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 M39-00275
Produccin de cuerpos de Mller por Cecropia obtusifolia (Cecropiaceae) con y sin hormigas Azteca
Mllerian body production by Cecropia obtusifolia (Cecropiaceae) with and without Azteca ants
Cecropia obtusifolia (Cecropiaceae) has a facultative mutualism with Azteca ants to minimize herbivory. In this study, Mllerian Bodies (MBs), glycoprotein bodies the plant produces to feed its mutualist ants, were
systematically removed, from C. obtusifolia. A second study tested at what rate the tree would decrease its production of MB if the ants were restricted from removing them. It was found that removal of the MBs did
result in an increase in MB production for two of the five trees in the treatment group. The C. obtusifolia where the ants were restricted from removing the MB, in all but one tree, showed a rapid decrease in MB
production in the first few days. The results were also looked at in terms percent canopy cover and herbivory. It was found that plants without ants suffered significantly higher herbivore damage (Mann
Whitney U-test, df=1, 2 = 5.51, p<.05). A trend between an increase in percent canopy cover and a decrease in MB production was noted but was not statistically significant. It seems that MB production is a costly process for C. obtusifolia and worth the trade off when they receive the protection from the Azteca ants but when ants are not present they will quickly divert that energy elsewhere.
Cecropia obtusifolia (Cecropiaceae) tiene un mutualismo facultativo con las hormigas del gnero Azteca para reducir la herbivora. En este estudio, los cuerpos de Muller (CM), glicoprotenas que la planta produce para alimentar a la hormiga mutualista, fueron sistemticamente removidos de C. obtusifolia. Un Segundo estudio probo a que tasa la planta decrece en la produccin de CM si se restringe el acceso a los mismos por parte de las hormigas. Encontr que la eliminacin de CM hizo lugar a un aumento en la produccin de los mismos para dos de los cinco rboles utilizados en este tratamiento. En las plantas en las que se restringi el acceso a las hormigas a los CM, en todos excepto uno, se demostr un rpido decrecimiento en la produccin de CM a los pocos das. Se analiz tambin el porcentaje de cobertura de dosel y la herbivora. Encontr que las plantas sin hormigas sufren mayor dao por herbivora (Mann Whitney U-test, df=1, 2 = 5.51, p<.05). Existe una tendencia a una menor produccin de CM al aumentar la cobertura de dosel, pero no es estadsticamente significativo. Parece que la produccin de CM es un proceso costoso para C. obtusifolia y vale la pena producirlo cuando tienen la proteccin por parte de las hormigas, pero cuando no estn estas presentes es mejor utilizar la energa en otros procesos.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone
Plantas--Resistencia a insectos
Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology Fall 2010
Trumpet tree (Cecropia obtusifolia)
Ecologa Tropical Otoo 2010
t Monteverde Institute : Tropical Ecology
MÃ¼llerian body production by Cecropia obtusifolia (Cecropiaceae) with and without Azteca ants Josh Neighbor Department of Biology, University of Washington ABSTRACT Cecropia obtusifolia (Cecropiaceae) has a facultative mutualism with Azteca ants to mini mize herbivory. In this study, MÃ¼llerian Bodies (MBs), glycoprotein bodies the plant produces to feed its mutualist ants, were systematically removed, from C. obtusifolia . A second study tested at what rate the tree would decrease its production of MB if t he ants were restricted from removing them. It as found that removal of the MBs did result in an increase in MB production for two of the five trees in the treatment group. The C. obtusifolia where the ants were restricted from removing the MB, in all but one tree, showed a rapid decrease in MB production in the first few days. The results were also looked at in terms percent canopy cover and herbivory. It was found that plants without ants suffered significantly higher herbivore damage (Mann Whitney U test , df=1, 2 = 5.51, p<.05). A trend between an increase in percent canopy cover and a decrease in MB production was noted but was not statistically significant. It seems that MB production is a costly process for C. obtusifolia and worth the trade off when they receive the protection from the Azteca ants but when ants are not present they will quickly divert that energy elsewhere RESUMEN Cecropia obtusifolia (Cecropiaceae) tiene un mutualismo facultative con las hormigas del gÃ©nero Azteca para reducir herbi vorÃa. En este estudio, cuerpos de Muller (CM), glicoproteÃnas que la planta produce para alimentar la hormiga mutualista, fueron sistemÃ¡ticamente removidos de C. obtusifolia . Un Segundo estudio probo a que tasa la planta decrece