40Ar-39Ar age determinations on the Owyhee basalt of the Columbia Plateau

40Ar-39Ar age determinations on the Owyhee basalt of the Columbia Plateau

Earth and Planetary Science Letters, 31 (1976) 75-84 © Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands 75 [41 4°Ar-3...

652KB Sizes 0 Downloads 8 Views

Earth and Planetary Science Letters, 31 (1976) 75-84 © Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

75

[41

4°Ar-39Ar AGE DETERMINATIONS ON THE OWYHEE BASALT OF THE COLUMBIA PLATEAU RICHARD J. BOTTOMLEY and DEREK YORK Geophysics Division, Department o f Physics, Univeristy of Toronto, Toronto, Ont. (Canada) Received September 29, 1975 Revised version received March 3, 1976 4°Ar/39Ar step-heating analyses have been performed on 11 samples of basalt from sites near Owyhee Reservoir of southeastern Oregon, U.S.A. These rocks were extruded during the great flood basalt episode of the Pacific Northwest. The whole-rock points are highly correlated on a plot of 4°Ar/36Ar versus 39Ar/36Ar, corresponding to a common age of the samples of 14.3 _+0.3 m.y. Inspite of this, individual "plateau" plots of the age versus fraction of 39Ar released do not give good plateaux. These age spectra exhibit to varying degrees a common structure in which lower age values are found at higher temperatures. This pattern may result from a closed-system redistribution of the argon isotopes. The usefulness of grinding the basalts in removing a loosely held atmospheric argon component is confirmed.

1. Introduction Initial attempts to date the Columbia River Plateau lavas were in general thwarted by high atmospheric argon contaminations and sometimes by failure to obtain reproducible ages on chunks of a single core [1,2]. The Steens Mountain material (15.1 -+0.3 m.y.) was a well-behaved exception to this [3]. While plots of 4°Ar/ 36Ar versus 4°K/36Ar for the erratic samples led York et al. [2] to propose that the Columbia Plateau basalts had been largely erupted between 14 and 18 m.y. ago, this proposal was only clearly substantiated by an interesting discovery o f Baksi. He showed that the erratic atmospheric argon component afflicting these rocks could be released on grinding, and concordant ages could readily be found in the 1 3 - 1 6 m.y. range by analyses of such ground material [4,5]. In view o f the significance o f the Columbia Plateau material for tectonic studies [6] and for the methodology o f K - A r dating, 4°Ar/39Ar analyses have been carried out on 11 whole-rock core samples from exposures on the east side o f the Owyhee Reservoir in Malheur County, Oregon (43°37'N, 117°17'W), about 70 km west o f Doise, Idaho. Originally the cores were part of a suite of sixteen flows sampled by Watkins [7] for a study of the paleomagnetic directions of the lava flows o f the

Pacific Northwest. Sample numbers as defined and used by Watkins are retained for the reader's convenience. The localities and magnetic data for these flows may be found in Watkins and Baksi [5]. Watkins and Baksi found a possible range in age o f 1 3 . 1 - 1 3 . 9 m.y. from four K - A r analyses on ground-up whole rocks from this section.

2. Experimental methods Several 40Ar/a9Ar analyses were done on each of four flows (NK1, 11, 14 and 16). Irradiations were performed in the enriched-uranium reactor at McMaster University. Six samples were crushed to - 1 2 , +200 mesh before irradiation, while three were irradiated as chunks and crushed just prior to fusion. The other two samples were irradiated and fused as chunks. All specimens were bal~ed out at approximately 200°C for at least forty hours in a bakeable fusion system. They were then usually fused in eight separate steps and the corresponding argon fractions were analyzed on an MS10 mass spectrometer. The resultant isotope ratios were normalized to an atmospheric argon ratio o f 4°Ar/ 36Ar = 295.5, and were corrected for atmospheric and neutron-generated argon isotope interference accord-

76 TABLE 1 Step-heating data Lab. no.

T(°C)

fl

f2

Atmos.

