Ptolemy's Almagest dominated astronomical thought from when it was written in the 2nd century CE, until the 16th century. Yet, as historically important as this work is, its nearly impossible to find out anything beyond a surface level understanding of what it contained.
I've spent nearly 7 years working through this text, making sure I understand every calculation. And now, I'll explain exactly what this book contains. In this series, I'll walk through the text - first explaining it conceptually, and then exploring select mathematical topics in future videos.
This video covers Book V in which Ptolemy updates the lunar model to account for discrepancies between the model derived in Book IV, and the moon's apparent position at first and third quarter. Book V also develops the calculations necessary to determine lunar parallax, and derives the sizes and distances to the sun and moon.
This video examines Book 5 of Ptolemy's *Almagest*, detailing the geometric refinements made to the lunar model to correct positional discrepancies during the first and third quarters. It explores the introduction of an eccentric deferent, the measurement of lunar parallax using zenith-distance instruments, and the creation of parallax tables to enable the accurate prediction of eclipses.
Written in the 2nd century CE, Claudius Ptolemy's Mathematicae Syntaxis,
better known as the Almagest, became the most important text in mathematical astronomy
for the next 1500 years. But despite the historical significance of this text,
it's nearly impossible to find detailed information on how its models were derived and
functioned. So much so that when I began studying historical astronomy, I decided it was best to sit
down and work through it myself. So in this series, I'll share what I learned about one of
the most important astronomical texts of all time. I'll begin by addressing things conceptually so
viewers that aren't interested in the detailed mathematical workings can get an appreciation for
Ptolemy's work. With that out of the way, I'll then explore select mathematical topics, taking
the understanding from conceptual to functional, allowing viewers to understand how the models
were practically applied. In this video, I'll be exploring Book 5. Here, Ptolemy explains an issue
that he discovered with the lunar model, develops a correction, and with his new model in hand,
examines how lunar parallax is determined. So join me as we learn how Ptolemy defined astronomy
before the telescope.
In the last book, Ptolemy made his first attempt at a lunar model. This one was simply the moon
going around an epicycle carried by a deferent tilted with respect to the ecliptic by about 5
degrees. However, he told us that it has problems. He stated that the model as it stands now works
pretty well for new and full moons, which happen at conjunction and opposition with the sun,
collectively known as syzygy. However, the current lunar model doesn't work well near 1st and 3rd
quarter. As such, Ptolemy is going to introduce a new complication to help resolve this issue.
To develop that first lunar model, Ptolemy exploited the geometry of eclipses to get the
coordinates of the moon on certain dates. But eclipses don't occur at 1st and 3rd quarter.
As such, Ptolemy is going to be forced to use some astronomical instrumentation
to directly determine the position of the moon. So, he opens book 5 by describing the
construction and use of what we would refer to as an armillary sphere, but which he called
an astrolabe. This instrument would best be explained with a hands-on demonstration,
and unfortunately, I don't have such an instrument. So, I'll have to skimp on the
details for now. I hope to visit St. John's College in Santa Fe, New Mexico at some point,
as they have an instrument based on Ptolemy's design. But for now, I'll simply say that the
instrument works by placing a ring in the plane of the ecliptic. There are then two sets of sites.
One would be aligned with an object of known position, such as the sun as calculated by the
solar model, or a fixed star, which we'll discuss when we come to books 7 and 8. The other site
would be moved to the object in question. This, then, gives the difference in their ecliptic
longitude. Thus, Ptolemy will be able to determine the positions of the moon necessary to calibrate
the revised lunar model. But what precisely does this revised model need to account for?
Ptolemy lays out the exact issue in chapter 2, telling us that the model works well sometimes
around first and third quarter. The discrepancy, he tells us, is very small or nothing when the
moon is at apogee or perigee of the epicycle, but reaches a maximum deviation when the moon
is near its mean speed, and thus the equation of the first anomaly is also at a maximum.
