Page:Notes and Queries - Series 11 - Volume 5.djvu/515

 US. V. JUKE 1,1912.]

NOTES AND QUERIES.

423

adequate in this matter, it is immaterial for what year (since the New Style) the dates are taken* The values given in the table have

been taken from ' Whitaker's Almanack '. for 1909, which was to hand at the time of writing.

Sign.

Rune

Month

Possible Dates.

Present Date,

Aquarius

p,

January

2, 9, 16, 23, 30

20

Pisces



February

1, 8, 15, 22

19

Aries

\>

March

3, 10, 17, 24, 31

21

Taurus

f

April

2, 9, 16, 23, 30

20

Gemini

May

3, 10, 17, 24, 31

21

Cancer

K

June

4, 11, 18, 25

22

Leo

p

July

5, 12, 19, 26

23

Virgo



August

5, 12, 19, 26

23

Libra

f>

September

5, 12, 19, 26

23

Scorpio

fc

October

5, 12, 19, 26

24

Sagittarius

1

November

4, 11, 18, 25

22

Capricornus

n

December

4, 11, 18, 25

22

Owing to the over-correction caused by making every fourth year a leap year, under the Old Style calendar the sun entered the signs before the correct date, so that all the values in the column" headed ' Possible Dates ' that are greater than the corre- sponding ' Present Date ' may be discarded. This eliminates all but the first three vertical columns of ' Possible Dates.' Subtract these three columns from the corresponding ' Present Date,' and the anticipations will be 18, 11, and 4 days respectively. In the twentieth century the Old Style calendar is only 13 days in error ; so that the first column, which gives an anticipation of 18 days, may also be rejected. The choice is now betwe?n the centuries in which the error of the calendar was 11 and 4. The error was 11 days during the period 1700-99, and 4 days during the period 700-899. The latter period is obviously out of the question, so that the probable date, from this information, is 1700-99. If the date of the introduction of the New Style be considered, this period can be materially shortened. The change was made hi Nor- way and Denmark in 1700, and in Sweden in 1753. It is very unlikely, therefore, that the calendar is later than 1753 ; and the possible range is now 1700-53. The fact that the original was prior to 1753 is confirmed to some extent by the use of a solar cycle of twenty-eight years.

The numerical values of the Golden Number and the epact enable the date to

be calculated somewhat roughly. This is due to the fact that the errors of the Metonic cycle necessitate an occasional revision of the epacts after one, two, or three centimes, On consulting .a table of epacts it will be found that the epacts calculated corre- sponded to the Golden Numbers assumed from 1500 to 1799, if the Old Style were retained during this period.

The evidence afforded by the saints who are commemorated on the calendar is not of much value. The latest one is St. Bur- git te, who was canonized in 1391. The calendar must therefore be subsequent to this date.

It will be observed that none of these sources of information contradicts any of the others. Taking them all into consideration, it is very probable that the original calendar was made between 1700 and 1753.

Before giving the full arrangement of the calendar, attention may be drawn to the errors which are to be found on it. In the first place, errors arise from the confusion of similar runes. The following groups con- tain runes which are frequently inter- changed :

1 1