Lakeside daisy (Hymenoxys herbacea) COSEWIC assessment and update update status report: chapter 6

• Previous page
• Table of contents
• Next page

Biology

General

Hymenoxys herbacea is a herbaceous perennial (Figure 1). It blooms from early May to early July, producing a single yellow inflorescence that is insect pollinated. Seeds are dispersed via gravity and wind vectors approximately three weeks after the inflorescence has finished blooming (De Mauro, 1993). There is no period of seed dormancy and new seedlings are produced late in the summer, during periods when the soil is moist (De Mauro, 1993). Flower buds are produced late in the summer and, as such, overwinter.

Studies of the life history of H. herbacea usually classify plants into one of 5 demographic stages (J. Windus, pers. comm.). These include:

1. seed - small, 5-angled, top-shaped, hairy achenes;
2. plantlet - a tiny shoot with 2 leaves and 1 narrow center leaf;
3. Juvenile 1 - single rosette, 4-6 leaves, less than 2.5 cm tall;
4. Juvenile 2 - more than 6 leaves, more than 2.5 cm tall;
5. Adult reproductive – rosette with flowering scape.

Reproduction

Modes of Reproduction

Hymenoxys herbacea plants are capable of reproducing both sexually, via a single capitulum, and asexually, via rhizomatous growth and/or branching of the woody caudex. The importance of these two modes of reproduction may vary among years and locations. However, in two populations of H. herbacea on the Bruce Peninsula, 23% of the rosettes reproduced asexually and between 12% and 24% of the population reproduced sexually over a one year period (Campbell, 2001).

Sexual Reproduction

Flowering plants produce an inflorescence (capitulum) that consists of many individual flowers (or florets) packed densely together. The number of florets per inflorescence averaged 87.2 (SE = 1.77) in 1999 and ranged from 38 to 150 (Campbell, 2001). Furthermore, coastal populations tended to have significantly more florets than those inland (t-test: t = 2.849, df= 10, p = 0.017).

Hymenoxys herbacea flowers are self-incompatible (De Mauro, 1993). Self-incompatibility is a genetically controlled mechanism whereby pollen with the same alleles as the pollen recipient plant, at the self-incompatibility gene, are recognized and rejected (de Nettancourt, 1977; Mulcahy and Mulcahy,1985). As a result, fertilization and subsequent seed production occurs only after pollinations between genetically distinct individuals. At the population level, sustained seed production requires at least 4 self-incompatibility alleles to be represented among the residents. Self-incompatibility has been confirmed through pollination experiments in 13 populations on the Bruce Peninsula and 6 Manitoulin Island populations (Campbell, 2001). Seed set averaged 0.2% in self-pollinations. This was confirmed by genetic analyses in two populations which show that 86% of all offspring were the result of cross-fertilizations (14% from self-fertilization). While mate diversity was negatively correlated with population size, in no population was seed production limited by a lack of diversity at the self-incompatibility gene (Campbell, 2001).

Seed Production

Only a subset of the available ovules in H. herbaceaever mature into seeds. Across the range, including U.S. populations, seed developed in 43.5% (range = 27.5 to 66.2%) of the available ovules (De Mauro, 1993; Campbell, 2001). Based on estimates of seed set and the number of florets per inflorescence, the mean number of seeds produced per inflorescence in 1999 was 42.6 and ranged from 23.8 to 59 (Campbell, 2001). Mean seed set of Bruce Peninsula populations was similar to that of populations on Manitoulin Island.

Pollen Limitation

Seed production is possibly limited by a number of ecological and genetic factors, including: 1) resource availability, 2) genetic sterility, or 3) pollen limitation (i.e. not enough compatible pollen deposited on the stigmas). The primary concern in past studies of H. herbacea has been pollen limitation (De Mauro, 1993; Moran-Palma and Snow, 1997; Campbell, 2001). Measured as the proportional increase in seed set due to the addition of pollen to open-pollinated florets, pollen limitation was negligible when averaged across all 12 populations examined (mean = 0.08). In other words, overall, adding more pollen had no effect on seed set. However, pollen limitation did vary from 0 to 0.54 among populations, on a scale from 0 to 1, and seed set in supplemental pollinations was significantly higher in one of the populations (Appendix 1, population CPL). Although variation in seed set among populations cannot be fully accounted for by pollen limitation there appears to be some potential for it to occur in H. herbacea due to its mating system. Pollen limitation in H. herbacea is extremely low compared to mean pollen limitation calculated in a survey of other angiosperm species (0.40) (Larson and Barrett, 2000). In fact, H. herbacea had unusually low pollen limitation for a self-incompatible plant, which in general have higher pollen limitation (mean pollen limitation = 0.59 (± 0.04)) than self-compatible plants (mean pollen limitation = 0.31 (± 0.03)). Clearly, H. herbacea is unusual in its ability to acquire sufficient compatible pollen under its ecological conditions.

