The conclusion of the argument is that "the marine biologists' hypothesis, therefore, is clearly wrong." What hypothesis? That lobsters kept together in lobster traps eat one another in response to hunger.
Why it it wrong? Because there are instances of lobsters sharing traps together for weeks, and, more specifically, "eight lobsters even shared one trap together for two months without eating one another." Right away you should be thinking that these counterexamples would only refute the marine biologists' hypothesis if the lobsters in these cases were hungry (i.e. there was a shortage of food).
Remember that on a Strengthen with Necessary Premise question, we need to first determine whether the answer choice strengthens the stimulus, and if it does, then we will negate it to make sure it is also necessary to the argument.
(A) is incorrect because it does not strengthen the argument. Even if lobsters not caught in lobster traps have been observed eating one another, we still don't know why this is happening. Is it because of hunger or another reason? We are trying to strengthen the theory that lobsters stuck in traps together do not eat one another in response to hunger. The negation of (A) would be, "Lobsters caught in lobster traps have been observed eating another," which is entirely consistent with the stimulus.
(B) is incorrect even though it somewhat strengthens the stimulus, because it is not necessary. If two months is the longest period during which eight or more lobsters have been trapped together, then the fact that they did not eat each other could show that they are not reduced to cannibalism due to hunger when stuck in a trap even for the longest known period. However, we do not know what the conditions in this trap were. What if the trap was sitting in an area with a lot of natural lobster food floating around readily available for the lobsters in the trap to eat? In that case, the lobsters were not actually hungry, because they were still eating. So this does not strengthen that lobsters do not eat each other in response to hunger when trapped. Moreover, when we negate this answer choice, "Two months is not the longest known period during which eight or more lobsters have been trapped together," we do not end up with a statement that makes our stimulus fall apart. The negation of this answer choice does not affect the stimulus because we still don't know whether or not the lobsters were hungry, which is key in this stimulus.
(C) is incorrect because it does not strengthen the argument. Who cares if it is unusual? It happened at least once before (i.e. in the example cited) and, when it did, lobsters did not eat each other.
(D) is incorrect because even though it strengthens, it is not necessary. Though it strengthens the stimulus to know that other marine species eat their own kind due to hunger, when the answer choice is negated, "members of other marine species never eat their own kind when no other food sources are available," does not destroy the argument because it is still ultimately referring to other marine species and we are focused on whether lobsters do this or not.
(E) is CORRECT because it not only strengthens, but it is also necessary. It strengthens by showing that the lobsters were in fact hungry, but yet they didn't eat each other. When we negate the answer choice, "Any food that the eight lobsters in the trap might have obtained was enough to ward off hunger," it destroys the argument. If the eight lobsters were able to get enough food to not be hungry, then the argument against the marine biologists' does not stand because the lobsters were not hungry, so the "eight lobsters" example would not be a counterexample to the marine biologists' hypothesis.