Furthermore, at either quadrature, when the first anomaly is subtractive, the moon's observed
position is at an even smaller longitude than that calculated by subtraction, the equation of
the first anomaly, but when the first anomaly is additive, its true position is even greater.
In other words, the effect of this discrepancy seems to magnify the effect of the epicycle
in creating the first anomaly. It doesn't have an effect when the moon is at apogee or perigee
on the epicycle because, at those times, there is no anomaly in position, so the magnifying effect
is essentially being multiplied by zero. Conversely, it is greatest when the moon is as far
away from its mean position as it can get. So how to achieve this effect? Ptolemy's solution
is to introduce a mechanism that physically moves the moon closer to the earth at first
and third quarter. Since, when things are closer to us, they look larger, this will have the effect
of magnifying the anomaly created by the epicycle, but have no effect at apogee or perigee since
those points are still along the line of sight. But how does Ptolemy change this model to achieve
this? The first step is to move the center of the deferent off of the earth, making its center
eccentric. However, that alone will not achieve the intended effect. That would only bring the
moon closer at certain times unrelated to the lunar phase. So, to ensure that it happens at
the correct times, he then places this center on a small circle of its own, which will rotate in
the opposite direction as the deferent. To understand why this works, let's temporarily
ignore the epicycle and only concern ourselves with its center. You can imagine the position of
the moon there for simplicity, and I'll draw a line to help us keep track of it relative to the
center of the deferent. I'll also draw in a blue line to help us keep track of the apogee of the
deferent, the point furthest away from the earth, which will always be along a line from the earth
through the now eccentric center of the deferent. Presently, the center of the epicycle, marked by
the moon, is at apogee, so you can't see this new line just yet. Next, let's imagine the position of
the sun being to the right. This means that the moon is currently in its new phase. As noted,
Ptolemy then sets things up such that the center of the deferent rotates in the opposite direction
of the deferent, but slightly slower. Let's go around first quarter to see why. As we discussed
in the last book, the lunar synodic month, the length of time it takes to go through the full
cycle of phases, is about 29 and a half days. That means that a quarter of that time, which brings
the moon from new to first quarter, is about 7 days and 3 hours. In that time, the sun has moved to
about here. If we looked up that same amount of time in the lunar mean motion table that we created
in the last book, we'd find that the moon has moved about the deferent by this much. However, we also
need to consider the motion of the center of the deferent and what that does to the apogee of the
deferent. In that same amount of time, it will have moved to here. Note that this is now directly
opposite the center of the epicycle, which means that the center of the epicycle is now at perigee.
In other words, it has pulled the moon closer to the earth at first quarter. To understand the
impact, let's temporarily draw in the epicycle, both with the deferent moved off-center and with
at its original position. First, let's imagine the moon located to the side on the epicycle,
such that it would have the maximum possible anomaly. When I draw in lines of sight from earth,
we can clearly see that the angle is greater and thus the effect of the anomaly is magnified as
compared to its unaltered position. Next, consider the moon at apogee or perigee about the epicycle
as viewed from earth. Here, changing the position doesn't actually change the angle as viewed from
earth, thus fulfilling the requirements that the effect of this change be at its greatest
when the anomaly is largest, but nothing when at apogee or perigee of the epicycle.
You'll also notice that the center of the epicycle had to move a bit more than 90 degrees
from its starting point to get to first quarter due to the sun also moving. Meanwhile, the apogee
of the deferent moved slightly less than 90 degrees by the same amount as the center of
the epicycle, so that it forms that straight line with the center of the epicycle at just
the right moment. Let's continue letting time pass until we reach the full phase.
Now the center of the epicycle is back at apogee, having moved a bit more than 180 degrees,
while the apogee of the deferent moved a bit less coming around to meet it from the other direction.
