Access to this full-text is provided by PLOS.
Content available from PLOS One
This content is subject to copyright.
RESEARCH ARTICLE
Fossil bacula of five species of Borophaginae
(Family: Canidae): Implications for their
reproductive biology
Daniel Varajão de LatorreID
1,2
*
1Department of Integrative Biology, University of California, Berkeley, Berkeley, California, United States of
America, 2University of California Museum of Paleontology, Berkeley, California, United States of America
*daniel.latorre@berkeley.edu
Abstract
The baculum of mammals offers the opportunity to study the reproductive biology of extinct
species given that it is a fossilizable part of the male genitalia and that its size and shape cor-
relate with several aspects of the reproductive biology of extant mammals. Fossil bacula,
however, are rare. Currently, bacula have been described from only two extinct species of
canids, one from the subfamily Caninae and the other from the extinct subfamily Hespero-
cyoninae. Here, I describe the bacula of five extinct species of Borophaginae, each of which
was found with other skeletal elements that have enabled identification to the species level.
Two specimens (Aelurodon ferox and Aelurodon stirtoni) are largely complete, while the
baculum from Carpocyon compressus is complete but still embedded in matrix that
obscures some of its features. The bacula of Paratomarctus euthos and Desmocyon thom-
soni are incomplete, but they provide useful information nonetheless. These borophagine
bacula are similar to extant canines in being robust, having a urethral groove, and a simple
distal end. These features suggest that the Borophaginae had long-lasting copulation and
possibly spontaneous ovulation, similar to the extant canines. However, unlike the straight
baculum of extant canines, borophagine bacula are ventrally curved (arched), which is also
observed in the hesperocyonine baculum. The implication of this curvature for the reproduc-
tive biology of these animals remains unknown.
Introduction
Studying the reproductive biology of extinct animal groups is challenging. While neontologists
can observe directly the morphology, behavior, endocrine responses, and life cycles of their
study organisms, it is extremely rare for paleontologists to find direct evidence of reproductive
biology [e.g., a fossilized pair of mating insects 1]. Instead, paleontologists must rely largely
on morphological traits that fossilize, and on the fidelity of correlations between these traits
and other aspects of reproduction seen among extant species. As a result, we have much less
information regarding the reproductive biology of extinct species than for living species. Con-
sistent with this, sexual selection and mating represent less than 5% of all the entries in the
PLOS ONE
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 1 / 15
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Varajão de Latorre D (2023) Fossil bacula
of five species of Borophaginae (Family: Canidae):
Implications for their reproductive biology. PLoS
ONE 18(1): e0280327. https://doi.org/10.1371/
journal.pone.0280327
Editor: Luca Pandolfi, Universita degli Studi della
Basilicata, ITALY
Received: April 23, 2022
Accepted: December 27, 2022
Published: January 17, 2023
Copyright: ©2023 Daniel Varajão de Latorre. This
is an open access article distributed under the
terms of the Creative Commons Attribution
License, which permits unrestricted use,
distribution, and reproduction in any medium,
provided the original author and source are
credited.
Data Availability Statement: All relevant data are
within the paper.
Funding: DVL received funding from the University
of California Museum of Paleontology Research
Award. ucmp.berkeley.edu. The funders had no
role in study design, data collection and analysis,
decision to publish, or preparation of the
manuscript.
Competing interests: The authors have declared
that no competing interests exist.
compendium of behaviors inferred from fossils [Hsieh and Plotnick, [2], tabulated Boucot and
Poinar 1’s, [3], compendium]. Nonetheless, when aspects of reproductive biology can be
inferred, they can provide important insights into observed changes in biodiversity. For exam-
ple, Bush et al. [4] showed that marine animals with direct fertilization drive the diversification
of marine fauna during the Cretaceous-Cenozoic while groups that spawn in the water column
have a relatively constant diversity over the same period.
In this context, the penis bone of mammals (baculum, os penis or os priapi) provides an
opportunity for drawing inferences regarding the reproductive biology of extinct species [e.g.,
5]. The baculum is present in 7 mammalian orders, within which it has been gained and lost
multiple times [6]. Interestingly, females of some species have ossa genitalia homologous to the
baculum, the baubellum or os clitoridis, but it is normally smaller than the male counterpart
[e.g.,7] and is largely understudied, to such an extent that it has not been associated with any
aspect of animals’ reproductive biology. The baculum can inform our understanding of the
reproductive biology of extinct species because it is a part of the male genitalia with potential
for fossilization and because its morphology is known to correlate with several aspects of the
reproductive biology of extant mammals. For example, in carnivorans induced ovulation
tends to occur in species that have bacula with a more complex distal tip [8]. Moreover, more
robust bacula are associated with prolonged copulation [9], and longer bacula are associated
with larger testis [8,10], which is indicative of stronger sperm competition [11].
A good fossil record for carnivoran bacula would allow us to explore how the reproductive
biology of this mammalian clade has changed through time. However, very few extinct species
have had bacula described [but see 5,12–17, for the description of the baculum in eight spe-
cies]. To date, bacula have only been described from two extinct canid species: gregarius in the
extinct subfamily Hesperocyoninae [16], and the dire wolf, Aenocyon dirus [17,18], in the still
extant subfamily Caninae. Here, I add to our knowledge of canid bacula by describing the
bacula of five species of Borophaginae, the other extinct subfamily of Canidae. All specimens
are from North America and were found associated with sufficient other skeletal elements that
allowed identification to the species level by Wang et al. [19]. Without the association with
other skeletal material, species level identification would not be possible, as fossil bacula are
rare and lack diagnostic traits. Although there is always a chance for misidentification of fossils
identified from associated elements, I consider that the chances are low for the specimens
described here, given previous work on these specimens [19] and considering that none of
them are from fossil rich localities or bone beds. In fact, most of the specimens are the only
ones from the localities in which they were found. These are the first descriptions of bacula
that can be reliably attributed to Borophaginae.
