ArticlePDF Available

Abstract and Figures

To enhance bite force at the canines, feliform carnivorans have short rostra relative to caniform carnivorans. Rostral reduction in feliforms results in less rostrocaudal space for the maxilloturbinals, the complex set of bones involved in conditioning inspired air and conserving water. It is unknown whether the maxilloturbinals might show adaptations to adjust for this loss, such as greater complexity than what is observed in longer snouted caniforms. To understand the impact of rostral shortening on turbinals in feliforms, we used high resolution CT scans to quantify turbinal surface areas (SA) in 16 feliforms and compared them with published data on 20 caniforms. Results indicate that feliforms have reduced maxilloturbinal SA for their body mass relative to caniforms, but comparable fronto-ethmoturbinal SA. However, anterior portions of the ethmoturbinals in feliforms extend forward into the snout and are positioned within the respiratory pathway. When the SA of these anterior ethmoturbinals is added to maxilloturbinal SA to produce an estimated respiratory SA, feliforms and caniforms are similar in respiratory SA. This transfer of ethmoturbinal SA to respiratory function results in feliforms having less estimated olfactory SA relative to caniforms. Previous work on canids found a positive association between olfactory surface area and diet, but this was not found for felids. Results are consistent with feliforms having somewhat reduced olfactory ability relative to caniforms. If confirmed by behavioral data, the relative reduction in olfactory SA in many feliforms may reflect a greater reliance on vision in foraging relative to caniforms. Anat Rec, 297:2065–2079, 2014. © 2014 Wiley Periodicals, Inc.
Content may be subject to copyright.
Respiratory and Olfactory Turbinals in
Feliform and Caniform Carnivorans:
The Influence of Snout Length
BLAIRE VAN VALKENBURGH,
1
* BENISON PANG,
1
DEBORAH BIRD,
1
ABIGAIL CURTIS,
1
KAREN YEE,
2
CHARLES WYSOCKI,
2
AND BRENT A. CRAVEN
3
1
Department of Ecology and Evolutionary Biology, UCLA, Los Angeles, California
2
Monell Chemical Senses Center, Philadelphia, Pennsylvania
3
Applied Research Laboratory, Department of Mechanical and Nuclear Engineering, The
Pennsylvania State University, University Park, Pennsylvania
ABSTRACT
To enhance bite force at the canines, feliform carnivorans have short
rostra relative to caniform carnivorans. Rostral reduction in feliforms
results in less rostrocaudal space for the maxilloturbinals, the complex
set of bones involved in conditioning inspired air and conserving water. It
is unknown whether the maxilloturbinals might show adaptations to
adjust for this loss, such as greater complexity than what is observed in
longer snouted caniforms. To understand the impact of rostral shortening
on turbinals in feliforms, we used high resolution CT scans to quantify
turbinal surface areas (SA) in 16 feliforms and compared them with pub-
lished data on 20 caniforms. Results indicate that feliforms have reduced
maxilloturbinal SA for their body mass relative to caniforms, but compa-
rable fronto-ethmoturbinal SA. However, anterior portions of the ethmo-
turbinals in feliforms extend forward into the snout and are positioned
within the respiratory pathway. When the SA of these anterior ethmotur-
binals is added to maxilloturbinal SA to produce an estimated respiratory
SA, feliforms and caniforms are similar in respiratory SA. This transfer
of ethmoturbinal SA to respiratory function results in feliforms having
less estimated olfactory SA relative to caniforms. Previous work on canids
found a positive association between olfactory surface area and diet, but
this was not found for felids. Results are consistent with feliforms having
somewhat reduced olfactory ability relative to caniforms. If confirmed by
behavioral data, the relative reduction in olfactory SA in many feliforms
may reflect a greater reliance on vision in foraging relative to caniforms.
Anat Rec, 297:2065–2079, 2014. V
C2014 Wiley Periodicals, Inc.
Key words: turbinal; turbinate; olfaction; respiration; carni-
vora; Feliformia; allometry
Grant sponsor: NSF; Grant numbers: IOB-0517748, IOS-
1119768; Grant sponsor: NIH-NIDCD Core Grant; Grant num-
ber: 1P30DC011735-01.
*Correspondence to: Blaire Van Valkenburgh, Department of
Ecology and Evolutionary Biology, University of California, Los
Angeles, CA 90095-7239. Fax: 310-206-3987.
E-mail: bvanval@ucla.edu
Received 24 June 2014; Accepted 25 June 2014.
DOI 10.1002/ar.23026
Published online in Wiley Online Library (wileyonlinelibrary.
com).
THE ANATOMICAL RECORD 297:2065–2079 (2014)
V
V
C2014 WILEY PERIODICALS, INC.
The mammalian skull is a complex structure that is
constrained by development, ancestry, and the multiple
functions it has to perform. In many instances, adapta-
tions for improvements in one key function, such as
feeding, may be opposed by selection for other critical
functions, such as olfaction or vision. For example,
among carnivorans, selection for enhanced bite force
has often produced rostral shortening to improve the
mechanical advantage of jaw closing muscles (Radinsky,
1981a, b; Biknevicius and Van Valkenburgh, 1996). This
is especially apparent among the feliform carnivorans,
the group that includes felids, hyaenids, herpestids,
viverrids, and euplerids (Figs. 1, 2). Relative to cani-
forms (e.g., canids, ursids), many feliforms have rela-
tively short snouts and reduced tooth rows (Radinsky,
1981a). As a consequence of this facial shortening, feli-
forms may also have less space for housing the respira-
tory turbinals that typically fill the rostral nasal
chamber. If so, the respiratory turbinals might show
adaptations to increase their size and surface area,
such as greater complexity or denser packing than
what is observed in longer snouted caniforms. More-
over, unlike caniforms in which olfactory epithelial sur-
face area is largely spatially separated from that of
respiratory epithelial surface area, these two functional
regions may overlap to a greater degree in short-
snouted feliforms, as is the case in some primates
(Smith et al., 2011).
Fig. 1. Time-calibrated phylogeny of the order Carnivora at the family level based on multiple nuclear
gene sequences, with millions of years on the horizontal axis and Cenozoic epochs on the near vertical
axis. Caniformia shown in red, Feliformia shown in blue. All but the following families are represented in
this study: Ailuridae, Odobenidae, Otariidae, Phocidae, and Eupleridae. Figure from Van Valkenburgh and
Wayne, 2010.
2066 VAN VALKENBURGH ET AL.
Fig. 2. Lateral views of the skull of (top) gray wolf (Canis lupus) and (bottom) domestic cat (Felis catus)
showing how skull length, snout length, and occiput-orbital lengths were measured. Skull length is 236
mm for the gray wolf and 83 mm for the domestic cat.
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2067
To understand the impact of rostral shortening on
respiratory and olfactory turbinals in feliforms, we used
high resolution CT scans to quantify turbinal surface
areas relative to body mass in 16 species spanning six
feliform families and compared them with our previously
published data on caniforms (Green et al., 2012). We
used the same methods and software to quantify turbi-
nal surface area as in our caniform study (Green et al.,
2012). Given the relative reduction in snout length,
defined as that portion of the skull anterior to the
orbital margin (Fig. 2), we expected to find that feliforms
had reduced maxilloturbinal surface areas relative to
caniforms. Our expectations concerning the turbinals
that typically bear olfactory epithelium, the frontal and
ethmoturbinals (hereafter referred to as fronto-ethmo-
turbinals), were less certain. Because these are posi-
tioned posterior to the snout, they might be less affected
by rostral shortening and thus caniforms and feliforms
might have similar sized fronto-ethmoturbinals. How-
ever, it is generally thought that some feliforms (e.g.,
cats) rely less on olfaction for foraging than caniforms
and there is anatomical evidence to support this (Kitch-
ener et al., 2010). For example, felids have reduced olfac-
tory bulbs relative to canids (Radinsky, 1975) and
domestic cats are known to have much smaller ethmo-
turbinals than domestic dogs (Dodd and Squirrel, 1980).
Given this, all felids might have smaller fronto-
ethmoturbinals and thus less olfactory epithelium. Other
feliforms, such as the larger hyaenids, appear to rely
more heavily on olfaction in both foraging and social
communication (Gorman, 1980; Mills et al., 1980; Mills
and Gorman, 1987), and thus might be expected to be
more similar to caniforms than felids in the extent of
olfactory epithelium. Behavioral data concerning forag-
ing are more limited for the remaining feliforms in our
sample (mongooses, linsang, palm civet) making it diffi-
cult to predict the relative size of the fronto-
ethmoturbinals.
A key assumption of our work is that olfactory turbi-
nal surface area is positively correlated with some or all
aspects of olfactory ability, such as sensitivity and the
ability to detect a wide array of odorants. This is sup-
ported by the fact that semi-aquatic and aquatic mam-
mals that do not use olfaction for foraging under water
have reduced olfactory turbinal surface areas relative to
terrestrial mammals (Pihlstr
om, 2008; Van Valkenburgh
et al., 2011). In the case of comparing aquatic and ter-
restrial foragers, the differences in apparent need for
olfactory performance appear to be large, and thus are
well-reflected by their olfactory skeletons. However, the
differences in olfactory performance requirements among
terrestrial mammals are likely to be less extreme, and
consequently it may be more difficult to detect them
using turbinal surface area measurements alone. Never-
theless, our previous work on the olfactory turbinals of
caniform carnivorans showed that both diet and body
size were indicative of olfactory turbinal surface area
and presumed olfactory sensitivity, in terms of the abil-
ity to detect odorants at very low thresholds. Carnivo-
rous species tend to have enlarged olfactory surface
areas relative to more omnivorous species, and this was
most apparent among larger, highly carnivorous species
that have large home ranges, such as wolverines, gray
wolves, and African wild dogs (Green et al., 2012). Nota-
bly, Gittleman (1991) also reported larger olfactory bulbs
among carnivorans with large home ranges. It may be
that enhanced olfactory ability is favored in these spe-
cies because they specialize on relatively large prey that
are widely dispersed. The ability to detect such prey
over large distances allows them to minimize energetic
costs associated with hunting. These same patterns may
emerge in our sample of feliforms, which includes many
highly carnivorous species (felids) as well as insectivo-
rous and omnivorous species.
MATERIALS AND METHODS
The feliform sample consists of 16 species from 6 fami-
lies, Felidae (7), Hyaenidae, (4), Herpestidae (2), Viverri-
dae (1), Prionodontidae (1), and Nandinidae (1). Data for
the comparative sample of caniforms are taken from
Green et al. (2012) and include 20 terrestrial caniform
species from five families: Canidae (10), Mephitidae (1),
Mustelidae (4), Procyonidae (2), and Ursidae (3) (Table 1).
