On Sept.  30, William Kirk requested information concerning the effects
of pollen size on the efficiency of transfer by bees.  In fact, there
seems to be little information on any biological aspect of pollen size
(other than the correlation with style length which has been known for
200 years, but seems not to be universal).  I have found correlations
between pollen size and removal from three plant species, but they
confuse, rather than elucidate (see Ecology 71:1110-1125).  In
particular, removal increased with pollen size for Mertensia paniculata
(Boraginaceae) but decreased with size for two congeners with rather
different flowers (Pedicularis contorta and P.  bracteosa:
Scrophulariaceae).  I suspect that a correlation with some unmeasured
variable may be involved.
 
To further cloud the issue, I have recently found that removal from
Pontederia cordata (Pontederiaceae) is affected by pollen size (at least
for mid- and long-level anthers - this species is tristylous), but the
effect depends on the time of day!  Pollen size increases by 1 um during
the day, presumably due to hydration.  Size does not affect removal
during the morning, but during the afternoon large pollen is more
recalcitrant to removal.  This is too bizarre even for a bee joke.
 
William, I hope this helps (?).