The claim that only a tiny fraction of whole-genome duplications ever stabilize in a lineage [3] rests on a numerator with no denominator behind it. A duplication that kills the cell, or the embryo, or the lineage two generations later, leaves nothing to sequence. What gets counted is the survivors, which is exactly the shape of the evidence when Jonathan Wendel says every seed plant alive carries at least one ancient duplication [6], or when barnacles, insects, trout and arachnids turn out to carry the same signatures [7]. None of that makes the suspected failure rate wrong; it does mean the figure is an inference from what lived rather than a count of what happened.
Douglas Soltis of the Florida Museum of Natural History puts it as a hitter who strikes out a lot and hits home runs when he connects, and calls polyploidy the most important process on the planet that hardly anyone knows anything about [5]. The analogy is honest about the variance and quiet about the average, because a batting average needs recorded at-bats. Even the most-cited home run is unsettled: whether the changes that produced jawed vertebrates trace to ancient duplications is still argued over [8].
The reason any of this is now legible is price. In the early 1990s Kenneth Wolfe's lab at University College Dublin joined an EU scheme paying 2 euros per nucleotide to sequence baker's yeast [10]; by 2022 sequencing ran about $0.000000006 per nucleotide [11]. That is a factor of roughly 300 million [12], and since it compares euros then with dollars now, treat it as an order of magnitude rather than a deflated series. Before sequencing, polyploidy was something you noticed by counting chromosomes under a microscope, and its consequences only became visible base pair by base pair [13] - what Wendel means when he says the tools opened up a world nobody knew was there [14].
Which brings the snail's value into focus. Under a million years is recent enough that the adjustment to a heavier genomic load is still underway rather than sanded flat by time [4], so the intermediate steps can be observed in one genome instead of reconstructed across hundreds of millions of years. What the snail's youth cannot supply is an explanation for why the doubling happened at all, and Neiman says so plainly: the answer is missing for her snails and for every other species [2]. Recurrence is a rate, and rates need the failures counted. My read, with its condition attached: the field will keep improving on the consequences of duplication and stay stuck on its trigger until someone can watch new duplications appear in a population before selection has filtered them. Genome doubling being the most radical mutation available in a single generation [9] is precisely what makes the failures invisible.
