A Study of the Ages of the Precambrian of Texas

Home , Barringer Hill, Bliss Formation, Llano Uplift

JOURNALOF GEOPHYSICALRESEARCH VOLUME67. NO. 10 SEPTEMBElt1962

G. J. W•sssa•us•

California Institute o[ Technology,Pasadena G. W.

Institute o/Geophysicsand Departme•t o• Geology University o• California at Los Angeles L. T. SILVER

California Institute o[ Technology,Pasadena P. T. FLAWN

Bureau o[ EconomicGeology, University o/ Texas

Abstract. Age determinationsusing the SrS7-RbS7,Ar4ø-K4ø, and Pb-U methodswere made on samplesof muscovite,biotite, amphibole,microcline, and zirconfrom igneousand metamorphicrocks from the Franklin Mountains,Hueco Mountains,Pump Station Hills, and Carrizo and Van Horn Mountains. In addition ageswere determinedon a number of basement cores from Texas and New Mexico. The results show that a belt of rocks of varied lithologyextending from E1 Pasoto east of the Llano uplift are all of the sameage. The general age by the strontium and argon methodsis 1000 to 1090 m.y.; and by the lead- uranium method on zirconsit is 1150 to 1200 m.y. This event is in the same time band as the 'Grenville' orogenyin Canadaand the northeasternUnited States and possiblyshould be consideredpart of the general 'Grenville' episode.All the data now available indicate that the orogenicevent at about 1000to 1200m.y. is the mostwidespread and pervasiveepi- sodeof Precambrianorogeny on the North Americancontinent for which adequateevidence has been presented.At least one and probably two older periodsof igneousactivity and metamorphismoccurring at 1250 and 1400 m.y. are found in the northern regionsof the Texas Precambrian basement. No evidence was found for any igneous event between the early Paleozoicand the 1000-m.y.episode.

INTRODUCTION (b) In the HuecoMountains the graniteis either The purposeof this studyis to attempt to use surroundedby unconsolidatedalluvial deposits mineral agesto correlateigneous and metamor- or in fault contactwith youngerbeds; general phic rocks'of Precambrianage in Texas. regionalevidence, however, suggests that it is Precambrianrocks in Texas crop out in the pre-Bliss (Cambro-Ordovician)in age. (c) Llano uplift area of central Texas and in a num- Rhyolitein the Pump StationHills is surrounded ber of separated mountain uplifts in Trans- by unconsolidatedalluvial deposits, and there is PecosTexas (Figure 1). Stratigraphicevidence no convincingstratigraphic evidence pertaining as to the age of these rocks is as follows: to its age.On the basisof regionalgeology earlier 1. Central Texas graniteswhich intrude meta- investigatorssuggested that the rocks are Pre- sedimentaryrocks are overlainby Upper Cam- cambrian [King and Flawn, 1953; Masson, brian strata (Hickory sandstonemember of the 1956]. (d) Metamorphicrocks and pegmatites Riley formation [Cloud, Barnes, and Bridge, in the Carrizo and Van Horn Mountains are 1945]). overlainby Permian (Wolfcamp)beds, but re- 2. In Trans-PecosTexas stratigraphic relations gional stratigraphicevidence suggests that the differin the separatemountain ranges. (a) In the rocks are pre-Van Horn sandstone(Precam- Franklin mountains,granite, volcanic rocks, and brian?), which is pre-Bliss (Cambro-Ordovi- metamorphic rocks lie beneath the Cambro- cian) in age. OrdovicianBliss formation [Harbour, 1960]. A number of chemical lead-uranium and lead-

4021 4022 WASSERBURG,WETHERILL, SILVER, AND FLAWN

N E W M E X I C

S O

ION

,1114 I

•1•1•Sample localities andnumbers I $cal•'

I .

Modifiedfrom F, 0 I. The Unlvers,IV o! TelLO•Publicolion 5:)01 Fig.1. Map showingthe exposuresof Precambrian rocks in Trans-PeeosTexas and the sample localities. thorironages up to 1100million years have been clearlyshow the existenceof plutonieigneous determined on the rare-earth minerals from the activityat about1100 m.y. in the south-central BarringerHill pegmatitein the Llano region United States. The correlation of this occur- [Bartell, 1917; Holmes,1931]. More recently reneewith 1000-m.y.plutonit activity in eastern age estimateshave been made on zirconsfrom North America some 1200 miles distant was not the Llano regionand from well coresin west specificallymade by theseworkers. Texasby the 'Pb-a' technique[Jaffe, Gottfried, In the presentstudy samples for agedetermi- Waring,and Worthington,1959]. Owingto the nation were obtainedboth from the exposedPre- manydifficulties both technicaland inherentin cambrian and from basement cores. Samples this nonisotopicmethod, the resultswill not be were collected from the various localities with considered in detail. the previousgeologic studies as a guide,and Modern age determinationsusing Ar4ø-K'ø, basementcores were selectedwith the study by Sr87-Rb87, Pb'ø•-U2. •, and Pb'ø*-U•5 methodshave Flawn [1956] as a guide,the purposebeing to beenmade by severalworkers on mineralsin the establish time correlations and compare the graniticrocks and associatedpegmatites in the various geochronometricmethods. Llano area [Aldrich,Wetherill, Davis, and Til- Results ton, 1958;Goldich, Nier, Baadsgaard,Hoffman, and Krueger,1961; Zartman,1962, 1963]. The results for the exposedrocks are pre- The da•a obtained by Aldrich et al., in the sentedby area, after which the basementcore Llano regionare summarizedin Table 1. They data are given.Thin sectionsof the rocksand AGES IN THE PRECAMBRIAN OF TEXAS 4023 TABLE 1. Age Determinationsfrom the Llano Region

Locality Mineral Ar4ø/K4ø Sr87/Rb87 pb2o6/u2s8pb2oT/u2s5 pb2o•/pb•o6

Petrick Quarry Biotite* 1060 1100 ...... (granite) Zircon* 950 990 1070 (pegmatite) Biotite]• 1(•) li•(• ...... Lone Grove Biotite, feldspar$ 1060 ...... Pluton (granite)

* After Aldrich, Wetherill,Davis, and Tilton [1958]. • After Goldich,Nier, Baadsgaard,Hoffman, and Krueger[1961]. •:After Zartman[1962a]. Average of severaldeterminations by both methods.All ageshave been re- calculatedfor the decayconstants used in this paper. the mineral separatesare describedin the ap- tial white micrographicintergrowth. This rock pendix. also shows distinct iron oxide stains on the freshest material. FRANKLIN MOUNTAINS A sampleof feldspar,TH-6, wastaken from a Precambrian rocks are exposed in North microcline-perthite-quartzpegmatite which was FranklinMountain in thevicinity of E1Paso (see seen to cut the granophyric granite. This dike Figure 2). These have been describedin a recent has a thickness of about a meter. No. fresh mica publicationby Harbour [1960]. He has pre- was found in this pegmatite. senteda sedimentarysequence in this area com- In addition to these materials a sample of prisedof the followingunits in order of decreas- a friable tourmaline-biotite-plagioclaseschist, ing age: (1) The Castnerlimestone, possibly TH-2, was collected. This material was not equivalent to the Allamore limestone of the Van foundin the measuredsection reported by Har- Horn area. Several diabase sills are included in bour for a traversefurther west (seeFigure 2). the Castner. (2) The Mundy breccia,which The unit is about 3 meters, thick and lies be- rests unconformablyon the Castner. (3) The tween a thick section of Castner limestone on the Lanoria quartzite.On top of this unit is (4) a west and a diabase sill to the east. The Castner rhyoliteextrusive. (5) The youngestrock in the showsbeautifully developedstromatolite struc- area underlying the Bliss sandstoneis a massive turesand mud cracksat this locality.The top of granite,which is seento surroundand engulf the limestone here shows distinct contact meta- large blocks of the sedimentarysection. This morphic effects from the granite. The diabase granite showsa variety of phases,the larger sill containsabundant biotite near the granite exposuresbeing a pink-white biotite granite, contact. The origin of the schistlayer is not whichis typicallyrather friable.One phase is a understood. gray-white biotite granite, which locally con- The ages for the Sb87-Rb87and Ar•ø-K•ø tains numerous reacted inclusions and shows a methodson variousmineral separates are pre- jointingstructure with the fracturesspaced at sentedin Table 2; the agesfor the zircons',in approximately 30 to 60 cm. These fractures all Table 3. An estimatedmaximum error is given show iron oxide stains and a pale green dis- for each of the ages. This estimate is based colorationeven in a fresh quarry face.We infer solely on the analytical data without consider- this alterationto be a deutericeffect. A sample ing possibleerrors in the decay constants.A of this phase,TH-1, wastaken from the interior more detaileddiscussion of errorsis presented of a boulderof about 120-cmdiameter (see in the appendixunder analyticalprocedures. Figure 2). Samplesof a third phase,a grano- The biotite and microclineseparated from phyric biotite-hornblendegranite, TH-5, were granite TH-1 gave the followingresults' the collectedin FusselmanCanyon. This rock has strontiumage of the biotiteis considerablyless a spottedappearance produced by the isolation than the argon age, and both are less than the of dark gray microclinecrystals by an intersti- strontiumage of the coexistingmicrocline. The 6 4024 WASSERBURG,WETHERILL, SILVER,AND FLAWN

i•_ x lu.i

'..

o.. AGES IN THE PRECAMBRIAN OF TEXAS 4025

TABLE 2. Age Determinations from the and a microclinefrom a crosscuttingpegmatite Franklin Mountains near E1 Paso, Texas TH-6. Strontium ages were determined on the feldspars,and an argon age was determinedon Age, m.y. hornblende[Hart, 1961]. No. Rock Mineral K-Ar Rb-Sr The youngerpegmatite gives a strontium age of 1087 m.y., which is greater than the results

TH-1 Granite Biotite 980 4- 20 930 4- 20 on the other feldspars.The differencebetween Feldspar 1040 4- 30 the feldsparsfrom TH-6 and TH-5 is well out- TH-2 Biotite side of experimentalerror. The differencebe- schist Biotite 860 4- 20 710 4- 50 tween TH-6 and TH-1 feldsparsis just at the TH-5 Granite Horn- limits of error. While this disagreement is blende 1000 4-20 Feldspar 970 4- 50 not large, it is certainly real and is yet an- TH-6 Pegmatite Feldspar 1090 4- 20 other exampleof the 'pegmatite-countryrock' discrepancyreported by previousworkers [Was- setburg,Wetherill, and Wright, 1959; Wetherill, per cent differencebetween the feldsparand the Davis, and Tilton, 1960; Gerling and Polkanov, biotite is well beyond analytical error. This age 1958]. pattern is rather typical of almost all the data. Two. zircon concentratesof differing uranium A biotite separated from the friable schist content were extracted from the granophyric TH-2 gave discordantages, the argo.nage being granite TH-5 for Pb-U analysis.The lechniques considerablyless than that obtainedon the bio- followed are described by Silver, McKinney, tite from TH-1. In both casesit was apparent Deutsch, and Bolinger [1962]. The analytical in the field that the rocks were not 'fresh.' None- results are given in Table 3, and the parent theless no characteristics were observed in thin daughterratios are plotted in Figure 3 in a con- sectionsof the rocks, or in the grain mounts of cordia diagram [Wetherill, 1956]. A straight the separated minerals, which might indicate line drawn through these points intersectsthe that the biotite was unsuitable for age determi- concordiacurve at about 1200 m.y. It has been nation. shown by Silver and Deutsch [1961] that dis- Samples of hornblende and microcline were cordant U-Pb systemsin cogeneticzircons may separated from the granophyric granite TH-5 generatelinear patterns on a concordiadiagram.

