Anim Cogn (2005) 8: 182–189DOI 10.1007/s10071-004-0244-9

ORIGINAL ARTICLE

Michael H. Ferkin · Andrew A. Pierce ·Robert O. Sealand · Javier delBarco-Trillo

Meadow voles, Microtus pennsylvanicus, can distinguish moreover-marks from fewer over-marks

Received: 1 July 2004 / Revised: 1 November 2004 / Accepted: 9 November 2004 / Published online: 3 December 2004C! Springer-Verlag 2005

Abstract Is it possible that voles have a sense of number?To address this question, we determined whether voles dis-criminate between two different scent-marking individualsand identify the individual whose scent marks was on topmore often than the other individual. We tested whethervoles show a preference for the individual whose scentmarks was on top most often. If so, the simplest explanationwas that voles can make a relative size judgement—suchas distinguishing an area containing more of one individ-ual’s over-marks as compared to less of another individual’sover-marks. We found that voles respond preferentially tothe donor that provided a greater number of over-marksas compared to the donor that provided a lesser numberof over-marks. Thus, we concluded that voles might dis-play the capacity for relative numerousness. Interestingly,female voles were better able than male voles to distin-guish between small differences in the relative number ofover-marks by the two scent donors.

Keywords Relative numerousness . Voles . Over-marks

Introduction

Most terrestrial mammals deposit scent marks alongwell-traversed paths and on prominent objects in an area,where they may be over-marked by conspecifics (Thiessenand Rice 1976; Gosling and Roberts 2001). Several studieshave shown that individuals can discriminate betweenthe donors of the top and bottom scents of an over-mark(Johnston et al. 1995; Johnston 1999, 2003; Kohli andFerkin 1999; Cohen et al. 2001). Briefly, Golden hamsters,Mesocricetus auratus, and meadow voles, Microtus

M. H. Ferkin (!) · A. A. Pierce · R. O. Sealand ·J. delBarco-TrilloDepartment of Biology, The University of Memphis,Ellington Hall,Memphis, TN 38152, USAe-mail: mhferkin@memphis.eduTel.: +901-678-3509Fax: +901-678-4746

pennsylvanicus spend more time investigating the markof the top-scent donor than that of the bottom-scent donorwhen the two marks are offered separately (Johnston et al.1997a, b; Ferkin 1999; Woodward et al. 2000). Voles andhamsters can discriminate and respond selectively to thetop-scent donor if the identity of the top- and bottom-scentdonors does not change.

In free-living populations of voles, however, the identityof the top- and bottom-scent donors may vary, especially ifthe two scent donors are in close proximity to one another.Thus, at any given time in a particular area, an individualmay have some of its marks on top of those depositedby another individual or below those deposited by thatindividual. It is not known whether voles can discriminatebetween these two different scent-marking individuals andidentify the individual whose scent marks was on top moreoften than the other individual. Is it possible that voleshave a sense of number? If so, the simplest explanationis that voles can make a relative size judgement—such asdistinguishing an area containing more of one individual’sover-marks as compared to less of another individual’s ofover-marks. If voles can identify the more frequent top-scent donor from the less frequent top-scent donor andrespond accordingly, it may be an indication that they havethe capacity for relative numerousness (Davis and Perusse1988; Gallistel 1990). An important feature of this abilityis that animals may not have literally counted the numberof objects in the two groups (Davis 1993).

One way to assess if animals display numerousness is toexpose them simultaneously to a set with relatively largenumbers of stimuli for one response and a set with a rel-atively small number of stimuli for another response, andthen give them a test to determine if they respond differ-ently to either set or if they respond selectively to the largerset (Boysen and Capaldi 1993; Shettleworth 1998; Hauser2000). If animals respond selectively to the larger set overthe smaller set, that would be an indication that they mayhave the ability to spontaneously discriminate betweentwo different numerosities and have a concept of relativenumerousness (reviewed by Davis and Perusse 1988;Boysen 1997; Shettleworth 1998; Hauser et al. 2003).

183

Several studies have shown that a variety of animals indifferent taxa have the ability of discriminating a group orentity containing more objects from one containing less ofthe same objects (Beran 2001; Machado and Keen 2003;Kilian et al. 2003; Lipton and Spelke 2003; Uller et al.2003).