en la producciÃ³n de CM s i se restringe el acceso a los mismos por parte de las hormigas. EncontrÃ© que la remociÃ³n de CM resulta en un aumento en la producciÃ³n de los mismos en dos de cinco Ã¡rboles utilizados en este tratamiento. En las plantas en las que se restringiÃ³ el acceso a las hormigas a los CM, en todos excepto uno, se desmostrÃ³ un rÃ¡pido decline en la producciÃ³n de CM a los pocos dÃas. Se analizÃ³ tambiÃ©n el porcentaje de cobertura de dosel y la herbivorÃa. EncontrÃ© que las plantas sin hormigas sufren mayor daÃ±o por he rbivorÃa (Mann Whitney U test, df=1, 2 = 5.51, p<.05). Existe una tendencia a una menor producciÃ³n de CM al aumentar la cobertura de dosel, pero no es estadÃsticamente significativa. Parece que la producciÃ³n de CM es un proceso costos para C. obtusifolia y vale la pena producirlo cuando ti enen la protecciÃ³n por parte de las hormigas, pero cuando no estÃ¡n estas presentes en mejor utilizar la energÃa en otros procesos. INTRODUCTION Plants employ an array of defenses to protect themselves against herbivore attacks, and these defenses can be both chemical and physical. An example is the facultative symbiotic relationship of many ant plants, in which ants provide an herbivore defense in exchange for receiving housing and food from the plant, though neither organism completely relies on the othe r for survival (Folgarait & Davidson 1995, Karban et al . 1999). Ants are an excellent herbivore defense for plants for several reasons: (1) they are aggressive and have well developed defense mechanisms such as biting and stinging, (2) they are willing to sacrifice themselves in defense of the colony, and (3) they can release chemical cues telling others of a disturbance (HÃ¶lldobler & Wilson 1990). The genus Cecropia (Cecropiaceae), a neotropical pioneer tree, has around 70 80 different species, of which 8 0% are myrmecophytic and inhabited by symbiotic ants
( Davidson & Fisher 1991) . The symbiotic ants include four subfamilies: Dolichoderinae, Formiciae, Myrmicinae and Ponerinae ( Folgarait et al . 1994). In Costa Rica there are only five species of Azteca (Do lichoderinae) ants that are mutualistic with Cecropia three of which: C.peltata , C. obtusifolia , and C. insignis , occur in the Monteverde region, where I conducted my study. Myrmecophytic Cecropia trees produce MÃ¼llerian bodies (MB), which are small ovoid structures, located on and produced by special petiolar pads called trichilia (Rickson 1976, Davidson & Fisher 1991). These MBs are full of glycogen and small amounts of protein, and are a main food source for Azteca . Along with food Cecropia also provides housing for the ants; the hollow internodes in the tree form domatia which the ants inhabit (Longino 1991). In this facultative relationship with Cecropia Azteca ants, in turn, contribute by helping protect the plant from herbivores (Janzen 1969, Schupp 1 986). This relationship clearly benefits the Cecropia in at least two ways. The sapling growth rate when occupied by ants is significantly greater than that of unoccupied trees (Schupp 1986). Additionally, there is a positive correlation between the number of 1992). While this facultative relationship increases the plants fitness when herbivores are present in the environment, it is costly to the plant in terms of energ y and nutrient expenditure. If herbivores are absent, producing MBs would be taxing and wasteful, and (Agrawal 1998). This cost benefit trade off leads to the idea that without herbivore pressure or ant occupation th e Cecropia would limit production of MB (Putz & Holbrook 1988, Scalley 1993, Karban et al . 1999). Studies have shown that when ants are not present, the Cecropia reduces production of MB but when ants return, the Cecropia will increase production of MBs (F olgarait et al . 1994, Karban 1999) but this has not been attempted outside of a greenhouse. These are indications of both the costliness of producing MBs and the ability of Cecropia Folgarait et al. 19 94). My study examined the facultative mutualism of C obtusifolia, and Azteca (Schupp 1986) ants in relation to canopy cover and herbivore damages, in the Monteverde region of Costa Rica The two main questions the study was designed to answer were: (1) A t what rate is MBs production increased by C. obtusifolia, when MBs are manually removed, simulating ant inhabitation, compared to C. obtusifolia in which MBs are not removed; (2) If ants are restricted from removing MBs, how quickly does C. obtusifolia re duce MB production? MATERIALS AND METHODS SITE DESCRIPTION. The study was conducted during the wet season in Monteverde, Costa Rica, between October and November 2010. Cecropia trees were encountered in Bajo del Tigre (1,300m), Santa Elena (1,400m), Ce rro Plano (1450m), and the biology station (1550m).
