Survival and Scatterhoarding of Frugivores-Dispersed Seeds as a Function of Forest Disturbance

The effect of forest disturbance on survival and secondary dispersal of an artificial seed shadow (N= 800) was studied at Brownsberg Natural Park, Suriname, South America. We scattered single seeds of the frugivore‐dispersed tree Virola kwatae (Myristicaceae), simulating loose dispersal by frugivores, in undisturbed and disturbed secondary forest habitats. Seed survival rate aboveground was high (69%) within 2 wk and was negatively correlated with scatterhoarding rate by rodents, the latter being significantly lower in the undisturbed forest (9%) than in the disturbed forest (20%). Postdispersal seed predation by vertebrates was low (3%) and infestation of seeds by invertebrates was almost zero in all instances. Therefore, secondary seed dispersal by rodents in forest is not as critical for recruitment as observed among other bruchid‐infested large‐seeded species. Secondary seed dispersal by rodents may, however, facilitate seedling recruitment whether cached seeds experience greater survival than seeds remaining above ground surface.     

Successful invasion of the neotropical species Piper aduncum in rain forests in Papua New Guinea

Abstract. Piper aduncum is a neotropical invasive species which has spread throughout Papua New Guinea over the past three decades. It has become a most successful alien woody plant in New Guinea, occurring from sea level up to 2000 m a.s.l. The species prefers initial stages of forest succession and is particularly common in recently abandoned gardens representative of a system of swidden agriculture. It often attains high cover, suppresses other pioneer species and becomes the absolute dominant species in these habitats. The species is now also spreading into naturally disturbed habitats far from direct human influence, such as natural tree‐fall gaps, landslides and frequently flooded stream banks. It has, however, never been found in a closed primary forest. The species germinates from faeces of mammal and bird species, and we conclude that dispersal through endozoochory contributes to this species’ extraordinary success in Papua New Guinea. A similar invasion behaviour has been documented over a large geographic area, from Malaysia to Fiji. Piper aduncum has attributes which are common amongst successful invasive species: (1) a large native geographic range; (2) aggressively colonizing disturbed habitats in its native area; (3) relatively small seeds; (4) a short juvenile period; (5) a large seed production every year.     

Seed survival of the tropical tree Cryptocarya aschersoniana (Lauraceae): Consequences of habitat disturbance

Seed survival in soil could be strongly influenced by habitat characteristics, but little is known about the behaviour of seeds sensitive to desiccation in seed banks installed in natural or disturbed habitats. Cryptocarya aschersoniana seeds disperse at the end of the rainy season but do not germinate immediately; thus, they may form seed banks in soil. This study evaluated the behaviour of C. aschersoniana seed banks induced in the natural environment of the species and in a disturbed area. Recently harvested C. aschersoniana germination units were characterized according to their water content, germination and viability. In 2011 and 2012, seed banks were established by burying samples of seeds in the understory of a semi‐deciduous forest. In 2012, samples were also buried in a disturbed area. The seed banks were sampled at certain time intervals, and the samples were characterized as described above. Precipitation and air temperature data were collected. As a result, seeds in the seed bank established in the natural environment form a transient seed bank and showed the same behaviour in both years studied. A germination peak was observed starting 210 days after burial (coinciding with the onset of the rainy season) and reached germination percentages higher than 80% at the end of the experiment for both years. Seed mortality did not exceeded 28% in the natural environment. However, in the disturbed environment, the seeds lost their viability more rapidly, with 90% of the seeds becoming unviable 240 days after burial. Germinated seeds in the disturbed environment (maximum 21%) were not able to establish seedlings. These results underscore the importance of maintaining a natural, undisturbed forest for the conservation of this species.     

