Chemical Geology, 97 (1992) %32 9 Elsevier Science Publishers B.V., Amsterdam

[2]

Evidence of magmatic activity related to Middle Jurassic and Lower Cretaceous rifting from northeastern Brazil

(Cearfi-Mirim): K/Ar age, palaeomagnetism, petrology and Sr-Nd isotope characteristics 1

G. Bell ieni a, M.H.F. Macedo b, R. Petr in i c, E.M. Piccir i l lo d, G. Cavazz in i e, P. Comin-Ch iaramont i f, M. Ernesto g, J .W.P. Macedo h, G. Mart ins g, A.J. Melf i g, I.G. Pacca g and A. De Min d

a Dipartimento di Mineralogia e Petrologia, University of Padova, Corso Garibaldi, 37, 1-35 I00 Padova. Italy bDepartamento de Geologia, Universidade Federal do Rio Grande do Norte (UFRN), Natal, Brazil

Clstituto di Geocronologia e Geochimica Isotopica, CNR, Pisa, Italy dlstituto di Mineralogia e Petrografia, University of Trieste, Trieste, Italy

eCentro Studio Orogeno Alpi Orientali, CNR, University of Padova, Padova, Italy fIstituto di Mineralogia, Petrografia e Geochimica, University of Palermo, Palermo, Italy

glnstituto Astronomico e Geofisico, University of Sgto Paulo (USP), Sdo Paulo, Brazil hDepartamento de Fisica Teorica e Sperimental, Universidade Federal do Rio Grande do Norte ((UFRN), Natal, Brazil

(Received June 18, 1990; revised and accepted August 6, 1991 )

ABSTRACT

Bellieni, G., Macedo, M.H.F., Petrini, R., Piccirillo, E.M., Cavazzini, G., Comin-Chiaramonti, P., Ernesto, M., Macedo, J.W.P., Martins, G., Melfi, A.J., Pacca, I.G. and De Min, A., 1992. Evidence of magmatic activity related to Middle Jurassic and Lower Cretaceous rifting from northeastern Brazil (Cear~i-Mirim): K/Ar age, palaeomagnetism, petrology and Sr-Nd isotope characteristics. Chem. Geol., 97: 9-32.

The Mesozoic magmatic activity in the easternmost part of NE Brazil (Cear~i-Mirim) is mainly represented by two- pyroxene tholeiitic dykes; only few dykes have alkaline character. K/Ar ages and palaeomagnetism data indicate that Cear~i-Mirim dykes are of Middle Jurassic ( 175-160 Ma) and Early Cretaceous ( 140-130 Ma ) age.

Both Middle Jurassic and Lower Cretaceous dykes have high incompatible-element concentrations and TiO2 with the lower contents generally confined to the Middle Jurassic dykes; rare tholeiitic dykes low in TiO2 and incompatible ele- ments are present.

Sr-Nd isotopic and other chemical data do not support appreciable crustal contamination and in a 875r/86Sr vs. 143Nd/ 144Nd diagram almost all the dykes plot in a narrow area of the enriched quadrant of the "mantle array".

Tholeiitic and alkaline Ceara-Mirim dykes are related to different parent magmas probably originating by different degrees of melting of garnet-peridotite mantle sources characterized by distinct Sr-Nd isotope compositions and small- scale chemical heterogeneities.

Cearfi-Mirim tholeiitic dykes show close mineralogical, chemical and isotopic similarities with the Lower Cretaceous h igh-TiO2 tholeiites from Maranh~o (Sardinha intrusives ) and Parami (Northern Province) basins.

In general, chemical and Sr-Nd isotope data for high- and low-TiO2 Brazilian Mesozoic tholeiites appear to be related to time-integrated enriched and depleted mantle components, respectively. Subcontinental mantle heterogeneity is be- lieved to be due to variable "enriched" components related to "metasomatic" processes.

The Cear~i-Mirim Jurassic dykes correlate well in terms of composition and tectonic setting with the analogue tholeiitic dyke swarms from the Benue trough, and can be related to the opening of the CentralAtlantic Ocean. The Lower Cretaceous Ceara-Mirim magmatic activity corresponds to the coeval one of the Benue trough and can be related to the early rifting events responsible for the opening of the Equatorial Domain of the South Atlantic whose oceanic crust formed between 120 and 100 Ma.

qGP Project No. 257.

0009-2541/92/$05.00 © 1992 Elsevier Science Publishers B.V. All rights reserved.

10

1. Introduction

Northeastern Brazil was affected by wide- spread basic magmatism which mainly oc- curred towards the present-day continental margin during the Mesozoic (Lower Jurassic to Lower Cretaceous; Fig. 1 ).

In the w e s t e r n a r e a s (Maranh~o and Piaui States) this magmatic activity is represented by extensive Lower-Middle Jurassic (200-170 Ma) tholeiitic flood basalts (Mosquito For- mation: Maranh~o/Parnaiba basin) and Lower Cretaceous (130-120 Ma) tholeiitic sills and dykes (Sardinha Formation ) (Aguiar and Nahass, 1969; Cordani, 1970, 1972; Nunes et al., 1973; Hasui et al., 1976; Sial, 1976; Bel- lieni et al., 1990). In the e a s t e r n a r e a s (Rio Grande do Norte and Cear~ States) the Me- sozoic magmatic activity is represented by a Middle Jurassic-Lower Cretaceous basaltic dyke swarm (Cear~i-Mirim magmatism;

G. BELL IENI ET AL.

Gomes et al., 1981 ), and scarce basalt flows ot Middle Jurassic (180-169 Ma: Lavras da Mangabeira; Bon, 1962; Priem et al., 1978) and Early Cretaceous (130 Ma; Alto de Tou- ros; Asmus and Guazzelli, 1981 ) ages.

The present paper reports new K/Ar ages, palaeomagnetic results, mineral and whole- rock chemistry, Sr-Nd isotopes on the Cear~i- Mirim dykes compared with other Mesozoic magmatic rocks in NE Brazil (Maranh~o basin) and SE Brazil (Parami basin ). The main aims are to establish: (a) relationships be- tween petrological and palaeomagnetic data; (b) basalt genesis and mantle source charac- teristics; and (c) relationships between NE Brazil magmatism and the opening of the At- lantic Ocean. It will be demonstrated that the Cear~i-Mirim dykes correlate with the Meso- zoic basic magmatism from the Benue trough, West Africa, and that they can represent early rifting events related to the Central and Sube- quatorial Atlantic opening.

4'40 4'20 4 ° . 3'B ° - X,s o Lu,s ATL, .,

• r ta l eza O C ' Z . x .. ,,, o < . ' ~ 4*

Mosquito ~ ". C EARA - ~ , _ .4 . . . . . . l r " ^ . ~ T e r e s ) n a __ ~ . .Macau A r 2C ;~'S / r o ~ l l l c ~ t i o "-o

O0 PIAUI . \,~'" I..'.~......~ Aq:Uo ~ I

~')~I~I~J'F ' °~l"ql~bl~'l#~'qlll"ll ' -- I RIO GRANDE 130 I Por t o ~ , i ~ ~ ~| ~ -~ • .qk~' " I ' ' / LAV~RA~DE - N O R T E ~ 1 F r a n c o ' I , - ~ ~ ' " b i P ° c o . . . . . . . . . . ~ l

11 b ~" ~- / P E R N A M B U C O I , 46 ° ~ 414° F o f f ~ 412. ~ 10 ° 3 ; " - 316°RecJfe ~ 8 iI

Fig. 1. Generalized location of Mesozoic magmatic rocks from NE Brazil. 1 = Mosquito tholeiitic flows (Maranh~o basin ); 2 = Sardinha tholeiitic intrusives (Maranhho basin ); 3 = Cear~i-Mirim dyke swarm.

MAGMATIC ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFI'ING 1 |

2. Geology, K/Ar age and palaeomagnetism

The Cear~i-Mirim basaltic dykes are con- centrated in the State of Rio Grande do Norte, where they trend about E-W. To the west (Cear~i State ) they are less abundant and their attitude progressively changes towards the NE-

SW. The investigated dykes subvertically cross- cut the Precambrian crystalline basement (Fig. 2 ), have a maximum thickness of ~ 100 m and may extent for more than 100 km (Rolff, 1965; Sial, 1976).

K/Ar (whole-rock) and apatite fission-track ages from the literature (Cordani, 1970; Sial,

3'o° .-'Mo soR

" ~ 7 - - A L T 0 S A N T 0 - - - - P E D R O ]OAO " 6 ° 3 0 ' ~ ~ A ~U A V E L I NO ,CAMARA

]AGUARETAMA '~ - --WWo - -~ . ~: : -

® ~ - - ~ ~ ~i~, . ANP. ICm S ~ .... o :,\..-~

SOLON6POLE~

F E R N A N D E S , + + + . ~ ,'~-,t,' , ~ ~, " ~

I 2 3 4 , , 5 b , . . . . . . 5 c

Fig. 2. Geological sketch-map of the investigated area. 1 = Mesozoic and Cenozoic sediments; 2, 3, 4 = Precambrian crys- talline basement: 2 = granitic rocks, 3 = Serid6 Complex (gneisses, schists and quartzites), 4 = Caic6 Complex (gneisses and migmatites); 5 = Cear~i-Mirim Mesozoic dykes: 5a = Lower Cretaceous tholeiitic dykes, 5b = Middle Jurassic tholei- itic dykes, 5c= Lower Cretaceous alkaline dykes. I, ..., Vrefer to palaeomagnetic sub-swarms.

TABLE 1

K/Ar whole-rock ages of Ceani-Mirim Mesozoic dykes

Sample Sub-swarm K 4°Arraa Arat m Age (%) ( 10 -6 cm 3 STP g-~) (%) (Ma)

6504 II 1.44 8.17 14.11 141 +-3 6506 II 1.69 14.87 16.34 214+ 11 6517 II 1.58 7.58 50.60 119+-8 6520 II 1.42 7.91 3.23 138 +- 3 6507 II1 1.61 9.48 12.15 145+-3 6511 III 1.36 7.99 21.46 145 +- 7 6512 III 1.46 7.38 34.48 126 +- 3 6521 III 1.12 6.09 14.53 134 __+ 5 6526 III 1.41 7.15 39.73 126 +- 6 6532 IV 1.59 9.47 11.15 147 +__ 7 6558 IV 0.21 11.10 4.50 1,002 +- 42 6530 V 1.99 10.79 7.45 134+-4

12 G. BELLIENI ET AL.

N N 340__ m 20

200 ~ 160 S (a) S (b)

Fig. 3. a. Sample mean magnetization directions/'or the Ceara-Mirim dykes. Open symbols = negative inclinations; solid symbols = positive inclinations. b. Palaeomagnetic poles computed for the Cear~i-Mirim dykes (sub-swarm II and the combination of sub-swarms I, III and IV) plotted along with other South American Mesozoic poles (Ernesto et al., 1992). The Lower Cretaceous poles are mainly from the Paran~i basin volcanic rocks. Circles and contours represent confidence levels.

1976; Gomes et al., 1981; Sial et al., 1981 ) and the present study (Table 1) reveal that the Cear~i-Mirim dykes have two age-frequency maxima at 175-160 Ma (Middle Jurassic) and at 145-125 Ma (Upper Jurassic-Lower Cre- taceous). This age-frequency distribution does not take into account the Precambrian K/Ar ages (Ebert and Bronchini, 1968; present study: Table 1 ) which reflect 4°Ar excess de- rived from the crystalline basement.

The existence of Jurassic and Cretaceous dykes in the Cear~i-Mirim region is supported by the detailed K / A r study of Horn et al. ( 1988 ) whose technique allow removal of the effects of 4°Ar excess. They found Jurassic (minimum and maximum ages 161 and 179 Ma, respectively) and Late Jurassic-Early Cretaceous (minimum age 145-130 Ma) ages. The Jurassic ages of Horn et al. (1988) exclu- sively refer to the 2nd dyke-subswarm of Er- nesto et al. (1992), while the Late Jurassic- Early Cretaceous ones apply to the other dyke sub-swarms.

It is interesting that basaltic magmatism of

Jurassic to Early Cretaceous age ( 193-141 Ma; Popoff et al., 1982; Bowden et al., 1984; Ra- haman et al., 1984) occurs in the Benue trough (Jos area 191-141 Ma, and Burashika area 147 + 7 Ma) and matches quite well that of Cear~i-Mirim at pre-Atlantic configuration. Note that Jurassic ages are reported by Dal- rymple et al. (1975), Westphal et al. (1979) and Vannucci et al. (1989) for the basaltic dyke swarms and other magmatic rock-types from Liberia (193-172 Ma), Morocco (200- 180 Ma) and NW Nigeria (157+7 Ma), respectively.

A detailed palaeomagnetic study was re- cently carried out by Ernesto et al. ( 1992 ) on a large number of samples representative of the Cear~i-Mirim dyke swarm. Results allowed the distinction of five sub-swarms (I-V, from north to south; Fig. 2) whose different direc- tions of magnetization indicate that the stud- ied dykes were emplaced at different times, and probably cooled with different rates.

Fig. 3 shows that the magnetization charac- teristics of the 2nd dyke sub-swarm (Jurassic)

MAGMAT1C ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFTING 1 3

are different from those of the other sub- swarms (Lower Cretaceous). Groups I and 4 display very similar magnetization directions although the magnetic declination angles are all < 180 ° in group 1, while they are > 180 ° in group 4. The mean magnetization direction of group 3 is roughly antiparallel to the mean of groups I and 4, suggesting that these dykes were emplaced under a reversed geomagnetic field. The combined directions of magnetizations of groups 1, 3 and 4 give a palaeomagnetic pole comparable to the Upper Jurassic/Lower Cre- taceous poles of South America. The charac- teristic magnetization of group 2 is not consis- tent with the palaeofield of these geological epochs, and gives a palaeopole corresponding to the Lower Jurassic, as indicated by the available data for South America (Ernesto and Pacca, 1988). However, it is worth to stress that the apparent polar wandering path for South America during the Mesozoic is still poorly defined (Ernesto and Pacca, 1988). Group 5 is defined by two alkaline dykes which show different remanent magnetization. The magnetization of these southernmost dykes (Fig. 2), although not consistent with any dykes of the groups 1, 3 and 4, may also be at- tributed to the Lower Cretaceous (Fig. 3). However, the southernmost alkaline dyke is characterized by an anomalous magnetization.

The "magnetic signature" of the Cear~i- Mirim dykes clearly indicates that each sub- swarm was emplaced at different times. It is interesting to note that the Jurassic dykes of group 2 constitute the most expressive sub- swarm (Fig. 2). Samples from the latter sub- swarm dykes tend to be less coarse-grained than those from the other sub-swarms, indicating a faster cooling rate.

Note that the Upper Jurassic-Lower Creta- ceous palaeomagnetic pole of the Cear~i-Mirim dykes is distinct and older than those corre- sponding to the Lower Cretaceous volcanic ac- tivity and dyke swarm emplacement (e.g., Ponta Grossa Arch) of the Paran~i Basin (SE Brazil ).

3. Petrographic and classification

The Cear~i-Mirim dykes have fine- to me- dium-grain size textures which vary from aphyric to moderately porphyritic in the mar- ginal portions of the thickest dykes. The latter usually show holocrystalline subophytic tex- tures in the inner parts.

Mineral assemblages commonly consist of plagioclase, augite, pigeonite, Ti-magnetite as- sociated with scarce ilmenite, apatite and sul- phides, and occasionally with rare crystals of titanite and zircon. Olivine is sporadic and present as small (<0.4 mm across) com- pletely altered crystals, excepting in a few al- kaline dykes where it is present as medium grain size (max. 1-2 mm) fresh crystals. Me- dium-grained holocrystalline dykes may have quartz-alkali feldspar intergrowths, and rare late-crystallized amphibole mantling the pyroxenes.

In the TAS (Zanettin, 1986) and RI-R2 (De La Roche et al., 1980) diagrams (Fig. 4) most dykes plot in subalkaline fields and correspond to andesi-basalts, latiandesites and latibasalt according to De La Roche's nomenclature. Only two samples from the southernmost dyke group (group 5, Fig. 2) are alkaline (tephrite and trachybasalt). In the AFM diagram most dykes plot between the alkaline and tholeiitic suites of Hawaii (MacDonald and Katsura, 1964).

4. Mineral chemistry

Microprobe analyses were made by means of an Etec®-Autoscan®-Autospec ® system op- erating at 15 kV and 5 mA. The Ortec ® Magic IV version of the Magic program (Colby, 1972) was used to convert X-ray counts into weight percent of the corresponding oxides. Results are considered accurate to within _+ 2- 5% for major elements and _+9% for minor elements.

A close inspection of the mineral composi- tions from the rock-types which constitute the

14 G. BELLIENI ET AL.

R2 CEAR,~-MIRIM MESOZOIC DYKES .~ HTi LTi 7 "~ TEPHRITE ~ BASALTIC \,

Middle Jurassic - - - - o ~, j \ TRACHY- \ "--t-holeiitic dyk-e, & :~C~BASANIT~L ANDESITE X Lower Cretaceous (5 + / T D A r I 4 y - \ ~'

! ~ A • qlp

2000 - - - [ " ° I

BASANITE / ,4 ~ .o.~C a ~ i , ~-o , BASALTIC r / / ASAL ' '

TEPHRITE~'~ "// ~ --~"~'" ~5, S ' 0 2 , w t ' / " / ~ ~ I 45 4 7 4 9 (51 53 ' ~55 5'7

YBA ESI-BASALT ~

J TRACHYANDESITE A N D E S I T

ooo ,

ANDESITE i

I , , s : J t ~ , I , t , , I

500 I000 1500 2000 RI

Fig. 4. Distribution of Cear~i-Mirim Mesozoic dykes in the R ~-R2 diagram of De La Roche et al. ( 1980 ), as modified by Bellieni et al. (1981). Inset: SiO~ vs. alkali (TAS) diagram (Zanettin, 1984). HTi and LTi=high- and low-TiO2, respectively.

different dyke sub-swarms revealed that the el- emental variations are within experimental errors.

4.1. Pyroxenes

Average (mean) microprobe compositions of early- and late-crystallized pyroxenes repre- sentative of the main rock-types of the Cear~i- Mirim dykes are listed in Table 2.

Late-crystallized augite and pigeonite from andesi-basalts and latiandesites have Fe* (Fe 2÷ +Mn+Fe 3+) higher than that for the early-crystallized ones (av. augite 26 vs. 21 Fs%, and pigeonite 43 vs. 38 Fs%). Fe* en- richment is accompanied by Ca decrease for augite (av. 35 vs. 33 Wo%) and Ca increase for pigeonite (av. 8 vs. 11 Wo%). On the whole,

Ca-rich pyroxenes plot below the Skmrgaard (Iceland) pyroxene trend excepting those from trachybasalt, which have higher Ca content (Fig. 5 ).

In some andesi- and latibasalts, augites may coexist with optically homogeneous subcalcic augites (Ca< 0.5 a.f.u.) which are chemically close to those found in some Paran~i dykes (Bellieni et al., 1988). These subcalcic augites are actually formed by fine-scale spinodal to lamellar exsolutions ofaugite and pigeonite (cf. Mellini et al., 1989). Therefore, they were not considered for the calculation of crystalliza- tion temperatures.

Application of Kretz's (1982) and Ishii's ( 1975 ) thermometers yielded average crystal- lization temperatures of 1120-1170 °C for au- gites and 1040-1070°C for pigeonites, respec- tively. These temperatures are consistent with

TA

BLE

2

Mic

rop

rob

e

av

era

ge

(a

v.)

co

mp

osit

ion

s

an

d sta

nd

ard

d

ev

iati

on

(s.d

.) of e

arl

y- (

E)

an

d la

te-

dyke

s

(L)

cry

sta

lliz

ed

p

yro

xe

ne

s fro

m C

ea

ra-M

irim

Me

sozo

ic

K

An

de

si-b

asa

lts

La

tia

nd

esi

tes

Tra

ch

yb

asa

lt

E

L E

L E

L (N

= 1

5)

(N=

15

) (N

= 1

4)

(N=

15

) (N

=I)

(N

=I

av.

s.d

. av.

s.d

. av.

s.d

. av.

s.d

. -

-

An

de

si-b

asa

lts

La

tia

nd

esit

es

E

L E

L (N

=3

) (N

=7

) (N

=7

) (N

=ll

) av.

s.d

. av.

s.d

. av.

s.d

. av.

s.d

.

7~

t~

t"

F~

SiO

z(W

t%)

51

.51

0.7

9

50

.98

0.8

7

51

.41

0.5

1

51

.61

0.4

4

44

.14

TiO

2

1.1

4 0

,28

1

.04

0.4

3

1.0

6 0

.18

0

.81

0.2

4

3.9

5

AI2

03

1

.97

0.6

8

1.8

9 0

.73

1

.73

0.5

4

1.2

0 0

.28

7

.95

Fe

Ot

12

.71

0.8

5

14

.83

1.3

3

13

.52

1.1

2

16

.14

1.9

3

9.7

1

Mn

O

0.3

8 0

.12

0

.39

0.1

3

0.3

9 0

.15

0

.45

0.1

3

0.1

8

Mg

O

15

.13

0,6

2

14

.74

0.7

8

15

.28

0.6

7

14

.29

0.8

9

11

.63

C

aO

1

7.1

6 0

.58

1

5.8

1 0.5

5

16

.42

0.7

8

15

.50

0.7

2

20

.90

N

a2

0

0.0

8 0

.19

0

.05

0.1

1

- 0

.01

0.0

4

1.3

3

Cr2

03

0

.12

0.1

3

0.1

0 0

.14

0

.12

0.1

0

0.1

6 0

.13

0

.05

Su

m

10

0.2

1

99

.83

9

9.9

3

10

0.1

7

99

.84

Fe2

03

*l

_ 0

.10

-

- 7

.95

Si

(afu

) 1

.92

8

1.9

27

1

.93

2

1.9

53

1

.64

9

AI T

M

0.0

72

0

.07

3

0.0

68

0

.04

7

0.3

50

Su

m

2.0

00

2

.00

0

2.0

00

2

.00

0

1.9

99

A1 v

l 0

,01

5

0.0

12

0

.00

9

0.0

07

-

Fe

2+

0

.39

8

0.4

66

0

.42

5

0.5

11

0

,22

3

Fe

3÷

-

0.0

03

-

- 0

.08

0

Cr

3+

0

.00

4

0.0

03

0

.00

4

0.0

05

0

.00

1

Mg

0

.84

4

0.8

30

0

.85

6

0.8

06

0

.64

8

Mn

0

.01

2

0.0

12

0

.01

2

0.0

14

0

.00

6

Ti

0.0

32

0

.03

0

0.0

30

0

.02

3

0.1

11

C

a

0.6

88

0

.64

0

0.6

61

0

.62

9

0.8

37

N

a

0.0

06

0

.00

4

0.0

01

0

.09

6

Su

m

1.9

98

2

.00

0

1.9

97

1

.99

5

2.0

00

46

.39

5

2.3

9 0

.85

5

1.3

5 0

.78

5

2.4

1 0

.56

5

1.2

8 1

.51

3

.43

0

.40

0.0

7

0.5

3 0

.26

0

.56

0.1

4

0.4

3 0

.21

6

.57

0

.33

0.0

9

0.6

3 0

.27

0

.36

0.2

0

0.9

2 1

.51

9

.98

2

3.1

4 2

.70

2

5.4

8 3

.81

2

3.5

5 1

.63

2

5.6

4 3

.00

0

.17

0

.64

0.0

4

0.8

5 0

.09

0

.69

0.1

1

0.8

0 0

.23

1

2.3

3

18

.81

2.3

2

15

.57

2.5

7

18

.11

1.8

4

15

.64

2.6

2

20

.74

3

.85

0.3

3

5.4

2 0

.76

4

.36

0.8

8

5.0

9 0

.87

0

.34

0

.08

0.1

3

..

