Ecological Research Volume 22 Issue 3 2007 [Doi 10.1007%2Fs11284-006-0037-5] Mayuko Suzuki; Yuhdai Kaya; Takahide a. Ishida; Kouhei Hattori; -- Flowering Phenology and Survival of Two Annual Plants Impatiens Noli-tanger

of 6
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.
Related Documents
Flowering Phenology and Survival of Two Annual Plants Impatiens Noli-tanger
  ORIGINAL ARTICLE Mayuko Suzuki   Yuhdai Kaya   Takahide A. IshidaKouhei Hattori   Kaori Miki   Takashi NakamuraMasahito T. Kimura Flowering phenology and survival of two annual plants  Impatiens noli-tangere   and  Persicaria thunbergii    co-occurring in streamsideenvironments Received: 9 November 2005/ Accepted: 27 July 2006/Published online: 15 September 2006   The Ecological Society of Japan 2006 Abstract  Flowering phenology of cleistogamous annualplants  Impatiens noli-tangere  and  Persicaria thunbergii  was studied with reference to their mortality in stream-side environments in northern Japan.  I. noli-tangere produced flower-buds and flowers from late June toOctober, whereas  P. thunbergii   did so from mid-Augustto October. Both species were often killed by rapidcurrent and submergence after heavy rainfall, which wasunpredictable in the timing and intensity.  P. thunbergii  was more tolerant to rapid current and submergencethan  I. noli-tangere.  Some individuals of   I. noli-tangere were killed by herbivory from sawfly larvae in summer,but  P. thunbergii   individuals were seldom killed byherbivory. In consequence, the percentage of individualsthat survived until the end of September was higher in  P.thunbergii   than in  I. noli-tangere . These situations areassumed to select a long reproductive life in  I. noli-tangere  and autumn flowering in  P. thunbergii  . Keywords  Annual plants  Æ  Flowering phenology  Æ Herbivory  Æ  Rapid current  Æ  Submergence  Æ Survival Introduction Annual plants are often found in floodplains, marshes,and sand dunes where disturbances often occur, andtheir life-history characteristics such as small size, rapidgrowth and large reproductive effort are suggested asadaptations to disturbances (Grime 1979). In addition,their reproductive timing, another important life-historytrait, is expected to be tuned to cope with disturbances.Plants that reproduce after the occurrence of distur-bances would suffer fitness loss due to severe mortalityduring the disturbances, but those that finish reproduc-tion prior to disturbances may suffer fitness loss due toshortened growth period. Satake et al. (2001), assuminga semelparous annual plant that reproduces once anddies, developed a theoretical model on the evolutionarystable reproductive timing under disturbance. Theypredicted that when disturbance is serious and its timingis predictable, plants evolve a timid strategy—plantsreproduce prior to the major disturbance season. Whenthe timing of disturbance is unpredictable, plants exhibitphenotypic variability in reproductive timing (a mixtureof early and late reproduction, or dates of reproductionspreading over a wide interval). On the contrary, whendisturbance is not serious, a bold strategy evolves— plants reproduce only when they are fully grown.Some of these predictions are confirmed in foregoingstudies. For example, Schemske (1984) reported that aforest population of   Impatiens pallida  subjected tocatastrophe (100% mortality) in July due to herbivoryfrom beetles reproduces in June, prior to catastrophe;i.e., it exhibits a timid strategy. On the other hand, Mackand Pyke (1983, 1984) and Watkinson et al. (1998, 2000) reported that annual grasses  Bromus tectorum  and  Vul- pia ciliata  reproduce at the end of the growing season(i.e., they exhibit a bold strategy), although they aresubjected to temporally and spatially variable distur-bances. It is assumed that the intensity of disturbanceson these two grasses is lower than the threshold abovewhich a mixture of early and late reproduction