Further support for a Cretaceous age for the feathered-dinosaur beds of Liaoning,China:New 40Ar÷39Ar dating of the Yixian and Tuchengzi Formations

Abstract

We report new 40Ar÷39Ar dating results obtained from total fusion and incremental-heating analyses of sanidine and biotite from three tuffs found
interbedded within the fossil-bearing deposits of Liaoning, northeast China. The first is a new sample of the Bed 6 Sihetun
tuff from the Yixian Formation, previously dated by our team as middle Early Cretaceous, and recently considered by Lo et
al., partially reset due to metamorphism from a nearby basaltic sill. The second is the Yixian Bed 9 tuff from Hengdaozi considered
by Lo et al. to be unaffected by metamorphism and whose age, based on total fusion 40Ar÷39Ar dating of biotite, argues for a Jurassic age for the Yixian Formation. The third tuff is a previously undated tuff from
the upper part of the underlying Tuchengzi Formation. Single crystal total fusion 40Ar÷39Ar analyses of the Sihetun sanidine showed homogeneous radiogenic Ar, Ca÷K ratios, excellent reproducibility and gave a mean
age of 125.0±0.18 (1SD)±0.04 (SE) Ma. Single sanidine crystal total fusion 40Ar÷39Ar analyses of the Hengdaozi tuff gave a mean age of 125.0±0.19 (1SD)±0.04 (SE) Ma, which is indistinguishable from the Sihetun
tuff. The Tuchengzi Formation tuff gave a mean age of 139.4±0.19 (1SD)±0.05 (SE) Ma. Detailed laser incremental-heating analyses
of biotite from Sihetun, Hengdaozi, and Tuchengzi tuffs show disturbed Ar release patterns and evidence of trapped argon components.
We conclude from these analyses that the total fusion dates on biotite by Lo et al. are erroneously old and isotopic dating
of both biotite and sanidine from tuffs of the Yixian Formation point to a middle Early Cretaceous age. The upper part of
the Tuchengzi Formation can be referred to the Early Cretaceous.