The objective of this study was to determine if voleswere able to distinguish between areas that contain differentnumbers of over-marks from two scent donors, suggesting acapacity for relative numerousness. In this study, we testedfor relative numerousness by allowing voles to explore anarea that contained a set of over-marks by one individualand a smaller or larger set of over-marks by another individ-ual. If voles displayed the capacity for distinguishing morefrom less, they would respond preferentially to the donorwho was the top-scent donor on more of the over-marks ascompared to the donor who was the bottom-scent on mostof the over-marks.

Methods

Animals

Female and male meadow voles were used in the follow-ing experiments. In each experiment, we used animals bornand raised under a long photoperiod (14L:10D, lights on at0800 hours, CST). This photoperiod simulates a day lengthtypical of the breeding season. Animals were first, second,and third generation offspring of field-caught animals cap-tured in Ohio and Kentucky, U.S.A. All voles used in thisexperiment were weaned at 21 days of age, housed withlittermates until 42 days of age, and then housed singly inclear plastic cages measuring 27.0"16.5"12.5 cm. Thesecages contained wood chip bedding, cotton nesting ma-terial, and ad libitum food (Laboratory Rodent Diet no.5008, PMI, St. Louis, Mo., U.S.A.) and water. Cages werecleaned once a week and the cotton nesting material wasreplaced every 2 weeks.

Subjects and scent donors

All voles used in these experiments were between 70 and120 days of age, sexually mature, but sexually inexperi-enced. Subject animals were 70 male and 70 female voles.Scent donors were an additional 40 male and 40 femalevoles. Female meadow voles are induced ovulators and donot undergo estrous cycles (Milligan 1982; Keller 1985).During the breeding season, female voles may be pregnantand lactating, pregnant, lactating, or not pregnant or lactat-ing. Long-photoperiod female voles that are not pregnantor lactating and long-photoperiod male voles, are sexuallyreceptive (Ferkin et al. 2004a). In the present study, weused voles that were virgins, and sexually receptive. Toeliminate litter effects, we used no more than two individ-uals from the same litter as either subjects or scent donors.Subjects and scent donors were unfamiliar and unrelated toone another.

Pre-exposure phase: placing and simulating multipleover-marks

We created multiple over-marks of two different same-sexconspecifics in a T-shaped arena that simulates a runwaythat a vole may encounter in the field (Ferkin et al. 2001,2004a, b). Depending on the experiment, we either placedsix, seven, or eight evenly spaced, simulated over-marks,each measuring 1.5 cm in diameter on specific locationson the filter paper substrate on the horizontal portion ofthe T-shaped arena. Over-marks were placed evenly onthe substrate with 5 cm separating contiguous over-marks(Fig. 1). Thus, for each experiment we were able to controlfor both the size of the over-marks and the total surfacearea they covered.

Over-marks were created by the investigators by rubbingfresh feces from the cage of the first scent donor (bottom-scent vole) on a specified portion of paper substrate in theT-shaped arena and allowing it to dry for 5 min, then rub-bing fresh feces from the cage of the second scent donor(top-scent vole) directly on top of it such that it made an“+” shape (Ferkin et al. 1999). Each scent mark was cre-ated by rubbing the feces against a plastic template withan opening 1.5 cm long and 0.5 cm wide that was placedin the desired location on the substrate. Thus, we wereable to control for the size of each scent mark and over-mark. The template was cleaned thoroughly with a solu-tion of alcohol and warm soapy water and dried betweenapplications. We chose to use feces in this series of ex-periments to eliminate our handling of the scent donors.However, anogenital area scents and urine may also beused as sources of scents for over-marks (Ferkin et al.2004a). We used black ink to trace the outline of the fe-ces marks of the first scent donor (bottom-scent vole), andred ink to trace the outline of the second scent donor (top-scent vole). Previous tests indicated that the presence orcolor of the ink used in tracing does not affect the behav-ior of voles investigating over-marks (Ferkin et al. 1999,2004a, b).

Exposure to multiple over-marks

Each subject was tested once, each with a unique pair of top-and bottom-scent donors. All pairs of donors were similarin weight (within 5 g), and unfamiliar and unrelated to eachother and to the subjects. We randomly assigned the top-and bottom-scent donors.

