My fascination is two-fold: How simple can it be made? How powerful can it be made? After researching several methods, many obscure even to those who enjoy calendar calculating, I've managed to merge two astounding versions that bring together the best combination of power and simplicity I've ever run across.
Effect
You start by bringing out a business card with this calendar design:
You then ask someone for the year, and then the month, of their birthday. With no further questions, you're instantly able to fill in the calendar for that month and year. For example, if the person said they were born in 1979 in September, you could quickly fill in the calendar from memory like this:
They can verify your calendar with any outside source, such as Wolfram|Alpha, and see that the calendar you created from memory is 100% accurate!
Even if you've become frustrated trying to master other day of the day of the week for any date formulas, I think you'll like the simplicity and flexibility of this version. You can start learning the method in the next tab!
Year Key
With this method, you ask for the year first, and you're going to calculate a date from it, ranging from 0 to 6.
In many methods, calculating a number from the year is the biggest hurdle, both because of the number of calculations involved, and the fact that many calculations involve increasing the amount represented by the year. In this method, however, the steps are few, and the method always involves making the year number smaller.
When given a year, the first thing you need to do is to break it up into the last 2 digits, which will be referred to as the year from here on out, and the remaining digits, which will be referred to as the century. For exam, if you're given the year 1975, you would think of 75 as the year and 19 as the century.
First, I'll explain the calculations to perform on the 2-digit year, and then I'll explain how to handle the century.
Step 1 - Partition: Split the 2-digit year into a sum, consisting of the largest multiple of 4 equal to or less than the given year, plus a remainder of 1, 2, or 3, if needed.
Example: 75 is split into 72 + 3. 72 is the largest number equal to or less than 75, and 3 is the remainder needed to make it a sum that totals 75.
Step 2 - Modify & solve: Divide the multiple of 4 by 2, and change the plus sign to a minus sign. Once you've modified the problem, the next step is to solve it.
Example: 72 (from 72 + 3 above) ÷ 2 = 36, so 72 + 3 becomes 36 - 3. Solving this, we get 36 - 3 = 33.
Example: The largest multiple of 7 equal to or less than 33 is 28, so we subtract 33 - 28 to get 5.
This may seem like a lot of work at first, but it can be done surprisingly quickly, once you become familiar with it.
Below is a simple widget to show you how this is done with various years. Simply select the first digit of the 2-digit year with the left menu, and the second digit with the right menu. You'll automatically see how the calculations below are affected by each change.
Starting year: 75
Partition: 72 + 3
Modify & solve: 36 - 3 = 33
Cast out 7s: 33 - 28 = 5
Note that years which are evenly divisible by 4 are easy to calculate. You'll quickly learn that multiples of 4 allow you to skip the partitioning step, and go straight to dividing a single number by 2.
Also, pay particular attention to the years 01, 02, 03, and 07. When these are partitioned, divided, and subtracted, they all result in negative numbers. In these few cases, you need to add 7 to make them positive. If you can remember that -1 + 7 = 6, -2 + 7 = 5, and -3 + 7 = 4, you can handle all of these situations.
The number you calculated from the year above will now be modified to take the century into account. Adjusting for the century only requires a little memory and a single addition.
If the requested century is in the 1700s, you don't need to add anything. For the 1800s, you'll add 2 to your running total. For the 1900s, you'll add 4, and for the 2000s, you'll add 5. For years after that, the pattern repeats every 400 years:
+ 0
+ 2
+ 4
+ 5
---
---
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
This table only goes back to 1500 because the calendar we currently use, the Gregorian calendar, was first adopted on October 15, 1582. In later lessons, you'll learn how to handle earlier dates.
To help you recall the right number, remember the nonsense mnemonic, “No tuna for Friday”. No represents +0, tuna represents +2, for represents +4, and Friday represents +5, since Friday is the 5th and last day of the traditional 5-day Monday through Friday work week.
How do you know which numbers go with which years? From right to left, they cover the 1700s, the 1800s, the 1900s, and the 2000s respectively. Bob Goddard (see Credits & Notes below) suggesting remembering that the mnemonics apply to the first 400 years of American history.
Once you can recall which modification goes with which year, you simply add the appropriate number for the century. Don't forget to cast out 7s if you get a number that's 7 or larger.
Going back to our 1975 example from above, we worked out a key number of 5 for 1975. To modify this number for the century, we add 4 for the 1900s, so 5 + 4 = 9. Casting out 7s, we subtract 9 - 7 = 2. So, 2 is our final key number for 1975.
What about, say, 1776? 76 doesn't need to be partitioned, since it's a multiple of 4. 76 ÷ 2 = 38, and casting out 7s we get 38 - 35 = 3. Since the 1700s have a century modification of +0, we simply don't add anything, and leave the total as 3. So, the year key for 1776 is 3.
Once you have this year key, what do you do with it? The next step is to take the key number you just calculated, and combine it with a particular month. You'll learn how to do that in the next tab.
Credits & Notes
The mathematical method for handling the 2-digit year comes directly from an article titled, “How to find the day of the week on which any particular date falls” by E. Rogent and W.W. Durbin in the August 1927 issue of The Linking Ring.
Now that you've calculated a key number based on the year and century, which will be treated as a date from here on out. The next step is to ask for a month. Each month is associated with a day of the week, and all you'll do to create the correct calendar for the month is to put the date you calculated from the year and century together with the day associated with the given month!
Months and Days
To help you recall with days are associated with which months, I developed 7 short rhymes to help make the connections memorable:
February, March, and November,
Three on Sunday to remember.
February, March, and November are all associated with Sunday. Sunday also happens to be the only day of the week associated with 3 different months. Each of the other days is associated with only 1 or 2 months each.
June day?
Moon-day!
This short rhyme makes it easy to remember that June is associated with Monday.
A skeptic needs clues,
but Sept./Dec. needs Tues.
Sept. is Septemeber, and Dec. is short for December, and they're both associated with Tues., or Tuesday.
Winds in April, Winds in July,
“Winds”-day’s the day, wet or dry.
This is simple enough. April and July are associated with Wednesday.
Jan’s and October’s 31,
Reminds us that Thursday won.
Jan, of course, refers to January. Not only are January and October both associated with Thursday, but they're the only set of months in a group which all happen to share the same length.
Month 5 is May,
Day 5 is Friday.
May being associate with Friday made for an easy rhyme, of course. Since May actually rhymes with every day of the week, I thought the additional association of the 5th month and the 5th and final day of the work week strengthened the association.
