A review of issues in seagrass seed dormancy and germination:implications for conservation and restoration

Please download to get full document.

View again

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
A review of issues in seagrass seed dormancy and germination:implications for conservation and restoration
Document Share
Document Tags
Document Transcript
  REVIEW Vol. 200: 277-288 2000 A review of issues in seagrass seed dormancy and germination: implications for conservation and restoration Robert J. Orth r , atthew C. Harwell, Eva M. Bailey, Aaron Bartholomew, Jennifer T. Jawad, Alfonso V. Lombana, Kenneth A. Moore, Jennifer M. Rhode, Helen E. Woods MARINE ECOLOGY PROGRESS SERIES Mar Ecol Prog Ser Virginia institute of Marine Science School of Marine Science College of William and Mary Gloucester Point Virginia 23062 USA Published July 14 ABSTRACT: Seagrasses have received considerable attention over the past 2 decades because of the multiple ecological roles they play in estuarine and coastal ecosystems and concerns over worldwide losses of seagrass habitat due to direct and indirect human impacts. Restoration and conservation efforts are underway in some areas of the world, but progress may be limited by the paucity of infor- mation on the role of seeds in bed dynamics. Although flowering occurs in most of the 58 seagrass species, seed germination data exist for only 19 of the 42 species that have some period of dormancy, with only 93 published references to field and/or laboratory studies. This review addresses critical issues in conservation and restoration of seagrasses involving seed dormancy (e.g. environmental vs physiological), existence and type of seed bank (transient or persistent), and factors influencing seed germination (e.g. salinity, temperature, light). Results ofmany earlier published studies relating seed germination to various environmental factors may need re-examination given more recent published data which show a confounding influence of oxygen level on the germination process. We highlight the importance of conducting ecologically meaningful germination studies, including germination experiments conducted in sediments. We also identify questions for future research that may figure prominently in landscape level questions regarding protected marine or estuarine reserves, habitat fragmentation, and restoration. KEY WORDS: Seagrasses Seeds Dormancy Seed banks Germination Restoration Conservation INTRODUCTION Seagrasses, marine angiosperms comprising 58 spe- cies in 11 genera, occur in all coastal waters of the world except the Antarctic (den Hartog 1970). They have been receiving increasing attention worldwide since the 1970s because of recognition of (1) heir mul- 'E-mail: jjorth@vims.edu The order of the 2 primary authors was assigned based on contribution to the manuscript. The remaining authors are arranged alphabetically tiple ecological functions in estuarine and coastal sys- tems, such as nursery and fish habitat (Coles et al. 1987, Heck et al. 1997) and regions ofhigh primary and secondary production (Larkum et al. 1989, Edgar 1990, Fredette et al. 1990), and (2) osses in many areas of the world due to anthropogenic inputs ofnutrients and sediments (Short Burdick 1996, Short Wyllie- Echeverria 1996). In addition, considerable seagrass area has been permanently lost to coastal develop- ment, primarily from dredge and fill operations (Short et al. 1991) or altered due to commercial/recre- ational activities such as propeller and anchor scarring O Inter-Research 2000 esale o ful article not permitted  27 8 Mar Ecol Prog Ser 200: 277-288, 2000 Walker et al. 1989, Sargent et al. 1995, Dawes et al.  1997). These issues have led to recent developments in efforts to both restore and conserve these habitats Fonseca et al. 1998). Although seagrass restoration has focused on mature or adult plants, discussions over the last decade on bio- diversity issues National Research Council 1995) that entail both conservation biology and restoration ecol- ogy suggest that seed ecology should be a critical research topic. Seagrass reproduction and seed output have been largely ignored in questions related to their potential contribution to population structure and genetics, plant demographics, bed maintenance, or in the development of new beds E lrkman Kuo 1990,  Orth et al. 1994, Williams Davis 1996, Piazzi et al.  1998, Marba Walker 1999, Orth 1999). We believe that for conservation and restoration of seagrasses to be effective, more research into their seed germination ecology must be conducted. Most of the 58 reported species