So, the archaeal RNase P enzyme contains an RNA that resembles those of Bacteria, but not eukaryotes, in both sequence and structure, and these RNAs can bind to bacterial RNase P protein to create a functional holoenzyme. The natural presumption, then, would be that the archaeal enzyme would contain a bacterial-like small, single protein.
But the complete genome sequences of several archaeal species had become available by that time, and none of these (or any of those that have become available since) contain sequences recognizably similar to any bacterial RNase P protein. Several groups have looked, and looked hard, in vain. We went so far as to do a genetic screen of archaeal genes in attempt to complement E. coli with a defective RNase P protein, also in vain.
That the archaeal RNase P had a protein component was fairly clear; RNase P enzymatic activity purified from cells was robust under 'normal' RNase P assay conditions. Removal of protein by phenol extraction resulted in RNase P activity that mirrored the very different activity of the RNA generated by in vitro transcription. For example, RNase P activity from Methanocaldococcus jannaschii is robust at low ionic strength, but disappears completely upon phenol extraction. RNase P activity from M. thermoautotrophicus is likewise robust at low ionic strength, but is dramatically reduced by phenol extraction, and becomes dependent on very high ammonium and Mg++ concentrations, and reduced temperature.
And so the nature of the protein in archaeal RNase P was a mystery....