..

..

0

.18

0

.10

0.1

0

0.0

4 0

.05

0

.07

0.1

5

0.0

9 0

.09

10

0.1

3

99

.74

9

9.8

7

10

0.1

1

99

.89

1.9

4

..

..

1.7

40

1

.98

7

1.9

79

1

.98

6

1.9

75

0

.26

0

0.0

13

0

.02

1

0.0

14

0

.02

5

2.0

00

2

.00

0

2.0

00

2

.00

0

2.0

00

0.0

31

0

.00

2

0.0

07

0

.00

2

0.0

16

0

.25

8

0.7

34

0

.82

1

0.7

46

0

.82

6

0.0

55

-

- 0

.00

5

0.0

03

0

.00

1

0.0

02

0

.00

3

0.6

90

1

.06

3

0.8

94

1

.02

3

0.8

98

0

.00

5

0.0

21

0

.02

8

0.0

22

0

,02

6

0.0

97

0

.01

1

0.0

15

0

.01

6

0.0

12

0

.83

4

0.1

56

0

.22

4

0.1

77

0

.21

0

0.0

25

0

.00

6

- -

-

2.0

00

1

.99

6

1.9

91

1.9

89

1.9

91

(7

(7

(7

t"

z (7

t-

O

r~

D3 ©

Ca

(a

t.%

) 3

5.4

3

32

.80

3

3.8

3

32

.09

4

6.6

8

Mg

4

3.4

6

42

.54

4

3.8

1

41

.12

3

6.1

4

Fe

.2

21

.11

2

4.6

6

22

.36

2

6.7

9

17

.18

45

.28

7

.90

1

1.3

9

8.9

9

10.7

1

37

.46

5

3.8

5

45

.45

5

1.9

8

45

.82

1

7.2

6

38

.25

4

3.1

6

39

.02

4

3.4

7

N=

n

um

be

r of s

am

ple

s.

*~Fe

20

3

=ca

lcu

late

d

acco

rdin

g

to P

ap

ike

et a

l. (

19

74

).

*ZFe

= F

e z +

+ M

n +

Fe

3 +

.

1 6 O. BELLIENI ET AL.

60/~TRACHYBASAL T ~ Ca , / I L A TIANDESITES IA 1/'1

/ ~NOESI-BASALTS] ~Y. r~ 60/ / [ Sze ~aa= o J40~ "-..'llf~ 9,4 ~.d~aa

,o/

/

/ zo~ / / /

lo/ /

Mg 90 8o

A A AA

ooo ° o oo.

I

90 80 70 60 so F e *

~L V V

70 60 50 40 F e *

Fig. 5. Ca-Mg-Fe* (Fe*=Fe 2+ +Mn+Fe 3+) (at%) of Ca-rich and Ca-poor pyroxenes from Cearfi-Mirim Mesozoic dykes. E and L = early- and late-crystallized pyroxenes, respectively.

those reported for pyroxenes from Paranfi ba- saltic flows and intrusives (Bellieni et al., 1988).

4.2. Plagioclase

Microprobe compositions reveal quite dif- ferent anorthite contents from early- to late- crystallized plagioclases.

Early- and late-crystallized plagioclases from andesi-basalts have average (mean) An57 (range: Anso-6o) and An32 (range: An10-49), respectively. Similar average compositional variations are also shown by plagioclases from latiandesites (early An56 range Anso-60; late An34 range An22_49). Early- and late-crystal- lized plagioclases from trachybasalt have an- orthite variation between An60_7~ and An20_45, respectively. In general, the orthoclase content in early-crystallized plagioclase is positively correlated with K20 of the host rock.

Average crystallization temperatures (Kudo and Weill, 1970; Mathez, 1973) for early- and late-crystallized plagioclases calculated for dry condition are 1113-1250 ° and 990-1028°C, respectively. These temperatures are similar to those obtained for the coexisting pyroxenes.

4.3, Fe- Ti-oxides

Ti-magnetite is quite abundant ( ~ 5-10% by volume) and may be present in large crystals (up to 4 mm across). Ilmenite is compara- tively scarce and is given by small elongated laths. Both magnetite and ilmenite quite com- monly show lamellar exsolutions.

Microprobe compositions of homogeneous magnetites and ilmenites are shown in Table 3. Temperatures andfo2-values for homogeneous magnetite-ilmenite pairs were calculated for three samples according to Buddington and Lindsley (1964). A temperature of 1215 °C [ - logfo2 = 7.8; NNO-HM (nickel/nickel ox- ide-hematite/magnetite) buffer condition] was obtained for the fine-grained specimen 6532; temperatures of 740 ° and 721 °C [ - l og fo2 = 17.4 and 16.9, respectively; QFM-MW (quartz/fayalite/magnetite-magnetite/wfis- tite) buffer condition] were obtained for the specimens 6518 and 6508, respectively.

5. Whole-rock compositions

Major- and trace-element contents were de- termined by X-ray fluorescence using the pro cedure described in Bellieni et al. ( 1983 ). Re- sults are considered accurate to within + 2-5%

MAGMATIC ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFTING 17

TABLE 3

Average (av.) microprobe analyses (in wt%) and standard deviations (s.d.) of early- (E) and late- (L) crystallized Ti-magnetites and ilmenites from Cear~-Mirim Mesozoic dykes

(A) Ti-magnetites

Andesi-basalts Latiandesites Trachybasalt

E L E L E L (N--19) (N=7) (N= 14) (N--7) (N=I ) (N=I )

av. s.d. av. s.d. av. s.d. av. s.d.

SiO2 0.18 0.04 0.20 0.05 0.19 0.05 0.18 0.05 0.25 0.17 TiO2 21.99 5.65 23.53 6.27 26.67 5.95 28.06 3.93 7.93 24.30 A1203 1.86 0.85 1.47 0.74 1.61 0.50 1.25 0.64 0.50 1.20 Cr203 0.30 0.20 0.30 0.32 0.34 0.34 0.20 0.19 - Fe203* 24.31 11.62 25.01 15.42 14.97 12.12 12.72 8.48 55.56 20.85 FeO 50.53 5.66 50.07 6.65 54.83 4.89 55.74 3.38 36.21 45.58 MnO 0.53 0.38 0.82 0.33 1.00 0.34 0.90 0.16 1.72 7.62 MgO 0.23 0.25 0.38 0.50 0.21 0.19 0.18 0.22 0.14 CaO 0.55 1.60 0.33 0.35 0.16 0.19 0.48 0.65 1.76 0.62

Sum 100.50 102.11 99.99 99.72 103.93 100.47

% ULV.* 62.81 67.29 76.12 79.96 22.72 68.91

(B) llmenites

Andesi-basalt Latiandesites Trachybasalt

L E L L (N=5) (N=7) (N=2) (N=I )

av. s.d. av. s.d. av. s.d.

SiO 2 0.24 0.02 0.19 0.06 0.11 0.06 0.19 TiO2 47.99 4.48 45.02 5.39 45.14 6.14 43.65 A1203 0.38 0.33 0.68 0.40 0.53 0.38 0.79 Cr203 0.21 0.12 0.14 0.16 - - 0.18 Fe203 *t 7.50 9.38 14.17 10.94 13.68 12.88 17.08 FeO 41.58 4.41 38.15 5.49 39.02 6.13 35.36 MnO 0.94 0.37 1.03 0.29 1,07 0.03 2.71 MgO 0.42 0.33 0.30 0.17 0,31 0.36 0.40 CaO 0.12 0.09 0.77 1.71 0.04 0.01 0.52

Sum 99.38 100.44 99.89 100.87

% R203" 7.95 14.62 13.87 17.55

N = number of samples. *Fe203, % ULV. (percent ulv6spinel ) and % R203 after Carmichael ( 1967 ).

for major elements, and better than _+ 10% for trace elements.

Selected chemical compositions of seventy samples representative of different dyke sub-

swarms and also of the lithological variation across the thickness, from margin to core, for the thickest dykes are given in Table 4. Sr-Nd isotope ratios for dykes of the 1 st to 5th dyke

TA

BL

E 4

Ma

jor-

(wt.

%) a

nd

tra

ce- (

pp

m) e

lem

en

t, an

d S

r-N

d is

oto

pe

co

mp

osi

tio

ns of s

ele

cte

d

Ce

ar~

i-M

irim

Me

sozo

ic dy

ke

s

Su

b-

I I

I sw

arm

S

am

ple

66

11

6500

65

01

SiO

2

53

.99

5

3.0

6

53

.27

T

iOz

2.6

0

3.1

6

2.9

0

A1

20

3

15

.34

1

4.0

6

15

.29

F

eO

t 1

0.8

1

12

.82

1

1.5

7

Mn

O

0.1

7

0.1

9

0.1

8

Mg

O

4.2

6

3.8

7

3.5

5

Ca

O

6.5

6

7.2

0

6.9

3

Na

20

2

.85

2

.67

3

.39

K

20

2

.58

2

.08

2

.03

P

20

5

0.8

4

0.9

1

0.8

9

Fe

20

3 *~

1.7

6

4.6

4

2.7

0

Fe

O *~

8

.99

8

.28

8

.89

L

OW

2

.05

2

.26

1

.57

mg

.2

0.4

4

0.3

8

0.3

8

Q.3

4

.97

6

.25

3

.87

O

1/H

y .3

-

_ H

e .3

_

_

Cr

34

3

6

35

N

i 2

6

20

2

0

Ba

8

88

9

19

9

58

R

b

18

6

46

4

5

Sr

48

8

55

4

57

3

La

4

6

61

5

2

Ce

9

8

11

8

10

6

Zr

32

9

36

9

33

3

Y 4

8

52

4

5

Nb

3

0

32

2

9

Nd

4

9

71

6

6

(87

5r/

86

Sr)

m

- 0

.70

64

7(2

) R

o(1

70

Ma

) -

- R

o(1

30

Ma

) -

0.7

06

05

(~

43

Nd

/ 0

.51

23

7(4

) 14

4Nd)

m

I I_

I II

6502

66

12

6612

b 65

83

6584

65

85

6506

c

en

tre

ma

rgin

ce

ntr

e i

nte

rm,

ma

rgin

53

.32

5

3.6

0

54

.09

5

4.6

8

54

.19

5

4.2

4

52

.34

3

.09

2

.95

3

.20

3

.17

3

.21

3

.23

3

.32

1

4.7

8

15

.67

1

5.1

6

13

.45

1

3.2

7

13

.30

1

4.7

3

12

.04

1

1.8

7

12

.46

1

2.5

9

13

.10

1

2.8

0

12

.73

0

.21

0

.17

0

.17

0

.19

0

.18

0

.19

0

.17

3

.48

2

.99

2

.88

2

.93

2

.95

2

.91

4

.41

7

.08

6

.75

6

.04

6

.88

6

.81

7

.01

6

.63

2

.81

3

.16

3

.07

3

.02

2

.96

2

.94

3

.09

2

.36

2

.02

2

.10

2

.17

2

.29

2

.34

2

.03

0

.85

0

.83

0

.83

0

.92

1

.05

1

.03

0

.55

3.8

0

4.1

9

4.7

0

4.0

8

4.5

9

4.2

2

4.1

5

8.2

8

7.8

3

7.9

3

8.6

7

8.7

6

8.7

6

8.7

0

1.8

8

2.1

8

2.3

2

1.7

8

1.9

0

1.7

7

1.6

7

0.3

7

0.3

4

0.3

2

0.3

2

0.3

1

0.3

2

0.4

1

5.6

3

5.8

6

7.4

7

8.1

6

7.6

3

7.6

2

2.3

5

45

3

2

29

3

6

39

3

6

40

2

3

26

2

8

26

1

8

22

3

4

88

9

66

6

69

4

62

3

69

3

65

1

73

4

58

4

9

54

5

7

58

5

9

64

5

64

4

95

4

80

5

68

5

78

5

82

3

02

5

3

52

5

1

46

4

7

51

4

9

10

5

10

0

10

2

89

8

2

80

8

7

35

1

33

0

34

0

33

9

34

2

34

2

30

2

46

4

9

49

4

9

50

4

8

48

3

0

31

3

2

34

3

5

35

2

6

66

4

9

57

4

4

42

4

4

57

- -

- 0

.70

62

3(2

) -

- -

0.7

04

74

..