August’s few holidays make you yearn,
For the day off named after Saturn.
In most English-speaking countries, August tends to be the month with the fewest nation-wide holidays. This made for an interesting rhyme to help you memorize that August is associated with “Saturn-day”, or Saturday.
Leap Year Exceptions
Leap years stand out in this method because they don't need partitioning like most other numbers. If you're given a leap year, work out the key as taught previously, and then make a mental note that it's a leap year.
January and February both change the days with which they are associated in leap years. January in a leap year moves 1 day back from Thursday to Wednesday. In other words, in a leap year, January, April, and July are all associated with Wednesday, and October is the only month associated with Thursday.
Similarly, February also moves 1 day back from Sunday to Saturday in a leap year. In leap years, then, February and August are both associated with Saturday, and only March and November are associated with Sunday.
These two leap year exceptions may seem difficult to manage at first, but if you just think of January and February moving their day associations back by 1 day, it quickly becomes natural.
Remember that any year which is evenly divisible by 4 is a leap year, except for years ending in 00. Years ending in 00 are only leap years if they're evenly divisible by 400. For example, 2000 was a leap year, but 1900 and 2100 are not.
The approach of associating each year and century with a date, and each month with a day of the week is also from “How to find the day of the week on which any particular date falls” by E. Rogent and W.W. Durbin in the August 1927 issue of The Linking Ring.
In the original article, however, Durbin and Rogent's month and day associations were based on dates in the 1800s. In order to make the math work properly with Bob Goddard's First Sunday Doomsday Algorithm, however, the month and date associations have been changed so that they're now based in the 1700s.
Since the month and day association have been changed, the rhymes for remembering each association are original with me.
Creating the Calendar
At this point, many calendar calculation methods would have you do one last calculation to work out the day of the week for the given date.
Since my goal here is to minimize the calculation, this step simply involves creating a calendar for the given month and year. Here's a PDF template for a set of blank calendars, sized for printing on the back of business cards. When you bring out a single one to use, it should appear like this:
This step is where all of your calculation and memory work come together. To show you how, let's say you ask for a year, and you're given 1982. As you write down 1982 in the Date section of the calendar, you run through the calculations in your head: 82 = 80 + 2, which becomes 40 - 2 = 38, cast out 7s so we have 3, add 4 for the 1900s, giving 7, casting out 7s, leaves us with 0. We now know 1982 is a 0 year.
For the month, we'll say that you're given April, which you should immediately associate with Wednesday (Remember? “Wind in April...”), as you write down April in the Date section.
At this point, you know that 1982 is a 0 year and April is associated with a Wednesday, yet the calendar looks like this:
Generally, the rule is to put the day of the week you recalled above the number you were given. If you have 5 and Monday, you'd put Monday directly above the 5 on the calendar. In this case, though, we have 0 and Wednesday, so what do you do?
Simple, think of the “0th” of a month as being 1 week before the 7th. In the case of 0, you always put the day of the week above the 7. Since April is associated with Wednesday, we put the abbreviation Wed above the 7th, like this:
From here, it's easy to see the remaining days of the week should be placed:
Don't forget to give the month the correct number of days. In our example, April is a 30-day month, so we add a 29 and 30 in the bottom row:
At this point, you're done creating the calendar! If this was someone's birthday, and they were born on April 17, 1982, a quick look at the calendar is all you need to see that they were born on a Saturday. You can circle that date and the day of the week, and now they have a souvenir of their birthday, and your amazing calendar skills!
Another Leap Year Reminder
In the previous tab, you learned that when working with leap years, and you're given either January or February as the month, you need to move the day associate back by 1 day. Let's see how this works in practice.
For this example, we'll choose 1992 as the year. Multiples of 4 aren't partitioned, so we start with 92 ÷ 2 = 46, and cast out 7s (46 - 42) to get 4. Adding 4 for the 1900s to that, we get 8, and we cast out 7s to get 1. You've worked out that 1992 is a 1 year, and also realize you need to keep alert if they ask for January and February.
Continuing with this example, we'll imagine that February is chosen as the month. When you hear this, you need to recall 2 concepts. First, February is associated with Saturday in a leap year, and second, that a leap-year February has 29 days. Putting all that together, the resulting calendar should appear like this when finished:
Leap-year Januarys and Februarys are probably the worst-case scenario you'll face in this method. Once you're comfortable with those, you're ready to perform this feat.
Verification by the Spectators
For this feat to really be impressive, they need to know that the calendar you created is correct. In the past, this meant carrying around cumbersome almanacs. Today, though, the handiest and most convenient way is carry an internet-connected mobile device around.
They can verify it with any kind of search, but I usually have them point their browser to Wolfram|Alpha (http://www.wolframalpha.com). Once you have the calendar created, have them enter the year and month followed by the word calendar, and Wolfram|Alpha will generate the calendar for that month!
Over in the Quick Calendar Month Creation Quiz, you can practice creating calendars by clicking on the Calendar Quiz button. This quiz is different than the previous quizzes, however, as it opens up full-screen in a new window.
There are two pull-down menus near the top. By default, the one on the top-right reads Gregorian: 1700-2099, which are the years you'll probably want to start with when you're first learning this feat.
Clicking on this menu will bring up other selections, which are: Gregorian: 1582-9999, which tests on years in the current calendar, but in a much wider range (the first 2 items on the tips page can help here), Julian: 1 AD - 1582 AD, and Julian: 45 BC - 1 BC. You won't need to worry about these until you've read the next page.
To use this quiz, select a range from the top-right menu, and then click on the New Date button on the top-left of the screen. Down by the word “Date:”, at the bottom, a random year from the selected range will appear, followed shortly by a random month.
The days of the week towards the top of the screen can be scrolled left or right by using the left and right arrow keys on your keyboard, or by swiping them left or right, if you're using a mobile device. What you want to do is use this ability to scroll the days to align the key you calculated from the given year with the day you associate with the given month.
For example, if you're given 2014, and you quickly work out a key of 2, and then you're given the month of May, which is associated with Friday, you want to scroll the days left or right, so that Friday is right above the 2.
In order to make sure the number of days are right, you use the pull-down menu at the top-left of the screen, which defaults to 28 days. You can also select 29 days, 30 days, and 31 days, and the added days will appear in the bottom row. Continuing with our 2014 May example, you'd want to select 31 days, since that's the length of May, so that 29, 30, and 31 appear in the bottom row.
