the rapid growth of paleontology in China, starting in 1937 China entered the Second Sino-Japanese War that ended in 1945 and then the Chinese civil war that ended in 1949, which resulted in the founding of the People's Republic of China. Paleontological study during these years became difficult. The National Geological Survey of China moved to Yunnan. Peking University also moved to Yunnan to merge with Tsinghua University and Nankai University, and together they became the National Southwestern Associated University. At this new conglomerate University, students continued to be taught geology and paleontology, and many of the graduates later became the some of the most important figures in Chinese paleontology (e.g., Hongzhen Wang, Enzhi Mu, Dongsheng Liu, Zhiwei Gu, Yichun Hao etc.). As a 1937 graduate from Peking University, Yanhao Lu, a Cambrian and trilobite expert, first worked in the National Southwestern Associated University and then moved to the National Geological Survey of China. In reality, Chinese paleontological study had not completely stopped. For instance, Zhongjian Yang continued his work on the Lufeng dinosaur fauna in Yunnan during the trying years from 1938 to 1945 and had discovered the Jurassic dinosaur *Lufengosaurus* Fauna [14]. *Lufengosaurus* also represents China's first dinosaur with a mounted and complete skeleton. In 1940, Young was appointed the head of the Laboratory of Vertebrate Palaeontology and the honorary head of the Cenozoic Department of the National Geological Survey of China. In addition to the work he did in Yunnan, he also carried out fieldwork in Northwestern China. During the same time, Zanxun Yin and Yanhao Lu studied the Paleozoic biostratigraphy in southwestern China.

In sum, after the fall of Qing Dynasty in the 1911 and before the Japanese invasion in 1937, China had witnessed a relatively fast period of economic growth and the construction of social infrastructure including railroads, communications etc. despite civil war and political instability. As a result, Chinese paleontology had experienced a rapid growth in both the number of paleontologists and paleontological research. Their research covered nearly all major animal phyla and plant divisions. Furthermore, their studies not only included taxonomic and biostratigraphic work, but also included discussions on the evolution of some specific taxonomic groups. Chinese paleontologists had also established a preliminary stratigraphic framework spanning from the Precambrian to the Quaternary. It is also notable that many of the first-generation paleontologists had some background in studying overseas or benefitted in some way from collaborating with some of the talented western paleontologists. However, despite the persistence of dedicated scientists, overall, the wars and social turmoil had slowed down the sometimes-fast development of Chinese paleontology.

#### **4. From 1949 to 1977—The Expanding Stage of Chinese Paleontology**

The founding of the People's Republic of China in 1949 had two immediate impacts: first, the newfound social stability and the secondly, the cessation of academic exchange with western countries. Learning from and modeling after the former Soviet Union became a trend. China's connection with countries other than the former Soviet Union was limited. For instance, Zhang Miman (Meemann Chang) spent a year visiting the Swedish Museum of Natural History in 1966 (Figure 3), but did not return there to receive her Ph.D. degree from Stockholm University until 1982.

**Figure 3.** Miman Zhang and her colleagues in Stockholm, Sweden in 1966. Front row, front the left: Erik Stensiö, Miman Zhang, E. Mark-Kurik. Back row, from the left: Hans-Peter Schultze, Tor Örvig, Hans Bjerring, Gareth Nelson, Ray Thorsteinsson, Erik Jarvik, and Hans Jessen (credit to IVPP).

To reboot its economy and social development following the civil war, the new Chinse government established the Chinese Academy of Sciences in 1950 which includes many different research institutes. Siguang Li was appointed the director of the Nanjing Institute of Paleontology in 1950, and the institute was officially founded in 1951, which also includes the Department of Vertebrate Paleontology in Beijing. The institute was based on staff merged from the paleontological group of the Institute of Geology, the Central Research Academy, the department of paleontology and the Cenozoic Research Laboratory (Beijing) of the National Geological Survey of China. In 1953, the Department of Vertebrate Paleontology, based in Beijing, was separated the from the Nanjing Institute of Geology and Paleontology and became affiliated directly with the Chinese Academy of Sciences. In 1957, the department was upgraded to an institute and assigned its current name, the

Institute of Vertebrate Paleontology and Paleoanthropology (IVPP), in 1960 with Zhongjian Yang as its first director.

In 1959, the Nanjing Institute of Paleontology was renamed as the Nanjing Institute of Geology and Paleontology (NIGPAS) and has since focused on the study of invertebrate paleontology, paleobotany, and stratigraphy, in order to be distinct from the IVPP, but both are still affiliated with the Chinese Academy of Sciences (CAS) today. The two research institutes of the CAS have since become the major forces in paleontological study in China. It is also noteworthy that the two institutes were also able to accept graduate students and helped to educate many of the outstanding young paleontologists of the next generation.

In addition to the positions available at the two research institutes of the CAS (NIGPAS and IVPP), many other paleontologists were mainly employed at universities such as Peking University, Nanjing University, Northwest University, Beijing Institute of Geology (now the China University of Geoscience in Beijing and Wuhan), Beijing Institute of Mining and Technology (now the China University of Mining and Technology), Changchun Institute of Geology (now merged into Jilin University), and Chengdu Institute of Geology (now the Chengdu University of Technology) etc. They have educated many students in paleontology and stratigraphy, in addition to conducting extensive studies on invertebrates, plants and stratigraphy.

In addition, the Geological Academy of Sciences affiliated to the Ministry of Geology and other institutes affiliated to the later established Ministry of Oil have also recruited many staff members who are working on paleontology and stratigraphy, mainly to fulfil the tasks of geological surveying and exploring for geological resources.

In 1953, the journal *Acta Paleontologica Sinica* was established, with Zunyi Yang as chair of the editorial board. Minzhen Zhou, a 1950 Ph.D. from Lehigh University in the USA, helped create the journal *Vertebrata PalAsiatica* in 1956, with Zhongjian Yang as the chair of the editorial board. In 1957, Zunyi Yang and Yichun Hao, graduates from the National Southwestern Associated University, published the first textbook on paleontology for Chinese students. In the same year, Yichun Hao also published the first Chinese textbook on micropaleontology.

During this time, the extensive geological surveys conducted due to the nationwide need to domestically locate oil, coal, and other mineral resources stimulated a rapid growth in the number of paleontologists around the country. Micropaleontology was quickly developed, largely in response to the practical demand for experts in the field. While academic exchange between China and western countries had nearly stopped, a new generation of students were sent to the former Soviet Union to study paleontology. Among the graduates were Yichun Hao, Pinxian Wang, and Miman Zhang (Meemann Chang). Chinese paleontologists had been asked to edit books on categories of fossils from various regions of China and made progress on the study of fossils and stratigraphy of nearly all Phanerozoic ages of China. For instance, dinosaur excavations at Shandong, Sichuan, Inner Mongolia and Xingjiang have produced many previously unknown taxa from the Jurassic and Cretaceous. In addition, large scale expeditions were organized to investigate the geology of Tibet.

