Englisc Rímbóc: The Anglo-Saxon Calendar
Bede's Account of the Calendar in De Temporum Ratione
One of the least studied things in ancient Anglo-Saxon culture is the old pagan calendar. Yet, it is an area of most interest for many. What we know of the calendar was handed down to us by Bede in his work De Temporum Ratione. Unfortunately, while Bede gave us much information, he also left us in quite a mystery about how the calendar worked. We know not from his information alone whether the months were reckoned by the phases of the moon, and if so, whether they began on the Full or New Moon. We are perplexed how a fixed solar date could be the start of the year in a calendar that appears to be solar lunar (a calendar using both the Sun to keep track of years and the Moon to keep track of months), and even more so by when that fixed date occurs. Still, the information Bede gave us, along with other clues from Anglo-Saxon culture, the practices of other cultures, and comparison with the Icelandic calendar can result in a reliable reconstruction.
Bede begins his account of the old Heathen calendar by saying
Antiqui autem anglorum populi (neque enim mihi congrum videtur aliarum gentium annalem observantiam dicere et meæ reticere) iutxa cursum lunæ suos menses computavere. Unde et a luna hebræorum et græcorum more nomen accipiuiunt; siduidem apud eos luna mona, mensis appellatur monath.
The ancient English peoples -- for it does not seem to me proper to explain the yearly observance of other nations, and to keep silence concerning my own -- reckoned their months by the course of the moon, just as they were named from the moon in Hebrew and Greek. (Charles W, Jones translation)
Thus the mystery begins from the start of his text. Most have thought that "cursum lunæ suos menses computavere (by the course of the moon calculated)" indicates that the months were determined by the phases of the Moon. However, besides keeping track of time through the phases of the Moon, one can also keep track of time by the path the Moon takes through the sky. Using the course of the Moon to keep track of time results in using what modern astronomers call a sideral month, which is 27 days 7 hours and 43 minutes long. Every 27 days the moon returns to the same position in the sky it was 27 days before. The scholar Vaster Guðmundsson believed that this was the form of month the Norse used, and used it in his theoretical reconstruction of the ancient Scandinavian calendar (Guðmundsson. 1924, p.88). It is possible then that the Anglo-Saxons also did the same. However, as Bede draws a comparison to the Greek and Hebrew calendars, we may want to assume that the Anglo-Saxons used a synodic month (a month measured from a phase of the Moon to the next time that phase of the Moon occurs). There are other clues in Bede's account, that indicate this was so, and I will touch on those later. Bede then goes on to name the months of the old Anglo-Saxon calendar and further gives the corresponding Roman month.
Primusqu eorum mesis, quem latini ianuarium vocant, dicitur giuli; deinde februarius, solmonah; martius, hredmonath; aprilis, eosturmonath; maius, thrimilchi; iunius lida; iulius, similiter lida; augustus, vveodmonath; september, halegmonath; october, vvinterfilleth; november, blodmonath; december, giuli eodem quo ianuarius nomine vocatur.
The first month, which the Romans name January, is with them Giuli. Then follow February, Solmónaþ;March, Hrédmónaþ;April, Éosturmónaþ; May, Þrimilchi; June, Líþa; July also Líþa; August, Wéodmónaþ; September, Háligmónaþ; October,Winterfylleþ; November,Blótmónaþ; Decemeber, Giuuli, the same as for January.(Charles W, Jones translation)
At first this may not seem important, however, it shows that the Anglo-Saxon months roughly followed the Roman ones, enough so that Bede could draw correspondences. This shows what we have suspected, that the calendar was a solar lunar one, not a straight lunar calendar. And while it does not rule out the use of the sideral month, it increases the odds that the Anglo-Saxons used a synodic month. Sideral months being shorter would move more within the solar year. Synodic months being closer in length to the Roman fixed months would make for a closer comparison, and not move as much in relation to the seasons as long as intercalary (leap) months were used.
Bede then touches on when the year started. Something he has already hinted at by naming Giuli, the month corresponding to January as the first month.
Incipiebant autem annum ab ocatavo kelendarum ianuariarum die, ubi nunc natalem domini celebramus. Et ipasm noctem nunc nobis sacrosanctam, tunc gentili vocabulo modranect, id est matrum noctem, appellabant, ob causam, ut suspicamur, ceremoniarum quas in ea pervigiles agebant.