Full text

Page 1

NOTES
Chinese Science BulletinVol. 47No. 2January2002 135
Further support for a
Cretaceous age for the
feathered-dinosaur beds of
Liaoning,China:New40Ar/39Ar
dating of the Yixian and
Tuchengzi Formations
C. C. Swisher III1, WANG Xiaolin2, ZHOU Zhonghe2,
WANG Yuanqing2, JIN Fan2, ZHANG Jiangyong2,
XU Xing2, ZHANG Fucheng2& WANG Yuan2
1. Department of Geological Sciences, Wright Labs, Rutgers University,
Piscataway, New Jersey 08854, USA (e-mail: cswish@rutgers.
rci.edu)
2. Institute of Vertebrate Paleontology and Paleoanthropology, Chinese
Academy of Sciences, Beijing 100044, China (e-mail: xlnwang@
263.net)
Abstract
tained from total fusion and incremental-heating analyses of
sanidine and biotite from three tuffs found interbedded
within the fossil-bearing deposits of Liaoning, northeast
China. The first is a new sample of the Bed 6 Sihetun tuff
from the Yixian Formation, previously dated by our team as
middle Early Cretaceous, and recently considered by Lo et
al., partially reset due to metamorphism from a nearby ba-
saltic sill. The second is the Yixian Bed 9 tuff from Heng-
daozi considered by Lo et al. to be unaffected by metamor-
phism and whose age, based on total fusion40Ar/39Ar dating
of biotite, argues for a Jurassic age for the Yixian Formation.
The third tuff is a previously undated tuff from the upper
part of the underlying Tuchengzi Formation. Single crystal
total fusion
showed homogeneous radiogenic Ar, Ca/K ratios, excellent
reproducibility and gave a mean age of 125.0 ± 0.18 (1SD) ±
0.04 (SE) Ma. Single sanidine crystal total fusion
analyses of the Hengdaozi tuff gave a mean age of 125.0 ±
0.19 (1SD) ± 0.04 (SE) Ma, which is indistinguishable from
the Sihetun tuff. The Tuchengzi Formation tuff gave a mean
age of 139.4 ± 0.19 (1SD) ± 0.05 (SE) Ma. Detailed laser in-
cremental-heating analyses of biotite from Sihetun, Heng-
daozi, and Tuchengzi tuffs show disturbed Ar release pat-
terns and evidence of trapped argon components. We con-
clude from these analyses that the total fusion dates on bio-
tite by Lo et al. are erroneously old and isotopic dating of
both biotite and sanidine from tuffs of the Yixian Formation
point to a middle Early Cretaceous age.
the Tuchengzi Formation can be referred to the Early Cre-
taceous.
We report new
40Ar/39Ar dating results ob-
40Ar/39Ar analyses of the Sihetun sanidine
40Ar/39Ar
The upper part of
Keywords:
chengzi Formation?western Liaoning.
40Ar/39Ar dating?Cretaceous?Yixian Formation?Tu-
Excavations in the lake beds of the Yixian Formation
of Liaoning, China, continue to unearth a wide variety of
spectacular fossils that further help elucidate views on
long lived controversies over bird-dinosaur relationships,
early diversification of birds and mammals and the evolu-
tion of flowering plants[1?4]. Although the “primitive”
