American Mineralogist, Volume 62, pages 966-978, 1977
Relationof nucleationand crystal-growthrate to the developmentof granitic textures
Senrurl E. SwnNsoNl
Tuttle-JahnsLaboralory for Experimental Petrology
and Geology Department, Stanford Uniuersity
S t anfo rd, Cal ifo rnia 94305
Abstract
Analysis of crystal growth as a function of temperaturefor synthetic granite and granodiorite compositions in the system KAlSigOa-NaAlSirO'-CaAlrsiros-Sio, under HzO-saturated and undersaturatedconditionsyields quantitativedata on the growth kineticsofquartz,
alkali-feldspar,and plagioclasein thesesystemsat 8 kbar and 400-900"C. Measuredgrowth
rates vary from 3 X l0 6 cm/sec (about 3 mm/day) to I X l0-'0 cmlsec (about I mm/yr),
while nucleationdensity(nucleationsites/unit volume) variesfrom 0 to over I X l08sites/cm3.
Crystal-growthratescommonly increasewith increasedundercooling.Maximum growth rates
are lower in systemswhich contain a HzO-rich vapor phasethan in systemsthat are undersaturatedwith respectto HrO.
Number of nuclei (nucleation density), growth rate, and morphology for each of the
minerals determinesthe texture of an igneous rock. Results of this study show that long
periods of time are not necessaryto produce the texturescommonly associatedwith igneous
rocks. A relativelysmall number of large crystalsmay be produced by crystallizationat a low
undercoolingin a few days. Volcanic (fine-grained)texturesmay be produced by crystallization at large undercoolings,which produces alarge number of small nuclei (high nucleation
density, low growth rate). Coincidenceof low nucleation density and high growth rate for
alkali feldspar may explain the large alkali-feldsparcrystals(phenocrystsor interstitial masses) sometimesfound in granodiorites.
Introduction
Theory
Development of textures in igneous rocks has received relatively little study compared to various aspects of phase equilibria. Crystallization of magma
involves the kinetics of nucleation, and crystal
growth and useful concepts concerning these variables were developed many years ago (Tammann,
1925) to describe the crystallization of organic liquids. Empirical use of crystallizationkinetics(Dowty
et al., 1974: Carmichael et al., 1974) places restrictions on some models of igneous textural evolution.
The purpose of this paper is to quantify nucleation
and crystal-growth kinetics for some specificgranitic
systems.A better understandingof crystallizationkinetics in geologicallyinterestingsystemsshould provide some models for the evolution of various isneous textures.
Theoretical analysis of nucleation and crystal
growth is the subject of a large amount of literature.
Tammann (1925) discussedthe theory of nucleation
and crystalgrowth, and outlined the various parameters involved in terms of, a probabilistic model involving a functional relation to pressure, temperature, and time. Many later studies deal with the
nucleation and growth of crystals on an atomistic
level.Although such studiesare very significant,data
on nucleation and crystal-growth kinetics are required for the system under study prior to developm e n t o f a t o m i s t i c m o d e l s .Q u a n t i t a t i v ed a t a o n n u cleation and crystal growth are available for only a
few systemsof geological importance, and the purpose of this section is to presenta brief introduction
to theories of nucleation and crystal growth as a
prelude to the resultsof crystal-growthexperiments.
Nucleation involvesthe addition of atoms or molecules to a nucleushaving the structure of the solid. If
the nucleusis larger than a certain critical size,it will
1 P r e s e n ta d d r e s s :G e o l o g y D e p a r t m e n t ,A p p a l a c h i a nS t a t e U n i versity,Boone,North Carolina 28608.
SWANSON: GRANITIC TEXTU RES
grow spontaneously,but if it is smaller than the critical size, the nucleus will be unstable.Critical size is
d e p e n d e n tu p o n t h e a m o u n t o f u n d e r c o o l i n g[ A 7 :
? ' ( l i q u i d u s ) - Z ( g r o w t h ) l a n d v a r i e sf r o m a n i n f i n i t e
size at liquidus temperatures to smaller sizes at
greater degreesof undercooling. Another factor in
the nucleation of crystalsfrom a melt is the mobility
of atoms and moleculesin the melt as measuredby
the diffusion coefficient. In glass-forming systems,
such as silicates, liquid diffusion coefficientsdrop
markedly with decreasingtemperature(Towers and
C h i p m a n , 1 9 5 7 ) .A s t h e t e m p e r a t u r ed r o p s b e l o w t h e
liquidus, nucleationwill increasefrom zero to a maximum at some undercooling.Diffusion rates are then
very low and nucleation decreaseswith further decrease in temperature. This explains the pattern of
nucleation in liquids where the liquid diffusivity decreaseswith temperature,and is consistentwith the
patterns observedin this study.
Crystal growth from a melt proceeds by attachment of chemicalspeciesfrom the liquid to the crystal
nuclei. Rate of growth is thus a function of mobility
of crystal-forming specieswithin the melt. Mobility,
as measuredby the diffusion coefficient,drops with
decreasingtemperature,and growth rate, like nucleation, is expectedto rise from zero at the liquidus to a
maximum at some undercooling and then decrease.
This is the expectedpattern of growth rate of crystals
f r o m a v i s c o u sl i q u i d ( T a m m a n n , 1 9 2 5 ) .F o r a n e x cellentreview of crystal growth in geologicalsystems,
the reader is referred to Kirkpatrick (1975).
Crystal morphology is a function of growth rate
( Y) and mobility of the slowestcrystal-formingcomp o n e n t i n t h e m e l t ( D ) ( K i r k p a t r i c k , 1 9 7 5 ) .D i f f u s i o n
coefficientsin a melt decreasewith increasedundercooling while, as previouslyexplained,crystal-growth
rate will increaseto a maximum and then decrease.
The net result is that D/Y decreaseswith increased
undercooling (Kirkpatrick, 1975). At small undercoolings D/Y is relatively large, and crystals are expected to have smooth, planar surfaces(Cahn, 1967;
K i r k p a t r i c k , 1 9 7 5 ) . W i t h i n c r e a s e du n d e r c o o l i n g
D/Y decreasescausing a breakdown of smooth,
planar crystal faces.Initially the breakdown of planar
crystal faces takes the form of widely-spacedprotuberanceson the crystal faces, and with increased
undercooling the spacing betweenthe protuberances
decreases.The pattern of crystal morphology is expectedto changefrom smooth, planar crystal facesat
low undercoolingsto spheruliticcrystals at large undercoolings (Keith and Padden, 1963; Kirkpatrick,
r9 7 5 ) .
967
Previous work
Researchon nucleation and crystal growth in silicate systems has been done primarily by material
scientistsin connection with the ceramic and glass
i n d u s t r i e s .S o m e o f t h e s e s t u d i e s ,e . g . M e i l i n g a n d
Uhlmann (1967), have used seedcrystalsin growthrate experimentsand thus are not concernedwith the
n u c l e a t i o no f c r y s t a l s .
