If you went out into the woods, and dug up the skeletons of a couple of raccoons that had been dead for 20 years, there might be some DNA left in their bone marrow or in their teeth. If you were a very good geneticist, you might be able to tell if these two raccoons were related from their DNA, and maybe even if one of them was the parent of the other. But you would have to be very good to tell this with certainty.
If you know of a way that you can show that some fossils were the literal parents of other fossils, how does it work without any guesswork? DNA is long gone.
How about if there was an ancestor/descendant relationship between a couple of fossils of completely different animals? How does anybody know it for sure? DNA is long gone.
How do you know if the two fossils that you say are ancestor/descendant, are not really two completely different animals that came about completely separately with no ancestor/descendant relationship at all? Saying that there had to be a relationship - THERE JUST HAD TO BE - doesn't cut it scientifically. Why not suggest that ALL the different animals came about separately, with absolutely NO ancestor/descendant relationship?
So, how do you determine it? If you don't have a really factual way, then nobody knows if there even is any kind of the evolution that evolution theory suggests exists. What is the way that tells us factually. Even Astargath hasn't been able to explain this. But maybe you are a better explainer than he is. So print it out right here in simple, brief language so we all know.
And don't go about copying and pasting some gibberish that some orator who doesn't know anything about evolution proof just spouted out. We want the proof principle in simple language that shows us the proof for evolution over the multiple thousands of years of ancient prehistory times.
From what you wrote there, you aren't exactly denying evolution, rather, you don't seem to understand how ancestry works.
I'll go back to dog breeds here -- both golden retrievers and rottweilers are dogs, would you say that they're completely different animals as well? Pretty much all dog breeds are "artificial" in a way, where we only let them breed with specific partners to let certain attributes become more dominant. They are still "dogs", as different as the breed might be, they all share the same ancestry.
I haven't worked at a lab, so I can't get too specific there, but fossils can and do hold DNA for a lot longer than 20y, they recently found an 800,000 years old one in Greenland, still had preserved genetic material. It's not so much a matter of a skilled individual, but more so how good our tools are. DNA is not the only "resource" used to link species.
Science doesn't magically provide facts and answers, it's a method to get to them.
Astargath's been explaining things pretty well, and less long-winded than myself, you're just dismissing things a priori instead of trying to understand the evidence logically.