of seagrasses are known to flower and produce fruit, but differ dramati- cally in seed anatomy and germination strategy den Hartog 1970). Kuo Kirkman 1996) characterized the seagrass genera into 3 classes based on seed anatomy, nutrient storage, and germination strategy Table 1). Seeds from 2 ofthese classes do not exhibit any type of dormancy, with 1 class, which comprises 4 species in 2 genera, exhibiting viviparous development. The third class, comprising 42 species in 7 genera, produces Table 2. Seed bank, dormancy, and germination citation counts from the published seagrass literature from 1938 to 2000) Genus Number of citations Zostera spp. Halophila spp. Cymodocea spp. Halodule spp. Phyllospadix spp. Syringodium sp. seeds with a hard seed coat and some period of dor- mancy. Despite more than a century of scientific research and over 2000 yr of recorded observations on factors influencing the dormancy and germination of seeds in terrestrial systems Baskin Baskin 1998), there is a dearth of information on seagrass seeds. Searches of the published literature from 1938 to 2000) revealed seed-based information on only 19 of the 42 seagrass species Table 2). These studies found some level of seed dormancy, with 93 references to field observa- tions or laboratory/field experiments on seed germina- tion Appendix l). This review, which concentrates on the group of 42 species that have some period of dor- mancy Kuo Kirkman 1996), addresses critical issues in conservation and restoration of seagrasses involving seed dormancy, existence and type of seed banks, and Table 1. Comparison of the type of fruit and seed covering, dormancy, and seed bank characteristics for the 12 genera of seagrasses modified from Kuo Kirkrnan 1996). Definition of dormancy after Fenner 1995), but modified to include seeds that appear to have a highly variable dormancy period. Viviparous = viviparous reproduction Genus Fruit/seed covering Dormancy Seed bank Length of dormancy Source Class Enhalus Membranous Not distinct Posidonia Membranous Not distinct Thalassia Membranous Not distinct Class mphi bolis Viviparous Not distinct Thalassodendron Viviparous Not distinct Class 3 Cymodocea Hard Distinct Halodule Hard Distinct Halophila Hard Distinct He terozostera Hard Distinct Ph yLlospadix Hard Distinct Syringodium Hard Distinct Zos tera Hard Distinct None None None None None Transient Persistent Persistent Unknown Transient Persistent Transient None None None None None 46 mo Up to 24 mo Unknown 2 wk-12 mo Kuo Kirkman 1996) Kuo Kirkman 1996) Kuo Kirkman 1996) Kuo Kirkman 1996) Kuo Kirkman 1996) Reyes et al. 1995), Pirc et al. 1986) McMillan 1991) McMillan 1991) Kuo Kirkman 1996) Turner 19831, Kuo et al. 1990), Reed et al. 1998) McMillan 1991) Orth Moore 19831, Moore et al. 1993)  01th et al : Issues in seagrass seed dormancy and germination 27 9 factors influencing seed germination. This review does not include freshwater submersed macrophytes, such as Ruppia spp. (for review see Kantrud 1991), which are considered euryhaline and often found co-occur- ring with seagrasses. DORMANCY Species which have some form of dormancy are able to disperse through time as well as space by physical and/or biological vectors (Chambers MacMahon 1994, Fenner 1995). Terrestrial plants exhibit a suite of dormancy strategies, from willow Salix spp., whose seeds germinate in days to weeks, to moth mullein Verbascum blattaria, whose seeds can stay dormant for 1000 yr (Fenner 1995, Baskin Baskin 1998 and refer- ences within). As both spatial and temporal distur- bances occur in seagrass communities, some form of seed dormancy in seagrasses can be important for long-term population persistence. Many definitions of dormancy have been introduced, based, in part, on the factors that control dormancy (Hilhorst Karssen 1992, Baskin Baskin 1998). Classification of dormancy can be assigned according to the residence time in the seed bank (Simpson 1990), he timing of dormancy ini- tiation (Hilhorst Toorop 1997), or the mechanisms preventing germination (Baskin Baskin 1998). Clas- sification of plant communities is often founded on the transient or persistent nature of seed banks. From the perspective of the mechanisms controlling dormancy and germination in a given species, the seed develop- mental stage at which dormancy is imposed is of importance. Dormancy initiated during seed develop- ment requires a physiological mechanism, such as embryo maturation, to initiate germination. Dormancy initiated at the time of seed release requires an envi- ronmentally mediated factor, such as temperature or oxygen stratification, to