View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by RERO DOC Digital Library BEHAVIOR Oviposition Preference of Botanophila Flies (Diptera: Anthomyiidae) Towards Stroma Size of Epichloe¨ (Hypocreales: Clavicipitaceae) Hosts 1,2 3 4 1 K. GO´ RZYN´ SKA, Z. OLSZANOWSKI, A. LEUCHTMANN, AND M. LEMBICZ Ann. Entomol. Soc. Am. 107(2): 532Ð538 (2014); DOI: http://dx.doi.org/10.1603/AN13094 ABSTRACT Stromata of grass-infecting fungi from the genus Epichloe¨ (Clavicipitaceae: Ascomy- cota) serve as a food source and egg-laying surface for ßies of genus Botanophila (Diptera: Antho- myiidae). Larger stromata should make it possible for ßies to lay more eggs and provide more food to offspring. This hypothesis was tested in four different grassÐfungus associations that occur in central Poland. In two of these associations, Epichloe¨ bromicola on Elymus repens and Epichloe¨ typhina on Puccinellia distans, ßies showed a preference for longer stromata, and egg density on these stromata was signiÞcantly higher than in the other two associations. A negative correlation between egg density and offspring success was observed in only one association, E. bromicola–El. repens. However, offspring success in this association did not differ signiÞcantly from offspring success in associations with lower egg density on the stromata, in which ßies showed no preference for the stroma length. Long-term observations (2000Ð2010) of ßyÐfungus interaction in the E. typhinaÐP. distans association showed that ßy preference toward stroma length may vary over time but with no clear tendency. No signiÞcant correlations were found between the larval density on a stroma and either larval weight or mortality. The results of the current study question our assumptions that egg laying depends on the stroma length and the fate of eggs laid (i.e., their hatching success and the condition, in terms of weight and survival, of the larvae) on egg density. It is possible that ßies choose stromata based on attributes other than size. KEY WORDS Anthomyiidae, Botanophila sp., Epichloe¨ sp., oviposition preference In nature, organisms always co-occur with other or- ßowers and, after hatching, larvae feed on developing ganisms from different taxa and most of these co- seeds (e.g., Pellmyr and Huth 1994, Anstett et al. occurring organisms interact with each other. At pres- 1997). Our research, which focuses on interactions ent, we know that relationships between organisms between Botanophila ßies and Epichloe¨ fungi (Leuch- present a continuum of intensity and of gains and tmann 2007; Go´rzyn´ ska et al. 2010, 2011; Lembicz et al. losses that vary in time and space (Bronstein 1994a,b; 2013), follows this line of study. 2001). Many studies show that there are places in The ancestors of those Botanophila ßies that later space where a species does not enter any interaction became connected with Epichloe¨ fungi, were likely (so called “cold spots”) and places in which the rela- grass herbivores initially, similar to many modern rep- tionship between partners is strongest (so called “hot resentatives of the family Anthomyiidae. Those indi- spots”; Gomulkiewicz et al. 2000, Thompson and Cun- viduals who also ate fungal fruiting bodies called stro- ningham 2002). Hence, interspeciÞc interactions are mata, which appeared on grasses at some point in time, difÞcult to deÞne unambiguously. gained an advantage over other individuals because A spectacular example of interactions that are cur- stromata were a richer source of nutrients than grass rently studied is the relationship between Botanophila tissues (Bultman 1995). At present, stromata of fungi ßies (Diptera: Anthomyiidae) and fungi from the ge- are the only known food source for some Botanophila nus Epichloe¨ (Clavicipitaceae: Ascomycota). This re- species (Kohlmeyer and Kohlmeyer 1974). However, lationship resembles interactions between some in- the Epichloe¨ÐBotanophila interaction is not entirely sects and ßowering plants in which insects both restricted to the feeding relationship. Botanophila fe- pollinate and parasitize a plant (pollinating seed par- males lay eggs on Epichloe¨ stromata and transfer sper- asites). During pollination, insects lay eggs in visited matia of two different types in their faeces, thus en- abling cross-fertilization of these heterothallic fungi 1 Department of Plant Taxonomy, A. Mickiewicz University, Umul- (Bultman and White 1988; Bultman et al. 1995, 1998; towska 89, 61-614 Poznan´ , Poland. Bultman and Leuchtmann 2008). Some volatile com- 2 Corresponding author, e-mail: [email protected]. pounds produced by Epichloe¨ fungi are also known to 3 Department of Animal Taxonomy and Ecology, A. Mickiewicz attract ßies (Steinebrunner et al. 2008b). University, Umultowska 89, 61-614 Poznan´ , Poland. 4 Plant Ecological Genetics, Institute of Integrative Biology (IBZ), Botanophila ßies have long been considered the Universita¨tstrasse 16, 8092 Zu¨ rich, Switzerland. main, if not only, vector of fungal spermatia (Bultman 0013-8746/14/0532Ð0538$04.00/0 ᭧ 2014 Entomological Society of America March 2014 GORZY´ NSKA´ ET AL.: OVIPOSITION PREFERENCE OF FLIES ON Epichloe¨ STROMATA 533 Table 1. Number of stromata collected in particular grass–fungus association and their origin GrassÐfungus No. of collected stromata Site Location Fly taxa association 2000 2001 2002 2003 2006 2008 2009 2010 D. glomerataÐE. typhina Morasko 52Њ 27.857Ј N Botanophila phrenione (Se´guy) 55 16Њ 55.868Ј E Botanophila dissecta (Meigen) Pakos´c´ 152Њ 47.531Ј N B. phrenione 30 30 55 127 29 130 18Њ 06.118Ј E P. distansÐE. typhina Janikowo 52Њ 46.384Ј N B. phrenione 67 30 30 30 144 145 147 18Њ 08.032Ј E Giebnia 52Њ 46.544Ј N B. phrenione 30 30 30 30 146 149 147 18Њ 06.190Ј E Pakos´c´152Њ 47.284Ј N n.d. 60 30 30 30 30 143 150 18Њ 06.257Ј E Pakos´c´ 252Њ 47.293Ј N B. phrenione 149 148 146 18Њ 06.721Ј E Taxon 1 El. repensÐE. bromicola Pakos´c´ 152Њ 47.397Ј N Botanophila lobata (Collin) 54 18Њ 06.064Ј E Taxon 1 H. lanatusÐE. clarkii Morasko 52؇ 27.910Ј N B. phrenione 67 16Њ 55.370Ј E et al. 1998); thus, the insectÐfungus interaction was m2, which were located within fungus-infected grass recognized as obligatory. At present, we know that populations. Stromata were collected at the end of stromata can be fertilized without the ßyÕs contribu- June, after the maximum number of eggs was laid. tion (Rao and Baumann 2004; Go´rzyn´ ska et al. 2010, Measurements. The presence or absence of evi- 2011); therefore, the fungus can reproduce sexually in dence of insect visits was determined based on the the absence of the insect. However, the Botanophila occurrence of eggs on stromata, which were observed life cycle depends strictly on the fungus because the under a microscope (Olympus SZ61-TR; Olympus, ßy larvae, like the adults, feed on fungal stromata until Tokyo, Japan). To determine the relationship between their metamorphosis. stroma length and egg number, stroma length was mea- The nutritional dependence of Botanophila larvae sured and the number of eggs on each stroma was de- on fungal stromata suggests that a pregnant ßy will termined. The number of eggs comprised both the num- select larger stromata on which to lay eggs, to ensure ber of unhatched eggs and larvae. In addition, density of that more food will be available to its offspring. Based insects on a stroma was determined as the mean number on this assumption, the aim of this study was to eval- of eggs per 1 mm of stroma. To characterize the effect of uate 1) the relationships between stroma size and both egg density on the offspring success, the following pa- ßy presence and the number of eggs laid, 2) the effect rameters were determined for each stroma: 1) egg den- of egg density on offspring success, and 3) the effect 2 of larval density on larval survival and weight. The sity per 1 mm , 2) offspring success, expressed as the investigations were conducted in four different grassÐ percentage of hatched eggs, and 3) stroma length. The fungus associations. One of these associations has been comparison of the aforementioned traits was con- observed since 2000, which made it possible to deter- ducted between the four grassÐfungus associations for mine whether this interaction has changed over time. the stromata collected in 2010, within the Puccinellia distans–Epichloe¨ typhina (P.d.ÐE.t.) association for the years 2000Ð2010, except 2004, 2005, and 2007, and Materials and Methods between the P.d.ÐE.t. and Dactylis glomerataÐE. Collection of Material. The research was conducted typhina (D.g.ÐE.t.) associations in the locality Pakos´c´ from 2000 to 2010 and involved three fungal species 1 for the years 2000, 2003, 2008, and 2009. The last occurring on four grass species in six localities situated comparison was conducted to show whether there are in central Poland (Table 1). In each of these localities, any differences in the studied traits of the ßyÐfungus Epichloe¨ stromata were sampled from study plots of 25 relationship, if this relationship occurs in the same Table 2. Presence of Botanophila eggs on fungal stromata and the relationship between stroma length and egg presence in each grass–fungus association in 2010 GrassÐfungus Egg presence Stroma lenght (mm) Ϯ SD N P LeveneÕs test t df P association n (%) Eggs present No eggs P.d.ÐE.t. 440 134 (30.4) 17.23 Ϯ 3.85 14.85 Ϯ 3.79 NS 6.04 438 Ͻ0.001 D.g.ÐE.t. 130 115 (88.5) 48.19 Ϯ 13.37 45.20 Ϯ 14.18 NS 0.81 128 NS El.r.ÐE.b. 54 54 (100) 28.50 Ϯ 4.91 Ð Ð NS Ð Ð H.l.ÐE.c. 67 11 (16.4) 31.36 Ϯ 5.16 28.63 Ϯ 6.81 NS 1.26 65 NS P.d.ÐE.t., Puccinellia distansÐEpichloe¨ typhina; D.g.ÐE.t., Dactylis glomerata–Epichloe¨ typhina; H.l.ÐE.c., Holcus lanatusÐEpichloe¨ clarkii; El.r.Ð E.b., Elymus repens–Epichloe¨ bromicola.