Estimating human-mediated dispersal of seeds within an Australian protected area

Dispersal is critical step in plant invasions but there is limited information about human-mediated long distance seed dispersal, including in protected areas. Seed dispersal by hikers was quantified for five invasive species (the native Acaena novae-zelandiae, and the non-native weeds Rumex acetosella Anthoxanthum odoratum, Dactylis glomerata and Festuca rubra) in part of Australia’s Kosciuszko National Park. The proportion of seeds remaining attached to trousers and socks was quantified for replicated short (150 m) and long (5,000 m) distance walks. Functions were fitted for each dataset, and parameters compared among species and between trousers and socks. Dispersal data were combined with attachment rates and the number of people undertaking walks to estimate the total number of weed seeds that might be dispersed. The power exponential function gave the best fit for the majority of datasets, indicating that detachment probability decreased with distance. Seeds of all five species were more tightly attached to socks than trousers, with some seeds still present on socks at 5,000 m. Anthoxanthum and Acaena seeds were more tightly attached to clothing than the other species. Theoretically 1.9 million seeds could be dispersed on socks or 2.4 million seeds on trousers through a season but the actual numbers are likely to be much lower because of limited weed seed at the start of the walks. Because of differences in attachment and detachment rates, seeds from Acaena were more likely to be dispersed longer distances. Long distance human-mediated seed dispersal is potentially a major cause of spread of invasive weeds into protected areas that favours some invasive species over others.     

Long‐distance dispersal of Black Spear Grass (Heteropogon contortus) seed on socks and trouser legs by walkers in Kakadu National Park

Humans can contribute to the long‐distance dispersal of many plants, including weeds. We assessed the distance for which seed remained attached to the socks and trouser legs of walkers. The experiment in Kakadu National Park, Australia used seed of the Australian native Black Spear Grass (Heteropogon contortus) as a surrogate for the potential dispersal of weed seed. Two models were fitted to the data, with a double exponential model fitting the data slightly better than a power exponential model. Although 19% of seed were dispersed within the first 5 m of walking, most (55%) seed remained attached to socks and trousers at 5 km. Humans may, therefore, unintentionally carry and eventually disperse this grass and other invasive grasses with similar long awns and pointed seeds over long distances. Implementing strategies to reduce the potential for seed to attach to walkers (gaiters) and reduce potential dispersal (removal and careful disposal of seed) would reduce the risks posed by this type of human‐mediated seed dispersal.     

Effects of Two Types of Clothing on the Day-Night Variation of Core Temperature and Salivary Immunoglobulin a

Circadian variations in core temperature, skin temperatures, heart rate, and salivary immunoglobulin A (IgA) were compared in subjects wearing two different types of clothing that covered, or left uncovered, their extremities. The experiments were carried out on six female subjects at an ambient temperature of 24 ± 0.5°C and relative humidity of 50 ± 5%. One type of clotiiing consisted of long-sleeved shirts, full-length trousers, and socks (Type L: 1042 g, 1.048 clo); the other was half-sleeved shirts and knee-length trousers (Type H: 747 g, 0.744 clo). The main results were as follows: (i) The level of rectal temperature during night sleep was significantly lower with Type H than Type L clothing, and cosinor analysis indicated a significantly higher circadian amplitude with Type H clothing, (ii) Skin temperatures in the lower extremities increased significantly more on retiring to sleep with Type H than Type L clothing, (iii) Heart rate was significantly lower with Type H than Type L clothing during the sleep period, (iv) The day-night variation of salivary IgA showed a pattern that was the inverse of that of rectal temperature (i.e., low in the daytime and high in the nighttime), and the concentration of salivary IgA was significantly higher with Type H than Type L clothing at 02:30. (v) Subjectively, the self-assessed sleep quality was better with Type H clothing. These results suggest that clothing that leaves the extremities uncovered might be regarded as favorable at the moderate temperature since it induces good sleep and activates the immune response. (Chronobiology International, 14(6), 607–617, 1997)     

Opening myrmecochory’s black box: what happens inside the ant nest?