To test for possible associations between diet and relative
olfactory turbinal size, species were classified into one of
three broad dietary categories based on the literature as
in Green et al. (2012). The three categories differ in the
relative proportion of vertebrate (i.e., meat) and non-
vertebrate (plants, invertebrates) foods in their diets: (1)
vertebrate—meat specialists whose diets are >70% verte-
brates; (2) vertebrate/non-vertebrate—moderate omni-
vores whose diets are 50–70% vertebrates with the
balance non-vertebrate foods, or (3) non-vertebrate—
omnivores whose diets are <50% vertebrate with non-
vertebrate foods predominating. Thus, for example, all
felids and a few large canids (gray wolves, wild dogs) are
category 1, vertebrate specialists. Coyotes and foxes tend
to be a bit more omnivorous, and so are category 2, and
most ursids and raccoons eat less meat than plant or
invertebrate foods and so are category 3. Although these
categories are broad, more refined categorizations are
problematic due to variation in species diets with season
and over their geographic ranges. Moreover, the broad
categories used here have been shown to correlate well
with various aspects of craniodental morphology and are
useful for tracking dietary evolution within lineages of
carnivores (e.g., Van Valkenburgh 1988, 1991; Evans
et al., 2007). Dietary classification was based on Wilson
et al. (2009) and Van Valkenburgh (1988).
Skulls of all specimens were scanned at the Univer-
sity of Texas High Resolution X-Ray CT Facility (http://
www.ctlab.geo.utexas.edu/). Whenever possible, two
individuals were sampled, one female and one male.
Specimens were selected on the basis of having well-
preserved turbinals and are listed in the Appendix
(Table A1). Scan thickness ranged from 0.02 mm (Galer-
ella sanguinea) to 0.24 mm (Panthera leo) in resolution.
Because turbinal bones are relatively thin, the con-
trast between bone and air is often less pronounced than
between thicker bones and air. To maximize contrast, we
used custom software (available on request from BVV)
that utilized contrast limited adaptive histogram equal-
ization (CLAHE; Jain, 1989) followed by anisotropic dif-
fusion, as described in more detail in Van Valkenburgh
et al. (2011). After processing with CLAHE, scans for
each specimen were imported into the 3D visualization
software Mimics 15.0 (Materialise), in order to quantify
turbinal surface area, as well as measure skull and
snout length. For a few species, scans of the entire skull
2068 VAN VALKENBURGH ET AL.
TABLE 1. Species sampled, number of specimens dietary classification, mean skull length, snout length, body mass, and turbinal surface areas
Code Superfamily Species N Diet
Skull
length
(mm)
Snout
length
(mm)
Body
mass
(kg)
a
TTSA
(mm
2
)
MTSA
(mm
2
)
ETSA
(mm
2
)
Estimated
RTSA
(mm
2
)
b
Estimated
OTSA
(mm
2
)
b
MME Caniformia Mephitis mephitis 2 NV 75.1 21.8 2.1 14887.29 4623 10264.3
GGU Caniformia Gulo gulo 2 V 162.9 60.7 14.5 76006.79 13910.46 62096.33
MFR Caniformia Mustela frenata 2 V 45 9.7 0.2 3036.17 679.51 2356.66
NVI Caniformia Neovison vison 1 VNV 73.5 13.6 1 7057.02 2343.06 4713.96
TTA Caniformia Taxidea taxus 2 VNV 125.4 43.2 7.1 31746.37 8645.32 23101.05
PFL Caniformia Potos flavus 2 NV 88.1 22.4 3 15841.47 3721.76 12119.71
PLO Caniformia Procyon lotor 2 NV 121.79 45 5.5 31727.13 10848.89 20878.24
UAM Caniformia Ursus americanus 2 NV 259.7 111.6 100 129128.2 57494.7 71633.5
UAR Caniformia Ursus arctos 2 NV 350.4 123.8 180 297294.31 143009.2 154285.11
UMA Caniformia Ursus maritimus 2 V 366.2 106.8 352.5 438821.5 175243.44 263578.06
CLA Caniformia Canis latrans 2 VNV 178 75.5 13.4 59267.81 16021.89 43245.92
CLUa Caniformia Canis lupus arctos 2 V 225.4 96.3 42.8 148159.64 31910.98 116248.66
LPI Caniformia Lycaon pictus 2 V 192.1 78.2 22.1 89481.02 19089.34 70391.68
NPR Caniformia Nyctereutes procyonoides 2 NV 125.7 53.3 4 24791.45 7641.59 17149.86
OME Caniformia Otocyon megalotis 2 NV 124.3 52.4 4.2 17597.85 5564.68 12033.17
SVE Caniformia Speothos venaticus 1 V 130 35.5 6 25307.9 5639.8 19668.1
UCI Caniformia Urocyon cinereoargenteus 2 NV 112.3 38.8 3.8 15723.17 3790.78 11932.39
VLA Caniformia Vulpes lagopus 2 V 129.4 44.7 6.3 38909.82 8486.99 30422.83
VMA Caniformia Vulpes macrotis 2 VNV 108.9 32.5 2.6 17675.07 4233.05 13442.02
VVU Caniformia Vulpes vulpes 2 VNV 138 55.5 7 41598.54 7046.23 34552.31
PPA Feliformia Panthera pardus 1 V 215.6 55.9 55 138774.84 25056.84 113718 62801.12 73781.4
LPA Feliformia Leopardus pardalis 2 V 124.6 24.8 12 42081.14 5136.58 36944.56 19077.2 21192.06
LRU Feliformia Lynx rufus 2 V 109.4 22.5 9 29114.45 3113.6 26000.85 14263.99 14166.02
AJU Feliformia Acinonyx jubatus 2 V 166.3 50 50 88078.43 18976.99 69101.44 56745.89 32985.37
PCO Feliformia Puma concolor 2 V 178.8 52.9 51.6 114141.85 16719.46 97422.39 40049.11 73419.06
FSY Feliformia Felis sylvestris 2 V 79.3 14.5 4.7 16619.53 1649.99 14969.54 7961.1 8730.8
PLE Feliformia Panthera leo 2 V 278.8 86.9 161.5 211718.83 47694.15 164024.68 109995.61 96963.65
CCR Feliformia Crocuta crocuta 2 V 232.5 78.9 60 132951.96 7581.17 125370.79 71903.32 61048.64
HBR Feliformia Parahyaena brunnea 2 V 232.9 84.2 41 161542.57 7228.93 154313.64 65645.97 94543.28
HHY Feliformia Hyaena hyaena 1 VNV 209.4 76.5 35 130860.6 7038.6 123822 59505.1 64866.06
PCR Feliformia Proteles cristata 1 NV 137.3 49.2 10 27624.6 1629.94 25994.66 14473.96 12361.48
ABI Feliformia Arctictis binturong 1 NV 153 46.4 13 39206.12 8298.02 30908.1 11742.9 26901.66
NBI Feliformia Nandinia binotata 2 NV 87.8 24.1 2 10967.37 700.73 10266.64 4181.37 6786
PLI Feliformia Prionodon linsang 1 VNV 73.9 20.7 0.7 6499.36 478.5 6020.86 1213.16 5286.2
GSA Feliformia Galerella sanguinea 2 VNV 61.1 13.7 0.6 3006.33 164.73 2841.6 857.62 2033.02
SSU Feliformia Suricatta suricatta 2 VNV 57.9 16.9 0.7 2366.83 130.62 2236.21 854.02 1372.59
Abbreviations: Species codes refer to abbreviations used in figures. N, number of specimens. Diet classification: V, Vertebrate; VNV, vertebrate/non-vertebrate;
NV, non-vertebrate. TTSA, total turbinal surface area; MTSA, maxilloturbinal surface area, ETSA, front-ethmoturbinal surface area.
a
Estimated body masses are from Smith et al. (2003), except for four species that appeared to be incorrect in that data set. Data for Vulpes vulpes,Crocuta cro-
cuta,Parahyaena brunnea, and Hyaena hyaena are from Wilson et al. (2009).
b
Estimated RTSA and OTSA values for the caniforms are equal to the their MTSA and ETSA measurements respectively as these two turbinal complexes are
well separated in caniforms (see Methods).
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2069
were not available and skull length was measured from
digital images of the skull in lateral view using ImageJ
software (NIH). In both cases, skull length was meas-
ured as the distance between the anterior alveolar mar-
gin of the central incisors and the posterior edge of the
occipital condyles. Snout length was measured in lateral
view as the difference between skull length and the dis-
tance between the anterior edge of the orbit and the pos-
terior edge of the occipital condyles (occipital-orbit
length, OOL; Fig. 2). Relative snout length in all the
feliform and caniform carnivorans was ascertained using
linear regression of snout length against skull length.
In our previous work on the nasal turbinals of cani-
form carnivorans (Van Valkenburgh et al., 2011; Green
et al., 2012) we assumed that the combined surface area
of the fronto-ethmoturbinals and nasoturbinals could
serve as a proxy for olfactory (sensory) epithelial surface
area. Similarly we assumed that the surface area of the
maxilloturbinals could serve as a proxy for respiratory
epithelial surface area. This was justified based on the
spatial separation of these two sets of turbinals as well
as published data on tissue distributions and nasal air-
flow patterns in a variety of mammals. For example,
Craven et al. (2007, 2009, 2010) used computational
fluid dynamics to model nasal airflow in dogs while
breathing normally and sniffing. Their simulations
showed that, due to the morphology of the nasal cham-
ber and the separation of the maxilloturbinals and
fronto-ethmoturbinals, inspired airflow splits into dis-
tinct respiratory and olfactory flow paths upon entering
the nose (Craven et al. 2010). A single airway (the dorsal
meatus) transports odorant-laden air to the posterior
olfactory region, while the anterior maxilloturbinal air-
ways direct the remaining airflow away from the olfac-
tory region, toward the internal choana where it exits
the nasal chamber. Thus, the spatial separation of the
maxilloturbinals and fronto-ethmoturbinals allows them
to perform distinct functions: respiration in the former,
and olfaction in the latter.
In short-snouted species such as haplorhine primates,
the spatial separation of these two sets of turbinals is
less clear, and anterior portions of the ethmoturbinals
may overlap the maxilloturbinals and be covered in non-
sensory respiratory epithelium (Smith et al., 2007).
Although data on air flow patterns in these primate spe-
cies are not yet available, it appears likely that these
anterior ethmoturbinals assist the maxilloturbinals in
conditioning inspired air. Consequently, the surface area
of the maxilloturbinals would not be a good proxy for
respiratory surface area in these species, and the same
appears to be true of short-snouted carnivorans, such as
felids. In most of our sampled feliforms there is an ante-
rior extension of the ethmoturbinals that lies above the
maxilloturbinals (Fig. 3). Viewed in sagittal section,
some of these ethmoturbinals are similar to the maxillo-
turbinals below them in that both are aligned in the
direction of airflow between the external and internal
choanae.
