JOURNAL
OF GEOPHYSICAL
RESEARCH,
VOL. 101, NO. B6, PAGES 13,689-13,700, JUNE 10, 1996
Geochronology of Galfipagos seamounts
ChristopherW. Sinton, David M. Christie, and Robert A. Duncan
Collegeof OceanicandAtmospheric
Sciences
OregonStateUniversity,Corvallis
Abstract. Lavas from seamounts
of the centralGalfipagosPlatformand the CarnegieRidge
increasein age with distancefrom the westernedgeof the platform,consistent
with a hotspot
modelfor the generationof thesefeatures.The arealdistributionof seamountagessuggests
that
thepatternof dispersedvolcanismseenon the presentislandsalsoprevailedbetween5 and6 Ma.
Age-distance
relationships
areconsistent
with a decrease
in the velocityof the Nazcaplaterelative
to the Galfipagoshotspotduringthe past9 m.y., mostlikely at -5 Ma whentherewasa changein
Pacific-hotspotmotion. Alternatively,the platevelocityremainedconstantandvolcanismat a
singleseamountenduredup to 7 m.y. Lavasfrom seamounts
alongthe Wolf-Darwin lineamentto
the northwestof the Galfipagos
Platformareyoung(<1 Ma), with theyoungestmeasuredages
closestto the platform.
Introduction
In addition to the emergent volcanoes,there are numerous
seamountswithin the Galfipagosregion. In this paper, we
The Galfipagos Archipelago, located just south of the
Galfipagos Spreading Center (GSC) in the eastern Pacific
Ocean (Figure 1), is the present volcanic expressionof the
Galfipagos hotspot. Active volcanism is concentrated on,
though not restricted to, the western Galfipagos Platform
islandsof Fernandinaand Isabela. The clusteringof epicenters
of recent earthquakesindicates the present position of the
hotspotbeneath Isla Fernandina,the largest and most active
presentresultsof K-Ar and40Ar-39Ardatingof dredgedlavas
0148-0227/96/96JB-00642509.00
seamount
from seamountsand submarineescarpmentsof the Galfipagos
Platform and western Carnegie Ridge. At 5-6 Ma, lavas
erupted over a region at least 140 km east-west,suggesting
that the present broad distribution of active volcanism has
persisted for at least 6 m.y. Seamounts of the central
Galfipagosplatform and western Carnegie Ridge increasein
age eastwardfrom the westernedge of the platform, consistent
volcanoon the platform. High 3He/4Hein someFernandina with a hotspot model for their formation. Age-distance
lavas also implies proximity to the hotspot[Kurz et al., 1993;
relationshipsof islandsand seamountson the centralplatform
Graham et al., 1993]. However, unlike a "simple" hotspot, indicateeither a decreasein Nazca plate velocity relative to the
such as Hawaii, where active volcanismoccursin a relatively
hotspotor an increasein the durationof volcanismat a single
restrictedarea, the Galfipagoshotspotdisplaysa dispersedand
seamount.
Seamounts
of the Wolf-Darwin
lineament
are
widespread pattern of volcanism; at least nine widely
young (<1 Ma), with the youngestmeasuredagesclosestto the
distributedvolcanoeshave been historically active (Figure 1)
platform.
[White et al., 1993]. East of Fernandina, there are five active
volcanoeson Isla Isabela, and farther east, lavas have erupted
Sample Description
within the last centuryon Isla Santiagoand within the last few
thousandyears on Islas Floreana and Santa Cruz [White et al.,
Samplesusedin this studywere obtainedduringLeg 2 of the
1993]. The easternmostisland, San Cristobal, displays a -3
PLUME (PL) expedition of the R/V Thomas Washington
m.y. span of intermittent volcanism[Geist et al., 1985; White
et al., 1993]. Despite the dispersed pattern of volcanism (Figure 1). Table 1 presentsthe sample locations, dredge
depths, and brief petrologicaldescriptions. Dredge sites on
displayed by the islands, the age of earliest volcanic activity
the Carnegie Ridge (PL1) and the easternGalfipagosPlatform
increasestoward the east, consistentwith Nazca plate motion
(PL4-PL8 and PL17) are all seamounts. The samplesfrom
over a fixed mantle plume [Cox and Dalrymple, 1966; Bailey,
these
sites are moderately altered basalts notably lacking
1976; White et al., 1993].
pillow
rim glass. They are predominantlyaphyric with some
To the north of the larger, central islands is a group of
smaller islands (Figure 1). Both Isla Pinta and Isla Marchena plagioclase-phyric lavas (PL4). Dredge site PL2 is on an
escarpment that, based on geographic position and
have eruptedwithin the last century and the youngestlavas on
bathymetry,
appearsto be normal oceanic crust generatedat
Isla Genovesa are probably only a few thousand years old
the GSC. Basalts from an escarpmentsouthwestof Isabela
[White et al., 1993]. Wolf and Darwin Islands, •-150 km north
of Isabela, are part of a NW trendinggroup of seamountsand (PL22, PL23, and PL24) are very fresh, plagioclase-olivineislandstermed the Wolf-Darwin lineament. K-Ar agesof lavas phyric basalts with glassy rims. Dredge PL20 recovered
variably altered, rounded, basalt pebbles from a small
from Wolf and Darwin islands are 0.9-1.6 Ma and 0.4 Ma,
seamountto the east of thesedredgesites.
respectively [White et al., 1993].
Five dredges from the Wolf-Darwin lineament (PL26 to
PL30) recovered very fresh, aphyric and plagioclaseCopyright
1996bytheAmerican
Geophysical
Union.
ultraphyric tholeiitic basalts with glassy pillow rims. Sites
PL26, PL27, and PL28 are on a large, elongate, composite
Paper number96JB00642.
13,689
at the SE end of the Wolf-Darwin
lineament
and
13,690
SINTON
ETAL.:GEOCHRONOLOGY
OFGALAPAC•S
SEAMOUNTS
3øN
0.8, 0.9
• '2 Ma •
2øN
'Darwin
•'•" '3 Ma
'6 Ma
0.3
GalfipagosSpreadingCenter
IøN
0.3
0.1
(---'--••2000m--
c':•he:'&7
14
•-3•- - -3Ma
Genøves•/'-•l
-.......
o
Santiago 16_
/'-•(' 113.2
11.1
Fernandina•
• '"'"'•'••antaCr••uzlOO0.,
10
• 5.• •
pi•'"•"•"••
o
•
"%
5.6
•
2
5
SantaFe
•S•:Cri:ra'
•7.3• 7.8••
løS
•.....
