[R] null plotting symbol ?

[Prof Brian Ripley]

On Tue, 6 May 2003 partha_bagchi at hgsi.com wrote:

> pch = 26 seems to produce a blank symbol.

More accurately, it plots nothing.  (A `blank' symbol' suggests to me that
it plots in the background colour.)  That is true *currently* (26-31 are
unassigned), but there is no guarantee that will remain so.

Why would one use that rather than NA, whose meaning is clear?

In brief: what exactly was your point?  Were you seriously suggesting 
using pch=26?

> Prof Brian Ripley <ripley at stats.ox.ac.uk>
> Sent by: r-help-bounces at stat.math.ethz.ch
> 05/05/2003 04:39 PM
> 
>  
>         To:     tblackw at umich.edu
>         cc:     r-help at stat.math.ethz.ch
>         Subject:        Re: [R] null plotting symbol ?
> 
> 
> NAs in pch cause the corresponding points to be omitted.
> 
> On Mon, 5 May 2003 tblackw at umich.edu wrote:
> 
> > I am calling  plot()  with argument pch as a vector of numeric
> > symbol codes, the same length as x and y.  Is there some code
> > which produces no symbol - a blank - so that I can come back
> > with a second call to  points()  and fill in these locations
> > using a different fill color and a different symbol size ?

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595