Behavioral Ecology Vol. 10 No. 5: 525-532
© 1999 International Society for Behavioral Ecology
Red deer females collect on male clumps at mating areas
Cátedra de Biología y Etología, Facultad de Veterinaria, Universidad de Extremadura, 10071 Cáceres, Spain
Address correspondence to J. Carranza. E-mail: carranza{at}unex.es .
Received 21 July 1998; revised 30 January 1999; accepted 8 February 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Mating strategies in mammalian herbivores are adapted to the dispersion of females, and female dispersion is mainly determined by resource dispersion, although it is frequently unclear whether females may also be influenced by the location of males. In the red deer (Cervus elaphus) the distribution of females before the rut predicts the places were males should establish territories and even their relative success. However, the number of females using the mating areas in Doñana increases during the rut. We observed 20 areas of meadows, used by grazing females before the rut. At the onset of the rut, the number of females increased in some of these areas and decreased in others, and the opposite pattern was found after the rutting period. Changes in the vegetation at mating and nonmating areas could not account for the changes in female distribution; even some of the highest quality meadows were vacated by females during rut. In selecting the mating areas, females avoided isolated small meadows within the scrub area and preferred larger meadows where a number of neighboring rutting males could be found. Females also avoided those areas heavily used by fallow deer (Dama dama), a competing sympatric species. We found that females suffered less sexual harassment when in larger harems and when their harem was surrounded by other harems. Our results, together with those in the literature about this population, indicate that red deer females collect during the early rut in mating areas containing several rutting males, although once there they may select particular sites based on availability of food rather than based on the presence of a particular male. By joining harems in large meadows they are less harassed, and at the same time they probably increase their chances of mating with highly competitive males. The results from Doñana support the idea that harassment avoidance may lead to female movements to areas with male territories without lek breeding or female comparison of male phenotypes and may bring an insight into those factors leading to clumps of male territories and leks.
Key words: Cervus elaphus, harassment, leks, red deer, mating systems, space use, resource defense, territories.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
When female reproductive success is mainly limited by access to resources and male reproductive success is mainly limited by access to females (Arnold and Duvall, 1994
;
Bateman, 1948;
Clutton-Brock and Vincent,
1991; Trivers,
1972), female dispersion is expected to be influenced by resources
(modified by predation and the benefits and costs of social living), whereas
male dispersion should primarily be influenced by female dispersion
(Davies, 1991;
Emlen and Oring, 1977).
Despite males developing a number of strategies to stay close to females,
in many cases females themselves change their space use at the time of mating
to meet males. Such female decisions and their role in mating systems are
receiving increasing attention
(Ahnesjö et
al., 1992). In many cases it is not clear whether the distribution
of females at the time of mating is simply caused by ecological factors or
whether it may also be influenced by the distribution of males (e.g.,
Lundberg and Alatalo, 1992).
By such patterns of movement at the time of mating, females are expected to
obtain either indirect (genetic) or direct benefits (resources that males
offer, safe places, oviposition sites, etc.)
(Carbone and Taborsky, 1996;
Clutton-Brock et al.,
1996).
Male mating strategies in large mammalian herbivores are adapted to the
dispersion, grouping, and predictability of the occurrence of females
(Jarman, 1974,
1983). Female movements are
mainly designed to optimize nutrient intake, and males commonly move from
their usual home ranges to meet receptive females, either by displaying at
places that females are likely to use (hot spots) or by defending groups of
females (harems) or the resources they need
(Clutton-Brock, 1989).
Red deer males (Cervus elaphus), like many other ungulates,
commonly spend most of the year in areas separated from those occupied by the
females (Clutton-Brock and Albon,
1989; Cluton-Brock et al.,
1982). At the beginning of the rutting season, males usually move
from their home ranges to the areas with females
(Clutton-Brock et al., 1982).
In these areas, adult stags try to monopolize a number of females, excluding
competitors either by defending harems
(Clutton-Brock et al., 1982)
or mating territories at places where females clump
(Carranza et al., 1990,
1996). Previous work at
Doñana National Park in southern Spain has
shown that the distribution of females during the summer (August), before the
onset of the rut, predicts both the location of the rutting males and their
mating success (Carranza et al.,
1990). Actually, mating territories are located at places with
high-quality grass, at routes of female movement, or both
(Carranza, 1995,
1996). In any case, mating
territories are in areas that females use before the rut. Therefore, it seems
that females are using the space to find food and males defend strategic
spatial patches when the distribution of resources that females need make them
worth defending. The evidence for resource defense is further supported
because territorial males lost females when they were experimentally forced to
defend a different territory, and the loss of females was largely explained by
a decrease in grass quality (Carranza,
1995).
