From scan
to latitude

A sundial is first a surface. The catalogue names the type of each object as cone, sphere, plane or cylinder, and for 194 entries it does not say. The 3D model lets the register test the statement: cone, sphere and cylinder are fitted to the same set of points on the stone.

For 75 of the 113 graded objects one model is clearly better than the others, the fit is decisive: 52 cones and 23 spheres.

The remaining objects are left open. A fragment with few points cannot decide between a cone and a sphere; the register says so instead of choosing. Among them are 5 cylinders, none decisive.

The catalogue lists 782 dialface entries. Some objects carry several dialfaces, so the catalogue describes 694 objects.

For 243 objects the register holds a measurement record; 113 of them have detected lines and a grade. The others are not worse; their measurement is not finished.

For 72 objects a design latitude is determined, 58 as a consensus of several independent construction witnesses and 14 from a single witness.

For 57 objects the cone angle is shown. For 34 objects it is withheld because the fitted cone does not hold (see the open points on the gateway page).

A conic sundial is designed for a latitude. The three noon marks fix the cone but not the latitude; the latitude enters only through the division of the hour lines and the dressing of the stone planes. The register therefore combines independent construction paths, its witnesses.

The dial ObjID 126 from the Archaeological Museum of Delos has 7 witnesses. 4 of them enter the consensus: the hour lines of summer and winter (without an assumed obliquity) and two further construction paths, I and II of paper S6.

The consensus is 37.97° ± 0.80°. The figure marks the klimata Delos and Athens as given in paper S6.

The paper on this dial (S6) states a latitude of 38.01°; the register gives 37.97°. The two values are not reconciled, and this page shows both.

The 72 latitudes, in classes of five degrees. The colour tells how each was determined: consensus of witnesses, a single witness, or provisional.

44 of the 72 values lie between 30° and 45°, the range of the Greek and Roman Mediterranean. The obliquity of the ecliptic in the year −100 is 23.71°; the astronomical picture on the gateway page uses it.

Values below 25° are a warning, not a result: 6 latitudes lie there and 0 of them is a consensus without the provisional flag.

For 28 latitudes the cone fitted to the day circles leaves a residual above 2 mm. The register marks them provisional. The clearest case is ObjID 18 in the British Museum with 8.87 mm against 0.39 to 0.65 mm for four control cones; its latitude of 41.45° counts as provisional until the issue is settled.