Viper Ax

HOLARCTIC ECOLOGY II: 77-80. Copenhagen 1988

Reproductive success, mortality and sexual size dimorphism in the adder, Vipera berus Thomas Madsen

Madsen, T, 1988. Reproductive success, mortality and sexual size dimorphism in the adder, Vipera berus. - Holarct. Ecol. 11: 77-80, From 1981 to 1986 an isolated adder population was studied in the extreme south of Sweden. During this period 48 adult males and 44 adult females were marked, Male adders did not grow as large as the females. Large males had a significantly higher annual mating success and were engaged in more combats than smaller males. The mean length of recaptured males was significantly lower than that of those not recaptured, indicating a higher motality of larger males. Females brood size was positively correlated with body size. In females there was no difference in mean length of recaptured vs not recaptured individuals. The adder is one of the few snake species with male combat where males are smaller than females. I suggest that this is due to stronger selective advantages for large body size in females than in males, T. Madsen, Dept of Ecology, Animal Ecology, University of Lund, Ecology Building, Hetgonav. 5, S-223 62 Lund, Sweden.

1. Introduction

Darwin (1871) suggested that sexual selection could result in the evolution of characteristics leading to an "advantage which certain individuals have over other individuals of the same sex and species, in exclusive relation to reproduction". Thus males' competition for females could result in selection for fighting abilities. If large males are more successful than smaller males and size is heritable then male size will increase until balanced by natural selection. As pointed out by Greenwood and Adams (1987) "extrapolation of Darwin's ideas have led to a matching pair of fallacious assumptions. First, if large size is advantageous in male-male competition then males must inevitably evolve to be larger than females. Second, if males are larger than females then this must inevitably be because of the advantage of large male size in male-male competition. Shine (1978) found a high correlation between the occurrence of male combat in snakes, and sexual size dimorphism in which the male was the larger sex. He suggested that the result strongly supported the hypothesis that large male size was an adaptation to malemale competition. In the adder, however, male combat

is well documented but still the males do not grow as large as the females (e,g, Andren 1986, Madsen 1987). Other explanations than male combat are obviously required to explain the sexual size dimorphism in the adder. Here I present data on male-female mortality and reproductive success, suggesting a stronger selective advantage in large size in female as compared with male adders. 2. Study area, materials and methods

The study area is situated in the absolutely southernmost part of Sweden (55°20'N; 13°22'E). It is approximately 800 X 50 m, bordered to the south by the Baltic sea, to the north by arable land, and to the east by a barbour. A village is situated about one km west of tbe area, Tbe nearest known adder population is found to the north, across 20 km af arable land. Thus the adder population is totally isolated. The vegetation consists of grass, herbs and some small stands of trees and bushes. Field work was carried out during April and May from 1981 to 1986 for a mean number of 22 days annually. The adders were collected by hand. Snout-vent and total length, weight and sex were recorded. The adders

Accepted 14 September 1987 ©

IIOLARCIIC ECOLOGY

© HOLARCTIC ECOLOGY

77

were individually marked by branding their ventral scales with a battery-charged soldering iron. In addition all snakes were marked on their dorsal scales with a red paint that permitted individual identification from a distance. Males with a snout-vent length of at least 40 cm were recorded courting females and were thus considered as adults. Altogether 48 adult males were marked. The small size of the study area and the fact that male adders are quite conspicuous during spring made them very easy to capture. During no year were unmarked adult males recorded after the thirteenth visit to the study area. I am therefore confident that almost 100% of the adult males were captured each year. Due to the high capture efficiency, and the fact that no male was ever recaptured without having been captured the previous year, a male was considered dead if he was not recaptured during a subsequent year. The smallest female recorded mating in the study area was 45 cm (snout-vent length). Females more than 45 cm were thus considered as adults. Altogether 44 adult females were captured. During spring female adders are much less conspicuous and harder to find than males (pers. obs.) and in contrast to the males, several females were recaptured every second year. Therefore female mortality is based on females captured from 1981 to 1984; if neither recaptured during 1985 nor 1986 a female was considered as dead. Data on male mating success are based on field work carried out from 1984 to 1986. In order to keep a continuous record of matings, females were monitored with 27 MHz radio-transmitters. The weight of the transmitters was 3.5 g and the size 20 x 12 X 8 mm. They were force-fed to the females which were released some after capture. During the mating seasons in 1984, 1985 and 1986,11,7 and 9 females, respectively, were monitored. The mating season starts when the males have shed for the first time of the year (Nilson 1980, Madsen 1987). The days immediately prior to shedding the snakes' eyes are opaque. When the first males showed this sign, I visited the study area daily and continued until no matings had been recorded for 4-6 days. During 1984 the study area was visited from 9 am to 6 pm. However, because of the mornings were chilly no matings were observed before 12 am. Therefore, in 1985 and 1986, the visits lasted from 11 am to 6 pm, unless bad weather reduced snake activity. Each female was located every 40 min. Median copulation time was one hour (pers. obs.) and thus I assume that most matings of the monitored females were recorded. Altogether 51 matings were recorded. Female brood size was recorded from seven females from the study area and from 15 females collected in other adder populations in southern Sweden. The females were kept in the laboratory when giving birth. Additional data on female brood size were obtained from Bernstrom (1943) and from females at the Museum of Zoology in Lund. 78

