Zoo Biology 6:11-20 (1987) Mother-Young Relationships in Captive Ungulates: Spatial and Temporal Patterns Katherine Rails, Barbara Lundrigan, and Karl Kranz Department of Zoological Research, National Zoological Park, Washington, DC(K.R., B.L) and The Zoological Society of Philadelphia, Philadelphia (K.K.) Ungulates are often divided into two groups?the followers and the hiders? characterized by differing spatial and temporal patterns of mother-young behavior. We assessed the relative roles of mother and young in maintaining these patterns by recording standardized quantitative measures on 37 mother-young pairs repre- senting eight species. We found no differences between followers and hiders on any of our measures. The mother was largely responsible for maintaining the species characteristic spacing distances between mother and young, although young sometimes contributed. The mother usually initiated activity bouts of the young. Activity bouts of young were longer and more frequent when the mother was active than when she was lying down. Young often stood when their mother was lying, but these activity bouts were brief. Young usually terminated their own activity bouts. Hinde's measure, a measure designed to quantify the relative roles of primate mother and young in maintaining proximity, proved less satisfactory for ungulates, due to the multiple interpretations possible for low values of the measure and the difficulty of adequately defining approaches and departures. Key words: follower, hider, ungulate, mother-young INTRODUCTION Most ungulate species can be placed into one of two groups, the "followers," in which young are in close proximity to their mother from birth, or the "hiders," in which young spend most of the first days or weeks of life at some distance from their mother, often concealed in vegetation [Warmer, 1961, 1964, 1965, 1966; Talbot et al, 1965; Eraser, 1968]. Mother and young of follower species interact frequently, but contact between hider mother and young is restricted to as few as two to three times per day during the hiding period [Lent, 1974]. What are the relative roles of mother and young in producing these characteristic spatial and temporal patterns? In hider species, young appear to select their own hiding places [Bubenik, 1965; McCullough, 1969; Jungius, 1970; Gosling, 1969; Schaller, 1967; Walther, 1979], which prompted Lent [1974] to suggest that young Received for publication March 27, 1986; accepted September 12, 1986. Address reprint requests to Katherine Rails, Department of Zoological Research, National Zoological Park, Washington, DC 20008. ? 1987 Alan R. Liss, Inc. 12 Rails, Lundrigan, and Kranz hiders make their own spatial decisions. Temporal patterns are usually defined in terms of the initiation and termination of activity periods. Hider mothers often approach their lying young to initiate nursing and are therefore thought to be largely responsible for initiating activity periods [Lent, 1974; Leuthold, 1977; Walther, 1979]. After nursing, the young actively move away from the mother [Walther, 1979], thus terminating activity periods. The roles of follower mother and young in determining spatial and temporal patterns are less clear. Lent [1974] views follower mother and young as continuously responding to each other's spatial position and suggests that follower young are more likely than hider young to initiate activities such as nursing and play on their own. Walther [1979] discusses following largely in terms of the behavior of the young; movements and vocalizations of the mother are thought to elicit the following response. The collections of the National Zoological Park (NZP), Washington, DC. and the Metropolitan Toronto Zoo (MTZ), Toronto, Ontario, provided us with the oppor- tunity to compare several species using standardized quantitative measures. Previ- ously, we reported on similarities and differences between followers and hiders on several measures of mother-young association and the percentage of time the young spent lying [Rails et al, 1986] and changes in these measures as the young matured; [Rails et al, 1987]. Here, we assess the relative roles of mother and young in maintaining spatial and temporal patterns. METHODS Thirty-seven mother-young pairs representing eight species and two families (Table 1) were