TABLE 3. West Texas and Llano Zircons

Apparent Ages, m. y.

Sample pb•o6/u•8 pb•o7/u•5 pb•oT/pb•o6

West Texas Samples Micrographic granite (TH-5), Franklin Mrs., E1 Paso Co. Fraction A 880 4- 20 950 4- 20 1095 -V 20 Fraction B 770 4- 20 860 4- 20 1080 4- 20 Rhyolite porphyry (TV-13), Pump Station Hills, Hudspeth Co. Fraction A 1065 4- 20 1105 4- 20 1175 4- 25 Fraction B 1005 4- 25 1050 4- 25 1140 4- 20 Granite well core, Phillips Puckett //lc, Pecos Co. 925 4- 20 970 4- 20 1065 4- 20 Muscovite-oligoclase-quartzschist (TV-1), Van Horn Mrs., Hudspeth Co., fine fraction 745 4- 20 875 4- 25 1220 4- 40 Llano Samples Town Mr. granite, Petrick quarry [Tilton et al., 1957] 950 4- 25 990 4- 15 1070 4- 25 Town Mr. granite, Granite Mr. quarry, [Silver et al., in preparation] 970 4- 20 1020 4- 20 1120 4- 20

Constants:•3s = 1.537 X 10-•ø;X•35 = 9.72 X 10-•ø; (atoms U•S•)/(atomsU TM) = 1/137.8. 4026 WASSERBURG, WETHERILL, SILVER, AND FLAWN

0.21

0.20

0.19

0.18

0.17

0.16

0.15

0.14.

0.13

0.121.2 1.3 1.4 1.5 t.6 1.7 1.8 1.9 2.0 Z.I Z.Z 23 2.4 pb2O7/u235 Fig. 3. The Pb•øø/U•Sand Pb2ø?/U•ratios for samplesof zirconsfrom the Precambrianof Texas. Throughoutthis paperthe linearpattern of the cloudingand in generalappears unaltered. A accumulated zircon data is used as an argument strontiumage of 1045 m.y. was determinedon a for contemporaneity.The intersectionof the sample (TV-14) (Table 4). patternwith the concordiacurve is interpreted PUMP STATION •ILLS as an originalage of crystallizationfo.r the zircons.A more completediscussion of this inter- A sampleof a granophyricrhyolite porphyry sectionis givenin a later section.The Pb2ø•-Pb'•ø• was collectedfrom this locality. This rock is ex- age must be taken as minimal assumingany tremely tough and resistantbut in thin section (nonfractionating)form of lead loss. shows considerablealteration. Samples of the The resultsobtained in this area by the vari- other phasesthat have been reportedwithin the ous methods indicate an event at about 1100 Pump Station Hills area [Masson, 1956] were m.y. Ito.wever,they scatter far outsideof ex- not collected.A 2.5-cm cube of rhyolite, TV-13, perimentalerror. The only visibleevidence for was crushedto pass 80 mesh, and a split was a postemplacementmetamorphism is the altera- taken for whole-rock strontium analysis. An- tion and recentweathering previously described. other strontium age was determinedon a sample The block faulting in this area doesnot appear of microclineseparated from another specimen. to have causedmore generalmetamorphic ef- The results,given in Table 4, are in goodagree- fects, but post-Paleozoicintrusive activity is evident in the region and its local effectscan- TABLE 4. Age Determinations from the not be estimated. Hueco Mountains and Pump Station Hills

I-IuEco MOUNTAINS Age, m.y., NO. Locality Rock Mineral Rb-Sr A samplewas collected from a low-lyingknob of red porphyritic hornblendegranite from the TV-14 Hueco foothills of the Hueco Mountains. No fresh Mrs. Granite Feldspar 1050 • 60 pegmatitescontaining mica were found in lhis TV-13 Pump locality. The granite showsa tendencyto granu- Station lar disintegration,presumably due lo weather- Hills Rhyolite Feldspar 1060 4- 40 Whole ing. The granitecould be describedas 'inelastic.' rock 1020 4- 60 In thin sectionthe microclineshows only slight AGES IN THE PRECAMBRIAN OF TEXAS 4027 ment and indicate a strontium age of 1060 m.y. TABLE 5. Age Determinations from Two zircon separates of differing uranium the Carrizo Mountains content were obtained from a hand specimen of this porphyry.The resultsare givenin Table Age, m.y. 3 and Figure 3. The two fractions give Pb2ø7- No. Rock Mineral K-Ar Rb-Sr Pb2ø• ages of 1175 and 1140 m.y. and lie closeto the sameline as the Franklin Mountain samples TV-10-a Schist Biotite 780 + 20 580 -v 20 (Figure 3). These results clearly indicate the TV-10-b Amphibo- Horn- Precambrianage of the rocks in the Pump Sta- lite blende 900 + 20 tion Hills. TV-12-b Schist Biotite 960 4- 60 TV-12-c Meta- Whole 1070 :k 70 CARRIZO •¾[OUNTAINS rhyolite rock This area is known to be structurally more complex than any of the others investigatedin the hornblendeand a distinctly younger result this work. It is bounded to the north by the on the biotite. The strontium age on the biotite Hillside fault and is possibly involved in the is much less than the argon age. Streeruwitz overthrust in this region. Flawn These measurementshave a pattern quite [King and Flawn, 1953] has describedthe Car- similar to that obtained in the Franklin Moun- rizo Mountain group as a seriesof steeplytilted tains, but becauseof the polymetamorphichis- metasedimentary and metaigneousrocks. The tory in the Carrizo Mountain area it is rather sedimentary rocks were intruded by rhyolites difficult to make any simplestatements, and the and basicdikes, and the sectionwas subsequently conclusions.must be regarded as tentative. It is deformed.The rhyolites,markedly deformedand evident that these rocks are at least 1000 m.y. sheared,are describedas metarhyolites.Toward old. The apparent age may reflectthe effectsof the northwest of the map area (Figure 4) the recrystallization during shearing, and the true metarhyolite is converted into a mylonite. The age may be considerablyhigher. It is also im- metasedimentary rocks are a relatively low- possibleto draw any conclusionsabout the time grade metamorphic assemblage,although there of thrusting from thesedata. If the biotite from is distinct evidencefor a polymetamorphichis- TV-12-b lost its radiogenicstrontium at the time tory. of shearing, the time of shearing would be Samplesof metarhyolite TV-12-c and an am- greater than 960 m.y., but whether complete phibolite schist TV-12-b were collected from loss would occur under these conditions is not the northwestern part of the area, where the known. Furthermore, the shearingneed not have rhyolite showsprofound evidenceof cataclasis been concurrent with the thrusting. Although and has a strong planar structure and linearion. experiencewith deformed rocks of this type is The schist.is only locally present and presum- very limited, we tentatively suggestthat the ably is a metaigneousrock. A 2.5-cm cube of whole-rock strontium age on the rhyolite is the the metarhyolite was crushedto. pass 80 mesh, most reliable estimate of the age of the rhyolite. and a fraction was taken for a total-rock stron- VA>• I-Ioa>• MO•?NTA•NS tium age determination.A biotite separatewas obtained from the schist. The results,given in A thick sectionof middle-grademetamorphic Table 5, indicate an age in the neighborhoodof rocks intruded by pegrnatitesis exposedin the •070 m.y. Mica Mine area of the Van Horn Mountains, In addition to these materials, samplesof a an area describedby Flawn [1951, 1956]. These fine-grainedhornblende-biotite-garnet schist TV- rocks are unconformably overlain by Permian 10a and a hornblende-epidoteschist TV-10b (Wolfeamp) rocks. The metamorphicsequence were collected from the southern part of the is comprisedof a feldspathie quartzite, a felds- area, farther from the locus of the overthrust. pathic muscoviteschist, and biotite schist and TV-10b occursin a small dike-like body about amphibolite.The pegmatitesoccur in bodiesof 13 cm thick. Biotite was extracted from TV-10a, varied shapethat conformto and cut acrossthe and hornblende from TV-10b. The results shown foliation of the host rock. in Table 5 indicatean argon age of 900 m.y. for A detailed comparisonof the various mineral 4028 WASSERBURG, WETHERILL, SILVER, AND FLAWN

Kc 30 o 04' Ph ( Ph' Qal "-., J

Qal

"•'/ "'II iii'i!:.

:Cl

: ß .,,-,/ p.CCq

00'

Ph' ..' / Qal : /

105000' I04ø55 .After Plote 1,The University SCALE of Texos Publicorion No. ,5501

EXPLANATION

j Qal p•Cq Alluvialdeposits Hueco limestone Metarhyolite QuartzoofeldspaJhic rocks

Intrusiveigneous rocks Van Horn sandstone Limestone Chlorite-mico schisl +TV12 Cox sandstone and Granodiorite Phyllite Samplelocality and number Campogrande limestone

Amphibolite Mixed units Fig. 4. Geologicmap of the Carrizo Mountains (after Flawn), showingthe sample localities Fig. 5. Geologic map of Mica Mine area in northwest Van Horn Mountains (after Flawn), showing sample localities. 4030 WASSERBURG, WETHERILL, SILVER, AND FLAWN ages was undertaken at this locality becauseof rather low argon age and a distinctly low stron- the wide variety of minerals in the various tium age. lithologies and the relative freshness of the The maximum age obtained on these meta- rocks. The locality provides an excellent op- morphic rocks was 1030 m.y. by the strontium- portunity to compare the ages of several coge- rubidium method on a muscovite from TV-1. netic metamorphicminerals and a presumably The total rock age on TV-1 was somewhat late-stage pegmatite. The sample localities are low compared with the muscovite result. The shown in Figure 5. zircon concentrateyielded a very discordantset A sample (TV-1) of the biotitie muscovite- of ages (see Table 3), the Pb'•ø•/Pb2ø• age being albite-oligoclaseschist was collected.This rock the highest observedamong the zirconsstudied. is quite friable, presumablybecause of the high The age pattern for the zircon stronglysuggests mica content. Mineral separates of muscovite, that they retained a.t least part of their original biotite, and zircon were made, and in additiona predetrital daughter products. The great dis- total-rock sample was obtained from a 4-cm cordance indicates that at least some of the zir- cube.Argon and strontium ageswere determined cons originated considerably more than 1220 on the mineral separates,and a strontium age m.y. ago. This is reflected in the location of the was determined on the total-rock sample. The point representingthe TV-1 zircons.in the con- zircons are variable in form and color and show cordia diagram in Figure 3. If the detrital markedly irregular forms due to corrosion. sources of the other minerals in TV-1 were riot The general suite has the appearance of significantlydifferent from those of the zircons, detrital origin modifiedby reaction in the meta- we must conclude from the whole-rock strontium morphic environment. age of 970 m.y. that the strontium of this sys- Hornblende sampleswere obtained from two tem was mixed with a common strontium amphibolites, TV-2 and TV-6. These samples reservoir at the time of metamorphism. are of very high purity, aad the argon agesare Samples of four pegmatites were collected: unaffected by any biotite or K-feldspar im- TV-3, TV4, TV-118, TV-252. TV-3 muscovite puffties. was obtained from a single crystal of 17-cm A sample of a coarse-grainedbiotite schist diameter. A cleavagefragment of microclinewas TV-5 was collected,and the biotite separatewas taken from the same outcrop. A fraction was analyzed. The results of these analyses (TV-1, ground to pass40 mesh (total F); the remainder TV-2, TV-5, TV-6) on the metamorphicrocks was ground, and the --28 +40 mesh fraction are presentedin Table 6. Except for the TV-1 was taken for analysis.The strontium and rubid- biotite sample, all the ages are in excellent ium contents of these two feldspar fractions agreement,the spread being just larger than are slightly different. The resulting ages are in the analytical error. The TV-1 biotite has a good agreement.The argon age of the coexist-

TABLE 6. Age Determinations from the Mica Mine Locality near Van Horn, Texas

Age, m.y.

No. Rock Mineral K-Ar Rb-Sr

TV-1 Muscovite schist Muscovite 980 q-20 1030 q- 30 Biotite 940 q- 20 890 q- 30 Whole rock 970 q- 50 TV-2 Amphibolite Hornblende 1000 q- 20 TV-5 Biotite schist Biotite 1020 q-20 990 q- 30 TV-6 Amphibolite Hornblende 990 q- 20 TV-3 Pegmatite Muscovite 1020 q- 20 1090 q- 20 Feldspar (sized) 1050 q- 30 Feldspar (total) 1040 q- 30 TV-4 Pegmatite Muscovite 1060 q- 20 TV-118 Pegmatite Muscovite 1090 q- 20 1090 q- 20 TV-252 Pegmatite Biotite 1010 q- 20 970 q- 20 AGES IN THE PRECAMBRIAN OF TEXAS 4031 ing mica agrees with the strontium ages of the appears to be preservedin the micas and am- feldsparsbut is significantlyless than the st.ron- phiboles. tium age of the mica. The feldsparages and the muscovite strontium age agree just within ex- BASEMENT CORES perimental error. The agreement of the two The minerals dated on the exposedPrecam- feldsparages indicates that there is no significant brian rocks of the Trans-Pecosregion indicate fractionationof parent or daughteron the scale that all theseminerals were formed or recrystal- of the sampling. lized at about 1100m.y. This result was obtained Muscovite (TV-4) was obtainedfrom crystals on a variety of lithologiesrepresenting a wide of 2-cm diameter from a graphic granite phase spectrum of metamorphic and igneousactivity. of a pegmatite. TV-118 was collected during Theseevents appear to be contemporaneouswith previous field work by Flawn. The argon age the metamorphismand plutonismof the Llano was first determined, and since it gave a result area some 500 miles distant. To delineate the oc- significantly higher than previous samplesthe currence of this activity further it was neces- strontium age was also.measured. It also yielded sary to use samplesof basementcores. Samples a high value of 1090 m.y., which is not different were chosenfrom wells lying between the Van from that obtained on the other pegmatitic Horn Mountains and the Llano uplift and adja- muscovites. cent regionsto test the continuity of the 1100- In one locality (TV-252) pegma.titicbiotite m.y. event. crystals were found. They have a diameter of For this purpose cores of granite were ob- about 3 cm and occur in a coarse microcline tained from Pecos,Schleicher, Williamson, and pegmatite as thin (0.1 to 0.2 cm) plates along Coleman counties.Their petrographyhas been fractures crosscuttingthe quartz and feldspar. describedby Flawn [1956]. The well locations The biotite is a late-stage phenomenonand and data are given in Figure 6 and Table 7. shows some iron stains around the borders of Enough of the Pecos County core was avail- the crystals'. Fresh elastic material was care- able t.o permit a zircon analysis.However, feld- fully selectedfor analysis. The argon age is in spar separatesmade from this sample had un- goodagreement, with the other argon results,but favorableRb-Sr ratios,and so no other analyses the strontium age is distinctly low, by 7 per were possible.The Pb•ø7-Pb•ø• age obtained on cent. the zirconswas 1070 m.y. The data plot in the From these results on the pegmatitic mate- concordia diagram close to the same line as the rials we conclude that the minimum and most previous samples. probable age of eraplacementof the pegmatites The feldspar separates obtained from the is 1090 m.y. These pegmatite results are defi- other cores had favorable Rb-Sr ratios and were nitely greater than the ages of the geologically used for age determination.The coresshow con- older metasedimentary rocks (970 to 1030 siderablefracturing and somealteration, but in m.y.). Although the discrepancyis not la.rge,it the samplespresented here the microclineap- appearsto be distinctlygreater than the analyti- pears fresh and only slightly cloudedby altera- cal error. This is an additional example of the tion. The feldsparin the SchleicherCounty core 'pegmatite-countryrock' discrepancy.Assuming showsmore staining and cloudingthan that in diffusion to be the cause of relative daughter the other samples.The biotite in some of the lossthis would suggestthat pegmatiticmaterials cores was typically altered and shredded and other than biotite have a characteristically did not appearsuitable for dating purposes. smaller D/a 2 value, possibly reflecting larger The ages obtained for these samples range units in the mosaic structure. The biotites in from 960 to 1050 m.y. This spread is beyond both the schistsand the pegmatitestend on oc- experimentalerror and is comparableto the dis- casion to give low results, particularly by the persion exhibited by the feldspars from the Sr-Rb method. The preceding data show that Franklin Mountains. during middle-grademetamorphism all the Sr•- Two corefragments from the ColemanCounty Rb 87and Ar4ø-K4øclocks in the parent minerals well represented a coarse-grainedleucogranite were completely reset during recrystaIlization, and a fine-grainedor aplitic phase. The results and no evidenceof the premetamorphichistory are in excellent agreement. 4032 WASSERBURG, WETHERILL, SILVER, AND FLAWN

NEW MEXICO ,r - •rExAs

, I ...... ' .... 'V•'O, !o "!.... w.• (•26'•'...... •% I '-,,,'' .., 'AMARILLO" I///VC ! ""..... ': _..,,,TERRAN.E..'."'-.; ....,, E'

•x ! ,/'" '"'-.""..CRATON ,/"i,•'"•. ARBUCKLE•.• ...... ' "-' ...... :_C.;.'3;i's ,, :\. !," • -.•--"...... ' ...... • •,•o...o....©...... i'•, _•>• ...... •'•' "• •,•2o ...... RED ©i"•:'e'•RWER ...... MOBILE BELT .... • .,<,,•< ,-- - '---......

.... • ...... j.' •.• •ORT WORTt •'ß• -.. <• I...'" ...... ''"-...... • •,"•.•. ,•, '•,;...... '•C•;•-•'"'"'",...TEXASCRATON ,•*,,...... ,•oœc•*./ ©,240I ,'"'"',•%", I'EL,ASO ...... •i.... N_E_W.._..M_•_X.!C_O...... J ", •-'-.,' FRANKLINMTNS•UMP/ STATION TEXAS \...... '(•) 050 •..•.• MTNS 1090•HUECOio5o •/HILLS-1060 TEXAS CRATON >'.-•-..• -SANANGELO v'4/V ß I10 ß.•r• (• •oo) ...•,•"' •'%.... •, %, z 'AUST•N

/• "•/•r•Tn3'••'{•'•' 'SANANTONIO Fig. 6. Map of part of Texas, showingthe sample locationsand the various terranesas des- ignated by Flawn [1956]. Well cores are numbered as in Table 7. The best age estimate for each locality is given.

TABLE 7. Age Determinations on Drill Core Samples

Age, m.y.

Locality Rock Terrane Mineral K-Ar Rb-Sr

1. SchleicherCo., Texas Alkaline granite Texas eraton Feldspar 960 q- 30 2. Coleman Co., Texas A. Aplite dike Texas craton Feldspar 1050 q- 30 B. Granite Texas craton Feldspar 1040 •: 40 3. Williamson Co., Texas Granite Texas craton Feldspar 1010 + 70 Gneiss 5. Eddy Co., N.M. Granite Texas craton Feldspar 1240 q- 60 6. Roosevelt Co., N.M. Hornblende Texas craton Biotite 1400 q- 40 Biotite gneiss Hornblende 1390 • 30 7. Roosevel• Co., N.M. Rhyolite Panhandle volcanic Feldspar 1180 q- 70 8. Potter Co., Texas Granite Panhandle volcanic Feldspar 1260 q- 40 Whole rock 1190 q- 70 9. Lubbock Co., Texas Rhyolite Panhandle volcanic Feldspar 1235 q- 120 10. Crosby Co., Texas Amphibolite Red River mobile belt? Hornblende 1320 + 30 Texas craton? 11. Cottie Co., Texas Biotite- Red River mobile amphibole belt Biotite 1400 •- 40 Gneiss 12. Giles Co., Tenn. Granite Feldspar 1120 q- 30 AGES IN THE PRECAMBRIAN OF TEXAS 4033

In addition to these samples several cores from the rhyolite porphyries,and the strontium were obtainedfrom other areasto gain informa- ages were measured.The results are 1180 and tion on the agesin otherparts of the basement 1235 m.y. as describedby Flaton [1956]. The Potter County sampleis a micrographic Flawn oufiined a number of structural and granite having some fine-grained phases. The lithologicdivisions of the Texasand southeast feldsparsare parfly sericitizedand clouded,the New Mexico basement on the basis of petro- degreeof alteration being similar to that of the graphicstudy of the availablebasement well Roosevelt CounW sample. The strontium age coresand samples.He distinguishedthe follow- was determined on a total-rock sample taken ingunits: Texas craton, Van Hornmobile belt, from a 3-cm cube. The result was 1190 m.y. For Red River mobilebelt, Panhandlevolcanic ter- a felspar separate from this rock a strontium rane,Fisher metasedimentary terrane, Swisher age of 1260 m.y. was obtained. gabbroicterrane, and Wichita igneous province. Red River mobile belt. Cores were obtained He alsoattempted to relate thesedivisions to from Crosby and Cottle counties,Texas. Flawn oneanother in time by meansof lithologiechar- indicated that the assignmentof the Crosby acteristics,structural behavior, and the available County sampleis uncertainand that it may be- agedeterminations, and he presented a tentative long to the Texas eraton. Both coresare quite correlation of these subdivisions. fresh. Some serieitization of the plagioelaseoc- From the limited number of cores available curs, but it is confinedto occasionalfractures. for this study a numberof favorablesamples Hornblende was obtained from the Crosby were selectedfor investigation.These samples, County core and biotite from the Cotfie County describedbelow, are grouped accordingto core.Biotite, mostly chloritized,is presentin the Flawn's divisions: Crosby County sample. The ages for the two Texas craton. In addition to the granite samplesare 1320 and 1400 m.y., respectively. samplesfrom Pecos, Williamson, Sehleicher, and Coleman countiespreviously described,cores DrscvssroN 0r SRS'-RBs* AND AR•ø-K • RESULTS fromEddy andRoosevelt counties, New Mexico, Comparisonof the resultsobtained on the out- were run. The micrographicgranite from Eddy crop materials, where there is more control than Countyyielded microcline suitable for strontium in the well cores,indicates several regularities. analysis.Whereas the biotite in this rock is The pegmatitic muscovites,although showing severelyaltered, all the feldsparsappear rather some scatter, have the highest strontium and fresh.The age obtainedwas 1240 m.y. argon age in a locality. The strontium ages on The sample from RooseveltCounty is de- microcline are usually in agreementwith those scribedby Flawn as a quartz dioritegneiss. This on muscovite in the one locality where there rock lies beneath a breeeiatedgranite found at are comparisons.The argon ageson amphiboles a shallowerdepth in the samewell. The rock is appear somewhatlow; relative to the highest quitefresh and containsboth biotite and horn- strontium ages they indicate a retentivity of blende. An argon age was determinedon the about 90 per cent. Biotites, particularly fine- hornblende,and a strontiumage on the biotite. grained material (_•0.5 mm), yield rather low The resultsare in very goodagreement at 1400 argon ages and even more disturbed strontium m.y. ages. Panhandle volcanic terrane. Cores from Conversely, data of other workers [Aldrich, RooseveltCounty, New Mexico,and Potter and Wetherill, Davis, and Tilton, 1958; Tilton, Weth- Lubbock counties,Texas, were obtained from erill, Davis, and Bass, 1960] have shownthat it this terrane. is also commonfor the strontium age of biotite The samplesfrom Rooseveltand Lubbock to be greater thaa• the argon age. From the countiesare rhyoliteporphyries. All the feldspars present observationswe concludethat no regu- in these rocks show characteristic discoloration lar relationshipis to be expectedbetween the and clouding.The Lubbock County sampleex- two agesin this mineral, in contrastto potassium hibits distinct flow structure. The alteration of feldspar where the argon age is invariably less the feldsparsis somewhatless than in the previ- than (or equal to) the strontium age. ous sample.Feldspar separates were obtained The strontium and argon ages from all the zt03•i WASSERBURG, WETHERILL, SILVER, AND FLAWN outcrop samples are illustrated in the histo- must be partly due to differential weathering grams of Figure 7. These results show a wide and to differencesin the nature of the crystal dispersionover a range of about 350 m.y. The mosaicand in the particular geometriclocation maximum age is 1090 m.y., and 77 per cent of of the crystal. The dispersionon a regionalscale the results are above 950 m.y. The maximum of a synchronousevent is a result of both the age in each of the five outcrop localities lies be- local effects and the grosserdifferences in geo- tween 1050 and 1090 m.y. All but one of the logic history. The superpositionof several epi- ages significantlyless than 1000 m.y. are from sodesof metamorphism (in the general sense) biotites' showing obviously discordant ages. at different times and intensities at different As previous experiencein this field indicates, localitieson a primary regionalevent produces pegmatitic feldspar and muscovite apparently dispersion.This may involve the local 'quench- give the best measureof the absoluteage by the ing' of a particular site or mineral early in a strontium and the strontium and argon methods, regional episode of considerableduration, in respectively.The scatter in the ages is due to. which case the dispersionshould give informa- time resetting mechanismsin which the daugh- tion about the time scale for a single-episode ter-parent ratio is diminishedin a mineral. regional metamorphism, or it may involve one The nature of the resetting mechanismsis or more completely independent later events. somewhat different for each mineral speciesin Severalexamples of two eventsclearly separated responseto the particular decayscheme and the in time have been found [Tilton, Davis, Weth- physical-chemicalcharacteristics of the. species. erill, and Aldrich, 1958; Campston,Je#ery, and The mechanisms'are poorly understood, and Riley, 1960; Wetherill, Kuovo, Tilton, and Gast, there are few a priori criteria for choosingmate- 1962]. These results are well known and were rials for dating. The general effects, however, elucidated by comparing the Pb-U ages of zir- are apparent in all the mineral speciesand are conswith the Sr87-Rb87 and Ar4ø-K4øages of the manifest on a very local scale (hand specimen, other minerals and by studying the Sr87-Rb• outcrop, and locality) as well as on a regional ages in whole-rock samplesand in the mineral scale for rocks that are the product of synchro- components.In the present discussionwe wish nousprimary events.The very local manifesta- to emphasizethe situation where the 'events' are tions of age dispersionof one mineral species not so disparate in time.

03 500 600 700 800 900 1000 1100 1200 1500 MOO 1500 "' ALL STRONTIUM AGES OF OUTCROP SAMPLES

Z 500 600 700 800 900 I000 I100 1200 1500 1400 1500 ALL ARGON AGES OF OUTCROP SAMPLES

o Indicates outcrop sample

600 7oo 800 950 •o• •oo o •3•o' 1400 1500 MAXIMUM AGE FOR ALL LOCALITIES AND CORES Fig. 7. YIistogramsshowing the age patterns by the strontium and argon methods on all the outcrop materials studied. The lowest diagram shows the maximum ages obtained at each locality by either method. AGES IN THE PRECAMBRIAN OF TEXAS 4O35

As a consequenceof these resettingprocesses, in a comparatively small time interval. Both the for regionM-scalephenomena we may expect to episodicand continuousdiffusion models of lead obtain a time band with the age distributionde- loss would give a primary age of 1150 to 1200 pendent on the details and time scale of the m.y. for both the Pump Station Hills and the processes,as has beendiscussed in more detail by Franklin Mountains, which is greater by about Wasserburg [1961]. The fact that the maximum 10 per cent than the maximum age obtainedby agesby the strontium and argon methodsat all the other methods. This effect is manifest in the outcrop localitieslie between 1050 and 1090 some other suites of data comparing zircon- m.y. strongly suggeststhat all the rocks are of uranium ages with the strontium and argon the sameage and that the dispersionobserved to ages.The reason for this discrepancyis not un- be commonto them, despite the diversity of derstood. their occurrence,is dominantly the product of local phenomena,although some of the other REGIONAL CONSIDERATIONS effectsmay be present.These age measurements From the comparisonof the results by the are indistinguishablefrom the results obtained various dating methods we conclude that the by previousworkers on the Llano graniteslAid- Franklin Mountains granite, Hueco Mountains rich, Wetherill, Davis, and Tilton, 1958; Goldich granite, Pump Station Hills rhyolite, Carrizo and et al., 1961; Zartman, 1962, 1963]. Van Horn metamorphic rocks and pegmatites, and the Llano granites originated nearly con- U•AN•u•-LE• A•Es oN Z•coNs temporaneously,possibly within a period of 100 The Pb2ø7/Pb2ø• ages on the zirconsare equal m.y. The different lithologiesand modes of oc- to or greater than the maximum ages obtained currence,therefore, representa wide variety of by the strontium and argonmethods. The zircon geologicenvironments in which there has been isotope ratios from the Trans-Pecos and the magmatic activity ranging from shallow intru- Llano localities lie close to a well-defined common sion and vulcanism to deep-seatedplutonic em- line in the concordiadiagram (Figure 3), suggest- placement and middle-rank metamorphism. ing approximate,if not precise,contemporaneity. These contemporaneousevents represent the It is probablethat the zircondata representa dis- manifestations of a single diversified orogenic tinct spreadin agesamong the variousareas but period.

0.21 0.20 Apl)alachi•n- A•ironda :kZircons 0.19

0.18

o.•?

•0.16

0.15

0.14

0,13 ,. /,,•.:--.,-TEX.•,SZIR._,ON C-lORD

0'121.2._ I.• 1.4 1.5 1.6 17 18 1.9 20 2.1 2.• 2.5 2.4 pb207/U235 Fig. 8. Concordia diagram illustrating the location of some Appalachian Grenville zircons. The best-fit straight line for the Texas zircons is shown for reference. 4036 WASSERBURG, WETHERILL, SILVER, AND FLAWN Various workers,including Flawn [1956] and eastern Tennesseeremains open. No basement Bass [1960], have used lithologic character and coresare available from this intervening region. distinctionsin lithologiccharacter to map the Through the generouscooperation of Manuel concealed basement and to make time-tectonic Bassone sampleof cuttingsfrom a well which correlations.The resultspresented above clearly penetrated granophyric granite was obtained indicate the difficultiesinherent in using litho- from south central Tennessee(see Figure 9). A logic characteras the principal basisof classifi- strontium age determinationon a feldsparsepa- cation in time-tectonicproblems. rate gave a result of 1120 m.y. (see Table 7). The zircon ages determined on the granite This result is in agreementwith an earlier inde- core from PecosCounty and the strontiumages pendent determinationby Bass (personalcom- on coresfrom Schleicher,Williamson, and Cole- munication). man countiesindicate that theserocks originated There is a scarcity of basement cores con- contemporaneouslywith all the outcropsstudied. taining mineralssuitable for dating in tile inter- This finding is highly suggestiveof an almost vening region.The availability of such coresin continuousbelt of magmatic and orogenicac- the future will be of great value in clarifying tivity extending from E1 Paso to beyond the this matter. Llano-Burner region--more than 500 miles. The The widespreadoccurrence of the 'Grenville' time of this orogenywas about 1090 m.y. by the event (about 1000 to 1100 by Sr-Rb) in north- Sr87-Rb87 and Ar'ø-K •ø methods and between 1150 eastern North America has been recently dis- and 1200 m.y. by the Pb-U method on zircons. cussedby Tilton, Wetherill, Davis, and Bass These agesare very similar to thoseobtained by [1960]. Silver [1961], has pointed out that previous workers on the Grenville orogenicbelt evidence for 1150 to 1200-m.y. magmatic ac- in the northeastern United States and Canada tivity appearsintermittently from Californiato [see Tilton, Wetherill, Davis, and Bass, 1960]. Greenland.To illustrate the geographicaldistri- Someof the agesin this regionare complicated bution of these ages we have compiled the by a Paleozoicmetamorphism, but where satis- available data, including our results, those of factory comparisonsare availableby the stron- Tilton et al. [1960], and somerecent agesfrom tium and argonmethods the resultsare directly easternCanada, southwest Greenland (Illimaus- comparablewith the data presentedfor Texas. saq), westernUnited States,and the Lake Su- Since the zircon ages are less affectedby the perior regionby other workers [compilationby younger metamorphism we have plotted the Lowdon, 1961; Moorbath, Webster, and Mor- availablezircon data for 'Grenville'samples from gan, 1960; Silver, 1960, 1961; Silver et al., 1962; the northeastern United States and the best-fit Goldichet al., 1961; Fairbairn, Bullwinkel,Pin- chord from the zircon results from Texas on a son, and Hurley, 1959]. These data are shown concordia diagram (Figure 8). Uranium-lead in Figure 9, together with some data in- systemsin Ontario and Quebecare generally dicatingolder terranes,to delineatethe bound- similar but contribute a somewhat broader dis- ary of the Grenville event. These do not persionto the data. All the resultslie essentially representcomplete coverage of the older rocks, on a commonline with a rather goodfit. The as- and none of the Phanerozoic orogenies are sembled age data from these geographically represented.The availabledata suggestthat the widespread regions indicate that the Grenville Grenville orogenicevent is the most widespread orogenyand the episodeof intrusion, extrusion, and pervasiveepisode of Precambrianorogeny and metamorphism found in Texas lie in the and magmatic activity on the North American same time band. Detailed discussion of the dis- continentfor which recognizableevidence is pre- persionof U-Pb systemdata for the type Gren- served. ville of Ontario and the Adirondack-Appala- chianswill be discussedby Silver (in prepara- THE I)ISCREPANCY BETWEEN METHODS tion). It is our opinionthat the orogenyfound in Some uncertainty remains about the time of Texas shouldbe consideredpart of the general the Grenville orogeny,in particular about the 'Grenville'orogenic episode. The questionof the age of the orogenyin Texas. It is manifestedin continuity of this orogenyin the buffed base- part by the dispersionof the strontiumand ar- ment betweenWilliamson County, Texas,and gon agesand by the differencebetween them and AGES IN THE PRECAMBRIAN OF TEXAS 4037

ß 450-600 o GRENVILLE x 1200-1800

ß > 2000

Fig. 9. The occu•ence of 1000- to 1200-m.y. events in North America. Open circles represent 1000- to 1200-m.y. events; crosses,events between 1200 and 2000 m.y.; squares, events greater than 2000 m.y. The solid circles represent the 525-m.y. ages found in the Wichita Mountains. the Pb-U age as determinedfrom the concordia the maximum strontium and argon agesof 1090 construction. As was seen from the data shown m.y. Up to this point we have used the common earlier, all the Pb-U ages on the zircons from linear relationshipexhibited by the zirconsand Texas show varying degrees of discordance. the characteristic age pattern given by the Nevertheless,all these results lie near a single strontium and argon methods to argue for con- straight line on the concordiadiagram with temporaneiW.However, the precisespecification upper intercept of 1150 to 1200 m.y. and an ap- of the age must remain in questionbecause of proximate lower intercept of 200 to 400 m.y. the differences between the methods. This effect From the point of view of episodicloss [Wether- is present in a similar comparisonof the differ- ill, 1956] this impliesthat thesezircons represent ent methods in the 'Grenville' of the eastern systemswith • primary age of 1150 to 1200 m.y. United States and Canada, where the intercept which were subjectto a reductionin their Pb/U on the concordia plot (see Figure 8) is again ratio 200 to 400 m.y. ago. The continuousdif- 1150 to 1200 m.y., whereas the oldest stron- fusion loss model [Tilton, 1960] for a 1150- to tium and argon ages on associatedminerals in 1200-m.y. primary age generates an almost this area are about 1070 m.y. identical line over the range for which the data Possibleexplanations of this discrepancyare are shownwith no inferred secondepisodic event, an error in the U =• half-life superimposedon and therefore the data do not distinguishbe- the 'normal' pattern of lead loss,a loss of inter- tween these mechanismsof discordance.The age mediate daughter (possiblyRn 2•2) from the U • 1150 to 1200 m.y. is significantly greater than chain superimposedon the 'normal' pattern of 4035 WASSERBURG, WETHERILL, SILVER, AND FLAWN lead loss,on a differentialresponse of the various constant of U 23•has been discussedby Aldrich minerals to loss of daughter products. and Wetherill[1958], who have emphasizedthat The effect of increasingthe decay constantfor the best determinationof this decay constantis U •'5 would be to generate a concordia curve the experimentby Fleming, Ghiorso,and Cun- displacedto the right of the curve characterized ningham[1952]. The experimentalerror quoted by the original choice of Xu•'5 and Xu•'8. For by these workers is 2.2 per cent. Aldrich and small changesthe shapeof the curve will not be Wetherill concluded that the uncertainty in drastically altered. It can be seen that the ku•s•is at least 2 per cent. An increaseof the constant fractional loss of an intermediate decayconstant by this uncertaintyfactor would daughterproduct in the U TMchain would give a essentially eliminate the discrepancybetween decreasedPb•ø6/U TM ratio for a given age. Such the zircon and the strontium and argon ages as changeswill decreasethe age determinedby the presentedin this work. There is no independent intersection of a line with the concordia curve. evidence, however, to justify such a change. This is illustrated in Figure 10 for a 2.2 per cent A considerationof the changesin the Rbs7 and increasein •v •. The best-fit line for the Texas K 4ødecay constants necessary to accountfor the zircon data is also shown on the graph. It is differencein the ages indicates that these are evident that the intercept on concordiais widely unlikely. shifted by this change,the 1200-m.y. intercept In this paper we have chosento use the geo- on curve A changingto a l100-m.y. intercept on logically determineddecay constantfor Rb8• curve B. This is due to the small angle of inter- [Aldrich, Wetherill, Tilton, and Davis, 1956]. sectionof the episodicloss line and the concordia There are several recent determinations of this curve. It is difficult to evaluate the possibility decayconstant by countingmethods [Flynn and of a constant fractional loss of intermediate Glendenin, 1959; MeNair and Wilson, 1961; daughter,but it seemsreasonable to expectsuch Egelkraut and Leutz, 1961; Beard and Kelly, lossesto be dependenton the detailedhistory of 1962]. They show a wide range of values,and it each sample and hence to be rather irregular. is not possibleto chooseone of them asdefinitive. The regularloss required for the samplesreported As a matter of convenience we have continued here would be surprising. to use the value ;t = 1.39 X 10-• yr -• for pre- The experimental uncertainty in the decay sentingthe age data. When a definitivedetermi-

O2O

O•,l9

018

017

J3016

015

014

013

0 121.2 15 14 15 16 17 18 19 20 21 22 25 24 pb207/U235

Fig. 10. The graph showsthe displacementof the concordiacurve for a 2.2 per cent increasein ku TM The straight line representsthe best fit to the Texas zircon data. AGES IN THE PRECAMBRIAN OF TEXAS 4039 nation of this constant becomesavailable, all the extensive granitic intrustion [Flawn, 1956, the strontium ages calculatedusing this value p. 68]. The metamorphic rocks of the Carrizo will have to be slightly revised by the same Mountains and Van Horn Mountains which factor. Flawn assignedto the Van Horn Mobile belt were If the decayconstant is significanfiydecreased interpreted as youngerthan the Texas craton be- below the value used here, the strontium and cause they appeared to have been deformed lead agesreported in this paper will be brought against a stable massto the north. The new data closerto concordance.This would in turn imply developedin this study indicate that the meta- a lower argon retentivity in the micas. Such a morphic rocks of Flawn's Van Horn mobile belt change is not suggestedby our evaluation of are the same age as someof the rocksassigned present counting results. If the decay constant to the Texas craton, namely about 1100 m.y. is increasedby as muchas 6 per cent [Flynn and Moreover, rocks in the Pump Station Hills, Glendenin,1959], the age measurementswould Hueco Mountains, and Franklin Mountains west indicate that radiogenicstrontium is preferen- of Flawn's Texas craton are of this sameage. tially lost in both feldsparsand micas in com- In the following discussionthe comparative parison with argon in micas. ages in the various terranes are evaluated by Althoughzircons very commonlydisplay dis- restricting considerationto the data obtainedby cordantages, they are also more likely to pre- the Rb-Sr method, which provides the most serve evidenceof their originalage during some completelocality coverage. metamorphicprocesses because of the peculiari- The micrographicgranite from Eddy County, ties of the two uranium decay schemes.It is New Mexico, was assignedto the Texas craton, thereforepossible that the differencein agesre- but the measuredstrontium age of 1240 m.y. is ported by the strontiumand argonmethods and distincfiy older than that obtainedon the pre- by the zircon lead-uranium method is due to the vious samples. This result is particularly in- fact that the zirconsform partially closedsys- teresting, since on lithologic grounds the ma- tems under higher temperature (or meta? terial is almost identical with many of the morphic) conditionsthan the feldspars,micas, granites of the craton. The metadiorite from and hornblendes.Thus, for example, in deep- Roosevelt County, New Mexico, comesfrom a seated rocks the zircons could possibly begin rather complex area that was assignedto the registering time before other minerals dur- Texas craton.The ageobtained for this material, ing a cooling period. This could possi- 1400 m.y., is the oldest found in our study. bly account for the observations. There is Similar older ages of 1320 and 1400 m.y. are some difiqculty in applying this argument found in metamorphic rocks in cores from to such shallow intrusions as occur in the Crosby and Cottle counties,respectively. These Franklin Mountains and Pump Station Hills. It localities have been assigned by Flawn to the is possiblethat strontium and argon lossesunder Red River mobilebelt. Previouswork by Tilton, normal near-surface conditionsare responsible Wetherill, and Davis [1962] on rocks from for this effect or that a more recent undetected the Arbuckle Mountains also gives agesof 1400 thermal event caused the losses. m.y. by the strontium,argon, and lead methods. The available data do not permit any conclu- Samplesof rhyolites from the Panhandlevol- sionsabout the causeof this discrepancybe- canicterrane in RooseveltCounty, New Mexico, tween the methods.It must be resolvedby fu- and Lubbock County, Texas, and a granite from ture studies. Potter County, Texas, give ages of 1180, 1235, and 1260 m.y., respectively.All these ages are P•EGIONAL GEOLOGICALIMPLICATIONS greater than the 1090 m.y. that appearsso com- The relationshipof the measuredages to the monly farther south acrossthe Texas craton. basement provinces of Flawn [1956] is com- No evidenceof the younger igneousevent of plicated. Flawn's fundamental basementelement 500 m.y. found in the Wichita Mountains, Okla- is the Texas craton, which was defined both on homa [Tilton, Wetherill, and Davis, 1962], has a lithologicbasis as consistinglargely of granitic been found in Texas. Roth [1960] has reported plutonic rocks and structurally on the basis of a whole-rockAr4ø-K•ø age of 650 m.y. on a rhyo- its apparent behavior as a stable element after lite core from Deaf Smith County, Texas. The 4040 WASSERBURG, WETHERILL, SILVER, AND FLAWN result reported by Roth is at best a minimum terrane to the south.These rocks are part of the age, since it is well known that argon is lost larger cratonic structure comprising all the from feldspars.The age he reported was un- buried basementstudied, the southernpart of substantiatedas to analytical accuracyand is this province being a dominantly granttic ter- an isolatedresult of questionablequality. Such rane that appearsto be a southwesternexten- a measurementcannot be seriouslyconsidered sion of the Grenville belt. To the north this evidenceof a younger event. larger cratonic structure contains older rocks of The rhyolites in the Panhandle volcanic ter- 1250- and 1400-m.y. age. The southern bound- rane pose a very puzzling problem. They in- aries are undefined.The precedingassignments clude extrusive rocks which Flawn interpreted are partly arbitrary, and they must be looked as extruded on rocks of the Texas craton and on as tentative until more materials have been therefore younger than the 1100-m.y. granite studied and the dating methods better under- rocks. The new data presented here indicate stood. that they are older. Petrographicallythey are In the oversimplifiedpicture presentedabove, similar to the younger 500-m.y. suite in the we may appear to have assumedthat the oro- Wichita Mountains, but they are also petro- genic events must actually be associatedwith graphically similar to the older Precambrian very sharply definedtimes. The problem of the rhyolites in Missouri. The samples from the dispersionof ages and the existence of time Panhandle volcanic terrain suggestthat this bands makes such a simplified interpretation terrain may be a true time unit. Sinceit is made doubtful. More data on adequatesamples may up almost exclusively of extrusive rocks, it clarify the question of the width of the time must rest on an older (earlier than 1250-m.y.) bands.Our knowledgeof the episodesof igneous basement. and metamorphic activity in the Phanerozoic All the resultsare illustrated in the histogram showsclearly that in certain regionsthere were of Figure 7. There is little doubt that the ages long intervals of time that were almost densely of 1090 and 1400 m.y. are well resolved.The populated with orogenies.Certain events may ages ranging from 1180 to 1260 m.y. are sig- have been more widespread and indicate a nificantly greater than the 1090-m.y. results.It greater intensity of activity, but they were not is not clear whether these 1090-m.y. results always widely separatedin time. If we consider represent a single event or a degradedor dis- the age distribution in the Phanerozoictrans- persedolder event. The older ages(1100 to 1200 lated back in time by 1000 m.y., we would see m.y.) obtained on zircons from the so-called that the events which are resolvedstratigraph- 1090-m.y. rocks also suggestthe possibilitythat ically and sometimesgeochronologically in the the true age of these rocks is about 1200 m.y. Phanerozoicmay (within the experimentalerror If so, the distinctionbetween the 1090- and the and natural dispersion) merge in the trans- 1180- to 1260-m.y. ages becomesuncertain. It lated assemblage,and from this point of view may be noted that all the 1180- to 1260-m.y. could appear more as a continuousdistribution. ages are the same to within analytical error. Our understandingof Phanerozoichisiory indi- From consideration of earlier discussions we cates that we need not always expect to find tentatively concludethat the 1180- to 1260-m.y. sharply defined,widely spacedorogenic episodes results represent a single event at about 1250 in the Precambrian.The only evidencesuggesi- m.y. and that this is dist.inctfrom the 1090-m.y, ing a change in orogenic phenomena between and the 1400-m.y.events. The 1320-m.y.age is the Phanerozoic and the Precambrian is the from a hornblende.The comparisonspresented fact that in North America where many age earlier indicate that this result may be low by measurements have been made there is no several per cent. We thus assign the 1320- to evidencefor an orogenythat occursbetween the 1400-m.y. samplesto a single episodeat about early Cambrian (about 500 to 550 m.y.) and 1400 m.y. The rocks of this age may represent the Grenvillian at about 1100 m.y. The present an o.rogenic belt extending to the Arbuckle study again fails to indicate any event in this Mountains. It now appearsthat Flawn's Texas wide (some 500-m.y.) time gap. Less complete craton is composedof older rocks, 1250 to 1400 data on other continentsindicate a gap of at m.y., to the north, and a younger 1090-m.y least between650 and 1000 m.y. The existence AGES IN THE PRECAMBRIAN OF TEXAS 4041 of this period of orogenicquiescence is a most was completed,the charcoaltraps were warmed remarkable fact deserving considerableatten- and the samplewas.mixed with Ar• tracer which tion. had been introducedpreviously. The gaseswere The outcrop and well core data for all the then purifiedover hot CuO and passedthrough samplesstudied indicate some relationships dif- a cold trap into a titanium furnace.After puri- ferent from those based on the regional litho- fication the argon residue was transferred to logic-structuralinterpretation of the basement the mass spectrometerusing a charcoalsample as presentedby Flawn [1956]. It is evidentthat tube. The samples were run on a 60ø sector a classificationbased solely on lithologic and massspectrometer with a 6-inch radiusof curva- structural criteria in the absenceof a reliable ture. This instrument has two collectors,one of stratigraphy may yield equivocalresults in in- which is a normal Faraday cup and the other an terpreting geologichistory. Flawn emphasized electronmultiplier. Ar'ø/Ar• data were taken on this earlier and pointed out that, in the ab- the simplecollector, Ar•6/AF 8 data on the multi- senceof age data, many of the time-tectoni(' plier. conclusionsare decidedly tentative and con- Discrimination was monitored by running jectural. Several difficultiesare inherent in such normal argonsamples. Samples TV-3M, TV-SBi, an approach.One is the fact that most rock and TV-6Hbd were each run in duplicate for types occur throughout the geologiccolumn, argon; the maximum differenceobserved was 2 and, consequently,a lithologictype is not neces- per cent. The Ar • tracers were made by a sarily a time type. This includesmuch finer pipetting techniquefollowing P. Hurley and J. correlations such as those based o.n accessory ZShringer (personal communication). The par- minerals and mode of occurrence,which some- ticular spike line used in this laboratory was times appear to be more exclusivecriteria. designedand tested by Dr. M. Lanphere,R. E. Perhaps the greatest difficulty with all our Zartman, and G. J. Wasserburg.This apparatus studiesis that a fundamental understandingof consists of a 2-liter reservoir and a double orogenic processesand driving forces is very pipette isolatedby two stopcocksand two mer- incomplete,and our ability to make consequen- cury cutoff reservoirs(Figure 11). The pipette tial predictionsin this field is singularlylacking. We in no way imply that agemeasurements will provide the critical answers,but in favorable circumstancesthey may permit a clearer interpretation of simultaneity (of crystallization) and possiblyadd mo.reknowledge to the mor- to phologyo.f orogenesis, so that somemore funda- so mental insight may provide a basis for under- system ,-.,4cc// standing.

APPENDIX

Analytical Procedures

The analytical procedures were essentially cm those reported in previous papers. Argon was extracted from the minerals by fusion in mo- 6 ( trocer) lybdenum crucibles with a radio-frequency 2 liters heater. All the sampleswere heated to a temperature of 1400øC,producing a fluid melt. One Fig. 11. Argon spike pipette. Drawing not to sample of biotite was heated to produce a scale. An aliquot of the spike is taken by raising sintered mat, and a split of the same material the left-hand column to one of the two cutoff was heated to 1400øC. The a'•'gonyields were positions. The right-hand column is lowered below the same. The gasesproduced during heating the cutoff and the gas is equilibrated. The aliquot is then isolated by raising the right-hand column. were frozen out continuously on charcoal at The fractional depletion per release is about 2 X liquid nitrogen temperature. After the fusion 10-=. 4042 WASSERBURG, WETYiERILL, SILVER, AND FLAWN

TABLE 8. Analytical Data for Rb-Sr and K-Ar Ages

Ar4O*, Sample K, % 10-s mole/g Ar4ø*/Ar4ø Rb, ppm Sr87', ppm Sr87*/SrS7

TH-1 (Bi) 6.19 1. 416 O. 97 2274 8.35 0.90 Tin1 (F) 304.9 1.253 0.36 TH-2 (Bi) 6'(• 1.i• o'• 243.5 O. 684 0.095 TH-5 (Hb) 0.937 0.2195 0.90 TH-5 (F) ...... o'i• TH-6 (F) 3483 15.02 0.95 TV-1 (M) 8'•'• 1.• 0'• 244.2 1. O01 0.50 TV-1 (Bi) 7.. 1.•4o o.. 44.o 1.557 0.61 TV-1 (Rk) 85.42 0.3274 0.14 TV-2 (Hb) 0'• o.6iil4 o'g• TV-3 (M) s.•s .071 0.. 10•' TV-3 (F) ...... 4..0 1.781 0.46 TV-3 (total F) ...... 422.8 1. 736 0.49 TV-4 (M) 663.0 2.791 0.84 TV-5 (Bi) 1.326 0.65 TV-6 (Hb) o..a5.92 o.o•8411.014 o.7.0.94 6•] 5 TV-10a (Bi) o'•:• TV-10b (Hb) o..s 0.0•4s o.so TV-12b (Bi) ...... l•'a o'•io o'i• TV-12c (Rk) ...... 10a.• 0 441 o. 097 TV-13 (F) ...... a74.• I 576 0.26 TV-13 (Rk) ...... lS0..lS4.s 0 724 0.13 TV-14 (F) 0 760 0.11 TV-118 (M) s'•;7.•0 .•g•1.so• 0'•o.•s s7•...7.s 3 796 0.93 TV-252 (Bi) 3 60 0.76

Drill Cores 1. Schleicher Co., Texas (F) ...... 471.1 1.782 0.33 2A. Coleman Co., Texas (F) ...... 676.7 2.807 0.57 2B. Coleman Co., Texas (F) ...... 736.8 3.026 0.26 3. Williamson Co., Texas (F) ...... 346.9 1. 382 0. 093 5. Eddy Co., N.M. (F) ...... 256.7 1.265 0.15 6. Roosevelt Co., N.M. (Hb) 1.19 0.433 0.95 (Bi) ...... •i•'• 7. Roosevelt Co., N.M. (P) ...... 291 .S 1.359 0.11 8. Potter Co., Texas (P) ...... •.s 2.669 0.26 (Rk) ...... 280.6 1.322 0.12 9. Lubbock Co., Texas (F) ...... 234.6 1.150 0. 077 10. Crosby Co.,

Texas (Hb) O. 869 O. 295 O. 87 ...... ß . . 11. Cottle Co., Texas (Bi) ...... 314.1 1.745 0.36 12. Giles Co., Tenn. (F) ...... 182.3 0.810 0.41 AGES IN THE PRECAMBRIAN OF TEXAS 4043 has a sufficientlylarge volume to overcomethe developedat the Department of Terrestrial problems associatedwith capillary tubes. The Magnetismof the CarnegieInstit.ution of Wash- double pipette permits the removal of two dif- ington.We shouldparticularly like to thank the ferent spike aliquots, dependingon the sample membersof that group for aiding us in setting size. The ends of the pipette are tapered to a up the procedures. 2-mm bore and are ground flat. Aliquots of the Samples were run by a surface ionization spike are transferredby expandinginto the sam- technique on a 60ø sector mass spectrometer ple systemand freezing on charcoal.The whole with a 12-inch radius of curvature, with an apparatus is built in a portable frame and can electronmultiplier as the collector.Discrimina- be conveniently moved after cutting the con- tion wasmonitored by runningnormal rubidium nection to the sample line. and strontium. Potassium analyses were done on a Perkin- The rubidium and strontium tracer solutions Elmer flamephotometer using a lithium internal were calibrated by mixing with known normal standard. In addition, all the hornblendeanaly- rubidium and strontium solutions. The normal seswere doneby isotopedilution. The resultsby rubidium solutions were made with spectro- the two methodsagreed to within experimental scopicallypure RbC1 purchasedfrom Johnson error. The estimated maximum error in the Ar•ø/ and Matthey and Co., Ltd., and with RbC1 K • ratio is taken to be 2.5 per cent. purifiedfrom C.P. gradesalt by the RbICl• pro- The atomic abundanceof K 4øwas taken, to be cedure as describedby Archibald [1932]. Both 1.19 X 10-4, and the decay constantsused are salts were analyzed by a conventionaloptical ke - 0.585 X 10-•ø yr -•, k • - 4.72 X 10-•ø yr -•. spectrographicmethod for cationic impurities The isotopedilution proceduresfor the stron- and were found to contain less than 0.01 per tium and rubidium analysesare essentiallythose cent for the maximum impurity. Rubidium and

TABLE 9. West Texas Zircon Analyses

Radiogenic Lead Concen- Isotopic Composition, trations, Observed Lead Isotope Ratios atom % ppm

pb•oo/pb•o4 pb•oo/pb•o* pb•.oo/pb•.OS pb•.OO Pb•o* pb•os Pb *,d U

Micrographic granite (TH-5), Franklin Mrs., E1 Paso Co. Fraction A 905.7 11.141 5. 459 81.80 6.16 12.05 50.4 330 Fraction B 711.5 10.544 4.277 79.38 5.93 14.49 71.8 527 Rhyolite porphyry (TV-13), Pump Sta- tion Hills, Huds- peth Co. Fraction A 351.4 8. 413 3.268 77.71 6.09 16.20 61.6 313 Fraction B 528.4 9. 616 3. 654 77.77 5.99 16.24 54.1 299 Granite well core, Phillips Puckett # lc, PecosCo. Total zircon 459.5 9. 482 4.852 82.98 6.16 10.86 32.4 205 Muscovite-oligoclase- quartz schist (TV-1), Van Horn Mrs., Hudspeth Co. Zircon fraction, pass 200 mesh 177.5 6.244 2.894 81.50 6.53 11.97 42.0 329 Common Pb correction 206/204 207/204 208/204 17.9 15.65 38.0 4044 WASSERBURG,WETHERILL, SILVER, AND FLAWN chlorine were determinedin duplicate on the of the massspectrometer. For a 10 per cent en- Johnsonand Matthey material. Rubidiumwas richment the maximum error is estimated as 6 determinedas the perchlorateafter a butyl percent. A morecomplete discussion of analyti- alcohol-ethylacetate extraction. The chloride cal errors is given by Zartman [1963]. was determinedby precipitatingAgC1 from a The decayconstant used for calculatingthe AgNO•solution. The RbC1was foundto be strontiumages is ,• • 1.39 X 10-• yr-L The iso- stoichiometricto within better than 0.1 per cent. topicratios for normalstrontium used in this The normal strontium solutionswere made work are Sr•/Srss -- 0.1194,Srs•/Sr ss • 0.0843. withspectroscopically pureSrC08 (Johnson and The ratio Sr•/Sr • givenin the last columnof Mattheyand Co., Ltd.) andreagent grade SrC08 Table 8 is the ratio of radiogenicSF • to total SF• (Merck).The cationicimpurities of bothsalts in the experiment. werenegligible. The saltswere analyzed in du- Descriptionoy Thin Sectionso/ Rocksand plicateby the followingprocedures. Strontium Mineral Separates wasprecipitated as the sulfatefrom acid solution containing50 per centby volumeC•H50H Van Horn Mountains (denatured),washed in 95 per centC•H5OH, TV 1. Quartz and albite-oligoclase(79'%), mus- and filtered; the precipitatewas dried to con- covite (15%), potassiumfeldspar (5%), biotite (1%), apatite(tr.), zircon(tr.). Paralleloriented stantweight at 700øC.The saltswere also con- plates of muscoviteand pale-brownbiotite occur verteddirectly to thesulfate. The results showed in a quartz-feldsparmosaic. Grain size: 0.1-0.3 that the SrCO• was stoichiometricto within 0.1 mm. Fabric: crystalloblastic.Rock: biotitic mus- per cent. covite albite-oligoclaseschist. Muscovite separate The interlaboratorystandards MIA-Std mi- (--40 •-80 mesh):98.5% musc., 15% qtz., bio, plag. Biotite separate(--40 -]-80 mesh): 89'%bio, 6% croclineand B-3203 biotite were analyzed.The muse.,1% partly chloritizedbiotite, 4% opaques, results obtained on MiA-Std are 6.78 ppm feldspar, quartz. Sr8•* and 1004ppm Rb. The resultsobtained TV 2. Plagioclase(56%), hornblende(40%), magnetite-ilmenite(4%), apatite (tr.). Blue-green previouslyby L. T. Aldrich(personal com- hornblende in sheavesand small prismatic grains, munication)are 6.90 ppm Sr•* and 1010ppm locallypoikiloblastic, forms a mosaicwith plagio- Rb. The results obtained on B-3203 Bi are clase(oligoclase-andesine). Grain size: 0.2-0.3ram, 1.701ppm SF •* and440.6 ppm Rb. G. W. Weth- hornblendeup to 2 mm. Fabric: crystalloblastic. erill (personalcommunication) had previouslyRock: amphibolite.]¾ornblende separate (--80 •100 mesh): 97% hbd. 0.2% muse.,0.8% opaques, obtained1.707 ppm Sr •* and439 ppm Rb. Our 2% plagioclaseand unidentifiedinclusions. valuesare in goodagreement with the data re- TV 3. Pegmatitic perthitc containingshreds portedby theseworkers. The dataon B-3203 of muscovite,small grainsof quartz, and a trace Bi are somewhatdifferent from the resultsby of apatite. Feldsparquite fresh.Some slight dis- P. Hurley (personalcommunication). coloration. TV 5. Quartz (48%), plagioclase(25%), bio- The CarnegieInstitution rubidium tracer con- tite (25%), muscovite(2%), garnet (tr.), hema- centrationwas determinedby us. Our results tite (tr.), apatite (tr.), zircon (tr.). Dark green agreedto within0.5 per centof thevalue mea- biotite and poikiloblasticsodic (?) plagioclaseoc- suredpreviously by Wetherillet al. (personalcurs in a mosaic of strained quartz. Plagioclaseis partly to completelyaltered to sericite.Grain communication). size: 0.2-3 mm. Fabric: crystalloblastic.Rock: Duplicateanalyses were done on several sam- biotite schist.Biotite separate(--40 -]-80 mesh): ples,including the following. Eddy County, New 96% bio., 1% muse.,1% opaques,2% plag. and Mexico,feldspar: 256.6 ppm Rb, 1.274ppm qtz. SF•*. TH-1 biotite: 2278 ppm Rb, 8.269 ppm TV 6. Plagioclase(51%), hornblende(45%), magnetite-ilmenite(4%), biotite (tr.), apatite(tr.). Sr•*. TV-1 biotite: 438.5 ppm Rb, 1.574ppm Prismatic to large poikiloblasticblue-green horn- SF•*. Theseresults are in goodagreement with blendegrains occur in an aggregatewith incipiently the first measurementsreported in Table 8. sericitizedplagioclase (oligoclase-andesine). Grain The estimatedmaximum error in the Sr•*/ size: 0.2-0.5, hornblendeup to 2 mm. Fabric: Rb• ratiois takenas 2 per centfor highlyradio- crystalloblastic.Rock: amphibolite.Hornblende separate (--40 -]-80 mesh): 96% hbd., 3.6% genicsamples. For samples only slightly enriched opaques, 0.4% apatite. in radiogenicstrontium the errorsare consider- TV 1œB. Plagioclase (45%), biotite (30%), ablyamplified, owing to changesin discrimination hornblende (10%), epidote(7%), microcline(2%), AGES IN THE PRECAMBRIAN OF TEXAS 4045 sphene (2%), magnetite-ilmenite (2%), quartz ? size: phenocrystsup to 3-5 mm, groundmassinter- (1%), apatite (1%). Plagioclase,partly altered to growth 0.2-0.4 mm. Fabric: porphyritic. Rock: epidote, occursas broken twinned subhedra and rhyolite porphyry. mosaic fragments. Green biotite forms a foliated TV. 14. Microcline microperthite (80%), plagio- mass and includes hornblende prisms and grains clase and quartz (15%), amphibole (5%), epidote of epidote. Sphene occurs as abundant clusters of (tr.), magnetite-ilmenite (tr.), zircon (tr.). Micro- small grains. A few large grains of microcline are cline microperthite occurs as large grains and as present, and quartz (?) occurs as a granular ag- smaller grains in an intergranular aggregate with gregate around some feldspar grains. Grain size: sodic plagioclase and quartz. Quartz occurs as average 0.1-0.5 mm, larger grains up to 2 mm. round grains in feldspar and in amphibole. The Fabric: relict hypidiomorphic granular-crystallo- amphibole is a deeply colored green to brown blastic. Rock: hornblende-epidote-biotite schist de- hornblende that occurs in large skeletal grains. rived from alteration of an igneousrock, probably Grain size: perthire I cm ñ, smaller grains 0.1- a diorite. Biotite separate (--100 -]-150): 73% 5 mm. Fabric: panallotriomorphic granular. Rock: bio., 2% hbd., 7% chlor., 2% qtz. and feldspar, hornblende granite. 12% opaques,4% epidote. Franklin Mountains TV 12-C. Plagioclase and microcline (10%), quartz (5%), groundmass(85%), calcite (tr.), seri- TH 1. Microcline microperthite (70%), sodic cite-muscovite (tr.), sphene (tr.), zircon (tr.). plagioclase (10%), quartz (10%), biotite (8%), Large grains of microcline and plagioclase sub- sericite (1%), fluorite (1%), magnetite-ilmenite hedra (phenocrysts) together with fragments of (tr.), zircon (tr.). Microcline microperthite occurs quartz mosaic occur in a crushed and sheared in large grains; plagioclase, as twinned subhedra. groundmassof alkali feldspar and quartz contain- Deeply colored brown biotite forms large irregu- ing sericite fibers. Quartz mosaic fragments may lar skeletal grains and also occurs in small plates. be fragments of broken quartz veins. Other quartz It is partly altered to a fibrous mass that may be veins occur parallel to the flow structure or folia- a mixture of bleached biotite, chlorite, and sericite. tion in the groundmass. Grain size: groundmass Fluorite is in large grains connected by thin vein- 0.01-0.10 mm, phenocrysts up to 2 mm. Fabric' lets. Zircon is abundant. Grain size: 0.5-5.0 mm. relict prophyritic-cataclastic.Rock: shearedrhyo- Fabric: hypidiomorphic granular. Rock: medium- lite porphyry or metarhyolite. to coarse-grained biotite granite. Biotite separate (--20 -t-40 mesh): 82% bio., 4% hbd., 9% Carrizo Mountains opaques,5% qtz., plag., microc. TV 10A. Quartz and albite-oligoclase (74%), TH 2. Plagioclase (60%), biotite (26%), tour- biotite (14%) partly altered to chlorite (20%), maline (8%), sphene (2%), ilmenite (2%), semi- muscovite (7%), garnet (2%), apatite (tr.), zir- opaque titanium (?) mineral (2%), microcline con (tr.), iron oxides (1%), tourmaline (tr.). (tr.), pyrite (tr.), apatite (tr.). Plagioclase (oligo- Thinly laminated schist with alternate coarser- clase?) and biotite occur in a poikiloblastic mo- grained quartz-rich laminas and finer-grained mica- saic. Foliation is well developed. Plagioclase is rich laminas. Orientation of both micas is trans- partly altered to sericite. Large poikiloblasts of verse to lamination and does not show good plagioclase and pink-brown tourmaline enclose orientation. Grain size: 0.1-0.5 mm. Fabric: crys- other rock constituents. The semiopaque mineral talloblastic laminated. Rock' garnet-biotite schist. may be mixed sphene and leucoxene. Grain size: Biotite separate (--100 •-150): 76% bio., 16% poikiloblasts 1.0-2.0 mm., groundmass 0.05-0.20 chlor., 3% qtz. and plag., 1% hbd., 1% musc., mm. Fabric: crystalloblastic-poikiloblastic. Rock: 3% opaques. tourmaline-biotite-plagioclase schist. This is an TV lOB. Itornblende (34%), plagioclase and interesting rock which shows disequilibrium rela- quartz (57%), epidote (6%), opaques(2%), sphene tions. Growth of poikiloblasts may be a reflection and apatite (1%), biotite (tr.). Blades of horn- of near-by granitic intrusion. Biotite separate (--40 blende randomly oriented in a finer-grained mo- -•80 mesh): 75% bio., 24% bio. with opaque in- saic of plagioclase and epidote. Separate clots of clusions,1% qtz. and feld. No chlorite observed. epidote and of hornblende occur irregularly TH 5. Microcline microperthite (75%), myr- through the rock. Magnetite in irregular coarseag- mekite (10%), biotite and amphibole (8%), gregates. Grain size: 0.1-2 mm. Fabric: Rock: quartz (7%), magnetite-ilmenite (tr.), apatite (tr.), amphibolite. Itornblende separate: 88% hbd., 10% zircon (tr.). Microcline microperthite occurs as iron stained hbd., 2% opaques. large grains. Micrographic intergrowths are rare. Ferromagnesian minerals occur as intergranular Hueco Pump Station and Pump Station Hills •nests.' Grain size: microperthite up to 1 cm, in- TV 13. Feldsparphenocrysts (15-20%), largely terstitial grains 0.2-1.0 mm. Fabric: granophyric. altered to clay (?) and rounded and corroded Rock' biotitic hornblende granite. I-Iornblende sep- quartz phenocrysts (75-80%) occur in a grano- arate (--60 -]-200): 69% hbd., 14% bio., 1.5% phyric groundmass.The groundmasscontains seri- microcl., 2% qtz., 8% zircon, 5.5% opaques. cite, chlorite, and biotite (?) as well as finely intergrown quartz and alkali feldspar. Biotite (?) Basement Cores occurs as tiny plates. Zircon (tr.) is present. Grain The majority of cores are described by Flawn 4046 WASSERBURG, WETHERILL, SILVER, AND FLAWN [1956]. The page referenceto the core description ogy by radioactive decay, Ann. Rev. Nuclear is given by the number (FXXX). $ci., 8, 257-298, 1958. 1. Schleicher Co., Texas. Humble No. I Spencer Aldrich, L. T., G. W. Wetherill, G. L. Davis, and 6909-10 feet (F-196). G. R. Tilton, Radioactive ages of micas from 2. Coleman Co., Texas. Naylor No. 1 Stone granitic rocks by I•b-Sr and K-A methods, 5335-38 feet (F-140). Trans. Am. Geophys. Union, 39, 1124-1134, 3. Williamson Co., Texas. Shell and Sinclair No. 1958. 1 Purcell 9474-79 feet (F-202). Aldrich, L. T., G. W. Wetherill, G. R. Tilton, and 4. Pecos Co., Texas. Phillips No. 1-C Puckett G. L. Davis, Half life of Rb s7,Phys. Rev., 103, 14924-30 feet (F-189). 1045-1047, 1956. 5. Eddy Co., New Mexico. Richardson and Archibald, E. H., The Preparation o• Pure Inor- Bass No. 1 Cobb-Federal 16459 feet (F-216). ganic Substances, John Wiley & Sons, New 6. Roosevelt Co., New Mexico. Spartan No. 1-36 York, 1932. State 7240-42 feet (F-232). Biotite separate (--40 Barrell, J., Rhythms and the measurements of •80 mesh): 91% bio., 8% hbd., 1% opaques,feld- geologictime, Bull. Geol. $oc. Am., 28, 858-871, spar, sphene. 1917. 7. Roosevelt Co., New Mexico. Shell Saunders Bass, M. N., Grenville boundary in Ohio, J. Geol., No. 1 8650 feet (F-231). 86% hbd., 6.5% bio., 2.7% 68(6), 673-677, 1960. alkaline feldspar, 2% quartz, 2.8% opaques,plagio- Beard, G. B., and W. It. Kelly, Nuclear Phys., in clase, apatite, zircon. press, 1962. 8. Potter Co., Texas. Shell Bivins Ranch No. Cloud, P. E., V. E. Barnes, and J. Bridge, Stratig- 1-207 located 1000 feet from west line and 660 feet raphy of the Ellenburger group in central Texas from north line, section 207, block 2, AB and M --a progress report, Univ. Texas Publ. 4301. Survey, micrographic granite. 133-161, 1943, 1945. 9. Lubbock Co., Texas. Farris No. 1--11 778-83 Compston, W., P.M. Jeffery, and G. H. Riley, feet (F-175). Age of emplacement of granites, Nature, 186, 10. Crosby Co., Texas. Humble No. 1 Irvin 702-703, 1960. 9947 feet (F-147). tIornblende separate (--80 •100 Egelkraut, K., and tI. Leutz, Z. Physik, 161, 13-19, mesh): 98% hbd., 0.2% feldspar, 1.8% opaques, 1961. epidote, sphene, and zircon. Fairbairn, H. W., H. J. Bullwinkel, W. H. Pinson, 11. Cottle Co., Texas. General Crude No. 13-1 and P.M. Hurley, Age investigations of syenites Swenson5719-20 feet (F-144). tIornblende separate from Coldwell Ontario (abstract), Bull. Geol. (--80 -]-150 mesh): 95% hbd., 3% biotite, 1.6% $oc. Am., 69(12), p. 1543, 1959. feldspar, 1.4% apatite, opaques. Flawn, P. T., Pegmatites of the Van Horn Moun- 12. Giles Co., Tennessee. California Company tains, Texas, Econ. Geol., 46(2), 163-192, 1951. No. I E. W. Beeler NE, NE, section 4-15S-29E Flawn, P. T., Basement rocks of Texas and south- •/• mile SW of Campbellsville 5620-5750. Cuttings. east New Mexico, Univ. Texas Publ. 5605, 1956. Massive granophyric porphyry. Microcline pheno- Fleming, E. tI., Jr., A. Ghiorso, and B. B. Cun- crysts. Contains a blue sodium amphibole (0.5%) ningha.m, The half life of U 2•, Phys. Rev., 88, plagioclase (tr.), the remainder being quartz 642, 1952. and microcline. Flynn, K. F., and L. E. Glendenin, Phys. Rev., 116, 744-748, 1959. Acknowledgments. We should like to acknowl- Gerling, E. K., and A. A. Polkanov, Problems of edge the help of Robert L. Harbour of the U.S. absolute ages of the Precambrian of the Baltic Geological Survey, who generouslyaided us in our shield, Geokhimiya, 8, 695, 1958. collecting trip to the Franklin Mountain area. Giletti, B. J., S. Moorbath, and R. St. J. Lam- Some of the core samples were made available bert, A geochronological study of the meta- to us through the kindness of Dr. 1•. G. Swanson morphic complexes of the Scottish Highlands, of the Shell Oil Company, Dr. Milton Williams Quart. J. Geol. $oc. London, 117, 233-272, 1961. of the Humble Oil and Refining Company, and Goldlob, S.S., A. O. Nier, H. Baadsgaard, J. tI. Dr. James F. Johnson of the Sinclair Oil Com- Hoffman, and H. W. Krueger, The Precambrian pany. geology and geochronologyof Minnesota, Minn. We should particularly like to thank Professor Geol. Survey Bull. 41, 1961. Manuel Bass, whose enthusiastic interest encour- Harbour, R. L., Precambrian rocks at north Frank- aged this work and who gave freely of his time lin Mountain, Texas, Bull. Am. Assoc. Petrol. and knowledge in discussingthe problems. Geologists, 44(11), 1785-1792, 1960. This work was supported by the National Sci- Hart, S. R., The use of hornblendesand pyroxenes ence Foundation grants NSF-G19084 and NSF- for K-Ar dating, J. Geophys. Research, 66(9), G15945, and the Atomic Energy Commission con- tract AT(11-1)-208. 2995-3001, 1961. Holmes, A., Bull. Natl. Research Council, 80, 330- REFERENCES 334, 1931. Jaffe, H. W., D. Gottfried, C. C. Waring, and Aldrich, L. T., and G. W. Wetherill, Geochronol- H. W. Worthington, Lead-alpha age determina- AGES IN THE PRECAMBRIAN OF TEXAS 4047

