The Condor 91:341-346 0 TheCooper Ornithological Society 1989

REVERSE MOUNTING IN ’

GARY L. NUECHTERLEIN Department of Zoolom, North Dakota State University,Fargo, ND 58105

ROBERT W. STORER Museum of Zoology and Department of Biology, Universityof Michigan, Ann Arbor, MI 48109

Abstract. Reversemounting (females mounting males) is commonin manygrebe . Of 328 mountingsobserved and filmed in Silvery Grebes( occipitalis), 27% were by the female.Similarly, for HoodedGrebes (P. gallardoi) observedduring the sameperiod, 15% of 95 mountingswere reversed. For both species,reverse mounting was significantly more common early in the season.Reverse mounting in Silvery Grebesalso was more frequentlyassociated with pre- and postmountingcourtship display activity at the nestthan werecopulations. We concludethat reversemounting is not aberrantbehavior in grebes, but constitutesa regularand integralpart oftheir courtshipbehavior. Since reverse mounting alsoappears to occur in a wide diversity of other monomorphic species,we suggest that using mounting behavior to determine sex in these speciesmay be unreliable. Key words: Hooded ;Silvery Grebe;Podiceps; reversemounting; copulation.

INTRODUCTION that there may be important behavioral differ- Although reversemounting hasbeen reported for ences between reverse mounting and “normal” many bird species,most authors appear to regard copulation sequences.In an analysis of filmed the behavior as unusual or aberrant. The evi- sequencesof the ( dom- dence from easily sexed, strongly dichromatic i&us), for example, Storer (1976) found signif- bird speciesgenerally lends support to this as- icant differences related to whether or not the sumption (Morris 1954, 1955; Ficken 1963; No- roles of male and female were reversed. lan 1978; Thompson and Lanyon 1979), but in Several nonmutually exclusive hypothesescan grebes(Podicipedidae) and several other mono- be proposed to explain why reverse mounting chromatic species the behavior is sufficiently occurs in grebes. Reverse mounting may (a) re- widespreadto warrant re-examination of this as- sult in fertilization, (b) result from mistaking a sumption. partner’s sex,(c) be an aberrant behavior, (d) play The prevalence of reverse mounting in grebes a role in courtship and pair formation, (e) indi- has become particularly evident in our research cate a reversal of dominance in the sexes(Storer on the comparative courtship behavior of the 1976), or (f) stimulate the ovaries of the female colonial grebes.Because grebe pairs often engage to grow (Storer 1976). in repeated mounting attempts (sometimes five The primary objectives of the fieldwork re- or six copulations in a 15- to 20-min period), ported in this paper were to examine the contexts reversal in the roles of the sexes during such and behavioral details of reverse mounting in bouts is particularly conspicuous(see Storer 1969, Silvery Grebes (Podiceps occipitalis) and Hooded 197 1, 1976). Within colonies, sufficient sample Grebes (P. gallardoi), asking specifically the fol- sizesof both copulations and reverse mountings lowing questions: (1) Is reverse mounting aber- can be obtained to compare statistically the con- rant behavior? (2) When in the pair-formation texts and structural details of the two forms of processdo reversemountings take place?(3) Does mounting behavior. sperm transfer take place during reverse mount- Most authors describing copulation in grebes ing? (4) Are there consistent differences in the have used the neutral terms “passive bird” and platform displays of reverse mountings vs. cop- “active bird,” to avoid implying sex-specificbe- ulation? havior, but this practice ignores the possibility METHODS As part of a 3-year comparative study, we con- I Received2 1 July 1988.Final acceptance3 January ducted fieldwork on Silvery and Hooded grebes 1989. on small (lessthan 2 km2) snow-melt lakes in the