..

0

.51

24

5(4

)

II

11

6606

65

19

ce

ntr

e

51

.27

5

2.7

9

3.5

2

3.2

2

15

.58

1

5.4

1

12

.62

1

2.0

6

0.1

7

0.1

7

3.8

8

3.5

8

7.6

9

7.2

6

3.0

6

2.8

2

1.7

0

2.1

3

0.5

2

0.5

5

2.8

2

3.3

9

9.8

5

8.7

3

1.7

1

1.7

3

0.3

8

0.3

8

2.3

5

4.7

7

54

4

1

55

4

1

50

6

69

1

47

5

2

49

1

51

2

30

4

2

62

8

8

25

3

27

5

39

4

7

24

2

2

34

5

0

- 0

.70

65

4(2

) -

0.7

05

83

- 0

.51

24

7(2

)

II

II

6518

65

89

6590

66

04

ma

rgin

ce

ntr

e m

arg

in

50

.69

5

4.7

5

53

.42

5

1.9

2

3.8

6

3.2

1

3.3

9

3.6

5

13

.60

1

5.2

7

14

.20

1

5.2

1

14

.52

1

0.9

7

12

.55

1

2.6

2

0.1

8

0.1

6

0.1

8

0.1

7

5.6

6

3.1

8

3.3

7

3.5

5

7.1

6

6.8

8

7.4

7

7.0

2

2.3

8

3.4

1

3.0

5

3.4

3

1.6

1

1.6

2

1.7

6

1.8

7

0.3

5

0.5

6

0.6

0

0.5

6

5.9

8

2.9

0

3.3

9

2.0

4

8.6

7

8.0

4

9.2

6

10

.52

2

.63

1

.81

1

.82

1

.72

0.4

4

0.3

7

0.3

5

0.3

6

3.0

9

7.2

7

6.1

1

2.2

1

67

2

3

31

4

9

58

2

6

32

5

3

69

3

55

3

50

9

61

4

38

3

6

38

5

3

47

0

53

9

55

5

49

0

40

4

6

34

3

1

78

8

0

59

7

3

28

5

27

8

29

8

25

1

47

2

5

47

45

4

4

39

2

8

29

2

3

51

3

0

33

TA

BL

E 4

(co

ntin

ued)

Su

b-

II

11

I1

swar

rl'l

Sa

mp

le

6593

65

17

6516

65

15

ce

ntr

e m

arg

in

II

6504

1I

II

II

I1

6595

65

94

6520

65

86

6588

65

87

6514

ce

ntr

e m

arg

in

ce

ntr

e in

term

, ma

rgin

ce

ntr

e

I1

I11

6513

66

07

6511

m

arg

in

ce

ntr

e

6510

65

09

inte

rm, m

arg

in

SiO

2

TiO

2

A12

03

Fe

O,

Mn

O

Mg

O

Ca

O

Na

20

k2

0

P2

05

F%

Os *

~

FeO

* LO

I °j

mg

*z

Q.3

O

I/H

y .3

N

e .3

Cr

Ni

Ba

Rb

S

r L

a

Ce

Z

r Y N

b

Nd

(87

Sr/

S6

Sr)

m

Ro

(17

0M

a)

Ro

( 13

0 M

a)

( ~4

SN

d/

14

4N

d)m

52

.28

52

.35

53

.30

52

.97

3

.84

3

.32

3

.00

3

.06

1

3.4

9 1

5.5

2 1

5.8

0 1

4.3

5

13

.61

12

.54

12

.23

12

.61

0

.19

0

.17

0

.17

0

.19

3

.79

3

.44

3

.44

3

.54

7

.50

7

.05

6

.92

7

.29

2

.88

3

.15

2

.99

3

.04

1

.77

1

.97

1

.96

2.

08

0.6

4

0.4

9

0.2

0

0.8

6

3.9

0

3.7

1

3.9

6

4.3

6

.98

0

8.9

4

8.2

8

8.3

8

2.07

1

.44

2

.21

1

.92

0.3

6

0.3

6

0.3

6

0.3

6

5.1

8

3.3

3

4.5

3

4.6

1

49

4

8

37

4

2

40

4

3

31

2

8

50

9

70

9

71

3

78

8

48

5

2

48

5

0

49

3

53

2

58

1

54

4

25

4

6

47

5

6

57

8

5

95

1

06

2

83

2

73

2

99

3

59

4

1

45

4

8

55

4

8

22

2

5

30

2

9

53

5

1

71

51

.16

3

.77

1

4.8

6

13

.86

0

.20

3

.66

7

.53

2

.60

1

.88

0

.48

4.3

8

9.5

4

2.0

9

0.3

5

3.9

9

53

.19

54

.48

5

2.4

0 5

3.1

3 5

2.7

0

53

.75

5

4.2

6

3,7

2

3.5

6

3.4

9

3.3

9

3.6

4

3.5

4

2.8

0

13

.66

14

.36

1

5.1

9 1

5.1

3 1

3.7

2

14

.91

1

5.8

3

13

.08

1

1.7

0

13

.25

12

.54

13

.34

1

1.8

7

11

.55

0

.19

0

.18

0

.19

0

.17

0

.19

0

.18

0

.17

3

.23

3

.35

2

.90

2

.85

3

.40

3

.16

2

.53

7

.70

6

.94

7

.35

7

.17

7

.71

7

.12

6

.81

2

.80

2

.99

2

.85

3

.39

2

.96

3

.20

3

.47

1

.85

1

.84

1

,89

1

,69

1

.74

1

.69

1

.98

0

.58

0

.59

0

.50

0

.54

0

.59

0

.57

0

.61

4.5

7

3.8

3

4.1

0

2.82

3

.08

2

.98

2

.82

8

.68

7

.86

9

.19

9

.78

1

0.3

4

8.8

8

8.8

8

2.0

7

2.3

1

2,0

9

1.6

4

1.6

1

1.3

9

1.4

3

0.3

3

0.3

7

0.3

1

0.3

1

0.3

4

0.3

5

0.3

1

6.9

1

8.2

3

5.3

1

4.9

2

5.5

3

5.6

8

5.5

1

59

5

3

37

5

1

45

4

1

82

8

51

6

58

4

40

4

6

46

4

55

5

36

5

04

4

3

38

5

2

72

6

2

85

2

75

2

87

2

82

4

8

42

4

6

25

2

9

29

5

1

34

5

0

- 0

.70

61

7(3

) 0.7

06

45

(2)

- 0

.70

55

2

0.7

05

83

49

4

2

47

4

4

30

4

4

45

3

8

41

3

4

67

9

50

4

48

8

55

5

75

7

44

4

3

46

4

6

49

5

31

4

83

5

03

4

97

5

58

4

5

36

2

4

46

4

9

79

7

5

58

8

1

95

3

00

2

86

2

82

2

74

3

23

4

4

42

4

0

43

5

1

27

2

6

28

2

8

27

5

2

37

3

5

48

6

0

- -

0.7

05

75

(2)

- -

- 0

.70

51

3

- -

- 0

.51

24

6(4

) 51

.60

5

5.6

6 5

1.0

8

3.6

0

2.5

2

2.9

8

13

.95

1

7.0

0 1

4.9

8

13

.10

9

.92

13

.01

0

.19

0

.14

0

.19

5

.26

2

.04

5

.37

7

.66

6

.00

6

.96

2

.48

3

.96

3

.18

1

.70

2

.13

1

.73

0

.47

0

.63

0

.54

3.3

7

1.5

3

3.8

6

9.7

8

8.3

8

9.1

7

1.6

6

1.5

0

2.1

6

0.4

5

0.2

9

0.4

5

4.0

0

5.9

0

- -

0.0

3

62

1

4

31

6

2

24

4

8

66

0

72

8

72

5

48

5

0

68

4

84

5

30

5

89

3

9

39

4

3

69

8

2

82

2

67

3

28

2

53

4

2

47

4

0

21

2

7

22

4

5

48

4

5

- 0

.70

72

0(2

)

- 0

.70

65

8

50

.96

5

0.9

9

3.0

3

3.1

2

15

.02

1

4.7

9

13

.37

1

3.3

0

0.19

0.

20

5.2

5

5.2

9

6.9

5

6.6

2

2.9

5

3.2

5

1.7

9

2.0

0

0.4

9

0.4

4

4.5

3

5.1

7

8.8

9

8.2

6

2.4

1

2.3

0

0.4

4

0.4

5

0.4

5

-

- 0

.17

48

4

3

54

4

8

72

9

73

2

57

6

5

61

8

63

3

40

4

0

82

6

7

23

8

23

6

40

4

3

22

2

1

44

4

8

7=

TA

BL

E 4

(co

ntin

ued)

ba

Su

b-

111

swa

rm

Sa

mp

le

6512

111

Ill

Ill

6526

65

25

6524

65

21

6507

ce

ntr

e

inle

rm,

ma

rgin

ce

ntr

e

Ill

Ill

IV

6508

65

29

6605

65

58

ma

rgin

c

en

tre

IV

IV

6557

66

01

6532

m

arg

in

W

IV

V

6600

65

30

6602

SiO

2

TiO

2

A1

20

3

Fe

O,

Mn

O

Mg

O

Ca

O

Na

20

K

20

P

20

5

Fe2

Os *m

F

eO

*~

LO

I *~

mg

.2

Q.3

Ol/

Hy

.3

Ne*

3

Cr

Ni

Ba

R

b

Sr La

C

e

Zr

Y Nb

N

d

(87

Sr/

51

.48

5

2.8

6

51

.07

5

0.0

0

50

.87

5

3,4

5

3.0

5

2.8

4

3.2

4

3.2

3

3.1

2

2,8

2

14

.41

1

4,8

1

14

.38

1

3.9

8

14

.93

1

6.0

1

13

.35

1

2.0

8

13

.55

1

4.2

8

13

.76

1

1.5

5

0.1

9

0.1

7

0.2

1

0.2

4

0.1

8

0.1

6

5.3

0

4,5

8

5.3

6

6.3

4

4,4

8

3.3

8

7.0

2

7,2

4

7.6

5

8.0

6

7.9

6

6.9

1

2.7

5

2,9

0

2.5

5

2.6

0

2.6

4

3.1

5

1.9

0

1.9

4

1.5

2

0.8

9

1.5

1

2.0

0

0.5

2

0.5

9

0.4

7

0.3

7

0.5

5

0.5

6

4.5

9

3.7

7

5.4

2

7.9

9

6.1

6

3.7

9

8.8

5

8.3

6

8.2

7

6.5

9

7.8

2

7,8

4

2,3

5

2,0

7

2,4

7

2.9

9

2,3

5

2,0

1

0.4

5

0,4

3

0.4

4

0.4

7

0,4

0

0.3

7

1.8

2

3.7

0

2.7

6

1.1

2

2,6

8

4.6

3

51

43

5

4

56

6

2

35

5

5

44

5

6

57

6

3

33

7

18

6

71

6

28

6

36

6

66

7

70

5

7

62

3

8

25

3

0

53

5

23

5

69

5

12

5

06

5

06

5

55

4

5

43

4

0

37

3

9

48

8

0

85

8

1

66

7

8

92

2

46

2

60

2

46

2

48

2

44

3

06

4

4

43

4

1

42

4

5

53

2

0

21

2

0

22

2

1

24

4

7

48

4

3

42

4

3

59

50

.07

5

2.1

6 5

3.2

3 5

5.8

8

3.9

4

3.4

1

2.6

4

1.4

9

13

.77

1

4.6

1 1

6,5

2 1

6,0

5

14

.62

1

3.7

5 1

1.9

1

9.8

4

0,22

0

.19

0

.17

0.

13

5,0

9

3.4

7

2,9

1

5.3

2

8.1

0

7,0

3

6.7

5

7.2

3

2.5

9

2.7

6

3,3

3

3.5

6

1,2

3

2.0

3

1.9

3

0.2

6

0.3

7

0.6

1

0.6

0

0.2

3

'6.1

9

3.7

9

3.4

8

3.0

3

8,5

7

10

.00

8

.45

6

.94

2

.73

1

.90

2

£8

1

.33

0,4

1

0.3

4

0.3

3

0.5

2

2,4

1

4,6

7

3.9

7

7.0

7

76

14

2

6

12

7

45

2

8

35

7

9

60

5

81

3

73

7

14

1

31

5

3

50

1

2

44

4

46

3

59

2

42

9

41

5

3

40

1

2

71

9

8

78

1

6

28

1

30

5

28

7

69

4

5

50

4

4

28

2

3

29

2

6

2

44

5

6

4l

21

g6

Sr)

m

.4

0,7

06

95

(2) 0

.71

07

2(2

) -

- 0

.70

64

7(2

) 0.7

06

14

(2)

- R

o(1

70

Ma

) .

..

..

..

R

o(1

30

Ma

) 0

.70

63

6

0,7

10

13

-

- 0

.70

61

5

0,7

05

63

-

'43

Nd

/ ~

44

Nd)m

*4 0

.51

24

5(3

) 0.5

12

44

(3)

..

..

.

- 0

.70

58

5(2

)

- 0

.70

57

0

- 0

.51

20

8(4

)

55

.88

5

3,5

4 5

2.7

7

1.5

0

3.3

1

3.1

4

15

.79

1

3.8

8 1

4.9

4

9.7

7

13

,36

13

.50

0

.13

0

,17

0

.19

5

.37

3

.37

3

.21

7

.22

6

.99

6

.39

3

.82

2

.85

3

.46

0

.28

1

.85

1

.85

0

.23

0

,66

0

,54

3.1

0

3.8

2

3.9

2

6.8

2

9.6

7

9.6

6

1.22

1.

83

1.79

0.5

3

0,3

4

0.3

3

5.9

1

7.0

6

3.2

6

133

12

2

5

63

3

2

36

3

67

6

88

7

96

1

4

50

5

2

43

7

45

2

49

1

12

4

1

54

2

3

79

9

8

70

3

11

2

98

2

3

50

5

0

2

31

2

8

1,2

44

60

- 0

.70

66

4(2

)

- 0

.70

60

7

- 0

.51

24

0(4

)

53

.48

46

,75

4

7.5

9

3.3

7

2,3

7

2.1

5

14

.62

17

,63

1

9.3

9

12

.98

12

.24

9

,73

0.

19

0.2

4

0.1

5

2.7

1

6.2

7

4.5

3

6.8

7

8.3

0

9.1

1

3.2

6

2.8

6

5.3

4

1,8

7

2.7

8

1.4

2

0.6

4

0.5

5

0,5

9

2,5

0

6.0

9

0.2

8

10

,50

6

.17

9

,43

1

.70

4

.33

0

,36

0,3

0

0.5

1

0,4

8

5,9

3

4,1

8

- -

12

.84

10

3

1

56

3

1

44

8

9

59

6

75

7

30

2

48

1

03

2

2

45

3

74

6

762

43

2

6

21

9

0

56

4

5

29

8

16

4

16

7

45

2

5

2l

30

3

5

34

4

5

28

1

5

0.7

08

46

(2) 0

.70

31

0(4

)

0.7

07

72

0

.70

29

4

0.5

12

75

(4) 0

.51

29

0(3

)

1 ....

. v r

efe

r to

pa

lae

om

ag

ne

tic

su

b-s

wa

rms.

R

~=

init

ial 8

7Sr/8

6Sr ra

tio

. * ~

Ori

gin

al va

lue

s.

*2A

tom

ic Mg

] (M

g +

Fe

2 +

) ( F

e2

03

/Fe

O

= 0

.15

).

*3C

IPW

-ao

rma

tiv

e

qu

art

z (Q

). o

liv

ine

/h,v

pcrs

ten

e

(Ol/

Hy

) ra

tio

an

d ne

ph

eli

ne

(Ne

).

*4M

ea

sure

d

(m)

rati

os.

z m >

r-

MAGMATIC ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFTING 21

sub-swarms, as defined on the basis of palaeo- magnetic data, are also given.

In general, the Ceara-Mirim tholeiitic dykes correspond to evolved basalt compositions with atomic Mg/ (Mg+Fe 2+ ) ratio (=mg) in the range 0.53-0.32 (mg based on Fe203/ FeO=0.15). All but one dyke (samples 6557 and 6558) are characterized by high contents of incompatible elements, and P205 and TiO2, e.g., Zr 230-369, Ba 480-958, La 23-61 and Ce 58-118 ppm, and P205 0.3-1.0 and TiO2 2.5-3.8 wt%. Dykes with comparatively low contents of incompatible elements (e.g., Zr 69- 70, Ba 141-367, La 12-16 and Ce 16-23 ppm) are also characterized by low P205 (0.23 wt%) and TiO2 ( 1.5 wt%).

No significant distinction was noted among Cretaceous tholeiitic dykes belonging to the 1 st, 3rd and 4th sub-swarms. The Jurassic tholei- itic dykes (2nd sub-swarm) tend instead to have, for similar MgO content, slightly lower amounts of incompatible elements (e.g., P205, K20, Y, Zr, La, Ce) (Figs. 6 and 7 ).

Variation diagrams show that both the Jur- assic and Cretaceous Cear~i-Mirim dykes have a large elemental scattering (Figs. 6 and 7). However, it is apparent that, with decreasing MgO, there is a general increase in SiO2, Na20 and incompatible elements, and a decrease of CaO, Ni, Cr (FeO~, TiO2), while Sr remains virtually constant. These variations are broadly compatible with gabbro fractionation. The chemical variations across the thickest dykes (Table 4 ) indicate that the central portions are usually more evolved (e.g., lower mg-values), relative to the outer margins.

6. Sr-Nd isotopes

Sr-Nd isotope compositions were deter- mined using the analytical procedure de- scribed in Bellieni et al. (1991). NBS-987 standard analyses gave average values of 0.71026_+ 0.00003; the average 143Nd/144Nd

ratio obtained for the La Jolla standard is 0.51187_+0.00002 and no correction was ap- plied for instrumental bias. Standard devia- tions of the isotopic ratios are expressed at the 95% confidence level (as 2a of the mean value); Sr blank values are < 2 ng.

Based on the radiometric ages and palaeo- magnetic results the initial SVSr/86Sr ratios (Ro) were calculated back to 170 and 130 Ma for the Jurassic and Cretaceous dykes, respectively.

The Jurassic dykes yielded R0 ranging from 0.7051 to 0,7058 (av. 0.7054+0.0005), while the measured 143Nd/144Nd ratios vary from 0.51245 to 0.51247. The Cretaceous tholeiitic dykes show instead wider R0 ranges (0.7057- 0.7101; av. 0.7067_+0.0014) and 143Nd/la4Nd (0.51208-0.51245; av. 0.51235+0.00015), while the two Cretaceous alkaline dykes are characterized by R0=0.7029 (tephrite) and 0.7077 (trachybasalt), and J43Nd/la4Nd= 0.51290 (tephrite) and 0.51275 (trachyba- salt). Note that the trachybasalt dyke is quite altered (high celadonite content).

In the present-day 143Nd/144Nd vs. ~VSr/S6Sr diagram (Fig. 8 ) most of the investigated dykes plot in the enriched quadrant of the "mantle array". Note that the tephrite dyke plots in the depleted quadrant of the "mantle array". Few samples of the Cear~i-Mirim dykes plot well outside of "mantle array", having either high aTSr/S6Sr ratios (sample 6526) or quite low ~43Nd/144Nd (sample 6558). These plots in- dicate that the corresponding magmas possi- bly suffered interaction with different crustal materials.

In terms of Sr-Nd isotope compositions both the Jurassic and Cretaceous dykes from Cear~i- Mirim correspond quite well to the Lower Cre- taceous tholeiites flood basalts and dykes from northern Paran~i (high- and low-TiO2 types; Petrini et al., 1987; Piccirillo et al., 1989, 1990), as well as to the Lower Cretaceous tho- leiites intrusives (high-TiO2 type) of the Mar- anhgo basin (Sardinha Formation). In con-

22 G. BELLIENI ET AL,

¢r Na20 ~ ~ "

• - . .

'-' & IIII fl I IT I1 '~al i '~,~ CaO ;

# zx , ~

~ 4 - _ _ , ,% I* • J

ss SiO 2 @ . " " . . . . "',, "

47L 1 *L t 7 6 S 4 3

Mg0

O.E

O,E

O.z

0.2

CEAR~,-MrRIM MESOZOIC DYKES Middte Jurassic

tholeiitic dykes Lower Cretaceous

t holeiitic dykes • alkaline dykes 1~

MARANHAO i~) II~p

NORTHERN PARANA O .-j SOUTHERN PARANA ( i

fT0 .

7 6 5 4 3

MgO

Fig. 6. MgO (wt.%) vs. major (wt.%) elements for selected Ceara-Mirim Mesozoic dykes. Fields indicate other high- Ti02 (solid lines) and low-TiO2 (dashed lines) Brazilian Mesozoic basaltic rocks. Major elements recalculated to 100% on a volatile-free basis.

trast, the Cear~i-Mirim tholeiitic dykes are isotopically distinct from the Lower Jurassic tholeiites (low-TiO2 type) of the Maranh~o basin and those (low-TiO2 type ) of Early Cre- taceous age from southern Paran~i. The latter Jurassic and Cretaceous low-Tio2 basalts usu- ally have high 87Sr/86Sr ratios (> 0.707; Fig. 8 ) and trend towards the depleted quadrant of the "mantle array" (Petrini et al., 1987; Cor- dani et al., 1988; Piccirillo et al., 1989).

7. Petrogenetic aspects

Major- and trace-element variations of the Cear~i-Mirim tholeiitic dykes (high-TiO2 type) are, in general, compatible with frac- tional crystallization processes (Figs. 6 and 7 ). Mass-balance calculations (major elements) indicate that the most evolved dykes can be derived from the least evolved types through ~30-50% fractionation of plagioclase (4-

MAGMAT1C ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS R IFT ING 23

CEARA-MIRIM MESOZOIC DYKES Middle Jurassic

tholeiitic dykes Lower Cretaceous

thoLeiitic dykes • alkaline dykes

MARANNAO (~ '~P NORTHERN PARANA 0 SOUTHERN PARANA (~

150 L - _

l Ni . . . . . . . . . . " u ~ ~ ~-~---,,

50

30

•

j I;

Nb ¢r

a5

300

200

100

40

20

70

50

30

Nd • z,,

m • • i x

i x •

• ~d~ oo ' ~ , ~ • oq~e

,# ~, A

i i

7 o s 4 MgO

nOgc Ce ~~~~ie/~/

50 / x

30

40 • ,'.~ ~ . , , ~ e'~ .~ R,~ /'

20 ,,TLfI11~Th i~lXlT~;~, . ' ~ - - - - - "

8o Rb . - - . . . . . " - ,

4 0

800 t ,#

I ,,r" ooor ,dil

.dClU

. . . . . ~_ .~.~

900

700

500

SF

7 6 5 4 3

MgO

Fig. 7. MgO (wt.%) vs. trace (ppm) elements for selected Cear~i-Mirim Mesozoic dykes. Symbols as in Fig. 7.

22%), augite (3-8%), pigeonite (0-14%) and magnetite (1-4%) (L-ResZ=0.15-0.92). Cal- culated incompatible-element contents (Ray- leigh model) are, on average, moderately higher than those observed (i.e. calc./obs. 1.18 __+ 0.30; range 0.75-1.67 ). The largest dif-

ferences were obtained for Cr and Ni (calc./ obs. 0.81 +0.82; range 0.04-2.40). These are believed to be due, at least in part, to sulphide phase(s) not considered in the calculations. However, it is to be stressed that important differences in many element concentrations

24 G. BELLIEN1 ET AL.

CEAR~-'MIRIM MESOZOIC DYKES HTJ LTi Middle Jurassi{

th0teirt ic dykes [ ower Creraceous

\ , ~ , rh01eiir ic dykes • 05130 """ ~ te-rVgUtl e a lka l ine dykes "!~

T" i ' ~. ~ A b r o l h o s . . . . . . ~ ; ~ " " " "' ~'~,,~ Fern ndo

~ ' ~ ' - . ~ . - -~.. %,~%~, de Noroflha

E \ .,-,. "r ° i

"<;:L e -c ^ - / - - , , I . . . . . . ",,t R idge / / ~ . - * ~ ' ~ ' . - - ~ . ~ / ~ ' ~ , e - ".,~V . ,s x x • r

- ~ S o u t h e r n Parana '~ , \ K ~ l o w - T i O 2 ~ _ ) _ ~ - - L " I

NLN_qr t h e rn ParanA [ + " ~ ' h l g h : and~ow-T~O 2 ) ( e T s / e 6 e . . , , I

O TOZO o zo6o o z o s o o z oo

Fig. 8. Measured (m) 878r/86Sr vs. 143Nd/J44Nd ratios for Ceara-Mirim Mesozoic dykes, Maranh~o basin (Bellieni et al., 1990), Paran~i basin (Petrini et al., 1987; Piccirillo et al., 1989), Walvis Ridge, South Atlantic (Richardson et al., 1984), Abrolhos Islands, South Atlantic (Fodor et al., 1989), St. Paul Island, Indian Ocean (Roden et al., 1984) and Fernando de Noronha Islands, equatorial Atlantic Ocean (Gerlach et al., 1987). B.E. =bulk Earth; HTi and LTi=high- and low-TiO2, respectively.

exist for dykes with the same mg value (cf. Figs. 6 and 7; Table 4). This indicates that these dykes are probably related to different paren- tal magmas. The chemical differences also in- clude ratios between incompatible elements which are not substantially affected by frac- tional crystallization and melting processes (e.g., Zr/Ce 2.8-4.8, Zr /Nd 4.9-10.0; Fig. 9). Note that the elemental variations relative to Sr and Nd isotope ratios indicate that, in gen- eral, crustal contamination did not play an ap- preciable role in the genesis of the investigated tholeiites.

In general, there is consensus that the gene- sis of tholeiitic magmas requires high degrees of partial melting (e.g., > 10%; cf. Frey et al., 1978; Jaques and Green, 1980; Takahashi and Kushiro, 1983). If we assume a common source, the generation of the high-TiO2 tholei- ites requires garnet peridotite source(s) for explaining the chemistry and the important differences in the incompatible-element ra- tios. In this case, the melting degree would vary, at equilibrium conditions, in the range 10-25% (Bellieni et al., 1984; Piccirillo et al., 1988a). On the other hand, the major-element similar-

ity of the dykes suggests that the melting de- gree was similar for the Cear~i-Mirim high- TiO2 tholeiites. Then the differences in incom- patible elements could be more probably re- lated to small-scale chemically heterogeneous mantle source (s). Different melting degrees can be more suitable for explaining the genesis of the rate tholeiitic dykes low in incompatible elements and TiO2 (e.g., 25% melting), as well as that of the rare alkaline dykes (e.g., 5% melting).

On the whole, the genesis of the high-TiO2 Jurassic and Cretaceous tholeiitic dykes from Ceani-Mirim appears to be compatible with a time-integrated enriched mantle source sub- stantially homogeneous in terms of Sr-Nd iso- tope compositions, but characterized by small- scale chemical heterogeneity. In contrast, a time-integrated depleted mantle source is re- quired for the genesis of the alkaline dykes.

8. Comparison between the Ceard-Mirim dykes and other Brazilian Mesozoic basaltic rocks

The Jurassic and Cretaceous tholeiitic dykes from Cear~i-Mirim show (Figs. 6 and 7 ) sub-

MAGMAT1C ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFTING 2 5

CEAR~,-MIRIM MESOZOIC DYKES HTi LTi

_Middle Jurassic tholeiitic dykes A

Lower Cretaceous tholeiiti¢ dykes • alkaline dykes ~-

150

50

150

50

Ni

cr **

20

J Nd , ~ ,

6o La ~ . / . ~ '

Zr I00 200 300

•

~oo Ce ,~

1oo Rb ~ " ~

. /

Ba . ' " ~, z~ zx I

/ ,> /

800

600

400

200

700 I

500

Sr '~"

q]l~o ~Zr

100 200 300

Fig. 9. Zr (ppm) vs. other trace elements (ppm) for Cear~i-Mirim Mesozoic dykes.

stantial similarities, both in their major and trace elements with the high-TiO2 Lower Cre- taceous tholeiitic intrusives of the eastern MaranhSo basin (Sardinha Formation) and with the Lower Cretaceous high-TiO2 flood basalts and tholeiitic dykes from northern Pa- ran~ (Bellieni et al., 1984; Piccirillo et al., 1989b, 1990). On the other hand, all the inves- tigated tholeiitic dykes, except one low in TiO2,

are distinctly different (Figs. 6 and 7 ) from the flood tholeiites of the Maranh~o basin (Juras- sic) and southern Paranfi (Cretaceous) since the latter two suites are low in incompatible elements and TiO2.

In terms of Sr-Nd isotope compositions we emphasize that the Brazilian Mesozoic tholei- ites appear to be related to heterogeneous mantle sources. In general, the tholeiites high

26 G. BELLIENI ET AL.

100

"~ 50

o E

E 5

i I I I I I I I I I I

CEARA-MIRIM MESOZOIC OYKES Middle ]ura ssic

~ / ~ ~ tholeiRes. HTi A Lower Cretaceous

,i~ ,~X, tholeiites,LTi i~.'#l'l'i~/ 'k tephrite *

../ " " * - t 1~=0,7062

'¢i" Ro= 0 . 7 0 2 9 ~ . . ~ • ~A R : 0 . 7 0 5 4

, . , ~ i ~ 9 ~ %

• ~ , ~ ~ '0 ,% ",

4>

(ppm)

o

100

~50

n~ zE

0

E

E~5 n~

Northern Parmn~ lSouthern Paran~

, Ro= 0 7056

• , / Ro=07059

Ik~" , l i ,~ '~',/i. / .~, Ro= 0.7054 _ ¥ , / ' 5 , \ / - - -

# \ % I

~ -'e ^ ~ Ro=0.7058 ,'I, ? - . .

Ro=0.7030/"\,/ \ "~ - - - - - . o:O O66)0 f

<.,.> 5.3 43

I i I I I 1 I I i l l i I 1 I f I I I I I I

Rb Ba K Nb La Ce Sr P Zr Ti Y Rb Ba K Nb La Ce Sr P Zr Ti Y

(a) (b)

o • A • 5.5 4.5 59 4.1

(ppm)

Fig. 10. Sample/primordial mantle (Wood et al., 1979) elemental ratios for Cear~t-Mirim Mesozoic dykes, Maranh~o (Bellieni et al., 1990) and Paran~t (Petrini et al., 1987; Piccirillo et al., 1989) thole••tic basalts. HTi and LTi=high- and low-TiO2, respectively.

in incompatible elements and TiO2 (i.e. flows and intrusives from northern Parami, includ- ing the scarce rock-types low in TiO2; the rare high-T••2 flows and dykes from southern Pa- ran~i, the Sardinha Formation intrusives from the Maranh~o basin and the Ceani-Mirim dykes) are related to time-integrated enriched mantle, and usually plot in the "mantle array" (Fig. 8). In contrast, the thole•ires low in in- compatible elements and TiO2 (i.e. flows and dykes from southern Paran~ and the Maran- h~o basin) usually plot outside the "mantle ar- ray" and define trends which merge to time- integrated depleted mantle sources, similar to some Jurassic thole•ires of southern Paran~ and the Maranh~o basin (Fig. 8 ). Thus the genesis

of the Brazilian Mesozoic mafic magmatism does not appear to be related to an isotopically homogeneous mantle source as believed by Fodor (1987) and Fodor et al. (1989), but implies time-integrated enriched and depleted mantle source components. A third mantle component in the genesis of the low-TiO2 tho- leiites may be suggested if their deviation from the "mantle array" (i.e. high 87Sr/a6Sr ratios; Fig. 8) is due to melting of mantle contami- nated by ancient subduction crustal compo- nents (Hawkesworth et al., 1988) rather than to low-pressure crustal contamination docu- mented by Bellieni et al. (1984, 1986, 1990), Fodor et al. ( 1985, 1989), and Piccirillo et al. (1990). Note that mantle contamination by

MAGMATIC ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFTING 27

Ba/Y ,oo Ba

North ParanA 22 HTi ~ | ].r,~

, , /

o Z .+~;---.~"" L tholeiitic dykes A

/." I " t ~ E-MORB Lower Cre taceous ,

2 / I " ' S ° ~ 4 N-MORB- I -thole,,ticdyWes • 9

2 4 6 8 ~o ZrlY Fig. 11. Zr/Y vs. Ba/Y and Ba vs. Zr (inset) for Cear~-Mirim Mesozoic dykes, Maranh~o (Bellieni et al., 1990), Paran~ (Petrini et al., 1987; Piccirillo et al., 1989), Walvis Ridge (Erlank et al., 1984) and Southwestern Indian Ridge (E-MORB and N-MORB: Le Roex et al., 1983).

crustal components of subducted slabs system- atically requires appropriate and specific geo- logical processes for the South American plat- form, while basaltic magma interaction with the continental crust during low-pressure dif- ferentiation is perhaps inescapable. Thus it is probably not fortuitous that petrogenetic models indicate that the Brazilian Mesozoic low-TiO2 tholeiites are related to quite higher melting degree relative to the high-TiO2 ana- logues (e.g., 25% vs. 10%), and that the high- est 87Sr/86Sr ratios are found for low-TiO2 ba- salts associated with acid volcanics (e.g., Paran~ basin). The latter magmas are geneti- cally related with melting processes involving crustal materials (Bellieni et al., 1986, 1990; Petrini et al., 1987; Piccirillo et al., 1988a, 1989).

Spidergrams for the Brazilian Mesozoic ba- sic rock-types which can be considered uncon- taminated or poorly contaminated by the crust are shown in Fig. 10A. It is clear that both the high-TiOz Jurassic and Cretaceous tholeiites

from Cear~-Mirim have similar variation pat- terns, and are characterized by negative Nb anomaly, relative to K and La, while the Cre- taceous tephritic dyke shows a distinct posi- tive Nb anomaly. These above-mentioned Nb anomalies probably reflect different mantle source compositions consistent with impor- tant differences in Sr-Nd isotope values (Fig. 8 ). The variation pattern of the low-TiO2 dyke, mainly in terms of a primordial-mantle-nor- malized Ce/P ratio, is different from the other Cear~i-Mirim dykes (i.e. 0.76 vs. 1.57-1.61 ). The positive Sr anomaly may reflect the effect of"cumulus" plagioclase.

Spidergrams relative to other Brazilian Me- sozoic rocks (Fig. 10B) show that the high- TiO2 tholeiites from the MaranhSo and Para- n~i basins have variation patterns very similar to those of the Cear~i-Mirim dyke analogues, including the scarce low-TiO2 tholeiites from northern Paran~i. In contrast, the low-TiO2 tholeiites of the MaranhSo basin and southern Paran~i with the lowest initial 87Sr/86Sr ratios

28 G. BELLIENI ET AL.

are distinct from the other tholeiites in terms of the chondrite-normalized La/Ce and K/Nb ratios, respectively, which are less than the unity.

The correlations among Zr, Y and Ba (Fig. 11) show that the Mesozoic "uncontami- nated" tholeiites from Brazil cannot be easily related to mixing models restricted to two components, and do not support significant contribution of N- and E-types mid-ocean ridge basalt (MORB) (Le Roex et al., 1983 ) in two- component mixing models. Note that the low- TiO2Brazilian tholeiites partially overlap the Iow-Ba/Y end of the Walvis Ridge field, off Namibia, where also plot (not shown in Fig. 11 ) the samples of "Dupal" E-type MORB of the Southwest Indian Ridge (Le Roex et al., 1989).

