REVIEW 4 major objections 4 minor 26 references
Bo\v{s}kovi\'{c}'s Spherical Trigonometric Solution for Determining the Axis and Rate of Solar Rotation by Observing Sunspots in 1777
T0 review · 4 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read This paper derives, for the first time, the closed-form spherical-trigonometric equations behind Bošković's 1785 verbal method, and reproduces his 1777 solar rotation elements from three sunspot positions.
desk verdict A genuinely useful first derivation of Bošković's spherical-trig method, but the 'no approximation' and 10^-5 arcsec claims need to be walked back. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the configuration of eight oblique spherical triangles (plus the short-solution triangles △10 and △11) built from the northern ecliptic pole $P$, the northern solar-equator pole $P'$, and three sunspot positions $C$, $C'$, $C''$ in ecliptic coordinates. Sides such as $CC'$ and $P'C''$ are obtained by repeated cosine and cotangent rules; midpoints $E$ and $E'$ create right triangles through which $P'E'$ is found; and mirror symmetry of triangles △9 and △6 sets $P'C' = P'C''$. The final elements come from the side $PP' = i$ via the cosine rule, the angle at $P$ giving the node $\Omega$, and the angle at $P'$ combined with elapsed mean solar time giving $T'$.
What would settle it
Compute high-precision heliographic positions of a single sunspot at three times, apply these equations, and compare with an independent least-squares rotation-element fit over a longer track; if the three-position values differ from the fit by more than the measurement uncertainties, the rigid-parallel assumption is the source of the bias.
Extended reading notes
Core claim
The paper's central claim is that Bošković's §VII, №76–№81 description corresponds to an exact chain of spherical triangles linking the northern ecliptic pole, the northern solar-equator pole, and three observed positions of one sunspot in ecliptic coordinates. Solving the chain with cosine and cotangent rules yields the arc between the poles as the inclination $i$, the angle at the ecliptic pole as the node $\Omega$, and the angle at the equator pole divided by the elapsed mean solar time as the sidereal period $T'$. The reported results are $i = 6.80728^\circ = 6^\circ48'26.20337''$, $\Omega = 74.04774^\circ = 74^\circ02'51.87646''$, and $T' = 26.806232$ days, closely matching Bošković's planar solution and his sidereal period. The same equations over all twenty triples of his six positions produce a wide spread in $i$ and $\Omega$, so the choice of well-separated positions matters, and the paper concludes that the spherical solution is complete and gives all three elements without approximation.
Load-bearing premise
The chain assumes the spot stays at one solar latitude and the Sun rotates rigidly between the observations, so the pole-to-spot distance is the same for the first and third positions and the triangles on either side of the midpoint are mirror images.
Editorial extensions
If this is right
- Three well-separated positions of one spot are enough to return $i$, $\Omega$, and $T'$ simultaneously from closed-form trigonometry, so no least-squares fit or initial guess is required.
- Applying the new equations to Bošković's six-day 1777 sequence reproduces his published values ($i \approx 6^\circ49'$, $\Omega \approx 74^\circ03'$, $T' \approx 26.8$ d), strengthening the case that his verbal procedure was numerically complete.
- Relative to the Carrington and Spörer elements for 1777, the spherical solution's errors are about 2–3% in $\Omega$, 2–7% in $i$, and under 1% in $T'$, making it a serviceable quick estimator for sparse historical data.
- Because the same chain works for any three tracked surface features, the paper's stated generalization to other rotating bodies such as stars and exoplanets follows directly.
Reading between the lines
- The derivation's rigid-parallel assumption means that differential rotation will leak into the inferred elements; the three period estimates in Table 4 (26.77, 26.81, 26.85 days) already bracket the effect, so feeding the equations modern latitude-resolved spot data would quantify the bias.
- A sharper historical test would re-reduce Bošković's apparent-disk measurements with modern ephemerides before applying the new equations; any shift in $\Omega$ and $i$ away from his 1785 values would measure how much log-table rounding contributed to the original result.