At third quarter, we're back to the center of the epicycle and the apogee of the deferent
being directly opposite one another, causing the epicycle's center to be pulled in closer to the
earth. And when we come back to the new phase, they're aligned again, returning the moon to
its default distance based on the first lunar model. To fully calibrate this model, we'll
first need to know just how big this effect is. Ptolemy works this out in chapter three.
To do so, Ptolemy looks for observations which, again, fit special criteria. First,
he requires that the moon's speed be close to its mean speed. This happens when the moon is the
furthest it can be to either side of its epicycle, which is where the anomaly caused by the epicycle
is at its greatest. Second, it needs to be around first or third quarter. Again, this comes from
the observation that the second anomaly is greatest around those times. And third, the moon needs to
have no longitudinal parallax. The reason for this is that parallax can cause the moon's apparent
position to shift in both ecliptic latitude and longitude. Thus, if we used an observation of the
moon when there was a longitudinal parallax, we wouldn't know if the discrepancy from the
calculated value was due to this second anomaly or the effects of parallax. Thus, we should
eliminate the parallax as a variable for this. Ptolemy comes up with a suitable observation and
determines that the moon, at the time of the observation, was 7 degrees and 40 minutes away
from its mean position. He then calculates what the anomaly from only an epicycle would have been
and determines it to have been 4 degrees and 57 minutes, which he rounds off to an even 5 degrees.
To drive the point home, he calculates the same using an observation made by Hipparchus.
In reality, he comes up with a difference of nearly 3 degrees, but calls it close enough
and proceeds forward with the value of 2 and 2 thirds degrees. With that in hand,
Ptolemy is ready to work out how large the small circle carrying the deferent's center
needs to be. This takes a bit of geometry, but Ptolemy pretty quickly determines that the center
of the moon's deferent would need to be offset by about 10 and 1 thirds parts, where the radius of
the moon's deferent is 60 parts and the radius of the epicycle is about 5 and 1 fourths parts.
You'll notice that the amount that this brings the moon closer is huge, more than 1 sixth of
the distance of the deferent alone, which means that the apparent size of the moon in the sky
should change considerably, but it doesn't. This is a major problem with the theory,
one that Ptolemy is suspiciously quiet about. When using the first lunar model, we were instructed
to determine the amount of anomaly by considering the distance of the moon about its epicycle from
apogee, but this new model complicates that. Previously, determining the apogee of the
epicycle was easy, as we could simply draw a line from the earth through the center of the
epicycle to its far side, but with the deferent's center being moved off of the earth, it's no
longer clear that we should do that. Should we instead be determining the point of apogee by
a line drawn from the deferent's new center? In chapter 5, Ptolemy demonstrates that neither
answer is actually correct. Instead, there's a third point, which he calls the mean apogee,
that we should draw the line from. That point is twice the distance from the earth as the
deferent's center along the same line. The proof is pretty involved, so I won't get into it here.
Another thing that the change in the model screws up is our ability to simply determine
the anomaly from knowing the position of the moon about its epicycle. So, chapter 6 is dedicated
to showing how we can calculate the actual anomaly with this new complication. Again,
the calculations are getting too complicated to explain in a video that's focused on a qualitative
understanding, but I will say that one of the new factors that we now have to account for
is the elongation, the distance around the ecliptic from the sun. This certainly makes
sense since the deviations this model is meant to explain were meant to only come into play
at certain lunar phases, which are directly dependent on the elongation.
Putting a slightly finer point on it, the real value we need is two times the elongation,
which is aptly named the double elongation. Understanding this, Ptolemy then tells us,
in chapter 7, how he's going to condense these various pieces of information into a table for
quick calculation. And that table is laid out in chapter 8.
Chapter 9 then gives an example of how to use this to calculate the lunar position.
This is just done as a set of instructions and not a specific example. First, we determine the
amount of time since the beginning of the epoch, and then break it up into segments that we can
find in the relevant tables. We'll then look up all of the values for the change in longitude,
elongation, anomaly, and latitude to determine the values for each at the time in question.
