Changing oviposition times of the crab spider Misumena vatia (Thomisidae) correlate with climate change.
In spite of the currently perceived importance of documented climatic change and much anecdotal information for many groups of organisms, rather few long-term measures of organismal response to yearly change currently exist over multiple trophic levels (Clark & May 2002; LeRoy et al. 2013). To the best of my knowledge no such information exists for spiders. However, in the process of a long-term study of the crab spider Misumena vatia (Clerck, 1757) (Thomisidae) in coastal Maine, USA (Morse 1979, 2007, 2014), I maintained large populations of adult females over a 32-year period (1979-2010) that allowed me to record initial dates of oviposition for each year. For 16 of these years, I also obtained initial flowering dates of their most important hunting site, inflorescences of common milkweed Asclepias syriaca, which attract great numbers of large insects that provide the main resources for Misumena reproduction (Morse 1982a; Morse & Fritz 1982). For most of the flowering years I also recorded the dates at which bumblebees, the spiders' most important prey, first recruited to the milkweed flowers. This body of information allowed me to establish whether the activity patterns of these three species shifted in accordance with the changing seasons and whether any such changes affected their relationship to each other. In turn, the results allowed me to assess the effects of climate change on these species over the period of study, as well as to project likely future consequences for their relationships.
Misumena vatia is a small sit-and-wait predator that typically hunts on flowers. Females weigh from 35-55 mg upon reaching adulthood, usually in the early summer following hatching in late summer two years earlier. They average 220 mg at laying their single clutch of eggs, with a range of 115 mg to over 400 mg. Their single clutch weighs approximately 65% of their mass immediately prior to egg laying (Morse 2007).
Each year, I maintained 50 or more field-collected adult females in order to gather various life history variables on the population and to obtain young for a wide range of experiments reported elsewhere. I collected the spiders in fields and roadsides at and near the Darling Marine Center, South Bristol, Lincoln County, Maine, U.S.A. (43[degrees]57'N, 69[degrees]33'W), mostly from common milkweed and pasture rose Rosa Carolina, with smaller numbers from ox-eye daisy Leucanthemum vulgare, common buttercup Ranunculus acris, cow vetch Vicia cracca and red clover Trifolium pratense.I sampled these adults independently of mass, but the sample size ensured that they included individuals likely to become among the earliest ovipositers. I maintained them in 7-dram vials (5 cm long, 3 cm diameter) at ambient temperature and natural day length in an unheated laboratory and fed them moths or large flies every other day. Generally these spiders will not take large prey on successive days (Morse 2007), so the feeding regime resembled that obtained at a high-quality hunting site like that of common milkweed.
The spiders typically refuse to feed one to two days before laying. At that point, I placed them on a milkweed leaf in the field, which resembled a natural nest site, and enclosed the plant with a loosely fitting nylon tricot bag. Most individuals laid either the following night or one night later. I resorted to the field regime because these spiders appeared reluctant to lay in the vials where they could not construct a normal nest. The bags themselves do not significantly affect the temperature within (Morse 1994a). I checked the spiders daily to ensure an accurate laying date.
Additionally, I recorded first observations of egg masses while conducting unrelated fieldwork between 1980 and 1994. They provided a comparison between laying dates in the field and those in the standardized routine described above.
I also recorded the initial flowering dates of milkweed, the most important hunting site for Misumena in my study area, during the 16 years (1979-1994) for which I possessed data on the spiders' oviposition date. A perennial, common milkweed grows in clones that vary greatly in size, but that routinely attract large numbers of bees, moths and butterflies when in bloom (Morse 1982a, 2007). Milkweed stems vary from 80-120 cm in height and bear multiple round inflorescences (umbels) that bloom sequentially (Morse 1985). A large clone may remain in flower for three weeks or more.
