Fig. 1. Map of DSDP/ODP drill sites for samples discussed in this paper. Also shown are selected magnetic anomalies from Cande et al. (1989). See Table 1 for details.

Fig. 2. Representative results from the IZZI experiment described in the text (see also Table 2). Circles are the NRM remaining versus pTRM gained at each temperature step. Darker (lighter) symbols are the ZI (IZ) steps. (Open) closed symbols are (excluded) included in the slope calculation. Triangles are the pTRM checks and squares are the pTRM-tail checks. Solid lines connect the highest and lowest data points used in the calculation and are not the “best-fit” lines. Insets are vector end-point diagrams of the zero-field steps. (a) The total remanence estimated by summing the vector difference between subsequent demagnetization steps (VDS) the dash-dot line. The fraction F_vds is the portion of the VDS used in the paleointensity calculation. δ_i and are the pTRM checks and pTRM tail checks. (b) Example of nearly ideal behavior. (c) Example of the zig–zag behavior detected in the IZZI experiment when blocking and unblocking temperatures are not the same. (d) Example of experimental data that barely meet acceptance criteria.

Fig. 3. Cumulative distributions of selected parameters defined in Table 2 calculated for 951 specimens of DSDP/ODP submarine basaltic glass. Vertical lines indicate cutoffs for acceptability where the solid curves are those accepted and the dashed curves are rejected.

Fig. 4. Cumulative distributions of depth below sea floor (a) and age (b) of accepted (solid) and rejected (dashed) specimens.

Fig. 5. Virtual axial dipole moments (Z=10²¹) for all specimens (+) and sample averages (dots) meeting minimum criteria for the submarine basaltic glass data compiled here. Also shown are the data of Riisager et al. (2003) from Site 1185. Sample data are listed in Table 3.

Fig. 6. Summary of VADM data versus age. Triangles are from the published literature compiled by Tauxe and Staudigel (2004) (those in red are the Troodos submarine basaltic glass data). Dots are the sample averages from Fig. 5. The present field is the dotted line. The thin solid line is the average dipole moment calculated assuming a log-normal distribution (see text). At the bottom is the Geomagnetic Polarity Time scale from Berggren et al. (1995) and Gradstein et al. (1995) used here.

Fig. 7. Histogram of all data shown in Fig. 6. The present day value is the vertical line. The data are clearly not normally distributed, but have a long high intensity tail.

Fig. 8. Cumulative distributions of the VADM data from Fig. 6.

Fig. 9. Theoretical effect of cooling rate on estimated paleointensity (B_est) relative to the ancient field B_anc as a function of relaxation time. The slower the cooling rate, the longer the effective relaxation time. The relaxation time during laboratory experiments is assumed to be approximately 100 s. The temperature labels for each line are blocking temperatures. A 20% difference between lava flow data and SBG could be caused if the lava flow cooled to about 400 ° C over a period of days to months. (Figure redrawn from Selkin et al., 2000, see also Halgedahl et al., 1980.)

Fig. 10. Data from the present study (open and closed circles); the compilation of Tauxe and Staudigel (2004) (open triangles and squares) and the Brunhes compilation of Tauxe and Love (2003) (solid triangles). The associated polarity chrons are noted.

Locations, ages, and paleolatitude informtion for the DSDP/ODP sites referred to in this study

Biostratigraphic constraints are in parentheses. [1] Nakanishi et al. (1992); [2] Gradstein et al. (1995); [3] Berggren et al. (1995); [4] Castillo et al. (1994); [5] Sager (2004); [6] Lee and Lawver (1995): reconstructed to stable Europe, λ^′ here used Besse and Courtillot (2002) for EU; [7] Besse and Courtillot (2002); [8] Steiner (1981); [9] Wallick and Steiner (1992); [10] Yan and Kroenke (1993); [11] Mahoney et al. (1993); [12] DeMets et al. (1990); [13] present latitude; [14] 0.4° /Myr. PA: Pacific; FB: Fiji Basin; IN: Indian; ANT: Antarctic; NA: North American; AF: African; EU: European; CO: Cocos; NZ: Nazca; LB: Lau Basin; SA: South American; EPR: East Pacific Rise (Axial Summit Caldera); l Alb.- e. Cen.: late Albian to early Cenomenian; e Camp: early Campanian; l Bath. - e Berri.: late Bathonian to early; Berriasian; WPN-15 (lower NP23).

Parameters

[1] Coe et al. (1978), [2] Tauxe and Staudigel (2004), [3] Tauxe (1998), [4] Kirschvink (1980).

Summary data for samples passing minimum selection criteria (see text)