We then use the value for the elongation and double that to enter into the table from chapter
8 to get a few adjustments for the actual anomaly and latitude which, then applied,
give the position of the moon from the earth. Well, mostly. We'll come back to that shortly.
This seems like it should complete what this book was all about. After all, we now have what
appears to be a working lunar model. But Ptolemy isn't done yet. Ptolemy next spends a chapter
going back to prove that the changes we made to the model don't impact its ability to function
at Syzygy's new and full moon. After all, it would be pretty bad for the model if it broke
the phases the original model did work for. Ptolemy writes,
Some people will suspect that the moon's eccentric circle might also have a considerable effect at
conjunctions and oppositions, and the eclipses occurring at them. Therefore, we shall try to
show that this difference cannot produce any considerable error in the calculations of the
phenomena at Syzygy's. Going through some geometry, Ptolemy demonstrates that the eccentric circle
renders the same results as the first lunar model. Ptolemy then returns to address a potential
problem with the calculation of lunar position that he laid out in chapter 9. Specifically,
that calculation assumes that the observer is at the center of the ecliptic. Unfortunately,
we're not exactly at the true center because the Earth has some size. While in book 1 Ptolemy
stated that the size of the Earth is negligible in relation to the size of the cosmos as a whole,
this is not true for the size of the Earth compared to the orbit of the moon.
As we've noted previously, the moon is close enough that parallax plays a significant role
in where it appears in the sky. If you'll recall, Ptolemy has worked quite hard to ensure that the
observations he chose would be free of the challenges of parallax, so the model is fine.
But if you want to use the model to predict where the moon would appear in the sky,
you will need to take parallax into account. This chapter ends up being free of math and is
generally Ptolemy laying out a bit of a roadmap for what we're going to need to determine parallax.
Most importantly, he tells us, we'll need the ratio of the distance to the moon and Earth's
radius. But how to know this? Ptolemy states that we can find it if we can determine the amount of
parallax in a particular situation. This seems like a bit of a catch-22 in which we need this
ratio to find the parallax, yet need the parallax to find the ratio. But as we've seen, Ptolemy is
quite skilled at finding particular arrangements that make things easier. We'll see another example
of this in the next chapter. Next, Ptolemy tells us that if we're not able to measure the parallax
of objects, we're generally not going to be able to measure their distance. Here, Ptolemy is
referring to the five planets, foreshadowing that we will not be able to derive their distances
in meaningful units. Their orbits will always remain in parts, tailored to each planet's
orbit. Lastly, in this chapter, Ptolemy briefly discusses how Hipparchus approached the question
of measuring the ratio of the lunar distance and the Earth's radius. He states that Hipparchus
made use of a total solar eclipse to set up the geometry to work out the ratio. However,
there's a major question that Hipparchus was unable to resolve. Does the sun exhibit parallax?
If so, it changes the geometry and subsequently the ratio we're looking for. Ptolemy states that
it is altogether uncertain, in the case of the sun, not only how great its parallax is, but
even whether it has any parallax at all. Ptolemy leaves this as an open question and closes out
the chapter there. Chapter 12 introduces us to a new instrument that Ptolemy uses for the
observations he's going to need. I won't go into it in too much detail because it's a pretty simple
instrument. Despite the name, this instrument isn't designed to measure parallax directly.
All it does is measure the angle of an object from the zenith. Well, what good is that? It turns out
there's something interesting about the way that parallax works that's directly related to the
zenith distance. To understand, let's make a quick diagram. Here, I'm drawing the Earth viewed side
on. Around it, we'll place the celestial sphere, really just to serve as reference.
