Selection retained in the landscape
Polaskia chichipe is a branched, tree-like cactus whose fruits are used by residents of the Tehuacán Valley. When a field is prepared, some desirable plants are left in place. This management is called in situ selection because the selected individuals continue growing in their existing habitat. Other plants are introduced into gardens, where they are cultivated. Scrub, fields and gardens therefore create different selective environments within one region. Studying their reproduction allows investigation of traits associated with early domestication, the alteration of inherited population characteristics through human use and selection. [Otero-Arnaiz et al. 2003]
Alejandro Otero-Arnaiz and colleagues compared nine populations near San Luis Atolotitlán, Caltepec and Los Reyes Metzontla in Puebla. They represented three management types, wild populations, plants retained in fields and plants cultivated in gardens. The researchers assessed flowering season, flower opening duration, fruit production following controlled pollination and seed germination. Individual experiments used their own samples, so nine populations describes the scope of the overall project rather than the sample size of every procedure. This design combines field observations with experiments separating the effects of self pollen, pollen from other plants and unrestricted animal visitation. [Otero-Arnaiz et al. 2003]
Flowering season changes opportunities for contact
Flowering began in January in all three groups, but its subsequent course differed. Wild populations finished flowering in March, field-retained populations in April and cultivated populations in May. The respective peaks occurred in February, March and April. Cultivated plants therefore had an extended reproductive season. Part of this period overlapped with flowering in wild populations, preserving opportunities for pollen exchange. Later flowers in gardens operated with a different set of available partners. Variation in flowering dates can consequently change the frequency of contacts among populations even when the distances separating plants remain the same. [Otero-Arnaiz et al. 2003]
The availability of an individual flower also changed. In winter it remained open for approximately 28 hours on average, compared with about 13 hours in spring. Longer opening included successive periods of day and night, whereas shorter opening narrowed the time available for animal contact. The observed rhythm must therefore be associated with the season of measurement. One value for flower duration would fail to describe both situations. This difference helps explain why the researchers conducted separate winter and spring experiments and compared daytime with nighttime access. Opportunities to visit a flower depend simultaneously on season and the course of its brief functional period. [Otero-Arnaiz et al. 2003]
Self-compatibility measured experimentally
Self-compatibility is the capacity of a flower to produce offspring after receiving pollen from the same plant. The researchers protected flowers from animal visits and compared autonomous self-pollination with manual application of self pollen. Following autonomous self-pollination, fruits developed from approximately 12% of flowers in wild populations, 27% in field-retained populations and 22% in cultivated populations. After manual self-pollination, the respective values were approximately 12%, 24% and 27%. These data indicate partial self-compatibility and differences among management types. The percentages concern flowers setting fruits under a specified treatment and experimental conditions. [Otero-Arnaiz et al. 2003]
Fruits resulting from self-pollination contained fewer seeds than fruits produced after pollination with another plant's pollen or unrestricted pollinator visits. Some offspring were nevertheless capable of further development, as germination trials confirmed. Assessing the reproductive system requires both levels of outcome, fruit production and the number and germination capacity of its seeds. Fruit presence alone describes only the first level. In this system, the capacity to use self pollen coexisted with more effective reproduction involving other plants. It provided an additional reproductive route where a suitable partner or pollinator might be less available. [Otero-Arnaiz et al. 2003]
Pollinators during day and night
Bees, hummingbirds and nocturnal hawkmoths were observed at flowers. This visitor list was supplemented by experiments excluding access during selected periods. In the winter trial, whole-day access resulted in 40% fruit set, while nighttime-only access produced 15%. Unrestricted access in that comparison gave 60%. In the spring trial, daytime and unrestricted access each produced 100% fruit set, whereas nighttime-only access produced 0%. Each treatment in these comparisons included 20 flowers. The results confirm the predominant daytime contribution and show that the contribution of nighttime access varied with season. [Otero-Arnaiz et al. 2003]
In separate reproductive-system trials, freely accessible flowers had fruit set exceeding 80% in every population type. Manual cross-pollination produced lower values, approximately 38–49% depending on the group. Comparing these results requires retaining the distinction between procedures. Bagging and handling a flower changes experimental conditions, while unrestricted visitors can deliver pollen repeatedly. Natural fruit set indicates high effectiveness of the overall pollination system during measurement. Observations of hummingbirds and the presence of pollen document their contact with flowers, whereas the daytime experiment includes all animals admitted during that part of the day. [Otero-Arnaiz et al. 2003]
Germination varies with seed origin
The next stage assessed seeds from the three population groups under identical conditions. The chamber alternated 27°C and 12°C, combined with 14 hours of light and 10 hours of darkness. After 34 days, 60.8% of seeds from wild plants had germinated, compared with 77.2% from field-retained plants and 76.3% from cultivated plants. Germination was also faster in the two managed groups. The common procedure allows differences to be associated with the origin of material within the study's scope. The outcome concerns germination achieved, while subsequent offspring survival would require a separate comparison. [Otero-Arnaiz et al. 2003]