The subfamily Borophaginae has a rich fossil record, spanning over 30 Ma, from the Orel-
lan (base around 34 Ma, early Oligocene) to the Blancan (top around 1.8 Ma, early Pleistocene)
with at least 66 recognized species [19]. They were the most diverse group of carnivores in the
middle Miocene in North America [20], and by that time had radiated from a small hypocarni-
vorous ancestor to exhibit a diverse range of body sizes and diets [21]. Functional morphologi-
cal studies have led to opposing conclusions regarding whether they formed groups to hunt or
not [22,23]. Currently, very little is known about their reproductive biology—the bacula
described here offers new insights into this aspect of their biology.
Bacula of extant canids
I use the bacula of the extant canids [all from subfamily Caninae, 24–26] as a reference for the
anatomy of Borophaginae bacula. Most extant canines have a fairly straight baculum in lateral
profile, although it may curve slightly ventrally (concave down) or dorsally (concave up). The
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 2 / 15
baculum is a distal continuation of the high pressure erectile tissue named corpora cavernosa
[27,28], and the base of the baculum (proximal end) shows marks of soft tissue attachment
(scars). On the ventral side there is a deep groove that runs from the proximal end to about
three quarters down the length of the shaft (Fig 1). The urethra sits within this groove. The
presence of the groove is associated with lateral expansions of the baculum along its ventral
margin, which also results in a distinctive proximo-distal dorsal crest. Most individuals have a
peak somewhere along the proximal-half of the dorsal crest, but it is absent in some specimens.
Further, the bulbus glandis, another high pressure erectile tissue present in extant canids and
involved in the copulatory tie of canids, engorges around the proximal three quarters of the
baculum, where the urethral groove is present [28]. The distal end (anterior) of the baculum is
simple in canids, having either a cylindrical rod-like shape, or being slightly expanded dorso-
ventrally. The cartilaginous distal-most portion of the baculum, the apex, can calcify with age
[25] but is rarely observed in zoological collections.
Descriptions of borophagine bacula
Museum Abbreviations: AMNH—American Museum of Natural History, New York, NY,
US; F:AM—Frick Collection, Department of Vertebrate Paleontology, American Museum of
Natural History, New York, NY, USA; USNM PAL—Smithsonian Institution National
Museum of Natural History Paleontology collections, Washington, D.C., USA; UNSM—Uni-
versity of Nebraska State Museum, Lincoln, NE, USA.
Paratomarctos euthos
Paratomarctus euthos was a coyote-sized species [29] that can be considered mesocarnivorous
based on its dentition [21], with a tendency towards hypercarnivory [19] and that could take
prey larger then themselves [22]. The distal portion of the baculum of Paratomarctus euthos
(Fig 2) was found associated with fragments of the skull and mandible as well as numerous
post-cranial bones [specimen F:AM 61088—19]. This specimen is from the late Clarendonian
(13.6-10.3Ma, middle Miocene) from the Merritt Dam Member of the Ash Hollow Formation,
Cherry Co., Nebraska. This partial skeleton is the only specimen from its locality in the
AMNH collections. The preserved portion of the baculum is 7cm long. Despite the absence of
Fig 1. Canine baculum. Left lateral view of the baculum of coyote (Canis latrans) with visible urethral groove on the ventral side, a dorsal ridge, and
soft tissue attachment scars on the base. (Specimen 74824 from the Museum of Vertebrate Zoology, Berkeley, CA, USA).
https://doi.org/10.1371/journal.pone.0280327.g001
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 3 / 15
the proximal end, it is clear that the baculum was robust. The ventral side of the proximal por-
tion exhibits a shallow and wide urethral groove, with a corresponding dorsal crest. Distally,
the groove becomes gradually shallower and narrower until it disappears about half way along
the preserved portion of the baculum. The distal end consists of a small conical projection of
rough bone. At the base of this cone there is an almost continuous circumferential ring of
bone. The baculum is sigmoidal in lateral profile. There is an overall ventral curvature (con-
cave down) that changes to a dorsal curvature (concave up) at the distal third of the specimen.
Given the aspect ratio (length to width) of the specimen it seems likely that the fragment repre-
sents at least 50% of the total length of the baculum.
Desmocyon thomsoni
Desmocyon thomsoni was a medium-sized [around 7kg—22] mesocarnivore species that ran-
ged from late Arikareean to early Hemingfordian. Its dental morphology is intermediate
between the earlier hypocarnivorous borophagines and the larger hypercarnivorous species
common in the later diversification of the subfamily [19,21,22]. Two partial bacula of
Desmocyon thomsoni (Fig 3) are associated with partial skeletons, including partial skulls and
mandibles as well as many elements of the post-crania skeletons [19]. Both are from the late
Arikareean (30.8—20.43Ma, Oligocene-Miocene) of Wyoming. Specimen F:AM 49097 was
found in the Lusk area, Upper Harrison beds in Goshen Co., Wyoming, and it is the only spec-
imen from its locality in the AMNH collections. Specimen F:AM 50213 is also from the Lusk
area, Upper Harrison beds, but in Niobrara Co., Wyoming. Only one other specimen (identi-
fied as Promartes) was found in the same locality of F:AM 50213.
Specimen F:AM 50213 has a complete proximal end where marks of soft tissue attachment
are present on the lateral surface. These marks form a 1.5 cm long diagonal line that extends
from the dorsal and proximal extremity to the ventral face of the shaft. On the lateral surface,
and dorsal to this diagonal line, the bone is rougher but transitions to smoother bone distally.
