Tamarins and Dung Beetles: An Efficient Diplochorous Dispersal System in the Peruvian Amazonia Laurence Culot 1,2,5 , Darren J. Mann 3 , Fernando J. J. Mu ˜ noz Lazo 4 , Marie-Claude Huynen 1 , and Eckhard W. Heymann 2 1 Behavioral Biology Unit, University of Lie ` ge, Quai van Beneden, 22 ba ˆ t. I1, 4020 Lie ` ge, Belgium 2 Department of Behavioral Ecology and Sociobiology, German Primate Centre (DPZ), Kellnerweg 4, 37077 Goettingen, Germany 3 Oxford University Museum of Natural History, Parks Road, Oxford OX13PW, UK 4 Department of Ecology and Fauna, National University of Peruvian Amazon, Pevas, Iquitos, Peru ABSTRACT Dung beetles fulfill several key functions in ecosystems but their role as secondary seed dispersers is probably one of the most complex ones. Various factors, such as seed characteristics, dispersal pattern induced by the primary disperser, season, and habitat, can affect the seed–beetle interaction. Particularly little is known about the fate of seeds primarily dispersed in small feces. The aim of this study was to investigate the effects of these factors on the dung beetle community (species composition, number and size of individuals) and its consequences on burial occurrence and depth of seeds primarily dispersed by two tamarin species. We captured dung beetles in a Peruvian rain forest with 299 dung-baited pitfall traps to characterize the dung beetle community. Seed burial occurrence and depth were assessed by marking in situ 551 dispersed seeds in feces placed in cages. Among these seeds, 22.5 percent were buried by dung beetles after 2 d. We observed a significant effect of the amount of dung, season, time of deposition, and habitat on the number of individuals and species of dung beetles, as well as on seed burial occurrence and depth, while the tamarin species significantly influenced only the number and the size of dung beetles. This seed dispersal loop is particularly important for forest regeneration: small to large seeds dispersed by tamarins in secondary forest can be buried by dung beetles. These seeds can thus benefit from a better protection against predation and a more suitable microenvironment for germination, potentially enhancing seedling recruitment. Abstract in Spanish is available at http://www.blackwell-synergy.com/loi/btp. Key words: Saguinus fuscicollis; Saguinus mystax; scarabaeinae; secondary forest; seed burial; seed dispersal. SEED DISPERSAL IS OFTEN ACCOMPLISHED through a series of consec- utive events. Primary seed dispersers, such as birds, bats, and pri- mates, allow the seeds to escape density dependent mortality by depositing them away from the parent plant. Secondary seed dis- persers such as scatter-hoarding rodents, ants, and dung beetles can bring the seeds to suitable microsites where the probability of seed- ling establishment is disproportionably high. Indeed, these second- ary seed dispersers often provide the seeds with better protection against predators and a better microenvironment for germination due to the removal of perishable pulp or fecal matter or due to seed burial (Vander Wall & Longland 2004). In the case of seeds pri- marily dispersed through mammal defecation, dung beetles play a key role in the multi-staged seed dispersal process. Dung beetles bury feces before using it for feeding or ovipo- sition (Hanski & Cambefort 1991). During this process, one or several seeds embedded in the dung may be accidentally buried in tunnels under the feces or in dung balls away from the dung source (Estrada & Coates-Estrada 1991, Shepherd & Chapman 1998, Feer 1999, Vulinec 2000, Andresen 2002a). From the beetle’s per- spective, seeds present in dung are contaminants (Andresen & Feer 2005). From the seed’s perspective, displacement and burial may significantly affect its survival (Chambers & MacMahon 1994, Vander Wall & Longland 2004). Buried seeds are more likely to escape detection by seed predators than those remaining on the surface (Estrada & Coates-Estrada 1991, Shepherd & Chapman 1998, Andresen 1999). In addition, they may end up in a more humid environment, appropriate for germination (Chambers & MacMahon 1994). However, deep seed burial by dung beetles can also prevent seedling emergence (Andresen 2001, Andresen & Levey 2004). As one event influences the next in the seed dispersal loop (Wang & Smith 2002), the defecation pattern of a vertebrate will have an important influence on the dung beetle community and on its handling of the feces, and hence, on the seeds deposited by the dispersers (Andresen 2002b, Ponce-Santizo et al. 2006). Most stud- ies on seed–beetle interactions have focused on relatively large feces (usually 20–25 g of fecal matter used for bait in pitfall traps; Estrada & Coates-Estrada 1991, Andresen 2001), and this might bias the understanding of this complex process toward dung beetle assem- blages attracted to large feces. One might expect that, on small fe- ces, the dung processing carried out by the beetle community or the dung beetle community itself might differ. Yet, it is known that certain foraging and dung processing behaviors, such as pellet roll- ing or perching at low height in the vegetation, are specific adap- tations to small feces (Howden & Nealis 1978, Hanski & Cambefort 1991). Indeed, the smaller the amount of fecal matter, the less likely it is that true dung balls will be formed or that large beetles will be attracted (Peck & Howden 1984, Hanski & Received 5 August 2009; revision accepted 10 March 2010. 5 Corresponding author; e-mail: laurence.culot@ulg.ac.be BIOTROPICA 43(1): 84–92 2011 10.1111/j.1744-7429.2010.00655.x 84 r 2010 The Author(s) Journal compilation r 2010 by The Association for Tropical Biology and Conservation