By eliminating the local neutron star explanation for gamma-ray bursts (GRBs) favored for twenty years, BATSE observations have elevated the origin of GRBs to one of the great outstanding problems in astrophysics. GRB sources are now thought to be distributed either in a large halo (or corona) around the Galaxy or else at cosmological distances. Discriminating between these possibilities and setting the GRB distance scale, now undetermined to five orders of magnitude, is a pivotal observational question unlikely to be solved by continued BATSE observations. Even if BATSE continues observing GRBs for another five years, its statistics will be insufficient to resolve this issue.

Coronal halo models of GRBs predict similar burst halos around other massive spiral galaxies. Andromeda is an imaging hard X-ray/soft gamma-ray instrument designed with sufficient sensitivity and sky coverage to detect such a corona if it exists around M31. The sensitivity and field of view are optimized so that coronal distributions which satisfy the angular moment constraints of BATSE typically predict an excess of ~ 30 bursts detectable by Andromeda towards M31 during the 1 year mission, relative to a control field. The absence of any excess towards M31 would prove that most GRBs are cosmological. In this case, our high-sensitivity observations will determine the shape of the number vs. flux distribution in a region well beyond BATSE's detection threshold, yielding an improved estimate of the GRB luminosity distribution and source redshifts.

Andromeda is a coded aperture gamma-ray telescope consisting of a hexagonal coded mask coupled to an alkali-halide imaging scintillation detector, a flight-proven technology adapted from the balloon-borne Caltech Gamma-Ray Imaging Payload (GRIP). A 2000 sq cm phoswich detector combines large area with background rejection, making Andromeda more sensitive that any gamma-ray imaging instrument previously flown. Andromeda is currently being considered for launch through the Department of Energy and will be proposed to NASA's Small Explorer program.

==============================================================
Energy range        10 - 200 keV                              
Energy resolution   19% FWHM @ 60 keV                         
Collecting area     2320 sq cm                                
Effective area      2000 sq cm @ 30 keV                       
Mask transparency   50%                                       
Sensitivity         3.4 x 10e-6 photon/(sq cm)/sec/keV        
  (continuum)       (3 sigma in 2 x 10e4 s, (delta E)/E = 50%)
Sensitivity         1.6 x 10e-9 erg/(sq cm)/s                 
  (transient)       (5.5 sigma in 10 s)                       
                    (10 - 200 keV, photon index = 2)          
Field of view       10 deg FWHM                               
Angular resolution  1.5 deg                                   
Timing capability   100 microsec (relative)                   
==============================================================