[3411 342 GARY L. NUECHTERLEIN AND ROBERT W. STORER

TABLE 1. Frequencyof copulations(male on female: larger than female bills (t = 6.0, df = 28, P < M/F), reverse mountings (F/M), and same sex mount- 0.001 for bill length; t = 3.9, df = 28, P -c 0.001 ings (M/M and F/F) in Silvery and Hooded grebes. for bill depth). When both measures were used in combination, there was almost no overlap be- Possible mountings M/F F/M M/M F/F tween male and female bill measurements. Al- though too few museum specimens exist for a Silvery Grebe 239 88 1 similar statistical analysis of bill size for the rarer 81 14 0 Hooded Grebe, a similar bill dimorphism was evident. foothills of the Andes in southern Patagonia, Ar- Most mounting activities were observed from gentina (52”S, 72”E). Observations on reverse 4 to 10 m distance using either binoculars or a mounting and copulation were most intensive in 15-60 x spotting scope. We also obtained films the 198 1-1982 season. of the behavior from within a distance of 5 m In November 198 1, we observed a group of using a 16-mm Beaulieu camera and the floating several hundred Silvery Grebes and 35 Hooded blind. For each mounting, we noted the sequence Grebes engagingin courtship activities on an un- of platform displays that occurred and whether named staging lake, which we named Laguna the active and passivebirds were male or female. Nevada. Unlike waterfowl (),grebes Display terminology used in this paper is that of cannot mount in open water (McAllister and Sto- Storer (1969). rer 1963), and we observed Silvery Grebes using In late November, the courting flocks of grebes flat, partly submergedrocks on the far north end on Laguna Nevada dispersedto other small lakes of the lake as copulation platforms. To obtain in the area for breeding. In December 198 1, closer observation and films of this activity, we Nuechterlein set up platforms on Laguna Blan- constructed platforms directly in front of our chillo, where over 100 Silvery Grebes and 30 camp. These platforms consistedof three woven Hooded Grebes eventually nested. Most basketsplaced in each comer of a floating trian- on this lake arrived already paired, so these ob- gular frame made of 3-m bamboo poles. The servations and those at the nesting colony pro- entire frame was then anchored in 1 m of water vided us with a late-season sample that could be and protected from wave action by floating wave compared to that from Laguna Nevada. breaks. These platforms were readily adopted by the RESULTS grebesand over a 2-year period, we were able to We observed 328 mounting attempts in Silvery film or observe more than 400 copulations or Grebes and 95 in Hooded Grebes. All attempts reverse mountings. Later in the season, mount- involved a male and female except for one case ing at nest platforms within breeding colonies in Silvery Grebes where a male attempted to was also observed from a floating blind. In Pat- mount another male (Table 1). This case also agonia, both specieswere very tame and allowed was unusual in that the passive male neither the close observation distances necessaryto sex Reared nor Invited. the birds reliably. Reverse mounting was common in both Prior to mounting, all pairs in the sample were species,accounting for 27% of the mountings in sexed visually using bill size, which is greater in Silvery Grebes and 15% in Hooded Grebes (Ta- males for most or all grebe species(Storer, un- ble 1). This was particularly evident early in the publ. data). Only pairs that could be easily dis- seasonon the temporary platforms, where 40% tinguished before making the behavioral obser- of 185 mounting attempts by Silvery Grebes were vations were included in the sample. This sexing by females. Later, in the -laying period, re- technique was verified for Silvery Grebes by R. verse mounting was significantly less frequent W. Storer by measuring the bills of male and (9.8% of 142 attempts, Table 2). A similar tally female specimens using dial calipers. Measure- of reverse mountings vs. copulations in Hooded ments taken included depth of the closed bill at Grebes showed that pairs just beginning to build the level of the posterior edge of the nostril and their nesting platforms (first day) engagedin sig- bill length from the anterior edge of the nostril nificantly more reverse mountings than did pairs to the tip, both to the nearest 0.1 mm. For both with well-formed cupped nests during the same measures,bills of males were highly significantly 3-day period (33% vs. 5%, Table 2). REVERSE MOUNTNG IN GREBES 343

TABLE 2. Frequencyof copulation vs. reverse mounting by: (A) Silvery Grebes on platforms during the pair- formation period (26-27 November, Laguna Nevada) vs. at nests during egg laying (11-12 December, Laguna Blanchillo); and (B) Hooded Grebes during early and late stagesof nest building.