appearance of some of these fossils have suggested tem-
poral links with the Tithonian (Late Jurassic) faunas of
Europe, such as the Archaeopteryx bearing Solnhofen
limestone of southern Germany, other fossil data indicate
a much younger, middle Early Cretaceous age. Support
for a Jurassic age of the Yixian fossils comes from K-Ar
and Rb-Sr dating of associated volcanics[5,6], whereas;
40Ar/39Ar dating of interbedded tephra and overlying lavas
lends support to an Early Cretaceous age[7—10]. As a result,
two competing camps have emerged; one that views the
Yixian fossils as Late Jurassic contemporaries of Archae-
opteryx and a second that argues that while many of the
fossils possess primitive characteristics, the Yixian fossils
are of middle Early Cretaceous age, some 20 million years
younger than deposits that produced Archaeopteryx[7].
1Background and debate
In 1999, members of our group[7]reported40Ar/39Ar
dates on volcanic layers interbedded within the fos-
sil-bearing layers of the Yixian Formation that indicate a
Barremian (middle Early Cretaceous) age. The 124.6 Ma
dates were based on replicate total fusion40Ar/39Ar analy-
ses of single sanidine crystals and incremental-heating
40Ar/39Ar analysis of a multiple crystal sample separated
from tuff layers of the Yixian Formation exposed at the
sites of Sihetun and Jianshangou. We concluded that our
dates are more reliable than prior K-Ar and Rb-Sr dates
on biotite and whole rock[5,6]because of the superior argon
retentivity of sanidine and the unambiguous relationships
of our dated units with fossiliferous horizons of the Yixian
Formation[7]. Support for our age for the Yixian Formation
comes from40Ar/39Ar dating of overlying lavas and intru-
sive sills by Smith et al.[8]that yielded dates averaging
121 Ma, and from 125.2 Ma U-Pb dates on zircon re-
ported by Wang et al.[9,10]on the same Bed 6 Sihetun tuff
dated by40Ar/39Ar methods reported in Swisher et al.[7].
The underlying basalt also gave
Ma[10]which further supports the Early Cretaceous age of
the Yixian Formation.
Recently, Lo et al.[11,12]have presented new dating
results that challenge our Cretaceous age for the Yixian
Formation.Their view comes from recent 147 Ma
40Ar/39Ar dates on biotite and whole rock from Yixian
tuffs. Jiang et al.[13]utilizing Lo et al.’s dates[11,12]argue
that although our team[7]dated volcanic horizons from
sites unambiguously associated with the main fossilifer-
ous horizons of the Yixian Formation, our sample locali-
ties had been affected by local contact metamorphism.
nearby 121 Ma intrusive basaltic sill was considered to
have thermally reset the age of the sanidine from the Si-
hetun Tuff dated by our team. In contrast, Jiang et al.[13]
point out that Lo et al.[11,12]dated samples from Hengdaozi,
40Ar/39Ar dates of 128
A

