survival costs of reproduction in a short-lived perennial plant: live

904

American Journal of Botany 96(5): 904–911. 2009.

Life-history theory assumes that the investment in one life function is traded off against investment in other functions ( Stearns, 1992 ). In plants, one of the trade-offs that has received most attention is the cost of reproduction, i.e., a decrease in cur-rent or future growth, future fecundity and/or future survival as a result of allocation of resources to current reproduction ( Williams, 1966 ; Bell, 1980 ; Tuomi et al., 1983 ; Stearns, 1992 ). In spite of a general agreement over the existence of a trade-off between current reproduction and future plant performance ( Obeso, 2002 ), studies conducted so far have produced incon-sistent results; some described negative effects of reproduction on future performance, but others failed to detect these negative effects ( Obeso, 2002 ; Reekie and Avila-Sakar, 2005 ).

Inconsistent evidence for a cost of reproduction in plants has often been attributed to the different experimental settings used to evaluate trade-offs ( Reznick, 1985 ; Ashman, 1994 ; Karlsson and M é ndez, 2005 ). Phenotypic costs of reproduction can be examined either through observational methods dealing with naturally occurring variation of reproductive allocation or by experimentally manipulating reproductive allocation (e.g., through fl ower removal or pollen addition) ( Reznick, 1985 ). In general, experimental manipulations are preferable because they induce greater variation in the reproductive allocation and avoid confounding effects of environmental heterogeneity or differences in resource status ( Reekie and Avila-Sakar, 2005 ).

It has also been suggested that the cost of reproduction should be more evident in stressful habitats such as those with low re-source availability ( Reznick, 1985 ). Failure to fi nd a cost of reproduction could be due to experimental settings in relatively benign environmental conditions. Increased probability of de-tecting a cost of reproduction with increasing environmental stress has been confi rmed at least in some cases ( Primack and Antonovics, 1982 ; Biere, 1995 ; Sandvik, 2001 ), but not in oth-ers ( Reekie, 1991 ; Cheplick, 1995 ; Saikkonen et al., 1998).

A little-explored possibility for the relatively frequent fail-ures to detect costs of reproduction may be related to the different types of costs (growth, fecundity, or survival) being expressed to different degrees by different species, depending on their life histories. For example, reproductive allocation is higher in annual than in perennial plants (reviewed in Karlsson and M é ndez, 2005 ) and higher in herbs than in woody plants ( Cheplick, 2005 ). A demographic classifi cation of life histories also indicates differences between short-lived herbs, forest herbs, shrubs, and trees in the relative importance of fecundity and survival for lifetime fi tness ( Franco and Silvertown, 1997 ). Thus, long-lived species might show costs of reproduction in terms of growth or fecundity but not in survival, which would be prioritized in this life strategy. On the other hand, short-lived species would be more prone to survival costs due to high re-productive investment and priority of fecundity and growth in this life strategy. Indeed, while the effects of current reproduc-tion on growth and future reproduction have been frequently studied, empirical data on trade-offs between current reproduc-tion and future survival is relatively scarce ( Obeso, 2002 ). Due to the strong implications of a survival cost of reproduction for life history evolution in perennial plants, as compared with growth or fecundity costs, it is surprising that survival costs are among the less well documented, perhaps because relatively few short-lived species have been studied (M. M é ndez, unpub-lished data).

We tested the aforementioned hypotheses on a short-lived perennial shrub, Helianthemum squamatum , a semiarid specialist

1 Manuscript received 2 July 2008; revision accepted 16 January 2009. The authors warmly thank R. Molleja for help with fi eld work, R. Torices

for collaboration with the nutrient digestion procedure, and Y. Vali ñ ani for technical assistance with the Skalar. They also thank two anonymous referees for their valuable comments on an earlier version of this manuscript. Support of the research projects EXTREM (CGL2006-09431) and REMEDINAL (S-0505/AMB/0335) is gratefully acknowledged. C.F.A. held a F.P.U. fellowship (AP2003-3662, MEC, Spain).

2 Author for correspondence (e-mail: [email protected]), phone +34 914888288, fax: +34 916647490

doi:10.3732/ajb.0800223

SURVIVAL COSTS OF REPRODUCTION IN A SHORT-LIVED PERENNIAL PLANT: LIVE HARD, DIE YOUNG 1

Cristina F. Arag ó n, 2 Marcos M é ndez, and Adri á n Escudero

Á rea de Biodiversidad y Conservaci ó n, Departamento de Biolog í a y Geolog í a, ESCET, Universidad Rey Juan Carlos, C/ Tulip á n s/n, M ó stoles, E- 28933, Madrid, Spain