They began the year with December 25, the day we now celebrate as Christmas; and the very night to which we attach special sanctity they designated by the heathen term módraniht, that is, the mothers' night --- a name bestowed, I suspect, on account of the ceremonies they performed while watching this night through. (Charles W, Jones translation)
At that time, under the old Julian calendar, December 25 (or eight days before the calends of January as Bede puts it) was also the winter solstice. The problem with this is that if the Anglo-Saxon calendar was a solar lunar calendar, then it could not have a starting point that was a fixed solar date (at least not have such a fixed date and operate with any accuracy). Bede may have confused an Anglo-Saxon Heathen mid-winter festival with a New Year's celebration. Or it is possible the actual start of the new year was on the Full or New Moon near the solstice. It is also possible the Anglo-Saxons used a sideral month and somehow managed to reconcile its differences with the solar year (though it would be as difficult to do this as reconciling the start of a lunar calendar using synodic months in such a fashion). Finally, the most distinct possibility perhaps is the Anglo-Saxon Heathens used more than one calendar for more than one purpose. That is they could have used a solar calendar separate from the one Bede presented, and it was the solar calendar whose new year began on December 25. There have been many theories as to why the year began on December 25, ranging from Roman borrowing to influences of Mithraism, but few have thought it the start of a separate calendar. The possibility is distinct however.
The Icelanders, for example, did use a solar calendar which they established in 930 CE in conjunction with the Norse lunar months (which ceased to be lunar), but reformed again in 955. The core of this calendar consisted of two seasons, winter and summer for a year of 364 days or 52 weeks. Each season was 26 weeks long. To this calendar was added one week every seven years as a "leap week," to keep it in line with the solar year (much as a day is added every four years to the Gregorian calendar today). New Year's Day was Veturnætur or "Winter Nights," a time near the Fall equinox. While some scholars have attributed the Icelanders adopting a solar calendar as being due to Classical and Christian influence (via trade relations with the then Christian Anglo-Saxons and Irish), one cannot rule out the possibility a solar calendar was a native idea to most Germanic peoples. The Anglo-Saxons like the Icelanders also had only two seasons, Summer and Winter, and this is even stated by Bede (though elsewhere he refers to four, a comment that can be attributed to his Church education which held there were four seasons, not two). It would not be surprising then if they used a separate solar calendar based on the equinoxes and solstices, perhaps even using weeks for time reckoning much like the Icelanders did. Further it would not be far fetched if they saw Módranect as some sort of marker in this calendar. In Bede's account the start of the year, he notes the start of Winter (while linked to a Full Moon, it would be close to what we know as autumn), the Winter solstice, and the fact months were added after the Summer solstice. It would be difficult therefore to assume that the Anglo-Saxons and Germanic peoples in general could not have kept track of time by the Sun. Bede then goes on to talk about the use of an intercalary month called Þrilíða.
cum vero temporibus, hoc est xiii mensium lunarium, annus occurreret, superfluum mensem æstati apponebant, ita ut tunc tres menses simul lida nomine vocarentur, et ob id annus ille thrilidi cognominabatur habens quattuor menses æstatis, ternos ut semper temporum cæsterorum.
When, however, an embolism occurred, that is, a year of thirteen lunar months, they added the intercalated month to the summer, so that in that case three months in succession were called Líða. Such a year was known as Þrilíði, having four months of summer, and three of the other seasons. (Charles W, Jones translation)
We know from this statement by Bede that Þrilíði was added in some years in the Summer to bring the calendar back in line with the solar cycle. This again is a clue that the calendar was indeed a solar lunar one, most likely using synodic months (using one phase of the Moon to its next occurrence as a way of measuring a month). The Chinese, Hebrews, and even the ancient Romans used similar methods to adjust their calendars.
Item principaliter annum totem in duo tempora, hiemis videlicet et æstatis, dispertiebant - sex illos menses quibus longiores sunt noctinus dies æstati tribuendo, sex reliquos hiemi. Unde et mensem quo hiemalia tempora incipiebant vvinterfilleth apbellabant, composito nomine ab jieme et plenilunio quia videlicet a plenilunio eiusdem mensis hiems sortiretur initium.
