RELEASE NOTES 

		spatstat 1.16-2

		22 August 2009

	We thank Bert Loosmore for contributions.
	
OVERVIEW
	Code for calculating areas of dilations and intersections. 
	Extensions to the 'envelope' function.
	New generic function 'pixellate'.  Numerous minor improvements.

NEW FUNCTIONS

	dilated.areas
		Areas of the morphological dilations of a given set 
		by a sequence of radii.

	discpartarea
		Area of the intersection between a disc and another set.

	pixellate.owin
		Converts a window to a pixel image by computing the
		area of intersection between the window and each pixel

	plot.bermantest
		Plots the result of Berman's goodness-of-fit test.

	areaGain, areaLoss
		The existing function 'areadiff' has been renamed 'areaGain'.
		It now handles multiple points 'u' and multiple radii 'r'.
		There is now a counterpart function 'areaLoss'.

IMPROVEMENTS/ENHANCEMENTS

	envelope
		It is now possible to compute the sample variance
		of the simulated functions, and to construct 
		critical envelopes of the form
			sample mean +/- nSD * standard deviation

	envelope
		The argument 'simulate' can now be an object
		of class "envelope" produced by calling envelope()
		with savepatterns=TRUE. The saved point patterns
		will be extracted and used as the simulated patterns
		for computing the new envelopes. This makes it possible
		to generate envelopes of two functions (eg Kest and Gest)
		based on exactly the same set of simulated point patterns.

	as.im
		All methods for 'as.im' now have arguments 'eps' 
		and 'xy' offering additional control over pixel resolution.

	as.ppp
		Now accepts a numeric vector of length 2
		and interprets it as the coordinates of a single point.

	model.matrix.ppm
		Improved column names of matrix.

	pixellate
	pixellate.ppp
		The old function 'pixellate' has become 'pixellate.ppp',
		a method for the new generic function 'pixellate'.

	bermantest
		The result of bermantest is now of class "bermantest"
		and has its own plot method.
BUGS

	[1] rmh, rmh.default, rmh.ppm
		When a marked point process was simulated
		with fixed numbers of points of each type 
		(control$fixall = TRUE), a spurious error report
		    "Error in rmhResolveTypes"
		arose if start$x.start was specified.
		(Detected by Bert Loosmore)
		Fixed.

	[2] as.im.owin
		Did not correctly handle non-integer values
		of the arguments 'value' and 'na.replace'.
		Fixed.