Such findings suggest that females are mostly interested in food, and as a
result they mate with highly competitive males simply because males have to
compete for females at food patches. Although male movement to female areas is
commonly acknowledged for the red deer mating system, there are some mentions
in the literature that females may increase the use of some areas during the
mating period (Clutton-Brock et al.,
1982). Carranza et al.
(1990) mentioned an increase
in the number of females in the mating areas in
Doñana as the rut progressed. If females
gather during the rut, they should decrease in other areas. There could be at
least two explanations for such movements: (1) they are simply reflecting
changes in the distribution of food as the season progresses. In this case it
would appear that females moved to mating areas but these areas become mating
areas because males have to go there to find females. (2) Females prefer to
mate at these areas for some other reasons related to factors involved in
mating behavior. These reasons are commonly investigated in lek breeding
ungulates, where females collect on clumps of male territories
(Höglund and
Alatalo, 1995). One key question here is whether mate choice or
harassment avoidance are involved in such female movements (see
Carbone and Taborsky, 1996,
and references herein). So far we have no firm evidence of mate choice in
territorial ungulates (Balmford,
1991; Balmford and Read,
1991). In fallow deer (Dama dama) at Petworth, England,
males retained females after being experimentally moved from their
territories, but even here the male's success may depend on his ability to
retain females, instead of female preferences
(Clutton-Brock et al., 1989).
On the other hand, female preferences cannot be discarded in any studied
species (Clutton-Brock et al.,
1996), so the question of why females prefer clumps of males is
still open.
The aim of this study was to determine whether red deer females collect on some areas during the mating season in Doñana National Park and, if so, to investigate the features of these areas and the likely benefits associated to such behavior.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The study area in Doñana National Park included a western zone with Mediterranean scrubland and an eastern zone with the marsh (which is dry during the months of study) of the Guadalquivir River, separated by an ecotone, a long narrow strip with meadows and rushes (Figure 1). The climate is typically Mediterranean, with hot, dry summers and mild, wet winters. Seasonality is strong, especially regarding the water level and its effects on the vegetation. After the winter-spring rains and throughout summer and early autumn, the waters recede and most habitats dry up. At this time, slight depressions in the overall plain area are the only zones remaining relatively moist, due to the reserves from the higher water table of the scrubland. This allows the presence of relatively green meadows at such places even in late summer and early autumn, where deer graze. These meadows occur mainly at the ecotone, between the higher area of scrub and the lower dry marsh, but also at some places within the scrubland where there are patches of meadows at small depressions or around areas with ponds. For more details on Doñana National Park, including the vegetation, see Allier et al. (1974
) and Rogers and Myers
(1980).
|
The rut in Doñana usually takes place
approximately between the 1st and 25th of September. During this period males
move from their home ranges to the areas used by females in the ecotone and
use either harem-holding or territorial tactics to monopolize females
(Carranza, 1995;
Carranza et al., 1990,
1996). Within the study area,
our observations were centered on the meadows used by red deer, either at the
ecotone or within the scrub area. Data were collected in August and September
1995, in September 1996, and from August to October 1997.
Climatic conditions during the years of field work changed dramatically from, 1995 to 1997. Rainfall in Doñana (Palacio de Doñana meteorological station) between September, 1994 and August, 1995 was only 254.4 mm, this year being at the end of a 5-year drought period. In, 1995-1996 rainfall increased to 1032.3 mm, and it was also high (885.3 mm) in 1996-1997. As a consequence, food availability was low in 1995 and high in 1996 and 1997. Also, deer populations in the study area decreased between 1995 and 1997, probably as a consequence of the effects of the previous drought period coupled with a great flood in 1996, when a number of red and many fallow deer drowned (Carranza and Valencia, personal observations). Mean number of red deer females persite observed in both years of study (see below for a description of the study sites) was 10.81 ± 8.81 SD in 1995, and it was only 4.41 ± 7.05 SD in 1997 (Wilcoxon test, Z = 3.107, N = 14, p =.0019). For fallow deer, mean number of individuals per study site was 21.36 ± 24.03 SD in 1995 and only 3.59 ± 4.67 SD in 1997 (Wilcoxon test, Z = 3.059, N = 12, p =.002).
During field work in 1995 and 1997, we made daily censuses by visual scan at dusk (between 1700 and 1900 h) at 20 areas of meadows (only 14 of them in 1995), half of which were at the ecotone and half of which were in the scrub area (Figure 1). These censuses were done in late August (pre-rut), in mid-September (peak rut), and in early October (post-rut, only in 1997); for analyses we used the arithmetic mean for the days of observation (between 5 and 8 days per monthly observation period) in August, September, and October. From 1995 to 1997 we also carried out a daily census during the rut (2-24 September) at four main mating places at the ecotone. In this case data were used to see the changes over the period, so we used daily totals for the four places.
