Assessment of dexamethasone suppression by salivary cortisol measurement is far less repeatable than the use of plasma cortisol. In the context of field studies of dexamethasone suppression, salivary cortisol measurements may only be appropriate for large numbers of subjects.

An isotope shift of 5 ppm per deuterium atom has been observed for the 59Co resonance of tris(ethylenediamine)cobalt(~~~) chloride ([Co(en),]Cl,) and hexaamrninecobalt(~~~) chloride ([Co(NH&] Cl,) after exchange of hydrogen for deuterium. The upfield direction of the shift can be readily understood as changes in the ligand-field strength causing changes in the nuclear paramagnetic screening.We always observe a downfield shoulder on the "Co resonance of tris-(ethylenediamine)cobalt(~~~) chloride ([Co(en),] Cl,) in D 2 0 solvent. In the past we assigned this shoulder to an impurity. However, n.m.r. of a freshly dissolved sample showed transient spectra of up to nine equally spaced resonances. The transient nature of these spectra lead us to conclude that it was a hydrogen-deuterium exchange phenomenon. A detailed study has confirmed this.Spectra of [Co(en),] Cl, in H 2 0 , D 2 0 and various H20-D20 mixtures, after hydrogen-deuterium exchange had reached equilibrium, are illustrated in Fig. la 99.8% D 2 0 at 273 K and obtaining a spectrum at 280 K where the amount of deuterium exchange is quite small during the few minutes required to obtain a spectrum. The resulting spectrum is illustrated in Fig. lb. The shift per exchanged hydrogen atom is 4.7 ppm upfield. It might be expected that, for instance, the shift of d2-[Co(en),] Cl, would be different depending on whether the deuterium atoms are geminal or further apart, This type of difference, if any, is sufficiently small so that

Valiev's suggestion of excitation of the vibrational modes of the octahedron is considered as a spin relaxation mechanism for 59Co spins in some symmetric, bidentate CoIII complexes. His theory is extended to the spin 7/2 system; the efficiency of the mechanism is very sensitive to the molecular size with larger complexes having much shorter T1 values. Experimental evidence in support of the mechanism is provided by the contradictory values of quadrupole coupling constants calculated experimentally by assuming rotational reorientation to be the dominant time process and by measurements on a variety of CoIII complexes. Rotational correlation times were measured from 13C T1 values of protonated carbons in the complex. An X-ray structure determination of tris(tropolonato)-cobalt(III) is reported in support of the suggested relaxation mechanism. The differences between T1 and T2 in some CoN6 complexes are shown to arise from unresolved 59Co-14N scalar coupling.