---

Now let's consider the kinds of naked eye observations that one could make at this time (and still today):

• Sun rises and sets each day careful observers would see that the sun doesn't always rise or set in the same place each day.

• If you made a timing device (like a container of sand with a hole in it) you might notice that the length of time the sun appears in the sky varies. even a crude timing device would show this. For instance, at the latitude of Eugene the length of time the sun is in the sky varies by 8 hours from winter to summer. So if you had a sand clock that say held 8 hours worth of sand , you would easily figure out that sometimes the sun is in the sky longer than one contain's worth of stand - a lot longer.

• Seasonal climate variations were certainly recognized at the time and ultimately this could easily be related to how long the sun stayed in the sky, and the position where the sun set or rose.

• The moon has phases - or more simply, sometimes there are times when there is no visible moon in the sky (where did it go?) it always does seem to come back: This pattern most easily lends itself to a calendar where a "moon" (time between full moons) is a unit.

• The stars rise and set more importantly, different patterns of stars (e.g. constellations) are seen at different times of the year. This changing pattern is really quite obvious to anyone that looks at the night sky.

  For instance, at our latitude in the summer, if we look overhead we would see this pattern of stars with some specific constellations shown:

  

  In the winter, this pattern has clearly changed and compare the orientation of the big dipper, the little dipper and cassiopea in this view compared to the summer view.

  

Hence it really was obvious that at times, different patterns of stars would be in the overhead sky compared to other times and, that pattern variation was repeatable. This would allow one to develop "constellation time", which is essentially equivalent to a year.

So then how do you make a model of the earth plus celestial sphere that is consistent with these observations?

A stationary earth with everything else rotating around it is the most operationally simple model and can qualitatively account for the rising and setting motion of objects, but that's about it.

• This model really can't easily explain the change in the amount of "time" the sun is present in the daytime sky.

• Its unclear if this model adequately can explain lunar phases.