1. The Morphological Interaction Betweenthe Nasal Cavity and Maxillary Sinusesin Living HumansNATHAN HOLTON,1,2TODD YOKLEY,3AND LAUREN BUTARIC41Department of Orthodontics, University of Iowa, University of Iowa, Iowa City, Iowa2Department of Anthropology, University of Iowa, University of Iowa, Iowa City, Iowa3Department of Sociology and Anthropology, Metropolitan State University of Denver,Denver, Colorado4Department of Anthropology, Texas A&M University, College Station, TexasABSTRACTTo understand how variation in nasal architecture accommodates theneed for effective conditioning of respired air, it is necessary to assess themorphological interaction between the nasal cavity and other aspects of thenasofacial skeleton. Previous studies indicate that the maxillary sinuses mayplay a key role in accommodating climatically induced nasal variation suchthat a decrease in nasal cavity volume is associated with a concomitantincrease in maxillary sinus volume. However, due to conflicting results inprevious studies, the precise interaction of the nasal cavity and maxillarysinuses, in humans, is unclear. This is likely due to the prior emphasis onnasal cavity size, whereas arguably, nasal cavity shape is more importantwith regard to the interaction with the maxillary sinuses. Using computedtomography scans of living human subjects (N540), the goal of this study isto assess the interaction between nasal cavity form and maxillary sinus vol-ume in European- and African-derived individuals with differences in nasalcavity morphology. First, we assessed whether there is an inverse relation-ship between nasal cavity and maxillary sinus volumes. Next, we examinedthe relationship between maxillary sinus volume and nasal cavity shapeusing multivariate regression. Our results show that there is a positive rela-tionship between nasal cavity and maxillary sinus volume, indicating thatthe maxillary sinuses do not accommodate variation in nasal cavity size.However, maxillary sinus volume is significantly correlated with variation inrelative internal nasal breadth. Thus, the maxillary sinuses appear to be im-portant for accommodating nasal cavity shape rather than size. Anat Rec,296:414–426, 2013. VC 2013 Wiley Periodicals, Inc.Key words: climate; computed tomography; human variation;pneumatizationThe nasofacial skeleton displays a wide range of mor-phological variation among recent modern humanpopulations. Variability in this region of the cranium hasbeen largely assessed within the context of climatic ad-aptation (Thomson and Buxton, 1923; Davies, 1932;Weiner, 1954; Wolpoff, 1968; Hiernaux and Froment,Grant sponsor: National Science Foundation; Grant number:BCS-0550036 (N.E.H.). Grant sponsor: L.S.B. LeakeyFoundation (T.R.Y).*Correspondence to: Dr. Nathan Holton, Department ofOrthodontics, S219 Dental Science Building, University of Iowa, IowaCity, IA 52242. Tel: 319-335-7288; E-mail email@example.comReceived 2 October 2012; Accepted 6 December 2012.DOI 10.1002/ar.22655Published online 5 February 2013 in Wiley Online Library(wileyonlinelibrary.com).THE ANATOMICAL RECORD 296:414–426 (2013)VVC 2013 WILEY PERIODICALS, INC.
2. 1976; Carey and Steegmann, 1981; Crognier, 1981a,b;Franciscus and Long, 1991; Franciscus, 1995; Roseman,2004; Yokley, 2006, 2009; Holton and Franciscus, 2008;Hubbe et al., 2009; Noback et al., 2011; Holton et al.,2011, 2012) and the ability to modify the heat and mois-ture content of respired air (Proetz, 1953; Walker et al.,1961; Franciscus and Trinkaus, 1988; Cole, 1992; Fran-ciscus, 1995; Yokley, 2006, 2009; Sahin-Yilmaz andNaclerio, 2011). Populations from contrasting climates(e.g., northern vs. equatorial regions) are characterizedby differences in gross nasal cavity dimensions that alterthe available mucosal surface area and nasal passagebreadth. Populations from colder climatic conditions, forexample, exhibit taller, narrower, and longer nasal pas-sages compared to populations from warmer, humidclimates (Charles, 1930; Franciscus, 1995; Yokley andFranciscus, 2005; Yokley, 2006, 2009; Holton and Fran-ciscus, 2008; Noback et al., 2011; Holton et al., 2012).To understand how the nasal region adapts to varia-tion in climate and the precise way in which internalnasal morphology alters to accommodate the need forthe proper conditioning of respired air, it is necessary toexamine the broader morphological interactions betweenthe nasal cavity and non-nasal regions of the facial skel-eton. Previous studies have documented that, withregard to other components of the facial skeleton,aspects of the nasal region are relatively modularized(Chierici et al., 1973; Carey and Steegmann, 1981;Ackermann, 2005; Polanski and Franciscus, 2006; Mit-terocker and Bookstein, 2008). This is particularlyevident with respect to morphological variation in nasalcavity breadth. For example, external measures of nasalbreadth vary independently of other transverse dimen-sions of the facial skeleton both across extant apes(Polanski and Franciscus, 2006; Mitterocker and Book-stein, 2008) and in narrower comparisons within genusHomo (Glanville, 1969; Holton and Franciscus, 2008).Similarly, while altered patterns of facial skeletal growthin humans due to artificial cranial deformation influencelarger measures of facial breadth (e.g., bimaxillarybreadth), the breadth of the nose is unaffected (Anton,1989; Rhode and Arriaza, 2006). While it is unclear ifthe internal nasal cavity exhibits the same level of mor-phological autonomy, the potential independence of thenasal region, as suggested by external nasal morphology,may be key to the nasal cavity’s ability to accommodatevariation in climate.The lack of covariation between transverse dimensionsof the nasal cavity and other measures of facial width,and the ability for nasal breadth to vary as a function ofclimate, is potentially due to the spatial and morphologi-cal relationships between the nasal cavity and themaxillary sinuses (cf., Shea, 1977). Anatomically, thenasal cavity and maxillary sinuses share a commonbony wall, and as such, the sinuses may act as zones ofaccommodation for variation in nasal cavity breadth. Ifthe maxillary sinuses do in fact accommodate nasal vari-ation, one would predict that populations characterizedby more capacious nasal dimensions would exhibitsmaller maxillary sinus volumes than populations withreduced nasal cavity size. The potential accommodativecapacity of the maxillary sinuses is particularly impor-tant with regard to variation in nasal morphology giventhat the relative independence of the nasal cavity mayvary across different regions of the nose. For example,although there are potentially important population-spe-cific differences in superior nasal breadth related torespiratory function and air-conditioning capacity of thenasal cavity (e.g., Franciscus, 2003), variation in thebreadth of the superior nasal region may be constrainedby neural structures and visual organs (Kean andHoughton, 1987). In contrast, air-filled spaces in place ofbone and vital organs (i.e., the maxillary sinuses) areunlikely to impose the same constraints on other aspectsof nasal morphology. The potential spatial accommoda-tion provided by the maxillary sinuses may therefore becrucial to ensuring that the nasal cavity is capable offacilitating the requirements of heat and moisture trans-fer during respiratory function.The ability for the maxillary sinuses to accommodatenasal cavity variation is predicated, in part, on the max-illary sinuses serving a largely nonfunctional role withregard to respiratory function and physiology. Indeed,various functions have been proposed for the maxillarysinuses and paranasal sinuses in general (See Butaricet al., 2010 and references therein). However, while theevolutionary history of the paranasal sinuses is not wellunderstood (e.g., Witmer, 1997, 1999), there is littledirect evidence to indicate that they serve an importantfunctional role, at least in primates, particularly withrespect to respiration. As such, they are likely“evolutionary baggage” and may develop as the result ofopportunistic invasion of the sinus mucosa during ontog-eny (Shea, 1977; Blaney, 1986; Witmer, 1997; Rae et al.,2003; Rae and Koppe, 2004; Zollikofer and Weissmann,2008; Zollikofer et al., 2008; Smith et al., 