With the results from CGRO, the number of gamma-ray pulsars has risen from 2 to at least 7, and several surprises have come to light. The new detections show that pulsars can be remarkably efficient producers of GeV photons, with 10% or more of the energy in the pulsed emissions for 10⁵-10⁶ year old pulsars. Furthermore, the detection of Geminga as a radio-quiet pulsar shows that radio and high-energy pulsars are overlapping subsets of the neutron star population, but with quite different beam patterns. The gamma-ray observations thus provide a complementary (and apparently more complete) sample of the young pulsar population. Although Geminga remains a sample of one, radio-quiet pulsars are expected to represent a sizable fraction of the unidentified high-energy gamma-ray sources. Thus studying the gamma-ray sample will greatly advance our understanding of the neutron star birthrate (and its relationship to the supernova rate). The site of the pulsar particle acceleration and the gamma radiation is still under investigation. Theoretical modeling has focused on acceleration at the polar caps and in vacuum regions in the outer magnetosphere. These models have advanced to the point where pulse profiles, luminosities, and spectral variations with pulsar phase can be computed; comparison with CGRO data on the brightest objects (Crab, Vela, and Geminga) have provided significant constraints. What these comparisons make clear is that the GeV emission directly probes the dynamics and geometry of the particle acceleration region where electron/ positron energies are inferred to exceed 10 TeV. A unique attraction of pulsar modeling arises from the fact that rotation brings different regions of the acceleration zone into view during the pulse; with sufficient statistics and energy coverage the rich temporal structure in pulsar spectra allow a tomographic analysis of physical conditions in the magnetospheric particle accelerator.  Finding more gamma-ray pulsars, both radio-loud and radio-quiet, will be essential to answering the outstanding questions. A crucial test of pulsar models will be their ability to predict which radio pulsars will be detected as gamma-ray pulsars in the future. When a source is a known radio pulsar, sensitive pulse searches and measuring high quality phase resolved spectra benefit from very long exposures. Thus the most important attributes of a future high-energy gamma-ray mission are large effective area coupled with large field-of-view. When a source is not identified with a radio pulsar, finding the pulsed emission directly in the gamma-ray data requires a high count rate. In both cases, high angular resolution (< 10') at GeV energies will be very important for isolating sources from the bright galactic background. Further, arcminute positions in the GeV range will enable powerful searches for counterparts with imaging X-ray telescopes and ground instruments. Such counterpart searches offer the best means to trace the origin of the galactic plane sources. To untangle the physics of the detected pulsars, high quality phase resolved spectra are crucial. Particularly important are extension of the sensitive range above 10 GeV where the pulses merge with unpulsed plerionic emission (and where Compton scattered photons may dominate the pulsed signal) and below 10 MeV where existing observations require a break from the flat GeV spectra and important phase variability is expected in many models.

Objectives: