Reconstructing the Antikythera Mechanism s Central Front dial parts Division and Placement of the Zodiac Dial ring

Andreas Vossinakis; Aristeidis Voulgaris; Christophoros Mouratidis

REVIEW 4 major objections 6 minor 19 references

Reconstructing the Antikythera Mechanism s Central Front dial parts Division and Placement of the Zodiac Dial ring

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that the Antikythera Mechanism's zodiac dial ring was a 365-day time scale, not a 360-degree angular ring.

desk verdict A serious, physically grounded reconstruction whose central 365-day claim is underdetermined by the preserved evidence and oversold in the abstract. read the letter →

arxiv 2505.08484 v1 pith:4NR2QJ7F submitted 2025-05-13 physics.hist-ph astro-ph.IM

Aristeidis Voulgaris , Christophoros Mouratidis , Andreas Vossinakis This is my paper · ORCID

classification physics.hist-phastro-ph.IM

keywords AntikytheraMechanismzodiacdialringEgyptiansolaranomaly365-dayyearGeminusseasonlengthsFragmentCfiducialline

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that the zodiac dial ring on the Antikythera Mechanism's front face was not a 360-degree angular scale but a 365-day calendar ring. On that reading, each of the twelve zodiac month sectors contains an unequal number of day-subdivisions, from 29 to 32, so the engraved ring itself encodes the solar anomaly, the Sun's varying apparent speed, and the unequal durations of the seasons. The authors support the claim by reconstructing the four parts of the central front dial from same-scale front/back images and CT scans of Fragment C, then matching their reconstruction to the preserved 30-subdivision months of Libra and Scorpio. The result matters because it makes the front dial a coherent time-measuring instrument, calibrated to an absolute starting date of 23 December 178 BC, without invoking extra gears or complex spacing assumptions.

What carries the argument

The load-bearing object is the 365-subdivision zodiac dial ring: 365 equally spaced marks around the circle, assigned to twelve zodiac months whose boundaries fall between marks so that month lengths are unequal. The assignment of lengths is carried by a rounding table built from Geminus' season durations, with the constraints that Sagittarius, the month containing perihelion, is shortest and Gemini, the month containing aphelion, is longest. The calibration geometry is carried by the fiducial line and the vertical line VLb, which put the 1st of Capricorn opposite the 18th of Hathyr and set the epoch. This machinery lets a single uniformly engraved ring do double duty: days for the Sun and, at about 1.8 hours per subdivision, a timing scale for lunar motion.

What would settle it

A direct observation that would settle the claim is to locate and measure any preserved subdivision of the missing zodiac months, Sagittarius/Capricorn or Taurus/Gemini/Cancer/Virgo, in the CT volume of Fragment C. If a complete month is found to contain exactly 30 equally spaced marks, or if the preserved Libra and Scorpio arcs turn out to be exact $30^\circ$ angular sectors with no adjacent month deviating from 30, the 365-subdivision claim fails; if Gemini and Cancer show 31 or 32 equal marks and Sagittarius and Capricorn show 28 or 29, it stands.

Watch

Extended reading notes

Core claim

The paper's central claim is that the zodiac dial ring was divided into 365 equal subdivisions representing days of a rounded solar tropical year of $365.25$ days, not into 360 degrees; hence each zodiac month occupies an unequal number of subdivisions, from 29 for Capricorn and Sagittarius to 32 for Gemini and Cancer under the preferred rounding. The unequal arcs are not an error or a crude approximation but the deliberate representation of the solar anomaly and of the season durations reported by the Hellenistic astronomer Geminus: autumn $88\frac{1}{8}$, winter $90\frac{1}{8}$, spring $94\frac{1}{2}$, and summer $92\frac{1}{2}$ days. The paper also claims the ring was calibrated by making the first subdivision of Capricorn point at the fiducial line, which places the initial date at 23 December 178 BC, with the 18th of Hathyr on the Egyptian ring aligned to the same line. In this design the front dial is a time-measuring device: the solar pointer crosses day-subdivisions, the parapegma star-event letters fall on days, and the Egyptian ring shifts by one subdivision every four solar years to track the quarter-day excess.