Pollinator Observations and Visitation Rates

The insect visitors of H. herbacea are diverse, a common feature of the insect visitors of many plants that flower in early spring (Thein et al., 1983; Godley and Smith, 1981). In a recent study of 13 populations on the Bruce Peninsula a total of 41 taxa, from eight families (Hymenoptera, Diptera, Lepidoptera, Neuroptera, Homoptera, Hemiptera, Coleoptera, Orthoptera; see Table 1) were observed on H. herbaceaflowers (Campbell, 2001); however, some of these were probably not pollinators. The number of taxa observed per 30-minute observation period averaged 2.68 and ranged from 0.86 to 5.17. However, this value was not correlated with either geographic isolation or population size (Campbell, 2001). In addition, the diversity of insects visiting each population ranged widely among populations and was negatively correlated with geographic distance to the nearest population. However, variation in insect diversity could not be linked to differences in seed set among populations (Campbell, 2001).

Each plant receives an average of 0.66 insect visits (SE = 0.24) per 30 minute observation period (Campbell, 2001). Plants in small populations tend to receive more visits than those in large populations. However, it is likely that not all insect visitors are effective pollen vectors. Studies of H. herbaceasuggest that bees (Apidae, Xylocopidae and Halictidae) are particularly important for pollination (De Mauro, 1993), although flies were much more prevalent flower visitors (Campbell, 2001). In 2001, bee visitation averaged 0.08 (SE=0.04) visits/plant/30-minute observation period, and three populations did not receive any visits from bees during 57 hours of observation (populations BC, FW, HL). The importance of bees is highlighted by the fact that the degree to which plants are pollen limited increases as bee visitation decreases. In general, pollinator visitation is susceptible to the vagaries of the environment, including temperature, wind, and precipitation and visitation by insects to H. herbacea is no exception. In years with more extreme weather conditions, especially those not conducive to pollen dehiscence, pollinator activity may affect pollen limitation more drastically than in years with good weather conditions for insect activity (Campbell, 2001).

Mate Limitation

Reproduction in self-incompatible plants may be limited by insufficient diversity of mating types in a population. The diversity of mating types (mate diversity), measured as the proportion of crosses that are genetically compatible within a population, has been measured in 12 Canadian H. herbaceapopulations (Campbell, 2001). Roughly speaking, between 17 and 58% of all pollinations were compatible in populations on the Bruce Peninsula. These values are similar or lower than the proportion of compatible crosses in a study of an H. herbaceapopulation in the U.S. by De Mauro (1993), where 58% of the within-population crosses were compatible. Mate diversity was strongly correlated with population size: as population size decreases, mating type diversity also drops (Campbell, 2001). However, mate diversity was apparently high enough in most populations as to not affect seed production (Campbell, 2001).

In summary, Canadian populations, on average, have enough pollinator activity and mate diversity to ensure that seed set is not pollen limited (Campbell, 2001). However, population size plays an important role in pollen limitation via mate diversity. Natural area managers must be aware that should the size of these populations of H. herbacea become smaller, pollen limitation could threaten their persistence.

Recruitment Rates

In a recent demographic study of two populations, recruitment over a one-year period occurred at a rate of 0.65 recruits per existing rosette. Of these recruits, 94% were produced via asexual recruitment and 6% were derived from sexual reproduction (Campbell, 2001).

Growth and Survival

Size Structure

Existing populations are composed of rosettes of different size and stage of development: Juvenile 1 (4-6 leaves/rosette); Juvenile 2 (>6 leaves/rosette) and Reproductive Adult (rosette with inflorescence). In a demographic study of two populations in 1999 and 2000, Juvenile 2 plants were most frequent (54 to 67% of all individuals), followed by Reproductive Adults (24 to 30%) and Juvenile 1 plants (8.8 to 15.9%) (Table 2) (Campbell, 2001).

Juvenile 1 individuals comprise a single rosette with 4-6 leaves; Juvenile 2 individuals have more than 6 leaves, and Adult individuals are reproductive. The first value in each cell represents the number of individuals of a particular stage in 1999 that occurred in a specific stage in 2000. The second value (in brackets) represents the proportion of individuals from 1999 that were observed in 2000. Proportions in each column sum to one.

Stage Transitions

Growth of individuals has been assessed by monitoring the changes in developmental stage over a one-year interval. From 1999 to 2000, most (53.4%) Juvenile 1 rosettes remained as Juvenile 1 plants. Of the remainder, 33% grew to become Juvenile 2 plants and 4% became Reproductive Adults. Most Juvenile 2 individuals (64.2%) remained as Juvenile 2 plants. However, 11% of Juvenile 2 plants reverted to Juvenile 1 status and 20% grew to become Reproductive Adults. Juvenile 2 individuals were the most likely of all stages to become Reproductive Adults. Over the same one-year interval, most Reproductive Adults (71.4%) reverted to the Juvenile 2 stage; only 6.8% remained as Reproductive Adults (Campbell, 2001).