2010). Follow-ing from this view, maxillary sinus form can beexplained as responding to the size and shape of the fa-cial skeleton and thus is a reflection of the surroundingcranial morphology.Shea (1977), in his examination of maxillary sinus vol-ume and climate in arctic populations, assessed thehypothesis [first proposed by Keith (1902)] that the max-illary sinuses accommodate variation in nasal cavityform. Using samples of cold-climate Inuit populationsalong with samples composed of Mongolian and Euro-pean-derived crania, Shea measured maxillary sinusvolume as a proportion of total maxillary volume. Hefound that temperature was positively correlated withrelative maxillary sinus volume such that an increase inmean temperature was associated with an increase inrelative maxillary sinus volume. In contrast, the propor-tion of nonsinus maxillary volume (i.e., the nasal cavityand surrounding maxillary bone) was inversely corre-lated with mean temperature values. Though Shea(1977) could not directly assess the relationship betweenmaxillary sinus and nasal cavity volume, he argued thatthe contrasting relationship between sinus volume andnonsinus maxillary volumes with temperature was afunction of the spatial interaction between the nasal cav-ity and maxillary sinuses.Subsequent analyses have offered mixed supportregarding the accommodative role of the maxillarysinuses. For example, Rae et al. (2003) assessed the rela-tionship between nasal cavity size and maxillary sinusvolume in a sample of climatically variable macaques.Consistent with Shea (1977), these researchers docu-mented a positive relationship between maxillary sinusvolume and temperature, along with a negative correla-tion between nasal cavity size and temperature.VARIATION IN NASAL ARCHITECTURE 415
3. Moreover, they found a negative relationship betweennasal cavity size and maxillary sinus volume indicatingthat a more capacious nasal cavity is associated withreduced maxillary sinus volume. Similarly, Marquez andLaitman (2008), also using macaques, documented nega-tive correlations between measures of nasal cavitybreadth and maxillary sinus volume. More recently,Butaric et al. (2010) examined the correlation betweendirect measures of maxillary sinus and nasal cavity vol-ume in a recent human sample encompassing a widerange of geographic and climatic variation. In contrastto previous studies, these researchers were unable todocument a meaningful relationship between maxillarysinus and nasal cavity volume in their climatically vari-able sample. Thus, contrary to Shea (1977), Butaricet al.’s (2010) results suggest that at least in humans,the maxillary sinuses may not act to accommodate nasalcavity variation across wider ranges of populationvariation.Based on the results of previous analyses, the preciserelationship between nasal cavity and maxillary sinusvolumes is unclear. However, with respect to the hypoth-esis that the maxillary sinuses accommodate nasalcavity variation, the shape of the nasal cavity is argu-ably more important than overall nasal cavity size. Agiven nasal cavity volume can be influenced by variationin nasal height, length, or breadth, with variation in allthree dimensions contributing to population differencesin external and internal nasal form (Thomson and Bux-ton, 1923; Davies, 1932; Weiner, 1954; Wolpoff, 1968;Hiernaux and Froment, 1976; Carey and Steegmann,1981; Crognier, 1981a,b; Franciscus and Long, 1991;Franciscus, 1995; Roseman, 2004; Yokley, 2006, 2009;Holton and Franciscus, 2008; Hubbe et al., 2009; Nobacket al., 2011; Holton et al., 2012). As such, a measure ofnasal cavity volume alone does not reflect how that vol-ume is distributed. Indeed, while Shea (1977) focusedhis assessment on nasal cavity volume, he underscoredthat variation in internal nasal breadth (potentiallyresulting from differences in nasal turbinate size) waslikely driving the relationship between the nasal cavityand maxillary sinuses suggesting the shape of the nasalcavity is more important than volume alone.In the present study, we test the hypothesis that themaxillary sinuses accommodate variation in nasal cavityform (i.e., size and shape) in samples of European- andAfrican-derived individuals. Previous research has estab-lished that these populations exhibit variation in nasalcavity morphology (e.g., Franciscus and Long, 1991;Franciscus, 1995; Yokley, 2006, 2009; Holton and Fran-ciscus, 2008; Holton et al., 2012) including keydifferences in both nasal cavity and piriform aperturebreadth. Moreover, with regard to maxillary sinus vol-ume, there is evidence to indicate that European-derivedpopulations are characterized by larger maxillary sinusvolumes (both absolute and scaled to facial size) whencompared with African-derived populations (Fernandes,2004a,b; Holton et al., 2011). As such, these populationsare particularly amenable to assessing the morphologicalinteraction between components of the nasofacialskeleton.Following from the results of previous studies (Shea,1977; Rae et al., 2003; Fernandes, 2004a,b; Marquez andLaitman, 2008; Holton et al., 2012), we make the follow-ing predictions. First, we predict that the European-derived sample will be characterized by both a signifi-cantly larger maxillary sinus volume and significantlysmaller nasal cavity volume when compared with theAfrican-derived sample. Second, we predict that therewill be a significant negative correlation between nasalcavity and maxillary sinus volumes such that individu-als with a larger nasal cavity are characterized bysmaller maxillary sinuses, while individuals with asmaller nasal cavity exhibit larger maxillary sinuses.Finally, we predict that maxillary sinus volume is signif-icantly correlated with variation in nasal cavity shape,particularly variation in internal nasal cavity breadth(e.g., Shea, 1977; Holton et al., 2011). Specifically, wepredict that narrower internal nasal passages are associ-ated with an increase in maxillary sinus volume, whilewider internal nasal passages are associated withreduced maxillary sinus volume.MATERIALS AND METHODSSamplesWe assessed the interaction between the nasal cavityand maxillary sinuses using computed tomography (CT)scans of a morphologically diverse sample composed ofEuropean- and African-derived living human subjects(N540). Our European-derived subsample (N520) con-sisted of N511 male and N59 female European-Americans, while our African-derived subsample (N520)was composed of N58 male and N512 female African-Americans and native South Africans. The subjectsincluded in this analysis were originally recruited forother studies related to the assessment of in vivo masti-catory function (Holton, 2009) and internal nasalmorphology (Yokley, 2006, 2009, 2010) and were used inour previous assessment of population variation in max-illary sinus volume (Holton et al., 2011). Subjects for thepresent study were selected from a larger pool of individ-uals and were included in the analysis based on theabsence of pathological conditions or medical proceduresthat affected the skeletal morphology of the nasal cavityand maxillary sinuses.Data Collection and AnalysisWe collected maxillary sinus and nasal cavity volumesusing Osirix imaging software (Rosset et al., 2004). Withrespect to the maxillary sinuses, we used unilateralright maxillary sinus volumetric measurements [origi-nally obtained for Holton et al., (2011)] by manuallysegmenting the sinus from coronally oriented CT images(Fig. 1). Similarly, we used coronally oriented CT imagesto calculate nasal cavity volume by manual segmenta-tion of the nasal cavity. For the purposes of thisanalysis, we defined the anterior–posterior borders ofthe nasal cavity as the region between the piriform aper-ture at the level of alare and the posterior border of thenasal cavity at the level of the choanae. Moreover, fol-lowing Butaric et al. (2010), we excluded the ethmoidalair cells in our calculation of nasal cavity volume.With respect to our first prediction, we tested for sig-nificant population differences in maxillary sinus andnasal cavity volumes using a nonparametric Mann–Whitney U-test. This included an assessment of bothabsolute maxillary sinus and nasal cavity volumes aswell as scaled volumes. We scaled nasal cavity and416 HOLTON ET AL.