Subjects were placed into the stem of the T-shaped arena,behind an opaque divider 10 min after the last over-markwas placed in the horizontal section of the maze (see above).After the subject was placed in the stem of the T-shapedarena for 2 min, the divider was raised and the subject wasallowed 10 min to explore the arena. All the subjects usedin this study explored the entire horizontal portion of thearena.

184

Fig. 1 An example of theover-mark scheme for a test ofthe 3 versus 4 testing condition.Over-mark placement wasassigned randomly, A/B isrepresented four times and B/Ais represented three times. In thecontrol conditions, twomatching rows of three or fourover-marks were used and A/Bwas used an equal number oftimes as B/A

T-shaped arena

This arena was used in previous studies of scent markingand over-marking in voles (Ferkin et al. 2001, 2004a, b).The sides of the arena were constructed of opaque greenacrylic plastic. Each arm and the stem of the arena were25"13"16 cm (length " width " height, respectively).The experimenter could block entry from the stem into thehorizontal portion of the arena by placing a solid, opaquepartition (made of the same material as the arena) acrossthe stem. A large sheet of white photocopy paper served asthe substrate of the arena. The voles were unfamiliar withthe arena at the time of testing.

Experimental condition 1

In this experiment, 10 male and 10 female subjects wereplaced in arenas that contained seven over-marks in whichthe top- and bottom position of the two scent donors’ marksin the over-mark remained the same. That is, donor A’s scentmarks over-marked each of donor B’s seven scent marks.This represented a situation of seven over-marks by donor Aand zero over-marks by donor B. The identity of the volesthat were designated donor A and donor B was random,except that they were the opposite-sex of the subjects andmet the criteria for odor donors listed above. Each subjectwas exposed to areas containing the scent marks of a uniquepair of odor donors in this and in subsequent experiments.

Experimental condition 2

In this experiment, 10 male and 10 female subjects wereplaced in arenas that contained seven over-marks in which

the top- and bottom position of the two scent donors’ marksin the over-mark varied as follows. Donor A’s scent marksover-marked six of donor B’s seven scent marks and one ofdonor B’s scent marks over-marked one of donor A’s sevenscent marks (six over-marks by donor A vs one over-markby donor B). The particular scent marks that were over-marked by donors A and B were determined randomly inthis and the subsequent experiments.

Experimental condition 3

In this experiment, 10 male and 10 female subjects wereplaced in arenas that contained seven over-marks in whichdonor A’s scent marks over-marked five of donor B’s sevenscent marks and two of donor B’s scent marks over-markedtwo of donor A’s seven scent marks (five over-marks bydonor A vs two over-marks by donor B).

Experimental condition 4

In this experiment, 10 male and 10 female subjects wereplaced in arenas that contained seven over-marks in whichdonor A’s scent marks over-marked four of donor B’s anddonor B over-marked three of donor A’s seven scent marks(four over-marks by donor A vs three over-marks by donorB; Fig. 1).

Experimental condition 5

In this experiment, 20 male and 20 female subjects wereplaced in arenas that contained an equal number of over-marks in which donor A was the top-scent donor and donor

185

B was the top-scent donor. These two control conditionsrepresented a situation of three over-marks by donor Aversus three over-marks by donor B and four over-marksby donor A versus four over-marks by donor B. There were10 different male and 10 different female subjects used inthe three and four over-mark tests, respectively.

Post exposure phase: odor preference test

Immediately after exploring the T-shaped arena, subjectswere returned to their home cage. Fifteen minutes after re-turning to their home cage, each subject underwent a single5-min attractivity test; this test has been used previously involes (Ferkin and Johnston 1995). Briefly, voles were pre-sented with a clean glass microscope slide (2.5"7.6 cm)that contained the fecal scent marks from two opposite-sexconspecifics. Each slide was divided into three equal sec-tions (each 2.5 cm long). One end section contained a scentmark from one donor. The other end section contained ascent mark from the other donor. The middle section con-tained no stimulus odor. The glass slide was suspended bya wire hook and a clasp 1 cm above the substrate in thesubject’s home cage, against the wall opposite the animal’snest. During each 5-min trial, we recorded continuously thetime each subject investigated the two scented sections ofthe slide, and the middle section of the slide. Criteria forinvestigation of a mark were that: (1) a vole was obviouslylicking or sniffing a stimulus odor or its nose was withinapproximately 1 cm of one end of the slide, (2) it investi-gated the two scented areas on the slide, and (3) it spentmore time investigating the two scented areas of the slidethan the clean middle section (Ferkin and Johnston 1995).