The general division of the year was into two seasons, winter and summer, summer comprising the six months in which the days are longer than the nights, and winter the others. Hence the month with which they began the winter season was called Winterfylleþ, a name compounded of the terms for winter and full moon, because from the full moon of that month winter was esteemed to begin. (Charles W, Jones translation)
It is worth noting that the time of Winterfylleþwas also the time the Old Norse started their year (Veturnætur or "Winter Nights"), and there has been speculation on whether or not the actual new year of the calendar presented by Bede did not start at this time also, that Bede was mistaken. That Winter could be used of the entire year and not just the season is apparent to anyone that has studied Old English. The Anglo-Saxons counted years by winters and even had such terms as wintergetel "a number of years" and wintersteal "a one year old stallion." Garman Lord in The Way of the Heathen even theorizes that perhaps the name Winterfylleþ could have meant something closer to "New Year's Day." More important though, what Bede has to say about Winterfylleþ is also a clue that the months began with the Full Moon. Bede goes on to discuss the meaning of the month names, and as such this is his extent of clues on how the calendar operated. For more information on exactly how the calendar worked we must go outside his text to Anglo-Saxon concepts of the day, as well as draw from other cultures with solar lunar calendars.
One of the primary problems with the calendar is determining when the months began. Since few ancient cultures used the sideral month in calendars of this type, we can safely assume perhaps that the ancient Angles, Saxons, and Jutes used a synodic month like the Chinese, Hebrews, ancient Romans, and most other peoples in the world. But then we are faced with the question of when did the months begin? The only clue Bede gives us is in the name of Winterfylleþ, when he states that Winter was said to start from the Full Moon of that month. If Winter started on the Full Moon of Winterfylleþ and each season only had six months (not portions of months but whole months) then we almost must assume that Winterfylleþ began on the Full Moon. Then again Bede states Winter began on the Full Moon of that month, meaning that the month its self might not begin on the Full Moon. For a solution to this dilemma we must look to other cultures with similar ways of seeing tides as well as closely related cultures.
A bronze plaque with a calendar used by the Gauls engraved on it was found in Coligny, France, in 1897, and dated to about 50 CE. This calendar consisted of 12 months with names of Celtic origin. Further these months began on the Full Moon. The Gauls had lived in close proximity to the Germanic peoples for centuries. At one point it is possible that some of the Germanic tribes had developed a fascination with Celtic culture (or alternately been subjugated by a Celtic people). Within the Germanic languages there are several Celtic borrowings of great antiquity. These borrowings were words for rulership and warfare. Among them are the modern English word iron, Old English ríce, "kingdom;" and Old English ambeht, "servant." It is possible then that due to Celtic influence the Anglo-Saxons used the Full Moon as the marker for the start of a new month. Most societies that use a solar lunar calendar however do not use the Full Moon as a starting point for months.
Indeed, there is just as much evidence when we look at other cultures for the Anglo-Saxon calendar beginning its months on the first crescent of the New Moon. Further the evidence is also more convincing when common sense is applied. The first and best argument is that it is difficult to determine precisely when the Full Moon is occurring in Northern Europe. Indeed, it can appear to be full for three days, when in truth, only one of these days is the Full Moon. The New Moon can be almost as difficult to determine, but not quite as difficult. One can always look to the First Crescent of the New Moon, something easy to spot if one is watching for it, and not easily mistakened for another phase. This is precisely the moon phase many cultures such as the Hebrew and Babylonian cultures used. Further, cultures that start their day at sunset, also usually begin their year in the Winter months and their months (if they use a solar lunar calendar) with the New Moon or the First Crescent. The ancient Roman calendar prior to revisions by the Republic also started its months with the First Crescent as did that of the ancient Greek cultures. Perhaps the best evidence arguing for the Anglo-Saxons using the First Crescent is that of the Lithuanian calendar. The Lithuanian calendar not only started its months on the First Crescent of the New Moon, but also had a midwinter celebration the same time as Yule. Oddly enough though it started its year in April (Straižys and Klimka). Perhaps of all Indo-European peoples the Balts have the most in common with the Germanic. Even their religion is very close to the ancient Germanic one with components such as a World Tree, and deities similar to our own such as the thunder god Perkunas who is not a far cry from the ever familiar Þunor (Thor), not to mention a Sun goddess and Moon god. Unfortunately, especially since it may be the closest comparison to the Anglo-Saxon calendar, not much has been written on the Lithuanian calendar in English. Yet considering the close relationship of the Balts to the Germanic peoples, it presents a good argument for the Anglo-Saxon months starting with the First Crescent, and not the Full Moon.
