The relationship between animals and their gut flora is simultaneously one of the most common and most complex symbioses on Earth. Despite its ubiquity, our understanding of this invisible but often critical relationship is still in its infancy. We employed adult Neotropical butterflies as a study system to ask three questions: First, how does gut microbial community composition vary across host individuals, species and dietary guilds? Second, how do gut microbiota compare to food microbial communities? Finally, are gut flora functionally adapted to the chemical makeup of host foods? To answer these questions we captured nearly 300 Costa Rican butterflies representing over 50 species, six families, and two feeding guilds: frugivores and nectivores. We characterized bacteria and fungi in guts, wild fruits, and wild nectars via amplicon sequencing and assessed the catabolic abilities of the gut microbiota via culture‐based assays. Gut communities were distinct from food communities, suggesting that the gut environment acts as a filter on potential colonists. Nevertheless, gut flora varied widely among individuals and species. On average, a pair of butterflies shared 21% of their bacterial species and 6% of their fungi. Host species explained 25–30% of variation in microbial communities while host diet explained 4%, suggesting that non‐dietary aspects of host biology play a large role in structuring the butterfly gut flora. Much of the variation between species correlated with host phylogeny. Host diet was related to gut microbial function: compared to frugivores, nectivores’ gut flora exhibited increased catabolism of sugars and sugar alcohols and decreased catabolism of amino acids, carboxylic acids, and dicarboxylic acids. Since fermented juice contains more amino acids and less sugar than nectar, it appears that host diet filters the gut flora by favoring microbes that digest compounds abundant in foods. By quantifying the degree to which gut communities vary among host individuals, species and dietary guilds and evaluating how gut microbial composition and catabolic potential are related to host diet, this study highlights the linkages between structure and function in one of the most complex and ubiquitous symbioses in the animal kingdom.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Submitted July 12, 2011; Accepted February 3, 2012; Electronically published April 25, 2012 Online enhancement: appendix. Dryad data: http://dx.doi.org/10.5061/dryad.144v45c6. The University of Chicago Press andabstract: How strong is selection for cheating in mutualisms? The answer depends on the type and magnitude of the costs of the mutualism. Here we investigated the direct and ecological costs of plant defense by ants in the association between Cordia nodosa, a myrmecophytic plant, and Allomerus octoarticulatus, a phytoecious ant. Cordia nodosa trees produce food and housing to reward ants that protect them against herbivores. For nearly 1 year, we manipulated the presence of A. octoarticulatus ants and most insect herbivores on C. nodosa in a full-factorial experiment. Ants increased plant growth when herbivores were present but decreased plant growth when herbivores were absent, indicating that hosting ants can be costly to plants. However, we did not detect a cost to ant colonies of defending host plants against herbivores. Although this asymmetry in costs suggests that the plants may be under stronger selection than the ants to cheat by withholding investment in their partner, the costs to C. nodosa are probably at least partly ecological, arising because ants tend scale insects on their host plants. We argue that ecological costs should favor resistance or traits other than cheating and thus that neither partner may face much temptation to cheat.

Summary1. Ants provide variable protection against herbivores to ant-plants (i.e. myrmecophytes and myrmecophiles). The ways in which ant-plants dynamically adjust both their direct (chemical and physical) and indirect (biotic) defences in response to varying levels of herbivory are not well understood. 2. We experimentally generated a broad range of ant-attendance levels and herbivory pressures in a tropical myrmecophyte, Cordia nodosa, which allowed exploration of the inducibility of and interactions between direct and indirect resistance traits. 3. In response to increased herbivory, host plants encouraged indirect (biotic) defence by increasing domatium volume, regardless of whether ants were present on the plant. When ants were present, larger domatia housed more workers, which in turn decreased herbivory on adjacent leaves. 4. Independent of the presence of ants, plants responded to increased herbivory by inducing both chemical (phenolics) and structural (leaf toughness, trichomes) resistance traits; these traits were associated with reduced palatability to a folivorous beetle. 5. Synthesis. Our results show that both direct and indirect defences are inducible in C. nodosa, which suggests that C. nodosa may retain direct defences as insurance against varying levels of protection from its ant bodyguards. Thus, the predictions of optimal defence theory are not violated: although C. nodosa invests in multiple forms of defence, they are not redundant.

Facultative, intracellular bacterial symbionts of arthropods are usually vertically transmitted with great fidelity, and may dramatically affect host biology and reproduction. The length of these symbiont-host associations may be thousands of years, and while symbiont loss is predicted, there have been very few observations of a decline of symbiont infection from high frequencies to low. In a population of the sweetpotato whitefly species (Bemisia tabaci MEAM1) in Arizona, USA, we documented the frequency decline of a strain of Rickettsia in the bellii clade from near-fixation in 2011 to 36% of whiteflies infected in 2017. In previous studies, Rickettsia had been shown to increase from 1% to 97% from 2000 to 2006 and remained at high frequency for at least five years. At that time, Rickettsia infection was associated with both fitness benefits and female bias. In the current study we established matrilines of whiteflies from the field (2016, Rickettsia infection frequency= 58%), and studied Rickettsia vertical transmission, fitness and sex ratios associated with Rickettsia infection, and symbiont titer. The vertical transmission rate was high, approximately 98%. Rickettsia infection in the matrilines was not associated with fitness benefits or sex ratio bias, and appeared to be slightly costly, as more Rickettsia-infected individuals produced non-hatching eggs. Overall, the titer of Rickettsia in the matrilines was lower in 2016 than in the whiteflies collected in 2011, but the titer distribution appeared bimodal, with high and low titer lines, and constancy within lines of the average titer over three generations. We found no association between Rickettsia titer and fitness benefits or sex ratio bias, and this change in the interaction between symbiont and host in 2016 whiteflies may explain the drop in symbiont frequency we observed.