A Tropical, Gregariously Semelparous Bamboo Shows No Seed Dormancy

BIOTROPICA 40(1): 28–31 2008

10.1111/j.1744-7429.2007.00336.x

A Tropical, Gregariously Semelparous Bamboo Shows No Seed Dormancy Sean M. Bellairs1 School of Science & Primary Industries, Charles Darwin University, Darwin Northern Territory 0909, Australia Donald C. Franklin and Nicholas J. Hogarth School for Environmental Research, Charles Darwin University, Darwin Northern Territory 0909, Australia

ABSTRACT Seed dormancy may be disadvantageous for gregariously semelparous plants because it disrupts the high levels of reproductive synchrony necessary for success. Alternately, it may provide a bet-hedging option for an otherwise ‘all eggs in the one basket’ reproductive strategy. Rapid germination of seeds upon hydration has been demonstrated for a range of tropical, semelparous bamboos, but the fate of seeds that failed to germinate promptly has been inadequately investigated. We demonstrate prompt germination of seeds upon hydration and the absence of a dormant seedbank in a long-lived, gregariously semelparous bamboo, Bambusa arnhemica, from monsoonal northern Australia. However, we refute the suggestion that seed dormancy is necessarily maladaptive in a gregariously semelparous plant. Rather, caryopsis dormancy may not be possible in a seasonally moist tropical climate. Given an inability to adjust or bet-hedge their germination, bamboo germinants must cope with the vagaries of the monsoonal climate, a factor that may contribute to the general restriction of bamboos to regions with higher rainfall. Key words: Bambusa arnhemica; bet-hedging; caryopsis; germination; monsoonal northern Australia; Northern Territory; reproductive biology; seed biology.

THE

EXTRAORDINARY GREGARIOUS SEMELPARITY OF MANY LONG-

offers few opportunities for bet-hedging against total reproductive failure. Not only do individuals (genets) flower once and then die after decades or even centuries of clonal growth (Janzen 1976), but neighbors who presumably include kin also do so at the same time. Furthermore, as flowering appears to be triggered by a biological clock that is impervious to immediate environmental conditions (Kawamura 1927, Franklin 2004), there is no obvious process by which the bamboo may respond to variation among years to ensure that suitable environmental conditions prevail at the time of sexual reproduction. Many plants hedge their reproductive ‘bets’ through seed dormancy mechanisms that stagger germination or delay it until suitable conditions prevail (Rees 1997). Seed dormancy mechanisms are prevalent in semelparous annuals but surprisingly rare in longlived semelparous species (Young & Augspurger 1991). Numerous studies of tropical bamboos report prompt and high germination rates upon hydration (often around 80%) and/or marked declines in seed viability within a matter of months (White 1947; Mohan Ram & Hari Gopal 1981; Wong 1981; Kondas 1982; Venkatesh 1984; Azmy 1994; Banik 1994; Ravikumar et al. 1998a,b; Koshy & Harikumar 2001; Rawat & Thapliyal 2003). From these, it has been inferred that semelparous bamboos lack seed dormancy mechanisms (Janzen 1976). However, in a perusal of this literature we could find no evidence concerning the fate of seeds that failed to germinate promptly, reflecting a prevailing horticultural rather than an ecological perspective. Seed dormancy has been demonstrated or robustly inferred in several temperate-zone bamboos (Matumura & Nakajima 1981, Taylor & Qin 1988). Short-term dormancy of a proportion of seeds could be a successful bet-hedging strategy in the north Australian monsoonal LIVED BAMBOOS

Received 18 January 2007; revision accepted 13 March 2007. author; e-mail: [email protected]

1 Corresponding

28

climate (Andrew & Mott 1983) where the early storms of the wet season are often followed by periods of intense heat and dryness before monsoonal rains set in (Cook & Heerdegen 2001). Longerterm dormancy—delaying germination of some seeds for a year— might be a viable bet-hedge against the failure of wet season rains or other reproductive catastrophes. However, Janzen (1976) argued that interannual delays would be selected against in gregariously semelparous bamboos because it would generate disadvantageous reproductive asynchrony in the subsequent generation. In this note, we examine the germination strategy of a gregariously semelparous bamboo, Bambusa arnhemica, from northern Australia.