In the process of seed dispersal by ants (myrmecochory), foragers bring diaspores back to their nest, then eat the elaiosome and usually reject viable seeds outside the nest. Here, we investigate what happens inside the nest, a barely known stage of the myrmecochory process, for two seed species (Viola odorata, Chelidonium majus) dispersed either by the insectivorous ant Myrmica rubra or by the aphid-tending ant Lasius niger. Globally, elaiosome detachment decreased ants’ interest towards seeds and increased their probability of rejecting them. However, we found marked differences in seed management by ants inside the nest. The dynamics of elaiosome detachment were ant- and plant-specific whereas the dynamic of seed rejection were mainly ant-specific. Seeds remained for a shorter period of time inside the nest of the carnivorous ant Myrmica rubra than in Lasius niger nest. Thus, elaiosome detachment and seed rejection were two competing dynamics whose relative efficiency leads to variable outcomes in terms of types of dispersed items and of nutrient benefit to the ants. This is why some seeds remained inside the nest even without an elaiosome, and conversely, some seeds were rejected with an elaiosome still attached. Fresh seeds may be deposited directly in contact with the larvae. However, the dynamics of larvae-seeds contacts were also highly variable among species. This study illustrates the complexity and variability of the ecological network of ant–seed interactions.     

Seed handling behaviours of native and invasive seed‐dispersing ants differentially influence seedling emergence in an introduced plant

相似文献   

Hydrodynamics and seed dispersal in the lower Amazon

相似文献   

Do plant-community-based grazing regimes lead to epizoochorous dispersal of high proportions of target species?

The contribution of large-herbivore epizoochory to the transfer of seeds within and between areas is thought to be significant. But often seeds of ubiquitous species are dispersed, which may enhance ruderalization processes. In order to study the dispersal of target species by sheep, we employed a community-based grazing approach followed by intra- and inter-area sheep transfers (maximum transfer distance 3 km). In case of inter-area transfers, well-developed target communities of an open inland sand ecosystem are used as “source”, linked to less-developed sand habitats (“sink”) via sheep. Also other factors determining which species become dispersed under field conditions were tested: seed surface structure, seed mass, plant seed-releasing height and animal behaviour. Finally the influence of animal movement on seed detachment and the actual arrival of seeds within a “sink” were studied.

Sheep transfers resulted in the dispersal of 56 seed species, dominated by Red List (seven species) and other target species. Quantitatively, most transported seeds belonged to target species, whereas graminoid competitors were highly under-represented. Morphological traits enhance the attachment probability regardless of seed mass. But for seeds without these epizoochory-facilitating traits, mass seems to affect attachment negatively. Plant height affected the number of species present in sheep coats but not the seed quantities. Probably certain species in the vegetation produced large numbers of seeds, e.g. low-growing Medicago minima with seed surface structures and high-growing species Verbascum phlomoides without seed surface structures. Also, although transfer half-times were three times lower than grazing half-times, naturally attached Stipa capillata and Agrimonia procera seeds showed no significant detachment rates during transfer, whereas considerable losses were found during grazing. Other 3-km sheep transfers did not result in significant losses of either epizoochorously transported seed or species quantities. Our study shows that community-based grazing can lead to the dispersal of especially target species. Besides that, different habitat fragments can be connected to each other via sheep without significant seed losses along the way.     

Seed dispersal in Cakile edentula var. lacustris : decoupling the fitness effects of density and distance from the home site

A factorial design of three densities of siblings at three local distances from seed parents was employed to distinguish effects of density from effects of dispersal distance on lifespan and fruit production of Cakile edentula var. lacustris, a plant with heteromorphic seeds. The segmented fruits produce two seed types: proximal and distal, with distal seeds having greater mass and greater dispersibility. Effects of longer distances (0.5 km and 30 km) on lifespan and fruit production were investigated using plants at low density. The prediction was tested that the greater seed mass of distal seeds increases fitness when seeds are dispersed into sites of unknown quality away from the home site or when seeds are dispersed to low density. High density caused earlier mortality and lower probability of reproduction. Distance from the maternal plant did not influence lifespan or reproduction at distances of 15 m or less, but lifespan was longer 0.5 km from the home site. No interaction was detected between the effects of density and distance on either lifespan or total fitness. Environmental conditions that influence fitness did not vary as a function of dispersal distance in this system, and favorable conditions at the home site did not persist between generations. Therefore, selection on dispersion patterns in natural conditions is likely to be through effects of density rather than dispersal distance. Proximal seeds had greater reproduction than distal seeds at the home site, and distal seeds had greater reproduction at the more distant sites (but not the most distant site), as expected, but these performance differences could not be attributed to differences in mass between the two seed types. Reduced seed mass was favored at the most distant site, but larger seed mass was favored most strongly at low density. Seeds that are dispersed to low density are larger, suggesting that although kin selection may limit the effectiveness of individual selection to increase seed mass under conditions of sibling competition, density-dependent individual selection on seed mass, rather than distant-dependent selection, also contributes to the observed associations among seed type, seed mass and dispersal ability. Received: 21 October 1996 / Accepted: 4 December 1996     