Page 2

NOTES
136Chinese Science BulletinVol. 47No. 2 January 2002
a site unaffected by the younger intrusive basalts, and
therefore, the older 147 Ma dates more accurately reflect
the geologic age of the Yixian Formation. Combined with
the previously published K-Ar date of (137 ± 7) Ma and
Rb-Sr date of (143 ± 4) Ma[5,6]from the Yixian Formation,
Jiang et al.[13]
conclude that the Yixian fossil-bearing
lake beds are Late Jurassic age, contemporaries of the
Archaeopteryx bearing Solnhofen limestone (note: these
workers use an age of around 135 Ma for the Jurassic/
Cretaceous boundary, although most workers now accept
an age for the boundary of around 144 Ma[14]).
Lo et al.’s[11,12]dating serves as the basis for Jiang et
al.’s arguments against an Early Cretaceous age for the
Yixian Formation. The dates are based on laser total fu-
sion
two samples of a tuff from Bed 9, situated 20—30 m
above the main Sihetun fossiliferous layers of the Yixian
Formation at Hengdaozi[15,16]. Individual biotite crystal
dates from the two samples range from 136 Ma to 155 Ma,
with means of (147.1 ± 5.5) Ma and (142.5 ± 3.6) Ma.
Plotting the
on isochron plots, Lo et al. derive isochron dates of (147.9
± 2.1) Ma and (142.2 ± 2.2) Ma[11,12].40Ar/36Ar intercept
values are within uncertainty of air ratio, which indicated
to Lo et al. that the isotopic systems of the biotites were
relatively closed, that there was no apparent trapped argon
component and no significant alteration. Combining the
data for the two samples gave an overall mean age of
(145.3 ± 4.4) Ma and a combined isochron date of (147.1
± 1.8) Ma[11,12].
Lo et al.[11,12]also report dates for groundmass
(whole rock) chips from these two samples. Individual
total fusion dates range from 116 to 130 Ma, with means
of (125.5 ± 4.2) Ma and (120.8 ± 3.1) Ma. An isochron
plot of the individual analyses of one of the samples gave
an air intercept and date of (127.8 ± 1.9) Ma. Although
they point out that this age is comparable to those reported
by our group[7]and by Smith et al.[8], the matrix was most
likely affected by slight alteration and therefore not geo-
logically meaningful[11,12]. Besides, they also report a
whole rock K-Ar date on a fine-grained volcanic ash from
Bed 6[15,16]of (147.3 ± 3.3) Ma, but because of its
fine-grained character they were unsuccessful in obtaining
40Ar/39Ar dates on this sample[11,12].
At first glance, Lo et al.’s biotite data[11,12]and the
arguments raised by Jiang et al.[13]lend credence to the
possibility that the Yixian Formation may indeed be of
Jurassic age. However, after close examination of Lo et
al.’s data and further40Ar/39Ar analyses of the biotite and
sanidine from the Hengdaozi tephra that was analyzed by
Lo et al.[11,12]by our group, we disagree with their conclu-
sions.
We argue that the intrusive sill at Sihetun is of insuf-
ficient proximity to our sample sites at Sihetun and Jian-
shangou to have any major adverse affect on the argon
40Ar/39Ar analyses of single crystals of biotite from
36Ar/40Ar and
39Ar/40Ar data from these data
systematics of the dated sanidine. We note from field ob-
servation that no physical evidence of metamorphism of
either the dated tephra or the adjoining fossiliferous sedi-
mentary beds occurs beyond about 0.5—1 m of the mar-
gins of the basalt sill. We also note no apparent distur-
bance of the argon systematics as observed in the release
pattern obtained from our incremental-heating analysis of
the sanidine[7].
In contrast, we point out that the single biotite crystal
dates reported by Lo et al.[11,12]show much greater spread
in age (136—155 Ma) than is expected from a single ho-
mogeneous mineral population. This leads us to believe
that the Ar systematics of the biotite is disturbed either by
alteration or by a trapped Ar component. Given that all of
these dates are based on the total fusion analyses, any
disturbance of the argon systematics or the determination
of trapped argon components cannot be fully examined.
Plotting the total fusion dates on a single isochron do
show a linear array with an atmospheric intercept, how-
ever, we believe this is a result of various crystal to crystal
amount of surficial atmospheric contamination released
from weak interstitial layers and altered surfaces of the
biotite. No incremental heating analyses were made that
could address low temperature surficial and higher tem-
perature trapped Ar components. As a result, these total
fusion
previously by total fusion K-Ar techniques.
2Samples, analyses and results
To directly address these issues, we analyzed
sanidine and biotite from three different tuff samples from
the Liaoning area. The first, for control purposes, is a new
sample (99L-S1) of the Bed 6 Sihetun tuff previously
dated by us, and reported by Lo et al.[11,12]to be affected
by metamorphism. The second sample (99L-HDZ1), is the
Bed 9 tuff from Hengdaozi dated by Lo et al.[11,12]report-
edly unaffected by metamorphism. The third tuff (99L-T1)
is a previously undated tuff from the underlying upper
part of the underlying Tuchengzi Formation.
40Ar/39Ar dating of Liaoning tuffs was accomplished
by replicate CO2 laser, total fusion analyses of single
sanidine crystals and incremental-heating of biotite fol-
lowing procedures outlined previously by us in Swisher et
al.[7]references therein.All analyses were made using a
Mass Analyzer Products 215-50 mass spectrometer at the
Berkeley Geochronology Center, in Berkeley California,
following procedures outlined by Renne et al.[17]and
references therein. The ages were calculated using an age
of 28.02 Ma for the co-irradiated Fish Canyon Sanidine
monitor mineral[17]. To control and minimize uncertainties
due to lateral neutron flux gradients during sample irra-
diation, each Liaoning sample was arranged in Al sample
holders bracketed between multiple samples of Fish Can-
yon Sanidine.Uncertainties reported are one standard
deviation of mean and standard error of the mean and do
40Ar/39Ar data result in similar ages as produced