Another clue to the months starting on the First Crescent is when the Anglo-Saxons began their day. Other cultures such as the Hebrew that use the First Crescent as the starting point for their months, start their day at sunset (the Lithuanians seem to be an exception). We find this too with the Anglo-Saxons. Wódenesniht to an ancient Engle was not Wednesday night but Tuesday night; Wednesday began on what we would think of as Tuesday evening. We can also see this start of the day at Sunset with the festivals, for example, Módraniht mentioned by Bede. Modern survivals of this include Halloween (a contraction of "All Hallows Evening"), New Years Eve, and Walpurgis Night. The Old English word niht, not dæg was used for counting, that is one would say "10 nihtas" not "10 dægas" (a modern survival of this is our term fortnight). This too was indicative of cultures that started their months on the First Crescent. We can therefore probably safely assume therefore that the Anglo-Saxons used the First Crescent as a marker for the start of a month, although we can never be one hundred per cent certain (barring finding a lost document dating from the period detailing such calendar information).
Other Potential Calendars
As mentioned above, a solar lunar calendar may not have been the only way that the Anglo-Saxons kept track of times longer than a day. There were the two seasons calledmissera in Old English. We know that they had two seasons, but we do not know if they were used in such a way as the Icelanders later did. For the Icelanders, themissera played a more important role than the solar year. Indeed, their calendar was not based so much on the solar year as it was the half year. As detailed above, their calendar consisted of the two seasons, numbered 26 weeks each. The two seasons were further broken down into weeks. The week has always been a problematic time unit however. Most scholars view it as a borrowing from the Romans, who in turn borrowed it from the Greeks. Ultimately it is seen as having come from the Semitic cultures of the Middle East with its sole purpose being religious. That is scholars feel the week is a manmade time unit with no bearing on astronomical or natural events (unlike the month, the year, the day). The problem is that within the written record of many peoples such as the Celts and Germanics, there is no sign of there not being a 7 day week, or at least a week of some sort. So while scholars can claim that the week is a borrowing, they cannot prove definitively it is a borrowing. This problem is further complicated when one looks at how a week can be used in time keeping. The synodic month is approximately 29 days and 12 hours long. Four seven day weeks could then be used as a rough division of the month (being only a day or so off). This is somewhat confirmed by the origins of the word week which scholars think comes from an Indo-European word meaning "to turn." In other words, the word week might have originally referred to the turning of the phases of the moon (and it also may have been a longer time unit than now). Even there we are on shaky ground as while some consider it a native Germanic term, others see it as coming from Latinvices "recurrences."
While the week as we know it now may be a borrowing, the concept of a unit of time lasting several days (but only a fraction of a month) may not be. The Lithuanians used a nine day week at one point, which makes for a good division of the the sideral month. As mentioned above a sideral month, is 27 days 7 hours and 43 minutes long. Therefore the week in some form may have always been a Germanic time keeping unit (however, it may not have always been seven days long). Vaster Guðmundsson believed that the Norse used a five day week prior to the borrowing of the seven day week, and used this in his reconstruction of the Norse calendar. However, his week has no bearing on astronomical events either. The solar year is more evenly divided by five week periods though, and there is an Old Norse legal term referring to such a five day period as a fimmt. There could no doubt be other time units similar to the week we do not know of that the ancient Germanic peoples, and thus the ancient Angles, Saxons, and Jutes may have used. That information though is unfortunately lost if it ever existed.
Regardless, that they could have used a solar calendar similar to the Icelandic one after Roman contact is a certainty. They would have had the time unit of the week, and had all the astronomical knowledge to use one (which they had well before Roman contact). It could well be that such a solar calendar is seen lurking behind the solar lunar one as presented by Bede. He does after all refer to solar events such as the start of the year on the winter solstice and the seasons in his description of the calendar. The only problem is, we can keep on saying "could," as there is no evidence truly for or against the Anglo-Saxons using a solely solar calendar like that of the Icelanders. What we do know is that missrera could also be used to mean years, as its Old Norse cognate missari could also. Indeed, the Norse did not even truly have a term for the whole year. If the missrera then played a more important part in Anglo-Saxon time keeping, they would most certainly had a solar calendar as well.
