SELECTIVE PREDATION FOR VERTEBRAL PHENOTYPE IN GASTEROSTEUS ACULEATUS: REVERSAL IN THE DIRECTION OF SELECTION AT DIFFERENT LARVAL SIZES

Variation in the number of vertebrae is widespread in fishes, and is partly genetic in origin. The adaptive significance of this variation was tested by exposing larvae of the threespine stickleback (Gasterosteus aculeatus) to predation by sunfish (Lepomis gibbosus). Two vertebral characters were considered: the total number (VN) and the ratio of abdominal to caudal vertebrae (VR). Predation was selective for both characters, but selection was more directly related to VR than to VN. The direction of selection depended on larval length: as length increased, optimal VR decreased. Total selection for VR was a combination of direct selection and an indirect effect of selection acting on a correlated trait, the ratio of precaudal to caudal length. Direct and indirect selection were in opposing directions at a given larval length. Variation in vertebral number may be maintained in populations partly because the strength of selection is reduced by opposing directions between direct and indirect selection, and between total selection at different larval lengths.     

THE EVOLUTIONARY ECOLOGY OF AN ANTIPREDATOR REACTION NORM: DAPHNIA PULEX AND CHAOBORUS AMERICANUS

Ponds containing the parthenogenetic zooplankter Daphnia pulex with and without chaoborid predators were sampled over the course of a season. A significant (P < 0.05) Spearman rank correlation was found between predator density and the expression of an antipredator defense (neckteeth) by the Daphnia. The reaction norms (percent induction of a single genotype versus predator density) of clones isolated from predator-free and predator-rich habitats were determined in a laboratory setting. There was a statistically significant different response among the six clones tested (P < 0.05). Clones isolated from chaoborid ponds showed significantly greater sensitivity to the presence of predator than clones from predator-free ponds (P < 0.05). In the laboratory, food levels under which prey were cultured affected induction of the antipredator response. Highest induction was found at the lowest food level used.     

Predation impact of Leptodora kindtii on population dynamics and morphology of Bosmina fatalis and B. longirostris in mesocosms

1. We assessed the impact of predation by the invertebrate predator Leptodora kindtii on Bosmina longirostris and B. fatalis, which show seasonal and reciprocal succession patterns in Lake Suwa, in a mesocosm experiment using 20‐L tanks with or without the predator under different food conditions. We also analysed morphological responses of the two Bosmina species to the predator in the tanks. 2. Bosmina fatalis dominated B. longirostris regardless of predator presence under high food density. However, the presence of Leptodora induced the dominance of B. fatalis more rapidly than its absence. On the contrary, no dominance of B. fatalis was observed in tanks with low food density, irrespective of the presence of the predator. Only B. fatalis showed morphological changes in response to the presence of Leptodora. 3. Mucrone length and antennule shape (angle between body and antennule and angle between antennules) showed marked responses at both high and low food densities, but antennule length responded only at high food density. Mucrone length seems to be a more effective defence against Leptodora. 4. The results suggest that B. fatalis is a superior competitor against B. longirostris and is more resistant to Leptodora predation, especially in good food conditions. The repeatedly observed seasonal succession of the two Bosmina species in the eutrophic Lake Suwa – the replacement of B. longirostris by B. fatalis following the occurrence of abundant Leptodora– seems to be caused by the selective predation of Leptodora on B. longirostris as well as the competitive ability of B. fatalis.     

Behavioral responses of a sex-role reversed pipefish to a gradient of perceived predation risk

Conspicuous behaviors such as courtship and mating often makeanimals susceptible to predation. When perceiving themselvesat an elevated level of risk, animals frequently reduce conspicuousbehaviors in tradeoff for a decrease in probability of beingpreyed upon. In the present study, we used two experiments toexamine the effect of perceived predation risk from cod (Gadusmorhud) on nonreproductive and reproductive behaviors in thesex-role reversed pipefish (Syngnathus typhle). In the firstexperiment, no differences due to predation risk were detectedin the nonreproductive behaviors of either males or females.In the second experiment, predation risk had significant effectson reproductive behaviors. Pipefish were allowed to court andcopulate at four different predation levels. We created predationlevels differing in perceived predation risk by controllingthe number of sensory modes through which pipefish could detectthe presence of a cod. As predation risk increased, pipefishcopulated and courted less frequently, swam alone (displayedand searched for conspecifics) less often, and waited longerbefore commencing courtship. These changes in behavior minimizedthe amount of time spent above the eelgrass and presumably reducedconspicuousness to visual predators. Pipefish also copulatedafter a smaller amount of courtship as predation risk increased,indicating that they may trade information concerning mate qualityfor a reduction in predation risk. No differences were foundin any response variable between males and females. The roleof operational sex ratios and intersexual competition in determiningwhich sex assumes greater costs in mate acquisition is questioned.     