Now, let's pick an observer at some latitude, let's say around here. I'll call the position
of the observer A. We can determine where that observer's zenith would be on the celestial sphere
by drawing a line from the center of the celestial sphere, i.e. the center of the Earth,
through the observer, which means that the zenith for that observer is here, which we'll call point
Z. Next, let's draw in an object. For now, I'll place it here, and we'll call that position point
D. Now, let's consider where this object appears to be from the point of view of the center of the
celestial sphere. To do that, we'll draw a line from the center, through the object, until it
hits the celestial sphere, and call that point H. This shows where the object would appear to be
without any parallax. Now, let's ask where the object would appear from the point of view of
the observer at A. We draw a line from them, through the object at D, to the celestial sphere
at T. And as you see, it's different from where it appeared from the center of the celestial sphere.
Its apparent position has changed by arc HT. This is the parallax. But what happens as we move that
object closer and closer to the zenith? As we move it, we can see that the arc that is the parallax
decreases. So what we can clearly see is that the further an object is from the zenith,
the greater the parallax is. You might be thinking to yourself, this is all well and
good for a flat diagram like this one, but the celestial sphere is three-dimensional.
And you'd be right, but it also doesn't matter. This diagram was based on three points,
the center of the celestial sphere, the observer, and the object. Three points define a flat plane,
so a two-dimensional diagram is always suitable. And what that means is that not only does the
parallax increase the further you get away from the zenith, but the direction in which it appears
to move is always directly away from the zenith. In other words, along a great circle drawn through
the zenith and the object, i.e. an altitude circle. Which is precisely why this instrument
will be so useful. It measures precisely the angle in question along the circle in question.
Ptolemy then gives an example of how this instrument might be used. Since we started
discussing the moon in the last book, I've repeatedly noted that the moon bounces up and
down with respect to the ecliptic by about 5 degrees. But how do we know this? Ptolemy uses
the answer to that question as his example. He sets out to take an observation of the moon as
across the meridian at a time when the moon was near its northern limit, i.e. its maximum ecliptic
latitude, and near one of the solstices on the ecliptic. Ptolemy imposed these requirements
because when a solstice point of the ecliptic is on the meridian, then the ecliptic is perpendicular
to the meridian, which makes the lines of ecliptic longitude parallel to the meridian.
In other words, although he's technically measuring an altitude circle, because the lines are
parallel, it's essentially the same as measuring differences in ecliptic latitude, allowing him to
freely mix values from different coordinate systems. Thus, he can subtract the angle between
the zenith and the moon from the latitude at which he was observing in Alexandria, Egypt.
This will tell him how far from the celestial equator the moon was at that moment, which can
then be compared to the distance of the ecliptic from the celestial equator at that same moment.
The leftover is the ecliptic latitude of the moon at that time, which Ptolemy finds to be
quite close to the 5-degree figure that we've been referencing. Incidentally,
due to the latitude Ptolemy observed from, again in Alexandria, Egypt, this meant that the moon
would be quite close to the zenith, which means that there would be very little parallax. Which
is good, because we don't want any parallax for this exercise, since we're only trying to measure
the true distance from the ecliptic. Now that he's explained the instrument, Ptolemy's ready
to start using it for the real purpose we set out, to determine parallax for the moon at a specific
point in time, which we can then use to calculate distance. To do this, Ptolemy describes an
observation he made of the moon as it crossed the meridian early 1 October, when the moon was near
the winter solstice. From this observation, Ptolemy is able to compare the observed position of the
moon to the position expected from his model to determine how much it was shifted due to parallax.
Having worked out the amount of parallax, Ptolemy can now do some geometry to work out the distance
to the moon in terms of Earth radii. He comes up with a value of 39 ¾ Earth radii away.
But keep in mind, this value is just for that moment in time. The moon moves closer and further
due to both the epicycle as well as the small circle on which the center of the deferent is
moved. Thus, we're not done yet. Ptolemy calculates the arrangement of all of the spheres for that
moment, as well as the moon's position about them, and determines how large they would have to be
to meet the additional criteria of placing the moon at the distance he just determined.