- The closed-form chain could be inverted symbolically to propagate measurement errors into $i$, $\Omega$, and $T'$ analytically, giving uncertainty estimates without Monte Carlo sampling.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reconstructs the trigonometric spherical solution that Ruđer Bošković described verbally in his 1785 Opuscule II for determining the solar rotation elements (i, Ω, and the sidereal period T′) from three observed positions of the same sunspot. The authors derive a sequence of spherical-trigonometric equations for the full and the “short” solutions, apply them to Bošković's September 1777 sunspot positions (positions 1, 3, and 6), and obtain i = 6.80728° (≈6°48′26″), Ω = 74.04774° (≈74°03′), and T′ = 26.806 d (≈26.81 d), closely reproducing Bošković's published values from the planar-trigonometric solution. They also compare with a contemporary method (Roša et al. 2021) and with their own unpublished vector formalism method, and they examine the sensitivity of the result to the choice of triple via a 20-combination table.
Significance. If the reconstruction is correct, the paper closes a historical gap by turning Bošković's verbal description into a modern algorithm, and it appears to be the first complete equation set for that method. The derivation is self-contained and the worked example is internally consistent, with no free parameters, and it reproduces the arcminute-level published values—a concrete success that supports the historical claim. The main value is historical-methodological rather than a new solar-physics measurement. The paper's stronger claims (arcsecond-level precision, “without any approximation,” a single unambiguous procedure, and independent VFM confirmation) are not supported by the data or by the derivation as written, and these claims should be revised. With those revisions the paper would be a sound contribution to the history of solar rotation measurements.
major comments (4)
- [§2.3, §2.4.1, Eqs. (26), (34), (43), (50)–(51); §4.2, Table 6] The derivation is exact only under the assumption that the three positions lie on a single heliographic parallel, i.e. P′C = P′C′ = P′C′′, combined with rigid rotation. This equality is asserted from the mirror-symmetry of triangles △9 and △6 (§2.4.1) and is not derived from the measurements. The conclusion's statement that the spherical solution consists of “closed equations without any approximation” is therefore unsupported: the trigonometric algebra is exact, but the physical model contains the constant-latitude, rigid-rotation approximation. The sensitivity of this assumption is visible in Table 6, where the VFM calculation over the 20 triples gives i between 3.512° and 18.696° and Ω between −41.197° and 87.811°, while Bošković's selected triple [136] gives i=6.807° and Ω=74.048°. The selection criteria in §4.2 (C1≈C6, C3 at minimum latitude) are precisely a search for a triple that approximately satisfies the same-parallel condition, so the method is not a general three-position inversion without additional assumptions.
- [§4.3, Tables 1, 2, and 4] The claimed precision of 10⁻⁵ arcseconds is not supported by the input data. Table 1 lists ecliptic longitudes and latitudes to arcminutes, and the historical measurement process is far coarser; nevertheless Table 4 reports i = 6.80728° = 6°48′26.20337″ and Ω = 74.04774° = 74°02′51.87646″. A 10⁻⁵ arcsecond digit is about nine orders of magnitude finer than the arcminute input. The statement in §4.3 that the results were “calculated using high precision and closed equations without loosing precision” conflates arithmetic precision with measurement precision. The paper should propagate the input uncertainties and quote i, Ω, and T′ to a justified number of significant digits; the agreement with Bošković's values is then at the arcminute level, which is sufficient for the historical claim but not for the advertised precision.
- [§4.1, Table 4] The VFM method used as a “contemporary method” check is unpublished (“paper is in preparation”). For the triple [136] it returns values identical to the spherical-trigonometric solution to eight significant figures (i=6.80727871°, Ω=74.0477436°, T′=26.8062322 d). If VFM is mathematically equivalent to the present derivation, this agreement is by construction and provides no independent confirmation; if it is not equivalent, the reader needs the VFM definitions to evaluate the comparison. The independent validation should rest on the comparison with Bošković's published values and with Roša et al. (2021); the VFM table should be moved to supplementary material or explicitly reported as an internal consistency check.
- [§2.2–2.4, Eqs. (7), (30), (42), (55)] The equations use inverse trigonometric functions (arccos, arctan) without specifying branch selection. For example, Eq. (30) determines B′′−D from its cosine, which has two possible signs; Eqs. (42) and (55) have the same issue, and Eq. (7) assumes that P lies inside the angle CC′C′′ so that the two angles sum rather than subtract. A reader applying the method to arbitrary triples—as the Conclusion invites for exoplanets and stars—cannot reproduce the results without additional quadrant rules or case distinctions. The claim of a single closed-form procedure is therefore incomplete as stated.
minor comments (4)
- [Table 2] The row for Eq. (6) appears twice with different values: first cos P C′′C′ = 0.107516 with angle 83.82785°, and later cos P C′′C′ = 0.108113 with angle 83.79345°. The second value is the one used in Eq. (28); the first appears to be a typographical error and should be harmonized.