The fragment is mostly straight in lateral profile, except for a slight ventral curvature that starts
about one centimeter from the broken end of the specimen. In ventral view there is a deep ure-
thral groove that is narrow and cylindrical. The proximal end of the baculum is laterally com-
pressed, corresponding to where the urethral groove disappears. This differs from most
Fig 2. Paratomarctos euthos. The partially preserved baculum of specimen F:AM 61088, exhibiting the complete distal end. A) Left lateral view
showing the sigmoidal profile of the bone. B) Ventral view showing the beginning of the ventral groove about half way down the specimen, which then
expands proximally. C) Dorsal view showing the lateral expansion where the ventral groove is present and the resulting dorsal crest. A,B and C are at
the same scale; anterior at left. D) Slightly oblique view of the broken end. E) Oblique view of distaltip. F) Inferred cross-sections at two points along
the specimen with caliper measurements indicated.
https://doi.org/10.1371/journal.pone.0280327.g002
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 4 / 15
modern canids, in which the urethral groove extends all the way to the proximal end, but is
similar to the baculum of the gray fox, Urocyon cinereoargenteus [26]. The dorsal surface of the
baculum presents a strong proximal-distal crest, which can be seen clearly in the cross section
of the broken surface (Fig 3E).
Specimen F:AM 49097 is slightly longer than F:AM 50213 (6.9 cm and 5.63 cm, respec-
tively), but it is more abraded and is missing both the proximal and the distal ends. This speci-
men preserves more of the distal end than F:AM 50213, revealing a ventral curvature (concave
down). The groove is not visible on the either the proximal or distal broken surfaces and is
thus restricted to the middle portion of the bone; it is filled with sediment, which obscures the
ends of the groove. The dorsal crest is present in the middle portion of the fragment, but not at
the proximal and distal ends. Using the proximal end of the crest and lack of the urethral
groove on the proximal break surface as reference points to align the two specimens, it is possi-
ble to infer that only a small portion of the proximal end of F:AM 49097 is missing. It is not
possible to see where the groove ends distally due to sediment and abrasion, but the groove is
not present on the distal break surface (as visible in Fig 3D). The specimen starts to taper dis-
tally and the dorsal crest is not present along the distal portion. Given these features of the dis-
tal portion and given that other canid bacula have fairly simple distal ends, it seems likely that
only a small fraction of the distal end of this specimen is missing.
Aelurodon ferox
Aelurodon ferox is among the largest described species of borophagines, with an estimated
body mass around 31kg [22]. It is considered a hypercarnivorous species that could take prey
larger then themselves and with adaptations for crushing bone [19,22]. One baculum of
Fig 3. Desmocyon thomsoni. Bacula from specimens F:AM 49097 (upper image in each panel) and F:AM 50213 (lower image in each panel). In each
panel the specimens are roughly aligned based on the proximal ends of dorsal crest—it appears that F:AM 49097 is only missing a small portion of its
proximal end, while F:AM 50213 is missing a larger portion of its distal end. A) Left lateral view, showing the ventral curvature in F:AM 49097 and soft
tissue attachment marks on the proximal end of F:AM 50213. B) Right lateral view. The right side of specimen F:AM 49097 is less abraded giving a
better view of the dorsal crest, which is absent from the proximal and distal ends. C) Dorsal view shows the dorsal crest of both specimens, and the
triangular distal end of F:AM 50213. D) Ventral view showing the urethral groove in both specimens, although it is filled with matrix in specimen F:AM
49097. The distal surface for this specimen also shows that the specimen broke distally to the distal end of the groove. E) Cross section of the baculum F:
AM 50213 where there is a clean break in the specimen. F) Inferred cross sections with caliper measurements indicated.
https://doi.org/10.1371/journal.pone.0280327.g003
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 5 / 15
Aelurodon ferox was found together with a partial skeleton, specimen F:AM 61723, which has
a partial mandible and numerous post cranial bones [19]. The specimen is considered Claren-
donian in age (13.6-10.3Ma, middle Miocene), from the Pojoaque Member of the Tesuque
Formation (Santa Cruz Grant, Santa Fe Co., New Mexico), and it is the only specimen from its
locality in the AMNH collections.
While Wang et al. [19] mentioned the presence of a “partial baculum”, the specimen
appears to be mostly complete, missing only a minimal portion of the proximal end. It is a
robust bone measuring 11.7 cm in length. When observed in lateral view, it curves ventrally
and tapers distally (Fig 4). Soft tissue attachment scars are not visible at the proximal end,
although the bone seems slightly rougher at that end. A urethral groove is not visible at the
Fig 4. Aelurodon ferox. The baculum of specimen F:AM 61723. A) Dorsal view showing the lateral expansion on the middle part of the shaft where the
ventral groove is present. B) Lateral view showing the ventral curvature and the tapering of the bone distally. C) Ventral view showing the urethral
groove. Same scale applies to A,B and C. D) Closer view of the distal tip (1cm scale bar). E) Inferred cross-sections at four points along the bone with
caliper measurements indicated.
https://doi.org/10.1371/journal.pone.0280327.g004
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 6 / 15
proximal end, but begins about 2cm from the proximal end and becomes deeper until about
two thirds along the shaft, where it again becomes shallow until it disappears where the distal
curvature becomes strongest. Where the groove is present, there is a lateral expansion of the
lower (ventral) half of the bone which leaves the dorsal crest visible. The dorsal (and ventral)
view reveals that the proximal end is laterally compressed where the groove is absent. The dis-
tal end is almost cylindrical in cross section. The distal face of the distal end shows rougher
bone, similar to some modern canids.
