It is broadly accepted that Supermassive Black Holes (SMBHs) are located in the centers of most massive galaxies, although there is still no convincing scenario for the origin of their massive seeds. It has been suggested that primordial black holes (PBHs) of masses $\gtrsim 10^{2} M_\odot$ may provide such seeds, which would grow to become SMBHs. We suggest an observational test to constrain this hypothesis: gas accretion around PBHs during the cosmic dark ages powers the emission of high energy photons which would modify the spin temperature as measured by 21cm Intensity Mapping (IM) observations. We model and compute their contribution to the standard sky-averaged signal and power spectrum of 21cm IM, accounting for its substructure and angular dependence for the first time. If PBHs exist, the sky-averaged 21cm IM signal in absorption would be higher, while we expect an increase in the power spectrum for $\ell~\gtrsim 10^2-10^3$. We also forecast PBH detectability and measurement errors in the abundance and Eddington ratios for different fiducial parameter configurations for various future experiments, ranging from SKA to a futuristic radio array on the dark side of the Moon. While the SKA could provide a detection, only a more ambitious experiment would provide accurate measurements.

19 pages, 14 Figures; Added figures 3 and 4, discussion extended, results and conclusions unchanged