Exam statistics: Mean: 63. Range: 30-96. Most scores are in the high 50s to mid 60s. Compared to exam 1, a few people did much better (2 scores in the 90s, one in the high 80s), fewer did really badly (not as many scores less than 50.) As you know, I don't curve individual tests, but if the final distribution of all points looked like it did on this test I would curve by 4 points (so 86 and above would be an A, 76 - 85 a B, and so forth.) Let me remind you again that I don't usually end up curving that much -- the final point distribution is usually higher. Also, let me note that this distribution of grades is very similar to what I see on this exam during a "normal" semester (not taught on the web) -- based on the average and on the range of scores you're not doing either better or worse than students do in a typical semester. Finally, remember your final exam percentage can substitute for one previous exam -- but only one.
Comments on some questions people had trouble with (to indicate what material from this test you're likely to see again on exam 3!) People had trouble with the questions on adaptive landscapes and correlated characters (questions 11-13). Remember that in an adaptive landscape the population evolves to have the traits that make up the fitness peak closest to the frequencies of traits with which the population starts out. Remember that when characters are correlated, a trait that decreases fitness slightly can evolve if it's correlated with a character that strongly increases fitness. This information also applies to the new material we'll be looking at on evolution of optimal traits and adaptation. People still had difficulty with mid-parent offspring regression tests of heritability (qu. 14) and with the calculation of wbar for the natural selection question (16). Many of you need to clarify your explanation of why derived, not primitive, traits provide evidence for phylogeny (question 18) and many of you need to clarify your explanations of why different forms of DNA evolve at different rates (question 20b.) Finally, a lot of you could not pick out the best supported phylogeny (20d). Be sure you can draw and identify best-supported phylogenies. Practice drawing the same phylogeny with different horizontal orders of the species -- I think some of you did not recognize the best supported phylogeny because it was drawn in a different order from the order you expected, but remember horizontal order does not mean anything (only the branching pattern tells us about relationship.) Being able to draw and interpret phylogenies will be crucial for this next section of the course!
Once you have received your exam and read the following key, please ask if you have questions. If there's a grading error or if you think your answer is as good as one that I've given in the key, give me a written explanation, and your exam (you can leave them in the Biol. Dept. Office) by 12 April.
Key to Exam 2
1. d.
2. c.
3. d. (because in this case individuals pollinated by different pollinators could arise and would be unlikely to reproduce with each other; in the other cases, all individuals are likely to reproduce -- there's no ecological factor to separate them.)
4. a. (because in b the differences are gradual, there aren't distinctive difference so the groups would be the same species by both concepts, and in c the groups would be different species by both concepts, not just the PSC.)
5. polymorphism
6. polyploidy
7. c. (they're different species by either concept since hybrids are sterile; the individual with the duplication could still produce functional gametes and be fertile since it has an even number of chromosome sets so pairing in meiosis can occur, but hybrids are sterile because they have an odd number of chromosomes.)
8. a. (when traits change randomly the population may be moved to near a new adaptive peak so that once the population grows, it will evolve to that new peak. Answer b is not correct because loss of genetic variation would make the population unlikely to evolve new distinctive traits so it wouldn't become a new species. Answer c is not correct; gene flow would prevent speciation.)
9. Negative frequency dependent selection
10. b.
11. Adaptive landscape (only half credit for coadapted gene complex or adaptive peak because these refer to the high fitness combinations not all possible combinations.)
12. a. (because the population will move to the closest adaptive peak, not necessarily the highest adaptive peak.)
13. b. (because long jaws are correlated with long fangs and long fangs have very high fitness and will evolve causing long jaws to evolve, too, even though they decrease fitness slightly.)
14. A plot of offspring toe length (trait value) versus mid-parent toe length (trait value) that shows a flat (horizontal) or nearly flat line.
15. d.
16. (0.8)2(1) + (2)(0.8)(0.2)(1) + (0.2)2(0.26).
17. Step 1: synthesis of simple organic molecules through non-biological processes. Evidence: Miller' recreated what he thought was the early atmosphere of earth -- methane, ammonia, hydrogen gas -- in a sealed chamber, and boiled water below the chamber; he used electric sparks to simulate lightning. He found amino acids formed in the water. This is plausible because it creates a system that models the early earth and shows that in those conditions, with only the non-biological processes possible in the experimental chamber, amino acids, which are simple organic molecules, could form. Other evidence for step 1: meteors have been found with organic molecules suggesting that they could form in space and be brought to earth; plausible as a way organic molecules could have been formed because evidence from volcanoes suggests the early atmosphere may have been high in carbon dioxide which would have cushioned the fall of meteors to earth and prevented destruction of these molecules.
Step 2: formation of polymers from simple organic molecules. Evidence: in experimental solutions with clay particles amino acids will spontaneously add to protein chains, suggesting in such conditions on early earth (in water washing over rocks, sediment, with clay) amino acids could have joined to make proteins.
Step 3: development of self-replication molecules. Evidence: RNA could plausibly have been a self-replicating molecule in early conditions since it contains genetic material and can act as a catalyst so it may be able to catalyze its own replication.
18. The reason traits provide evidence for phylogeny is that when two species have the same trait, a likely explanation is that they have it because they inherited it from their ancestor. If the trait is a primitive trait, it was present in the ancestor to the entire ingroup, so it two species have this trait in common it probably came from the ancestor to the entire ingroup. It does not, as a result, show anything expect that these species came from the ancestor to the whole ingroup -- since all ingroup species came from that ancestor, this does not tell anything about whether the species that have the primitive trait are more or less related to each other or to other ingroup species, so it doesn't give information about phylogeny. In contrast, if the trait species have in common is a derived trait, then it evolved within the ingroup. If two species have the derived trait in common this is evidence that they evolved from the species that evolved that trait, and this was a more recent species than the ancestor to the whole ingroup, so it suggests that they came from this more recent ancestor, which means they are more related to each other than to species in the ingroup that did not evolve from this ancestral species.
19. To determine the rate of DNA evolution, find two modern species with a fossil record good enough so the date of speciation (when those two species separated from each other) can be determined by radioactive dating of the fossil ancestor to these two species. Obtain DNA from the two modern species; count the number of DNA differences between them. The number of DNA differences between these modern species divided by the time since speciation determined from the fossils is the rate of evolution of this DNA (assuming it evolves at a constant rate.)
b. e. (both a and c show pairs of species that are equally related, so should have the same number of DNA differences between them if the molecular clock hypothesis is true.)
20. (a) b.
(b) DNA that codes for ribosomes evolves the most slowly because most mutations that occur will harm the function of ribosomes, which are crucial for protein synthesis, so these can not evolve. Because only few mutations result in surviving individuals, there are only a few mutations that could evolve so evolution is very slow. Protein coding evolves somewhat faster; there are more areas where mutations could occur without harming the function of the protein, so there are more mutations that could possibly evolve. Non-coding evolves fastest; all mutations are of equal fitness so any mutation that arises could potentially evolve.
(c) If you chose DNA that evolved too slowly you would not find enough differences among species to be able to study phylogeny. [NOTE: many of you said "chance convergence" -- that would be a problem if you chose DNA that evolved too fast, not too slow.]
(d) The upper middle phylogeny is correct (the phylogeny looks like this:)
Please note that the three phylogenies that show "Slow" as being more related to one of the ingroup species than to others can NOT be correct -- the outgroup is equally related to all ingroup species.
(e) C.I. = # derived states/ # evolutionary changes = 5/6 (this is the CI for the correct tree; if you chose an incorrect tree this answer was graded based on the tree you actually chose.)
(f) Shell pattern shows homology. It supports the hypothesis that the blue, green, and orange turtles are more closely related to each other than they are to the red (or slow) turtles. (Note: this is the answer based on the correct tree; if you chose an incorrect tree this answer was graded based on the tree you actually chose -- for some of the incorrect trees, this trait shows convergent evolution.)
(g) The blue turtle and orange turtle are each other's closest relatives. The green turtle is more closely related to the blue and orange turtles than it is to the red turtle. (The blue, orange, green, and red turtles are more related to each other than they are to the slow turtle.)