Why do the ice-cubes from the bottom ice-tray fracture? 
 When emptying stacked ice-cube trays from the freezer, the ice-cubes
in the tray on the bottom is always fractured and difficult to free.
It is not caused from a specific tray, if I switch the trays, the
ice-cubes from the bottom tray still comes out fractured, and if I lay
the trays side by side, the ice-cubes all are fine, ie: solid,
semi-transparent, and easily "popped" free of the tray. Could it have
something to do with the cold mass of the upper tray above?

This explanation seems plausible to me:

"Wwater will expand when it freezes. Ice made in plastic trays will
have a little mound on the top of each cube because the top edges will
freeze first, and then as the rest of the water solidifies it will
push up through the 'weaker' center. This may push the h2o up against
the bottom of the other tray, causing fractures in the crystaline
structure of the ice. When you try to pop the cubes out, they will
separate at the fractures, making "broken bits" and irking you."

http://www.chowhound.com/topics/show/312355

I will likely regret going against a Pinkfreud statement but my guess
has to do with speed of freezing. Faster freezing means larger crystal
structures and weaker cleavage lines.

Or is it vice versa?

Just a SWAG.

I thought that faster freezing meant smaller crystal. Less cell damge
with liq. N2 than -20 freeze.

"Could it have something to do with the cold mass of the upper tray above?"

I think it could have more to do with the LESS cold mass above when
you put stacked ice trays in the freezer.  The bottom tray sits on the
shelf with the cooling liquid tubes under it, freezing the water at
the bottom quite fast.  The water in the tray above keeps the water at
the top of the bottom tray from freezing as quickly as it would
without the tray on top, upsetting the formation of ice, as Mark
suggests.

Ice is formed first in the top layer of water, and as every one knows
ice has more volume than water.. on top of the tray as water freezes,
it has room to expand and expands forming ice, no cracks, once the top
layer changes to ice,
the bottom layer slowly changes to ice but the room for expansion is
nil so the ice starts to break struggling against the top, causing
cracks in botto, part of tray, the thickly fit top often stays strong
and may bulge out or move up but does not break,

This is a simple materials problem which is often overlooked. Some of
the comments are partially correct. When your water freezes it tends
to freeze from the outside inwards. Hence the surfaces (top, bottom
and sides will freeze first. Ice has a lower density and hence bigger
volume (the reason it floats). THis means that when frozen it expands.
So the water in the middle of your ice cube is trying to expand but is
prevented from doing so by the freezing of the outer surfaces. This
tends to build up very high stresses in the ice cuber which are
releived by cracking with slight temperature changes or when the ice
cube is relased from the mould. The bottom trays tend to freeze first
due to being in contact with the cooling compartment and the heat
rising to the other trays. Hence the bottom trays have the most built
in stress.
So there you go.

Knickers provides an elegant explanation of the concept. But "if I lay
the trays side by side, the ice-cubes all are fine, ie: solid,
semi-transparent" In this situation, the surface area (and by
extension cooling rate)is increased, with less evidence of "cracking
with slight temperature changes or when the icecube is relased from
the mould"