evolves.However, available data on the intensity of disturbancesare not sufficient to test this assumption.In this study, we assessed the timing and intensity of disturbances in two annual species—  Impatiens noli- M. Suzuki  Æ  Y. Kaya  Æ  T. A. Ishida  Æ  K. Hattori  Æ  K. Miki  Æ T. Nakamura  Æ  M. T. Kimura ( & )Graduate School of Environmental Earth Science,Hokkaido University, Sapporo, Hokkaido 060-0810, JapanE-mail: +81-11-7062236Fax: +81-11-7062225 Present address : T. A. IshidaGraduate School of Agricultural and Life Sciences,Asian Natural Environmental Science Center,The University of Tokyo, Midoricho 1-1-8,Nishitokyo, Tokyo 188-0002, JapanEcol Res (2007) 22: 496–501DOI 10.1007/s11284-006-0037-5  tangere  L. (Balsaminaceae) and  Persicaria thunbergii  (Sieb. et Zucc.) (Polygonaceae)—to test the model of Satake et al. (2001). These two species often co-occuron streamsides or roadsides in central and northernJapan, but  I. noli-tangere  reproduces from early summerto autumn, whereas  P. thunbergii   does in autumn(Hiratsuka 1984; Kawano et al. 1990; Momose and Inoue 1993; Masuda and Yahara 1994; Hatcher 2003). Materials and methods Study area and flowering phenologyThe study was carried out in a cool–temperate deciduousforest in Nopporo Forest Park (43.1  N, 141.5  E), lo-cated about 15 km east of the center of Sapporo City. Inthis forest,  I. noli-tangere  and  P. thunbergii   were ob-served along streams, paths and roads. Other majorundergrowth plants in this forest are  Sasa senanensis (Franch. et Savat.),  Cryptotaenia japonica  Hassk., Chrysosplenium graynum  Maxim.,  Petasites japonicus (Siebold et Zucc.),  Smilacina japonica  Gray,  Anemone flaccida  Fr. Schm. and  Lilium cordatum  Koidz.The flowering phenology of   I. noli-tangere  was stud-ied at seven sites along streams in 2004 (Fig. 1). At eachsite, a 50 cm  ·  50 cm plot was set in mid May. A total of 65 individuals grew in these seven plots, and they weremonitored for the formation of flower buds once a week.However, three individuals died before flower-bud for-mation, and then data were obtained from 62 individu-als.The flowering phenology of   P. thunbergii   was studiedat ten sites in 2004 (Fig. 1). At each site, more than 100individuals grew, and at a number of sites the densitywas very high (about ten individuals per 100 cm 2 ). Insuch dense patches, checking of the presence or absenceof flower buds sometimes damages plants. Then, once aweek, 50 individuals were randomly chosen at each siteand checked for the formation of flower buds.Survival of natural populationsSurvival of these two annual plants was studied in areasA and B along streams (Fig. 1). Area A was locatedunder dense canopy with small gaps. The relative lightintensity was measured at 11 points in this area on 19August 2003 using quantum sensor (LI-190SA, LI-COR, Lincoln, USA); it ranged from 0.8 to 2.6% of theregime in open area. In this area,  I. noli-tangere  growssingly or in small patches, and  P. thunbergii   forms pat-ches variable in size and density. Area B was located in arelatively large gap. The relative light intensity in thisarea was 3.6–3.9% of the regime in open area. The twostudy species formed large mixed colonies in this area.The study was made with 50 cm  ·  50 cm plots, whichwere set in mid May when seedlings of the study speciesappeared. In area A,  I. noli-tangere  plots were set in all(five) patches with >5 individuals in 1999 and 2000, andin 63 patches that varied in the density in 2001 (Table 1).In 2002,  I. noli-tangere  was not studied since only a fewindividuals were observed in this area, probably due toserious damage by heavy rain in the last year (see Re-sults). For  P. thunbergii  , a number of plots were set inpatches with various densities in 1999, 2001 and 2002(Table 1). In 2000,  P. thunbergii   was not studied becauseof our circumstances. In area B, several plots were set inmixed patches in 2001 and 2002.The study plots were divided into two categoriesaccording to the height from the water level of streams:those at high locations (>25 cm above the water level in Up streamN0 50 mDown streamArea BArea A Fig. 1  Study sites.  