Because the first ethmoturbinal is known to possess
both olfactory and respiratory mucosa in many mam-
mals (e.g., Negus, 1958; Kratzing, 1978; Rowe et al.,
2005; Smith et al., 2007), we undertook histology to
establish mucosal boundaries along the ethmoturbinals
using domestic cat (Felis catus) and bobcat (Lynx rufus)
specimens (see Appendix). After fixing in 4% parafor-
maldehyde in phosphate buffered saline (PBS), the
heads were decalcified in Sorenson’s solution (5% EDTA
in phosphate buffer, pH 56.8) and cryoprotected in 10,
20, and 30% sucrose series. The lower jaws and muscles
were removed and the noses were frozen in M1 embed-
ding matrix (Shandon Lipshaw, Pittsburgh, PA). Each
specimen was sectioned at intervals between 160 and
200 lm and stained with Alcian blue. Preliminary his-
tological analyses of the domestic cat and bobcat indi-
cate that part of the first ethmoturbinal is covered in
respiratory epithelium (Fig. 4). Microscopic examina-
tion specifically established that in these species, the
anterior lamellae of the first ethmoturbinal were nonol-
factory where they overlapped with the maxilloturbinal
(Fig. 4E). This suggests that to accurately measure
Fig. 3. Sagittal CT scans of the nasal passages of (A) long-snouted
arctic fox (Alopex lagopus) and (B,C) short-snouted cheetah (Acino-
nyx jubatus). In the long-snouted arctic fox (A), there is clear spatial
separation between the ethmoturbinals (blue) and maxilloturbinals
(yellow), whereas they overlap in the short-snouted cheetah (B). In C,
the portion of ethmoturbinals in the cheetah that we assumed to be
covered in respiratory rather than sensory epithelium is colored in
orange.
2070 VAN VALKENBURGH ET AL.
Fig. 4. (A) Coronal histological section of a domestic cat nose,
showing outlines of ethmoturbinals (ET) and maxilloturbinals (MT) cov-
ered by respiratory epithelium (green) with sensory epithelium (red) on
the nasoturbinals (NT) above the septum (S). At this section of the
nasal passage, the maxilloturbinals are still present, showing an over-
lap of ethmoturbinals covered with respiratory epithelium. Maxilloturbi-
nals are also covered with respiratory epithelium. Histological slides
were stained with Alcian Blue to label goblet cells and Bowman’s
glands. Non-sensory respiratory epithelium and sensory olfactory epi-
thelium were identified by well-defined histological characteristics (i.e.,
epithelial thickness, presence or absence of Alcian blue stained goblet
cells, and absence or presence of Bowman’s glands). (B) Sagittal CT
scan of a cat with red line indicating the approximate location of the
histological section. (CE) Higher magnification images of the epithe-
lium taken at locations indicated by arrows in A. Sensory epithelium
(C,D) is composed of several cell layers and Bowman’s glands
(bg).The transition between sensory to non-sensory epithelium (D)
shows a decrease in epithelial thickness and lack of Bowman’s
glands. Respiratory epithelium (D,E) is thinner and includes goblet
cells (gc) interspersed among ciliated cells. Scale bar 525 lm.
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2071
respiratory surface area, we would need to include the
overlapping anterior ethmoturbinals that are also cov-
ered by respiratory epithelium.
Given that we did not have data on epithelial tissue
distributions for all sampled species, we estimated respi-
ratory turbinal surface area (estimated RTSA) in the
feliforms as the combined surface area of the maxillotur-
binals and all coextensive parts of the ethmoturbinals
that were positioned in the same airflow, as determined
by visual inspection. To do so, we inspected the 3D
reconstructions of the entire turbinal complex (maxillo-
turbinal 1ethmoturbinals) in a sagittal view, and sepa-
rated them along a plane that roughly separates the
ethmoturbinals of the olfactory recess from the ethmo-
turbinals that overlap the maxilloturbinals. In all spe-
cies, we used the transverse lamina to define the extent
of the olfactory recess and the ethmoturbinals contained
within (Fig. 3B,C), given the significance of the trans-
verse lamina in separating olfactory and respiratory
regions (Craven et al., 2010; Lawson et al., 2012).
To maximize consistency, this process of qualitatively
assigning ethmoturbinals to respiratory function was
done by the same person (BP). We then estimated olfac-
tory turbinal surface area (estimated OTSA) as the
combined surface area of the nasoturbinals and remain-
ing fronto-ethmoturbinals. We also present data on the
surface areas of the maxilloturbinals (MTSA) and com-
bined nasoturbinals and fronto-ethmoturbinals (ETSA)
for comparison with previously published data for
caniforms.
The relationships between the various surface area
measurements and body mass were explored using spe-
cies means with ordinary least squares (OLS) linear
regression in the software package SPSS. Following
Smith (2009), we used OLS rather than reduced major
axis (RMA) regression because we were interested in
y-values (e.g., turbinal surface area) given a particular
X-value (e.g., body mass). Furthermore, because the
r-squared values for all our OLS regressions were high
(0.91–0.97), the slopes calculated for OLS and RMA
regressions were similar. To compare regression slopes
between different groups (e.g., canids vs. felids), we
used an analysis of covariance (ANCOVA) test to deter-
mine if slopes were significantly different. To test for
differences in mean residual values among groups (e.g.,
dietary categories), we used non-parametric Mann–
Whitney tests.
Phylogenetic Comparative Analysis
We performed regressions of the same variables using
phylogenetic generalized least squares (PGLS) to
account for covariance in traits owing to shared ances-
try. We also performed a phylogenetic analysis of var-
iance (ANOVA) to assess differences between dietary
groups. For both, we used a phylogeny of the Carnivora
based on that published in Slater et al. (2012), pruning
the tree to remove all species that were not included in
our study (see Appendix). The PGLS regressions were
run in R 3.0.1 (R Development Core Team, 2013) using
the caper (Orne et al., 2012) and ape packages (Paradis
et al., 2004). As the PGLS regression results do not dif-
fer significantly from OLS regressions, we report OLS
results in this article. The phylogenetic ANOVA was also
run in R 3.0.1, using the geiger package (Harmon et al.,
2008). PGLS regression statistics and the phylogenetic
tree used in this phylogenetic analysis are included in
the Appendix (Fig. A1, Tables A2, A3).
RESULTS
Relative Snout Length
In both caniforms and feliforms, snout length is posi-
tively allometric (slope 51.26) and consequently larger
species have relatively longer rostra than smaller spe-
cies (Fig. 5A). The slopes of the feliform and caniform
lines are not significantly different, but, as expected,
feliforms have significantly shorter snouts than cani-
forms of similar skull length, based on a Mann–Whitney
test of residual values (P<0.01). This difference is most
pronounced between the cats (felids) and the dogs (can-
ids), with members of the other families showing more
overlap. Among the caniforms, the mink (Neovison
vison) is remarkable for having a relatively short snout.
Relative Maxilloturbinal Size
As expected, feliforms have reduced MTSA relative to
their body mass in comparison with caniforms (Fig. 5B).
The difference in MTSA is less at large body size
because of stronger positive allometry in the feliforms
(Table 2). Whereas the slope of the line for the Canifor-
mia (slope 50.73, 95% CI 50.66–0.81) is not signifi-
cantly different from that expected for geometric
similarity (0.66), it is for the Feliformia (slope 50.95,
95% CI 50.8–1.09). Among the feliforms, the four hyena
and two herpestid species stand out as having particu-
larly reduced MTSA for their body size. Because the her-
pestids are the smallest feliforms, their unusual
morphology substantially influences the slope of the
regression, making it more positive. If they are excluded
from the analysis, the slope declines from 0.95 to 0.83,
but is still significantly different from isometry.
Relative Fronto-ethmoturbinal Size
The relationship between ETSA and body mass is similar
among caniforms and feliforms, especially if the two small
herpestids are excluded (Fig. 5C). As was the case for the
maxilloturbinals, the slope of the feliform line is more posi-
tive than that of the caniforms, but the two slopes do not dif-
fer significantly, and neither deviates significantly from that
expected for isometry (Table 2). Among the caniforms, the
three ursid species deviate from the usual caniform pattern
and appear to have relatively reduced ethmoturbinals, more
similar to that expected for feliforms of their size. Species
that appear to have relatively large ETSA include a mix of
caniforms and feliforms, including the gray wolf (Canis
lupus), red fox (Vulpes vulpes), brown (Parahyaena brun-
nea), and striped hyenas (Hyaena hyaena), and leopard
(Panthera pardus). The meerkat (Suricatta suricatta)has
unusually reduced fronto-ethmoturbinals and contrasts
markedly with another similar-sized feliform, the linsang
(Prionodon linsang).
Estimated Respiratory Turbinal Surface Area
When we estimate the respiratory area to include por-
tions of the anterior ethmoturbinals that are above the
maxilloturbinals and appear to lie within the respiratory
2072 VAN VALKENBURGH ET AL.
airflow path, caniforms and feliforms appear to have
similar estimated RTSA for their body mass (Fig. 6A).
The correlation coefficient is high (r
2
50.96) and species
within each superfamily fall above and below the com-
mon regression line, although there is a tendency for
larger feliforms to have greater estimated RTSA than
caniforms of similar size. The four hyaenid species that
had very reduced maxilloturbinals for their size no lon-
ger appear unusual, now plotting closer to other feli-
forms; however, the two herpestid species, meerkat and
slender mongoose (Galerella sanguinea), remain as out-
liers, with very low estimated RTSA for their size.
Estimated Olfactory Turbinal Surface Area
The transfer of portions of the anterior ethmoturbi-
nals to respiratory function results in a reduced esti-
mated OTSA for most feliforms relative to similar sized
caniforms (Fig. 6B). There are some interesting excep-
tions, including the linsang, brown and striped hyenas,
all of which have expanded estimated OTSA and plot
closer to the caniform line than the feliform line. There
are a few feliforms that stand out as having quite
reduced estimated OTSA, including the meerkat and
cheetah (Acinonyx jubatus), and to a lesser extent, the
aardwolf (Proteles cristata) and lion (Panthera leo).
Diet and Olfactory Turbinal Surface Area
Among caniforms, and especially the family Canidae,
large-bodied highly carnivorous species have expanded
olfactory surface areas for their body size (Green et al.,
2012). This does not seem to be true for feliforms. In the
plot of estimated OTSA against body mass (Fig. 6B), the
slope of the line is positive for feliforms (0.73, 95% CI 5
0.60–0.85), but does not differ significantly from that
expected for isometry (0.66). An ANCOVA showed that
the feliform slope did not differ significantly from that
observed in canids (0.86, 95%CI 50.67–1.05), and a phy-
logenetic ANOVA of residual values showed no signifi-
cant differences between dietary groups.
Looking more closely at how estimated OTSA relative
to body mass varies with dietary category among feli-
forms, it is apparent that some highly carnivorous spe-
cies fall above (brown hyena, striped hyena) and some
fall below (African lion, cheetah) the common feliform
regression line (Fig. 6C). The linsang, with a mass of
less than one kilogram, appears to have an exceptionally
large olfactory surface area for its small size and is
somewhat omnivorous. It is interesting that the sole
insectivorous hyaenid, the aardwolf, that often uses
auditory cues when hunting (Kruuk and Sands, 1972),
has a relatively reduced olfactory surface area compared
with its larger relatives, the striped, brown and spotted
hyenas, all of which are much more carnivorous. If we
restrict the analysis to the family Felidae, all of which
are hypercarnivorous, we still do not see the strong
Fig. 5. Log
10
/Log
10
plots of (A) Snout length against skull length, (B)
Maxilloturbinal surface area (MTSA) against body mass, and (C)
Fronto-ethmoturbinal surface area (ETSA) against body mass. Cani-
form species are red triangles, while feliform species are blue circles.
Regression lines for caniforms and feliforms are in red and blue,
respectively. The common regression line is black. Regression statis-
tics are in Table 2 and species abbreviations are in Table 1.