Despite the general practical need of the geological surveys to explore mineral resources, there was still some scientific consideration for the paleontological development in China thanks to the enduring visions of the first generation of Chinese paleontologists (Figure 4). For instance, Zhongjian Yang had outlined the focus and directions of the IVPP as "four origins and two deposits" in 1955. The four origins include the origins of vertebrates, tetrapods, and mammals, as well as humans, primates and cultures from the perspective of biology. The two deposits represent the fossil bearing red beds in southern China and the soil-like deposits in northern China from the perspective of geology. In 1958, Yang further proposed the goals of "three gaps to be filled" for the IVPP, i.e., the evolutionary gap, regional gap, and the stratigraphic gap. In the 1960s, the discovery of human fossils had been remarkable, including some new specimens of *Homo erectus*, and *Homo sapiens*.

**Figure 4.** Group photo of officials elected at the first National Congress of the Chinese Paleontological Society in 1956. Front row, front the left: Senxun Yue, Zanxun Yin, Xiaohe Zhou, Zhongjian Yang, Yunzhu Sun, Xingjian Si, Jinke Zhao, and Zunyi Yang. Back row, from the left: Hongzhen Wang, Zhiwei Gu, Yuanren Qu, Shicheng Huo, Ren Xu, Minzhen Zhou, Yu Wang, Enzhi Mu, and Longqing Chang (credit to NIGPAS). The Chinese characters means the photo was taken at the first National Congress of the Chinese Paleontological Society on June 16th, 1956.

International collaboration between China and the former Soviet Union is not only limited to the education of Chinese students by experts from the former Soviet Union. In 1959, in adherence to an agreement between the Academies of Sciences of the two countries, a joint expedition to Central Asia was formed (Figure 5), and was led on the Chinese side by Zhou Minzhen from the IVPP. Unfortunately, the joint expedition did not last long and was ended in 1960 due to the deterioration of the political relationship between the two countries.

**Figure 5.** Field camp of the Sino–Soviet Union joint expedition in Inner Mongolia in 1959 (credit to IVPP).

During the Cultural Revolution (1966–1976), much like many other scientific disciplines, Chinese paleontology was seriously disrupted. In particular, the Chinese paleontological community was nearly completely isolated from the outside world. Despite the turmoil, field excavations continued to produce some exciting finds, and the empirical paleontological and stratigraphic work carried out during that time filled many gaps in our

understanding of both the evolution of life and the stratigraphic record. In addition, micropaleontology witnessed a stage of rapid development due to its application in surveying and exploring for geological resources.

In summation, with the second generation of paleontologists entering the mainstage during the period from 1949 to 1977, the Chinese paleontological community was greatly expanded. On one hand, this was under the guidance of the first generation of paleontologists and their international vision. On the other hand, it was also impacted by the influence of the former Soviet Union and the growth of Chinese paleontology had been closely related to the national industrial requirement for extensive geological surveys and prospecting with a focus on the collection of fossils from across the country spanning from the Precambrian to the Quaternary. A preliminary biostratigraphic frame was established, and the study of the Chinese paleontologists covered nearly all the taxonomic categories of paleontology. Overall, Chinese paleontology had generally been geology-oriented, although the biological significance of some fossils had also been discussed to a degree.

#### **5. From 1977 to 2020**

In 1977, the National College Entrance Exam was reinstated after it had been abandoned for ten years and as a result many more talented students were able to study paleontology in college. A new generation of graduate from Chinese universities were able to be recruited into the NIGPAS, IVPP, and other research institutes as well universities, and they remain the backbone of Chinese paleontology today.

With the beginning of the open-door policy in 1978, the connection between China and western countries was able to be re-established. While some senior paleontologists were able to communicate with their international colleagues and had more and more visits from both sides, some Chinese students were able to pursue their degrees in western universities. Meanwhile, new ideas and the most recent developments including scientific methods in western countries such as cladistics and paleoecology began to be introduced into China [15,16].

Since the 1980s, Chinese paleontologists have been frequently invited to participate in the editing of the Treatise on Invertebrate Paleontology. In addition, Chinese paleontologists have published many books about their systematic work on various groups such as trilobites, brachiopods, graptolites, corals, bivalves, cephalopod, insects, conchostrans etc.

Starting in the late 1990s, Chinese paleontology began to gradually enter its "golden age". First, some of the western educated students came back to China and were awarded enough funding for their research to establish their own research labs. Secondly, many domestically educated students graduated from colleges and joined research institutes such as the NIGPAS and IVPP. The third generation of paleontologists, or the "reform and open generation", had the advantage of having a better grasp of the English language and were able to more easily learn new technique and methods. It is notable that some of the Chinese students chose to stay in western countries after earning their Ph.D. degrees continued their paleontological career, and they helped to foster collaborations between China and various western countries in order to both educate students and sponsor further collaborative projects. More international meetings were hosted in China and more Chinese paleontologists were able to participate in meetings outside China.

Starting in 2000, the scale of funding from the government for basic scientific research began to be increased. The National Natural Science Foundation, which was founded in 1986, has been growing steadily and has become the main source of support for Chinese paleontologists since the 1990s. For instance, over two dozen of the third and fourth generation of paleontologists have been supported by the Distinguished Young Scientist Fund, and paleontologists from the NIGPAS, IVPP, China University of Geosciences, and Northwest University have also been awarded the Innovation Research Group Fund to ensure that their research is being supported in the long term.

In addition to the funding from the NSFC, Chinese paleontologists had a chance to be awarded large grants from other government agencies to help organize a large group of paleontologists to work together on a major scientific endeavor. For instance, the Special Funds for Basic Research of China ("973" projects) by the Ministry of Sciences and Technology provided grants amounting to twenty million Chinese yuan for a five year's project [17]. Since 2000, several paleontologists from the NIGPAS, IVPP, and the University of Geosciences (Wuhan, Beijing) have been awarded this grant. In addition, Chinese paleontologists have recently been able to secure other major grant, i.e., the Strategic Priority Research Program from the Chinese Academy of Sciences, to investigate the coevolution of life and paleoenvironment during major critical intervals of Earth's history. Most recently, a 10-year multidiscipline project was awarded to a paleontologist by the NSFC. It is intended to study the Mesozoic terrestrial biota and its tectonic background from the perspective of Earth system science, and it has drawn participants from paleontology, geochemistry, geophysics, and sedimentology. Furthermore, even more paleontologists have been invited to participate in several other multidiscipline projects, e.g., the new Tibetan Plateau Expeditions.

The state key laboratory administrated by the Ministry of Sciences and Technology is another way to support basic scientific research in China. The State Key Laboratory of Paleobiology and Stratigraphy, based at the Nanjing Institute of Geology and paleontology, and the State Key Laboratory of Biogeology and Environmental Geology, based at the Chinese University of Geosciences were established in 2002 and 2012, respectively. These labs are annually fiscally supported not only in scientific research and collaboration, but also in the acquisition of new equipment.

Due to the increasing funding capacity available to the promising young generation of paleontologists, Chinese paleontology has been growing at an unprecedent rate, and their research scope and production have been greatly expanded, spanning from traditional paleontology and stratigraphy, to paleobiology and some newly developed fields. Since 2000, some students and postdocs from western countries started to join the Chinese institutions and universities, which have further fostered the international collaborations between China and its international community.