Load-bearing premise

The load-bearing premise is that the season lengths the builder encoded are the ones attributed to Geminus (autumn $88\frac{1}{8}$, winter $90\frac{1}{8}$, spring $94\frac{1}{2}$, and summer $92\frac{1}{2}$ days); the paper itself notes in Section 2.4.1 that it is unclear whether these were Geminus' own values or borrowed from earlier astronomers, so if a different ancient season table was used, the 29-to-32 day month pattern and the calibration dates would shift.

Editorial extensions

If this is right

The two fully preserved zodiac months, Libra and Scorpio, each with 30 subdivisions, cease to be evidence for a 360-degree ring: five of the six best roundings in Table 3 keep those months at 30 day-subdivisions, so the surviving fragments are consistent with a 365-day ring.
Because the parapegma letter-events are date events, putting them on a day-subdivision ring makes the star calendar and the zodiac ring share the same unit, so a pointer reading on the zodiac ring corresponds directly to a parapegma date.
A user can convert any date between the Egyptian and zodiac calendars by reading the two concentric rings together; after four full solar turns, the Egyptian ring is shifted one subdivision, mechanically reproducing the $365.25$-day tropical-year excess over the 365-day Egyptian year.
With the preferred rounding and the 23 December 178 BC calibration, the paper generates absolute date ranges for all twelve zodiac months, placing perihelion in Sagittarius around 29 November and aphelion in Gemini around 30 May for that epoch.
The front dial can be reconstructed as a time-measuring device with fewer added components than a 360-degree unequal-spacing scheme, because the solar anomaly is carried by the unequal month lengths rather than by irregularly spaced degree marks.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A testable extension: if the 365-day division is correct, future imaging of the missing Gemini/Cancer and Sagittarius/Capricorn sectors should reveal 31 or 32 subdivisions for the first pair and 28 or 29 for the second, rather than 30 equal marks.
The same criterion could be applied to other ancient dials: whenever a graduated ring sits next to a calendar scale and is marked with event letters that refer to dates, counting its subdivisions as days rather than degrees is the hypothesis with the fewest extra assumptions.
The one-subdivision counter-rotation of the Egyptian ring every four years is the same correction that the Julian calendar codified as a leap day; if this reconstruction is right, the Antikythera Mechanism is an early mechanical embodiment of that correction, though the paper does not draw the comparison.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

The thing to know is that this is an honest, physically grounded reconstruction that oversells its central result in the abstract. The authors build a bronze replica of the front dial, give detailed tables of possible daily subdivisions per zodiac month, and explicitly admit in Section 8.3 that the decisive parts of the ring are lost. That candor deserves credit. What they have not established is that the ring was divided into 365 days rather than 360 degrees. The preserved evidence—two months with 30 subdivisions each—fits both schemes. Their own Table 3 shows five of six rounding scenarios produce 30 for Libra and Scorpio, so that's consistency, not positive evidence.

The day-count tables are the real new contribution, along with the Eudoxus-based scenarios and the physical bronze reconstruction. But those tables are fitted, not derived. The Geminus season durations are selected because the authors assert a correlation between Geminus and the Mechanism, while admitting it's unclear whether Geminus borrowed from earlier astronomers. Using Meton or Euctemon instead would shift the month lengths. The 'uniform increment' assumption from Sagittarius to Gemini is a modeling choice with no surviving support. And the calibration date of 23 December 178 BC is imported from their earlier work, so the new claims inherit that dependence.

Soft spots in proportion: the abstract says the solar anomaly and unequal seasons 'were well represented' as if established; the paper's own caveat in Section 8.3 is more careful. Some reconstruction details—the locking pin, the square pin-drivers, the sliding catches—are presented as definite when they are hypotheses. The appendix's construction protocol and the airplane anecdote are tangential to the main argument.

Who should read this: anyone working on the Antikythera Mechanism or on ancient instrument reconstruction. It is a serious attempt that advances the physical understanding of the front plate, even if the 365-day division remains one plausible hypothesis alongside 360. I would send it to peer review, with the expectation that the authors qualify the claim and quantify measurement uncertainty. A good referee can push them toward presenting the 365-day scheme as a testable possibility, not a conclusion.