(%)

4OAr* f(39) (× 10 -7 cm 3

Age (m.y.)

STP/g)

Sample NKI-IL nos. 1 7 - 2 5 17 18 19 20 21 22 23 24 25

500 600 700 850 950 1040 1200 1400 1600

-0.000158 -0.000198 -0.000756 -0.000890 -0.00140 -0.00310 -0.0157 -0.0147 -0.0138

0.000134 0.000987 0.0158 0.0469 0.0206 0.0171 0.0188 -0.00865 -0.0123

Total

85.0 62.1 29.5 15.1 38.6 56.7 87.6 151.3 106.0

0.55 3.03 3.55 2.08 1.55 0.36 0.04 -0.12 -0.03

53.2

11.01

0.053 0.273 0.303 0.191 0.144 0.029 0.005 0.001 0.001 integrated age

13.7 14.6 15.4 14.3 14.2 16.5 9.04 -124.3 -78.2 14.5

-+ 1.3 ± 1.1 -+ 0.2 -+ 0.2 + 0.1 + 0.3 + 5.4 -* 70.5 + 1270 ±

Standard (40/39) s = 21.91 Mass: = 1.86 g

Sample NKll-2~ nos. 2 9 - 3 3 29 30 31 32 33

885 1020 1130 1300 1400

-0.00204 -0.00131 -0.00426 -0.00973 -0.0104

-0.00162 -0.000130 -0.00171 -0.00438 -0.00629

Total

97.7 90.0 93.7 96.5 97.7

3.62 1.64 0.03 0.31 0.05

96.9

5.64

93.0 82.6 65.9 44.8 43.1 52.7 85.6 93.8

0.60 0.18 2.91 2.00 3.91 0.89 0.11 0.12

67.7

10.69

81.3 66.4 49.5 28.0 25.5

0.12 1.51 1.69 0.97 1.61

0.621 0.265 0.006 0.092 0.016 integrated age

14.4+7.3 15.2±1.2 13.1±l.7 8.2±2.1 7.5±7.5 13.9±0.6

Standard: (40/39) s = 25.52 Mass = 1.79 g

Sample NKI-IL nos. 4 2 - 4 9 42 43 44 45 46 47 48 49

400 500 600 700 865 1000 1400 1600

-0.000151 -0.0000931 -0.000475 -0.000988 -0.000997 -0.00175 -0.00307 -0.0232

-0.0000208 0.0000982 0.00168 0.0102 0.0113 0.0140 0.00520 0.00165

Total

0.067 0.017 0.250 0.182 0.359 0.088 0.018 0.018 integrated age

11.6 ± 1.3 13.7 ± 26.9 15.1 ± 0.7 14.3 + 0.6 14.1 ±-+ 0.1 13.1 ± 0.2 8 . 3 + 1.4 8.2 + 8.1 13.9 ± 0.2

Standard: (40/39) s = 23.63 Mass: ~ 1.3 g

Sample NKll-2L nos. 6 2 - 6 9 62 63 64 65 66

400 500 600 700 850

-0.000212 -0.000285 -0.000794 -0.00126 -0.00122

0.000065 0.000695 0.00427 0.0193 0.0227

0.016 0.241 0.249 0.148 0.259

17.6 14.8 16.0 15.6 14.7

± -+ +-+ _+

2.3 0.2 0.1 0.2 0.1

0.3

77

TABLE 1 (continued) Lab. no.

T(°C)

fl

Atmos.

(%)

4OAr* f(39) (X 10 -7 cm 3

Age (m.y.)