According to life-history theory, reproductive investments involve costs in terms of growth, future fecundity, and/or survival. However, studies to date have often failed to detect costs of reproduction, with survival costs among the less documented. We in-vestigated the cost of reproduction in Helianthemum squamatum (Cistaceae), a short-lived perennial of semiarid Mediterranean environments. After experimental fl ower removal, we evaluated next season ’ s growth, reproduction, and survival of the plants. We also monitored an indicator of plant physiological status ( F v / F m ) and leaf nutrient concentration at key phenological stages during reproduction. Survival rate in deblossomed plants was signifi cantly higher than in control plants. As far as we know, this is the fi rst experimental evidence of a survival cost of reproduction in a perennial plant. In contrast, no cost to growth or reproduction was found during the next season, and no signifi cant differences in F v / F m or leaf nutrients were found between control and deblos-somed plants. Helianthemum squamatum ’ s success in semiarid Mediterranean ecosystems seems to rely on a persistent seed bank, combined with a sustained high reproductive output at the expense of survival. We conclude that this strategy might be more com-mon than previously thought among short-lived shrubby plants growing in stressful Mediterranean areas.

Key words: Cistaceae; cost of reproduction; Helianthemum squamatum ; life-history theory; Mediterranean; reproductive phenology; semiarid; stress.

905May 2009] Arag ó n et al. — Survival costs of reproduction in short-lived perennial

season: before fl owering (May 16), at the onset (June 13) and at the peak (June 24) of fl owering, and at the peak of fruit ripening (July 9). The sampling dates were chosen according to fi eld observations and phenological data from previ-ous years ( Arag ó n et al., 2007 ; Arag ó n et al., 2008 ).

To estimate plant physiological status, we determined the maximum quantum effi ciency of photosynthetic energy conversion of photosystem II, F v / F m , which is equal to ( F m – F o ) / F m , where F m and F o are the maximum and minimum fl uorescence of chlorophyll, respectively, measured in dark-adapted leaves ( Maxwell and Johnson, 2000 ). Measurements were performed with a portable pulse modulated chlorophyll fl uorimeter (FMS2, Hansatech, Norfolk, UK) from 1200 to 1400 hours GMT in three mature, healthy leaves per plant, dark-adapted for 30 min before each measurement with appropriate leaf-clips.

To evaluate leaf nutrient content, we collected three mature, healthy leaves per plant around noon, transported them on ice to the laboratory for immediate storage at – 80 ° C until processing. Those leaves were pooled for the analysis, providing a single measurement per plant. Total N and P were determined, after micro-Kjeldhal digestion, in a Skalar segmented-fl ow nutrient autoanalyzer (Skalar Analytics, Breda, Netherlands). To account for the potential confound-ing effects caused by water availability per se, since water is considered one of the most limiting factors for plant performance in semiarid Mediterranean sce-narios, we monitored soil moisture content during the study. Volumetric soil moisture content was measured with a soil moisture probe (Thetaprobe ML2 × , Delta T, Cambridge, UK) at 6 cm in depth for all the tagged plants, from 1000 to 1200 hours GMT. We took measurements on the northern and southern side of each plant within a 20 cm radius, then calculated the mean.

Plant growth was calculated as the difference in plant size between years. Before the 2005 and 2006 fl owering seasons (around mid-May), we measured plant height ( h ), maximum crown diameter ( M ), and its maximum perpendicu-lar length ( m ) and estimated plant size as an approximation to the volume of an ellipsoid (2 / 3 ⋅ π hMm ). This estimate of plant size, proven useful in other stud-ies with the same species, is closely correlated to the reproductive variables considered ( Arag ó n et al., 2007 , 2008 ) and to the index F v / F m ( Arag ó n et al., 2008 ).

In 2006, plants were allowed to fl ower and set fruit freely. At the end of the reproductive period of 2006 (early August), we counted the total number of infl orescences per plant and collected 10 of them (when available), to estimate the mean number of fl owers and fruits produced per infl orescence and per plant. With these variables, we calculated the proportion of fl owers setting fruits (fruit set). We dissected 10 fruits per plant to estimate the mean number of viable and aborted seeds produced per fruit and per plant. Intrafruit abortion rate was cal-culated as the proportion of aborted seeds to total number of seeds per fruit. We considered a seed viable when it appeared turgid and aborted when it had lost water and was dry and fl at.

Data analysis — To assess the effect of the experimental fl ower removal on future survival, reproductive output, and vegetative growth, we fi tted general-ized linear models (GLMs; McCullagh and Nelder, 1989) as implemented by the PROC GENMOD of SAS (SAS Institute, 1990). GLMs are an extension of the linear modeling process that allow the modeling of data having any type of probability distribution aside from the normal distribution, such as the binomial Poisson. We modeled survival and growth, assuming a binomial error distribu-tion (link function: logit) and a normal error distribution (link function: iden-tity), respectively. The reproductive variables modeled were total number of fl owers and total number of seeds per plant (both with a Poisson error distribu-tion and log link function), and fruit set and intrafruit seed abortion (both with a binomial error distribution and logit link function). The set of predictors in-cluded treatment (deblossomed vs. control plants), slope aspect, the interaction between them, and plant size as a statistical control of the individual plant performance.