A few crystal-growth studies at one atmosphere
pressure with anhydrous bulk compositions have
dealt with applications to naturally occurring crystafs. Vogt (1923)recognizedthe rapid growth rate (on
the order of I mm/min) of akermanite, diopside,
fayalite,and rhodonite from a silicateliquid. Winkler
(1947) measuredthe growth rate of nephelinein the
system NarCOr-Alros-SiOr-LiF. More recently,
Kirkpatrick (1974) studied the growth rate of pyroxene, anorthite, and melilite in the system
CaMgSirOu-CaAl,SiO..Gibb (1974)found the nucleation of plagioclasevaried as a function of temperature, time, and cooling rate. Kirkpatrick et al.
(1975) measuredthe rate of growth of diopside and
a n o r t h i t e f r o m m e l t s o f t h e i r o w n c o m p o s i t i o n ,a n d
also found that diopside nucleatesmore readily than
plagioclase.Extension of the crystal-growth studies
to high-pressuresystemsseemsnatural, but has been
tried by only a few researchers.
Most crystal-growth studies at elevatedpressures
have been concerned with textural relations rather
than nucleation and crystal-growth kinetics. Jahns
a n d B u r n h a m ( 1 9 5 8 ) g r e w c r y s t a l sf r o m a h y d r o u s
s i l i c a t el i q u i d f o r m e d b y m e l t i n g a s a m p l e o f n a tural pegmatite, as pictured in Wyllie (1963).
Mustart (1969) grew crystals of albite in the system NaAlSi.Or-NarSirOu-H20. Substitution of
KAlSisO8 or SiO, for NaAlSirO, allowed growth.of
potassiumfeldspar and quartz respectively(Mustart,
1969, 1972). Swanson et al. (1972) grew crystals of
plagioclase,alkali-feldspar, and quartz in hydrous
haplogranitic systems.Lofgren (1974a,b)has grown
plagioclase crystals in the system NaAlSirOrCaAl,Si,O,-H,O. Donaldson (19'14) studied the
texture of olivine crystals grown from feldspathic
peridotite and quartz-normative basalt systems.
Lofgren et al. (1975) have investigatedthe development of porphyritic texture in terrestrial and
lunar basaltic compositions. Naney (1975) has
grown biotite and clinopyroxene from haplogranitic
liquids. Thesestudiesshow the feasibilityof studying
crystal growth in silicatesystemsof geologic importance.
Kinetics of nucleation and srowth of alkali-feld-
SWANSON: GRANITIC TEXTURES
spars from a hydrous silicate liquid in the system
KAlSiaos-NaAlSirOr-HrO have been intensively
studied by Fenn (1972, 1973, 1976), following the
method of Mustart (1969). Experimentswere made
with both HrO-undersaturatedand saturatedsilicate
liquids. Growth rates were relatively high (up to 6.8
X l0-Gcm/sec). Nucleation lag times of up to 48
hours indicate that the measured growth rates are
m i n i m u m v a l u e s .G r o w t h r a t e s i n t h e l i q u i d p l u s
aqueous vapor assemblageswere lower than in the
undersaturatedbulk compositions.
The relation of crystal-growthrate and nucleation
density to igneoustextureswas describedby Swanson
(1974) in relation to a haplogranodiorite composit i o n . G r o w t h r a t e sr a n g e df r o m 1 . 4 X l 0 ' t o
I X
l0 10cm,/secfor the plagioclasecrystals, while the
quartz and potassiumfeldsparcrystalshad maximum
growth rates of about 2 X l0-? cm,/sec.Relative
positions of the growth rate and nucleation curves
were used to explain igneous textures.
effect of pressure,temperature,and water content on
the phase relations of these bulk compositions,and
Figure I shows the results of the 8000 bar experiments and the systemsselectedfor study.
Experimental method
Experimentswere made in internally-heatedpressure vesselsmodified considerablyfrom the designof
Yoder (1950). Temperatures were measured with
PtlPt-10 Rh thermocouples periodically calibrated
againstthe melting points of gold (1062.5'C) and
sodium chloride (800.5"C). Pressurecorrectionswere
made in thermocouple calibration using the data of
C l a r k ( 1 9 5 9 ) a n d A k e l l a a n d K e n n e d y ( 1 9 7 1 ) .T e m perature measurements are probably precise to
+lOoc, and the pressure measurementsmade with
m a n g a n i nc o i l s a r e t h o u g h t t o b e t l 0 0 b a r s .
Dehydrated silicate gels (Luth and Ingamells,
I965) were used as the starting materials.Gels were
d r i e d i n a v a c u u m o v e n p r i o r t o l o a d i n g .W a t e r d i s tilled, deionized, and boiled just prior to use was
Selection of bulk compositions
added to platinum or Ag.oPd,ocapsules (2.0 mm
Preliminary experiments (Swanson et al., 1972) O . D . ) s e a l e d a t o n e e n d . A c a l c u l a t e da m o u n t o f
had shown the feasibility of growing crystals from
dried gel was added and the capsulesealedby weldh y d r o u s s i l i c a t el i q u i d s i n t h e h a p l o g r a n i t i cs y s t e m s . ing. Capsuleswere then placed in a vacuum oven to
In order to assessthe role of anhydrousbulk compo- check for leaks and to distribute the fluid homosition on nucleation and crystal growth, synthetic geneouslythroughout the charge. Two capsulesof
granite and granodiorite compositions were selected the same composition were loaded and run in each
for study as representinga variety of phase assem- experimentto allow for the low probability of nucleblages.These compositionswere modeled in the sys- ation in some compositions ('experiment' refers to a
tem NaAlSi.Os-KAlSirOr-CaAlrSirOr-SiO, after the group of capsulesthat were contained in the same
'charge'refersto
granite and granodiorite in the hornblende-biotite pressurevesselat the sametime, and
s e r i e s o f N o c k o l d s ' ( 1 9 5 4 ) . T h e c o m p o s i t i o n s a r e t h e c o n t e n t o f a s i n g l ec a p s u l e ) .
given in Table l. Whitney (1975) has studied the
Capsuleswere allowed to homogenizefor not less
t h a n 9 6 h o u r s a b o v et h e l i q u i d u s a t l 0 0 0 o C ( F i g . l ) .