Aelurodon stirtoni
Aelurodon stirtoni was also a hypercarnivorous species that likely crushed bones when feeding
and that could take prey larger than themselves potentially as social hunters, but was smaller
than A. ferox with approximately 20kg [22]. The baculum of specimen USNM PAL 215320
(Fig 5) was found with an almost complete skeleton, with good preservation quality, including
a partial skull and mandible with all teeth preserved [19], as well as stomach content. It is from
the late late Barstovian (15.97-13.6Ma, middle Miocene) and was found in the Burge Member
of the Valentine Formation, Cherry Co., Nebraska [19]. Although several parts of the skeleton
have already been described and imaged [30], the baculum has not. My description of the
baculum is based on pictures provided by the Smithsonian National Museum of Natural His-
tory. The specimen is on display at that museum and the mount added to position the baculum
obscures the dorsal side of the proximal end.
Fig 5. Aelurodon stirtoni. The baculum of specimen USNM PAL 215320. Dashed area indicates material added to the specimen so that it could be
mounted as part of an exhibit. A) Dorsal view, which indicates the lack of a lateral expansion seen in other species where the urethral groove is present.
The mount obscures details of the proximal end. B) Left lateral view showing the ventrally curved profile. C) Ventral view showing the presence of a
shallow groove along most of the length of the shaft but that disappears at both the distal and proximal ends. D) View of the distal end. There was
insufficient information to infer cross-section shapes. Images courtesy of the Smithsonian Institution.
https://doi.org/10.1371/journal.pone.0280327.g005
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 7 / 15
The specimen is complete and has a simple shape. In lateral profile it curves ventrally, show-
ing little variation in dorso-ventral height along the shaft. On the ventral side a shallow groove
is present along most of the shaft, disappearing just before the proximal and distal ends. In
contrast to the bacula described above, the baculum in this species lacks the lateral expansions
typically associated with the urethral groove, and thus there is no dorsal crest. The proximal
end is triangular and displays rougher bone that is typically associated with soft tissue attach-
ment. The distal end has an approximately circular circumference (although slightly pinched
on the dorsal side) and consists of rough and porous bone.
Carpocyon compressus
Carpocyon compressus was a medium sized species [about 16kg, 22] with dental adaptations
indicating a reversal to a hypocarnivorous diet [19]. The baculum of one specimen (UNSM
128864, Fig 6) was found with an almost complete and associated skeleton, between the hind
limbs and near the pelvic bones. The specimen also has a complete skull and mandible that
allowed its identification [19, where they refer to the specimen field number UNSM 2556-90].
It is late Barstovian (15.97-13.6Ma, middle Miocene), from the Valentine Formation, Brown
Fig 6. Carpocyon compressus. The baculum of specimen UNSM 128864. A) Ventral view showing a deep urethral groove. B) Right lateral view, with
the ventral side facing up on the picture, showing the baculum’s ventral curvature. C) Distal end in left lateral view. There was insufficient information
to infer cross-section shapes.
https://doi.org/10.1371/journal.pone.0280327.g006
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 8 / 15
County, Nebraska [19]. Here I describe the baculum based on photographs provided by the
University of Nebraska State Museum. The baculum is partially embedded in matrix along
with other post-cranial bones.
The specimen appears to be complete but has one break close to the proximal end. This
break slightly alters the lateral profile, but a ventral curvature is clear. The ventral side contains
a deep urethral groove that extends for almost the entire length of the bone, ending about 1 cm
before the distal end and about 1 cm from the proximal end. Although the dorsal side is still
embedded in matrix, in lateral view lateral expansions associated with the groove can be seen,
as can the corresponding dorsal crest. The matrix covering the proximal end makes it difficult
to determine if there were any soft tissue attachment scars. The distal tip is laterally com-
pressed but has a height similar to the height along most of the shaft. This is in contrast to
most canid bacula, which tend to show compressed and cylindrical distal ends.
Discussion
The bacula of the five borophagine species described here display similarities to each other and
to the bacula of all other canids. Specifically, the bacula of extant canines, the extinct hespero-
cyonine H. gregarius, and borophagines are similar in terms of: i) relatively large size; and, ii) a
ventral urethral groove along much of the length of the shaft but not at the distal tip. In two
traits the five borophagine bacula are similar to H. gregarius but differ from the extant canines:
i) their ventral curvature (concave down) in lateral profile; and ii) the closed urethral groove at
the proximal end. Finally, the simple and mostly straight distal end of borophagine bacula is
similar to extant canids but differs from the up curved “hook like” distal end of H. gregarius’
baculum [16]. Fig 7 shows lateral view silhouettes of these bacula for comparative purposes,
with example bacula of species in other families included to capture some of the variation of
bacular morphology. Below I discuss how morphological features of borophagine bacula may
have implications for their reproductive biology.
Relative size
Similar to modern canids, the bacula of the Borophaginae were large and robust bones. This
supports the observation from Wyss and Flynn [36] that a large baculum is generally diagnos-
tic for the Suborder Caniformia (which includes Canidae, Ursidae, Musteloidea, Pinnipedia,
and Amphicyonidae) and contrast with the generally reduced bacula of the sister group, the
Feliformia.
Implications for reproductive biology. Several studies have explored the relationship
between baculum length and the reproductive biology of the living carnivorans [e.g.,8–10,37–
40]. These relationships can thus be used to draw inferences about the reproductive biology of
the Borophaginaes. From a functional standpoint, the baculum adds rigidity to the erect penis,
therefore the longer and more robust it is, the more it can aid in the maintenance of long-last-
ing copulations. For example, in extant carnivoran species with long-lasting copulations
(defined as more than 3 minutes) tend to have longer bacula [whether phylogeny is considered
40, or not 37], but this relation-ship is not significant when using duration of copulation as a
continuous character [9,38].
Among extant carnivorans, there is a weak correlation between longer bacula and larger
testis mass, when both body mass and phylogeny are accounted for [10,39, but see 8, who
found just a weak signal for this relationship]. Larger testis mass is frequently interpreted as a
good indicator of the strength of sperm competition because increased sperm production
increases the chances of fertilization when females may mate with more than one male during
a single round of reproduction [reviewed in 11,41].