Because measurements during the 2005 reproductive season were repeated on the same plants, we expected them to be correlated. Thus, to evaluate the effect of the experimental fl ower removal on F v / F m and leaf nutrient concentra-tion, we used generalized estimating equations (GEEs; Liang and Zeger, 1986), a generalization of GLMs that takes into account the autocorrelation of the data. We used the REPEATED statement of the GENMOD procedure of SAS. The three response variables, the index F v / F m , and the concentrations of total N and total P (mmol/g dry mass), were assumed to follow a Poisson error distribution (link function: log). The explanatory variables were treatment, slope aspect, and date, and all the possible two-way interactions among them, together with plant size as a fi xed covariate. When the interaction between any two variables was signifi cant, we performed posthoc tests using the LSMEANS statement of PROC GENMOD.

plant endemic to the Iberian Peninsula. We experimentally re-moved fl oral buds in a single season and recorded the effects on plant performance in the following year. This perennial plant, restricted to extremely poor gypsum soils, has a very short life expectancy ( Caballero, 2006 ), which makes it ideal to study short-term effects of current reproduction. Furthermore, repro-duction in H. squamatum takes place during the Mediterranean summer drought period ( Arag ó n et al., 2007 ), characterized by an extreme water shortage and high temperatures and irradi-ances. The implications of stress for fi nal reproductive output in this species have been shown to differ depending on the repro-ductive stage of the plant at the time of the physiological stress ( Arag ó n et al., 2008 ). For instance, low values of a chlorophyll fl uorescence-based indicator of plant stress ( F v / F m ) at the beginning of the reproductive period were related to a lower production of fl owers and seeds, while low values of F v / F m at the peak of fruit ripening were related to lower fruit sets ( Arag ó n et al., 2008 ). It has been hypothesized that reproduction could make plants more prone to stress-based mortality ( Ehrl é n and Van Groenendael, 2001 ) and that nutrient pools in vegetative compartments can be depleted during reproductive events if the needs for reproductive structures exceed current uptake ( Chapin et al., 1990 ). Thus, we phenologically monitored indicators of plant physiological status ( F v / F m ) and leaf nutrient concentra-tion to evaluate the potential physiological and resource costs of reproduction.

MATERIALS AND METHODS

Species and study area — Helianthemum squamatum (L.) Dum. Cours (Cistaceae) is a perennial, nonclonal subshrub (10 – 40 cm) with a life span ranging from 4 to 6 years ( Caballero, 2006 ). This species is relatively abun-dant in the gypsum landscapes of the Iberian Peninsula, being a diagnostic species for Iberian gypsum vegetation. The growing period of this species extends throughout winter and early spring and does not overlap with the re-productive period, which takes place during late spring and summer. In this sense, H. squamatum is a “ phenophase sequencer ” ( Castro-D í ez and Montserrat-Mart í , 1998 ). The flowering period of this late-flowering plant occurs from the end of May to the end of July ( Arag ó n et al., 2007 ), coinciding mostly with summer drought. Floral organs develop during the same growing season they are deployed (C. F. Arag ó n, personal observation). Flowers, arranged in dense infl orescences (scorpioid cymes) at the tip of new branches, are her-maphroditic and yellow, and fruits are multiseeded capsules 3 mm in diameter ( L ó pez-Gonz á lez, 1993 ). This species, predominantly allogamous ( Arag ó n and Escudero, 2008 ), has an extremely high fecundity even under very harsh conditions ( Arag ó n et al., 2007 ).

The study was performed in two nearby sites with contrasting slope aspects (south- vs. north-facing), representing extremes of the local irradiance stress gradient, in a semiarid Mediterranean gypsum landscape (Central Spain, south of Madrid). The south-facing slope was covered by a sparse, patchy perennial shrubland where patches of genuine gypsophytes were interspersed in a bare matrix covered by a very rich biological gypsum crust ( Mart í nez et al., 2006 ). The north-facing slope was covered by a richer and denser shrub community, where together with some strict gypsophytes, more generalist plants or gyps-ovags are dominant. Climatic conditions at the study site during 2005 and 2006 are shown in Fig. 1A . Air temperature in each slope was recorded hourly during the study period with data loggers (ProH8032, Onset HOBO, Pocasset, Mas-sachusetts, USA) placed half a meter above the ground ( Fig. 1B ).