METHODS STUDY AREA AND SPECIES.—Bambusa arnhemica is a pachymorph (clumping) bamboo endemic to the northwest of the Northern Territory of Australia (Franklin 2003). The region has a monsoonal tropical climate with almost all rain falling between October and April. Bambusa arnhemica occurs in areas with a mean annual rainfall of 1200–1750 mm, mostly in riparian forests but occasionally on rocky hillsides (Franklin & Bowman 2004). The species flowers gregariously at the end of a genet life span estimated to be 40–50 yr (Franklin 2004). Flowering commences in the dry season and continues until the early wet season. Production and drop of the ca 20 mg caryopses is concentrated at about the time of the first storms of the wet season in October to December and numerous seedlings appear promptly after substantial storms. Seed production is prolific such that the ground below seeding clumps may be carpeted with seeds (D. Franklin & N. Hogarth, pers. obs.). FIELD WORK.—Seed and soil samples were collected from four locations in the 2005–2006 wet season. The four locations were: Mt


C 2007 The Author(s) C 2007 by The Association for Tropical Biology and Conservation Journal compilation


No Seed Dormancy in a Bamboo

Bundy Creek (MBC: 12◦ 50 S, 131◦ 35 E), Annaburroo Station (AS: 12◦ 53 S, 131◦ 42 E), and two sites at the Mary River Wildlife Ranch (MR1 & MR2: 1.0 km apart; 13◦ 36 S, 132◦ 13 E). MBC, MR1, and MR2 were on riverbanks, whereas the AS site was a rocky hillside. For each site, seeds and associated dry florets (with or without seeds) were collected from at least four B. arnhemica clumps in November 2005 and pooled. Topsoil samples were collected in March 2006, replicate samples being collected from within 1 m of the base of four B. arnhemica clumps that had seeded per site, each sample containing 20 cores of 8 cm diameter by 5 cm depth. Seedlings were counted in March 2006 within 1 m of the base of each of four clumps per site, each clump being sampled with three 20 × 10 cm quadrats. At one site where the seedling density was very low, quadrat size was increased to 1 × 1 m and placed within 2 m of the clump base. LABORATORY WORK.—Seeds and associated dry florets were dried in an air-conditioned room for 2 weeks and then stored at 25◦ C in an airtight container with a silica bag to reduce humidity until needed for viability and germination tests. Upon retrieval for viability and germination tests, florets containing seeds were identified by their swollen firmness compared to empty dry florets. After removal of the husks, four sub samples of 50 seeds per site were weighed on an electronic balance to 0.1 mg. Seed viability was assessed ca 1 mo and again ca 6 mo after collection by staining the embryos with a 1 percent solution of 2,3,5-triphenyl tetrazolium chloride. Four replicates of 50 seeds were selected for each site for the first test and four replicates of 25 seeds for the second test. Water uptake was not affected by the presence of the seed coat or husk (S. M. Bellairs, pers. obs.). Seeds with husks attached were soaked in deionized water for 24 h, immersed in the tetrazolium chloride solution, and placed in lightexcluding boxes in an incubator at 30◦ C. After 18 h in the incubator, seeds were dissected longitudinally through the embryo and the staining patterns were assessed under a dissecting microscope. If the embryo was entirely stained red and the endosperm pink, a seed was considered viable; if the embryo was pink and the endosperm at least light pink, it was considered possibly viable; if either the embryo or endosperm remained white it was considered nonviable. Petri-dish germination tests were carried out on seeds from each site ca 1 mo after collection, and repeated 4–6 mo after collection. Tests were duplicated using seeds alternately contained within dry florets and with the husks removed. Four replicates of 50 seeds per site were used for each treatment. Seeds were placed on two 1-mm thick 84-mm diameter seed testing filter papers in 90-mm-diameter glass Petri dishes. Seeds were hydrated and kept moist with deionized water containing 0.8 g/L Banrot (Banrot 400WP: 250 g/kg Thiophanate-methyl, 150 g/kg Etridiazole) fungicide solution. The seeds were incubated in an illuminated and temperature-controlled cabinet (12 h light, 12 h dark; 30◦ C) and were inspected approximately twice weekly until no potentially viable seeds remained. Germinated seeds (where the radicle had emerged through the seed coat) or dead (soft, nonfirm) seeds were counted and removed. Two 150 g sub samples of each topsoil sample were examined for the presence of B. arnhemica seeds. The soil samples were sieved