Page 3

NOTES
Chinese Science Bulletin Vol. 47No. 2January 2002137
not take into account uncertainties in the “absolute” age of
the monitor mineral.
24 single crystal total fusion
the Sihetun sanidine show homogeneous radiogenic Ar,
Ca/K ratios, excellent reproducibility and analytical un-
certainty ranging in age from 124.7 to 125.3 (Plate?(a)).
Our mean age of 125.0 ± 0.18 (1SD) ± 0.04 (SE) Ma is
slightly older than the 124.6 Ma we reported[7]previously
and attribute this slight age difference to improved stan-
dard calibration and control of lateral gradients induced
during the two irradiations. We consider this new date a
more accurate age for Sihetun and Hengdaozi than previ-
ously reported by our group. The sanidine from the Bed 9
Hengdaozi tuff located 20—30 m above the Bed 6 tuff
gave indistinguishable results from the Sihetun tuff. 24
single sanidine crystal total fusion
this sanidine gave a mean age of 125.0 ± 0.19 (1SD) ±
0.04 (SE) Ma (Plate?).
The tuff from the upper part of the underlying Tu-
chengzi Formation gave a mean40Ar/39Ar age of 139.4 ±
0.19 (1SD) ± 0.05 (SE) Ma, based on 19 single sanidine
crystal analyses (Plate?(a)). This represents the first iso-
topic dating for the Tuchengzi Formation and although
considered previously middle Jurassic in age, at least the
upper part of the formation can be referred to the Early
Cretaceous.The new date leads further credence to our
Cretaceous age for the overlying Yixian Formation and
limits the previously recognized hiatus between the two
formations to less than 15 million years.
stratigraphic location of the continental Jurassic-Creta-
ceous boundary in northeastern China must be sought for
within the lower part of the Tuchengzi Formation or
within underlying formations.
Our new analyses indicate that all three tuffs give
dates consistent with a Cretaceous age for the Yixian
Formation, as well as for the underlying upper Tuchengzi
Formation. The excellent agreement between the sample
from Sihetun and that from Hengdaozi supports our view
that the Sihetun sanidine was not affected by contact
metamorphism event. The congruence between our new
age for the Sihtuen tuff of 125.0 ± 0.18 (1SD) ± 0.04 (SE)
based on40Ar/39Ar dating of sanidine and the U-Pb zircon
dates reported by Wang et al.
argues that given the vastly different isotope system be-
havior and different closure temperature of the zircon and
sanidine, the Sihetun tuff was unaffected by metamor-
phism.
3 Comparison and discussion
40Ar/39Ar analyses of
40Ar/39Ar analyses of
As a result, the
[9]of (125.2 ? 0.9) Ma
What, then is the explanation for the erroneously old
total fusion biotite and whole rock dates? To examine this
issue more closely, we performed detailed laser incre-
mental-heating analyses of biotite from Sihetun, Heng-
daozi, and Tuchengzi tuffs (Plate?(b)). In all three cases,
the incremental-heating analyses show disturbed Ar re-
lease patterns. Of importance, both the Sihetun and Tu-
chengzi biotites give integrated ages of 152 Ma and the
Hengdaozi biotite 135 Ma. These ages are comparable to
the reported total fusion data[11,12]and lead to an explana-
tion of the old dates. Plotting the
39Ar/40Ar ratios obtained for the individual temperature
increments on an isochron (Plate ?), the Sihetun and
Hengdaozi biotites both indicate excess-trapped Ar com-
ponents[18]. For the Sihetun biotite, the lower temperature
steps give an isochron age of 154 Ma with a
intercept of 332; however, the higher temperature steps
give an age of 127 Ma with an intercept of 940 (Plate ?
(a)). Similarly the lower temperature steps for Hengdaozi
give an isochron age of 151 Ma with40Ar/36Ar intercept of
279 and the higher temperature more radiogenic steps
give an age of 127 Ma with a40Ar/36Ar intercept of 910
(Plate ?(b)). Whereas the Sihetun biotite shows a sharp
intersection of the two slopes on the isochron plot, the
Hengdaozi biotite shows a broader mixing interval with
an estimated age of about 134 Ma (Plate ?).
Contrary to the conclusion drawn by Jiang et al.[13],
we note that our detailed incremental-heating analyses of
the biotite show evidence of trapped argon components
and conclude that the total fusion dates of Lo et al.[11,12]
are erroneously old. All isochrons gave intercept values
greater than the expected air ratio of 295.5, thus making
all total fusion dates too old. Wang et al.[9]also discovered
in their XRD analysis that the biotite was almost entirely
weathered into the vermiculite, the K element largely ran
off during the weathering state, which led to an open K-Ar
system, therefore the Ar-Ar age of the biotite has little
geological meaning.
Our new data help explain the prior K-Ar dates on
biotite as well as the intermediate but younger age of the
whole rock chips, in the latter case, the total Ar being
dominated by the contribution from sanidine and glass
rather than biotite. We conclude that the K-Ar and total
fusion
biotite and whole rock by Lo et al.[11,12]and other work-
ers[5,6]were insufficient to characterize the disturbed argon
systematics of the Liaoning biotites.
As a result, we disagree with the conclusions reached
by Jiang et al.[13]and conclude that all isotopic dating of
tuffs from the Yixian Formation point to a middle Early
Cretaceous age. Reasons for the anatomical similarities of
certain Yixian taxa with Late Jurassic taxa elsewhere will
have to focus on biological and ecological explanations.
Acknowledgements
We thank Mee-mann Chang for her assistance
and Wang Songshan for reading manuscript. This work was supported by
the Special Fund for the Major State Basic Research Projects of China
(Grant No. G000077700), the Chinese Academy of Sciences (Grant No.
KZCX3-J-03), the National Natural Science Foundation of China (Grant
No. 49832002), the Hundred Talents Project of the CAS, the National
Science Fund for Distinguished Young Scholars of China (Grant No.
40025208) to Z. Z., and the National Science Foundation, Earth Science
Program to C. C. S.
36Ar/40Ar against the
40Ar/36Ar
40Ar/39Ar dating techniques employed to date the

