ontogenetic characteristics of the vomeronasal organ in saguinus

Ontogenetic Characteristics of the Vomeronasal Organin Saguinus geoffroyi and Leontopithecus rosalia, WithComparisons to Other Primates

Timothy D. Smith,1,2* Kunwar P. Bhatnagar,3 Christopher J. Bonar,4 Kristin L. Shimp,1

Mark P. Mooney,2,5 and Michael I. Siegel2

1School of Physical Therapy, Slippery Rock University, Slippery Rock, Pennsylvania 160572Department of Anthropology, University of Pittsburgh, Pittsburgh, Pennsylvania 152603Department of Anatomical Sciences and Neurobiology, University of Louisville, Louisville, Kentucky 402924Cleveland Metroparks Zoo, 3900 Wildlife Way, Cleveland, Ohio 441095Department of Oral Medicine and Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania 15261

KEY WORDS vomeronasal; nasopalatine; callitrichid; Saguinus geoffroyi; Leontopithecus

ABSTRACT It has been suggested that the variabilityof the primate vomeronasal organ (VNO) may be greaterthan previously thought, especially among New Worldmonkeys. It is not clear to what extent VNO variationreflects ontogenetic, functional, or phylogenetic differ-ences among primates. The present study investigatedVNO anatomy in an ontogenetic series of two genera ofcallitrichid primates, in order to assess recent attempts todevelop VNO character states and to examine the evi-dence for VNO functionality at different life stages. Asample of six Leontopithecus rosalia, one L. chrysomelas,and six Saguinus geoffroyi was serially sectioned andstained using various methods. Two adult Callithrix jac-chus were also sectioned for comparative purposes. TheVNO of each primate was examined by light microscopyalong its entire rostrocaudal extent. VNOs of the tamarinswere described to determine whether they fit into 1 of 3character states recently attributed to various New Worldmonkeys. At birth, the two species of tamarins differed inthe nature of communication between the VNO and naso-palatine duct (NPD). Two of 3 neonatal S. geoffroyi exhib-ited a fused VNO duct in a more dorsal position (adjacentto the nasal cavity) compared to that of L. rosalia. TheVNO duct communicated with the NPD and was patent inneonatal L. rosalia. Both species appeared to have anage-related increase in the amount of sensory epitheliumin the VNO. Subadult L. rosalia had caudal regions of theVNO that were exceptionally well-developed, similar to

those of strepsirhine primates. Compared to subadults, alladult callitrichids appeared to have more ventral commu-nications of the VNO duct directly into the NPD. Adult S.geoffroyi and L. chrysomelas both had VNO sensory epi-thelium separated by multiple patches of nonsensory ep-ithelium. This contrasted with the VNOs of C. jacchus,which had a nearly continuous distribution of receptors onall surfaces of the VNO. The findings indicate that tama-rins have delayed maturation of the VNO epithelium, andthat some species have little or no perinatal function.These results also suggest that ontogenetic changes incraniofacial form may alter the position of the VNO intamarins. The present study supports the use of at leasttwo character states to categorize the VNO of variouscallitrichids, but it is suggested that one of these, previ-ously called “reduced sensory epithelium” should be in-stead termed “interrupted sensory epithelium.” The dis-tribution of VNO sensory epithelium does not appear toreflect phylogenetic influences; it is more likely a func-tional characteristic that varies throughout postnatal life.Therefore, this chemosensory system has a high degree ofplasticity relating to age and function, which in someinstances can confound the use of characteristics as phy-logenetic traits. Further study is needed to quantify VNOreceptors in various species to determine if functionaldifferences exist and if some species have more precociousVNO function than others. Am J Phys Anthropol 121:342–353, 2002. © 2002 Wiley-Liss, Inc.

Primate vomeronasal organs (VNOs) are highlyvariable epithelial structures found bilaterally inthe mucosa of the nasal septum. VNOs are generallybelieved to be functional structures in strepsirhinesand New World monkeys, but vestigial or absent incatarrhine primates (Ankel-Simons, 2000; Maier,2000). Smith et al. (2001a) suggested that VNO vari-ability in haplorhine primates has been underesti-mated, and stressed the need for further investiga-tion of this group to determine the phylogenetic andfunctional significance of VNO variation.

Most often, VNO function has been inferred basedon its morphology (e.g., presence or absence of re-

*Correspondence to: Timothy D. Smith, School of Physical Therapy,Slippery Rock University, Slippery Rock, PA 16057.E-mail: [email protected]

Received 26 March 2002; accepted 3 July 2002.

DOI 10.1002/ajpa.10165

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 121:342–353 (2002)

© 2002 WILEY-LISS, INC.

ceptors). The VNO of most strepsirhine primates isdivided into nonsensory and sensory regions (Looand Kanagasuntheram, 1972; Hunter et al., 1984),but extreme variability exists among haplorhines(see Smith et al., 2001a). Recently, Smith and Bhat-nagar (2000) and Smith et al. (2001a) described theontogenetic basis for the extremes of primate VNOvariability, most notably the VNO of some homi-noids (Bhatnagar and Smith, 2001; Smith et al.,2001b). The VNOs of humans and chimpanzees arequite atypical among primates, showing a uniformrespiratory-like epithelium, and a superiorly dis-placed position (Smith et al., 1998, 2001b, c; Bhat-nagar and Smith, 2001). The rare detailed descrip-tions of the VNO in New World monkeys suggestthat their morphology differs markedly from moststrepsirhine primates (Hunter et al., 1984; Tanigu-chi et al., 1992; Evans and Grigorieva, 1994). Inseveral species, a more or less uniform epitheliumhas been described, as seen in the human/chimpan-zee VNO. Unlike the latter species, receptors havebeen described in the VNOs of most New Worldprimates that have been studied (Maier, 1980;Hunter et al., 1984; Taniguchi et al., 1992; Mendozaet al., 1994).

