To effect these calculations a fairly detailed knowledge of the different periods of the Sun and Moon was required so as at least to be able to fix the date of Easter. It should be remembered that in those days not only was it impossible to acquire a calendar but the actual spread of information, from one province to another and one nation to another, was considered so unreliable it was generally thought every episcopal seat could be left to deal with the matter in its own way.
If we accept that knowledge of the physical significance of eclipses of the Sun and Moon had not been completely forgotten in the late Latin period, even if the art of forecasting them had practically died out, we should also accept that the teaching of astronomy must have included a physics, or more precisely a geometry, part. It is possible there was also mention of the irregularity of planet motion along the ecliptic and more especially of the phenomena of planetary station and retrogradation.
Astrology was still the kind referred to as "Latin astrology", as explained in the treatise of Aratus (c.315-c.240 BC) or of Martianus Capella (IV-Vth century) and in the various "Latin spheres". The forecasts, of a generic nature, were based on somewhat elementary data such as the day of the week the new year began on, integrated by numbers drawn, by a variety of rules, from the letters making up the name of a person (7).
The Codex Angelica 123 allows us to test the accuracy of such hypotheses for Bologna, at least as regards computation. The first 16 pages of this Codex, compiled in the third decade of the XIth century, do in fact contain a fairly complex calendar covering the period 1039 to 1120, and the rules for its computation, according to the four year cycle of leap years, the weekly cycle of twenty-eight years - marked with the more archaic concurrentes than with the litterae dominicales - the 19 year lunar cycle and the 15 year indiction cycle.
The Ogdoadas and Hendecadas are also indicated in the calendar though there is only one brief reference in the text, not enough to be able to work them out. It is reasonable to assume that the presence of this latter cycle, with the characteristic Greek names used in the eastern church but rarely seen in western calendars, is evidence of the relationship between the Bolognese diocese and that of Ravenna which it was then subject to; it also indicates it was Ravenna that actually supplied the calendar even though the exact meaning of it as well as the rules for its computation were not known in Bologna.
Interspersed among the rules of computation in our text are astronomical ideas and quotations from Bede’s De natura rerum and Isidore’s Etymologiae. There are also some specific references to its didactic use in the dialogic forms proper to the prose part ("si vis scire", 5r and many other places, "vide ne oblivescaris..", 6v), gratification with the work done ("avidus calculandi inquisitor artis..", 8v), new discoveries ("Item argumentum nuper inventum..", 15r), divergent opinions ("non est verum quod de luna aliqui coniciunt..", 10v), and the possibility of experimental control of the data ("Horologi intentus qui perbene lineas - certas eorum comprobabit.." 14r) (8).
The computational examples (14v, 15v, 16r) all have 1029 as their year of reference. A lack of correspondence in the marking of leap years, in the concurrentes between text and calendar and a few errors in the calendar, together with a disagreement between the calendar and the computus starting dates, suggest that this codex preserves the transcription of an original of a few years before, compiled what is now almost a thousand years ago.