xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00162
Seleccin para la sobrevivencia: preferencia por el tamao del grano de tierra y el tamao corporal en las larvas wormlion (sp vermilio.) afectan la construccin de la trampa del hoyo.
Selection for survival: soil grain size preference and body size in wormlion (Vermilio sp.) larvae affect pit trap construction
Sessile predatory larvae rely on their habitats to provide biotic and abiotic factors needed for sufficient prey capture.
I examined the hypothesis that for Vermilio sp. (wormlion) larvae, body size and soil grain size affect pit trap size
and prey capture efficiency. At the Estacin Biolgica de Monteverde in Costa Rica. I measured and compared larva
size versus pit trap size, put larvae in containers with half fine grain soil (< 2 mm) and half coarse grain soil (2 mm
to 6 mm), tested for preference, and compared size of pit traps constructed in the two soil types. I also dropped ants
into pits and measured time to escape and frequency of capture by the larvae. Pit trap size was positively correlated
with larva size. Larvae preferred fine grain soil over coarse grain soil ( 2 = 45.08, df = 1, P < 0.01) and constructed
traps less often in coarse grain soil. I found no difference between pit trap sizes in fine and coarse grain soil, yet
sample size was small and I noted the trend that ants remained trapped in fine grain soil longer than in coarse grain
soil. Preference for fine grain soil indicates habitat selection of larval Vermilio sp. for areas that increase prey
capture, therefore promoting the survival of the species.
Las larvas depredadoras ssiles confan en sus hbitats para que provean los factores biticos y abiticos necesarios para capturar suficientes presas. Yo examin la hiptesis que en Vermilio spp. (Gusanos len) el tamao de las larvas y el tamao del grano de la tierra afectaran el tamao del hoyo de sus trampas y la eficiencia de la captura de sus presas en la Estacin Biolgica de Monteverde, Costa Rica.
Text in English.
Predator & prey
Depredador y presa
Tropical Ecology 2007
Ecologa Tropical 2007
t Monteverde Institute : Tropical Ecology
1Selection for survival: soil grain size preference and body size in wormlion ( Vermilio sp .) larvae affect pit trap construction Alice McCarthy Department of Zoology, Univer sity of Wisconsin Madison ABSTRACT Sessile predatory larvae rely on their habitats to provide bi otic and abiotic factors needed for sufficient prey capture. I examined the hypothesis that for Vermilio sp (wormlion) larvae, body size and soil grain size affect pit trap size and prey capture efficiency. At the Estacin Biolgica de Monteverde in Co sta Rica I measured and compared larva size versus pit trap size, put larvae in containers with half fine grain soil (< 2 mm) and half coarse grain soil (2 mm to 6 mm), tested for preference, and compared size of pit tr aps constructed in the two soil types. I also dropped ants into pits and measured time to escape and frequency of capture by the larvae. Pit trap size was positively correlated with larva size. Larvae preferred fine grain soil over coarse grain soil ( 2 = 45.08, df = 1, P < 0.01) and constructed traps less often in coarse grain soil. I found no difference between pit trap sizes in fine and coarse grain soil, yet sample size was small and I noted the trend that ants remained trapped in fine grain soil longer than in coarse grain soil. Preference for fine grain soil indicates habitat selection of larval Vermilio sp for areas that increase prey capture, therefore promoting th e survival of the species. RESUMEN Las larvas depredadoras y sesiles dependen de factores bi ticos y abiticos para capturar sus presas. Yo examin la hiptesis que en Vermilio spp (Gusanos len) el tamao de las larvas y el tamao del grano del suelo afectaran el tamao de sus trampas fosoriales cnicas y la eficiencia de la captura de sus presas. Para lograr este objetivo,med y compare el tamao de las trampas, puse larvas en recipientes que contenian suelo de grano fino (< 2mm) y suelo de grano grueso (2 6 mm), luego compar el tamao de las trampas que fueron construidas en los dos tipos de suelo. Tambin puse hormigas en las trampas y med cuanto tiempo tardaban las hormigas en escapar de las trampas y la frecuencia de captura de las larvas. Los ta maos de las trampas se relacionaron positivamente con los tamaos de las larvas. Las larvas tuvieron una preferencia por el suelo de granos finos sobre el suelo de granos speros ( 2 = 45.08, gf = 1, P << 0.01) y con solamente una eleccin del tipo de tierra, las larvas construyeron trampas con menos frecuencia en el suelo de granos