Stones as shelter for Mammillaria seedlings

The microhabitat forms beside the plant itself

For a seedling a few millimetres in size, conditions at the ground surface matter. These include temperature, water availability, light and contact with herbivores. A neighbouring shrub can change all these factors. A beneficial effect of a neighbour on the establishment of a young plant is termed facilitation. An inanimate element, particularly a stone, may serve a similar function. Its presence changes the geometry of illumination and conditions immediately beside the substrate. In ecological terms, such an object shelters a seedling, with its importance assessed through survival and development. [Peters et al. 2008, pp. 593–595]

Peters, Martorell and Ezcurra studied this phenomenon in Mammillaria in the Tehuacán–Cuicatlán region of Mexico. They combined observations of cactus distribution with a field experiment on Mammillaria pectinifera. The first part indicates where plants occur more often than expected from the availability of a given site. The second tests whether placing seedlings in different microhabitats actually changes their survival. Combining the two approaches allows adult plant distribution to be linked to the fate of the young generation. [Peters et al. 2008, pp. 595–596, 599]

How the cactus association with stones was identified

The 2008 paper distinguished eight taxa. Ten sites were studied for most, two for material named M. lanata, and one non-cliff site for M. supertexta. On randomly placed 50 m transects, plants were counted beneath a shrub or tree, beside a stone and on the remaining substrate. Transect width was 1 or 4 m, depending on the group studied. Association with a stone was defined as a distance of less than 2 cm from a stone embedded in the ground and at least 20 cm in diameter. [Peters et al. 2008, p. 595, Table 1, p. 596]

At each site, an additional 60 points were selected at random along three transects. They showed what proportion of available terrain belonged to each microhabitat category. Cactus abundance in a category was compared with the abundance expected under random distribution. This allows frequent occurrence beside stones to be distinguished from a situation in which stones simply cover most of the terrain. The tests concerned spatial relationships in the populations observed. [Peters et al. 2008, pp. 595–596]

The current Plants of the World Online treatment combines M. lanata with M. supertexta. The table below retains the researchers' two field groups separately, with the accepted name and historical name in parentheses. Their different results are part of the experiment described. Combining taxonomic names requires data provenance and site numbers to be retained because it does not remove differences between the populations observed. [2-WWW] [Peters et al. 2008, record POWO and Table 1, p. 596]

Different Mammillaria occupy different places

A positive association with stones was detected in four of the eight historically distinguished groups. One, M. kraehenbuehlii, was positively associated with shrubs. For M. pectinifera, the association with stones was statistically significant, with p = 0,0010. Its distribution relative to shrubs gave p = 0,7919. The results show variation in how members of the genus occupy terrain. [Peters et al. 2008, p. 596, Table 1]

Stone size in the definition matters for the result. The authors observed many M. pectinifera beside stones smaller than 20 cm, which the spatial test did not include in this category. In M. dixanthocentron, some close associations resulted from growth directly on rocks. For this taxon, the positive result may therefore also describe properties of the rocky substrate and the way the plant is anchored. The survival experiment was conducted separately for M. pectinifera. [Peters et al. 2008, pp. 596, 599]

The experiment with Mammillaria pectinifera seedlings

Seedlings were planted on exposed ground, on ground under artificial shade from a nylon cover, and beside a stone at least 30 cm in diameter. The larger stone size allowed a group of seedlings to fit within its shade. Each treatment was compared with and without protection from herbivores. The protection combined mesh with insecticide. The authors distinguished three-day-old and six-month-old plants. The design comprised 36 experimental units, with three replicates of each combination of conditions and 45 of the youngest or 35 older seedlings per unit. [Peters et al. 2008, p. 596]

Survival was analysed with a Weibull model, which allows the risk of death to change over time and accounts for plants still alive at the final reading. The fitted shape parameter was 0,87 and differed from 1, the value corresponding to constant risk. The result indicates greater risk at the beginning of observation and a decline in risk among plants that survived. The curves also show differences between seedling ages, microhabitat types and the use of protection from herbivores. [Peters et al. 2008, pp. 596–597, Fig. 1 and Table 2]

Seedling age changed the effectiveness of shelter

All plants planted as three-day-old seedlings eventually died. The comparison of their initial persistence beside a stone and under artificial shade had p = 0,2276. In this group, no advantage of one of the two shelters was established. In six-month-old seedlings, a stone favoured longer survival. The most favourable treatment combined older seedlings, a stone and exclusion of herbivores. In this treatment, modelled life expectancy exceeded one year. The difference between age groups was the largest component of the analysis presented. [Peters et al. 2008, pp. 596–598, Fig. 1–2 and Table 2]

The authors link this change to development of the plant surface. A newly germinated seedling is small and spineless, while a six-month-old plant already has areoles and stem cover. This structural change may increase tolerance of radiation. The older plant can then benefit from substrate properties beside a stone while retaining access to more light. This is a biological interpretation of the survival results. Individual components of the water balance, temperature and gas exchange were not measured in this experiment. [Peters et al. 2008, p. 599]

Height of surviving plants and significance for conservation

In the final comparison, surviving plants beside stones had a mean height of 5,84 ± 0,14 mm, and those under an artificial cover 5,30 ± 0,18 mm. The mean and standard error were reported. The difference was significant, p = 0,026. Only the groups of older seedlings beside stones and under cover provided enough plants for size analysis. These values describe the height of plants that survived and need to be read together with the earlier mortality. [Peters et al. 2008, p. 598]

M. pectinifera has a stem partly buried in the soil, so its exposed photosynthetic surface is small. The authors propose that strong, prolonged shade may limit the supply of photosynthetically active radiation. Beside a stone, water and thermal conditions may be milder, and the extent of shading changes with the time of day, position and plant size. This explanation links cactus structure with the geometry of its surroundings. The direct results concern increased survival of older seedlings and their final height. [Peters et al. 2008, pp. 599–600]

In reintroducing this species, microhabitat selection and plant stage are therefore part of the conditions for success. Producing seedlings in a greenhouse begins the subsequent stage of field establishment. The study indicates the usefulness of stones and protection of young plants from herbivores under the conditions employed. The experiment concerns one species, and the distribution table shows differences among the other Mammillaria. Selecting a site for another species requires consideration of its own ecology and population observations. [Peters et al. 2008, pp. 598–600]