Here, the author is trying to find a total by multiplying one sample set by the number of sets. This would be like counting the number of people in my house by the total number of houses on my street to find the total number of people on the street. This relies on an assumption that the number of people in my sample (the well-studied class in the argument, the number of people in my house in my example) is representative of the norm or average for the whole group. For instance, what if I live alone, but most houses on my street turn out to have 5 people? Then my estimate will be only about 20% of the right answer.
B addresses this assumption. If the class in question turns out to be atypical, then multiplying by the number of classes could give a very inaccurate result.
A references a claim about the species within the class, but there isn't any claim made about that.
C isn't a flaw. Sure, there might be other ways to get the total, but that doesn't mean that this one wouldn't work. Similarly, if I say that you could get somewhere by bus, my argument isn't flawed just because you could also ride a bike.
D is inaccurate. It's saying that the author doesn't acknowledge the difference between identified species and total species, but they talk all about that. This is why they want to look at a class in which almost all the species have been identified. If it were true that all the classes were roughly equal, this would give us a good idea of the number of remaining unidentified species per class.
E is saying that the author doesn't recognize that some classes have more species than others. Again, this is something the author does recognize. They realize that some classes have a lot of unidentified species, but they think they can use more well-studied classes to overcome this problem.