speros. No obtuve ninguna diferencia entre el tamao de las trampas construidas en el suelo de granos finos y de granos speros. No obstante, yo observe diferencia de tipo visual entre ellas y las hormigas se quedaron ms tiempo en las trampas construidas en el suelo de granos finos que en las trampas construidas en el suelo de granos speros. La preferencia por el suelo de granos finos indica que la seleccin de hbitat por Vermilio sp en la etapa larval podra aumentar las posibilidades de capturar presas, aumentando la supervivencia de los individuos. INTRODUCTION Habitat selection is important for the survival of all species and even more vital for a sessile species. Predatory larvae that are sessil e present an interesting opportunity to study the importance of habitat selection b ecause the duration of the larval stage and size of the adult depend on food availability in the larval stage (Gotelli 1997). Tiger beetles ( Pseudoxychila spp.), antlions ( Myrmelion spp.), and wormlions ( Vermilio spp .) are all sessile, sit and wait predators in the larval stage (McClure 1983, Palmer 1983, Zumbado 2006). Habitat selection
2 has been studied in tiger beetles (Pearson a nd Mury 1979, Mury -Meyer 1987) and antlions (Farji-Brener 2003, Lucas 1989), yet no studies have been documented for wormlions. Populations of wormlions (Family: Vermilioni dae) have been found in the United States and Costa Rica, but they are uncommon and so lit tle is known about them. Similar to antlion larvae, wormlion larvae build pit traps in dry dusty soil beneath projecting rocks or other overhanging objects (Zumbado 2006). The two have evolved independently but they both use the same mechanism for capturing prey by digging a conical pit trap in the ground to trap ants and other small arthropods (Zumbado 2006). Past studies found efficiency of pit traps in ant lions to depend on biotic factors, such as arth ropod availability and larvae size (Griffiths 1986), as well as abiotic factors, such as soil moistu re content (McClure 1976) and soil grain size (FarjiBrener 2003). Past experiments looking at Myrmeleon crudelis revealed a positive correlation between larva size and pit trap size (Griffiths 1986), a stat istical preference for fine grain soil (< 2 mm) over coarse grain soil (2 to 6 mm), larger pit sizes in fine grai n soil, and higher prey retention efficiency after capture (Farji-Brener 2003) Although antlion larvae differ greatly in morphology from wormlion larvae, the pit traps of these organisms are very similar. This study examines the hypothesis that for Vermilio sp ., larvae size and soil grain size affect pit trap size and prey capture efficiency. Therefore, I tested to see if wormlions have the same characteristics found in antlions by investigating if: (a) as larva size increases the pit trap size increases, (b) larva will build pit traps in fine grain soil more often than in coarse grain soil when exposed to both soil types simultaneously, (c) pit traps constructe d in fine grain soil will be larger than those constructed in coarse grain soil, and (d) pit traps in fine grain soil will be more efficient at preventing trapped prey from escaping than coarse grain soil pit traps. METHODS I collected Vermilionidae larvae, soil, and ants at the Estacin Biolgica in Monteverde, Costa Rica (1500 m elevation). Us ing screen sieves, I separated the dry sandy soil into < 2 mm soil grains, which I refer to as fine grain soil, and 2 to 6 mm soil grains, or coarse grain soil (Farji-Brener 2003). I used plastic containers 13 cm in diameter, filled 4 cm deep with soil for experiments. They were covered with plastic wrap to prevent disturbance and possible food sources from entering. These contai ners of soil were left undisturbe d for at least two days before experimentation. This allowed the wormlion la rvae to construct pits; the larvae were then removed from the containers. Larva size and trap size I found 50 wormlion pit traps in the field and measured pit diameter and depth before collecting the larvae. Pit traps were on an inclin e so, using calipers, I measured trap dimensions both parallel and perpendicular to the incline and averaged the m easurements. I measured mass, to the nearest thousandth of a gram, and length, to the nearest millimeter, of the wormlions and tested for a significant regression between larv a length and pit angle, larva length and pit diameter, larva mass and pit angle, and larva mass and pit diameter using linear regression analyses. I also repeated these steps for the 50 wormlion larvae that I placed in experimental conditions of fine or coarse-grain sand for five days.
