| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
2Department of Biological and Environmental Science, University of Jyväskylä, P.O. Box 35, FI-40351 Jyväskylä, Finland; 3Oulanka Research Station, University of Oulu, Liikasenvaarantie 134, FI-93999, Kuusamo, Finland
Received for publication February 15, 2005. Accepted for publication July 25, 2005.
ABSTRACT
We studied inbreeding depression in a perennial plant, Lychnis viscaria, in three populations differing in their inbreeding history and population size by measuring several traits at two nutrient levels over the plant's life cycle. The observed levels of inbreeding depression (cumulative inbreeding depression, from –0.057 to 0.629) were high for a plant with a mixed mating system. As expected, the population with a low level of isozyme variation expressed the least inbreeding depression for seed germination. Highest inbreeding depression for germination was found in the largest and genetically most variable population. No clear differences between populations in expression of inbreeding depression in the later life stages were found. The population level inbreeding depression varied with the nutrient conditions and among populations and life stages, but we found no evidence that inbreeding depression increased with lower nutrient availability. These results emphasize the importance of measuring inbreeding depression under several environmental conditions and over life stages.
Key Words: Caryophyllaceae • environmental quality • inbreeding depression • nutrient availability
Understanding inbreeding depression and its fitness consequences remains an important and challenging topic in evolutionary biology because of its potential importance in the evolution of mating systems, life history traits, and its implications for conservation and plant breeding. As population sizes of endangered species decrease, a consequent reduction in genetic diversity may endanger the ability of populations to adapt to changing environments and thus may affect the long-term persistence of the population, but the reduction in viability of individuals that results from inbreeding depression increases the risk of extinction even in the short term (e.g., Saccheri et al., 1998
).
The negative effects of inbreeding have been studied and documented widely (Charlesworth and Charlesworth, 1979
; Lloyd, 1979
; Lande and Schemske, 1985
, and references therein), but predicting the magnitude of inbreeding depression in any given population remains difficult (Husband and Schemske, 1996
). There are few generalities about the genetics of inbreeding depression; inbreeding depression may differ in magnitude from genealogical lineage to lineage, and because of different types and degrees of dominance and different degrees of epistasis. The magnitude of inbreeding depression is expected to be influenced by such things as mating history, life span, current and past population size (bottlenecks, founder effects), environmental conditions, and possibly by interactions between these factors. In plant populations with a mixed mating system, inbreeding may be increased not only as a result of a small number of potential mates, which evidently leads to mating with relatives, but also through greater selfing from elevated levels of autogamy and geitonogamy (Olesen and Jain, 1994
).
Inbreeding depression may be caused by the expression of deleterious recessive alleles or over dominant loci, or both of these factors in combination. Other factors, such as epistasis, may also be affecting the genetic basis and subsequent expression of inbreeding depression. Epistasis can either enhance or inhibit inbreeding depression. It causes the decline in fitness to be nonlinear, and thus makes estimating and predicting the magnitude of inbreeding depression difficult. Current evidence suggests that inbreeding depression results primarily from partially recessive deleterious alleles (Charlesworth and Charlesworth, 1987
; Johnston and Schoen, 1995
; Willis, 1999
; Carr and Dudash, 2003
). In a population going through generations of inbreeding, deleterious recessive alleles responsible for inbreeding depression are exposed to selection as homozygotes and, therefore, may be purged (Lande and Schemske, 1985
). As a result, selfing populations or populations with a history of inbreeding are expected to have lower levels of inbreeding depression than populations that have been predominantly outcrossing. Inbred and outbred species may also differ in the timing of inbreeding depression (Husband and Schemske, 1996
); outcrossers have strong inbreeding depression during early stages of their life-cycle, while selfers have milder inbreeding depression in later life stages. Inbreeding depression expressed in early life history traits may be caused by a few recessive lethals in one or a few loci that are effectively purged from selfing populations (Husband and Schemske, 1996
). Traits expressed in later life stages are often under polygenic control, and inbreeding depression in these traits is largely caused by mildly deleterious alleles. Thus, a mutation load of such characters is not easily purged. Because the expression of inbreeding depression is known to vary throughout the life cycle (Crnokrak and Roff, 1999
; Koelewinj et al., 1999
), measurements over the whole life span are needed for a reliable measure of inbreeding depression (Charlesworth and Charlesworth, 1987
). Several authors have reported reduction in inbreeding depression in inbred populations (Barrett and Charlesworth, 1991
; Dole and Ritland, 1993
; Johnston and Schoen, 1996
) or species (Husband and Schemske, 1996
; Goodwillie, 2000
). However, recent evidence suggests that selfing populations can also have considerable amounts of inbreeding depression (Byers and Waller, 1999
, and references therein; Willis, 1999
) and that purging may effectively remove only a minor proportion of alleles that cause inbreeding depression. Different interactions between loci, such as epistasis, can also impede the extent to which genetic load can be purged. Therefore, more information is needed on the levels of inbreeding depression in populations with different mating histories.