3. Results

Female adders grew to a larger size than did males (Fig. 1). The ten largest females were significantly larger than the ten largest males (t = 4.78, p<0.001). There was, however, a great variation in male growth rate (Fig. 1). Some males, first captured when 38-42 cm (snout-vent length), attained a length of >50 cm within 2-4 yr. Other males, initially captured at the same

40

Fig. 1. Annual growth of 8 male (dashed- and continuous lines) and 5 female adders (dotted lines).

en c

b

<= 5 -

° 4 321-

40

45

50 Snout-vent length (cm)

55

Fig. 2. Number of male matings in relation to size (Spearman rank correlation, t = 2,52, p<0.02, df = 46; df equal to number of different males, t calculated according to Siegel (1956)). Circle size denotes 1-15 males. © HOLARCTIC ECOLOGY

Tab. 1. Mean length (snout-vent (cm)) of recaptured and not recaptured male and female adders. Mean length of not recaptured

Mean length of recaptured

Males Females

adders

SD

45.5 50.4

3.20 3.35

length, had only reached 44-46 cm during the same period (Fig. 1). There was a significant positive correlation between male size and number of matings (Fig. 2). There was also a significant positive correlation between male size and number of combats (Fig. 3). The mean length of

96 42

adders

SD

47.1 50.4

2.98 3.02

23 26

2.36 0.01

p<0.05 n.s

Tab. 2. Mean annual log weight/log length ratio of male adders captured in 1981 to 1986.

Mean Log weight/log length SD N

1981

1982

1983

1984

1985 1986

1.12 0.03 19

1.10 0.08 17

1.13 0.05 23

1.14 0.05 29

1.12 0.05 25

1.11 0.04 25

F = 1.75, d f = 5, 132, p>0.1

•

8 o d

recaptured males was significantly lower than those that were not recaptured, indicating a higher mortality of the larger than of the smaller males, whereas there was no such difference in females (Tab. 1). There was a significant positive correlation between female size and brood size (Fig. 4). There was no significant difference in mean annual log weight/log length ratio of males adders captured during 1981 to 1986 (Tab 2).

• •

3]

1

55

50

45

40

Snout - vent length (cm) Fig. 3. Number of combats in relation to male size (Spearman rank correlation, t = 3.09, p<0.01, df = 46; df equal to number of different males, t calculated according to Siegel (1956)). Circle size denotes 1-7 males.

15

45

50

55

60

65

Snout-vent length (cm) Fig. 4. Female brood size in relation to body length (r = 0.41, p<0.01, df = 42). Circle size denotes 1-4 females. © HOLARCTIC ECOLOGY

4. Discussion

In contrast to most other "combat species" where males in most cases grow to a larger size than females (Shine 1978), the male adders did not grow as large as the females (Fig. 1). Large males performed a significantly higher number of matings than did small males (Fig. 2). Selection would thus favour males growing to a large size. However, the mean length of the recaptured males was significantly lower than those not recaptured (Tab. 1) indicating an increased mortality of the larger compared with the smaller males. Due to the great difference in growth rates between the fast and the slow growing males, size is not a good indicator of male age (Fig. 1). Thus the increased mortality of large males was most likely not just a result of them being older. Food shortage have been suggested to increase the risk of mortality of large as compared with small male adders (Madsen 1987). However, there was no significant difference in mean annual relative thickness of the males (Tab. 2). This indicates that the nutritional condition of the male adders did not differ between years. Thus the relatively higher mortality of larger males could hardly be a result of reduced prey availability. During the mating season the males were very active whereas the females were quite sedentary (Madsen 1987). This resulted in that most of the females, and 79

