We have observed galactic λ2.6mm and λ1.3mm CO absorption and emission along nine lines of sight toward compact extragalactic mm-wave continuum sources, using the IRAM Plateau de Bure Interferometer and NRAO 12m telescopes. In absorption we detected some two dozen kinematic components, and nearly every feature known from HCO
+ spectra (Lucas & Liszt,
1996A&A...307..237L) has a direct CO counterpart: a few CO lines are missing or very weak in emission even when τ(CO)>0.5. The column densities of CO and HCO
+ are well correlated, but not linearly related. The widths of the CO lines are typically 15% smaller than those of HCO
+ (on average 0.75km/s vs. 0.86km/s). We derive
12CO column densities 0.1≲N(
12CO)≲20x10
15cm
–2 which are in all cases very small compared to the column of carbon nuclei expected for 1 magnitude of visual extinction, even allowing for substantial depletion. The partial thermal pressure of H
2 is inferred to be 1000≤n(H
2)T
K≤12,000cm
–3.K, with a median p/k=3.2x10
3cm
–3.K. Thus the clouds are likely warm (T
K≃ tens of K), somewhat diffuse (n(H
2)≃50-300cm
–3), with the majority of the gas-phase carbon in the form of C
+ and perhaps even with a substantial fraction of H I in the thinnest cases. The isotope ratios in the CO usually differ strongly from the local interstellar ratio which we have separately measured in these clouds to be
12C/
13C=60 (Lucas & Liszt,
1998A&A...337..246L); we find 15≤N(
12CO)/N(
13CO)≤54, declining with increasing N(
12CO). The
13CO/C
18O ratio seen in emission or absorption is typically 25 (instead of 8) and C
18O is very difficult to detect in emission even when T
R*(
12CO)/T
R*(
13CO)<10. Apparently, the relative abundance of
13CO is typically greatly enhanced, even at very low extinction, and never diminished by selective photodissociation. One effect of this enhancement is that lines of
12CO are substantially less optically thick than might otherwise have been inferred. There is little evidence for a general selective depletion of C
18O.