From this, Ptolemy determines that the center of the epicycle is 59 Earth radii away when at its
greatest distances, at conjunction and opposition, but gets brought as close as 38 ¾ Earth radii
when pulled closer during 1st and 3rd quarter. Furthermore, the epicycle must have a radius of
5 1⁶ Earth radii. With supposed values for the distance of the moon at various points in its
orbit, Ptolemy wants to work towards a distance to the Sun. But he doesn't have all of the
information he's going to need to be able to do so just yet. In addition to the distances,
he's going to need the apparent angular size of the Sun and Moon at conjunction and opposition.
Thus, chapter 14 opens with Ptolemy briefly discussing and rejecting various methods by
which astronomers prior to him attempted to measure these values. In general, they appear
to have been by determining the amount of time that it took these objects to cross the horizon
using water clocks. Then, by understanding the rate at which their spheres turned,
they could estimate the angles involved. Instead, Ptolemy tells us that he followed
Hipparchus' method and constructed an instrument that he refers to as a dioptera. Ptolemy gives a
bit more detail on this instrument a bit later, stating that it has a plate covering the object
being sighted which runs down the rod described. Using this instrument, Ptolemy found that the
Sun's diameter always subtends approximately the same angle, there being no noticeable difference
due to its distance, but that the Moon subtends the same angle as the Sun only when it is at
its greatest distance from the Earth, i.e. at the apogee of the epicycle, at full moon,
in contradiction to the hypothesis of my predecessors who assumed that it subtends
the same angle as the Sun at full moon when it is at its mean distance. What Ptolemy is
stating here is that when the Moon is at its maximum distance, both with the deferent at
its maximum distance and the Moon at its furthest point about the epicycle, it will just barely
cover the Sun, and when brought closer, either by the deferent being pulled closer or the Moon
being closer to us about the epicycle, then it will cover the Sun all the more easily.
This is an interesting claim since it assumes that annular solar eclipses, in which the Moon's
apparent size is smaller than the Sun and thus doesn't cover it completely, are impossible.
Given such eclipses are possible, we have to wonder how Ptolemy discounted them. But
while Ptolemy firmly holds the conclusion that the Moon's apparent diameter matches the Sun
at the lunar maximum distance, he doesn't fully trust the diopter, and instead develops a new
method to determine the lunar diameter. For this, he again turns to lunar eclipses,
but this time selects a pair of partial eclipses. Because the eclipses are partial, this means that
the Moon's center wasn't quite back to the nodes, which happen when the Moon is 90 degrees away from
its northernmost point. Ptolemy can therefore use the lunar model to determine how far along the
Moon's inclined path it was from the node. Ptolemy selects one of his eclipses in which the Moon's
disk was half-obscured. This means that the arc drawn between the lunar circle and the ecliptic
would, in this case, also be measuring the angular width of the Earth's shadow. For the
second eclipse, Ptolemy states that one-fourth of the Moon's width was covered. Ptolemy can then
look at that same arc between the lunar path and the ecliptic and ask how much longer it was for
this eclipse. That difference, then, is one-fourth of the diameter of the Moon. Thus, he concludes
that the Moon's width is 31 minutes and 20 seconds of arc, just over half a degree when at its
distance. Using this, Ptolemy can also ask how large the Earth's shadow is at this distance,
and Ptolemy finds to be about two-and-three-fifths times as wide as the full Moon.
As a quick reminder, these videos are meant to communicate Ptolemy's work, not to critique it.
That being said, there's a lot of issues with this chapter that I hope to return to in a future
video. And yet, somehow, the value Ptolemy arrives at is quite reasonable, falling squarely in the
middle of the Moon's actual variation in angular size. With this value for the angular size of the
Moon in hand, Ptolemy is ready to determine the distance to the Sun. To do so, he begins with a
schematic diagram depicting the Sun, Moon, and Earth all arranged in such a manner as they would
be to create a solar eclipse with the Moon at its maximum distance. Again, this is the case in which
Ptolemy takes the angular size of the Moon to be equal to the angular diameter of the Sun.