For 11 of these years (1979-1991, data missing for 1989 and 1990) I also recorded the dates at which bumblebees first recruited to milkweed. Bumblebees, mostly Bombus terricola and B. vagans, were the most common large diurnal insects that visited milkweed flowers and made up by far the spiders' most important prey (Morse & Fritz 1982).
The vicinity of the study site has undergone measurable climatic change over the 32-year period of data gathering. Using the estimate of Jacobson et al. (2009) of a 0.25[degrees] F/decade increase in temperature for coastal Maine in 1975-2009, a period similar to that of this 1979-2010 study, I estimated that the mean temperature of the study area increased 0.44[degrees]C between 1979 and 2010, a rate that considerably exceeds the estimated global rate for the entire twentieth century of 0.6[degrees]C (Houghton et al. 2001). Precipitation also increased by 6.50 cm during this time (Jacobson et al. 2009). Fernandez et al. (2015) also noted that the warm season in this area (defined as days with a mean daily temperature above 0[degrees] C) increased by two weeks over the past century, with both earlier spring and later fall seasons. Increase in winter temperatures exceeded the increase in summer temperatures.
Initially I ran one-way ANOVAs between each one of the three variables (oviposition of spiders, flowering of milkweeds, and recruitment of bumblebees) against the ongoing years to search for whether these variables exhibited systematic change over the period of the study. I first ran these variables separately because of the different lengths of data runs for each of them. In addition to the three separate analyses, I ran a three-way ANOVA that incorporated data from the 11 years (1979 through 1991) for which I possessed data sets for crab spiders, milkweeds, and bumblebees. Analyses were carried out in R Version 2.13.0 (R Development Core Team 2011).
The initial date of oviposition shifted earlier by approximately one month over the 32-year period from 1979 to 2010 (Fig. 1). Although exhibiting considerable year-to-year differences, the highly significant slope moved consistently toward earlier laying dates ([R.sup.2] = 0.524, [F.sub.1,30] = 35.05, P <0.0001 in a one-way ANOVA), as did dates of the egg masses found in the field ([R.sup.2] = 0.828, [F.sub.1,13] = 62.56, P <0.0001 in a one-way ANOVA: Fig. 2). Flowering dates of the milkweed from 1979 to 1994 generally resembled the oviposition pattern of the spiders ([R.sup.2] = 0.334, [F.sub.1,14] = 7.03, P = 0.019 in a one-way ANOVA, although considerably weaker. Similarly, recruitment dates of the bumblebees to the milkweed flowers shifted earlier over the period of 1979 to 1991 ([R.sup.2] = 0.485, [F.sub.1,9] = 8.46, P = 0.017 in a one-way ANOVA.
During 1979-1991, when I obtained data on oviposition times of spiders, flowering times of milkweed, and recruitment times of bumblebees (Table 1), the overall relationship among these variables was sig nificant ([R.sup.2] = 0.945, [F.sub.3,7] = 25.51, P =0.011 in a three-way ANOVA), with significant contributions of both oviposition time and flowering time. Recruitment date was not significant, but interactions between oviposition time and flowering time and between flowering time and recruitment time suggested further differences among the three variables (Table 1). The slope of oviposition date over the 1979-1991 period ([R.sup.2] = 0.534) closely resembled that measure for the entire 32-year period.
The initial laying date of crab spiders shifted nearly a month earlier (from late July to late June) during this 32-year analysis. The field-laid broods exhibited an even stronger change during the 15 years in which I recorded them, although a likely outlier contributed to this difference. The procedures used in the two measures also differed significantly, preventing a direct comparison. However, results of the field-laid sample resemble the primary analysis in showing a strong shift toward earlier initial laying dates over time.