4. maxillary sinus volumes to overall facial centroid size(mm) calculated from the following landmarks that takeinto account the height, width, and depth of the facialskeleton: nasion, prosthion, staphylion, frontomalar tem-porale (bilaterally), and ectomolare (bilaterally).Next, we tested the prediction that the volume of thenasal cavity is inversely related to maxillary sinus vol-ume using Pearson’s correlation coefficient. In ourprevious study (Holton et al., 2011), we documented pop-ulation differences in the bivariate relationship betweenthe maxillary sinuses and facial size such that for agiven facial size, European-derived individuals exhibitlarger maxillary sinuses compared to African-derivedindividuals. As such, it is possible that there are popula-tion differences in the bivariate relationship betweenmaxillary sinus and nasal cavity volume. This may infact account for the lack of a significant relationshipbetween sinus and nasal cavity volume reported byButaric et al. (2010). Therefore, we assessed the rela-tionship between our dependent and independentvariables across the total sample and within our individ-ual European- and African-derived samples. We usedANCOVA to test for potential significant differences inthe bivariate relationships between maxillary sinus andnasal cavity volume.We assessed our third prediction, that maxillary sinusvolume is correlated with nasal cavity shape, by firstquantifying the shape of the nasal cavity using a seriesof unilateral (right-side) three-dimensional coordinatelandmarks recorded from three planes along the lengthof the nasal cavity (Fig. 2a; Table 1). The first plane,Plane A, consisted of landmarks defining the externalskeletal nose (Fig. 2b): nasion, alare, and the tip of theanterior nasal spine (landmarks 1–3). The second plane,Plane B, was defined as a coronally oriented slice locatedat the midpoint between the anterior nasal spine andFig. 1. Maxillary sinus and nasal cavity segmentation. To calculate maxillary sinus and nasal cavity vol-umes, we manually segmented these regions from individual coronally oriented CT slices. The individualsegments were then reconstructed into three-dimensional renderings from which we obtained our volu-metric data.VARIATION IN NASAL ARCHITECTURE 417
5. staphylion (i.e., 50% nasal cavity length). Landmarks inthis region consisted of the superior-most extent of thenasal cavity (intersection between the perpendicularplate and the cribriform plate of the ethmoid), the later-almost aspect of the nasal wall (located in the region ofthe inferior turbinate), and the inferolateral and inferiorTABLE 1. Landmarks used to assess nasal cavity variationLandmarknumber Plane Description1 A Nasion2 A Alare3 A Anterior nasal spine4 B Superior aspect of the nasal cavity where the perpendicular plate ofthe ethmoid intersects the cribriform plate of the ethmoid5 B Lateral-most aspect of the lateral nasal wall6 B Inferior-most aspect of the lateral nasal wall at the intersection withthe nasal floor7 B Inferior aspect of the nasal cavity where the nasal septum articulateswith the nasal floor8 C Superior aspect of the posterior nasal aperture where the vomerarticulates with the body of the sphenoid9 C Lateral-most aspect of the posterior nasal aperture10 C Inferior-most aspect of the posterior nasal aperture at the intersectionwith the nasal floor11 C Inferior aspect of the posterior nasal aperture where the nasalseptum articulates with the nasal floorSee Fig. 2 for landmark and plane illustration.Fig. 2. Three-dimensional coordinate landmarks. (a) The landmarkdata were collected to assess nasal cavity shape from three planes.(b) The first plane, Plane A, consisted of landmarks used to define theexternal nasal aperture. (c) The second pane, Plane B, is located atthe midpoint of nasal floor length (i.e., 50% of anterior nasal spine-staphylion length). The landmarks in Plane B were used to assess theshape of the internal nasal cavity. (d) The third plane, Plane C, islocated at the posterior nasal aperture. Descriptions of the landmarksare found in Table 1.418 HOLTON ET AL.
6. aspect of the cavity in the midsagittal plane (Fig. 2c,landmarks 4–7). Finally, Plane C was defined as a coro-nally oriented slice located at the level of the posteriornasal aperture and included landmarks corresponding tothose found in Plane B (Fig. 2d, landmarks 8–11).To quantitatively examine the relationship betweenmaxillary sinus volume and nasal cavity shape, we firstaligned and scaled the coordinate landmarks using gen-eralized Procrustes analysis on the pooled sample. Next,we conducted a multivariate regression analysis withthe Procrustes coordinates and log-transformed maxil-lary sinus volume and used permutation tests (N51,000)to test for independence between the variables. To con-trol for between-group variation and the potentialconfounding effects of morphological covariance due tofactors such as shared population history, we used apooled within-group regression in MorphoJ (Klingen-berg, 2008–2010). We visually assessed the componentsof nasal cavity shape that are correlated with maxillarysinus volume using wireframe models.RESULTSThe descriptive statistics for nasal cavity and maxil-lary sinus volumes are presented in Table 2. Withrespect to our first prediction, we found that there is asignificant difference in maxillary sinus volume betweenour groups with the European-derived sample exhibitinga larger mean maxillary sinus volume than the African-derived sample (P0.001). On average, maxillary sinusvolume in the European-derived sample was 17.68 cm3compared to an average sinus volume of 11.34 cm3forthe African-derived sample. As evidenced by the box plotin Fig. 3, there is little overlap between the two distribu-tions, although one of the African-derived individualsfalls well within the European-derived distribution andexceeds the European interquartile range. Similarly,when scaled to facial centroid size (which was not signif-icantly different between the two samples; Table 2), theEuropean-derived sample was still characterized by asignificantly larger maxillary sinus volume (P0.001;Fig. 4).With regard to nasal cavity volume, the European-derived sample mean was smaller than the African-derived sample mean (45.00 and 47.43 cm3, respectively;Table 2; Fig. 5). Nevertheless, there was no significant dif-ference between the two samples. However, when scaledto facial centroid size, the African-derived sample exhib-ited a significantly larger nasal cavity volume (P50.040)compared to the European-derived sample (Fig. 6). It isalso important to note that there were no significant dif-ferences between the African-American and South Africangroups that compose the African-derived sample for maxil-lary sinus volume (P50.353), scaled maxillary sinusvolume (P50.547), nasal cavity volume (P50.602), orscaled nasal cavity volume (P50.841)With respect to our prediction that maxillary sinusvolume is inversely correlated with the size of the nasalcavity, we assessed the bivariate relationship betweenmaxillary sinus and nasal cavity volume both in ourcombined sample and within our individual European-and African-derived samples. Across the sample, therewas a low, albeit significant correlation between the var-iables (r50.338; P50.033). However, the individualsample correlation coefficients were greater than thecombined sample correlation coefficient. Within the Eu-ropean-derived sample, the correlation betweenmaxillary sinus volume and nasal cavity volume wasr50.760 (P50.001), while the correlation within the Afri-can-derived sample was r50.515 (P50.021). Contrary toour prediction, between- and within-group correlationsbetween volumetric variables were positive. Thus, anincrease in nasal cavity volume does not correspond to aconcomitant decrease in maxillary sinus volume.Figure 7 is a scatter plot illustrating the bivariate rela-tionship between maxillary sinus volume and nasalcavity volume with least squares regression lines fittedto the individual European- and African-derived samples.Results from the ANCOVA indicate that there is a signif-icant difference in the slopes of the regression linesbetween our samples (F512.05; P50.001). Given this dif-ference, we are unable to test for significant differencesin the Y-intercepts of the regression lines. Nevertheless,visual inspection of the scatter plot indicates that thereis a clear distinction in the relationship between nasalcavity and maxillary sinus volume in the two samples.For a given nasal cavity volume, the European-derivedsample is characterized by a larger maxillary sinus vol-ume when compared with the African-derived sample.With respect to our third prediction, the results of ourmultivariate regression analysis indicate that there is asignificant relationship between maxillary sinus volumeand nasal cavity shape (P0.001), specifically internalnasal cavity breadth. As maxillary sinus volumeincreases in our sample, there is a concomitant reduc-tion in internal nasal cavity breadth. Conversely, areduction in maxillary sinus volume is associated withan increase in the breadth of the internal nasal cavity.This is evident in the comparisons of the wireframemodels in Fig. 8. Relative to the mean shape (gray wire-frame), variation is restricted largely to landmark 5,which represents the lateral-most aspect of the internalnasal cavity in Plane B (see Fig. 2 for reference). Asmaxillary sinus volume increases, the lateral internalnasal cavity is medially displaced resulting in a relativereduction in internal nasal breadth. Conversely, asTABLE 2. Descriptive statistics and Mann–Whitney U-test resultsMeasurementsEuropean-derived African-derivedPMean (SD) Min–max Mean (SD) Min–maxMaxillary sinus volume (cm3) 17.68 (4.97) 11.96–30.94 11.34 (4.87) 3.29–23.00 0.001Nasal cavity volume (cm3) 45.00 (4.97) 36.23–64.34 47.43 (5.53) 37.76–61.27 0.157Facial centroid size (mm) 125.07 (7.05) 112.39–136.05 124.20 (7.12) 107.23–133.55 0.705Scaled maxillary sinus volume (%) 14.02 (3.34) 9.88–22.74 8.99 (3.60) 2.80–17.62 0.001Scaled nasal cavity volume (%) 35.81 (4.10) 31.10–47.09 38.02 (3.63) 32.08–43.13 0.040VARIATION IN NASAL ARCHITECTURE 419
7. maxillary sinus volume decreases, there is a lateral dis-placement of the lateral internal nasal cavity resultingin a relative increase in internal nasal breadth. It isnoteworthy that mediolateral shape variation associatedwith maxillary sinus volume is confined to the internalnasal cavity (Plane B). That is, with regard to maxillarysinus volume, anterior and posterior nasal aperturebreadths are invariant.DISCUSSIONOur ability to understand the evolutionary and devel-opmental causal dynamics that underlie morphologicaladaptations of the nasal region requires an understand-ing of the complex interaction between the nasal cavityand other components of the facial skeleton. Thisincludes a continued assessment of the potential interac-tion between the nasal cavity and maxillary sinuses.While the precise relationship between these spaces isnot well understood, our results indicate that acrossrecent European- and African-derived populations, themaxillary sinuses accommodate population variation ininternal nasal morphology.In our analysis, we first assessed whether there is aninverse relationship between nasal cavity and maxillarysinus volumes. With respect to our first prediction, wefound that there was a significant difference in maxillarysinus volume between our samples. On average, the Eu-ropean-derived sample was characterized by a maxillarysinus volume that was 36% larger than the African-derived sample. The considerable difference in maxillarysinus volume was retained when scaling sinus volume tofacial centroid size. This result is in contrast to the max-illary sinus volume values reported by Butaric et al.Fig. 3. Box plot comparison of maxillary sinus volume (cm3) in ourEuropean- and African-derived samples. The difference between thesamples was statistically significant (P0.001).Fig. 4. Box plot comparison of scaled maxillary sinus volume (%) inour European- and African-derived samples. The difference betweenthe samples was statistically significant (P0.001).Fig. 5. Box plot comparison of nasal cavity volume (cm3) in our Eu-ropean- and African-derived samples. There was no significant differ-ence between our samples (P50.157).Fig. 6. Box plot comparison of scaled nasal cavity volume (%) inour European- and African-derived samples. The difference betweenthe samples was statistically significant (P50.040).420 HOLTON ET AL.