Ptolemy also considers the Earth's shadow, which allows him to geometrically determine that,
if we take Ptolemy's previous proof in which he concluded that the center of the Moon's epicycle
is 59 Earth radii away, then the Sun's distance is 1,210 Earth radii away, and the apex of the
Earth's shadow is 268 Earth radii away. And knowing this, it's simple to work out the sizes and volumes
of the various bodies. Doing so, Ptolemy concludes that the Earth's diameter is 3 2 5 times the
diameter of the Moon, and the Sun's diameter is 18 4 5 times the diameter of the Moon. Obviously,
these values are wildly incorrect, but they don't really play into anything further.
Instead, Ptolemy turns his attention back to the calculation of parallax.
While Ptolemy has certainly demonstrated the method for calculating the parallax,
it's an annoying calculation to go through. As such, Ptolemy tells us he wants to find a
convenient method of calculating the parallax for distances at intermediate positions between
apogee and perigee from the parallaxes tabulated above. In short, Ptolemy is going to do what we've
done so many times now, create a convenient table for quick reference in which we can look up the
values we need instead of having to calculate them from scratch each and every time. As we've
learned in this video, the parallax is primarily dependent on two things, how close the object is
and how far it is from the observer's zenith. For the distance, Ptolemy tells us that although the
Sun's distance does vary, since its circle is eccentric, the amount by which it varies compared
to its distance is sufficiently small that we can pretend that it's at a constant distance.
But, as we've seen for the Moon, its distance varies wildly in Ptolemy's model. Thus,
Ptolemy considers four special positions, when the Moon's deferent circle is at its greatest and
least distances and, for each of those, when the Moon is at its closest and furthest points about
the epicycle. He then walks us through a sample calculation of each of these luminaries being
30 degrees from the zenith for each of the situations just mentioned. And, as usual,
Ptolemy does the calculations for other zenith distances behind the scenes to save us the work.
This being done for the four situations we just described, Ptolemy then calculates some additional
columns for the table to account for the Moon being at intermediate positions about the epicycle
and for the varying distances of the deferent. The results of these calculations are then presented
in chapter 18 as a large table. With the theoretical underpinnings now firmly established,
Ptolemy then instructs readers on how to actually make use of the table he just constructed to
calculate the parallax. To do so, we're going to need to know the zenith distance of the Sun
or Moon at the time we're calculating. While Ptolemy has given us an instrument by which
we could measure that, that does us no good if we're trying to calculate at a time in the past
or future. As such, determination of the zenith distance has to be done by calculation using the
lunar or solar model. Fortunately, we actually already have everything we need to do this.
Specifically, at the end of Book 2, we created an entire table to calculate the distance of
points along the ecliptic from the zenith for various hours before or after noon. So,
to determine lunar parallax, we'll need to calculate where the Moon is along the ecliptic
using the updated lunar model, and then use the table from Book 2 to determine the zenith distance.
This value then gets looked up in the parallax table just created. However, the process is
somewhat different for the Sun vs the Moon, since we assumed that the Sun's distance is constant
while the Moon's distance varies. Ptolemy first provides instructions on how to calculate the
lunar parallax. The value from the zenith angle table from Book 2 gets looked up in the parallax
table, and the values for the four special cases written down. Then we need to apply those
correction columns I mentioned, which take into account the position of the Moon about the
epicycle as well as the deferent being moved closer and further. The motion about the epicycle
is something that is part of the calculation of the lunar position before even considering
parallax, so we will already have that handy. For the motion of the deferent, recall that it was
designed such that the Moon was pulled closer at first and third quarter, and pushed back to the
distance from the first lunar model at the new and full phase. The key piece of information here is
that the distance is tied to the lunar phase, which is a function of how far the Moon is from
the Sun. This is why Ptolemy inserted values into his tables, which help us quickly determine the
distance from the Sun about the ecliptic, known as the elongation. Thus, this value is something
that we get from these tables as well. The math to combine all of these pieces of information is
rather complicated, so we won't get into it here. Rather, we'll just note that everything
is distilled down to basic operations, addition, subtraction, multiplication, and division.