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2073
positive relationship between olfactory surface area and
body mass (and correlated home range size) that was
observed by Green et al. (2012) for the Canidae.
DISCUSSION
Respiratory Turbinal Surface Area
In general, feliforms have shorter snouts than simi-
larly sized caniforms. As expected, the shortened snouts
are associated with reduced maxilloturbinal surface
areas relative to body mass in feliforms compared with
caniforms. This might indicate a functional difference
between the two groups such as a reduced need for
water conservation and/or the heating of inspired air in
feliforms, but this appears unlikely. The feliform and
caniform species within our sample span a similar range
of habitats and locomotor types and thus their respira-
tory needs should not differ greatly. For example, both
the spotted hyena (Crocuta crocuta, mean body mass 5
60 kg) and African wild dog (Lycaon pictus, mean body
mass 522 kg) are cursorial predators that coexist and
yet the former has a much smaller maxilloturbinal sur-
face area than the latter. Instead, it seems more likely
that respiratory turbinal surface areas in feliforms are
underestimated when based solely on maxilloturbinals.
After adjusting our estimate of respiratory turbinal
surface area for feliforms through the addition of ante-
rior ethmoturbinals that are positioned within the
anticipated respiratory airflow path, feliforms and cani-
forms no longer differ greatly in this parameter. All four
of the hyena species are no longer outliers and instead
are positioned slightly above the common regression line
with greater respiratory surface areas than caniforms of
similar size (Fig. 6A). This may indicate that we have
now slightly overestimated respiratory surface area for
feliforms. The only way to resolve this and get a more
accurate estimate of functional surface areas is by analy-
sis of the distribution of respiratory versus sensory epi-
thelium in all these species, a task that will require
many fresh specimens and considerable labor.
Olfactory Turbinal Surface Area
Unlike maxilloturbinal surface area, total fronto-
ethmoturbinal surface area relative to body mass was
similar in caniforms and feliforms. This makes sense
given that the fronto-ethmoturbinals are located poste-
rior to the snout and thus the area available to them is
less affected by snout length. However, when we sub-
tract the anterior ethmoturbinals that are anticipated to
lie within the respiratory rather than the olfactory air-
flow path and add them to the maxilloturbinals to pro-
duce an estimate of respiratory surface area, the
situation changes. In general, feliforms have reduced
estimated olfactory turbinal surface areas (as estimated
by the surface area of the remaining fronto-ethmoturbi-
nals) relative to caniforms of similar size. This result is
consistent with previous estimates of olfactory organ
size in canids and felids based on olfactory epithelial
surface area (Dodd and Squirrell, 1980; Laruschkus
1942 in Pihlstr
om et al., 2005) and cribriform plate sur-
face area (Pihlstr
om et al., 2005; Bird et al., this issue).
There are four feliforms that seem to have especially
reduced estimated olfactory surface areas, the meerkat,
aardwolf, cheetah, and lion, but there are no obvious
functional similarities among these four species that
could explain a possible reduced reliance on olfaction.
Similarly, there is no obvious feature uniting three feli-
forms with relatively expanded olfactory surface areas,
the brown hyena (P. brunnea), striped hyena (H.
hyaena), and the much smaller linsang (P. linsang).
In general, cats are considered to rely more on vision
and less on olfaction than canids when foraging for prey
(Kitchener et al., 2010). They have relatively larger eyes
(Nummela et al., 2013) and smaller olfactory organs, as
noted above. This is probably not the case for all the
other feliforms in our sample however. The three large
hyena species appear to be more similar to large canids
in foraging behavior (Mills, 1990; Richardson, 2001) and,
with the exception of the spotted hyena, they are more
similar to canids in relative olfactory surface area than
they are to other feliforms. The remaining feliforms
include various small to medium size species, all of
which are omnivores except for the insectivorous aard-
wolf, and it is not clear why they might rely less on
olfaction than similar sized caniforms. It is certainly pos-
sible that our estimates of olfactory surface area based
on the turbinal bones may not be very precise proxies
for olfactory ability or the number of olfactory receptors.
Species with reduced turbinal surface areas might com-
pensate by having higher densities of olfactory receptor
neurons within the sensory epithelium. Again, resolving
this will require histological analysis but at least our
study highlights species that would be especially inter-
esting to target for such a study, such as the cheetah,
meerkat, and aardwolf.
TABLE 2. Regression statistics for measures of turbinal surface area against body size
Regression Group n Slope y-intercept r
2
95% CI P
a
Maxilloturbinal SA All 36 0.79 3.02 0.76 0.64–0.94 N.S.
Feliformia 16 0.95 2.54 0.93 0.8–1.09 <0.01
Caniformia 20 0.73 3.33 0.96 0.66–0.81 N.S.
Ethmoturbinal SA All 36 0.69 3.79 0.94 0.63–0.75 N.S.
Feliformia 16 0.77 3.7 0.95 0.67–0.86 N.S.
Caniformia 20 0.62 3.87 0.92 0.53–0.71 N.S.
Estimated RTSA All 36 0.81 3.29 0.96 0.76–0.87 <0.01
Feliformia 16 0.9 3.22 0.97 0.82–0.98 <0.01
Caniformia 20 0.73 3.33 0.96 0.66–0.81 N.S.
Estimated OTSA All 36 0.66 3.72 0.88 0.58–0.74 N.S.
Feliformia 16 0.73 3.53 0.92 0.61–0.84 N.S.
Caniformia 20 0.62 3.87 0.92 0.53–0.71 N.S.
a
Probability that the slope differs significantly from isometry. N.S., not significant.
2074 VAN VALKENBURGH ET AL.
In our previous article on caniform carnivorans
(Green et al., 2012), large-bodied hypercarnivores, such
as the wolverine (Gulo gulo), wolf (Canis lupus), and
African wild dog (Lycaon pictus), had larger ethmoturbi-
nals for their body mass than more omnivorous species.
The potentially enhanced olfactory ability of these spe-
cies was argued to be consistent with a need to detect
widely dispersed prey over an expansive home range.
Among carnivorans, home range size increases with both
body mass and a shift to a highly carnivorous diet (Git-
tleman and Harvey, 1982). Given this, the same might
be expected of the larger carnivorous feliforms, such as
the hyenas and big cats. However, our results did not
confirm this. There was no consistent association
between diet and relative estimated olfactory turbinal
surface area within feliforms. Moreover, larger species
with larger home ranges did not have significantly more
estimated olfactory surface area than smaller species.
Within the family Hyaenidae, there was some suggestion
of diet playing a role in that the smallest species with
an insectivorous diet, the aardwolf, had reduced esti-
mated olfactory surface area relative to its larger, more
carnivorous sister species.
The lack of association between diet, home range size
and olfactory surface area in the cats might reflect the
fact that felids rely more heavily on vision than olfaction
when foraging. Felids ambush their prey and often hunt
in particular parts of their home range that maximize
the likelihood of hunting success, such as lions frequent-
ing watering holes or areas of thicker vegetation (David-
son et al., 2013; Loarie et al., 2013). Large canids appear
to move more broadly over their home ranges and may
be more likely to use scent to decide on foraging direc-
tion. Hyenas would seem to be more similar to canids
than felids in foraging mode. They are cursorial rather
than ambush hunters, and also travel widely across
their home range. Thus, they should have expanded
olfactory surface areas to detect dispersed prey. While
this seems to be true of the striped and brown hyenas,
both of which appear to rely heavily on scent when for-
aging (Mills et al., 1980; Mills, 1990; Richardson, 2001),
it was not the case for the most carnivorous of the three,
the spotted hyena. However, both spotted hyenas and
lions are known to respond readily to auditory cues
when hunting, as evidenced by their moving towards the
sounds of distressed antelope and/or the sounds of hye-
nas or lions squabbling at a carcass (Mills et al., 2001).
It is clear that predators will use all their senses to seek
out prey, and thus it is difficult to easily classify one as
more visual or more olfactory.
Limitations and Implications for Future
Research
Our quantitative assessment of relative turbinal size
in carnivorans confirms that the size of the maxilloturbi-
nals can be constrained by having a short snout, but
Fig. 6. Log
10
/Log
10
plots of: (A) Estimated respiratory turbinal sur-
face area against mass, (B) Estimated olfactory turbinal surface area
against mass, and (C) Estimated OTSA against mass for feliforms
only. For A and B, species codes and colors as in Fig. 4. In C, species
are coded by diet: vertebrate (solid squares), vertebrate/non-verte-
brate (diamonds), and non-vertebrate (inverted triangles). Regression
statistics are in Table 2 and species abbreviations are in Table 1.
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2075
this effect is much more pronounced in feliforms than
caniforms. For example, the smallest species in our cani-
form sample, the mink and long-tailed weasel, have very
short rostra but much greater maxilloturbinal surface
area than similarly sized feliforms. These short-snouted
caniforms maximize the surface area of the maxilloturbi-
nals through their architecture. As is true of all cani-
forms, their maxilloturbinals exhibit a dendritic,
branching pattern, whereas the maxilloturbinals of most
feliforms tend to be relatively simple scrolls (Fig. 7).
These architectural differences suggest a strong phyloge-
netic influence on the anatomy of the turbinals, with all
feliforms exhibiting substantial spatial overlap between
maxilloturbinals and fronto-ethmoturbinals whereas
caniforms tend to have minimal or no overlap. The over-
lap is apparent in even the most basal extant feliform in
our sample, the palm civet Nandinia binotata, suggest-
ing that it may have evolved very early, near the split
between Caniformia and Feliformia more than 40 mil-
lion years ago (Wesley-Hunt and Flynn, 2005). To main-
tain a similar scaling relationship between the area of
the turbinals devoted to respiratory function and body
mass, the anterior most portion of the fronto-
ethmoturbinals are at least partially positioned within
the path of respiratory air flow, as is also true in some
primates. The fact that these fronto-ethmoturbinals are
in the respiratory flow path could and should be con-
firmed by computational fluid dynamics simulations of
airflow through the nose of some feliforms, as has been
done for canids and a few other mammal species (Cra-
ven et al., 2009, 2010; Zhao et al., this volume).
The area of the turbinals devoted to olfaction seems
to be much more variable within the Carnivora, sug-
gesting that there are significant differences in olfac-
tory ability. In some cases, the functional differences
are obvious, such as the distinction between terrestrial
and semi-aquatic carnivorans that forage under water
and consequently have reduced olfactory abilities (e.g.,
Pihlstr
om et al., 2005; Pihlstr
om, 2008; Van Valken-
burgh et al., 2011). In other cases, such as the compar-
ison made here among terrestrial species, the
functional differences are much harder to assess.
Although there are some data suggesting that felids
have a reduced sense of smell relative to canids, there
are very limited data on other terrestrial carnivorans.