T a b l e l . C h e m i c a l c o m o o s i t i o n ss t u d i e d
Then the temperature was dropped to the isotherm
SynEhet ic
Synthetic
0xide
selectedfor nucleationand growth. Time required for
granite
granod iorite
the drop to the growth temperaturevaried with the
73.98
si02
7 0. 3 4
a m o u n t o f u n d e r c o o l i n gb u t w a s o n t h e o r d e r o f a
AI^O^
15.07
18.00
few minutes. During quench to the growth isotherm,
pressure could not be maintained isobaric due to
Ca0
1 .5 0
3.97
limitations of the pumping system,pressuredrops as
Na-O
3 . 75
4. 3 7
large as 2000 bars resulted. Following the drop in
K^O
5.70
3.32
temperature to the growth isotherm, pressure was
readjusted to 8000 bars. Capsules remained at the
Modified
normtive
growth isotherm for 24 to 144 hours, and the experiequivalent
ment was then quenched to room temperature by
zo.)
23.r
a
turning off the furnace power. This resulted in a
Or
34.0
1 9 .8
temperature drop of 300"C after l0 minutes; room
was reachedwithin 20 minutes.
temperature
Ab
32.O
3 7. 3
It may be argued that the pressure used in this
An
7.5
19.8
study (8000 bars) limits the applicability of the re-
SWA NSON,' GRANITIC TEXTURES
S Y N T H E T I CG R A N O D I O R I T E
S Y N T H E T I CG R A N I T E
pt+At+
969
!O+L
AtlcO+L
Pl+At+qO+L
w
Af+( O+L+V
aaa
Pl+Af+qO+L
aa
7 0H 2 O
Fig. l. Bulk compositions selectedfor crystal growth. (a) Temperature Dr wt 7o H,O phase-assemblagediagram for synthetic granite
( W h i t n e y , 1 9 7 5 ) .( b ) T e m p e r a t u r e u s . w t V o H r O p h a s e - a s s e m b l a gdei a g r a m f o r s y n t h e t i cg r a n o d i o r i t e ( W h i t n e y , 1 9 7 5 ) .
sults. This pressurewas selected:(l) to take advantage of the depressionof liquidus temperaturesat
high pressure,and (2) becausethe phaserelationsfor
these compositions were well known at this isobar
(Whitney, 1975). Reconnaissanceexperiments at
2000 bars suggestthe form of the curves shown on
Figures 2 and 3 is little changedby pressure,though
t h e a b s o l u t ev a l u e sm a v b e c h a n s e d .
R"rult,
Experimentswere done at 50o intervals from 900o
to 600'C and at l00o intervalsfrom 600'C to 400'C,
and are more successfulwith the granodiorite compositions than with the granite compositions.At higher
water contents the granite compositions often fail to
nucleate any crystals,and only the granite composition plus 3.5 weight percentwater yield sufficientdata
f o r a m e a n i n g l u la n a l y s i s .
Nucleation
After the capsuleswere opened, the chargeswere
examined with the microscope and the number of
nucleation sites were counted or estimated.Charges
and quartz grains were embedded in epoxy, and a
thin section was prepared with the quartz used to
gauge the thickness. Nucleation sites were then
counted in a portion of the charge,and the use of the
thicknessof the thin sectionresultedin an estimateof
the nucleation sites as a function of volume. In
charges with only a few nucleation sites the total
number of siteswere counted, and this was divided by
the averagevolume of the charges(about 0.032 cm3)
to obtain the nucleationdensity.Resultsof the nucleation,densitymeasurementsare shown on Figure 2 as
a function of temperature. Attempts to determine
rates of nucleation were made by determining the
nucleation density as a function of time for a given
isotherm and bulk composition. However, lack of
nucleation during the experiment made it very difficult to determinenucleationrates(Table 2). Thus the
number of nucleation sites present at the end of the
experiment was the only nucleation parameter that
could be routinely determined for all charges with
crystals. When nucleation-rate data could be obtained, they followed the general pattern shown by
the nucleation density on Figure 2.
When the chargesare examined, the probabilistic
nature of the nucleation is readily apparent. Variation in growth time probably accountsfor the considerable scatter in the nucleation density data; for ex-
SWANSON: GRANITIC TEXTURES
b
aT(oc)
o
Nucleation
Oonsity
(sitos/cm3)
lxlOl
.6,T("c)
D Pl- Pbgiocl.!.
O Af- Alkrli Foldspar
a
O- Ouenz
^aT{oc}
F i g . 2 . N u c l e a t i o n d e n s i t y a s a f u n c t i o n o f u n d e r c o o l i n g ( A Z ) ( a ) S y n t h e t i c g r a n i t e p l u s 3 . 5 w t V a H r O . ( b ) S y n t h e t i cg r a n o d i o r i t e
p l u s 6 . 5 w t V o H 2 O . ( c ) S y n t h e t i cg r a n o d i o r i t e p l u s l 2 w t V oH z O . ( c ) S y n t h e t i c g r a n o d i o r i t e p l u s l 2 w t V oH z O .
ample in short-durationexperimentsone charge
oftenshowsnucleation
whereas
theotherchargedoes
not containany crystals(Table2). Curvesfor nucleation densityshown on Figure 2 have been constructedfrom the maximumvaluesonly in an attempt
to escapethe scatter.Thesecurvesprobablyreflect
the form of the nucleationdensity,but due to the
considerable
scatterof data the absolutemaximum
and locationof the curvescannotbe determined.
Time lag betweenquenchto a growthisothermand
first appearance
of crystals(nucleationlag time) is
relatedto the proximityof thegrowthisothermto the
liquidusof the bulk composition.
For example,no
nucleationappearedin the run productsafter 144
hours at 850"C or 24 hours at 800'C for the bulk
compositionlabeledb on Figure 2, whereasan experimentof 48 hoursat 800'C producedcrystals.
Nucleationof a solid from a melt without the aid
of foreignmaterials,
suchasnucleior containerwalls,
nucleationand is contrasted
is termedhomogeneous
wherethe crystalsnuclenucleation
to heterogeneous
ate on some foreign substrate(Flemmings,1974).
nucleationis difficultto prove,since
Homogeneous
Capsulewalls somethe nucleiare submicroscopic.
times provide a substratefor spherulitenucleation.
When crystalsnucleateon capsulewalls,they producelargercrystalsthan the crystalsthat nucleated
within the charge,indicatingthat the largercrystals
(externallynucleatedcrystals)may nucleatebefore
the smallercrystalswithin the charge(internallynucleatedcrystals).Suchexternallynucleatedcrystals
are not usedin the nucleationor growth-ratecurves
of Figures2-3.
to act assitesfor
Vapor bubblesmight be expected
nucleation.Bubblesof an HrO-rich
heterogeneous
vapor phaseare presentin chargesof certainbulk
Presence
of thesebubblesdid not afcompositions.
fect the nucleationof eitherthe feldsparsor quartz.
SWANSON.' CRANITIC TEXTURES
971
b
lx
1'10-t
Growth
Rate
(cm./sec)
1x1O-t
lxlO-rc
aT(oc)
E Pl- Plagioclas€
O Af- Alkali Feldsp!r
A
O- Ouartz
200
^r("c)
F i g 3 C r y s t a l - g r o w t h r a t e s a s a f u n c t i o n o f u n d e r c o o l i n g ( A I ) . ( a ) S y n t h e t i c g r a n i t e p l u s 3 . 5 w t V oH r O . ( b ) S y n t h e t i c g r a n o d i o r i t e
p l u s 6 . 5 w L V oH 2 O . ( c ) S y n t h e t i c g r a n o d i o r i t e p l u s l 2 w t V o H z O .