29

using a 2.36-mm aperture sieve. The fraction greater than 2.36 mm had any soil peds broken up by hand, was visually checked for seeds and discarded. For the 2.36-mm soil fraction, dust was removed by sieving the material with a 0.60-mm aperture sieve. The remainder of the retained fraction of the 150-g sample was added to a ca 200-ml solution of deionized water containing 140 g of magnesium sulphate, 10 g of sodium hexametaphosphate, and 5 g of sodium bicarbonate. Samples in solution were stirred vigorously for 1 min and then left to settle. After approximately 15 min, the organic matter had floated to the top, while the mineral soil had settled to the bottom. The organic matter was removed, placed in water on a white tray, and the florets were separated. In a simulation test using these methods, B. arnhemica seeds and dry florets containing B. arnhemica seeds mixed in a soil sample were easily detected, with recovery from six samples, of which the seed content was variable and unknown to the observer, being greater than 90 percent. Slight pressure was applied to the recovered florets to detect whether a seed was present and if so whether it was firm and thus potentially viable, or soft and decayed. All recovered seeds were examined under a dissecting microscope to determine whether they were whole or damaged. Germination and emergence of seeds in soil were assessed for a 1500 g sub sample of each replicate topsoil sample. Seedling trays (34 × 29 × 5 cm deep) were filled with a 2:1 sterile mix of coarse sand and coco peat. A second tray was placed on the top of the first, filled to within 2 cm of the top of the tray with the sterile mix, then the soil sample was spread on the top. Trays were placed in a shadehouse (75% shade), watered daily, and the emergence of B. arnhemica seedlings monitored every 2 d for a week, then weekly for 2 weeks, then fortnightly (the diminishing rate reflecting the absence of emergence), from mid-March until May 2006.

RESULTS Mean seed mass was 45 percent higher at site MBC than at site AS. Other sites were intermediate but seeds from MR1 were scarcely larger than those from AS (Table 1). Viability and Petri-dish germination proportions were generally greater than 60 percent after 1 mo of dry storage, exceptions being at AS (12–15%) and the germination of seeds with the husk removed at MR2 (32%) (Table 1). The proportions of seeds that germinated within husks were similar to the assessed viability. There was no consistent direction to the mostly small differences in germination rates between seeds with the husk present and removed. Over 90 percent of seeds that germinated did so within 1 week of being hydrated, and all seeds that germinated did so within 2 weeks with the possible exception of four seeds from MR1 (Fig. 1). Soil collected 4 mo after the initial seed collection yielded 294 dry florets containing seeds but no potentially viable seeds and no germinants (Table 1). After 4 mo of dry storage, low-to-moderate rates of germination (0.5–26%) occurred in seeds from the two sites tested at that time, but no seeds germinated after 5 or 6 mo of dry storage in the two sites tested at the later date (Table 1). After 6 mo of dry storage, no seeds were assessed as possibly viable.

30

Bellairs, Franklin, and Hogarth

TABLE 1.

Characteristics of Bambusa arnhemica seed lots from four sites. Months are the interval after seed collection that the assessment was made or test commenced. Data are means ± SE calculated from four samples or sub samples; the standard error for seed mass being calculated from samples of 50 seeds. nd = no data Sites

Data after 1 mo

MBC

MR1

MR2

AS

Seed mass (mg) (husk removed) Viability assessment (%)

19.3 ± 0.2 68.5 ± 2.5

13.7 ± 0.2 75.0 ± 1.9

17.5 ± 0.2 64.0 ± 3.4

13.3 ± 0.2 nd

Germination test - seed with husk (%) - with husk removed (%)

72.5 ± 4.3 88.5 ± 4.1

82.0 ± 2.4 65.0 ± 1.9

66.5 ± 5.4 31.5 ± 5.0

15.0 ± 1.3 12.0 ± 3.2

Data after 4 mo Seedling density (/m2 )

333 ± 96

1296 ± 307

1083 ± 155

0.7 ± 0.5

0 0

0 0

0 0

0 0

Soil samples - potentially viable seeds found - seeds germinated in soil Data after 6 mo Viability assessment (%)

0.0 ± 0.0

0.0 ± 0.0

0.0 ± 0.0

0.0 ± 0.0

Germination test - time since collection (mo) - seed with husk (%)

4 26.0 ± 3.7

5 0.0 ± 0.0

6 0.0 ± 0.0

4 0.5 ± 0.5

- with husk removed (%)

9.5 ± 1.3

0.0 ± 0.0

0.0 ± 0.0

0.5 ± 0.5

DISCUSSION The low viability and germination rates of seeds collected from the rocky hillside site on Annaburroo Station was a notable exception to otherwise high rates, and corresponded with low seedling densities

FIGURE 1.

