The substantial lowering of tropical snowlines at the Last Glacial Maximum (LGM), circa 21 kya, is examined using a modified version of the single-cell tropical climate model of Betts and Ridgway <1989>. These authors concluded that it was difficult to reconcile the large depression of snowlines at the LGM with the small reduction in mean tropical sea-surface temperature (SST) of the Climate: Long-Range Investigation, Mapping, and Prediction (CLIMAP) reconstruction. Here, climatic implications of the snowline depression are compared, not with CLIMAP, but with temperature proxies (δ18O and Sr/Ca ratios in corals, noble gas concentrations in aquifers, alkenone and Mg/Ca ratios in deep-sea sediments and recent faunal reconstructions) whose LGM values have been interpreted as implying lower surface temperatures in the tropics. It proves difficult, in the framework of this model, to reconcile the coldest of the indicated paleotemperatures with the observed snowline depression, which by itself is found to be consistent with an SST reduction of ~3 K. A cooling of this magnitude corresponds most closely to recent warm-pool LGM sea-surface temperature estimates based on Mg/Ca paleothermometry. Discordance among the various proxy reconstructions may result, at least in part, from regional variations in surface temperature change at the LGM.