Some very small crystals are observedon the inside
walls of the vapor bubbles,but theseare presumedto
be material that was dissolvedin the vapor and crystallized on the quench to room temperature.Bubbles
did not act as nucleationsitesfor feldsparsor quartz
and are not found in the center of any of thesecrystals.
Generalizations regarding the nucleation density
curves are difficult. Tammann (1925) showed that
nucleation would rise to a maximum and then fall
with increasedundercooling. This same nucleation
pattern is found for the granitic systems(Fig. 2). The
maximum density of nucleation sites is approximately the same in the granodiorite systen.i,whereas
the granite with 3.5 weight percent water has a maximum nucleation density an order of magnitude
g r e a t e r( F i g . 2 ) .
Within a melt the number of nucleiis dependent
upon temperature, and this fact may be used to explain the time betweenquench to a growth isotherm
and nucleation of crystals.Quench to a growth temperature causes a rapid decrease in temperature
w i t h i n t h e s y s t e m ,a n d t h e d i s t r i b u t i o no f n u c l e im u s t
readjust to this new temperature before nucleation
can start (Turnbull, 1948). This period of readjustment may explain the nucleation lag time.
Table 2. Example of experimental results from the synthetic
g r a n o d i o r i t ew i t h 6 . 5 w t 1 o H , A c r y s t a l l i z e da t 8 0 0 ' C ( A Z : 1 5 0 " )
t-wtfr
l:_me (nrsl
24
24
48
48
96
Il+l+
*
Charge
tr
Growth
Rate
(cm,/sec)
}fucleation
Density
^
(sites/cm')
ro52
'I
NN+
?i *x
NN
NN
?1
^<?
r790
L79I
LZO6
z. I xlv-o
Z. 8xt0-z
2.IrJ.u--7
1l+61
NN - No nucl-eation
re+ ilnrrcs.nrris
f,o
'l
nrcleetion
site
ner
eharpe
972
SWANSON: GRANITIC TEXTURES
Crystal-growth rqte
When the chargeswere examined after the experiment, the maximum amount of crystal growth was
d e t e r m i n e d ,a n d t h i s w a s d i v i d e d b y t h e a m o u n t o f
time allowed for crystal growth to obtain the growth
rate. Measurementof the amount of crystal growth
for spheruliteswas relatively simple, since the center
of the spherulite where crystallization started could
be easilyidentified.For chargescontaining individual
crystalsone half of the crystal length is assumedto be
the amount of growth. The largest problem in this
method is a lack of data on the exact length of time
involved in the crystal growth. For example, if a
charge does not nucleate any crystals in a 24 hour
experiment but does nucleate crystals in a 48 hour
experiment, what is the time of growth? All of the
growth in the 48 hour run could have occurred in the
last 6 hours or even 6 minutes. Becausegrowth times
are not exactly known, the growth rates on Figure 3
represent minimum values. At the highest growth
rates,crystalsoften hit the side ofthe capsuleor other
crystals,and this interfereswith the length of growth
and acts to reduce the calculatedgrowth rate.
Growth-rate cupvesshown on Figure 3 are drawn
to include all the data points from internally nucleated crystals.Scatter in the growth-rate data can be
clearly seenon Figure 3. Growth rates measuredon
externally nucleated crystals were greater than the
growth rates for internally-nucleatedcrystals; this
supports the idea of faster nucleation (and hence
longer growing time) for the crystals nucleated on
foreign materials.
Growth-rate data suffer largely from a lack of
k n o w l e d g ec o n c e r n i n gt h e a m o u n t o f t i m e i n v o l v e d
in the growth of a given crystal, but other factors may
contribute errors. Interpretation of the textural relations can introduce another source of error in determination of growth rate. For example, in a semiradial arrangement of crystals, did a given crystal
start growth at one end or did it start growth in the
center and grow in both directions? The assumed
answer will make a factor-of-2 difference in the
growth rate.
Severalgeneralizationscan be made with regard to
the growth rates measured in this study. As the
amount of undercooling increases,the growth rate
for an individual phaseincreasesto a maximum and
(Fig. 3). This is the expectedbehavior
then decreases
of growth rate measured in a viscous liquid (Tamm a n n , 1 9 2 5 ) .W i t h a g i v e n a n h y d r o u sb u l k c o m p o s i tion, the maximum growth rate decreasesas the
amount of water in the system increases(Fig. 3).
Fenn (1976) also observed a decreasein the maximum growth rate with increasingwater content, and
this relation is as yet unexplained.Addition of water
to a silicatesystemlowers the viscosityof the silicate
l i q u i d ( F r i e d m a ne t a l . , 1 9 6 3 ; S h a w , 1 9 6 3 ) .A g r e a t e r
mobility of components and hence a faster crystal
growth rate is expected as a result of lowered viscosity. Decreasein crystal-growth rate with increasing water content is not compatible with the petrogenetic theory equating the growth of large crystals
with the evolution of a vapor phase(Jahnsand Burnham, 1958).
A note of caution must be issuedabout the plots
s h o w n o f F i g u r e s2 a n d 3 . W i t h i n s p h e r u l i t e si ,n d i v i d ual phasesare very difficult and sometimesimpossible to identify. Spherulitic growth is dominant at
small undercoolingsin the granite composition and is
also found within the granodiorite composition.
Given this limitation on phase identification in
spherulites,the resultsshown of Figures 2 and 3 are
probably more accuratefor the granodiorite than the
granite systems,and more reliable at larger undercoolingswithin the granite system.However, it is felt
that the results presentedon Figures 2 and 3 represent a vast improvement over nucleationand growthrate pfots that are schematic (Dowty et al., 1974;
Carmichael et al., 1974) despite any limitations on
t h e a b s o l u t en u m b e r s .
Textural relations
Texture describesthe relationsofcrystals and glass
in these experiments in terms of size, shape, and
arrangement.Crystal morphology refersto the shape
of individual crystals and, as explainedpreviously,is
controlled by the growth rate and the diffusivity of
the least mobile component in the system. Crystal
morphology should thus vary with undercooling
(Kirkpatrick, 1975). Since nucleation and growth
rate also vary with undercooling, textures are expected to vary with undercooling.