Arrows  indicate sites where flowering phenologywas studied ( solid arrows ;  I. noli-tangere ,  broken arrows ;  P.thunbergii  ). Survival was studied in areas A and B Table 1  Numbers of high (>25 cm above the water level in lateMay) and low (<25 cm) study plots in areas A and B, and therange of numbers of individuals in the plots (in parentheses)Area A Area BHigh Low High Low Impatiens noli-tangere 1999 5 (6–21) – – – 2000 5 (6–24) – – – 2001 47 (1–20) 16 (1–6) 3 (33–76) 4 (32–91)2002 – – 5 (4–29) 1 (20) Persicaria thunbergii  1999 3 (15–56) – – – 2001 44 (1–146) 33 (1–164) 3 (12–60) 4 (68–230)2002 7 (18–33) 8 (9–41) 5 (4–44) 2 (13–52)497  late May) and those at low locations (<25 cm). On plotsat low locations, only a few plants were found apartfrom sparsely distributed individuals and populations of the present study species, probably because of frequentsubmergence. Plots at high locations were usually cov-ered by vegetation, including some perennials. Thenumber of plots and the range of individual number inplots are shown in Table 1. Survival and leaf damage of  I. noli-tangere  and  P. thunbergii   in the study plots weremonitored every 1 or 2 weeks from early May until theend of October, when almost all individuals died.The difference in survival was analysed by  v 2 andtwo-dimensional log likelihood ratio tests with JMP 4.0(SAS Institute, Cary, IN, USA).Mortality factorsFour different types of mortality factors were classified.1. Rapid current and submergence: In the presentstreams, heavy rainfall often caused rapid current andincreased levels of water for a few to several days, anda number of individuals were carried away or buriedunder silt. In addition, some fallen individuals expe-rienced submergence and died. It was determined thatthey had been killed by rapid current and submer-gence.2. Herbivory:  Impatiens noli-tangere  leaves were oftenheavily attacked by larvae of two sawfly species. InJapan, two species of sawfly have been reported toattack  Impatiens  leaves:  Aglaostigma neburosa  (And-re ´) and  Siobla ferox  (Smith) (Okutani 1967). Larvaeobserved in the study area were considered to be thesetwo species according to their morphology, but thiswas not confirmed since they were not successfullyraised to adults. Some lepidopteran larvae were ob-served on  I. noli-tangere  and  P. thunbergii  , but theirattacks were much less serious. These lepidopteranlarvae were not identified as to species. In this study,individuals of   I. noli-tangere  and  P. thunbergii   whichdisappeared after complete defoliation by these in-sects were assumed to have been killed by herbivory.Individuals that died due to rapid current and sub-mergence before they were completely defoliated wereassumed to have been killed by rapid current andsubmergence. Rapid current, submergence and her-bivory are also expected to lower the fitness of plantsby retarding their growth, but these effects were notdetermined in this study.3. Aging: In mid- to late October, most plants withered.Some of them were infected by pathogens, but it wasnot determined whether the infection caused death ornot. In this study, they were categorized to have dieddue to aging.4. Unidentified factors: Shading and competition wouldalso be important mortality factors, but it was diffi-cult to determine whether these factors acted as adirect cause of death or not. In this study, mortalityfactors other than rapid current, submergence, her-bivory and aging were categorized as ‘‘unidentifiedfactors’’.Submergence experimentsTolerance to submergence was compared between  I.noli-tangere  and  P. thunbergii  . Seedlings of   I. noli-tang-ere  were collected in the campus of Hokkaido Univer-sity, and those of   P. thunbergii   were collected fromNopporo Forest