Once you're satisfied that you've created the month, click the Verify button at the top right to see whether you are correct. It can inform you whether the number of days in the month is wrong, whether the days aren't labeled correctly, or whether you got it correct. If nothing is correct, it will simply say, “Oops. Try again, please.”
To try again, simply click the New Date button, and you're given a brand new random year and month!
Once you're comfortable working with years in the Gregorian calendar, the next tab will show you how to handle years in the Julian calendar, all the way back to 45 B.C.!
Credits & Notes
Creating a calendar month as an alternative to doing calculations involving an individual date is an approach I developed when I created Day One. In fact, the calendar PDF linked in the top of this tab was originally developed for Day One.
The quiz seen here is also a modified version of the calendar quiz I originally developed and included with Day One.
Other countries didn't adopt the Gregorian calendar until later. Britain and its colonies, including America at the time, adopted it in 1752, jumping from the Julian date of September 2 to the Gregorian date of September 14 on the following day.
In the Julian calendar, leap years happened every 4 years, including years ending in 00. With the Gregorian calendar, years ending in 00 are only leap years if they were evenly divisible by 400. For example, 2000 was a leap year, but 1900 was not and 2100 will not be.
With this shifting ahead of days, and different leap year rules, you might think it would be difficult to account for this change in a calendar calculation. I have good news for you. If you want to calculate a Julian date with this method, it's astonishingly simple.
To adjust for a Julian date, the only modification you need to do concerns the century. Instead of using the 0-2-4-5 “No tuna for Friday” approach, simply add the century number itself, casting out 7s if needed.
For example, if someone asked about 1066, you would start with 66 just as before. 66 = 64 + 2, which is modified and solved as 32 - 2 = 30. Casting out 7s from 30 leaves 2. For the Julian calendar, we just add the century number (10): 2 + 10 = 12. Casting out 7s from 12 leaves us with 5.
Next, if someone asks for the month of October, we recall that October is associated with Thursday (“Jan’s and October’s 31...“), so we can create a Julian calendar month for October 1066, by starting with a Thursday the 5th. Yes, Wolfram|Alpha can verify Julian calendar dates, too.
As long as you don't accept dates in 1582 from October 5th to October 14th, you can now handle dates as far back as 1 A.D., and as far into the future as you like!
To practice creating calendar for Julian years and months from 1 A.D. to 1582 A.D., you can go the the Quick Calendar Month Creation Quiz page, and click on the Calendar Quiz button. When the Calendar Quiz opens, select Julian: 1 AD - 1582 AD from the top-right menu to practice these dates.
B.C. Dates
The Julian calendar was first put in place by Julius Caesar in 45 B.C., so wouldn't it be nice to handle those first few dates, as well?
The first thing to remember is that the calendar we use today went from 1 B.C. directly to 1 A.D., so there was no year 0.
Handling dates such as these only requires one additional adjustment. When given a B.C. date, subtract the year from 57, and then treat it like any other Julian year.
If you're asked to create a calendar in, say, 25 B.C., the first step is to subtract it from 57: 57 - 25 = 32. Now you can handle that as a Julian year. 32 doesn't need to be partitioned, so we just do 32 ÷ 2 = 16, and casting out 7s from 16 leaves 2. Since the century is effectively 0, we just leave the key at 2.
Now, if someone asks for June, we recall that June is associated with a Monday (“June day? Moon-day!”), so June in 25 B.C. must've had a Monday the 2nd, which Wolfram|Alpha verifies for us!
When dealing with Julian and/or B.C. dates, don't forget the leap year adjustment, if needed. For example, since 25 B.C. is a leap year, you'd still work out that the key number for it is 2, but if someone asked for January, you need to recall that it's associated with Wednesday in a leap year. That way, you know January in 25 B.C. had a Wednesday the 2nd.
You can also practice BC dates by going to the Quick Calendar Month Creation Quiz page, and clicking on the Calendar Quiz button. After the Calendar Quiz opens, select Julian: 45 BC - 1 BC from the top-right menu to practice these dates.
Once you're comfortable working with Julian calendars and calendars back to 45 B.C., check out these tips to help your performance along.
Credits & Notes
The Julian and B.C. date modifications both come directly from the approach used in Bob Goddard's First Sunday Doomsday Algorithm. Basing the Gregorian century modifications on the 1700s is what makes this possible.
Tips
• Only the last 4 digits of a year matter. Working with the year 15739562 is no different than working with the year 9562.
• Another way to think about the centuries, especially when dealing with years far away from the 1700s, 1800s, 1900s, and 2000s:
If it's divisible by 400, like the 2000s, add 5.
If it's even, but not evenly divisible by 400, like the 1800s, add 2.
If it's just after a multiple of 400, like the 1700s, don't add anything.
If it's just before a multiple of 400, like the 1900s, add 4.
• If you're comfortable with using the Mnemonic Major System, you can speed up this feat even further. Since dates within a century repeat in a pattern every 28 years, all you have to do is memorize the 2-digit year keys for the years 00-27. You not only have quick recall of those years, but of all the other years as well:
Years ending in 28 through 55: Subtract 28, then recall the corresponding key.
Years ending in 56 through 83: Subtract 56, then recall the corresponding key.
Years ending in 84 through 99: Subtract 84, then recall the corresponding key.
• You don't have to use the business card calendar PDF. Any calendar that allows you to display an entire month AND adjust which days go with which months will work. Some examples:
• There's a simple formula for working out how many days needed to be added to the Julian calendar to adjust it to the Gregorian calendar. You can learn it from my Changing Calendars Mentally post. When working with the Julian calendar, this can make an impressive additional detail.
• Which months have a Friday the 13th? Once you've taken the year and century into account and you have your key number, ask yourself which day of the week is that many days after Saturday, and then ask yourself which months are associated with that day of the week. The months will be the only ones with a Friday the 13th!
For example, if you're given a year whose key you calculate as 3, ask yourself what day is 3 days after Saturday? The answer, of course, is Tuesday. Which months are associated with Tuesday? September and December are associated with that day, so you can state that September and December are the only months that year with a Friday the 13th!
This makes an impressive extra feat to add in to your performance. Note: As always, don't forget that January's and February's associations move 1 day back in leap years!
• What do you do if someone asks for a year before 45 B.C.? Explain that the pre-Julian Roman calendars were just too different. For example, one version had only 304 days each year, and another had 355 days each year, with leap years making some years as long as 378 days. The last pre-Julian year, 46 B.C., was adjusted to 445 days long!