There's no need to deal with solving triangles or anything more complicated,
as would have been necessary if we were finding the parallax by geometrical means.
However, Ptolemy is not quite done. While we should know how to calculate the parallax for
the Sun and Moon, we only know the total distance the object would appear to move
along a great circle drawn through the object in the zenith. This alone isn't particularly helpful
because that great circle isn't generally aligned with the preferred coordinate system.
Thus, we need to break the apparent change in position down into its component pieces
in ecliptic longitude and latitude. However, Ptolemy's math here gets a bit dodgy.
To explain, let's sketch a diagram. First, let's draw a segment of the ecliptic and a segment of
Moon's inclined circle, making sure to include the node where they intersect. Now, we'll draw
in the true position of the Moon along its circle, and then draw a line from it perpendicular to the
ecliptic, which will be the line describing ecliptic latitude for the true lunar position.
Now, we'll draw in the zenith and a segment of the great circle from it through the Moon
until it hits the ecliptic. The lunar parallax will shift the Moon's apparent position along
this line away from the zenith. So, we can now mark the position of the Moon's apparent position
after parallax. Having that, we can draw in a short line parallel to the ecliptic that will
be the component of the change in longitude due to parallax. On the other side of the triangle
is the component of the change in latitude. This little triangle is the key to breaking the change
in position due to parallax into its components as it's a nice little right triangle in which
know the hypotenuse. Thus, if we know one of the non-right angles, we should be able to solve it,
right? Well, not quite. For starters, we don't actually have one of the angles in this triangle.
What Ptolemy told us to do was refer back to the table of angles we derived at the end of Book 2,
which also contained information on the angle between the ecliptic and an altitude circle.
But what altitude circle? What we want is the altitude circle that goes through the zenith
and the Moon to the ecliptic, but the way that the table Ptolemy tells us to refer to was
constructed, it's actually the altitude circle between the zenith and the point on the ecliptic
of the Moon's actual ecliptic longitude. Thus, we're finding this angle, when the angle that we
really need is this angle. These angles are close, but they're not the same. Fortunately, Ptolemy is
able to work up a correction to get us to the angle we actually want. But there's another
problem. While knowing a side and an angle is sufficient for planar triangles, ones drawn on
flat surfaces, things are different when you're working on curved surfaces like the surface of
the celestial sphere. However, Ptolemy tells us that we should go ahead and calculate it as if it
were planar, essentially constructing a triangle of the chords of these points cutting through the
This is not strictly correct, but for such small triangles, there's not enough curvature on the
surface for it to be substantially different. As Ptolemy said, circular arcs of such small size
are not noticeably different from straight lines. Having walked us through this conceptually, Ptolemy
then takes us through an example of the calculation of the corrections that he just described and
This was a longer book, so let's review what we've covered.
First, Ptolemy derived an update to the lunar model developed in Book 4.
This was done to address the fact that, at first and third quarter, the amount that the
moon's position deviated from the mean position was more than could be explained by an epicycle
alone. Ptolemy's solution was another small circle that pulled the deferent closer at
Ptolemy then explored parallax for both the moon and sun, which required determining the
distances to them, which in turn required finding the angular sizes of both. Then,
he showed us how to calculate the parallax using a new table he derived for just this purpose.
This may all seem like we're getting into very small details that aren't really that important,
but in Book 6, we're going to put the solar and lunar model together to be able to predict
eclipses, and the characteristics of eclipses are significantly impacted by these small details,
which is why Ptolemy has spent so much effort in getting them right.
So, join me next time as we continue to explore how Ptolemy imagined astronomy before the telescope.