Copulations

A. Silvery Grebe Platforms 111 14 G* = 40.4, df = 1, P < 0.001 Nests 128 14 B. Hooded Grebe On new nesting platforms 22 11 G2 = 13.4, df = 1, P < 0.001 On cupped nests 59 3

A behavioral analysis of copulation and re- quences males were significantly more likely to verse-mounting sequencesin Silvery Grebes in- Rear and Wing-quiver (Table 3A) prior to In- dicates many significant differences(Table 3). In viting the female to mount. Postmounting dis- contrast to copulatory sequences, ejaculation play activity, particularly Head-flicking and never occurred in reverse mountings, although Habit-, also was more frequent when sample size for this behavior was somewhat lim- females mounted males (Table 3C). ited (n = 21) owing to the necessity for close observations of birds that were facing directly DISCUSSION away from the observer. Tail thrusting by the On the basis of our data from the Silvery, Hood- female, however, occurred in 29% of 72 reverse ed, and other grebes,we can reject several of the mounts, and the cloaca of the male often was proposed hypotheses. If fertilization were in- everted during reverse mounts (Table 3B). volved, we would predict that reverse mounting Whether male or female, the passive bird of would be most frequent during the egg-laying opposite-sex mountings always Invited mount- period and that ejaculation would occur. Instead, ing by holding its head low over the water with data from studies of the Silvery Grebe show that crest flattened. There were differences,however, the frequency of reverse mounting is much great- in two displays that frequently preceded this in- er during early pair formation than during or vitation to mount. In reverse-mounting se- immediately preceding egglaying. Furthermore,

TABLE 3. Comparison of behavior occurring (A) before, (B) during, and (C) after copulations vs. reverse mountings in Silvery Grebes. Behaviorsare listed in sequence,with those of the passive(mounted) bird denoted by *. Postmounting displays often were given by both birds, but only those of the active bird were tallied.

Copulation Reversemounting Behavior’ Did Did not Did Did not More frequentin: Significance’

