Intrinsically weak magnetic fields are difficult to identify, since flux measurements (magnetograms) cannot by themselves distinguish between filling‐factor and field‐strength effects. The first real determinations of intrinsically weak (less than kG) fields have in fact been made only this year, using the StokesVprofiles of an infrared line pair near 1.56 &mgr;m. Many cases of discrete magnetic elements with field strengths as low as 0.4 kG have been found, immediately adjacent (within a couple of arcsec) to the normal strong‐field fluxtubes that have strengths in the range 1.4–1.6 kG and a magnetic polarity that can be both the same or opposite to that of the adjacent magnetic component.There appears to be a continuous sequence of bipolar magnetic regions of various scales, down to the spatial resolution limit, from active regions to ephemeral regions and inner‐network fields. It seems likely that this sequence continues in the form of a subarcsec mixed‐polarity or ‘‘turbulent’’ field that permeates the 99% of the photospheric volume not occupied by the kG flux tubes in the network. A one‐sigma upper limit of 100 G to the strength of this hitherto ‘‘invisible’’ field has been set from line‐broadening constraints, which indicates that this small‐scale field is intrinsically weak. Arguments are given why the spatial spectrum of flux emergence should saturate when scales approaching the photon mean free path in the photosphere (about 100 km) are approached, which is the range of scales that may be opened up to exploration by LEST and OSL.It is shown how correlations (‘‘active longitudes’’) in the pattern of small‐scale flux emergence lead to a replenishment of the global or ‘‘background’’ magnetic‐field pattern at high heliographic latitudes in a time as short as weeks, more than two orders of magnitude faster than predicted by numerical models of the Babcock‐Leighton type. There is thus a close link between the small‐scale dynamics and the global solar‐cycle evolution.