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 9 / 15
Both of these relationships—copulatory duration and likelihood of sperm competition—
suggest that the baculum plays a role in postcopulatory sexual selection, especially in species in
which longer copulations may function as a form of mate-guarding that prevents a female
from mating with additional males [42] and/or greater sperm production to increase the odds
of paternity over competitors. Given that these mechanisms of postcopulatory sexual selection
are correlated with large bacula, the relatively large size of bacula within the Borophaginae sug-
gests that similar mechanisms may have been present in these extinct taxa. Given the limited
sample size for the Borophaginae and the relatively weak relationships between baculum
length and these aspects of postcopulatory competition in extant carnivorans [8], I have
refrained from using these relationships to estimate actual values for copulatory duration and
testis mass for the Borophaginae.
Fig 7. Lateral view silhouettes of bacula of borophagines and selected species within the suborder Caniformia.The
specimens represented were chosen to illustrate a substantial range of morphological variation of the canid bacula as
well as a few representative morphologies observed in other families. [For further examples of morphological variation
of bacula refer to 24,26,31]. Drawings for the Borophaginae are based on photographs in this paper. Silhouettes from
extant species are based on specimens observed at the Museum of Vertebrate Zoology, Berkeley, California. Drawings
from extinct species are based on published images for Amphicyon ingens [14], Daphoenus vetus and Hesperocyon
gregarius [32]. The underlying phylogeny is for illustrative purposes and it is based on: Slater [21] for the relationship
between the Borophaginae, Caninae and Hesperocyoninae; Slater and Friscia [33] for the position of Ursidae (grizzly
bear), Mustelidae (marine otter and american mink), and Procyonidae (coati); Paterson et al. [34] for position of the
Pinnipedia; and Tomiya and Tseng [35] for the position of the extinct Amphicyonidae (Amphicyon and Daphoenus).
Time scale in million years ago, with ranges of extinct species projected to last fossil occurrences of the taxon.
https://doi.org/10.1371/journal.pone.0280327.g007
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 10 / 15
Urethral groove
The urethral groove is present in all bacula described from canids, including H. gregarius [16],
the borophagine specimens examined here, as well as the extant and extinct canines [24–26].
This suggests that this trait is plesiomorphic for the family.
Interestingly, a groove is also present in other families within the Suborder Caniformia,
including the extinct and basal Amphicyonidae [13,14], some mustelids, and pinnipeds.
Canids, however, appear to have a unique configuration of the urethral groove. Based on pub-
lished descriptions [e.g. 24,26] and my own observations of the bacula of the other caniform
families, the canids appear to be the only taxon in which the urethral groove is deep, runs for
most of the length of the baculum, and is absent at the distal end. While the configuration of
the canid urethral groove appears to be unique, it does show variation, mostly in its depth,
width, and whether it terminates before the proximal end or not. For example the groove of
Aelurodon is shallower than in other canids.
In contrast, in the members of other carnivoran families that have a deep groove, the groove
continues to the distal most tip of the baculum, and may even be restricted to the distal tip, for
example, extant otters (Lontra,Lutra, and Pteronuria), weasels (Mustela), and bearded seals
(Erignathus barbatus) (personal observation). The deep distal groove is also observed in two
fossil bacula described from the Amphicyonidae [Amphicyon ingens and Daphoenus,13,14,
respectively], a stem family of the Caniformia [35]. Finally, some bears (Family Ursidae—e.g.,
spectacled bear Tremarctos ornatus, and polar bear Ursus maritimus, personal observation)
have a urethral groove in a similar position (on the middle of the shaft, absent at the distal
end), but much shallower than that of canids.
Finally, while most canine species have an open urethral groove at the proximal end [e.g.
26], there are two exceptions, the gray fox [Urocyon cinereoargenteus—26] and the bat-
eared fox (Otocyon megalotis—personal observation), both of which are more similar to the
borophagine condition of having a closed proximal end. Interestingly, the gray fox is the
most basal species in phylogenetic reconstructions of canines that include fossils [e.g.,43],
but see [44], for a discussion of the position of the root of extant canids phylogeny], suggest-
ing that the transition from a closed to an open proximal end might have occurred within
the Caninae.
Implications for reproductive biology. It has been suggested that the urethral
groove protects the urethra during copulation, facilitating the delivery of sperm [37].
Functional experiments support this inferred function in some bats [45], although the shape
and position of the groove in these species is different from that of canids. In silico func-
tional models of grooves more similar to those seen in canids [9] have demonstrated that
the groove increases the robustness of bacula subject to dorso-ventral forces. Given this
result, and given that robustness is correlated with long-lasting copulations [9], it seems
likely that a reinforced baculum with a urethral groove that aided in the delivery of sperm
and enabled prolonged copulation was ancestral to the divergence of the three subfamilies
of canids.
The presence of a urethral groove may also have implications for borophagine mating sys-
tems. The urethral groove is a major contributor of bacula complexity when its morphology is
quantified from 3D digital scans [8,46]. Brassey et al. [8] found that socially monogamous spe-
cies tend to have bacula that are slightly more complex than solitary species, while species that
live in groups with several breeding females tend to have bacula less complex than the other
two mating systems. If this relationship also applies to the borophagines then their deep ure-
thral groove would suggest that they did not have a mating system with multiple breeding
females, although this inference remains tentative.
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 11 / 15
Distal end
The distal end of the borophagine baculum shows a simple morphology, either mostly cylin-
drical or laterally compressed, which is similar to the simple distal end observed in extant
canids [26]. In contrast, the extinct hesperocyonine H. gregarius had a dorsally-curved distal
tip (Fig 7, [16,32]), which is much more complex than other known canid bacula and more
similar to that of some extant mustelids and procyonids.
