University of Virginia Library

CHAPTER IV
Appendix III

ECLIPSES DURING THE REIGN OF EMPEROR WEN

During this period, six eclipses were recorded in the SC and HS.
We discuss them in chronological order.

i. In the second year, the eleventh month, on the day kuei-mao, the
last day of the month, an eclipse was recorded (4: 8b). P. Hoang gives
this date as Jan. 2, 178 B.C., on which Oppolzer calculated his solar
eclipse no. 2447. This was a partial eclipse; calculation shows that at
Ch'ang-an it reached a magnitude of 0.20 at 2:48 p.m., local time. The
longitude of the sun was 279.0° = 278.2° R.A. HS 27 Cb: 13b says
that it was one degree in the constellation Wu-nü, of whose stars ε and μ
Aquarii were then in 281.7° and 283.0° R.A. respectively.

ii. In the SC, after the record of the above eclipse, there is the following
statement (cf. Mh II, 461), "In the twelfth month, on the day of full
moon [the fifteenth of the month], there was an eclipse of the sun."
This date was, according to Hoang, Jan. 17, 178 B.C.; but eclipses of
the sun cannot happen on the day of full moon. They happen only at
new moon, whereas eclipses of the moon happen at full moon. Hsü
Kuang (352-425) says that the HS does not record this eclipse [it is now
neither in chap. 4 nor chap. 27], and tells that one copy [of the SC]
says it was an eclipse of the moon, but that the histories do not record
eclipses of the moon. The conjectural emenation of that copy is correct
—Oppolzer calculated his lunar eclipse no. 1580 on the evening of Jan. 16,
178 B.C. at Ch'ang-an. If the historian misread in the astronomical
records a poorly written [OMITTED] as being [OMITTED], this mistake could easily occur.
This passage seems to show that the court astronomers kept a record of
lunar as well as of solar eclipses. The discrepancy in dates indicates that
Hoang's calendar is a day in error here.

iii. In the third year, the tenth month, on the day ting-yu, the last
day of the month, a third eclipse was recorded (4: 11a). This date was
Dec. 22, 178 B.C., for which Oppolzer calculated his solar eclipse no.
2449. He charts the umbra of the moon as passing through the Malay
peninsula; calculation shows that the eclipse at Ch'ang-an reached a
magnitude of 0.35 at 1:58 p.m., local time. The sun was in long.
267.4° = 267.1° R.A. HS 27 Cb: 13b adds, "It was 23 degrees in
[the constellation] Tou," of whose stars μ, λ, φ, σ, τ and ζ Aquarii were
then in R.A. 241.1°, 243.5°, 247.2°, 249.8°, 253.3° and 250.6°, respectively.

iv. In the third year, the eleventh month, on the day ting-mao, the
last day of the month, a fourth eclipse is listed (4: 11a). P. Hoang gives
this date as Jan. 20, 177 B.C., the first day of the twelfth month. HS
27 Cb: 13b adds, "It was 8 degrees in [the constellation] Hsü," whose
stars, α Equulei and β Aquarii, were then in R.A. 291.4° and 293.4°.
The Han-chi 7: 7b dates this eclipse on "[the day] yi-mao, the last
day of the month," 12 days before the HS's date and on the 18th day
of that month.

There was no eclipse on that date, for Oppolzer gives none. In the
22 years from the preceding eclipse to the next correctly dated eclipse
in 154 B.C., in which period the HS lists 3 eclipses, there occurred 56
eclipses, of which only 10 were visible in China.[1]

In this period, 9 umbral eclipses were visible in China: (1) no. 2452,
on June 6, 176 B.C., 23 days after a ting-mao day, sun in long. 70.7° =
69.3° R.A.; (2) no. 2459, on Oct. 10, 174 B.C., 39 days after such a day,
sun in long. 193.2°; (3) no. 2460, on May 4, 173 B.C., 36 days after such
a day, sun in long. 10.6°; (4) no. 2470, on July 17, 169 B.C., 41 days
after such a day, sun in long. 110.2°; (5) no. 2475, on May 28, 167 B.C.,
1 day after such a day, sun in long. 61.9°; (6) no. 2477, on May 17, 166
B.C., 55 days after such a day, sun in long. 51.4°; (7) no. 2486, on Mar. 5,
162 B.C., 3 days after such a day, sun in long. 340.9°; but the umbral
path of this eclipse ran so far south that I calculated it, and found it
invisible in the present Ch'ang-an, Pei-p'ing, Shantung, and Ch'ang-sha.
In Wu (present Soochow), the eclipse reached a magnitude of only 0.02
at 4:34 p.m. local time, so that it was practically invisible; (8) no. 2489,
on Aug. 16, 161 B.C., 53 days after such a day, sun in long. 140.1° =
142.5° R.A.; the umbral path of this eclipse ran so far north that I
calculated it and found that in Ch'ang-an it reached a magnitude of
0.17 at 5:16 a.m. local time; (9) no. 2505, on Oct. 10, 155 B.C., 19 days
after such a day, sun in long. 193.4°; the umbral path of this eclipse ran
so far south that I calculated it and found that in Ch'ang-an the eclipse
reached a magnitude of 0.17 at 9:32 a.m., local time. Westwards and
southwards its magnitude was greater.