A radial array of crystals is the most commonly
observed texture in the hypersolidus region of the
b u l k c o m p o s i t i o n ss t u d i e d .T h e r a d i a l a r r a y m a y i n volve the large crystals found in the granodiorite
composition at low undercoolings(Fig. 4) or classical
spherulite development seen in the synthetic granite
and any combination in between. Indeed no clear
boundary existsbetweenany of the crystal morphologies observed in this study. With an increase in
undercooling, large single crystals give way to an
open array of coarse skeletal to dendritic crystals,
and finally a more compact radial mass of coarse
SWANSON: GRANITIC TEXTURES
973
E
: 5 0 " C , c r y s t a l l i z a t i o nt i m e 1 4 4h r s '
F i g . 4 . p h o t o m i c r o g r a p h so f s y n t h e t i cg r a n o d i o r i t ep l u s 6 . 5 w t 7 oH , O . ( a ) P l a g i o c l a s ec r y s t a l s ,A T
W h i t e b a r i s l 0 m m . ( b ) p l a g i o c l a s e s p h e r u l i t e , A Z =1 5 0 ' C , c r y s t a l l i z a t i o n t i m e 2 4 h r s . W h i t e b a r i s l . 0 m m . ( c ) D e n d r i t i c p l a g i o c l a s e
w i t h s m a l l e r c r y s t a l so f a l k a l i c r y s t a l s ,A ? : 1 5 0 . C , c r y s t a l l i z a t i o nt i m e 1 4 4 h r s . W h i t e b a r i s 0 2 5 m m . ( d ) S p h e r u l i t eo f p l a g i o c l a s e
(
h
e
x
a
g
o
n a lp l a t e s ) a n d a l k a l i - f e l d s p a r
q
u
a
r
t
z
(
e
)
o
f
C
r
y
s
t
a
l
s
:
1
.
0
m
m
i
s
W
h
i
t
e
b
a
r
2
4
h
r
s
.
300"C, crystallizationtime
f " i d s p u r ,A I
( i l C r v s t a l so f q u a r t z , a l k a l i - f e l d s p a r a, n d p l a g i o c l a s e ,
( t a b u l a r c r y s t a l s ) ,A I = 3 0 0 . C , c r y s t a l l i z a t i o nt i m e 2 4 h r s . W h i t e b a r i s 0 . 2 5 m m
AZ : 350'C, crystallization time 24 hrs. White bar is 0 I mm
SWANSON' GRANITIC TEXTURES
crystalsresults(Fig. a). This samesequenceof crystal
morphologieshas been observedin other experimental studies (Lofgren 1974a,Fenn 1976) and in rocks
(Bryan, 1972)and is directly relatedto the amount of
u n d e r c o o l i n g( K i r k p a t r i c k , 1 9 75 ) .
Plagioclaseand alkali-feldspar crystals grown in
this study show the same growth forms as the alkalifeldsparsgrown by Fenn (1973, 1976). This is not
surprising in view of the slight structural difference
between the alkali and the plagioclase feldspars.
Crystals are elongatedparallel to the a axis with the
e l o n g a t es i d e sb o u n d e d b y ( 0 0 1 ) a n d t e r m i n a t e db y
the (100) face. Relative growth velocitiesalong the
axesareq)c)b.
Quartz crystals grown in this study commonly lie
on the (0001) plane, resulting in hexagonal plates,
although some crystalsare elongatedparallel to the c
axis. Relative growth rates for the different orientations of quartz are c ) a.
Radial massesshow growth along a preferredcrystallographic axis, low angle branching of crystals,
and a distinctiveextinction cross.AII theseproperties
are characteristicof spherulitic growth (Keith and
P a d d e n , 1 9 6 3 ) .D e v e l o p m e n to f s p h e r u l i t e si s t o b e
expectedfrom these melts due to the high viscosity,
slow growth rate, and low number of nuclei (Keith
and Padden, 1963). Even for the small crystals,
formed at large undercoolings (400-500'C), high
magnification revealsthe fibrous nature of the crystals. These small fibrous crystals probably represent
s m a l l s p h e r u l i t e sW
. i t h i n a c r y s t a l ,i n d i v i d u a l f i b e r s
have the same crystallographicorientation. Crystals
composed of individual fibers are not a common
feature of igneous rocks, and presumably with further heating the fiberswill anneal to form singlesolid
crystals.
The patterns of crystallization for the synthetic
granite and granodiorite are quite different. In the
granodiorite composition, crystallization of large
crystals is followed by dendritic morphologies,
spherulitecrystallization,and finally growth of small
fibrous single crystals with increased undercooling
(Figure 4). In the syntheticgranite the crystallization
of spherulites is followed by small fibrous crystals
with increasingundercooling,the large crystalsfound
in the granodiorite at small undercoolirrgsare not
found in the granite.
Temperature is a logical variable to explain the
change in crystal morphologies. Crystal morphologies show a progressivechange with undercooling;
on the basis of crystal growth in the system
NaAlSi,O.-CaAlrSi,O.-H,O Lofgren (1974a) pro-
posed a general relationship between crystal morphology and undercooling, uiz., that plagioclase
crystallized from a silicate melt will show a change
from tabular to skeletaland dendritic growth forms
and finally result in a spherulitic morphology with
increasedundercooling. Care was taken to describe
the crystal morphologies, and the gradual change
from one morphology to another was also noted,
although for sodium-rich bulk compositions the
plagioclase crystals showed a change from tabular
directly to spherulitic morphologiesapparentlywithout any intermediate growth forms. The pattern of
crystallization observed in the granodiorite in this
study is similar to the pattern suggestedby Lofgren.
He did not observevery small crystals,but this could
be attributed to a lack of sufficientundercooling.The
crystallization sequenceof spherulites to small fibrous crystalsobservedin the syntheticgranite omits
the tabular, skeletal,and dendritic morphologiesthat
Lofgren includesin his description of crystal-growth
forms.
Developmentof igneoustextures
Textures resulting from the crystallizationof a silicate melt are dependentupon the crystal nucleation
and growth rate, and theseparametersvary with the
temperature-pressurepath followed by the melt duri n g c r y s t a l l i z a t i o nA
. ddition of iron and magnesium
will add new phasesto thesesystemsand thus change
t h e k i n e t i c so f n u c l e a t i o na n d g r o w t h ( N a n e y , 1 9 7 5 ) ,
but the generalitiesdevelopedhere for feldsparsand
quartz should hold true for thesesystemsevenwith a
small quantityof mafic components.
Observationof growth of reasonablylarge crystals
(severalmillimeters)in a short period of time geologically (several days) is one of the most significant
resultsof this study. Such growth rates are found in
systems containing one crystalline phase (Mustart,
1972; Fenn, 1973, 1976: Lofgren, 1974b),but this is
the first time such growth rates have been measured
in the polyphasegranitic systems.Rapid crystallization of feldspar and quartz in thesesystemssuggests
reevaluationof the length of time requiredto produce
the large crystals of the granitic rocks. Increase in
water content produces a decreasein the maximum
apparent growth rate for the system(Fig. 3), and this
should lead to a reinterpretation of the role of the
vapor phase in producing large crystals.