There is a growing body of evidence showing that inbreeding depression is environmentally dependent (e.g., review by Henry et al., 2003
). As a result of the more pronounced environmental sensitivity of inbred plants, inbreeding depression is likely to increase in adverse environmental conditions (Dudash, 1990
; Hoffmann and Parsons, 1991
). In earlier studies, more severe inbreeding depression was found in harsher environmental conditions, such as in the field as opposed to greenhouse environments (Dudash, 1990
; Eckert and Barret, 1994
), under competition (Schmitt and Ehrhardt, 1990
; Cheptou et al., 2000
), in the presence of herbivores (Carr and Eubanks, 2001
), under natural as opposed to captive conditions (Crnokrak and Roff, 1999
), or with lower food availability (Keller et al., 2002
). However, in some cases, higher inbreeding depression was observed in laboratory conditions than in natural conditions. For example, Henry et al. (2003)
reported higher inbreeding depression for growth and survival in the freshwater snail in the laboratory, while outbreeding depression was evident in the field. Therefore, to get a more reliable measure of inbreeding depression, the studies should be conducted in natural habitats (Schemske, 1983
; Eckert and Barret, 1994
) or preferably in different environmental conditions (e.g., Schemske, 1983
; Dudash, 1990
). Additionally, experiments manipulating both the level of inbreeding and environmental conditions are needed to investigate the interaction between inbreeding depression and environmental conditions.
In this study, the expression and timing of inbreeding depression over the plants' life histories were examined in three peripheral Lychnis viscaria populations that differed in population size and level of heterozygosity. We tested the prediction that less inbreeding depression would occur in more homozygous and/or smaller populations that are likely to have experienced higher levels of inbreeding during the earlier generations. In addition, we used two nutrient levels to study the effect of nutrient availability on the expression of inbreeding depression. A major part of the experiment was conducted in the field instead of the greenhouse to let the more realistic environmental conditions (diseases, herbivores, and weather) influence plant growth and reproduction. We expected inbreeding depression to be most pronounced in the largest and most heterozygous population and to be expressed in the early stages of the life cycle, while the homozygous and smaller populations were expected to have lower levels of inbreeding, mainly at later life history stages. We also expected the fertilized individuals to express less inbreeding depression than the unfertilized ones, as the fertilized individuals have more resources for their growth and development.
MATERIALS AND METHODS
Lychnis viscaria L. (Caryophyllaceae) is a perennial herb, occurring in open, sunny habitats like dry meadows and south-facing, rocky cliffs. It is found in most parts of central Europe, and its main distribution area extends to northern Scandinavia to the 62nd latitude. A few isolated populations are found up to the 68th latitude (Hulten, 1971
). In the surroundings of Jyväskylä (62°15' N, 25°45' E) where the populations studied were located, L. viscaria occurs in its northern range in rather small and isolated patches.
Plants produce 1–50 inflorescences, each bearing about 20–25 purple flowers. The flowers are protandrous and pollinated by insects, mainly bumblebees and butterflies (Jennersten, 1988
). Despite protandry, autogamous self-pollination is possible, but seed set is low after autogamy (Kwak and Jennersten, 1986
). The seeds are dispersed by gravity.
In our previous study (Siikamäki and Lammi, 1998
; Lammi et al., 1999
), we estimated the isozyme variation in 11 populations of L. viscaria, 3 from the central distribution area and 8 from the peripheral area, by using starch gel electrophoresis as described in Wendel and Weeden (1989
; for details see Siikamäki and Lammi, 1998
). The mean population sizes in peripheral and central areas were 122 (N = 8, SD = 102) and 700 (N = 3, SD = 360) flowering individuals, and the corresponding expected heterozygosities were 0.034 (N = 8, SD = 0.010) and 0.114 (N = 3, SD = 0.019), respectively (Lammi et al., 1999
). Determining the isozyme variation (Siikamäki and Lammi, 1998
; Lammi et al., 1999
) allowed us to predict the levels of previous inbreeding in these populations. For this study, three populations (Iso-Salmijärvi, Kanavuori, and Vaarunvuori) were selected from the periphery of the distribution area of L. viscaria in Finland to have approximately the same climatic and other environmental conditions, but with as much difference in the population sizes and genetic variability as possible (Table 1). The habitats of all populations were open, east-west facing, sparsely vegetated, exposed rock habitats. There is evidence for partial self-incompatibility in some populations (Wilson et al., 1995
), but the studied populations are self-compatible (Table 1).
|
). The seeds of L. viscaria germinate readily (>90%) without any pretreatment in autumn or spring, and there is no evidence of seed dormancy. Thus, the seed bank probably does not play an important role in the population dynamics of L. viscaria, and sampling the flowering individuals gives a relatively reliable estimate of population size.
Hand pollination in field populations
To estimate the level of inbreeding depression for germination, growth, and reproduction, hand pollinations were conducted in early July 1996 in three populations with different sizes and different levels of isozyme variation. According to historical data (first records from 1916), none of these populations have had recent bottlenecks in population sizes (Välivaara et al., 1991
, references therein).
During hand pollinations within each of the three study populations, in 14– 15 families, an average of 40% of the flowers (five flowers per each cross type) were either self-pollinated (geitonogamy) or cross-pollinated (allogamy). The pollinated flowers had been covered with nylon bags for 3 wk before the pollinations and were emasculated 3 d in advance. The two pollination treatments were conducted randomly within the inflorescence by rubbing removed anthers from the same individual (self-) or from a random individual within the population (cross-) to the receptive stigmas. Only flowers at an early, but mature female stage were accepted as pollen recipients. Each cross-pollinated flower within an individual plant was pollinated by one father, and each of these flowers had a different father.
After the hand-pollinations, the experimental plants were then again covered with nylon bags to exclude pollinators for 3 wk. Due to environmental conditions in population Vaarunjyrkkä, a proportion of the pollinations failed and maternal family data was thus incomplete. Therefore, for some families only outcrossed progeny were available for analysis, and for others only inbred progeny were available.