Experimental procedures — In May 2005, before the onset of fl owering, we selected 30 plants in each slope aspect and randomly divided them into two groups. In the experimental group, fl ower buds were clipped before anthesis; the control group was left undisturbed. We monitored the plants every 2 – 3 d during fl owering and clipped any new fl owers appearing in the experimental group.

We evaluated plant physiological status, leaf nutrient content, and soil mois-ture content in four relevant phenological moments during 2005 reproductive

906 American Journal of Botany [Vol. 96

ferences between deblossomed and control plants were detected in terms of growth or reproductive output ( Table 1 , Fig. 2 ). Slope aspect was not a signifi cant predictor in any of the models ( Table 1 ).

No signifi cant differences in F v / F m or leaf nutrient concen-trations between deblossomed and control plants were found ( P ≥ 0.17, for all GEEs). F v / F m measured at midday remained

RESULTS

Only one and three control plants in the south- and the north-facing slope, respectively, did not fl ower in 2005. All the sur-viving plants in 2006 blossomed. Survival rate in 2006 was signifi cantly higher for plants that had been deblossomed in 2005 (0.50, deblossomed; 0.24, control), but no signifi cant dif-

Fig. 1. Climatic conditions at the study site (Chinch ó n, South of Madrid, 40.08 º N 3.26 º W). (A) Bars and lines represent monthly mean precipitation and mean temperature, respectively, along the period 1972 – 2004 and in the studied years, 2005 and 2006; (B) Lines show hourly air temperatures at plant height during the study period (May – August 2005 and 2006) on the south- and north-facing slopes.


Fig. 2. Survival rate, growth, and reproductive output (number of fl owers and seeds per plant, fruit-set and intrafruit seed abortion) for Helianthemum squamatum deblossomed and control plants. These data were recorded in 2006 reproductive season, on the south- and north-facing slopes. Bars represent mean ± SE. Numbers above the bars indicate sample sizes. Asterisks indicate signifi cant differences between treatments ( P < 0.05).


stressful scenario for this species. This endurance of the south-facing slope conditions might be related to the existence of a well-developed biological soil crust in this exposure ( Mart í nez et al., 2006 ), where only a few number of genuine gypsophyte species are able to germinate, surpass the gypsum crust, and grow ( Rom ã o and Escudero, 2005 ). Higher competition for resources on the north-facing slope, where aside from some strict gypsophytes other calcicole plants become dominant, may explain to some extent a slightly worse plant physiologi-cal status in this exposure, similar to what has been reported in shaded vs. sunny Mediterranean habitats ( Valladares and Pearcy, 2002 ).

Likewise, the depletion of mineral resources was not more important in control plants, contrary to what has been reported for other shrubs (Pakonen et al., 1988). On the other hand, the decline in N and P leaf concentrations as the season advanced is an expected result because tissue concentrations of growth-lim-iting nutrients often decline during water stress ( Chapin, 1991 ). The existence of compensatory mechanisms by which plants would enhance resource acquisition during reproduction could explain the lack of evident costs of reproduction related to al-location of resources ( Obeso, 2002 ; Karlsson and M é ndez, 2005 ). With regard to nutrients, this compensatory response seems to be unlikely because reproductive structures cannot be autonomous for mineral nutrients, in contrast to a sometimes feasible self-supported carbon economy, and nutrient uptake is assumed to be negligible during the dry season (Zotz and Richter, 2006). However, the role of storage in reproduction has not been studied in this species. Storage in roots or stems, rather than in leaves, could differ between treatments.

It is somewhat surprising that we did not detect somatic costs of reproduction associated to the observed survival cost, be-cause it has been proposed that fecundity and survival costs should be mediated by costs in terms of growth ( Kozlowski, 1991 ; Reekie and Avila-Sakar, 2005 ). Costs in terms of reproduction were not found either. The most obvious explanation is the lack of statistical power due to the reduced sample size in the second year as a consequence of the high plant mortality. We need to take into account that the different types of costs are expressed in a sequential way and that plants that died could not be in-cluded in the assessment of the costs in terms of future growth and reproduction. Although speculative in the view of our re-sults, we point out some characteristics of H. squamatum that suggest the possibility that not only statistical power problems underlie the absence of somatic and reproductive costs. First, the lack of signifi cant differences in leaf nutrient content be-tween control and deblossomed plants assessed in our initial set

close to the optimum level (0.83 according to Maxwell and Johnson, 2000 ) in both slopes at all the sampling dates, reach-ing minimum values at the onset of fl owering ( Fig. 3 ). As the season advanced, leaf nutrient concentrations decreased simi-larly in both slopes although more gradually in the north-facing slope ( Fig. 3 ).

Deblossomed and control plants had similar soil water avail-ability at all dates. Soil moisture content remained extremely low, close to 0% throughout the study period.