(1500m). Each is Premontane Moist Forest according to Holdridge. Mean annual temperature is 17 24 C and the mean annual rainfall is 2000 4000 mm (Haber 2000). I chose C. obtusifolia that were 1.5 2.5m ta ll. Once the trees were found, the trunks were shaken to determine if ants were present. Seven trees with ants were found and labeled A2 A8. Ten trees with out ants were found and labeled B1 B10. MB REMOVAL EXPERIMENT. Ten C. obtusifolia without ant inh abitants were located, divided into a control (B3,5,6,7,9) and treatment (B1,2,4,8,10) group of five each. Vaseline was placed around the bottom of the trunk of each tree to prevent new ants or any other insect from reaching trichilia. In the treatment gro up, the presence of ants was simulated by manually removing MBs daily. In the control group, MBs were allowed to accumulate on plants, this treatment simulating the absence of ants. Because MB production by C. obtusifolia normally peaks just after dusk (Da vidson & Fisher 1991) I counted the number of MB every morning on the three youngest leaves of each plant. The three youngest leaves were chosen because I observed only the three youngest trichilia tended to produce MB. I record the number of new MBs -whi ch were detectable by their brighter white color (Figure 1). The production of MB was monitored for 16 days, from November 3 to November 18, 2010. ANT REMOVAL EXPERIMENT. On the C. obtusifolia that had ants (n=7) I used a combination of Vaseline and mosq uito netting (Figure 2) to prevent ants from removing MBs on the three youngest leaves by covering the trichilia. I returned every morning and recorded the new MBs that were produced for 13 days between Nov. 5 and 18, 2010. The new MBs were again identifie d by their brighter white color. CANOPY COVER. Percent canopy cover was measured for each plant by using a Densitometer. In order to gage the amount of direct sunlight each tree was receiving in relation to each other. HERBIVORY. Herbivory was calculat ed for each plant by placing each leaf under a transparent grid with 1cm 2 squares. Younger leaves were chosen because they tend to FIGURE 1. Newly produced MÃ¼llerian bodies on the trichila of C. obtusifolia from a plant in Monteverde, Costa Rica. FIGURE 2. Vaseline and mosquito netting to prevent ant removal of MÃ¼llerian bodies
have higher herbivory (Garnsey 1999), normally the third of forth leaf down on the plant. The leaf that was chosen seemed to be a good representation of the overall herbivore damage to the plant. First the total area within the perimeter of the leaf was calculated. Then the total number of squares missing leaf tissue was counted. Percent herbivory was equal to the total number o f squares missing leaf tissue divided by the total number of squares within the perimeter of the leaf multiplied by 100. ANT RATIO. In order to gage the ant population of each tree, I shook each tree vigorously, counted to ten and then recorded that n umber of ants along a 10cm portion of the trunk between the first and second leaves. RESULTS MB REMOVAL EXPERIMENT. In the treatment group, over the 16 day study period in which MBs were removed from each plant daily, the average MB production per day did not show an overall change. For trees B1 and B8 there was actually a decrease (Table 1). A linear regression analyses for MB production by day was run for each tree in both treatment and control groups. All the trees in the control group had positive s lopes and the equation of the line was significant (p<.05). Two C. obtusifolia in the treatment group demonstrated a strong positive correlation between MB production and day. There was a two fold and five fold increase in the slope over the largest slope in the control group, respectively, for trees B2 and B10. Tree B2 (Figure. 