The preference test began when the slide was placed intothe cage of the subject. Each slide was used in only one trialand discarded. We randomly placed the stimulus odors onthe left or right side of the slide. We used paired t-tests todetermine if significant differences existed in the amountof time subjects in each experiment spent investigating theodor of the two scent donors. We used a Binomial test todetermine if significant differences existed in the number ofsubjects that spent more time investigating the odor of thetop-scent donor to that of the bottom-scent donor. For bothtests, significant differences were accepted at !=0.05. Ifsubjects spent significantly more time investigating a markfrom either of the two scent donors, we considered the moreinvestigated mark as being preferred (Ferkin and Johnston1995).

Collection of stimulus odors for the preference test

Fresh fecal scent marks were obtained for each trial fromeach scent donor. We collected scent marks by rubbing afresh fecal bolus against a clean glass microscope slide for3–5 s. The experimenter wore disposable latex gloves tominimize human scent transfer while handling all slides.Feces contain sexually distinct odor cues for meadow voles(Ferkin and Johnston 1995).

Results

Meadow voles discriminated between the top-scent donorand bottom-scent donor of multiple over-marks. Thus,voles showed a capacity for relative numerousness. Inter-estingly, sex differences that were biased towards femalesexisted in the level of numerousness displayed by voles.

Experimental condition 1

Male and female voles exposed to seven over-marks bydonor A and to zero over-marks by donor B later spentsignificantly more time in the preference test investigat-ing the odor of donor A as compared to that of donor B(for males, t=3.01, df=9, P<0.020; for females, t=2.64,df=9, P<0.025). That is, males and females spent moretime investigating the odor of the donor who was the top-scent donor of the over-marks (Fig. 2 top). In addition, 9 of10 male subjects and 9 of 10 female subjects spent moretime investigating the odor of the top-scent donor as com-pared to that of the bottom-scent donor (both males andfemales, binomial tests P=0.021).

Fig. 2 Mean (+SEM) amount of time(s) that female and male volesexposed to (top panel) seven over-marks by donor A and to zeroover-marks by donor B and (bottom panel) six over-marks by donorA and to one over-mark by donor B, later spent investigating the odorof these two scent donors

186

Experimental condition 2

Male and female voles exposed to six over-marks by donorA and to one over-mark by donor B later spent signifi-cantly more time investigating the odor of donor A thanthat of donor B (for males, t=2.8, df=9, P<0.020; for fe-males, t=3.99, df=9, P<0.003). Both males and femalesspent more time investigating the odor of the donor whosescent mark was in the top position of an over-mark most fre-quently (Fig. 2 bottom). Additionally, 9 of 10 male subjectsand 9 of 10 female subjects spent more time investigatingthe odor of the top-scent donor as compared to that of thebottom-scent donor (both males and females, binomial testsP=0.021).

Experimental condition 3

Female voles exposed to five over-marks by donor A and totwo over-marks by donor B later spent significantly moretime investigating the odor of donor A as compared to thatof donor B (t=4.18, df=9, P<0.003). In addition, 9 of 10female subjects spent more time investigating the odor ofthe top-scent donor as compared to that of the bottom-scentdonor (binomial test P=0.021). In contrast, males exposedto five over-marks by donor A and two over-marks by donorB later spent similar amounts of time investigating the odorof donor A and that of donor B (t=0.56, df=9, P>0.1).In addition, only 6 of 10 male subjects spent more timeinvestigating the odor of the top-scent donor as comparedto that of the bottom-scent donor (binomial test P=0.754).Thus, only females spent more time investigating the odorof the donor whose scent mark was in the top position ofan over-mark most frequently (Fig. 3 top).

Experimental condition 4

Female but not male voles exposed to four over-marks bydonor A and three over-marks by donor B spent signifi-cantly more time in the preference test investigating theodor of donor A as compared to that of donor B (for males,t=1.77, df=9, P>0.1; for females, t=3.78, df=9, P<0.04).Only females spent more time investigating the odor of thedonor whose scent mark was in the top position of an over-mark most frequently (Fig. 3 bottom). In addition, 9 of 10female subjects spent more time investigating the odor ofthe top-scent donor than that of the bottom-scent donor (bi-nomial test P=0.021), whereas 5 of 10 male subjects spentmore time investigating the odor of the top-scent donor thanthat of the bottom-scent donor (binomial test P=1.00).