Implications for reproductive biology. The distal end of the baculum of extant carnivor-
ans is associated with the mode of ovulation. While species with more complex distal tips tend
to show induced ovulation, those with simple distal tips are associated with spontaneous ovula-
tion [8]. Given that most extant canids have spontaneous ovulation, the bacula described here
suggest a similar strategy for the borophagines. Larivière and Ferguson [47] hypothesized that
induced ovulation is advantageous for species with low population densities and with a multi-
male mating system that live in highly seasonal environments [see also 48]. Although their
hypothesis does not entail that high population densities are associated with spontaneous ovu-
lation, species with higher population density have higher chances of mate encounters during
the restricted time window of spontaneous ovulation. The fossil record of borophagines is par-
ticularly rich [22], consistent with the potential for high population densities in these animals.
Together with the bacular structure of these animals, available evidence, although limited, sug-
gests that the borophagines were spontaneous ovulators.
Curvature
All bacula described here have a primarily ventral curvature (arched) in lateral profile (Fig 7),
with one species (P. euthos) having a secondary dorsal curvature on the distal end. Only two
specimens (both Aelurodon) have complete bacula without any indication of preservational
distortion (such as bending or warping), while for the other three species the bacula are either
incomplete or had the lateral profile slightly altered by a break (C. compressus). Nevertheless, a
primarily ventral curvature is seen in all specimens. In this regard, the bacula of the Borophagi-
nae differ from the mostly straight or even dorsally curved bacula of almost all extant canids
(Fig 7). For the very few extant canids (such as the gray fox) with a ventrally curved baculum,
it is not as curved as in Aelurodon. The degree of ventral curvature in the borophagines is pres-
ent in some bears and pinnipeds, although more data are necessary to establish the generality
of this observation.
Implications for reproductive biology. Using in silico biomechanical analyses, Brassey
et al. [9] showed that a ventral curvature leads to less robust bacula when a dorso-ventral force
is applied to its distal tip. In turn, species with less robust bacula show shorter duration of
copulation [9]. Based on this feature alone, it is possible to hypothesize that the baculum of
borophagines would not withstand as much force during copulation as modern canids and
thus had a shorter duration of copulation than extant canids. However, other morphological
features may have increased the robustness of borophagine baculum, which would suggest a
long copulatory duration. For example, the baculum of Aelurodon ferox (Fig 4) is a massive
bone, with a thick dorsal ridge and a dorso-ventral height that is larger than observed in mod-
ern carnivorans. Further quantitative analysis similar to those of Brassey et al. [9] should help
clarify the robustness of borophagine bacula, and thus inferences about their copulatory
duration.
Conclusion
The baculum of borophagines shows an overall similarity with extant canines, in relative size, a
simple distal end, and presence and configuration of a urethral groove. Drawing upon
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 12 / 15
correlations between bacular morphology and reproductive biology of the extant carnivorans,
it is possible to generate preliminary inferences regarding some aspects of borophagine repro-
ductive biology. Specifically, based on comparative bacular morphology, I suggest the boro-
phagines likely had long copulatory durations and spontaneous ovulation, based on the size
and simple distal tip of their bacula, respectively. It is possible that with the description of
more bacula and further quantitative analysis more robust inferences can be made. The study
of reproductive biology in extinct species is challenging and depends on well established rela-
tionships between morphology and behaviour or physiology. Nevertheless, further inferences
for extinct species could improve our understanding of the selective pressures and consequent
evolutionary dynamics of reproductive strategies changing through time. In the long term, it
might be possible to better distinguish the role of reproductive function from “phylogenetic
inertia” in determining the morphology of the baculum.
Acknowledgments
I would like to thank Charles Marshall for general advising as well as comments to the manu-
script; Eileen Lacey for detailed comments that greatly improved the manuscript; Jack Tseng,
Pat Holroyd, Ashley Poust, and Debora Y. C. Brandt for general advice; Judy Galkin and
Alana Gishlick for providing access to the AMNH collections; Amanda Millhouse from the
Smithsonian National Museum of Natural History for providing photographs of Aelurodon
stirtoni; and, Shane Tucker and George Corner from the University of Nebraska State Museum
for providing photographs for the specimen of Carpocyon compressus. Their generous assis-
tance allowed my research to proceed even during the time of travel restrictions imposed by
the Covid-19 pandemic.
Author Contributions
Conceptualization: Daniel Varajão de Latorre.
Data curation: Daniel Varajão de Latorre.
Funding acquisition: Daniel Varajão de Latorre.
Investigation: Daniel Varajão de Latorre.
Methodology: Daniel Varajão de Latorre.
Visualization: Daniel Varajão de Latorre.
Writing – original draft: Daniel Varajão de Latorre.
Writing – review & editing: Daniel Varajão de Latorre.
References
1. Fischer TC, Ho
¨rnig MK. Mating moths (Tineidae, ditrysia, lepidoptera) preserved as frozen behavior
inclusion in baltic amber (eocene). Palaeontologia Electronica. 2019; 22(1):1–10. https://doi.org/10.
26879/829
2. Hsieh S, Plotnick RE. The representation of animal behaviour in the fossil record. Animal Behaviour.
2020; 169:65–80. https://doi.org/10.1016/j.anbehav.2020.09.010
3. Boucot AJ, Poinar Jr GO. Fossil Behavior Compendium. First edit ed. CRC Press; 2010. Available from:
https://www.taylorfrancis.com/books/9781439810590.