• Hans-Christian Solka, author of the Day of Week Calculation Blog, also suggests the following related online resources for further reading:
Before you learn how to determine the moon's phases in your head, you'll need to learn and remember some important moon-related terms, so you can describe each phase.
Moon's Age
No, this doesn't refer to the moon being roughly 4.5 billion years old. It refers to the number of days since the most recent new moon, and ranges from 0 to about 30 days.
Over at cycletourist.com, there's a Moon age calculator that uses the date and your time zone to give the exact moon age. Try a few different dates on that calculator, so you can get an idea of the different ages.
Northern vs. Southern Hemisphere
The way the moon proceeds through its phases appears differently, depending on whether you're in the Northern or Southern hemisphere. Here's how the moon's cycle appears in the Northern Hemisphere:
To make this progression more memorable, we'll just focus on stages 3, 5, and 8 from the above diagram:
These phases look like the word “DOC”. The D is shaped like the half-moon, the O is shaped like the full moon, and the C is shaped like the crescent.
In the Southern hemisphere, however, the moon starts its phases from the left side, and proceeds until it disappears on the right, like this:
Looking at only stages 2, 5, and 7 from this picture, we get:
You've probably already guessed that this progression can be remembered using the word COD.
To remember that DOC applies to the Northern hemisphere, use the fact that north is up on most maps, and just think of Bugs Bunny's classic saying, "What's up, Doc?" For COD, recall that a codpiece is designed to be worn in a man's “Southern hemisphere.”
Events vs. Phases
In the 8 stages shown the above pictures, 4 are really only short one-day events, while the rest are part of phases, each of which are about a week long. I'll start with the events, since most people learn about those first.
Events: New Moon vs. Full Moon
The full moon is simple enough to remember. It's when the moon is as full, or as brightly lit as its going to get in its current cycle, as shown in stage 5 of either of the above graphics.
Some people confuse the concept of a new moon with a full moon. If you think of NEW MOON as being the same as NO MOON, that shouldn't be a problem. Image 1 in either of the above graphics shows a new moon.
Events: First Quarter vs. Third Quarter Moon
The term quarter moon often seems confusing, especially since they refer to what most people think of as a half-moon. Quarter, in this context, refers to the point in the cycle, not the shape itself.
Think of quarters in a moon cycle like quarters of a year. After a new year, just like a new moon, comes the first quarter of the year. The third quarter of a year is the last to happen before you head into the next new year, so a third quarter moon is the last you see before the next new moon.
In the above graphics, stage 3 is the first quarter moon, and stage 7 is the third quarter moon. Since the side that's lit in either quarter varies with the hemisphere from which it's being viewed, the DOC and COD mnemonics are very important.
Phases: Waxing vs. Waning
Waxing is the phase in which the moon is getting brighter from night to night. Stages 2, 3, and 4 all show a moon in a waxing phase. If you think of either ear-wax buildup, or waxy yellow buildup on a linoleum floor, this will help you remember that waxing refers to the moon building up, or getting brighter.
Waning, on the other hand, refers to the moon getting darker and darker, as in stages 6, 7, and 8 above. To help you remember this, you can think of Lil Wayne, and use Lil to help you recall that waning means that the light on the moon is getting “littler.” Alternatively, you might think of Wayne Newton as “Wayne Newmoon,” to remind you that waning moons are getting closer and closer to the new moon.
Phases: Crescent vs. Gibbous
The crescent phase is easy to recall, especially if you're familiar with crescent rolls or crescent wrenches. When use to refer to the moon, crescent simply means that less than half the moon is illuminated, as in stages 8, 1, and 2 above.
Gibbous, on the other hand, is an unfamiliar word to many. It refers to the phase when half or more of the moon is illuminated, as in stages 3, 4, 5, 6, and 7 above. Imagine you're trying to negotiate a deal with someone, and they keep saying, “Gibbous more than half or no deal!”
Quick Notes Before Moving On
Waxing or waning, when combined with crescent or gibbous, identifies the exact phase of the moon at any given time. The waxing crescent phase runs from the new moon to the first quarter moon. Next comes the waxing gibbous moon, which runs from the first quarter moon to the full moon.
After that, there's the waning gibbous, from the full moon to the third quarter moon, and the waning crescent, from the third quarter moon back to the new moon.
The line dividing the dark part of the moon from the light part is known as the terminator, which can make from some amusing sci-fi movie references.
Once you're comfortable with all the terms, you're ready to start learning the formula for determining the moon phase in your head, which begins in the next tab.
Just like the day of the week feat, you're going to translate the year, date (1-31), century, and year into key numbers, and add them up.
In the day of the week feat, you then performed a modulo 7 operation on this number, but for the roughly 30-day moon cycle, you're going to perform modulo 30. For a clear explanation of this concept, read BetterExplained's post, Fun With Modular Arithmetic.
Unlike the calendar formula, the moon phase formula is not exact. The real moon's orbit is quite complex, and exact calculations include corrections for every detail.
The formula you'll learn is simple enough to do in your head, and will be accurate to within about a day or so. You'll learn how to deal with this in the Presentation section of the Practice tab.
Month Key
The month key is very easy to remember. For March to December, you simply use the standard month number. For example, March's key number is 3, April's key number is 4, and so on up to December's key number, which is 12.
January and February aren't hard to recall. January's key number is 3, and February's key number is 4. If you can remember the month numbers, and these numbers for the first two months, you'll be able to remember the year key without trouble.
Date Key
If you thought the month key was easy, the date key is even easier! All you use is the number of the date, 1 through 31, as the date key.
So, your first step is to look at the month and date, and add the two keys together. July 12th? That's 7 + 12 = 19. February 7th? That's 4 + 7 = 11. January 19th? That's 3 + 19 = 22.
Once you have a total of the month and day keys, it's usually a good idea to add 30 to the total before working with the century and year keys. With those, you're going to be doing some subtracting, and adding 30 will help ensure that the running total remains a positive number.
When performing this feat, it's best to use a date that's meaningful, such as a birthday or anniversary. So, all you really need to know are keys for the 1900s and 2000s.
If you're given a date in the 1900s, you simply subtract 4 from the running total. If you're given a date in the 2000s, you'll subtract 8 instead. That's it!
Year Key
Unlike the previous keys, there are a few steps to working out the year key. Don't worry, they can still be done in your head.
First, you'll need to know your multiples of 19: 0, 19, 38, 57, 76, and 95.
