We propose to monitor one extreme BL Lac object (0716+714 or 1749+701,
depending on launch date) for 12 days with a sampling of two
observations per day. Both objects are known to show strong intraday
varibility both in the radio and optical wavelength domains; for
0716+714 there are strong indications that these variations are
correlated between the radio and optical. Similar correlations seem
to exist in 1749+701 and a number of other sources, but are extremely
difficult to prove without coverage of the infrared range. Rapid
variations present a serious challenge to the standard relativistic
jet model, since they imply either Doppler factors as high as 200, or
exotic (e.g. stimulated) emission mechanisms.

By obtaining light curves at 6 wavelengths ranging from 5 to 160 micron
we will be able to answer the following questions:

What is the wavelength dependence of the variability amplitude? Is
there a break at mid-IR wavelengths?

Are there correlated variations through the radio - IR - optical range?

Are there time delays between the light curves at different
wavelengths?

To what extent (if any) are the radio light curves affected by inter-
stellar scattering, which might mimic intrinsic variations?

We will thus be able to model in much more detail than presently
possible the physical conditions in the compact jets of blazars, and
to understand the radiation mechanisms responsible for the emission
coming from these regions.