Experimental condition 5

Male and female voles exposed to three over-marks bydonor A and three over-marks by donor B (Fig. 4 top) orfour over-marks by donor A and four over-marks by donorB (Fig. 4 bottom) later spent similar amounts of time in-

Fig. 3 Mean (+SEM) amount of time(s) that female and male volesexposed to (top panel) five over-marks by donor A and to two over-marks by donor B and (bottom panel) four over-marks by donor Aand to three over-marks by donor B, later spent investigating the odorof these two scent donors

vestigating the odor of donor A as compared to that ofdonor B (for males in 3 vs 3, t=1.47, df=9, P>0.1; for fe-males in 3 vs 3, t=0.034, df=9, P>0.1; for males in 4 vs 4,t=1.27, df=9, P>0.1; for females in 4 vs 4, t=0.78, df=9,P>0.1). Thus, females and males spent similar amountsof time investigating odors of the two donors when theyover-marked the same number of marks relative to one an-other (Fig. 4 top and bottom). Additionally, 5 of 10 femalesubjects spent more time investigating the odor of the top-scent donor than that of the bottom-scent donor in both the3 versus 3 situation and 4 versus 4 situation (both situations,binomial tests P=1.00). In the 3 versus 3 situation, 5 of 10male subjects spent more time investigating the odor of thetop-scent donor than that of the bottom-scent donor (bino-mial test P=1.00), whereas in the 4 versus 4 situation, 6of 10 male subjects spent more time investigating the odorof the top-scent donor than that of the bottom-scent donor(binomial test P=0.754).

Discussion

In this study, we determined if voles could discriminatebetween the top- and bottom-scent donors and respond

187

Fig. 4 Mean (+SEM) amount of time(s) that female and male volesexposed to (top panel) four over-marks by donor A and to four over-marks by donor B and (bottom panel) three over-marks by donor Aand to three over-marks by donor B, later spent investigating the odorof these two scent donors

accordingly in areas containing multiple over-marks inwhich the identity of the top- and bottom-scent donors var-ied. Female and male voles spent more time investigatingthe odor of the more frequent top-scent donor as comparedto the less frequent top-scent donor after they exploredan area in which the top- and bottom scent donors variedwith respect to one another. Previous work has shown thatvoles spend more time investigating the top-scent donor ascompared to the bottom-scent donor when they encountermultiple over-marks in which the top and bottom posi-tion of the donors does not change (Johnston et al. 1997a;Ferkin 1999; Woodward et al. 2000). The present resultsextend this finding, showing that in a given area, voles candiscriminate between donors that deposit more over-marksas compared to those that deposit fewer over-marks. Suchdiscriminations between more than and less than may beall that voles, like other animals need to be able to performin nature (Davis 1993).

Does spontaneous discrimination of more over-marksfrom fewer over-marks indicate that voles have a capacityfor relative numerousness? It could be argued thatspontaneous discrimination of more over-marks relative tofewer over-marks by voles may be based on non-numericalfeatures. For example, it is possible that the numerical

attributes of the presentation of the over-marks co-varywith other features such as their density, surface area, size,and hedonistic value (Davis and Perusse 1988; Honig andStewart 1989; Brannon and Terrace 2000). This wouldsuggest that voles have not exercised numerousness,per se. In our study, we attempted to control for thesenon-numerical factors. We used donors in each of ourcomparisons that were of similar hedonistic value relativeto one another; both donors were sexually receptive andlike-sex conspecifics. In addition, although we presentedover-marks of both donors simultaneously to investigatingvoles, these marks were similar in size, surface area thatthey covered, and density.

It could also be argued that in our study voles are re-sponding to differences in the intensity of the compoundsin the scent marks and the volume of the scent marks. Al-though this is possible, we think it unlikely. Previous workhas shown that voles spend more time investigating the top-scent mark as compared to that of the bottom-scent mark,independent of differences in the amount and surface areathat the top-scent mark overlaps the bottom-scent mark.Voles do not differ in the amount of time that they spendinvestigating different amounts of the same odor from thesame conspecific, suggesting that they may not be respond-ing to more dispersed density of odors (Ferkin and Johnston1995; Ferkin unpublished data).