This is because the first step in working out the year key is finding out how close you are to the nearest multiple of 19, whether that multiple is more or less than the year number.
Let's use 2008 as an example, focusing on the year as simply 8. 8 is between 0 and 19, but obviously closer to 0. To get 8 starting from 0, you'd add 0 + 8, so think only of +8 at this stage.
It's important to note that this was just the first step to finding 2008's key, so +8 is the number we need for the 2nd step, not the key itself.
Before moving on to the 2nd step, though, let's try another example year, such as 2013. Working with 13, we see that it's between 0 and 19, but closer to 19. Starting from 19, you'd get to 13 by working our 19 - 6, so 13's number at this stage should be thought of as -6.
This is what I mean about finding the distance to the closest multiple of 19. 8 was 8 MORE than 0, so we think of it as PLUS 8. 13 was 6 LESS than 19, so we think of 13 as MINUS 6.
What about, say, 59? That would be +2, because 57 is the nearest multiple of 19, and 57 + 2 = 59. How about 88? That becomes -7. Do you understand why?
This first step seems complicated when you first learn about it, but with practice, it's not hard to determine the number you need quickly and easily. Regardless of the year, notice that you'll always get a number ranging from -9 to +9 as a result.
Now it's time to explain the second step. First, I'll explain the long way of working through this step, then I'll give you a shortcut that gives the same result.
LONG VERSION: Whether the number you got in the first step was positive or negative, you're going to work through this step as if it's a positive number, and then restore the proper sign at the end.
Going back to our 2008 example, we got +8 in the first step, and we remember that plus sign. Next, we multiply that by 11, and get 88. Finally, 88 modulo 30 is 28, and we recall that the original number was positive, so we know that the 28 is positive, as well. This is the final number we need, so we would add 28 to our running total.
What happens with 2013? Applying the first step to 13 gave us -6 (remember?), so we'll work through this process as just 6, remembering the minus sign for later. 6 × 11 = 66, and 66 modulo 30 is 6. Restoring the minus sign from earlier, we get -6. This simply means that we'd subtract 6 from our running total.
The first thing to notice here is that neither the sign nor the ones digit (the rightmost digit) change. It's also easy to see that the multiples of 3 don't change at all.
Just like in the long version, when you do the modulo 3, you need to temporarily drop the sign, then put it back afterwards, for this formula to work out properly.
Let's work through the shortcut method with a year ending in 53, as an example. In the first step, 53 becomes -4 (you should still understand why). We temporarily drop the minus sign, and perform 4 mod 3 to get 1, so we put a 1 in there to make -14. This means we need to subtract 14 from our running total.
The more you practice and perform this feat, the more quickly these numbers will come to mind. Sooner than you expect, the numbers will come from your memory quicker than from calculation.
Now that you understand how to find each key, it's time to put them all together and get the final result!
Simply add each of the key numbers together. If the number is greater than 30, take the answer modulo 30.
You'll wind up with a number ranging from 0 to 29, and this number represents the moon's age in days. The chart below shows how the age corresponds to the four major moon events:
To see how this works, let's work through an example date, April 1, 1994. April gives 4 and the 1st gives 1, so that's a total of 5. As I mentioned earlier, I usually like to add 30 at this point to prevent working with negative numbers, so 5 + 30 = 35.
The key for the 1900s is -4, so 35 - 4 = 31. Running 94 through the year key process gives us -11 (Because 95 - 1 = 94, and -1 * 11 = -11, remember?), so 31 - 11 = 20. In this case, 20 is below 30, so we don't have to work out modulo 30.
Our estimate, then, is that the moon's age is about 20 days old (±1 day). That suggests we're about 5 days past the full moon, so the moon would be getting darker each night, but a majority of the moon's face would still be illuminated. In technical terms, the moon was in a waning gibbous phase.
If you enter April 1, 1994 into Wolfram|Alpha and scroll down to the bottom, you can verify that the moon was indeed in a waning gibbous phase on that date!
As a matter of fact, Wolfram|Alpha provides a great way to test your ability to work through this formula. Get a random date in the 20th or 21st century by clicking here, then use that date to work through the moon phase formula in your head. When you've got an answer, click the date in the Result pod, and scroll to the bottom to see the actual phase of the moon on that date.
Presentation
Due to the margin of error of this formula, you can't pin down the specific events (new moon, 1st quarter moon, full moon, and 3rd quarter moon) exactly. Since the various phases are each about a week long, and overlap, you can use this to your advantage.
I often use the difference in technical terms used by astronomers (and on Wolfram|Alpha) in contrast with an informal description of the moon, including how average people would refer to the moon. This is usually enough to be considered accurate and verifiable with an outside source.
Don't forget the importance of having an outside source to verify your claim on the moon. If you can't have someone look up the date online, at least carry an almanac or other reference in which your audience can verify the moon phase.
If you make sure people verify your claim with Wolfram|Alpha, you can also usually state your estimate of the moon's age openly. Wolfram|Alpha, at least at this writing, doesn't include this information.
Here's a general idea of what to say for any given estimate:
29, 0, or 1 days:“On that date, the moon was in a crescent phase, as the 3rd quarter moon happened about a week ago. In fact, so little of the moon is lit, most people would just refer to it as a new moon.”
2 through 6 days:“The new moon was just about (2, 3, 4, 5, or 6) days before that date, so it's getting brighter every night, but less than half the moon is lit on the right side. It's known as the waxing crescent phase.”
7 or 8 days:“It's been about a week since the new moon, so the moon is in a waxing phase - in other words, getting brighter each night. The average person might even call it a half-moon or 1st-quarter moon.”
9 through 13 days:“The 1st quarter moon was about (1 to 2, 2 to 3, 3 to 4, 4 to 5, or 5 to 6) days before that date, so the moon is getting brighter each night, and more than half is lit. This is called the waxing gibbous phase.”
14, 15, or 16 days:“It's been just about a week since the 1st quarter moon, and so much of the moon is lit that most people will say things like, 'All the crazies are out, it's a full moon!'”
17 through 21 days:“Just about (1 to 2, 2 to 3, 3 to 4, or 4 to 5) days before that was the full moon, so most of the moon is still illuminated, but it's getting darker each night. The technical term for this is the waning gibbous phase.”
22 or 23 days:“On that date, it's been about a week since the full moon, so around half of the left side of the moon is illuminated. Many would refer to this as a half-moon or 3rd quarter moon.”
