One of the questions biologists try to answer is the presence of variance among individuals in populations. Why would variance in genetic composition or looks even be an issue? To make a long story short, it is not to far stretched to interpret natural selection as leading to evolution favoring a specific variant, or form of the population in question (in its most well known version “survival of the fittest” where the “fittest” form is one specific morphology or character). Assuming that that is the case, all other forms would be inferior and loose in the Darwinian battle between genes and gene combinations. Given wide distributions of species, or large differences in the environment species live in, the variance we actually see within species could quite easily be explained by different advantages to different gene variants in different parts of the environment. As an example, an Anthias from the Red sea would probably show some physiological difference to an Anthias from the same species from the Maldives, as the salt concentration of the Red sea is a good deal above that in the Maldives, selecting for adaptations to the different environmental situations in the two locations.
The problem is somewhat harder to explain when individuals of the same species but of widely different looks co-occur. An example of that is the widely varying color or pattern of some species of tube- building polychaete worms, such as the Christmas tree worms or the sabellid duster worms. These worms co-occur, often on the same coral head, with individuals of widely different colors or patterns. Why isn´t there one pattern that is the best, which all of them have?

Two sable dusters of the same species with different colors and patterns

Two sabellid dusters of the same species with different colors and patterns

I guess, or at least I hope, that I am not the only one looking for a particular item, say a book, which I believed had a yellow cover, and couldn´t find that book even if I knew it was in the room I was looking in. After giving up, and probably uttering a few expletives, I found the book in plain sight, but with a red cover. This is a great example of the phenomenon of search image. One looks for a specific color or pattern, and in order to do so efficiently, all other colors or patterns are ignored. It is afterwards often hard to understand, or even believe that the red book was in plain sight all the time, but our signal detection capacities are pretty limited, leading us to focus on few signals and reducing noise from the surrounding. Predators seem to do that as well just in the same way I do when I search for books, and will often ignore rare colors or patterns to focus and be able to detect the more common ones. This means that predators will search for and target common forms and probably not even detect less common or rare variants. So, when an earlier common pattern has been depleted to low numbers, two things happen. First of all, individuals with a formerly rare pattern or color will have been able to reproduce freely, and increase in numbers. Second, the predator will change its search pattern, now looking for pattern that is now common. Such shifts in predator preference will then change selective forces over time, in a way that sometimes favors certain patterns and sometimes other patterns, ensuring that variance is kept in the population in question. In other words, the “fittest” will not have the same genetics over time, but will vary over time as predators shift preferences.
Sabellid worm surviving another day

Sabellid worm surviving another day