Humans get roughly half their genes from each parent, but mothers affect their offspring in additional ways. First, there is mtDNA which is inherited only from one's mother. Second, there is epigenetic inheritance via mothers' eggs, which covers all their non-genetic qualities of these cells which grow up (post-fertilization) to become humans.
If a maternally inherited trait reduces the genetic fitness of a woman, then it will be under negative selection, and will be weeded out. If, on the other hand, it reduces the genetic fitness of a man, then it will not be affected at all: this reduction in fitness has no evolutionary effect since maternally inherited traits (e.g., mtDNA) are doomed in male bodies anyway.
Not surprisingly, such traits have been implicated in male sperm quality conditions, with e.g., specific haplogroups leading to reduced sperm count or mobility.
Sexual selection theory suggests that humans pick their mates because of their "good genes" (see other recent post). But this raises this issue: if males with good genes are selected for in each generation, then how come is a great reproductive skew maintained in the human species: why do some men produce many offspring while many produce none or a few? And, why do women often "cheat" on their mates, having children with others than their official mates.
Maternal inheritance explains this paradox: male reproductive variation due to the Y-chromosome or the autosomes can be shaped by evolution to produce males with good (well-adapted) genes, but maternally inherited factors cannot.
UPDATE: Interestingly, this may solve the paradox of non-inheritance of male attractiveness. While sexy parents have sexy daughters, apparently they don't tend to have especially attractive sons. This may be due to male-expressed maternally inherited traits. Such traits don't make their mothers' attractive (they are male expressed), and they are not inherited from their fathers.
Genetica. 2008 Sep;134(1):45-54.
Maternal inheritance, epigenetics and the evolution of polyandry
Zeh JA, Zeh DW.
Growing evidence indicates that females actively engage in polyandry either to avoid genetic incompatibility or to bias paternity in favor of genetically superior males. Despite empirical support for the intrinsic male quality hypothesis, the maintenance of variation in male fitness remains a conundrum for traditional "good genes" models of sexual selection. Here, we discuss two mechanisms of non-Mendelian inheritance, maternal inheritance of mitochondria and epigenetic regulation of gene expression, which may explain the persistence of variation in male fitness traits important in post-copulatory sexual selection. The inability of males to transmit mitochondria precludes any direct evolutionary response to selection on mitochondrial mutations that reduce or enhance male fitness. Consequently, mitochondrial-based variation in sperm traits is likely to persist, even in the face of intense sperm competition. Indeed, mitochondrial nucleotide substitutions, deletions and insertions are now known to be a primary cause of low sperm count and poor sperm motility in humans. Paradoxically, in the field of sexual selection, female-limited response to selection has been largely overlooked. Similarly, the contribution of epigenetics (e.g., DNA methylation, histone modifications and non-coding RNAs) to heritable variation in male fitness has received little attention from evolutionary theorists. Unlike DNA sequence based variation, epigenetic variation can be strongly influenced by environmental and stochastic effects experienced during the lifetime of an individual. Remarkably, in some cases, acquired epigenetic changes can be stably transmitted to offspring. A recent study indicates that sperm exhibit particularly high levels of epigenetic variation both within and between individuals. We suggest that such epigenetic variation may have important implications for post-copulatory sexual selection and may account for recent findings linking sperm competitive ability to offspring fitness.