How is mimicry not a just-so explanation for similarities b/w species?

[tomsecret]

how is mimicry not a "just-so" explanation for similarities b/w species?

ok obviously its not but this is what it seems like to me..cant wrap my head around this for some reason, please explain:
I am drawing a distinction between behavioral and appearance mimicry.. So how is mimicry (in appearance) not a "just-so" explanation of similarity between species? What i mean is, couldn't we just as easily argue say that cats are mimicing monkeys by having a tail that enables them to climb tall trees to escape predators? or that ravens are mimicing crows by having wings to escape cats?
Lots of great examples of behavioral mimicry in nature, but how can a phenotype actually evolve with respect to mimicry? If organism do change appearance, then shouldn't we therefore see many similarities within extant species and more diversity in the fossil record? I am not talking about instant change like that octopus that can change color to camoflauge its appearance but rather permanent morphological (??) changes? The famous butterfly example has since been updated to reflect mullerian mimicry but still, --where/how do we draw the line between 2 species that simply look similar, and 2 species actually adapting model/mimic roles? thanks for any help/clarification!

 

[Ygggdrasil]

Determining whether traits in different organisms come from a shared evolutionary history or whether they evolved independently is a major challenge in biology. There are are a few lines of evidence that biologists can examine to determine the difference between homologous structures (structures with a shared evolutionary history) and analogous structures (structures that evolved independently). First, we can look at whether the similarities go beyond superficial similarities. For example, wings of crows and ravens share a similar bone structure and develop in a similar way in the embryo. In contrast, the bone structure and development of bat wings or insect wings is very different from those of avian wings.

One can also incorporate information about the evolutionary history of the species. Crows and ravens are very closely related and all closely related species around them in the evolutionary tree also have wings, implying that the common ancestor of these animals had wings. In contrast, flies are very distantly related to birds, and because many species descending from the common ancestor of flies and birds do not have wings, it is unlikely that the common ancestor of these species possessed wings.

Finally, molecular evidence (e.g. DNA sequences) can provide evidence whether traits in different organisms are evolutionary related or not. For example, if one studies the DNA and protein sequences involved in the development of wings in crows and ravens, one would likely find that these sequences are very similar and analyzing the similarity of the sequences is a good way to demonstrate that they are evolutionarily related. The wings of flies, however, likely use very different types of genes and proteins in their development.

 

[epenguin]

Surely mimicry means just looking the same? And in particular looking the same for a biological purpose, usually purposes of deceit. That means looking the same enough to deceive another animal. E.g. looking like something innocent when you are a predator, or like something frightening, or uninteresting (a stick), or disgusting or poisonous to a predator. There is just no reason for that purpose why you need to be closely related to what you are simulating and I am sure that often enough the relation is quite distant. Looks are superficial! No doubt if you are related to something else it can be an easier series of steps to get to look the same.

 

[atyy]

There have to be some checks as to whether the phenotype is an adaptation or not.

The usual method in science is to use the hypothesis to make as many predictions as possible, and test those predictions.

Naively, one could test this by spraying paint, changing the appearance of the mimic and seeing whether the mimic loses fitness. Of course, this is probably too coarse a manipulation and the mimic would lose fitness for other reasons, like getting poisoned by the paint.

However, there are more refined tests. For example, there should be no advantage where the mimic and model don't geographically overlap, so one would predict no mimicry in those cases. That argues against cats' tails, since most people don't keep monkeys at home. Also, mimicry is only predicted to be useful only when the mimic is rare compared to the model, which yields a testable prediction of less similarity between mimic and model in geographical regions in which both are common.

Apparently whether mimicry is the correct explanation or not is still not known in many of the cases discussed in http://www.univet.hu/users/jkis/education/Visokol/Mimicry.pdf[/URL]

Unfortunately not free, but [URL]http://www.nature.com/nrg/journal/v12/n11/full/nrg3015.html[/URL] reviews the general problem of whether something is an adaptation or a spandrel ("just so story").