Referee Report

4 major / 6 minor

Summary. The paper proposes a reconstruction of the Antikythera Mechanism's central front dial, arguing that the zodiac dial ring was divided into 365 equal day-subdivisions rather than the traditional 360 degrees. The authors present a bronze reconstruction based on photographs and CT scans of Fragment C, and they use ancient season durations (particularly those of Geminus) to assign unequal day counts (29--32) to the twelve zodiac months. They further claim that the ring was calibrated to 23 December 178 BC via the fiducial line and the first subdivision of Capricorn. The paper explicitly acknowledges that only about 20% of the zodiac ring is preserved and that the decisive sectors are lost (Section 8.3), but the abstract and conclusions nevertheless present the 365-day division as an established result.

Significance. If the 365-day division were established, the paper would contribute a concrete physical reconstruction and a novel reading of the zodiac dial as a time scale rather than an angular scale. The authors deserve credit for the detailed photographic and CT-based documentation of Fragment C, for building working bronze reconstructions, and for transparently reporting the missing sectors that limit inference. However, the central evidential claim is not currently supported: the preserved 30-subdivision months of Libra and Scorpius are compatible with both the 360-degree and 365-day schemes, and the paper's own Section 8.3 concedes that the critical sectors (Sagittarius/Capricorn and Taurus/Gemini/Cancer/Virgo) are lost. The additional assumptions about Geminus' season durations, the rounding rule, and the calibration date are not independently justified, so the specific month lengths and the 365-day ring remain a plausible hypothesis rather than a demonstrated reconstruction.

major comments (4)

[Abstract and Section 8.3] The preserved evidence does not distinguish between a 360-part and a 365-part zodiac ring. The authors concede in Section 8.3 that the critical sectors (Sagittarius/Capricorn and Taurus/Gemini/Cancer/Virgo) are lost, and the two fully preserved months, Libra (ΧΗΛΑΙ) and Scorpius (ΣΚΟΡΠΙΟΣ), each have 30 subdivisions under both schemes. The statement that "five of the six 'Best rounding' options also presented Libra and Scorpio with 30 subdivisions" shows only consistency, not positive evidence for 365. The abstract's assertion that the ring "was correlated with 365 equal subdivisions-days" therefore outruns the evidence; at most the paper can claim that a 365-day division is compatible with the preserved fragments.
[Section 2.4.1 and Table 3] The derivation of specific month lengths depends entirely on adopting Geminus' season durations (Autumn 88⅛, Winter 90⅛, Spring 94½, Summer 92½). The authors justify this choice by invoking "a strong correlation between Geminus' astronomy and the Antikythera Mechanism," but they also admit it is unclear whether Geminus' numbers came from his own observations or from earlier astronomers. If the values of Meton, Euctemon, Callippus, or Eudoxus were used instead, the proposed month lengths change, and even the necessity of a 365-day ring is affected. No independent argument is given that the Mechanism's builder specifically encoded Geminus' fractional season lengths, so the 29--32 day month assignments are not established by the evidence presented.
[Section 3 and Table 4] The calibration date of 23 December 178 BC is imported from the authors' earlier work (Voulgaris et al., 2023a) and then used to assign concrete dates to each zodiac month in Table 4. The present paper does not re-derive or test this date, and the fiducial-line argument shown in Figure 12 depends on assumptions about the original ring positions that are not independently verified. If the earlier dating is insecure, the month-date assignments in Table 4 inherit that uncertainty, and the claim that the design "directly connects" the Mechanism to that specific date is stronger than the evidence supplied here.
[Section 2.2 and Section 8.3] The functional preference for a day-scale (parapegma events, Egyptian-ring correlation) does not logically require physically engraving 365 day-ticks on the zodiac ring. One could mark the same month boundaries on a 360-degree ring and still interpret the subdivisions as time units, as the authors themselves note the possibility of unequal degree divisions. The paper conflates the unit of measurement with the number of physical subdivisions. To establish the 365-day division, positive evidence of 365 ticks (or of months with 31 or 32 subdivisions) is needed, not merely agreement of the preserved 30-subdivision months with a reconstructed 365-day scheme.

minor comments (6)