STP/g) 67 68 69

960 1050 1600

-0.00234 -0.00384 -0.0300

0.0134 0.00781 -0.0241

Tot~

52.7 72.0 98.3

0.30 0.11 0.05

60.4

6.36

103.6 83.6 102.3 67.2 41.2 56.6 88.3 59.4

-0.005 0.07 -0.0008 0.27 1.27 0.53 0.47 0.20

70.0

2.81

0.054 0.020 0.014 integrated age

13.2 -+ 0.5 12.8 ± 1.0 8.8 +_ 18.4 15.0 +, 0.3

S t a n d a r d : ( 4 0 / 3 9 ) s = 31.31 Mass:= 1.30 g

Sample NKll-2L nos. 7 0 - 7 7 70 71 72 73 74 75 76 77

400 500 600 700 750 785 875 1600

0.00 -0.00172 -0.00253 -0.00179 -0.00128 -0.00177 -0.00504 -0.00140

0.0 -0.000889 -0.000725 0.00281 0.00862 0.00730 0.000463 0.00602

TotM

0.001 0.010 0.001 0.080 0.393 0.203 0.222 0.089 integrated age

-29.7 40.6 3.3 19.0 18.4 14.9 12.1 12.8

+_ 291.4 _+ 26.5 +- 57.9 +, 0.4 +, 1.0 _+ 0.5 +- 0.7 _+ 2.2

16.0 _+

0.7

Standard: (40/39)s = 31.31 Mass: -~ 1.0 g

Sample NKll-IL nos. 8 0 - 8 7 80 81 82 83 84 85 86 87

400 500 600 700 800 845 935 1600

-0.000846 -0.00260 -0.00396 -0.00263 -0.00178 -0.00161 -0.00193 -0.00608

-0.000595 -0.000430 0.0387 0.00680 0.0115 0.0179 0.0246 0.0126

Total

99.4 89.4 11.7 68.7 46.3 34.2 31.2 71.2

0.0005 0.05 0.31 0.22 1.30 0.95 0.58 0.95

54.8

4.37

-2.4 95.2 132.4 71.1 39.8 33.0

0.07 0.009 -0.14 0.15 1.22 1.64

0.001 0.013 0.016 0.062 0.300 0.216 0.131 0.261 integrated age

1.8 13.6 63.7 12.1 14.7 15.0 15.0 12.4

-+ 10 l° +- 8.9 + 1.5 +- 1.4 -+ 0.1 +- 0.1 + 0.9 _+ 0.3

14.8 -+ 1.5

Standard: (40/39) s = 32.88 Mass: = 1.79 g

Sample NKll-lL nos. 8 9 - 9 6 89 90 91 92 93 94

400 500 600 650 750 800

-0.000005 -0.000773 -0.00120 -0.00134 -0.00178 -0.00159

-0.000090 0.000113 0.000568 0.00294 0.0145 0.0187

0.001 0.006 0.027 0.037 0.229 0.324

226.8 4.0 -16.0 12.1 15.6 14.9

+ 186.1 +- 3.2 +- 3.0 -+ 1.5 +- 0.4 +- 0.5

78

TABLE 1 (continued) Lab. no.

T(°C)

/"1

Atmos.

(%)

4OAr, f(39) (X 10 -7 cm 3

Age (m.y.)