DISCUSSION

Observational studies have detected costs of reproduction in terms of survival in woody plants ( Bond and Maze, 1999 ), palms ( Pi ñ ero et al., 1982 ), giant rosettes ( Smith and Young, 1982 ), herbaceous perennials ( Sohn and Policansky, 1977 ; Karlsson et al., 1990 ), biennials ( Paige and Whitham, 1987 ; Gillman and Crawley, 1990 ), and annuals ( Law, 1979 ). How-ever, experimental studies have failed to detect survival costs except for a biennial plant ( Paige and Whitham, 1987 ). There-fore, the survival cost of reproduction reported here is, to the best of our knowledge, the fi rst evidence of this type of cost detected in a perennial plant using an experimental approach. It is remarkable that costs of reproduction in terms of survival have been so rarely detected in plants because of their assumed importance for life histories ( Stearns, 1989 , 1992 ). The possi-bility that survival costs are mainly present in short-lived plants, while long-lived perennials mainly show growth and fecundity costs, should be seriously considered in future studies.

Although reproduction involved costs that affected plants ’ chances to survive until the following season, we could not demonstrate any weakening of the control plants, either to plant physiological status or resource level, compared to the deblossomed plants at key reproductive stages. Moreover, midday F v / F m values remained considerably high during all the reproductive season for both groups of plants, near the op-timum level of 0.83 ( Maxwell and Johnson, 2000 ), indicating a mild photoinhibition under conditions where it is usually very pronounced ( Valladares and Pearcy, 1997 ). However, this high resistance to photoinhibitory damage during summer drought along the sun exposure stress gradient considered has been already reported for this species under fi eld conditions ( Arag ó n et al., 2008 ). Contrary to the widely accepted view of north-facing slopes as more favorable environments for plant performance in Mediterranean habitats ( Kutiel, 1992 ; Esc ó s et al., 2000 ), the south-facing slope did not represent a more

Table 1. Results of the generalized linear models for the effects of treatment (deblossomed vs. control), slope aspect and their interaction, together with plant size, on next season ’ s survival, growth, and reproduction (fl owers and seeds per plant, fruit set, and abortion rate per fruit) of Helianthemum squamatum . When modeling the reproductive variables, recorded in 2006, we used plant size measured in the same year, which is known to have an important effect in the reproductive output. When modeling growth and survival, also recorded in 2006, we had to use plant size measured in the previous year (2005). Signifi cant effects ( P < 0.05) are in boldface type.

Survival Growth Flowers Seeds Fruit set Abortion Rate

Source of variation χ 2 P χ 2 P χ 2 P χ 2 P χ 2 P χ 2 p

Treatment 4.99 0.02 1.21 0.28 0.13 0.72 0.05 0.81 0.08 0.77 0.36 0.56Aspect 2.87 0.09 0.12 0.73 1.71 0.21 0.17 0.69 0.06 0.81 0.34 0.55Treatment × Aspect 0.10 0.76 0.29 0.59 0.06 0.82 0.04 0.85 0.02 0.89 0.29 0.59Plant size 0.00 0.99 7.83 0.01 4.40 0.03 4.02 0.04 1.35 0.24 0.15 0.70Distribution Binomial Normal Poisson Poisson Binomial BinomialLink function Logit Identity Log Log Logit LogitNo. of cases 60 22 22 21 22 21


Fig. 3. Maximum photochemical effi ciency of PSII ( F v / F m ) in Helianthemum squamatum at midday and leaf concentrations of N and P (mmol/g dry mass) at different phenological stages (before fl owering, at the onset and the peak of fl owering, and at the peak of fruiting) measured during the reproduc-tive period of 2005 in 30 plants per slope. Data points represent mean ± SE. Horizontal lines indicate signifi cant differences between dates ( P < 0.05); homogeneous groups of means appear under the same horizontal line. No differences between treatments were found at any date.

of 60 plants is consistent with the possibility that growth was not dramatically affected by reproduction, because leaf nitro-gen and phosphorus contents are directly related to growth rate ( Cornelissen et al., 1997 ). Nevertheless, this argument must be

taken with caution since we do not really know the potential role of storage on nutrient dynamics in this species. Second, it seems that providing that the plant survives, reproduction takes place again regardless of the previous reproductive investment


(C. F. Arag ó n et al., unpublished manuscript). However, whether somatic costs are expressed in this species requires fur-ther study.