In the observed areas, we measured the grass quality in August, September, and October. All measurements at each area were done along the same fixed linear transect, at 10 sampling points 10 m apart. At sampling points, and in an area described by approximately 1 m around the point, we randomly dropped a square of 30 x 30 cm divided in four quarters. At every sampling point we measured the following: (1) grass cover: a visual estimate of percent area covered by grass in the square of 30 x 30 cm; (2) green index: the four corner tips of the square touched a leaf blade, which could be either green or brown. The green index was the number of tips at green leaves (from 0 to 4) with respect to the tips touching any blade (0 to 4). If no tip touched a grass blade despite there being some grass cover, we recorded the greenness of the nearest leaf to any of the four tips; (3) grass length: at one point with most cover within the square, we measured the length of the longest aerial part of the grass. To test the relationship between the grass measurements used and biomass, we cut all the aerial parts of the grass included in the square, in a sample of the measures. Grass length times grass cover was highly correlated with dry mass [both variables were transformed by log(x + 1) before analysis; r =.736, N = 40, p =.0001].
The area of observation sites was measured on the basis of digitalized aerial photographs by using Canvas on a Macintosh computer. The surface of the isolated meadows was clearly delimited by the meadow/scrub boundaries. For the ecotone meadows, on the contrary, the surface was taken as the area of meadow under observation, although in many cases the meadow continued farther. Neither vegetation nor surface were measured at sites 11 and 13 because they were narrow strips along the trail where grasses were highly mixed with different plant communities.
At the four main mating areas (numbers 2, 3, 5, and 6 in Figure 1), we made observations of sexual harassment from fixed hidden observatories during the rut in 1996 and 1997. We considered sexual harassment all interactions in which a male approached a female who avoided him, usually by rapid walking or running. Most of these chases were part of the courtship and herding behavior of stags. In gathering data on sexual harassment, we made focal group observations of groups of females or harems of different sizes, recording for every harassment that took place its duration in seconds, the size of the female group, whether the female was included in a harem, and, if so, whether the harassment was done by the harem holder or by any other male (mostly subadult males) and whether the harem had neighboring harems at one side or at two sides (due to the rather linear arrangement of rutting males along the ecotone). Harems were considered neighboring when there was less than 100 m of unoccupied land between them. Observation times ranged between 10 and 60 min (mean 29.80 min in 1996 and 16.84 min in 1997), and were interrupted when a change occurred in the composition of the group. Shorter observations were excluded from the analysis. We assumed that individual females in a group cannot be considered as independent cases. Therefore, to avoid pseudoreplication and ensure statistical independence of cases, we considered as such every group with a particular composition and only once per observation day. Thus, the rate of harassment in that group was considered as a single case and obtained after dividing the observations per unit time and females in the group. Likewise, for each case harassment duration was the mean of the duration of the harassments registered during the observation of one group.
Statistical analysis
Variables were tested for normality by using the Kolmogorov-Smirnov
normality test in Statview 4.5 on a Macintosh computer. Some variables were
normally distributed without transformation because data were arithmetic
means. Total harassment rates were transformed by log[(x +
0.5)
], and duration was log-transformed. When harassment
rates were split into those performed by harem-holders and those by other
males, their distribution departed from normality, hence nonparametric tests
were used in these cases. We avoided the use of parametric tests for very
small sample sizes.
The relationship between variables was tested by using polynomial and linear regressions, according to which of them explained a higher proportion of the variance. The results in the text refer to linear regressions unless otherwise stated. Spearman rank correlation was used for non-normally distributed variables. Probabilities were two tailed, and differences were considered significant at p <.05. Means ± SDs are presented. The analyses were done using Statview 2.5 and 4.5.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Deer numbers at mating areas and nonmating areas
The number of females using the main mating areas in Doñana increased from the early rut in 1995 and 1996, almost doubling by the peak of the rut and decreasing toward the end (polynomial regressions: 1995: r =.663, N = 22, p =.004; 1996: r =.646, N = 22, p =.006; Figure 2). In 1997 the number of females was much lower (see Methods) and variable throughout the days (r =.170, N = 21, p =.768). The same area was used by fallow deer, which compete for the same grass with red deer. To see whether the area was attractive for herbivores in general at this time, we also counted the number of fallow deer using the area during the red deer rut. Fallow deer did not follow the same pattern, increasing to the end of the rut (1995: r =.907, N = 22, p =.0001; 1996: r =.740, N = 22, p =.0001). In fact, the rutting period of fallow deer occurs immediately after that of the red deer, typically overlapping with the end of it, so fallow deer were actually increasing at the beginning of their mating period.
