The crystal structure for a diphenyldialkoxysulfurane, (C6H5)2S[OC(CF3)2C6H5]2, reveals an approximate trigonal bipyramidal geometry about sulfur, with a lone pair considered to occupy one equatorial position, the two phenyl ligands occupying the other two equatorial positions (C-S-O angles range from 86.4(2) to 91.2(2)°; the C-S-C angle is 104.4(3)°), and with the electronegative alkoxy ligands occupying apical positions (O-S-O angle is 175.1(2)°w ith the distortion from linearity in the direction of the equatorial phenyl rings). The S-0 bond lengths (1.889(4) and 1.916(4) Á) are approximately 0.2 Á longer than the sum of the covalent radii as would be expected from an S-O bond order less than unity. The crystals are triclinic, the space group is Pi, and there are two molecules in a unit cell of dimensions a = 10.026(3), b = 14.268(3), c = 10.802(3) Á, a = 109°5 0'(2'), ß = 92°5 2'(2'), and y = 95°2 0'(2')• The structure was solved by Patterson-heavy atom methods based on sulfur, and has been refined to a conventional R of 0.070 for 3364 nonzero reflections measured by counter methods. Some comparisons of the bonding scheme with those for tetracovalent sulfur compounds containing halogen ligands are made.

Elderly and college-age subjects were compared in two experiments, one involving episodic memory and one involving semantic memory. Responses were generally slower for older subjects; however, in some conditions, older subjects were as good as or better than young subjects, even in terms of response time. The results suggest that older subjects have no difficulty with memory tasks that do not require exact memory-matches or careful inspection of retrieved propositions. It is argued that careful inspection is a much more costly process for older adults than it is for young adults, but that plausibility judgments and feature overlap processes are equally easy for both age groups. The suggestion is made that older subjects also tend to modify their performance in ways that would minimize the detrimental effects of forgetting specific facts.A salient attribute associated with increasing age, from young adulthood to retirement and beyond, is a decrement in memory performance (see Birren & Schaie, 1977;Hunt &Hertzog, 1981;Poon, Fozard, Cermak, Arenberg, & Thompson, 1980, for extensive reviews). Although there is almost universal agreement that memory performance declines with age in certain situations, there is less consensus as to the nature of the decline or the cause. Researchers have argued as to whether the deficits are primarily encoding-based, storage-based or retrieval-based, or some combination of these, and whether deficits should be thought of as a reduction in capacity, loss of processing power, loss of processing speed, or loss of mental flexibility (see Craik, 1977, Kausler, 1982, and Salthouse, 1982, for extensive reviews).Arguments have been made that the more difficult the memory task, the greater the disparity in performance between older and younger adults (e.g., Craik, 1968;Laurence, 1967). On this, there is relative consensus. It is not surprising, for example, that the disparity in performance between younger and older adults would be greater for recall than for recognition (e.g., Botwinick & Storandt, 1974;Erber, 1974;Schonfield& Robertson, 1966). Recall is less likely to suffer from ceiling effects and requires more processes than does recognition, so there are more processes or stages for an age advantage to emerge. More surprising are claims that the disparity in performance increases with more "meaningful"

Further evidence is presented (8) that changes in optical rotation at the D-line of sodium which occur with changes in solvent largely reflect changes in conformational equilibria.Changes in the nuclear magnetic resonance spectrum of 4,4,5,5-tetradeuterio-2-methoxytetrahydropyran and in the rotation of s(+)-2-methoxytetrahydropyran with changes in solvent are interpreted to show that solvents capable of donating a hydrogen to a hydrogen bond have an effect on the magnitude of the anomeric effect which is greater than that resulting from the change in the dielectric constant of the solvent. In the case of methyl 2-deoxypyranosides with the methoxy group in axial orientation, water appears to have an especially profound influence in stabilizing that orientation of the methyl group which brings it into garrche relationship with both the C2-grouping and the anomeric hydrogen; that is, counter to the anomeric effect.The methyl 2-deoxy-e-L-erythro-pentopyranoside was synthesized from di-0-acetyl-L-arabinal by way of an iodomethoxylation. Deuterolysis of the resulting 1,2-trans-acetylated methyl 2-deoxy-2-iodopentosides is shown to proceed in one case with extensive inversion of configuration and with extensive retention in the other case and offers a route to the preparation of 2-deoxy-2-deuterio-erythropentose with a high proportion (> 75%) of the diastereoisomer with the arabino-configuration.

No abstract

Well correlated ab initio electornic structure calculations have been used to investigate the structure of the trifluoride ion. It is calculated to be stable with respect to F and Fz by about 11 kcal/mol. While this represents a relatively weak bonding of F to F2 to form F<, a significant interaction is evidenced by a 0.3-A lengthening of the F2 bond (from 1.41 to 1.70 %.) upon forming the Dmk F3-. This lengthening of the F-F bond distance decreases the repulsive *-type interactions of the unshared electron pairs of the fluorines of F2, and it is postulated that this provides a major stabilizing influence on the symmetrical trifluoride ion. The calculated centrosymmetric equilibrium structure, along with the calculated vibrational constants, is generally consistent with matrix isolation experiments reported earlier by Ault and Andrews. The nature of the bonding in this unique main group hypervalent molecule is discussed.The dsp3 hybridization scheme has traditionally been invoked to rationalize the bonding in electron-rich compounds of the main group elements which formally violate G. N. Lewis' octet rule, e.g., PF,, SF4, CIF,, and XeF2.2s3 Although molecular structures are usually reliably predicted with this model, estimated electron distributions are inconsistent with NQR4 and Mossbauer5 data for several such species that have been studied. In addition, a b initio calculations of the electronic structures of many of these molecules indicate that d-orbitals are not populated to the extent that the hybridization schemes imply.6 W e consider the widely accepted 3-center 4-electron (3c-4e) bond model of Rundle7 and Pimente18 to be a better approximate description of the electronic structure of these molecules. Musherg termed these molecules hyperualent in order to distinguish them from molecules of the main group elements which have either electron deficient 3-center 2-electron bonds or simple electron pair bonds.The term "hypervalent" has been redefined several times, most recently through the use of ab initio methods which attempt to calculate a valence for each atom in a molecule based on the calculated population of orbitals centered on the atom.I0 The approximation which we use for the bonding scheme for hypervalent species with a linear 3c-4e bond involves a u-delocalized bond as in the LCAO-MO diagrams pictured below for the 4-H-2 bifluoride anion and the 10-F-2 trifluoride ion. Both involve cr-delocalized bonds which are u analogues to the 3c-4e *-bond of the allyl anion. The formal valence shell expansion of the hydrogen of the bifluoride anion and of the central fluorine in the trifluoride ion represented in this way involves the +L2 orbital which places two electrons in an orbital essentially localized on the apical ligands. A hypervalent species is one in which this symmetrical structure is an energy minimum. Our definition,]' which remains