Other Methods of Telling Time Shorter than a Day
On quite firmer ground than the possible use of a solar calendar is how
the Anglo-Saxons divided the day. It would appear that they, like their Norse
cousins divided the day into eight even divisions, or stundas(sometimes eferred to as a tíd in Old English). Early
Anglo-Saxon sundials asol-merca or d&selig;gmæl) show only four
evenly spaced marks for telling time during daylight (these four divisions of
daylight are paralleled by four at night). The names of these eight divisions
are seen throughout Old English literature. They are: úht (roughly 3
am to 6 am), morgen (roughly 6 am to 9 am),undern (roughly 9 am
to noon),middæg (roughly noon to 3 pm), gelotendæg (roughly 3
pm to 6 pm), æfen (roughly 6 pm to 9 pm), niht (roughly 9 pm
to midnight), and midniht (roughly midnight to 3 am). Our usage of
such words as morning, noon, and evening to divide up the day are but
vestiges of this time keeping method. There is a very good article on how the
Norse used these divisions for time keeping at: http://hea-www.harvard.edu/ECT/Daymarks/
called Telling Time without a Clock: Scandinavian Daymarks. The
methods described in it are no doubt the same as used by the ancient
The earliest calendars of most ancient societies seem to have been established to keep track of religious observances. Agriculture, the other great preoccupation of ancient peoples had no real need for a calendar as the time to plant or reap was fairly obvious by observing the weather. Bede mentions several potential Anglo-Saxon holytides but did not unfortunately elaborate on them. Snorri does say if the Old Norse in the Heimskringla that:
þá skyldi blóta í móti vetri til árs, enn at miðjum vetri blóta
til gróðrar, it þriðja at sumri, þat var sigrblót
"On winter day there should be blood-sacrifice for a good year, and in the middle of winter for a good crop; and the third sacrifice should be on summer day, for victory in battle." (Ynglinga Saga Chapter 8)
The ancient Anglo-Saxon Heathens seem to have paralleled these three great holy tides of the Norse. Libermann in The National Assembly in the Anglo-Saxon Period notes that the Anglo-Saxon witanagemót met most often on St. Martin's Day (November 10th), Christmas, and Easter or Whitsunday. These dates correspond to when Anglo-Saxon kings are reported to have worn their crowns (Chaney, Cult of Anglo-Saxon Kingship, p. 65). If we accept Bede's description of the Anglo-Saxon pagan calendar in De Temporum Ratione there may have been more Anglo-Saxon pagan holy tides. Bede as stated above started the Heathen year withModranect, the “Mothers Night.” It fell between Ærra Geola, our December and Æfterra Geola, or January, and is the period today we know as Yule (which is now no more than a synonym for Christmas for most people). Of Solmonað, roughly our February, Bede says the Anglo-Saxons offered cakes to their Gods, and thus it was named the month of cakes; he also mentions Hreðmonað, roughly our March as when the Goddess Hreda was worshipped, followed by Éastremonaþ when the Goddess Éostre was worshipped. He does not name Líða as a sacred month, however, that Midsummer falls within it, there may have been a holy day corresponding to Mid-Winter or Yule. This is pretty much confirmed by Midsummer celebrations that survived into modern times in England, not to mention much folklore surrounding St. John's Day which falls near the summer solstice. Bede then mentions Haligmónaþ, roughly our September, which was called "holy" as in Bede's words "because our ancestors, when they were heathen, paid their devil tribute in that month." The next potential holy tide mentioned by Bede is Blótmonaþ, roughly our November. The name its self means "sacrifice month" and was the time when animals were slaughtered for the coming winter. It follows Winterfylleþ which corresponds to the Norse Winter Nights or Winter's Day, the time the ancients reckoned winter to have started. That the relatively rescent holidays of All Hallows, St. Martin's Day, and Guy Fawkes Day, all important English holidays fall in this period may be a memory that an actual holy tide took place near or at the junction of the two months. One of the problems in reconstructing the Anglo-Saxon Heathen calendar is that we do not know the criteria for dating a holy tide year to year with the exception of Modranect (and of course if we accept a similar holy tide on Midsummer). Heathen Easter for example may have been dated by the Spring equinox, the same way the Christian Easter is by the first Full Moon after the spring equinox, or by the first New Moon of Spring.