Effects of population density on male and female reproductive success in the wind-pollinated, wind-dispersed tree species Betula maximowicziana

The effects of population density on male and female reproductive success of Betula maximowicziana were evaluated in two mixed and two post-fire stands, with various population densities, ranging from 1.9 to 300.0 trees per ha, in central Hokkaido, Japan. First, we investigated ecological determinants of reproductive success (seed set and germination) of both seeds collected from the trees (tree seeds) and dispersed seeds collected from seed traps (dispersed seeds). We then evaluated the effects of population density on seed set and germination of tree seeds and dispersed seeds using a generalized linear mixed model (GLMM). Subsequently, we genotyped 950 seeds collected from mother trees and 940 seeds trapped after dispersal derived from tree seeds and dispersed seeds, respectively, using eleven microsatellite loci. Using the acquired data, we then evaluated the outcrossing rate and effective number of pollen donors (N _ep) of the tree seeds, and the genetic structure of both pollen pools and dispersed seed populations. The seed set and germination rate of dispersed seeds was significantly lower both in the lowest-density stand and in the highest-density stand. The GLMM revealed that seed set and germination rates of dispersed seeds may be maximal at approximately 120 trees per hectare (optimal density). Outcrossing rates were consistently high (t _m = 0.995), regardless of the population density. In contrast, N _ep was lower in the lowest-density stands. Significant genetic structure of the dispersed seed population was found in two low-density stands, probably due to the limitation of overlapping seed shadows.     

Effects of Intercropping on Growth and Reproductive Capacity of Late-emerging Senecio vulgaris L., with Special Reference to Competition for Light

Due to increased emphasis on long-term management of weed populationsin cropping systems with a reduced reliance on herbicides, theproduction of seeds by weeds that emerge after the criticalperiod for weed control is increasingly important. It was hypothesizedthat increased soil cover and light interception by a crop canopywould shorten the critical period for weed control and reducegrowth and fecundity of late-emerging weeds. This hypothesiswas tested in a series of field and glasshouse experiments inwhich competition for light was manipulated. Senecio vulgaris,an important weed in vegetable production systems, was chosenas the target plant, and canopies of pure and mixed stands ofleek and celery were used to provide shade. The time courseof light interception differed among the crop canopies. Increasingcompetition for light caused morphological changes to S. vulgaris,including a vertical shift in leaf area distribution. Increasedshading reduced biomass, capitula:shoot ratio and seed productionof S. vulgaris. However, the viability of seeds produced bythe shaded weed plants was not affected. Results indicate thatintercropping can increase light interception in a weakly competitivecrop such as leek and can contribute to weed suppression ina long-term strategy for weed management. Copyright 2001 Annalsof Botany Company Competition for light, late-emerging weeds, critical period, Apium graveolens L., celery, Allium porrum L., leek, Senecio vulgaris L., common groundsel, seed production, weed management, intercropping     