The adaptive significance of colour pattern polymorphism in the Australian scincid lizard Lampropholis delicata

Females of Lampropholis delicata are dimorphic for colour pattern, the difference between morphs being the presence or absence of a distinct white mid-lateral stripe. A less distinct striped morph occurs also in males. We evaluated alternative hypotheses for the maintenance of this polymorphism by examining temporal and spatial variation in morph frequency, testing for differential selection among morphs using data on body size and reproductive traits from preserved specimens, and experimentally manipulating colour pattern in free-ranging lizards of both sexes, to assess the influence of the lateral stripe on survival rates. We found that the relative frequency of striped individuals varied among populations and decreased from north to south in both sexes, coincident with an increasing incidence of regenerated tails. Morph frequencies did not change through time within a population. Striped gravid females appeared to survive better and produced larger clutches than did non-striped females. In our experimental study, the relationship between survival and colour morph differed between the two sexes; males painted with a white lateral stripe had lower survival than control (brown stripe) males, but survival did not differ between striped and control females. The different response in the two sexes may be due partly to differences in temperature and microhabitat selection. We propose that the white lateral stripe decreases susceptibility to predators in gravid females but increases risk of predation in males, especially in combination with low temperatures. The polymorphism might be maintained by: (1) opposing fitness consequences of the stripe in males and females; (2) sex-specific habitat selection; and (3) gene flow in combination with spatial variation in relative fitness of the two morphs.     

Demography and life history of free-rangingPropithecus diadema edwardsi in ranomafana national park,madagascar

I conducted the first long- term study of the life history patterns of Propithecus diadema edwardsi—Milne- Edward’s sifaka— in the rain forests of southeastern Madagascar, beginning in 1986. I report behavioral observations on a total of 33 individuals from three groups over a 9- year span. We captured,marked, and released 21 individuals. Individual group size ranged from three to nine sifakas. Two breeding females lived in groups I and II until 1993. A newly formed group (III) had one breeding female. Age at first reproduction is 4 years for females and 5 years for males. Gestation length is 179 days (n =2). Most births occurred in June (n = 17), but infants were also born in May (n = 2) and July (n =2). Nine of 21 (43%) infants born died before the age of 1 year, and 15 (67%) died before the age of reproduction. One female bred in her natal group after the death of the resident male and the immigration of an adult male. Another two females disappeared at 4 and 5 years of age;they could have emigrated or died. All 5- to- 6- year- old males (n = 4) have emigrated from their natal groups to adjacent groups. Two have committed infanticide. Five or more individuals were killed by Cryptoprocta ferox.Despite high mortality and offspring dispersal, the number of individuals in the two main groups remained nearly the same over the 9- year study.     

Intraspecific competition in immature Amblyseius fallacis,Amblyseius andersoni,Typhlodromus occidentalis and Typhlodromus pyri (Acari: Phytoseiidae)

