- Measurements of the stellar mass in galaxies to increasingly high redshifts provide us one of the most powerful ways of gauging early galaxy build-up. Combining ultra- deep IRAC imaging observations with deep WFC3/IR observations have allowed astronomers both to identify and estimate masses for sources out to redshifts as high as z~8. However, it has been difficult to extend such studies to even higher redshifts due to the incredible faintness and rarity of typical z~9-10 galaxies. One method for stepping beyond the normal S/N limitations is to take advantage of gravitational lensing by massive galaxy clusters to magnify faint z>8 sources, but to do so, we must know where the magnified sources lie. Fortunately, utilizing the huge investment of HST time in the 524-orbit CLASH program, we have been able to identify 100s of z~5-8 galaxies and even 4 extremely tantalizing z~9-11 galaxy candidates. To extend our measurement of galaxy masses to the earliest possible times, we require ultra-deep Spitzer observations on these candidates to accurately measure their rest-frame optical properties. While two of the z~9-11 candidates are the subject of deep IRAC observations as a result of an approved program, the other two will not be the subject of such observations. Given the considerable galaxy-to-galaxy variations in the mass-to-light ratio, ultra-deep Spitzer observations are needed for all four z~9-11 candidates to establish the typical properties of galaxies in this epoch. We propose to obtain such observations over the two z~9 candidates lacking such observations. Not only will our proposed observations substantially improve our ability to estimate the stellar mass density at z~9-11, but it will play a pivotal role in characterizing the stellar masses and other properties of 20-30 other gravitationally lensed z~4-8 galaxies behind these clusters.