A. Before Rear* 15 113 20 47 Reverse mounting P < 0.005 Wing-quiver* 3 13 13 4 Reverse mounting P < 0.005 Invite* 141 0 75 0 - ns B. During Evert cloaca* 52 1 6 8 Copulation P < 0.001 Tail-thrust 132 9 51 21 Copulation P < 0.001 Ejaculation 23 18 0 21 Copulation P < 0.001 C. After Patter 160 2 72 0 - Head-flick 3 111 12 42 Reverse mounting P 3.001 Face partner 41 63 24 31 - Head-turn 121 32 50 24 Reverse mounting? P <:.06 Habit-preen 6 108 17 37 Reverse mounting P < 0.001 I Display terminologyfrom Storer(I 969). i Significancelevels are for G-tests(df = 1 m all cases)of the null hypothesisthat behavioral frequenciesfor copulatorymounting vs. reverse mountmgare not different. 344 GARY L. NUECHTERLEIN AND ROBERT W. STORER we did not observe ejaculation in any reverse nificantly in frequency. In reverse-mounting se- mountings. Likewise, in the case of the Horned quences, Inviting was more frequently preceded Grebe (Storer 1969) reverse mounting was most by Rearing and Wing-quivering, and mounting frequently observed on early platforms that were was followed more frequently by Head-flicking not used for egg laying, and tine films of both and Habit-preening. All of these are ritualized Horned and Least grebesalso showed no cloaca1 displays commonly used in courtship sequences contact (Storer 1976). during early pair formation. In contrast, the func- If reverse mounting were aberrant behavior, it tional copulatory movements of everting the should be rare. However, 27% of 328 mountings cloaca, tail-thrusting, and ejaculation were all by Silvery Grebes and 15% of 95 mountings by significantly less frequent during reverse mount- Hooded Grebes were by the female. Reverse ing (Table 3). mounting is also common in at least Horned Finally, all grebes that have been studied are (Podicepsauritus) (Storer 1969) Great Crested monogamous, monochromatic, and engage in (P. cuistatus)(Selous 1901) Least (Storer 1976), mutual or reciprocal displays. In the Least Grebe Eared (P. nigricollis), and Western grebes(Aech- (Storer 1976), reversemounting differs from cop- mophorus occidentalis) (Nuechterlein, unpubl. ulation in the lack of cloaca1 contact, the less data), and it has been reported by Storer (1976) regularalternation of postcopulatorydisplays, and in Red-necked (P. grisegena) and Pied-billed a difference in the reaction time between post- grebes ( podiceps), and in the New copulatory displays. We conclude that reverse Zealand Dabchick ( rufopectus). mounting is a regular and integral part of the This suggeststhat far from being aberrant be- “courtship” behavior of grebes. Whether dom- havior, it is of regular occurrencein many, if not inance reversal or stimulation of the ovaries or all, grebe species. both are involved cannot be determined from If reversemounting were a matter of mistaking our data. the sex of the partner, one would predict that The extent to which reverse mounting in other mounting by birds of the same sex also would birds is functionally similar to that in the grebes be common. However, of 327 mountings by Sil- is unclear. Reverse mounting in at least two non- very Grebes,including 88 reversemountings, only passerinesspecies appears to occur in different one was by a bird of the same sex as the bird contexts from that in grebes. In the mounted (both males). We know of no other rec- (Scopusumbretta), “repeated false mounting is ords of same-sex mountings in grebes. performed, apparently largely indiscriminantly If reverse mounting has been incorporated as among groups of up to 8-l 0 birds near the nest” part of courtship, we can make three predictions: (Brown et al. 1982); and in the Acorn Wood- (1) that reverse mounting would be most fre- pecker (Melanerpes formicivorus), preroosting quent early in the seasonduring pair formation, mounting has been reported by MacRoberts and (2) that some functional details of the move- MacRoberts (1976). The latter is apparently nei- ments involved in copulatory mounting may ther age- nor sex-related, and its function is un- show signsof becoming cursory or ritualized, and known, although it may be related to reducing (3) that reverse mounting in birds would be most aggressivenessprior to group roosting in holes common in monogamous,monomorphic species, where birds may be crowded together. in which male and female commonly engagein The difficulty of determining the sex of mono- other reciprocal or mutual courtship displays. morphic birds is an important obstacle for re- The first of these predictions is supported by searchersstudying their behavior. A common our data from the Silvery Grebe (Table 2), and method used by ornithologists is to color-mark Storer (1969) reported that reverse mounting in individual birds, then “sex” them by noting which at least the Homed Grebe occurs “early in the bird mounts during “copulation.” The research- season.” Field studies presently underway show er thereby avoids the disturbance and difficulties a similar bias towards early season reverse associated with live-capture and laparotomy mounting in Eared Grebes (G. Nuechterlein and techniques. Applications of this procedure, how- D. Buitron, unpubl. data). ever, entails an important implicit assumption: In the Silvery Grebe, Inviting invariably pre- that in birds, males mount females and not vice cededboth copulation and reversemounting, but versa. other pre- and postmounting displays varied sig- This assumption may not be valid, not only REVERSE MOUNTNG IN GREBES 345

TABLE 4. Preliminary list of speciesin which reverse mounting has been reliably observed. Relative degree of sexual dichromatism is indicated (M = monochromatic, M’ = nearly monochromatic, D = dichromatic).