Survival

Survival is generally high for H. herbacea plants from one year to the next. In one demographic study (Campbell, 2001), fewer than 5% of rosettes died. While survival was high for all rosettes, it tended to be higher in Juvenile 1 plants (99%) than Juvenile 2 (96.6%) and Reproductive Adult (97.8%) plants.

Population Growth Rate

Using the above information on growth, survival and reproduction, population growth rates (l= N_t+1 / N_t) have been estimated for two populations of H. herbacea. The growth rates were 0.486 for population HL and 0.903 for population CPL indicating that both populations were declining in size (l= 1, stable population size). It is not clear how representative these estimates are for H. herbaceapopulations in general. Both of these populations are in high-use areas; population HL is on the Bruce Trail and population CPL is in an area that is relatively popular with climbing or scrambling enthusiasts.

Generation Length

Generation length (L), defined as the mean age at which new plants produce offspring (asexual or sexual) (Yonezawa, 1997), averaged 16 and ranged from 10.78 to 21.08 years (Campbell, 2001).

Movements/Dispersal

The achenes are dispersed by gravity or wind approximately four to six weeks after fertilization (DeMauro, 1993). Although dispersal distance is unknown, seedlings are most dense within one meter of adult plants (De Mauro, 1993).

Migration between populations has been measured for H. herbacea by examining the distribution of genetic diversity within and among populations. Based on a genetic analysis of 12 populations from the Bruce Peninsula, Ontario, migration (Nm) averaged 0.56 migrants per generation (Campbell, 2001). This is a low level of migration compared to many plants of similar life history, and could lead to significant genetic differentiation among populations.

Nutrition and Interspecific Interactions

There are two interspecific interactions that are of important consequence for H. herbacea: herbivory and human trampling.

Herbivory

Herbivory has been observed on the peduncles, florets, receptacle and achenes by insects, white-tailed deer, seed-eating birds and eastern cottontail rabbits. In some cases herbivory has entirely prevented seed production in affected plants (De Mauro, 1993; Campbell, 2001). The amount of herbivory likely varies among populations, regions (Manitoulin Island versus Bruce Peninsula) and years depending on the size of herbivore populations and availability of other foodstuffs. In all 7 Manitoulin Island populations visited in 1999, there was severe damage imposed by a seed-eating larva, which often reduced seed production to zero. The same damage, however, was evident in only 1 of 12 populations surveyed the same year on the Bruce Peninsula, Ontario. Immature grasshoppers were more often observed on the Bruce Peninsula plants and were virtually non-existent on Manitoulin Island. Herbivory by rabbits was more commonly seen in inland populations while herbivory by seed-eating birds was more commonly seen in lakeshore habitats. Herbivory by rabbits, birds and deer however was much less severe than the damage imposed by larvae or grasshoppers (Campbell, 2001). The intensity of herbivory and the impact on population growth rates has not been measured quantitatively.

Trampling

Eight of the 13 populations on the Bruce Peninsula (Cypress Lake and Halfway Log Dump of Bruce Peninsula National Park) are on hiking trails and popular scrambling areas. Two populations (SC and CPL), in particular, were seen to decline in numbers over the two years of observation and, although untested, it seems likely that this is due to damage in part from human traffic. Other populations (i.e., HL or LC), however, were seemingly unaffected by human traffic. Seven of the populations on Manitoulin Island are also in heavily traveled areas; however their population sizes have not been monitored. Many populations exist on or near roads and hiking trails. It is uncertain whether the existence and maintenance of the roads affects the survival of the populations positively (maintaining open spaces) or negatively (compaction of soil and damaging the plants) (Campbell, 2001).

Genetic Diversity

The genetic diversity of H. herbacea has been measured using enzyme electrophoresis. All of the 13 populations sampled on the Bruce Peninsula had variation in at least one locus, with an average of 1.33 alleles per locus (Campbell, 2001). The percentage of polymorphic loci ranged from 11.11% to 44.44%, with a population average of 30.77%.

The magnitude of genetic diversity within H. herbacea populations was similar to that in other perennial, endemic, outcrossing, animal-pollinated plants (Hamrick, 1990). As H. herbacea is a self-incompatible plant and hence obligate outcrosser, populations are predicted to be relatively undifferentiated with respect to genetic variation.

Measures are based on an a survey of allozymes using cellulose acetate electrophoresis. A minimum of 15 individuals from each population were screened at 9 enzyme loci. P = Percentage of polymorphic loci; A = Mean number of alleles / locus; AP = Mean number of polymorphic alleles per locus; HN = Nei’s measure of expected heterozygosity (standard error).