We have observed the most distant ( = 0.829 ) cluster of galaxies in the Einstein Extended Medium Sensitivity Survey ( EMSS ) , with the ASCA and ROSAT satellites .
We find an X-ray temperature of 12.3 ^ { +3.1 } _ { -2.2 } { keV } for this cluster , and the ROSAT map reveals significant substructure .
The high temperature of MS1054-0321 is consistent with both its approximate velocity dispersion , based on the redshifts of 12 cluster members we have obtained at the Keck and the Canada-France-Hawaii telescopes , and with its weak lensing signature .
The X-ray temperature of this cluster implies a virial mass \sim 7.4 \times 10 ^ { 14 } h ^ { -1 } M _ { \odot } , if the mean matter density in the universe equals the critical value ( \Omega _ { 0 } = 1 ) , or larger if \Omega _ { 0 } < 1 .
Finding such a hot , massive cluster in the EMSS is extremely improbable if clusters grew from Gaussian perturbations in an \Omega _ { 0 } = 1 universe .
Combining the assumptions that \Omega _ { 0 } = 1 and that the intial perturbations were Gaussian with the observed X-ray temperature function at low redshift , we show that this probability of this cluster occurring in the volume sampled by the EMSS is less than a few times 10 ^ { -5 } .
Nor is MS1054-0321 the only hot cluster at high redshift ; the only two other z > 0.5 EMSS clusters already observed with ASCA also have temperatures exceeding 8 keV .
Assuming again that the initial perturbations were Gaussian and \Omega _ { 0 } = 1 , we find that each one is improbable at the < 10 ^ { -2 } level .
These observations , along with the fact that these luminosities and temperatures of the high- z clusters all agree with the low- z L _ { X } - T _ { X } relation , argue strongly that \Omega _ { 0 } < 1 .
Otherwise , the initial perturbations must be non-Gaussian , if these clusters ’ temperatures do indeed reflect their gravitational potentials .