I did not present anything on the 2011 China plant systematics conference.  That may be a little pity. However, I still get reward.  Some of presentations provide me inspiration for my thinking on biogeography and evolution.  One of such inspiration is regarding the dispersals.  By looking at several disjunction patterns of plant groups created by rivers captures, Prof. Sun summarized that dispersal seems not as easy as people thought.  And he also mentioned that there are actually many other examples which may show similar pattern. His evidence is solid and obviously trustable.  I think I have no reason to reject the conclusion that dispersal is not that easy because many plant groups show genotypic pattern of vicariance.  But that makes me think that it does not mean dispersal we need deny the importance of dispersal.  It is true that geneflow are not possible for many groups because they can not dispersal through the barrier after the events of vicariance (the revier capture). But we still do not know what does “many” mean?  I guess part of plants still can cross the barrier and make their distribution complex enough. AND because dispersal is hard, we can still see some pattern created by vicariance.  However, if look at the whole flora of the region, I bet “many” of the plants crossed the barrier. I do not know what are the methods of the crossing.  But as whole, part of them may be a results of “chance”.  This could be also true for long distantly disjucnted taxa such as the transocean ones.  Because dispersal is so hard so that we do not see that many disjunctions between asia and southamerica.  If vicariance is responsible for the long distance disjunctions, too much extinctions are needed to explain the pattern because for many of the cases, it is recorded for the case itself but not for a broad relatives of the group. And because dispersal is hard enough, we do not record groups jump forward and back.  It may be facilitated by chance by air, ocean currents, and could not be possible to come back again.