Leaf-cutting ant nests support less dense and impoverished seed assemblages in a human-modified Caatinga dry forest

Regenerating forests make up an increasingly large portion of tropical landscapes worldwide and regeneration dynamics may be influenced by leaf-cutting ants (LCA), which proliferate in disturbed areas and collect seeds for fungus culturing. Here, we investigate how LCA influences seed fate in human-modified areas of Caatinga dry forest. We evaluate the seed deposition and predation on Atta opaciceps nests, foraging habitat surrounding nest and control habitat away of nest influence of 15 colonies located along a forest cover gradient during the rainy and dry seasons. For each habitat, four 50-cm² plots were established and all seeds on the soil surface were collected along 1 year. We recorded 13,628 seeds distributed among 47 species and 36.57% of the total seeds did not show any sign of predation. Nest mound habitats supported low-density and species-poor seed assemblages, which were taxonomically distinct from the control habitats. These effects only occurred in the rainy season. The proportion of undamaged seeds were similar across the habitats. While forest cover did not influence seed assemblage in terms of species richness or seed predation, it did interact with habitat type via increments in seed abundance as forest cover increased across the nests. Forest cover also affected seed composition, but only in the rainy season. These results indicate that LCA decrease seed deposition in areas under their influence, particularly on the nest mounds. As LCA profit from human disturbance in the Caatinga, their role as seed ‘sinks’ should be enhanced in disturbed Caatinga patches, particularly during the rainy season, when most of the plant recruitment occurs. Our findings reinforce the importance of LCA as drivers of forest dynamics and resilience in human-modified landscapes.     

Response of ants to human‐altered habitats with reference to seed dispersal of the myrmecochore Corydalis giraldii Fedde (Papaveraceae)

Anthropogenic habitat disturbance has potential consequences for ant communities. However, there is limited information on the effects of ant responses on associated ecological processes such as seed dispersal. We investigated the effect of disturbance on the abundance, richness, and composition of ant communities and the resulting seed‐dispersal services for a herbaceous myrmecochore, Corydalis giraldii (Papaveraceae), in an undisturbed habitat (forest understory), moderately disturbed habitat (abandoned arable field), and highly disturbed habitat (road verge) on Qinling Mountains, China. In total, we recorded 13 ant species, and five out of these were observed to transport seeds. The community composition of dispersers was significantly different amongst habitats. The richness of the dispersers did not differ among the habitats, but their total abundance varied significantly across habitats and was 21% lower in the road verge than in the abandoned arable fields. The major seed‐dispersing ant species in both the forest understory and the abandoned arable field were large‐bodied (Myrmica sp. and Formica fusca, respectively), whereas the major seed‐dispersing ants found in the road verge were the small‐bodied Lasius alienus. This difference resulted in lower seed removal rates and dispersal distances in the road verge than in the other two habitats. The different dispersal patterns were attributed primarily to differences in dispersing ant abundance and identity, most likely in response to habitats with different degree of anthropogenic disturbance. The possible influence of disturbance on the ecological specialization of ant‐seed dispersal interaction is also discussed.     

Seed Dispersal by Kerama Deer (Cervus nippon keramae) on Aka Island,the Ryukyu Archipelago,Japan1

Endozoochorous seed dispersal by kerama deer (Cervus nippon keramae) was investigated at four sites on Aka Island, the Ryukyu Archipelago, Japan. Kerama deer feces contained seeds of 35 plant species. Most of the germinated seeds were small (1.3 ± 0.18 mm) and had a hard coat, comparable to the germinated seeds in several other studies of endozoochorous seed dispersal by herbivores. Both the numbers and species compositions of seeds differed among the four sites, reflecting differences in the food available to kerama deer among habitats. Although many graminoid seeds were found in the two open habitats (Sites A and B) and in the adjoining forest habitat (Site C), woody plant seeds dispersed endozoochorously by birds and mammals were dominant in the forest (Site D), away from the marsh. Although a majority of the graminoid species was growing in open habitats such as the marsh and open fields, few were growing in the forest. Therefore, the kerama deer spread many open‐habitat graminoid seeds to the open and forest habitats. If a high density of kerama deer persists for a long time and gaps in forests are created by browsing kerama deer or by other means, graminoid species may spread substantially on Aka Island. Therefore, it is possible that seed dispersal by forest deer contributes to the expansion of grasslands.     