Germination
Mature, unopened capsules were collected from the self and cross pollinations in late July 1996 and kept at 5°C for 3 mo before the germination experiments. To study germination rate and germination percentage, 50 seeds from both cross types per family were germinated on wet paper in petri dishes. To get an averaged sample over pollen donors, all seed capsules obtained from a single maternal family were put together; seeds were mixed and a random sample of seeds was chosen. The germination was conducted at 22°C with continuous light (1991 W/m2). The seeds were monitored daily for 3 wk, then the final germination percentage was determined. The seeds were considered to have germinated when the cotyledons became visible.
Seedling growth
Seeds from the outcross and self pollinations in the three populations were sown in 4 x 4 cm pots, 3–4 seeds per pot, 5–10 pots per each maternal family for each treatment and population. The soil consisted of a mixture of peat moss and vermiculite. The pots were covered with plastic foil and kept at uniform conditions at 19°C, 160 W/m light for 12 h followed by 12 h of dark. The pots were thinned twice, the first time at 3 wk after sowing to two seedlings per pot and, after another week, to one seedling per pot. Four wk after sowing, the boxes were transferred to a greenhouse (temperature 20°C, 120 W/m light), and the plastic cover was removed. The pots were watered when necessary.
Because L. viscaria is sensitive to competition (Wilson et al., 1995
), seedling growth and establishment are probably critical periods in the individual life cycle (Wilson et al., 1995
). Therefore, seedling growth rate was examined with special care by measuring the length of the longest leaf four times: 6 (19 November), 9 (12 December), 15 (29 January), and 28 (19 May) wk from sowing. The growth rate was calculated as the mean growth in mm per week, (
xi/ti)/N (ti = time from the beginning of the experiment in weeks, xi = leaf length at time ti, and N = number of measurements). This measure was used because it gives weight to fast growth in the beginning, a key factor for outcompeting other seedlings during early growth.
From our field observations, we know that the size and reproductive output of L. viscaria individuals is closely connected to the habitat and soil type: in rocky and sandy places with a thin and nutrient-poor soil, the plants are much smaller and produce much fewer seeds than plants in meadow habitats with a deeper, nutrient-rich soil layer (K. Mustajärvi et al., unpublished data). To examine whether the expression of inbreeding is affected by the availability of resources, half of the outcrossed and selfed seedlings of each population were fertilized (Puutarhan PK fertilizer 27–17, N 2.0%, P 7.0%, K 16.6%, Ca 5.4%, Mg 2.5%) after the second growth measurement, 9 wk after sowing. Each pot received 25 mL of (1 g/100 mL H20) fertilizer. The treatment was repeated a week later. Fifteen wk after sowing (29 January), seedling growth was measured for the third time, and leaf number was also measured. The pots were then transferred to a cold room (4°C, relative humidity 37%) for a 3-mo vernalization. In April 1997, the plants were transferred back to the greenhouse for a month. In May, seedling survival ratio, the ratio to the number of sowed pots was measured. Then the seedlings from the three populations were planted separately in three uniform locations about 200 m from each other in a 100-ha field in Laukaa Research and Elite Plant Station (62°15' N, 25°30' E). The station is situated relatively close to the original populations (Table 1). The vegetation surrounding the common garden consisted of cultivated Phleum pratense, Trifolium hybridum, Hordeum vulgare, and Brassica rapa, as well as naturally occurring meadow species. The habitat is relatively similar to the heathy grasslands and dry meadows where L. viscaria naturally occurs in its central distribution area, but is probably more fertile than the original habitats of the experimental populations.
The two nutrient levels were maintained by fertilizing only the previously fertilized seedlings with organic fertilizer sticks (N 4%, P 1%, K 2%, Biolan Ruukun voima (Biolan oy, Kauttua, Finland). The sticks were inserted close to the roots of the plants a few days after planting and again at the beginning of the 1998 growing season. The nutrient treatments are referred to as fertilized and unfertilized (= control) treatments throughout the paper.
Measurements in the common garden
In 1997, only a small fraction of the plants flowered, and we only recorded whether a plant produced flowers or not to study if flowering in 1997 would affect reproductive success in 1998. The number of overwintering, surviving plants was recorded at the beginning of the 1998 flowering season. The total reproductive success of the plants was estimated, only in 1998, as the number of flowering stems and number of produced capsules (in naturally, insect-pollinated flower stems) in the longest flowering stem. The total capsule production per individual was then estimated by multiplying the number of stems by the number of capsules in the longest stem. This is a reliable estimate due to the architecture of L. viscaria plants. The flowering stems within a single plant are similar enough to yield a reliable estimate with this method.
Hand pollination in the common garden
In 1998, part of the plants (20 individuals, each family represented, from each cross x fertilization treatment in each population) were chosen for the hand-pollination experiment to examine the effect of outcrossing and selfing on the next generation and the maternal effect of inbreeding depression on seed production (seed number per capsule). These plants were covered with transparent nylon sheet cages before flowering. When the flowers opened, 5–8 flowers at an early male stage were emasculated. When the stigmas of these flowers were receptive, the flowers in plants that had been grown from selfed seeds were selfed and flowers in plants that were grown from outcrossed seeds were crossed with any random individual from the same artificial population. To test if the hand pollination per se would affect seed yield, the rest of the plants were left to be naturally pollinated by insects. Five mature unopened capsules were collected from both open-pollinated and hand-pollinated plants and weighed. To estimate the mean mass of a single seed of an individual, a random subset of 20 seeds from each plant was weighed to the nearest 0.001 mg. Seed set per capsule was calculated by dividing the average mass of five capsules from a plant by the average mass of an individual seed.