Seed-transmission in the ecology of nematode-borne viruses

Virus-free populations of vector nematodes can acquire tomato black ring (TBRV), raspberry ringspot (RRV) and arabis mosaic (AMV) viruses from weed seedlings grown from virus-carrying seed. When soils from fields where nematode-borne viruses occurred naturally were air-dried to kill vector nematodes and then moistened, TBRV and RRV occurred commonly in the weed seedlings that grew, but AMV occurred only rarely. Similar tests did not detect tobacco ringspot, grapevine fanleaf or tobacco rattle viruses in weed seeds in the single soil studied in each instance, although these three viruses are also seed-borne in some of their hosts. Many weed species, when infected experimentally, readily transmit TBRV and RRV to their seed, but the viruses were much commoner in naturally occurring seed of some of these species than of others. These discrepancies between the frequency of seed-transmission of viruses from experimentally infected plants and the extent of natural occurrence of infected seed seem largely to reflect the host preferences of the vectors. Infective Longidorus elongatus kept in fallow soil retained TBRV and RRV only up to 9 weeks. When weed seeds in the soil were then allowed to germinate, the nematodes reacquired virus from the infected seedlings. Some weed species were better than others as sources of virus. Persistence of these viruses in fields through periods of fallow or fasting of the vector therefore depends on a continuing supply of infected seedlings produced by virus-containing weed seeds. This is probably less true of viruses like AMV and grapevine fanleaf, which persist for 8 months or more in their vectors (Xiphinema spp.). A few seeds containing TBRV and RRV were found in soils free of vector nematodes, suggesting that the viruses are disseminated in weed seed. This probably explains how TBRV and RRV have reached a large proportion of L. elongatus populations in eastern Scotland.     

Functional uniqueness of a small carnivore as seed dispersal agents: a case study of the common palm civets in the Tabin Wildlife Reserve, Sabah, Malaysia

Many carnivorous mammals consume fruits and disperse the intact seeds to specific sites. Few studies have attempted to quantify this seed dispersal or evaluate its effectiveness, despite its potential importance and functional uniqueness. In the study reported here, we found that a frugivorous carnivore, the common palm civet (Paradoxurus hermaphroditus), generated seed shadows that are distinct from those of the sympatric frugivore, the pig-tailed macaque (Macaca nemestrina), and played a unique and important role in the regeneration of Leea aculeata (Leeaceae). We found that macaques dispersed the seeds randomly, while civets dispersed them non-randomly to sites such as the banks of small rivers, rain-flow paths, abandoned trails, and treefall gaps, which are characterized by low stem density and canopy cover. Seeds of L. aculeata that were dispersed by civets to the banks of rivers and gaps had significantly higher survival and growth rates than those dispersed to rain-flow paths or abandoned trails. Seeds dispersed by macaques or to random locations also had low survival. Although the effects of the civets on seed fate were not straightforward, compared with macaques and random dispersal, civets significantly enhanced the survival and growth of L. aculeata seeds after 1 year. These results indicate that non-random dispersal by civets is important for the persistence of L. aculeata. Civets may disperse other plant species and thus could have profound effects on forest dynamics.     

The relationship between seed dormancy,seed size and weediness,in Crepis tectorum (Asteraceae)

Summary I examined the germination characteristics of weed and outcrop populations of Crepis tectorum to test the hypothesis that the presumably more ephemeral weed habitat favors the highest levels of seed dormancy. The winter annual habit characterizing most plants of this species was reflected in a rapid germination of seeds sown in late summer. A slightly higher fraction of surface-sown seeds of weed plants delayed germination. Buried seeds of weed plants also survived better than seeds produced by plants in most outcrop populations, supporting the idea that weediness favors seed dormancy and a persistent seed bank. However, the differences in seed dormancy between the two ecotypes were small and not entirely consistent. Furthermore, high levels of seed dormancy were induced during burial in the outcrop group, suggesting that there is a potential for a dormant seed population in this habitat as well. Demographic data from one of the outcrop populations verified the presence of a large between-year seed bank. Possible environmental factors favoring seed dormancy in outcrop populations are discussed. The unusually large seeds of weedy Crepis contrasts with the relatively small difference in seed dormancy between the two ecotypes.     

Evolution of seed dispersal in North American <Emphasis Type="Italic">Ephedra</Emphasis>

Related plants often produce seeds that are dispersed in very different ways, raising questions of how and why plants undergo adaptive shifts in key aspects of their reproductive ecology. Here we analyze the evolution of seed dispersal syndromes in an ancient group of plants. Ephedra (Gymnospermae; Gnetales; Ephedraceae) is a genus containing ≈50 species in semiarid ecosystems worldwide and with three distinct types of cones. We collected mature cones and seeds of ten species of Ephedra in southwestern United States and measured nine morphological traits for each species. Principal component analysis and other data characterized three types of Ephedra cones and seeds. Species with dry, winged cone bracts are dispersed by wind (i.e., E. torreyana and E. trifurca), those with succulent, brightly-colored cone bracts are dispersed by frugivorous birds (i.e., E. antisyphilitica), and those with small, dry cone bracts and large seeds are dispersed by seed-caching rodents (e.g., E. viridis and E. californica). Two species (E. funerea and E. nevadensis) have cone and seed morphologies intermediate between two seed dispersal syndromes. Seed and cones traits were mapped onto two recent phylogenies to help reveal the evolutionary history of seed dispersal syndromes. Bird dispersal is thought to be the ancestral form of seed dispersal in ephedras as it is common in the Old World where Ephedra originated, but the three North American species dispersed by birds are not monophyletic. The two wind dispersed species in North America also do not cluster together, suggesting separate origins. Seed dispersal by seed-caching rodents is common in North America and appears to have evolved several times, but this syndrome is absent form other continents. The evolutionary history of Ephedra in North America suggests that the means of seed dispersal has been malleable. Evolutionary shifts were likely linked to changes in ecological conditions.     

