The auroral events of the year 1716 most clearly announced that the prolonged solar and auroral calm that we now call the Maunder Minimum [Eddy, 1976a] had ended. But the onset of renewed auroral activity was noted already in the previous solar cycle. In 1707 an aurora was seen in Berlin and recorded in the journal of the Berlin Academy. Curiously, in New England, which is closer to the auroral zone than is London, Paris, or Berlin, the aurora returned suddenly in 1719. Contemporary accounts put the first recorded appearance of an aurora in Italy in the 1720's. In summary, prior to the twentieth century, long-term auroral variations were well established, and their investigation was an important subject under the general heading of auroral research. However, to the extent that the aim of the investigation was to determine the law of the variability it was not particularly successful. The 11-year variation was first discovered in the sunspot data, although it was then quickly found to be presen• in the auroral data also. The attempt to determine the period of the secular variation, which had yielded such a variety of results, was not long satisfied with the answer of 55.5 years. After further collection and examination of sunspot data, Wolf finally favored a secular period of 83 years, and a roughly 80-year period was subsequently assigned to the auroral data [Schove, 1955;Gleissberg, 1965;Link, 1968].In addition to increasing the suggested period for the secular variation to roughly 80 years, twentieth-century research led to suggestions of variations characterized by longer periods, namely 200 years [Schove, 1955] and 400 years [Link, 1968]. Attributing to these variations the property of periodicity is, however, less warranted than in the case of the 80-year cycle. The number of events in the pre-eighteenth-century auroral record has continued to grow as the result of historical searches in the twentieth century with additions from eastern and western sources. At the present time the longest-term variations emerge as a clear signal simply by plotting the recorded auroral frequency as a function of time. There are now enough events in the aurorally rich intervals to see distinct 11-year cycles. It is evident that the paucity of auroral reports during the Maunder Minimum does not reflect a lack of a familiar precedent. All of the early reports after the resumption of displays