Isabela
23
Espafiola l/m
5 N •
"•1oreana
..:o 5.3,•.8/5 •••2000m
•
DSDP
157
O
20
2øS
92 ø
8.7,9.0
•3000m_
90øW
88 ø
86 ø
Figure 1. The GalfipagosArchipelago showing PLUME2 dredge sites (solid circles) and DSDP Site 157
(open circle). Dredge site numbersare labeled and radiometricagesare in italics. Active emergentvolcanoes
are representedby solid triangles. The inset is a magnification of the Wolf-Darwin lineament. The light
dashed lines are approximate positions of magnetic anomaly isochrons, and the heavy dashed line is a
pseudofaulttrace from the westwardpropagationof the GalfipagosSpreadingCenter (GSC) [Wilson and Hey,
1995]. The pseudofaultjuxtaposesoceaniccrust with ,--1.8m.y. differencein age. Bathymetry and location of
the GSC are basedon work byWilsonand Hey (1995) and our unpublishedSeaBeamdata.
sites PL29 and PL30 are on small, conical seamounts located
towards the NW end of the Wolf-Darwin
and Darwin Islands(Figure 1).
lineament near Wolf
of the older rocks (sites PL1-PL8) were moderately altered
throughout. For a suite of rocks from the same dredge,we
generally selected the most differentiated samples(highest
to assuremeasurable
radiogenic
40At). TheK20
Tholeiitic pillow basalts from three seamountsoff the K20 contents
contentsof the site PL13 lavas were too low K20 (0-0.06 %)
northeasternedge of the platform (PL9-PL14) range from
slightly to moderatelyalteredwith both glassyand glass-free for radiometricdating.
We used both conventional K-At and 40Ar-39Ar incremental
pillow rims. Roundedcobblesfrom sites PLll and PL13
indicate that these seamounts were at one time subaerial
heating methods. One problem associated with dating
[Christie et al., 1992].
Lavas from the northeastern submarine basalts is that low-temperature alteration in
seawatermay causeK additionor radiogenic
40Atlossfromthe
seamountsare rather primitive, and most have MgO > 8% in
glassand relatively low K20 abundances
(<0.10% [Sinton, sample, resulting in measuredages that are significantlyless
1992]). Most are aphyric, though several plagioclase- than the crystallizationages. In this respect,40Ar-39Ar
ultraphyric or olivine-spinel-phyric lavas were recovered. incremental heating experiments are consideredto be more
Two plagioclase-phyricbasalts and an olivine-plagioclase- reliable in dating altered submarinerocks [e.g., Duncan and
phyric basaltwere recoveredfrom dredgesite PL16, which was Hargraves, 1990]. The lower-temperature steps release Ar
located on a small seamount southeast of Genovesa.
from grain boundariesand other low-retentionsitesin the rock
(fine-grained matrix and alteration phases,such as clays and
zeolites) while higher-temperature steps release Ar from
unaltered crystal interiors, giving Ar isotopic compositions
Analytical Methods
that more closely reflect the crystallization age. With these
Where possible, we took samplesof the freshestand best criteria in mind, we analyzedthe fresh lavas from the southern
by K-Ar. All othersampleswereanalyzedby 40Arcrystallized material, as far from the pillow rim as possible escarpment
and from the least alteredportion of the rock. However, some 39At,andsomewereanalyzedby bothmethods.
SINTONETAL.:GEOCHRONOLOGY
OFGAL,•d:'AGOS
SEAMOUNTS
13,691
Table1. SampleLocations
andDescriptions
Sample
PLl-12
PL 1-46
Location
?
Description
02 ø 01.47'S
85 ø 40.34'W
02 ø 01.47'S
Depth*
(m)
seamount;aphyric;vesicular
2570-2137
seamount;aphyric;vesicular
2570-2137
escarpment;
aphyric;slightlyvesicular
2610-2140
seamount;largeplagioclasephenocrysts;
slightlyvesicular
1882-1250
seamount;aphyric;vesicular
875-570
seamount;aphyric;highlyvesicular
1004-917
seamount;plagioclaseandolivinemicroglomerocrysts;
vesicular
1425-935
seamount;aphyric;vesicular
1141-1006
seamount;aphyric;vesicular
1309-702
seamount;sparseplagioclaseandolivinemicrophenocrysts;
vesicular
1378-1123
seamount;sparseplagioclaseandolivinephenocrysts;
non-vesicular
1639-1090
seamount;plagioclaseandolivine microphenocrysts;
non-vesicular
1688-1320
seamount;
plagioclase-phyric;
non-vesicular
640-294
seamount;aphyric;highlyvesicular
640-294
seamount;aphyric;slightlyvesicular
965-310
escarpment;
sparseplagioclaseandolivinephenocrysts;
non-vesicular
1842-1245
escarpment;
sparseplagioclaseandolivinephenocrysts;
non-vesicular
3397-3356
escarpment;
plagioclaseandolivinephenocrysts;
slightlyvesicular
3039-2399
seamount;sparseplagioclasemicrophenocrysts;
vesicular
851-500
seamount;aphyric;vesicular
1650-1464
seamount;
plagioclase-phyric;
vesicular
1454-911
seamount;aphyric;slightlyvesicular
1863-1582
seamount;
aphyric;slightlyvesicular
1863-1582
seamount;
non-vesicular;
sparseplagioclase
andolivine
2395-2116
85 ø 40.34'W
PL2-02
PL4-19
PL5-01
PL8-01
PL9-50
PL 10-05
PL 11-2
PL12-1
PL 14-1
PL 16-02
PL17-01
PL17-04
PL20-02
00 ø 22.40'S
85 ø 09.70'W
00 ø 55.92'S
7 ø 07.46'W
01 o 17.40'S
88 ø 21.20'W
01 ø 17.40'S
88 ø 21.20'W
00 ø 28.13'S
88 ø 32.11'W
00 ø 34.04'S
88 ø 35.44'W
00 ø 11.70'S
88 ø 40.10'W
00 ø 14.45'S
88 ø 38.20'W
00 ø 06.00'N
89 ø 05.80'W
00 ø 06.44'S
89 ø 49.26'W
01 ø 11.38'S
89 ø 06.58'W
01 ø 11.38'S
89 ø 06.58'W
01 ø 37.73'S
90ø 10.70'W
PL22-01
PL23-01
PL24-01
PL26-25
PL27-17
PL28-10
PL29-02
PL29-48
PL30-01
01 ø 18.23'S
90 ø 44.92'W
01 ø 20.03'S
90 ø 59.84'W
01 ø 18.60'S
90 ø 55.60'W
00 ø 50.50'N
91 ø 17.30'W
00 ø 53.80'N
91 ø 19.80'W
01 ø 00.19'N
91 ø 33.34'W
01 ø 54.50'N
92 ø 10.70'W
01 ø 54.50'N
92 ø 10.70'W
01 ø 33.63'N
91 ø 55.66'W
microphenocrysts
in glassy
matrix
?Position
on-bottom.
*On-bottom
tooff-bottom
depths.
