Terminology quibble. Usually, the term "historical science" -- which includes astronomy, geology, anthropology, paleontology, climatology, archaeology, cosmology, criminology, and of course, evolution -- is used in contrast to "experimental science." Your use of the term "natural science" (instead of "experimental science") makes it seem as if the historical sciences are somehow un-natural. They are not.
Well PH, of course the historical sciences are not "unnatural!!!" Human beings engage in them far too much to say that they could in any way be regarded as unnatural. Engaging them is a huge part of what it is to be a thinking human being.
So if you prefer the term "experimental science," I don't have the least quibble. In fact if anything, IMHO it affords greater clarity.
Earlier you wrote to suggest that once you know the tools that were available to work on a [historical] problem, then you can deduce all the rest.
Let's see. Let's say I own a hammer. Let's say you can ascertain that I, in fact, do own a hammer; and more, you can ascertain that, indeed, I do know how to use a hammer. Now, from that secure knowledge base, how do you get to the knowledge that allows you to state what projects on which I have actually used that hammer? Or how long it took me to complete those projects?
I really liked the article general_re bumped me to. I thought it was a very rational attempt to explain how a thing could be done, based on the set of evidence selected for the purpose. But that is not what it takes to show that that was how a thing actually was done. See what I mean? We have a big "gap" here.
Thanks for writing, PH.