[Abstract] The abstract contains typos and formatting errors, including "Mechanism s Central" and "scacs," which should be corrected before publication.
[Table 3] In the 'Best rounding 5' row, the season-duration line appears to omit the Autumn value (showing only "90 94 93"); please check and complete the entry.
[Table 1] The zodiac dial row lists "360 degrees or 365 days" as alternatives, but the paper later considers a 360-degree ring with unequal subdivisions as a third possibility; the table should be revised to reflect the full range of proposals.
[Figures 8 and 9] The statement that the line of solstices does not cross the center because of the unequal season lengths is potentially confusing, since the line is a chord between two points on the ring; please add a sentence explaining the geometry.
[Section 8.1] The claim that "the solar anomaly does not have to be represented" on a 360-degree division is inaccurate, because Evans et al. (2010) proposed 360 unequal period subdivisions precisely to represent the anomaly; the contrast should be stated more carefully.
[References] Some online sources include access dates while others do not; please standardize the reference format for URLs.

Circularity Check

3 steps flagged · score 6.0 of 10

The zodiac ring's 365-day division is a construction whose stated confirmations reduce to the preserved 30-subdivision months and the authors' own calibration date.

fitted input called prediction [Section 2.4.1, Table 3 and caption; Section 8.3]
"Only about 20% of the zodiac dial ring is preserved, so we do not know the exact durations of the four seasons, nor the day distribution per zodiac month, except for the months ΧΗΛΑΙ and ΣΚΟΡΠΙΟΣ, which have 30 subdivisions/days (Bitsakis and Jones, 2016a). ... Note that five out of six options of the best rounding present Libra and Scorpio with 30 subdivisions-days, as are also preserved for Libra and Scorpio in Fragment C."

Table 3 begins its day-distribution row with Libra and Scorpio set to 30, the only two fully preserved months, before any rounding is applied. The caption then cites the agreement of the rounded rows with those same preserved 30 subdivisions as support for the 365-day scheme. That agreement is an artifact of the construction, not an independent check. Moreover, the traditional 360° division also predicts 30 subdivisions in Libra and Scorpio, so the match has no discriminative power between 365 and 360; Section 8.3 concedes that the decisive sectors are lost. The claimed confirmation is thus built from the very data it is supposed to validate.
self citation load bearing [Section 2.4.1 (Table 4), Section 3, Note 8]
"By applying the 'Best rounding 1' for the initial calibration date of the Antikythera Mechanism (23 December 178 BC—Winter Solstice—1st Day of Capricorn; Voulgaris et al., 2023a; 2023c), the dates for each zodiac month can be deduced (Table 4)."

The epoch 23 December 178 BC is imported from the authors' own earlier work (Voulgaris et al., 2023a, 2023c) and is not re-established in this paper. Note 8 itself acknowledges that other researchers arrived at different starting dates. The date is then used as an unexamined premise to generate Table 4 and the whole calibration narrative, so that portion of the argument rests on a self-citation rather than on evidence presented here. This is not the basis of the 365-vs-360 division, but it is load-bearing for the claimed placement and month dates.

1 more flagged steps

self definitional [Abstract and Section 2.4.1]
"Since there is a strong correlation between Geminus' astronomy and the Antikythera Mechanism ... the season duration mentioned by Geminus ... was selected for the analysis in Table 3. ... The different number of days per zodiac month created 12 unequal central angles on the zodiac dial ring and therefore the solar anomaly and the unequal time span of the astronomical seasons were well represented on the Antikythera Mechanism."

The Geminus season lengths (Autumn 88⅛, Winter 90⅛, Spring 94½, Summer 92½) already contain the solar anomaly and the unequal seasons; the reconstruction engraves those values as month subdivisions. The abstract's 'therefore ... well represented' restates the chosen input rather than reporting a measurement from the ring, which is only about 20% preserved. The resulting 29–32 day month lengths are a translation of Geminus into dial markings, so the claimed representation of the solar anomaly holds by construction, not by independent confirmation.

