Sir

Paul Hoffman and Adam Maloof1, in discussing the paper by D. M. Williams et al.2, misconstrue the large-obliquity hypothesis for low-latitude glaciation near sea level during the Proterozoic3 and overlook relevant geological evidence.

Seasonality would be greatly amplified globally with a large obliquity (>54°). The solsticial climate would be severe in high latitudes but more benign near the Equator, where mean annual insolation would be minimal and where substantial snowfall could occur. The net accumulation of solsticial snow through successive equinoxes, necessary for low-latitude glaciation, would be aided by the high albedo of snow and ice. Importantly, Neoproterozoic glaciogenic deposits in Australia that formed near the palaeoequator4 contain seasonal freeze-thaw wedge structures in permafrost regolith that indicate marked seasonal changes of temperature3,4.

The ‘snowball Earth’ hypothesis for low-latitude glaciation5 is difficult to simulate6,7 and conflicts with geological evidence. The late Neoproterozoic stratigraphic record8 shows no sign of the drastic lowering of sea level that would accompany global glaciation. Moreover, seasonal freeze-thaw structures could not form near the Equator on a snowball Earth because the very low global temperatures would inhibit seasonal variation9.