Relationships among the seed rain,seed bank and vegetation of a Hawaiian forest

Abstract. Hawaiian ecosystems are prone to invasion by alien plant species. I compared the seed rain, seed bank, and vegetation of a native Hawaiian forest to examine the potential role that seed ecology plays in allowing alien species to invade native forest. Absolute cover of seed plants in the forest was 126 %, annual seed rain was 5 713 seeds m^-2 yr^-1, and the mean density of seedlings emerging from the seed bank averaged across four seasons was 1 020/m². The endemic tree Metrosideros polymorpha was the most abundant species in the vegetation, seed rain and winter seed bank. Overall, native seed plants comprised 95 % of the relative cover in the vegetation and 99 % of the seeds in the seed rain, but alien species comprised 67 % of the seeds in the seed bank. Alien species tended to form persistent seed banks while native species formed transient or pseudo-persistent seed banks. Dominance of the seed bank by alien species with persistent seed banks suggests that aliens are favorably placed to increase in abundance in the vegetation if the forest is disturbed.     

格氏栲天然林林窗和林下种子散布及幼苗更新研究

以格氏栲(Castanopsis kawakamii)天然林为研究对象,探讨了林窗和林下格氏栲种子雨、种子库的分布特征及幼苗更新状况。结果表明:林窗种子雨总量和完好种子密度高于林下,未成熟种子比例低于林下;林窗和林下种子雨高峰期掉落数量分别占种子雨总量的77.13%和74.5%;林窗种子库储量低于林下,种子库中以全食或捡拾种子比例最高,其中种子库储量垂直分布表现为枯落物层(约占2/3)>腐殖质层(0～5 cm)(约占1/3)>心土层(5～10 cm)(小于1%),以格氏栲种子占绝对优势;格氏栲从种子到幼苗的转化率低,林窗中格氏栲完好种子密度与幼苗密度均高于林下。林窗微生境提高了种子散布过程中格氏栲成熟种子的密度和比例,有利于促进格氏栲的幼苗更新,表明林窗在格氏栲种群恢复过程中扮演着重要角色。     

Post‐Dispersal Seed Removal and Germination Selected Tree Species Dispersed by Cebus capucinus on Barro Colorado Island,Panama1

Dispersal quality, an important component of seed disperser effectiveness, may strongly affect the rate of plant recruitment. Here we evaluated the quality of Cebus monkey dispersal by comparing the secondary removal fate and germination of fresh and Cebus‐ingested seeds of nine tree species on Barro Colorado Island, Panama. Overall, rates of secondary seed removal by vertebrates were low, with most Cebus defecations remaining undisturbed for extended periods on the forest floor. Only four of 30 feces were completely buried by dung beetles, and we found significantly higher vertebrate removal of defecated seeds than control seeds for only one species, Cordia bicolor. Seed germination varied greatly between plant taxa. Seeds of 3 out of 9 species showed significantly higher percent germination after monkey gut passage than control fresh seeds. Germination times tended to be shorter for defecated than for control seeds but were significantly different only for one of nine species, Cecropia insignis. Low rates of seed removal from Cebus feces, coupled with high germination probabilities, suggest high dispersal effectiveness for Cebus and contrasts strongly with patterns of post‐dispersal seed fate recorded for other primate species.     

