Questions and postings pertaining to the usage of ImageMagick regardless of the interface. This includes the command-line utilities, as well as the C and C++ APIs. Usage questions are like "How do I use ImageMagick to create drop shadows?".
The -size setting is only used by some image creation and reading options. It is NOT use by montage.
You can not directly control montage output. You can control the size and the number of the tiles that go to make up the montage using -geometry and -tile options. The gap setting of the -geometry argument is used to set a border type spacing around each tile. -frame and -shadow also add extra spacing around each tile.
The nul: image tag is only used to add extra tiles to an image otherwise the miss use of geometry to generate a negative spacing also will clip the edges of the line of images. Just as a positive space adds extra space to each tile in the form of a border, negative spacing removed (shrinks) each tile.
The tile handling is only used for size and offset calculations when generate a canvas onto which the individual images are overlaid. This is the only reason the overlapping image example actually even works. It is in some ways much like how the more DIY "Programmed Positioning of Images Example" works... http://www.imagemagick.org/Usage/layers/#example
However not only are images processed and sized before overlaying, but the canvas size is also pre-calculated from the images, where as the above DIY example uses the newer '-layers merge" operator to do that canvas size calculation, and thus prevent any clipping of the images.
It would only take a little work to convert the above example into a more exact "montage" equivalent that does not 'clip' edges.