In the first part of this paper palaeomagnetic and rock magnetism investigations were developed in volcanic samples from the Northeast of Brasil. The age of the samples spans the Jurassic and Cretaceous periods. To accomplish this task four areas were studied and a total of 495 samples from 56 cites were analyzed. A portable drilling machine with 2.5 cm core diameter was used to collect the samples. The orientation of the samples were obtained by means of a magnetic compass, and a clinometer. The specimens were submitted to alternating field demagnetization and in, a few cases, to thermal demagnetization. Giving unit weight to each site the mean direction of the characteristic remanent magnetization of each one of the studied areas were determined. The volcanic rocks from Jurassic, lying in the western part of the Maranhão Basin (Porto Franco - Estreito) , yielded the mean direction: declination D=3.9º, inclination I=-17.9°, with the circle of 95% of confidence α95=9.3º, precision parameter k=17.9, number of sites N=15. All sites showed normal polarity. For this area was determined a palaeomagnetic pole with coordinates 85.3°N, 82.5°E (circle of 95% of confidence A95 = 6.9º) that is situated near other known palaeomagnetic poles for this period. The Lower Cretaceous rocks from the eastern part of the Maranhão Basin (Teresina-Picos-Floriano) yielded a mean direction for the characteristic remanent magnetization having D= 174.7º, I=6.0°, α95=2.8º, k=122, N=21. All sites showed reversed polarity. The calculated palaeomagnetic pole associated with these rocks has coordinates 83.6°N, 261.0ºE (A95= 1.9°) and is in agreement with other South American poles of the same age. In Rio Grande do Norte a swarm of Lower Cretaceous tholeiitic dikes was studied having a characteristic mean direction with D= 186.6º, I= 20.6º with α95= 14.0º, k= 12.9, and N= 10. The sites in this area showed mixed polarity. The computed palaeomagnetic pole is located at 80.6ºN and 94.8ºE with A95= 9.5º. The study of the volcanic rocks of the magmatic province of Cape Santo Agostinho yielded the following values for the characteristic remanent magnetization D= 0.4º, I=-20.6º with α95= 4.8º, k=114, N=9. All sites showed normal polarity and the calculated paleomagnetic pole has the coodinates: 87.6ºN 135ºE with A95= 4.5. The secular variation of the obtained directions was discussed so that each pole presented in this paper is really a palaeomagnetic pole. The analysis of the magnetic minerals of these samples was done by thermomagnetic curves and by X-ray diffraction. In most cases the magnetic phase in the rocks is mainly titanomagnetite with poor titanium content. Maghemite and sometimes hematite, usually a product of weathering, did not obscure the initial thermoremanent magnetization of these rocks. Generally the determined Curie temperature lies between 500-600º C. Frequently it was observed that the exsolved titanomagnetite has a phase near magnetite and another phase rich in titanium, near ilmenite, as a result of high temperature oxidation. The second part of this paper deals with the determination of the time of the opening of the South Atlantic ocean by means of palaeomagnetic data. In this paper, however, instead of using the polar wandering paths of the continents (the usual method) statistical tests were applied that give the probability that a certain configuration for the two continents be consistent or not with the palaeomagnetic data for a chosen period. For the Triassic, Jurassic, Lower Cretaceous and Middle-Upper Cretaceous periods the palaeomagnetic poles for Africa were compared with the respective poles of South America in pre-drift configuration by means of an F-test. Other configurations that indicate some separation between the two continents were also tested. The results of the tests showed that the pre-drift reconstruction after Martin et al (1981) is consistent with the palaeomagnetic data for Triassic, but there is a significant difference between the respective Jurassic, Lower Cretaceous and Middle-Upper Cretaceous palaeopoles for the two continents, with a probability of error of less than 5%. Other pre-drift reconstructions were tested and the results were the same. Comparing the pole positions for South America and Africa in a configuration that indicates a small separation between the two continents, as the one suggested by Sclater et al (1977) for 110 m.y.B.P. one finds a significant difference for the Triassic data. For the Jurassic and Lower Cretaceous palaeomagnetic poles, which are, however, earlier than the suggested date of the reconstruction, the results are consistent with that separation of the continents with a probability of error of less than 5%. The reconstruction for 80 m.y.B.P., after Francheteau (1973), indicanting a larger separation between the continents, is consistent with the Middle-Upper Cretaceous palaeomagnetic poles. Assuming the premise of the movements of crustal blocks relative to each other as rigid blocks, the results of the F-test indicated that South America and Africa were close together during Triassic. There was, nevertheless, a small separation between the continents in Jurassic, probably due to an earlier rifting event, and this separation was stationary until Lower Cretaceous time. This result is different from the most part of the papers that discuss the openning of the South Atlantic ocean. The Middle-Upper Cretaceous data are compatible with a fast and significant spreading of the continents in that period.
|
