Kepler-105 has two planets that straddle the exoplanet “radius gap,” but in a rare architecture
with the larger planet (Kepler-105b) closer to the host star. This architecture makes Kepler-105
a particularly good system to test the limits of photoevaporation. Photoevaporation is a process
where X-ray and ultraviolet (XUV) radiation from host stars strips away the atmospheres of planets.
This process is one of the common explanations for the observed exoplanet radius gap: a paucity
in the occurrence rate of exoplanets between approximately 1.5R? and 2.0R?. The closer a planet
is to its host star, the more high-energy photons it receives. It is therefore surprising that Kepler-
105b managed to retain a significant gaseous envelope, while its more distant companion (Kepler-
105c) did not. To test whether photoevaporation can explain the unusual architecture of Kepler-
105, we collected 92 radial velocity (RV) observations with the High Resolution Echelle Spectrometer
(HIRES) on Keck I and combined them with the transit timing variations (TTVs) produced by Kepler
cadence photometry from Q1-Q17. Using this combined RV and TTV analysis, we measure masses
of 10.8 ± 2.3M? and 5.6 ± 1.2M? for planets Kepler-105b and Kepler-105c, respectively. We find
with 74% confidence that the Kepler-105 planets are consistent with atmospheric mass-loss primarily
sculpted by photoevaporation. Thus, even in a testbed case for probing the limits of photoevaporation,
we find that it is more likely than not that the Kepler-105 planetary architecture is consistent with
photoevaporation.