observed for a total of over 600 hours between 1977 and 1982. Common names of these species are used in the text and tables; scientific names are given in Table 1. Pairs were observed at NZP; NZPs Conservation and Research Center (CRC), Front Royal, Virginia; and MTZ. Enclosures were large (ranging from 0.11 ha to 12.14 ha) and contained ample hiding places (eg, tall grass, fallen logs, and rocky outcrops). Social groupings in captivity resembled those in the wild. An attempt was made to observe each mother-young pair for 1 continuous hour per day, at least 3 days per week, for the first 7 weeks after birth of the young, but scheduled observations were occasionally missed, and individual animals were not always visible for the entire observation period. The total number of hours during which both mother and young were visible for each species is shown in Table 1. Observations were made in the morning (0800-1000) or in late afternoon (1600- 1800), when the animals tended to be most active. Thirty-two trained volunteers collected the data; each observed one or more mother-young pairs from the birth of the young until the end of the seven-week observation period. Data were recorded on a simple checksheet. Because the species observed were relatively large and slow moving and the behaviors scored were familiar and easily recognized, we achieved better than 92% agreement between observers. An analysis of the mean weekly coefficients of variation for several measures of mother-young association showed that these were not consistently related to the number of observers contributing data for a particular species, the number of pairs of each species observed, or the time of day at which observations were made [Rails et al, 1983]. Mother-Young Relationships in Ungulates 13 TABLE 1. Scientific names of species, number of mother-young pairs of each species observed, total number of hours both mother and young were visible, and type of mother-young relationship* Type of mother-young No. of pairs Hours visible relationship Taxon Rails et al Lit. References Cervidae Rangifer tarandus 5 64 F F Espmark, 1971 (reindeer) Bovidae Bison bison 3 58 F F McHugh, 1958 (bison) Hippotragus niger 4 97 H H Grobler, 1974 (sable) Oryx dammah 9 117 H H Walther, 1979 (oryx) Connochaetes taurinus 3 90 F F Estes and Estes, (wildebeest) 1979 Gazella dorcas 7 147 H H Walther, 1966, (Dorcas gazelle) 1979 Hemitragus jemlahicus 3 41 F I Schaller, 1977 (Himalayan tahr) Capra ibex 3 50 F I Schaller, 1977 (Ibex) F Walther, 1979 *Based on Rails et al [1986] and according to the literature. Species listed in taxonomic order. F, follower; H, hider; I, intermediate. At the end of each minute, observers recorded whether mother and young were lying or standing (which included moving) and the distance between them in terms of mother-lengths. Changes in activity state (as from lying to standing) and approaches and departures of one pair member to or from the other were recorded whenever they occurred. An approach was scored if an animal moved to within one mother-length of the other, a departure if it moved beyond this distance. A measure developed for primates by Hinde [1974], "the percentage of approaches by the young minus the percentage of departures by the young" (Hinde's measure) was calculated as an indicator of the relative roles of mother and young in maintaining proximity. Positive values of Hinde's measure are thought to indicate that the young plays a greater role in maintaining proximity between mother and young; negative values that the mother plays a greater role [Hinde, 1974]. The Mann-Whitney U-test was used to examine differences between followers and hiders [Siegel, 1956]. The sign test [Siegel, 1956] and test of proportions [Neter et al, 1978] were used for differences between mothers and young. RESULTS Species listed in Tables 2-7 are classified as "followers" or "hiders" according to a cluster analysis based on three measures of mother-young association recorded during the first week after birth [Rails et al, 1986]. In six of the eight species, our 14 Rails, Lundrigan, and Kranz TABLE 2. Total number of approaches and departures by mother and young, percentage made by the young, and value of Hinde's measure during week 1 and weeks 2-7 after birth* Type of mother-young Approaches Departures %By %By Hinde's relationship" Total young Total young measure Week 1 Ibex F 53 58 56 71 -13 Oryx H 72 21 67 23 -2 Wildebeest F 23 39 22 41 -2 Reindeer F 107 19 104 20 -1 Sable H 3 0 3 0 0 Gazelle H 68 41 74 31 10 Tahr F 50 36 55 25 11 Bison F 36 44 42 29 15 Weeks 2-7 Oryx H 545 51 533 48 3 Tahr F 106 51 111 43 8 Sable H 138 70 137 59 11 Reindeer F 424 62 410 46 16 Bison F 91 64 80 45 19 Gazelle H 586 70 584 47 23 Wildebeest F 275 70 275 40 30 Ibex F 285 74 271 39 35 *Ranked on Hinde's measure. aF, follower; H, hider. assignments agree with those in the literature (Table 1). There is no consensus in the literature regarding the placement of species in the tribe Caprini (goats and sheep). We classified the ibex as a follower; others have considered it either a follower [Walther, 1979] or an intermediate between followers and hiders [Schaller, 1977]. We also classified the tahr as a follower although Schaller [1977] considers it intermediate. There were no significant differences between followers and hiders on any of the measures reported here (Tables 2-7, Mann-Whitney U-test); thus we combined the two groups when examining behavioral differences between mothers and young. Spatial Patterns Hinde's measure. In week 1, values of Hinde's measure fell within a relatively narrow range (?13 to +15, Table 2) despite considerable variation across species in the percentage of approaches and departures made by the young. In subsequent weeks, young made a greater percentage of approaches (Table 2), resulting in higher values of Hinde's measure for all species but the tahr. Distance between mother and young when both were lying. The tendency for mother and young of follower species to lie much closer together than those of hider species was apparent throughout the 7-week study period, although spacing between mother and young changed over time [Rails et al, 1987]. A detailed week- by-week tabulation of the data for each species indicated that these characteristic spacing patterns were maintained regardless of whether mother or young was the second to lie (data summarized in Table 3). Mother-Young Relationships in Ungulates 15 TABLE 3. Percentage of times mother or young was second to lie at various distances from the other Distance (ML)" <1 1-2 >2 Followers Mother 54 15 31 Young 73 12 15 Hiders Mother 11 13 76 Young 14 24 62 "ML, mother-lengths. TABLE 4. Percentage of times mother or young was the first of the pair to stand* Type of mother-young relationship" First to stand (%) Sign Species Mother Young testb Oryx H 96 4 + Wildebeest F 95 5 + Reindeer F 81 19 + Bison F 79 21 + Sable H 72 28 + Gazelle H 71 29 + Tahr F 69 31 + Ibex F 66 34 + * Ranked on "mother first to stand." aH, hider; F, follower. bFor the sign test, +, mother was first to stand more often than young (P = .008, two-tailed). Temporal Patterns Initiation of activity periods. The mother was usually the first of the pair to stand after a period when both animals had been lying (Table 4). We calculated the proportion of time the young "followed" its mother, ie, stood within five mintues of the mother standing; the mean across-species was 47% (range 33-74%, data not shown). With the exception of the ibex, young were more likely to stand if their mother was standing than if she was lying (Table 5). Furthermore, young of all species tended to stand for longer periods when the mother was standing than when she was lying (Table 6). We examined the possiblity that this was due to mothers standing for longer periods than they lay; there was no significant difference in the average duration of mothers' standing and lying bouts (P = .29, data not shown). Termination of activity periods. The young was usually the first of the pair to lie down after a period when both mother and young had been standing (Table 7). In reindeer, oryx, bison, sable, and wildebeest, the young was first to lie more than 90% of the time. Mothers "followed" their young, ie, lay within 5 minutes after their young, an average of only 19% of the time (range 10-30%, data not shown). 