We report the discovery of a massive planet OGLE-2008-BLG-355Lb .
The light curve analysis indicates a planet : host mass ratio of q = 0.0118 \pm 0.0006 at a separation of 0.877 \pm 0.010 Einstein radii .
We do not measure a significant microlensing parallax signal and do not have high angular resolution images that could detect the planetary host star .
Therefore , we do not have a direct measurement of the host star mass .
A Bayesian analysis , assuming that all host stars have equal probability to host a planet with the measured mass ratio implies a host star mass of M _ { h } = 0.37 _ { -0.17 } ^ { +0.30 } M _ { \odot } and a companion of mass M _ { P } = 4.6 ^ { +3.7 } _ { -2.2 } ~ { } M _ { J } , at a projected separation of r _ { \perp } = 1.70 ^ { +0.29 } _ { -0.30 } AU .
The implied distance to the planetary system is D _ { L } = 6.8 \pm 1.1 kpc .
A planetary system with the properties preferred by the Bayesian analysis would be a challenge to the core-accretion model of planet formation , as the core-accretion model predicts that massive planets are far more likely to form around more massive host stars .
This core accretion model prediction is not consistent with our Bayesian prior of an equal probability of host stars of all masses to host a planet with the measured mass ratio .
So , if the core accretion model prediction is right , we should expect that follow-up high angular resolution observations will detect a host star with a mass in the upper part of the range allowed by the Bayesian analysis .
That is , the host would probably be a K or G dwarf .