- [§4.2] The numerical values quoted for C1 and C6 (20°27′ and 22°45′) do not match the entries in Table 1 as the table is currently formatted (20°37′ and 22°45′, if read literally); please reconcile the text with the table.
- [Eq. (37)] The quantity A = 365.25 days should be defined explicitly as the Earth's orbital (or tropical/sidereal) period used in converting between synodic and sidereal rotation periods; the present text introduces it only in a footnote-like parenthetical.
- [Figure 5] The scanned historical page in Figure 5 is difficult to read in the arXiv version; a transcription of the relevant numbered paragraphs (№76–№81) would make the mapping from Bošković's text to the equations easier to verify.
Circularity Check
No significant circularity: the spherical-trigonometric reconstruction is self-contained and the historical validation is an external benchmark, not an input-output identity.
full rationale
The derivation chain is self-contained: equations (1)-(57) are spherical-trigonometric identities applied to the three observed ecliptic positions and mean solar times, and i, Omega, and T' are not used as inputs. No free parameter is fitted to Bošković's published values; the reproduction of his 1785 results is a consistency check of the reconstruction against an external historical benchmark, and the agreement with Carrington/Spörer values is an independent external comparison. The same-parallel/mirror-symmetry step (P'C = P'C' = P'C'', Sections 2.3 and 2.4.1) is a physical modeling premise of the rigid-rotation method, not a consequence of the outputs, so it creates a correctness/robustness limitation (visible in the Table 6 scatter) but not circularity. The unpublished VFM method is a same-author corroboration and is not load-bearing: the central reconstruction does not depend on it. The '10^-5 arcsec precision' statement refers to numerical internal consistency of the closed-form computation, not to validation accuracy, and is a precision-claim issue rather than a circular step. Therefore no significant circularity is found.
Assumptions & free parameters
assumptions (4)
- domain assumption The three sunspot positions lie on a single heliographic parallel (same heliographic latitude), i.e., P'C = P'C' = P'C''.
- domain assumption The Sun's rotation is uniform between observations, so the angular travel around the solar axis is proportional to elapsed time (Equation 35).
- domain assumption The three positions are of the same physical spot and the ecliptic coordinates from Bošković's Tables are accurate to about an arcminute.
- standard math Standard spherical trigonometry (cosine rule, sine rule, cotangent rule) applies to the eight constructed triangles.
Cite this review
Pith. "Pith review of Bo\v{s}kovi\'{c}'s Spherical Trigonometric Solution for Determining the Axis and Rate of Solar Rotation by Observing Sunspots in 1777." pith.science (2026). https://pith.science/paper/ORDDSAG5
@misc{pith2026250720383,
author = {Pith},
title = {Pith review of: Bo\vskovi\'c's Spherical Trigonometric Solution for Determining the Axis and Rate of Solar Rotation by Observing Sunspots in 1777},
year = {2026},
howpublished = {\url{https://pith.science/paper/ORDDSAG5}},
note = {Machine review of arXiv:2507.20383}
}
abstract
In September 1777 Ru\dj er Bo\v{s}kovi\'{c} observed and measured the sun-spot positions to determine the solar rotation elements. In 1785, among other methods, he described a trigonometric spherical solution for the determination of the position of the axis and rate of the solar rotation using three sunspot positions, but without equations. For the first time, we derive the equations that are applicable to modern computers for calculating the solar rotation elements, as they were described by Bo\v{s}kovi\'{c}. We recalculated Bo\v{s}kovi\'{c}'s original example using his measurements of sunspot positions from 1777 and the equations developed here, confirming his results from 1785. Bo\v{s}kovi\'{c}'s methodology of arithmetic means determines $i$, $\Omega$, and sidereal period $T'$ separately, while the planar trigonometric solution determines $i$ and $\Omega$ together. His spherical trigonometric solution calculates $i$, $\Omega$, and the sidereal period $T'$ in a single procedure. Keywords: Ru\dj er Bo\v{s}kovi\'{c}, Sunspots, Solar rotation, Spherical trigonometry
