The supposed tendency of particular types of weather to occur at about the same period every year, independently of and often in sharp contrast to the regular march of the seasons, has been described by R. Abercromby under the name of “recurrence,” and there is a large literature on the subject; especially in connection with periods of unseasonable temperature. While Indian summer is the most discussed example of recurrence in the American weather calendar, in the Old World more attention, both popular and scientific, has been devoted to a frosty period supposed to recur in May. With the elaboration of the ecclesiastical calendar, the frosts in question became definitely associated with the days dedicated to Saints Mamertus, Pancras, and Servatius (May 11, 12, 13), or, in south-central Europe, Saints Pancras, Servatius, and Boniface (May 12, 13, 14), hence known as the Ice Saints. These saints and their days are called in French _saints de glace_, and in German _Eisheiligen_, _Eismänner_, or _gestrenge Herren_. Yet other examples of the elusive phenomenon of recurrence are the “January thaw” of New England, the April “blackthorn winter” of England, and the June “sheep-cold” (_Schafkälte_) of Germany, dangerous to newly shorn sheep. In the middle of the last century the cold weather of the Ice Saints was variously ascribed to the melting of the ice and snow of high latitudes, the passage of periodic meteor showers between the earth and the sun, and other far-reaching terrestrial or cosmical causes. FitzRoy believed that the liberation of latent heat in autumn during the formation of ice in the circumpolar regions was accountable for Indian summer. A review of the whole body of literature concerning supposed recurrent irregularities in the annual march of temperature will be found in the “Monthly Weather Review” (Washington) for August, 1919. Whether recurrence, in Abercromby’s sense of the term, is a real phenomenon is still an unsettled question. Many periods of unseasonable weather occur in the course of each year, and it is easy for the uncritical observer to identify one of them with the Ice Saints, another with Indian summer, and so on. About the best that meteorologists can do at present is to explain each particular instance of such weather by reference to barometric and other conditions shown on the daily weather map. GLOSSARY The vast vocabulary of meteorology is very inadequately represented in ordinary dictionaries, and has never been made the subject of a comprehensive special glossary. The writer of this book has been gathering material toward such a glossary for some years, and from the material now in hand it appears that an approximately complete English meteorological dictionary, embracing both scientific and nonscientific terms relating to the atmosphere and its phenomena, would contain upward of fifteen thousand definitions.