The high reproductive output of H. squamatum plants even in very unfavorable years ( Arag ó n et al., 2007 ), a strategy more characteristic of semelparous species ( Hautek è ete et al., 2001 ), becomes “ suicidal ” given the observed costs of reproduction for plant survival. This pattern of investment may contribute to the observed short life span of this species ( Caballero, 2006 ), rather unusual for a perennial plant in a semiarid environment ( Agami, 1987 ; Cody, 2000 ). Nevertheless, it is also worth men-tioning that 2005 summer conditions were extreme, with tem-peratures above those of 2003 when the greatest heat wave registered in Europe in the last 150 years took place ( Sch ä r and Jendritzky, 2004 ). Reproduction may have made plants more sensitive to extrinsic sources of mortality not related to plant status and thereby not expressed as somatic costs of reproduction.

Survival in H. squamatum was remarkably low, even in the deblossomed plants. Apparently, extrinsic (i.e., condition-in-dependent) mortality sources are important in the demography of H. squamatum . In semiarid and arid climates, characterized by a high environmental stochasticity, perennial shrubs usu-ally rely on their longevity to persist over time ( Garc í a and Zamora, 2003 ) because the effort invested in new plant estab-lishment seems to be very high and the allocation of energy to stress tolerance mechanisms is supposed to increase survival but to reduce energy for reproduction ( Esc ó s et al., 2000 ). This very short life cycle of H. squamatum plants should be classi-fi ed as an “ annual-like behavior ” following Agami (1987) , who also reported it for other perennials of the same genus. Persistence of genetic individuals can be achieved either by a long life span of established individuals or by a long-lived seed bank ( Ehrl é n and Lehtil ä , 2002 ). Like other species in semiarid climates, H. squamatum has a highly structured and persistent seed bank ( Caballero et al., 2005 ), a trait that has been shown to be negatively correlated with adult longevity ( Arroyo et al., 2006 ).

To conclude, we have shown exceptionally high levels of extrinsic mortality for H. squamatum , a perennial plant spe-cialized for gypsum, semiarid ecosystems. In agreement with the framework proposed by Franco and Silvertown (1997) , the success of H. squamatum in such habitats relies on a persistent seed bank, combined with a sustained high reproductive output in a wide range of environmental conditions ( Arag ó n et al., 2007 ) at the expense of survival, that together leads to a rapid turnover of plants. Our study, together with other evidence in perennial Helianthemum ( Agami, 1987 ), suggests that this strategy might be more common than previously thought among short-lived shrubby plants growing in stressful Medi-terranean areas.

LITERATURE CITED

Agami , M. 1987 . The establishment, growth, phenology and longevity of two species of Helianthemum in the Negev Desert Highlands, Israel. Oecologia 71 : 486 – 490 .

Arag ó n , C. F. , M. J. Albert , L. Gim é nez-Benavides , A. L. Luzuriaga , and A. Escudero . 2007 . Environmental scales on the reproduction of a gypsophyte: A hierarchical approach. Annals of Botany 99 : 519 – 527 .

Arag ó n , C. F. , and A. Escudero . 2008 . Mating system of Helianthemum squamatum (Cistaceae), a gypsophile specialist of semi-arid Mediterranean environments. Botanica Helvetica 118 : 129 – 137 .

Arag ó n , C. F. , A. Escudero , and F. Valladares . 2008 . Stress-induced dynamic adjustments in reproduction differentially affect fi tness com-ponents in a semiarid plant. Journal of Ecology 96 : 222 – 229 .

Arroyo , M. T. K. , P. Chac ó n , and L. A. Cavieres . 2006 . Relationship between seed bank expression, adult longevity and aridity in species of Chaetanthera (Asteraceae) in central Chile. Annals of Botany 98 : 591 – 600 .

Ashman , T.-L. 1994 . A dynamic perspective on the physiological cost of reproduction in plants. American Naturalist 144 : 300 – 316 .

Bell , G. 1980 . The costs of reproduction and their consequences. American Naturalist 116 : 45 – 76 .

Biere , A. 1995 . Genotypic and plastic variation in plant size: Effects on fecundity and allocation patterns in Lychnis fl os-cuculi along a gradi-ent of natural soil fertility. Journal of Ecology 83 : 629 – 642 .

Bond , W. J. , and K. E. Maze . 1999 . Survival costs and reproductive benefi ts of fl oral display in a sexually dimorphic dioecious shrub, Leucadendron xanthoconus. Evolutionary Ecology 13 : 1 – 18 .

Caballero , I. 2006 . Estructura espacio-temporal de un banco de semillas: Las comunidades gips í colas del centro de la Pen í nsula Ib é rica. PhD dissertation, Universidad del Pa í s Vasco, Bilbao, Spain.

Caballero , I. , J. M. Olano , A. L. Luzuriaga , and A. Escudero . 2005 . Spatial coherence between seasonal seed banks in a semiarid gypsum community: Density changes but structure does not. Seed Science Research 15 : 153 – 160 .