|
When comparing the number of females using the areas before the onset of the rut with those using the same areas during the rut, it appeared that hind numbers increased in some of the areas while they decreased in others, and the opposite trend was observed when comparing the rut and the period after the rut. This resulted in a negative relationship between female increments on both sides of the rutting period for the 20 areas (1997 data: r =.516, N = 20, p =.02). If we look at the relationship between the distribution of females in these areas and the number of adult males that settled there later on (at the beginning of the rut), we should conclude that the number of adult males at mating areas could be predicted by the number of females previously using these areas (1995: r =.667, N = 14, p =.009; 1997: r =.745, N = 20, p =.0002). However, by the time males settled, females moved from some areas to other areas, and there was a positive relationship between female increment and number of rutting males (1995: r =.651, N = 14, p =.012; 1997: r =.801, N = 20, p =.0001). As a consequence, male and female areas overlapped during the rut, and the distribution of one sex now better explained the variance in the distribution of the other sex (1995: r =.887, N = 14, p =.0001; 1997: r =.817, N = 20, p =.0001).
Therefore, the question raised is what benefits do females get by preferring such areas during the mating season? Some not mutually exclusive hypotheses on the potential factors affecting the distribution of females during the rut are examined below.
Traditional mating areas
The distribution of females in the study sites at the peak rut in 1995
correlated with that in 1997 (r =.766, N = 14, p
=.0014), suggesting that the overall distribution of females during the mating
season was consistent from one year to another. This means that females
gathered at some traditional mating sites, although site selection needs not
be arbitrary.
Changes in grass quality
Changes in grass quality did not account for the increment of red deer
females, unless we accept that females were avoiding good-quality meadows. In
particular, changes in cover and greenness of meadows were negatively
correlated with female increment at the onset of the rut in 1997 (cover
increment: r = -.729, N = 18, p =.0006; green
increment: r = -.787, N = 18, p =.0001), and the
grass height was not significantly related to female increment (r =
-.157, N = 18, p =.535). In 1995, data were similar for
green index increment (r = -.652, N = 14, p =.012)
and not significant for cover and height (cover increment: r = -.070,
N = 14, p =.813; height increment: r = -.039,
N = 14, p =.894). In fact, and surprisingly, females
decreased in areas with higher grass quality and increased in areas with lower
grass quality. So female increment at the onset of the rut appeared negatively
related to the values of grass cover and green index during the rut, both in
1995 (cover: r = -.764, N = 14, p =.002; green:
r = -.804, N = 14, p =.0005; height: r =
-.170, N = 14, p =.562) and 1997 (cover: r = -.506,
N = 18, p =.032; green: r = -.557, N = 18,
p =.016; height: r = -.186, N = 18, p
=.459). Thus, the results do not suggest that the movement of females was
caused by changes in the vegetation. On the contrary, females left good
feeding areas when moving to mating areas, suggesting that it could be a cost
for females adopting a suboptimal distribution with respect to food
availability by means of an increase in feeding competition.
Large meadows
Among the observed areas, some (6 in 1995 and 10 in 1997) were isolated
patches of meadows within the scrub, and the remaining (8 in 1995 and 10 in
1997) were at the ecotone. Although there were differences between the size of
our study sites at the ecotone (mean size 24.96 ± 15.26 ha, N
= 10) and those in the scrub (mean size 7.72 ± 11.85 ha, N =
8; Mann-Whitney U test, Z = 2.577, N1 =
10, N2 = 8, p =.010), the main difference was
that the ecotone meadows were continuous over a big surface (see
figure 1).
In 1995, the changes in female density from before to after the onset of the rut were different for areas within the scrub or at the ecotone (Mann-Whitney U test, Z = 3.098, N1 = 8, N2 = 6, p =.0019). In the areas within the scrub, the density of hinds significantly decreased from before to after the onset of the rut (mean decrease -1.37 ± 1.19 females/ha, Wilcoxon test, Z = 2.201, N = 6, p =.028). In the ecotone, the nonsignificant tendency was to increase (mean increment 0.31 ± 0.53 females/ha, Wilcoxon test, Z = 1.400, N = 8, p =.161), and the increment of female density was negatively correlated with the number of fallow deer using the area (rs = -.857, N = 8, p =.023). These results indicate that females avoided small meadows within the scrubland and those containing large numbers of fallow deer.