On the whole, the data support the view that the Mesozoic Brazilian magmatism is proba- bly related to mantle components isotopically enriched and depleted, relative to the bulk Earth, and with variable contents of "en- riched" components introduced by "metaso- matic processes" (cf. Richardson et al., 1982; Erlank et al., 1987; Menzies et al., 1987).

9. Concluding remarks

(1) The Mesozoic magmatism in the east- ernmost part of NE Brazil (Cear~i-Mirim magmatism) is represented by a mafic dyke swarm trending E-W to NE-SW and com- posed of prevailing two-pyroxene tholeiites and rare alkaline rock-types (tephrite and trachybasalt ).

(2) K/Ar ages and palaeomagnetic data al- low the Cear~i-Mirim dykes to be distin- guished in sub-swarms of Middle Jurassic ( 179-161 Ma) and Late Jurassic-Early Cre- taceous ( 145-130 Ma) ages, respectively.

(3) The Cear~-Mirim dyke magmatism is essentially represented by evolved tholeiites with high content of incompatible elements and TiO2. No significant chemical differences exist among the Cretaceous dyke sub-swarms, while

the Jurassic one shows slightly lower concen- trations of incompatible elements (e.g., Zr, La, Ba, Y).

(4) Chemical and Sr-Nd isotope data indi- cate that crustal contamination did not play an appreciable role in the genesis of the Cear~l- Mirim basalts. The 87Sr/86Sr vs. 143Nd/t44Nd diagram reveals that most dykes plot in a nar- row field of the enriched quadrant of the "mantle array". The tephritic dyke is distinct in that it plots in the depleted quadrant of the "mantle array".

(5) Important chemical variations of the Cear~i-Mirim tholeiitic dykes require different parent magmas, probably related to garnet peridotite source (s) substantially homogene- ous in Sr-Nd isotopes but with small-scale chemical heterogeneity.

(6) Cear~i-Mirim dykes show close miner- alogical, chemical and isotopic (Sr, Nd) simi- larities with the Lower Cretaceous tholeiitic intrusives from the eastern Maranh~o basin (NE Brazil; Sardinha Formation), the flood basalts from northern Paran~ (SE Brazil ) and the scarce high-TiO2 basalt and dykes from southern Paran~i. In contrast, the investigated dykes are chemically and isotopically different from the flood basalts low in incompatible ele- ments (and TiO2) from the Maranh~o basin (Jurassic) and those from southern Paran~i (Lower Cretaceous ).

(7) Chemical and Sr-Nd isotope data sug- gest that in Brazil the Mesozoic basaltic mag- matism high in incompatible elements is re- lated to time-integrated enriched mantle, while that low in incompatible elements requires also a time-integrated depleted mantle component. The related subcontinental mantle heteroge- neity is believed to be due to variable content of "enriched" components introduced by "metasomatic" processes.

(8) The Jurassic dykes from Cear~i-Mirim correspond quite well to those from the Benue trough (and Liberia, Morocco and NW Ni- geria), and can be in general associated to the rifting processes of the central North Atlantic

MAGMATIC ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFTING 29

opening. The younger age ( ~ 10 Ma) for the Cearfi-Mirim (and Benue trough) Jurassic magmatic rocks relative to that of West Africa ( e.g., Liberia and Morocco ) may reflect an ad- vanced stage of the rifting processes prior the formation of oceanic crust.

(9) The Lower Cretaceous basaltic magma- tism of the Cearfi-Mirim corresponds quite well to the coeval one of the Benue trough. The palaeomagnetic data indicate that the Cearfi- Mirim dykes are older (up to Upper Jurassic) than the dykes of the Paran~i basin. Therefore, they may be probably related to the early rift- ing events of the Equatorial Domain of the South Atlantic (Popoff, 1988)whose oceanic crust formed between ~ 120 Ma (Pindell, 1988) and ~ 100 Ma (Popoff, 1988). Accord- ing to Popoff ( 1988 ) the time elapsed from the co-rift to oceanization stages in Equatorial Do- main of the South Atlantic was ~ 20-25 Ma.

Acknowledgements

The authors benefitted from the financial support of CNPq, FAPESP, FINEP (Brazilian agencies), and MPI and CNR (Italian agen- cies). A special thank to A. Cundari and M. Popoff for the helpful discussions, suggestions and critical review of the manuscript. The au- thors wish to thank G. Mezzacasa, A. Giaretta and P. Da Roit (University of Padova), and R. Zettin (University of Trieste) for their pre- cious collaboration in the analytical and tech- nical work. O.A. Figueiredo Filho is acknowl- edged for field assistance.

References

Aguiar, G.A. and Nahass, S., 1969. General geologic map of the Maranh~o basin, scale 1 : 1,000.000. Petrobras, Renor-Direx, Rio de Janeiro, Rep. No. 371.

Asmus, H.E. and Guazzelli, M., 1981. Descri9~o sumaria das estruturas da margem continental brasileira e das areas oceanicas e continentais adjacentes, hipoteses sobre o tectonismo causador e implica~oes para os prognosticos de recurcos minerais. Projeto Remac 9: Estruturas e tectonismo de margem continental brasi- leira, e suas implica~oes nos processos sedimentares e

na aval ia~o do potencial de recursos minerais - - Re- latorio final. Petrobras, Cenpes/Dintep, Rio de Ja- neiro, pp. 187-269.

Bellieni, G., Piccirillo, E.M. and Zanettin, B., 1981. Clas- sification and nomenclature of basalts. I.U.G.S. (Int. Union Geol. Sci.) Subcomm. on Systematics of Ig- neous Rocks, Circ. 34, Contrib. 87, pp. 1-19.

Bellieni, G., Brotzu, P., Comin-Chiaramonti, P., Ernesto, M., Melfi, A.J., Pacca, I.G., Piccirillo, E.M. and Stolfa, D., 1983. Petrological and paleomagnetic data on the plateau basalts to rhyolite sequences of the southern Paran~i basin (Brazil). An. Acad. Bras. Cienc., 55: 355- 383.

Bellieni, G., Comin-Chiaramonti, P., Marques, L.S., Melfi, A.J., Nardy, A.J.R., Piccirillo, E.M. and Roisenberg, A., 1984. High- and low-TiO2 flood basalts from the Paran~i plateau (Brazil): petrology and geochemical aspects bearing on their mantle origin. Neues Jahrb. Mineral., Abh,, 150: 273-306.

Bellieni, G., Comin-Chiaramonti, P., Marques, L.S., Melfi, A.J., Nardy, A.J.R., Papatrechas, C., Piccirillo, E.M., Roisenberg, A. and Stolfa, D., 1986. Petrogenetic as- pects of acid and basaltic lavas from the Paran~i pla- teau (Brazil); geological, mineralogical and petro- chemical relationships. J. Petrol., 27: 915-944.

Bellieni, G., Piccirillo, E.M., Comin-Chiaramonti, P., Melfi, A.J. and Da Roit, P., 1988. Mineral chemistry of continental stratoid volcanics and related intrusives from the Paran~i basin (Brazil). In: E.M. Piccirillo and A.J. Melfi (Editors), The Mesozoic Flood Volcanism of the Parana Basin: Petrogenetic and Geophysical As- pects. Inst. Astron. Geofis. Univ. ofS~o Paulo, pp. 73- 92.

Bellieni, G., Piceirillo, E.M,, Cavazanni, G., Petrini, R., Comin-Chiaramonti, P., Nardy, A.J.R., Civetta, L., Melfi, A.J. and Zantedeschi, P., 1990. Low- and high- TiO2 Mesozoic tholeiitic magmatism of the Maranh~o basin (NE-Brazil): K/Ar age, geochemistry, petrol- ogy, isotope characteristics and relationships with Me- sozoic low- and high-TiO2 flood basalts of the Paran~i basin (SE-Brazil). Neues Jahrb. Mineral., Abh., 162: 1-33.

Bellieni, G., Petrini, R., Piccirillo, E.M., Cavazzini, G., Civetta, L., Comin-Chiaramonti, P,, Melfi, A.J., Ber- tolo, S. and De Min, A., 1991. Proterozoic mafic dyke swarms of the Silo Francisco Craton (SE~Bahia State, Brazil): petrology and Sr-Nd isotopes. Eur. J. Min- eral., 3: 429-449.

Bon, E.H., 1962. Relatorio preliminar sobre trabalho geo- logico executado no sul do Cear~i. Div. Geol., Sudene (unpublished).

Bowden, P., Kinnaird, J.A., Abba, S.I., Ike, E.C. and Tur- aki, U.M., 1984. Geology and mineralization of the Nigerian anorogenic ring complexes. Geol. Jahrb., 56: 1-65.

Buddington, A.F. and Lindsley, D.H., 1964. Iron-tita-

30 O. BELLIENI ET AL.

nium oxide minerals and synthetic equivalents. J. Pe- trol., 5: 310-357.

Carmichael, I.S.E., 1967. The iron-titanium oxides and their associated ferro-magnesian silicates. Control. Mineral. Petrol., 14: 36-64.

Colby, J.W., 1972. Magic IV, a computer program for quantitative electron microprobe analysis. Bell Tele- phone Laboratories, Inc., Allentown, PA.

Cordani, U.G., 1970. Idade do vulcanismo no Oceano At- lantico Sul. Bol. Inst. Geoc. Astron., (S~o Paulo Univ. ), 1: 9-75.

Cordani, U.G., 1972. Arquivo Direx-Petrobras, Renor, Rio de Janeiro (unpublished).

Cordani, U.G., Civetta, L., Mantovani, M.S.M., Petrini, R., Kawashita, K., Hawkesworth, C.J., Taylor, P., Cavazzini, G. and Piccirillo, E.M., 1988. Isotope geo- chemistry of flood volcanics from the Paran~i basin (Brazil). In: E.M. Piccirillo and A.J. Melfi (Editors), The Mesozoic Flood Volcanism of the Paranfi Basin: Petrogenetic and Geophysical Aspects. Inst. Astron. Geofis., Univ. ofS~.o Paulo, pp. 157-178.

Dalrymple, G.B., Gromme, C.S. and White, R.W., 1975. Potassium-argon age and paleomagnetism of diabase dikes in Liberia: initiation of Central Atlantic rifting. Geol. Soc. Am. Bull., 86:399-411.

De La Roche, H., Leterrier, P., Grandclaude, P, and Mar- chal. M., 1980. A classification of volcanic and plu- tonic rocks using Rt-R, diagram and major-element analyses - - Its relationships with current nomencla- ture. Chem. Geol., 29: 183-210.

Ebert, H. and Bronchini, M.F,, 1968. Estudos estratigra- ficos e geocronologicos no escudo cristalino brasileiro. Cienc. Cult., 20:621-625.

Erlank, A.J., Waters, F.G., Hawkesworth, C.J., Haggerty, S.E., Allsopp, H.L., Richard, R.S. and Menzies, M.A., 1987. Evidence for mantle metasomatism in perido- lite nodules from Kimberley pipes, South Africa. ln: M.A. Menzies and C.J. Hawkesworth (Editors), Man- tle Metasomatism. Academic Press, London, pp. 221- 309.

Ernesto, M. and Pacca, I.G., 1988. Paleomagnetism of the Paran~i basin flood volcanics, southern Brazil. In: E.M. Piccirillo and A.J. Melfi (Editors), The Mesozoic flood volcanism of the Paranfi basin: petrogenetic and geo- physical aspects, Inst. Astron. Geofis., $8o Paulo Univ., pp. 229-255.

Erneslo, M., Furtatdo, M.N., Macedo, J.W.P., Martins, G. and Pacca, I.G., 1992. Paleomagnetic data from a Mesozoic dyke swarm in northeastern Brazil. Geo- phys. J. Int. (in press).

Fodor. R.V., 1987. Low- and high-TiO2 flood basalts of southern Brazil: origin from picritic parentage and a common mantle source. Earth Planet. Sci. Lett., 84: 423-430.