As the receptor organ for the vomeronasal system,the VNO responds to nonvolatile chemicals, whereasthe olfactory epithelium is specialized for respond-ing to volatile signals (reviewed in Wysocki, 1979;Keverne, 1999). Scent-marking in callitrichidswould seem to provide indirect evidence for VNOfunction, since it is accomplished using extensiveglands that produce fluid (i.e., nonvolatile) secre-tions (Epple, 1981). Nonetheless, it is not clear towhat extent such stimuli may be detected via gusti-catory sensation, or may have associated airborneodorants that are detected by olfaction. Therefore, itis important to assess the specific importance of thevomeronasal pathway by anatomical or physiologi-cal examination.

Experimental evidence for the nature of VNOfunction is strong in Microcebus murinus (Schillinget al., 1990; Aujard, 1997), but has only been ob-tained on marmosets among New World primates

(Dixson, 1998). Even in light of findings on marmo-sets, it is uncertain how the accessory olfactory(vomeronasal) system responds differently than themain olfactory system. Experiments on callitrichidshave involved combined ablation of olfactory andvomeronasal pathways (Barret et al., 1993; Dixson,1993), rendering it difficult to determine the relativeimportance of vomeronasal vs. olfactory systems tochemosensation. Because pheromonal communica-tion may be mediated by olfactory rather than vome-ronasal chemoreception in some mammals (Dorrieset al., 1997; Kelliher et al., 2001), it should not beassumed that VNOs of all New World monkeys func-tion in pheromonal reception. In the absence of con-vincing experimental evidence, morphological as-sessment of VNO epithelia can indicate thefeasibility of VNO functionality in different NewWorld monkeys.

Based on functional characteristics of the VNOepithelium, Smith et al. (2001a) categorized pri-mates into five character states (Table 1). In somecases, these categories appear to typify large taxo-nomic groups (with some exceptions), such as strep-sirhines (well-developed) or VNO absence (all adultOld World monkeys studied to date). New Worldmonkeys may have the greatest variability in VNO,yet it is not clear to what extent VNO variationreflects ontogenetic, functional, or phylogenetic dif-ferences (Smith et al., 2001a). Not only do speciesappear to differ, but pronounced age-related differ-ences in the VNO receptor cell population may existpostnatally in at least some callitrichids (Evans andGrigorieva, 1994). The present study investigatesthe VNO in an ontogenetic series of two genera ofcallitrichid primates, in order to test the hypothesisthat tamarins exhibit relatively reduced VNO sen-sory epithelium at birth compared to juveniles oradults, and to assess the significance and adequacyof character states proposed in our recent review(Smith et al., 2001a).

MATERIALS AND METHODS

The histological sample included 6 Leontopithecusrosalia (4 neonates, 1 juvenile, and 1 adult), 1 adult

TABLE 1. Character states of primate VNO1

Character state Description

Well-developed VNO resembles that of rodents and other mammals exhibiting clear demarcation between a SEand NSE; found in most strepsirhines (Fig. 4H)

Sensory epithelium only Scant or no NSE observed; found in C. jacchus (Fig. 4F)Sensory epithelium reduced2 Where SE may be found on all sides of VNO, but is interrupted by multiple patches of NSE;

found in S. geoffroyi (Fig. 4B,C)Displaced VNO Superiorly displaced VNO (spatially separated from the paraseptal cartilages); simplified,

respiratory-like epithelium in VNO; patches of ciliated cells but no reliable reports ofreceptor cells; found in Homo sapiens, Pan troglodytes (Smith et al., 2001b)

VNO absence No structure resembling a VNO of any morphology is found in a typical or displaced position;refers to postnatal state; seen in all cercopithecoids examined to date (review in Bhatnagarand Meisami, 1998; Maier, 2000)

1 From Smith et al., (2001a).2 Term “interrupted” sensory epithelium is proposed here to replace “reduced” sensory epithelium. SE, sensory epithelium (a.k.a.vomeronasal neuroepithelium); NSE, nonsensory epithelium (note that this is similar to the “receptor-free epithelium” described byBreipohl et al. (1979), except for the usual absence of cilia).

ONTOGENY OF VNO IN TAMARINS 343

L. chrysomelas, and 6 Saguinus geoffroyi (3 neo-nates, 2 juveniles, and 1 adult) (Table 2). Two adultCallithrix jacchus were also sectioned to provide acomparison to the callitrichid for which the bestevidence for VNO functionality exists (Taniguchi etal., 1992; Dixson, 1993). All specimens were pre-pared as described in Roslinski et al. (2000) exceptregarding sectional thickness. Briefly, tissues weredissected free of the nasal cavity after fixation in10% buffered formalin, decalcified using a formicacid-sodium citrate solution, dehydrated in a gradedseries of ethanols, and embedded in paraffin. Blockswere sectioned serially at 10–14 �m (10–12 �m inthe region of the VNO), and every fifth section wasstained alternately with Gomori trichrome, hema-toxylin-eosin, or alcian blue-periodic acid-Schiff; in-tervening sections were saved or stained as needed.

Serial sections were examined by light micros-copy, using a Leica photomicroscope at the School ofPhysical Therapy, Slippery Rock University. TheVNO of each primate was examined from its rostralopening to its caudal termination into glandularducts. VNOs of tamarins were assessed to determinewhether they fit into 1 of 3 character states attrib-uted to various New World monkeys by Smith et al.(2001a): well-developed, sensory epithelium only, orsensory epithelium reduced (Table 1). The locationof the VNO duct (rostral opening) was described inregard to point of communication (nasal cavity orNPD). In all neonates and juveniles, the patency ofthe VNO duct was assessed by examining the rostralregion of the VNO at high magnification (�600–�1,000), and ducts were considered fused if a con-tinuous margin of epithelial cells closed off the VNOfrom any external communication. This was verifiedby mounting and staining intervening sections asneeded.

In neonates and juveniles, the relative (rostrocau-dal) position of specific structures was determinedduring examination by noting start and stop pointswithin the sectional series. Structures included the

premaxilla (from first rostral appearance to end atthe level of the palate), nasopalatine duct, vomero-nasal organ (from first appearance of vomeronasalduct lumen to the end of the vomeronasal neuroep-ithelium, excluding caudal gland ducts), vomer, firstdeciduous incisor (di1), second deciduous incisor(di2), and deciduous canine. In adults, a muchsmaller portion of tissue was sectioned, and start/stop points only were determined for the VNO. VNOlength also was determined by (except in two speci-mens that had epithelial damage to part of the VNO)by multiplying the number of sections containingthe VNO by sectional thickness.

