Bivariate Binning into Hexagonal Cells
DESCRIPTION:
Creates an object of class "hexbin". Its basic components are a cell identifier and a count of the points falling into each occupied cell.
USAGE: 
hexbin(x, y, xbins=30, shape=1, xlim=range(x), ylim=range(y))
REQUIRED ARGUMENTS:
x:
numeric vector. Usually the first (horizontal) coordinate of bivariate data to be binned.
y:
numeric vector. Usually the second (vertical) coordinate of bivariate data to be binned.
OPTIONAL ARGUMENTS:
xbins:
number of bins partitioning the range of x values.
shape:
height to width ratio for the resulting plotting region. This parameter is used in determining the number of bins in the y-direction given xbins, the number of bins in the x-direction. The default value shape = 1 makes the hexagons appear equal-sided when plotted.
xlim:
the horizontal limits of the binning region in units of x. By default these are the minimum and maximum values of x.
ylim:
the vertical limits of the binning region in y units. This defaults to the minimum and maximum values of y.
VALUE:
an object of class "hexbin". This is a data frame with the following columns:
cell:
vector of cell identifiers that can be mapped into the bin centers in data units
count:
a vector with the points count for each corresponding cell.
xcenter:
the x center of mass (average of x values) for the cell.
ycenter:
the y center of mass (average of y values) for the cell.

The returned data frame has attributes:

class:
the class of the returned object, "hexbin".
call:
the original call to hexbin which generated the object.
xbins:
number of hexagons along the x axis. Same as the input value for xbins. Hexagons inner diameter equals diff(xlim)/xbins in x units
dims:
the i-th and j-th limits of count if treated as a matrix (count[i,j]).
xlim:
same as the input value for xlim.
ylim:
same as the input value for ylim.
shape:
same as the input value for shape.
DETAILS:
Returns counts for non-empty cells. The plot shape must be maintained for hexagons to appear with equal sides. Calculations are in single precision.
REFERENCES:
Carr, D. B., Littlefield, R. J., Nicholson, W. L. and Littlefield, J. S. (1987). Scatterplot matrix techniques for large N. Journal American Statistical Association, 83, 424-436.
SEE ALSO:
summary.hexbin , plot.hexbin , identify.hexbin , smooth.hexbin, erode.hexbin , cell2xy , xy2cell .
EXAMPLES: 
x <- rnorm(10000)
y <- rnorm(10000)


bin1 <- hexbin(x,y)
trellis.device(motif)
plot(bin1,style="nested.centroids")


# Lower resolution binning and overplotting with counts
bin2 <- hexbin(x,y,xbins=25)
binpar <- plot(bin2, style="latt" ,minarea=1, maxarea=1, density=0,
        border=T)
oldpar <- par(binpar)     # reset graphics to the plot on the screen
xy <- cell2xy(bin2)
text(xy$x, xy$y, as.character(bin2$count), adj=.5, cex=.3)
par(oldpar)     # restore old graphics parameters


maples <- lansing[lansing$species=="maple",]
plot(maples$x,maples$y)
maple.bin <- hexbin(maples$x,maples$y,xbins=10)
hexagons(maple.bin,border=T,dens=0)