Most K-Ar and40Ar-39Arsamples
werecrushedin bulk in a
ceramic jaw crusher and sieved to a uniform 0.5-1 mm grain
at least 12 hours. Samples were heated in outgassedMo
crucibles by radio frequency induction. K-Ar sampleswere
size. The exceptions
arethe40Ar-39Aranalysesof PL26-PL30 completelyfusedin one stepand the 40Ar_39Arincremental
rocks, for which we used 100 mg minicores. All sampleswere
ultrasonically cleaned in distilled water. For the K-At
analyses,an aliquot of samplewas powderedand analyzedfor
K20 by atomic absorptionspectrophotometry
and about 5 g of
granularmaterial was fused and the releasedgas was measured
by argon isotopic dilution [Dalrymple and Lanphere, 1969].
heating samples were heated in progressively higher
temperature
stepsuntil fusion. Ideally,each40Ar_39Ar
sample
should have as many heating steps as possible without
jeopardizing the precision of each step. Becauseof the low
K20 and relatively young age of the samples, they were
analyzedin four or five heatingsteps. Ar compositions
were
For the 40Ar-39Aranalyses,
0.5-1.0 g of eachgranularsample then measuredusingan AssociatedElectricalIndustries(AEI)
and the mini-coreswere sealedin an evacuatedquartzglassvial
and irradiated for 6 hours at the Oregon State University
TRIGA nuclear reactor facility. We monitored neutron flux
during irradiation using the Mmhb-1 hornblende (520.4 Ma
[Samson and Alexander, 1987]) and FCT-3 biotite (27.7 Ma
[Hurfurd and Hammerschmidt,1985]) standards.
For both the K-Ar and 40Ar-39Arexperiments(exceptthe
minicores), Ar extraction was done in a conventional glass
extraction systemafter the systemwas baked out at 180øC for
MS-10S mass spectrometer.
Each minicore was heated in a temperature-controlled
resistance furnace.
In order to minimize atmospheric
contamination,the sampleswere preheatedto 450øC and the
releasedgaswaspumpedaway. Thenthe sampleswereheated
in five sequential150øCstepsfrom 600ø to 1400øC(fusion).
Ar from the minicoresampleswasmeasured
usinga low-blank
Mass Analyzer Products(MAP) 215-50 massspectrometer.
K-Ar ageswere calculatedusingthe methodsdescribedby
13,692
SINTONETAL.:GEOCHRONOLOGY
OFGAL.4PAC•S
SEAMOUNTS
Table2. Results
from
K-Arand4øAr-39Ar
Radiometric
Dating
Sample
K-ArAge*
_+1o (Ma)
PLl-12
PL1-46
PL2-02
PL4-19
PL5-01
PL8-01
PL9-50
PL10~5
PL11~2
5.78 + 0.43
5.07 + 0.22
0.89 + 0.07
----1.10+ 0.11
--
PL12~l
PL14-1
0.63 +0.11
8.95 + 1.65
PL16-2
PL17~l
PL17-4
PL20-2
-~~~
~~
PL22-1
PL23~l
PL24-1
1.69 + 0.06
0.18 + 0.04
0.14 + 0.02
PL26-25
PL27~17
PL28-10
PL29~2
PL29-48
PL30-1
0.03
0.13
0.34
0.03
1.60
0.25
+
+
+
+
+
+
0.13
0.16
0.17
0.05
0.07
0.05
Plateau
Age*
_+1o(Ma)
8.7 +
9.1 +
11.1 +
7.8 +
8.0 +
7.3 +
5.7 +
5.6+
3.2 +
0.4
0.4
1.4
1.2
1.5
0.5
0.6
1.4
0.5
%39Ar
100
100
80.5
100
100
100
86.2
82.8
100
Isochron
+ 1c•(Ma)
8.1 + 0.5
9.0 + 0.1
10.2 + 1.6
10.4 + 5.5
8.0 + 8.3
7.0 + 0.3
4.2 + 2.1
6.0+0.4
3.2 + 0.3
N?
4
4
3
4
5
4
3
3
5
SUMS/ 40Ar/36Ar
iõ
(N-2)
0.63
0.05
0.17
0.50
6.30
0.02
0.74
0.01
0.88
+ 1c•
297.0 + 1.9
295.7 + 1.2
298.6 + 13.3
292.3 + 2.0
287.2 + 4.2
296.2 + 7.3
301.0 + 9.8
293.9+8.8
296.6 + 2.9
............
............
5.6
5.3
5.8
2.8
+
+
+
+
0.3
0.4
0.8
0.3
0.1
0.1
0.1
0.1
0.2
0.1
624
624
__
453
314
330
295
295
280
__
__
93.8
95.2
100
94.3
5.2 +
5.1 +
5.5 +
3.1 +
0.1
0.3
0.6
0.2
4
4
5
4
0.05
0.51
0.63
0.13
302.6
298.0
298.0
291.6
+
+
+
+
6.2
7.0
7.0
7.6
87.7
68.1
92.0
47.9
66.7
52.4
-0.1 +
-0.2 +
0.4 +
0.7 +
1.2 +
4.3 +
0.1
0.1
0.5
0.2
1.4
1.7
4
3
4
3
3
3
2.08
0.02
0.94
0.29
1.50
0.82
299.1
296.2
293.3
293.6
290.5
276.0
+
+
+
+
+
+
1.9
3.1
3.0
4.4
4.7
3.4
............
............
............
0.1 +
0.3 +
0.3 +
0.9 +
0.8 +
1.7 +
Distance(km)
165
235
235
107
43
10
16
__
*Ages
are
calculated
using
)•e=
0.581
x10© yr-1and
)•[•=
4.962
x10
-10
yr-1;
K-Ar
ages
are
calculated
using
40K/K=l.167x10-4
mol/mol.
tNumber
ofsteps
used
intheplateau
and
isochron
agecalculations.
õInitial
40Ar/36Ar
based
ontheinverse
ofthey-intercept
oftheregression.
Dalrymple and Lanphere [1969] after correction for mass PL26-PL30. The northeast volcanoes are the islands of Pinta,
fractionation(Table 2). Individual agesfor each 40Ar-39Ar Marchena, and Genovesa and the seamountdredge sitesPL9step were calculated using correctionsfor background,mass PL14. The rest of the islandsand dredgesites(exceptPL2) are
fractionation, and isotopic interferences. From the heating included in the central platform and western Carnegie Ridge
steps,agespectrum
diagrams(ageversus% 39Arreleased)were (Figure 1). The PL9-PL10 seamount,which could be in either
generated (Figure 2). We calculated plateau ages from of the latter two regions, is included with the northeast
consecutivestepsthat are concordantwithin lc• errorusingthe volcanoes becauseits lavas are similar in compositionto the
proceduredescribedin Dalrymple et al. [1988], in which step other volcanoesin this group [Sinton, 1992].
ages were weighted by the inverse of their variance (Table 2).
Ridge
It is possible,however, that the lavas did not completely The Central Platform/Carnegie
equilibratewith an atmospheric
Ar composition
(40Ar/36Ar=
SamplesPLl-12 andPL1-46 gavereliable40Ar-39Aragesof
295.5) upon crystallization. We investigatedthis possibility
by plotting isotope correlation diagrams (36Ar/40Ar versus about 9 Ma with statistically indistinguishable plateau and
initial 40Ar/36Arratios
39Ar/40Ar)in whichthe slopeis proportional
to the age and isochronagesand nearlyatmospheric
(Table
2
and
Figure
2).
