1. IntroductionAlthough Uranus was 'officially' discovered by William Herschel (1738-1822) on 13 March 1781, it had previously been sighted on 22 occasions between 1690 and 1771 by four different astronomers.' But there is reason to think that a naked-eye observation of the planet was made in Antiquity.The systematic error of -1 ° found in the longitudes of the Almagest star catalogue has raised many controversies about the identity of its author. Several ingenious theories2 have been suggested to solve the problem, favouring either Ptolemy or Hipparchus. However the key to the mystery may lie in the description of a star pattern in the constellation Virgo.All the versions of the Almagest agree about the presence of a quadrilateral in the left thigh of the Virgin, a part of the constellation Virgo located approximately three degrees northeast of Spica. While the coordinates given for the two stars forming the northern side of the quadrangle also agree, they do not for the two stars forming the southern side. The reason is obvious: there is only one star of the same brightness at the lower left corner of the figure. The purpose of this article is to show that the object (Baily 513 or Virgo 17) occupying the bottom right corner of the quadrangle was none other than the planet Uranus.The fundamental argument of this paper is the following:(1) There is a widespread consensus among experts (Toomer, Kunitzsch, Peters and Knobel, etc.) that the four stars in the quadrilateral in the left thigh of Virgo are as shown in Table 1.(2) There is an equally widespread realization within this consensus that this result is awkward and troubling. The quadrilateral does not look like a quadrilateral, and the coordinates do not match well for the supposed Virgo 17 = 76 Vir (see Figure 1 (left)).(3) There is a simple way to fix things. For Virgo 19, read s = -4 rather than -3, in agreement with a couple of Almagest manuscripts. Then the Vir 19 coordinates match very well to 76 Vir.(4) But then what matches to Virgo 17? There is a very good match in coordinates and magnitude to Virgo 17 if we assume that this celestial body was the planet Uranus observed by Hipparchus about 128 b.c., and the quadrilateral is now more apparent (see Figure 1 (right)).In Tables 1 and 2, Flamsteed numbers are used in column Identity; visual magnitudes are indicated in column V; and the resulting errors are given in arc minutes. As will be discussed in more detail below, the versions of the Almagest supporting the Uranus solution are the so-called Arabic versions while those favouring the Consensus solution are Greek.In the remainder of this article, all dates are in the Julian calendar and unless otherwise stated, the years before Christ are written in the manner of astronomers (for instance 128 b.c. corresponds to -127), celestial positions are referred to the mean equinox of the year -128.0 and the time scale used is Universal Time.2. The Quadrilateral in the Left Thigh of the VirginBefore examining the appearance and the localization of this asterism in detail, let us first attempt to define the term quadrilateral (in Greek isiparcAsupov) which is rather puzzling because this shape may designate any convex polygon with four sides. The whole of the star catalogue found in Books 7 and 8 of the Almagest contains 14 such figures of which 12 are clearly identifiable. Among these 12 figures, 6 refer to a rectangle (Dra, Lep, Psc, Ser, UMa, UMi), 4 to a trapezium (Cet, Cnc, Ori, Sgr), 1 to a rhombus (Del), and 1 to a parallelogram (Tau) formed by four stars of nearly the same brightness. Apart from the unresolved quadrangle in Virgo, another ambiguous figure lies in the constellation Cetus. However, in that case, the problem is due the presence of several candidate stars in the vicinity.This clarification made, let us now settle the position of the quadrangle with the help of two different celestial charts: an ancient and a modern one. …
Read full abstract