RESULTS

In both genera of tamarins, the VNO was seen tohave a rostral opening and a caudal union withgland ducts. Absolute length of the VNO appeared tovary with age, generally appearing to be longer injuveniles than in neonates, and was greatest in theadult S. geoffroyi. The position of the rostral openingvaried according to age and species. In general, theVNOs of neonates opened more dorsally than injuveniles. The opening was likewise more dorsal injuvenile S. geoffroyi compared to the adult. In neo-natal S. geoffroyi, VNOs actually opened (if patent)

TABLE 2. Age, source, and vomeronasal organ length of callitrichid specimens1

Sp. no. Species Age Sex Source Right VNO length (in �m)

SG3 Saguinus geoffroyi n(0) F CZ 1,650 �mSG4 Saguinus geoffroyi n(0) F CZ 1,980 �mSG5 Saguinus geoffroyi n(0) F CZ ?MM105 Saguinus geoffroyi 1 month M CZ 2,800 �mMM101 Saguinus geoffroyi 2 month F CZ 2,900 �mSG6 Saguinus geoffroyi 2.75 years F CZ 3,900 �mLR1 Leontopithecus rosalia n(3) F CZ 1,300 �mLR2 Leontopithecus rosalia n(4) M CZ 1,150 �mLR3 Leontopithecus rosalia 4 months F CZ 1,800 �mLR4 Leontopithecus rosalia n(9) F CZ 1,200 �mLR5 Leontopithecus chrysomelas 13 years M CZ ?LR6 Leontopithecus rosalia n(0) F CZ ?LR7 Leontopithecus rosalia Adult F SB ?CJ21 Callithrix jacchus 21 years M DM 2,700 �mCJ16 Callithrix jacchus 8 years M DM 2,880 �m

1 Sp. no., specimen number; Age, n � neonatal (days postpartum); Sex: F, female; M, male; Source: CZ, Cleveland Metroparks Zoo; DM,Duke University Medical Center; SB, Department of Anatomical Sciences, State University of New York at Stony Brook; ?, epithelialdamage to start or endpoints prevented accurate measurement; all sectioned tissues currently housed at School of Physical Therapy,Slippery Rock University.

Fig. 1. Schematic tracings, based on coronal sections throughneonatal (A–C, D–F) and juvenile (G–I, 2-month-old) S. geoffroyiat selected intervals (left � rostral-most section). Left columnshows starting point of NPD (double arrows); middle columnshows site of VNO duct opening (small open arrow); right columnshows first section where VNO (large open arrow) diverges fromits rostral communication. Note that in the neonate in top row,region of VNO duct opening was completely fused (B), whereas inthe other neonate (E), VNO duct opened into nasal cavity, with nocommunication to NPD. In juveniles, VNO duct opened into dor-sal (nasal) half of NPD. pmx, premaxilla; ns, nasal septum; vnc,vomeronasal cartilage; vom, vomer; pi, permanent incisor toothbud. Rostrocaudal distance between serial sections is as follows:A–B, 200 �m; B–C, 100 �m; D–E and E–F, 100 �m; G–H, 500 �m;H–I, 100 �m. Scales � 1 mm.

344 T.D. SMITH ET AL.

Fig. 1.


into the base of the nasal cavity rather than in theNPD (see VNO opening in Fig. 1E, and more caudalnasal opening of NPD in Fig. 1F). In two neonatal S.geoffroyi, VNO openings were found at the samelevel, but were completely fused (Fig. 1 B,C). Fusionwas verified by examining every serial section andverifying epithelial continuity across the region ofthe VNO duct opening (compare Fig. 1B,E). Neona-tal L. rosalia exhibited VNOs that communicatedwith the NPD alone (Fig. 2B), or with both the NPDand nasal cavity (Fig. 2E). A 9-day-old L. rosalia hada VNO that had communication to the NPD near thenasal cavity on one side, whereas the other side hada duct that joined the NPD more ventrally. In all L.rosalia and S. geoffroyi juveniles, the VNO duct wasobserved to open directly into the NPD very near thebase of the nasal cavity (Figs. 1, 2).

The VNO duct (i.e., a stratified, nonsensory epi-thelium connecting the VNO to the NPD or nasalcavity) was an extremely short segment of the VNOin both tamarins. In neonates, the VNO opening wasmost often a lateral gap in the lumen, often caudal tothe beginning of the VNO. In juvenile S. geoffroyi,the duct was found just rostral to a dilated portion ofthe VNO in which sensory epithelium could be seen.In the juvenile L. rosalia, there was barely any VNOduct at all (i.e., sensory epithelium was seen nearthe NPD). Caudal to the VNO duct, the position ofthe VNO varied according to age and species (Figs.1, 2). VNOs were found directly adjacent to the lowerseptal cartilage margin in neonatal Leontopithecusspp., whereas in the juvenile, the VNO was spatiallymore distant from (ventral to) the septal tip. Incontrast, the VNOs of S. geoffroyi were completelyventral to the inferior septal tip at all ages.

In the rostrocaudal position, spatial relationshipsof the VNO to other midfacial structures differedbetween species in neonates and juveniles. In S.geoffroyi, the VNO was spatially overlapped by thevomer bone (Fig. 1), whereas the vomer only over-lapped the caudal half of the VNO in L. rosalia. Inneonates, the VNO appeared to closely overlap therostrocaudal position of the deciduous canines in S.geoffroyi, whereas the VNO was relatively more ros-tral in L. rosalia. A marked increase in palatalheight was seen in juvenile tamarins compared toneonates, along with greater size of permanent in-cisor tooth buds (Figs. 1G, 2G).

