Spatial mosaic evolution of snail defensive traits

Abstract

Background: Recent models suggest that escalating reciprocal selection among antagonistically interacting species is predicted to occur in areas of higher resource productivity. In a putatively coevolved interaction between a freshwater snail (Mexipyrgus churinceanus) and a molluscivorous cichlid (Herichthys minckleyi), we examined three components of this interaction: 1) spatial variation in two putative defensive traits, crushing resistance and shell pigmentation; 2) whether abiotic variables or frequency of molariform cichlids are associated with spatial patterns of crushing resistance and shell pigmentation and 3) whether variation in primary productivity accounted for small-scale variation in these defensive traits.

Results: Using spatial autocorrelation to account for genetic and geographic divergence among populations, we found no autocorrelation among populations at small geographic and genetic distances for the two defensive traits. There was also no correlation between abiotic variables (temperature and conductivity) and snail defensive traits. However, crushing resistance and frequency of pigmented shells were negatively correlated with molariform frequency. Crushing resistance and levels of pigmentation were significantly higher in habitats dominated by aquatic macrophytes, and both traits are phenotypically correlated.

Conclusion: Crushing resistance and pigmentation of M. churinceanus exhibit striking variation at small spatial scales often associated with differences in primary productivity, substrate coloration and the frequency of molariform cichlids. These local geographic differences may result from among-habitat variation in how resource productivity interacts to promote escalation in prey defenses.

Figure 1

Collecting sites of Mexipyrgus churinceanus from spring-fed habitats in the Cuatro Ciénegas valley located in the center of the Chihuahuan desert of northeastern Mexico. These pools and streams are arrayed around the Sierra de San Marcos that bisects the center of the valley. Population numbers are shown on the map and referred to in Table 1.

Figure 2

Representative images (A-C) of the hydrobiid snail, Mexipyrgus churinceanus, collected from nearby populations to illustrate small scale variation in size and shell pigmentation. Herichthys minckleyi papilliforms (right figure D) exhibit gill arches modified into more gracile pharyngeal jaws with small muscles and pointed teeth that are ineffective at crushing snails while H. minckleyi molariforms (left figure D) have robust muscles and enlarged crushing teeth on their pharyngeal jaws that seem clearly modified to crush snails. Figure E portrays heterogeneity in habitats with Nymphaea (top) and without Nymphaea (bottom).

Figure 3

A) Average size-adjusted crushing resistance (95% Confidence Intervals) for M. churinceanus populations. B) Frequency of banded snails for M. churinceanus populations. Populations along the Y-axis are arranged on a transect from southwestern populations to the northern Rio Mesquite down to the southeastern lobe.

Figure 4

A and B Relationship between Moran's I and cytochrome b pairwise distance classes (correlograms) for two Mexipyrgus traits: size-adjusted crushing resistance (triangles and dashed line) and frequency of pigmented shells (closed circles and solid line).

Figure 5

Relationship between frequency of molariform cichlids and both adjusted marginal mean crushing resistance (triangles and dashed line) and frequency of pigmented shells (circles and solid line).