full rationale

The paper is a coherent and honest reconstruction: it states that only ~20% of the zodiac ring survives, openly lists the lost critical sectors in Section 8.3, and uses functional arguments (parapegma event dates, the 365-hole bearing ring, and the Egyptian-ring analogy) to motivate a day-unit zodiac scale. Those parts are not circular. However, three load-bearing steps reduce the paper's stated validations to their own inputs. First, the preserved 30-subdivision counts for Libra and Scorpio are used as the starting allocation in Table 3 and then cited as evidence that the 365-day scheme is consistent with the preserved data; since the 360° scheme makes the same prediction, this agreement is both built in and non-discriminating. Second, the 178 BC calibration date is imported from the authors' own prior papers and used to produce the month-date table, with the paper itself noting that other researchers obtain different epochs. Third, the Geminus season durations are selected as input and then the solar-anomaly representation is announced as a conclusion; that conclusion is contained in the chosen astronomical parameters. The remaining underdetermination between 365 and 360 subdivisions is an evidence/correctness issue rather than circularity, and I have not scored it as such by itself. Because the central 365-day hypothesis retains independent functional content but several of the paper's confirmations reduce by construction or by self-citation, the appropriate score is 6: partial circularity.

Assumptions & free parameters 4 free parameters · 6 assumptions · 4 invented entities

The paper's reconstruction rests on several adopted values and assumptions, many from the authors' own prior publications, and the specific day counts are fitted to the preserved 30-subdivision data points under flexibility.

free parameters (4)

Days per zodiac month (Best Rounding 1) = Cap 29, Aqu 30, Psc 31, Ari 31, Tau 31, Gem 32, Cnc 32, Leo 30, Vir 30, Lib 30, Sco 30, Sgr 29
Selected from six equally valid roundings in Table 3 to match the preserved 30 subdivisions of Libra and Scorpio and the Geminus season totals; this is a post hoc choice.
Season durations = Autumn 88, Winter 90, Spring 94, Summer 92 (rounded from Geminus 88⅛, 90⅛, 94½, 92½)
Adopted from Geminus rather than Meton, Euctemon, Callippus, Eudoxus, or Cleomedes; the paper justifies this by an asserted correlation between Geminus and the Mechanism. Not derived from the artifact.
Rounding rule for fractional season days = round to nearest or current day
The paper permits '94½ can be rounded to 94 or to 95', adding flexibility that makes the six best roundings fit.
Initial calibration date = 23 December 178 BC (1st day of Capricorn, 18th of Hathyr)
Taken from the authors' earlier work (Voulgaris et al., 2023a); shifts all derived month dates in Table 4, and is not independently established in this paper.

assumptions (6)

domain assumption The zodiac dial ring represents the Ecliptic and one full pointer turn equals one solar tropical year of 365.25 days.
Sections 2.2 and Appendix; standard ancient astronomy, but the specific 365.25 value is an input.
ad hoc to paper The Egyptian dial ring is a 365-day time scale and the zodiac ring shares equal design and common center, so the zodiac ring is also expected to be a 365-day time scale.
Section 2.3; analogy is used to support the 365-day hypothesis, but functional symmetry does not force the same unit system.
ad hoc to paper Geminus' season durations (Autumn 88⅛, Winter 90⅛, Spring 94½, Summer 92½) are the correct ancient values for the Mechanism's zodiac month lengths.
Section 2.4.1; the paper itself admits it is not clear if Geminus used his own observations or earlier astronomers, and other ancient sources give different values.
domain assumption The index letters engraved on zodiac subdivisions correspond to parapegma events dated in days, so the subdivisions must be days rather than degrees.
Section 2.2 and 8.3; the inference depends on the interpretation that parapegma letters on the ring are day markers, not degree markers.
domain assumption The preserved 30 subdivisions each for Libra and Scorpio are accurately counted and representative.
Section 2.4.1, Table 3; the count is taken from Bitsakis and Jones (2016a), and no measurement uncertainty is given despite the fragment's deformation.
domain assumption The initial calibration date of 23 December 178 BC derived in the authors' prior work is correct.
Section 3, Table 4; the date is loaded from Voulgaris et al., 2023a and used to assign zodiac month dates.

invented entities (4)