4) and B10 demonstrated that MB production per day was greater for those trees with daily removal than those in the control, though this trend only help true for two of the five tre es. It should be noted that the weather changed and the last five days of the FIGURE 4. Relationship between the number of MÃ¼llerian Bodies found on C. obtusifolia plant A2 per day this trend is constant for six of the seven sampled (Nov. 3 18, 2010). A regression fit log to log was run. Lny= 4.07 0.67ln(x) . One way ANOVA, F= 19.8, R 2 =0.60, n=16, p<0.05. Data taken in Monteverde, Costa Rica. FIGURE 3. Mean percent herbivory ( SD) of C. obtusifolia trees with (n=7, 5.57 4.69) and without (n=10, 15.2 8.23) ants in Monteverde, Costa Rica. Percent herbivory was significantly different between the two groups (Mann Whitney U test, df=1, 2 = 5. 51, p<.05) [on November 18, 2010] NO ANTS ANTS
experiment were considerably warmer and sunnier. ANT REMOVAL EXPERIMENT. During the 13 day sampling of C. obtusifolia, where ants were restricted from removing the MB, all but one tree showed an overall decrease in MB production (Table 2). Five of the seve n trees had a significant relationship (p<.05) between MB production and day, when both were log transformed. Tree A5 was close to being significant with p=.08. Figure 3 demonstrates the rapid decline in MB production over the first few days for tree A2, t his trend is similarly represented for trees A4, A5, A6 and A7, thought not as pronounced. In all the trees that followed this trend within around 3 days they all dropped and hit a baseline MB production of usually around 15 or so MB. CANOPY COVER. Canopy cover had a trend but no significant impact on MB production for plants without ants (n=5) and no MBs removed manually (F=0.68, R 2 =0.18 p=.46 see Figure 6). In the treatment group the tree that had the most MB produced with 1,811, had the lowest percent c anopy cover 7.5% and the tree that had the least with only 20 MB, had the highest percent canopy cover (76%). HERBIVORY. The percent herbivory of trees with ants (mean Sd= 5.57 4.69) and without ants (mean sd = 15.2 8.23) were significantly different (Mann Whitney U test, df=1, 2 = 5.51, p<.05 see Figure 3). Plants with ants had from 1% to 13% leaf tissue removed, and those without had from 2% to 28% removed. Of the ant plants sampled, all appeared to have substantial populations of ant inhabitants. Equation R 2 n F p B3 c Y= 4.48 + 1.75x 0.40 16 9.45 .0083 B5 c Y= 4 + 2.02x 0.73 16 38.75 <.0001 B6 c Y= 1.33 + 0.85x 0 .72 16 36.68 <.0001 B7 c Y= 1.68 + .6x 0.34 16 7.36 .016 B9 c Y= 5.8 + 1.11x 0.40 16 9.30 .0087 B1 t Y= 2.85 0.31x 0.37 16 8.07 .013 B2 t Y=8.08 + 3.96x 0.50 16 13.88 .0023 B4 t Y= 1.5 + 1.28x 0.23 16 4.30 .057 B8 t Y= 9.1 0.31x 0.058 16 0.86 .3 7 B10 t Y= 29.01 + 9.9x 0.45 16 11.62 .0042 TABLE 1: Results of linear regression analyses for MÃ¼llerian bodies (MB) production in Monteverde, Costa Rica over 16 days for Cecropia obtusifolia without ants. B1 10 each represents an individual C. obtusifolia , followe treatment group, respectively. Trees with a significant relationship between time and MB production are in bold (p<.05).