Grebes, Podicipedidae (many species) M This paper Wedge-tailed Shearwater,Pujinus pacz~cus M Shallenbereer1973 Red-footed Booby, Sula stda M Vemer 19gl Peruvian Booby, Sula variegata M Nelson 1978 Northern Gannet, Sula bassana M Nelson 1965 Great , Phalacrocoraxcarbo M Kortlandt 1942 Brandt’s Cormorant, Phalacrocoraxpenicillatus M B. Boekelheide, pers. comm. Shag,Phalacrocorax aristotelis M E. Cullen in Nelson 1965 , M Skead 1966, Blaker 1969, Lancaster 1970 Hamerkop, Scopusumbretta M Brown et al. 1982 Hadada Ibis, Bostrychiahagedush M D. S. DeCourcey, pers. comm. Moorhen, Gallinula chloropus M Huxley 1914, Howard 1940, Anderson 1975 Piping Plover, Charadriusmelodus M S. Haig, pers. comm. Southern Brown Skua, Catharacta skua M P. F. Jenkins, pers. comm. Pigeon Guillemot, Cepphuscolumba M Storer 1945 Rock Dove, Columba livia M Huxley 1914, N. Burley, pers. comm. Pied Barbet, Tricholaemaleucomelan M Curio 1978 Red-headed Woodpecker, Melanerpes erythrocephalus M Brackbill 1969 Acorn Woodpecker, Melanerpesformicivorus M’ MacRoberts and MacRoberts 1976 Lewis’ Woodpecker, Melanerpeslewis M Bock 1970 Red-bellied Woodpecker, Melanerpescarolinus M’ Kilham 1958, 1961; Hauser 1959 Northwestern Crow, Corvuscaurinus M James 1983 Rook, Corvusfrugilegus M Coombs 1978 European , Sturnusvulgaris M Glick 1954 Prairie Warbler, Dendroicadiscolor D Nolan 1978 Painted Bunting, Passerinaciris D’ Thompson and Lanyon 1979 American Redstart, Setophagaruticiila D Ficken 1963 Zebra Finch, Poephilaguttata D Morris 1954 I Reportedcase involved a youngmale in subadultplumage.

for grebes, but for many other monochromatic not been previously published. Although this table un- bird species. In recent years, the increased focus doubtedly provides a woefully incomplete picture of the extent of reversemounting in birds, we hope it may on marking and observing known individuals stimulate others to make the critical observationsnec- has provided documented instances of reverse essaryto document this behavior in other species. mounting for a wide diversity of species,partic- ularly early in pair formation (Table 4). Our re- LITERATURE CITED searchand review suggestthat reverse mounting ANDERSON,A. 1975. A method of sexing Moorhens. in many speciesmay be more common than fre- Wildfowl 26177-82. quently supposed, and that observations of BLAKER,D. 1969. Behaviour of the Ar- deola ibis. Ostrich 40:75-129. mounting behavior should not be used indis- BOCK, C. E. 1970. The ecology and behavior of the criminately as primafacie evidence for sex de- Lewis Woodpecker (Asyndesmuslewis). Univ. termination. Calif. Publ. Zool. 92: l-100. BRACKBILL,H. 1969. Reverse mounting by the Red- ACKNOWLEDGMENTS headed Woodpecker. Bird-Banding 40:255-256. BROWN,L. H., E. K. URBAN,AND K. NEWMAN.1982. Our studiesin Argentina were funded by the National The birds of . Vol. 1. Academic Press,Lon- Geographic Society and the International Council on don. Bird Preservation (Pan American Section). We thank COOMSS,F. 1978. The crows, a study of the corvids of our coworker,Andres Johnson, and the FundacionVida Europe. Batsford, London. SilvestreArgentina for their continual help and support CURIO, E. 1978. Scheinpaarungeneines Paars von during our stay in Argentina. Deborah Buitron and Rotstirnbartvogeln (Tricholaema 1. leucomelan). Frank McKinney provided many helpful suggestions J. Omithol. 119:331-333. for improving the manuscript. We would also like to FICKEN,M. S. 1963. Courtship of the American Red- thank our many colleagueswho, during our poster ses- start. Auk 80:307-3 17. sion on this subject,provided the information and ob- GLICK, B. 1954. Reverse mounting in the Starling servationssummarized in Table 4, much of which has (Sturnusvulgaris). Auk 7 1~204. 346 GARY L. NUECHTERLEIN AND ROBERT W. STORER