Small perpendicular pin with a head for locking the Egyptian dial ring
purpose: Prevents accidental rotation of the Egyptian ring; inserted into one of the 365 holes during use.
Section 3: 'Rotation of the Egyptian dial ring by mistake was probably avoided by using a small perpendicular pin with a head, which perforated the ring and was inserted in one of the 365 holes'. No such pin is preserved.
Square pin-drivers on the internal wooden casement
purpose: Precise positioning of the square front plate via square locking holes.
Section 2.1: 'Originally square pin-drivers in these square holes were attached to the internal wooden casement'. Only one square hole is preserved; the drivers are hypothetical.
Four sliding catches on the square front plate
purpose: Detach the front plate for access to internal gears.
Section 2.1, Figure 3(A); one preserved sliding catch on Fragment C is used to infer four catches symmetrically. The authors themselves use symmetry to reconstruct.
Lost cover of b1 gear with lunar age scale
purpose: Provides a 29.5-day lunar phase scale for the lunar pointer.
Appendix 8.1 and Figure 7 caption: 'The lost cover of the b1-gear is also presented, in which six circles are engraved...' The existence of the cover and its scale is hypothetical.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Reconstructing the Antikythera Mechanism s Central Front dial parts Division and Placement of the Zodiac Dial ring." pith.science (2026). https://pith.science/paper/4NR2QJ7F

@misc{pith2026250508484,
  author       = {Pith},
  title        = {Pith review of: Reconstructing the Antikythera Mechanism s Central Front dial parts Division and Placement of the Zodiac Dial ring},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4NR2QJ7F}},
  note         = {Machine review of arXiv:2505.08484}
}

read the original abstract

In this paper we analyze, discuss and present the design of the Antikythera Mechanism s central front parts. Based on the aligned and of same scale visual images of Fragment C front/back face and the X-ray CT scacs, we designed and reconstructed in bronze, the four independent parts comprising the central front dial. We then correlated the zodiac dial ring with 365 equal subdivisions-days and we investigated the number of days per astronomical season and per zodiac month. Then, we adopted a specific number of equal subdivisions/days per each zodiac month and we engraved these on the bronze zodiac month ring. The different number of days per zodiac month created 12 unequal epicenter angles on the zodiac dial ring and therefore the solar anomaly and the unequal time span of the astronomical seasons were well represented on the Antikythera Mechanism. In this way, the functionality of the central front dial of the Mechanism was achieved by adopting the minimum number of hypotheses.

Figures

Figures reproduced from arXiv: 2505.08484 by the authors.

**Figure 3.** The four parts of the central front dial plate, reconstructed of bronze (94%Cu + 6%Sn). (A) The square front plate of the Mechanism. The ancient craftsman adapted four sliding catches on the small oblong holes, so that he could easily detach the square front plate to have a direct look at the moving parts inside (i.e. the gears and axles). (B) The bearing base ring (fourth part/third ring). The outer perimeter/arc l… view at source ↗

**Figure 14.** A close-up of the solar ray and the zodiac and Egyptian ring subdivisions on a reconstructed copy of the Antikythera Mechanism. The ΗΛΙΟΥ ΑΚΤΙΝ points to the 27th day of Cancer and at the same time it points to the 20 th day of Thoth (photograph: the authors) [PITH_FULL_IMAGE:figures/full_fig_p020_14.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 19 canonical work pages

[1]

Voulgaris et al. (2022) make a case for the existence of the missing Draconic cycle that should exist on the Antikythera Mech - anism on the grounds of the Draconic gear- ing (as an engagement of gears b1 –a1, r1/Fragment D and one hypothetical small gear s1; and revised in gears s1, s2, t1 in Voulgaris et al., 2024), which represent the fourth lunar cycl...

work page 2022
[2]

The hours of eclipse events can be easily calculated by using only the Antiky thera Mechanism and without any external infor- mation (unrelated to the Mechanism): by measuring the number of the zodiac ring subdivisions from New Moon/Full Moon to the next New/Full Moon, and convert - ing the m into days/hours (Voulgaris et al., 2023b: 22–25)

work page
[3]

E.g. on Fragment C the lunar cylinder is stuck underneath the Egyptian/zodiac dial rings, parapegma plate 1 is stuck in front of the lunar cylinder, and one piece of parapegma plate 2 is stuck to the reverse face and in the opposite direction on para- pegma plate 1. See also Figure 6 for a part of the front dial plate that is trapped inside the petrified ...