8. (2010). In their study, they found that the bilateral max-illary sinus volumes for their Liberian (25.44 cm3; N55)and German (25.21 cm3; N55) samples were nearlyequal. However, our results are consistent with Fer-nandes (2004a,b) who also found, in an analysis of N553dry crania, that Europeans are characterized by signifi-cantly larger maxillary sinuses compared to Africans.We also found that nasal cavity volume was, on aver-age, smaller in the European-derived sample. Withregard to absolute nasal cavity volume, the differencebetween the samples was 5.0% but was not statisticallysignificant. When scaled to facial centroid size however,the difference in nasal cavity volume was significantlydifferent. This result, coupled with the differences inmaxillary sinus volume, is broadly consistent with thehypothesis that the maxillary sinuses accommodate vari-ation in nasal cavity size. However, we found that incontrast to the prediction from Shea’s (1977) model,nasal cavity volume is positively rather than negativelycorrelated with maxillary sinus volume both across thesample as a whole (albeit weakly) and within our indi-vidual European- and African-derived samples. A morecapacious internal nasal cavity, therefore, is associatedwith a larger rather than smaller maxillary sinusvolume.Given these results, variation in maxillary sinus vol-ume cannot be explained as a function of nasal sizeaccommodation, at least in our European- and African-derived comparisons. The positive correlations betweennasal cavity volume and maxillary sinus volume withineach sample suggest that this relationship may bedriven by an underlying facial size factor. Numerousstudies have documented that the volume of the maxil-lary sinuses correlates with measures of facial sizeacross primates (Koppe and Nagai, 1997; Koppe et al.,1999; Rae and Koppe, 2000) and in narrower compari-sons within humans (Holton et al., 2011; Rae et al.,2011). Moreover, Butaric et al. (2010) have shown thatnasal cavity volume is positively correlated with facialsize in recent humans. Thus, our positive correlationbetween maxillary sinus and nasal cavity volume islikely reflective of this size dynamic and may be influ-enced, in part, by potential variation in sexualdimorphism in maxillary sinus and nasal cavity volumesas evidenced by the scatter plot in Fig. 7 (see also Bastiret al., 2011). At least with regard to the European-derived sample, the females fall at the lower end of theregression line relative to the males. The same howeveris not true of the African-derived sample, although thismay be due to reduced control over sampling proceduresin using a living human sample.While the positive correlation between maxillary sinusvolume and nasal cavity volume within our European-and African-derived samples is likely a function ofunderlying facial size differences, variation in facial sizecannot explain the differences in the bivariate relation-ship between our samples. That is, for a given facial sizeand nasal cavity volume, the European-derived sampleis characterized by a greater maxillary sinus volumewhen compared with the African-derived sample. Thedifferences in maxillary sinus volume relative to nasalcavity volume in our samples may underlie the incon-gruity between our bivariate results and those presentedby Butaric et al. (2010). In their study, the authors wereunable to document a significant correlation betweenmaxillary sinus volume and nasal cavity volume using ageographically diverse sample of recent humans thatincluded European- and African-derived samples. How-ever, Butaric et al. (2010) focused their bivariateanalysis across human populations rather than assess-ing variation within populations, as in the presentbivariate analysis. As documented in our study, the cor-relation between nasal cavity and maxillary sinusvolume across the samples, while significant, is lowerthan the correlation within the individual samples.Nevertheless, given that each of the samples in Butaricet al. (2010) was composed of N55–6 individuals, it ispossible that they would not have had large enoughsample sizes to identify significant intrapopulationcorrelations.It is of note, that with regard to the different bivariaterelationships in our samples, the spread about theregression line in Fig. 5 is greater for the African-derived sample relative to the European-derived sample.This greater dispersion is potentially due to the inclu-sion of African-American and native African individualsin our African-derived sample. Because of the effects ofgenetic admixture, African-Americans span the range ofEuropean and African nasofacial variation (FranciscusRG, Foster AD, Nasofacial skeletal differentiation amongequatorial Africans, Europeans and African-Americans,in prep.). Thus, while we have documented statisticallysignificant differences between our European- and Afri-can-derived samples, these differences may be ratherconservative and are likely to be greater with the use ofa sample composed solely of native African individuals.With respect to our third prediction, we documented asignificant correlation between maxillary sinus volumeFig. 7. Scatter plot of ln maxillary sinus volume on ln nasal cavity vol-ume. European-derived individuals are represented by closed circles(males) and closed squares (females), while African-derived individualsare represented by open circles (males) and open squares (females).There is a significant positive correlation between nasal cavity volumeand maxillary sinus volume both across (r50.338; P50.033) and withinthe European- (r50.760; P50.001) and African-derived samples(r50.515; P50.021). As is evidenced by the regression lines, there aredifferent scaling relationships between the European- and African-derived samples. For a given nasal cavity volume, the European-derivedsample is characterized by a larger maxillary sinus volume.VARIATION IN NASAL ARCHITECTURE 421
9. and the shape of the nasal cavity. The result of our mul-tivariate regression analysis of Procrustes-scaled nasallandmarks and log-transformed maxillary sinus volumeindicates that a larger maxillary sinus is associated witha decrease in the transverse dimensions of the internalnasal cavity. In contrast, a smaller maxillary sinus isassociated with increased internal transverse nasaldimensions. As such, our results indicate that theFig. 8. Wireframe renderings of nasal cavity shape correlated withmaxillary sinus volume (gray wireframe5mean shape; black wirefra-me5deviations from mean shape). (a) Anteromedial view of the nasalcavity (unilateral right side). (b) Anterosuperior view of the nasal cavity(unilateral right side). For orientation, the three planes defined in Fig. 2are labeled on the left wireframes in both A and B, while externalnasal landmarks (Plane A) are labeled on the right wireframes in bothA and B. Nasal cavity variation is restricted primarily to the mediolat-eral displacement of the lateral nasal cavity in Plane B. As maxillarysinus volume increases, the lateral nasal wall is displaced mediallyresulting in a relative reduction in internal nasal breadth (left). In con-trast, a small maxillary sinus is associated with a lateral displacementof the lateral nasal wall resulting in a relative increase in internal nasalbreadth (right).422 HOLTON ET AL.