STP/g) 95 96

870 1600

-0.00151 -0.00841

0.0159 0.0342

Tot~

37.0 57.4

0.95 0.80

48.3

4.69

0.202 0.173 integrated age

13.9 +13.5 +

0.4 0.1

13.8 ±

0.7

Standard: (40/39) s = 32,88 Mass: = 1.61 g

Sample NK 14-1, nos. 9 8 - 1 0 5 98 99 100 101 102 103 104 105

400 500 600 650 700 740 820 1600

-0.00576 -0.000636 -0.00138 -0.00163 -0.00334 -0.00607 -0.00692 -0.0101

-0.00346 -0.000371 -0.000546 0.000440 0.00313 -0.000890 0.00261 0.0227

Total

51.1 88.3 97.3 81.5 75.5 89.8 82.7 79.3

0.04 0.02 0.01 0.24 0.68 0.47 0.27 0.38

0.001 0.005 0.017 0.084 0.255 0.222 0.112 0.304

83.5

2.12

81.4 72.6 73.6 83.8 69.8 82.1 89.9 85.6

0.53 0.79 0.54 0.24 0.15 0.05 0.05 0.17

78.5

2.52

98.3 94.7 78.9 63.0 42.2 59.0 74.3

0.02 0.06 0.45 0.18 0.67 0.69

0.009 0.029 0.148 0.061 0.217 0.255

8.2 10.7 15.1 14.6 15.3 13.4

0.70

0.281

12.4 -+ 0.2

73.0

2.77

integrated age

217.9 30.4 4.5 17.1 16.0 12.7 14.5 7.5

± 260.0 ± 21.2 ± 172.1 ± 1.0 ± 0.3 ± 0.4 ± 0.9 ± 0.3

12.7 -+

0.3

Standard: (40/39) s = 32,88 Mass: ~ 1.5g

SampleNK14-1, nos. 1 0 6 - 1 1 3 106 107 108 109 110 111 112 113

600 700 790 850 900 960 1100 1600

-0.00146 -0.00266 -0.00376 -0.00451 -0.00435 -0.00424 -0.00497 -0.00808

0.000105 0.00180 0.00214 0.000879 0.00583 0.00356 0.00467 0.00933

Total

0.176 0.248 0.167 0.107 0.058 0.031 ,0.059 0.154 integrated age

17.5 18.4 18.7 13.3 15.1 9.7 4.8 6.3

± 1.1 -+ 0.8 -+ 0.6 ± 0.3 -+ 0.8 ± 1.7 -+ 0.4 ± 0.6

14.6 -+ 0.6

Standard: (40/39) s = 38.33 Mass: = 1.79 g

Sample NK11-1L nos. 1 1 4 - 1 2 0 114 115 116 117 118 119 120

600 700 750 870 940 1100 1600

Total Standard (40/39) s = 38.33 Mass: = 1.71 g

-0.00201 -0.00212 -0.00196 -0.00131 -0.00134 -0.00207 -0.00553

-0.00165 -0.00121 0.000874 0.00299 0.00844 0.00669 0.00703

integrated age

± 10.5 ± 3.2 ± 0.4 -+ 0.1 ± 0.2 + 0.2

13.7 -+ 0.2

79

TABLE 1 (continued) Lab. no.

T(°C)

fx

f2

Atmos. (%)

4°Ar* f(39) (X 10-7 cm 3 STP/g)

Age (m.y.)