Timing of germination of Bambusa arnhemica seeds that germi-

nated in a test commenced ca 1 mo after seed collection. Symbols are days of assessment. Sample sizes (seeds that germinated) are: MBC: 322; MR1: 294; MR2: 196; AS: 54.

in the field. Although seeds from AS were small, they were similar in weight to those from site MR1 in which viability and germination rates were high. We hypothesize that low rainfall during the premonsoonal storm period may have stressed flowering clumps at this site, lowering reproductive success. Franklin (2004) provided phenological evidence that hillside stands of B. arnhemica suffer much greater moisture stress than riparian stands, and consequent reproductive failure may go some way to explaining why nonriparian stands are rare. Bambusa arnhemica does not maintain a dormant soil seedbank; far from hedging an otherwise ‘all eggs in the one basket’ reproductive strategy, its seed biology consolidates this strategy as suggested by Janzen (1976). Within 1 mo of seed maturity, we have demonstrated a complete absence of dormancy. All germinants did so within 18 d of hydration of the seed, no viable seeds remaining ungerminated after 4 mo in the field, and no viable seeds remained after 6 mo of dry storage. In that germination rates are high and rapid following hydration, B. arnhemica is similar to other tropical bamboos. However, our demonstration of a lack of dormancy provides an important additional perspective on the life history of a long-lived gregariously semelparous plant. It may seem surprising that there is no evidence of bet-hedging in the germination strategy of B. arnhemica. One possibility, favored by Janzen (1976; p. 372), is that the internal calendar in bamboo is set at germination and the selection for synchronous reproduction is much stronger than any selection to spread germination risks. We feel that selection against dormancy would only occur if a high proportion of seeds remained dormant, or if seedlings produced in a later year were strongly competitive against those of an earlier year.

No Seed Dormancy in a Bamboo

We propose instead that dormancy would be maladaptive in tropical bamboos because the high temperatures and humidity of a tropical wet season are unsuitable conditions for the survival of grass seeds, as noted by Loch et al. (2004). In its mostly flood-prone habitats, B. arnhemica seeds would be exposed to saturated warm conditions in the soil during the wet season. The documented dormancy in bamboos is an obligate requirement for vernalization in montane or temperate-zone species (Matumura & Nakajima 1981, Taylor & Qin 1988) that prevents seeds from germinating immediately prior to winter, though we acknowledge that this would not disrupt subsequent reproductive synchrony. If the survival of bamboo seeds through a tropical wet season is not or scarcely feasible, dormancy to extend germination across years would be a poor evolutionary option and selection should favor prompt germination and mechanisms that promote seedling survival. Franklin and Bowman (2003) noted that B. arnhemica seedlings are exceptionally resilient to flooding soon after germination and to fire during their first dry season. The absence of short-term dormancy is less easy to explain. It may be that the establishment of rhizomes and storage of nutrients during the seedling’s first wet season demands maximum use of the wet season growing period. Tropical bamboos are largely restricted to higher rainfall regions where the wet season is relatively reliable and extended (Gadgil & Prasad 1984, Hassan et al. 1988, Judziewicz et al. 1999). Semelparous species, and in particular long-lived and gregariously semelparous plants, cannot afford catastrophic failure during flowering, seed production, germination, or subsequent growth. The ability of germinants to survive dry periods of up to several months following germination, and the mechanisms facilitating this, are of considerable interest.

ACKNOWLEDGMENTS We thank K. Gleeson and L. Saunders for permission to work on their properties, C. Wagg for undertaking some of the viability testing, T. Knaggs and M. Wallace for assisting in the field, and H. Brodie-Hall for assisting with laboratory work. This work was funded by an internal grant from Charles Darwin University.

LITERATURE CITED ANDREW, M. H., AND J. J. MOTT. 1983. Annuals with transient seed banks: The population biology of indigenous Sorghum species of tropical north-west Australia. Aust. J. Ecol. 8: 265–276. AZMY, H. M. 1994. Germination and fertilization of Gigantochloa ligulata seedlings. In Proceedings 4th International Bamboo Workshop on Bamboo in Asia and the Pacific held at Chiangmai, Thailand, November 27–30, 1991, pp. 120–122. Food and Agriculture Organization of the United Nations, Rome, Italy. BANIK, R. L. 1994. Studies on seed germination, seedling growth and nursery management of Melocanna baccifera (Roxb.) Kurz. In Proceedings 4th International Bamboo Workshop on Bamboo in Asia and the Pacific held at Chiangmai, Thailand, November 27–30, 1991, pp. 113– 119. Food and Agriculture Organization of the United Nations, Rome, Italy.