The neonate/juvenile tamarins all had indistinctsensory and nonsensory epithelium in the most ros-tral part of the VNO, i.e., receptor cells were seen inscattered patches on all borders of the VNO epithe-lium. This typified the entire rostrocaudal extent ofthe VNO in S. geoffroyi (Figs. 4A,B) and the rostralthree-fourths of the VNO in L. rosalia (Fig. 4D).Distinct zones were seen that were mostly thin andmostly nonsensory, with sparse receptor cells. Cau-dally, the VNO of three of the neonatal and thejuvenile L. rosalia resembled the strepsirhine con-dition, with distinct medial sensory epithelium andlateral nonsensory epithelium (Fig. 4E). One neona-

tal L. rosalia had damage to the snout due to ma-ternal biting, which resulted in damage to the pal-atal region. However, the possibility of a palatal cleftalso existed and was investigated histologically. Inthis neonate, palatal continuity was observed on theright, and on the left side an apparent cleft palatewas seen. The NPD was patent, but the communi-cation with the VNO duct was either obscured bytissue damage or distorted by palatal malformationsin this specimen. Some cartilaginous abnormalitieswere noted, and well-defined VNOs were not foundin this specimen, although structures consistentwith vomeronasal nerves were seen encircled byvomeronasal cartilages.

The VNO duct was observed to enter the NPD atthe middle of the NPD in adult S. geoffroyi (Fig 3A).A similar morphology was observed in the adult C.jacchus (Fig. 3B). The precise location of the VNOopening in the adult L. rosalia and L. chrysomelaswas not discernable due to tissue damage, but sincethe nasal cavity floor was visible in both cases, theopenings were surmised to occur below the level ofthe nasal cavity. In adult L. chrysomelas, S. geof-froyi, and C. jacchus, the VNO epithelium had re-ceptors on all sides of the VNO, and none of thesespecies had nonsensory epithelium in an exclusivelylateral position. However, patches of nonsensory ep-ithelium were evident in L. chrysomelas and S. geof-froyi (Fig. 4C), and appeared in various positions(dorsal, ventral, or lateral). The VNO was mostlyabsent from sections in adult L. rosalia (due to anerror during dissection), but could be observed tohave both sensory epithelium and nonsensory epi-thelium dorsally. The receptor population appearedto be largest in C. jacchus, where the sensory epi-thelium was continuous (Fig. 4F), in contrast to theinterrupted segments of sensory epithelium seen inthe other adult callitrichid VNOs.

DISCUSSIONFusion/patency and position of VNO duct

Positional variation of the primate VNO is strik-ing (Bhatnagar and Meisami, 1998; Smith et al.,1998; Smith and Bhatnagar, 2000). When compar-ing the VNOs of humans or chimpanzees to any

Fig. 2. Schematic tracings based on coronal sections throughneonatal (A–C, D–F) and juvenile (G–I, 4-month-old) L. rosalia atselected intervals (left � rostral-most section). Left column showsstarting point of NPD (double arrows). Middle column shows siteof VNO duct opening (small open arrow). Right column showsfirst section where VNO (large open arrow) diverges from itsrostral communication. Although both neonates had VNO ductscommunicating with NPD very close to nasal cavity, note slightlydifferent appearances of VNO duct communication points in ne-onate in top row (B) compared to second neonate (E), which maybe due to slight differences in sectional planes. In the juvenile,VNO duct opened slightly more ventrally into the NPD. pmx,premaxilla; ns, nasal septum; vnc, vomeronasal cartilage; pi,permanent incisor tooth bud. Rostrocaudal distance betweenserial sections is as follows: A–B and B–C, 100 �m; D–Eand E–F, 120 �m; G–H and H–I, 240 �m. Scales �1 mm.


Fig. 2.


strepsirhine examined to date, the rostral openingand location of the VNO itself differ markedly(Smith et al., 2001b). The unusual position of thehuman VNO (and probably that of chimpanzees) isbest understood as the retention of a primitive em-bryonic position. After its invagination into an epi-thelial tube in stage 17 embryos, the VNO of Homo,Microcebus, and other mammals opens into the na-sal cavity alone (Fig. 3C; Yoshida et al., 1993; Smithand Bhatnagar, 2000); the VNO opening into theNPD is a secondary communication that occurs incertain mammals after the secondary palate ele-vates (Fig. 3D; Smith and Bhatnagar, 2000). Thepresent study shows that not all New World pri-mates exhibit NPD/VNO duct communication atbirth. Notably, this communication is not estab-lished in S. geoffroyi, but develops sometime duringinfancy. Likewise, neonatal Leontopithecus rosalia

exhibited a VNO duct opening that was very close tothe nasal cavity, although not to the same degree asin S. geoffroyi. It is unfortunate that no prenatalspecimens of these species were available for study,but one might speculate that if neonatal S. geoffroyilacked a communication of the VNO to the NPD, ithad not yet been established during prenatal/peri-natal ontogeny.

Wohrmann-Repenning and Bergmann (2001) in-terpreted the nasal opening of the VNO in someadult primates (e.g., Tarsius) as occurring into afunnel-shaped, dorsal extension of the NPD. Thiswas clearly not the case in S. geoffroyi, where theVNO opening was seen to be spatially separatedfrom the NPD (rostral and dorsal to it) in the coronalplane. Although minute changes in the plane of sec-tioning might influence the apparent proximity ofstructures (Fig. 2A,D), the completely fused VNOopening in two S. geoffroyi eliminated the possibilitythat the VNO and NPD communicated. Thus, fluidspassing through the NPD do not have direct accessto the VNO at birth in this species. Aside from L.rosalia, patent VNO ducts have been described in S.labiatus and Cebuella pygmaea (both as early as 2days postnatal); both species have VNOs that com-municate directly into the upper NPD at this age(Evans and Grigorieva, 1994). This emphasizes theinteresting possibility that both the communicationsite of the VNO duct and the timing of patency (andthus access to chemostimuli) vary among cal-litrichids. Furthermore, age-related differences be-tween tamarin species (see below) suggest that thematuration of the VNO sensory epithelium may becorrelated with the timing of VNO duct patency,which is clearly not the case in mice (Coppola et al.,1993).