It is also possible that voles prefer the odor of the mostfrequent top-scent donor to the least frequent top-scentdonor because they were exposed to the former odors moreoften and became more familiar with them. This wouldbe the case when voles were exposed to conditions inwhich one donor was the top-scent donor more frequentlythan the second donor. However, this explanation is notconsistent with findings from a previous study on voles(Ferkin et al. 1999). In that study, voles were exposed toan over-mark in which the bottom-scent mark covered anentire glass microscope slide and the top-scent was placedin the center of the slide, where it overlapped only 5% ofthe bottom-scent mark. In the subsequent odor preferencetest, voles responded preferentially to the top-scent donor’smark (Ferkin et al. 1999). Voles did not respond preferen-tially to the bottom-scent mark, although it was present ingreater quantity, and potentially more familiar to them, ascompared to the top-scent mark. Such findings suggest thatvoles exposed to over-marks are discriminating between thescent marks of the two donors by their relative position toone another in an over-mark and not by relative differencesin the intensity, volume, or familiarity of the marks.

Although we cannot completely rule out all the non-numerical factors or control for the nature and extent of thechemical information that may affect the discriminationof the top- and bottom-scent donors in an over-mark, wespeculate that the present data may be in line with thenotion that voles may spontaneously discriminate areas thatcontain more over-marks by one individual as compared tothose of another individual. Our findings suggest that volesmay have the capacity for relative numerousness, whichadds them to the growing list of other animals, includinghumans that do so (Beran 2001; Machado and Keen 2003;

188

Kilian et al. 2003; Lipton and Spelke 2003; Uller et al.2003). If this speculation is supported, our results may beconsistent with the view that the mechanisms underlyingrelative numerousness may be shared across different taxa(Boysen and Capaldi 1993; Boysen 1997; Hauser 2000).

We found that female voles were better than males inspontaneous discrimination of more frequent versus lessfrequent top-scent donors. Females were capable of spon-taneously discriminating between areas in which donor A’smarks were on top of donor B’s marks four times relativeto the three times donor B’s marks were on top of donor A’smarks. Males could make a similar discrimination if donorA’s marks were on top of donor B’s marks six times relativeto the one time donor B’s mark was on top of one of donorA’s marks. This type of asymmetry in relative numerous-ness by voles is interesting in that the literature is repletewith studies suggesting that sex differences exist in math-ematical skills that favor males over females, particularlyin primates and humans (reviewed in Geary 1996). Oneargument is that males have better spatial and navigationalabilities than females, which provides them with a greaterneed and capacity to solve problems in geometry and othermathematics-related activities (Geary 1996). The fact thatmale meadow voles have better spatial ability as comparedto female meadow voles (Gaulin et al. 1990), but the formerhas poorer prenumerical ability than the latter is interest-ing (this study). Our findings suggest that sex differencesin distinguishing more over-marks from fewer over-marksare somehow separate from the spatial ability of meadowvoles. Moreover, for voles it appears that spatial abilitymay be a poor predictor of relative numerousness, and viceversa. Female biases in relative numerousness may be thedevelopmental link for more complex numerical processes,such as subitizing, estimation, and counting, and arithmeti-cal reasoning in this species (i.e., Gallistel 1990; Boysenand Capaldi 1993; Hauser 2000; Hauser et al. 2003).

If female voles are more sensitive than male voles indistinguishing between more and less over-marks, howmight this attribute be associated with their socio-sexualbehavior? Previous work has shown that asymmetries ininvestigation time for a particular scent donor reflect anindividual’s preference for that donor relative to anotherdonor (Ferkin and Johnston 1995), and that this preferencemay reflect a preference for a potential mate (Woodwardet al. 2000). An individual’s preference for the top-scentdonor is also concomitant with a selective memory for thatdonor (Johnston et al. 1997a, b; Ferkin et al. 1999). Femalevoles may have a better selective memory for traits orsets of traits that differ in relative numerousness. Femalemeadow voles, similar to females in other species thatexert female choice for males, may be predisposed or havea sensory bias allowing them to distinguish between smallasymmetries among males (Ryan 1997). This may allowfemales to distinguish between males, and identify thosemales that are either currently nearby or socially dominant.Recency in areas and dominance status are both features ofthe top-scent donor (Rozenfeld et al. 1987; Rich and Hurst1999), and attributes of males that sire offspring (Boonstraet al. 1993; Ferkin 1999). In that male voles wander