The
K-Ar
ages
from the samesamples
the inverse of the y-intercept gives the initial 40Ar/36Ar
are,
however,
significantly
younger
than
the 40Ar-39Arages.
compositionin the crystallizedrock. For each sample,steps
from the plateau age were used in the regression. The
goodness-of-fitparameter SUMS [York, 1969] has a Z 2distributionwith (N-2) degreesof freedom,N beingthe number
of steps included in the isochron regression(Table 2). For
example,a value of SUMS/(N-2) = 2.6 for N=5 shouldonly be
exceeded for one sample out of 20 based on Gaussian
distributionof measurement
error; analogousvaluesare 3.0 for
N=4 and 3.8 for N=3. Of all the samples,only PL5-1 exceeded
the 95%
Results
confidence
limit.
Judging from the altered condition of the samples (clays
filling vesicles and cracks), it is likely that the K-Ar ages
reflect lossof radiogenic
4øArand/oradditionof secondary
K20 duringlow-temperature
alteration,andtherefore
the40Ar39Ar agesare the more accurate. Planktonicforaminiferain
the sedimentsoverlying the basaltic basement at Deep Sea
Drilling Project (DSDP) Site 157 (near site PL1) are of the
Globorotalia plesiotumida zone [Kaneps, 1973], giving a
minimumage of 11 to 6 Ma [Berggrenet al., 1996]. However,
poor preservation in the sediments above the basement
suggests that the basaltic crust could be older. This is
indicatedby magneticisochronsfrom Lonsdale and Klitgord
[1978]
which show that the PL1 seamount overlies crust of
For the purposesof discussion,we divide the Galfipagos magnetic chron 5Cn (17 to 16 Ma) [Berggren et al., 1996].
Platform into three regions: the Wolf-Darwin lineament, the Assumingthat the magneticisochronsaccuratelyrepresentthe
northeast seamounts and islands, and the central
age of the oceaniccrust,the PL1 lavas are 7 to 8 m.y. younger
platform/Carnegie Ridge (Figure 1). The Wolf-Darwin than the underlying crust.
lineamentincludesWolf and Darwin Islandsand dredgesites
Samples from dredge sites on the central platform (PL4,
SINTONET AL.:GEOCHRONOLOGY
OFGAL/•PAC•SSEAMOUNTS
Basalt
PLl-12
Basalt
15.0
15.0
12.0
12.0
9.0
9.0
6.0
(D
8.7+/-0.4 Ma
e•o
<
3.0
'
0.0
10
'
20
'
'
30
40
'
'
50
60
'
70
'
80
PL 1-46
9.1 +/- 0.4 Ma
6.0
3.0
'
90
13,693
i
0.0
100
0
i
10
20
i
i
30
40
i
50
i
60
I
i
I
70
80
90
100
%39Ar released
%39Ar released
0.005
0.005
-
8.1 +/- 0.5 Ma
/
•
-
9.0 +/- 0.1 Ma
,.
0.000
0.000
0.00
i
i
i
PL2-2
Basalt
50.0
20.0
40.0
16.0
i
i
i
i
i
0.02
39Ar/40Ar
0.00
0.08
Basalt
i
PL4-19
7.8 +/- 1.2 Ma
30.0
12.0
20.0
•
11.1 +/-1.4
8.0
-
4.0
lO.O
I
0.0
0
10
!
i
i
i
i
i
i
i
20
30
40
50
60
70
80
90
0.0
100
0
10
%39Ar released
20
30
40
50
60
70
80
90
100
%39Ar released
0.005
0.005
<
10.4 +/- 5.5 Ma
10.2+/1.6Ma/
0.000
'
0.00
'
'
'
0.000
.............
39Ar/40A
r
0.02
ß
0.00
39Ar/40A
r
0.04
Figure 2. Stepagespectraandisotopecorrelation
diagrams
for each40Ar-39Aranalysis.The lengthof each
stepagerefersto the fractionof the total39Arreleasedandthe thickness
of the stepis the + l c•error. The
light horizontallinesadjacentto the plateauagespanthe stepsusedin the agecalculation.The solidsquares
in the correlationdiagramsindicatethe datapointsusedin a weightedlinear regressionto obtainan isochron
age. Open squareswere excludedfrom the regression.The inverseof the y-interceptof the regressiongives
the initial 40Ar/36Ar ratio.
13,694
S•TON ETAL.:GEOCHRONOLOGY
OFGALAPAGOS
SEAMOUNTS
Basalt
Basalt
PL5-1
20.0
20.0
16.0
16.0
12.0
12.0
PL8-1
7.3 +/-0.5
Ma
8.0
8.0
4.0
i
i
i
i
i
i
10
20
30
40
50
60
0.0
0.0
10
20
30
40
50
60
70
80
90
100
i
70
i
i
80
90
100
%39Ar released
%39Ar released
0.005
0.005
8.0 +/- 8.3 Ma
7.0 +/- 0.3 Ma
0.00O
i
i
i
i
0.00
i
i
i
0.000
i
i
Basalt
I
I
I
0.00
0.15
39Ar/40Ar
,
I
I
'
i
•
0.04
39Ar/40Ar
PL9-50
Basalt
PL10-5
80.0
20.0
1
16.0
5.7 +/-0.6
12.0
48.0
8.0
32.0
4.0
16.0
5.6 +/- 1.4
0.0
0.0
10
20
30
40
50
60
70
80
90
10
100
20
30
40
50
60
70
80
90
I(X)
%39Ar released
%39Ar released
0.005
0.005
<
-
-'u
<
4.2+/-2.1
0.000
I
0.00
I
I
I
I
Ma
I
39Ar/40Ar
'
'
'
0.000
0.06
Figure 2.
PL5, PL8, PL16, andPL17) yieldedreliable40Ar-39Arages,
althoughthe isochronagesfromPL4-19andPL5-1havehigh
uncertainties
(Figure2 andTable2). LavasfromPL22, PL23,
andPL24 were analyzedby K-Ar only. Althoughthereare no
internal checkson the reliability of their measuredages,the
resultsare mostlikely meaningfulbecauseof the freshness
of
the samples,the high K20 content,and the similarityof the
agesto nearbyIsabelaIsland[Whiteet al., 1993].
I
I
I
0.00
6.0+/-0.4
Ma
'
•
'
39Ar/40Ar
'
'
I
O.Ol
(continued)
Northeast
Volcanoes
Because of the low K20 contents of the northeastern
seamountlavas,40Ar-39Arexperiments
wereperformedon the
mostdifferentiatedsamplesfrom PL9, PL10, PL11, andPL16.
