Effectiveness of rodents as local seed dispersers of Holm oaks

Abstract

In this study we assessed the effectiveness of rodents as dispersers of Quercus ilex in a patchy landscape in southeastern Spain. We experimentally followed the fates of 3,200 marked and weighed acorns from dispersal through the time of seedling emergence over three years. Rodents handled about 99% of acorns, and dispersed 67% and cached 7.4% of the dispersed acorns. Most caches were recovered and consumed, and only 1.3% of the original experimental acorns were found alive in caches the following spring. Dispersal distances were short (mean = 356.2 cm, median = 157 cm) and strongly right-skewed. Heavier acorns were dispersed further and were more likely to be cached and survive than lighter acorns. All caches were in litter or soil, and each contained a single acorn. Rodents moved acorns nonrandomly, mostly to oaks and pines. Most surviving acorns were either in oaks, a poor microhabitat for oak recruitment, or shrubs, a suitable microhabitat for oak recruitment. Our results suggest that rodents, by burying a relatively high proportion of acorns singly in shrubs and pines, act as moderately effective dispersers of Q. ilex. Nonetheless, this dispersal comes at a very heavy cost.

Tree-shrub associations in grazed woodlands: first rodents, then cattle?

“…The reported predation of unburied acorns under shrubs was similarly high (Li and Zhang 2003;Pons and Pausas 2007;Pérez-Ramos and Marañón 2008), probably due to predator avoidance and associated higher mice densities under shrubs. Also other seed tracking experiments with rodents as dispersal agent found similar low acorn survival varying from 0.4 to 1.4% (Takahashi et al 2006;Gómez et al 2008). On the one hand, such low survival percentages lead to the questioning of the effectiveness of wood mice as dispersal agent.…”

“…We pooled the data over all the four plots as the number of observations for each microhabitat was too low to perform Chisquare tests for each plot separately. Expected values (number of acorns per microhabitat) were calculated by assuming that for undirected (random) dispersal the amount of acorns per microhabitat is proportional to the area that it covers (Gómez et al 2008). Significant deviations from expected values would indicate directed dispersal towards a particular microhabitat.…”

“…C1 year (%) with Gómez et al (2008) who reported less than expected dispersal toward shrubs in a Mediterranean (not-cattle grazed) oak woodland. This difference may be explained by the shorter mean dispersal distance as observed in their study (3.6 vs. 7.9 m in our study), preventing acorns to reach microhabitats other than where initially placed in.…”

“…On the one hand, such low survival percentages lead to the questioning of the effectiveness of wood mice as dispersal agent. On the other hand, even very low survival percentages are important for plant recruitment (Levey and Byrne 1993;Vander Wall et al 2005) and may add up to a considerable recruitment over time, especially for long-lived trees such as Quercus robur (Takahashi et al 2006;Gómez et al 2008). …”

“…Those studies that measured high post-dispersal predation of acorns (Li and Zhang 2003;Iida 2006), therefore questioned the effectiveness of wood mice as dispersers. In contrast, many other studies underline the importance of seed dispersal by wood mice for tree regeneration despite high predation rates, as even a tiny fraction of acorns that escapes from post-dispersal seed predation may be sufficient for a substantial tree recruitment over time (Gómez 2004;Den Ouden et al 2005;Vander Wall et al 2005;Gómez et al 2008;Smit et al 2009). …”