This shift to earlier laying dates should enhance overwintering survival of the resulting offspring. Middle and late instars overwinter more successfully than early instars (Morse 2007, 2012). However, the factor most important to the spiders, namely the recruitment of bees to milkweed flowers, should depend to some degree on flowering time, though not in the simplest way, because bumblebees do not recruit to the very earliest milkweed flowers, perhaps because the bees at that time are foraging for pollen to supply the colony's brood (Morse 1982a, 2007). They cannot obtain pollen from milkweeds, because milkweed encapsulates its pollen in pollinia that pollinators can transmit, but cannot exploit (Woodson 1954; Morse 1985). Nevertheless, the bumblebees' recruitment time was significantly related to flowering time, though the significant lag in recruitment to milkweed deprives them of a considerable amount of nectar (Morse 1982b). This lag deprives the spiders as well, since it curtails their time to forage profitably at a highly attractive hunting site (Morse & Fritz 1982).
Intheinitialyearsofthiscensus,spidershadlittle opportunity to produce a second brood due to time limitations, but by the end of the census period oviposition had moved earlier enough to provide the opportunity for a second brood. Indeed, the appearance of several second nests subsequent to rearing a successful first brood suggests the ability to rear a second brood at the study site. Rapid weight gains of females following removal from their first brood followed by supplementary feeding demonstrated that these spiders can produce a second brood if time allows (Morse 1994b), although over the study period I did not find a successful second brood in the field.
Possible advantages of a second brood are open to question because of the stage at which a brood would reach overwintering. A second brood that emerged just before the end of summer would experience a greater overwintering loss than members of first broods, since even members of the latest first broods experience a much lower overwintering success than older juveniles (Morse 1993, 2012). In an analogous situation, Neff & Simpson (1992) found that occasional second broods of the anthophorid bee Diadasia rinconis usually failed because of the lack of an adequate flower source. The timing might thus lead second Misumena broods into a developmental trap, in which a vulnerable instar reaches the season's end in a stage unfavorable for or fatal to survival (Fielding 2006; Van Dyck et al. 2015).
The lengthened season might appear to provide the opportunity for the spiders to reproduce a year earlier; that is, rather than overwintering a second time, to reproduce the season following their birth. Such offspring, however, would face the problem of immediately encountering the winter, which brings with it low survival rates, especially to the earlier instars conventionally produced (from 3rd year females). Any possible survivors should, however, enhance the fitness of their parents, although producing a second brood might entail countering costs for them, such as increased predation on their first broods resulting from a decreased period of guarding.
The relationship of the ongoing years with spider oviposition, milkweed flowering, and bumblebee recruitment suggests that changing conditions over the study period have affected all of these factors. Regressions for the three species were roughly similar, but if played out over longer periods, their synchrony would decline significantly (Bale et al. 2002). Oviposition date was weakly affected by recruitment date, a reflection of the bumblebees' importance as a food source for the spiders, in spite of the lack of a significant interaction between oviposition and flowering dates. This pattern reflects the response of the spiders directly to their prey, rather than to the flowers themselves, as reported earlier (Morse & Fritz 1982). The one-way analyses confirmed this difference: oviposition and flowering regressions were 0.524 and 0.334, respectively. If continued, this trend would progressively force the spiders to find sites other than milkweed for hunting. Flowering date significantly affected recruitment date; thus the two factors remained relatively in step, perhaps because the bumblebees' nectar sources became available earlier as well. These results all indicate ongoing changes, some of which will continue to decrease the synchrony of these species (Barton et al. 2009; Laws & Joern 2013).
Over the period presented here I have also kept less extended phenological data for several other species in the study area that are minimally related or unrelated to the spider-milkweed-bumblebee system, but show similar trends: emergence dates of adult Japanese beetles Popilia japonica, flight dates of Harris' checkerspot butterfly Chlosyne harrisii, and eclosion dates of the fern moth Herpetogramma theseusalis (Crambidae), the moth's common parasitoid wasp Alabagrus texanus (Braconidae), and its hyperparasitoid Aprostocetus sp. (Eulophidae) (D.H. Morse unpubl. data). Thus, although Misumena, milkweed, and bumblebees have demonstrated striking shifts in phenology, they do not appear unusual in their pattern of change.