Shifts in an invasive rodent community favoring Black rats (Rattus rattus) following restoration of native forest

One potential, unintended ecological consequence accompanying forest restoration is a shift in invasive animal populations, potentially impacting conservation targets. Eighteen years after initial restoration (ungulate exclusion, invasive plant control, and out planting native species) at a 4 ha site on Maui, Hawai'i, we compared invasive rodent communities in a restored native dry forest and adjacent non‐native grassland. Quarterly for 1 year, we trapped rodents on three replicate transects (107 rodent traps) in each habitat type for three consecutive nights. While repeated trapping may have reduced the rat (Black rat, Rattus rattus) population in the forest, it did not appear to reduce the mouse (House mouse, Mus musculus) population in the grassland. In unrestored grassland, mouse captures outnumbered rat captures 220:1, with mice averaging 54.9 indiv./night versus rats averaging 0.25 indiv./night. In contrast, in restored native forest, rat captures outnumbered mouse captures by nearly 5:1, averaging 9.0 indiv./night versus 1.9 indiv./night for mice. Therefore, relatively recent native forest restoration increased Black rat abundance and also increased their total biomass in the restored ecosystem 36‐fold while reducing House mouse biomass 35‐fold. Such a community shift is worrisome because Black rats pose a much greater threat than do mice to native birds and plants, perhaps especially to large‐seeded tree species. Land managers should be aware that forest restoration (i.e. converting grassland to native forest) can invoke shifts in invasive rodent populations, potentially favoring Black rats. Without intervention, this shift may pose risks for intended conservation targets and modify future forest restoration trajectories.     

Non-native mammals are weak candidates to substitute ecological function of native avian seed dispersers in an island ecosystem

Although prominent examples exist of non-native species causing substantial ecological harm, many have neutral or positive effects, including filling surrogate roles once performed by extinct native organisms. We tested the ecological roles of two non-native mammals as seed dispersers or seed predators in Guåhan, which, due to invasive brown tree snakes (Boiga irregularis), is devoid of native seed dispersers–birds and bats. We conducted feeding trials with captive rats (Rattus spp.), which are present but uncommon due to predation by snakes, and pigs (Sus scrofa), which are abundant. We examined if and how they interacted with common forest fruits. We then compared how any gut-passed or animal-handled seeds germinated compared to seeds left in whole fruit or depulped seeds. Rats and pigs interacted with most of the fruits and seeds (>80%) that they were fed. Of those, most seeds were destroyed—78% for rats and 90% for pigs, across both native and non-native plant species. Compared to seeds germinating within whole fruits, rats improved germination of the seeds that they handled without ingesting, while pigs diminished the germination of seeds that they handled. The small percentage of seeds (approximately 1.5% for rats and 5% for pigs) that survived gut passage germinated in higher proportions than those in whole fruits. Percentages of seed survival to germination are lower than found in similar studies with native avian frugivores. Our results indicate that pigs and rats have mixed effects on seeds, but are not suitable surrogates for native seed dispersers.     

Native and exotic invasive plants have fundamentally similar carbon capture strategies

1.  Leaf trait relationships of native and exotic invasive species from a range of habitats were compared to assess consistency across habitats and the role of disturbance.
2.  One hundred and twenty-two native and exotic species were sampled in five habitats in eastern Australia. Specific leaf area, foliar nitrogen ( N _mass), assimilation rate ( A _mass) and dark respiration ( R _mass) were measured for each species. Plants were classified into four types: native undisturbed, native disturbed, exotic invasive undisturbed and exotic invasive disturbed.
3.  All traits were positively correlated and slopes were homogeneous within habitats. Significant differences between plant types in slope elevation were found in only two of 18 cases. There were significant shifts in group means along a common slope between plant types within habitats. These shifts were associated with disturbed vs. undisturbed areas, with plant types from disturbed areas having higher trait values.
4.   Synthesis . Exotic invasive and native species do not have fundamentally different carbon capture strategies. The carbon capture strategy of a species is strongly associated with disturbance, with species from disturbed sites having traits that confer capacity for fast growth. Thus, differences between exotic invasives and natives may reflect differences in the environmental conditions of the sites where they occur rather than differences between exotic invasives and natives per se .     