Behavioral data will be extremely difficult to collect
for most of these species; therefore, we will have to
rely on anatomical or genetic proxies. Analyses of the
bony surface area devoted to olfactory function do
seem to provide information on olfactory sensitivity. It
appears that an enlarged olfactory skeleton increases
the probability that odorants will be detected at very
low concentrations, perhaps by allowing a wider distri-
bution and greater abundance of various olfactory
receptors. However, our estimates of epithelial surface
area based on bone need to be improved through the
addition of data on the actual distribution of olfactory
epithelium in the nose (as exemplified by Smith et al.,
2011, 2012; DeLeon and Smith, this volume), as well
as studies of the density and distribution of olfactory
receptors within the epithelium in a broad array of
species.Itmaybethatwecanuseolfactorygenerep-
ertoires as a surrogate for some aspects of olfactory
ability such as the breadth of odorants that can be
detected (e.g., Hayden et al., 2010), but that will also
need to be validated by behavioral and anatomical
studies. Clearly, our understanding of the nose, its
anatomy, function, and variation among species, is still
at an early stage.
ACKNOWLEDGEMENTS
The authors thank M. Colbert, R. Ketcham, and J. Mai-
sano of the University of Texas HRCT Digital Morphol-
ogy group for their dedication and skills in producing
the CT scans, C. Packer for the loan of lion skulls, and
the multiple curators and collection managers that
allowed us to borrow skulls for scanning. The manu-
script was greatly improved by the helpful comments of
Fig. 7. Coronal CT scans of the respiratory region of the nasal pas-
sageway of: (A) Long-tailed Weasel (Mustela frenata) and (B) Slender
mongoose (Galerella sanguinea). In the short-snouted caniform (wea-
sel), the turbinals are complex and branching, whereas the maxillotur-
binals of the short-snouted feliform tend to be relatively simple folded
scrolls.
2076 VAN VALKENBURGH ET AL.
M. Muchlinski and T. D. Smith. Some histology and
microscopy was performed at the Monell Histology and
Cellular Localization Core.
APPENDIX
Phylogenetic Reconstruction
We obtained a phylogeny of the Carnivora from Slater
and Friscia (unpublished), which is an extension of
the caniform phylogeny published in Slater et al. (2012).
The tree was pruned in R 3.0.1 (R Development
Core Team, 2013) using the APE (Paradis et al., 2004)
package such that only the species from our study
remained in the final tree (Fig. A1). PGLS regressions
were performed using the CAPER package (Orne et al.,
2012).
Histological Methods: Additional Information
Olfactory tissue from two adult domestic cats, a 10-
year-old male with normal genotype and phenotype and
a 6.5-month-old male with heterozygous mucopolysac-
charidoses VI genotype and normal phenotype. Both tis-
sue specimens were obtained at autopsy from animals
housed and euthanized at the University of Pennsylva-
nia, School of Veterinary Medicine in accordance with
protocols approved by its Institutional Animal Care and
Use Committee and the guidelines of the American Vet-
erinary Medical Association. The bobcat specimen was
acquired from a hunter in northeastern Pennsylvania
(PA) in accordance with the regulations of the PA Game
Commission. Experiments described in this work con-
formed to National Institute of Health guide for the care
and use of laboratory animals (NIH Publications 80-23,
revised 1978).
Sectioning of the heads and subsequent staining is
described in the article. Selected stained sections were
digitally scanned at 1,200 dpi resolution on a HP flatbed
scanner to generate an 800 31000 printed hard copy of
each section. The septum, maxilloturbinal, nasoturbinal,
and fronto-ethmoturbinals were identified based on loca-
tion on each section. Non-sensory olfactory epithelium
(including respiratory, transitional, and stratified epithe-
lia) and sensory olfactory epithelium were identified
under 403and 603objective magnifications by well-
defined histological characteristics (i.e., epithelial thick-
ness, presence or absence of Alcian blue stained goblet
Fig. A1. Phylogeny of the carnivora.
TABLE A2. Regression statistics for phylogenetic gen-
eralized least squares regression of turbinal surface
areas against mass for feliforms
Regression DF Slope Intercept r
2
Maxilloturbinal SA 11 0.87 2.65 0.95
Ethmoturbinal SA 11 0.71 3.8 0.95
Estimated RSA 11 0.88 3.26 0.97
Estimated OSA 11 0.66 3.64 0.93
Results for the caniform carnivorans in this study can be
found in Green et al., (2012).
TABLE A1. Feliform specimens used in this study
Genus Species Sex ID
Crocuta crocuta M USNM181527
Crocuta crocuta F USNM164506
Parahyaena brunnea M FMNH34584
Parahyaena brunnea U MVZ117842
Hyaena hyaena M USNM182034
Proteles cristata F USNM368497
Galerella sanguinea M USNM539732
Galerella sanguinea F MVZ118450
Suricatta suricatta M USNM384037
Suricatta suricatta F MVZ118450
Panthera pardus M LACM11704
Leopardus pardalis M FMNH34339
Leopardus pardalis F LACM26789
Lynx rufus M UCLA10115
Lynx rufus F LACM15254
Acinonyx jubatus M FMNH29635
Acinonyx jubatus F FMNH127834
Puma concolor M LACM87430
Puma concolor F LACM85440
Felis sylvestris M LACM14480
Felis sylvestris F LACM14474
Panthera leo M MMNH17537
Panthera leo F MMNH17533
Neofelis nebulosa M LACM31155
Arctictis binturong F USNM259101
Nandinia binotata M USNM450440
Nandinia binotata F USNM20397
Prionodon linsang M USNM303036
All caniform specimens were the same as used in Green
et al., (2012). M, male; F, female; U, unknown. Museum or
source abbreviations are listed as: LACM, Museum of Natu-
ral History, Los Angeles; MVZ, Museum of Vertebrate Zool-
ogy, Berkeley; UCLA, Donald R. Dickey Collection, UCLA;
USNM, United States National Museum; FMNH, Field
Museum of Natural History, Chicago; MMNH; University of
Minnesota Museum of Natural History.
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2077
cells, and presence or absence of Bowman’s glands in the
underlining lamina propria).
LITERATURE CITED
Biknevicius A, Van Valkenburgh B. 1996. Design for killing: cranio-
dental adaptations of predators. In: Gittleman JL, editor. Carni-
vore behavior, ecology, and evolution. Vol. II. Ithaca, NY: Cornell
University Press. p 393–428.
Bird D, Amirkhanian A, Pang B, Van Valkenburgh B. 2014. Quanti-
fying the cribriform plate: influences of allometry, function, and
phylogeny. Anat Rec 297:2080–2092.
Craven B, Neuberger T, Paterson EG, Webb AG, Josephson AM,
Morrison EE, Settles GS. 2007. Reconstruction and morphometric
analysis of the nasal airway of the dog (Canis familiaris) and
implications regarding olfactory airflow. Anat Rec 290:1325–1340.
Craven BA, Paterson EG, Settles GS, Lawson MJ. 2009. Develop-
ment and verification of a high-fidelity computational fluid
dynamics mode of canine nasal airflow. J Biomech Eng 290:
1325–1340.
Craven BA, Paterson EG, Settles GS, Lawson MJ. 2010. The fluid
dynamics of canine olfaction: unique nasal airflow patterns as an
explanation of macrosmia. J R Soc Interface 7:933–943.
Davidson Z, Maleix M, Van Kesteren F, Loveridge AJ, Hunt EJ,
Murindagamo F, Macdonald DW. 2013. Seasonal diet and prey
preference of the African lion in a waterhole-driven semi-arid
savanna. PLoS One 8:e55182.
DeLeon VB, Smith TD. 2014. Mapping the nasal airways: using his-
tology to enhance CT-based three-dimensional reconstruction in
Nycticebus. Anat Rec 297:2113–2120.
Dodd GH, Squirrell DJ. 1980. Structure and mechanism in the
mammalian olfactory system. Symp Zool Soc Lond 45:35–56.
Evans AR, Wilson GP, Fortelius M, Jernvall JC. 2007. High-level
similarity of dentitions in carnivorans and rodents. Nature 445:
78–81.
Gittleman JL, Harvey PH. 1982. Carnivore home-range size, meta-
bolic needs and ecology. Behav Ecol Sociobiol 10:57–63.
Gittleman J. 1991. Carnivore olfactory bulb size: allometry, phylog-
eny and ecology. J Zool 225:253–272.
Gorman ML. 1980. Sweaty mongooses and other smelly carnivores.
Symp Zool Soc Lond 45:87–106.
Green P, Van Valkenburgh B, Pang B, Bird D, Rowe T, Curtis A.
2012. Respiratory and olfactory turbinal size in canid and arctoid
carnivorans. J Anat 221:609–621.
Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W. 2008. GEIGER:
investigating evolutionary radiations. Bioinformatics 24:129–131.
Hayden S, Bekaert M, Crider TA, Mariani S, Murphy WJ, Teeling
EC. 2010. Ecological adaptation determines functional mamma-
lian olfactory subgenomes. Genome Res 20:1–9.
Jain AK. 1989. Fundamentals of digital image processing. Upper
Saddle River, NJ: Prentice Hall, Inc. 569 p.
Kitchener A, Van Valkenburgh B, Yamaguchi N. 2010. Felid form
and function. In: Macdonald, DW, Loveridge A, editors. Biology
and conservation of wild felids. Oxford: Oxford University Press.
p 83–106.
Kratzing J. 1978. The olfactory apparatus of the bandicoot (Isoodon
macrourus): fine structure and presence of a septal olfactory
organ. J Anat 125:601–613.
Kruuk H, Sands WA. 1972. The aardwolf (Proteles cristatus Sparr-
man) 1783 as predator of termites. E Afr Wildlife J 10:211–227.
Lawson MJ, Craven BA, Paterson EG, Settles GS. 2012. A computa-
tional study of odorant transport and deposition in the canine
nasal cavity: implications for olfaction. Chem Senses 37:553–566.
Loarie SR, Tambling CJ, Asner GC. 2013. Lion hunting behaviour and
vegetation structure in an African savanna. Anim Behav 85:899–906.
Mills MGL. 1990. Kalahari hyaenas: comparative behavioural ecol-
ogy of two species. London, UK: Unwin Hyman Ltd. 320 p.
Mills MGL, Gorman ML. 1987. The scent marking behaviour of the
spotted hyaena, Crocuta crocuta, in the Southern Kalahari. J Zool
212:483–497.
Mills MGL, Gorman ML, Mills MEJ. 1980. The scent marking
behaviour of the brown hyaena, Hyaena brunnea, in the Southern
Kalahari. South Afr J Zool 15:240–248.
Mills MGL, Juritz JM, Zucchini W. 2001. Estimating the size of
spotted hyaena (Crocuta crocuta) populations through playback
recordings allowing for non-response. Anim Cons 4:335–343.
Negus VE. 1958. The comparative anatomy and physiology of the
nose and paranasal sinuses. London: Livingstone.
Nummela S, Pihlstrom H, Puolamaki K, Fortelius M, Hemila S, Reuter
T. 2013. Exploring the mammalian sensory space: co-operations and
trade-offs among senses. J Comp Physiol A 199:1077–1092.
Orne D, Freckleton R, Thomas G, Petzoldt T, Fritz S, Isaac N,
Pearse W. 2012. Caper: comparative analyses of phylogenetics
and evolution in R. R package version 0.5. http://CRAN.R-project.
org/package-caper.
Paradis E, Claude J, Strimmer K. 2004. APE: analyses of phyloge-
netics and evolution in R language. Bioinformatics 20:289–290.