Castro-D í ez , P. , and G. Montserrat-Mart í . 1998 . Phenological pattern of fi fteen Mediterranean phanaerophytes from Quercus ilex commu-nities of NE-Spain. Plant Ecology 139 : 103 – 112 .

Chapin , F. S. 1991 . Effects of multiple environmental stresses on nutri-ent availability and use. In H. A. Mooney, W. E. Winner, and E. J. Pell [eds.], Response of plants to multiple stresses, 67 – 88. Academic Press, San Diego, California, USA.

Chapin , F. S. III , E. Schulze , and H. A. Mooney . 1990 . The ecology and economics of storage in plants. Annual Review of Ecology and Systematics 21 : 423 – 447 .

Cheplick , G. P. 1995 . Life history trade-offs in Amphibromus scabrival-vis (Poaceae): Allocation to clonal growth, storage, and cleistogamous reproduction. American Journal of Botany 82 : 621 – 629 .

Cheplick , G. P. 2005 . The allometry of reproductive allocation. In E. G. Reekie and F. Bazzaz [eds.], Reproductive allocation in plants, 97 – 128. Elsevier Academic Press, New York, New York, USA.

Cody , M. L. 2000 . Slow-motion population dynamics in Mojave desert perennial plants. Journal of Vegetation Science 11 : 351 – 358 .

Cornelissen , J. H. C. , M. J. A. Werger , P. Castro-D í ez , J. W. A. van Rheenen , and A. P. Rowland . 1997 . Foliar nutrients in relation to growth, allocation and leaf traits in seedlings of a wide range of woody plant species and types. Oecologia 111 : 460 – 469 .

Ehrl é n , J. , and K. Lehtil ä . 2002 . How perennial are perennial plants? Oikos 98 : 308 – 322 .

Ehrl é n , J. , and J. Van Groenendael . 2001 . Storage and the delayed cost of reproduction in the understorey perennial Lathyrus vernus. Journal of Ecology 89 : 237 – 246 .

Esc ó s , J. , C. L. Alados , F. J. Pugnaire , J. Puigdef á bregas , and J. Emlen . 2000 . Stress resistance strategy in an arid land shrub: Interactions between developmental instability and fractal dimension. Journal of Arid Environments 45 : 325 – 336 .

Franco , M. , and J. Silvertown . 1997 . Life history variation in plants: An exploration of the fast-slow continuum hypothesis. In J. Silvertown, M. Franco, and J. L. Harper [eds.], Plant life histories: Ecology, phylogeny and evolution, 210 – 227. Cambridge University Press, Cambridge, UK.

Garc í a , D. , and R. Zamora . 2003 . Persistence, multiple demographic strategies and conservation in long-lived Mediterranean plants. Journal of Vegetation Science 14 : 921 – 926 .

Gillman , M. P. , and M. J. Crawley . 1990 . The cost of sexual reproduc-tion in ragwort ( Senecio jacobaea L.). Functional Ecology 4 : 585 – 589 .

Hautek è ete , N.-C. , Y. Piquot , and H. Van Dick . 2001 . Investment in sur-vival and reproduction along a semelparity – iteroparity gradient in the Beta species complex. Journal of Evolutionary Biology 14 : 795 – 804 .


Karlsson , P. S. , and M. M é ndez . 2005 . The resource economy of plant reproduction. In E. G. Reekie and F. Bazzaz [eds.], Reproductive allocation in plants, 1 – 49. Elsevier Academic Press, New York, USA.

Karlsson , P. S. , B. M. Svensson , B. Å . Carlsson , and K. O. Nordell . 1990 . Resource investment in reproduction and its consequences in three Pinguicula species. Oikos 59 : 393 – 398 .

Kozlowski , J. 1991 . Optimal energy allocation models — An alternative to the concepts of reproductive effort and cost of reproduction. Acta Oecologica 12 : 11 – 33 .

Kutiel , P. 1992 . Slope aspect effect on soil and vegetation in a Mediterranean ecosystem. Israel Journal of Botany 41 : 243 – 250 .

Law , R. 1979 . The cost of reproduction in an annual meadow grass. American Naturalist 113 : 3 – 16 .

Liang , K. Y. , and S. L. Zeger . 1986 . Longitudinal analysis using gener-alized linear models. Biometrika 73 : 13 – 22 .

L ó pez-Gonz á lez , G. 1993 . Helianthemum Mill. In S. Castroviejo, C. Aedo, S. Cirujano, M. Lainz, P. Montserrat, R. Morales, F. Mu ñ oz-Garmendia, et al. [eds.], Flora Iberica III, 365 – 421. Centro Superior de Investigaciones Cient í fi cas, Madrid, Spain.

Mart í nez , I. , A. Escudero , F. T. Maestre , A. de la Cruz , C. Guerrero , and A. Rubio . 2006 . Small-scale patterns of abundant of mosses and lichens forming biological soil crusts in two semi-arid gypsum envi-ronments. Australian Journal of Botany 54 : 339 – 348 .