In traditional paleontology and stratigraphy, Chinese paleontologists have greatly extended their scope of excavations of fossils to nearly all areas of China, spanning from Precambrian to the Quaternary. As a result, the rate of discovery of new taxa in various biological groups has significantly increased. For instance, the rate of commonly accepted dinosaur species named from China now outnumbers that of any other country. Starting 2015, supported by the "Special Research Program of Basic Science and Technology of the Ministry of Science and Technology", Chinese paleontologists launched an ambitious project to publish a complete series of *Palaeovertebrata Sinica*, which comprises three volumes and 23 fascicles on the taxonomy of all the vertebrate fossil species (nearly 10 thousand) in China. Currently, 13 fascicles have been published, and several more are nearly finished.

Among many of the discoveries, there are several world famous Lagerstätte are probably best known to the international paleontological community, i.e., the Neoprotozoic Wengan Biota (580 my) in Guizhou, Early Cambrian Chengjiang Biota (520 my) in Yunnan, the late Jurassic Yanliao Biota (160 my), and the Early Cretaceous Jehol Biota (125 my) in western Liaoning and neighboring areas, which have produced many world-known exceptionally preserved fossils bearing a lot evolutionary significant data on the early evolution of early life including the evolution of animals [18–21], origin of birds, and regarding the early evolution of mammals, birds, pterosaurs, and flowering plants etc. [22–25]. In addition, these discoveries have also drawn great attention from the media and public, hence increasing the profile of Chinese paleontology in popular scientific communities. In 2001, *Science* magazine published a special issue introducing the highlights of the discoveries and research done in Chinese paleontology.

Thanks to sufficient funding and the great efforts put into field investigations, many other Cambrian fauna have also been discovered, including the Neoprotozoic Lantian Biota (600 my) in Anhui [26] and the Miaohe Biota in Hubei, the terminal Ediacaran Shibantan Biota of Yangtze Gorges, South China [27], the Edicarian fauna in Guizhou [18,28], the Early Cambrian Burgess Shale-type fossil Lagerstätte the Qingjiang Biota [29], and the Middle Cambrian Burgess Shale-type fossil Lagerstätte the Lingyi Biota [30] (Figure 6), etc. Newly discovered Silurian fishes from Chongqing have produced many exceptionally preserved articulated jawed fished, among many other discoveries. Devonian plants from Yunnan and other regions produced unknown vascular plants [31]. The Permian Pompeii (about 300 my) preserved remarkable plants from coal in Wuhai, Inner Mongolia [32]. Many new species of Middle Triassic marine reptiles and fishes often represented by complete skeletons, including the earliest turtle, have also been discovered in Guizhou and Yunnan provinces [33]. The Miocene Hezheng Fauna has also produced abundant and diverse mammals and birds, sometimes with exceptional preservation of soft tissues and gut contents [34]. Lastly, the Miocene Zhangpu Biota from Fujian represents another amber treasure trove, recording a diverse fauna and flora of the tropical forest ecosystem [35].

**Figure 6.** Reconstruction of the latest discovered Middle Cambrian Lagerstätte Lingyi Biota in Shandong (credit to Fangcheng Zhao).

Chinese paleontologists have also made significant contributions to several of the major evolutionary issues, such as the Cambrian explosive radiation and its background [36], the origin of animals [37], the study on the Great Ordovician Radiation and the end-Ordovician mass extinction [38,39], the study of early vertebrates [40], the P-T extinctions [41], the biodiversity changes based on big data [42], the origin of birds and their flight [23,43], the early evolution of Mesozoic mammals including middle ears [22,44], the Mesozoic insect–plant coevolution [45], sexual selection [46,47], the interaction of the Cenozoic biota and flora with the uprising of the Tibet Plateau [48,49], the evolution and dispersal of modern and archaic humans in China, etc. [50–52].

In the study of stratigraphy, Chinese paleontologists have now witnessed the ratification of 11 global boundary stratotype sections and points (GSSP, or the so-called golden spikes) in China, including the GSSP of the boundary between the Paleozoic and Mesozoic in Zhejiang Province [53,54]. From the study of terrestrial strata, the Paleogene and Neogene biochronologic frameworks in China were preliminarily established [55]. New integrative stratigraphy and timescales for 13 geological periods (Ediacaran–Quaternary) in China were published in the special issue of SCIENCE CHINA Earth Sciences co-edited by Shen and Rong [56]. This research summarized the latest advances in stratigraphy and timescale and discussed the correlation among different blocks in China with international timescales.

High-resolution stratigraphic studies have also been helpful in oil and gas explorations as well as marine geology. Chinese paleontologists, particularly invertebrate paleontologists and micropaleontologists, continue to have work in close collaboration with oil companies for the prospecting of oil and gas resources in which the Silurian division by graptolites has played a key role [57].

Paleogeographic, paleoecological, and paleoenvironmental research has also been active. For instance, paleogeographic and paleoecological reconstruction has been useful in studying the controversial rising history of the Tibetan Plateau [49,58], and the paleobiogeographic and paleogeographic evolution of blocks in the Qinghai–Tibet Plateau [59].

It is notable that the geochronological progress in China has made a great contribution to the establishment of stratigraphic frame, which is critical for the discussion of various geological and evolutionary questions. In particular, the dating of the volcanic ashes interbedded in the terrestrial sediments provided a rare chance for the precise correlation of deposits as well as the fossils in them [60–63]. Furthermore, the precise dating of the fossilbearing deposits enabled us to better relate the evolution of the Jehol Biota to the major tectonic background [64,65]. Exciting dating results have been obtained from deposits ranging from the Edicaran to the Quaternary, thus providing a solid ground for discussing the interactions between the evolution of different forms of life and their geological and paleoenvironmental background.

Geobiology, derived from the study of geomicrobiology, has also been developing quickly in China [66], and shows a promising future as it can better bridge the geological and biological processes in Earth's history by taking advantage of the latest advances in the study of geochemistry and molecular biology. It is also noteworthy that an institute called the Institute of Geo-Biology was founded in Beijing in 1940, with P. Teilhard de Chardin as the honorary president and zoologist P. Leroy as the director. It was based on the collections and laboratories of the Huangho-Paiho Museum (now the Tianjin Natural History Museum) founded by F. Licent in 1915.

The application of the modern genomic technique to the study of fossils has quickened the development of the discipline of molecular paleobiology, particularly regarding the study of fossil DNA. In the past decade, Chinese paleontologists have succeeded in extracting the DNA sequences from bones of various fossils of *Homo sapiens*, as well that of Denisovan from sediments [67,68]. The rapid progress in this area has also been well connected with the study of archaeology and will certainly make a big impact on the study of the early history of modern humans in China as well as the cultural exchanges between various human populations in the past 100 ka [69,70]. In addition, the extraction of fossil DNA has been increasingly often used in the study of Quaternary fossil mammals such as the Giant Panda.

The study of molecular paleobiology is not limited to the study of fossil DNA. In fact, the study of fossil proteins in deep time has also shown a promising future. A recently published work on the paleoproteomics of the mysterious primate *Gigantopithecus*, dating back 1.9 million years, provided some interesting phylogenetic information [71]. It is also notable that Chinese paleontologists have also been working hard on the identification and detailed analysis of keratin from fossil feathers of dinosaurs and birds from the Jurassic and Cretaceous [72].