10. maxillary sinuses in European- and African-derived pop-ulations respond to population variation in nasal cavityshape (i.e., nasal cavity breadth) rather than to varia-tion in nasal cavity volume. The interaction betweennasal cavity shape and maxillary sinus volume explainsthe difference in the bivariate relationship betweennasal cavity volume and maxillary sinus volume in oursamples. That is, the European-derived sample exhibitsa larger absolute and relative maxillary sinus volumebecause it is characterized by a relatively narrower in-ternal nasal cavity. Conversely, the absolute and relativesmaller maxillary sinus volumes in the African-derivedsample are a function of increased internal nasal cavitybreadth. Our result is consistent with Shea (1977) who,while hypothesizing the influence of nasal cavity volumeon maxillary sinus volume, emphasized that variation ininternal nasal breadth dimensions likely drives the rela-tionship between the nasal cavity volume and the size ofthe maxillary sinuses.While the functional and evolutionary history of para-nasal sinuses is incompletely understood (e.g., Witmer,1997, 1999), their retention in humans, at least withregard to the maxillary sinuses, appears to be a key fac-tor in the nasal cavity’s ability to adapt to differentclimatic conditions and thereby maintain an effective airconditioning capacity. A primary function of the nasalcavity is the heating and humidification of air during in-spiration and the recapturing of heat and moistureduring expiration (Proetz, 1953; Walker et al., 1961;Cole, 1992; Hillenius, 1992, 1994; Franciscus, 1995;Keck et al., 2000a,b; Lindemann et al., 2001a,b,2007,2009; Wolf et al., 2004; Yokley, 2006, 2009; Elad et al.,2008; Sahin-Yilmaz and Naclerio, 2011; Holton et al.,2012). Effective heat and moisture exchange withrespired air is, in part, a function of the total mucosalsurface area of the nasal passages and breadth of therespired airstream (Schmidt-Nielsen et al., 1970; Collinset al., 1971; Hanna and Scherer, 1986; Schroter andWatkins, 1989; Yokley, 2006, 2009; Lindemann et al.,2009). In most mammalian species, temperature andmoisture regulation of respired air is accomplished viaextensive maxilloturbinates characterized by elaboratebranching patterns. These structures function toincrease the available nasal mucosal surface area bydividing the respired airstream into numerous smallerchannels (Negus, 1958; Hillenius, 1992, 1994; Van Val-kenburgh et al., 2004). Thus, in most mammals,variation in the ability to regulate heat and moisturecontent during respiration is a function of variation inthe density of maxilloturbinate bone and the resultingdecrease in the distance between mucosal surfaces (i.e.,gap width; Collins et al., 1971; Schroter and Watkins,1989). Primate maxilloturbinates however, includingthose of Homo sapiens, are much less complex than thetypical mammalian configuration and have been reducedto simple, scroll-like structures. Furthermore, decreasedreliance on olfaction among primates has allowed themajority of their ethmoturbinates, which primarilyhouse olfactory receptors in other mammalian groups(Hillenius, 1992, 1994), to be co-opted for air-condition-ing purposes. Given their simple turbinates and the factthat all but a small portion of their nasal passages arecovered with respiratory mucosa, humans must modifythe gross architecture of their nasal cavities to increaseefficiency of nasal heat and moisture exchange.Among humans, the need to modify internal nasalarchitecture to accommodate climatic variation is evi-dent in measures of perimeter and area of coronal crosssections through the nasal cavity, which have beenshown to be the most useful for assessing variation inmucosal surface area (Hanna and Scherer, 1986; Yokley,2006, 2009). For a given point along the passages, theperimeter of the internal nasal cross-sectional measuresmucosal surface area, while the perimeter-to-area ratiogauges the surface area of the mucosa relative to size ofthe nasal airway. Yokley (2006, 2009) found that Euro-pean-Americans have greater perimeter-to-area ratiosthan African-Americans at the anteroposterior midpointof the nasal passages (approximately Plane B in Fig. 2)when a decongested state is reconstructed. Thus, basedon expected differences in air conditioning capacity inpopulations derived from equatorial versus northern lat-itudes, the European-Americans were characterized by apredicted increase in available mucosal area comparedto African-Americans. Comparisons of perimeter-to-arearatios with linear measurements of the nasal cavityrevealed that these key functional differences in nasalmorphology are largely the result of variation in breadthdimensions of the nasal cavity.1The importance of the maxillary sinuses in accommo-dating nasal cavity variation in mammals with reducedturbinate structure is further evident in macaques. Rela-tive to other anthropoids, cercopithecoids arecharacterized by the loss of maxillary sinuses (e.g., Raeet al., 2002; Rae and Koppe, 2003). While maxillarypneumatization has reappeared in multiple fossil cerco-pithecoid taxa (Kuykendall and Rae, 2008; Rae, 2008),macaques are the only extant Old World monkeys thatdeviate from the typical condition (Rae et al., 2002).Whether maxillary sinus evolution in macaques wasdriven by climatic adaptation resulting from their wide-spread distribution is unclear. Nevertheless, as inhumans, the maxillary sinuses of macaques morphologi-cally respond to nasal cavity variation (Rae et al., 2003;Marquez and Laitman, 2008). Rae et al. (2003), forexample, found that there is an inverse relationshipbetween the size of the nasal cavity and maxillary sinusvolume in Macaca fuscata. However, since theseresearchers only assessed nasal cavity size, it is unclearif there are concomitant shape changes in the nasal cav-ity (e.g., allometric changes in nasal cavity breadth).Marquez and Laitman (2008), however, documented aninverse relationship between maxillary sinus volume1While assessments of internal nasal breadth have failed to findsignificant differences between European- and African-derivedsamples (Charles, 1930; Franciscus, 1995), these results arepotentially due to how breadth was measured. Charles (1930) andFranciscus (1995) both measured the greatest distance betweenthe lateral walls of the cavum nasi (Charles, 1930: p. 187). Thelack of significance between European- and African-derived sam-ples simply indicates that they each have a point along the nasalcavity where they become similarly wide. The location of thispoint, however, is variable. On the other hand, univariate analy-ses of additional measures of nasal cavity breadth (nasal breadth,superior and inferior ethmoidal breadths, choanal breadth) andmultivariate analyses of nasal cavity variation (Franciscus, 1995;Yokley and Franciscus, 2005; Yokley, 2009; Noback et al., 2011)have revealed significant differences in nasal cavity breadthdimensions in European- and African-derived groups (with theEuropean-derived samples being narrower in all cases).VARIATION IN NASAL ARCHITECTURE 423