Pihlstr
om H. 2008. Comparative anatomy and physiology of chemi-
cal senses in aquatic mammals. In: Thewissen JGM, Nummela S,
editors. Sensory evolution on the threshold. Berkeley: University
of California Press. p 95–109.
Pihlstr
om H, Fortelius M, Hemila S, Forsman R, Reuter T. 2005.
Scaling of mammalian ethmoid bones can predict olfactory organ
size and performance. Proc R Soc B 272:957–962.
R Development Core Team. 2013. R: a language and environment
for statistical computing. R foundation for statistical computing,
Vienna, Austria. ISBN 3-90051-07-0. Available at: URL http://
www.R-project.org
Radinsky L. 1975. Evolution of the felid brain. Brain Behav Evol
11:214–254.
TABLE A3. Regression statistics for reduced major axis (RMA) regressions of turbinal surface area against
mass for all species, caniforms, and feliforms
Regression Group n Slope y-intercept r
2
P
a
Maxilloturbinal SA All 36 0.93 2.88 0.74 0
Caniformia 16 0.75 3.32 0.96 0.04
Feliformia 20 0.98 2.50 0.93 0
Ethmoturbinal SA All 36 0.73 3.76 0.93 N.S.
Caniformia 16 0.65 3.84 0.92 N.S.
Feliformia 20 0.79 3.68 0.95 0.02
Estimated RSA All 36 0.85 3.25 0.96 0
Caniformia 16 0.75 3.32 0.96 0.04
Feliformia 20 0.91 3.21 0.97 0
Estimated OSA All 36 0.85 3.25 0.96 0
Caniformia 16 0.65 3.84 0.92 N.S.
Feliformia 20 0.76 3.50 0.92 N.S.
a
Probability that the slope differs significantly from isometry. N.S., not significant.
2078 VAN VALKENBURGH ET AL.
Radinsky L. 1981a. Evolution of skull shape in carnivores. 1. Repre-
sentative modern carnivores. Bio J Linn Soc 15:369–388.
Radinsky L. 1981b. Evolution of skull shape in carnivores. 2. Addi-
tional modern carnivores. Bio J Linn Soc 16:337–355.
Richardson PRK. 2001. Hyena family. In: Macdonald DW, editor.
The new encyclopedia of mammals. Oxfordshire, UK: Andromeda
Oxford Ltd. p 140–145.
Rowe TB, Eiting TP, Macrini TE, Ketcham RA. 2005. Organization
of the olfactory and respiratory skeleton in the nose of the gray
short-tailed opossum Monodelphis domestica. J Mamm Evol 12:
303–336.
Slater GJ, Harmon LJ, Alfaro ME. 2012. Integrating fossils with
molecular phylogenies improves inference of trait evolution. Evo-
lution 66:3931–3944.
Smith FA, Lyons SK, Morgan E, Ernest SKM, Jones KE, Kaufman
DM, Dayan T, Marquet PA, Brown JH, Haskel JP. 2003. Body
mass of late Quaternary mammals. Ecology 84:3403.
Smith RJ. 2009. Use and misuse of the reduced major axis for line-
fitting. Am J Phys Anthropol 140:476–486.
Smith TD, Eiting TP, Bhatnagar KP. 2012. A quantitative study of
olfactory, nonolfactory, and vomeronasal epithelia surface area in
the nasal fossa of the bat Megaderma lyra. J Mamm Evol 19:
27–41.
Smith TD, Eiting TP, Rossie JB. 2011. Distribution of olfactory and
nonolfactory surface area in the nasal fossa of Microcebus muri-
nus: implications for microcomputed tomography and airflow
studies. Anat Rec 294:1217–1225.
Smith TD, Rossie JB, Bhatnagar KP. 2007. Evolution of the nose
and nasal skeleton in primates. Evol Anthropol 16:132–146.
Van Valkenburgh B. 1988. Trophic diversity in past and
present guilds of large predatory mammals. Paleobiology 14:155–
173.
Van Valkenburgh B. 1991. Iterative evolution of hypercarnivory in
canids (Mammalia: Carnivora): evolutionary interactions among
sympatric predators. Paleobiology 17:340–362.
Van Valkenburgh B, Curtis A, Samuels JX, Bird D, Fulkerson, B,
Meachen-Samuels J, Slater G. 2011. Aquatic adaptations in the
nose of carnivorans: evidence from the turbinates. J Anat 218:
298–310.
Van Valkenburgh B, Wayne RK. 2010. Primer: carnivores. Curr Biol
20:915–919.
Wesley-Hunt GD, Flynn JJ. 2005. Phylogeny of the Carnivora: basal
relationships among the Carnivoramorphans, and assessment of
the position of “Miacoidea” relative to Carnivora. J Syst Paleontol
3:3–28.
Wilson DE, Russell A, Mittermeier, editors. 2009. Handbook of the
mammals of the world—volume 1, carnivores. Barcelona: Lynx
Edicions. 728 p.
Zhao K. 2014. Computational fluid dynamics (CFD) for surgical
planning: a pilot study. Anat Rec 297:2187–2195.
TURBINALS IN FELIFORM AND CANIFORM CARNIVORA 2079
... The nasoturbinal is a rostrocaudally elongated structure that extends from near the naris to the dorsal side of the nasal cavity (Moore 1981). The maxilloturbinal is located on the rostroventral side of the nasal cavity and has a complex branched structure in some mammals (Negus 1958;Van Valkenburgh et al. 2004, 2014bMaier and Ruf 2014;Smith et al. 2015). The nasoturbinal and maxilloturbinal are covered with a respiratory epithelium, which serves to moisten and warm the inhaled air (Negus 1958;Smith and Rossie 2008;Van Valkenburgh et al. 2014b;Martinez et al. 2018;Wagner and Ruf 2019). ...
... The maxilloturbinal is located on the rostroventral side of the nasal cavity and has a complex branched structure in some mammals (Negus 1958;Van Valkenburgh et al. 2004, 2014bMaier and Ruf 2014;Smith et al. 2015). The nasoturbinal and maxilloturbinal are covered with a respiratory epithelium, which serves to moisten and warm the inhaled air (Negus 1958;Smith and Rossie 2008;Van Valkenburgh et al. 2014b;Martinez et al. 2018;Wagner and Ruf 2019). ...
... The basic structure of the turbinal and the lamina project into the nasal capsule (Dieulafe 1906;Ruf 2020). Some cartilages of the turbinal and the lamina are formed when the nasal capsule is formed, and some protrude from the inner wall of the nasal capsule after the formation of the nasal capsule (de Beer 1937;Smith and Rossie 2008;Van Valkenburgh et al. 2014b). ...
Article
Full-text available
The phylogenetic relationships of major groups within the Order Eulipotyphla was once highly disputed, but the advent of molecular studies has greatly improved our understanding about the diversification history of talpids, soricids, erinaceids, and solenodontids. Their resolved phylogenetic relationships now allow us to revisit the turbinal and lamina evolution of this group. The inner structure of the nasal cavity of mammals is highly complicated and the homologies of the turbinals among mammalian species are still largely unsettled. In this regard, investigation on fetal anatomy and ontogenetic changes of the nasal capsule allows us to evaluate the homologies of the turbinals and laminae. We observed various fetuses and adults of talpids and soricids using high-resolution diffusible iodine-based contrast-enhanced computed tomography (diceCT) and reviewed previous reports on erinaceids, solenodontids, and other laurasiatherians. Although the turbinal and lamina morphology was previsouly considered to be similar among eulipotyphlans, we found phylogenetic patterns for talpids and soricids. The nasoturbinal of the common ancestor of talpids and soricids was most likely rostrocaudally elongated. The epiturbinal at the ethmoturbinal II disappeared in soricids independently. Finally, we propose two possible scenarios for the maxilloturbinal development: 1) the maxilloturbinal of talpids and soricids became small independently with a limited number of lamellae as a result of convergent evolution, or 2) the common ancestor of talpids and soricids already had a small and simple maxilloturbinal.
... Van Valkenburg and her team widely developed quantitative analyses based on turbinal data, acquired by micro-CT (e.g. Van Valkenburg et al. 2004, 2011, 2014a, Green et al. 2012. One of the biggest challenges is now the required time to process the data and in particular the segmentation (= isolation of an area of interest). ...
... Apparently, turtles are the only amniote to lack turbinals even at a very early embryonic stage (Parsons 1971). In general, turbinals are more developed in endotherms and therefore have a huge morphological disparity related to their phylogenetic relationships, skull morphology, and ecology (Bang 1961, 1965, 1968, Parsons 1971, Hillenius 1994, Van Valkenburg et al. 2004, 2011, 2014a, Green et al. 2012, Ruf 2014, Martinez et al. 2018. Birds generally possess two respiratory and one olfactory turbinals. ...
... In mammals, turbinals generally occupied a large portion of the nasal cavity (Parsons 1971, Hillenius 1994, Van Valkenburg et al. 2004, 2011, 2014a. Mammalian turbinals are usually discriminated in two distinct functional parts: the anterior respiratory turbinals and the posterior olfactory turbinals (Fig. 3). ...
Thesis
In most tetrapods, the nasal cavity houses a bony or a cartilaginous system (i.e. turbinals or turbinates) supporting epithelium and sensory organs involved in either olfaction or heat and moisture conservation. Among extant tetrapods, mammals have on average, the largest turbinals to skull length ratio. Despite some studies in primates, Carnivora, bats, lagomorphs as well asrodents, our understanding of the selective pressures affecting turbinals remains imprecise.This PhD aims to unravel the evolutionary processes responsible for the large anatomical and morphological variations of turbinals among mammals. In the course of our work we acquired an extensive dataset of three-dimensional micro-computed tomography scans (micro-CT) in rodents and other small terrestrial mammals. We were then able to statistically test hypotheses linking turbinal morphology to ecology (e.g. diet or ecotype) and evolutionary patterns such asconvergence or evolutionary trade-off (e.g. conflict for space in the nasal cavity between different organs).The present dissertation provides a non-exhaustive review of the olfaction. In the light of our works, we discussed the methodological and conceptual limits of the field. Indeed, olfaction is a complex function relying on multifactorial processes, under various selective pressures. Olfaction may be tackled by resorting to different approaches (e.g. morphology, histology, genomics) and anatomical proxies (e.g. turbinals, olfactory bulb, vomeronasal organ). In thiscontext, our ongoing projects try to refine current functional hypotheses in studying covariation in olfactory-related organs using different anatomical proxies, immunohistochemistry, and transcriptomic.
... The olfactory epithelium covers turbinal bones (turbinates), delicate, scroll-like arrangements of approximately five bones, whose shapes can change the surface area for potential odorant deposition. Olfactory turbinals are highly convoluted and variable in shape (Ruf 2014;van Valkenburgh et al. 2014;Curtis and Simmons 2016;Lundeen and Kirk 2019), but microcomputed tomography (µCT) scanning and image analysis now makes large-scale comparative analyses of these complex structures tractable . Evidence for selection shaping the size, shape, and relative orientations of turbinates is emerging, including convergent expansion of turbinates in worm-feeding rodents (Martinez et al. 2018) and convergent signatures of trade-offs of olfactory and respiratory turbinates in amphibious rodents (Martinez et al. 2020). ...