The rich paleontological and stratigraphic resources that remain to be further investigated and the unique geological history of the Chinese continent provide numerous chances for the future development of Chinese paleontology, particularly when paleontology is regarded as an integrated part of the Earth System. For instance, the collision between the Indian plate and Eurasian plate and the subsequent rise of the Tibetan Plateau in the Cenozoic has been one of the major tectonic events that has made an important impact on the geography, climate, and history of China. The study of the evolution of life, including human evolution in the Tibetan Plateau against its unique geological background, is expected to produce more exciting multidiscipline results. The subduction of the paleo-Pacific plate towards the Eastern Asia during the late Mesozoic in East Asia has been regarded as related to the evolution of the Yanliao and Jehol Biota, and the study of the control of deep tectonic activities on the surface geological processes and its impact on the evolution of terrestrial life has now drawn attention from paleontologists, geologists, and geochemists.

Technical applications, such as synchrotron, have also played a key role in the study of paleontology in China [73–75]. Several Chinese institutes and paleontological labs in universities have now been equipped with high resolution CT in addition to SEM, TEM, etc. The Synchrotron facilities in Shanghai, Beijing, and Hefei as well as those in Taiwan have also been used by Chinese paleontologists, bringing forward more opportunities and research directions for the younger generation of paleontologists [76–78].

The NIGPA has constructed an impressive Geobiodiversity Database (GBDB) thanks to the longstanding support of the State Key Laboratory of Paleobiology and Stratigraphy and other major projects. Recently, a high-resolution summary of Cambrian to Early-Triassic marine invertebrate biodiversity curves with an imputed temporal resolution of 26 ± 14.9 thousand years was published based on the Geobiodiversity Database (GBDB), which used quantitative data from 11,000 marine fossil species collected from more than 3000 stratigraphic sections in China [42].

Benefitted by the rapid economic growth and the need for popular science education, many new museums of natural history or geology have been constructed in China, which have created more chances for the employment of graduates of paleontology. While the NIGPAS and IVPP remain the two biggest paleontological centers in China, a few universities, e.g., the Chinese University of Geosciences, Peking University, Nanjing University, Northwest University, Lanzhou University, and the Geological Academy of Sciences remain to have a strong paleontological program. New paleontological programs have been established and are growing in several other universities, such as the Yunnan University, Sun Yat-Sen University, Capital Normal University in Beijing, Shenyang Normal University in Liaoning, Hefei University of Technology in Anhui, and Lingyi University in Shandong, etc. In addition, paleontologists are also active in some of the biological institutes, e.g., the Xishuangbanna Tropical Botanical Garden, Institute of Botany of the CAS, and many provincial museums.

Despite the remarkable progress made during the past three decades, the advancement of Chinese paleontology is also held back by several challenges. For instance, the illegal collecting and marketing of vertebrate fossils remain unsolved issues, while scientific collecting often meets great difficulty mainly due to the immature administrative management of fossil resources and lack of local interest.

In sum, due to over 40 years of reform and the open-door policy of China in addition to the rapid growth of the nation's economy, paleontology in China has largely merged into the global paleontological community and became a major force that constantly produces exciting new discoveries of fossils that arouse public interest. Furthermore, it has contributed important evidence to ongoing evolutionary research, which has added to our understanding of the tree of life in deep time. The integration of paleontology with biological and geological sciences has indisputably proved its importance.

#### **6. Future Directions and Challenges**

The new generations of paleontologists in China, who have benefited from the success of their predecessors against the background of digital age and several decades of the fast growing of Chinese economy, seem to be more optimistic and confident in using new technology and methods in paleontological studies. In addition, they are more willing and prepared to participate in multidisciplinary research that often involves a background in geochemistry as well as in molecular and developmental biology. More urgent global changes also encourage them to pay more attention to the interactions between life evolution and paleoenvironment during deep time.

During the digital age, more attention has been paid to the construction of Big Data and the application of AI technology to paleontological study. Paleontological data will probably be better integrated with geochronological and geochemical data, which may help produce more interesting results regarding paleobiodiversity and paleogeography, as well as the impact of paleoenviroment, on the evolution of life on Earth.

Considering the population of China, in the future, there will be even more museums or universities with paleontology programs to increase the employment chances for paleontology students and arouse more public interest in this discipline. Young paleontologists can find their positions either in the department of Earth sciences or biology. Their expertise in the history of the Earth and the evolution of life on Earth will be a necessity not only for students of Earth sciences and biological majors, but also for any students who may be interested in the history of the Earth and life, or the theory of evolution (Figure 7).

**Figure 7.** Group photo of participants of the conference celebrating the 70th anniversary of the Nanjing Institute of Geology and Paleontology in Nanjing in 2021 (credit to NIGPAS). The Chinese characters means a warm congratulation to the 70th anniversary of the Nanjing Institute of Geology and Paleontology of the Chinese Academy of Sciences.

With the growing need for science communication in order the increase the scientific literacy of the public, Chinese paleontologists should be optimistic and enthusiastic about the great potential of paleontology in attracting both kids and adults. It is important to note that when natural sciences are further integrated with liberal sciences, the knowledge gained in paleontology and evolutionary biology will be beneficial to a wide range of readers and future scientists.

However, challenges remain in Chinese paleontology. The practical philosophy of Chinese culture has hindered the development of basic scientific research. Some of the students chose paleontology because they saw it as an advantageous career rather than out of their personal interest. The NIGPAS and IVPP, as CAS's two institutes, seem to be under some pressure to focus on institutional organized projects, rather than small individual projects born out of pure curiosity.

Due to historical and cultural reasons, it seems that Chinese paleontologists are shy to propose new hypotheses, attempt contribute more to evolutionary theory from paleontological evidence, or to try to integrate paleontological data with biological data. Cross-discipline interaction is never easy, and it only happens when the scientific culture is suitable for scientists to focus on purely scientific pursuits. Yu recently provided an interesting account of the social, cultural, and disciplinary factors that influenced the reception and appropriation of Darwinism by China's first-generation paleontologists [79]. With the development of young generations of paleontologists in the new century, it is hoped that Chinese paleontology will continue to make more contributions to world's paleontological community. This is accomplished not only by making more exciting discoveries, but also by engaging in new studies that enrich our understanding of the evolutionary mechanisms of life on Earth and provide more clues to our understanding of the impact of global changes in biodiversity on human evolution.

**Funding:** This research was funded by National Natural Science Foundation of China grant number 42288201.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** I wish to thank Mary H. Schweitzer and Ferhat Kaya for their kind invitation to contribute to the special issue "Paleontology in the 21st Century". Yinmai O'Connor helped improve the English. I also thank Renbin Zhan for providing help with the figures. In addition, three referees provided many valuable suggestions that helped improve the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Commentary* **A Plea for a New Synthesis: From Twentieth-Century Paleobiology to Twenty-First-Century Paleontology and Back Again**

**Marco Tamborini**

Department of Philosophy, Technische Universität Darmstadt, Marktplatz 15 (Residenzschloss), 64283 Darmstadt, Germany; marco.tamborini@tu-darmstadt.de

**Simple Summary:** This article examines the relationship between twentieth- and twenty-first-century paleobiology. After summarizing the disciplinary problem of paleontology in the mid-twentieth century, I focus on five representative research topics in contemporary paleontology. In doing so, I outline twenty-first-century paleontology as a science that seeks more data, more technology, and more integration. At the end of the paper, I highlight a possible new paleobiological revolution: it would give paleontologists strong political representation to deal with issues such as expanded evolutionary synthesis, conservation of the Earth's environment, and global climate change.