Textures of igneousrocks are controlled largely by
t h e k i n e t i c so f n u c l e a t i o na n d g r o w t h o f t h e c r y s t a l s .
For the compositions used in this study, crystallization at small undercoolingsproduces relatively few
SWANSON: GRANITIC TEXTURES
nucleationsiteswith large crystals,as is characteristic
of plutonic rocks, whereas large undercoolingsproduce a large number of nucleation sites with small
crystals,as found in volcanic rocks. Obviously such
extremesof temperaturechangeare rarely found, and
more commonly, plutonic rocks crystallize from a
slightly undercooled melt that is continuing to cool.
This would result in a texture in which different
phaseswould have different grain sizes,dependingon
the nucleation and growth rates. Zoning would also
be expectedin phasesof variable composition. Volcanic rocks often contain phenocrystsset in a finegrained groundmass,and this texture has been interpreted as resulting from movement of a magma that
was crystallizing al. a low undercooling with a few
nucleation sites to an environment with a lower undercooling to produce the fine-grainedgroundmass.
Similar models have been developed to explain the
porphyritic textures in intrusive igneous rocks.
Combination of the experimentalresultsshown on
Figures2 and 3 (Fig. 5) suggestsan interestingmodel
for the developmentof porphyritic texturesin granodioritic compositions. At low undercooling temperatures,plagioclasehas a rapid growth rate and a low
nucleation density,which may lead to formation of a
few large, euhedral plagioclasecrystals. Continued
crystallization at greater undercoolingswould result
in theseplagioclasecrystalsbeing a substratefor new
growth, in addition to the growth produced on new
plagioclasenucleation sites. Plagioclasecrystals nucleatedearly in the crystallizationprocesswould have
large euhedral cores,whereascrystals that nucleated
later would have smaller,more irregular cores,due to
interferencewith other crystals and a slower growth
rate (Fig. 4).
Plagioclasedoes not commonly form phenocrysts
in granodiorites,and yet the high growth rate combined with a relatively small number of nuclei at low
undercoolingssuggestsideal conditions for their development.There may be severalreasonsfor the lack
of plagioclasephenocrystsin granodiorites.Crystals
fbrmed early in the cooling processmay be removed
from the liquid by some differentiation process.Decreasein growth rate coupled with an increasein the
number of nuclei may mask the early plagioclase
crystals during later crystallization. Finally, most
granodiorite magmas may never be entirely liquid,
but insteadconsistof a mixture of crystals(including
plagioclase)and liquid. In this case, crystallization
would not produce the few large plagioclasecrystals
predictedfrom Figure 5; insteadthe early crystallization would take place on a substrateof preexisting
975
plagioclasecrystals. However, in some dikes (Bateman and Wahrhaftig, 1966) plagioclasephenocrysts
are found in granodiorites,and theseoccurrencesare
easily explained using the information on Figure 5.
Development of alkali-feldspar phenocrysts in
granodiorites and adamelliteshas been discussedby
Bateman et al. (1963) and Kerrick (1969). Phenocrysts vary widely in size,the largestbeing over l0 cm
in length. Morphology of the phenocrysts ranges
from euhedral to anhedral in different plutons. Inclusions of other minerals are found within the
phenocrystsin zones paralleling the crystal faces of
the alkali-feldspar crystals. The mineral inclusions
appearto have beenpushedout in front of and finally
included in the advancing face of the alkali-feldspar
phenocryst, suggestingthat these phenocrystsdeveloped while there was at least some silicate liquid
remaining in the system.Movement of water in front
of an advancingalkali-feldsparcrystal face has been
describedby Fenn and Luth (1973),and the development of concentration gradients in front of an advancing crystal face seemsa reasonableexplanation
for the inclusions in the alkali-feldsparphenocrysts,
and supportsa model of crystallizationfrom a silicate
liquid.
D e v e l o p m e n t o f a l k a l i - f e l d s p a rp h e n o c r y s t s i n
granodiorites may be explained from the relation of
the nucldation density and growth rate curves for
alkali-feldsparshown on Figure 5. The growth rate of
the alkali-feldsparreachesa maximum while the nucleationdensity for this phaseis relativelylow. Development of a few large alkali-feldsparphenocrystsis
expectedfrom the granodioritic magma. Maximum
developmentof the alkali-feldsparphenocrystsis expected near the solidus temperatureunder thesepressure conditions (Figs. I and 5). Alkali-feldspar
phenocrysts may push other phases aside during
growth or include thesepreexistingphaseswithin the
phenocrysts.ln some casesthe magma may be nearly
solid when the alkali-feldsparstartsto crystallize,and
due to the limited space,thesephenocrystswould be
anhedral and even interstitial. Basedon the crystallization kinetics of alkali feldspar in granodioritic
magmas, phenocrystsof this late-crystallizingphase
are expectedin granodiorites.
Development of large crystalsof quartz in granodiorites is not expected.Maximum growth of quartz
correspondswith the maximum in nucleationdensity,
and this combination would produce a large number
of relatively small quartz grains. A similar argument
may be made for all of the crystalline phasesin the
syntheticgranite.
SIIANSON : GRANITIC TEXTURES
Crystallizationkinetics of the granite systemshave
been combined from Figures 2 and 3 and are shown
i n F i g u r e 5 . M a x i m u m g r o w t h r a t e sa n d n u c l e a t i o n
densitiescoincide for quartz and plagioclasein the
granite composition (Fig. 5). As in the caseof quartz
in the granodiorite system,coincidenceof high nucleation densitiesand growth rates resultsin crystallizat i o n o f m a n y r e l a t i v e l y s m a l l c r y s t a l s .T h e h i g h
growth rate of alkali-feldspar might be expectedto
produce large crystals in the granite system(Fig. 5).
Nucleation density is high for all phases, and the
growth rate of the alkali-feldspar is two orders of
magnitude greater than growth rates measured on
other crystalline phases.Under theseconditions the
alkali-feldsparis expectedto be much larger than the
other crystals. From these results, granites crystallized at low undercoolingsare expectedto form alkali-feldspar phenocrystswhile granites crystallized
a t l a r g eu n d e r c o o l i n gt e m p e r a t u r ew
s ill be equigranufar. Bateman et al. (1963) have found that the gran-
ites of the Sierra Nevada batholith in Californra are
equigranular, although Williams et al. (1954) find
alkali-feldsparphenocrysts common in granites.
Basedon the crystallizationkinetics shown in Figure
5, either equigranular or porphyritic rocks are expectedfrom granite systems,dependingon the undercoolings.
Conclusions
Textures of igneous rocks are largely dependent
upon the crystallizationkinetics of the system.Development of nuclei and the subsequentrate of growth
will determine the appearanceof individual crystalline phases.Since the texture of a rock is the mosaic
of mineral grains, igneous-rocktextureswill be determined by the crystal-growthkineticsof the individual
phases.Crystallization kinetics of a particular phase
vary from one bulk composition to another, and this
helps to explain why different igneous rocks sometimes show different textures.