Since 2 of the 3 neonatal S. geoffroyi had fusedVNO ducts in the present study, duct patency maybe roughly timed as a perinatal event (this issue isless certain in L. rosalia, since the youngest noncleftspecimens died 3–9 days after birth). Wohrmann-

Fig. 3. Coronal sections, showing communication of VNOduct (open arrows) with nasopalatine duct (npd) in adult S. geof-froyi (A) and C. jacchus (B). Note similar location of communica-tion site, directly into NPD. In primates such as M. murinus(C, D: 16-mm embryo and 37-mm fetus, respectively, from Blunt-schli Collection, Department of Mammalogy, American Museumof Natural History), this communication develops prenatally,since embryos (C) exhibit a primitive VNO communication tonasal cavity (open arrow), and communication to NPD developsafter secondary palate elevates, as shown in fetal M. murinus (D).ns, nasal septum; vnc, vomeronasal cartilage. Scale bars: A, B �500 �m; C, D � 300 �m.

Fig. 4. Coronal sections, in which variations of VNO epithe-lium show distribution of sensory epithelium (open arrows) andnonsensory epithelium (small double arrows) in vomeronasal or-gan (vno). A: In neonatal S. geoffroyi, VNO epithelium was ex-tremely thin, with few visible receptors (L, lumen of VNO). B:Juvenile S. geoffroyi had patches of thicker sensory epitheliumseparated by thinner regions of nonsensory epithelium. C: AdultS. geoffroyi had sensory epithelium on all sides of VNO, withsmall intervening regions of nonsensory epithelium. Rostrally,neonatal and juvenile (D) L. rosalia had sensory epithelium on allsides of VNO, but caudally (E), VNO resembled that of moststrepsirhines in having a thick medial sensory epithelium and amedial nonsensory epithelium. F: Neither species closely resem-bled VNOs of adult C. jacchus, which had uniformly thick sensoryepithelium in all aspects (open arrows indicate receptor layer ofcells). In other primates such as M. murinus, a uniformly thickVNO epithelium is seen early in embryonic stages (G), whereasfetal VNO (H) exhibits partitioning of VNO into medial sensoryepithelium and lateral nonsensory epithelium (same embryo andfetus as in Fig. 3). ns, nasal septum; vnc, vomeronasal cartilage.Scale bars: A, B � 250 �m; C, D � 500 �m; E � 250 �m; F, G �75 �m; H � 20 �m.


Fig. 4.


Repenning and Barth-Mueller (1994) presented aninteresting hypothesis regarding the timing of thepatency of ducts leading to the VNO in mammals.They proposed that the early (prenatal) patency ex-hibited by some mammals relates to earlier auton-omy of the young (precociality). The inverse of thishypothesis, that altricial young have prolonged fu-sion of the NPD, does not appear to be true. Cal-litrichids have highly dependent young, with infantsbeing continually carried by one or more adults forthe first 3 weeks (Moynihan, 1970; Hoage, 1982;Yamamoto, 1993). Yet the present study indicatesvariable timing of VNO ductal patency for theserelatively altricial mammals. These results parallelfindings on laboratory rodents that have altricialyoung but may either develop VNO ductal patencyprenatally (rats) or postnatally (mice) (Coppola etal., 1993; Coppola and Millar, 1994).

Such variation has clear functional implications(nonpatency eliminates the possibility of chemo-stimulus access), but the significance of the locationof the VNO opening is less clear. In most primates,the VNO has been described to open within theNPD, although relatively dorsally in New Worldspecies (e.g., Evans and Grigorieva 1994; Mendozaet al., 1994). Wohrmann-Repenning and Bergmann(2001) described the VNO of Arctocebus calabarensisas opening far more dorsally than in any other strep-sirhine, arguably at the interface of the nasal cavityand NPD. The VNO of adult Tarsius bancanusopens into the base of the nasal cavity (Wohrmann-Repenning and Bergmann, 2001). The same appearsto be true in Pithecia spp., both in subadults (Maier,1980) and adults (authors’ unpublished data). Suchvariability in primates, combined with disagree-ment regarding the primitive vs. derived VNO com-munication site of mammals in general (Wohrmann-Repenning, 1993; Giere et al., 1999; Sanchez-Villagra, 2001), suggest that it may not be a verydependable phylogenetic characteristic in primates.The resultant (adult) location of the VNO opening

may serve as a viable distinction among higher taxa(e.g., perhaps hominoids vs. New World primatesand strepsirhines). However, the present study il-lustrates that some species have a varying commu-nication site across ages.

The age-related variability of the communicationsite in S. geoffroyi and L. rosalia suggests that theposition of the VNO opening may reflect a subtle on-togenetic phenomenon. Data from the present studyimply that the VNO opening undergoes a spatial tran-sition from a more dorsal (nasal) opening to a commu-nication nearer to the middle of the NPD. Since thisalso occurs in other mammals, this displacement dur-ing growth would not be unusual in callitrichids. How-ever, this process occurs prenatally in mouse lemurs(Smith and Bhatnagar, 2000), whereas the presentstudy suggests similar postnatal changes in tamarins.A notable difference in juvenile tamarins compared toneonates included a markedly greater palatal height(dorsoventrally), which was accompanied by greatersize of the permanent incisor tooth buds (Figs. 1G, 2G).This increase in palatal height likely influences thelength of the NPD and position of the VNO duct rela-tive to the NPD, perhaps surrounding both ducts dur-ing a differential growth process. The rostrocaudalrelationship of craniofacial elements further suggeststhat as the premaxilla enlarges, the VNO and relatedcartilages remain in a more caudal position whilemaintaining a communication with the NPD. Finally,it appeared that the VNO of S. geoffroyi had moreoverlap with the vomer bone, perhaps relating to therelatively reduced midfacial region in Saguinus spp.(Ackermann and Cheverud, 2002). The relationshipbetween midfacial reduction and positional variabilityof the VNO should be investigated further in a broadspectrum of New World monkeys.