work page
[4]

AMRP radiography is Advanced Medical Radiography Practitioner radiography, and involves a senor radiographer with an ad- vanced level of knowledge and skill

work page
[5]

Small parts of the parapegma plates with inscriptions are also preserved on Frag- ments 9, 20, 22, 28 ( Bitsakis and Jones, 2016a)

work page
[6]

On the back cover inscription, the text ΠΡΟΕΧΟΝ ΑΥΤΟΥ ΓΝΩΜΟΝΙΟΝ Σ[ΕΛΗΝΗΣ], and a (lunar) pointer project- ing from it, are preserved (see Bitsakis and Jones, 2016b: 232)

work page
[7]

The golden sphere –Sun should be linked to the b1 gear or the lost cover of the b1 gear (see Voulgaris et al., 2019b: Figure 1)

ΗΛΙ[ΟΥ] ΑΚΤΙΝ, solar ray (see Bitsakis and Jones, 2016b: 233). The golden sphere –Sun should be linked to the b1 gear or the lost cover of the b1 gear (see Voulgaris et al., 2019b: Figure 1). The phrase ‘solar ray’ can be related to the solar pointer, which travels across the zodiac months and their subdivisions

work page
[8]

Other researchers followed dif - ferent approach es and they arrived at dif - ferent starting dates, see Carman and Evans, 2014; Freeth, 2014; Iversen, 2017; Jones, 2020

This date resulted from combining Gemin- us’ definition of the exeligmos (Saros) cycle and the specific prominent and critical placement of the Winter Solstice inscrip - tion at the top-left corner of the Antikythera Mechanism (see Voulgaris et al., 2023c: Figure 5). Other researchers followed dif - ferent approach es and they arrived at dif - ferent star...

work page 2014

Show all 19 references

[9]

The axis of the b1 gear is located at the center of the front plate of the Mechanism (e.g. at the intersection of the two diagon - als of the rectangle shape of the Mech - anism), which is the center of the lunar cylinder and the common center of the Egyptian and zodiac dial r...
[10]

A dif - ferent number of holes would cause the Mechanism’s measurements and operation to malfunctions

The specific number of 365 holes is man - datory for the proper function ing of the Egyptian dial ring (see Section 3 ). A dif - ferent number of holes would cause the Mechanism’s measurements and operation to malfunctions
[11]

Wright (2005 : Figures 10 and 11; 2011: 12) suggests that it was a part of the A

Freeth and Jones (2012) suggest that this sliding catch was part of the back-cover plate. Wright (2005 : Figures 10 and 11; 2011: 12) suggests that it was a part of the A. Voulgaris, C. Mouratidis and A. Vossinakis The Antikythera Mechanism’s Zodiac Dial Ring ~ 272 ~ back plat...

2012
[12]

The position of the draconic pointer on the Antikythera Mechanism re presents the po- sition of the Moon in the zodiac belt. When the draconic pointer lies between the Eclip- tic limits a solar or a lunar eclipse occurs (depending on the relative position of the lunar pointer/...

2006
[13]

Therefore, 365.25 days/365 subdivisions ≈ 1.0006849 days/subdivision (≈ 24h 0.986m), and this difference is practically undetect - able when taking into account minor divis- ion errors, mechanical mismatches, and the width of the engraving tool’s nose
[14]

How did the ancient crafts- man kn ow when eclipses would occur?

During one full turn of the lunar pointer = one sidereal cycle of 27.321675 days ≈ 655.72 hours (according Geminus ’ des- cription of exeligmos and lunar cycles in Chapter XVIII; see, also, Voulgaris et al., 2023b), the lunar pointer transited the 365 subdivisions of the zodia...
[15]

A large number of astronomical clocks of the Medieval period present a circular lunar age scale divided into 29.5 subdivisions/days (see Guilbaud, n.d.)

