blueman wrote:
I'm building a bookcase with adjustable shelves with a 46" span ...
I am making the shelves 1.5" thick (double 3/4" plywood).

- Can I assume that all other things being equal, 1/4" pins will support
a heavier load than 5mm pins (since the diameter is larger)?

Some points haven't been covered by other replies; first, in terms of
stiffness, your
double 3/4" plywood has the same stiffness as double 1/4" plywood with
a center
space (glued lightweight ribs or hardwood at the pin-stress points) of
about 1.25 inch.
Overall height of the sandwich is then 1.75 i nch... but there's a
lot less shelf
weight and cost.

Also, the pins are stressed downward at the face of the upright, BUT
they tilt slightly
and the pins actually press upward at their deepest point, so your load
is carried on the
surface area of (roughly) the width of the pin times half its depth in
the upright. That is
why a pressed-in collar is such a good idea (the area of the collar is
larger, and if the
tilt of the pin doesn't deform the collar, the useful area is doubled
by comparison).

For heavy timbers, where stresses are near the material limits, pins
aren't used;
instead one makes a circular cut and presses a ring (like a pipe
section) into it. The shelf
depth would tolerate 3/4" pipe sections instead of little pins, and
with suitable (hardwood)
trim at the composite plywood endcaps, Forstner drill/routing a recess
to hide the pin
would be easy enough.

I've always hated high-stress points, like metal pins in wood, because
age, accidents,
and changes in moisture can crush the wood so easily. Even if it
doesn't fail soon
enough to catch in a test, it WILL fail.