In conclusion, I have demonstrated long-term shifts in the oviposition times of crab spiders Misumena vatia that parallel patterns of climate change. The lengthening of the season increases the feasibility of becoming double-brooded, although such a second brood would face extreme difficulties. This change in oviposition time parallels changes of other species upon which the spiders depend, but the differences in rate of change among these species suggests that the spiders' relationship with them will shift over time and result in future dependence on other species.
A generation of Brown University undergraduates assisted in the collection and care of the spiders. Several US National Science Foundation grants supported this work. I thank the Darling Marine Center of the University of Maine for use of the study site and K.J. Eckelbarger, T.E. Miller, L. Healy, and other staff members for facilitating fieldwork on the premises.
Bale, J.S., G.J. Masters, I.D. Hodkinson, C. Awmack, T.M. Bezemer, V.K. Brown et al. 2002. Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Global Change Biology 8:1-16.
Barton, B.T., A.P. Beckerman & O.J. Schmitz. 2009. Climate warming strengthens indirect interactions in an old-field food web. Ecology 90:2346-2351.
Bewick, S., R.S. Cantrell, C. Cosner & W.F. Fagan. 2016. How resource phenology affects consumer population dynamics. American Naturalist 187:151-166.
Clark, J.A. & R.M. May. 2002. Taxonomic bias in conservation research. Science 297:191b-192b.
Fernandez, I.J., C.V. Schmitt, S.D. Birkel, E. Stancioff, A.J. Pershing, J.T. Kelley et al. 2015. Maine's Climate Future: 2015 Update. University of Maine, Orono, Maine.
Fielding, D. 2006. Optimal diapause strategies of a grasshopper, Melanoplus sanguinipes. Journal of Insect Science 6:2.
Houghton, J.T., Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, D. Xiaosu et al. Climate Change 2001: The Scientific Basis. Cambridge University Press, Cambridge, UK.
Jacobson, G., I. Fernandez, S. Jain, K. Maasch, P. Mayewski & S. Norton. 2009. Maine's climate yesterday, today, and tomorrow. Pp. 8-16. In Maine's Climate Future: an Initial Assessment. (G.L.
Jacobson, I.J. Fernandez, P.A. Mayewski & C.V. Schmitt, eds.) University of Maine, Orono, Maine.
Johansson, J., N.P. Kristensen, J.-A. Nilsson & N. Jonzen. 2015. The eco-evolutionary consequences of interspecific phenological asynchrony - a theoretical perspective. Oikos 124:102-112.
Laws, A.N. & A. Joern. 2013. Predator-prey interactions in a grassland food chain vary with temperature and food quality. Oikos 122:977-986.
LeRoy, B., M. Paschetta, A. Canad, M. Bakkenes, M. Isaia & F. Ysnel. 2013. First assessment of effects of global change on threatened spiders: potential impacts on Dolomedes okabtarius (Clerck) and its conservation plans. Biological Conservation 161:155-163.
Mason, S.C., G. Palmer, R. Fox, S. Gillings, J.K. Hill, C.D. Thomas et al. 2015. Geographical range margins of many taxonomic groups continue to shift polewards. Biological Journal of the Linnean Society 115:586-597.
Memmott, J., P.G. Craze, N.M. Waser & M.V. Price. 2007. Global warming and the disruption of plant-pollinator interactions. Ecology Letters 10:710-717.
Morse, D.H. 1979. Prey capture by the crab spider Misumena calycina (Araneae: Thomisidae) Oecologia 39:309-319.
Morse, D.H. 1982a. The turnover of milkweed pollinia on bumble bees, and implications for outcrossing. Oecologia 53:187-196. Morse, D.H. 1982b. Behavior and ecology of bumble bees. Pp. 245-322. In Social Insects, vol. 3. (H.R. Hermann, ed.). Academic Press, New York.
Morse, D.H. 1985. Milkweeds and their visitors. Scientific American 253(1):112-119.