Interspecific variation in persistence of buried weed seeds follows trade‐offs among physiological,chemical, and physical seed defenses

相似文献   

Variation in Weed Seed Fate Fed to Different Holstein Cattle Groups

Weed seeds may maintain their viability when passing through the digestive tract of cattle and can be therefore dispersed by animal movement or the application of manure. Whether different cattle types of the same species can cause differential weed seed fate is largely unknown to us particularly under non-grazed systems similar to Holstein-Friesian dairy farming. We investigated the effect on the seed survival of four weed species in the digestive tracts of four groups of Holstein cattle: lactating cows, feedlot male calves, dry cows and growing heifers. The weed species used were Cuscuta campestris, Polygonum aviculare, Rumex crispus and Sorghum halepense. Cattle excretion was sampled for recovery and viability of seeds at four 24 hourly intervals after seed intake. The highest seed recovery occurred two days after seed intake in all cattle groups. Averaged over weed species, dry and lactating cows had the lowest and highest seed recovery of 36.4% and 74.4% respectively. No significant differences were observed in seed recovery of the four weed species when their seeds were fed to dry cows. Based on a power model fitted to seed viability data, the estimated time to 50% viability loss after seed intake, over all cattle groups ranged from 65 h (R. crispus) to 76 h (P. aviculare). Recovered seeds from the dung of feedlot male calves showed the highest mortality among cattle groups. Significant correlation was found between seed viability and ruminal pH (r = 0.86; P<0.05). This study shows that management programs aiming to minimize weed infestation caused by livestock should account for the variation amongst cattle groups in seed persistence. Our findings can be used as a guideline for evaluating the potential risk of the spread of weeds via the application of cattle manure.     

Body size determines rates of seed dispersal by giant king crickets

New Zealand is home to giant king crickets called weta, which are the only insects known to consume fleshy-fruits and disperse seeds after gut passage. Although they disperse seeds in viable condition after consumption, their importance as seed dispersers is unknown. We conducted a series of field observations and laboratory experiments to investigate intraspecific variation in the capacity of Wellington tree weta (Hemidenina crassidens) to disperse seeds of tree fuchsia (Fuchsia excorticata). We asked three questions. How frequently do weta disperse fuchsia seeds? Do seed passage rates differ between sexes and different-sized weta? Might weta select for particular seed sizes via differential seed mortality after ingestion? A total of 2,272 F. excorticata seedlings germinated from 241 scats (i.e., faecal pellets) that were collected from the field. Experimental results showed that, on average, 15% of seeds ingested by weta successfully germinated, whereas 75% germinated in control trials. Larger weta dispersed greater numbers of seeds in experimental trials, while no differences in dispersal rates were observed between sexes. Regardless of sex and size, weta preferentially dispersed larger seeds. When interpreted collectively, results indicate that (1) weta are frequent seed dispersers of F. excorticata, although many seeds are destroyed during ingestion, (2) larger-bodied weta consistently disperse greater quantities of seeds, which is unusual in seed dispersal mutualisms, and (3) weta preferentially disperse larger seeds, suggesting that they might interact evolutionarily with New Zealand plants.     

