Eusocial insects (ants, bees, wasps, termites) have many female workers who never reproduce despite being genetically capable of doing so. How can such a disasterous trait (for the individual) as never reproducing be selected for?
In milder forms, altrustic behavior is seen in many species, and sometimes it clearly endangers the individual who practices it. Why isn't altruism selected against?
The most obvious example of kin selection is parental care.
Kin selection can work even when individuals cannot identify their own kin, as long as they interact more often with kin than non-kin. Kin selection may be stronger when individuals can recognize their own kin.
The concept of inclusive fitness extends the idea of an individual's fitness by adding in the gene copies identical to his that are possessed by his relatives. For example, an individual's inclusive fitness would consist of his own fitness, plus half the fitness of each sibling, plus a quarter of the fitness of each half sibling, and so on through different degrees of relation. Usually we cannot calculate the whole thing, but the contribution of distantly related individuals is low anyway.
A gene which has a greater inclusive fitness than its competitors can increase even if its individual (non-inclusive) fitness is lower than theirs. For example, if you help your sister to have two extra children, the same average number of your genes is passed on as if you had had one child of your own. If you help her to have three extra children, you will increase the representation of your genes.
Social insects are haplodiploid: females have two sets of chromosomes, males come from unfertilized eggs and have only one set. If a queen bee mates with only one drone, her daughters all get the drone's 1 set of chromosomes, and on average share half the queen's genes, so they have relatedness of 3/4 (rather than the 1/2 of mammalian sisters). Haplodiploid species may have an easier time evolving eusocial behavior. A bee's sister shares 3/4 of her genes and so aiding a sister (such as a new young queen) is nearly as good as having offspring of one's own.
However, haplodiploidy cannot explain the eusocial behavior of naked mole rats, mammals who have a bee-like social system with only one breeding female. Clearly even diploid animals can get a big enough selective advantage that it's worthwhile assisting kin rather than raising your own offspring.
The most drastic form of eusocial behavior is the behavior of a multi-cellular animal's cells. Once, we presume, their ancestors were single cells who looked out for themselves. But now most of our cells sacrifice their chance to reproduce in order to help the germ line reproduce. How did this situation evolve? Cancer can be thought of as a breakdown, where cells reproduce for themselves at the expense of the body.
Another possible explanation of altruistic behavior (behavior that helps someone else at your expense) is that it may actually have a hidden advantage for the individual. The individual may get a direct advantage from the other creature's well-being, or it may hope that the other creature will reciprocate in the future.
This is probably behind many cases of symbiosis. The two organisms aid each other not because they are related, but because each one depends on the other and therefore has a stake in the other's survival.
There are interesting questions about how reciprocal altruism can arise in a population of selfish organisms. Many models require either kin selection or some ability to recognize cheaters in order to get started. It's difficult being the only altruist in a population of non-altruists! This type of question is addressed by the mathematical discipline of game theory.
It is controversial whether a trait that is bad for the individual and does not its kin, but is good for the group, can evolve and spread by natural selection.
Computer simulations show cases when this can happen. The best scenario for group selection is to form new groups every generation using a random mix of individuals. Groups are saved or killed based on the frequency of the selected gene in them. At the end of the generation offspring are reshuffled into new groups. In this case, even a gene that is bad for individuals can be selectively advantageous if it makes the group survive.
However, it is unusual for organisms to be grouped randomly. If they are grouped with kin, kin selection is probably a much stronger factor than group selection. Also, it seems unusual for whole groups to be constantly wiped out and replaced by other groups, which is needed to overcome the disadvantage to the individual.
In the fossil record, some species produce many daughter species, while others persist a long time but don't speciate into many new forms. In the long run, we will see more organisms derived from the fast-speciating species than the slow-speciating ones, because they will colonize more niches.
This could be considered an example of group selection if the traits that make a species speciate are bad for individuals within the species, but it's usually not clear if this is the case.
One trait for which this argument is often made is mutation rate. Organisms have some genetic control over their mutation rate; they can invest more or less energy in proofreading and correction. An organism that mutates rapidly may have an individual disadvantage because of deleterious mutations, but also a higher chance to produce advantageous new forms. Could there be group or species selection in favor of higher mutation rate? Or can individual selection preserve a trait that is very rarely highly advantageous (a new benificial mutation) but usually harmful?
Another such trait is sexual reproduction. Asexual females can concentrate their effort on producing 100% female offspring containing 100% of their genes. This gives them a two-fold advantage over sexuals (we will talk about this more later in the quarter). Yet most creatures are sexual. Could this be because of species selection, where asexual species become extinct easily and seldom speciate, whereas sexual species last longer and speciate more? Or is there a hidden advantage of sex to the individual? This is an important and not completely solved question.