Statistical analysis
When only a maternal mean of 50 seeds per cross was available, the germination percentage was analyzed with a split-plot ANOVA model, with cross and population as a fixed effects and family as random effect (block) within population. The seedling growth rate, number of leaves, and total capsule production was analyzed with ANOVA, in which population, fertilization, and cross were considered as fixed factors, and family as a random factor nested within population. In the analysis of the two types of survival rates, the ANOVA was performed for proportion of the family surviving or flowering within treatments. In the analysis of seed set per capsule, the pollination type (hand pollination vs. open pollination) was also included in the model as a fixed factor. The data on seed production was first analyzed with a full model including all the possible three-and four-way interactions between the fixed effects, but because none of these interactions were significant (P > 0.100) and their removal did not affect the outcome of the analysis, they were excluded from the analysis to make it easier to interpret. For all the ANOVA tests described, data were log-transformed so that interactions between cross-type and other factors tested for variation in the ratio of selfed to outcrossed values were apparent, rather than for their differences (Johnston and Schoen, 1994
). Because only a small fraction of plants flowered in 1997, the effects of fertilization and cross and the differences between populations in probability of flowering during first year of growth were analyzed with chi-square tests.
Analysis of inbreeding depression
To illustrate the expression of inbreeding depression in the three populations at different life stages at the two nutrient levels, we calculated the population level inbreeding depression as 
= (wo – ws)/wo. Positive values indicate that the performance of outcrossed individuals exceeds that of selfed individuals (inbreeding depression), and negative values indicate that the performance of selfed individuals exceeds that of outcrossed individuals (outbreeding depression). Inbreeding depression was estimated for germination percentage, seedling growth rate, leaf number, seedling survival, overwintering survival, capsule production, seed set per capsule, and for cumulative fitness for each population, separately for both fertilization treatments. Cumulative fitness was calculated for each cross type in each population for both fertilization treatments as w = germination percentage x seedling survival rate x overwintering survival rate x total capsule production. Population inbreeding depression, rather than an estimate based upon family means, was calculated (Johnston and Schoen, 1994
). However, because studying the variation in family-level inbreeding depression is very important for understanding the dynamics of inbreeding depression with genes that modify the mating system (Holsinger, 1988
, 1991
; Uyenoyama and Waller, 1991a
, b
), inbreeding depression was also measured at the family level. Student's t tests were used to check if the family level inbreeding depression estimates for the life history traits deviated significantly from zero. Because cumulative inbreeding depression using family data yielded only a few valid replicates in Vaarunjyrkkä, where the progeny data was sparse already in the beginning of the experiment, the cumulative inbreeding depression was calculated at the population level only.
RESULTS
Germination success
The inbreeding had an effect on the germination success of seeds (Table 2, Fig. 1), but as suspected, the populations differed in their response to selfing: outcrossed seeds outperformed the selfed seeds in Kanavuori (t15 = 2.58, P = 0.021) and Vaarunjyrkkä (t14 = 2.67, P = 0.018), while no difference was found in Iso-Salmijärvi (t16 = –1.09, P = 0.293). The populations also differed in germination success (Table 2, Fig. 1), and this difference was significant between Iso-Salmijärvi and Vaarunjyrkkä (post hoc Bonferroni, mean difference 0.255, P = 0.008).
|
|
|
|
Fertilized individuals had more leaves than unfertilized individuals (Table 2, Fig. 2B). This result was consistent over all populations (mixed model ANOVA: Iso-Salmijärvi, fertilization, F1,258 = 52.59, P < 0.001; Kanavuori, fertilization, F1,276 = 51.22, P < 0.001; Vaarunjyrkkä, fertilization, F1,264 = 201.10, P < 0.001). The effect of fertilization was extremely strong in Vaarunjyrkkä, causing significant population x fertilization interaction (Table 2). In population Iso-Salmijärvi, however (cross x fertilization: F1,276 = 22.67, P < 0.001, Fig. 2B), no differences were found among the selfed seedlings in their reaction to fertilization (t95 = 1.54, P = 0.127), causing a significant three-way interaction in Table 2.
Seedling survival was significantly lower for fertilized seedlings than for the unfertilized ones, but no differences among populations or between selfed and crossed seedlings (Table 2, Fig. 2C) or interactions between these factors were found.
Flowering during the first growing season and overwintering success
After the first year of growth in 1997, a higher proportion of fertilized than unfertilized individuals flowered (Fig. 3A, df = 1, 
2 = 85.1, P < 0.001), but cross had no significant effect on flowering (Fig. 3A, df = 1, 2 = 1.35, P = 0.141). No interaction between cross and fertilization was found (heterogeneity chi-square of samples [Zar, 1999
]: df = 2, 2 = 0.02, P > 0.1), but populations differed in their flowering probability (Fig. 3A, df = 2, 2 = 15.9, P < 0.001); significantly fewer plants from Iso-Salmijärvi than from Kanavuori (multiple comparisons Tukey test [Zar, 1999
]: q
,3 = 5.71, P < 0.05) and Vaarunjyrkkä (q,3 = 3.56, P < 0.05) flowered. There was no trade-off for flowering during the first growing season, instead plants that flowered during the first year had higher probability of also flowering in 1998 (df = 1, 2 = 8.32, P = 0.005), and reproductive success (total seed set in the following year in 1998) was not dependent on the flowering in 1997 (mean ± SE for plants that flowered in 1997 = 651 ± 46, mean ± SE for plants that flowered only in 1998 = 614 ± 26, ANOVA: F1,401 = 0.477, P = 0.490). The tests presented here are for the pooled data, but the trend was consistent within populations (tests not shown).