24 through 28 days:“Around (1 to 2, 2 to 3, 3 to 4, 4 to 5, or 5 to 6) days before that date was the 3rd quarter moon, so the moon is getting darker each night, and less than half the moon is illuminated. Astronomer call this the waning crescent phase.”
The above statements were written with the assumption that the speaker was talking about the Northern hemisphere. For the Southern hemisphere, just switch all the references to left and right.
Once you've practiced both the formula itself and the presentation of your moon estimate, you're ready to perform this feat for someone else.
• Be alert to key number combinations that total 0, to cut your calculation time:
Years ending in a multiple of 19 (00, 19, 38, 57, 76, 95): 0 (year) = 0
The 4th of any month in the 1900s: +4 (date) -4 (1900s) = 0
The 8th of any month in the 2000s: +8 (date) -8 (2000s) = 0
Any February or April in the 1900s: +4 (month) -4 (1900s) = 0
Any August in the 2000s: +8 (month) -8 (2000s) = 0
Use the Moon Century/Year Combo Key Generator widget below to find years that require no adjustment, such as 2017.
• Instead of trying to describe the moon, you might try having images of 10 to 12 different phases of the moon on your mobile device. Since you have less than half of the 30 different images, it's reasonable that the photo you're showing isn't an exact match to the moon. There's even a moon-shaped toy called Moon In My Room that can display 12 different phases of the moon, which is perfect for this presentation.
• If you want to work with other centuries, I've developed a Wolfram|Alpha widget that will work out the century keys for anywhere from the 1800s to the 3000s. Simply select the century from the menu, then click Submit.
The widget will return 2 numbers, r and s. You can use either one of these as the key for your chosen century, but one if often simpler to use than the other. For example, selecting the 2900s will return r = 6 and s = -24. This means that you can use either +6 or -24 as your key for the 2900s. In this example, adding 6 is the easier way to go.
• Should you decide to focus on a single year, you can use the widget below to get a single key that compensates for the century and the year at once. If you can memorize several keys, you can use the widget to lookup each year and that range.
Use the pull-down menus to select any year from 1800 to 3000, and then click the Submit button. Like the above widget, you're given two choices, r and s, and you can decide which one is easier to use. You can ignore the variable y, as it's simply the last two digits of the year you input.
• If you have Day One, and you're familiar with the Major/Peg System for memorizing numbers, it's not hard to develop a Day One-type approach to the Moon Phase feat.
Use the Moon Century/Year Combo Key Generator widget above to determine keys for each of the years 2000 through 2018, and link from the same year images I use in Day One to the Major/Peg images for each adjustment (I add a spear to the peg image to represent the minus sign, so I know when to subtract). The years 2019 - 2099 can be reduced to these images via modulo 19, similar to the use of modulo 28 in Day One. To adjust for a date the 1900s, simply add 4.
I jokingly refer to this approach as Night One.
• Here are some excellent videos that can provide helpful background knowledge:
While some may prefer to use the formula without understanding how it works, I've added this section for people who want to understand it better.
Why Modulo 30? If you were to draw a line from the center of the Earth through the center of the moon to a fixed star, it would take about 27.321661 days (27 days, 7 hours, 43 minutes, 11.5 seconds, known as a sidereal month) to do so. However, it wouldn't return to the exact same phase, since the Earth has moved roughly 45 million miles in that same amount of time.
To return to the same moon phase takes longer, an average of about 29.530588 days (29 days, 12 hours, 44 minutes, 2.9 seconds, known as a synodic month).
The phase of the moon in this 29.530588-day cycle is approximated through the use of modulo 30, which is easier to work with mentally.
Why Focus on Groups of 19 Years? When estimating moon phases, it's useful to know how long it will be before the moon will be in the same phase on the same date. As it turns out, the answer is surprisingly accurate 19 years and 2 hours! This is known as the Metonic cycle.
Over a 100-year period, the difference is only about 10.5 hours, so working in groups of 19 works very well in the estimate. The estimation itself adjusts every 100 years, and takes this into account.
Why Multiply the Years by 11? If the moon revolved exactly 12 times around the Earth every tropical year (the technical name for the year as most people know it), we wouldn't need to make any adjustment for years, as the moon would always be in the same phase on the same day.
If it took the moon, say, 2 days short of a tropical year to revolve exactly 12 times, then we'd have to adjust by adding 2 days for every year. This would be done simply by multiplying by 2.
Why Do Years Range From -9 to 9, Instead of 0 to 18? Over the 19-year Metonic cycle, that small 1⁄8 day difference can start to add up. 18 years × 1⁄8 day = 18⁄8 day = 21⁄8 days!
When originally developing the moon phase estimation formula, John Conway worked out the rather ingenious -9 to +9 approach you learned. What's so ingenious about it?
Go to this Desmos Calculator page, and drag the slider back and forth from 0 to 18 years. The red bar shows the 1⁄8-day difference building up over that time. The numbers on the y-axis (the vertical axis) represent the total number of days in error.
By comparison the blue bar, representing the -9 to +9 approach, through the first 9 years, and then begins to decrease dramatically as you add more years!
Hopefully, this gives you a better idea of how this reduces the margin of error.
What is the Margin of Error?
With all the approximate numbers we're using, the resulting accuracy of this estimation is quite amazing.
To give you an idea of just how accurate it can be, I've provided the following calendar. You can use the drop-down menus to select any month from January 1900 to December 2099, and the calendar will instantly change.
Est., in each day, represents the estimated age of the moon you get by working through the method in this tutorial.
As I've modified them, these calculations reflect the phase of the moon as it appears at noon in Greenwich Mean Time on each given day. The moon's age in degrees, from 0° to 359°, is converted into its age in days in the synodic month, and then rounded to the nearest whole number.
Somebody has mixed up the calendar! Can you help straighten it out?
Just like the 15 puzzle, the object of this puzzle is to return all the pieces to their correct order. Instead of 15 pieces to move around, however, the puzzle below has 41 pieces. In addition, the calendar aspect adds an extra dimension to the challenge!
Note the year and month above the calendar puzzle itself. Your challenge is to arrange the pieces so that they form the correct calendar for that month and year. The following rules apply:
• The blue square marked with a “1” will always be located somewhere in the top row. This should make sense. You don't start any calendar on the 2nd week.
• The blue numbered tiles are arranged on their corresponding days in the month. If the 1st fell on a Tuesday in the given month, then the blue “1” needs to be placed in the Tuesday column, and so on.