In 1997, the results were similar, but we were able to compare the changes in the distribution of females not only at the onset of the rut but also at its end. In the isolated scrub meadows, the changes in female distribution differed (Wilcoxon test, Z = 2.805, N = 10, p =.005) between the beginning (mean increment: -0.31 ± 1.40) and the end of the rutting period (mean increment: 1.89 ± 2.75). In the continuous ecotone meadows, the changes were in the opposite direction (mean increment at the beginning: 2.38 ± 3.76; mean increment at the end: -1.62 ± 4.02), although the difference did not reach significance (Wilcoxon test, Z = 1.581, N = 10, p =.114), and the increment of female density was again related to the number of fallow deer using the area (rs = -.758, N = 10, p =.023; Figure 3).
|
The number of adult males in, 1997 increased at the beginning of the rut in the ecotone meadows (mean increment: 1.625 ± 1.332) and decreased at the end of the rut (mean increment: -1.542 ± 1.378), and the difference was significant (Wilcoxon test, Z = 2.805, N = 10, p =.005). In the scrub meadows the changes were slight and not significant (pre-rut: 0.400 ± 1.076; post-rut: -0.033 ± 0.934; Wilcoxon test, Z = 0.059, N = 10, p =.953). This kind of comparison of changes at the beginning and at the end of the rut was not significant for subadult males (ecotone meadows: pre-rut: -1.058 ± 2.469, post-rut: -0.05 ± 0.806, Wilcoxon test, Z = 1.186, N = 10, p =.236; scrub meadows: pre-rut: -0.317 ± 0.647, post-rut: 0.175 ± 0.791, Wilcoxon test, Z = 1.367, N = 10, p =.172).
Also, the changes in female distribution at the beginning of the rut were related to the changes in the distribution of adult males (r =.783, N = 20, p =.0001) and not to changes in the distribution of subadult males (r =.235, N = 20, p =.318), while at the end of the rut, when adult males left the rutting areas, the changes in the distribution of females were not significantly related to those of adult males (r =.364, N = 20, p =.115), but they then appeared related to the changes in the distribution of subadult males that probably follow them at that moment (r =.715, N = 20, p =.0004).
Avoiding harassment
Harassment was performed by any male toward females outside harems and also
by the harem holder toward the females included in his harem. In harems,
females may have received less harassment by external males simply because the
harem holder tried to keep other males away. In the two periods of study,
females received more harassment from non-haremholder males when they were
outside harems than when they were in harems (Mann-Whitney U test:
1996: Z = 5.069, N1 = 34, N2
= 21, p =.0001; 1997: 4.788, N1 = 64,
N2 = 21, p =.0001;
Figure 4). Within harems, on
the other hand, females suffered a higher rate of harassment from the harem
holder than from other males (Wilcoxon test: 1996: Z = 3.776,
N = 32, p =.0002; 1997: Z = 3.191, N = 50,
p =.0014; Figure 4).
In spite of that, however, the total harassment rate was much lower for
females within harems than for females outside harems
(Table 1;
Figure 4). The difference
between the per capita harassment rate for females outside and inside harems
depended on the year (see interaction in
Table 1). It was much higher in
1996 (Figure 4), when there
were more deer in the study area (see
Figure 2) and mean harem size
was higher (7.71 ± 4.46 females per harem in 1996; 3.86 ± 2.45
in 1997; Mann-Whitney U test, Z = 4.211,
N1 = 34, N2 = 64, p =.0001).
The per capita rate at which females were harassed was negatively correlated
with the number of females in the group (1996: rs =
-0.287, N = 55, p =.035; 1997: rs =
-.323, N = 85, p =.003). Incidents of harassment lasted
longer outside harems compared to within harems (1996: outside: 4.17 ±
5.98 min, inside: 2.04 ± 1.74; 1997: outside: 1.22 ± 1.31,
inside: 0.34 ± 1.12; Table
1), and the differences between years were also significant
(Table 1).
|
|
Furthermore, females in a harem suffered a lower harassment rate if the harem was bordered by other harems at one or at both sides (Figure 5; Table 1). Harassment performed by the harem holder and by other males followed the same trend of decreasing as the harems were bordered at one or two sides, although the differences for each kind of male separately did not reach significance for the 2 years included in the study (Kruskal-Wallis tests: 1996: harem holder: H = 11.947, df = 2, N = 32, p =.0025; other males: H = 1.616, df = 2, N = 32, p =.446; 1997: harem holder: H = 0.202, df = 2, N = 64, p =.904; other males: H = 5.904, df = 2, N = 64, p =.052; Figure 5). Mean duration of harassment incidents tended to be longer in isolated harems in 1996 (3.92 ± 2.28 min; 1.40 ± 0.55, and 1.17 ± 0.41 for none, one, or two neighbors, respectively) but not in 1997 (0.15 ± 0.21; 0.59 ± 1.80, and 0.41 ± 0.20 for none, one, or two neighbors, respectively), so the pattern was not clear in this case (Table 1).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The results show that in red deer, not only males but also females collect on mating areas at the onset of the rut, while some other areas previously used by females appear to be avoided during the mating season. It is generally acknowledged that patterns of female distribution predict the mating systems and the distribution of males at the mating period (Davies, 1991
;
Emlen and Oring, 1977). This
appears to be the case of mating systems for many mammals
(Clutton-Brock, 1989),
including red deer (Clutton-Brock et al.,
1982). Previous work in Donana showed that female distribution
predicts the location of male rutting areas and territories
(Carranza et al., 1990). Our
data corroborated this point because the number of females before the rut
correlated with the number of rutting males that settled there the next month,
although this is not the whole story. Areas most heavily used by females were
chosen by males as rutting areas, but a number of females (near 50% in our
data) that remained feeding at other areas moved to the main areas during the
early rut. Changes in space use by females at the time of mating have been
reported for other red deer populations. In the Rum population, for instance,
females restrict their home ranges, making heavier use of short grasses, where
males join them and form harems during the mating period
(Clutton-Brock et al., 1982),
but to our knowledge this has never been studied in detail nor has any
attention been paid to its adaptive meaning.