Intraspecific competition in immature Amblyseius fallacis, Amblyseius andersoni, Typhlodromus occidentalis and Typhlodromus pyri was examined in the laboratory using small cages at five different predator densities (two, four, eight, 16 and 32) in the absence and presence of prey 100 eggs of two-spotted spider mite, Tetranychus urticae (Koch), at 25 ± 1°C, 80% RH and 16L:8D photoperiod. In the absence of spider mite prey, some individuals of immature phytoseiids showed increased development and surival with increasing predator densities up to certain limits, but none survived to the adult stage, except for a single male each of A. andersoni and A. fallacis who completed development by cannibalizing on conspecifics at a density of 32 predators per cage. In the absence of spider mite prey, the mean immature survival time was independent of the initial predator density, but the variance of survival time increased with predator density. In the presence of prey, the proportion of immatures surviving to adulthood generally decreased with initial predator density and dropped sharply to almost none at the predator density of 32 for A. fallacis, eight for A. andersoni, 16 for T. occidentalis and four for T. pyri. The number of prey consumed per predator during the first day generally decreased with predator density in all four species, as prey available per predator decreased and the competition for food increased with predator density. Our data indicate that scramble competition is operating in these four species. Although cannibalism was occasionally observed, especially after the exhaustion of prey and in the generalist predators such as A. andersoni, the immatures of these phytoseiids were less influenced by the interference of conspecifics than by the increasing difficulty of finding food at high predator densities. The implications of this study for understanding phytoseiid population dynamics and their use in biological control are discussed.     

Habitat-specific predation susceptibilities of a littoral rotifer to two invertebrate predators

The rotifer Euchlanis dilatata lives associated with submerged vegetation in the littoral zone of freshwater lakes and ponds. I assessed habitat-specific predation susceptibilities for this rotifer in the presence of three aquatic macrophytes (Myriophyllum exalbescens, Elodea canadensis, and Ceratophyllum demersum) and two predators (damselfly nymphs — Enallagma carunculata; and cnidarians — Hydra). Rotifer survival was greatest on Myriophyllum in the presence of both predators. Conversely, the presence of the other macrophyte species actually increase rotifer suspectibility to predation by damselfly nymphs. I also manipulated plant structural complexity. As predicted, decreasing the relative complexity of each plant resulted in lower rotifer survival.     

Variation in growth and age at maturity in bluegill sunfish: genetic or environmental effects?

Populations of bluegill sunfish Lepomis macrochirus , experiencing heavy juvenile predation, showed increased growth rates and increased age and size at maturity relative to populations experiencing decreased predation on juveniles but increased predation on adults. This study examined bluegills experimentally from both types of populations and a cross between them in a common environment to determine if variation in growth and age at maturity is genetically or environmentally induced. Two factorial experiments, varying strain of bluegills and resource availability, were used to evaluate differences in growth rate. One experiment, varying strain of bluegills, was used to assess differences in age at maturity. Growth was strongly influenced by resource level, but growth rate did not vary among populations. Nearly all bluegills in each population matured at 1 year of age in a common environment. Thus, variation observed in source populations must be mostly attributable to differences in the environment between populations. At least three factors could potentially cause differences in growth and age at maturity: (1) variation in resource availability; (2) variation in demographic structure; and (3) variation in size-specific mortality rates caused by differences in predator abundance between populations. Observed patterns of variation between populations are best explained by effects of differences in predator populations.     

THE IMPACT OF TREE SQUIRRELS (TAMIASCIURUS) ON LIMBER PINE SEED DISPERSAL ADAPTATIONS

Tree squirrels (Tamiasciurus) are important selective agents on conifer reproductive strategies (Smith 1970, 1975). Although this is well established for wind-dispersed pines, the impact of tree squirrels on bird-dispersed pines has been largely ignored. I assessed the impact of tree squirrels on the allocation of reproductive energy in the bird-dispersed limber pine (Pinus flexilis) by comparing its cone and seed traits from three sites in the Rocky Mountains where tree squirrels (Tamiasciurus) are present to those from three mountain ranges in the Great Basin where tree squirrels are absent. As predicted, differences between the two regions in individual cone and seed traits are consistent with the hypothesis that tree squirrels are important selective agents on these traits. In the absence of tree squirrels, limber pine allocates more than twice as much energy to kernel compared with that invested in putative seed defenses (cone, resin, and seed coat) as does limber pine where tree squirrels are present. Such a large difference is particularly striking, because tree squirrels may have become extinct in the Great Basin in only the last 12,000 yr. Although many factors influence the allocation of energy to cones and seeds, no single factor other than the presence of tree squirrels is compatible with the large and consistent differences between limber pine in the Rocky Mountains and Great Basin. These results show that tree squirrels are an important constraint on the evolution of cone and seed traits that promote the dispersal of seeds by birds.