• The blue numbered tiles must be arranged in numerical order, reading from left to right, then top to bottom.
• The white lettered tiles are used to fill in the remaining spaces not used by the dates, and must be arrange in alphabetical order, reading from left to right, then top to bottom.
For example, if you're given a 30-day month that begins on Tuesday, the goal would be to arrange the puzzle like this:
Note, in this example, that A and B are used to fill the first Sunday and Monday respectively, since the month begins on a Tuesday. After the month ends, the remaining lettered pieces, C through K, are used to fill the remaining spaces.
The Year Range selection menus can be used to choose the range of years for your next puzzle (the default range is 2000 to 2099), and the New Puzzle button will generate a new month and year, as well as shuffle the pieces.
To apply them to this particular puzzle, consider the year and month you're given, along with the first of the month. Once you figure out on which day of the week the first day falls, you also know which lettered tiles (if any) come before it.
Once you get the first lettered and numbered tiles in place, it's mostly like solving the original 15 puzzle.
When you get to the last two rows, the tricky part becomes working out which letter goes below which other piece. If you take into account where the letters in the top row (again, if any) left off, and how many days are in the month, this isn't too difficult.
Yes, you could cheat and look up the given month calendar online, but it's far more impressive to do it without looking.
This tutorial is meant as a sequel to my unit circle tutorial. Much of this tutorial assumes you have already been through that one, so if haven't already done so, please go through the previous tutorial now.
As a matter of fact, all the measurements on the unit circle work this way. Sine and cosine can be applied to real world measurements with a single multiplication to scale them up or down.
That's why, in the unit circle, all the trigonometric functions do a sort of double duty. They're ratios, as you've been taught, but in the idealized world of the unit circle, they can also be treated as absolutes, since they result in useful coordinates. For example, sine and cosine are both ratios that relate in different ways to the length of the hypotenuse, but also give exact coordinates to graph.
In this tutorial, we'll go beyond sine and cosine, and explore the other trigonometric functions of tangent, cotangent, secant, and cosecant. The goal of this tutorial is to explain them in a way that can be easily understood, comprehended, and remembered.
To explain these, we're going to scale the triangle up beyond the bounds of the unit circle. In the previous tutorial, the hypotenuse always had a length of 1 because it was representing the radius of the unit circle.
The hypotenuse of a right triangle is, by definition, the longest of the three sides, so this meant that, with any given angle, the other two side must always fall on or inside the unit circle itself, because they have to be 1 (in the case of 90 degree or π/2 angles) or less.
What happens, though, if we scale up a right triangle in the unit circle so that the width or height is 1 instead of the hypotenuse? In this tutorial, we'll try that out, and examine the useful measurements that result.
“Change in elevation” over “change in horizontal distance” is usually stated in a way that's catchier and easier to remember:
For example, take a 45 degree angle in the unit circle. When the hypotenuse was 1 (the radius of the unit circle), what was the rise and what was the run? If you remember the hand trick from the video, it shouldn't be too hard to recall:
So, for a 45-degree angle the slope is 1. In other words, for every 1 unit you move horizontally, you're going to move 1 unit vertically, as well. This makes sense for a 45-degree right triangle, since the two shorter sides are the same length.
What about that 30-degree angle I asked about earlier? Lets work through that problem:
The number is less exact than in our 45-degree answer, but the meaning remains the same. With a 30-degree angle, for every 1 unit you run (move horizontally), you're going to rise (move vertically) 0.57735... units. That's more than half a unit, but less than a slope of 0.60.
Ever been told by someone that you're going off on a tangent? Ultimately, the complaint is that you're going off on a line that will only take the discussion farther and farther from the main point.
It's almost exactly the same definition in math. One mathematical definition of tangent is a line that touches a circle (a unit circle, for our purposes) at only one point. Starting at that one point, and then traveling along the tangent would only take you farther and farther away from that main point.
As you can see, a tangent is easy to draw. Draw a radius line at any angle, and at the point where it touches the circle, draw a line perpendicular (at a 90° angle) to the radius, and that's the tangent.
To keep things simple in this tutorial, we're only going to consider the horizontal and vertical tangents. In this particular section, we'll only focus on the vertical tangent.
In the previous tutorial, we always created the length of the hypotenuse (the longest side) to be 1. Imagine that, instead of the hypotenuse having length 1, we had the base have a length of 1.
Here's a picture of the situation:
Here's a question for you: If we tried scaling up a 45° right triangle in this manner, what would the coordinates be where the hypotenuse of this new larger triangle intersects the tangent line?
Let's think about this. Because the tangent is defined as x=1, the x coordinate where the hypotenuse intersects the tangent will also obviously be x=1. So, the coordinates we have so far are (1, something). That's half the work already done!
In the previous section, we worked out that the slope of a 45° angle was 1, at that meant that for every 1 unit you moved horizontally, you moved 1 unit vertically. As it happens, a base width of 1 means we are moving over 1 unit horizontally, so we should obviously move up 1 unit vertically!
That means, for every 1 unit we travel horizontally, we travel 0.57735... units vertically. Yep, the coordinates where the hypotenuse meets the tangent line is (1, 0.57735...).
See the pattern? Whenever a right triangle with a given angle has a base length of 1 in the unit circle, the coordinates where the hypotenuse will intersect the vertical tangent line is (1, slope of that angle)!
As with sine and cosine in the unit circle, the slope is doing double duty. It's both the slope itself, and the y coordinate where the hypotenuse intersects the tangent at x=1. In fact, we could just call this number the tangent.
Remember SOHCAHTOA? That helps remind us, among other things that the formula for the tangent is the opposite side's length over the adjacent side's length. Also, remember that those lengths on the unit circle are worked out by figuring sine (the height) and the cosine (the width)
Let's take a closer look at the formulas for slope and tangent:
With a horizontal tangent, we're simply dealing with what happens when y=1, instead of x=1.
Remember how we determined the tangent (the same as the slope) of a 45° angle was 1, so the coordinates where the hypotenuse met the vertical tangent was (1,1)? For a 45°, this works out nicely, since it intersects the horizontal tangent in the same place.
Let's go back to our 30° example, and get a better idea of what changes. The slope is 0.57735..., so let's view what happens when we try this out:
Here's another look at the same situation, via Wolfram|Alpha (albeit slightly distorted). Even without knowing the exact coordinates, we can see that the hypotenuse, in the 30° case, is MUCH longer than where it intersected the vertical tangent.