31

COOK, G. D., AND R. G. HEERDEGEN. 2001. Spatial variation in the duration of the rainy season in monsoonal Australia. Int. J. Climatol. 21: 1723– 1732. FRANKLIN, D. C. 2003. Morphology and taxonomy of the top end bamboo Bambusa arnhemica F. Muell., a little-known bamboo from northern Australia. Bamboo Science and Culture 17: 44–54. FRANKLIN, D. C. 2004. Synchrony and asynchrony: Observations and hypotheses for the flowering wave in a long-lived semelparous bamboo. J. Biogeogr. 31: 773–786. FRANKLIN, D. C, AND D. M. J. S. BOWMAN. 2003. Bamboo, fire and flood: Regeneration of Bambusa arnhemica (Bambuseae: Poaceae) after massflowering and die-off at contrasting sites in monsoonal northern Australia. Aust. J. Bot. 51: 529–542. FRANKLIN, D. C, AND D. M. J. S. BOWMAN. 2004. A multi-scale biogeographic analysis of Bambusa arnhemica, a bamboo from monsoonal northern Australia. J. Biogeogr. 31: 1335–1353. GADGIL, M., AND S. N. PRASAD. 1984. Ecological determinants of life history evolution of two Indian bamboo species. Biotropica 16: 161–172. HASSAN, M. M., M. K. ALAM, AND A. H. MAZUMDER. 1988. Distribution of bamboo under the edaphic and climatic conditions of Bangladesh. Indian Forester 114: 505–513. JANZEN, D. H. 1976. Why bamboos wait so long to flower. Annu. Rev. Ecol. Syst. 7: 347–391. JUDZIEWICZ, E. J, L. G. CLARK, X. LONDONO, AND M. J. STERN. 1999. American bamboos. Smithsonian Institution Press, Washington, DC. KAWAMURA, S. 1927. On the periodical flowering of the bamboo. Jap. J. Bot. 3: 335–349. KONDAS, S. 1982. Biology of two Indian bamboos, their culm potential and problems of cultivation. Indian Forester 108: 179–188. KOSHY, K. C., AND D. HARIKUMAR. 2001. Reproductive biology of Ochlandra scriptoria, an endemic reed bamboo of the Western Ghats, India. Bamboo Science and Culture 15: 1–7. LOCH, D., S. ADKINS, M. HESTLEHURST, M. PATERSON, AND S. BELLAIRS. 2004. Seed formation, development and germination. In L. E. Moser, B. L. Burson, and L. E. Sollenberger (Eds.). Warm season (C4) grasses, pp. 95–143. American Society of Agronomy Inc., Madison, Wisconsin. MATUMURA, M., AND N. NAKAJIMA. 1981. Fundamental studies on artificial propagation by seeding useful grasses in Japan. VIII. Some observations concerning the seed propagation of the dwarf bamboo, Sasa senanensis Rehd. Research Bulletin, Faculty of Agriculture, Gifu University 45: 289–297. MOHAN RAM, H. Y., AND B. HARI GOPAL. 1981. Some observations on the flowering of bamboos in Mizoram. Curr. Sci. 50: 708–710. RAVIKUMAR, R., G. ANANTHAKRISHNAN, AND A. GANAPATHI. 1998a. Viability of bamboo (Dendrocalamus strictus Nees) seeds under four different storage conditions. Seed Res. 26: 43–46. RAVIKUMAR, R., G. ANANTHAKRISHNAN, A. GANAPATHI, AND T. APPASAMY. 1998b. Biochemical changes induced by accelerated ageing in Bambusa bamboos seeds. Biol. Plant. 40: 459–464. RAWAT, M. M. S., AND R. C. THAPLIYAL. 2003. Storage behaviour of bamboo (Dendrocalamus membranaceus) seeds. Seed Sci. Technol. 31: 397– 403. REES, M. 1997. Seed dormancy. In M. J. Crawley (Ed.). Plant ecology, 2nd Edition, pp. 214–238. Blackwell Science, Oxford, UK. TAYLOR, A. H., AND Qin ZISHENG. 1988. Regeneration from seed of Sinarundinaria fangiana, a bamboo, in the Wolong Giant Panda Reserve, Sichuan, China. Am. J. Bot. 75: 1065–1073. VENKATESH, C. S. 1984. Dichogamy and breeding system in a tropical bamboo Ochlandra travancorica. Biotropica 16: 309–312. WHITE, D. G. 1947. Longevity of bamboo seed under different storage conditions. Trop. Agric. 24: 51–53. WONG, K. M. 1981. Flowering, fruiting and germination of the bamboo Schizostachyum zollingeri in Perlis. Malaysian Forester 44: 453–463. YOUNG, T. P., AND C. K. AUGSPURGER. 1991. Ecology and evolution of long-lived semelparous plants. Trends Ecol. Evol. 6: 285–289.