-- -;=zi-1"I
I
{r
'Pl-----v
if
.A'T{oc)
I
I
l/
ti
tt
lt
Pl- Plagioclase
Al- Alkali Feldspar
O- Ouartz
Solid Lines- Growth Rates
Oashed Lines- Nucleetiog Oonsity
tiz
A'T(ocl
F i g . 5 . C o m b i n e d n u c l e a t i o n d e n s i t y a n d c r y s t a l - g r o w t h r a t e c u r v e s . ( a ) S y n t h e t i c g r a n i t e p l u s 3 . 5Hwr tOV o( b ) S y n t h e t i c g r a n o d i o r i t e
p l u s 6 . 5 w t V o H r O . ( c ) S y n t h e t i cg r a n o d i o r i t e p l u s 1 2 w t 7 o H r O
SWANSON: GRANITIC TEXTURES
977
Acknowledgments
This studyis concerned
with theconceptof time of
The author thanks W. C. Luth and P. M. Fenn for assistance
crystallization
and texturein igneousrocks.Conceptually, plutonicrocks are coarse-grained
becauseof and advice during the course of this study. Special thanks are due
to P. R. Gordon for technical assistance in the operation and
long periods of slow cooling and crystallization, maintenance of the high pressure equipment. P. M. Fenn, H. R.
whereasvolcanicrocks are fine grainedbecauseof Shaw, and an anonymous reviewer made many helpful comments
very rapid cooling and crystallization.Nucleation on an early version ofthis manuscript. This work was supported by
densityand growth-rateplots shown on Figure 5 N S F g r a n t s G A 1 6 8 4 ( t o R . H . J a h n s a n d W . C . L u t h ) a n d G A
offer an alternativeexplanationfor the development 4 1 7 3 1 ( t o W . C . L u t h , G . E . B r o w n a n d W . A . T i l l e r ) .
of plutonic and volcanic textures.Crystal-growth
References
rateswithin thesesystemsreachmaximumvalueson
Akella, J. and C. G. Kennedy (1971) Melting of gold, silver,and
per day.Obviously,at
theorderof severalmillimeters
copper-proposal for a new high-pressure calibration scale. "I.
this ratelong time periodsarenot neededto produce
Geophys Res., 76, 4969-4977.
the type of crystalscharacteristic
of plutonic rocks. B a t e m a n ,P . C , L . D . C l a r k , N . K . H u b e r , J . G . M o o r e a n d C . D .
R i n e h a r t ( 1 9 6 3 ) T h e S i e r r a N e v a d a b a t h o l i t h - a s y n t h e s i so f
within the
This does not imply that temperatures
recent work across the central part. U S. Geol. Suro. Prof. Pap.
earthfall so much[?'(liquidus) (solidus) 300o]
4t4- D.
in sucha short periodof time, but it doesillustrate
, a n d C . W a h r h a f t i g ( 1 9 6 6 )G e o l o g y o f t h e S i e r r aN e v a d a .
for the develthat long time periodsare not necessary
In E. H. Bailey, Ed., Geology of Northern California, p. 107-172.
opmentof plutonictextures.Rapidcrystallization
to
Cald. Diu. Mines Geol Bull. 190.
producea plutonictexturemay accountfor the plu- B r y a n , W . B . ( 1 9 7 2 ) M o r p h o l o g y o f c r y s t a l si n s u b m a r i n eb a s a l t s .
J G e o p h y s R e s, 7 7 , 5 8 1 2 - 5 8 1 9 .
tonic rocks that crystallizedat shallow depths in
Cahn, J. W (1967) On the morphological stability of growing
volcanic
(Tabor
1969).
some
areas
and Crowder,
c r y s t a l s .I n H . S . P e i s e r ,E d . , C r y s t a l G r o w t h , p . 6 8 1 - 6 9 0 . P e r Volcanictexturesresult when magmais quickly
gamon Press,Oxford.
subjectedto large undercoolingtemperaturesby ex- Carmichael, I. S. E., F. J. Turner and J Verhoogen(1974) Igneous
trusion. Figure 5 showsthat at large undercooling Petrology McGraw-Hill, New York.
values,nucleationdensityis high and growthrate is C l a r k , S P . , J r . ( 1 9 5 9 ) E f f e c t o f p r e s s u r eo n t h e m e l t i n g p o i n t s o f
e i g h t a l k a l i h a l i d e s .J . C h e m P h y s , 3 l , 1 5 2 6 - 1 5 3 1 .
low, whichproducesa largenumberof smallcrystals. D o n a l d s o n , C . H ( 1 9 7 4 ) E x p e r i m e n t a l i n v e s t i g a t i o n o f o l i v i n e
Volcanictexturesare producedby the crystallization m o r p h o l o g y ( a b s t r ) . G e o l . S o c . A m . A b s t r . P r o g r . , 6 , 7 1 2 .
values.
Dowty, E., K. Keil and P. Martin (1974) Lunar pyroxene-phyric
kineticsat largeundercooling
basalts: crystallization under supercooled conditions. J Petrol ,
Developmentof porphyritictexturesdependson
t5,4t9-454.
the associationof a low nucleationdensityand a
Fenn, P. M. (1972) Nucleation and growth of alkali feldsparsfrom
rapid growth rate. Phenocrystshave beendescribed s y n t h e t i cm e l t s ( a b s t r . ) . T r a n s .A m G e o p h y s .U n i o n , 5 3 , 1 1 2 7 .
as mineralsthat crystallizedearly and were carried (1973) Nucleation and Growth of Alkali Feldspars from
with the magmato a new environment,resultingin
Melts in the System NaAlSisOa-KAlShO"-H20. Ph.D. Thesis,
Stanford University, Stanford, California.
rapid crystallizationof the groundmass.However,
(1976) Nucleation and growth of alkali feldspar from hythe possibledevelopmentof phenocrystsof alkalidrous melts. Can. Mineral , in press.
feldsparin granodioriteshas been previouslydis- - ,
a n d W . C . L u t h ( 1 9 7 3 ) H a z a r d s i n t h e i n t e r p r e t a t i o no f
cussed,
and that modeldoesnot dependon any twoprimary fluid inclusions in magmatic minerals (abstr.). Geol.
stagecrystallizationprocess."Porphyritic" textures
Soc Am. Abstr.Progr.,5,617.
wereproducedin a numberof charges
in thisstudyin Flemmings, M. C. (1974) Solidifcation Processing.McGraw-Hill,
New York.
wherelargergrains of a
a one-stepcrystallization,
Friedman, L, W. Long and R. L. Smith (1963) Viscosity and water
faster-growingphasecoexistwith smallergrainsof a
content of rhyolite glass. "/. Geophys Res.,68,6523-6535.
slower growing phase(Figure 4). Developmentof G i b b , F . G . F . ( 1 9 7 4 ) S u p e r c o o l i n g a n d t h e c r y s t a l l i z a t i o n o f
plagioclasefrom a basaltic magma. Mineral. Mag., 39,641-653.
temperature-induced
zoning is found in many
J
a
h n s , R . H . a n d C . W . B u r n h a m ( 1 9 5 8 ) E x p e r i m e n t a ls t u d i e so f
phenocrysts.