Differences in germination rate contribute to the picture of populations under management, alongside phenology and partial self-compatibility. Their ecological significance depends on the conditions in which a seedling begins growth. In this trial, all seeds received the same temperature, light and access to water. The researchers could therefore compare responses of seeds originating from different populations. Such an experiment clarifies relationships while leaving the success of young plants in a variable habitat as a separate question. The association between faster germination and domestication is interpreted together with other plant traits and the way people select plants. [Otero-Arnaiz et al. 2003]
Tracing paternity with DNA
A study published in 2005 shifted attention from fertilization capacity to the actual parentage of offspring. Adult plants in nine populations were genotyped using five microsatellite markers. These are short, repeated DNA regions whose variants help distinguish individuals. Offspring came from 63 maternal plants in seven populations. Comparing variants in an offspring, its known mother and candidate fathers allowed pollen to be assigned to plants. The analysis covered local distances between mating partners and cases in which the father came from outside the maternal population. [Otero-Arnaiz et al. 2005]
A genotype match has a defined resolving power. The authors reported an exclusion probability of 0.96 for the marker set used, meaning a high capacity to exclude an incorrect father. Some candidates could nevertheless share compatible variants, and an unsampled individual outside the population could happen to resemble a local plant genetically. The proportion of external pollen was consequently treated as a minimum estimate. This property explains why assigned fathers and pollen immigration rates require joint interpretation. Identified fertilizations document gene movement, while every flight an animal makes between flowers does not necessarily leave offspring available for analysis. [Otero-Arnaiz et al. 2005]
Local mating and movement among populations
Most identified local matings occurred between plants less than 40 m apart. The mean distance of local pollen movement was approximately 42 m. Including identified contacts among populations increased the mean to about 113 m. At the same time, at least approximately 27% of offspring involved pollen from outside the maternal population. These findings describe different parts of the same process. Numerous short movements concentrate reproduction in the neighbourhood, while less frequent distant contacts connect separate plant groups. Wild cacti and garden plants can consequently participate in a shared flow of genes. [Otero-Arnaiz et al. 2005]
Realized selfing across the sample was approximately 9%. Estimates for cultivated, field-retained and wild populations differed numerically, but those differences were statistically nonsignificant. Comparison with the earlier experiment separates the capacity for self-fertilization from how often it occurs under natural pollen access. A plant can accept its own pollen while producing most offspring with neighbouring plants. Greater self-compatibility observed in some experiments on managed populations therefore does not automatically imply that selfing predominates in their actual reproduction. [Otero-Arnaiz et al. 2005] [Otero-Arnaiz et al. 2003]
Inherited change with connectivity maintained
Gene flow can distribute inherited variants among populations and limit differentiation. Repeated human selection of desirable plants operates at the same time in a managed landscape. The authors interpreted the persistence of differentiation in managed populations as an outcome of selection occurring alongside continuing genetic exchange. Unequal contributions of adult plants to offspring paternity also matter. Flowering individuals do not all transmit genes at similar frequencies. Partner distribution, flowering dates and effective pollen transfer can therefore influence which variants enter the next generation at a particular location. [Otero-Arnaiz et al. 2005]
The biology of Polaskia chichipe describes domestication operating within an open population system. Retaining plants in fields maintains them where they are used, cultivation changes the membership of local groups, and pollination connects those plants with surrounding scrub. Flower experiments reveal differences in reproductive capacity, while DNA markers establish the parentage of offspring actually produced. Conserving genetic resources therefore involves the whole arrangement of sites and reproductive contacts. In the region studied, traditional management coexisted with pollen movement among wild and managed populations of the same species. [Otero-Arnaiz et al. 2003] [Otero-Arnaiz et al. 2005]
Evolution of Polaskia chichipe (Cactaceae) under domestication in the Tehuacán Valley, central Mexico: Reproductive biology
Otero-Arnaiz A., Casas A., Bartolo C., Pérez-Negrón E., Valiente-Banuet A., Evolution of Polaskia chichipe (Cactaceae) under domestication in the Tehuacán Valley, central Mexico: Reproductive biology, American Journal of Botany, 2003, 90(4):593-602, doi: 10.3732/ajb.90.4.593
Comparison of wild, field-retained and cultivated Polaskia chichipe populations in the Tehuacán Valley covers phenology, flower opening, controlled pollination and seed germination. Some flowers produced fruits after self-pollination, with results varying among management types. Naturally pollinated flowers had high fruit set, while experiments allowing daytime or nighttime access indicated a predominant daytime contribution. Seeds from managed populations germinated faster and in higher proportions under the laboratory conditions used.
Direct and indirect estimates of gene flow among wild and managed populations of Polaskia chichipe, an endemic columnar cactus in Central Mexico
Otero-Arnaiz A., Casas A., Hamrick J. L., Direct and indirect estimates of gene flow among wild and managed populations of Polaskia chichipe, an endemic columnar cactus in Central Mexico, Molecular Ecology, 2005, 14(14):4313-4322, doi: 10.1111/j.1365-294X.2005.02762.x
Analysis of five microsatellite markers combines genetic structure in nine Polaskia chichipe populations with offspring paternity assignment. Most identified matings occurred over short distances, while some pollen originated outside the maternal population. Realized selfing was limited and did not differ significantly among management types. Gene flow connected wild plants, plants retained in fields and plants cultivated in gardens.