K-Ar analyseswere also performedon samplesfrom PL10,
PL12, and PL14. The K-Ar agefrom sitePL10 is youngerthan
thecorresponding
40Ar-39Ar
age,soweusethe40Ar-39Ar
age
SINTONET AL.'GEOCHRONOLOGY
OFGAL/•tPAGOS
SEAMOUNTS
Basalt
Basalt
PLll-2
13,695
PL16-2
35.0
20.0
28.0
15.0
3.2 +/-0.5
Ma
I
21.0
lO.O
14.0-
5.0
5.6 +/- 0.3
0.0
7.0
0.0
-5.0
0
10
20
30
40
50
60
70
80
90
0
100
'
'
'
'
'
'
10
20
30
40
50
60
'
'
'
70
80
90
100
%39Ar released
% 39Ai•elei/S6d
0.005
0.005
•
5.2 +/- 0.1
3.2+/-0.3
Ma
/
0.000
i
i
i
i
i
i
i
i
I
39Ar/40Ar
0.00
Basalt
i0
0.000
0.20
I
I
I
0.00
I
I
•
I
•
39Ar/40A
r
PL17-1
Basalt
25.0
I
0.50
PL17-4
25.0
20.0
20.0
5.3 +/- 0.4 Ma
15.0
5.8 +/- 0.8 Ma
15.0
lO.O
10.0
5.0
5.0
o.o
0.0
10
20
30
40
50
60
70
80
90
0
100
10
20
30
40
50
60
70
80
90
100
%39Ar released
%39Ar released
0.005
0.005
'
0.000
5.1 +/-0.3
i
0.00
i
i
i
i
Ma
i
39Ar/40Ar
5.5 +/- 0.6 Ma
i
i
0.30
Figure 2.
as representativeof the crystallizationage. Basedon these
0.000
0.00
•
'
'
'
'
'
I
39Ar/40Ar
I
I
o.3o
(continued)
Ar is small, and so even a small amount of retained mantle-
derivedAr wouldhavea significanteffecton the measured
age.
formedby 5.7 Ma, 3.2 Ma, and5.6 Ma, respectively(Table 2).
Lineament
PL14-1 gavea K-Ar ageof 9.0 Ma, considerably
olderthanthe Wolf-Darwin
magneticage of the underlyingcrust(Figure 1) [Wilson and
The K-Ar agesfrom sitesPL26-PL28 (0.03-0.34 Ma) are
Hey, 1995]. This sample apparently contained excess
radiogenic40Ar. Becauseof the low K20 of thislava,andits within lo errorof the 4OAr-39Arages(0.1-0.3 Ma) (Table 2).
necessarilyyoung age, the amountof K-supportedradiogenic However,thetwo methodsgivedifferentagesfor the sitePL29
results, the PL9-PL10, PL11-PL12, and PL16 seamountswere
13,696
SINTONETAL.:GEOCHRONOLOGY
OFGAL/•tPA•S SEAMOUNTS
Basalt
Basalt
PL20-2
PL26-25
5.0
12.0
3.6
9.6
7.2
2.8 +/- 0.3
4.8
0.1 + 0.1 Ma--
O.8
-0.6
2.4
-2.0
0.0
10
20
30
40
50
60
70
80
90
100
10
20
%39Ar released
30
40
50
60
70
80
90
100
%39Ar released
0.005
0.005
•
3.1+/-0.2 Ma
/
-o.1 +/- o.1 Ma
0.000
0.00
'
I
I
I
I
•
•
1
i
39Ar/40Ar
Basalt
o.ooo
O. lO
ß
o.oo
0.20
39Ar/40Ar
PL27-17
Basalt
PL28-10
10.0
7.0
0.3 + 0.1 Ma
4.0
0.3 + 0.1 Ma -1.0
-1.0
-2.0
-3.0
-5.0
-5.0
0
10
20
30
40
50
60
70
80
90
0
100
l0
20
%39Ar released
30
40
50
60
70
80
90
100
%39Ar released
0.005
0.005
-0.2 +/-0.1
0.4 +/- 0.5 Ma
Ma
0.000
i
0.000
0.00
39Ar/40Ar
0.00
0.04
Figure 2.
and PL30 lavas. Two PL29 lavasyieldedK-Ar agesof 0.03 +
i
i
i
i
i
/
i
39Ar/40A
r
i
i
0.04
(continued)
0.05 and 1.60 + 0.07 Ma. The 40Ar-39Arage spectrafrom
samestepsas the plateausare concordantwith the plateauages
(within analytical uncertainty) and near-atmosphericinitial
both thesesamplesare disturbedand show evidenceof excess
40Ar/36Arratios. Together,the data indicatea crystallization
radiogenic40Ar in the first andlast steps. Despitethis,the age of 0.8-0.9 Ma for theselavas. The disturbedage spectrum
middlestepsfor both samplesgive similarintegratedplateau from PL30-1 and the disagreementbetweenthe K-Ar age (0.25
ages(0.9 + 0.1 Ma and0.8 + 0.2 Ma). Isochronagesusingthe
+ 0.05 Ma), the integrated age for the first two stepsof the
SINTONETAL.:GEOCHRONOLOGY
OFGAL/•PAGOS
SEAMOUNTS
Basalt
PL29-2
Basalt
13,697
PL29-48
10.0
9.0
8.0
6.(}
0.9 + 0.1 Ma
4.0
3.0
0.8 + 0.2 Ma-
l.O
-1.0
10
20
30
40
50
60
70
80
90
100
10
20
%39Ar released
30
40
50
60
70
80
90
i
i
100
%39Ar released
0.005
0.005
.
\
0.000
I
I
0.7 +/- 0.2 Ma
I
I
0.00
I
I
/
•
1.2 +/- 1.4 Ma
•
Basalt
0.00
0.05
39Ar/40Ar
I
0.000
i
i
i
i
i
i
i
/
0.04
39Ar/40Ar
PL30-1
20.0
16.0
12.0
8.0
-1.7+0.1
?Ma--
4.0
0.0
I
I
I
I
I
I
10
20
30
40
50
60
•
70
•
•
80
90
100
%39Ar released
0.005
\
0.000
0.00
ß
i
i
/
4.3 +/- 1.77 Ma
i
i
i
i
i
39Ar/40Ar
Figure 2. (continued)
initial 4OAr/36Ar
ratios(Table2 andFigure
4OAr-39Arspectrum(1.7 + 0.1 Ma), andthe isochronage(4.4 near-atmospheric
+ 0.5 Ma; from the first three steps)castsdoubt on whether 2). Like the PL1 samples,the K-Ar age is significantly
youngerthanthe4OAr-39Ar
age,andthereforewe consider
the
any of theseagesare meaningful.