DISCUSSION Overall the treatment group did not show an increase in MB production as expected. Two of them even stopped producing MBs all together. Folgarait et al. (19 94) did a similar study in a green house where they controlled for certain abiotic factors. In two of their three experiments they found a significant increase in MB production for the plants in which they systematically removed the MBs. Trees B2 (Figure 5 ) and B10 (Table 1) in my treatment group did demonstrate that what Folgarait et al. (1994) found in their greenhouse experiment could be replicated outside of a greenhouse. In nature there are many more variables to consider. Canopy cover could possibly c ontribute to why some of these plants did not produce many MBs. The tree in my treatment group with the greatest canopy cover produced the least MBs overall, and the tree that produced the most MBs also received the most direct sunlight. It would seem that percent canopy cover plays into MB production. While there was more herbivore damage to the trees without ants a study by Agrawal & Rutter (1998) showed that in the case of C. obtusifolia it did not appear that damage to the plant to reduce the MB product ion significantly. The control group had an overall increase in MB production. I can only think to attribute this to the fact that the weather in the last few days of the experiment was considerably sunnier and warmer. Equation R 2 n F p A2 Ln(y)=4.07 .67ln(x) 0.60 15 19.8 .0006 A3 Ln(y)= 2.64 + 1.7ln(x) 0.53 13 4.58 .01 A4 Ln(y)= 5.14 2.34ln(x) 0.91 13 60.03 .0002 A5 Ln(y)= 2.81 1.29ln(x) 0.85 13 11.61 .080 A6 Ln(y)= 3.85 0.57ln(x) 0.56 13 13.99 .0033 A7 Ln(y)= 2.68 1.03ln(x) 0.73 13 16.60 .0065 A8 Ln(y)= .46 0.31ln(x) 0.09 13 .86 .38 TABLE 2: Regression analyses for MÃ¼llerian body (MB) production in Monteverde, Costa Rica over 13 day s for Cecropia obtusifolia whose ants were restricted from removing them. MB production and day were log transformed. A2 8 each represents one of the 7 individual C. obtusifolia sampled. Trees with a significant relationship between time and MB production are in bold (p<.05). FIGURE 5. Relationship between the total number of MÃ¼llerian bodies (MB) produced per C. obtusifolia tree per day. Tree B2 was samp led in Monteverde, Costa Rica from Nov.3 18, 2010 (y= 8.08 + 3.96x, R 2 = 0.50, p<.05).
This trend did not hold for all the plants in the treatment group, possibly due to other contributing factors such as lower nutrients in the soil or just the small sample size. Folgarait et al. (1994) did find that there was no significant difference in their control and treatment groups whe n the MB removal experiments were done in low nutrient soil. In the second test where the ants were restricted from removing the MB, there was an overall rapid decrease in MB production. This is an ideal example of how these inducible defenses increase p lant fitness in the presence of herbivores (Agrawal 1998) but in the absence of ants the plant will allocate these resources elsewhere, such as to growth, reproduction or other defenses (Karban et al . 1997). Generally, within three days the production of M the absence of ants and reduce production very rapidly. Herbivory damage was significantly higher on unoccupied plants. These results differ from a previous study done in Monteverd e last spring, where they found no significant difference (Riha 2010). The discrepancy in these results could be contributed to several factors: (1) that there has been observed about a two fold increase in herbivore damage from the dry season to the wet ( Coley 1983), (2) that in Cecropia, ant number is positively associated with resistance against herbivores (Rocha & Bergallo 1992). All of the ant plants sampled had substantial population of ants, which would have aided in protection. And perhaps the ants are more helpful for the trees during the wet season. Three main factors seem to contribute to MB production: weather, leaf age and rate of MB accumulation. Weather, being the amount of sunlight and energy the plant receives was able to cause a substanti al increase in MB production. Leaf age contributes as well, where only the youngest three leaves or so, produce MB. Since the youngest leaves tend to suffer more to herbivory, the plant is possibly trying to draw the ants up towards the younger leaves. And finally the rate of accumulation of MBs plays a major role in MB production, the more MBs left on the Trichilia the less MBs the plant will produce. It appears from this study that C. obtusifolia removal and accumulation of M Bs and adjust accordingly. This adjustment is rapid, with obvious signs seen within about a day or two, an example of an induced defense. The energy saved when the C. obtusifolia decrease its MB production is most likely diverted elsewhere, to growth or se condary metabolites for defenses. I would expect the plant to boost its other defense when ants are not present, since when comparing herbivore damage between trees inhabited by ants and those without, the ant mutualism appears FIGURE 6. Relationship between the total number of MÃ¼llerian bodies (MB) produced over a 16 day span (Nov. 3 18, 2010) and the percent canopy cover for the control group of C. obtusifolia (n=5), in Monteverde, Costa Rica. (F=0.68, R 2 =0.18 p=.46)