4. Bush AM, Hunt G, Bambach RK. Sex and the shifting biodiversity dynamics of marine animals in deep
time. Proceedings of the National Academy of Sciences of the United States of America. 2016; 113
(49):14073–14078. https://doi.org/10.1073/pnas.1610726113 PMID: 27821755
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 13 / 15
5. Abella J, Valenciano A, Pe
´rez-Ramos A, Montoya P, Morales J. On the Socio-Sexual Behaviour of the
Extinct Ursid Indarctos arctoides: An Approach Based on Its Baculum Size and Morphology. PLoS
ONE. 2013; 8(9):1–5. https://doi.org/10.1371/journal.pone.0073711 PMID: 24058484
6. Schultz NG, Lough-Stevens M, Abreu E, Orr T, Dean MD. The Baculum was Gained and Lost Multiple
Times during Mammalian Evolution. Integrative and Comparative Biology. 2016; 56(4):644–656.
https://doi.org/10.1093/icb/icw034 PMID: 27252214
7. Spani F, Morigi MP, Bettuzzi M, Scalici M, Gentile G, Carosi M. The ultimate database to (re)set the
evolutionary history of primate genital bones. Scientific Reports. 2021; 11(1):1–15. https://doi.org/10.
1038/s41598-021-90787-2 PMID: 34045627
8. Brassey CA, Behnsen J, Gardiner JD. Postcopulatory sexual selection and the evolution of shape com-
plexity in the carnivoran baculum. Proceedings of the Royal Society B: Biological Sciences. 2020; 287
(1936):20201883. https://doi.org/10.1098/rspb.2020.1883 PMID: 33049172
9. Brassey CA, Gardiner JD, Kitchener AC. Testing hypotheses for the function of the carnivoran baculum
using finite-element analysis. Proceedings of the Royal Society B: Biological Sciences. 2018; 285
(1887). https://doi.org/10.1098/rspb.2018.1473 PMID: 30232157
10. Ramm SA. Sexual selection and genital evolution in mammals: A phylogenetic analysis of baculum
length. American Naturalist. 2007; 169(3):360–369. https://doi.org/10.1086/510688 PMID: 17238128
11. Lu¨pold S, De Boer RA, Evans JP, Tomkins JL, Fitzpatrick JL. How sperm competition shapes the evolu-
tion of testes and sperm: A meta-analysis: Sperm competition meta-analysis. Philosophical Transac-
tions of the Royal Society B: Biological Sciences. 2020; 375 (1813). https://doi.org/10.1098/rstb.2020.
0064 PMID: 33070733
12. Hatcher JB. Oligocene Canidae. Memoirs of the Carnegie Museum. 1902; 1:65–108. https://doi.org/10.
5962/p.234819
13. Peterson OA. Description of new carnivores from the Miocene of western Nebraska. Mem of the Carne-
gie Museum. 1910; 4(5):205–278. https://doi.org/10.5962/p.48329
14. Olsen SJ. The baculum of the Miocene carnivore Amphicyon. Journal of Paleontology. 1959; 33
(3):449–450.
15. Harrison JA. The Baculum of Plesiogulo (Carnivora: Mustelidae). Journal of Paleontology. 1982; 56
(5):1266–1273.
16. Wang X. Phylogenetic Systematics of the Hesperocyoninae (Carnivora:Canidae). Bulletin of the Ameri-
can Museum of Natural HIstory. 1994;(221).
17. Dundas RG. Quaternary records of the dire wolf, Canis dirus, in North and South America. Boreas.
1999; 28(3):375–385. https://doi.org/10.1080/030094899422109
18. Hartstone-Rose A, Dundas RG, Boyde B, Long RC, Farrell AB, Shaw CA. The Bacula of Rancho La
Brea. Natural History Museum of Los Angeles County Science Series. 2015; 42:53–63.
19. Wang X, Teaford MF, Taylor BE. Systematics of Borophaginae. Bulletin of the American Museum of
Natural History. 1999; 243:1–391.
20. Silvestro D, Antonelli A, Salamin N, Quental TB. The role of clade competition in the diversification of
North American canids. Proceedings of the National Academy of Sciences. 2015; 112(28):8684–8689.
https://doi.org/10.1073/pnas.1502803112 PMID: 26124128
21. Slater GJ. Iterative adaptive radiations of fossil canids show no evidence for diversity-dependent trait
evolution. Proceedings of the National Academy of Sciences. 2015; 112(16):4897–4902. https://doi.
org/10.1073/pnas.1403666111 PMID: 25901311
22. Van Valkenburgh B, Sacco T, Wang X. Pack hunting in Miocene borophagine dogs: Evidence from cra-
niodental morphology and body size. Bulletin of the American Museum of Natural History. 2003; 2003
(279):147–162. https://doi.org/10.1206/0003-0090(2003)279%3C0147:C%3E2.0.CO;2
23. Andersson K. Were there pack-hunting canids in the Tertiary, and how can we know? Paleobiology.
2005; 31(1):56–72.
24. Didier R. Etude syste
´matique de l’os penien des mammifères. Mammalia. 1948; 12(2):67–93. https://
doi.org/10.1515/mamm.1948.12.2.67
25. Hildebrand M. Morphology of the skeleton and integument in recent canidae. Univ. of California, Berke-
ley; 1951.
26. Burt W. Bacula of North American Mammals. Miscellaneous Publications Museum of Zoology, Univer-
sity of Michigan. 1960;(113). https://doi.org/10.3998/mpub.9690254
27. Kelly DA. Anatomy of the baculum-corpus cavernosum interface in the norway rat (Rattus norvegicus),
and implications for force transfer during copulation. Journal of morphology. 2000; 244(1):69–77.
https://doi.org/10.1002/(SICI)1097-4687(200004)244:1%3C69::AID-JMOR7%3E3.0.CO;2-%23 PMID:
10723082
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 14 / 15
28. Evans HE, de Lahunta A, Miller ME. Miller’s anatomy of the dog. 4th ed. St. Louis (Mo): Elsevier Saun-
ders; 2013.