initiate germination. Primary dormancy occurs when dormancy is im- posed on the developing seed attached to the parent plant, while secondary dormancy is associated with dormancy regulation factors that occur in the seed bank (Hilhorst Toorop 1997). The seagrass literature suggests that dormancy is initiated during seed devel- opment; however, we are unaware of any studies which test if seeds of those species in the third class (Table 1 will germinate immediately after release from the parent plant. While dormancy is not a life history strategy for sev- eral seagrass genera (notably Posidonia, Thalassia, and Amphibolis), dormancy has been demonstrated in others and ranges from a few weeks (Phyllospadix tor- reyi, Reed et al. 1998) to up to 4 yr (Syringodium fiLi- forme, McMillan 1983a). A critical question that has 5 20-Apr 20-May 19-Jun 19-Jul 18-Aug 17-Sep 17-0ct ime o Seed Release Fig. 1. Correlation between timing of seed release and lati- tude in 9 populations of Zostera marina along the east coast of North America. Notations (in order of increasing latitude): North Carolina, Phillips (1972); Virginia, Moore et al. (1993); 3~ew ersey, Phillips (1972); 4New York, Phillips (1972); Rhode Island, Churchill (1983); 6Massachusetts Addy (1947, Phillips (1972); Maine, Phillips (1972); Nova Scotia, Phillips (1972); New Brunswick, Keddy Patriquin (1978) not been addressed in these studies is whether the environment or some physiological characteristic of the seed is responsible for preventing germination (i.e. environmental vs organic [physiological dormancy], Baskin Baskin 1998). As in studies on other plant physiology systems, there can be a high degree of interaction between factors (Thomas 1992, Ungar 1995, Hilhorst Toorop 1997, Khan Ungar 1997). Harrison (1991) used seeds collected from parent plants and from the seed bank (1 mo after seed release) to examine dormancy in Zostera marina. His exper- iments, conducted under various temperature and salinity conditions with scarified (i.e. scored) seed coats, concluded that both physiological and physical dormancy exist in this species. Published data (Addy 1947, Phillips 1972, Keddy Patriquin 1978, Churchill  1983, Moore et al. 1993) from this species latitudinal range along the North American coast from North Carolina to Canada shows a progressively later period of seed release with increasing latitude (Fig. 1). How- ever, evidence from 3 locations (Virginia, New York, Massachusetts; Fig. 1) suggests that germination oc- curs in the fall regardless of latitude, suggesting that dormancy may be under environmental rather than inherent physiological control (Baskin Baskin 1998). Other observational studies have addressed issues of timing of flowering (Jacobs Pierson 1981) and seed production (Phillips et al. 1983, Silberhorn et al. 1983), and seed release (Fig. 1) along latitudinal gradients; however, seed dormancy has not been tested along a latitudinal gradient under rigorous experimental con- ditions. The maintenance of dormancy in terrestrial seeds can be attributed to a variety of factors (Simpson 1990, Hilhorst Toorop 1997, and references within both), including hormones, secondary plant compounds (i.e.  280 Mar Ecol Prog Ser for allelopathy), ight, temperature, salinity, gases (COz and 02 , nd nutrients. For seagrasses, studies of these factors have examined their influence on seagrass seed germination rather than dormancy maintenance. Res- earch on maintaining seagrass seed dormancy has been restricted mainly to studies of the retention of seeds for artificial seed reserves (e.g. McMillan 1991, Brenchley Probert 1998). SEED BANKS Seed banks have been studied extensively in many plant ecosystems (Skoglund 1992 and references within) with the notable exception of seagrass mead- ows. Seed banks are often classified by the type of dormancy present, and they can vary greatly in their characteristics. The seed bank for a given species is population-specific and can be either transient (turn over in less than 1 yr) or persistent (seeds that remain longer than 1 yr) (Simpson 1990). The length of time a seed remains in the seed bank can be critical as long- term population stability is influenced by whether or not the seed reserve exists through the next reproduc- tive cycle (i.e. a true seed bank). Seed banks are characterized best by a number of generalities. First, distribution of seeds in a seed bank is extremely patchy (Fenner 1995) and temporal vari- ability of seeds in the seed bank can exist in a plant community having either transient or persistent spe- cies. This variability makes analyzing seed