Each pointer was accompanied by its cor - responding calibrated scale; the lunar disc probably had its ‘personal’ scale of 29.5 day-subdivisions based on the synodic cycle engraved on the (lost) cover of b1 gear (see Figure 7 ). A large number of astronomical clocks of the Med...
[16]

For the directions of the winds see Ana- stasiou et al., 2016 : 205; Wood and Sy- mons, 1894: 81, Figures 1 and 2)

2016
[17]

when the synodic, anomalistic and draconic cycles are all lo - cated at their beginnings ( Voulgaris et al., 2023a: 8–16)

According to Geminus, the Exeligmos (Saros) starts with the apokatastasis of the (three) lunar cycles, i.e. when the synodic, anomalistic and draconic cycles are all lo - cated at their beginnings ( Voulgaris et al., 2023a: 8–16)
[24]

and Saturn (line 25) he spent only one line on each of these . In that one line and in just 70–87 letters he described the planet’s sphere, its color, its position, its pointer and its oper - ation, and he also mentioned which com - ponent w ould be presented on the next line....

2003 arXiv
[90]

… the Sun stops there

or [93] 91 365d Geminus (Elem. Astron.) 90⅛ 94½ 92½ 88⅛ 365.25d Geminus (Par. collection) 89 95 92 89 365d Cleomedes 90¼ 94½ 92½ 88 365.25d itude), and the parapegma units-days would be in better agreement if the zodiac subdivisions were in days, since the index letters of the...

2013

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Voulgaris et al. (2022) make a case for the existence of the missing Draconic cycle that should exist on the Antikythera Mech - anism on the grounds of the Draconic gear- ing (as an engagement of gears b1 –a1, r1/Fragment D and one hypothetical small gear s1; and revised in gears s1, s2, t1 in Voulgaris et al., 2024), which represent the fourth lunar cycl...

work page 2022

[2] [2]

The hours of eclipse events can be easily calculated by using only the Antiky thera Mechanism and without any external infor- mation (unrelated to the Mechanism): by measuring the number of the zodiac ring subdivisions from New Moon/Full Moon to the next New/Full Moon, and convert - ing the m into days/hours (Voulgaris et al., 2023b: 22–25)

work page

[3] [3]

E.g. on Fragment C the lunar cylinder is stuck underneath the Egyptian/zodiac dial rings, parapegma plate 1 is stuck in front of the lunar cylinder, and one piece of parapegma plate 2 is stuck to the reverse face and in the opposite direction on para- pegma plate 1. See also Figure 6 for a part of the front dial plate that is trapped inside the petrified ...

work page

[4] [4]

AMRP radiography is Advanced Medical Radiography Practitioner radiography, and involves a senor radiographer with an ad- vanced level of knowledge and skill

work page

[5] [5]

Small parts of the parapegma plates with inscriptions are also preserved on Frag- ments 9, 20, 22, 28 ( Bitsakis and Jones, 2016a)

work page

[6] [6]

On the back cover inscription, the text ΠΡΟΕΧΟΝ ΑΥΤΟΥ ΓΝΩΜΟΝΙΟΝ Σ[ΕΛΗΝΗΣ], and a (lunar) pointer project- ing from it, are preserved (see Bitsakis and Jones, 2016b: 232)

work page

[7] [7]

The golden sphere –Sun should be linked to the b1 gear or the lost cover of the b1 gear (see Voulgaris et al., 2019b: Figure 1)

ΗΛΙ[ΟΥ] ΑΚΤΙΝ, solar ray (see Bitsakis and Jones, 2016b: 233). The golden sphere –Sun should be linked to the b1 gear or the lost cover of the b1 gear (see Voulgaris et al., 2019b: Figure 1). The phrase ‘solar ray’ can be related to the solar pointer, which travels across the zodiac months and their subdivisions

work page

[8] [8]

Other researchers followed dif - ferent approach es and they arrived at dif - ferent starting dates, see Carman and Evans, 2014; Freeth, 2014; Iversen, 2017; Jones, 2020

This date resulted from combining Gemin- us’ definition of the exeligmos (Saros) cycle and the specific prominent and critical placement of the Winter Solstice inscrip - tion at the top-left corner of the Antikythera Mechanism (see Voulgaris et al., 2023c: Figure 5). Other researchers followed dif - ferent approach es and they arrived at dif - ferent star...

work page 2014