Distinguish a stepping-stone pattern of gene flow among populations from an island pattern. Under which of these schemes, if either, would we see a pairwise pattern of differentiation in neutral genes in which two populations closer to each other are more similar genetically than two populations separated by greater distances?

Is overdominance in fitness likely to be an important explanation for the maintenance of large amounts of genetic variation in nature?

Describe the ways in frequency-dependent fitnesses appear to arise.

How can we set broad magnitudes of genotypic and environmental variance and the intensity of selection in order to obtain a pattern of gradual, continuous change in the average value of a phenotypic character? How about if we wish to see rapid, almost step-like change in average character values?

Which are the conditions under which hitchhiking is likely to occur?

Under which specific population-genetic conditions can recombination act to increase the genotypic variation in a population?

How does the magnitude of the environmental variance affect the rate at which a continuous trait will evolve in response to natural selection?

How can strong natural selection be acting on a trait and yet cause no evolutionary change in the average value of that trait in a population?

If a group of conspecific populations are not exchanging migrants at all, should we see any spatial pattern in the level of pairwise divergence among them?

You know that in a finite population the probability that a neutral allele will achieve fixation from this point in time on is simply its relative frequency in the population. Summon your powers of deductive reason ("Here, Sherlock, time to solve one") and tell us what is the probability that a brand new neutral mutant will be fixed in a population of N haploid individuals.

Consider the following situation. Two linked loci, "A" and "B" each have two alleles, A and a at the "A" locus and B and b at the "B" locus. Alleles A and B are favored by natural selection. There is an initial positive disequilibrium wherein A and B are positively associated. At a third locus, "C", one allele c acts to increase the recombination rate between "A" and "B" but has no other phenotypic effects. Is it likely that the frequency of c will increase over time? Why or why not?

What are the specific genetic and phenotypic effects of inbreeding?

What is the evolutionary significance of genetic hitchhiking?

Consider a gene "A" in a large, random-mating plant population with these properties:

• Genotypes: A1A1 A1A2 A2A2
• Probability of survival: 0.90 0.60 0.80
• Avg. number of offspring: 4.00 4.00 4.00

1. What is the fate of the A1A2 genotype (eventually)?

2. If the A1 and A2 alleles were equally frequent in the population and if the genotype frequencies were at the Hardy-Weinberg equilibrium point, what would the mean absolute fitness be in the population? How about the mean relative fitness?

3. If all reproduction occurred via selfing, would the ultimate fate of the A1A2 genotype from your answer in (a) change? If so, how? If not, why not? If not, does the rate at which that outcome is attained change? Why or why not?

4. For bonus, calculate the relative frequency of the A1 allele in the next generation (starting from the condition described in part (b)) and assuming random mating).

5. Redo parts (a)-(e) but this time let the heterozygote's average number of offspring be equal to 6.50. Does anything change?

6. Consider the conditions described in part (e) as occurring in two populations, conveniently denoted I and II. Population I has about 10,000 individuals; population II has about 50 individuals. In principle, should the outcome of selection be the same in each population? In practice, what is likely to happen?

7. In population I under condition (f), at equilibrium, which homozygote will be more common? Would the answer change if A2A2 homozygote's probability of survival were 0.70?

Consider this situation describing the properties of individuals with each of the three genotypes at the "K" locus (I'm tired of A's), which affects body size, in a population of cockroaches:

• Genotypes: K1K1 K1K2 K2K2
• Probability of survival to maturity: 0.30 0.20 0.60
• Avg. number of offspring: 6.00 9.00 3.00

It turns out that survival probabilities are associated with body size (larger roaches at a given age are less likely to survive because humans are better at seeing and swatting them), but so are offspring production rates (bigger males get more mates, and bigger females make more eggs). You can deduce the rank order of genotypes in body size.

1. Will the average body size evolve through natural selection in this population?

2. Will average body size change within a single generation?

3. BIG TIME BONUS FOR THE TRULY DEDICATED! It turns out "K" is a supergene. Two real genes exist, "Q" and "R", the "1" alleles at each locus increase body size, and K1K2 is really a double heterozygote. What's the sign of the disequilibrium coefficient, D, and do the two genuine genes have equal effects on body size variation?