0.0000167 0.00132 0.00410 0.00374 0.00316 0.00495 0.00536 0.0162

90.4 71.8 56.7 58.7 58.0 54.1 68.9 74.6

0.17 0.48 1.07 0.86 0.76 0.66 0.39 0.19

8.5 14.6 15.7 14.6 14.7 14.8 13.8 14.4

65.5

4.58

Sample NK16-2i, nos. 123-130 123 124 125 126 127 128 129 130

500 600 700 750 790 910 1100 1600

-0.000669 -0.00110 -0.00138 -0.00127 -0.00105 -0.00134 -0.00320 -0.0178

Total

0.063 0.104 0.213 0.185 0.163 0.140 0.089 0.042 integrated age

+-0.7 + 0.3 +-0.2 +-0.3 -+0.2 -+-0.1 +-0.7 +- 1.1

14.4 + 0.2

Standard: (40/39) s = 38.33 Mass: -- 1.42 g

ing to the expression: 40" _39K

4(.~)M ( 1 - - f l ) - - 2 9 5 . 5

3(~)M ( l - - f 2 ) - - 4(~-~)K

(1) where 40* = radiogenic 4°At; 39 K = 39Ar from 39I(; ( )M = argon isotope ratio measured during analysis of a gas fraction from rock; ( )K = argon isotope ratio in gas extracted from irradiated potassium; f l = 1/[1 -- (37/39)Ca/(37/39)Ml;f2 = [1 -- (36/39)Ca/(36/39)M]/[1 -- (37/39)Ca/(37/39)M]; ( )Ca = argon isotope ratio in gas extracted from irradiated calcium. The formulae assume all corrections for 37mrdecay have been made. Analyses o f irradiated CaF2 and KzSOa during this investigation yielded values of (39/ 37)c a = (6.51 -+ 0.31) × 10-4, (36/37)C a = (2.54 -+0.09) × 10 -4 and (40/39)K = (1.56 +- 0.04) × 10 -2 and these were adopted in eq. 1. The use of eq. 1 was proposed by York and its advantages are explained in Mak et al.

[81. Samples of a basalt, RS-1, were used as a standard. K - A r age o f this was determined by six flame photometric potassium analyses and two argon analyses to be 27.2 + 0.2 m.y. Two samples of standard were included with every batch of unknowns. In three irradiations the standards indicated a flux variation among

the samples of less than 4.5%, so an average value for the standards in that batch was taken for the associated unknowns. In two irradiations the standards recorded differences in 40"/39 K ratios of 7.3% and 9.3% respectively. F o r samples from these two cans the 40"/39 K ratio of the nearest standard was employed in the age calculations. Errors due to the scatter in the mass spectrometer runs are quoted at the 1-sigma level and do not include estimates of neutron flux variation. Early analyses revealed large methane and methanerelated peaks in some gas fractions which sometimes were big enough to cause significant pressure scattering. The problem was solved effectively for later samples by installing a resistance-heated titanium getter on the MS10 inlet system. Three samples of core were also analyzed with the conventional K - A r technique, one (NK1 4-1) being fused as a chunk, two (NK5-2, NK16-2i) being fused as powders. 3. Results The results are displayed in Tables 1 - 3 and Figs. 1, 2. The age spectra for the individual samples are shown in Fig. 1, while the integrated values for the whole rocks are shown on a plot of 4°Ar/36Ar versus 39Ar/ 36Ar in Fig. 2. In keeping with previous experience [2] two of the

80 2O

03 20 13:: < uJ >- 18 u_ 16 o 14 03

03

rr 18 ,,~ uJ >- 16

..........................L

u.. 14 0 J 03 12

Z

o

Z

o

10

...a I0

__1 8 NKI-li

u_l 6[

<

I

1

0

or)

12

._J

LAB # 1 7 - 2 5 I

I

I

I

i,i (..9 I

,<

J

.4

.2

8 6

1.0

4 -6 .8 FRACTION 39

2

0

1.0 ~.1400°C 16000C

.6 .8 FRACTION 3 9

20-

cr 18<:~ uJ >- 16it 0 14 03 12 Z 0 10 _J

--

8

NKI-li

LAB:#:42-49

uJ 6 0

1

I

.2

0

I

I

I

I

.6

4

I

.8

FRACTION

I

1.0

39

03 201 rr

< i,i >- 16~

g

LL

............. 1

oo Z

~, '2~ 0 I0

NK I-li LAB # 6 2 - 6 9

N8

LAB 4# 7 0 - - 7 7

NK 11-2i

L~J (_9

6 0

12

d I!f

._1

~ (-9

16

o

<~ I

I

.2

I

I

I

1

I

I

4 .6 .8 FRACTION 3 9

Fig. 1 (for caption see p. 81).

1.0

I

0

!

I

2

I

I

I

I

I

I

4 .6 .8 FRACTION 3 9

1

I

1.0

81 63.7 m.y. 0320 r~

< w ~- 1 2 ~ it OK? co 7 8 0 -56 _J "~ 4

,

LIHIIHZlHH.HHH~

w I

I

I

.2

1

1

.4

1

I .8

I

.6

. I

NKI6-2i

1

0

I .2

1

FRACTION 5 9

14 0 03 12

JE 6"

I .6

I

1 .8

i

1

1.0

LAB # 9 8 - 1 0 5

6~

4 6 -8 FRACTION 3 9

1.0

N

0

KLAB I # 140 6 - - 113 1 .2

.4 .6 8 FRACTION 3 9

1.0

20 u3 < 18 uJ >- 16

.##mv~lllJ/lllllllllll.,.

