Pre-adult mortality: should we care about it, and what can we do about it?

In a wide range of taxa, most individuals have zero fitness: they die before reproducing. In this Commentary, I first confirm that across taxa - from Drosophila to elephant seals to trees - pre-adult mortality is the norm, with ∼60-90% of offspring dying before reproduction. Two seemingly opposite, though not mutually exclusive, hypotheses explain who dies: (1) that this is simply due to stochastic events, a matter of chance or luck, or (2) that it involves selective disappearance, with the loss of low-quality individuals with specific phenotypic traits associated with low survival. I then review (a) what we know about (physiological) phenotypes early in development, at independence, (b) whether these might become fixed in early development, and (c) whether these traits are repeatable or labile during ontogeny, forming targets of selection determining fitness (cf. adult phenotype). I highlight four reasons to care about pre-adult mortality in current, experimental studies: (1) identifying the phenotypic traits (and physiology) determining life's winners and losers is a significant knowledge gap and worthy research goal; (2) it should matter if our study populations comprise a random sample of individuals (chance) or a 'biased' high-quality subset of individuals (selective disappearance); (3) we typically create conditions to minimize mortality in laboratory populations, but these are then totally different from natural populations (with high pre-reproductive mortality); and (4) if individuals that make it to reproduction are all high-quality individuals, the 'best of the best', this might explain the seeming absence of, or failure to detect, trade-offs and costs.