Page 4

NOTES
138Chinese Science BulletinVol. 47No. 2 January2002
References
1. Zhou, Z. H., Wang, X. L., Zhang, F. C. et al., Important features of
Caudipteryx—evidence from two nearly complete new specimens,
Vert. PalAsiat., 2000, 38(4): 241.
2. Wang, X. L., Xu, X., A new iguanodontid (Jinzhousaurus yangi
gen. et sp. nov.) from the Yixian Formation of western Liaoning,
China, Chinese Sci. Bull., 2001, 46(19): 1669.
3. Zhang, F. C., Zhou, Z. H., A primitive enantiornithine bird and the
origin of feathers, Science, 2000, 290: 1955.
4. Li, J. L., Wang, Y., Wang, Y. Q. et al., A new family of primitive
mammal from the Mesozoic of western Liaoning, China, Chinese
Sci. Bull., 2001, 46(9): 782.
5. Wang, D. F., On the age of the Rehe Group in western Liaoning
Province, China, Bull. ChineseAcad. Geol. Sci., 1983, 7: 65.
6. Wang, D. F., Diao, N. C., Geochronology of Jura-Cretaceous vol-
canics in west Liaoning, China, in Scientific Papers on Geology
for International Exchange (1), Beijing: Geological Publishing
House, 1984, 1—12.
7. Swisher III, C.C. Wang, Y. Q., Wang, X L. et al., Cretaceous Age
for the feathered dinosaurs of Liaoning, China, Nature, 1999, 400,
58.
8. Smith, P. E., Evensen, N. M., York, D. et al., Dates and rates in
ancient lakes:40Ar-39Ar evidence for an Early Cretaceous age for
the Jehol Group, Northeast China, Can. J. Earth Sci., 1995, 32,
1426.
9. Wang, S. S., Wang, Y.Q., Hu, H. G. et al., The existing time of Si-
hetun vertebrate in western Liaoning–evidence from U-Ph dating
of zircon, Chinese Sci. Bull., 2001, 46 (9): 779.
10. Wang, S. S., Hu, H. G, Li P X. et al., Further discussion on geo-
logic age of Sihetun vertebrate assemblage in western Liaoning,
China: evidence from Ar-Ar dating, Act. Petrolog. Sin., 2001,
17(4): 663
11. Lo, C. H., Chen, P. J., Tsou, T. Y. et al., Age of Sinosauropteryx
and Confuciusornis ——
dating of the Yixian Formation, N.E. China, Geochemica, 1999,
28(4), 405.
12. Lo, C. H., Chen, P. J., Tsou, T. Y. et al.,40Ar/39Ar laser single-grain
and K-Ar dating of the Yixian Formation, NE China, Palaeoworld,
1999(11): 328.
13. Jiang, F. X., Chen, J. H., Cao, M. Z. et al., A discussion on the age
of the feathered dinosaur——bearing beds of Liaoning, China,
Acta Palaeont. Sin., 2000, 39(2): 307.
14. Gradstein, F. M., Agterberg, F. P., Ogg, J. G. et al., A Triassic, Ju-
rassic and Cretaceous times scale. In: Berggren, W. A., Kent, D. V.,
Aubry, M. P. et al., eds. Geochronology, Time Scales and Global
Stratigraphic Correlation. S.E.P.M. Spec. Publ., 1999, 54: 95.
15. Wang, X. L., Wang, Y. O., Wang, Y. et al., Stratigraphic sequence
and vertebrate-bearing beds of the lower part of the Yixian Forma-
tion in Sihetun and neighboring area, western Liaoning, China,
Vert. PalAsiat., 1998, 36(2): 81.
16. Wang, X. L., Wang, Y. O., Jin, F. et al., The Sihetun fossil verte-
brate assemblage and its geological setting, western Liaoning,
China, Palaeoworld, 1999(11): 310.
17. Renne, P. R., Swisher, C. C, Deino, A. L. et al., Intercalibration of
standards, absolute ages and uncertainties in
Chemical Geology, 1998, 145: 117.
18. McDougall, I., Harrison, T. M., Geochronology and Thermochro-
nology, New York: Oxford University Press, 1988.
(Received September 11, 2001; accepted October 29, 2001)
40Ar/39Ar laser single-grain and K-Ar
40Ar/39Ar dating,