As a followup to a previous posting regarding the win.metafile() function, I've been having some problems as well. In my case, I've been converting some scripts from S-SPLUS to R that create graphs in order to take advantage of R's plotmath capabilities.
In the sample code shown below, the axis labels display the units for the data. In particular, the units for the y axis are "grams carbon per meter squared per month" or "gC m-2 m-1". I have noticed that using the following code, if one prints directly from the graph window, the plot prints correctly with the proper spacing between the "gC", "m-2", and "m-1". However, when I use either the "copy as metafile" feature of the Windows RGui or add a direct call to win.metafile(), the subsequent .wmf file does not have spaces between the components of the axis labels. #win.metafile(filename= "Rplot%02d.emf", height=11, width=8.5, pointsize=12) par(mar= c(5,4,4,5)+0.1) x <- 1:10 y <- 0.8*x+1 plot(x, y, type="n", axes=F, xlab="", ylab="") box() points(x, y, pch=1, col=1, cex=1.2) axis(side=2) mtext(side=2, line=2.5, expression(paste("monthly (gC ", m^-2, " ", m^-1, ")"))) axis(side=1, at=1:12) par(new=T)#, xaxs="d") plot(x, y, type="n", axes=F, xlab="", ylab="") y <- 0.5*x-1 points(x, y, pch=16, col=1, cex=1.2) axis(side=4) mtext(side=4, line=3, expression(paste("cumulative (gC ", m^-2, ")"))) I have been working with R 1.8.0 on a computer runing Windows XP. However, I've also tried the code under R 1.7.1, and the behavior appears to be the same. Also, the "correct" spacing when printing the graph window directly does not appear to be related to what kind of print driver is used because both PostScript and PCL drivers print correctly. Moreover, I've also tried to paste the metafile into a program other than MS Word, but the results are the same as with MS Word. So I am guessing that the lack of spaces between the axis label components is related to the creation of the windows metafile. A workaround is that an extra space can be added in the call to expression(), but it certainly would be nice not to need two different statements that depended on whether the graph was printed directly or saved as a metafile. So could this be a bug, or is it some sort of known limitation when creating metafiles? Thanks for your time and consideration, -Paul Schwarz ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help