3 diameter height Grain size preference To test soil size preference for building pit trap s, I filled 49 containers with soil collected from the wormlion habitat. In each container one half of the soil was fine grained, and the other half was coarse grained. I placed one larva in ea ch of the containers on the line where fine grain meets coarse grain soil. After five days I reco rded in which soil type each larva created a trap and tested for preference of soil size for trap construction using a goodness of fit Chi-square test. Grain size versus trap size To analyze the size differences in traps constructed in fine grain soil and traps constructed in coarse grain soil I measured the trap s that were constructed in fine grain soil in the preference experiment. Then I removed each la rva, measured its length and mass, confirmed that it was still alive, and then placed it in its own container filled with only course grain soil. After five days I measured diameter and depth of the new traps, and calculated pit trap angle as = tan-1(diameter/ (2* depth)) in degrees (Fig.1) I compared diameter and depth for traps built in coarse grain soil and tested for si gnificant difference using a Mann-Whitney U test. Grain size and trap efficiency In order to examine the efficiency of prey captu re for pit traps in fine versus coarse grain soil, I collected small ants from the perimeter of the wormlion habitat. All ants were the same species and ranged from 2 mm to 4 mm long. For traps in fine gr ain soil, I dropped an ant into each trap and noted the time it stayed in the trap and whether it was caught by the larva. There were very few traps in coarse grain soil, and so I dropped three ants, one at a time, into each pit and measured time remaining in the trap and whethe r or not it was caught. If an ant remained in a trap for ten minutes I recorded time as 10+ minutes. RESULTS Linear regression analysis showed that worm lion larva length related positively with pit trap angle (Fig. 2 a ). There was a positive correlation betw een wormlion larvae length and mass (R2 = 0.64, p<<0.01, n=46), and between pit trap diameter and pit angle (R2 = 1.0, p<<0.01, n=46). For this reason larva mass and trap diameter were positively related (Fig. 2b), as was larva mass and trap angle (R2 = 0.29, F = 17.674, p-value << 0.01, df = 1, 44), and larva length and trap diameter was positively related (R2 = 0.40, F = 29.503, p-value << 0.01, df = 1, 44). The same trends were apparent for larvae and pit traps in the field but they were more prone to outside disturbances; thus I concentrated on experimental larvae in the fine grain soil experiments. FIGURE 1. Measurements of pit traps. Includes diameter, height, and calculated angle ( = tan-1(diameter/ (2* depth)).
4 0 50 100 150 200 250 0 0.5 1 1.5 2 2.5Larva length (cm) 0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.010.020.030.040.05Larva mass (g) The wormlions preferred fine grain soil for th eir pit traps compared to course grain soil (Fig. 3). When larvae were exposed to only one soil type, the number of traps built in fine grain soil over a five-day time period was also higher th an that in coarse gr ain soil (goodness of fit, 2 = 43.49, df = 1, p<< 0.01). Along with lower frequenc y of pit traps construc ted, I noted that the of the course grain traps a ppeared to be partially lined with fine grain soil, FIGURE 2. Increase in pit tr ap size with increase in larva size. Traps were measured for larvae in fine grain soil after five days. (a) Positive relation between wormlion pit trap angle and wormlion larva length (y = 63.209x + 42.302). Linear regression analysis shows statistical significance (R2 = 0.46, F = 29.503, p<<0.01, df = 1, 44). (b) Positive relation between wormlion pit trap diameter and wormlion larva mass. Linear regression analysis shows statistical significance (R2 = 0.29, F = 17.674, p-value << 0.01, df = 1, 44). (a) (b) ( y = 63.209x + 42.302 ) ( y = 24.84x + 1.7342 )
5 0 10 20 30 40 50 60 Fine Coarse Soil grain type Although pit traps in fine versus coarse grai n soil were not different in diameter (MannWhitney U, U = 112.50, P = 0.32, n1 = 50, n2 = 6) or pit trap angle (Mann-Whitney U, U = 112.50, P = 0.32, n1 = 50, n2 = 6), pits in coarse grain appeared to have larger angles observation. There was also an apparent difference in efficiency of trapping ants for the two soils. All ants in fine grain, 23 out of 23 (100%), soil pits remained in the pit for 10+ minutes, while only five of 18 (33 %) of the ants in coarse-grain soil rema ined in the pit for over 10 minutes; the other thirteen were trapped from 0.17 to 6 minutes. Larvae showed a trend of trapping the prey, squeezing it for several minutes, releasing it, and then trapping it again. This process was sometimes repeated. Larvae captured ants by bendi ng their anterior, or n on-foot-like end, in half, and in effect trapping and squeezing its prey. The posterior foot-like end remained anchored in the soil. DISCUSSION Habitat selection for the sit and wait predator, Vermilio spp., must be important for effectively building a pit trap a nd capturing prey. My hypothesis, that larva size and soil grain size affect pit trap size and pr ey capture efficiency, was part ially supported. This study demonstrated that, similar to antlion larvae (G riffiths 1986), increase in wormlion size relates positively with pit trap size increase. Als o, like antlion larvae, wormlion larvae showed a preference for constructing pits in fine grain soil (< 2 mm) over coarse gr ain soil (2 6 mm). Wormlions may choose fine grain soil for p it construction because it is more energy efficient; possibly costing less energy to cons truct and maintain pits, resulting in higher efficiency of prey capture (Lucas 1982, FarjiBrener 2003). There was no difference in angle size of the pits between the two soil grain type s, which is important because it affects prey capture in antlions (Farji-Brene r 2003) and may do the same for worm lions. However, there was a trend that showed that the traps in coarse grain soil appeared to have greater angles. Had I been able to increase the sample size I may have found a difference. Soil grain size might also FIGURE 3. Soil grain size preference in fine ve rsus coarse. Larvae in containers with fine and coarse grain soil showed statistical preferen ce for building pit traps in fine grain soil ( 2 test, 2 = 45.08, df = 1, P << 0.01).