Park in early May. They were plantedin flowerpots (ten individuals per flowerpot) and raisedin a grove located in the campus of Hokkaido Univer-sity. The flowerpots with plants were submerged in apond in the campus of Hokkaido University on 28 May2001 for 1, 3, 8, 16 or 32 days, and plants were examinedfor survival after they were allowed to recover for1 month. A flowerpot with ten individuals was used ineach treatment for each species. Plant height at the timeof submergence was 15–20 cm in  P. thunbergii   and 20– 30 cm in  I. noli-tangere . Water temperature during theexperiment was 17–20  C. Results Flowering phenologyIn 2004,  I. noli-tangere  produced flower buds and/orflowers from late June to October, whereas  P. thunbergii  from mid-August to October (Fig. 2).Survival and mortality factorsIn the study areas, most individuals of   I. noli-tangere and  P. thunbergii   germinated in May, and the number of survivors gradually decreased until October (Fig. 3).The percentage of individuals which survived until theend of September was always higher in  P. thunbergii  than in  I. noli-tangere  (Table 2), and the difference wassignificant ( v 2 test with sequential Bonferroni correction, 020406080100July August I. noli-tangere P. thunbergii  June    I  n   d   i  v   i   d  u  a   l  s  w   i   t   h   f   l  o  w  e  r  s  a  n   d   /  o  r   f   l  o  w  e  r   b  u   d  s   (   %   ) Fig. 2  Seasonal changes in the percentage of individuals withflowers and/or flower buds in  I. noli-tangere  and  P. thunbergii   in2004498  P <0.01) except in the case of low plots in 2002. Rapidcurrent and submergence after heavy rainfall causedvery high mortality to both species in September 2001,and moderate mortality from July to October in 2002. In2000, no individual of   I. noli-tangere  was killed by rapidcurrent or submergence, although heavy rainfall oc-curred in July (data not shown). This is because all of thestudy individuals occurred at high locations (>25 cmfrom the water level of the streams) (Table 1). In 1999,some  I. noli-tangere  individuals were killed by rapidcurrent and submergence, although they occurred athigh locations (Table 1).After heavy rainfall in September 2001, mortality wassignificantly higher at lower plots (<25 cm from thewater level of the streams) than at higher plots (>25 cm)in both species (two-dimensional log likelihood ratiotest,  P <0.01 for areas A and B), and also higher in  I.noli-tangere  than in  P. thunbergii   on areas A ( P <0.001)and B ( P =0.02) (Fig. 4).Some  I. noli-tangere  individuals were killed by her-bivory from sawfly larvae in summer (Figs. 3, 5). The & submergence 0501001500501001500501000200400600050100150M J J A S O herbivoresundeterminedaging I. noli-tangere P. thunbergii  1999200020012002 1999 100020003000 2001 M J J A S O 2002    N  u  m   b  e  r  o   f   i  n   d   i  v   i   d  u  a   l  s 02004006008000 rapid current Fig. 3  Survival of   I. noli-tangere  and  P. thunbergii  .Mortality factors were alsoshown.  White bars  indicatesurvivors. Data on all studyplots were pooled Table 2  The percentage of individuals survived until the end of September in  I. noli-tangere  and  P. thunbergii   at high and low plotsHigh plots Low plots Impatiens noli-tangere 1999 37.1 (62) – 2000 9.2 (65) – 2001 4.9 (408) 0 (171)2002 17.2 (64) 30.8 (39) Persicaria thunbergii  1999 64.5 (110) – 2001 33.5 (1,264) 9.3 (1,121)2002 48.3 (236) 37.8 (246)Data on plots along areas A and B were pooled. Numbers inparentheses refer to the total number of individuals that germi-nated in the plots 020406080100 High Low High Low    M  o  r   t  a   l   i   t  y    (   %    ) I. noli-tangereP. thunbergii  Fig. 4  Mortality due to rapid current and submergence at high andlow plots in area A ( white bars ) and area B ( shaded bars ) inSeptember 2001499


Jul 23, 2017


Jul 23, 2017
Similar documents
View more...
Related Search
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