In this period, 11 partial eclipses occurred in the northern hemisphere,
of which only one was visible in ancient China: no. 2481, on Mar. 26,
164 B.C., 13 days after a ting-mao day, sun in long. 165.7°. At Ch'ang-an
the eclipse reached a magnitude of only 0.02 at 7:12 a.m., with visibility

better at places north and west. On the longitude of Ch'ang-an, at latitude
40°, in the Ordos region, the magnitude of the eclipse reached 0.14.[2]

It is probable that the eclipse intended by the text was the first umbral
eclipse listed above, on June 6, 176 B.C. Then the ting [OMITTED]- mao of the
text is an error for hsin [OMITTED]- mao (a natural mistake) and Hoang's calendar
is two days in error; this date was the first day of the fifth month, in
the fourth year. The heavenly location given for this eclipse is then
greatly in error.

v. In the latter part of the reign, the fourth year, the fourth month,
on the day ping-yin, the last day of the month, a fifth eclipse is listed
(4: 17b). The Han-chi 8: 15a has the same reading. But there was
no ping-yin day in that month. HS 27 Cb: 13b dates it on the day
ping-ch'en, and adds, "It was 13 degrees in [the constellation] Tung-ching."
This latter date was June 9, 160 B.C., according to P. Hoang.
The first star in Tung-ching, μ Gemini, was then in 63.4° R.A. But
Oppolzer lists no eclipse on this date.

For the eclipses occurring about this time, cf. the discussion under
the preceding eclipse. As to their cyclical days, ping-yin is the day
before ting-mao, and ping-ch'en is 11 days before ting-mao.

Very possibly the eighth umbral eclipse mentioned previously, on
Aug. 16, 161 B.C., is the eclipse referred to in the text. It occurred on a
keng-shen day, in the third year, the sixth month, the day before the
last day of the month, according to Hoang's calendar. This identification
would require no alteration in the order of events of HS ch. 4,
merely a redating of one event; the cyclical day is already in doubt
because of the difference between the two recordings. The location of
the eclipse in the heavens is about 60° in error.

vi. HS 27 Cb: 13b adds, "In the seventh year, the first month, on
[the day] hsin-wei, the first day of the month, there was an eclipse of
the sun." The Han-chi has the same recording. P. Hoang gives this
date as Feb. 9, 157 B.C. But no eclipse occurred on that date.

For eclipses occurring about this time, cf. sub eclipse iv. Hsin-wei is
4 days after ting-mao. There was no eclipse visible in China in Emperor
Wen's reign after the date we adopted for the preceding eclipse. It
seems that Pan Ku considered this eclipse doubtful, for he did not put
it into his "Annals," although it got into his chapter on the "Five Elements,"

and was copied by the Han-chi in its chronological account.
It looks as though someone considered that an eclipse was due at this
time in order to predict the death of Emperor Wen, so placed an eclipse
five months (about an eclipse season) before his death. Pan Ku's rejection
of this eclipse in his Annals is a good testimony to his historical
judgment.

But the solar eclipse of Oct. 10, 155 B.C., which Hoang dates in Emperor
Ching's second year, the ninth month, the day yi-yu, the last day
of the month, was the only eclipse visible in China after the one identified
for the preceding eclipse and before the next and correctly dated
eclipse. Since Hoang's calendar may be a day in error, this date may
have been the tenth month, the first day, the day ping-hsü. It seems
stretching things to identify this eclipse with the one listed in ch. 27, yet
it is peculiar that, in a period during which only one eclipse was listed,
there should have been only one eclipse visible.