Microscopical anatomy of VNO epithelia

Based on the detailed descriptions that exist of theVNO in New World monkeys (Table 3), function ofthe VNO and timing of function may differ among

TABLE 3. Epithelial characteristics reported for the VNO of various New World primates1

Species Morphological descriptionCharacter

state References

Cebuella pygmaea2 “Complete neurosensory lining”3 SEO 1Saguinus fuscicollis Lumen “limited medially and laterally” by SE SEO? 2, 4Saguinus geoffroyi Interrupted SE ISE 7Saguinus labiatus “Complete neurosensory lining”2 SEO 1Saimiri sciureus2 “No clear distinction” between SE/NSE; “irregular height” of epithelium ISE? 3Leontopithecus rosalia Interrupted SE; more SE caudally ISE 7Leontopithecus chrysomelas Interrupted SE ISE 7Callithrix jacchus2 “Uniformly lined” with SE SEO 5, 6, 7Aotus trivergatus2 VNO epithelium is “similar to that described in other New World primates”4 SEO? 2Alouatta caraya Specimen with damaged epithelium; SE likely existed based on present

VNN? 6

Ateles geoffroyi2 VNO epithelium “similar to that lining the” NPD ? 2

1 SE, sensory epithelium; NSE, nonsensory epithelium; character state: SEO, sensory epithelium only; ISE, interrupted sensoryepithelium (revised from “reduced” sensory epithelium in Smith et al., 2001a); References: 1, Evans and Grigorieva (1994); 2, Hunteret al., (1984); 3, Maier (1980); 4, Mendoza et al. (1994); 5, Taniguchi et al. (1992); 6, Smith et al. (2002); 7, data from present study.2 Species for which an accessory olfactory bulb was reported by Stephan et al. (1981). 3 But note description of irregular epithelialcharacteristics in subadults by Evans and Grigorieva (1994). 4 But note appearance of mediolateral differences in VNO epitheliumshown in Figure 3c of Hunter et al. (1984).


taxa. Most species have been described as having avariable degree of sensory epithelium comprisingthe VNO (Maier, 1980; Mendoza et al., 1994), butreceptor populations may not be fully established atbirth in at least some callitrichids (Evans and Grig-orieva, 1994). Ateles geoffroyi may not have a che-mosensory VNO at all, according to Hunter et al.(1984). Thus VNO epithelial characteristics acrosstaxa are not well-explained by phylogenetic relation-ships, although further study may reveal patternswithin certain primate taxa.

In agreement with the findings of Hunter et al.(1984) and Evans and Grigorieva (1994), our mor-phological evidence indicates that the receptor pop-ulation (i.e., the major constituents of the sensoryepithelium) varies among species and may vary ac-cording to age. A striking aspect of the VNO inneonatal tamarins was the sparseness of receptorcells, as well as their intermittent distributionacross various sides of the VNO wall. Compared toS. geoffroyi, neonatal and juvenile L. rosalia ap-peared to have a larger population of receptors cau-dally, as well as partitioning of lateral nonsensoryand medial sensory epithelia. Since a division of theVNO epithelium into two zones is common amongmost mammals (including strepsirhine primates),juvenile L. rosalia had a more primitive VNO epi-thelial organization than any other callitrichid de-scribed thus far (no data exist to determine if thisalso typifies adult L. rosalia). Such variations com-plicate an understanding of the phylogenetic signif-icance of VNO morphology in haplorhines.

It is instructive to compare primates with thegreatest extremes in postnatal morphology, such asHomo sapiens and Microcebus spp., at differentstages of development (Roslinski et al., 2000; Smithand Bhatnagar, 2000; Smith et al., 2001d). In bothspecies, the embryonic VNO is initially of uniformthickness (Fig. 4G), and appears to be entirely com-prised of receptor epithelium. By late embryonic/early fetal stages, Microcebus exhibits distinct non-sensory/sensory epithelia (Fig. 4H), whereas Homoretains a homogeneous epithelial lining of the VNO.An exception is that some human embryos (but notfetuses) have VNOs with a thinner lateral regionresembling a nonsensory epithelium (Smith andBhatnagar, 2000). The present study extends thisobservation to other haplorhines, suggesting thatpartitioning of the VNO into sensory/nonsensory ep-ithelia may be a typical developmental phase in L.rosalia. In some species (such as Homo sapiens) thismorphology is inconsistent and might best be con-sidered aberrant. To discern phylogenetic VNOcharacter-states in haplorhines is therefore an intri-cate endeavor. It appears that most adult haplo-rhines do not possess a partitioning of the VNOepithelium; this may include adult tarsiers, since acompletely sensory VNO epithelium was describedin Tarsius bancanus (Wohrmann-Repenning andBergmann, 2001). Since characteristics appear tovary according to age, further ontogenetic investiga-

tions of diverse taxa are sorely needed to determinewhich primates retain an embryologically primitive,homogeneous VNO epithelium, and which do not.

Although the minimum number of VNO receptorsneeded for pheromonal reception is unclear, receptorpopulations can estimate the magnitude of functionin chemosensory systems (discussed in Dawley,1998). The present study suggested interesting on-togenetic changes in the capacity for VNO functionin the two tamarins under study, supporting thesuggestion by Evans and Grigorieva (1994) of rela-tively delayed neurogenesis in the VNO of somecallitrichid primates compared to other mammals.An apparent age-related increase in VNO sensoryepithelium (in L. rosalia) and VNO length (in S.geoffroyi) in both tamarins contradicts the notionthat the VNO is relatively degenerative in cal-litrichid primates (Maier, 1980; Hunter et al., 1984).Qualitative observations also suggest that neonatalL. rosalia exhibits a larger population of VNO recep-tors, although the average age of the two sampleswas separated by several days. The descriptions ofneonatal S. labiatus and C. pygmaea also show spe-cies differences in the amount of receptors, but bothspecies showed indirect (histochemical) evidence forVNO functionality at birth (Evans and Grigorieva,1994). Both the data from the present study andthose of Evans and Grigorieva (1994) provide indi-rect evidence that VNO function may be increas-ingly important after infancy (in juveniles or adoles-cents). Quantitative studies are needed to verify anage-related increase in receptor populations. Due tothe discontinuity of the sensory epithelium in tama-rins, quantitative data from immunohistochemicallylabeled VNO receptors may offer the best option forcomparing a broad range of species, rather thanVNO length or volume.

