A series of Locusta adipokinetic hormone I (AKH-I), < QLNFTPNWGTa, analogues, were synthesized with modifications at the C-terminal threonine residue using a combination of solid- and liquid-phase methodology and evaluated in Locusta migratoria, in a lipid mobilization assay in vivo and an acetate uptake assay in vitro. Modifications at Thr10 of AKH-I involved replacement of its C-terminal amide by the groups -OH, -OCH3, -NHCH3, -N(CH3)2, and -NHC6H5; the last three groups were also applied to the amide of AKH-I-[Thr(Bzl)10]. The methyl ester, monomethyl, and dimethyl analogues were all of lower activity than the parent in the lipid mobilization assay, but lost less than two orders of potency. In the acetate uptake assay, again the methyl ester analogue showed the greatest retention of biological activity of all modified peptides. A cyclic analogue, cyclo (PLNFTPNWGT), was active in both assays, but only at very high concentrations. Almost all analogues were more active in the acetate uptake assay than in the lipid assay, but unusually, AKH-I-NHCH, and AKH-I-N(CH3)2, together with cyclo(PLNFTPNWGT), were more active in the lipid mobilization assay. In addition, the acid AKH-I analogue did not suffer as large a loss in potency in the lipid mobilization assay as in the acetate uptake assay, although it was less potent in the former. The relative potencies of these two methyl analogues contrast with those for AKH-I[Thr(Bzl)10]-NHCH3 and AKH-I-[Thr(Bzl)10]-N(CH3)2, which, together with both phenyl analogues, were significantly more active in the acetate uptake assay. We conclude that the acetate uptake assay has a greater preference for a hydrophobic C-terminus, compared with the lipid mobilization assay.