16 Rails, Lundrigan, and Kranz TABLE 5. Percentage of standing bouts of young when mother was standing and when mother was lying* Mother-young Mother (%) Sign testb Species relationship" Standing Lying Reindeer F 94 6 + Wildebeest F 90 10 + Oryx H 87 13 + Gazelle H 70 30 + Tahr F 65 35 + Bison F 58 42 + Sable H 53 47 + Ibex F 45 55 - * Ranked on "young stand while mother standing." "F, follower; H, hider. bFor the sign test, +, young standing bouts were more frequent when mother was standing than when mother was lying; the sign, -, indicates that young standing bouts were less frequent when mother was standing than when she was lying in this species (P = .07, two-tailed). TABLE 6. Mean duration (in minutes) of standing bouts of young when mother was when mother was lying down* standing 1 and Type of mother-young relationship" Mother Sign Species Standing Lying testb Wildebeest Gazelle Sable Oryx Reindeer Bison Tahr Ibex F H H H F F F F 6.8 6.3 5.8 5.8 4.8 4.4 2.9 2.8 2.0 + 3.0 + 3.0 + 2.3 + 2.4 + 2.6 + 1.4 + 2.0 + *Ranked on "young stand while mother standing." aF, follower; H, hider. bFor the sign test, +, young standing bouts were of greater duration when mother standing than when mother lying (P = .008, two-tailed). TABLE 7. Percentage of time mother or young was the first of the pair to lie* Type of mother-young First to lie (%) Sign Species relationship" Mother Young testb Reindeer F 2 98 + Oryx H 5 95 + Bison F 6 94 + Sable H 6 94 + Wildebeest F 9 91 + Tahr F 13 87 + Gazelle H 24 76 + Ibex F 25 75 + *Ranked on "young first to lie." "F, follower; H, hider. bFor the sign test, +, young was the first to lie more often than mother (P = .008, two-tailed). Mother-Young Relationships in Ungulates 17 DISCUSSION Spatial Patterns Hinde's measure. Hinde's measure was designed to quantify the relative roles of primate mother and young in maintaining proximity. In the rhesus monkey, it changes from negative to positive as an infant matures, indicating that maintenance of proximity is due mostly to the mother in the early weeks and to the infant in later weeks [Hinde and Atkinson, 1970]. Our values for Hinde's measure and the percentages of approaches and departures made by the young (Table 2) suggest that the young played a greater role in maintaining proximity after week 1. Although this result seems intuitively reasonable, we believe that it should not be accepted without further research as we encountered serious methodological problems with these measures. Hinde's measure proved less satisfactory for ungulates than for primates for two reasons: 1) There are three possible interpretations of small values, and 2) it is difficult to adequately define approaches and departures. Our values for Hinde's measure during the first week were relatively small. The three sets of circumstances that can lead to small values are 1) The mother makes the majority of both approaches and departures, 2) the young makes the majority of both approaches and departures, and 3) each of the pair makes about the same number approaches and departures. Our week 1 data contained examples of all three. The mother made the majority of both approaches and departures in oryx, reindeer, and sable, the young made the majority of both in ibex, and mother and young contributed about equally in the wildebeest. In Hinde's rhesus study, an approach was scored when the distance between mother and young changed from more than 60 cm to less than 60 cm and a departure when the distance between them changed from less than 60 cm to more than 60 cm. We modified these definitions for ungulates by scoring an approach when the distance between mother and young decreased to less than 1 mother-length and a departure when this distance increased to more than 1 mother-length. Lickliter [1984] independently adopted these definitions in his study of the domestic goat, and apparently encountered no problems with them. However, we found both definitions inadequate. Mothers of many hider species do not approach to within one mother-length before making contact with their young. Rather, they pause several meters away [Walther, 1979; Murdock et al, 1983] and make soft vocalizations, apparently calling the young out of hiding [Lent, 1974; Leuthold, 1977; Walther, 1979; Murdock et al, 1983]. In followers also, maternal cues such as tail-wagging or head-bobbing may induce the young to approach [Lent, 1974; Walther, 1979]. In both cases, we would credit the young with an approach, although the mother initiated the interaction. Thus, we strongly suspect that the use of this definition underestimates the role of