through large home ranges that encompass the territoriesof one or more females (Madison 1980), resident femalesmay be able to select potential mates based on asymmetriesin the number of times that a particular male was the top-or bottom-scent donor in their territory. Females may needto distinguish between and remember the individuals thatmay enter their territories in order to keep track of them.In contrast, males are not territorial (Madison 1980) andmay not need to respond to over-marks, as do female voles(Ferkin et al. 2001, 2004a, b). Thus, female voles may havea better “memory for scent marks” relative to male voles.Alternatively, females voles may be more responsive thanmale voles to asymmetries in the relative number of top andbottom-scent marks in an area because the top-scent markconveys greater or more salient information to femalesrelative to that of the bottom-scent marks (Johnston et al.1997a; Ferkin et al. 1999; Woodward et al. 2000).

It is not clear why male voles cannot distinguish smallasymmetries in the relative number of over-marks of fe-male conspecifics. Perhaps males do not frequently enterareas marked by multiple adult females. Alternatively, malevoles may gain similar benefits if they encounter the resi-dent female or a female that was passing through anotherfemale’s territory. Males may only need to distinguish be-tween females if there are large asymmetries in the relativenumber of over-marks between them. Larger numbers ofover-marks by one female relative to another female mayindicate that the former is the current resident (Ferkin 1999;Rich and Hurst 1999; Ferkin et al. 2001).

Acknowledgements We thank Augusta Okaputo, WilliamWilliams, and Hong Li for technical assistance. Funding wasprovided by the NIH BRIDGE program at The University ofMemphis and NIH grant AG 16954-01 to M.H.F. and NIH grant tothe Tennessee Mouse Mutagenesis Consortium.

References

Beran MJ (2001) Summation and numerousness judgements ofsequentially presented sets of items by chimpanzees (Pantroglodytes). J Comp Psychol 115:181–191

Boonstra R, Xia X, Pavone L (1993) Mating system of the meadowvole, Microtus pennsylvanicus. Behav Ecol 4:83–89

Boysen ST (1997) Representation of quantities by apes. Adv StudBehav 26:435–462

Boysen ST, Capaldi EJ (1993) The development of numerical com-petence: animal and human models. Erlbaum, Hillsdale, N.J.

Brannon EM, Terrace HS (2000) Representation of the numerosities1–9 by rhesus macaques (Macaca mulatta). J Exp PsycholAnim Behav Process 26:31–49

Cohen AB, Johnston RE, Kwon A (2001) How golden hamsters(Mesocricetus auratus) discriminate top from bottom flankscents in over-marks. J Comp Psychol 115:241–247

Davis H (1993) Numerical competence in animals: life beyondclever Hans. In: Boysen ST, Capaldi EJ (eds) The developmentof numerical competence. Erlbaum, Hillsdale, N.J., pp109–125

Davis H, Perusse R (1988) Numerical competence in animals:definitional issues, current evidence, and a new researchagenda. Behav Brain Sci 11:561–615

Ferkin MH (1999) Scent over-marking as a competitive form ofchemical communication in voles. In: Johnston RE, Muller-Schwarze D, Sorenson P (eds) Advances in chemical signals invertebrates, number 8. Plenum, New York, pp 239–246

189

Ferkin MH, Johnston RE (1995) Meadow voles, Microtuspennsylvanicus, use multiple sources of scent for sex recogni-tion. Anim Behav 49:37–44

Ferkin MH, Dunsavage J, Johnston RE (1999) What kind ofinformation do meadow voles, Microtus pennsylvanicus, useto distinguish between the odors of the top and bottom-scentdonors of an over-mark? J Comp Psychol 113:43–51

Ferkin MH, Lee DN, Leonard ST (2004a) The reproductive stateof female voles affects their scent marking behavior andthe responses of male conspecifics to such marks. Ethology110:257–272

Ferkin MH, Li HZ, Leonard ST (2004b) Meadow voles and prairievoles differ in the percentage of conspecific marks that theyover-mark. Acta Ethol (in press)

Ferkin MH, Mech SG, Paz-y-Mino CG (2001) Scent marking inmeadow voles and prairie voles: a test of three hypotheses.Behaviour 138:1319–1336

Gallistel CR (1990) The organization of learning. MIT Press,Cambridge, Mass.

