METAMORPHIC DIFFERENTIATION IN SOME BIOTITE SCHISTS FROM NORTH-EAST ANTRIM, IRELAND

Absect r Biotite in some biotite grade pelitic rocks from north-east Antrim, Ireland, is concentrated along S2 cleavage zones and is absent, or present only in small amounts, in areas between the cleavage zones. Thus, the SA2 cleavage zones form biotite-rich domains, while the intervening areas form bionte-poor domains in the rocks. I t is shown that the various models proposed to explain metamorphic differentiation, involving passive concentration of more dense, less soluble phases by solution and selective removal of more soluble phases, cannot be used to accountfor tkdifferentiated feature observed in the studied rocks. T o explain the segregation and the mechanism of metamorphic differentiation, i t is suggested that biotite had preferentially nucleated and grown in the S cleavage zones. 2


Introduction
The term metamophic differentiation denotes collectively the various processes by which contrasted mineral assemblages develop from a rock in which the minerals are more or less uniformly distributed.It is also used in a wider sense t o include p w t h of porph oblasts and reaction between

4"
compositionally contrasting adjacent layers?.The differentiation of a metamorphic rock into mineralogical and/or chemically different portions is contrary to the generally accepted concept that metamorphism usually tends to obliterate compositional differences 9f premetamorphic textures.schmidt8 attempted to explain metamorphic differentiation purely on mechanical grounds.He suggested a process of tectonic unmixing similar to that found in rolled wrought iron, which is partially unmixed during deformation into layers of relatively pure iron alternating with carbon-rich iron.According to him, shearing of rocks causes more ductile minerals (eg.micas and amphiboles) to concentrate in layers of intense shear, whereas less ductile minerals (eg.quartz and feldspars) tend t o remain in Iess sheared portions of the rock.~i c h o l s o n ~ described metamorphic differentiation associated with crenulation cleavages in quartz-mica schists which showed concentrations of quartz at the crests of crenulations and development of mica-rich zones at the limbs of the crenulations.He suggested that migation of silica from areas of high strain and recrystallization of quartz at sites of minimum strain develop this type of metamorphic differentiation.Talbot and ~o b b s ~ and ~r a ~~ &lso described metamorphic differentiation associated with discrete crenulation cleavages.Gray produced analytical data to support his conclusion that the discrete .crenulationcleavages were solution planes and he suggested that the differentiated nature was caused by passive concentration of relatively insoluble materials, such as micas and chlorite, due to solution and removal of quartz along the cleavages.
Hyndman4 discussed the fornation of quartzo-feIdspathic layers parallel to schiotosity and axial surfaces of folds in a series of silt and shale beds.He proposed that the metamorphic differentiation process involves migration of some compounds (eg. to form quartz and feldspars) to lower pressure surfaces parallel to schistosity and migration of other components less of this tendency to the higher pressure sites (eg. to form mafic minerals) during metamolphism and deformation.
In this paper, metamorphic differentiation associated with crenulation cleavages in some biotite schists is described and on the basis of'petrographic data, it is proposed that biotite may have nucleated and grown preferentially in S2 cleavage zones.'It is also shown that the above described processes such as solution, recrystallization and tectonic unmixing cannot explain this differentiated feature.

Geological setting of the study area
The-samples examined in the present study were collected from the Dalradian rocks of north-east Antrim, Ireland (Figure 1).These rocks represent biotite grade metamorphism and have been affected by three phases of deformatiun.The first deformation phase (Dl ) produced a slaty cleavage (S1 ) and the second phase deformed these rocks intensively and produced the regional schistosity which is a discrete crenulation cleavage (S2).The third deformation phase D3 produced small-scale folding and crenulation of S2 to form c