Figures
Figures from the paper (2 more)
Reference graph
Works this paper leans on
-
[1]
barticle Arlt , R. , Fr \"o hlich , H.-E. : 2012 , The solar differential rotation in the 18th century . 543 , A7 . https://doi.org/10.1051/0004-6361/201219266 . 2012A&A...543A...7A . barticle
-
[2]
barticle Arlt , R. , Vaquero , J.M. : 2020 , Historical sunspot records . Living Reviews in Solar Physics 17 , 1 . https://doi.org/10.1007/s41116-020-0023-y . 2020LRSP...17....1A . barticle
-
[3]
: 1736 , De maculis solaribus exercitatio astronomica habita
bbook Boscovich , R.J. : 1736 , De maculis solaribus exercitatio astronomica habita. 1736dmse.book.....B . bbook
-
[5]
bchapter Boscovich , R.J. : 1785 b, Opuscule II . Sur les éléments de la rotation du soleil sur son axe déterminés par l'observation de ses taches , Venetiis apud Remondini , Bassano , p. 75 . https://doi.org/10.3931/e-rara-1010 . https://www.e-rara.ch/zut/content/zoom/278121 . bchapter
-
[6]
: 1617 , Logarithmorum Chilias Prima , London
bbook Briggs , H. : 1617 , Logarithmorum Chilias Prima , London . 1617lcp..book.....B . https://archive.org/details/bim_early-english-books-1475-1640_logarithmorum-chilias-pr_briggs-henry_1617/page/n13/mode/2up . bbook
-
[7]
bbook Briggs , H. : 1624 , Arithmetica logarithmica sive logarithmorum chiliades triginta, pro numeris naturali serie crescentibus ab unitate ad 20000 : et a 90000 ad 100000, quorum ope multa perficiuntur arithmetica problemata et geometrica hos numeros primus invenit clarissimus vir Iohannes Neperus Baro Merchistonij, eos autem ex eiusdem sententia mutav...
-
[8]
, : 1633 , Trigonometria britannica , Il Giardino di Archimede
bbook Briggs , H. , : 1633 , Trigonometria britannica , Il Giardino di Archimede . 1633trbr.book.....B . bbook
-
[9]
u rgi , J. : 1620 , Aritmetische vnd Geometrische Progress Tabulen, sambt gr \
bbook B \"u rgi , J. : 1620 , Aritmetische vnd Geometrische Progress Tabulen, sambt gr \"u ndlichem vnterricht, wie solche n \"u tzlich in allerley Rechnungen zugebrauchen, vnd verstanden werden sol , Sessen . 1620agp..book.....B . https://books.google.hr/books?hl=hr&lr=&id=sZ9vd3zclXUC&oi=fnd&pg=PA2&dq=Aritmetische+vnd+Geometrische+Progress+Tabulen:+samb...
Show all 26 references
-
[10]
: 1863 , Observations of the spots on the Sun: from November 9, 1853, to March 24, 1861, made at Redhill
bbook Carrington , R.C. : 1863 , Observations of the spots on the Sun: from November 9, 1853, to March 24, 1861, made at Redhill . 1863oss..book.....C . bbook
-
[11]
, Vaquero , J.M
barticle Casas , R. , Vaquero , J.M. , Vazquez , M. : 2006 , Solar Rotation in the 17th century . 234 , 379 . https://doi.org/10.1007/s11207-006-0036-2 . 2006SoPh..234..379C . barticle
2006 doi
-
[12]
: 1976 , The Maunder Minimum
barticle Eddy , J.A. : 1976 , The Maunder Minimum . Science 192 , 1189 . https://doi.org/10.1126/science.192.4245.1189 . 1976Sci...192.1189E . barticle
1976
-
[13]
, Gilman , P.A
barticle Eddy , J.A. , Gilman , P.A. , Trotter , D.E. : 1977 , Anomalous Solar Rotation in the Early 17th Century . Science 198 , 824 . https://doi.org/10.1126/science.198.4319.824 . 1977Sci...198..824E . barticle
1977 doi
-
[14]
, Welser , M
bbook Galilei , G. , Welser , M. , de Filiis , A. : 1613 , Istoria E dimostrazioni intorno alle macchie solari E loro accidenti comprese in tre lettere scritte all'illvstrissimo signor Marco Velseri ... 1613idim.book.....G . bbook
-
[15]
, Braj s a , R
barticle Husak , M. , Braj s a , R. , S poljari \'c , D. : 2021 a, On the Determination of the Solar Rotation Elements i , and Period using Sunspot Observations by Ru er Bo s kovi \'c in 1777 . The Mining-Geological-Petroleum Engineering Bulletin / Rudarsko-geološko-naftni zbo...