banks diffi- cult because they require high spatial and temporal sampling resolution (Fenner 1995 and references within). Second, the size of the seed is often directly related to its persistence in the seed bank. This prop- erty appears to be a function of per capita expenditure of energy, ability of the seed to percolate into the sedi- ment, and potential to avoid predation (Fenner 1995). Third, abundance of seeds in a seed bank varies by plant community type. Seed banks in seagrass communities have not been studied extensively. Estimates of seed bank size for seagrasses are comparable to those of most plant community types (Table 3) (Fenner 1995 and refer- ences within); however, Inglis (2000) found large het- erogeneity in Halodule uninervis seed banks at multi- ple spatial scales. Harrison (1993) documented the dynamics of a Zostera marina seed bank in the Nether- lands, showing high spatial and temporal variability in a single population. A significant portion of the seeds in the seed bank were lost to autonomous death and a small percentage of the seeds contributed to a persis- tent seed bank. McMillan (1991) reported seeds of the following seagrass species collected from various sources surviving more than 12 mo under laboratory conditions: Syringodium filiforme 49 mo; H uninervis 41 mo; Halophila engelmannii 24 mo; Halodule wrightii up to 46 mo. Several studies have described the conditions neces- sary to store seeds, an indirect approach for under- standing natural seed banks. Conacher et al. (1994b) reported a maximum of a 50 d storage period for Zostera capricorni seeds under conditions of aeration and low temperatures (5 to 10°C). Churchill (1983) found high survival (8 mo) of Z marina seeds stored in water at room temperature. Reed et al. (1998) found that storage of Phyllospadix torreyi seeds under cold, dark conditions could delay germination by up to 83 d. SEED GERMINATION Successful seed germination following some period of dormancy results from an interaction between phys- iological and genetic factors (internal) and environ- mental factors (external), including sediment, light, and temperature (Baskin Baskin 1998). Both internal and external processes have the potential to create a seed bank for a given population. Critical environmen- tal cues that influence germination can vary over short horizontal and vertical distances as characteristics of the physical environment (e.g. sediment type, pH, organic matter) and biological environment (e.g. abun- dance and type of sediment-dwelling animals, physical arrangement of plant and animals in space and time) interact to create a mosaic of microclimates (Hamrick Lee 1987, Woodin et al. 1998). Termination of dormancy or initiation of germina- tionkeedling growth is influenced by hormones, light (photoperiod), temperature, water (e.g. imbibition, osmotic changes, salinity), nutrients, or mechanical cues (Hilhorst Karssen 1992, Gutterman 1994). Fac- tors critical to this process can be seed-specific, so the intraspecific range for each factor may span several orders of magnitude (Hilhorst Toorop 1997). Thus, exact mechanisms are often considered nebulous as Table 3 Size of seed bank for a variety of plant community types (Fenner 1995) compared to seagrasses Plant community type Arable communities Grasslands/moorlands Temperate forests Tropical forests Salt marshes Subarctidalpine forests Seagrassesa Seed bank size m2) aSee references in Appendix 1 for seed bank studies  Orth et al.: Issues in seagrass seed dormancy and germination 28 the existence ofmultiple factors can preclude identify- ing the exact mechanism of dormancy termination/ germination initiation (Hilhorst Karssen 1992, Bew- ley 1997). Additionally there are often interactions between several factors (e.g. salinity and temperature on halophytes, Ungar 1995, Khan Ungar 1997). Field and laboratory studies of seagrass germination have focused primarily on salinity, temperature, light, scarification, and, more recently, the sediment where seeds germinate (e.g. oxygen [oxygen-reduction pro- files or Eh], Moore et al. 1993, Brenchley Probert 1998), as critical cues influencing germination pro- cesses (Appendix 1). However, no work has been con- ducted on the precise physiological mechanisms lead- ing to seagrass seed germination. More importantly, it is critical that all seagrass seed germination studies identify a standard 'definition' of germination (e.g. Brenchley Probert 1998) for comparison to other work, including not just rupture of the seed coat but' emergence and growth ofthe cotyledon (Churchill 1983, Brenchley Probert 1998). Most