tions of accessory minerals of igneous rocks, Mineral ages from the Wichita and Arbuckle U.S. Geol. Survey Bull. 1097B, 131-133, 1959. Mountains. Oklahoma. and the St. Francis King, P. B., and P. T. Flawn, Geology and min- Mountains, Missouri, J. Geophys. Research, eral deposits of Precambrian rocks of the Van 67 (10), 1962. Horn area, Texas, Univ. Texas Publ. 5301, 1953. Tilton, G. R., G. W. Wetherill, G. L. Davis, and Lowdon, J. A., Age determinations by the Geo- M. N. Bass, 1000-•n.y.-old minerals from the logical Survey of Canada, Geol. Survey Can. eastern United States and Canada, J. Geophys. Paper 61-17, 127 pp., 1961. Research, 65(12), 4173-4179, 1960. Masson, P. //., Age of igneous rocks at Pump Tilton, G. R., G. W. Wetherill, G. L. Davis, and Station //ills, //udspeth Co., Texas, Bull. Am. C. A. Hopson, Ages of minerals from the Balti- Assoc. Petrol. Geologists, 40(3), 501-518, 1956. more gneiss, Bull. Geol. Soc. o[ America, 67, McNair, A., and l:I. W. Wilson, Phil. Mag., 6, 1469-1474, 1958. 563-573, 1961. Wanless, R. K., J. A. Lowdon, R. D. Stevens, Moorbath, S., R. K. Webster, and J. W. Morgan, R. J. Traill, J. Y. I-I. Rimsaite, and S. Abbey, Absolute age determination in southwest Green- Age determinations by the Geological Survey of land, Medd. Grpnland, 162(9), 5-14, 1960. Canada, Geol. Survey Can. Paper 61-17, 127 pp., Roth, R., Swisher gabbroic terrane of the Texas 1961. Panhandle, Bull. Am. Assoc. Petrol. Geologists, Wasserburg, G. J., Crustal history and the Pre- 4J(11), 1775-1784, 1960. cambrian time scale, Ann. N.Y. Acad. Sci., 91, Silver, L. T., Age determinations on Precambrian 583-594, 1961. diabase differentiates in the Sierra Ancha, Gila Wasserburg, G. J., G. W. Wetherill, and L. A. County, Arizona (abstract), Bull. Geol. Soc. Wright, Ages in the Precambrian terrane of Am., 71 (12), 1973, 1960. Death Valley, California, J. Geol., 67(6), 1959. Silver, L. T., Major gabbroic magmatic events and Wetherill, G. W., Discordant uranium lead ages, geochronology, J. Geophys. Research, 66(8), Trans. Am. Geophys. Union, 37, 320-326, 1956. 2560, 1961. Wetherill, G. W., G. L. Davis, G. R. Tilton, Age Silver, L. T., and S. Deutsch, Uranium-lead method measurements on minerals from the Cutler on zircons, Ann. N.Y. A cad. Sci., 91, 279-283, batholith, Cutler, Ontario, J. Geophys. Research, 1961. 65, 2461-2466, 1960. Silver, L. T., C. R. McKinney, S. Deutsch, and Wetherill, G. W., O. Kouvo, G. R. Tilton, and J. Bolinger, Precambrian age determinations in P. W. Gast, Age measurements on rocks from the western San Gabriel Mountains, California, the Finnish Precambrian, J. Geol., 70, 74-88, J. Geol., in press,1962. 1962. Tilton, G. R., Volume diffusion as a mechanism Zartman, R. E., Thesis, California In•.itute of for discordant lead ages, J. Geophys. Research, Technology, 1963. 65, 2933-2945, 1960. Zartman, R. E., and G. J. Wasserburg, A geo- Tilton, G. R., G. L. Davis, G. W. Wetherill, and chronologic study of a granitic pluton from the L. T. Aldrich, Isotopic ages of zircon from gran- Llano uplift, Texas (abstract), J. Geophys. Re- ites and pegmatites, Trans. Am. Geophys. Un- search,67, 1664, 1962. ion, 33, 360-371, 1957. (Manuscript received May 7, 1962; Tilton, G. R., G. W. Wetherill, and G. L. Davis, revised June 19, 1962.)