[R] Connecting lines across missing data points, xyplot

[Carl Witthoft]

Just had to comment that I thought this question was unintentionally 
funny,  seeing as many of us have gone thru a lot of pain to force Excel 
NOT to join lines across NA-type points in our data.

Back to the subject:  if you use is.na(),  just be careful to remove the 
corresponding values in the x-data as well as the y-data :-) .  Probably 
mucking with something like
  >if (y[j]=NA) y[j]<-approx({local x and y data fit})