the mother. The definiton of a departure proved unsatisfactory because a distance of one mother-length is too small to distinguish movements of the young during activity periods from departures that terminate these periods. Hinde recognized a similar problem in primates. As the young mature, "the mere crossing of an arbitrary boundary could become potentially misleading as a guide to who was responsible for the change in the proximity" [Hinde, 1974]. This problem may be more severe in young ungulates as they are born at a more advanced 18 Rails, Lundrigan, and Kranz stage of development than young primates, and long distance communication between mother and young begins early in life. Distance between mother and young when both were lying. As suggested by Walther [1979], the distance between mother and young when both were lying proved more useful in separating followers from hiders than the distance between them, irrespective of activity state [Rails et al, 1986,1987]. During the first week after birth, follower mother and young lie close together, and hider mother and young lie far apart. In both followers and hiders, mothers were more often the second of the pair to lie and thus appeared to be largely responsible for these characteristic spacing patterns. Temporal Patterns Our data on changes in activity state suggest that the mother plays the major role in initiating activity periods in both followers and hiders. Mothers were usually the first to stand after both animals had been lying and young often stood within the next 5 minutes. It is likely that young respond to their mother's stand in part because it represents a potential suckling opportunity. The activity state of the mother influenced the duration of the young's activity bouts. Most bouts were very short (less than 3 minutes), although some were considerably longer (5 to 15 minutes). Short bouts occurred regardless of the mothers activity state, but longer bouts usually occurred when the mother was standing. During activity bouts, young are particularly vulnerable to detection by predators. By restricting activity to times when the mother is active (thus relative vigilant), young may reduce their risk of predation. Walther [1979] states that young hiders normally do not stand until the mother approaches the hiding place to initiate an activity period. However, many of the short bouts (< 3 min) we observed in hider young (37% in the gazelle and 56% in the sable) occurred while the mother was lying. During such bouts, the young stood, stretched, and repositioned itself before lying down again. During longer bouts it suckled, played, and interacted with its mother. Young were usually the first to lie down after both mother and young had been standing, suggesting that they terminate their own activity periods when they tire. Mothers rarely followed suit by lying within 5 minutes (in contrast to the high frequency with which young stood within 5 minutes of their mother). CONCLUSIONS 1) Despite striking differences between followers and hiders in spacing between mother and young and the frequency of mother-young interactions, we find no differences between the two groups in the way spatial and temporal patterns are maintained. 2) Follower mother and young lie close together and hiders far apart. The mother is largely responsible for maintaining these characteristic spacing patterns, although appropriate distances are maintained when the young is the second to lie. 3) The mother usually initiates activity bouts of the young. 4) Activity bouts of the young are longer and more frequent when the mother is standing than when she is lying. 5) Young often stand when their mother is lying, but these activity bouts tend to be relatively short (< 3 min). Mother-Young Relationships in Ungulates 19 6) Young usually terminate their own activity bouts. 