**Abstract:** In this paper, I will briefly discuss the elements of novelty and continuity between twentiethcentury paleobiology and twenty-first-century paleontology. First, I will outline the heated debate over the disciplinary status of paleontology in the mid-twentieth century. Second, I will analyze the main theoretical issue behind this debate by considering two prominent case studies within the broader paleobiology agenda. Third, I will turn to twenty-first century paleontology and address five representative research topics. In doing so, I will characterize twenty-first century paleontology as a science that strives for more data, more technology, and more integration. Finally, I will outline what twenty-first-century paleontology might inherit from twentieth-century paleobiology: the pursuit of and plea for a new synthesis that could lead to a second paleobiological revolution. Following in the footsteps of the paleobiological revolution of the 1960s and 1970s, the paleobiological revolution of the twenty-first century would enable paleontologists to gain strong political representation and argue with a decisive voice at the "high table" on issues such as the expanded evolutionary synthesis, the conservation of Earth's environment, and global climate change.

**Keywords:** paleontology; paleobiology; history and philosophy of paleontology; twenty-first-century paleontology; paleobiological revolution; technoscience and global issues

#### **1. Introduction**

At the beginning of his book *Tempo and Mode in Evolution* (1944), North American paleontologist George Gaylord Simpson (1902–1984) provided an interesting picture of the difficulties and lack of interaction between the paleontological and biological communities during the middle of the twentieth century. He wrote, "not long ago [ ... ] geneticists said that paleontology had no further contributions to make to biology, that [ ... ] it was a subject too purely descriptive to merit the name "science". The paleontologist, they believed, is like a man who undertakes to study the principles of the internal combustion engine by standing on a street corner and watching the motor cars whiz by" [1].

The same opinion was shared by prominent biologists. For instance, British geneticist John Maynard Smith (1920–2004) noted that that during the 1960s there were continuous frictions between paleontologists and biologists. As he recalled in 1984, "[at] that time, the attitude of populations genetics to any paleontologist rash enough to offer a contribution to evolutionary theory has been to tell him go away and find another fossil, and not to bother

**Citation:** Tamborini, M. A Plea for a New Synthesis: From Twentieth-Century Paleobiology to Twenty-First-Century Paleontology and Back Again. *Biology* **2022**, *11*, 1120. https://doi.org/10.3390/ biology11081120

Academic Editors: Mary H. Schweitzer and Ferhat Kaya

Received: 27 June 2022 Accepted: 24 July 2022 Published: 26 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the grownups" [2]. However, in the same piece, Maynard Smith added two lines, in effect showing the different perception and value of paleontology. He admitted, "Palaeontology has been too long absent from the high table. Welcome back" [2].

In welcoming paleontology to the table that matters, the geneticist paved the way for some historical and theoretical questions: What happened between the 1960s and 1980s that was so imprinted as to make his judgment change? Does twenty-first-century paleontology still sit at the high-table? And how does the scientific agenda of today's paleontology relate to its scientific past and future?

In this paper, I will briefly address these questions, in effect reflecting on the elements of novelty and continuity between twentieth-century paleobiology and twenty-first-century paleontology. First, I will expose the heated debate about the disciplinary status of paleontology during the mid-twentieth century. Second, I will analyze the main theoretical issue behind this debate by looking at two prominent case studies within the broader paleobiological agenda. Third, I will move to twenty-first-century paleontology and address five representative research topics. By doing so, I will characterize twenty-first-century paleontology as a science that seeks *more data, more technology, and more integration*. Moreover, I will I focus on the notion of the "second digital revolution" as it occurs in paleontology. In the conclusion, I will state what twenty-first-century paleontology might take from twentieth-century paleobiology: the aspiration of and plea for a new synthesis that may lead to a second paleobiological revolution.

#### **2. Paleobiology vs. Paleontology**

"We are paleontologists, so we need a name to contrast ourselves with all you folks who study modern organisms in human or ecological time. You therefore become neontologists" [3]. With his classical provocative and quite "parochialism" jargon, North American paleontologist Stephan Jay Gould (1941–2002) called attention to two important elements that characterized the methodology and disciplinary status of paleontology throughout the twentieth century. First, Gould insisted that a different name was necessary to differentiate two quite different approaches to evolution. According to Gould, the term 'paleontology' was too weak and too historically laden to achieve the aim of promoting paleontologists as genuine biologists and not as geologists. Together with several US colleagues, such as Niles Eldredge, David Raup, Jack Sepkoski, Steven Stanley, and others, he chose the term "paleobiology" as the perfect name to describe paleontological interests in evolutionary biology [4,5]. Paleobiology had its heyday between 1970 and 1985. This name though goes back to Austrian paleontologist and biologist Othenio Abel (1875–1946). He coined the term "paleobiology" to emphasize the biological meaning of this discipline. Paleobiology, so Abel, was able to investigate evolutionary mechanisms, in effect providing the deeptime perspective to evolutionary study [6–8]. Hence, paleontology should be considered a biological sub-discipline and not practiced in geological departments.

Second, Gould contrasted between evolutionary investigations of living organisms (what he called neontological analyses) and the paleo(bio)ontological ones, which study what happened in the deep past. Indeed, Gould's statement derived from the empirical results he himself and other paleontologists were obtaining during the 1960s and 1970s. These scientists were keen to present their discipline as a rigorous investigation of evolutionary mechanisms. By uncovering deep-time patterns and processes, paleontologists were indeed able to contribute to and expand the evolutionary mechanisms set during the modern synthesis of evolution. Gould coupled this theoretical aim with a broader idea of science, paleontology, and evolutionary time [5,9–11].

As noted, the second ground behind Gould's provocative statement was his emphasis on the importance of deep-time investigations as defining characteristic of paleontology. Although the discovery and conquest of deep time was a classical argument for the importance of paleontology since the seventeenth century [12–16], this dimension acquired extra (biological) value in the twentieth century. In 1985, Gould published a paper in the journal *Paleobiology*, arguing that "Nature's discontinuities occur at different scales of time

or tiers". He distinguished three distinct temporal tiers. The first tier includes events at the ecological movement; the second tier encompasses events which occur during "millions of years in "normal" geological time"; whereas the most exciting subject in paleontology lies in our recognition that one of our best-recognized and most puzzling phenomena, mass extinction, is not merely more and quicker of the same, but a third distinct tier with rules and principles of its own [17].

Successively, he stated that "whatever accumulates at the first tier is sufficiently reversed, undone, or overridden by processes of the higher tiers" [17]. For instance, mass extinction occurs at the third tier. It "works by different rules and may undo whatever the lower tiers had accumulated" [17]. Evolution is an extremely hierarchical phenomenon. Hence, Gould noted that only paleobiologists are able to investigate what happened in the second and third tier; whereas "neontologists" study "modern organisms in human or ecological time" [17].

Gould's idea was that evolutionary time could be seen as a system of distinct tiers and the problem of transpacific evolution requires an explicit study of their interaction. Darwinian tradition leads us to deny this kind of structuring, to view time as a continuous, and to seek the source of causality at all scales in observable events and processes at smallest [9,18].