Influence of Invasive Tree Kill Rates on Native and Invasive Plant Establishment in a Hawaiian Forest

Fire tree (Morella faya) has invaded extensive areas of wet and mesic forest on the Island of Hawai’i, forming nearly monospecific stands. Our objective was to identify a method of controlling M. faya, which would allow native plants to establish while minimizing establishment by invasive plants. Treatments (logging all trees, trees left standing but girdled, and incremental girdling over 20 months) were selected to kill M. faya stands at different rates, thereby creating different conditions for species establishment. Leaf litter was either removed or left in place; seeds and seedlings of three native pioneer species, three native forest species, and three alien invasive species were then added to determine their ability to establish. Native pioneer species established best in the log and girdle treatments, whereas seedling emergence of native forest species was higher in the girdle and incremental girdle treatments. Seedlings of invasive species emerged faster than the natives, but each of them responded differently to the stand treatments. Leaf litter reduced seedling emergence for all species, with small‐seeded species (<1 mg/seed) affected most under low light conditions. No single method eliminated all invaders, but girdling of M. faya provided suitable conditions for most native species. If combined with selective removal of the most disruptive alien species and native seed additions, girdling could be an effective general strategy for restoring native forests that have been overwhelmed by woody invaders.     

Assessing the importance of disturbance, site conditions, and the biotic barrier for dandelion invasion in an Alpine habitat

Several factors have been identified as relevant in determining the abundance of non-native invasive species. Nevertheless, the relative importance of these factors will vary depending on the invaded habitat and the characteristics of the invasive species. Due to their harsh environmental conditions and remoteness, high-alpine habitats are often considered to be at low risk of plant invasion. However, an increasing number of reports have shown the presence and spread of non-native plant species in alpine habitats; thus, it is important to study which factors control the invasion process in these harsh habitats. In this study, we assessed the role of disturbance, soil characteristics, biotic resistance and seed rain in the establishment and abundance of the non-native invasive species Taraxacum officinale (dandelion) in the Andes of central Chile. By focusing on human-disturbed patches, naturally disturbed patches, and undisturbed patches, we did not find that disturbance per se, or its origin, affected the establishment and abundance of T. officinale. The abundance of this non-native invasive species was not negatively related to the diversity of native species at local scales, indicating no biotic resistance to invasion; instead, some positive relationships were found. Our results indicate that propagule pressure (assessed by the seed rain) and the abiotic soil characteristics are the main factors related to the abundance of this non-native invasive species. Hence, in contrast to what has been found for more benign habitats, disturbance and biotic resistance have little influence on the invasibility of T. officinale in this high-alpine habitat.     

Patterns of Seed Dispersal and Dispersal Failure in a Hawaiian Dry Forest Having Only Introduced Birds

Dry forests are among the most endangered natural communities in the Hawaiian Islands. Most have been reduced to isolated trees and small forest fragments in which native tree species reproduce poorly. The replacement of native birds by introduced generalists may be contributing to dry forest decline through modification of seed dispersal patterns. To document seed dispersal by introduced birds, we conducted foraging observations on fleshy-fruited trees and measured seed rain under trees and in adjacent open areas for 1 year in a dry forest dominated by native trees. Although trees covered only 15.2 percent of the study area, 96.9 percent of the bird-dispersed seeds were deposited beneath them. The Japanese white-eye (Zosterops japonicus) was the principal dispersal agent. Among bird-dispersed seeds, those of the invasive tree Bocconia frutescens accounted for 75 percent of all seeds collected beneath trees (14.8 seeds/m²/yr) and the invasive shrub Lantana camara accounted for 17 percent. Although nearly 60 percent of the reserve's native woody species possess fleshy fruits, introduced birds rarely disperse their seeds. Native trees accounted for <8 percent of all bird-dispersed seeds and are consequently experiencing dispersal failure by falling directly under parent trees. Smaller-seeded non-native plants, in contrast, may be benefiting from dispersal by introduced birds. Current dispersal patterns suggest that these readily disseminated non-native plants may eventually replace the remaining native flora.     