Morse, D.H. 1993. Some determinants of dispersal in crab spiderlings. Ecology 74:427-432.
Morse, D.H. 1994a. The role of self-pollen in the female reproductive success of common milkweed (Asclepias syriaca: Asclepiadaceae). American Journal of Botany 81:322-330.
Morse, D.H. 1994b. Numbers of broods produced by the crab spider Misumena vatia (Araneae, Thomisidae). Journal of Arachnology 22:195-199.
Morse, D.H. 2007. Predator Upon a Flower: Life History and Fitness in a Crab Spider. Harvard University Press, Cambridge, MA.
Morse, D.H. 2012. Reproductive output of female sit-and-wait spiders: the relative impacts of mating failure, natural enemies and resource availability. Entomologia Experimentalis et Applicata 146:141-148.
Morse, D.H. 2014. Locomotor performance of adult male crab spiders, with implications for the importance of small size. Evolutionary Ecology 28:23-36.
Morse, D.H. & R.S. Fritz. 1982. Experimental and observational studies of patch-choice at different scales by the crab spider Misumena vatia. Ecology 63:172-182.
Neff, J.L. & B.B. Simpson. 1992. Partial bivoltinism in a ground-nesting bee: the biology of Diadasia rinconis in Texas (Hymenoptera, Anthophoridae). Journal of the Kansas Entomological Society 65:377-392.
Parmesan, C. 2006. Ecological and evolutionary responses to recent climate change. Annual Review of Ecology, Evolution, and Systematics 37:637-669.
Primack, R.B., I. Ibanez, H. Higuchi, S.D. Lee, A.J. Miller-Rushing, A.M. Wilson et al. 2009. Spatial and interspecific variability in phenological responses to warming temperatures. Biological Conservation 142:2569-2577.
R Development Core Team. 2011. R: a language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria.
Thackeray, S.J., T.H. Sparks, M. Frederickson, S. Burthe, P.J. Bacon, J.R. Bell et al. 2010. Trophic level asynchrony in rates of phenological change for marine, freshwater and terrestrial environments. Global Change Biology 16:3304-3313.
Van Dyck, H., D. Bonte, R. Puls, K. Gotthard & D. Maes. 2015. The lost generation hypothesis: could climate change drive ectotherms into a developmental trap? Oikos 124:54-61.
Woodson, R.E. 1954. The North American species of Asclepias. Annals of the Miissouri Botanical Garden 41:1-211.
Manuscript received 3 June 2017, revised 10 October 2017.
Douglass H. Morse: Department of Ecology & Evolutionary Biology, Box G-W, Brown University, Providence, RI 02912 USA. E-mail: email@example.com
Table 1.--Results from three-way ANOVA of year vs. oviposition date (*) flowering date (*) recruitment date over 1979-1991 period. Df Sum Sq Mean Sq F value Pr(>F) Oviposition date 1 86.340 86.340 117.3094 0.001684 (**) Flowering date 1 9.352 9.352 12.7071 0.037694 (*) Recruitment date 1 3.525 3.525 4.7898 0.116412 Oviposition date: flowering date 1 2.921 2.921 3.9693 0.140368 Oviposition date: recruitment date 1 8.495 8.495 11.5418 0.042545 (*) Flowering date: recruitment date 1 19.209 19.209 26.0999 0.014509 (*) Oviposition date: flowering date: recruitment date 1 1.585 1.585 2.1540 0.238493 Residuals 3 2.208 0.736
|Printer friendly Cite/link Email Feedback|
|Author:||Morse, Douglass H.|
|Publication:||The Journal of Arachnology|
|Date:||Jan 1, 2018|
|Previous Article:||Circadian rhythms of locomotor activity in Metazygia wittfeldae (Araneae: Araneidae).|
|Next Article:||The egg sac of Benoitia lepida (Araneae: Agelenidae): structure, placement and the function of its layers.|