4.2 | Implication by the biased discard of dummy seeds
Although we could not track where the genuine seeds discarded because of their small-size, the author M. Y. observed a part of the genuine seeds carried into the nests are left in a corner of the nest (data not shown). Thus, the ant nest could be one of the seed directions. The dummy seeds coated with mixed odors were carried into the nest and subsequently discarded out of the nest by the ants. Although 13 dummy seeds were left inside the nest (17 were discarded outside), it appears that seeds staying in the nest can also be considered as dispersed sinceN. flavipes nests are located under litter or on rotting trees (Kallal & LaPolla, 2012). The direction of discards was biased, and the dummy seeds were discarded mostly onto the water-moistened glass beads. In the ant species Myrmica rubra , it is known that workers discard inert items and corpses in different destinations (Diez et al., 2012). In our study, it was also observed that ant corpses were dumped into cotton wool compartments, where the dummy seeds were not dumped. These results suggests that the ants differently use the discard destinations depending on what they are discarding, and the seeds could be discarded specific locations by the ants.
Mycorrhizal fungi of Russula species as the hosts of M. humile inhabit the top shallow layer of the forest floor (Rachel, 2004; Courty et al., 2008), and Imamura and Kurogi (2003) reported thatM. humile roots inhabit 5-10-cm depth as parasites on Russulaceae fungi. Thus, where the ant nests are located and where the seeds are discarded (moist soil near the surface) is considered to be close to their hosts and to be suitable for germination, growth and survival of the M. humile seeds although the mycorrhizal roots have not been quantified before.
The dummy seeds were left inside the nest or most frequently discarded onto the water-moistened glass beads section. This observation can support two seed-dispersal hypotheses: 1) seeds relocated to the moist environments including ant nests can escape desiccation and have enhanced germination and seedling survival; 2) seeds relocated to ant nests can also escape predators (modified from Hanzawa et al 1988).
Our results suggest that seed dispersal of M. humile by ants is directional. Directed seed dispersal and its ecological significance have rarely been verified in achlorophyllous herbaceous species (reviewed by Giladi, 2006). With regards to myco-heterotrophic plants, large parts of them produce dust-like and anemochorous seeds, which require host fungus in their germination, and which could disperse over long distances (Merckx, 2013). Thus, there is a trade-off of seed dispersal between distance and direction. Since the genus Monotropastrum , producing fleshy fruit, is thought to evolve from the relative and anemochorous genus Monotropa , producing capsule fruit (Bidartondo & Bruns, 2001), this speciation is considered as from anemochory (distance dispersal) to zoochory including myrmecochory. Therefore, the verification of the role of myco-heterotrophic herbaceous species in the temperate forests of Asia, and of the role of seeds without specialized rewards for ants, presents a new aspect of directed seed dispersal of achlorophyllous and myco-heterotrophic plants involving ants.