11. and external nasal breadth in Macaca fascicularis, indi-cating that maxillary sinuses in macaques may respondto nasal cavity shape as well.The results of our analysis indicate that the clinal dis-tribution of maxillary sinus volume in our samples isdue to the interactive effects of nasal cavity shape.Indeed, this distribution, also documented in other stud-ies (Fernandes, 2004a,b; Holton et al., 2011), differs fromprevious analyses that have shown a contrasting pattern(i.e., hyperpneumatization in warmer climates) in bothcold-climate Inuit populations (Shea, 1977) and maca-ques (Rae and Koppe, 2000; Marquez and Laitman,2008). Similarly, there is evidence to suggest that ratsraised in cold climate conditions exhibit maxillary sinusreduction resulting from developmentally plasticchanges in facial morphology (Rae et al., 2006). In theabsence of a more detailed assessment of the interactionbetween nasal cavity shape and maxillary sinus volumein these other populations/taxa, the precise reason forthe differences in clinal distributions is unclear. It may,however, lie in the use of the expanded geographic rangeof European- and African-derived populations comparedto previous analyses, which have focused largely on vari-ation within northern latitude populations. Additionally,there may be population differences in nasal cavity–maxillary sinus integration along with taxonomic differ-ences across nonhuman primates and other mammalsthat at present we do not understand. Moreover, giventhat the maxillary sinuses are influenced by variousother factors such as overall facial size (Koppe andNagai, 1997; Koppe et al., 1999; Rae and Koppe, 2000;Rae et al., 2011; Holton et al., 2011) or potentially bypopulation and taxonomic differences in the infraorbitalmorphology (e.g., Maddux and Franciscus, 2009; Swen-son et al., 2009), it is likely that different (i.e., non-nasal) factors also influence these distributions. This isparticularly relevant with regard to sinus reduction inrodents (Rae et al., 2006) given that they exhibit anelaborate nasal turbinate structure (e.g., Schmidt-Niel-sen et al., 1970) and therefore may not be a particularlyuseful model for nasal cavity–maxillary sinus interac-tions in primates.Finally, the results of our analysis show that whilemaxillary sinus volume covaries with transverse nasaldimensions, this morphological relationship is restrictedto the breadth of the internal nasal cavity. The results ofour multivariate regression indicate that the breadth ofthe anterior and posterior nasal apertures is uncorre-lated with maxillary sinus volume. Whereas the resultsof previous studies have documented that external nasalbreadth varies independently of other transverse meas-ures of the facial skeleton (Anton, 1989; Polanski andFranciscus, 2006; Rhode and Arriaza, 2006; Holton andFranciscus, 2008; Mitterocker and Bookstein, 2008), ourresults indicate that, unlike the internal nasal cavity,differences in the external nasal aperture breadth arenot reflected in the maxillary sinuses. This suggests thatexternal and internal nasal breadth dimensions arepotentially responding to different dynamics. Indeed,while internal nasal breadth is likely responding to theneed to properly heat and humidify respired air (seeabove; e.g., Yokley, 2006, 2009), variation in externalnasal breadth, which is likely less important with regardto climatic adaptation (Yokley, 2006, 2009; Holton andFranciscus, 2008), responds, in part, to variation infacial prognathism. Holton and Franciscus (2008), forexample, demonstrated that variation in external nasalaperture breadth, within recent humans (i.e., Europeanand African populations) and across genus Homo, ispartly a secondary response to variation in the anterior–posterior dimensions of the facial skeleton. Nevertheless,we cannot rule out that variation in internal nasalbreadth dimension and the resulting effect on maxillarysinus volume are also influenced by variation in facialprognathism.CONCLUSIONSIn contrast to most mammals, the ability to properlycondition respired air in the human nose is dependenton modifications of the gross architecture of the internalnasal cavity. The ability to alter nasal cavity dimensions,particularly internal nasal breadth, appears to be due tothe maxillary sinuses acting as zones of accommodationfor the nasal cavity, whereby air-filled spaces in place ofbone allow for variation in the position of the lateralnasal walls (e.g., Shea, 1977; Enlow, 1990). Thus, ourresults suggest that while the sinuses are unlikely toplay a direct role in nasorespiratory function, they areimportant with regard to the accommodation of climati-cally relevant changes in internal nasal shape. However,there may be important population/taxonomic differen-ces in the relationship as evidenced by the results ofprevious studies (Shea, 1977; Rae et al., 2003, 2006;Marquez and Laitman, 2008; Butaric et al., 2010). Givenour focus on variation in these dynamics in European-and African-derived samples, it is unknown whetherthese results can be extended to other populations. Gen-erally, climatic variation in the internal nasal cavity ofnon-European and non-African populations has not beenas intensively studied. As such, our understanding ofthe means by which the human nose has adapted to var-iation in climate will benefit from continued assessmentof the interaction of the nasal cavity and non-nasal com-ponents of the facial skeleton across broader populationcomparisons.ACKNOWLEDGEMENTSThe authors acknowledge our anonymous reviewers fortheir valuable comments and suggestions.LITERATURE CITEDAckermann RR. 2005. Ontogenetic integration of the hominoid face.J Hum Evol 48:175–197.Anton SC. 1989. Intentional cranial vault deformation and inducedchanges of the cranial base and face. Am J Phys Anthropol79:253–267.Bastir M, Godoy P, Rosas A. 2011. Common features of sexualdimorphism I the cranial airways of different human populations.Am J Phys Anthropol 146:414–422.Blaney SPA. 1986. An allometric study of the frontal sinus in Go-rilla, Pan and Pongo. Folia Primatol 47:81–96.Butaric LN, McCarthy RC, Broadfield DC. 2010. A preliminary 3Dcomputed tomography study of the human maxillary sinus andnasal cavity. Am J Phys Anthropol 143:426–436.Carey JW, Steegmann AT. 1981. Human nasal protrusion, latitude,and climate. Am J Phys Anthropol 56:313–319.Charles CM. 1930. The cavum nasi of the American Negro. Am JPhys Anthropol 14:177–253.424 HOLTON ET AL.