Although few adult tamarins were examined, itwas noteworthy that adult S. geoffroyi and L. chry-somelas were characterized by sensory epitheliumseparated by patches of nonsensory epithelium, asnoted for the majority of VNOs in juveniles. This isin strong contrast to the VNO morphology (homoge-neous sensory epithelium) in C. jacchus, and sug-gests that these taxa have truly distinct characterstates. Based on the literature to date, specieswithin the same genera (e.g., Saguinus, Table 3)apparently have differing character states. This em-phasizes the likelihood that functional or ontoge-netic factors may have a strong influence on VNOmorphology. Experimental evidence on other mam-mals suggests that endocrine factors can influencemorphology of the VNO and related structures (Se-govia and Guillamon, 1993). Among primates, lifestage, sex, and reproductive status could potentiallyinfluence VNO morphology and should be consid-ered in comparative studies wherever possible.

CONCLUSIONS

In a recent review, Smith et al. (2001a) groupedprimate VNOs into a series of five character states,


based on functional characteristics of the VNO epi-thelium (primarily receptor distribution and pres-ence/absence). However, one character state (“re-duced sensory epithelium”) was developed based ondescriptions of juvenile callitrichids. The results ofthe present study support the use of this characterstate to describe at least some adult callitrichids,but the diminished functional attributes implied by“reduced sensory epithelium” remain unverified andmay be inaccurate, based on the present study; abetter term to describe the morphology of the VNOseen in both tamarin species examined would be“interrupted sensory epithelium.” As the term “mi-crosmatic” is under some reconsideration in NewWorld monkeys (Laska et al., 2000), so should theidea that the VNO is relatively degenerative in cal-litrichid primates (Maier, 1980; Hunter et al., 1984).

The present study also indicates that there arechanges in the amount of sensory epitheliumthroughout postnatal life. Quantitative comparisonsof VNO receptor counts between callitrichids will berequired to determine the functional differences (ifany) between species that exhibit these contrastingmorphologies and to examine the interesting possi-bility that some species have more precocious VNOfunction than others. In future comparisons, exam-ination of different-aged specimens will be critical,since it appears that certain morphological charac-teristics may vary, based on ontogeny as it relates tomidfacial form and/or VNO function at a particularlife stage.

ACKNOWLEDGMENTS

The authors are grateful to J.H. Kinzinger forsectioning one of the specimens used in the study.We are grateful to A.B. Taylor and C. Vinyard, whoarranged our access to tissues from adult marmo-sets, and the faculty at the Department of Anatom-ical Sciences, State University of New York at StonyBrook, and especially C.F. Ross, for providing tissuefrom an adult lion tamarin. P.M. Mikkelsen of theAmerican Museum of Natural History graciouslymade her microscope imaging system available formicrographs of the embryos and fetuses of theBluntschli Collection. We also thank two anony-mous reviewers for constructive, helpful comments,and J.C. Dennis for helpful discussions on the topic.

LITERATURE CITED

Ackermann RR, Cheverud JM. 2002. Discerning evolutionaryprocesses in patterns of tamarin (genus Saguinus) craniofacialvariation. Am J Phys Anthropol 117:260–271.

Ankel-Simons F. 2000. Primate anatomy: an introduction. NewYork: Academic Press.

Aujard F. 1997. Effect of vomeronasal organ removal on malesocio-sexual responses to female in a prosimian primate (Mi-crocebus murinus). Physiol Behav 62:1003–1008.

Barret J, Abbot DH, George LM. 1993. Sensory cues and thesuppression of reproduction in subordinate female marmosetmonkeys, Callithrix jacchus. J Reprod Fertil 97:301–310.

Bhatnagar KP, Meisami E. 1998. Vomeronasal organ in bats andprimates: extremes of structural variability and its phyloge-netic implications. Microsc Res Techn 43:465–475.

Bhatnagar KP, Smith TD. 2001. The human vomeronasal organ.Part III: postnatal development from infancy through the ninthdecade. J Anat 199:289–302.

Breipohl W, Bhatnagar KP, Mendoza A. 1979. Fine structure ofthe receptor-free epithelium in the vomeronasal organ of therat. Cell Tissue Res 200:383–395.

Coppola DM, Millar LC. 1994. Stimulus access to the accessoryolfactory system in the prenatal and perinatal rat. Neuro-science 60:463–468.

Coppola DM, Budde J, Millar L. 1993. The vomeronasal duct hasa protracted postnatal development in the mouse. J Morph218:59–64.

Dawley EM. 1998. Species, sex, and seasonal differences in VNOsize. Microsc Res Techn 41:506–518.

Dixson AF. 1993. Callitrichid mating systems: laboratory andfield approaches to studies of monogamy and polyandry. In:Rylands AB, editor. Marmosets and tamarins: systematics, be-haviour and ecology. Oxford: Oxford University Press. p 221–227.

Dixson AF. 1998. Primate sexuality: comparative studies of theprosimians, monkeys, apes, and human beings. Oxford: OxfordUniversity Press.

Dorries KM, Adkins-Regan E, Halpern BP. 1997. Sensitivity andbehavioral responses to the pheromone androstenone are notmediated by the vomeronasal organ in domestic pigs. BrainBehav Evol 49:53–62.

Epple G. 1981. Effects of prepubertal castration on the develop-ment of the scent glands, scent marking and gonadal state inthe saddle-back tamarin (Saguinus fuscicollis) (Callitrichidae,Primates). Horm Behav 15:54–67.

Evans CS, Grigorieva EF. 1994. Morphology of the vomeronasalorgan in two South American monkeys (Saguinus labiatus andCebuella pygmaea, Callitrichidae): histology and lectin histo-chemistry. Adv Biosci 93:31–42.

Giere P, Freyer C, Zeller U. 1999. Opening of the mammalianvomeronasal organ with respect to the Glires hypothesis: acladistic reconstruction of the Therian morphotype. Mitt MusNaturk Berl Zool Reihe 75:247–255.

Hoage RJ. 1982. Social and physical maturation in captive liontamarins, Leontopithecus rosalia rosalia (Primates: Callitrichi-dae). Smithson Contr Zool 354:1–56.