A Reconstruction of the Heathen Anglo-Saxon Calendar
A reliable reconstruction of the Heathen Anglo-Saxon calendar is possible (though whether it comes close to how the ancient one was truly reckoned is another matter). First we must make some assumptions based on Bede as we have few facts on how the calendar operated. These assumptions are: a. the month began on the First Crescent of the New Moon; b. the month was a synodic month; c. leap months were periodically used to pull the lunar calendar back in line with the solar year; d. the months of Líða and Géola are double months (that is they are roughly 57 days long). The first two assumptions are based on comparisons with other cultures that use a lunar or solar lunar calendar, the last two based on the statements by Bede on the month of Þrilíða and the names of the months. In addition we a clue provided by Bede as to how the calendar should operate. Modranect is a set solar event that will not change year from year and therefore the calendar must accommodate it in some way. In order to make a solar lunar calendar work, we must first make up for the difference between the lunar and solar years. A lunar year using synodic months is 354 days long while a solar year is about 365. Using straight lunar months with no leap days, weeks, or months would result in the calendar being off by over an entire month in under four years. The Muslim calendar does this, allowing its months and holy tides to float around the year. However, as indicated by Bede (as well as the month names) this does not seem to be the case for the Anglo-Saxons. Therefore one needs to use leap months to keep the two calendars (solar and lunar in line). One method that works for this is to use the Metonic Cycle. The Metonic cycle was discovered by Meton of Athens (about. 440 BCE) who noticed that 235 lunar months made up almost exactly 19 solar years. Thus phases of the Moon fall on exactly the same solar dates every 19 years. By adding Þrilíða in the 3rd, 6th, 8th, 11th, 14th, 17th, and 19th years of this cycle one can keep the lunar months roughly in line with the solar calendar. Another method developed by Professor Kenneth Harrison involved using the the octaëteris or ogdoas. The octaëteris or ogdoas is a span of eight years of 99 lunar months at the start of the Metonic cycle following the pattern of OOEOOEOE where O is an ordinary year and E is a leap year. This method follows roughly the same pattern as the one above only over a shorter period. John Robert Stone based on Harrison's model created several rules that perhaps are more practical. These are (quoted exactly): 1. The next month is intercalary if the first crescent of the after Líða is observed on or before July 4, the eleventh evening after Midsummer Eve (June 23). 2. The next month is intercalary if the first crescent of the after Líða is observed before Midsummer. 3. The next summer will contain a third Líða if the first crescent of the after Géola is observed within the eves of Christmastide (December 24 to January 4). Stone's method does work but it also has problems. The first is he is working on the idea of an Ærra Géola and Æftera Géola as well as Ærra Líða and Æftera Líða. Bede does not mention these month names, and while the names for Géola are paralleled in the Gothic month names ofFruma Jiuleis and Aftuma Jiulea, there is little else to suggest these were separate months and not double months. The primary problemwith using Ærra Géola, Æftera Géola and Ærra Líða, Æftera Líða is it is near impossible to construct a calendar that will sandwich the solar dates of Yule and Midsummer between their respective months. This is perhaps only possible if one uses double months for Líða and Géola. The Norse once used a calendar of only six 59 day months (or double months) which indicates the idea was not alien to Germanic peoples. If one makes Líða and Géola double months, one can then ensure that Yule and Midsummer always fall within their respective months regardless of which set of rules one uses. The next problem is reconciling the start of the year with the beginning of the months. Bede clearly states that the year began on Módraniht and gives a date of December 25th, the old winter solstice date under the Julian calendar. It is not truly possible for the first month of the year to begin on this date every year as the First Crescent of the New Moon will not fall on December 25th each and every year. Indeed, the majority of the time the First Crescent will not fall on December 25th. We must assume then that perhaps Bede was mistaken, and that Módraniht merely fell near the start of the new month (perhaps within 12 days of it). It is also possible that the Anglo-Saxon new year was actually in the Fall, and that it began with Winterfylleþ. The Norse started their year at this time and referred to it as Veturnætur or "Winter Nights." In closing one can create a reconstruction by considering the following:
1) Líða and Géola are considered double months.
2) Módraniht fell within 12 nights of the First Crescent of the New Moon.
3) Intercalatory months were added in the 3rd, 6th, 8th, 11th, 14th, 17th, and 19th years of the Metonic Cycle.
Baity E.C.Archaeoastronomy and Ethnoastronomy so far // Current Anthropology. 1973, 14. p.389-449.
Harrison, Kenneth, The Framework of Anglo-Saxon History to A.D. 900 Cambridge: Cambridge University Press 1976
Hastrup, Kirsten, Culture and History in Medieval Iceland. Oxford: Claredon Press 1985
Hutton, Ronald The Pagan Religions of the Ancient British Isles; Their Nature and Legacy
Nilsson, Martin P. Primitive time-reckoning, Lund: C. W. K. Gleerup, 1920
Straižys, Vytautas and Klimka, Libertas (12, Vilnius 2600) Chapter 5. Natural rythms and calendar, COSMOLOGY OF THE ANCIENT BALTS Retrieved 2 Sept., 2004 from http://www.lithuanian.net/mitai/cosmos/baltai5.htm
Stone, John Robert, Observing Bede's Anglo-Saxon Calendar Retrieved 2, Sept., 2004
Download a reconstruction using the Metonic cycle of this calendar here.