thus male mating activity, was restricted to a fairly small area which could attract predators. During combats, males were extremely conspicuous and they often did not react when approached (pers. obs.) which certainly could increase the risk of predation. On two occasions were crows, Corvus corone, recorded attacking male adders during the mating season (both males had a snout-vent length of 50 cm). As the larger males were engaged in more combats than the smaller males (Fig. 3), the higher mortality rate of larger, than of smaller males might be a result of increased predation on large males associated with their high combat activity. Thus selection for large body size could be countered by an increased mortality of the larger males. In southern Sweden the adders' mating season normally lasts for three weeks (pers. obs.). The short mating season is most likely an adaptation to the short activity season in northern latitudes, and the females' need to give birth before autumn. A short mating season could prevent the largest males from monopolizing all receptive females and thus increase the chances for smaller males to mate. This would decrease the advantage of large male size as a mean of getting access to females. Consequently, the increased mortality of large compared with small males and the short mating season will most likely reduce the life-time reproductive success of fast growing as compared with the slow growing males. As in many other species of snakes, e.g. grass snakes Natrix natrix (Madsen 1983), the brood size of female adders was positively correlated with size (Fig. 4). As for the males, selection could thus favour females growing to a large size. However, in contrast to the males, there was no difference in mean length of recaptured vs not recaptured females which suggests that the mortality rate of large females was not higher than for small ones (Tab. 1). Thus for female adders there was no cost (measured in terms of increased mortality) in growing to a large size. Consequently, in females, increased life-time reproductive success is positively correlated with increased body size. Why do not female adders grow even larger than they do? As mentioned above, food shortage has been reported to increase the relative mortality of large male adders (Madsen 1987), therefore reduced prey availability will most likely also reduce the advantage of large female size. Consequently female

80

size could be a result of a selective balance between increased reproductive output and an increased risk of mortality caused by food shortage. Both adult male and adult female adders in the study population feed on field voles Microtus agrestis. Thus the difference in size between males and females can hardly be an adaptation to reduce competition for food between the two sexes. For the adder I suggest that the increased mortality on large males and a short mating season, could decrease the selective advantage of large body size in males as compared with females. The results thus support Shine's (1978) suggestion that in the "combat species", where males are smaller than females this could be a result of the fact that "other factors select for even larger female size" and thus "the mere presence of male combat in a species may not automatically select for large male size". Acknowledgements - I thank J. Loman, S. Eriinge, R. Shine, M. Osterkamp and T. Halliday for comments on the manuscript. I am also very greatful to T. Hakansson and J. Loman for their help with the field work during 1984. The work was partly supported by grants from the Royal Swedish Academy of Science. References Andren, C. 1986. Courtship, mating and agonistic behaviour in a free-living population of adders, Vipera berus. - Amphibia-Reptilia 7: 353-383. Bernstrom, J. 1943. Till kannedom om huggeormen Vipera berus berus (Linn6). - Goteborgs Kungl. Vetenskaps- och Vitterhets-Samhalles Handlingar. Sjatte foljden. Ser. B. Band 2. No. 10. (meddelanden fran Goteborgs Musei Zoologiska Avdelning 103): 2-43. (In Swedish). Greenwood, P. J. and Adams, J. 1987. Sexual selection, size dimorphism and a fallacy. - Oikos 48: 106-108. Darwin, C. 1871. The descent of man and selection in relation to sex. - John Murray, London. Madsen, T. 1983. Growth rates, maturation and sexual size dimorphism in a population of grass snakes, Natrix natrix, in southern Sweden. - Oikos 40: 277-282. - 1987. Natural and sexual selection in grass snakes, Natrix natrix, and adders, Vipera berus. - Ph.D. thesis. University of Lund. Nilson, G. 1980. Male reproductive cycle of the european adder, Vipera berus, and its relation to annual activity periods. - Copeia 1980: 729-737. Shine, R. 1978. Sexual size dimorphism and male combat in snakes. - Oecologia (Berl.) 33: 269-277. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. - McGraw-Hill Kogakusha, Ltd, Tokyo.

© HOLARCTIC ECOLOGY