In the course of studying Quasi-biennial (QB) variations in the global sea surface temperature (SST) field, an apparent solar signal was detected. Specifically, the amplitude of variations in the QB frequency band appeared to be modulated at the 11 year solar cycle at least over the last four cycles. The sense of the relation is that high solar activity goes with low amplitude variations in the QB frequency band. In a series of recent papers, van Loon and Libitzke (1988a,b) and Libitzke and van Loon (1988) have found in the same frequency band a similar amplitude modulation in atmospheric fields of both Hemispheres also over the last four solar cycles. In an effort to study the apparent relation over more solar cycles, the analysis of the time series of SST was extended back to 1884. The solar-SST relation found above was found to hold back to about 1925. Before that time, remnants of the SST signal can still be observed but the relation to the solar cycle becomes more confused. This lack of consistency through the 10 solar cycles examined here may be due to the fact that the apparent relationships are fictitious. They may also be partially due to poor SST data before 1920; the major failure in the solar-SST relation occurring in the data sparse 1915--1920 period. It may also be that the weak solar variability at the turn of the century could not force an observable response in the SST field. ¿ American Geophysical Union 1989