7) Hinde's measure proves less satisfactory for ungulates than for primates due to the multiple interpretations possible for low values of the measure and difficulty in adequately defining approaches and departures. ACKNOWLEDGMENTS This project would not have been possible without the help of the Friends of the National Zoo, Washington, D.C., which provided many of the volunteer observers as well as financial support. We thank L.W. Cahill, director of live collections at MTZ, for facilitating observations there and Elizabeth Merritt for preliminary data reduction and analysis. We are especially grateful to all those who observed animals or helped in other ways at NZP and MTZ and regret that space does not allow us to recognize them individually. We also thank Kent Redford, Chris Wemmer, and Susan Lumpkin for helpful comments on an earlier draft of the manuscript. REFERENCES Bubenik, A.B. Beitrag zur Geburtskunde und den Mutter-Kind Beziehungen des Reh-und Rot- wildes. ZEITSCHRIFT FUR SAUGETIER- KUNDE 30:65-128, 1965. Espmark, Y. Mother-young relationships ontog- eny of behavior in reindeer (Rangifer tarandus L.). ZEITSCHRIFT FUR TIERPSYCHOLOGIE 29:42-81, 1971. Estes, R.D.; Estes, R.K. The birth and survival of wildebeest calves. ZEITSCHRIFT FUR TIERP- SYCHOLOGIE 50:45-95, 1979. Eraser, A.F. REPRODUCTIVE BEHAVIOUR IN UNGULATES. London, Academic Press, 1968. Gosling, L.M. Parturition and related behaviour in Coke's hartebeest, Alcelaphus buselaphus cokei Gunther. JOURNAL OF REPRODUCTIVE FERTILITY (SUPPL)6:265-286, 1969. Grobler, J.H. Aspects of the biology, population ecology, and behavior of the sable Hippotragus niger niger (Harris, 1838) in the Rhodes Matapos National Park, Rhodesia. ARNOLDIA (RHO- DESIA) 7:1-36, 1974. Hinde, R.A. BIOLOGICAL BASES OF HUMAN SOCIAL BEHAVIOUR. New York, McGraw- Hill, 1974. Hinde, R.A.; Atkinson, S. Assessing the roles of social partners in maintaining mutual proximity, as exemplified by mother-infant relations in rhe- sus monkeys. ANIMAL BEHAVIOUR 18:169- 176, 1970. Jungius, H. Studies on the breeding biology of the reedbuck (Redunca arundinum Boddaert, 1785) in the Kruger National Park. ZEITSCHRIFT FUR SAUGETIERKUNDE 35:129-146, 1970. Lent, PC. Mother-infant relationships in ungu- lates, pp. 14-55 in THE BEHAVIOR OF UN- GULATES AND ITS RELATION TO MANAGEMENT. V. Geist; F. Warmer, eds. Morges, Switzerland, International Union for the Conservation of Nature, 1974. Leuthold, W. AFRICAN UNGULATES: A COMPARATIVE REVIEW OF THEIR ETH- OLOGY AND BEHAVIORAL ECOLOGY. New York, Springer-Verlag, 1977. Lickliter, R. Mother-infant spatial relationships in domestic goats. APPLIED ANIMAL BEHAV- IOR SCIENCE 13:93-100, 1984. McCullough, DR. THE TULE ELK: ITS HIS- TORY, BEHAVIOR AND ECOLOGY. Berkley, University of California Publications in Zoology no. 88, 1969. McHugh, T. Social behavior of the American buf- falo (Bison bison bison). ZOOLOGICA 43:1-40, 1958. Murdock, O.K.; Stine, W.W.; Maple, T.L. Obser- vations of maternal-infant interactions in a cap- tive herd of sable antelope (Hippotragus niger). ZOO BIOLOGY 2:215-224, 1983. Neter, J.; Wasserman, W.; Whitmore, G.A. AP- PLIED STATISTICS. Boston, Allyn and Bacon, 1978. Rails, K.; Lundrigan, B.; Kranz, K. Variability of behavioral data recorded by volunteer observers. INTERNATIONAL ZOO YEARBOOK 22:244- 249, 1983. Rails, K.; Kranz, K.; Lundrigan, B. Mother-young relationships in captive ungulates: Variability and clustering. ANIMAL BEHAVIOUR 34:134-145, 1986. Rails, K.; Lundrigan, B.; Kranz, K. Mother-young relationships in captive ungulates: Behavioral changes over time. ETHOLOGY, in press. Schaller, G. THE DEER AND THE TIGER. Chi- cago, University of Chicago Press, 1967. Schaller, G. MOUNTAIN MONARCHS. Chi- cago, University of Chicago Press, 1977. Siegel, S. NON-PARAMETRIC STATISTICS FOR THE BEHAVIORAL SCIENCES. New York, McGraw-Hill, 1956. Talbot, L.M.; Payne, W.J.A.; Leger, HP.; Verd- court, L.D.; Talbot, M.H. The meat production potential of wild animals in Africa. Common- 20 Rails, Lundrigan, and Kranz wealth Agricultural Bureau Farnam Royal En- OGIE 21:393-467, 1964. gland. COMMONWEALTH BUREAU ANI- Walther, F. Verhaltensstudien an der Grantgazelle MAL BREEDING GENETICS 16, 1965. (Gazella grand Brooke, 1872) im Ngorongoro- Walther, F. Einige Verhaltensbeobachtungen am Krater. ZEITSCHRIFT FUR TIERPSYCHOL- Bergwild des Georg von Opel Freigehege. TIER- OGIE 22:167-208, 1965. FORSCHUNG 1960-1961,53-89, 1961. Walther, F. MIT HORN UND HUF. Hamburg Walther, F. Verhaltensstudien an der Gattung Tra- und Berlin, Parey, 1966. gelaphus de Blainville (1816) in Gefangenschaft, Walther, F. Das Verbal ten der Horntrager (Bovi- unter besonderer Berucksichtigung des Sozialver- dae). HANDBUCH DER ZOOLOGIE. Band 8, haltens. ZEITSCHRIFT FUR TIERPSYCHOL- Lieferung 54, 10(30): 1-184, 1979.