12 z _0 I0 ._1 -J 8 uJ 6 (.9 < 0

NK 14.- I .2

0

2O u3 ¢'r" ,< 18 uJ >..- 16 It 0 14

1

03 12 Z 0 -3 I0 d

I0

Ld

1 .4

< 18 LIJ >- 16 m it 0 14

it

8-

89-96

FRACTION 3 9

2O (/3 r r 18 < w )-- 16

--

LAB#

2

I

1.0

k...,,,,,..,,,,,.

i'

8

I.tJ 2 ,< 0

IIIIIIIIIIIIIIIIIIIIIIIIIJ

16 it 0 14 (/3

z 12 0 .._q IO d

LAB # 8 0 - 8 7

NK ll-li

< 18 t~

..............

I t 14 0 or) 12 Z I0 _1 ___J 8

N K I 6 - 2 i LAB # 1 2 3 - 1 3 0

I

I

.2

I

I

4

1

I

I

-6

FRACTION

I

.8 39

N K I I - l i L A B # 114-120

3;

1

I

1.0

itl 6 £.9 < 0

I

1

.2

1

1

4

I

I

I

.6

I

.8

i

i

1.0

FRACTION 3 9

Fig. 1.4°Ar-~9Ar age spectra including error estimates, for all the samples studied. NK11-2i, Lab. nos. 29-34, is subject to the large errors shown because of problems with methane contamination. The first two fractions, presumed small, were lost. This problem was eliminated with the installation of a getter on the mass spectrometer inlet.

82 TABLE 2 Conventional K-Ar ages Sample

Lab. no.

Atmos. (%)

4OAr* (X 10-7 cm3 STP/g)

Mass (g)

Age (m.y.)

NK14-1 NK16-2i NK5-2

131 132 133

93 26 64

5.5 5.5 6.5

1.38 (chunk) 0.50 (powder) 0.61 (powder)

22.1 _+2.0 14.4 _+0.3 14.0 +_0.6

three sets of chunks analyzed had very large atmospheric contaminations. Thus, chunks of NK11-2i and NK14-1 (Table 3) contained 1.8 × 10-s and 0.8 X 10-s cm 3 STP/g of atmospheric 4°Ar respectively. These samples further displayed the p h e n o m e n o n discovered by Baksi [4] of loss of much of this contamination on grinding. For instance, as may be seen from Table 3, grinding reduced the atmospheric contamination of NK11-2i by approximately a factor of 20, while the reduction was by a factor of 8 for ground-up samples of NK14,1. In contrast, chunks of N K I - l i displayed a significantly lower atmospheric contamination (Table 3), 0.13 X 10-s cm 3 STP/g, which in one case was slightly reduced on grinding and in another case actually doubled when ground up. The results for the integrated 4°Ar-39Ar ages on eight of the nine powdered samples, and the two conventionally determined K - A r ages, all fall within 12.7 -+ 1.0 and 15.0 +- 0.5 m.y. This is in accordance with Baksi's findings, that analyses of ground samples of

whole basalt from the Columbia Plateau yield closely grouped results [4,5]. We illustrate this graphically in Fig. 2, where the integrated ratios for all the whole rocks are displayed on a plot of 4°Ar/36Ar versus 39Ar/ 36Ar. These ratios are obviously highly correlated and the best straight line through the set [9] corresponds to a common age for the samples of 14.3 m.y. and a 4°Ar/36Ar ratio in the contaminating argon of 295.1. Consideration of the weighted sum of the squares of the residuals about the line shows that the points are in fact scattered significantly more than would be expected on the basis of the mass spectrometry errors alone. If, however, allowance is made for the flux gradient during irradiation, the scatter about the line falls within expected experimental error and the best estimates of age and intercept become 14.3 -+ 0.3 m.y. and 295.1 -+ 0.9. Any systematic error in the age due to uncertainty in the age of the standard should not exceed 2%. Within these error limits this age cannot be distinguished from the average of all four conventional

TABLE 3 Effects of grinding on atmospheric 4°Ar Sample

Lab. no.