Gaulin SJ, Fitzgerald RW, Wartell MS (1990) Sex differences in spa-tial ability and activity in two vole species (Microtus ochrogasterand M. pennsylvanicus). J Comp Psychol 104:83–93

Geary DC (1996) Sexual selection and sex differences in mathemat-ical abilities. Behav Brain Sci 19:229–284

Gosling LM, Roberts SC (2001) Scent marking in male mammals:cheat-proof signals to competitors and mates. Adv Stud Behav30:169–217

Hauser MD (2000) What do animals think about numbers? Am Sci88:144–151

Hauser MD, Tsao F, Garcia P, Spelke ES (2003) Evolutionary foun-dations of number: spontaneous representation of numericalmagnitudes by cotton-top tamarins. Proc R Lond B Biol Sci270:1441–1446

Honig WK, Stewart K (1989) Discrimination of relative numerosityin pigeons. Anim Learn Behav 21:346–354

Johnston RE (2003) Chemical communication in rodents: frompheromones to individual recognition. J Mamm 84:1141–1162

Johnston RE (1999) Scent over-marking: how do hamsters knowwhose scent is on top and why should it matter? In: JohnstonRE, Muller-Schwarze D, Sorenson P (eds) Plenum, New York,pp 227–238

Johnston RE, Munver R., Tung C (1995) Scent counter marks:selective memory for the top scent by golden hamsters. AnimBehav 49:1435–1442

Johnston ER, Sorokin ES, Ferkin MH (1997a) Scent counter-marking by male meadow voles: females prefer the top-scentmale. Ethology 103:443–453

Johnston RE, Sorokin ES, Ferkin MH (1997b) Female volesdiscriminate males’ over-marks and prefer top-scent males.Anim Behav 54:679–690

Keller BL (1985) Reproductive patterns. In: Tamarin RH (ed)Biology of New World Microtus, 8th edn. American Society ofMammalogists. Special Publication, Lawrence, Kan., pp 725–778

Kilian A, Yana S, von Fersen L, Gunturkun O (2003) A bottlenosedolphin discriminates visual stimuli differing in numerosity.Learn Behav 31:133–142

Kohli KL, Ferkin MH (1999) Over-marking and adjacent markingare influenced by sibship in prairie voles, Microtus ochrogaster.Ethology 104:1–11

Lipton JS, Spelke ES (2003) Origins of number sense: large-number discrimination in human infants. Psychol Sci 14:396–408

Machado A, Keen R (2003) Relative numerosity discrimination inthe pigeon: further tests of the linear-exponential-ratio model.Behav Process 57:131–148

Madison DM (1980) An integrated view of the social biologyof meadow voles, Microtus pennsylvanicus. Biologist 62:20–33

Milligan SR (1982) Induced ovulation in mammals. Oxf Rev Reprod4:1–46

Rich TJ, Hurst JL (1999) The competitive counter-marking hypoth-esis: reliable assessment of competitive ability by potentialmates. Anim Behav 58:1027–1037

Rozenfeld FM, LeBoulenge E, Rasmont R (1987) Urine markingby male bank voles (Clethrionomys glareolus Schreber, 1780;Microtidae, Rodentia) in relation to social rank. Can J Zool65:2549–2601

Ryan MJ (1997) Sexual selection and mate choice. In: KrebsJR, Davies NB (eds) Behavioural ecology: an evolutionaryapproach. Blackwell, Boston, Mass., pp 179–202

Shettleworth S (1998) Cognition, evolution and behavior. OxfordUniversity Press, New York

Thiessen DD, Rice M (1976) Mammalian scent gland marking andsocial behavior. Psychol Bull 83:505–539

Uller C, Jaegar R, Guidry G, Martin C (2003) Salamanders(Plethodon cinerus) go for more: rudiments of number in anamphibian. Anim Cogn 6:105–112

Woodward RL Jr, Bartos K, Ferkin MH (2000) Different responseof meadow voles (Microtus pennsylvanicus) and prairie voles(M. ochrogaster) to over-marks from opposite- and same-sexconspecifics. Ethology 106:979–992