Measurable Differences May Have Obscure Causes: Genetic Drift and the Founder Effect

INTRODUCTION

Consider if you will the imaginary island of Serpentia. More than an island, actually—Serpentia is the central island of a small tropical archipelago, and is surrounded by a myriad of smaller, unnamed islands. Serpentia supports a rich tropical fauna and flora, but its dominant animal is the eponymous serpent, Mastigodryas johnsoni, the checker-bellied tree racer. Checker-bellied tree racers are large, deep-indigo or violet colored snakes with a striking blue and gray checkerboard pattern on their ventral (belly) scales. They live in the trees of Serpentia and feed mainly on birds, bird eggs, and lizards, with an occasional unlucky bat thrown in for variety. One unusual thing about M. johnsoni is that the number of vertebrae in a full-grown adult is quite variable. The population mean is 200 vertebrae (see the histogram for the main island in simulation), but the spread around this mean is controlled by strange gods (you) that say what the variance is. Vertebral number is highly heritable (it is completely genetically controlled), but appears to have no effect on the reproductive or ecological success of a given snake, and thus is selectively neutral. The number of vertebrae is also subject to relatively frequent mutations, with the offspring of a snake having a 0.03% probability of differing from their parent by +/– 4 vertebrae, a 0.7% probability of differing from their parent by +/– 3 vertebrae, a 1.2% probability of differing from their parent by +/– 2 vertebrae, and a 2.5% probability of differing from their parent by +/– 1 vertebra. Consequently, 95.57% of offspring will be the same as their parent, but some small proportion will have mutated.

M. johnsoni also primarily reproduces by parthenogenesis (uniparentally from unfertilized eggs) and so sex is rare enough that it can be discounted in our simulation. When a given female reproduces, she lays a clutch of 10–20 eggs, and these offspring are likely to share vertebral number with their parent although there is a chance that a mutation will occur that either increases or decreases the number of vertebrae, as described above.

Another thing you should know about Serpentia is that it is an island with almost no beaches, forming instead a high mesa or plateau topped by wet tropical forest and surrounded on all sides by 50-meter cliffs that drop straight into the azure depths of the sea. The tops of these cliffs are draped with lush jungle vegetation that often breaks away and falls into the water below, sometimes taking with it whole trees and the denizens living therein, including our friends, the checker-bellied tree racers.

This sets up an interesting dynamic because the other islands in the archipelago have all of the ecological necessities to support tree racers, but they don’t have any tree racers. Our simulation will be based on what happens when a tree from Serpentia falls from a cliff and floats to one of the other islands, carrying with it a small group of checker-bellied tree racers. The main island of Serpentia supports a population of 100,000 snakes and the smaller islands can potentially harbor 10,000 snakes. The simulation will show the resulting distributions of vertebral number on all of the islands after they have been colonized and have then reproduced until the small islands (islets) have reached carrying capacity. The parameters that can be adjusted in the model are the number of snakes that colonize an island (islet immigration size), and the standard deviation of the main island population (this is an indication of how spread out around the mean the data are).

SIMULATION

QUESTIONS

 

Question 1. What does eponymous mean?

Set the main island population standard deviation to 2 and press Initialize.

 

Question 2. Approximately what is the range of vertebrae count present on the island after you do this?

Set the main island population standard deviation to 7 and press Initialize.

 

Question 3. Approximately what is the range of vertebrae count present on the island after you do this?

Set the main island population standard deviation to 5 and the small island immigration size to 2. Press Initialize and then run the simulation.

 

Question 4. Are any of the resulting small island distributions bimodal (with two “humps” in the distribution)?

Rerun the simulation 5 times using the settings in Question 3.

 

Question 5. On average, how many of the small island distributions (out of five) appear bimodal using these settings?

 

Question 6. If a scientist were studying the small island snake population by measuring them and he/she found that there was a bimodal distribution of lengths (length is directly correlated with vertebral number), would he/she be justified to hypothesize that this distribution was due to sexual dimorphism (males and females being different sizes)?

 

Question 7. From what you’ve read in Chapter 7, name the effect that is responsible for the distribution observed by the scientist in Question 5.

Set the immigration size to 25 and the standard deviation to 5 and run the simulation a few times.

 

Question 8. In the results of the simulation, the main island distribution is superimposed over the resulting small island distributions as a black curve. Do the small island populations seem to be generally the same as the large island?

Now set the immigration size to 3 and the standard deviation to 5 and again run the simulation a few times.

 

Question 9. Do the small island populations still seem to be generally the same as the large island?

Run the simulation with a set immigration size and change the main island population standard deviation.

 

Question 10. Are you more likely to see a founder effect with a large or a smaller value of main island population standard deviation?

 

Question 11. As a scientist, is it important to keep genetic drift effects like the founder effect in mind when you are studying nature?