To put it simply, Gould and colleagues meant to "(1) make paleontology more theoretical and less descriptive; (2) introduce models and quantitative analysis into paleontological methodology; (3) import ideas and techniques from other disciplines (especially biology) into paleontology; (4) emphasize the evolutionary implications of the fossil record" [5]. Or, as philosopher Derek Turner put it in seven slogans, "(1) Paleontology has more to contribute to biology than to geology; (2) Study fossils in bulk—individual specimens don't tell you much about evolution; (3) Paleontology needs theories; (4) If you can't experiment, then simulate; (5) don't assume that the fossil record is incomplete; analyze the incompleteness; (6) resist reductionism; (7) don't shy away from raising big questions about evolution" [19]. This does not mean, though, that all these features were literally invented by paleobiologists, but rather that Gould and colleagues put these elements at the center of their research programs.

Following these starting points, Gould asserted that paleontology was a legitimate biological discipline able to uncover patterns and mechanisms which could be found at the second and tier tiers of time [20]. For instance, at the second tier, Gould individuated phenomena which occur primary within a punctuated equilibrium-pattern (as he had formulated years earlier together with Niles Eldredge [21]), whereas the third tier is dominated by mass extinction phenomena. These patterns and mechanisms complete, expand, and in part revise the neo-Darwinian picture of evolution.

Nine months after his article "The Paradox of the first Tier" was submitted to *Paleobiology*, Gould wrote to the journal's editor, Jack Sepkoski, explaining the necessity of an interaction between different and autonomous layers of the evolutionary theory:

Hierarchy, as here discussed in its genealogical context, is an 'internalistic' theory about evolution dynamics. And we need to formulate it properly if we are to tackle this internal dynamic with the other great mover of life's patterns—the externals of geological history especially mass extinctions, that so impact life's history ... in other words, all the data that you and your colleagues are treating in such new and exciting ways. Hierarchy confronts the geological dynamic, and we will not get it right until we reformulate both sides. Gould to Sepkoski, 13 August 1985 in [5].

In this letter, Gould clearly affirmed the importance of reformulating the hierarchical model of evolution in the light of the momentous and exciting research techniques used in the investigation of mass extinctions. The reformulation Gould had in mind, however, concerned the entire paleontological discipline. He suggested a "nomothetic and idiographic" approach to the fossil record based upon David Raup and Sepkoski's studies on the structure of the mass extinction [22–24].

The Kantian philosopher Wilhelm Windelband (1848–1915) coined the nomothetic vs. idiographic distinction. On May 4, 1894, he gave his rectoral address on the methodological differences between History and Natural Science at the Kaiser-Wilhelms-Universität Strasburg. During that address, he distinguished historical from natural sciences by focusing on the "formal character of their cognitive goals" [25]. Nomothetic disciplines aimed at formulating general laws and general judgments, while the ideographic sciences merely collected historical facts. The former were sciences of laws, the latter sciences of events: "the former teach what always is, the latter what once was" [25]. This distinction was not about the contents of the two sciences; but rather it was about how scientists produce knowledge. That means that the main difference between nomothetic and ideographical disciplines was methodological. With his programmatic statements expressed publicly in a paper, Gould intended to bridge the methodological gap between the natural and bio-historical sciences [20].

To accomplish the reformulation, Gould promoted a methodological synthesis. He clearly suggested this in commenting on Sepkoski's study on Phanerozoic diversity [26,27]. Gould affirmed that, "Here we see an interesting and fruitful interaction of nomothetics and ideographics. The form of the model remains nomothetic—the "real" pattern arises as an interaction between two general curves of the same form, but with different parameters. Ideographic factors determine the parameters and then enter as boundary conditions into a nomothetic model" [20].

The ideographic factors mentioned by Gould derive from Sepkoski's famous Compendium. It gave the required data a mathematical treatment of data. This in turn made visible what is invisible: the structure and development of the Phanerozoic diversity [28,29].

Hence, the contrast between paleontology and paleobiology was mainly based on disciplinary and methodological issues. Gould and colleagues first created a disciplinary space in which to insert their research agenda (they named their approach "paleobiology"). Successively, they sought to a possible methodological and disciplinary synthesis, in effect avoiding possible dichotomies. Famously and quoting German philosopher Immanuel Kant, Gould wrote "with all biology and no geology, paleontology is empty; but with geology alone, it is blind" [20]. Furthermore, by calling for a synthesis, another theoretical issue was tackled by twentieth-century paleobiology: the possible over- and under-determination of paleontological explanations.

#### **3. Over- and Under-Determination**

It is quite difficult to find well-preserved fossils that immediately resemble living organisms that can also be exhibited, used, and taken at face value. In fact, once an organism dies, it is subjected to the several taphonomic processes. These processes destroy and change the features of the original organism. For instance, it is rare to find fossils with their soft parts preserved. There are two ways to practically address this problem: (1) focus on structures that are more frequently preserved; (2) focus on exceptionally wellpreserved sites. These provide more selective observations that can be essential, but have their limitations when studying rapid changes during evolution (I thank an anonymous reviewer who pointed this out to me. See [30,31])

The imperfect and incomplete nature of the paleontological record gave several paleontologists cause to reflect on the epistemic aspect of the fossil record. As a result, paleontologists came up with practices intended to overcome the incomplete nature of the records of the past. In addition to working with imperfect and incomplete data, paleontologists must face another difficulty: the so-called over and underdetermination issue.

As philosopher of science Carol E. Cleland has pointed out, historical sciences, such as paleontology, are subjected to the asymmetry of overdetermination [32,33]. They are in the same condition as the investigator who is trying to reconstruct what, exactly, shattered a window starting from the traces on the floor. Let us imagine that three different people throw different objects at the same window at the same time. In that case, "the breaking of the window is overdetermined by numerous sub collections of shards of glass lying on the kitchen floor. That overdetermination of earlier facts by later traces occurs whenever a window breaks" [34].

Following Turner, we can develop the thought experiment a bit further. "The owners of the house sweep up the shards, throw the baseball in the bin, and eventually repair the window. A few weeks later, the only traces of the event that remain are a few shards of glass under the refrigerator. The housecleaning and repair are examples of what Sober (1988, 3) calls *information-destroying processes*" [34].

Let us follow Turner again in this line of reasoning. Let us assume that a future investigator discovers glass shards on the kitchen floor. He will then ask what kind of shards they are—are they pieces of a glass, a window, a vase, etc.? "Even if the historical investigator recognizes the traces for what they are", noted Turner, "rival hypotheses about earlier events and processes will often be underdetermined by the available traces. After studying the shards under the refrigerator, the historical investigator will be completely confused: The evidence does not permit her to discriminate at all between incompatible opposing hypotheses (window vs. wine glass, football vs. baseball, etc.) In other words, she confronts a local underdetermination problem" [34].

Therefore, the present event (a broken window, or, in paleobiology, a mass extinction) over- or under-determines its possible causes. Replaying the tape of time, we cannot be sure to identify the correct sequence of cause–effect, since there are many possible causal chains backwards from the local event.

Hence, first, the record of the past is always imperfect and incomplete, and second, we are not able to state whether our imperfect and incomplete data overdetermine or underdetermine the phenomena paleontologist would like to investigate.