|
Reproductive success
Cross, fertilization, or population had no significant effect on capsule production, but populations reacted differently to fertilization (Table 2, Fig. 4A). In Iso-Salmijärvi (ANOVA, fertilization: F1,129 = 6.25, P = 0.014), fertilized plants produced more capsules than unfertilized plants, and in Kanavuori (mixed-model ANOVA, fertilization: F1,125 = 10.17, P = 0.002), unfertilized plants outperformed fertilized ones. No differences were found in Vaarunjyrkkä (mixed-model ANOVA, fertilization: F1,110 = 1.66, P = 0.201).
|
|
from zero) in several traits in Kanavuori, while in Iso-Salmijärvi no significant inbreeding was observed (Fig. 5). In Vaarunjyrkkä, only a few observations (N = 3) were available and thus the t tests were not reliable to test the level of inbreeding depression in this population.
|
The highest cumulative population level inbreeding was expressed in the individuals from the largest and most variable population, Vaarunjyrkkä (unfertilized: 0.629; fertilized: 0.425). Cumulative inbreeding depression for individuals from Kanavuori was 0.308 and 0.122 for the unfertilized and fertilized plants, respectively. For both of these populations the cumulative inbreeding depression was thus higher for unfertilized plants. In Iso-Salmijärvi, however, the cumulative inbreeding depression for unfertilized individuals was absent (–0.057), while fertilized plants had relatively high inbreeding depression (0.217).
DISCUSSION
Differences between populations in the expression of inbreeding
Our results support the prediction that the difference in the magnitude of inbreeding depression between inbred and outbred populations is strongest in early life stages. The expression of inbreeding depression in L. viscaria populations was strongest in germination, where the cross type had the most significant effect on plant performance. The expression of inbreeding depression in germination after the first generation of selfing was lower in the homozygous population Iso-Salmijärvi than in Kanavuori and Vaarunjyrkkä. This is in accordance with previous studies in which inbreeding depression was lower in more inbred populations, especially for early traits (reviewed by Husband and Schemske, 1996
). Inbreeding depression at early stages of development, such as germination, is suggested to be largely due to recessive alleles of lethal or highly deleterious effects (Husband and Schemske, 1996
). Thus, inbreeding depression for germination rate might have been effectively purged from the homozygous population Iso-Salmijärvi, and to some extent, even from the small population Kanavuori. Willis (1999)
found that in Mimulus guttatus <50% inbreeding depression in germination was caused by alleles with a large deleterious effect, while in other traits only 8–30% of inbreeding depression could have been due to strongly deleterious alleles. In M. guttatus, purging through artificial inbreeding over several generations acted most efficiently on germination (Willis, 1999
).
The strong expression of inbreeding depression in the earliest traits may, however, be partly due to genetic maternal effects. Usually, literature's discussion of maternal effects confounding estimation of inbreeding depression concerns non-genetic effects, i.e., whether mothers differ in the amount of resources they have available due to microhabitat differences. Due to our experimental design, the effect of environmental maternal factors is not of concern, but genetic effects are likely to occur. If differential maternal provisioning to inbred and outbred offspring occurs, the differences in inbreeding depression in early stages may be exaggerated or masked by maternal effects, rather than reflect true purging. The differential provisioning is likely to act when resources are limiting or when offspring are sufficiently heterogeneous to create resource sinks of different strengths (Bookman, 1984
; Lee and Bazzaz, 1986
). Thus, differences in inbreeding depression in germination percentage between populations can be affected by habitat quality and genetic differences between inbred and outbred progeny. However, experimental evidence for differential maternal provisioning to inbred and outbred offspring is currently scarce. While del Castillo (1998)
found some evidence in Phacelia dubia, Husband and Gurney (1998)
found no evidence for differential maternal provisioning for Epilobium angustifolium.
However, our results on seed production per capsule suggest that genetic maternal effects do exist to some extent in L. viscaria in seed production, because the selfed progeny were not as capable as outcrossed progeny in producing seeds when pollinated naturally by insects (open-pollinated plants). This genetic maternal effect can be caused by differences in pollination success or in number of ovules or by the physiological differences in seed production ability between self and outcross progeny. However, the difference between selfed and outcrossed progeny increased when selfed progeny were pollinated with self pollen and outcrossed progeny with outcross pollen, indicating that the genetic quality of the seed also affects seed production per capsule. However, the interpretations should be made cautiously, because due to limited resources, we lacked the controls for hand pollinations in which inbred individuals were crossed and outbred individuals selfed. Thus, the design cannot rule out completely if there is some artifact caused by the hand-pollination procedure.
Seed production was the only later life history trait in which a significant difference between selfed and outcrossed progeny was detected, but for all the other traits measured, the overall mean of outcrossed progeny was higher than that of selfed progeny. At later life history stages, the differences between populations in the expression of inbreeding were not related to the level of assumed history of inbreeding or to population. Although total cumulative inbreeding depression was highest for the large and probably most outbred population Vaarunjyrkkä, high levels of inbreeding were also found in Kanavuori and in fertilized plants from Iso-Salmijärvi. This result suggests that inbred populations can possess harmful alleles affecting the later life history stages, although purging may have lowered inbreeding depression for traits expressed earlier in the life cycle (Husband and Schemske, 1996
).