29. McGrew PO. A new Cynodesmus from the lower Pliocene of Nebraska with notes on the phylogeny of
the dogs. University of California publications. Bulletin of the Department of Geological Sciences; v. 23,
no. 10. Berkeley, CA: University of California Press; 1935.
30. Munthe K. The skeleton of the Borophaginae (Carnivora, Canidae): morphology and function. vol. 133.
Univ of California Press; 1989.
31. Baryshnikov GF, Bininda-Emonds ORPP, Abramov AV. Morphological variability and evolution of the
baculum (os penis) in Mustelidae (Carnivora). Journal of Mammalogy. 2003; 84(2):673–690. https://doi.
org/10.1644/1545-1542(2003)084%3C0673:MVAEOT%3E2.0.CO;2
32. Scott WB, Jepsen GL. The Mammalian Fauna of the White River Oligocene: Part I. Insectivora and Car-
nivora. Transactions of the American Philosophical Society. 1936; 28(1):1. https://doi.org/10.2307/
1005507
33. Slater GJ, Friscia AR. Hierarchy in adaptive radiation: A case study using the Carnivora (Mammalia).
Evolution. 2019; 73(3):524–539. https://doi.org/10.1111/evo.13689 PMID: 30690704
34. Paterson RS, Rybczynski N, Kohno N, Maddin HC. A Total Evidence Phylogenetic Analysis of Pinniped
Phylogeny and the Possibility of Parallel Evolution Within a Monophyletic Framework. Frontiers in Ecol-
ogy and Evolution. 2020; 7(January). https://doi.org/10.3389/fevo.2019.00457
35. Tomiya S, Tseng ZJ. Whence the beardogs? Reappraisal of the middle to late eocene ‘miacis’ from
Texas, USA, and the origin of amphicyonidae (mammalia, carnivora). Royal Society Open Science.
2016; 3(10). https://doi.org/10.1098/rsos.160518 PMID: 27853569
36. Wyss AR, Flynn JJ. A phylogenetic analysis and definition of the Carnivora. In: Szalay F, Novacek M,
McKenna M, editors. Mammal Phylogeny: Placentals. New York: Springer Verlag, New York; 1993. p.
32–52.
37. Dixson AF. Baculum length and copulatory behaviour in carnivores and pinnipeds (Grand Order Ferae).
Journal of Zoology. 1995; 235(1):67–76. https://doi.org/10.1111/j.1469-7998.1995.tb05128.x
38. Larivière S, Ferguson SH. On the evolution of the mammalian baculum: Vaginal friction, prolonged
intromission or induced ovulation? Mammal Review. 2002; 32(4):283–294. https://doi.org/10.1046/j.
1365-2907.2002.00112.x
39. Fitzpatrick JL, Almbro M, Gonzalez-Voyer A, Kolm N, Simmons LW. Male Contest Competition And
The Coevolution Of Weaponry And Testes In Pinnipeds. Evolution. 2012; 66(11):3595–3604. https://
doi.org/10.1111/j.1558-5646.2012.01713.x PMID: 23106721
40. Brindle M, Opie C. Postcopulatory sexual selection influences baculum evolution in primates and carni-
vores. Proceedings of the Royal Society of London B: Biological Sciences. 2016; 283 (1844). https://
doi.org/10.1098/rspb.2016.1736 PMID: 27974519
41. Parker GA, Pizzari T. Sperm competition and ejaculate economics. Biological Reviews. 2010; 85
(4):897–934. PMID: 20560928
42. Alcock J. Postinsemination associations between males and females in insects: the mate-guarding
hypothesis. Annual review of entomology. 1994; 39:1–21. https://doi.org/10.1146/annurev.en.39.
010194.000245
43. Zrzavy
´J, Duda P, Robovsky
´J, Okřinova
´I, Pavelkova
´R
ˇiča
´nkova
´V. Phylogeny of the Caninae (Carniv-
ora): Combining morphology, behaviour, genes and fossils. Zoologica Scripta. 2018; 47(4):373–389.
https://doi.org/10.1111/zsc.12293
44. Lamarca AP, Schrago CG. Fast speciations and slow genes: Uncovering the root of living canids. Bio-
logical Journal of the Linnean Society. 2020; 129(2):492–504. https://doi.org/10.1093/biolinnean/blz181
45. Herdina AN, Kelly DA, Jahelkova
´H, Lina PHC, Hora
´ček I, Metscher BD. Testing hypotheses of bat
baculum function with 3D models derived from microCT. Journal of Anatomy. 2015; 226(3):229–235.
https://doi.org/10.1111/joa.12274 PMID: 25655647
46. Gardiner JD, Behnsen J, Brassey CA. Alpha shapes: Determining 3D shape complexity across morpho-
logically diverse structures. BMC Evolutionary Biology. 2018; 18(1):1–16. https://doi.org/10.1186/
s12862-018-1305-z PMID: 30518326
47. Larivière S, Ferguson SH. Evolution of induced ovulation in North American carnivores. Journal of
Mammalogy. 2003; 84(3):937–947. https://doi.org/10.1644/BME-003
48. Heldstab SA, Mu¨ller DWH, Graber SM, Bingaman Lackey L, Rensch E, Hatt JM, et al. Geographical
Origin, Delayed Implantation, and Induced Ovulation Explain Reproductive Seasonality in the Carniv-
ora. Journal of Biological Rhythms. 2018; 33(4):402–419. https://doi.org/10.1177/0748730418773620
PMID: 29730962
PLOS ONE
Borophaginae bacula
PLOS ONE | https://doi.org/10.1371/journal.pone.0280327 January 17, 2023 15 / 15
Available via license: CC BY
Content may be subject to copyright.