... The genetic controls of olfactory turbinate morphogenesis are unrelated to OR genes (although related the olfactory bulb) (Treloar et al. 2010), but the expansion of olfactory epithelium surface area directly increases the neural epithelial space in which olfactory receptor neurons can express OR genes. Although the expression of OR genes is monoallelic and stochastic per sensory neuron (Rodriguez 2013;Monahan and Lomvardas 2015;Abdus-Saboor et al. 2016), there is zonal organization of expression within the turbinates associated with different OR subfamilies. ...
Article
While evolvability of genes and traits may promote specialization during species diversification, how ecology subsequently restricts such variation remains unclear. Chemosensation requires animals to decipher a complex chemical background to locate fitness‐related resources, and thus the underlying genomic architecture and morphology must cope with constant exposure to a changing odorant landscape; detecting adaptation amidst extensive chemosensory diversity is an open challenge. In phyllostomid bats, an ecologically diverse clade that evolved plant‐visiting from an insectivorous ancestor, the evolution of novel food detection mechanisms is suggested to be a key innovation, as plant‐visiting species rely strongly on olfaction, supplementarily using echolocation. If this is true, exceptional variation in underlying olfactory genes and phenotypes may have preceded dietary diversification. We compared olfactory receptor (OR) genes sequenced from olfactory epithelium transcriptomes and olfactory epithelium surface area of bats with differing diets. Surprisingly, although OR evolution rates were quite variable and generally high, they are largely independent of diet. Olfactory epithelial surface area, however, is relatively larger in plant‐visiting bats and there is an inverse relationship between OR evolution rates and surface area. Relatively larger surface areas suggest greater reliance on olfactory detection and stronger constraint on maintaining an already diverse OR repertoire. Instead of the typical case in which specialization and elaboration are coupled with rapid diversification of associated genes, here the relevant genes are already evolving so quickly that increased reliance on smell has led to stabilizing selection, presumably to maintain the ability to consistently discriminate among specific odorants — a potential ecological constraint on sensory evolution. This article is protected by copyright. All rights reserved
... The nasal cavity anatomy of various animals have been reported, including that of monkeys (Schreider & Raabe, 1981;Yeh, Brinker, Harkema, & Muggenburg, 1997), goats (Gupta, Sharma, & Bhardwaj, 1992;Singh, Kumar, & Singh, 1992), buffalo calves (Ganganaik, Ananda, Kavitharani, & Khandare, 2009) and adults (Eman, Mohamed, Hatem, & Ahmed, 2019), pigs (Parkash, Kumar, & Singh, 2016;Yang, Dai, Yu, & Yang, 2017), sheep (Arthur, Pablo, & Ignacio, 2014;Ganganaik, Jain, & Kumar, 2004), Mizo local pigs (Kalita, 2014), yaks (Sharma & Gupta, 1991), camels (Eman et al., 2019), donkey (El-Gendy & Alsafy, 2010;Eman et al., 2019), marmosets (Smith, Eiting, Bonar, & Craven, 2014), fish (Abel et al., 2010;Holmes et al., 2011), cats (Pang et al., 2016;van Valkenburgh et al., 2014;Yee, Craven, Wysocki, & Van Valkenburgh, 2016), dogs (Craven et al., 2007;Schreider & Raabe, 1981;van Valkenburgh et al., 2014;Yee et al., 2016;Wagner and Ruf, 2018), whitetailed deer (Ranslow et al., 2014;Yee et al., 2016), blue bulls (Kumar, Sanjeev, Subha, Kavita, & Nikhil, 2017), humans (Ménache et al., 1997;Yeh et al., 1997), birds (Casteleyn et al., 2018), rabbits (Xi et al., 2016), and pygmy slow loris (Smith, Craven, Engel, Bonar, & DeLeon, 2019). Studies in rodents include shrews (Larochelle & Baron, 1989), squirrels , rats and mice (Gross, Swenberg, Fields, & Popp, 1982;Harkema et al., 2006;Ménache et al., 1997;Patra, Menache, Shaka, & Gooya, 1987;Schreider & Raabe, 1981), and chinchillas (Jurciseka, Joan, Durbina, & Kusewittb, 2003). ...
... The nasal cavity anatomy of various animals have been reported, including that of monkeys (Schreider & Raabe, 1981;Yeh, Brinker, Harkema, & Muggenburg, 1997), goats (Gupta, Sharma, & Bhardwaj, 1992;Singh, Kumar, & Singh, 1992), buffalo calves (Ganganaik, Ananda, Kavitharani, & Khandare, 2009) and adults (Eman, Mohamed, Hatem, & Ahmed, 2019), pigs (Parkash, Kumar, & Singh, 2016;Yang, Dai, Yu, & Yang, 2017), sheep (Arthur, Pablo, & Ignacio, 2014;Ganganaik, Jain, & Kumar, 2004), Mizo local pigs (Kalita, 2014), yaks (Sharma & Gupta, 1991), camels (Eman et al., 2019), donkey (El-Gendy & Alsafy, 2010;Eman et al., 2019), marmosets (Smith, Eiting, Bonar, & Craven, 2014), fish (Abel et al., 2010;Holmes et al., 2011), cats (Pang et al., 2016;van Valkenburgh et al., 2014;Yee, Craven, Wysocki, & Van Valkenburgh, 2016), dogs (Craven et al., 2007;Schreider & Raabe, 1981;van Valkenburgh et al., 2014;Yee et al., 2016;Wagner and Ruf, 2018), whitetailed deer (Ranslow et al., 2014;Yee et al., 2016), blue bulls (Kumar, Sanjeev, Subha, Kavita, & Nikhil, 2017), humans (Ménache et al., 1997;Yeh et al., 1997), birds (Casteleyn et al., 2018), rabbits (Xi et al., 2016), and pygmy slow loris (Smith, Craven, Engel, Bonar, & DeLeon, 2019). Studies in rodents include shrews (Larochelle & Baron, 1989), squirrels , rats and mice (Gross, Swenberg, Fields, & Popp, 1982;Harkema et al., 2006;Ménache et al., 1997;Patra, Menache, Shaka, & Gooya, 1987;Schreider & Raabe, 1981), and chinchillas (Jurciseka, Joan, Durbina, & Kusewittb, 2003). ...
Article
The nose is a structurally and functionally complex organ in the upper respiratory tract. It not only serves as the principal organ for the sense of smell, but also functions to efficiently filter, warm, and humidify inhaled air before the air enters the more delicate distal tracheobronchial airways and alveolar parenchyma of the lungs. Despite the volume of published studies on the biology of rodents, there is no information on the gross upper respiratory morphology of the African giant rat (AGR) in the available literature. Hence, this study aimed to examine the anatomy of the turbinates, their meatuses, and the morphometry of the nasal cavity. The following were found and reported in this study: (1) There were three nasal conchae in AGR: the nasoturbinate, which was the largest; the ethmoturbinate, which was composed of one well‐developed ectoturbinate and three well‐developed endoturbinates; and the maxilloturbinate, which was fusiform, short, and branched. (2) Three major meatuses were observed: the dorsal nasal meatus, which was the longest and widest; the middle nasal meatus, which was without limbs but had a deep oval caudal recess; and the ventral nasal meatus, which directly continued caudally into the nasopharyngeal meatus. (3) Four ethmoturbinates with four slit‐like meatuses were observed, each with dorsal and ventral limbs; the first contacted the middle nasal meatus but not the nasopharyngeal meatus. (4) There were three paranasal sinuses: one sphenoid, two frontal, and two palatine sinuses. The data obtained are relevant to pathologists and eco‐morphologists, considering the burrowing habitat and behaviors of AGR, and provide baseline data for more investigative studies. This article is protected by copyright. All rights reserved.
... Because the turbinates are not part of the encephalon (grouping together cerebellum, cerebrum, and olfactory bulbs), they are often omitted in published drawings. However, some observations on living species indicate that these structures may reflect hunting strategies (Blanton and Biggs 1969;Van Valkenburgh et al. 2014;Negus 1954). ...
... Abbreviations: C., Cartierodon; E., Eurotherium, H., Hyaenodon, M., Matthodon, P., Prodissopsalis. of the black-backed jackal, which has an acute sense of smell (Fox 1971). Therefore, an olfactory membrane with a large air contact surface is common among scavengers, and some are even nocturnal and have very developed turbinates; this is, for instance, the case in hyaenids (Van Valkenburgh et al. 2014). The turbinates seen in E. theriodis thus testify to a strong expansion of the olfactory membrane of the sinuses. ...
Article
Hyaenodonts are extinct placental mammals with a carnivorous diet. Their phylogenetic position among mammals and the relationships within Hyaenodonta are at present partially unresolved. The endocranium is a structure that has rarely been studied in this clade. Using 3D tomography, we studied the endocranium of the European hyaenodont Eurotherium theriodis, discovered in Egerkingen (Switzerland, Lutetian, middle Eocene). Eurotherium theriodis has an endocranium morphology that supports an increase in size of the cerebrum relative to the cerebellum over time within the Hyaenodontoidea. The endocranium also supports a complexification of the cerebrum (i.e., at least two furrows per cerebral hemisphere) within the Hyaenodontoidea and allows us to envisage an increase of the encephalization quotient (EQ), over time. Based on morphology, we consider that its endocranium does not depart from that of the hyaenodontoids known in the Lutetian, Bartonian, and Priabonian of Europe, being less complex than that of the hypercarnivorous Hyaenodon. However, the morphology of its olfactory bulbs and turbinates is similar to that observed in Hyaenodon. The large size of the turbinates of E. theriodis is regarded to be the result of a possible scavenger ecology and agrees with the meat/bone diet envisaged based on the analysis of the morphology of the skull and teeth.
... Thus, the P. antarcticum specimens analyzed here had a calculated body mass of 20-32 g. In mammals, the SSA has been measured in many species [54][55][56][57][58][59]. Although a review of the SSA of mammals is out of the scope of this work, from the partial literature cited, several mammalian species have a total SSA of the same order of magnitude as that of P. antarcticum: 2 species from the order Afrosorcida (range of SSA about 51-82 mm 2 , range of average body mass about 80-100 g), 2 species from Chiroptera (range of SSA about 32-68 mm 2 , range of average body mass about 16-28 g), 12 species from Eulipotyphla (range of SSA about 52-92 mm 2 , range of average body mass about 6-35 g), 1 species from Primates (SSA about 55 mm 2 , average body mass about 246 g), and 29 species from Rodentia (range of SSA about 29-104 mm 2 , average body mass about 7-195 g). ...
