Among the basaltic rocks sampled by submersible or dredged at the East-Pacific Rise axis around 21°N latitude (French–American–Mexican legs CYAMEX and RISE), 22 samples have been selected according to macroscopic, oriented criteria (elongation plane of feeder tubes in sheeted lavas, bubbles and gas cavities, stalactites, septa) from which the polarity of a sample, acquired in situ during its cooling, could be established.The natural remanent magnetization intensity of these generally voluminous samples (up to about 350 cm3) has a mean value of 11.35 ± 2.31 A∙m−1 (10−3 emu∙cm−3); the scatter of the values can be explained by the varying proportion of nonmagnetic material (glass, Mn) in the samples.The magnetic viscosity is always negligible and the stability of the remanent magnetization direction after alternating field demagnetization indicates that the stable direction determined is the direction of the original thermoremanence.Despite the relatively low latitude, which makes the occurrence of negative magnetic inclinations within a normal polarity period (Brunhes, in this case) more probable, no reverse polarity has been found. The mean inclination of the 22 samples is I = +41.8 ± 16.7° and becomes I = +42.2 ± 19.4° if the samples are grouped in 15 pin-point sampling sites; these values are not significantly different from those of the local dipole field, I = ±37.4°, or from those of the actual local field, I ~ +47°, which may be the result of effusions from the youngest lava flows.The scatter of the magnetic inclination values is more tightly bound to the quality of the polarity criteria than to the secular variation of the Earth ' s magnetic field, the latter interfering in only a few cases. It is obvious that the validity of the reorientation depends on (1) the morphological quality of the criteria used: those that are well developed, well represented in the sample, and apparently the most crystallized (glassy criterion has proved to be doubtful), permit a better reorientation of the sample; and (2) the principle of the criterion itself: criteria referring to vertical axes (stalactites) or vertical planes (septa) are more reliable than those based upon the determination of a horizontal plane (elongation plane of tubes in sheeted lavas).It appears from this study that macroscopic polarity criteria may be used in an almost systematic way to find the polarity of an in situ sample.

You do not currently have access to this article.