Since measurements began in the late nineteenth century, there has been a secular increase (with superposed ripples due to solar cycles) of the aa geomagnetic index <Mayaud, 1972>. Starting from this observation, Lockwood et al. <1999a, 1999b> conclude that the total open solar magnetic flux has increased by 41% from 1964 to 1995 and by 130% over all but the last 5 years of the twentieth century. However, solar data for more than two solar cycles - Carrington maps from Mount Wilson, and Wilcox Solar Observatories and newly reanalyzed data from the National Solar Observatory - show no secular trend in overall photospheric flux. More importantly, the magnetic flux open to interplanetary space (as calculated from photospheric measurements and assuming potential fields to a height of 2.5 R⊙) fails to show evidence of a secular increase over the last two solar cycles. Like Lockwood et al., we do not explicitly take account of transient events. Thus both data and calculations imply that the Sun's average coronal magnetic flux has not increased over the last two solar cycles. Analysis of simulations with the potential field source surface model shows that the interplanetary magnetic flux is not simply related to the erupted photospheric solar magnetic flux. Both results are in agreement with the findings of Wang et al. <2000>. The topology, rather than the strength, of the emergent solar magnetic field may be a major determinant of the interplanetary magnetic field experienced at Earth.