MASSIVE STARS LIE AT THE CENTER OF THE WEB OF PHYSICAL PROCESSES THAT HAS SHAPED THE UNIVERSE AS WE KNOW IT GOVERNING THE EVOLUTION OF THE INTERSTELLAR MEDIUM OF GALAXIES PRODUCING A MAJORITY OF THE HEAVY ELEMENTS AND THEREBY DETERMINING THE EVOLUTION OF GALAXIES. MASSIVE STARS ARE ALSO IMPORTANT AS SIGNPOSTS SINCE THEY PRODUCE MOST OF THE LIGHT AND ALMOST ALL THE IONIZING RADIATION IN REGIONS OF ACTIVE STAR FORMATION. A SIGNIFICANT FRACTION OF ALL STARS FORM IN MASSIVE CLUSTERS WHICH WILL BE OBSERVABLE THROUGHOUT THE VISIBLE UNIVERSE WITH JWST. THEIR LUMINOSITIES ARE SO HIGH THAT THE PRESSURE OF THEIR LIGHT ON INTERSTELLAR DUST GRAINS IS LIKELY THE DOMINANT FEEDBACK MECHANISM REGULATING THEIR FORMATION. WHILE THIS PROCESS HAS BEEN STUDIED IN THE LOCAL UNIVERSE MUCH LESS ATTENTION HAS BEEN FOCUSED ON HOW IT BEHAVES AT HIGH REDSHIFT WHERE THE DUST ABUNDANCE IS MUCH LOWER DUE TO THE OVERALL LOWER ABUNDANCE OF HEAVY ELEMENTS. THE HIGH REDSHIFT UNIVERSE ALSO DIFFERS FROM THE NEARBY ONE IN THAT OBSERVATIONS IMPLY THAT HIGH REDSHIFT STAR FORMATION OCCURS AT SIGNIFICANTLY HIGHER DENSITIES THAN ARE TYPICALLY FOUND LOCALLY. WE PROPOSE TO SIMULATE THE FORMATION OF INDIVIDUAL MASSIVE STARS FROM THE HIGH REDSHIFT UNIVERSE TO THE PRESENT DAY UNIVERSE SPANNING METALLICITIES RANGING FROM 0.001 TO 1.0 AND COLUMN DENSITIES FROM 0.1TO 30.0 G/CM2 FOCUSING ON HOW THE PROCESS DEPENDS ON BOTH THE DUST ABUNDANCE AND ON THE DENSITY OF THE STAR-FORMING GAS. THESE SIMULATIONS WILL BE AMONG THE FIRST TO TREAT THE FORMATION OF POPULATION II STARS WHICH FORM IN REGIONS OF LOW METALLICITY. BASED ON THESE RESULTS WE SHALL THEN SIMULATE THE FORMATION OF CLUSTERS OF STARS ACROSS ALSO COSMIC TIME BOTH OF MODERATE MASS SUCH AS THE ORION NEBULA CLUSTER AND OF HIGH MASS SUCH AS THE SUPER STAR CLUSTERS SEEN IN STARBURST GALAXIES. THESE STATE-OF-THE-ART SIMULATIONS WILL BE CARRIED OUT USING OUR NEWLY DEVELOPED ADVANCED TECHNIQUES IN OUR RADIATION-MAGNETO-HYDRODYNAMIC AMR CODE ORION FOR RADIATIVE TRANSFER WITH BOTH IONIZING AND NON-IONIZING RADIATION THAT ACCURATELY HANDLE BOTH THE DIRECT RADIATION FROM STARS AND THE DIFFUSE INFRARED RADIATION FIELD THAT BUILDS UP WHEN DIRECT RADIATION IS REPROCESSED BY DUST GRAINS. OUR SIMULATIONS INCLUDE ALL OF THE RELEVANT FEEDBACK EFFECTS SUCH AS RADIATIVE HEATING RADIATION PRESSURE PHOTODISSOCIATION AND PHOTOIONIZATION PROTOSTELLAR OUTFLOWS AND STELLAR WINDS. THE CHALLENGE IN SIMULATING THE FORMATION OF MASSIVE STARS AND MASSIVE CLUSTERS IS TO INCLUDE ALL THESE FEEDBACK EFFECTS SELF-CONSISTENTLY AS THEY OCCUR COLLECTIVELY. WE ARE IN AN EXCELLENT POSITION TO DO SO. THE RESULTS OF THESE SIMULATIONS WILL BE DIRECTLY RELEVANT TO THE INTERPRETATION OF OBSERVATIONS WITH JWST WHICH WILL PROBE CLUSTER FORMATION IN BOTH THE NEARBY AND DISTANT UNIVERSE AND WITH SOFIA WHICH CAN OBSERVE HIGH-MASS STAR FORMATION IN THE GALAXY. WE SHALL MAKE DIRECT COMPARISON WITH OBSERVATIONS OF MASSIVE PROTOSTARS IN THE GALACTIC DISK. WE SHALL ALSO COMPARE WITH OBSERVATIONS OF STAR CLUSTERS THAT FORM IN DENSE ENVIRONMENTS SUCH AS THE GALACTIC CENTER AND IN MERGING GALAXIES (E.G. THE ANTENNAE) AND IN LOW METALLICITY ENVIRONMENTS SUCH AS THE DWARF STARBURST GALAXY I ZW 18. ONCE OUR SIMULATIONS HAVE BEEN BENCHMARKED WITH OBSERVATIONS OF MASSIVE PROTOSTARS IN THE GALAXY AND MASSIVE PROTOCLUSTERS IN THE LOCAL UNIVERSE THEY WILL PROVIDE THE THEORETICAL BASIS FOR INTERPRETING OBSERVATIONS OF THE FORMATION OF MASSIVE STAR CLUSTERS AT HIGH REDSHIFT WITH JWST. WHAT DETERMINES THE MAXIMUM MASS OF A STAR? HOW DOES STELLAR FEEDBACK AFFECT THE FORMATION OF INDIVIDUAL STARS AND THE FORMATION OF MASSIVE STAR CLUSTERS AND HOW THE ANSWERS TO THESE QUESTIONS EVOLVE WITH COSMIC TIME. THE PROPOSED RESEARCH WILL PROVIDE HIGH-RESOLUTION INPUT TO THE STUDY OF STELLAR FEEDBACK ON GALAXY FORMATION WITH A SIGNIFICANTLY MORE ACCURATE TREATMENT OF THE PHYSICS PARTICULARLY THE RADIATIVE TRANSFER THAT IS SO IMPORTANT FOR FEEDBACK.