Is the notion that natural selection is slow and moderate supported by data?

One explanation for the large amount of genetic variation seen in natural populations is that this variation represents the balance between the rate at which mutation creates new but slightly deleterious alleles and the rate at which selection eliminates those alleles. But mutation rates, per gamete per locus per generation, are so low . . . . Is this "mutation-selection balance" explanation for genetic variation a realistic one?

In which range of allele frequencies does selection act most effectively? In which range does genetic drift act most effectively?

Define and explain the effective population size.

Define and describe marginal overdominance in fitness and offer a brief explanation of the context in which the concept arises.

Selection on a continuous trait ought to erode the level of genetic variation for that trait. Under which conditions will that erosion occur, and under which conditions will that erosion be minimal?

Why is the expected substitution rate for neutral alleles equal to the mutation rate?

At mutation-drift equilibrium, will the number of segregating alleles at a locus be greater in a large or a small population?

Many families of angiosperms include species whose growth form is a climbing vine. In some families the vines use tendrils to hold onto other plants. In the Passifloraceae the tendrils are modified stipules, whereas in the Bignoniaceae the tendrils are modified leaflets (parts of a compund leaf), and in the Ranunculaceae the tendrils are modified entire leaves.

1. In what sense might these cases be interpreted as "convergent evolution"?

2. Offer some explanations as to why different structures might have been modified in the different groups.

You are at a swank soiree regaling some of your fellow guests with a lucid discussion of how gametic disequilibrium is generated by finite population size when a casual listener says "Isn't it true that disequilibrium could also be generated quickly by natural selection?" Answer this person and offer a lucid explanation of why you answered as you did.

Figure 5.14 in the textbook illustrates an interesting pattern of step-like change in cell size over about 3000 generations of E. coli. The text says that this pattern resulted from strong directional selection that eliminated genetic variation quickly and each "step" up is the result of a new beneficial mutation that was then fixed. This result suggests that the rate of adaptive evolution is limited by the rate at which beneficial mutants appear, which is not the theme we've emphasized in lecture (which is that there is lots of hidden genetic variation whose release causes "steps" upward). It would seem that this result refutes the conclusions we drew from Castle's experiment about hidden genetic variation.

1. Which circumstances will cause a step-like pattern of long-term directional change in which the release of hidden genetic variation is necessary to "step up"? Which circumstances will cause a more continuous upward climb in mean trait values? In the latter case, is there likely to be no role for the release of hidden genetic variation?

2. Why are the steps in the E. coli experiment caused by new mutations? Why couldn't they be caused by the release of hidden genetic variation?

What is the "founder effect"?

Why do so many natural populations seem to have mechanisms to minimize inbreeding? Under what conditions might inbreeding be favored by natural selection?

Look at Figure 5.21 in the textbook.

1. Explain why there is greater variation in allele frequencies among young populations than among populations of intermediate age.

2. Do you think the textbook's explanation of why there is greater variation among old populations than among populations of intermediate age makes sense? Can you offer an alternative explanation to the one in the textbook?

In the large land snail Helix delicious, individuals may have no bands on the shell (A₂A₂), three bands (A₁A₂), or five bands (A₁A₁). There are no environmental effects on the numbers of bands on the shell. This species has a restricted range, and there are 16 populations arranged in the landscape in a 4 x 4 square like this:

A     b     C     d

e     F     g     H

I      j      K     l

m     N     o     P

Upper-case letters signify populations in which nearly all (over 95%) of the snails have five bands; lower-case letters signify populations in which nearly all (over 95%) of the snails have no bands.

1. Could genetic drift be responsible for this pattern? If not, why not, and if so, which qualitative conditions (magnitude of population size, magnitude of any selection coefficient on the alleles for banding, migration rate and pattern) must be fulfilled for genetic drift to be responsible for the pattern?

2. Under which circumstances, in the absence of any migration among the populations, would natural selection alone be able to generate this pattern?

3. If you knew that migration rates were high, and the migration pattern were an island one, then which force (selection or drift) is more likely to be responsible for the pattern?

4. If you knew that migration rates were low and were in a stepping-stone pattern in which snails, if they moved at all, moved only latitudinally (across a row) or longitudinally (up/down a column) to the nearest population location (i.e. no diagonal movement), then which force (selection or drift) is more likely to be responsible for the pattern? Or perhaps either is equally likely, given the information you have?

5. Just for fun, what do you think are the chances that, ultimately, two species would form from these populations, one species of unbanded snails, and one of five-banded snails? Can you speculate about the circumstances that would drive such a process (after 3/15 you should be able to answer in the affirmative and not merely be speculating)?

While standing in line for burgers at a barbecue, you overhear an argument over which factors determine the amount of neutral genetic variation within natural populations.

1. One of the disputants says that she has seen graphs that show that the amount of neutral genetic variation, measured as the number of alleles at a locus, increases as the effective population size increases. Explain why this is so.

2. The other disputant says that he read an article in National Geographic about a species of predatory feline whose numbers are now almost at 100,000 and yet seems to show almost no genetic variation (measured the same way as above). Let's presume the fact is correct. Is this fact consistent with your answer to part (a)? If not, resolve the paradox that it seems to represent.

There is a gene that affects running speed in rabbits, and the two alleles at this locus confer the following properties on each individual of each genotype:

• Genotype: A1A1 A1A2 A2A2
• Avg. running speed (mph): 15 20 25
• Probability of survival to reproduce: 0.60 0.70 0.80
• Avg. number of offspring per survivor: 10 10 10

When we begin to observe the population, the relative frequency of the A1A1 genotype is 0.16, that of the heterozygotes is 0.48, and that of the A2A2 genotype is 0.36. The population contains over 100,000 rabbits when we begin to observe them.

1. What are the allele frequencies at this point in time?

2. Are the genotype frequencies in Hardy-Weinberg equilibrium?

3. What is the average running speed and the genetic variance in running speed?

4. As natural selection begins to act, in which direction will the allele frequencies change? What will be the magnitude of the change in the relative frequency of the A2 allele between now and the next generation?

5. What is the ultimate fate of each allele in this population if natural selection continues to act in the same fashion in every generation as it's acting now?

6. How will the average running speed change between generations?

7. At the start of this process, the ratio of the genetic to the phenotypic variance is 0.80. In the first generation after our process of selection begins, how, if at all, will this ratio have changed? Why?

8. If the ratio of the genetic to the phenotypic variance were 0.30, would the average running speed and allele frequencies be changing at a different rate between generations than they do when that ratio is 0.80?

9. Now let's imagine that we start at the same relative frequencies at which we began this problem but that the heterozygote has a different running speed and probability of survival than those with which we've been working. Specifically, the values for the two homozygotes are just as before but now imagine that the heterozygote has an average running speed of 30 mph and a probability of survival of 0.90.

   What will be the magnitude of the change in the relative frequency of the A2 allele between now and the next generation?

10. When allele frequencies stop changing, what will be the value of each allele's relative frequency?

Your roommate is observing allele frequency changes in a population of flour beetles in your kitchen. He is selecting for longer antennae; antennal length variation is controlled by one locus, "A". There are two alleles and no environmental effects on antennal length variation. The A1A1 genotype has the longest antennae and A2A2 the shortest; the heterozygote's are exactly intermediate. When your roommate began the selection regime, the relative frequency of the A1 allele was 0.10. He found that the A1 allele became fixed in 20 generations, at which point he trashed the colony when you demanded he clean up the kitchen. Your roommate's girlfriend, who looked over his shoulder during the study, tells you she noticed that the relative frequency of an allele for brown eye color, which is controlled by a different gene than the one controlling antennal length, increased during your roommate's experiment from 0.05 to 0.50.

1. Offer two distinct possible explanations for the rise in frequency of the allele for brown eye color; each must be consistent with the facts stated above. If one explanation seems more likely than the other, say so and say why.

2. Which information would you need to obtain about your roommate's study or about the animal to enable you to diagnose which of the explanations you offered above is more likely to be correct? In other words, which circumstances would have to have existed to make one of those explanations correct, and which circumstances would make the other one correct?