Further evidence for the idea of quarterly festivals can be found throughout Scandinavian and Northern European sources which repeatedly place them at intervals which we may call “the start of winter”, “midwinter”, “the start of summer” and “midsummer”. Estonian and Finnish staff calendars regularly mark the years by quarters. For example, summer (suvipäive) occurs about April 14, midsummer (keskikesä) occurs on 13-14 of July, winter (talvipäive) occurs on or about October 14th, and midwinter (talvenapa, ‘winter breeding’) falls on the 13-14 of January. These dates regularly appear in folktales, sagas, provincial laws and other “every day contexts”. Nordberg argues that these fixed dates stem from a pre-Julian method of counting 28 days (exactly 4 weeks) after each solstice or equinox. After adjusting them to the Gregorian calendar, and comparing them to their corresponding equinox/solstice, he demonstrates that they do indeed occur exactly 28 days following the solar event, except for in one case, which showed only a two day discrepancy. Nordberg provides the following dates.
Autumnal equinox 21 Sept. +28 Days= Winter Nights start 20 Oct.
Winter solstice 21 Dec. +28 Days= Midwinter starts 19 Jan.
Vernal equinox 20 March +30 Days= First day of summer starts 20 April
Summer solstice 21 June +28 Days= Midsummer starts 20 July
There is some difficulty with his argument, however. Nils Lithberg believed that the shift from one quarter to the next originally took place during the first full moon following the solstice/equinox, and that they became fixed into the Julian calendar when it was introduced to the Nordic countries. He argued that this would account for the fact that winternights regularly appear as three consecutive days. This can be seen in Vala-ljóts saga, “the third winternight” (hinar þriðju veturnætur), and we can see evidence of the three day pattern extending to the other calendric festivals. Snorri says that the heathen Yule began at midwinter night, and that Yule itself lasted for three nights after. The Dalalagen refers to both winter and summer nights in plural. Both Lithberg and Nordberg agree that these three day periods, observed most readily in winternights (veturnætur), had originally existed at the transitional period from one seasonal quarter to the next. Over time these dates would have been standardized into one day instead of three. Árni Björnsson notes that various Icelandic bishops attempted to shorten the three day period into one, and Gunnell observes that these three were a “liminal period belonging to neither season.” Lithberg explains these three days originally coincided with the three days that the moon was full in the middle of the lunar month.
Nordberg himself agrees that “[heathen religious] festivals were held at the time of a new or full moon,” and attempts to reconcile this information with his argument for fixed days by proposing that while the actual start of the seasonal quarters began 28 days after the solstice/equinox, the festivals related to these seasonal changes would have been celebrated on the actual full moon following solstice/equinox. His explanation for why the official start of each quarter was fixed to four weeks after a solstice/equinox is that otherwise the start of the quarters would be tied to the moons, and would therefore shift up to about a month in different years. It seems to me that if people had no problem celebrating their seasonal festivals on the full moon (which shifted up to a month), then they would have had no problem with the official shift in seasons taking place at that time as well. It seems only natural that this three night shift between seasons would coincide with the three nights on which the moon shone its brightest.
The ritual year began with the coming of winter and the winternights (veturnætur). The period of darkness that followed encompassed the Yule festivities, around midwinter. The gradual return of the sun in later months brought the thaw, the first days of summer, and the planting and ploughing. Festivals would have taken place at intervals that were relevant to the turning seasons and rituals related to the particular time of year would have been enacted. In Ynglinga saga, Snorri says that “there should be a sacrifice at the start of winter for a good year; one in the middle of winter for good growth; the third in the summer: that was a victory sacrifice.”  The Ágrip af Noregs Konnungasögum regarding Ólafr Tryggvason says:
He abolished blót and blót-drinking, in place of which, as a favor to the people, he ordained holiday drinking at Yule and Easter, St John’s Mass ale and an autumn-ale at Michael’s Mass.