Article
Full-text available
The olfactory system is constituted in a consistent way across vertebrates. Nasal structures allow water/air to enter an olfactory cavity, conveying the odorants to a sensory surface. There, the olfactory neurons form, with their axons, a sensory nerve projecting to the telencephalic zone—named the olfactory bulb. This organization comes with many different arrangements, whose meaning is still a matter of debate. A morphological description of the olfactory system of many teleost species is present in the literature; nevertheless, morphological investigations rarely provide a quantitative approach that would help to provide a deeper understanding of the structures where sensory and elaborating events happen. In this study, the peripheral olfactory system of the Antarctic silverfish, which is a keystone species in coastal Antarctica ecosystems, has also been described, employing some quantitative methods. The olfactory chamber of this species is connected to accessory nasal sacs, which probably aid water movements in the chamber; thus, the head of the Antarctic silverfish is specialized to assure that the olfactory organ keeps in contact with a large volume of water—even when the fish is not actively swimming. Each olfactory organ, shaped like an asymmetric rosette, has, in adult fish, a sensory surface area of about 25 mm2, while each olfactory bulb contains about 100,000 neurons. The sensory surface area and the number of neurons in the primary olfactory brain region show that this fish invests energy in the detection and elaboration of olfactory signals and allow comparisons among different species. The mouse, for example—which is considered a macrosmatic vertebrate—has a sensory surface area of the same order of magnitude as that of the Antarctic silverfish, but ten times more neurons in the olfactory bulb. Catsharks, on the other hand, have a sensory surface area that is two orders of magnitude higher than that of the Antarctic silverfish, while the number of neurons has the same order of magnitude. The Antarctic silverfish is therefore likely to rely considerably on olfaction.
... All muroids possess a nasoturbinal, a maxilloturbinal, and three ethmoturbinals (or four, depending on convention; see below); these are named based on their bony articulation in adults, although they all originate from the fetal cartilaginous capsule (Maier 1993a, b). This arrangement is found in many mammals of other orders, including most non anthropoid primates (Kollmann and Papin 1925;Smith and Rossie 2008;Smith et al. 2019; Lundeen and Kirk 2019), scandentians (Wible 2011), lagomorphs (Negus 1958;Ruf 2014) and many carnivorans (Negus 1958;Van Valkenburgh et al. 2014a). Given the frequency of this arrangement concerning the larger turbinals that project close to the nasal septum, this arrangement may be a plesiomorphic for mammals (but see Lundeen and Kirk, 2019). ...
Article
Full-text available
Nasal anatomy in rodents is well-studied, but most current knowledge is based on small-bodied muroid species. Nasal anatomy and histology of hystricognaths, the largest living rodents, remains poorly understood. Here, we describe the nasal cavity of agoutis ( Dasyprocta spp.), the first large-bodied South American rodents to be studied histologically throughout the nasal cavity. Two adult agoutis were studied using microcomputed tomography, and in one of these, half the snout was serially sectioned and stained for microscopic study. Certain features are notable in Dasyprocta . The frontal recess has five turbinals within it, the most in this space compared to other rodents that have been studied. The nasoturbinal is particularly large in dorsoventral and rostrocaudal dimensions and is entirely non-olfactory in function, in apparent contrast to known muroids. Whether this relates solely to body size scaling or perhaps also relates to directing airflow or conditioning inspired air requires further study. In addition, olfactory epithelium appears more restricted to the olfactory and frontal recesses compared to muroids. At the same time, the rostral tips of the olfactory turbinals bear at least some non-olfactory epithelium. The findings of this study support the hypothesis that turbinals are multifunctional structures, indicating investigators should use caution when categorizing turbinals as specialized for one function (e.g., olfaction or respiratory air-conditioning). Caution may be especially appropriate in the case of large-bodied mammals, in which the different scaling characteristics of respiratory and olfactory mucosa result in relative more of the former type as body size increases.
... A combination of diceCT and µCT of the same specimens will allow a fuller understanding of what type(s) of mucosa line each turbinal. This would provide a firmer basis, or cautionary caveats, for the use of individual bones such as turbinals as proxies for a particular function (e.g., Van Valkenburgh et al., 2014;Martinez et al., 2018). This also has important application to future quantitative studies to further our understanding of the link between OE surface area and ecological variables (e.g., Yee et al., 2016), and in the study of fluid dynamics in the nasal airways (Craven, Paterson & Settles, 2010;Ranslow et al., 2014). ...
Article
Full-text available
Diffusible iodine-based contrast-enhanced computed tomography (diceCT) has emerged as a viable tool for discriminating soft tissues in serial CT slices, which can then be used for three-dimensional analysis. This technique has some potential to supplant histology as a tool for identification of body tissues. Here, we studied the head of an adult fruit bat (Cynopterus sphinx) and a late fetal vampire bat (Desmodus rotundus) using diceCT and µCT. Subsequently, we decalcified, serially sectioned and stained the same heads. The two CT volumes were rotated so that the sectional plane of the slice series closely matched that of histological sections, yielding the ideal opportunity to relate CT observations to corresponding histology. Olfactory epithelium is typically thicker, on average, than respiratory epithelium in both bats. Thus, one investigator (SK), blind to the histological sections, examined the diceCT slice series for both bats and annotated changes in thickness of epithelium on the first ethmoturbinal (ET I), the roof of the nasal fossa, and the nasal septum. A second trial was conducted with an added criterion: radioopacity of the lamina propria as an indicator of Bowman's glands. Then, a second investigator (TS) annotated images of matching histological sections based on microscopic observation of epithelial type, and transferred these annotations to matching CT slices. Measurements of slices annotated according to changes in epithelial thickness alone closely track measurements of slices based on histologically-informed annotations; matching histological sections confirm blind annotations were effective based on epithelial thickness alone, except for a patch of unusually thick non-OE, mistaken for OE in one of the specimens. When characteristics of the lamina propria were added in the second trial, the blind annotations excluded the thick non-OE. Moreover, in the fetal bat the use of evidence for Bowman's glands improved detection of olfactory mucosa, perhaps because the epithelium itself was thin enough at its margins to escape detection. We conclude that diceCT can by itself be highly effective in identifying distribution of OE, especially where observations are confirmed by histology from at least one specimen of the species. Our findings also establish that iodine staining, followed by stain removal, does not interfere with subsequent histological staining of the same specimen.
Article
Our knowledge of nasal cavity anatomy has grown considerably with the advent of micro‐computed tomography (CT). More recently, a technique called diffusible iodine‐based contrast‐enhanced CT (diceCT) has rendered it possible to study nasal soft tissues. Using diceCT and histology, we aim to (a) explore the utility of these techniques for inferring the presence of venous sinuses that typify respiratory mucosa and (b) inquire whether distribution of vascular mucosa may relate to specialization for derived functions of the nasal cavity (i.e., nasal‐emission of echolocation sounds) in bats. Matching histology and diceCT data indicate that diceCT can detect venous sinuses as either darkened, “empty” spaces, or radio‐opaque islands when blood cells are present. Thus, we show that diceCT provides reliable information on vascular distribution in the mucosa of the nasal airways. Among the bats studied, a nonecholocating pteropodid (Cynopterus sphinx) and an oral‐emitter of echolocation sounds (Eptesicus fuscus) possess venous sinus networks that drain into the sphenopalatine vein rostral to the nasopharynx. In contrast, nasopharyngeal passageways of nasal‐emitting hipposiderids are notably packed with venous sinuses. The mucosae of the nasopharyngeal passageways are far less vascular in nasal‐emitting phyllostomids, in which vascular mucosae are more widely distributed in the nasal cavity, and in some nectar‐feeding species, a particularly large venous sinus is adjacent to the vomeronasal organ. Therefore, we do not find a common pattern of venous sinus distribution associated with nasal emission of sounds in phyllostomids and hipposiderids. Instead, vascular mucosa is more likely critical for air‐conditioning and sometimes vomeronasal function in all bats.
Article
Full-text available
Events during the Late Oligocene of North America and the Pleistocene of South America, provide an opportunity for exploring possible causes of the evolution of hypercarnivory in canids. Plots of generic diversity gainst time for North American predators reveal a roughly inverse relationship between the number of hypercarnivorous canid taxa and the numbers of other hypercarnivores, such as creodonts, nimravids, mustelids, and amphicyonids. Similarly, the radiation of hypercarnivorous canids in South America occurred at a time of relatively low diversity of other hypercarnivores. Analysis of trophic diversity within the North American carnivore paleoguild before, during, and after the Late Oligocene reveals considerable taxonomic turnover among carnivores because of immigration and speciation. Late Oligocene hypercarnivorous canids appear to have been replaced first by amphicyonids and large mustelids, and then by felids. -from Author
Article
The deposition onto grass stalks of two distinct, strong-smelling substances produced in the anal scent pouch, is the most common form of scent marking in the brown hyaena (Hyaena brunnea). It is called pasting. The behaviour associated with pasting is described, as is the related functional anatomy of the scent pouch. The dispersion pattern of pastings within a group territory and the rate of marking in different parts of the territory were ascertained by direct observations on radio collared hyaenas. The data were analysed by the computer programs SYMAP and SYMVU which graphically display the data as a three dimensional map. Brown hyaenas leave most pastings in those areas in which they spend most time. This is in the central part of the territory. When they visit the boundaries, however, the frequency of pasting increases. GLC analyses of the pastings from two known individuals show distinct differences in the relative concentrations of the many compounds in the pastings of each. Behavioural observations show that the hyaenas are able to recognize different individuals' pastings. Pasting could function to inform group members of each other's movements as well as to inform outsiders that the territory is occupied.Die deponering op grasstlngels van twee afsonderlike, sterkruikende afskeidings wat in die anale ruiksak geproduseer is, is die algemeenste vorm van ruikmerking in die bruin hiëna (Hyaena brunnea). Dit word ‘pasting’ genoem. Die gedrag wat met ‘pasting’ verbonde is word hier beskryf sowel as die funksionele anatomie van die ruiksak. Die verspreidingspatroon van die afskeidings binne 'n groepterritorium en die tempo van merking in verskillende dele van die territorium is vasgestel deur direkte waarnemings op hiënas met radionekbande. Die data is deur die rekenaarprogramme SYMAP en SYMVU, wat grafies die data as 'n driedimensionele kaart vertoon, ontleed. Bruin hiënas deponeer die meeste van hulle afskeidings in die gebiede waar hulle die meeste tyd deurbring. Dit is in die sentrale deel van hulle territorium. Wanneer hulle egter die grense besoek, vermeerder die frekwensie van merking. GKV-ontledings van die afskeidings van twee bekende individue toon duidelike verskille in die relatiewe konsentrasies van die baie verbindings in nie afskeidings van elkeen. Gedragswaarnemings toon dat die hiënas verskillende individue se afskeidings kan herken. ‘Pasting’ mag funksioneer om groepslede van mekaar se bewegings te vergewis, sowel as om buitestanders te laat weet dat die territorium beset is.
Article
Fifteen variables, selected primarily to reflect functionally significant aspects of cranial morphology, were measured on one skull each of 62 species of modern carnivores, including viverrids, canids, mustelids and felids. To allow comparisons between species of different sizes without the potentially confounding effects of allometric shape changes, the measurements were transformed to dimmensionless variables, based on the residuals from allometric equations. Fourteen out of 15 of the transformed variables distinguish one or more of the four family groups and the rotated first two axes of a principal components analysis distinguish all four families from each other. The following functional hypotheses are proposed: mustelids and felids have the most powerful bites and canids the weakest among the four family groups studied; mustelids and, to a lesser degree, felids have more powerful neck musculature than do canids and viverrids; and visual abilities are best developed among felids and least developed among mustelids. The first two functional hypotheses suggest possible differences in killing behaviour, which are supported by a preliminary survey of the literature on such behaviour. Allometric analysis of the 15 cranial measures shows that the neurocranial components scale with negative allometry, while most of the other measures scale approximately isometrically.