4OAr-39Arageto representthe crystallization
age. Dredgesite
Site
PL2
PL2 is neara possible
pseudofault
according
to the magnetic
isochronmap of Wilsonand Hey [1995]. To the westof the
Sample PL2-2 gave a reliable 4OAr-39Ar age with site,the ageof the crustis 7-8 Ma (chron4n) [Berggrenet al.,
statisticallyindistinguishable
plateauand isochronagesand 1996], much youngerthan the 11.1 + 1.4 Ma age of the PL2
13,698
SINTONET AL.:GEOCHRONOLOGY
OF GALAPAC•S SEAMOUNTS
lava. About 70 km to the east, chron 5An (12.5-12.0 Ma) was
the farthest island from the hotspot (180-200 km) and, at 5-6
identifiedas by Lonsdaleand Klitgord [1978]. If the PL2 lava
Ma,
is associated with this crustal section, then it must have
Nevertheless, for the last 6 m.y., maximum age increases
linearly eastward from the hotspot; these maximum ages
define the hotspot-Nazcaplate velocity for the interval 6 to 0
Ma. In calculatingthis rate, the < 6 Ma dredge ages (PL11,
PL16, and PL22-PL24) are combinedwith the oldestK-Ar ages
originated at the Costa Rica rift.
Discussion
Past
Patterns
of
Volcanism
By examining the distribution of coeval lavas on the
Galfipagos Platform, we can assess the past patterns of
volcanism and compare these to the present pattern. Lavas
from three seamountsfrom the central platform (sites PL16,
PL9-PL10, and PL17) have ages between 5 and 6 Ma (Figure
1).
These locations are distributed over an area that is
comparable to the present distribution of active volcanism
(excluding the Wolf-Darwin lineament). This implies that
widespreadand dispersedvolcanismwas a characteristicof the
Galfipagoshotspot between 5 and 6 Ma, when the spreading
center was centeredover the hotspot[Hey et al., 1977; Wilson
and Hey, 1995]. We infer that dispersedvolcanismmay be a
long-term characteristicof the Galfipagoshotspot.
Hotspot-Nazca
Plate
Velocity
[Christie et al., 1992; White et al., 1993]. These data are
consistent with Nazca plate motion over a stationary mantle
plume presently located beneath Fernandina volcano and can
be used to estimate Nazca plate-hotspot velocities over the
past 9 m.y.
Previous estimates of the velocity of the Nazca plate
relative to the hotspot have been calculated by adding the
hotspot-Pacificmotion to the Pacific-Nazca motion. Minster
lavas
have
a broad
areal
distribution.
from each of the central islands (Floreana, 1.52 Ma; Pinzon,
1.4 Ma; and Santa Fe, 2.8 Ma) [White et al., 1993]. A
regressionweightedto the inverseof the error for eachsample
is fitted to the data pointsand the inverseof the slopeyields a
velocity of 39.2 + 0.2 km/m.y., within error of the 37+ 7
km/m.y. velocity calculated by Gripp and Gordon [1990].
Sample PL16 falls well to the left of this line, indicating that
it eruptedas much as 50 km west of the hotspot.
It is apparentfrom Figure 3 that the 39 km/m.y. line passes
well above someof the older lavas. This can be interpretedas
reflecting either a decreasein the Nazca-hotspotvelocity at 56 Ma or as an artifact of increasinglyprolongedvolcanismat
each site as well as the preferentialsamplingof youngerlavas
inherent in dredging. The dark shadedregion on Figure 3 is
the field of lavas that would have eruptedup to 180 km eastof
the hotspot (the approximate distance of San Cristobal from
Isla Isabela) assuming a constant 39 km/m.y. velocity over
time.
The ages of submarine samples from the central platform
and CarnegieRidge increasetoward the east (Figures 1 and 3),
extending the general age progression seen in the islands
seamount
While this region encompassesthe data points from
most of the submarine lavas, it does not include the site PL1
samples. For this site, a constant plate velocity of 39
km/m.y. would predict an age of 15.9 Ma (the approximateage
of the underlying crust) and, consequently,a 7 m.y. span of
volcanism.
So either
the duration
of volcanism
was much
longer at >9 Ma than it is presently or, alternatively, there has
been a decreasein the Nazca-hotspotvelocity.
A changein the hotspot-Pacificmotion at 5 Ma is evident
from a changein the Pacific-Juande Fuca relative motion and
the suddenonset of deformationand uplift along the San
Andreas fault [Cox and Engebretson,1985, and references
and Jordan [1978] calculateda velocity of 49 + 8 km/m.y. therein]. Cox and Engebretson[1985] calculated
the angular
using their Pacific-Nazca velocity from magneticanomalies velocityof hotspot-Pacific
rotationpolesas 0.73ø/m.y.(from
averagedover the past 3 m.y. and a hotspot-Pacificvelocity 19.5 Ma to 5 Ma) and 0.97ø/m.y.(5-0 Ma). They suggested
from K-Ar ages(<10 Ma) alongPacificislandchains. Gripp that this change in plate motion could be due to the
and Gordon [1990] useda differentPacific-Nazcavelocity detachmentof a subductingslabbeneaththe New Hebridesarc.
[DeMets et al., 1990] (also averagedover the past3 m.y.) and Becausethere was no apparentchangein the Pacific-Nazca
the samehotspot-Pacificvelocity [Minster and Jordan, 1978] spreadingrate betweenthe Oligoceneand the Plio-Pleistocene
to derive a Nazca-hotspotvelocityof 37 + 7 km/m.y., lower [Handschumacher,1976], the increasein the WNW movement
but within error of the Minster and Jordan [ 1978] estimate.
of the Pacific plate shouldhave resultedin a 32% decreasein
The Nazca-hotspotvelocity can be directly estimatedby the eastwardvelocity motion of the Nazca plate over the
plotting the ages of the Gal•pagos lavas relative to the Gal•pagoshotspot. Assuminga discretechangein plate
distancefrom the hotspot. BecauseGal•pagosvolcanismis motion and a constantlength-scaleof dispersedvolcanism
dispersed, simple backtracking of a sample site in the (-180 km), we can estimatethe Nazca-hotspot
plate velocity
direction of plate motion will not necessarilyarrive at Isla priorto 5 Ma from Figure3 by pivotingthe envelopeat 5 Ma
Fernandina. In addition,volcanismcan persistat a single until the shadedarea encompasses
site PL1. This yields a
island for at least 3 m.y. [White et al., 1993], so lavas can minimumplate velocityof about57 km/m.y. (Figure3). A
erupt well to the east of their contemporaneous
hotspot 32% decreasein this value givesa 5-0 Ma Nazca-hotspot
location. This can be a particularproblemwith the seamount velocityof 39 km/m.y. and agreeswith the changein plate
lavas becausethere is no stratigraphicinformation obtained motionput forth by Cox and Engebretson[1985].
by dredging.
