Sets of aeolian cross-strata within the Cretaceous Etjo Formation of NW Namibia are bounded by a hierarchy of surfaces, the origin of which are ascribed to one of four processes related to aeolian bedform and erg behaviour. The base of the main aeolian succession is characterized by a basin-wide erosional supersurface that formed in response to a period of aeolian de¯ation before the onset of the main phase of erg building. Interdune migration surfaces formed by draa migration are planar in sections parallel to the palaeowind and are inclined at up to 5°in an upwind direction (SW). Perpendicular to the palaeowind, interdune surfaces form 500-m-wide troughs, signifying crestline sinuosity within the original bedforms. Superimposition surfaces are inclined at 5±10°in a downwind direction and indicate the migration of crescentic oblique dunes over larger, slipfaceless transverse draa. Reactivation surfaces associated with minor changes in dune slipface orientation are distinct from other bounding surface types because overlying cross-strata lie parallel to them, rather than downlap onto them. Analysis of the geometry of these bounding surfaces, together with the orientation of the cross-strata within the sets that they bound, has enabled the detailed morphology of the original bedforms to be reconstructed. The maximum preserved thickness of individual aeolian sets varies systematically across the basin, from 52 m in the basin depocentre to only 8 m at the basin margin. The set architecture indicates that this spatial variation is primarily the result of decreased angles of bedform climb at the basin margin, rather than the presence of smaller bedforms. Similarly, a temporal reduction in the angle-of-climb, rather than a reduction in bedform size, is considered to be responsible for an upward decrease in preserved set thickness. Reductions in bedform climb angle re¯ect progressive loss of accommodation space as the accumulating erg ®lled the basin.

The relationship between pregnancy wantedness and adverse pregnancy outcomes was studied using data from 2,828 mothers who participated in the Missouri Maternal and Infant Health Survey. The wantedness of a pregnancy was measured using traditional classifications of mistimed and unwanted, as well as additional measures gauging how the woman felt about the pregnancy while she was pregnant. Fifty-eight percent of the very low birth weight infants and 59% of the moderately low birth weight infants resulted from unintended pregnancies, as did 62% of the normal-birth-weight infants. Logistic regression showed that mothers of very low birth weight infants were significantly more likely than those who had a normal-weight baby to report that they had felt unhappy about the pregnancy (odds ratio of 1.53). Very low birth weight was also associated with early denial of the pregnancy (1.54). Odds ratios associating these two unwantedness categories with low-birth-weight babies were higher among Medicaid recipients than among women not receiving Medicaid. Associations between very low birth weight and the denial variable were also significant among white women when very low birth weight outcomes were compared with normal outcomes, but there was no significant association among black women. There were no significant associations between low birth weight and the traditional unwantedness variables.

Antibodies in the Kidd blood group system show a great deal of serological variability, are notoriously elusive and hence evoke difficulties in detection. However, they have been regularly reported as causing severe immediate or delayed haemolytic transfusion reactions and this clinical potential has been largely attributed to their complement binding ability. In initial investigations on 43 anti-Jka/Jkb sera with a range of titres of IgG antibody only a few seemed to fix complement, though following repeated tests on 20 of these sera a further five were shown to bind complement, making a total of 12 (27.9%) showing evidence of complement binding. Twenty-three sera were unavailable for re-testing. Subsequent tests revealed that only those sera which showed direct agglutination or were positive with an anti-IgM reagent in an indirect antiglobulin test (IAT) fixed complement. Evaluations on the IgG fractions of six selected potent anti-Jka sera failed to reveal any complement-fixing ability although all the original sera bound complement avidly and contained variable amounts of IgM antibody, some at very low subagglutinating levels. These findings challenge past perceptions and give cause for reflection on the changing methodologies and strategies which could unduly compromise the detection of these potentially clinically damaging antibodies.

The editorial group responsible for this previously published document have withdrawn it from publication.