In their review, Byers and Waller (1999)
list possible explanations for the difficulty in detecting reduction of genetic load in more inbred lines: (1) lack of sensitivity in experimental design to detect changes in inbreeding depression, (2) studied populations are too similar in mating history or in population size, (3) long-lived species may be unlikely to shed their genetic load, and (4), most importantly, the extent to which purging occurs depends on the on the genetic background such as selective effects and the degree of dominance (or over dominance) and interactions among loci, such as epistasis. The experimental design of this study may not have been sufficient for quantification of inbreeding depression for traits expressed late in the life cycle, which are likely to be under polygenetic control. Further, the study populations might have been too similar in their mating history and size for us to effectively detect the differences. All populations originated from the periphery of the distribution area of L. viscaria in Finland were quite small and were able to produce seeds autogamously (Table 1). Thus, selfing may be quite common in the studied L. viscaria populations and differences among populations may not be large enough to allow detection of changes in inbreeding depression levels at later stages. Furthermore, L. viscaria is a perennial plant, which slows purging through inbreeding.
Although the idea of reduced expression of inbreeding depression in selfing populations has received support from several comparisons of species or populations (Holtsford and Ellstrand, 1990
; Dole and Ritland, 1993
; Parker et al., 1995
; Carr and Dudash, 1996
), ambiguous results have also been found (Latta and Ritland, 1994
; Johnston and Schoen, 1996
). Byers and Waller (1999)
concluded in their review that purging may be an inconsistent force within populations.
The absence of purging may also suggest that the inbreeding depression in L. viscaria may partly be due to overdominant alleles. Recent studies on the genetic basis of inbreeding depression, however, suggest that the majority of inbreeding depression is caused by mildly deleterious alleles with minor effects with intermediate degrees of dominance (Johnston and Schoen, 1996
; Willis, 1999
; Dudash and Fenster 2000
; Carr and Dudash, 2003
), which may be difficult or even impossible to purge through inbreeding. Thus, inbreeding populations may also express relatively strong inbreeding depression similar to the populations of L. viscaria studied here, and the genetic basis of inbreeding may still be due to dominant alleles. Slightly deleterious alleles may also become fixed in small populations and lower the viability of individuals.
Family level variation
In addition to population level differences in inbreeding depression, there were apparently differences between the family level responses to the pollination treatment, even though our experimental set-up was not specifically planned to focus on the variation in family level inbreeding depression. The family level variation in inbreeding depression is a common feature in plant populations (see a recent review by Kelly, 2005
). Holsinger (1988
, 1991
) and Uyenoyama and Waller (1991a
, b
) stress the importance of variation in inbreeding depression among families in mating system evolution. If inbreeding depression is due to partially recessive deleterious alleles, associations may develop such as alleles that promote selfing become associated with alleles related to fitness. This may allow selfing rate to increase despite the inbreeding depression it causes. Variance in family level responses to inbreeding in L. viscaria may thus suggest that there is potential for evolution toward higher selfing rates.
The effect of fertilization
In L. viscaria, inbreeding depression was not more pronounced in harsher conditions (here nutrient-low environment), as found in several other species (Dudash, 1990
; Schmitt and Ehrhardt, 1990
; Hauser and Loeschcke, 1996
; Daehler, 1999
). In at least one species, the opposite result has been found; in Schiedea lydgatei fertilization increased inbreeding depression (Norman et al., 1995
). In harsher environmental conditions, inbreeding may even be less pronounced if environmental noise obscures even appreciable fitness differences (Waller, 1984
; Mitchell-Olds and Waller, 1985
) or if harsh environments repress growth. Our environmental conditions (fertilization treatment) also affected the expression of inbreeding depression in L. viscaria; fertilization increased inbreeding depression in leaf number, the only life history trait for which significant interaction between fertilization and cross was observed. However, cumulative inbreeding depression was higher in unfertilized plants in populations Vaarunjyrkkä and Kanavuori. Instead, in Iso-Salmijärvi, the fertilized plants expressed relatively high cumulative inbreeding depression, while none was observed for unfertilized plants. Thus, no clear trend of fertilization either predominantly increasing or decreasing the expression of depression could be found. However, although the fertilization clearly affected the performance of the plants before planting to the field (seedling growth; Fig. 2A, flowering percentage in 1997; Fig. 3A) at later stages, the beneficial effect of fertilization was not evident in the field conditions later. This may be due to the fact that the common garden was already so fertile that the fertilized plants in the field did not acquire extra benefit, thus confounding the effects of environmental conditions on the expression of inbreeding at later stages. However, the establishment of the fertilized plants benefited as can be seen by their increased probability to flower in first flowering season (Fig. 3A).
These results indicate that predictions and generalizations on the magnitude of inbreeding depression in different environmental conditions should be made cautiously. Moreover, our results further strengthen the view that the expression of inbreeding depression should be measured over the whole life cycle and in variable environmental conditions. In conclusion, our results emphasize, together with other recent studies (e.g., Cheptou et al., 2000
; Henry et al., 2003
), that the environmental dependency of inbreeding depression needs to be included in theoretical models for the evolution of mating systems. Additionally, both empirical and theoretical work is needed to understand the reaction norms of inbreeding depression.
FOOTNOTES
1 The authors thank K. Holopainen and A. Lammi for help with the field work and in the greenhouse; V. Salonen, H. Siitari, L. Vertainen, H. Väre, J. Ågren, P. Mutikainen, and anonymous reviewers for comments on this manuscript; Laukaa Research and Elite Plant Station for facilities; and the Academy of Finland (to P.S.), the Ellen and Artturi Nyyssönen Foundation (to K.M.), and the Ella and Georg Ehnroot Foundation (to K.M.) for funding. This study is a part of the Finnish Biodiversity Programme (FIBRE 1997– 2002). 