Although the data are consistentwith a changein the
To producethe ageversusdistanceplot of Figure3, we make hotspot-Nazca
velocityat -5 Ma, this conclusion
hingeson
the assumption
thatall sampledlavaswereeruptedalonga line one criticalsite andfurtherdatingof lavasfrom the Carnegie
of longitude(91øW) throughthe eastcoastof Isla Isabela. The Ridge would be needed to test this assertion. The alternative
distancefrom this line to the sampleis measuredalongthe
azimuth of Nazca plate motion (102ø). Figure 3 showsthat
interpretation
of the age-distance
relationsis thatthe hotspotNazca plate velocityhas remainedconstantand Galfipagos
there has been a broad areal distribution of coeval lavas for at
hotspotvolcanismat 16 to 9 Ma was more dispersed
thanit
leastthe past5-6 m.y. Volcanismhasclearlybeenlong-lived was9-0 Ma. For SanCristobal,K-Ar agesrangefrom 2.35 to
at some locations,most notably Isla San Cristobal, which is
0.05 Ma [White et al., 1993], but if we assumethat the island
SINTON
ETAL.:GEOCHRONOLOGY
OFGAL/•AGOS
SEAMOUNTS
13,699
14-
12-
10-
_
PL16
_.
_
Seamount 40Ar-39Ar
._
I
100
200
300
400
500
I
Seamount K-At
ß
Island K-Ar
I
600
I
700
800
Distance (km)
Figure 3. Plot of age versusdistancefrom the eastcoastof Isla Isabelafor the dredgedlavasand the oldest
eastwardislandlavas. The datasetis madeof thedredge40Ar_39Ar
ages(opensquares)
andK-Ar ages(solid
squares)andthe K-Ar agesfrom the islands(solidcircles[Whiteet al., 1993]). Errorbarsare lcr uncertainty.
The solidline is a weightedregression
fitted to the < 6 Ma dredgeagesandthe islandages. The darkershaded
region is the field of lavas that would have eruptedup to 180 km from Isla Isabelaassuminga constant
velocityof 39 km/m.y. The lightershadedregionrepresents
the possibleagerangeof lavaseruptedup to 180
km from Isla Isabelaprior to 5 Ma if the plate velocitywere --57 km/m.y.
initially formed directly over the hotspot, volcanism would
have spannedalmost 5 m.y., shorterthan the 7 m.y. spanof
volcanismimplied for site PL1 by a constantplate speed.
A wider dispersionand longer span of hotspotvolcanism
could presumablyhave occurredif the GalfipagosSpreading
Center resided over the hotspotduring this interval, as this
would provide a channelingmechanismfor volcanismaway
from the plume. Accordingto the platereconstructions
of Hey
[1977], the hotspotwas producinga volcanic trail solely on
the Cocosplatefrom <15 to 10 Ma andnot on the Nazcaplate.
He inferred a lack of hotspotvolcanismon the Nazca plate at
this time from the "saddle," i.e., the deeper bathymetry,
between the Carnegie Ridge and the eastern Galapagos
Platform. This arguesagainstwider dispersionof volcanism
due to spreadingridge channelingbetween16 and 9 Ma at site
lineament at Darwin Island [Wilson and Hey, 1995]. The
northern part of the Wolf-Darwin lineament encompassesfour
conical volcanoes, Wolf and Darwin Islands, a small seamount
(site PL29; -10 km diameter, summit depth -1350 m) NW of
Darwin, and a small seamount (site PL30; --3 km diameter,
summit depth --2000 m) between Wolf and Darwin. The
southernpart of the lineament is an elongate seamount(sites
PL26-PL28; --70 km long, summit depth --400 m) that is
morphologicallydistinct from the NW Wolf-Darwin lineament
volcanoes and slightly offset in a right-stepping en echelon
pattern. Rather than a conical shape, this seamount has a
hummocky morphology and appears to be a composite
structure, constructed of numerous smaller volcanic cones
(unpublished SeaBeam data). The lavas from this seamount
(PL27 and PL26) are very young, indicating active or very
PL1.
recent volcanism. This is consistentwith epicentersof recent
Taken together,the presentdata are consistentwith either a earthquakes
in the area [seeFeighnerand Richards,1994].
decreasein the hotspot-Nazcaplate velocity or an increasein
The mechanism(s) controlling the location and orientation
the areal region of hotspotvolcanism. At this point, neither of the Wolf-Darwin lineament is not clearly understood.
of the interpretationsis more credible than the other.
Morgan [1978] proposed that the volcanoes of the WolfDarwin lineament formed at a spreading ridge and that the
Wolf-Darwin
Lineament
linear alignment of the volcanoesis a vector sum of northward
The Wolf-Darwin lineament is a roughly linear array of jumps of the GSC and the absolutemotion of the Nazca plate.
seamounts and two small islands. A pseudofault of the This model predicts that the seamount ages should increase
93øW/95.5øW propagating rift cuts obliquely through the away from the spreadingcenter and towardsthe hotspot,from
13,700
SINTONET AL.:GEOCHRONOLOGY
OFGALAPAC•SSEg.
MOUNTS
NW to SE. Feighner and Richards [1994] suggested,on the
basis of gravity observations,that the Wolf-Darwin lineament
overlies a "lithosphericfault", althoughthere is no indication
of strike-slip or dip-slip offset. This suggestsa model in
which volcanism occurs in response to lithospheric tearing
that allows mantle upwelling and melting. In this model,
volcanism can occur randomly along the lineament.
Our new radiometricagesconstrainsuchmodels. The PL29
lavas (0.8 and 0.9 Ma), which lie 30-40 km from the ridge
crest, could have erupted at the GSC, consistentwith either
hypothesis. However, the young (probably still active)
volcanism at the SE end of the Wolf-Darwin
lineament and the
Christie,D.M., R.A. Duncan,A.R. McBirney, M.A. Richards,W.M.
White, K.S. Harpp, C.G. and Fox, Drowned islands downstream
fromtheGalfipagos
hotspotimplyextended
speciation
times,Nature,
355, 246-248, 1992.
Cox, A., and G.B. Dalrymple,Paleomagnetism
and potassium-argon
agesof somevolcanicrocksfrom the GalfipagosIslands,Nature,
209, 776-777, 1966.
Cox, A., D. and Engebretson,Changein motionof Pacificplate at 5
M.y.r BP, Nature, 313, 472-474, 1985.
Dalrymple, G.B., M.A. and Lanphere,PotassiumArgon Dating:
Principles,Techniques,
andApplicationto Geochronology,
258 pp.,
W.H. Freeman, New York, 1969.
Dalrymple, G. B., D. A. Clague, T. L. Vallier, and H. W. Menard,
40Ar/39Ar age, petrology,and tectonicsignificance
of some
seamountsin the Gulf of Alaska, in Seamounts,Islands, and Atolls,
K-Ar ages from the Wolf and Darwin Islands ( 0.9-1.6 Ma and
Geophys.Monogr.Ser.,vol. 43, editedby B. H. Keating,P. Fryer,R.