In Doñana, the changes in female
distribution during the rut cannot be explained by changes in the vegetation.
Red deer in southern Spain feed more on grass when it is green
(Rodriguez-Berrocal, 1978),
and green grass is scarce during the summer and early autumn, so that food
availability in late summer and early autumn becomes a limiting factor for
ungulates (Rodriguez-Berrocal,
1978). Even so, red deer females in
Doñana left some good-quality meadows at the
end of the summer. Mating areas are places with relatively good grass
(Carranza, 1995;
Carranza et al., 1990,
1996), but not all the places
with good grass are used by females at this time, whereas they were otherwise
used before and after the rut. This strongly suggests that there must be other
benefits for this behavior, and as far as such changes in female distribution
coincide with the mating period, the benefits are likely to be related to
mating.
Traditional mating areas selected by red deer in successive years in
Doñana share some common features. First, they
have moderate densities of fallow deer. One likely reason for red deer hinds
to avoid competition with fallow deer is the short height at which a smaller
species, the fallow deer, cuts the sward when grazing, as has been reported
for African antelopes of different body sizes
(Bell, 1970;
Jarman, 1974). Second, main
mating areas are large patches of connected meadows, so that male rutting
areas are continuously distributed over a big surface. Rutting areas of red
deer stags in Doñana include defended
territories (0.5-2 ha) and areas where males tend to remain with their harems
(1-4 ha) (Carranza, 1995;
Carranza et al., 1996).
Avoided meadows were isolated and small, and they could potentially make room
for less than two or three areas of rutting males. This suggests that the
presence of several neighboring rutting males acts as a threshold required by
females to chose an area during the mating season.
But why do red deer hinds collect on areas with several rutting males? This
is a question commonly asked about lekking species, for which it is assumed
that females move to areas with several males to compare among them, and
females gain direct or indirect benefits by comparing potential mates
(Carbone and Taborsky, 1996;
Kirkpatrick and Ryan, 1991;
Queller, 1987). However, other
reasons such as harassment behavior of males may push female ungulates to move
to areas with several neighboring males
(Clutton-Brock et al., 1992,
1993,
1996;
Stillman et al., 1993).
Indirect (genetic) benefits may result either by mate choice or by
promoting male-male competition before accepting a mate. Female choice on
roaring rates has been reported after experiments with farmed red deer
(McComb, 1991), but this does
not necessarily imply any indirect benefits for females. For instance, females
may be selecting good places yielding some kind of direct benefits (food,
safety, etc.) by using male roars as a cue. Male and territory features are
frequently intercorrelated, and it becomes difficult to separate them
(Balmford, 1991). In
Doñana, territorial males compete for best
swards, and when some males were experimentally forced to move to places with
lower grass quality, they lost females in proportion to the losses in grass
quality (Carranza, 1995).
Therefore, we have evidence that females, once in the mating areas, are mainly
selecting feeding sites instead of males. However, because males compete for
good sites, this also suggests that females might be obtaining genetic
benefits for their offspring by choosing the best places in some areas were
neighboring males should compete. This could explain why females should be
reluctant to choose good-quality swards in small areas where the possibilities
for male competition are low. Therefore, that female tactics might be at least
partly modeled by selection based on indirect benefits cannot be ruled out
fully.