Hunter AJ, Fleming D, Dixson AF. 1984. The structure of thevomeronasal organ and nasopalatine ducts in Aotus trivirgatusand some other primate species. J Anat 138:217–225.

Kelliher KR, Baum MJ, Meredith M. 2001. The ferret’s vomero-nasal organ and accessory olfactory bulb: effect of hormonemanipulation in adult males and females. Anat Rec 263:280–288.

Keverne EB. 1999. The vomeronasal organ. Science 286:716–720.Laska M, Seibt A, Weber A. 2000. “Microsmatic” primates revis-

ited: olfactory sensitivity in the squirrel monkey. Chem Senses25:47–53.

Loo SK, Kanagasuntheram R. 1972. The vomeronasal organ intree shrew and slow loris. J Anat 112:165–172.

Maier W. 1980. Nasal structures in Old and New World primates.In: Ciochon RL, Chiarelli AB, editors. Evolutionary biology ofthe New World monkeys and continental drift. New York: Ple-num Press. p 219–241.

Maier W. 2000. Ontogeny of the nasal capsule in cercopithecoids:a contribution to the comparative and evolutionary morphologyof catarrhines. In: Whitehead PF, Jolly CJ, editors. Old Worldmonkeys. Cambridge: Cambridge University Press. p 99–132.

Mendoza AS, Kuderling I, Kuhn HJ, Kuhnel W. 1994. The vome-ronasal organ of the New World monkey Saguinus fuscicollis(Callitrichidae). A light and transmission electron microscopestudy. Ann Anat 176:217–222.

Moynihan M. 1970. Some behavior patterns of platyrrhine mon-keys II. Saguinus geoffroyi and some other tamarins. SmithsonContr Zool 28:1–77.

Roslinski DL, Bhatnagar KP, Burrows AM, Smith TD. 2000.Comparative morphology and histochemistry of glands associ-ated with the vomeronasal organ in humans, mouse lemurs,and voles. Anat Rec 260:92–101.


Sanchez-Villagra MR. 2001. Ontogenetic and phylogenetic trans-formations of the vomeronasal complex and nasal floor ele-ments in marsupial mammals. Zool J Linn Soc 131:459–479.

Schilling A, Serviere J, Gendrot G, Perret M. 1990. Vomeronasalactivation by urine in the primate Microcebus murinus: anexperimental 2DG study. Exp Brain Res 81:609–618.

Segovia S, Guillamon A. 1993. Sexual dimorphism in the vomer-onasal pathway and sex differences in reproductive behaviors.Brain Res Rev 18:51–74.

Smith TD, Bhatnagar KP. 2000. The human vomeronasal organ.Part II: prenatal development. J Anat 197:421–436.

Smith TD, Siegel MI, Burrows AM, Mooney MP, Burdi AR, Fab-rizio PA, Clemente FR. 1998. Searching for the vomeronasalorgan of adult humans: preliminary findings on location, struc-ture, and size. Microsc Res Techn 41:483–491.

Smith TD, Siegel MI, Bhatnagar KP. 2001a. A reappraisal ofontogeny, morphology, functionality, and persisting questionson the vomeronasal system of catarrhine primates. Anat Rec(New Anat) 265:176–192.

Smith TD, Siegel MI, Bonar CJ, Bhatnagar KP, Mooney MP,Burrows AM, Smith MA, Maico LM. 2001b. The existence of thevomeronasal organ in postnatal chimpanzees and evidence forits homology to that of humans. J Anat 198:77–82.

Smith TD, Buttery TA, Bhatnagar KP, Burrows AM, Mooney MP,Siegel MI. 2001c. Anatomical position of the vomeronasal organin postnatal humans. Ann Anat 183:475–479.

Smith TD, Mooney MP, Burrows AM, Bhatnagar KP, Siegel MI.2001d. Prenatal growth and adult size of the vomeronasal organin mouse lemurs and humans. In: Marchlewska-Koj A, Lepri JJ,Muller-Schwarze D, editors. Chemical signals in vertebrates IX.New York: Kluwer Academic/Plenum Publishers. p 93–99.

Smith TD, Bhatnagar KP, Shimp KL, Kinzinger JH, Bonar CJ,Burrows AM, Mooney MP, Siegel MI. 2002. Histological def-inition of the vomeronasal organ in humans and chimpan-zees, with a comparison to other primates. Anat Rec 267:166 –176.

Stephan H, Frahm H, Baron G. 1981. New and revised data onvolume of brain structures in insectivores and primates. FoliaPrimatol (Basel) 35:1–29.

Taniguchi K, Matsusaki Y, Ogawa K, Saito TR. 1992. Fine struc-ture of the vomeronasal organ in the common marmoset (Cal-lithrix jacchus). Folia Primatol (Basel) 59:169–176.

Wohrmann-Repenning A. 1993. The anatomy of the vomeronasalcomplex of the fox (Vulpes vulpes (L.)). Under phylogenetic andfunctional aspects. Zool Jb Anat 123:353–361.

Wohrmann-Repenning A, Barth-Mueller U. 1994. Functionalanatomy of the vomeronasal complex in the embryonic devel-opment of the pig (Sus-scrofa dom.). Acta Theriol 39:323.

Wohrmann-Repenning A, Bergmann M. 2001. The vomeronasalcomplex in strepsirhine primates and Tarsius. Mamm Biol66:257–258.

Wysocki CJ. 1979. Neurobehavioral evidence for the involvementof the vomeronasal system in mammalian reproduction. Neu-rosci Biobehav Rev 3:301–341.

Yamamoto ME. 1993. From dependence to sexual maturity: thebehavioral ontogeny of callitrichidae. In: Rylands AB, editor.Marmosets and tamarins: systematics, behaviour and ecology.Oxford: Oxford University Press. p 235–254.

Yoshida J, Kimura J, Tsukise A, Okano M. 1993. Developmentalstudy on the vomeronasal organ in the rat fetus. J Reprod Dev39:47–54.


ontogenetic characteristics of the vomeronasal organ in saguinus

Documents