0.4 Ma, respectively) [White et al., 1993] are not consistent
Batiza,and G. W. Boehlert,pp. 297-315,AGU, Washington,
D.C.,
1988.
with the Morgan (1978) model unlessthese volcanoesformed
at the ridge and continued erupting as they moved from the DeMets, C., R.G. Gordon, D.F. Argus, and S. Stein, Current plate
motions,Geophys.J. Int., 101,425-478, 1990.
ridge. If this were the case, then volcanism at the SE end of
Duncan,R. A., andR.B. Hargraves,
40Ar/39Argeochronology
of
the Wolf-Darwin
lineament
must have been continuous since 6
Ma, the approximateage of the crust at this point (Figure 3)
[Wilson and Hey, 1995]. This seemsimprobablefor sucha
small edifice. Overall, the lack of a clearly defined age
progression along the lineament is most consistent with
lithospheric weaknessescontrolling the location volcanism.
basementrocks from the MascarenePlateau,ChagosBank, and
Maldives Ridge, Proc. OceanDrilling Program,Sci. Results,115,
43-52, 1990.
Feighner, M.A., and M.A. Richards, Lithospheric structure and
compensationmechanismsof the GalfipagosArchipelago, J.
Geophys.Res., 99, 6711-6729, 1994.
Geist,D.J., W.M. White, and R.A. Duncan,Geologyandpetrogenesis
of lavas from San CristobalIsland,GalfipagosArchipelago,Geol.
Soc. Am. Bull., 97, 555-566, 1985.
Conclusions
We have usedradiometricagesof lavas dredgedfrom the
Graham,D.W., D.M. Christie,K.S. Harpp, and j.E. Lupton, Mantle
plume helium in submarinebasaltsfrom the GalfipagosPlatform,
Science, 262, 2023-2026, 1993.
Galfipagos
Platformandwestern
Carnegie
Ridgeto identify Gripp, A.E., and R.G. Gordon, Current plate velocities relative to
pastvolcanicpatterns
as well as constrain
the hotspot-Nazca hotspotsincorporatingthe NUVEL-1 globalplate model,Geophys.
Res. Lett., 17, 1109-1112, 1990.
velocityoverthe past9 m.y. Widespread
volcanism
on the
Handschumacher,D.W., Post-Eoceneplate tectonicsof the eastern
GalfipagosPlatform at 5-6 Ma indicatesthat the current
Pacific, in The Geophysicsof the Pacific Ocean Basin, Geophys.
dispersed
patternof volcanism
datesbackto at leastthistime.
Monogr.$er., vol. 19, editedby G. Sutton,M.H. Manghnani,andR.
The agesof the earliestlavasfrom the centralGalfipagos
Moberly, pp. 177-202, AGU, Washington,D.C., 1976.
Platform and the Carnegie Ridge increaseprogressively Hey, R., Tectonicevolutionof the Cocos-Nazcaspreadingcenter,Geol.
eastward from the western edge of the platform, consistent
with a hotspot model for the generation of these features.
Age-distance relationships suggesteither a decreasein the
Nazca-hotspotvelocity, most likely at 5 Ma, or a 7-m.y. span
of volcanismon the oldestof the sampledseamounts.Rocks
dredgedfrom the entire length of the Wolf-Darwin lineament
are young (<1 Ma). Although the youngestmeasuredlavas are
farthest from the GSC, there is no clear trend along the
lineament. The pattern of volcanism along the Wolf-Darwin
lineament does not fit the hotspot model; perhaps here
volcanismis controlledby the stresspatternaroundthe ridgetransform environment,with thermal and material input from
the hotspot.
Acknowledgments. This researchwas supportedby
NSF.
We would like thank A. Grunder and D. Graham for
comments on an earlier version of the manuscriptand L.
Hogan for assistance in the lab. Valuable reviews were
providedby D. Wilson,A. Basu,andJ. Ogg. Samplesusedin
this study are curatedin the OregonState Universitycore
repositorywhich is supportedby NSF grantOCE94-02298.
References
Bailey,K, Potassium-argon
agesfromtheGalfipagos
Islands,
Science,
192, 465-466, 1976.
Soc. Am. Bull., 88, 1404-1420, 1977.
Hey R., G.L. Johnson,and A. Lowrie, Recent plate motions in the
Galfipagosarea, Geol. $oc. Am. Bull., 88, 1385-1403, 1977.
Hurfurd,A.J., andK. Hammerschmidt,
40Ar/39Arnd K-Ar datingof
the Bishopand Fish CanyonTuffs: Calibrationagesfor fission-track
datingstandards,Chem.Geol., 58, 23-32, 1985.
Kaneps,A.G., Cenezoicplanktonicformaminifers,easternequatorial
PacificOcean,Initial Rep.Deep SeaDrill. Proj. 16, 713-745, 1973.
Kurz, M.D., D.P. Kammer, T.C. Kenna, and D. Geist, Dynamicsand
evolutionof the Galfipagoshotspotfrom heliumisotopes,
Eos Trans.
AGU, 74 (43), Fall Meet. Suppl.,632, 1993.
Lonsdale,P., and K.D. Klitgord, Structureand tectonichistoryof the
eastern Panama basin, Geol. Soc. Am. Bull., 89, 981-999, 1978.
Minster,J.B., and T.H. Jordan,Present-day
plate motions,J. Geophys.
Res., 83, 533!-5354, 1978.
Morgan, W.J., Rodriguez, Darwin, Amsterdam
.... a secondtype of
hotspotisland,J. Geophys.Res.,83, 5355-5360, 1978.
Samson,S.D., andE.C. Alexander,Jr, Calibrationof the inter-laboratory
40Ar/39Ar
datingstandard
MMhb-1,Chem.Geol.,66,27-34,1987.
Sinton,C.W., The evolutionof the GalfipagosPlatform:Resultsfrom
radiometricdatingandexperimental
petrology,M.S. thesis,117 pp.,
Oreg. StateUniv., Corvallis,1992.
White, W.M., A.R. McBirney, and R.A. Duncan, Petrology and
geochemistryof the GalfipagosIslands:Portraitof a pathological
mantleplume,J. Geophys.Res.,98, 19533-19563,1993.
Wilson, D.S., and R.N. Hey, History of rift propogation and
magnetizationintensity for the Cocos-Nazcaspreadingcenter, J.
Geophys.Res., 100, 10041-10056, 1995.
York, D., Least-squaresfitting of a straightline with correlatederrors,
Earth Planet. Sci. Lett., 5,320-324, 1969.
Berggren,
W. A., D. V. Kent,C. C. Swisher,
andM.-P.Aubrey,, A
D.M. Christie,R.A. Duncan,andC.W. Sinton,Collegeof Oceanic
revisedCenezoicgeochronology
and chronostratigraphy,
in and AtmosphericSciences,OregonStateUniversity,104 Ocean.
Geochronology,
TimeScales
andGlobalStratigraphic
Correlations: Admin.Bldg.,Corvallis,
OR97331-5503.
(email:csinton@oce.orst.edu)
A Unified
Temporal
Framework
foranHistorical
Geology,
edited
by
W. A. Berggren,D. V Kent,. and J. Hardenbol,Soc. of Econ.
Paleontolo.
andMineral.,Tulsa,Okla.,in press,1996.
(ReceivedAugust28, 1995;revisedDecember15, 1995;
acceptedFebruary 13, 1996)