Exercise 13.1

An Example of Escape from the Coevolutionary Arms Race in Snakes and Salamanders

(This exercise is based on Hanifin, C. T., E. D. Brodie Jr.,and E. D. Brodie III. 2008. Phenotypic mismatches reveal escape from arms-race coevolution. PLOS Biology 6: 471–482.)

(Note: The reference above links directly to the article on the journal’s website. In order to access the full text of the article, you may need to be on your institution’s network [or logged in remotely], so that you can use your institution’s access privileges.)


In Chapter 13 you read about many examples of coevolved species complexes. One of the examples was the predator-prey interaction between garter snakes in the genus Thamnophis and salamanders in the genus Taricha. In this paper by Hanifin et al., the authors describe a study of populations of these animals that live on the west coast of North America. The prey are newts (a kind of salamander) that have evolved a strong poison called tetrodotoxin (TTX) in their skin, presumably as a defense against predators. These newts are preyed upon by garter snakes that have evolved a physiological resistance to the newts’ skin toxin. The authors measured both the newts’ toxicity and the snakes’ resistance to this poison across the range where both species co-occur.

What they found is that while there is strong evidence of coevolution between these species in some areas of their range, the degree to which this is happening is not consistent across the range, and in some areas it doesn’t seem to be happening at all. Furthermore, they found that when one species outpaced the other in this evolutionary arms race, it was invariably the predator and never the prey.

The authors suggest that the pattern of phenotypic variation related to tetrodotoxin in this assemblage of predators and prey can be explained by three factors. Mismatched populations (populations where elevated values of a trait in one species are not matched by a corresponding elevation in the other species) are attributed to a lack of reciprocal selection resulting in a corresponding lack of coevolution. In areas with low values of snake TTX resistance, non-toxic newts tended to be present. In areas with intermediate values of snake TTX resistance, strong reciprocal selection leads to a coevolutionary arms race and thus elevated phenotypic expression of TTX-related traits in newts (matched populations). Finally they found that in some areas the newts were toxic, but the snakes were almost completely resistant to this defense, which the authors attributed to fixation of the genes for TTX resistance.



Question 1. Explain what the authors mean when they talk about “reciprocal selection.”


Question 2. Why do the authors refer to a breakdown of reciprocal selection in some areas of the geographic range that they studied?

Figure 1 Newt toxicity plotted against snakes’ resistance to toxins. The diagonal dashed line represents the toxin level at which the snake is reduced to 50% of its performance. Further details can be found in the figure caption from the paper (Hanifin et al. 2008.).


Question 3. Considering your answers to Questions 1 and 2, what can you infer about the pattern apparent in Figure 1? In your answer, discuss how you might arbitrarily divide the populations into three categories and talk about what those categories represent.


Question 4. Do you agree with the authors’ proposed explanation of why prey never escapes the arms race? Justify your answer.


Question 5. When reading scientific papers, and more importantly when writing them, one always needs to be cautious about avoiding the tunnel vision inherent in concentrating on smaller parts of a greater whole. It never hurts to step back and think about the bigger picture. With this in mind, can you come up with another feasible explanation of the pattern discussed above in Question 4?


Figure 2 Representative interaction gradients for localities across the range of mismatches. The solid line S-shaped curve is the generalized dose response curve relating whole-snake resistance in snakes to quantity of TTX in mg. The shaded regions correspond to the zones of mismatch evaluated based on phenotypic distributions of predator and prey phenotypes. From each zone (no localities were observed occupying the orange region) interaction gradients are shown for two example localities varying in newt toxicity. (Caption modified from Hanifin et al. 2008; further detail should be sought from the original figure caption in the paper.)

Question 6. Refer to Figure 2 above. No populations were observed that would be plotted in the orange area of the central graph, but if there were a hypothetical population in that region, what would the interaction gradient graph (similar to the 4 smaller graphs in the figure) look like? Specifically, propose a reasonable guess as to the value of % performance (snake) and a value of the slope (β).