Integrating revegetation with management methods to rehabilitate coastal vegetation invaded by Bitou bush (Chrysanthemoides monilifera ssp. rotundata) in Australia

Two multi‐year field experiments investigated the effects of integrating revegetation with invasive plant management methods to rehabilitate coastal dune and woodland vegetation invaded by Bitou bush (Chrysanthemoides monilifera (L.) Norl. ssp. rotundata (DC.) Norl.) in New South Wales, Australia. The revegetation technique used was to sow directly seeds of three native species common to coastal habitats. Management treatments involved combinations of prescribed fire, manual removal of Bitou bush and an application of herbicide. Addition of native seeds significantly increased density of native species in both habitats. The benefits of manually removing Bitou bush were observed only where densities of native species were at their lowest. Fire increased densities of some native species in the woodland, but decreased those of others in the dune. Densities of Acacia longifolia ssp. sophorae (Labill.) Court (woodland) and of Banksia integrifolia L.f. (woodland and dune) were significantly reduced within 4 months of herbicide application, alone or in combination with other treatments. The majority of these effects, however, did not persist. Manual removal in both habitats and addition of seed in the woodland were most effective in reducing Bitou bush densities when applied post‐fire. Herbicide treatment on its own or in combination with other treatments did not significantly reduce Bitou bush densities by the end of the experiments. We conclude that restoration of coastal ecosystems invaded by a major invasive plant species requires a whole‐of‐system approach involving revegetation in combination with known management methods to assist recovery of native species in the longer term.     

Seed Dispersal by Bats and Birds in Forest and Disturbed Habitats of Chiapas,Mexico1

We examined seed dispersal by bats and birds in four habitats of the Selva Lacandona tropical rain forest region, Chiapas, Mexico. The four habitats represented a disturbance gradient: active cornfield, ten-year-old abandoned cornfield, cacao plantation, and forest. Using seed traps examined before sunrise (0400 h) and before sunset (1800 h), we compared volant vertebrate seed dispersal, assuming that seeds found at the end of the night were dispersed by bats and those found at the end of the day were dispersed by birds. We did not find seeds from other frugivores such as monkeys or opossums. In all habitats bats dispersed more seeds than birds. In most months bats also dispersed more seeds than birds, except in December when no seeds were found in the traps. Bats also consistently dispersed more species of seeds than birds, although a x2 comparison showed differences not to be significant. Fifty percent of the species represented in the dispersed seeds in all habitats were pioneer species. Cecropia seeds represented a high percentage (up to 87% of those dispersed by bats and up to 83% by birds) of dispersed seeds that fell in our traps. The influence of bats and birds on secondary successional processes is likely to be fundamental for the establishment of vegetation. Since bats dispersed more seeds than birds (primarily to disturbed areas and consisting primarily of pioneer species), they are likely to play an important role in successional and restoration processes among habitats as structurally and vegetationally different as cornfields, old fields, cacao plantations, and forest.     

Forest edges and fire ants alter the seed shadow of an ant-dispersed plant

Exotic species invade fragmented, edge-rich habitats readily, yet the distinct impacts of habitat edges and invaders on native biota are rarely distinguished. Both appear detrimental to ant-dispersed plants such as bloodroot, Sanguinaria canadensis. Working in northeastern Georgia (USA), an area characterized by a rich ant-dispersed flora, fragmented forests, and invasions by the red imported fire ant, Solenopsis invicta , I monitored the interactions between ants and S. canadensis seeds in uninvaded forest interiors, uninvaded forest edges, invaded forest interiors, and invaded forest edges. I observed 95% of the seed dispersal events that occurred within the 60-min observation intervals. Seed collection rates were similar among all four (habitat × invasion) groups. The presence of invasive ants had a strong effect on seed dispersal distance: S. invicta collected most seeds in invaded sites, but was a poorer disperser than four of five native ant taxa. Habitat type (interior versus edge) had no effect on seed dispersal distance, but it had a strong effect on seed dispersal direction. Dispersal towards the edge was disproportionately rare in uninvaded forest edges, and ants in those habitats moved the average dispersed seed approximately 70 cm away from that edge. Dispersal direction was also skewed away from the edge in uninvaded forest interiors and invaded forest edges, albeit non-significantly. This biased dispersal may help explain the rarity of myrmecochorous plants in younger forests and edges, and their poor ability to disperse between fragments. This is the first demonstration that forest edges and S. invicta invasion influence seed dispersal destination and distance, respectively. These forces act independently.     

