916 EMLLIANI AND FLINT [CHAP. 34 



glacial times, when sea-ice probably covered a vast area all around Antarctica. 

 Conversely the coarse glacial-marine sediments are likely to have been 

 deposited during interglacial times (as the occurrence of such sediments in the 

 top portions of the cores indicates) when sea-ice was greatly reduced. Radium 

 measurements by Urry (1949) on three of the four cores indicate either greatly 

 varying rates of sedimentation, or radium migration, or both. Important 

 radium migration appears to have taken place in at least two of the cores (nos. 

 N-4 and N-5), where anomalous radium concentrations were observed at some 

 depth in the cores (Urry, 1949, table II; cf. Pettersson, 1951; Kroll, 1954, 1955). 

 Again, the detailed chronology published by Hough (1950) is open to question. 

 The contention that coarse sediments represent interglacial conditions and the 

 very fine sediments glacial conditions is supported by the observation that 

 diatoms and Radiolaria are abundant in the former and absent from the 

 latter (Thomas, 1959). 



Altogether, the deep-sea cores from the Atlantic and adjacent basins have 

 yielded a consistent picture of climatic change. The Pacific and Indian Ocean 

 cores have not yielded as good a picture partly because the Pleistocene tem- 

 peratvu'e oscillations may have been considerably smaller than in the Atlantic 

 and adjacent basins, partly because most of these cores have not yet been 

 studied by means of micropaleontological and oxygen-isotopic analysis. 



c. Chronology of deep-sea cores and correlation with the continental Pleistocene 

 stratigraphy 



14C, 230Th and the ratio 23ipa/230Th have been used for absolute dating of 

 deep-sea cores. i'*C measurements have been made on the foraminiferal com- 

 ponent alone, on bulk core material and on the fine carbonate fraction, using 

 modern shells for the contemporary assay. Measurements on the finer carbonate 

 fraction gave ages up to 10% greater than the ages obtained from the associated 

 foraminiferal component (Rubin and Suess, 1955; Ericson et al., 1956), indicating 

 the occasional presence of some reworked material. Generally, however, the 

 amount of reworked material in the fine carbonate fraction does not seem to be 

 large enough to affect importantly the ages obtained from bulk core material. 

 These, and the ages obtained from the foraminiferal component alone have 

 yielded a consistent chronology for various deep-sea cores, ranging back in 

 time to approximately 40,000 years B.P. (Rubin and Suess, 1955, 1956; Ericson 

 et al., 1956; Broecker and Kulp, 1957; Broecker et al., 1958). 



The ssoTh and 23ipa/230Xh methods are based on the fact that thorium and 

 protactinium isotopes produced by 238U ^nd 235U existing in solution in sea- 

 water are rapidly removed by adsorption on particles and deposited on the 

 bottom. The half-lives of 230Th, 23ipa, and of the ratio 23ipa/230Th are, respec- 

 tively, 80,000, 34,300 and 60,100 years. If the concentration of uranium in the 

 ocean remained constant with time, and if the rate of sedimentation of the clay 

 component also remained constant, the activity of 230x11 alone would provide 

 a reliable dating method. Urry (Piggot and Urry, 1942, 1942a; Urry, 1949) 

 dated various deep-sea cores from the North Atlantic, the Caribbean and the 



