Future progress in understanding star formation in galaxies will require detailed studies of individual regions as well as a statistical examination of suitably chosen samples. Very few galaxies have angular sizes large enough that they can be mapped by IRAS. M31 has been described by Habing at al. (1984); a few others, including M33, are being presented by Walter Rice (this volume). Wainscot, de Jong, and Wesselius (1986) have used the chopped photometric channel on IRAS to produce scans along the disks of several edge-on galaxies. In the case of NGC 891 they find good agreement among the 50 µm, 21 cm continuum, and CO profiles, all of which are considerably narrower than the 21-cm H I emission.
The best far-infrared resolution achieved to date on an extragalactic source is the study of the central regions of M51 by Lester, Harvey, and Joy (1986) using a 24 arcsec slit in the Kuiper Airborne Observatory. They find that much of the far-infrared luminosity of this galaxy comes from a small, sharply bounded region within a 700 pc radius of the nucleus. Finer detail than this can only be achieved by moving to shorter wavelengths and using large ground-based telescopes such as the IRTF. Maps at 10 µm of the central regions of galaxies such as NGC 2903 (Wynn-Williams and Becklin 1985), and NGC 3310 (Telesco and Gatley 1984) tend to show agreement in general, but not in detail, between the regions of strongest radio, infrared, and optical emission. Much more mapping at infrared wavelengths will be needed to disentangle the effects of dust extinction and of variations in the stellar formation histories in these regions.
Infrared spectroscopy is another area where important contributions can be expected from ground-based observations in the next few years. Roche and Aitken (1985) have shown that the 8-13 µm spectra of spiral galaxy nuclei show much stronger 11.6 µm features than are seen in H II regions in our Galaxy. It remains to be seen whether this effect is also seen in extragalactic spiral-arm star formation regions.