In the absence
this
type
of
of corrobopegmatite genesis: melting and crystallization of granite and
ratingevidence,
thereis little reasonto believethat a
pegmatite (abstr.) Geol. Soc. Am. Bull.,69, 1592-1593.
porphyritictexturewasproducedduring a multistage K e i t h , H . D . a n d F . J . P a d d e n J r . ( 1 9 6 3 ) A p h e n o m e n o l o g i c a l
process.
crystallization
theory of spherulite crystallization. J Appl. Phys., 34,
Crystallizationkineticsinvolving the nucleation 2409-242t.
froma porphyritic
magacrysts
and growthof crystalsin magmasareimportanttools Kerrick,D. M. (1969)K-feldspar
quartzmonzonite,centralSierraNevada,California.Am. Minin understandingthe developmentof igneoustexeral..54. 839-848.
tures.Extensionof this type of studyto other systems Kirkpatrick, R. J. (1974)Kineticsof crystalgrowth in the system
will furnishvaluableinformationon the development CaMgSirO"-CaAlrSiO6.
Am. J. Sci.,274, 215-242.
(1975)Crystal growth from the melt: a review.Am. Minof texturesin other typesof igneousrock.
978
SWANSON, GRANITIC TEXTURES
e r a l . .6 0 . 7 9 8 - 8 1 4 .
C. R. Robinson and J. F. Hays (1975) Kinetics of crystal
growth from silicate melts: diopside and anorthite (abstr.). Geol
Soc Am Abstr. Progr , 7, 1147.
Lofgren, G. E. (1914a) An experimental study of plagioclasecrystal morphology: isothermal crystallization. Am. J. Sci., 274,
243-273.
(1974b) Temperature induced zoning in synthetic plagioc l a s ef e l d s p a r I n W . S . M a c K e n z i e a n d J . Z u s s m a n , E d s . , T h e
Feldspars,p. 362-376. Manchester Univ. Press, Manchester.
-.
R. J Williams. C. H Donaldson and T M. Usselman
(1975) Experimental investigation of porphyritic texture (abstr.).
Geol. Soc. Am. Abstr. Progr.,7, ll73-ll'74
Luth, W. C. and C. O. Ingamells (1965) Gel preparation for
hydrothermal experimentation. Am Mineral., 50, 255-258.
Meiling, G. S. and D. R. Uhlmann (1967) Crystallization and
melting krnetics of sodium disilicate. Phys. Chem. Glasses,8,
62-68.
M u s t a r t , D . A . ( 1 9 6 9 ) H y d r o t h e r m a l s y n t h e s i so f l a r g e s i n g l e
crystals of albite and potassium feldspar (abstr ). Trans. Am.
Geophys Union, 50,675
(1972) Phase Relations in the Peralkaline Pottion of the
System NarO-Al2Oa-SiO,-H20. Ph D. Thesis, Stanford University, Stanford, California
Naney, M. T. (1975) Nucleation and growth of biotite and clinopyroxene from haplogranitic magmas (abstr.) Trans Am
Geophys. Union, 56, 1075
Nockolds, S. R. (1954) Average chemical composition of some
i g n e o u s r o c k s . G e o l . S o c .A m . B u l l . , 6 5 , 1 0 0 7 - 1 0 3 2 .
S h a w , H . R . ( 1 9 6 3 )O b s i d i a n - H r O v i s c o s i t i e sa t 1 0 0 0a n d 2 0 0 0b a r s
i n t h e t e m p e r a t u r er a n g e 7 0 0 o t o 9 0 0 " C . J G e o p h y s .R e s . , 6 8 ,
-,
-6343.
6337
Swanson, S. E (1974) The effect of crystal growth rate and nucleation density on the teKture of granodiorites (abstr.). Geol Soc.
Am Abstr. Progr , 6,979,
-,
J. A. Whitney and W. C. Luth (1972) Growth of large
quartz and feldspar crystals from synthetic granitic liquids
( a b s t r ) . T r a n s .A m . G e o p h y s .U n i o n , 5 3 , l l 2 ' 7 .
Tabor, R. W and D F. Crowder (1969)On batholiths and volcanoes-intrusion and eruption of late Cenozoic magmas in the
Glacier Peak area, north Cascades, Washington. U.S. Geol.
Suru, Prof. Pap ,604
Tammann, G (1925) The States of Aggregation Van Nostrand,
New York.
T o w e r s , H a n d J . C h i p m a n ( 1 9 5 7 )D i f f u s i o n o f c a l c i u m a n d s i l i c o n
in a lime-alumina-silica slag. Trans. Am. Inst. Min. Eng., 209,
769-',773.
Turnbull, D (1948) Transient nucleation Trans Am. Inst. Min
Eng., 175,'174-783.
V o g t , J . H L . ( 1 9 2 3 ) T h e p h y s i c a lc h e m i s t r yo f t h e c r y s t a l l i z a t i o n
and magmatic differentiation of igneous rocks: VII viscosity,
i n f l u e n c eo f t i m e , u n d e r c o o l i n g( o r s u p e r s a t u r a t i o n ) ,a n d e q u i Iibrium of mix-crystals betweenthe solid and liquid phase,etc. J.
G e o l . .3 l . 2 3 3 - 2 5 2 .
W h i t n e y , J . A . ( 1 9 7 5 ) T h e e f f e c t so f p r e s s u r e ,t e m p e r a t u r e ,a n d
X^"o on phase assemblagein four synthetic rock compositions
J Geol.8J. l-31.
Williams, H., F. J. Turner and C. M. Gilbert (1955)Petrography;
An Introduction to the Study of Rocks in t..in Sections. Freeman,
San Francisco.
Winkler, H. G. F. (1947) Kristallogrosse und Abkuhlung. Heidelberger Beitr. Mineral. Petrol., I ,86-104.
W y l l i e , P . J . ( 1 9 6 3 ) A p p l i c a t i o n s o f h i g h p r e s s u r es t u d i e st o t h e
earth sciences.In, R. S. Bradley, Ed., High PressurePhysicsand
Chemistry, Vol. 2 Academic Press,New York.
Yoder, H. S., Jr. (1950) High-low quartz inversionup to 10,000
bars Trans Am Geophys Union,3l,827-835.
Manuscript receiued, February 4, 1976; accepted
for publication, March 31, 1977.