The Jurassic magnetic Quiet Zone (JQZ) contains magnetic lineations, but their low amplitudes make correlation and interpretation difficult. Part of the problem is the separation of source and sensor for old, deep ocean crust. We increased anomaly amplitudes by collecting magnetic data along two deep-tow profiles over western Pacific JQZ lineations. A magnetic polarity reversal timescale was constructed by matching deep-tow anomalies with a simple, rectangular block magnetization model for oceanic crust. The polarity sequence covers -11 m.y. (156--167.5 Ma) and contains 88 pre-M29 polarity chrons extending to Chron M41. A limitation of this model is its poor representation of the oldest anomalies upward continued to sea level. On deep-tow profiles these anomalies have both long- and short-wavelength components, but only the latter are easily modeled on a datum close to the source. An alternative polarity model was constructed to match the anomalies upward continued to sea level. This model retains only 44% of the deep-tow model polarity chrons because of short-wavelength attenuation by upward continuation. Because of the inferred periods and magnetization contrasts, we think many of the short-wavelength anomalies represent paleofield intensity fluctuations. In contrast, polarity reversals have been documented by prior magnetostratigraphic work for the younger part of the timescale covered by our model. Thus our data may show a transition from a geomagnetic field behavior dominated by intensity fluctuations to one dominated by reversals. ¿ 1998 American Geophysical Union