Atmospheric 4°Ar (× 10-s cm3 STP/g)

Age (m.y.)

Sample form *

NKI-li

17-25 62-69 42-49 29-33 70 77 131 106 113 98-105

0.125 0.097 0.225 1.77 0.066 0.770 0.092 0.108

14.5 -+0.3 15.0 -+0.3 13.9 ± 0.2 13.9 -+0.6 16.0 ± 0.7 22.1 ± 2.0 ** 14.6 _+0.6 12.7 + 0.3

chunks powder powder chunks powder chunks powder powder

NK11-2i NK14-1

* Powder (B) = powdered before irradiation; powder (A) = powdered after irradiation. ** Age by conventional K Ar analysis. Remaining ages are all 4 0 Ar/ 3 9 Ar integrated ages. he = 0.584 X 10-1° yr-1; h# = 4.72 X 10-1° yr-1 . 4°K/K = 0.0119 atom%.

(B) (A) (B) (B) (A)

83 80C

143~

zoc

6oc 5oc 4OAr 36Ar 4c(

~

,-'[ 80-87 t25-130

62-69

42-49

70-77 106 115 114-120

29-54

20c

~IINTERCEPT

295 I

ioc I tO00 39 Ar K

56Ar Fig. 2.4°Ar/36Ar versus 39ArK/36Ar correlation plot for the integrated whole-rock ratios. 1a errors are either less than the black dot in size or are shown by error bars. The 39ArK/36Ar ratios have been normalized to allow for the different fluxes in the various irradiations.

ages reported by Watkins and Baksi [5]. Given this degree of consistency in age among the samples it might well be anticipated that the 4°Ar-39Ar age spectra would be good approximations to horizontal straight lines. In fact, the interesting new result is found that all the whole rock spectra exhibit departures from good plateaux (see Fig. 1). Only two samples (NK16-2i, runs 123-130; NKI-li, runs 17-25)even approach the appearance of reasonable plateaux. The remainder shows spectra with pronounced structures. Despite variations in detail, all the spectra display a diminution of age at higher temperatures. In some samples (NK14-1, 11-2i) this effect is extremely pronounced. Yet despite this age spectra evidence of isotopic disturbance, there remains the concordance of the integrated ages and the degree of correlation seen in Fig. 2. Such spectra could result from closed-system disturbances of varying degrees as a result of which some 4°At* was removed from its original retentive sites to loosely bound positions. Thus, in the most perturbed spectra, those of NK14-1 (Fig. 1), the highest low-temperature age ( 1 6 - 1 9 m.y.) and the lowest high-temperature ages ( 5 - 7 m.y.) are found. The mechanism for such a supposed closed system