2021 arXiv
-
[16]
, Braj s a , R
barticle Husak , M. , Braj s a , R. , S poljari \'c , D. : 2021 b, Solar rotation elements i , and period determined using sunspot observations by Ru er Bo s kovi \'c in 1777 . Central European Astrophysical Bulletin 45 , 1 . https://doi.org/10.48550/arXiv.2203.11745 . 2022arX...
-
[17]
, Braj s a , R
barticle Husak , M. , Braj s a , R. , S poljari \'c , D. , Krajnovi \'c , D. , Ru z djak , D. , Skoki \'c , I. , Ro s a , D. , Hr z ina , D. : 2023 , Bo s kovi \'c 's Method for Determining the Axis and Rate of Solar Rotation by Observing Sunspots in 1777 . 298 , 122 . https:/...
2023 doi
-
[18]
: 1614 , Mirifici Logarithmorum Canonis Descriptio , E
bbook Napier , J. : 1614 , Mirifici Logarithmorum Canonis Descriptio , E. Rabanum . 1614mlcd.book.....N . bbook
-
[19]
, Hr z ina , D
bchapter Ro s a , D. , Hr z ina , D. , Husak , M. , Braj s a , R. , S poljari \'c , D. , Skoki \'c , I. , Mari c i \'c , D. , S terc , F. , Rom s tajn , I. : 2021 , Determination of the Solar Rotation Elements and Period from Ru er Bo s kovi \'c 's Sunspot Observations in 1777...
2021 doi
-
[20]
: 1630 , Rosa Ursina
bbook Scheiner , C. : 1630 , Rosa Ursina . https://doi.org/10.3931/e-rara-556 . 1630rour.book.....S . bbook
-
[21]
: 1874 , Beobachtungen der Sonnenflecken zu Anclam
bbook Sp \"o rer , G.F.W. : 1874 , Beobachtungen der Sonnenflecken zu Anclam . 1874bsza.book.....S . bbook
-
[22]
: 2002 , Rotation , Springer Berlin Heidelberg , Berlin, Heidelberg , 277
bbook Stix , M. : 2002 , Rotation , Springer Berlin Heidelberg , Berlin, Heidelberg , 277 . 978-3-642-56042-2 . https://doi.org/10.1007/978-3-642-56042-2_7 . 2002tsai.book.....S . https://doi.org/10.1007/978-3-642-56042-2_7 . bbook
2002 doi
-
[23]
, Braj s a , R
barticle Sudar , D. , Braj s a , R. : 2022 , Solar Rotation in the Period 1611 - 1631 Determined Using Observations of Christoph Scheiner . 297 , 132 . https://doi.org/10.1007/s11207-022-02068-4 . 2022SoPh..297..132S . barticle
2022 doi
-
[24]
: 2006 , Coordinate systems for solar image data
barticle Thompson , W.T. : 2006 , Coordinate systems for solar image data . 449 , 791 . https://doi.org/10.1051/0004-6361:20054262 . 2006A&A...449..791T . barticle
2006 doi
-
[25]
: 1955 , Ergebnisse und Probleme der Sonnenforschung
bbook Waldmeier , M. : 1955 , Ergebnisse und Probleme der Sonnenforschung. 1955epds.book.....W . bbook
1955
-
[26]
: 1978 , On the Solar Rotation Elements i and as Determined by Doppler Velocity Measurements of the Solar Plasma
barticle W \"o hl , H. : 1978 , On the Solar Rotation Elements i and as Determined by Doppler Velocity Measurements of the Solar Plasma. 62 , 165 . 1978A&A....62..165W . barticle
1978
-
[27]
, Hohenkerk , C
barticle Yallop , B.D. , Hohenkerk , C. , Murdin , L. , Clark , D.H. : 1982 , Solar Rotation from 17th-Century Records . 23 , 213 . 1982QJRAS..23..213Y . barticle
1982
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.