12. Chierici G, Harrold P, Vargervik K. 1973. Morphogenetic experi-ments in facial asymmetry: the nasal cavity. Am J Phys Anthro-pol 14:177–253.Cole P. 1992. Nasal and oral airflow resistors. Arch OtolaryngolHead Neck Surg 118:790–793.Collins JC, Pilkington TC, Schmidt-Nielsen K. 1971. A model of respi-ratory heat transfer in a small mammal. Biophys J 11:886–914.Crognier E. 1981a. Climate and anthropometric variations inEurope and the Mediterranean area. Ann Hum Biol 8:99–107.Crognier, E. 1981b. The influence of climate on the physical diver-sity of European and Mediterranean populations. J Hum Evol10:611–614.Davies A. 1932. A resurvey of the nose in relation to climate. J RAnthropol Inst 62:337–359.Elad D, Wolf M, Keck T. 2008. Air-conditioning in the human nasalcavity. Respir Physiol Neurobiol 163:121–127.Enlow DH. 1990. Facial growth. 3rd ed. Philadelphia: Saunders.Fernandes CL. 2004a. Forensic ethnic identification of crania: therole of the maxillary Sinus—a new approach. Am J Foren MedPathol 25:302–313.Fernandes CL. 2004b. Volumetric analysis of maxillary sinuses ofZulu and European crania by helical, multislice computed tomog-raphy. J Laryngol Otol 118:877–881.Franciscus RG. 1995. Later Pleistocene nasofacial variation in west-ern Europe and Africa and modern human origins. Ph.D. Disser-tation, University of New Mexico.Franciscus, RG. 2003. Comparing internal nasal fossa dimensionsand classical measures of the external nasal skeleton in recenthumans: inferences for respiratory airflow dynamics and climaticadaptation. Am J Phys Anthrop Suppl 36:96–97.Franciscus RG, JC Long. 1991. Variation in human nasal heightand breadth. Am J Phys Anthropol 85:19–427.Franciscus RG, Trinkaus E. 1988. The Neandertal nose. Am J PhysAnthropol Suppl 75:209–210.Glanville EW. 1969. Nasal shape, prognathism and adaptation inman. Am J Phys Anthropol 30:29–38.Hanna LM, Scherer PW. 1986. A theoretical model of localized heatand water vapor transport in the human respiratory tract. J Bio-mech Eng 108:19–27.Hiernaux J, Froment A. 1976. The correlations between anthropo-biological and climatic variables in Sub-Saharan Africa: revisedestimates. Hum Biol 48:757–767.Hillenius WJ. 1992. The evolution of nasal turbinates and mamma-lian endothermy. Paleobiology 18:17–29.Hillenius WJ. 1994. Turbinates in therapsids: evidence for Late Per-mian origins of mammalian endothermy. Evolution 48:207–229.Holton NE. 2009. Modeling of masticatory biomechanics in livinghumans as a baseline for testing functional hypotheses in Nean-dertals. Ph.D. Thesis, University of Iowa, Iowa City.Holton NE, Franciscus RG. 2008. The paradox of a wide nasal aper-ture in cold-adapted Neandertals: a causal assessment. J HumEvol 55:942–951.Holton NE, Yokley TR, Franciscus RG. 2011. Climatic adaptationand Neandertal facial evolution: a comment on Rae et al. (2011).J Hum Evol 61:624–627.Holton NE, Yokley TR, Figueroa A. 2012. Nasal septal and craniofa-cial form in European- and African-derived populations. J Anat221:263–274.Hubbe M, Hanihara T, Harvati K. 2009. Climate signatures in themorphological differentiation of worldwide modern human popula-tions. Anat Rec 292:1720–1733.Kean MR, Houghton P. 1987. The role of function in the developmentof human craniofacial form—a perspective. Anat Rec 218:107–110.Keck T, Leiacher R, Riechelmann H, Rettinger G. 2000a. Tempera-ture profile in the nasal cavity. Laryngoscope 110:651–654.Keck T, Leiacher R, Heinrich A, K€uhnemann S, Rettinger G. 2000b.Humidity and temperature profile in the nasal cavity. Rhinology38:167–171.Keith A. 1902. Inflation of the nasal canal in the skulls of adultgorillas and chimpanzees, and the relative development of thesinus maxillaris and inferior meatus in man and apes. J AnatPhysiol 36:47–51.Klingenberg CP. 2008–2010. MorphoJ. Manchester: University ofManchester.Koppe T, Nagai H. 1997. Growth pattern of the maxillary sinus inthe Japanese macaque (Macaca fuscata): reflections on the struc-tural role of the paranasal sinuses. J Anat 190:533–544.Koppe T, Rae TC, Swindler DR. 1999. Influence of craniofacial mor-phology on primate paranasal pneumatization. Ann Anat 181:77–80.Lindemann J, K€uhnemann S, Stehmer V, Leiacker R, Rettinger G,Keck T. 2001a. Temperature and humidity profile of the anteriornasal airways of patients with nasal septal perforation. Rhinology39:202–206.Lindemann J, Leiacker R, Stehmer V, Rettinger G, Keck T. 2001b.Intranasal temperature and humidity profile in patients withnasal septal perforation before and after surgical closure. ClinOtolaryngol 26:433–437.Lindemann J, Keck T, Scheithauer M, Leiacker R, Weismiller K.2007. Nasal mucosal temperature in relation to nasal airflow asmeasured by rhinomanometry. Am J Rhinol 21:46–49.Lindemann, Joerg, Tsakiropoulou, Evangelia, Keck, Tilman,Leiacker, Richard, Wiesmiller, Kerstin M. 2009. Nasal air condi-tioning in relation to acoustic rhinometry values. Am J RhinolAllergy 23:575–577.Maddux SD, Franciscus RG. 2009. Allometric scaling of infraorbitalsurface topography in Homo. J Hum Evol 56:161–174.Marquez S, Laitman JT. 2008. Climatic effects on the nasal com-plex: a CT imaging, comparative anatomical, and morphometricinvestigation of Macaca mulatta and Macaca fascicularis. AnatRec 