6 affect how easily an arthropod can escape in othe r ways. Fine grain soil is more prone to landslides, making it more difficult for the prey to leave (Lucas 1982) Traps in coarse grain soil did appear to be partially lined w ith fine grain soil, but less so than fine grain soil traps. When I dropped small ants into pit traps a few escaped from coarse grain soil traps while no ants escaped from fine grain soil traps. This could indicate that fine grain soil traps are more efficient at trapping small arthropods. The greater efficiency of prey trapping is a likely explanation fo r the apparent tendency to choose fine grain soil for p it trap construction. The wormli on larvae captured ants by moving the anterior end of its body up from below the botto m of the pit, and bending the anterior end in half enclosing the ant and squeezing it, this is sometimes called a boa technique. The larvae squeezed the prey for several minutes, let go, and then trapped it again; sometimes this process was repeated. Therefore, it is important that the prey remain in the trap long enough for the larva to sense and capture it, and then recapture it after release. This study supports that wormli on larvae prefer to construct pit traps in specific soil types. It also reveals that these larvae have habitat preferences, which affect survival to adulthood and therefore the species survival. They share the same preferences for fine grain soil as antlions. Therefore, these two similar orga nisms independently evolved the same method of constructing a pit for prey capture (Zumbado 2006). Other similarities to antlions are likely to be found, as this is just the beginning of e xploring the abilities and preferences of Vermilio sp .. Future studies will add to what I have found a bout the habitat selection of wormlions, and may help explain more about this species natural history. ACKNOWLEDGEMENTS This study was made possible because of the help and advice of Tania Chavarra, Pablo Allen, and Karen Masters. Thank you Taegan McMahon for comments along the way and to my colleagues, Nathan Ebert, Nicole Williams, Dr ew Moore, Sarah Green, and Zachary Sheff for your work during our earlier experimentation with Vermilio sp., which helped pave the way. Also, I appreciated being able to complete this experiment at the Estacin Biolgica de Monteverde in Costa Rica. LITERATURE CITED Fafji-Brener, A. G. 2003. Microha bitat selection by antlion larvae, Myrmeleon crudelis : effect of soil particle size on pit-trap design and pr ey capture. Journal of Insect Behavior 16: 783-796. Gotelli, N. J. 1997. Competition and coexistence of larval ant lions. Ecology 78(6): 1761-1773. Griffiths, D. 1986. Pit construction by ant-lion larvae: a cost-benefit analysis. Journal of Animal Ecology 55: 37-59. Lucas, J. 1982. The biophysics of pit constr uction by ant-lion la rvae (Neuroptera: Myrmeleontidae). Animal Behavior 30: 651-664.
7 Lucas, J. R. 1989. Differences in habitat use between two pit-building antlion species: causes and consequences. American Mi dland Naturalist 121: 84-98. McClure, M. S. 1976. Spatial distribution of it-making ant-lion larvae (Neuroptera: Myrmeleontidae): density e ffects. Biotropica 8: 179-183. McClure, M. S. 1983. Myrmeleon (Hormiga Leon, Ant Lions). In: Costa Rican Natural History D. H. Janzen (ed). The University of Chicago Press, Chicago, IL, pp. 742. MuryMeyer, E. J. 1987. Asymmetric resource use in two syntopic species of larval tiger beetles (Cicindelidae). Oikos 50: 167-175. Palmer, M. K. 1983. Pseudoxychilia tarsalis (Abejon Tigre, Tiger Beetle). In: Costa Rican Natural History D. H. Janzen (ed). The University of Chicago Press, Chicago, IL, pp. 742. Pearson, D. L. and Mury E. J. 1979. Character divergence and convergence among tiger beetles (Coleoptera: Cicindelida e). Ecology 60: 557-566. Zumbado, M. A. 2006. Diptera of Costa Rica and the New World tr opics. Instituto Nacional de Biodiversidad. Vermileondidae. Santo Do mingo de Heredia, Costa Rica, pp. 92-93.