quite beneficial. Abiotic work could be done to fully understand how each plays into the species fitness. A similar larger scale study could help conclude how this variation in MB accumulation likely to impa ct the mutualism. While ant plant mutualisms may appear ideal we must remember that ants and chemical defenses are similar in that their numbers (concentrations) are dependent on various environmental factors (Karban et al. 1999). There will always be thi s constant cost benefit trade off between energy put towards defenses versus herbivory damage, where the plant is trying to maximize its fitness. To conclude the costliness of MB production is worth the trade off for the defenses received from the Azteca a nts, though as soon as the MB are no longer removed the plant will quickly divert that energy elsewhere. ACKNOWLEDGMENTS I would like to give special thanks to A. Masters for helping me come up with and execute my project, P. Allen for his humor and insig ht with statistics, M. Calderon for his overall assistance in everything and buying snacks. And everyone else for making this an amazing trip. LITERATURE CITED Agrawal A.A. 1998. Induced responses to herbivory and increased plant performance. Science 27 9:1201 1202 Agrawal, A.A. and M.T. Rutter. 1998. Dynamic anti herbivore defense in ant plants: the role of induced responses. Oikos 83: 227 236 Coley, P.D. 1983. Herbivory and defensive characteristics of tree species in a lowland tropical forest. Ecologic al Monographs. Vol. 53 no. 2 pp.209 229 Davidson, D.W. and B.L. Fisher. 1991. Symbiosis of ants with Cecropia as a function of light regime. Ant Plant Interactions (eds. C. Huxley & D. K. Cutler), pp. 289 309. Oxford University Press, New York. Folgarait, P. J., and D. W. Davidson. 1995. Myrmecophytic Cecropia : antiherbivore defenses under different nutrient treatments. Oecologi a . 104.2: 189 206. Folgarait, P.J, H.L Johnson, and D.W Davidson. 1994. Responses of Cecropia to experimental removal of MÃ¼llerian bodies. Functional Ecology. 8.1: 22 28. Garnsey, J.J. 1999. Structural and chemical changes in leaves of Cecropia (Cecropiaceae) with plant age. CIEE Tropical Ecology and Conservation Program Fall 1999. Haber, W.A. 2000. Monteverde ecology and conservation of a tropical cloud forest. Oxford University Press Inc. New York, NY. HÃ¶lldobler B, and E.O. Wilson. 1990. The ants. Belknap Press, Cambridge, Mass Janzen D.H. 1969. Allelopathy by myrmecophytes: the ant Azteca as an allelopathic agent of Cecropia . Ecolo gy 50:147 153 Janzen, D.H. 1973. Dissolution of mutualism between Cecropia and its Azteca ants. Biotropica. 5.1: 15 28. Karban, R., A.A. Agrawal, and M. Mangel. 1997. The benefits of induced defenses against herbivores. Ecology 78: 1351 1355 Karban, R, A.A Agrawal, J.S Thaler, and L.S Adler. 1999. Induced plant responses and information content about risk of herbivory. Trends in Ecology & Evolution. 14.11: 443. Longino, J.T. 1991. Azteca ants in Cecropia trees: taxonomy, colony structure, and behavior. In A nt plant interactions, edited by C.R. Huxley and D.F. Cutler. Oxford: Oxford University Press. Putz, F.E. and N.M. Holbrook. 1988. Further observations on the dissolution of mutualism between Cecropia and its ants: the Malaysian case. OIKOS 53: 121 125 Ric kson, F. R. 1976. Anatomical development of the leaf trichilium and MÃ¼llerian Bodies of Cecropia
peltata L. American Journal of Botany. 63.9: 1266 1271. Riha, K.A. 2010. Herbivory and alkaloid concentration in Cecropia obtusifolia (Cecropiaceae) trees with and without ants. CIEE Tropical Ecology and Conservation Program Spring 2010 Rocha CFD, and H.G. Bergallo. 1992. Bigger ant colonies reduce herbivory and herbivore residence time on leaves of an ant plant: Azteca muelleri vs. Coelomera ruficornis on Cecr opia pachystachya . Oecologia 91:249 252 Scalley, M. 1993. Effectiveness of defense tradeoffs in two species of cecropia. EAP Tropical Biology Program 73 8 Schupp, E.W. 1986. Azteca protection of Cecropia : ant occupation benefits juvenile trees. Oecologia . 70.3: 379 385.