291:1420–1445.Mitterocker P, Bookstein F. 2008. The evolutionary role of modular-ity and integration in the hominoid cranium. Evolution 62:943–958.Negus V. 1958. The comparative anatomy and physiology of thenose and paranasal sinuses. Edinburgh: ES Livingstone.Noback ML, Harvati K, Spoor F. 2011. Climate-related variation ofthe human nasal cavity. Am J Phys Anthropol 145:599–614.Polanski JM, Franciscus RG. 2006. Patterns of craniofacial integra-tion in extant Homo, Pan, and Gorilla. Am J Phys Anthropol131:38–49.Proetz AW. 1953. Applied physiology of the nose. St. Louis: AnnalsPublishing.Rae TC. 2008. Paranasal pneumatization in extant and fossil Cerco-pithecoidea. J Hum Evol 54:279–286.Rae TC, Koppe T. 2000. Isometric scaling of maxillary sinus volumein hominoids. J Hum Evol 38:411–423.Rae TC, Koppe T. 2004. Holes in the head: evolutionary interpreta-tions of the paranasal sinuses in catarrhines. Evol Anthropol13:211–233..Rae TC, Koppe T, Spoor F, Benefit B, McCrossin M. 2002. Ancestralloss of the maxillary sinus in Old World monkeys and independ-ent acquisition in Macaca. Am J Phys Anthropol 117:293–296.Rae TC, Hill RA, Hamada Y, Koppe T. 2003. Clinal variation ofmaxillary sinus volume in Japanese macaques (Macaca fuscata).Am J Primatol 59:153–158.Rae TC. 2008. Paranasal pneumatization in extant and fossil Cerco-pithecoidea. J Hum Evol. 54:279–286.Rae TC, Koppe T. 2003. The term “lateral recess” and craniofacialpneumatization in old world monkeys (Mammalia, Primates, Cer-copithecoidea). J Morphol 258:139–199.Rae TC, Viðarsodttir US, Jeffery N, Steegman AT, Jr. 2006. Devel-opmental response to cold stress in cranial morphology of Rattus:implications for the interpretation of climatic adaptation in fossilhominins. Proc R Soc B 273:2605–2610.Rae TC, Koppe T, Stringer CB. 2011. The Neandertal face is notcold adapted. J Hum Evol 60:234–239.Rhode MP, Arriaza BT. 2006. Influence of cranial deformation on fa-cial morphology among prehistoric South Central Andean popula-tions. Am J Phys Anthropol 130:462–470.Roseman CC. 2004. Detecting interregionally diversifying naturalselection on modern human cranial form by using matched molec-ular and morphometric data. Proc Natl Acad Sci 101:12824–12829.VARIATION IN NASAL ARCHITECTURE 425
13. Rosset A, Spadola L, Ratib O. 2004. OsiriX: an open-source softwarefor navigating in multidimensional DICOM images. J Digit Imag-ing 17:205–216.Sahin-Yilmaz A, Naclerio RM. 2011. Anatomy and physiology of theupper airway. Proc Am Thorac Soc 8:31–39.Schmidt-Nielsen K, Hainsworth FR, Murrish DE. 1970. Counter-current heat exchange in the respiratory passages: effect on waterand heat balance. Respir Physiol 9:263–276.Schroter RC, Watkins NV. 1989. Respiratory heat exchange inmammals. Resp Physiol 78:357–368.Shea BT. 1977. Eskimo craniofacial morphology, cold stress and themaxillary sinus. Am J Phys Anthropol 47:289–300.Smith TD, Rossie JB, Cooper GM, Carmody KA, Schmieg RM,Bonar CJ, Mooney MP, Siegel MI. 2010. The maxillary sinus inthree genera of New World monkeys: factors that constrain sec-ondary pneumatization. Anat Rec 293:91–107.Swenson KE, Holton NE, Franciscus RG, Marshall SD, Nieves MA,Southard T. 2009. Facial suture plating alters maxillary sinus inSus scrofa. J Dent Res 88 (Spec Iss A):902.Thomson A, Buxton LHD. 1923. Man’s nasal index in relation tocertain climatic conditions. J R Anthropol Inst 53:92–122.Van Valkenburgh B, Theodor J, Friscia A, Pollack A, Rowe T. 2004.Respiratory turbinates of canids and felids: a quantitative com-parison. J Zool Lond 264:281–293.Walker JEC, Wells RE, Merrill EW. 1961. Heat and water exchangein the respiratory tract. Am J Med 30:259–267.Weiner JS. 1954. Nose shape and climate. Am J Phys Anthropol12:615–618.Witmer L. 1997. The evolution of the antorbital cavity of archosaurs: astudy in soft-tissue reconstruction in the fossil record with an analy-sis of the function of pneumaticity. Soc Vert Paleo Mem 3:1–73.Witmer LM. 1999. The phylogenetic history of paranasal sinuses.In: Koppe T, Nagai H, Alt KW, editors. The paranasal sinuses ofhigher primates: development, function and evolution. Berlin:Quintessance. p 21–34.Wolf MW, Naftali S, Schroter RC, Elad D. 2004. Air-conditioningcharacteristics of the human nose. J Laryngol Otol 118:87–92.Wolpoff MH. 1968. Climatic influence on the nasal aperture. Am JPhys Anthropol 29:405–423.Yokley TR. 2006. The functional and adaptive significance of ana-tomical variation in recent and fossil human nasal passages.Ph.D. Thesis, Duke University, Durham.Yokley TR. 2009. Ecogeographic variation in human nasal passages.Am J Phys Anthropol 138:11–22.Yokley TR. 2010. Cross-sectional variation in human nasal passagesand the effect on heat and moisture exchange. Am J Phys Anthro-pol Suppl 50:249.Yokley TR, Franciscus RG. 2005. Variation in nasal passage sur-face-area-to-volume ratios of recent and fossil humans. PaleoAn-thropol PAS Abstracts. A12.Zollikofer CPE, Weissmann JD. 2008. A morphogenetic model ofcranial pneumatization based on the invasive tissue hypothesis.Anat Rec 291:1446–1454.Zollikofer CPE, Ponce de Leon MS, Schmitz RW, Stringer CB. 2008.New insights into mid-late Pleistocene fossil hominin paranasalsinus morphology. Anat Rec 291:1506–1516.426 HOLTON ET AL.