published work on seed germination has been conducted under aerobic, non-sediment conditions, studying temperature and salinity and their interac- tions (Appendix 1). The majority ofstudies have demonstrated increased germination at reduced salin- ities Zostera noltii, Hootsmans et al. 1987, Loques et al. 1990; Cymodocea nodosa, Caye Meinesz 1986). Other studies have shown no effect of salinity on ger- mination Syringodium filiforme, McMillan 1981; Halodule wrightii, McMillan 1981; Z, marina [peren- nial form], Hootsmans et al. 1987; Z. marina [annual form], McMillan 1983b; Z. capricorni, Conacher et al. 1994b). Seeds of Z. capricorni (Conacher et al. 1994b) germinated at all tested temperatures (15 to 30°C) at low salinity (1, 5, and 10 0), hile seeds held at higher salinities (20, 30, and 40 0) erminated only at lower temperatures. Several studies suggest that tempera- ture stratification may be critical for germination ofseeds, but it is not clear whether stratification is a nec- essary condition for seagrass seed germination, as in many terrestrial species (Baskin Baskin 1998), because ofthe potential confounding issues created by germinating seeds without sediment (e.g. different oxygen environment, Moore et al. 1993). Further, sea- grass seed germination experiments conducted at low salinities (e.g. O from Conacher et al. 1994b) may not reflect ambient field conditions. Both photoperiod and specific wavelengths of light are important for germination in terrestrial systems (Thomas 1992, Gutterman 1994, Fenner 1995, Hilhorst  Toorop 1997). McMillan (1987, 1988b) found light to be an important inducer of germination for both Halophila engelmanii and H. decipiens, while Birch (1981) reported germination of H. spinulosa seeds under both light and dark conditions. In contrast, light has not been considered important in the germina- tion process for Zostera japonica (Bigley 1981) and Z. marina (Tutin 1938, Hootsmans et al. 1987, Harrison 1991, Moore et al. 1993). Scarification of the seed coat resulted in increased germination rates for Zostera marina (Harrison 1991), Z. noltii (Loques et al. 1990), and Z. capricorni (Con- acher et al. 1994b), suggesting some form ofphysical dormancy. These studies also found an interaction between salinity and scarification, with germination increasing at lower salinities. Removal ofthe seed coat of Cymodocea nodosa resulted in increased germina- tion at a lower salinity (20 0) han a higher one (38 0) (Caye et al. 1992). McMillan (1987, 1988b) and Birch (1981) found no effect of seed coat removal for Halo- phila engelmanii, H. decipiens, and H. spinulosa. Hormones can influence seed germination in terres- trial plants through either stimulating seed germina- tion (e.g. gibberellic acid, Hilhorst Toorop 1997; thiourea, Conacher et al. 1994b) or preventing the ter- mination of dormancy (e.g. abscissic acid, Hilhorst Karssen 1992). While a 1 ppm solution ofgibberellin (GA7 and KN03) had no effect on Zostera noltii seeds (Loques et al. 1990), more concentrated solutions ofgibberellin (50 ppm, GbA3 and KN03) and thiourea (50 ppm) promoted germination of Z. capricorni (Conacher et al. 1994b). One factor common to many ofthe above studies is that they were conducted in either small aquaria or petri dishes without sediment and did not report oxy- gen concentrations (Appendix 1). Oxygen levels or some other property of sediments may be the most important factor affecting seed germination (Kawasaki 1993, Moore et al. 1993, Brenchley Probert 1998). Moore et al. (1993) showed faster germination of Zostera marina seeds held in both sediments with no oxygen (as measured by redox potential) and vials with water where oxygen was removed compared to seeds held in similar vials with oxygen. Kawasaki et al. (1988) and Kawasaki (1993) also showed that Z. marina seed germination was promoted by either covering seeds with sediment or placing them in water at very low dissolved oxygen levels (below 1 ppm), suggesting a reducing environment was also important for this species. Brenchley Probert (1998) found similar results for Z. capricorni with little apparent difference in seeds held at 2 salinities (15 and 30 ). In contrast, Church111 (1992) found that seeds of Z. marina germi- nated and retained under anaerobic conditions did not develop properly. Hootsmans et al. (1987) also found little survival of Z. marina seeds after 30 d under anaerobic conditions. Here, seeds were covered with some form of organic material for 1 mo, which may have resulted in toxic sulfide conditions (not mea-
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks