The samples were ground in an agate Tema swing mill at the University of Edinburgh, and splits of identical powders were used for all the different analyses noted below.

Major elements were analyzed at the Geological Survey of Denmark and Greenland, Copenhagen. Most elements were determined by X-ray fluorescence (XRF) analysis on fused glass discs. The rock powders were dried at 110ºC for 2 hr and ignited at 1000ºC for 1 hr with subsequent determination of the loss on ignition (LOI). The ignited samples were mixed with sodium tetraborate in the ratio 0.7500 g sample to 5.2500 g borate and fused in Pt/Au crucibles under continuous agitation for 1-1½ hr. After inspection for homogeneity, the melts were poured into Pt/Au molds creating glass discs with 32 mm diameter. The XRF spectrometer used was a Phillips PW1606 multichannel instrument with a Rh-anode X-ray tube. Calibration and correction for background and line overlaps were calculated from measurements on synthetic single-element glass discs, and corrections for matrix effects were calculated either from measurements on synthetic glass discs or from the absorption coefficients of Heinrich (1966).

Na₂O was determined by atomic absorption spectrometry (AAS). Each dried sample (0.25-0.5 g) was treated with hydrofluoric acid in a PTFE beaker on a hotplate. After evaporation to dryness, the residue was dissolved in a 50 ml hydrochloric acid/potassium chloride solution and sodium measured on a Perkin Elmer PE2280 AAS instrument.

For FeO determination, each dried sample (0.1 g) was treated with ammonium vanadate/hydrofluoric acid overnight. Boric acid and a measured amount of iron(II) was added. Surplus iron(II) was determined by automatic potentiometric titration using Cr(VII) as titrant. The method used is a modification of that used by Wilson (1955).

Volatiles were calculated as the weight LOI corrected for the calculated gain of weight created by the oxidation of iron(II) to iron(III) during ignition.

As shown by L.M. Larsen et al. (1998b), the accordance between the major element analyses carried out in Copenhagen and Edinburgh is excellent, ensuring full compatibility between the present major element data set and that for the Leg 152 samples obtained in Edinburgh (Fitton et al., 1998b).

Trace elements were analyzed by XRF on pressed powder pellets at University of Edinburgh, in the same laboratory and with the same methods as described by Fitton et al. (1998b) for the Leg 152 samples.

A subset of samples was also analyzed at the University of Leicester by instrumental neutron activation analysis and at the NERC Isotope Geoscience Laboratory for Sr, Nd, and Pb isotopes; these results are presented by Saunders et al. (Chap. 8 this volume).