4 Author for correspondence (e-mail: kaisa.mustajarvi{at}iki.fi
) present address: Penttilänkatu 12 c 14, FI-33820 Tampere, Finland; phone: + 358 50 5360488
LITERATURE CITED
Barrett S. C. H. D. Charlesworth 1991 Effects of a change in the level of inbreeding on the genetic load. Nature 352: 522-524[CrossRef][Medline]
Bookman S. S. 1984 Evidence for selective fruit abortion in Asclepias. Evolution 38: 72-86[CrossRef][ISI]
Byers D. L. D. M. Waller 1999 Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annual Review of Ecology and Systematics 30: 479-513
Carr D. E. M. R. Dudash 1996 Inbreeding depression in two species of Mimulus (Scophulariaceae) with contrasting mating systems. American Journal of Botany 83: 586-593[CrossRef][ISI]
Carr D. E. M. R. Dudash 2003 Recent approaches into the genetic basis of inbreeding depression in plants. Philosophical Transactions of the Royal Society of London, B, Biological Sciences 358: 1071-1084[CrossRef]
Carr D. E. M. D. Eubanks 2001 Inbreeding alters resistance to insect herbivory and host plant quality in Mimulus guttatus (Scrophulariaceae). Evolution 56: 22-30[ISI]
Charlesworth D. B. Charlesworth 1979 The evolutionary genetics of sexual systems in flowering plants. Proceedings of the Royal Society of London, B, Biological Sciences 205: 513-530
Charlesworth D. B. Charlesworth 1987 Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237-268[CrossRef][ISI]
Cheptou P.-O. E. Imbert J. Lepart J. Escarre 2000 Effects of competition on lifetime estimates of inbreeding depression in the outcrossing plant Crepis sancta (Asteraceae). Journal of Evolutionary Biology 13: 522-531[CrossRef][ISI]
Crnokrak P. D. A. Roff 1999 Inbreeding depression in the wild. Heredity 83: 260-270
Daehler C. C. 1999 Inbreeding depression in smoot cordgrass (Spartina alterniflora, Poaceae) invading San Francisco Bay. American Journal of Botany 86: 131-139
Del Castillo R. F. 1998 Fitness consequences of maternal and non maternal components of inbreeding in gynodioecious Phacelia dubia. Evolution 52: 44-60[CrossRef][ISI]
Dole J. K. Ritland 1993 Inbreeding depression in two Mimulus taxa measured by multigenerational changes in the inbreeding coefficient. Evolution 47: 361-373[CrossRef][ISI]
Dudash M. R. 1990 Relative fitness of selfed and outcrossed progeny in a self-compatible, protandrous species, Sabatia angularis L. (Gentianaceae): a comparison in three environments. Evolution 44: 1129-1139[CrossRef][ISI]
Dudash M. R. C. B. Fenster 2000 Inbreeding and outbreeding in fragmented populations. In A. G. Young and G. M. Clarke [eds.], Genetics, demography and viablity of fragmented populations, 35–53. Cambridge University Press, Cambridge, UK
Eckert C. G. S. C. H. Barrett 1994 Inbreeding depression in partially self-fertilising Decodon verticillatus (Lythraceae): population-genetic and experimental analyses. Evolution 48: 952-964[CrossRef][ISI]
Ellstrand N. C. 1992 Gene flow by pollen: implications for plant conservation genetics. Oikos 63: 77-86[CrossRef][ISI]
Goodwillie C. 2000 Inbreeding depression and mating systems in two species of Linanthus (Polemoniaceae). Heredity 84: 283-292
Hauser T. P. V. Loeschcke 1996 Drought stress and inbreeding depression in Lychnis flos-cuculi (Caryophyllaceae). Evolution 50: 1119-1126[CrossRef][ISI]
Henry P.-Y. R. Pradel P. Jarne 2003 Environment-dependent inbreeding depression in a hermaphroditic freshwater snail. Journal of Evolutionary Biology 16: 1211-1222[CrossRef][ISI][Medline]
Hoffmann A. A. P. A. Parsons 1991 Evolutionary genetics and environmental stress. Oxford University Press, Oxford, UK
Holsinger K. E. 1988 Inbreeding depression doesn't matter: the genetic basis of mating system evolution. Evolution 42: 1234-1244
Holsinger K. E. 1991 Inbreeding depression and the evolution of plant mating systems. Trends in Ecology and Evolution 6: 307-308[CrossRef]
Holstford T. P. N. C. Ellstrand 1990 Inbreeding effects in Clarkia tembloriensis (Onagraceae) populations with different natural outcrossing rates. Evolution 44: 2031-2046[CrossRef][ISI]
Hultén E. 1971 Atlas of the distribution of vascular plants in northwestern Europe. Generalstabens litografiska anstalts förlag, Stockholm, Sweden
Husband B. C. J. E. Gurney 1998 Offspring fitness and parental effects as a function of inbreeding in Epilobium angustifolium (Onagraceae). Heredity 80: 173-179[CrossRef][ISI]
Husband B. C. D. W. Schemske 1996 Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50: 54-70
Jennersten O. 1988 Pollination of Viscaria vulgaris (Caryophyllaceae): the contributions of diurnal and nocturnal insects to seed set and seed predation. Oikos 52: 319-327[CrossRef][ISI]
Johnston M. O. D. J. Schoen 1994 On the measurement of inbreeding depression. Evolution 48: 1735-1741[CrossRef][ISI]
Johnston M. O. J. Schoen 1995 Mutation rates and dominance levels of genes affecting total fitness in two angiosperm species. Science 267: 226-228
Johnston M. O. D. J. Schoen 1996 Correlated evolution of self-fertilization and inbreeding depression: an experimental study of nine populations of Amsinckia (Boraginaceae). Evolution 50: 1478[CrossRef][ISI]
