There are three groups of routines in the ellipse package.  The first
consists of those involved with plotcorr, which implements the plots
described in Murdoch and Chow (1996). These display correlations using
ellipses, whose shape is that of the contours of a bivariate normal
distribution with matching correlation.
    
The second group implements a version of the profile plots described in
Bates and Watts (1988), Nonlinear Regression Analysis and its
Applications.

The last group provide the basis for the others, drawing ellipses based on
various S objects, including scalar correlations, covariance matrices,
arima, lm, and nls fits.

Reference:
Murdoch, D.J. and Chow, E.D. (1996).  A graphical display of
large correlation matrices. {\em The American Statistician} {\bf
50},178-180.

ellipse                  Generic function to make an ellipse
ellipse.default          Produces an ellipse from a covariance matrix or
                         a correlation
ellipse.arima0           Outline approximate pairwise confidence region
                         for arima0 object
ellipse.glm              Outline an approximate pairwise confidence
                         region for a glm object          
ellipse.lm               Outline a pairwise confidence region for a
                         linear model fit object
ellipse.nls              Outline an approximate pairwise confidence
                         region for an nls object
ellipse.profile          Profile sketch
ellipse.profile.nls      Profile sketch	for nonlinear models
ellipse.profile.glm	 Profile sketch for GLM models
pairs.profile	         Profile pairs 
plotcorr                 Plot correlation matrix ellipses