So, what are the coordinates? We start in a manner similar to before, with the coordinates (something, 1), because the horizontal tangent formula is y=1 (instead of x=1 for the vertical tangent).
In the case of the vertical tangent, we just multiplied 1 (the x coordinate) times the slope to get the y coordinate. Since we're still talking about the same 30° angle, the slope is still the same 0.57735... we used before. We'll just have to come at it from the other way:
You can actually see several important things here. First, we found the coordinates for where our 30° angle intersects the horizontal tangent at (1.73205...,1). Second, while the tangent relationship of y = x × 0.57735... still holds, it would seem easier just to state the relationship the other way around, as in x = y × 1.73205..., especially when that lets you multiply by 1.
Finally, note that you can always find where the hypotenuse intersects the horizontal tangent by dividing 1 by the tangent of the same angle. That's why this number is given the name cotangent.
If you think about it, there are several ways to find the cotangent:
We keep talking about the slope of the hypotenuse and the coordinates where the hypotenuse intersects the tangent and cotangent lines, but what happens to the length of the hypotenuse as triangles are scaled up to meet the tangent and cotangent lines? That's discussed in the next section.
Hypotenuse Length
We're going to switch from focusing on the slope of the hypotenuse to its length.
This almost means a return to our old friend from the right-triangle, the Pythagorean Theorem: a2 + b2 = c2.
When the hypotenuse is 1 unit long, the squares of the other two sides must add up to 1. For example, in our 30° angle, where the cosine (width) is 0.8660... and the sine (height) is 0.5, we get:
However, the hypotenuse must get longer to meet the tangent lines as we've seen. Let's start as before, scaling up to meet the tangent line (that's the vertical tangent). We've already seen that the coordinates of a 30° angle gives us coordinates of (1, 0.57735...). As in the above example, we note that the coordinates are also the lengths of the two smaller sides. That being the case, let's figure out how long the hypotenuse is when it meets the vertical tangent:
So, we see that the hypotenuse is now 1.1547... units long. Since the original hypotenuse was only 1 unit long, this is a sort of scaling factor for the hypotenuse.
Take another look, though. When we scaled the width of a 30° right triangle from 0.866... to 1, that's a factor of 1.1547... times, as well. The height went from 0.5 units to 0.57735... units, which is also a scaling factor of 1.1547... times!
So, this one factor, when scaling a right triangle up to meet the tangent line tells us how to scale the lengths of all the sides in order to do so! Since we're scaling up from a hypotenuse of 1, this factor also gives us the exact length of the hypotenuse after being scaled up to the tangent line.
In trigonometry, this factor is called the secant. Since we scale the width of the triangle (the cosine, in the unit triangle) up to 1, it shouldn't be surprising that we can find the secant in this manner:
The hypotenuse length was 1 before scaling it up, so you could also work it out by dividing the hypotenuse by the width:
Meeting the Cotangent Line
Let's not forget the length of hypotenuse when scaled up to meet the cotangent line (that's the horizontal tangent line).
Our example 30° right triangle, when scaled up to meet the cotangent line, we found the coordinates (and side lengths) of (1.73205...,1). Let's run the numbers just as before, and find the hypotenuse length:
Oh! The hypotenuse in this case is exactly 2 units long. We took the height (the sine) from 0.5 to 1, which is a factor of exactly 2 units, so we shouldn't be surprised when the hypotenuse (and the width, for that matter) scales up by a factor of 2 units.
Since this scaling factor deals with scaling the height up to the cotangent line, this factor is naturally called the cosecant. The formulas for cosecant are as follows:
At this point, you should understand all 6 trigonometric functions: sine, cosine, tangent, cotangent, secant and cosecant. Once understood, however, they can be easily confused. In our final section, I'll offer some mnemonics to help keep them all straight.
Mnemonics
When you have the explanations and formulas in front of you, it's much easier to keep these things straight. How is it possible to keep all this straight in your head?
Let's break things up several ways. First, by the pairs in which you learned them.
Sine and cosine can be thought of as the simple lengths of the sides in the unit circle, when the hypotenuse is 1. Think of signs that tell you the length of a road.
Tangent and cotangent deal with the unit coordinates when the triangles are scaled up to meet the tangent and the cotangent lines respectively. This is its own mnemonic: tangents deal with tangents.
Secant and cosecant are the scaling factors to the tangent and cotangent lines respectively. Think secant means “secaling” factor.
Once you get the functions themselves straight, it can seem tricky to remember which of each pair deals with horizontal information and which ones deal with vertical information. However, there's an almost built-in mnemonic: look for the “o”.
Sine is the vertical length of the triangle. Tangent deals with coordinates when scaled up to the vertical tangent line. Secant deals with scaling factors up to the vertical tangent.
On the other hand, cosine is the horizontal length of the triangle. Cotangent deals with coordinates when scaled up to the horizontal tangent line. Cosecant deals with scaling factors up to the horizontal tangent.
Do you see what I mean by “looking for the o”? The words horizontal, cosine, cotangent, and cosecant all have an o in them, so they all go together.
Similarly, none of the words vertical, sine, tangent, and secant feature an o in them, so they all go together.
Looking for the letter o is an appropriate mnemonic for a circle, don't you think?
Another challenge is remembering the respective formulas. As mentioned before, SOHCAHTOA is the classic way to remember this, but that only gives 3 formulas.
Here's a lesser known yet more amusing mnemonic that gives all 6 formulas. Start by writing down the dividing lines, and writing “OOH AAH” above them:
“OOH AAH” is, no doubt, the sound you made upon first learning about unit circles. On the bottom, you're going to write “OOH AAH” again, but this time from right to left (as in “HAA HOO”):
From left to right, these are the formulas for the three vertical functions, sine (O/H), tangent (O/A), secant (H/A), followed by the three horizontal functions in the same order, cosine (A/H), cotangent (A/O), and cosecant (H/O).
You've also noticed that you can get some of these ratios simply by taking the inverse (dividing 1 by another number) of other factors. Tangent and cotangent are easy:
The ones beginning with t's are easy, so they can be thought of together. The ones featuring c's and s's are a little trickier:
When focusing on secant or cosecant, ignore the first s or c you see, and look for the second s or c. In the case of secant, you'd look at it as seCant. This C lets you know that it's the inverse of the function that begins with C - cosine!
With cosecant, you'd look at it as coSecant, and remember that this S it telling you that it's the inverse of the other S - sine!
Posts