Keller L. P. R. Grant B. R. Grant K. Petren 2002 Environmental conditions affect the magnitude of inbreeding depression in survival of Darwin's finches. Evolution 56: 1229-1239[CrossRef][ISI][Medline]
Kelly J. E. 2005 Family level inbreeding depression and the evolution of plant mating systems. New Phytologist 165: 55-62[CrossRef][ISI][Medline]
Koelewinj H. P. V. Koski O. Savolainen 1999 Magnitude and timing of inbreeding depression in scots pine (Pinus sylvestris L). Evolution 53: 758-768[CrossRef][ISI]
Kwak M. O. Jennersten 1986 The significance of pollination time and frequency, and purity of pollen loads for seed set in Rhinanthus angustifolius (Schrophulariaceae) and Viscaria vulgaris (Caryophyllaceae). Oecologia 70: 502-507[CrossRef][ISI]
Lammi A. P. Siikamäki K. Mustajärvi 1999 Genetic diversity, population size, and fitness in central and peripheral populations of a rare plant Lychnis viscaria. Conservation Biology 13: 1069-1078[CrossRef][ISI]
Lande R. D. W. Schemske 1985 The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39: 24-40[CrossRef][ISI]
Latta R. K. Ritland 1994 The relationship between inbreeding depression and prior inbreeding among populations of four Mimulus taxa. Evolution 48: 806-817[CrossRef][ISI]
Lee T. D. F. A. Bazzaz 1986 Maternal regulation of fecundity: non-random ovule abortion in Cassia fasciculata Michx. Oecologia 68: 459-465[CrossRef][ISI]
Lloyd D. G. 1979 Some reproductive factors affecting the selection of self-fertilisation in plants. American Naturalist 113: 67-79[CrossRef][ISI]
Mitchell-Olds T. D. M. Waller 1985 Relative performance of selfed and outcrossed progeny in Impatiens capensis. Evolution 39: 533-544[CrossRef][ISI]
Norman J. N. A. K. Sakai S. G. Weller T. E. Dawson 1995 Inbreeding depression in morphological and physiological traits of Schiedea lydgatei. Evolution 49: 297-306[CrossRef][ISI]
Olesen J. M. S. K. Jain 1994 Fragmented plant populations and their lost interactions. In V. Loescke and S. K. Jain [eds.], Conservation genetics, 417–426. Birkhäuser, Basel, Switzerland
Parker I. M. R. R. Nakamura D. W. Schemske 1995 Reproductive allocation and the fitness consequences of selfing in two sympatric species of Epilobium (Onagraceae) with contrasting mating systems. American Journal of Botany 82: 1007-1016[CrossRef][ISI]
Saccheri I. M. Kuussaari M. Kankare P. Vikman W. Fortelius I. Hanski 1998 Inbreeding and extinction in a butterfly metapopulation. Nature 392: 491-494[CrossRef]
Schemske D. W. 1983 Breeding systems and habitat effects on fitness components in three neotropical Costus (Zingiberaceae). Evolution 37: 523-539[CrossRef][ISI]
Schmitt J. D. W. Ehrhardt 1990 Enhancement of inbreeding depression by dominance and suppression in Impatiens capensis. Evolution 44: 269-278[CrossRef][ISI]
Siikamäki P. A. Lammi 1998 Fluctuating asymmetry in central and marginal populations of Lychnis viscaria in relation to genetic and environmental factors. Evolution 52: 1258-1292
Uyenoyama M. K. D. M. Waller 1991a Coevolution of self fertilization and inbreeding depression. I. Mutation-selection balance at one and two loci. Theoretical Population Biology 40: 14-46[CrossRef][ISI][Medline]
Uyenoyama M. K. D. M. Waller 1991b Coevolution of self fertilization and inbreeding depression. III. Homozygous lethal mutations at multiple loci. Theoretical Population Biology 40: 173-210[CrossRef][ISI][Medline]
Välivaara R. T. Raatikainen V. Saari P. Halinen M. Salminen M. Raatikainen 1991 Uhanalaiset kasvit Keski-Suomessa. The threatened vascular plants in central Finland. Julkaisu A2 1991. Keski-Suomen Liitto, Jyväskylä, Finland
Waller D. M. 1984 Differences in fitness between seedlings derived from cleistogamous and chasmogamous flowers in Impatiens capensis. Evolution 38: 427-440
Wendel J. F. N. F. Weeden 1989 Visualization and interpretation of plant isozymes. In D. E. Soltis and P. S. Soltis [eds.], Isozymes in plant biology, 5–45. Dioscorides, Portland, Oregon, USA
Willis J. H. 1999 The role of genes of large effect on inbreeding depression in Mimulus Guttatus. Evolution 53: 1678-1692[CrossRef][ISI]
Wilson G. B. J. Wright P. Lusby W. J. Whittington R. N. Humphries 1995 Biological flora of the British Isles, no. 188. Lychnis viscaria L. (Viscaria vulgaris Bernh). Journal of Ecology 83: 1039-1051[CrossRef]
Zar J. H. 1999 Biostatistical analysis. Prentice-Hall, Upper Saddle River, New Jersey, USA
This article has been cited by other articles:
|
|
 |
|
  C. GOODWILLIE and M. C. KNIGHT Inbreeding Depression and Mixed Mating in Leptosiphon jepsonii: A Comparison of Three Populations Ann. Bot., August 1, 2006; 98(2): 351 - 360. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||
= fertilized; • = unfertilized) and open-pollinated plants are presented. Positive index values indicate inbreeding depression and negative ones outbreeding depression. The asterisks indicate significant deviation from zero (two-way t test for family means of inbreeding depression)