文峪河上游华北落叶松林的种子雨、种子库与幼苗更新

华北落叶松林下更新不良,为探究其制约因素,开展了山西省文峪河上游5个华北落叶松林分的种子雨、土壤种子库和幼苗更新的研究。结果表明:(1)华北落叶松种子主要集中于9—10月散落。2011年为华北落叶松种子丰年:种子产量高,种子雨密度达(961.93±377.40)粒/m2;种子质量高,完整种子占(89.31±16.13)%。2012年为种子平年,种子产量低,种子雨密度为(252.73±115.12)粒/m2。华北落叶松种子雨主要源于毗邻树木,华北落叶松纯林和落叶松云杉林的种子雨密度显著高于其他3个针阔混交林。(2)土壤种子库主要由上年种子雨组成,2012年4月的土壤种子库密度为(695.18±297.23)粒/m2,完整种子占(59.73±9.56)%。种子自然萌发前,约(78.98±24.76)粒/m2具发芽力,基本可满足更新需要。但种子活力保持期少于2 a,只能形成短期持久土壤种子库。(3)华北落叶松更新不良,种子年后仍难以实现幼苗建成,当年生幼苗的出现频度平均为1.6%,且林下难以存活。幼苗发生与种子储量关联性不强,种源条件不是制约华北落叶松更新的主要因素。     

Long-distance Dispersal of Invasive Grasses by Logging Vehicles in a Tropical Dry Forest

Predicting responses of vegetation to environmental factors in human-altered tropical ecosystems requires an understanding of the controls on plant population expansion across landscapes (i.e., long-distance dispersal) as well as of factors affecting recruitment at local scales (i.e., microsite conditions). We studied the roles of light availability, habitat type, soil disturbance, and seed dispersal in a selectively logged forest in lowland Bolivia where the exotic forage grass Urochloa (Panicum) maxima is abundant on roads and log landings but does not invade felling gaps or unlogged forest. Shade-house trials and seed addition experiments with U. maxima revealed that this C₄ grass thrives in high light but also grows in partial shade (10% full sun, but not 1% full sun), and that felling gaps, but not undisturbed forest, are suitable for grass establishment. To determine if seed dispersal by logging vehicles explains the discrepancy between actual and potential grass recruitment sites, we collected grass seeds that fell from trucks onto log landings located long distances (>500 m) from established grass populations. Trucks dispersed an estimated 1800 alien grass seeds per log landing during the early dry season; automobiles also transported seeds of grass (135 seeds/vehicle). The seeds collected (and relative abundances) were the exotics U. (Panicum) maxima (97%) and Urochloa (Brachiaria) brizantha (2%), and the pan-tropical weeds Sorghum halapense (1%) and Rottboellia cochinchinensis (0.2%). Grasses invade this forest where disturbance coincides with seed dispersal by motor vehicles, while dispersal limitation apparently prevents invasion of many sites otherwise suitable for grass recruitment (i.e., felling and natural gaps).     

Small mammals as potential seed dispersers in New Zealand

Weed invasion success is strongly influenced by availability of seed dispersal vectors, which may include animals. We examined the potential of several small introduced mammals (mice, kiore, ship rats and possums) to disperse germinable seeds in New Zealand. Captive animals were fed fleshy fruit of weeds (Berberis glaucocarpa, Cotoneaster spp., Crataegus monogyna, Ilex aquifolium, Leycesteria formosa, Ligustrum sinense, Lonicera japonica, Passiflora mollissima, Pyracantha angustifolia, Sorbus hupehensis) and native species (Coprosma spp., Prumnopitys ferruginea and Solanum aviculare). We recorded the percentage of fruit consumed, seed ingested and gut passage time. Faeces were collected and the seeds extracted and tested for germination potential in an unheated glasshouse (two weed species) or under controlled conditions (11 species). The smallest rodents (mice and kiore) generally destroyed all seeds eaten. Large numbers of viable seeds of the small‐seeded (<1 mg) species, L. formosa and S. aviculare, passed through ship rats. Possums consumed the seeds of all adventive and native fruits except P. ferruginea. The proportion of seeds recovered intact from possum faeces varied with plant species and ranged from 6 to 83%. The time required for 50% of all seeds to be passed by possums ranged from 2.5 to 5.5 days with an average of 3.7 days, and was generally unrelated to simple fruit parameters such as percentage pulp and moisture content. For seeds where germination also occurred in the uneaten controls, the germination of seed from possums ranged from 3 to 78%. Germination was mostly lower in seeds from possums than in the controls, where differences were significant. Possums have major potential to disperse a wide range of fleshy fruit‐producing native and introduced plant species. Ship rats have the potential to disperse those with very small seeds.