On the other hand, females might seek direct benefits. One direct benefit
could be grass, but it cannot explain why females collect on mating areas in
Doñana, provided that there is not a relative
increase of grass quality there compared to other avoided areas. Mating areas
may be at places safer from predation. Although we lack firm evidence that
female ungulates suffer less predation at clusters of males
(Apollonio, 1989;
Balmford and Turyaho, 1992), in
some lekking antelopes, males and females collect in areas with higher
visibility (topi Damaliscus lunatus:
Gosling, 1986;
Gosling and Petrie, 1990;
Uganda kob Kobus kob thomasi:
Deutsch and Weeks, 1992). In
our population, distance to bushes where potential predators could hide is
clearly higher in the ecotone areas or larger meadows. Although the
possibility of an antipredator strategy cannot be ruled out, it seems unlikely
and would not explain why females would avoid the proximity to bushes only
during the mating season.
Another common direct benefit for estrous females in many ungulates is the
avoidance of sexual harassment. This has been recently put forward as one main
reason that females prefer to collect on clumps of male territories or leks
(Clutton-Brock et al., 1993).
For example, there is evidence in kafue lechwe Kobus leche that
disruption of mating sequences was more frequent in mixed herds compared to
disruption in leks (Nefdt,
1995) and that fallow does forced to move from leks were chased
more by young bucks (Clutton-Brock et al.,
1992). Unfortunately, we have no data on harassment outside the
mating areas, which would have provided more direct information on why females
avoid such areas, partly because little sexual behavior can be observed there.
Nevertheless, our data indicate that females are less harassed when included
in a harem, that the per capita levels of harassment are inversely related to
harem size, and that females in harems are less harassed when surrounded by
other harems, so they should prefer harems in large meadows instead of
isolated harems in a small meadow within the scrubland. The avoidance of
sexual harassment can therefore explain female preferences for areas with
several neighboring rutting males.
Our results may also have some implications for current ideas on the origin
of lek breeding in ungulates. Different reasons such as female preferences for
particular males and safe places (predation or harassment) might explain why
females change their space use and collect on leks
(Clutton-Brock et al., 1993,
1996). Female choice for
particular males and differences in male ability to retain females can be
ruled out as relevant in Doñana, because when
males were experimentally moved from their territories, they lost females in
the proportion of their losses in grass quality
(Carranza, 1995). Then, safe
places (from a harassment point of view) appears to remain as the more likely
reason for female collecting on male clumps in
Doñana, although eventual indirect benefits
from mating with highly competitive males cannot be discarded. This suggests,
therefore, that sexual harassment even without female mate choice for
particular male phenotypes may be enough to explain why females collect on
clumps of male territories or leks. But in addition, harassment outside the
clumps of male territories does not necessarily lead to a lek-king system. One
possible view is that the more clumped the male territories are, the greater
the benefits from harassment avoidance, but the greater the costs from feeding
competition, especially when food is as scarce as it is during the red deer
rut in southern Spain. As a consequence, territories remain more spaced and
linked to food distribution than for ungulate leks. This view is further
supported by data from some red deer populations with extremely high local
densities, where a number of male territories are very small and contained no
food resources, producing a leklike male clump
(Carranza, 1992;
Carranza et al., 1995).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We thank Miguel Delibes and Miguel Ferrer, consecutive directors of the Estación Biológica de Doñana, and people at the Doñana Biological Reserve for facilities in the course of the study. Marta Aguilar, Antonio Castellanos, Camilo González, Nieves González, Natalia López, Gregorio Rocha, Marta Rueda, Cristina Sánchez, and Jerónimo Torres assisted in field work. Marco Festa-Bianchet and two anonymous referees made comments on earlier versions of the manuscript. Financial support was from Spanish Ministry of Education, DGICYT project PB94-1028.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Allier C, González F, Ramirez L,1974 . Mapa Ecológico/Ecological map. Reserva Biológica de Doñana. Sevilla: Estación Biológica de Doñana.
Ahnesjö I, Vincent A, Alatalo R, Halliday T, Sutherland WJ, 1992. The role of females in influencing mating patterns. Behav Ecol 4:187-189.
Apollonio M, 1989. Lekking in fallow deer: just a matter of density? Ethol Ecol Evol 1:291-294.
Arnold SJ, Duvall D, 1994. Animal mating systems: a synthesis based on selection theory. Am Nat 143:317-348.[Web of Science]
Balmford AP, 1991. Mate choice on leks. Trends Ecol Evol 6:87-92.
Balmford AP, Read AF, 1991. Testing alternative models of sexual selection through female choice. Trends Ecol Evol 6:274-276.
Balmford AP, Turyaho M, 1992. Predation risk and lek-breeding in an African antelope. Anim Behav 44:117-127.
Bateman AJ, 1948. Intra-sexual selection in Drosophila. Heredity 2:349-368.[Web of Science][Medline]
Bell RHV, 1970. The use of the herb layer by grazing ungulates in the Serengeti. In: Animal populations in relation to their food resources (Watson A, ed). Oxford: Blackwell Scientific Publications; 111-124.