Petrography
The examined sampies are dark grey, fine grained rocks with abundant whitelgrey albite porphyroblasts which gave a spotted appearance to these rocks.S2 crenulation cleavage is well developed but S3 is weak or may be absent.
The petrogrqphic microscope reveals that the rocks are composed of albite, ~nuscovite, biotite, quartz, chlorite, apatite, tourmaline and ore minerals an5 the modal percentages of the minerals indicate that they belong to the category of pelites.
Albite .alaite porphyroblasts (best seen under a microscope) are abundant in the studied rocks.A characterisric feature of the albite porphyrobIast is that +hey contain numerous trails of inclusions which define various microstructures including S2 and S crenulations.In s6me specimens, albite porphyroblasts sre reversely zoned at the rim.A much larger area which included nort'l-east At;ltrim was studied and the age relationships of albite porphyroblast growth established.The albite porphyroblast grew during two static periods following D2 and D, phases of defor~nation.~Biotite : Two types of biotite crystals occur: fabric-forming type and slightly larger crystals (microporphyroblasts).The Yabric-forming type crystals are dimensionally oriented parallel to S1 and S2 cleavages but they are  Chlorite : primary chlorite is generally absent in the rocks studied.However, when primary chlorite is present, it occurs as dimensionally aligned crystals or the cryytals may be clustered together.
Quartz : The specimens described in this study contain little or no quartz.Therefore, the biotite-rich domains as well as the biotite-poor domains are poor in quartz.

Discussion
The petrographic study reveals that the rocks studied have been differentiated, on a microscopic scale, into biotite-rich and biotite-poor domains and thebiotite-rich domains coincide with S2 cleavage zones while the biotite-poor domains lie between them.Both types of damains contain little or no quartz.Also, albite which occurs as porphyrobIasts show no definite spatial relationship with kspect to the two domains.In this respect, the above described segregation differs horn the commonly reported differentiation associated with crenulation cleavage (Figure 3).schmidt's8 tectonic unmixing model cannot be used to account for the observed textural and mineralogical feature because the minerals inv~lved are mainly biotite and muscovite micas which are considered to have similar mechanical propel-ties.~i c h o l s o n ~ and ~r a ~~ considered that the segregation associated 4 t h crenulation cleavage was formed by selective solution and removal of silica resulting in the concentration of minerals such as chlorite, micas and amphiboles.However, there is no petrographic evidence in the studied rocks for selective solution and recrystallization of silica.Hyndman's4 model explains the contrasted mineral banding formed by quartz and feldspam on one hand and mafic minerals on the other hand but the model fails t o explain adequately the observed segregation involving m d y biotite and muscovite micas.
There is no convincing textural evidence in the rocks indicating the nature of the biotite forming reaction.However, on the basis'of mineralogy of the rocks and general absence of primary chloride in them, it is suggested that the biotite producing reaction is of the form, muscovitel + chlorite -t biotite + muscovite, + quartz + H 2 0 Therefore, if the biotite producing phases were segregated along the SZ cleavage zones'due to any one or more of the above discussed mechanical or chemical processes, then the growth of biotite would have produced the observed differentiation.However, this explanation seems t o be unlikely since chlorite-bearing, biotite-free rare rocks do not show any segregation of crilorite and muscovite along the S2 cleavages.
To explain the differentiation, it is suggested on the basis of petrographic evidence that biotite is preferentially nucleated and grown in S2 cleavage zones where the phyllosilicates have their highest interval strain energy.Ease of nucleation and diffusion of biotite forming components towards the nuclei in the highly strained regions, reduce the probability of nucleation of biotite in other regions.As the nucleation and growth of :tiotite proceed, these regions become separated from the other regions.RecentIy, ~ramwell' has described concentration of kyanite crystals in S3 cleavage zones in some pelitic rocks from Switzerland (Figure 4).The kyanite crystals in the S3 zone's are compositionally zoned with respect t o F e 2 0 3 .He has used textural and analytical data to show that kyanite first nucleated in the S 3 zones and later in the regions outside these zones.BramwelI suggested that the differentiation was produced by preferred nucleation and growth of kyanite crystals in the Sg zones.Thus preferred nucleation and growth of minerals in highly strained regions may be an important process causing differentiation.

Figure 1 :
Figure 1 : Location of the study area