Given these two issues, how can paleontologists bring out patterns and mechanisms that might expand evolutionary theory? Are these two issues not a death kiss for paleontology? And more broadly, what is paleontological business about? To answer these questions, I will briefly recall some research results Gould himself put in the middle of his agenda.

One main topic of twentieth-century paleobiology was the debate about the importance and dynamics of mass extinction [5,35]. One central research result was the famous paper (and graph) representing the periodicity in mass extinctions within geological time. It was used by Gould to indicate a classical phenomenon which happens on the third temporal tier. This representation has many interesting peculiarities, which are related to the notion of paleobiological data [16,29,36,37].

Due to Alvarez's team discovery (1978) of the iridium anomaly in rocks formation at K–T boundary, mass extinction became the hot topic of paleontology in the 1980s. One of the main questions during those years was concerning the number and intensity of the mass extinctions.

This issue was resolved in two famous papers written by David Raup and Jack Sepkoski. Using Sepkoski's database (1982), Raup and Sepkoski identified the number of mass extinctions and the periodicity of this phenomenon. Sepkoski and Raup plotted a huge number of fossil marine families against geological time and found that five mass extinctions clearly occurred in the history of the Earth as "statistically distinct from the background extinction levels". These "five extinction events are seen as sharp drops in standing diversity", and therefore they were easily detected. A mass extinction is thus an event in which a "large number of organisms have disappeared over relatively short time" [22,23].

This case study is emblematic since it shows the *highly integrative practice* used in twentieth-century paleobiology. As for the seminal investigation of morphogenesis conducted by David Raup and Adolf Seilacher during the 1960s and 1970s [11,38], also in this case paleobiologists sought to integrate data and technologies to produce possible coherent scenarios of the past. Since a direct access to the deep past is impossible, scientists

technically recreate possible scenarios which tell the scientists that something happened in the past. Hence, to overcome the issue of over- and underdetermination, paleobiologists stretched and elaborated their data with the help of technology (such as computer, electron microscope, databases, etc.). The key methodological insight was indeed *the fruitful combination of science and technology*. This implied that phenomena such as mass-extinctions or a life-like display of extinct specimens in a museum's hall depend both on the correct use of technological devices and on the interplay between these devices and theories. Hence, paleobiology can be seen as a phenomena-lead discipline. These "investigations are such because the relevance of evidence turns on relationships between phenomena, the hypotheses pertaining to them" [39] and, I would add, the technologies used to stage deep time. Shades of glass on the ground or fossil excavated in a particular region are evidence if these can be used into this integrative process [40].

#### **4. Twenty-First-Century Paleontology: More Data, More Technology, and More Integration**

What is left of the paleobiological research agenda? As I will detail in the conclusion, twentieth-century paleobiology shares with twenty-first-century paleontology the needs for integration and synthesis. In this section, I would like to briefly single out five promising research topics of current paleontology and connect them with twentieth-century paleobiology. I would like to characterize the research of twenty-first-century paleontology with a slogan: more data, more technology, and more integration.

The first topic I chose is about the emergence of paleocolor as a testable field of enquiry. To achieve this aim, it is important to have more clean data. In addition to the publication of studies on feather taphonomy, scientists are teaming up to use ion beam scanning electron microscopy to investigate the preserved melanin pigments. By integrating more data with more technologies, paleocolor can provided precious insights into the behavior and ecology of extinct organisms [34,41–48].

Second, as noted, one main issue paleontologists have to face is the lack of appropriate data. The clarification around the preservability of organic chemicals in fossils, and more broadly, the mechanisms behind taphonomy, is another key research program of twentyfirst-century paleontology. Also in this case, the choice and use of appropriate technologies (such as mass spectrometry methodology, appropriate databases, etc.) is essential to obtain more data and thus pose new questions on phylogenetic hypotheses [49–54].

Third, another promising topic of twenty-first-century paleontology is given by the intersection between morphology, evolutionary theory, and various technologies. The intersection between robotics and morphology provides one compelling example. Scientists are working together with engineers to model and construct bio-inspired robots to understand evolution. One example above all (for another example published recently, see [55]) is given by research on the morphology of *Orobates pabsti*. This is a four-legged vertebrate organism that went extinct about 300 million years ago. The study of the morphology of this well-preserved specimen is very important because it could offer valuable insights into the evolution of terrestrial vertebrates. *Orobates* are an early evolution of the lineage that led to amniotes. These made the vertebrate transition to land by becoming independent of open water during the early stages of development. Thus, studying and understanding how these species were able to transition from water to land is essential to better understand one of the major transitions in vertebrate evolution.

To perform this research, scientists designed the OroBOT robot. It was designed to account for the locomotion dynamics of *Orobates*. The OroBot was built in collaboration with bioengineers at the École Polytechnique Fédérale de Lausanne (EPFL) in Lausanne. The spine of the OroBot was segmented into eight joints: two for the neck, four for the trunk, and two for the tail. The feet consisted of three passive and compliant joints. The designed parts of OroBOT were made of polyamide plastic material and created by selective laser sintering. Working in vivo with these robots, Nyakatura's team was able to reach

conclusions about the complex form-function that characterized the vertebrates' transition to land [56–59].

The same logic is behind the creation of a chewing machine, an artificial mechanical chewing machine, used to study the micro-wear process and the possible correlation with diet and the rate of tooth wear. As the authors noted, "the aim was a simplified system that would allow changing various factors in mastication, for example different standardized feeds, to test whether expectations based on a simplistic interpretation of microscopic and macroscopic wear could be confirmed. The aim of this study was to produce microwear features seen in nature, i.e., pits and scratches, and to quantify macroscopic wear with real teeth and diets, in order to achieve a detailed picture of the wear process and its components" [60]. As in the case of the OroBot, the use of technology and various machines here also brings the investigation of the past (micro-wear process) close to engineering [59,61].

This synergy between technology and paleontological research has also taken different forms in the present day. Among them, one is virtual paleontology [62–67]. This leads to a second digital revolution in paleontology. This second digital revolution is signed by the passage from bio-robotics, or nature-inspired robotics, to robotics-inspired biology. This transition implies bridging of the gap between technology and nature. Form changes should now be studied through in vivo investigations (such as the classical anatomical dissection), in silico (as, for example, through CT scanners or computer simulations), and eventually again in a hybrid and highly integrated in vivo–silico–robotic environment through bio-robotics [61]. The full integration of these methodological levels would help illustrate the structural interplay of elements that characterizes form change.

The fourth hot topic of twenty-first-century paleontology is deeply rooted in the twentieth-century paleobiology. As noted in the introduction, paleobiology was launched to reorient the paleontological agenda toward broader evolutionary problems. This was the main reason for the paleobiological revolution of the 1960s and 1970s—think of Eldredge and Gould's call for punctuated equilibrium. The same search for new evolutionary mechanisms pervaded paleobiological research during the first encounters with the emerging evo-devo community. Or rather, the evolution of evo-devo was shaped by paleobiological questions from the beginning, and vice versa. In fact, at the 1981 Dahlem conference on "Evolution and Development", which will be considered as the grounding meeting of evolutionary developmental biology as an autonomous evolutionary discipline, biologists and paleontologists discussed together the relationship between evolution and development. For instance, in the working group on "The Role of Development in Macroevolutionary Change" biologists, such as Jim Murray, Pere Alberch, Brian Goodwin, Gunther Wagner, Tony Hoffman, and David Wake, defined with paleontologists Stephen Gould, Adolf Seilacher, and David Raup the role of constrains in evolution [11,68–70].