Promotion of exotic weed establishment by endangered giant kangaroo rats (Dipodomys ingens) in a California grassland

Giant kangaroo rats (Dipodomys ingens) continually modify their burrow precincts by digging tunnels, clipping plants, and other activities. In the valley grasslands of the Carrizo Plain Natural Area (San Luis Obispo County, California), this chronic disturbance to soil and vegetation promoted the establishment of exotic ruderal and early successional plant species. Erodium cicutarium, Bromus madritensis ssp. rubens, and other Mediterranean annuals were found to constitute a very large proportion of the vegetation on giant kangaroo rat precincts. When vegetation on precincts was compared with the vegetation in less disturbed intermediate areas located between precincts, species richness, cover and frequency of exotic plants were significantly greater on precincts. The reverse was found for native species. In addition, exotic species encountered in this study had significantly larger seeds than did native species, suggesting that these granivorous kangaroo rats preferentially cache large weed seeds on their precincts. Since the kangaroo rats depend on exotic plants for food and the exotic plants depend upon the kangaroo rats to disturb their habitat continually, the weed-kangaroo rat relationship is mutualistic. This strong relationship may also inhibit population growth of native grassland plants which occupy disturbed habitats but have difficulty competing with exotic weeds for resources. From a conservation perspective, this mutualism presents an intractable management dilemma. Restoration of valley grasslands where endangered giant kangaroo rats occur, to conditions where native species dominate, may be impossible.     

Seed rain of fleshy and dry propagules in different habitats in the temperate rainforests of Chiloé Island,Chile

In temperate rainforests on Chiloé Island in southern Chile (42°S), most canopy trees bear fleshy, avian‐dispersed propagules, whereas emergent tree species have dry, wind‐borne propagules. In the present study, the following hypothesis was tested: regardless of species, fleshy propagules are deposited in greater numbers in canopy gaps and in forest margins and hence have a more heterogeneous seed shadow than wind‐dispersed propagules. To test this hypothesis, the seed rains of these two types of propagules were compared in the following forest habitats: (i) tree‐fall gaps (edges and centre); (ii) forest margins with adjacent pastures; and (iii) under closed canopy (forest interior). Seed collectors (30‐cm diameter) were placed in two (15 and 100 ha) remnant forest patches (n = 60–100 seed collectors per patch) distributed in the four habitats. Seeds were retrieved monthly from each collector during two reproductive seasons (1996, 1997). In both years, the seed rain was numerically dominated by two species with dry propagules (Laureliopsis philippiana and Nothofagus nitida) and three species with fleshy fruits (Drimys winteri, Amomyrtus luma, and Amomyrtus meli). The seed shadows of the two species with dry, wind‐dispersed seeds differed markedly. Seeds of L. philippiana were deposited predominantly in canopy openings, whereas N. nitida seeds fell almost entirely in the forest interior. The fleshy‐fruited species, Drimys and Amomyrtus spp., had similar seed deposition patterns in the various habitats studied, but the between‐year differences in seed rain were greater in Drimys winteri than in Amomyrtus spp. Although no more than 10% of fleshy‐fruited propagules reached the margins of the patch, approximately 7% of these were carried there by birds. Every year, canopy gaps (pooling data from edges and centres) concentrated approximately 60% of the total seed rain of both propagule types in both forest patches. Forest margins received less than 20% of the total seed rain, which was largely dominated by fleshy‐fruited species. Seed shadows were a species‐specific attribute rather than a trait associated with propagule type and dispersal mode.