Fodor, R.V., Corvin, C. and Roisenberg, A., 1985. Petrol- ogy ofSerra Geral (Paran~i) continental flood basalts,

southern Brazil: crustal contamination, source mate- rial and South Atlantic magmatism. Contrib. Mineral. Petrol., 91: 54-65.

Fodor, R.V., Mukasa, S.B., Gomes, C.B. and Cordani, U,G., 1989. Ti-rich Eocene basaltic rocks, Abrolhos platform, Offshore Brazil, 18°S: petrology with re- spect to South Atlantic magmatism. J. Petrol., 30: 763- 786.

Frey, F.A., Green, D.H. and Roy, S.D., 1978. Integrated models of basalt petrogenesis: a study of quartz lholei- ires to olivine melilitites from south eastern Australia utilizing geochemical and experimental petrological data. J. Petrol., 19:463-513.

Gerlach, D.C., Stormer, Jr., J.C. and Mueller, P.A., 1987. Isotopic geochemistry of Fernando de Noronha. Earth Planet. Sci. Lett., 85: 129-144.

Gomes, J.R.C., Gatto, C.M.P.P., Souza, G.M.C. de, Luz, D.S. da, Pires, J.L., Teixeira, W., Franca, F.A.B. de, Cabral, E.M.A., Menor, E.A., Monteiro, M., Barros, M.J.G., Ribeiro, E.G., Lima, E.A. de and Fonseca, R.A. da, 1981. Geologia das Folhas SB-24/25, Jaguaribe e Natal. Proj. Radambrasil, Rio de Janeiro, MME/Secr. Geral, Lev. Rec. Nat., 23: 27-300.

Hasui, H., Hennies, W.T. and lwanuch, W., 1976. Idade do magmatismo basico Mesozoico do Brasil central. XXIX Congr. Bras. Geol., pp. 181-182 (abstract).

Hawkesworth, C.J., Mantovani, M.S.M. and Peate, D., 1988. Lithosphere remobilization during CFB mag- matism. In: Oceanic and Continental Lithosphere: Similarities and Differences. J. Petrol., Spec. Vol., pp. 205-224.

Horn, P., Mueller-Sohnius, D. and Schult, A., 1988. Po- tassium-argon ages on a Mesozoic tholeiitic dyke swarm in Rio Grande do Norte, Brazil, Rev. Bras. Geoc., 18: 50-53.

Ishii, T., 1975. The relations between temperature and composition of pigeonite in some lavas and their ap- plication to geothermometry. Mineral. J., 8: 48-57.

Jacques, A.L. and Green, D.H., 1980. Anydrous melting of peridotite at 0-15 Kb pressure and the genesis of tholeiitic melts. Contrib. Mineral. Petrol., 73:287-310.

Kretz, R., 1982. Transfer and exchange equilibria in a portion of the pyroxene quadrilateral as deduced from natural and experimental data. Geochim. Cosmo- chim. Acta, 46:411-421.

Kudo, A.M. and Weill, D.F., 1970. An igneous plagioclase thermometer. Contrib. Mineral. Petrol., 26: 52-65.

Le Roex, A.P., Dick, H.J.B., Erlank, A.J., Ried, A.M., Frey, F.A. and Hart, S.R., 1983. Geochemistry, mineralogy and petrogenesis of lavas erupted along the southeast Indian ridge between Bouvet triple junction and 11 degrees east. J. Petrol., 24: 267-318.

Le Roex, A.P., Dick, H.J.B. and Fisher, R.L., 1989. Pe- trology and geochemistry of MORB from 25 ° E to 46 ° E along the southwest Indian ridge: evidence for con- trasting style of mantle enrichment. J. Petrol., 30: 947- 986.

MAGMATIC ACTIVITY RELATED TO MIDDLE JURASSIC AND LOWER CRETACEOUS RIFFING 31

Macdonald, G.A. and Katsura, T., 1964. Chemical com- position of Hawaiian lavas. J. Petrol., 5: 82-133.

Mathez, E.A., 1973. Refinement of the Kudo-Weill pla- gioclase thermometer and its application to basaltic rocks. Contrib. Mineral. Petrol., 41:61-72.

Mellini, M., Carbonin, S., Dal Negro, A. and Piccirillo, E.M., 1989. Tholeiitic hypabyssal dykes: how many clinopyroxenes? Lithos, 22:127-134.

Menzies, M.A., Rogers, N.W., Tindle, A. and Hawkes- worth, C,J., 1987. Metasomatic and enrichment pro- cesses in lithospheric peridotites, an effect of asthen- osphere-lithosphere interaction. In: M.A. Menzies and C.J. Hawkesworth (Editors), Mantle Metasomatism. Academic Press, London, pp. 221-309.

Mysen, B.O. and Kushiro, I., 1977. Compositional varia- tions of coexisting phases with degree of melting of peridotite in the upper mantle. Am. Mineral., 62: 843- 865.

Nunes, A.B., Lima, R.F., Negreiros, B. and Filho, C., 1973. Geologia da Folha SB23 Teresina e parte da Folha SB24 Jaguaribe, Proj. Radam, 2: 1-30.

Papike, J.J., Cameron, K. and Baldwin, K., 1974. Amphi- boles and pyroxenes: characterization of other than quadrilateral components and estimates of ferric iron from microprobe data. Geol. Soc. Am., 6: 1053-1054.

Petrini. R., Civetta, L., Piccirillo, E.M., Bellieni, G., Comin-Chiaramonti, P., Marques, L.S. and Melfi, A.J., 1987. Mantle heterogeneity and crustal contamination in the genesis of low-Ti continental flood basalts from the Parana plateau (Brazil): Sr-Nd isotope and geo- chemical evidence. J. Petrol., 28: 701-726.

Piccirillo, E.M., Comin-Chiaramonti, P., Bellieni, G., Civetta, L., Marques, L.S., Melfi, A.J., Petrini, R., Ra- poso, M.1.B. and Stolfa, D., 1988a. Petrogenetic as- pects of continental flood basalt-rhyolite suites from the Paran~ basin (Brazil). In: E.M. Piccirillo and A.J. Melfi (Editors), The Mesozoic Flood Volcanism of the Parana Basin: Petrogenetic and Geophysical Aspects. Inst. Astron. Geofis., Univ. of Silo Paulo, pp. 179-206.

Piccirillo, E.M., Comin-Chiaramonti, P., Melfi, A.J., Stolfa, D., Bellieni, G., Marques, L.S., Giaretta, A., Nardy, A.J.R., Pinese, J.P.P., Raposo, M.I.B. and Ro- isenberg, A., 1988b. Petrochemistry of continental flood basalt-rhyolite suites and related intrusives from the Paranfi basin (Brazil). In: E.M. Piccirillo and A.J. Melt: (Editors), The Mesozoic Flood Volcanism of the Paranfi Basin: Petrogenetic and Geophysical Aspects. Inst. Astron. Geofis., Univ. ofS~o Paulo, pp. 107-156.

Piccirillo, E.M., Civetta, L., Petrini, R., Longinelli, A., Bellieni, G., Comin-Chiaramonti, P., Marques, L.S. and Melfi, A.J., 1989. Regional variations within the Paran~i flood basalts (southern Brazil): evidence for subcontinental mantle heterogeneity and crustal con- tamination. Chem. Geol., 75: 103-122.

Piccirillo, E.M., Bellieni, G., Cavazzini, G., Comin- Chiaramonti, P., Petrini, R., Melfi, A.J. Pinese, J.P.P.,

Zantedeschi, P. and De Min, A., 1990. Lower Creta- ceous tholeiitic dyke swarms from the Ponta Grossa Arch (southeast Brazil ): petrology, Sr-Nd isotopes and genetic relationships with the Paran~i flood volcanics. Chem. Geol., 89: 19-48.

Pindell, J.L., Cande, S.C., Pittman, W.C., Powley, D.B., Dewer, J.F., Brecque, J. and Haxby, W., 1988. A plate- kinematic framework for models of Caribbean evolu- tion. Tectonophysics, 155:121-138.

Popoff, M., 1988. Du Gondwana/: l'Atlantique sud: les connexions du foss6 de la B6nou6 avec les bassins du nord-est br6silien jusqu'~ l'ouverture du golfe de Gui- n6e au Cr6tace inf6rieur. J. Afr. Earth Sci., 7: 409-431.

Popoff, M., Kampunzu, A.B., Coulon, C. and Esquevin, J., 1982, D6couverte d'un volcanisme m6sozoique dans le NE du Nigeria: datations absolues, caractbres mag- matiques et significations dans l'6volution du rift de la B6nou6. In: M. Popoff and J.J. Tiercelin (Editors), Rifts et foss6s anciens. R6s. Comm. Tray. Lab. Sci. Terre St.-J6rome, Marseille, B 19: 47-49.

Priem, H.N.A., Boelrijk, N.A.I.M., Verschure, R.H., He- beda, E.H., Verdurmen, E.A.T.H. and Bon, E.H., 1978. K-Ar dating of a basaltic layer in the sedimentary La- vras basin, northeastern Brazil. Rev. Bras. Geoc., 8: 262-269.

Rahaman, M.A., Bennet, J.B., Van Breemen, O. and Bow- den, P., 1984. Detailed age study of the migration of younger granites ring complexes in northern Nigeria. J. Geol,, 92: 173-184.

Richardson, S.H., Erlank, A.J., Duncan, A.R. and Reid, D.L., 1982. Correlated Nd, Sr and Pb isotope varia- tion in Walvis Ridge basalts and implications for the evolution of their mantle source. Earth Planet. Sci. Lett., 59: 327-342.

Richardson, S.H., Erlank, A.J., Reid, D.L and Duncan, A,R., 1984. Major and trace elements and Nd and Sr isotope geochemistry of basalts from Deep Sea Drill- ing Project Leg 74 Walvis Ridge Transect. Init. Rep. Deep Sea Drill. Proj., 74: 739-754.

Roeden, M.H., Hart, S.R., Frey, F.A. and Melson, W.G., 1984. Sr, Nd and Pb isotopic and REE geochemistry of St. Paul's rocks: the metamorphic and metasomatic development of an alkali basalt mantle source. Con- trol. Mineral. Petrol., 85: 376-390.

Rolff, P.A.A., 1965. O pico vulcanico do Cabugi, Rio Grande do Norte. Dep. Nac. Miner. Div. Geol. Min- eral., Notas Prelim., Rio de Janeiro, Estudos, 126: 1- 26.

Sial, A.N., 1976. The post-Paleozoic volcanism of north- east Brazil and its tectonic significance. An. Acad. Bras, Cienc., 48:299-311.

Sial, A.N., Long, L.E., Pessoa, D.A.R. and Kawashita, K., 1981. Potassium-argon ages and strontium isotope geochemistry of Mesozoic and Tertiary basaltic rocks, northeastern Brazil. An. Acad. Bras. Cienc., 53:116- 122.

32 G. BELLIENI ET AL.

Takahashi, E. and Kushiro, I., 1983. Melting of a dry per- idotite at high pressure and basalt magma genesis. Am. Mineral., 68: 859-879.

Vannucci, R., Calvino, F., Cortesogno, L. and Tolomeo, L., 1989. Jurassic findings in Sokoto State (NW-Ni- geria). J. Afr. Earth Sci., 9: 245-258.

Westphal, M., Montigny, R., Thuizat, R., Bardon, C., Bossert, A., Hamzeh, R. and Rolley, J.P., 1979. Pal6o- magnetisme et datation du volcanisme Permien, Tria-

sique et Cr6tace du Maroc. Can. J. Earth Sci., 16:2150- 2164.

Wood, D.A., Joron, J.L., Treuil, M., Norry, M. and Tar- ney, J., 1979. Elemental and Sr isotope variations in basic lavas from Iceland and the surrounding ocean floor. Control. Mineral. Petrol., 70:319-339.

Zanettin, B., 1984. Proposed new chemical classification of volcanic rocks. Episodes, 7:19-20.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.