The same line of continuity and interaction denotes current paleontology, which even formalizes the paleontological contribution to evo-devo as Paleo-Evo-Devo. Therefore, the paleontology of the twenty-first century is also characterized by the search for a strong integration in this fourth case [71–79].

The fifth topic I have singled out is also anchored in the twentieth-century paleobiological agenda: the discussion of abiotic and biotic factors in evolution. Recall that I opened this paper with a quote from Maynard Smith. He welcomed the return of paleontology to the high table as a result of Raup and Sepkoski's findings on mass extinction and the consequent focus on biotic elements in evolution. This research, in turn, had emerged from the study of the dynamics of Phanerozoic marine paleodiversity (and the proposed model analyses for studying diversity) and the Red Queen model proposed by Leigh Van Valen [5].

Twenty-first-century paleontology is capitalizing on this line of research. At the same time, paleontologists are trying to synthetize biotic and abiotic factors. For instance, Mike Benton notes, "The realization that the Red Queen and Court Jester models may be scaledependent, and that evolution may be pluralistic, opens opportunities for dialog" [80]. He

goes on asserting, "methods are shared by paleontologists and neontologists, and this allows direct communication on the patterns and processes of macroevolution" [80]. Fortelius and colleagues are also proposing "a tentative synthesis, characterized by interdependence between physical forcing and biotic interactions" [81].

Along these lines, paleontologist Tyler Faith and his colleagues emphasized the key methodological elements of current hominin evolution. First, paleoanthropology, as paleontology, is all about data and time scales (or as Gould put it tiers): "There is no universally 'correct' scale of observation, but to address questions linking ecology and evolution, the scales of the processes of interest must align with the scales of the available data" [82]. Second, paleoanthropology is desperately in need of more theory (recall Turner's third point for describing paleobiology), or rather of a new hypothetic-deductive methodology [83]: "Incorporating a stronger theoretical framework into the agenda of hominin paleoecology will allow researchers to reverse the typical direction of inference (i.e., from data to hypothesis) by generating theoretically informed predictions that are tested with the data, and then determining if a hypothesis should be modified or rejected" [82]. By granting a new methodology, a new balance may emerge "between inferring evolutionary narratives from the data and testing process-based hypotheses using those data" [82]. Hence, again, twenty-first-century paleo-research is about more data, more technology, and more integration.

#### **5. Conclusions: A Plea for a New Synthesis**

What do the five topics just discussed tell us about the elements of continuity between twentieth-century paleobiology and twenty-first-century paleontology? First, both enterprises called paleontologists' attention to a *cooperative effort* to understand the (deep) past. As paleontological data are always imperfect and incomplete, paleontologists should omnivorously assimilate every method to successfully and opportunistically work on and with deep time [40,84,85]. Molecular approaches to the deep past are therefore supplementary not antagonistic to classical paleontological analyses of forms [86]. Sometimes, they allow paleontologists to see things better or, at least, in a different fine grade (e.g., [87]). How, however, can paleontologists confidently use data, names, technology, and knowledge outside their field of application if they do not know the correct boundaries of application of such tools in their own? Critical integration of different methods can help mediate and hopefully overcome technical issues. This should be paired with inter- and multidisciplinary programs to enable students and scholars to fruitfully learn and apply different methods—that was one of insights of the paleobiological revolution of the 1960s and 1970s.

Indeed, this implies a return to the sprit embodied by the paleobiological revolution of the mid twentieth-century. At the end, what Gould and colleagues were asking for was a genuine synthesis of knowledge. They asked for blurring the disciplinary borders between natural and historical sciences as well as between science and technology. This spirit should be put at the center of twenty-first-century paleontology – this agenda is also perused by other technoscientific disciplines such as biorobotics, synthetic biology, nano(bio)technology, etc. See, for instance, [59,61,88–91]. This enterprise should be characterized by a cooperative model of knowledge production. Instead of supporting a never-ending disciplinary struggle to define what paleontology is (or is not), to draw a sharp line between paleontology and neontology, or between morphological and molecular phylogenetics, scholars should work together and ideally share their data and method to address new challenges, as has happened during periods of major theoretical transitions in the history of paleontology [5,29,92]. The results provided by Perri et al. [93] are a clear example of the continuum of approaches that characterizes paleontology and biology.

A future task for philosophers and historians of paleontology would be to understand the various practices and reasons for the paleontological plea for synthesis and integration in the last century. In other words, the analysis should focus on what (and why) paleontologists sought (and are seeking) collaboration with neontologists and promote the circulation of knowledge and technologies [94] (an important starting point would be the analysis

of how paleontologists and neontologists are publishing together or taking part in joint conference. I thank one referee for this point. See, for instance, the papers gathered in the special issue "Crossing the Palaeontological-Ecological Gap" published in *Methods in Ecology and Evolution* 2016, 7. Furthermore, it is perhaps worth noting that Wolfgang Kiessling was the first paleontologist to play a major role in the Intergovernmental Panel on Climate Change (IPCC) report on climate change, working on "Climate Change 2021: Impacts, Adaptation and Vulnerability", thus bridging the gap between the two communities - I thank one referee for this. See also [35,95–97]).

This point will make visible the role of the so-called invisible technicians, i.e., all those who work (silently and without much recognition) on the production of paleontological knowledge. In addition to the flood of data and the transformation of the paleontologist into a data scientist (and vice versa), *field and preparatory work needs to be recognized as it remains at the heart of paleontology today* [98–100]. Today, more than ever, the classic metaphor of the earth as an archive of knowledge and data is still valid. In fact, the greatest amount of data is still buried in the field. On the importance of field work for paleontology [101–106]. This focus will create an awareness of the new social hierarchies in twenty-first century paleontology that result from the massive use of technology.

Hence, and to conclude, if constructively interpreted and read through the recent history of paleontology, the five topics singled out may provide some potentialities for a new synthesis between genetic and morphological approaches to the evolution of forms in the attempt to overcome the limits, issues, and problems of phylogenetic reconstructions. The major implication would be a broader reflection on what paleontology might become in light of recent technical and molecular revolutions. This would imply a second paleobiological revolution. In fact, as David Sepkoski put it, "the paleobiological "revolution" was more like a political contest in which one group perceives itself to be disenfranchised and agitates for greater representation in government than a contest of lofty ideas. The 1970s was a period of revolution in paleontology because paleobiologists saw themselves, and described what they were doing, as revolutionary" [5]. Echoing the paleobiological revolution of the 1960s and 1970s, a twenty-first-century paleobiological revolution would enable paleontologists to gain strong political representation and to argue with a decisive voice at "the high table" on topics such as the extended evolutionary synthesis, the Anthropocene, conservation of Earth's environment, and global climate change.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** I thank the guest editors of the Special Issue and the three anonymous referees for their valuable feedback.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Biology* Editorial Office E-mail: biology@mdpi.com www.mdpi.com/journal/biology

Academic Open Access Publishing

www.mdpi.com ISBN 978-3-0365-7573-5