Carbone C, Taborsky M, 1996. Mate choice or harassment
avoidance? A question of female control at the lek. Behav Ecol
7:370-378.
Carranza J, 1992. Lekking in red deer?—A comment on the concept of lek. Etología 2:83-90.
Carranza J, 1995. Female attraction by males versus sites in territorial rutting red deer. Anim Behav 50:445-453.
Carranza J, Alvarez F, Redondo T, 1990. Territoriality as a mating strategy in red deer. Anim Behav 40:79-88.
Carranza J, Fernandez-Llario P, Gomendio M, 1996. Correlates of territoriality in rutting red deer. Ethology 102:793-805.[Web of Science]
Carranza J, García-Muñoz AJ, Vargas JD, 1995. Experimental shifting from harem defence to territoriality in rutting red deer. Anim Behav 49:551-554.
Clutton-Brock TH, 1989. Mammalian mating systems.Proc R Soc Lond 236:339-372.[Medline]
Clutton-Brock TH, Albon SD, 1989. Red deer in the Highlands. Oxford: BSP Professional Books.
Clutton-Brock TH, Deutsch JC, Nefdt JC, 1993. The evolution of ungulate leks. Anim Behav, 46:1121-1138.
Clutton-Brock TH, Guinness FE, Albon SD, 1982.Red deer. Behaviour and eecology of two sexes . Edinburg: Edinburgh University Press.
Clutton-Brock TH, McComb KE, Deutsch JC, 1996.
Multiple factors affect the distribution of females in lek-breeding ungulates:
a rejoinder to Carbone and Taborsky. Behav Ecol
7:373-378.
Clutton-Brock TH, Price OF, MacColl ADC, 1992. Mate
retention, harassment, and the evolution of ungulate leks. Behav
Ecol
3:234-242.
Clutton-Brock TH, Vincent ACJ, 1991. Sexual selection and the potential reproductive rates of males and females.Nature 351:58-60.[Medline]
Davies NB, 1991. Mating systems. In:Behavioural ecology (Krebs JR, Davies NB, eds). Oxford: Blackwell Scientific; 263-294.
Deutsch J, Weeks P, 1992. Uganda kob prefer
high-visibility leks and territories. Behav Ecol
3:223-233.
Emlen ST, Oring LW, 1977. Ecology, sexual selection
and the evolution of mating systems. Science
197:215-223.
Gosling LM, 1986. The evolution of mating strategies in male antelopes. In: Ecological aspects of social evolution (Rubenstein DI, Wrangham RW, eds). Princeton, New Jersey: Princeton University Press; 244-281.
Gosling LM, Petrie M, 1990. Lekking in topi: a consequence of satellite behaviour by small males at hotspots. Anim Behav 40:272-287.
Höglund J, Alatalo RV,1995 . Leks. Princeton, New Jersey: Princeton University Press.
Jarman PJ, 1974. The social organization of antelope in relation to their ecology. Behaviour 48:215-267.[Web of Science]
Jarman PJ, 1983. Mating system and sexual dimorphism in large, terrestrial, mammalian herbivores. Biol Rev 58:485-520.
Kirkpatrick M, Ryan MJ, 1991. The evolution of mating preferences and the paradox of the lek. Nature 350:33-38.
Lundberg A, Alatalo R, 1992. The pied flycatcher. London: Poyser.
McComb K, 1991. Female choice for high roaring rates in red deer, Cervus elaphus. Anim Behav 41:79-88.
Nefdt RJC, 1995. Disruptions of matings, harassment and lek-breeding in Kafue lechwe antelope. Anim Behav 49:419-429.
Queller DC, 1987. The evolution of leks through female choice. Anim Behav 35:1424-1432.
Rodriguez-Berrocal J, 1978. Introducción al estudio y valoración de los recursos forestales y arbustivos para el ciervo en el área ecológica de Sierra Morena: I. Estudio de la dieta del ciervo. Arch Zootec 27:73-82.
Rogers PM, Myers K, 1980. Animal distributions, landscape classification and wildlife management. Coto Doñana, Spain. J Appl Ecol 17:545-565.
Stillman RA, Clutton-Brock TH, Sutherland WJ, 1993.
Black holes, mate retention, and the evolution of ungulate leks. Behav
Ecol
4:1-6.
Trivers RA, 1972. Parental investment and sexual selection. In: Sexual selection and the descent of man 1871-1971 (Campbell B, ed). Chicago: Aldine;136 -179.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. M. Hamilton, M. P. Haesler, and M. Taborsky Predators, reproductive parasites, and the persistence of poor males on leks Behav. Ecol., January 1, 2006; 17(1): 97 - 107. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||