1 | INTRODUCTION
Various seed and fruit traits such as size and color tend to be correlated, forming what are known as “dispersal syndromes” that may have arisen to attract particular dispersers (Brodie, 2017; Valenta and Nevo, 2020) or may facilitate transport from their parent plants and move by wind or currents (Nilsson et al., 1991; Bullock & Claeke, 2000; Ohnishi et al., 2008; Nathan et al., 2011). Among seed dispersal by animals, dispersal depending on ants is specifically called myrmecochory. Myrmecochory is widespread among angiosperms and ecologically important; over 11,000 species of myrmecochorous plants in 77 families and 334 genera, participate in myrmecochory across various ecosystems (Lengyel et al., 2010). One of the evolutionary significance of myrmecochory is considered as directed seed dispersal. The advantages of the directed ant-dispersal hypothesis are that ant nests maintain moist conditions and the seeds carried to the nest can escape desiccation and have a higher seedling survival (Levey & Byrne, 1993), and the ants bury the seeds at depths where humidity and temperature are suitable for germination (Gibson, 1993).
Seeds of many myrmecochorous plants possess appendages, such as elaiosomes containing various fatty acids and proteins. These are likely to be nutritional rewards for ants who disperse the seeds (Brew et al., 1989; Lanza et al., 1992). If the seeds are carried to an ant nest, the elaiosomes will likely be consumed inside the nest before the seeds are discarded (Culver & Beattie, 1978). Some granivorous ants, such as species of genera Messor,Pheidole, and Tetramorium , lose or abandon their seeds during transportation and thus also disperse seeds without elaiosomes (Retana et al., 2004; Kobayashi, 2009). Various studies have shown that nutritional contents, such as lipids, amino acids, and proteins, in the elaiosome are important for inducing the seed dispersal. In contrast, how seeds without elaiosomes enable efficient dispersal by ants is less investigated, although some chemical signaling might be involved (Youngsteadt et al., 2008). It is also unclear whether the seeds are discarded on suitable sites in a directed manner in species without elaiosomes. In this study, we investigated seed dispersal without specialized nutritional rewards to ants. We also tested the possibility of directed seed dispersal by ants by examining whether abandonment of dummy seeds is directionally biased towards suitable location using an achlorophyllous plant whose seeds do not have elaiosomes.
We used the plant Monotropastrum humile (D. Don) H. Hara (Ericaceae) and the ant Nylanderia flavipes Smith (Formicidae) to test the hypothesis. M. humile is an achlorophyllous and myco-heterotrophic plant that inhabits forest floors in the temperate regions of Asia (Ohashi et al. 2016). The plant is about 10 cm height and does not perform photosynthesis and parasitizes the fungi ofRussula and Lactarius species of Russulaceae (Bidartondo and Bruns, 2001). M. humile is a full myco-heterotrophic species (Merckx, 2013). Seeds are smaller than 0.5 mm in length (Uehara & Sugiura, 2017) and require their host fungi for germination, thus, seed destination is a determinant of survival. Their host, the mycorrhizal fungi of Russula species, inhabit the shallow areas of the forest floor (Rachel, 2004; Courty et al., 2008), and Imamura and Kurogi (2003) reported that M. humile roots inhabit the forest floor to a depth of 5–10 cm. For seed dispersal, M. humile uses the insects wandering on the ground when their berries fall to the ground during the fruiting season. Cockroaches (Uehara & Sugiura, 2017) and camel crickets (Suetsugu, 2017) have been reported as their seed dispersers. According to these studies, the M. humile berries fall to the ground during the fruiting season, the wandering insects are presumed to consume pulpy flesh around the seeds and consequently contribute to the dispersal of seeds involved in their foraging behavior.
Although M. humile seeds have no elaiosomes and seem to be non-myrmecocohorous plants (Suetsugu, 2017; Uehara & Sugiura, 2017), Suetsugu (2017) and Uehara and Sugiura (2017) reported that several ant species, including N . flavipes , frequently visited theM. humile fruits. In addition, we have observed, by using a fixed-point camera, that N. flavipes ants were attracted to the fruits and seeds of M. humile and that a part of the seeds, although not many, were certainly carried by them. Since the seeds are coated with a phlegmatic organic layer, it is possible thatM. humile attract the ants with the fruit- and seed odors and that their seed are carried by the ants. The seeds might be abandoned after the outer components are consumed. Consequently, the seed could be dispersed to the suitable, host-inhabited sites on the forest floor.
To validate directed seed dispersal by ants, it is necessary to track where the seeds are carried to and discarded. However, the seeds of myrmecochorous plants are small and difficult to track in the field. Thus, at first, we performed the bioassay using intact and odorless seeds to verify whether the seed odors of M. humile are key signals for dispersal by ants. Afterwards, bioassays were performed using small pieces of the paper coated with the odor component ofM. humile seeds as dummy seeds. We tested whether the dummy seeds were carried into the nests and were carried out of the nests by the ants.