Maxwell , K. , and G. N. Johnson . 2000 . Chlorophyll fl uorescence: A practical guide. Journal of Experimental Botany 51 : 659 – 668 .

McCullagh , P. , and J. A. Nelder . 1989 . Generalized linear models. Chapman and Hall, New York, New York, USA.

Obeso , J. R. 2002 . The costs of reproduction in plants. New Phytologist 155 : 321 – 348 .

Paige , K. N. , and T. G. Whitham . 1987 . Flexible life history traits: Shifts by scarlet gilia in response to pollinator abundance. Ecology 68 : 1691 – 1695 .

Pakonen , T. K. Laine, P. Havas, and E. Saari . 1988 . Effects of berry production and deblossoming on growth, carbohydrate and nitrogen compounds in Vaccinium myrtillus in North Finland. Acta Botanica Fennica 136 : 37 – 42 .

Pi ñ ero , D. , J. Sarukh á n , and P. Alberdi . 1982 . The costs of reproduction in a tropical palm, Astrocaryum mexicanum. Journal of Ecology 70 : 473 – 481 .

Primack , R. B. , and J. Antonovics . 1982 . Experimental ecological genet-ics in Plantago . VII. Reproductive effort in populations of P. lanceo-lata L. Evolution 36 : 742 – 752 .

Reekie , E. G. 1991 . Cost of seed versus rhizome production in Agropyron repens. Canadian Journal of Botany 69 : 2678 – 2683 .

Reekie , E. G. , and G. Avila-Sakar . 2005 . The shape of the trade-off func-tion between reproduction and growth. In E. G. Reekie and F. Bazzaz [eds.], Reproductive allocation in plants, 189 – 214. Elsevier Academic Press, New York, New York, USA.

Reznick , R. 1985 . Cost of reproduction: An evaluation of the empirical evidence. Oikos 44 : 257 – 267 .

Rom ã o , R. L. , and A. Escudero . 2005 . Gypsum physical soil crusts and the existence of gypsophytes in semi-arid central Spain. Plant Ecology 181 : 127 – 137 .

Saikkonen , K. , S. Koivunen , T. Vuorisalo , and P. Mutikainen . 1998 . Interactive effects of pollination and heavy metals on resource allocation in Potentilla anserina L. Ecology 79 : 1620 – 1629 .

Sandvik , S. M. 2001 . Somatic and demographic costs under differ-ent temperature regimes in the late-fl owering alpine perennial herb Saxifraga stellaris (Saxifragaceae). Oikos 93 : 303 – 311 .

SAS Institute . 1990 . SAS statistical package, version 9.0. SAS Institute, Cary, North Carolina, USA.

Sch ä r , C. , and G. Jendritzky . 2004 . Hot news from summer 2003. Nature 432 : 559 – 560 .

Smith , A. P. , and T. P. Young . 1982 . The cost of reproduction in Senecio keniodendron , a giant rosette species from Mt. Kenia. Oecologia 55 : 243 – 247 .

Sohn , J. J. , and D. Policansky . 1977 . The costs of reproduction in the mayapple Podophyllum peltatum (Berberidaceae). Ecology 58 : 1366 – 1374 .

Stearns , S. C. 1989 . Trade-offs in life-history evolution. Functional Ecology 3 : 259 – 268 .

Stearns , S. C. 1992 . The evolution of life histories. Oxford University Press, New York, New York, USA.

Tuomi , J. , T. Hakala , and E. Haukioja . 1983 . Alternative concepts of re-productive effort, cost of reproduction, and selection in life-history evolution. American Zoologist 23 : 25 – 34 .

Valladares , F. , and R. W. Pearcy . 1997 . Interactions between water stress, sun-shade acclimation, heat tolerance and photoinhibition in the sclero-phyll Heteromeles arbutifolia. Plant, Cell & Environment 20 : 25 – 36 .

Valladares , F. , and R. W. Pearcy . 2002 . Drought can be more critical in the shade than in the sun: a fi eld study of carbon gain and photo-inhibition in a Californian shrub during El Ni ñ o year. Plant, Cell & Environment 25 : 749 – 759 .

Williams , G. C. 1966 . Natural selection, the costs of reproduction, and refi nement of Lack ’ s principle. American Naturalist 100 : 687 – 690 .

Zotz , G. , and A. Richter . 2006 . Changes in carbohydrate and nutrient contents throughout a reproductive cycle indicate that phosphorus is a limiting nutrient in the epiphytic bromeliad, Werauhia sanguinolenta. Annals of Botany 97 : 745 – 754 .

survival costs of reproduction in a short-lived perennial plant: live

Documents