redistribution is not yet clear. The Owyhee samples are principally composed of plagioclase, pyroxene, olivine and some glass. The pyroxenes and olivines show varying degrees of alteration to what may be chlorophaeite. Removal of 4°Ar* during this alteration to relatively mobile sites may explain the spectra. Some support for this comes from the observation that the sample displaying the least perturbed spectrum (NK 1-1 i) is least altered, while the one with most chlorophaeite (?) (NK14-1) has the most disturbed spectra (Fig. 1). Somewhat similar spectra, with higher than "expected" low-temperature ages and lower than "expected" high-temperature ages were reported for a few samples of matrix and microbreccia clasts from Apollo 14 breccias by Turner et al. [10] and York et al. [1 I]. The former noted that despite the spectra shapes, the integrated ages agreed with R b - S r internal isochron ages on Apollo 14 basalts. They therefore suggested that there might have been closed-system redistribution of argon in the samples. The Owyhee sampies, therefore, present further evidence that such redistribution without argon loss is possible. If these postulated argon migrations occurred in the pre-laboratory history of the samples, then their observations in such differing lunar and terrestrial environments suggest closed system natural disturbances may not be uncommon. However, it is also possible that the perturbations were produced during the neutron irradiation. Further experiments involving mineral separates are needed to resolve the problem. The results presented emphasize the fact that there are interesting problems of interpretation still to be unravelled in 4°Ar-39Ar dating. The work of Berger [12] and others [13] demonstrates that good plateaux on age spectra may not always be immediately interpreted to give the dates of significant events. The results on the Owyhee samples suggest that apparently very complicated spectra may in fact sometimes be integrated to give useful ages.

Acknowledgments The Owyhee material was kindly provided by Professor N.D. Watkins. W.J. Kenyon and R.J. Doyle assisted irf the laboratory and J.A. Hanes made helpful comments on the thin sections. Financial support was provided by the National Research Council of Canada.

84 References 1 J. Gray and L.R. Kittleman, Geochronology of the Columbia River basalt and associated floras of eastern Washington and western Idaho, Am. J. Sci. 265 (1967) 257. 2 D. York, N.D. Watkins and A.K. Baksi, The Columbia volcanic episode: evidence for major volcanism during a limited part of the middle Miocene (abstract), Trans. Am. Geophys. Union 52 (1971) 382. 3 A.K. Baksi, D. York and N.D. Watkins, The age of the Steens Mountain geomagnetic polarity transition, J. Geophys. Res. 72 (1967) 6299. 4 A.K. Baksi, Isotopic fractionation of a loosely held atmospheric argon component in the Picture Gorge basalts, Earth Planet. Sci. Lett. 21 (1974) 431. 5 N.D. Watkins and A.K. Baksi, Magnetostratigraphy and oroclinal folding of the Columbia River, Steens, and Owyhee basalts in Oregon, Washington, and Idaho, Am. J. Sci. 274 (1974) 148. 6 R.L. Christiansen and P.W. Lipman, Cenozoic volcanism and plate-tectonic evolution of the Western United States,

7 8

9 10

11

12

13

I1. Late Cenozoic, Philos. Trans. R. Soc. London., Ser. A, 271 (1972) 249. N.D. Watkins, Paleomagnetism of the Columbia Plateaus. J. Geophys. Res. 70 (1965) 1379. E.K. Mak, D. York, R.A.F. Grieve and M.R. Dence, The age of the Lake Mistastin crater, Labrador, Canada (submitted to Earth Planet. Sci. Lett.). D. York, Least squares fitting of a straight line with correlated erros, Earth Planet. Sci. Lett. 5 (1969) 320. G. Turner, J.C. Huneke, F.A. Podosek and G.J. Wasserburg: 4°Ar-39Ar ages and cosmic ray exposure ages of Apollo 14 samples, Earth Planet. Sci. Lett. 12 (1971) 19. D. York, W.J. Kenyon and R.J. Doyle, 4°Ar-39Ar ages of Apollo 14 and 15 samples, Geochim. Cosmochim. Acta, Suppl. 3, 2 (1972) 1613. G.W. Berger, 4°Ar/39Ar step heating of thermally overprinted biolite hornblende and potassium feldspar from Eldora, Colorado, Earth Planet. Sci. Lett. 26 (1975) 387. R.J. Pankhurst, S. Moorbath, D.C. Rex and G. Turner, Mineral age patterns in ca. 3700 m.y. old rocks from West Greenland, Earth Planet. Sci. Lett. 20 (1973) 157.