On 2015 Mar 12, Donald Forsdyke commented:

ADAPTATION DECOUPLED FROM SPECIATION. The authors begin by noting that gene flow among populations, rather than assisting speciation, has a “greater impact” when acting “as a homogenizing force, reuniting populations that might otherwise have had separate evolutionary trajectories.” This blending effect, which thwarts speciation, is prevented by some form of reproductive isolation dependent upon an internal or external barrier. The nature of the barrier varies with time. The present work reflects on the order in which barriers arise, an early barrier being superceded by a later.

The two fern types, having diverged 60 million years ago, can still transfer gametes abiotically (so there are no genic incompatibilities conferring prezygotic isolation), and the resulting zygotes can still develop (so there are no genic incompatibilities conferring hybrid inviability). Thus, they are reproductively isolated to a degree sufficient to prevent blending, solely by virtue of hybrid sterility.

Following a line of reasoning that dates back to Romanes (1886), a failure of meiotic pairing, due to the accumulation of base differences in parental DNA sequences (that would not necessarily affect genes), results in sterile hybrids. The line cannot advance, so that in evolutionary terms the parents are reproductively isolated from each other. Over 60 million years, this chromosome pairing barrier would have been elevated by addition of macroscopically observable deletions, duplications or inversions. Accompanying genic changes would have affected phenotypic functions other than those affecting fertilization and development. Thus secondary adaptations would be decoupled from the primary speciation event. This is in keeping with the recent conclusion of Hedges et al. (2015) that “if adaptation is largely decoupled from speciation, we should not expect it to be a driver of speciation” (see: http://1.usa.gov/1ESNM0W where further references may be found).

On 2015 Mar 12, Donald Forsdyke commented:

ADAPTATION DECOUPLED FROM SPECIATION. The authors begin by noting that gene flow among populations, rather than assisting speciation, has a “greater impact” when acting “as a homogenizing force, reuniting populations that might otherwise have had separate evolutionary trajectories.” This blending effect, which thwarts speciation, is prevented by some form of reproductive isolation dependent upon an internal or external barrier. The nature of the barrier varies with time. The present work reflects on the order in which barriers arise, an early barrier being superceded by a later.

The two fern types, having diverged 60 million years ago, can still transfer gametes abiotically (so there are no genic incompatibilities conferring prezygotic isolation), and the resulting zygotes can still develop (so there are no genic incompatibilities conferring hybrid inviability). Thus, they are reproductively isolated to a degree sufficient to prevent blending, solely by virtue of hybrid sterility.

Following a line of reasoning that dates back to Romanes (1886), a failure of meiotic pairing, due to the accumulation of base differences in parental DNA sequences (that would not necessarily affect genes), results in sterile hybrids. The line cannot advance, so that in evolutionary terms the parents are reproductively isolated from each other. Over 60 million years, this chromosome pairing barrier would have been elevated by addition of macroscopically observable deletions, duplications or inversions. Accompanying genic changes would have affected phenotypic functions other than those affecting fertilization and development. Thus secondary adaptations would be decoupled from the primary speciation event. This is in keeping with the recent conclusion of Hedges et al. (2015) that “if adaptation is largely decoupled from speciation, we should not expect it to be a driver of speciation” (see: http://1.usa.gov/1ESNM0W where further references may be found).