Nikolai has written to me from Denmark to tell me something I didn't know. He wrote: Hi D, Excellent method for making box and whisker diagrams (boxplots) in Excel. Regarding the missing cross, it’s no big deal really since the dash is preferable to the cross as a median symbol. Anyway, in order to see the cross and other ‘missing’ symbols, background colour has to be set to ‘no colour’ and voilà, all missing markers are now available. The diagrams will look more professional if the markers for Q1 and Q3 are omitted, that is set marker to none. And to make the box plot statistically correct the whiskers should not extend to the minimum and maximum values but to the smallest and largest observations within 1,5*IQR (interquartile range, Q3-Q1). Observations between 1,5*IQR and 3*IQR are termed mild outliers and are marked with a circle (for example), whereas observations that fall outside of 3*IQR are termed extreme outliers and are marked with a cross. Creating a box plot that reflects true IQR, mild and extreme outliers will demand a bit more manual work as min and max observations will have to be compared to <1,5*IQR and then set as the range of the whiskers, and if there are observations >1,5 and 3*IQR, they will have to be included in the data table used for the box plots. Regards, Nikolai Graae I checked my work and found that all of my sources agreed with my method and I with theirs. Then I found a book that I have that agrees with Nikolai. So, very soon there will be a BoxPlots Revisions Page that will explain what Nikolai has shown me, how to use his knowledge and what it all means. Watch this space ... DW