I'm no creationist...but here's a serious question back at you concerning evolution.
It doesn't work. Not logically. No, really....hear me out.
Monarch butterfly...started as a worm, but then at some point,over a period of time, one family, over generations, mutated to metamorphosis into a flying insect. I can buy that, sorta. But then they went and developed poisonous wings. I am told this poison prevents other animals, namely birds, from eating them. And that the process for this happening is that, say, one was born that tasted bad, which was a favorable trait that was then passed on to it's progeny. But wait...how does anything know it tasted bad? Until, of course...they eat it...which is not so good for the butterfly, and it's would have been progeny, yes? So just how DID the monarch butterfly become poisonous, and how exactly DOES/DID that help it to survive? In the end, the only sign that it is are it's bright colors, which, in nature, often means danger. But that completely invalidates the use of the actual poisonous nature, as all it would REALLY need to be is brightly colored, and not poisonous to eat.
No, the idea that these traits came about over time just doesn't jive with logic, as none of these traits would prevent the death of their hosts from another species that isn't already aware of those traits. Nothing eats poison dart frogs because they know they are poisonous...millions of years of passed on internal info sees to that. What started that information? And what good at all is the poison, then? Yes, I eat a poison dart frog, and as a result, I die. But uh...so does the frog. Which now does not live to pass on it's traits.
Ok, so there are a few possible answers to the conundrum you've posed.
To be clear on the question, I'm going to use butterflies as an example and birds as the predator in question, but you can extrapolate to the frog. You're asking how the trait not to eat the butterfly could have ever been passed on, if every bird that ate a specific butterfly died.
I think you may also be asking how the first poisonous butterfly would have passed on the poisonous trait if it was eaten and both butterfly and bird died.
As always feel free to correct me as it's not my intention to misrepresent you're argument.
You are making several assumptions and haven't considered several possibilities.
First, and probably the simplest explanation is, that it's poison didn't always kill the predator in question. Perhaps it just tasted bad, maybe even made the birds sick. Over time the bird learns not to eat the "red ones" and passes that trait on to it's offspring when teaching it what to eat and what not to eat. hundreds or thousands or even millions of years later that bad taste has evolved to a stronger and deadly poison. Thus any bird that eats the butterfly now, definitely won't pass on that trait.
Let's expand this just a bit. Let's say the environment changed (let's say were in a closed system like an island) and butterflies were all there were to eat, forced to eat or die perhaps given enough time a few birds might have a trait that makes them immune to the butterflies poison and presto, all the birds without that trait die, those with it eat otherwise poisonous butterflies. On that island the butterflies might get wiped out.
And we see this in nature, specific adaptations in closed environments. Iguana's that swim in the ocean, finches with thicker stronger beaks ect....
Another possibility is that some birds simply didn't eat butterflies and others did. Those that did eat them died out and those that did not survived.
As far as the first butterfly born with poison, how did it live long enough to pass on it's trait? Well, perhaps the environment it was originally in didn't have any predators at all, which would make the poison a non-specific trait. That is, it is nether helpful, nor harmful. Humans have tons of these traits, things like different hair color, un/connected ear lobes, being able to roll your tongue, hair on your chest ect (i'm sure there are a lot more that would have more to do with potential survival, I just don't know of any)....These are traits and they don't help and don't hurt, but (hypothetically), what if the trait for blonde hair required a specific protein in the body, Now a virulent illness with a 100% mortality rate comes along, but does not effect those with the blonde hair protein. Suddenly, the non-specific trait (blonde hair), becomes specific in that it is necessary to survival. 100 years from now virtually everyone would have blonde hair and if not for the digital and film age, looking back people might have been tempted to believe that blonde hair was all their ever was.