HAUSER,D. C. 1959. Reverse mounting in the Red- MORRIS,D. 1955. The causationof pseudofemaleand bellied Woodpeckers.Auk 76:36 1. pseudomale behaviour: a further comment. Be- HOWARD,E. 1940. A Waterhen’s worlds. Cambridge haviour 8:46-56. Univ. Press, London. NELSON, J. B. 1965. The behaviour of the Gannet. Br. HUXLEY, J. 1914. The courtship-habits of the Great Birds 58:233-288. Crested Grebe (Podicepscristatus); with an addi- NELSON,J. B. 1978. The :gannets and boobies. tion to the theory of sexual selection. Proc. Zool. Oxford Univ. Press, London. Sot. London, 1914:49 l-562. NOLAN,V., JR. 1978. The ecologyand behavior of the JAMES,P. C. 1983. Reverse mounting in the Northwest Prairie Warbler Dendroica discolor. Ornithol. Crow. J. Field Omithol. 54:4 18-4 19. Monogr. No. 26. American Ornithologists’ Union, K~LHAM,L. 1958. Pair formation, mutual tapping and Washington, DC. nest hole selection of Red-bellied Woodpeckers. SELOUS,E. 1901. An observationaldiary ofthe habits- Auk 75:318-329. mostly domestic-of the K~LHAM,L. 1961. Reproductive behavior of Red-bel- (Podiceps cristutus). Zoologist (1901): 161-183, lied Woodpeckers.Wilson Bull. 73:237-354. 339-350. KORTLANDT,A. 1942. Levensloop, samenstelling en SHALLENBERGER,R. J. 1973. Breedingbiology, homing structuur der Nederlandse aalscholverbevolking. behavior, and communication patterns of the 31:175-280. Wedge-tailed Shearwater, Pujfinus pacijicus. LANCASTER,D. A. 1970. The breeding behavior of the Ph.D.diss. Univ. of . Los Anaeles. Cattle Egret in Colombia. Living Bird 9: 167-l 94. SKEAD,C. J. 1966. A study of the Cattle Egret, Ardeola LAWRENCE,L. DE K. 1967. A comparative life-history ibis L. Proc. 2nd Pan African Omithol. Congr., study of four species of woodpeckers. Ornithol. Ostrich Suppl. 6: 109-139. Monogr. No. 5. American Ornithologists’ Union, STORER,R. W. 1945. Structural modifications in the Washington, DC. hind limb in the Alcidae. Ibis 1945:433-456. MACROBERTS,M. H., AND B. R. MACROBERTS.1976. STORER,R. W. 1969.The behavior of the Homed Grebe Social organization and behavior of the Acorn in spring. Condor 7 1:180-205. Woodpecker in central coastal California. Orni- STORER,R. W. 1971. The behaviour of the New Zea- thol. Monogr. No. 2 1. American Ornithologists’ land Dabchick. Notomis 18:175-l 86. Union, Washington, DC. STORER,R. W. 1976. The behavior and relationships MCALLISTER,N. M., AND R. W. STORER.1963. Cop- of the Least Grebe. Trans. San Diego Sot. Nat. ulation in the Pied-billed Grebe. Wilson Bull. 75: Hist. 18:113-126. 166-173. THOMPSON,C. F., AND S. M. LANYON.1979. Reverse MORRIS,D. 1954. The reproductive behaviour of the mounting in the Painted Bunting. Auk 96:417- Zebra Finch (Poephih guttutu), with special ref- 418. erence to pseudofemale behaviour and displace- VERNER,J. 1961. Nesting activities of the Red-footed ment activities. Behaviour 6:271-322. Booby in British Honduras. Auk 78~573-594.