Snorri’s account does not mention a midsummer festival, but the other three events that he notes correlate with the Ágrip, describing heathen rituals that were tied to the changing seasons of the year. The practice of replacing heathen festivals with “Christian” themes was common, and reflects the strategy described by Pope Gregory in a letter sent to the abbot Mellitus who was trying to convert the Anglo Saxons in the 7th century:
Tell him (Augustine) what I have, upon mature deliberation on the affair of the English, determined upon, that the temples of the idols in that nation ought not to be destroyed; but let the idols that are in them be destroyed; let holy water be made and sprinkled in the said temples, let altars be erected, and relics placed. For if those temples are well built, it is requisite that they be converted from the worship of devils to the service of the true God; that the nation, seeing that their temples are not destroyed, may remove error from their hearts, and knowing and adoring the true God, may the more familiarly resort to the places to which they have been accustomed. 
Essentially, the best way to convert heathen populations was to allow them to continue their festivals under a more acceptable, Christian guise, or to reduce their significance from religion or ritual to ‘story’ or ‘game’. It is also important to note that the observances that took place over the course of the year took many forms, and changed over time and place. At times it may be difficult to draw a line of distinction between naturally “mundane games” or “religious rituals” whether they are sporting events and competitions, drinking bouts, feasts, dramatic reenactments, or the sacrifice of objects and animals. In part, this may be the result of Christianization, but it can also be that no distinction was made between “mundane” and “religious”. Terry Gunnell notes that in other contexts, games and play activities seem to have been linked to religious holidays from a very early period in time and notes that even today we see traditional games such as football and tug-of-war taking place during the Yule season in Orkney and Shetland. “Games meetings” (leikmót) are often mentioned in the sagas, and they are often associated with seasonal festivals. In Egils saga a ball game (knattleikur) is held near the winternights. Eyrbyggja saga similarly states that the same type of game, knattleikur, was played as an autumn tradition among the Breiðvík men during “vetrnætur”. Of course this does not always mean that every game or dance recorded in the sagas was a religious ritual, and certainly any time there is a festival, that would imply the occurrence of games and social activities. But in many ways, games seemed to have formed a common part of the holiday on which they occurred. In addition, we will see other games and activities which clearly seem to have formed a direct part of a religious ritual.
 – Ágrip of Nóregskonungasögum 19, pp. 32-33.
 – Vilkuna, 1962, p. 43; Nordberg 2006, p. 41.
 – Nordberg, 2006, ch. 2 and p. 150.
 – Nordberg, 2006, explains the two day discrepancy in the relationship between the vernal equinox and the first day of summer as follows: “Of the four astronomical fix points, both solstices are the most straightforward to observe. The easiest way of determining the equinoxes is to assume that they occur halfway between the solstice; this is true of the autumnal equinox, but the astronomical spring equinox occurs a couple of days earlier than its assumed date. This is not discernable to the naked eye, however, and we can assume that the pre-Christian Nordic quarters started four weeks after the dates that were assumed to be the astronomically correct solstices and equinoxes.” p. 151.
 – Nordberg, 2006, p. 151.
 – Lithberg, 1921, p. 155, pp. 165-168.
 – Valla-Ljóts saga, in Íslendinga sögur, 1987, p. 1832.
 – Hákonar saga Góða, in Heimskringla, 1944, p. 97. “En áðr var jólahald hafit hökunótt, þat var miðsvetrarnótt, ok haldin þriggja nátta jól.”
 – Cited in Nordberg, 2006, p. 41.
 – Árni Björnnson, 1995, pp. 59-61; Gunnell, 2000, p. 128.
 – Gunnell, 2000, p. 128.
 – Nordberg, 2006, p. 153.
 – Ynglinga saga, in Heimskringla, 1944, pp. 9-10.
 – Ágrip af Nóregskonungasögum, 1995, pp. 30-32.
 – Colgrave & Mynors, 1981, pp. 107-109.
 – Wessén, 1921, p. 120; Gunnell, 2000, p. 32.
 – Gunnell, 2000, p. 33.
 – Gunnell, 2000, p. 129.
 – Egils saga, 2013, ch. 40, p. 77.
 – Eyrbyggja saga, 1935, p. 115.