CrossRoads:
Have you seen 16 MHz crystals? I'm not sure any are that small, plus they need capacitors.
Resonators:
Crystals, Oscillators, Resonators | Electronic Components Distributor DigiKey
Look at some of the crystal packages here, see if any will be even close:
Electronic Components and Parts Search | DigiKey Electronics
A 2 SMD is 5mm x 3.2mm, that might fit.
You can right click the part on the Eagle design from the products page, see what package it says the resonator is. I posted the Uno parts last night, see if you can find that list.
Thanks, I'll check to see if I can find that list.
As far as crystal loading capacitors, I know they're supposed to be needed, but I've found that they seem to work just fine without the caps (and for that matter, without the 1 meg oscillator bias resistor).
I've tried "bare" crystals (no caps or resistor) with both low power and full swing fuse settings and they always work.
I can even stick my scope probe (at 10:1) on either crystal lead, on low power fuse, and see a nice sine wave of around 2.5 volts P-P (more or less centered between VCC and GND). With full swing fuse, I get the same waveform but with about 4.0 to 4.5 v P-P sine wave.
Since the crystal seems to be running solidly, even with a scope probe touching it, I'm quite sure it's working well and not just "barely working".
Besides, 20pF or so loading caps seem almost irrelevant when you think that the PCB and it's traces must present a several to tens of pF load just by being conductors (the traces) separated by an insulator (the board).
(edit to add): I guess I was right - the package is called "3-smd".
Unfortunately, the crystals have 4 leads instead of 2. The crystal itself is connected to diagonally opposite corners, and the other two pins are labelled "ground". I need to find a package that has only 2 pins!